diff options
298 files changed, 13645 insertions, 8495 deletions
diff --git a/Documentation/DMA-ISA-LPC.txt b/Documentation/DMA-ISA-LPC.txt index b1a19835e907..c41331398752 100644 --- a/Documentation/DMA-ISA-LPC.txt +++ b/Documentation/DMA-ISA-LPC.txt @@ -42,7 +42,7 @@ requirements you pass the flag GFP_DMA to kmalloc. Unfortunately the memory available for ISA DMA is scarce so unless you allocate the memory during boot-up it's a good idea to also pass -__GFP_REPEAT and __GFP_NOWARN to make the allocater try a bit harder. +__GFP_REPEAT and __GFP_NOWARN to make the allocator try a bit harder. (This scarcity also means that you should allocate the buffer as early as possible and not release it until the driver is unloaded.) diff --git a/Documentation/DocBook/Makefile b/Documentation/DocBook/Makefile index 5fd8f5effd0c..60a17b7da834 100644 --- a/Documentation/DocBook/Makefile +++ b/Documentation/DocBook/Makefile @@ -13,7 +13,7 @@ DOCBOOKS := z8530book.xml \ gadget.xml libata.xml mtdnand.xml librs.xml rapidio.xml \ genericirq.xml s390-drivers.xml scsi.xml \ sh.xml regulator.xml w1.xml \ - writing_musb_glue_layer.xml iio.xml + writing_musb_glue_layer.xml ifeq ($(DOCBOOKS),) @@ -71,6 +71,7 @@ installmandocs: mandocs # no-op for the DocBook toolchain epubdocs: latexdocs: +linkcheckdocs: ### #External programs used @@ -272,6 +273,6 @@ cleandocs: $(Q)rm -rf $(call objectify, $(clean-dirs)) # Declare the contents of the .PHONY variable as phony. We keep that -# information in a variable se we can use it in if_changed and friends. +# information in a variable so we can use it in if_changed and friends. .PHONY: $(PHONY) diff --git a/Documentation/DocBook/deviceiobook.tmpl b/Documentation/DocBook/deviceiobook.tmpl deleted file mode 100644 index 54199a0dcf9a..000000000000 --- a/Documentation/DocBook/deviceiobook.tmpl +++ /dev/null @@ -1,323 +0,0 @@ -<?xml version="1.0" encoding="UTF-8"?> -<!DOCTYPE book PUBLIC "-//OASIS//DTD DocBook XML V4.1.2//EN" - "http://www.oasis-open.org/docbook/xml/4.1.2/docbookx.dtd" []> - -<book id="DoingIO"> - <bookinfo> - <title>Bus-Independent Device Accesses</title> - - <authorgroup> - <author> - <firstname>Matthew</firstname> - <surname>Wilcox</surname> - <affiliation> - <address> - <email>matthew@wil.cx</email> - </address> - </affiliation> - </author> - </authorgroup> - - <authorgroup> - <author> - <firstname>Alan</firstname> - <surname>Cox</surname> - <affiliation> - <address> - <email>alan@lxorguk.ukuu.org.uk</email> - </address> - </affiliation> - </author> - </authorgroup> - - <copyright> - <year>2001</year> - <holder>Matthew Wilcox</holder> - </copyright> - - <legalnotice> - <para> - This documentation is free software; you can redistribute - it and/or modify it under the terms of the GNU General Public - License as published by the Free Software Foundation; either - version 2 of the License, or (at your option) any later - version. - </para> - - <para> - This program is distributed in the hope that it will be - useful, but WITHOUT ANY WARRANTY; without even the implied - warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. - See the GNU General Public License for more details. - </para> - - <para> - You should have received a copy of the GNU General Public - License along with this program; if not, write to the Free - Software Foundation, Inc., 59 Temple Place, Suite 330, Boston, - MA 02111-1307 USA - </para> - - <para> - For more details see the file COPYING in the source - distribution of Linux. - </para> - </legalnotice> - </bookinfo> - -<toc></toc> - - <chapter id="intro"> - <title>Introduction</title> - <para> - Linux provides an API which abstracts performing IO across all busses - and devices, allowing device drivers to be written independently of - bus type. - </para> - </chapter> - - <chapter id="bugs"> - <title>Known Bugs And Assumptions</title> - <para> - None. - </para> - </chapter> - - <chapter id="mmio"> - <title>Memory Mapped IO</title> - <sect1 id="getting_access_to_the_device"> - <title>Getting Access to the Device</title> - <para> - The most widely supported form of IO is memory mapped IO. - That is, a part of the CPU's address space is interpreted - not as accesses to memory, but as accesses to a device. Some - architectures define devices to be at a fixed address, but most - have some method of discovering devices. The PCI bus walk is a - good example of such a scheme. This document does not cover how - to receive such an address, but assumes you are starting with one. - Physical addresses are of type unsigned long. - </para> - - <para> - This address should not be used directly. Instead, to get an - address suitable for passing to the accessor functions described - below, you should call <function>ioremap</function>. - An address suitable for accessing the device will be returned to you. - </para> - - <para> - After you've finished using the device (say, in your module's - exit routine), call <function>iounmap</function> in order to return - the address space to the kernel. Most architectures allocate new - address space each time you call <function>ioremap</function>, and - they can run out unless you call <function>iounmap</function>. - </para> - </sect1> - - <sect1 id="accessing_the_device"> - <title>Accessing the device</title> - <para> - The part of the interface most used by drivers is reading and - writing memory-mapped registers on the device. Linux provides - interfaces to read and write 8-bit, 16-bit, 32-bit and 64-bit - quantities. Due to a historical accident, these are named byte, - word, long and quad accesses. Both read and write accesses are - supported; there is no prefetch support at this time. - </para> - - <para> - The functions are named <function>readb</function>, - <function>readw</function>, <function>readl</function>, - <function>readq</function>, <function>readb_relaxed</function>, - <function>readw_relaxed</function>, <function>readl_relaxed</function>, - <function>readq_relaxed</function>, <function>writeb</function>, - <function>writew</function>, <function>writel</function> and - <function>writeq</function>. - </para> - - <para> - Some devices (such as framebuffers) would like to use larger - transfers than 8 bytes at a time. For these devices, the - <function>memcpy_toio</function>, <function>memcpy_fromio</function> - and <function>memset_io</function> functions are provided. - Do not use memset or memcpy on IO addresses; they - are not guaranteed to copy data in order. - </para> - - <para> - The read and write functions are defined to be ordered. That is the - compiler is not permitted to reorder the I/O sequence. When the - ordering can be compiler optimised, you can use <function> - __readb</function> and friends to indicate the relaxed ordering. Use - this with care. - </para> - - <para> - While the basic functions are defined to be synchronous with respect - to each other and ordered with respect to each other the busses the - devices sit on may themselves have asynchronicity. In particular many - authors are burned by the fact that PCI bus writes are posted - asynchronously. A driver author must issue a read from the same - device to ensure that writes have occurred in the specific cases the - author cares. This kind of property cannot be hidden from driver - writers in the API. In some cases, the read used to flush the device - may be expected to fail (if the card is resetting, for example). In - that case, the read should be done from config space, which is - guaranteed to soft-fail if the card doesn't respond. - </para> - - <para> - The following is an example of flushing a write to a device when - the driver would like to ensure the write's effects are visible prior - to continuing execution. - </para> - -<programlisting> -static inline void -qla1280_disable_intrs(struct scsi_qla_host *ha) -{ - struct device_reg *reg; - - reg = ha->iobase; - /* disable risc and host interrupts */ - WRT_REG_WORD(&reg->ictrl, 0); - /* - * The following read will ensure that the above write - * has been received by the device before we return from this - * function. - */ - RD_REG_WORD(&reg->ictrl); - ha->flags.ints_enabled = 0; -} -</programlisting> - - <para> - In addition to write posting, on some large multiprocessing systems - (e.g. SGI Challenge, Origin and Altix machines) posted writes won't - be strongly ordered coming from different CPUs. Thus it's important - to properly protect parts of your driver that do memory-mapped writes - with locks and use the <function>mmiowb</function> to make sure they - arrive in the order intended. Issuing a regular <function>readX - </function> will also ensure write ordering, but should only be used - when the driver has to be sure that the write has actually arrived - at the device (not that it's simply ordered with respect to other - writes), since a full <function>readX</function> is a relatively - expensive operation. - </para> - - <para> - Generally, one should use <function>mmiowb</function> prior to - releasing a spinlock that protects regions using <function>writeb - </function> or similar functions that aren't surrounded by <function> - readb</function> calls, which will ensure ordering and flushing. The - following pseudocode illustrates what might occur if write ordering - isn't guaranteed via <function>mmiowb</function> or one of the - <function>readX</function> functions. - </para> - -<programlisting> -CPU A: spin_lock_irqsave(&dev_lock, flags) -CPU A: ... -CPU A: writel(newval, ring_ptr); -CPU A: spin_unlock_irqrestore(&dev_lock, flags) - ... -CPU B: spin_lock_irqsave(&dev_lock, flags) -CPU B: writel(newval2, ring_ptr); -CPU B: ... -CPU B: spin_unlock_irqrestore(&dev_lock, flags) -</programlisting> - - <para> - In the case above, newval2 could be written to ring_ptr before - newval. Fixing it is easy though: - </para> - -<programlisting> -CPU A: spin_lock_irqsave(&dev_lock, flags) -CPU A: ... -CPU A: writel(newval, ring_ptr); -CPU A: mmiowb(); /* ensure no other writes beat us to the device */ -CPU A: spin_unlock_irqrestore(&dev_lock, flags) - ... -CPU B: spin_lock_irqsave(&dev_lock, flags) -CPU B: writel(newval2, ring_ptr); -CPU B: ... -CPU B: mmiowb(); -CPU B: spin_unlock_irqrestore(&dev_lock, flags) -</programlisting> - - <para> - See tg3.c for a real world example of how to use <function>mmiowb - </function> - </para> - - <para> - PCI ordering rules also guarantee that PIO read responses arrive - after any outstanding DMA writes from that bus, since for some devices - the result of a <function>readb</function> call may signal to the - driver that a DMA transaction is complete. In many cases, however, - the driver may want to indicate that the next - <function>readb</function> call has no relation to any previous DMA - writes performed by the device. The driver can use - <function>readb_relaxed</function> for these cases, although only - some platforms will honor the relaxed semantics. Using the relaxed - read functions will provide significant performance benefits on - platforms that support it. The qla2xxx driver provides examples - of how to use <function>readX_relaxed</function>. In many cases, - a majority of the driver's <function>readX</function> calls can - safely be converted to <function>readX_relaxed</function> calls, since - only a few will indicate or depend on DMA completion. - </para> - </sect1> - - </chapter> - - <chapter id="port_space_accesses"> - <title>Port Space Accesses</title> - <sect1 id="port_space_explained"> - <title>Port Space Explained</title> - - <para> - Another form of IO commonly supported is Port Space. This is a - range of addresses separate to the normal memory address space. - Access to these addresses is generally not as fast as accesses - to the memory mapped addresses, and it also has a potentially - smaller address space. - </para> - - <para> - Unlike memory mapped IO, no preparation is required - to access port space. - </para> - - </sect1> - <sect1 id="accessing_port_space"> - <title>Accessing Port Space</title> - <para> - Accesses to this space are provided through a set of functions - which allow 8-bit, 16-bit and 32-bit accesses; also - known as byte, word and long. These functions are - <function>inb</function>, <function>inw</function>, - <function>inl</function>, <function>outb</function>, - <function>outw</function> and <function>outl</function>. - </para> - - <para> - Some variants are provided for these functions. Some devices - require that accesses to their ports are slowed down. This - functionality is provided by appending a <function>_p</function> - to the end of the function. There are also equivalents to memcpy. - The <function>ins</function> and <function>outs</function> - functions copy bytes, words or longs to the given port. - </para> - </sect1> - - </chapter> - - <chapter id="pubfunctions"> - <title>Public Functions Provided</title> -!Iarch/x86/include/asm/io.h -!Elib/pci_iomap.c - </chapter> - -</book> diff --git a/Documentation/DocBook/iio.tmpl b/Documentation/DocBook/iio.tmpl deleted file mode 100644 index e2ab6a1f223e..000000000000 --- a/Documentation/DocBook/iio.tmpl +++ /dev/null @@ -1,697 +0,0 @@ -<?xml version="1.0" encoding="UTF-8"?> -<!DOCTYPE book PUBLIC "-//OASIS//DTD DocBook XML V4.1.2//EN" - "http://www.oasis-open.org/docbook/xml/4.1.2/docbookx.dtd" []> - -<book id="iioid"> - <bookinfo> - <title>Industrial I/O driver developer's guide </title> - - <authorgroup> - <author> - <firstname>Daniel</firstname> - <surname>Baluta</surname> - <affiliation> - <address> - <email>daniel.baluta@intel.com</email> - </address> - </affiliation> - </author> - </authorgroup> - - <copyright> - <year>2015</year> - <holder>Intel Corporation</holder> - </copyright> - - <legalnotice> - <para> - This documentation is free software; you can redistribute - it and/or modify it under the terms of the GNU General Public - License version 2. - </para> - </legalnotice> - </bookinfo> - - <toc></toc> - - <chapter id="intro"> - <title>Introduction</title> - <para> - The main purpose of the Industrial I/O subsystem (IIO) is to provide - support for devices that in some sense perform either analog-to-digital - conversion (ADC) or digital-to-analog conversion (DAC) or both. The aim - is to fill the gap between the somewhat similar hwmon and input - subsystems. - Hwmon is directed at low sample rate sensors used to monitor and - control the system itself, like fan speed control or temperature - measurement. Input is, as its name suggests, focused on human interaction - input devices (keyboard, mouse, touchscreen). In some cases there is - considerable overlap between these and IIO. - </para> - <para> - Devices that fall into this category include: - <itemizedlist> - <listitem> - analog to digital converters (ADCs) - </listitem> - <listitem> - accelerometers - </listitem> - <listitem> - capacitance to digital converters (CDCs) - </listitem> - <listitem> - digital to analog converters (DACs) - </listitem> - <listitem> - gyroscopes - </listitem> - <listitem> - inertial measurement units (IMUs) - </listitem> - <listitem> - color and light sensors - </listitem> - <listitem> - magnetometers - </listitem> - <listitem> - pressure sensors - </listitem> - <listitem> - proximity sensors - </listitem> - <listitem> - temperature sensors - </listitem> - </itemizedlist> - Usually these sensors are connected via SPI or I2C. A common use case of the - sensors devices is to have combined functionality (e.g. light plus proximity - sensor). - </para> - </chapter> - <chapter id='iiosubsys'> - <title>Industrial I/O core</title> - <para> - The Industrial I/O core offers: - <itemizedlist> - <listitem> - a unified framework for writing drivers for many different types of - embedded sensors. - </listitem> - <listitem> - a standard interface to user space applications manipulating sensors. - </listitem> - </itemizedlist> - The implementation can be found under <filename> - drivers/iio/industrialio-*</filename> - </para> - <sect1 id="iiodevice"> - <title> Industrial I/O devices </title> - -!Finclude/linux/iio/iio.h iio_dev -!Fdrivers/iio/industrialio-core.c iio_device_alloc -!Fdrivers/iio/industrialio-core.c iio_device_free -!Fdrivers/iio/industrialio-core.c iio_device_register -!Fdrivers/iio/industrialio-core.c iio_device_unregister - - <para> - An IIO device usually corresponds to a single hardware sensor and it - provides all the information needed by a driver handling a device. - Let's first have a look at the functionality embedded in an IIO - device then we will show how a device driver makes use of an IIO - device. - </para> - <para> - There are two ways for a user space application to interact - with an IIO driver. - <itemizedlist> - <listitem> - <filename>/sys/bus/iio/iio:deviceX/</filename>, this - represents a hardware sensor and groups together the data - channels of the same chip. - </listitem> - <listitem> - <filename>/dev/iio:deviceX</filename>, character device node - interface used for buffered data transfer and for events information - retrieval. - </listitem> - </itemizedlist> - </para> - A typical IIO driver will register itself as an I2C or SPI driver and will - create two routines, <function> probe </function> and <function> remove - </function>. At <function>probe</function>: - <itemizedlist> - <listitem>call <function>iio_device_alloc</function>, which allocates memory - for an IIO device. - </listitem> - <listitem> initialize IIO device fields with driver specific information - (e.g. device name, device channels). - </listitem> - <listitem>call <function> iio_device_register</function>, this registers the - device with the IIO core. After this call the device is ready to accept - requests from user space applications. - </listitem> - </itemizedlist> - At <function>remove</function>, we free the resources allocated in - <function>probe</function> in reverse order: - <itemizedlist> - <listitem><function>iio_device_unregister</function>, unregister the device - from the IIO core. - </listitem> - <listitem><function>iio_device_free</function>, free the memory allocated - for the IIO device. - </listitem> - </itemizedlist> - - <sect2 id="iioattr"> <title> IIO device sysfs interface </title> - <para> - Attributes are sysfs files used to expose chip info and also allowing - applications to set various configuration parameters. For device - with index X, attributes can be found under - <filename>/sys/bus/iio/iio:deviceX/ </filename> directory. - Common attributes are: - <itemizedlist> - <listitem><filename>name</filename>, description of the physical - chip. - </listitem> - <listitem><filename>dev</filename>, shows the major:minor pair - associated with <filename>/dev/iio:deviceX</filename> node. - </listitem> - <listitem><filename>sampling_frequency_available</filename>, - available discrete set of sampling frequency values for - device. - </listitem> - </itemizedlist> - Available standard attributes for IIO devices are described in the - <filename>Documentation/ABI/testing/sysfs-bus-iio </filename> file - in the Linux kernel sources. - </para> - </sect2> - <sect2 id="iiochannel"> <title> IIO device channels </title> -!Finclude/linux/iio/iio.h iio_chan_spec structure. - <para> - An IIO device channel is a representation of a data channel. An - IIO device can have one or multiple channels. For example: - <itemizedlist> - <listitem> - a thermometer sensor has one channel representing the - temperature measurement. - </listitem> - <listitem> - a light sensor with two channels indicating the measurements in - the visible and infrared spectrum. - </listitem> - <listitem> - an accelerometer can have up to 3 channels representing - acceleration on X, Y and Z axes. - </listitem> - </itemizedlist> - An IIO channel is described by the <type> struct iio_chan_spec - </type>. A thermometer driver for the temperature sensor in the - example above would have to describe its channel as follows: - <programlisting> - static const struct iio_chan_spec temp_channel[] = { - { - .type = IIO_TEMP, - .info_mask_separate = BIT(IIO_CHAN_INFO_PROCESSED), - }, - }; - - </programlisting> - Channel sysfs attributes exposed to userspace are specified in - the form of <emphasis>bitmasks</emphasis>. Depending on their - shared info, attributes can be set in one of the following masks: - <itemizedlist> - <listitem><emphasis>info_mask_separate</emphasis>, attributes will - be specific to this channel</listitem> - <listitem><emphasis>info_mask_shared_by_type</emphasis>, - attributes are shared by all channels of the same type</listitem> - <listitem><emphasis>info_mask_shared_by_dir</emphasis>, attributes - are shared by all channels of the same direction </listitem> - <listitem><emphasis>info_mask_shared_by_all</emphasis>, - attributes are shared by all channels</listitem> - </itemizedlist> - When there are multiple data channels per channel type we have two - ways to distinguish between them: - <itemizedlist> - <listitem> set <emphasis> .modified</emphasis> field of <type> - iio_chan_spec</type> to 1. Modifiers are specified using - <emphasis>.channel2</emphasis> field of the same - <type>iio_chan_spec</type> structure and are used to indicate a - physically unique characteristic of the channel such as its direction - or spectral response. For example, a light sensor can have two channels, - one for infrared light and one for both infrared and visible light. - </listitem> - <listitem> set <emphasis>.indexed </emphasis> field of - <type>iio_chan_spec</type> to 1. In this case the channel is - simply another instance with an index specified by the - <emphasis>.channel</emphasis> field. - </listitem> - </itemizedlist> - Here is how we can make use of the channel's modifiers: - <programlisting> - static const struct iio_chan_spec light_channels[] = { - { - .type = IIO_INTENSITY, - .modified = 1, - .channel2 = IIO_MOD_LIGHT_IR, - .info_mask_separate = BIT(IIO_CHAN_INFO_RAW), - .info_mask_shared = BIT(IIO_CHAN_INFO_SAMP_FREQ), - }, - { - .type = IIO_INTENSITY, - .modified = 1, - .channel2 = IIO_MOD_LIGHT_BOTH, - .info_mask_separate = BIT(IIO_CHAN_INFO_RAW), - .info_mask_shared = BIT(IIO_CHAN_INFO_SAMP_FREQ), - }, - { - .type = IIO_LIGHT, - .info_mask_separate = BIT(IIO_CHAN_INFO_PROCESSED), - .info_mask_shared = BIT(IIO_CHAN_INFO_SAMP_FREQ), - }, - - } - </programlisting> - This channel's definition will generate two separate sysfs files - for raw data retrieval: - <itemizedlist> - <listitem> - <filename>/sys/bus/iio/iio:deviceX/in_intensity_ir_raw</filename> - </listitem> - <listitem> - <filename>/sys/bus/iio/iio:deviceX/in_intensity_both_raw</filename> - </listitem> - </itemizedlist> - one file for processed data: - <itemizedlist> - <listitem> - <filename>/sys/bus/iio/iio:deviceX/in_illuminance_input - </filename> - </listitem> - </itemizedlist> - and one shared sysfs file for sampling frequency: - <itemizedlist> - <listitem> - <filename>/sys/bus/iio/iio:deviceX/sampling_frequency. - </filename> - </listitem> - </itemizedlist> - </para> - <para> - Here is how we can make use of the channel's indexing: - <programlisting> - static const struct iio_chan_spec light_channels[] = { - { - .type = IIO_VOLTAGE, - .indexed = 1, - .channel = 0, - .info_mask_separate = BIT(IIO_CHAN_INFO_RAW), - }, - { - .type = IIO_VOLTAGE, - .indexed = 1, - .channel = 1, - .info_mask_separate = BIT(IIO_CHAN_INFO_RAW), - }, - } - </programlisting> - This will generate two separate attributes files for raw data - retrieval: - <itemizedlist> - <listitem> - <filename>/sys/bus/iio/devices/iio:deviceX/in_voltage0_raw</filename>, - representing voltage measurement for channel 0. - </listitem> - <listitem> - <filename>/sys/bus/iio/devices/iio:deviceX/in_voltage1_raw</filename>, - representing voltage measurement for channel 1. - </listitem> - </itemizedlist> - </para> - </sect2> - </sect1> - - <sect1 id="iiobuffer"> <title> Industrial I/O buffers </title> -!Finclude/linux/iio/buffer.h iio_buffer -!Edrivers/iio/industrialio-buffer.c - - <para> - The Industrial I/O core offers a way for continuous data capture - based on a trigger source. Multiple data channels can be read at once - from <filename>/dev/iio:deviceX</filename> character device node, - thus reducing the CPU load. - </para> - - <sect2 id="iiobuffersysfs"> - <title>IIO buffer sysfs interface </title> - <para> - An IIO buffer has an associated attributes directory under <filename> - /sys/bus/iio/iio:deviceX/buffer/</filename>. Here are the existing - attributes: - <itemizedlist> - <listitem> - <emphasis>length</emphasis>, the total number of data samples - (capacity) that can be stored by the buffer. - </listitem> - <listitem> - <emphasis>enable</emphasis>, activate buffer capture. - </listitem> - </itemizedlist> - - </para> - </sect2> - <sect2 id="iiobuffersetup"> <title> IIO buffer setup </title> - <para>The meta information associated with a channel reading - placed in a buffer is called a <emphasis> scan element </emphasis>. - The important bits configuring scan elements are exposed to - userspace applications via the <filename> - /sys/bus/iio/iio:deviceX/scan_elements/</filename> directory. This - file contains attributes of the following form: - <itemizedlist> - <listitem><emphasis>enable</emphasis>, used for enabling a channel. - If and only if its attribute is non zero, then a triggered capture - will contain data samples for this channel. - </listitem> - <listitem><emphasis>type</emphasis>, description of the scan element - data storage within the buffer and hence the form in which it is - read from user space. Format is <emphasis> - [be|le]:[s|u]bits/storagebitsXrepeat[>>shift] </emphasis>. - <itemizedlist> - <listitem> <emphasis>be</emphasis> or <emphasis>le</emphasis>, specifies - big or little endian. - </listitem> - <listitem> - <emphasis>s </emphasis>or <emphasis>u</emphasis>, specifies if - signed (2's complement) or unsigned. - </listitem> - <listitem><emphasis>bits</emphasis>, is the number of valid data - bits. - </listitem> - <listitem><emphasis>storagebits</emphasis>, is the number of bits - (after padding) that it occupies in the buffer. - </listitem> - <listitem> - <emphasis>shift</emphasis>, if specified, is the shift that needs - to be applied prior to masking out unused bits. - </listitem> - <listitem> - <emphasis>repeat</emphasis>, specifies the number of bits/storagebits - repetitions. When the repeat element is 0 or 1, then the repeat - value is omitted. - </listitem> - </itemizedlist> - </listitem> - </itemizedlist> - For example, a driver for a 3-axis accelerometer with 12 bit - resolution where data is stored in two 8-bits registers as - follows: - <programlisting> - 7 6 5 4 3 2 1 0 - +---+---+---+---+---+---+---+---+ - |D3 |D2 |D1 |D0 | X | X | X | X | (LOW byte, address 0x06) - +---+---+---+---+---+---+---+---+ - - 7 6 5 4 3 2 1 0 - +---+---+---+---+---+---+---+---+ - |D11|D10|D9 |D8 |D7 |D6 |D5 |D4 | (HIGH byte, address 0x07) - +---+---+---+---+---+---+---+---+ - </programlisting> - - will have the following scan element type for each axis: - <programlisting> - $ cat /sys/bus/iio/devices/iio:device0/scan_elements/in_accel_y_type - le:s12/16>>4 - </programlisting> - A user space application will interpret data samples read from the - buffer as two byte little endian signed data, that needs a 4 bits - right shift before masking out the 12 valid bits of data. - </para> - <para> - For implementing buffer support a driver should initialize the following - fields in <type>iio_chan_spec</type> definition: - <programlisting> - struct iio_chan_spec { - /* other members */ - int scan_index - struct { - char sign; - u8 realbits; - u8 storagebits; - u8 shift; - u8 repeat; - enum iio_endian endianness; - } scan_type; - }; - </programlisting> - The driver implementing the accelerometer described above will - have the following channel definition: - <programlisting> - struct struct iio_chan_spec accel_channels[] = { - { - .type = IIO_ACCEL, - .modified = 1, - .channel2 = IIO_MOD_X, - /* other stuff here */ - .scan_index = 0, - .scan_type = { - .sign = 's', - .realbits = 12, - .storagebits = 16, - .shift = 4, - .endianness = IIO_LE, - }, - } - /* similar for Y (with channel2 = IIO_MOD_Y, scan_index = 1) - * and Z (with channel2 = IIO_MOD_Z, scan_index = 2) axis - */ - } - </programlisting> - </para> - <para> - Here <emphasis> scan_index </emphasis> defines the order in which - the enabled channels are placed inside the buffer. Channels with a lower - scan_index will be placed before channels with a higher index. Each - channel needs to have a unique scan_index. - </para> - <para> - Setting scan_index to -1 can be used to indicate that the specific - channel does not support buffered capture. In this case no entries will - be created for the channel in the scan_elements directory. - </para> - </sect2> - </sect1> - - <sect1 id="iiotrigger"> <title> Industrial I/O triggers </title> -!Finclude/linux/iio/trigger.h iio_trigger -!Edrivers/iio/industrialio-trigger.c - <para> - In many situations it is useful for a driver to be able to - capture data based on some external event (trigger) as opposed - to periodically polling for data. An IIO trigger can be provided - by a device driver that also has an IIO device based on hardware - generated events (e.g. data ready or threshold exceeded) or - provided by a separate driver from an independent interrupt - source (e.g. GPIO line connected to some external system, timer - interrupt or user space writing a specific file in sysfs). A - trigger may initiate data capture for a number of sensors and - also it may be completely unrelated to the sensor itself. - </para> - - <sect2 id="iiotrigsysfs"> <title> IIO trigger sysfs interface </title> - There are two locations in sysfs related to triggers: - <itemizedlist> - <listitem><filename>/sys/bus/iio/devices/triggerY</filename>, - this file is created once an IIO trigger is registered with - the IIO core and corresponds to trigger with index Y. Because - triggers can be very different depending on type there are few - standard attributes that we can describe here: - <itemizedlist> - <listitem> - <emphasis>name</emphasis>, trigger name that can be later - used for association with a device. - </listitem> - <listitem> - <emphasis>sampling_frequency</emphasis>, some timer based - triggers use this attribute to specify the frequency for - trigger calls. - </listitem> - </itemizedlist> - </listitem> - <listitem> - <filename>/sys/bus/iio/devices/iio:deviceX/trigger/</filename>, this - directory is created once the device supports a triggered - buffer. We can associate a trigger with our device by writing - the trigger's name in the <filename>current_trigger</filename> file. - </listitem> - </itemizedlist> - </sect2> - - <sect2 id="iiotrigattr"> <title> IIO trigger setup</title> - - <para> - Let's see a simple example of how to setup a trigger to be used - by a driver. - - <programlisting> - struct iio_trigger_ops trigger_ops = { - .set_trigger_state = sample_trigger_state, - .validate_device = sample_validate_device, - } - - struct iio_trigger *trig; - - /* first, allocate memory for our trigger */ - trig = iio_trigger_alloc(dev, "trig-%s-%d", name, idx); - - /* setup trigger operations field */ - trig->ops = &trigger_ops; - - /* now register the trigger with the IIO core */ - iio_trigger_register(trig); - </programlisting> - </para> - </sect2> - - <sect2 id="iiotrigsetup"> <title> IIO trigger ops</title> -!Finclude/linux/iio/trigger.h iio_trigger_ops - <para> - Notice that a trigger has a set of operations attached: - <itemizedlist> - <listitem> - <function>set_trigger_state</function>, switch the trigger on/off - on demand. - </listitem> - <listitem> - <function>validate_device</function>, function to validate the - device when the current trigger gets changed. - </listitem> - </itemizedlist> - </para> - </sect2> - </sect1> - <sect1 id="iiotriggered_buffer"> - <title> Industrial I/O triggered buffers </title> - <para> - Now that we know what buffers and triggers are let's see how they - work together. - </para> - <sect2 id="iiotrigbufsetup"> <title> IIO triggered buffer setup</title> -!Edrivers/iio/buffer/industrialio-triggered-buffer.c -!Finclude/linux/iio/iio.h iio_buffer_setup_ops - - - <para> - A typical triggered buffer setup looks like this: - <programlisting> - const struct iio_buffer_setup_ops sensor_buffer_setup_ops = { - .preenable = sensor_buffer_preenable, - .postenable = sensor_buffer_postenable, - .postdisable = sensor_buffer_postdisable, - .predisable = sensor_buffer_predisable, - }; - - irqreturn_t sensor_iio_pollfunc(int irq, void *p) - { - pf->timestamp = iio_get_time_ns((struct indio_dev *)p); - return IRQ_WAKE_THREAD; - } - - irqreturn_t sensor_trigger_handler(int irq, void *p) - { - u16 buf[8]; - int i = 0; - - /* read data for each active channel */ - for_each_set_bit(bit, active_scan_mask, masklength) - buf[i++] = sensor_get_data(bit) - - iio_push_to_buffers_with_timestamp(indio_dev, buf, timestamp); - - iio_trigger_notify_done(trigger); - return IRQ_HANDLED; - } - - /* setup triggered buffer, usually in probe function */ - iio_triggered_buffer_setup(indio_dev, sensor_iio_polfunc, - sensor_trigger_handler, - sensor_buffer_setup_ops); - </programlisting> - </para> - The important things to notice here are: - <itemizedlist> - <listitem><function> iio_buffer_setup_ops</function>, the buffer setup - functions to be called at predefined points in the buffer configuration - sequence (e.g. before enable, after disable). If not specified, the - IIO core uses the default <type>iio_triggered_buffer_setup_ops</type>. - </listitem> - <listitem><function>sensor_iio_pollfunc</function>, the function that - will be used as top half of poll function. It should do as little - processing as possible, because it runs in interrupt context. The most - common operation is recording of the current timestamp and for this reason - one can use the IIO core defined <function>iio_pollfunc_store_time - </function> function. - </listitem> - <listitem><function>sensor_trigger_handler</function>, the function that - will be used as bottom half of the poll function. This runs in the - context of a kernel thread and all the processing takes place here. - It usually reads data from the device and stores it in the internal - buffer together with the timestamp recorded in the top half. - </listitem> - </itemizedlist> - </sect2> - </sect1> - </chapter> - <chapter id='iioresources'> - <title> Resources </title> - IIO core may change during time so the best documentation to read is the - source code. There are several locations where you should look: - <itemizedlist> - <listitem> - <filename>drivers/iio/</filename>, contains the IIO core plus - and directories for each sensor type (e.g. accel, magnetometer, - etc.) - </listitem> - <listitem> - <filename>include/linux/iio/</filename>, contains the header - files, nice to read for the internal kernel interfaces. - </listitem> - <listitem> - <filename>include/uapi/linux/iio/</filename>, contains files to be - used by user space applications. - </listitem> - <listitem> - <filename>tools/iio/</filename>, contains tools for rapidly - testing buffers, events and device creation. - </listitem> - <listitem> - <filename>drivers/staging/iio/</filename>, contains code for some - drivers or experimental features that are not yet mature enough - to be moved out. - </listitem> - </itemizedlist> - <para> - Besides the code, there are some good online documentation sources: - <itemizedlist> - <listitem> - <ulink url="http://marc.info/?l=linux-iio"> Industrial I/O mailing - list </ulink> - </listitem> - <listitem> - <ulink url="http://wiki.analog.com/software/linux/docs/iio/iio"> - Analog Device IIO wiki page </ulink> - </listitem> - <listitem> - <ulink url="https://fosdem.org/2015/schedule/event/iiosdr/"> - Using the Linux IIO framework for SDR, Lars-Peter Clausen's - presentation at FOSDEM </ulink> - </listitem> - </itemizedlist> - </para> - </chapter> -</book> - -<!-- -vim: softtabstop=2:shiftwidth=2:expandtab:textwidth=72 ---> diff --git a/Documentation/DocBook/regulator.tmpl b/Documentation/DocBook/regulator.tmpl deleted file mode 100644 index 3b08a085d2c7..000000000000 --- a/Documentation/DocBook/regulator.tmpl +++ /dev/null @@ -1,304 +0,0 @@ -<?xml version="1.0" encoding="UTF-8"?> -<!DOCTYPE book PUBLIC "-//OASIS//DTD DocBook XML V4.1.2//EN" - "http://www.oasis-open.org/docbook/xml/4.1.2/docbookx.dtd" []> - -<book id="regulator-api"> - <bookinfo> - <title>Voltage and current regulator API</title> - - <authorgroup> - <author> - <firstname>Liam</firstname> - <surname>Girdwood</surname> - <affiliation> - <address> - <email>lrg@slimlogic.co.uk</email> - </address> - </affiliation> - </author> - <author> - <firstname>Mark</firstname> - <surname>Brown</surname> - <affiliation> - <orgname>Wolfson Microelectronics</orgname> - <address> - <email>broonie@opensource.wolfsonmicro.com</email> - </address> - </affiliation> - </author> - </authorgroup> - - <copyright> - <year>2007-2008</year> - <holder>Wolfson Microelectronics</holder> - </copyright> - <copyright> - <year>2008</year> - <holder>Liam Girdwood</holder> - </copyright> - - <legalnotice> - <para> - This documentation is free software; you can redistribute - it and/or modify it under the terms of the GNU General Public - License version 2 as published by the Free Software Foundation. - </para> - - <para> - This program is distributed in the hope that it will be - useful, but WITHOUT ANY WARRANTY; without even the implied - warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. - See the GNU General Public License for more details. - </para> - - <para> - You should have received a copy of the GNU General Public - License along with this program; if not, write to the Free - Software Foundation, Inc., 59 Temple Place, Suite 330, Boston, - MA 02111-1307 USA - </para> - - <para> - For more details see the file COPYING in the source - distribution of Linux. - </para> - </legalnotice> - </bookinfo> - -<toc></toc> - - <chapter id="intro"> - <title>Introduction</title> - <para> - This framework is designed to provide a standard kernel - interface to control voltage and current regulators. - </para> - <para> - The intention is to allow systems to dynamically control - regulator power output in order to save power and prolong - battery life. This applies to both voltage regulators (where - voltage output is controllable) and current sinks (where current - limit is controllable). - </para> - <para> - Note that additional (and currently more complete) documentation - is available in the Linux kernel source under - <filename>Documentation/power/regulator</filename>. - </para> - - <sect1 id="glossary"> - <title>Glossary</title> - <para> - The regulator API uses a number of terms which may not be - familiar: - </para> - <glossary> - - <glossentry> - <glossterm>Regulator</glossterm> - <glossdef> - <para> - Electronic device that supplies power to other devices. Most - regulators can enable and disable their output and some can also - control their output voltage or current. - </para> - </glossdef> - </glossentry> - - <glossentry> - <glossterm>Consumer</glossterm> - <glossdef> - <para> - Electronic device which consumes power provided by a regulator. - These may either be static, requiring only a fixed supply, or - dynamic, requiring active management of the regulator at - runtime. - </para> - </glossdef> - </glossentry> - - <glossentry> - <glossterm>Power Domain</glossterm> - <glossdef> - <para> - The electronic circuit supplied by a given regulator, including - the regulator and all consumer devices. The configuration of - the regulator is shared between all the components in the - circuit. - </para> - </glossdef> - </glossentry> - - <glossentry> - <glossterm>Power Management Integrated Circuit</glossterm> - <acronym>PMIC</acronym> - <glossdef> - <para> - An IC which contains numerous regulators and often also other - subsystems. In an embedded system the primary PMIC is often - equivalent to a combination of the PSU and southbridge in a - desktop system. - </para> - </glossdef> - </glossentry> - </glossary> - </sect1> - </chapter> - - <chapter id="consumer"> - <title>Consumer driver interface</title> - <para> - This offers a similar API to the kernel clock framework. - Consumer drivers use <link - linkend='API-regulator-get'>get</link> and <link - linkend='API-regulator-put'>put</link> operations to acquire and - release regulators. Functions are - provided to <link linkend='API-regulator-enable'>enable</link> - and <link linkend='API-regulator-disable'>disable</link> the - regulator and to get and set the runtime parameters of the - regulator. - </para> - <para> - When requesting regulators consumers use symbolic names for their - supplies, such as "Vcc", which are mapped into actual regulator - devices by the machine interface. - </para> - <para> - A stub version of this API is provided when the regulator - framework is not in use in order to minimise the need to use - ifdefs. - </para> - - <sect1 id="consumer-enable"> - <title>Enabling and disabling</title> - <para> - The regulator API provides reference counted enabling and - disabling of regulators. Consumer devices use the <function><link - linkend='API-regulator-enable'>regulator_enable</link></function> - and <function><link - linkend='API-regulator-disable'>regulator_disable</link> - </function> functions to enable and disable regulators. Calls - to the two functions must be balanced. - </para> - <para> - Note that since multiple consumers may be using a regulator and - machine constraints may not allow the regulator to be disabled - there is no guarantee that calling - <function>regulator_disable</function> will actually cause the - supply provided by the regulator to be disabled. Consumer - drivers should assume that the regulator may be enabled at all - times. - </para> - </sect1> - - <sect1 id="consumer-config"> - <title>Configuration</title> - <para> - Some consumer devices may need to be able to dynamically - configure their supplies. For example, MMC drivers may need to - select the correct operating voltage for their cards. This may - be done while the regulator is enabled or disabled. - </para> - <para> - The <function><link - linkend='API-regulator-set-voltage'>regulator_set_voltage</link> - </function> and <function><link - linkend='API-regulator-set-current-limit' - >regulator_set_current_limit</link> - </function> functions provide the primary interface for this. - Both take ranges of voltages and currents, supporting drivers - that do not require a specific value (eg, CPU frequency scaling - normally permits the CPU to use a wider range of supply - voltages at lower frequencies but does not require that the - supply voltage be lowered). Where an exact value is required - both minimum and maximum values should be identical. - </para> - </sect1> - - <sect1 id="consumer-callback"> - <title>Callbacks</title> - <para> - Callbacks may also be <link - linkend='API-regulator-register-notifier'>registered</link> - for events such as regulation failures. - </para> - </sect1> - </chapter> - - <chapter id="driver"> - <title>Regulator driver interface</title> - <para> - Drivers for regulator chips <link - linkend='API-regulator-register'>register</link> the regulators - with the regulator core, providing operations structures to the - core. A <link - linkend='API-regulator-notifier-call-chain'>notifier</link> interface - allows error conditions to be reported to the core. - </para> - <para> - Registration should be triggered by explicit setup done by the - platform, supplying a <link - linkend='API-struct-regulator-init-data'>struct - regulator_init_data</link> for the regulator containing - <link linkend='machine-constraint'>constraint</link> and - <link linkend='machine-supply'>supply</link> information. - </para> - </chapter> - - <chapter id="machine"> - <title>Machine interface</title> - <para> - This interface provides a way to define how regulators are - connected to consumers on a given system and what the valid - operating parameters are for the system. - </para> - - <sect1 id="machine-supply"> - <title>Supplies</title> - <para> - Regulator supplies are specified using <link - linkend='API-struct-regulator-consumer-supply'>struct - regulator_consumer_supply</link>. This is done at - <link linkend='driver'>driver registration - time</link> as part of the machine constraints. - </para> - </sect1> - - <sect1 id="machine-constraint"> - <title>Constraints</title> - <para> - As well as defining the connections the machine interface - also provides constraints defining the operations that - clients are allowed to perform and the parameters that may be - set. This is required since generally regulator devices will - offer more flexibility than it is safe to use on a given - system, for example supporting higher supply voltages than the - consumers are rated for. - </para> - <para> - This is done at <link linkend='driver'>driver - registration time</link> by providing a <link - linkend='API-struct-regulation-constraints'>struct - regulation_constraints</link>. - </para> - <para> - The constraints may also specify an initial configuration for the - regulator in the constraints, which is particularly useful for - use with static consumers. - </para> - </sect1> - </chapter> - - <chapter id="api"> - <title>API reference</title> - <para> - Due to limitations of the kernel documentation framework and the - existing layout of the source code the entire regulator API is - documented here. - </para> -!Iinclude/linux/regulator/consumer.h -!Iinclude/linux/regulator/machine.h -!Iinclude/linux/regulator/driver.h -!Edrivers/regulator/core.c - </chapter> -</book> diff --git a/Documentation/Makefile.sphinx b/Documentation/Makefile.sphinx index 707c65337ebf..bcf529f6cf9b 100644 --- a/Documentation/Makefile.sphinx +++ b/Documentation/Makefile.sphinx @@ -43,7 +43,7 @@ ALLSPHINXOPTS = $(KERNELDOC_CONF) $(PAPEROPT_$(PAPER)) $(SPHINXOPTS) I18NSPHINXOPTS = $(PAPEROPT_$(PAPER)) $(SPHINXOPTS) . # commands; the 'cmd' from scripts/Kbuild.include is not *loopable* -loop_cmd = $(echo-cmd) $(cmd_$(1)) +loop_cmd = $(echo-cmd) $(cmd_$(1)) || exit; # $2 sphinx builder e.g. "html" # $3 name of the build subfolder / e.g. "media", used as: @@ -54,7 +54,8 @@ loop_cmd = $(echo-cmd) $(cmd_$(1)) # e.g. "media" for the linux-tv book-set at ./Documentation/media quiet_cmd_sphinx = SPHINX $@ --> file://$(abspath $(BUILDDIR)/$3/$4) - cmd_sphinx = $(MAKE) BUILDDIR=$(abspath $(BUILDDIR)) $(build)=Documentation/media $2;\ + cmd_sphinx = $(MAKE) BUILDDIR=$(abspath $(BUILDDIR)) $(build)=Documentation/media $2 && \ + PYTHONDONTWRITEBYTECODE=1 \ BUILDDIR=$(abspath $(BUILDDIR)) SPHINX_CONF=$(abspath $(srctree)/$(src)/$5/$(SPHINX_CONF)) \ $(SPHINXBUILD) \ -b $2 \ @@ -63,13 +64,16 @@ quiet_cmd_sphinx = SPHINX $@ --> file://$(abspath $(BUILDDIR)/$3/$4) -D version=$(KERNELVERSION) -D release=$(KERNELRELEASE) \ $(ALLSPHINXOPTS) \ $(abspath $(srctree)/$(src)/$5) \ - $(abspath $(BUILDDIR)/$3/$4); + $(abspath $(BUILDDIR)/$3/$4) htmldocs: - @$(foreach var,$(SPHINXDIRS),$(call loop_cmd,sphinx,html,$(var),,$(var))) + @+$(foreach var,$(SPHINXDIRS),$(call loop_cmd,sphinx,html,$(var),,$(var))) + +linkcheckdocs: + @$(foreach var,$(SPHINXDIRS),$(call loop_cmd,sphinx,linkcheck,$(var),,$(var))) latexdocs: - @$(foreach var,$(SPHINXDIRS),$(call loop_cmd,sphinx,latex,$(var),latex,$(var))) + @+$(foreach var,$(SPHINXDIRS),$(call loop_cmd,sphinx,latex,$(var),latex,$(var))) ifeq ($(HAVE_PDFLATEX),0) @@ -80,27 +84,34 @@ pdfdocs: else # HAVE_PDFLATEX pdfdocs: latexdocs - $(foreach var,$(SPHINXDIRS), $(MAKE) PDFLATEX=$(PDFLATEX) LATEXOPTS="$(LATEXOPTS)" -C $(BUILDDIR)/$(var)/latex;) + $(foreach var,$(SPHINXDIRS), $(MAKE) PDFLATEX=$(PDFLATEX) LATEXOPTS="$(LATEXOPTS)" -C $(BUILDDIR)/$(var)/latex || exit;) endif # HAVE_PDFLATEX epubdocs: - @$(foreach var,$(SPHINXDIRS),$(call loop_cmd,sphinx,epub,$(var),epub,$(var))) + @+$(foreach var,$(SPHINXDIRS),$(call loop_cmd,sphinx,epub,$(var),epub,$(var))) xmldocs: - @$(foreach var,$(SPHINXDIRS),$(call loop_cmd,sphinx,xml,$(var),xml,$(var))) + @+$(foreach var,$(SPHINXDIRS),$(call loop_cmd,sphinx,xml,$(var),xml,$(var))) + +endif # HAVE_SPHINX + +# The following targets are independent of HAVE_SPHINX, and the rules should +# work or silently pass without Sphinx. # no-ops for the Sphinx toolchain sgmldocs: + @: psdocs: + @: mandocs: + @: installmandocs: + @: cleandocs: $(Q)rm -rf $(BUILDDIR) - $(Q)$(MAKE) BUILDDIR=$(abspath $(BUILDDIR)) -C Documentation/media clean - -endif # HAVE_SPHINX + $(Q)$(MAKE) BUILDDIR=$(abspath $(BUILDDIR)) $(build)=Documentation/media clean dochelp: @echo ' Linux kernel internal documentation in different formats (Sphinx):' @@ -109,6 +120,7 @@ dochelp: @echo ' pdfdocs - PDF' @echo ' epubdocs - EPUB' @echo ' xmldocs - XML' + @echo ' linkcheckdocs - check for broken external links (will connect to external hosts)' @echo ' cleandocs - clean all generated files' @echo @echo ' make SPHINXDIRS="s1 s2" [target] Generate only docs of folder s1, s2' diff --git a/Documentation/admin-guide/README.rst b/Documentation/admin-guide/README.rst index 1b6dfb2b3adb..697a00ccec25 100644 --- a/Documentation/admin-guide/README.rst +++ b/Documentation/admin-guide/README.rst @@ -17,7 +17,7 @@ What is Linux? loading, shared copy-on-write executables, proper memory management, and multistack networking including IPv4 and IPv6. - It is distributed under the GNU General Public License - see the + It is distributed under the GNU General Public License v2 - see the accompanying COPYING file for more details. On what hardware does it run? @@ -236,7 +236,7 @@ Configuring the kernel - Having unnecessary drivers will make the kernel bigger, and can under some circumstances lead to problems: probing for a - nonexistent controller card may confuse your other controllers + nonexistent controller card may confuse your other controllers. - A kernel with math-emulation compiled in will still use the coprocessor if one is present: the math emulation will just diff --git a/Documentation/admin-guide/dynamic-debug-howto.rst b/Documentation/admin-guide/dynamic-debug-howto.rst index 88adcfdf5b2b..12278a926370 100644 --- a/Documentation/admin-guide/dynamic-debug-howto.rst +++ b/Documentation/admin-guide/dynamic-debug-howto.rst @@ -93,9 +93,9 @@ Command Language Reference At the lexical level, a command comprises a sequence of words separated by spaces or tabs. So these are all equivalent:: - nullarbor:~ # echo -c 'file svcsock.c line 1603 +p' > + nullarbor:~ # echo -n 'file svcsock.c line 1603 +p' > <debugfs>/dynamic_debug/control - nullarbor:~ # echo -c ' file svcsock.c line 1603 +p ' > + nullarbor:~ # echo -n ' file svcsock.c line 1603 +p ' > <debugfs>/dynamic_debug/control nullarbor:~ # echo -n 'file svcsock.c line 1603 +p' > <debugfs>/dynamic_debug/control diff --git a/Documentation/block/pr.txt b/Documentation/block/pr.txt index d3eb1ca65051..ac9b8e70e64b 100644 --- a/Documentation/block/pr.txt +++ b/Documentation/block/pr.txt @@ -90,7 +90,7 @@ and thus removes any access restriction implied by it. 4. IOC_PR_PREEMPT This ioctl command releases the existing reservation referred to by -old_key and replaces it with a a new reservation of type for the +old_key and replaces it with a new reservation of type for the reservation key new_key. diff --git a/Documentation/cgroup-v1/cpusets.txt b/Documentation/cgroup-v1/cpusets.txt index e5ac5da86682..8402dd6de8df 100644 --- a/Documentation/cgroup-v1/cpusets.txt +++ b/Documentation/cgroup-v1/cpusets.txt @@ -615,7 +615,7 @@ to allocate a page of memory for that task. If a cpuset has its 'cpuset.cpus' modified, then each task in that cpuset will have its allowed CPU placement changed immediately. Similarly, -if a task's pid is written to another cpusets 'cpuset.tasks' file, then its +if a task's pid is written to another cpuset's 'tasks' file, then its allowed CPU placement is changed immediately. If such a task had been bound to some subset of its cpuset using the sched_setaffinity() call, the task will be allowed to run on any CPU allowed in its new cpuset, diff --git a/Documentation/conf.py b/Documentation/conf.py index 1ac958c0333d..f6823cf01275 100644 --- a/Documentation/conf.py +++ b/Documentation/conf.py @@ -58,7 +58,7 @@ master_doc = 'index' # General information about the project. project = 'The Linux Kernel' -copyright = '2016, The kernel development community' +copyright = 'The kernel development community' author = 'The kernel development community' # The version info for the project you're documenting, acts as replacement for diff --git a/Documentation/core-api/cpu_hotplug.rst b/Documentation/core-api/cpu_hotplug.rst new file mode 100644 index 000000000000..4a50ab7817f7 --- /dev/null +++ b/Documentation/core-api/cpu_hotplug.rst @@ -0,0 +1,372 @@ +========================= +CPU hotplug in the Kernel +========================= + +:Date: December, 2016 +:Author: Sebastian Andrzej Siewior <bigeasy@linutronix.de>, + Rusty Russell <rusty@rustcorp.com.au>, + Srivatsa Vaddagiri <vatsa@in.ibm.com>, + Ashok Raj <ashok.raj@intel.com>, + Joel Schopp <jschopp@austin.ibm.com> + +Introduction +============ + +Modern advances in system architectures have introduced advanced error +reporting and correction capabilities in processors. There are couple OEMS that +support NUMA hardware which are hot pluggable as well, where physical node +insertion and removal require support for CPU hotplug. + +Such advances require CPUs available to a kernel to be removed either for +provisioning reasons, or for RAS purposes to keep an offending CPU off +system execution path. Hence the need for CPU hotplug support in the +Linux kernel. + +A more novel use of CPU-hotplug support is its use today in suspend resume +support for SMP. Dual-core and HT support makes even a laptop run SMP kernels +which didn't support these methods. + + +Command Line Switches +===================== +``maxcpus=n`` + Restrict boot time CPUs to *n*. Say if you have fourV CPUs, using + ``maxcpus=2`` will only boot two. You can choose to bring the + other CPUs later online. + +``nr_cpus=n`` + Restrict the total amount CPUs the kernel will support. If the number + supplied here is lower than the number of physically available CPUs than + those CPUs can not be brought online later. + +``additional_cpus=n`` + Use this to limit hotpluggable CPUs. This option sets + ``cpu_possible_mask = cpu_present_mask + additional_cpus`` + + This option is limited to the IA64 architecture. + +``possible_cpus=n`` + This option sets ``possible_cpus`` bits in ``cpu_possible_mask``. + + This option is limited to the X86 and S390 architecture. + +``cede_offline={"off","on"}`` + Use this option to disable/enable putting offlined processors to an extended + ``H_CEDE`` state on supported pseries platforms. If nothing is specified, + ``cede_offline`` is set to "on". + + This option is limited to the PowerPC architecture. + +``cpu0_hotplug`` + Allow to shutdown CPU0. + + This option is limited to the X86 architecture. + +CPU maps +======== + +``cpu_possible_mask`` + Bitmap of possible CPUs that can ever be available in the + system. This is used to allocate some boot time memory for per_cpu variables + that aren't designed to grow/shrink as CPUs are made available or removed. + Once set during boot time discovery phase, the map is static, i.e no bits + are added or removed anytime. Trimming it accurately for your system needs + upfront can save some boot time memory. + +``cpu_online_mask`` + Bitmap of all CPUs currently online. Its set in ``__cpu_up()`` + after a CPU is available for kernel scheduling and ready to receive + interrupts from devices. Its cleared when a CPU is brought down using + ``__cpu_disable()``, before which all OS services including interrupts are + migrated to another target CPU. + +``cpu_present_mask`` + Bitmap of CPUs currently present in the system. Not all + of them may be online. When physical hotplug is processed by the relevant + subsystem (e.g ACPI) can change and new bit either be added or removed + from the map depending on the event is hot-add/hot-remove. There are currently + no locking rules as of now. Typical usage is to init topology during boot, + at which time hotplug is disabled. + +You really don't need to manipulate any of the system CPU maps. They should +be read-only for most use. When setting up per-cpu resources almost always use +``cpu_possible_mask`` or ``for_each_possible_cpu()`` to iterate. To macro +``for_each_cpu()`` can be used to iterate over a custom CPU mask. + +Never use anything other than ``cpumask_t`` to represent bitmap of CPUs. + + +Using CPU hotplug +================= +The kernel option *CONFIG_HOTPLUG_CPU* needs to be enabled. It is currently +available on multiple architectures including ARM, MIPS, PowerPC and X86. The +configuration is done via the sysfs interface: :: + + $ ls -lh /sys/devices/system/cpu + total 0 + drwxr-xr-x 9 root root 0 Dec 21 16:33 cpu0 + drwxr-xr-x 9 root root 0 Dec 21 16:33 cpu1 + drwxr-xr-x 9 root root 0 Dec 21 16:33 cpu2 + drwxr-xr-x 9 root root 0 Dec 21 16:33 cpu3 + drwxr-xr-x 9 root root 0 Dec 21 16:33 cpu4 + drwxr-xr-x 9 root root 0 Dec 21 16:33 cpu5 + drwxr-xr-x 9 root root 0 Dec 21 16:33 cpu6 + drwxr-xr-x 9 root root 0 Dec 21 16:33 cpu7 + drwxr-xr-x 2 root root 0 Dec 21 16:33 hotplug + -r--r--r-- 1 root root 4.0K Dec 21 16:33 offline + -r--r--r-- 1 root root 4.0K Dec 21 16:33 online + -r--r--r-- 1 root root 4.0K Dec 21 16:33 possible + -r--r--r-- 1 root root 4.0K Dec 21 16:33 present + +The files *offline*, *online*, *possible*, *present* represent the CPU masks. +Each CPU folder contains an *online* file which controls the logical on (1) and +off (0) state. To logically shutdown CPU4: :: + + $ echo 0 > /sys/devices/system/cpu/cpu4/online + smpboot: CPU 4 is now offline + +Once the CPU is shutdown, it will be removed from */proc/interrupts*, +*/proc/cpuinfo* and should also not be shown visible by the *top* command. To +bring CPU4 back online: :: + + $ echo 1 > /sys/devices/system/cpu/cpu4/online + smpboot: Booting Node 0 Processor 4 APIC 0x1 + +The CPU is usable again. This should work on all CPUs. CPU0 is often special +and excluded from CPU hotplug. On X86 the kernel option +*CONFIG_BOOTPARAM_HOTPLUG_CPU0* has to be enabled in order to be able to +shutdown CPU0. Alternatively the kernel command option *cpu0_hotplug* can be +used. Some known dependencies of CPU0: + +* Resume from hibernate/suspend. Hibernate/suspend will fail if CPU0 is offline. +* PIC interrupts. CPU0 can't be removed if a PIC interrupt is detected. + +Please let Fenghua Yu <fenghua.yu@intel.com> know if you find any dependencies +on CPU0. + +The CPU hotplug coordination +============================ + +The offline case +---------------- +Once a CPU has been logically shutdown the teardown callbacks of registered +hotplug states will be invoked, starting with ``CPUHP_ONLINE`` and terminating +at state ``CPUHP_OFFLINE``. This includes: + +* If tasks are frozen due to a suspend operation then *cpuhp_tasks_frozen* + will be set to true. +* All processes are migrated away from this outgoing CPU to new CPUs. + The new CPU is chosen from each process' current cpuset, which may be + a subset of all online CPUs. +* All interrupts targeted to this CPU are migrated to a new CPU +* timers are also migrated to a new CPU +* Once all services are migrated, kernel calls an arch specific routine + ``__cpu_disable()`` to perform arch specific cleanup. + +Using the hotplug API +--------------------- +It is possible to receive notifications once a CPU is offline or onlined. This +might be important to certain drivers which need to perform some kind of setup +or clean up functions based on the number of available CPUs: :: + + #include <linux/cpuhotplug.h> + + ret = cpuhp_setup_state(CPUHP_AP_ONLINE_DYN, "X/Y:online", + Y_online, Y_prepare_down); + +*X* is the subsystem and *Y* the particular driver. The *Y_online* callback +will be invoked during registration on all online CPUs. If an error +occurs during the online callback the *Y_prepare_down* callback will be +invoked on all CPUs on which the online callback was previously invoked. +After registration completed, the *Y_online* callback will be invoked +once a CPU is brought online and *Y_prepare_down* will be invoked when a +CPU is shutdown. All resources which were previously allocated in +*Y_online* should be released in *Y_prepare_down*. +The return value *ret* is negative if an error occurred during the +registration process. Otherwise a positive value is returned which +contains the allocated hotplug for dynamically allocated states +(*CPUHP_AP_ONLINE_DYN*). It will return zero for predefined states. + +The callback can be remove by invoking ``cpuhp_remove_state()``. In case of a +dynamically allocated state (*CPUHP_AP_ONLINE_DYN*) use the returned state. +During the removal of a hotplug state the teardown callback will be invoked. + +Multiple instances +~~~~~~~~~~~~~~~~~~ +If a driver has multiple instances and each instance needs to perform the +callback independently then it is likely that a ''multi-state'' should be used. +First a multi-state state needs to be registered: :: + + ret = cpuhp_setup_state_multi(CPUHP_AP_ONLINE_DYN, "X/Y:online, + Y_online, Y_prepare_down); + Y_hp_online = ret; + +The ``cpuhp_setup_state_multi()`` behaves similar to ``cpuhp_setup_state()`` +except it prepares the callbacks for a multi state and does not invoke +the callbacks. This is a one time setup. +Once a new instance is allocated, you need to register this new instance: :: + + ret = cpuhp_state_add_instance(Y_hp_online, &d->node); + +This function will add this instance to your previously allocated +*Y_hp_online* state and invoke the previously registered callback +(*Y_online*) on all online CPUs. The *node* element is a ``struct +hlist_node`` member of your per-instance data structure. + +On removal of the instance: :: + cpuhp_state_remove_instance(Y_hp_online, &d->node) + +should be invoked which will invoke the teardown callback on all online +CPUs. + +Manual setup +~~~~~~~~~~~~ +Usually it is handy to invoke setup and teardown callbacks on registration or +removal of a state because usually the operation needs to performed once a CPU +goes online (offline) and during initial setup (shutdown) of the driver. However +each registration and removal function is also available with a ``_nocalls`` +suffix which does not invoke the provided callbacks if the invocation of the +callbacks is not desired. During the manual setup (or teardown) the functions +``get_online_cpus()`` and ``put_online_cpus()`` should be used to inhibit CPU +hotplug operations. + + +The ordering of the events +-------------------------- +The hotplug states are defined in ``include/linux/cpuhotplug.h``: + +* The states *CPUHP_OFFLINE* … *CPUHP_AP_OFFLINE* are invoked before the + CPU is up. +* The states *CPUHP_AP_OFFLINE* … *CPUHP_AP_ONLINE* are invoked + just the after the CPU has been brought up. The interrupts are off and + the scheduler is not yet active on this CPU. Starting with *CPUHP_AP_OFFLINE* + the callbacks are invoked on the target CPU. +* The states between *CPUHP_AP_ONLINE_DYN* and *CPUHP_AP_ONLINE_DYN_END* are + reserved for the dynamic allocation. +* The states are invoked in the reverse order on CPU shutdown starting with + *CPUHP_ONLINE* and stopping at *CPUHP_OFFLINE*. Here the callbacks are + invoked on the CPU that will be shutdown until *CPUHP_AP_OFFLINE*. + +A dynamically allocated state via *CPUHP_AP_ONLINE_DYN* is often enough. +However if an earlier invocation during the bring up or shutdown is required +then an explicit state should be acquired. An explicit state might also be +required if the hotplug event requires specific ordering in respect to +another hotplug event. + +Testing of hotplug states +========================= +One way to verify whether a custom state is working as expected or not is to +shutdown a CPU and then put it online again. It is also possible to put the CPU +to certain state (for instance *CPUHP_AP_ONLINE*) and then go back to +*CPUHP_ONLINE*. This would simulate an error one state after *CPUHP_AP_ONLINE* +which would lead to rollback to the online state. + +All registered states are enumerated in ``/sys/devices/system/cpu/hotplug/states``: :: + + $ tail /sys/devices/system/cpu/hotplug/states + 138: mm/vmscan:online + 139: mm/vmstat:online + 140: lib/percpu_cnt:online + 141: acpi/cpu-drv:online + 142: base/cacheinfo:online + 143: virtio/net:online + 144: x86/mce:online + 145: printk:online + 168: sched:active + 169: online + +To rollback CPU4 to ``lib/percpu_cnt:online`` and back online just issue: :: + + $ cat /sys/devices/system/cpu/cpu4/hotplug/state + 169 + $ echo 140 > /sys/devices/system/cpu/cpu4/hotplug/target + $ cat /sys/devices/system/cpu/cpu4/hotplug/state + 140 + +It is important to note that the teardown callbac of state 140 have been +invoked. And now get back online: :: + + $ echo 169 > /sys/devices/system/cpu/cpu4/hotplug/target + $ cat /sys/devices/system/cpu/cpu4/hotplug/state + 169 + +With trace events enabled, the individual steps are visible, too: :: + + # TASK-PID CPU# TIMESTAMP FUNCTION + # | | | | | + bash-394 [001] 22.976: cpuhp_enter: cpu: 0004 target: 140 step: 169 (cpuhp_kick_ap_work) + cpuhp/4-31 [004] 22.977: cpuhp_enter: cpu: 0004 target: 140 step: 168 (sched_cpu_deactivate) + cpuhp/4-31 [004] 22.990: cpuhp_exit: cpu: 0004 state: 168 step: 168 ret: 0 + cpuhp/4-31 [004] 22.991: cpuhp_enter: cpu: 0004 target: 140 step: 144 (mce_cpu_pre_down) + cpuhp/4-31 [004] 22.992: cpuhp_exit: cpu: 0004 state: 144 step: 144 ret: 0 + cpuhp/4-31 [004] 22.993: cpuhp_multi_enter: cpu: 0004 target: 140 step: 143 (virtnet_cpu_down_prep) + cpuhp/4-31 [004] 22.994: cpuhp_exit: cpu: 0004 state: 143 step: 143 ret: 0 + cpuhp/4-31 [004] 22.995: cpuhp_enter: cpu: 0004 target: 140 step: 142 (cacheinfo_cpu_pre_down) + cpuhp/4-31 [004] 22.996: cpuhp_exit: cpu: 0004 state: 142 step: 142 ret: 0 + bash-394 [001] 22.997: cpuhp_exit: cpu: 0004 state: 140 step: 169 ret: 0 + bash-394 [005] 95.540: cpuhp_enter: cpu: 0004 target: 169 step: 140 (cpuhp_kick_ap_work) + cpuhp/4-31 [004] 95.541: cpuhp_enter: cpu: 0004 target: 169 step: 141 (acpi_soft_cpu_online) + cpuhp/4-31 [004] 95.542: cpuhp_exit: cpu: 0004 state: 141 step: 141 ret: 0 + cpuhp/4-31 [004] 95.543: cpuhp_enter: cpu: 0004 target: 169 step: 142 (cacheinfo_cpu_online) + cpuhp/4-31 [004] 95.544: cpuhp_exit: cpu: 0004 state: 142 step: 142 ret: 0 + cpuhp/4-31 [004] 95.545: cpuhp_multi_enter: cpu: 0004 target: 169 step: 143 (virtnet_cpu_online) + cpuhp/4-31 [004] 95.546: cpuhp_exit: cpu: 0004 state: 143 step: 143 ret: 0 + cpuhp/4-31 [004] 95.547: cpuhp_enter: cpu: 0004 target: 169 step: 144 (mce_cpu_online) + cpuhp/4-31 [004] 95.548: cpuhp_exit: cpu: 0004 state: 144 step: 144 ret: 0 + cpuhp/4-31 [004] 95.549: cpuhp_enter: cpu: 0004 target: 169 step: 145 (console_cpu_notify) + cpuhp/4-31 [004] 95.550: cpuhp_exit: cpu: 0004 state: 145 step: 145 ret: 0 + cpuhp/4-31 [004] 95.551: cpuhp_enter: cpu: 0004 target: 169 step: 168 (sched_cpu_activate) + cpuhp/4-31 [004] 95.552: cpuhp_exit: cpu: 0004 state: 168 step: 168 ret: 0 + bash-394 [005] 95.553: cpuhp_exit: cpu: 0004 state: 169 step: 140 ret: 0 + +As it an be seen, CPU4 went down until timestamp 22.996 and then back up until +95.552. All invoked callbacks including their return codes are visible in the +trace. + +Architecture's requirements +=========================== +The following functions and configurations are required: + +``CONFIG_HOTPLUG_CPU`` + This entry needs to be enabled in Kconfig + +``__cpu_up()`` + Arch interface to bring up a CPU + +``__cpu_disable()`` + Arch interface to shutdown a CPU, no more interrupts can be handled by the + kernel after the routine returns. This includes the shutdown of the timer. + +``__cpu_die()`` + This actually supposed to ensure death of the CPU. Actually look at some + example code in other arch that implement CPU hotplug. The processor is taken + down from the ``idle()`` loop for that specific architecture. ``__cpu_die()`` + typically waits for some per_cpu state to be set, to ensure the processor dead + routine is called to be sure positively. + +User Space Notification +======================= +After CPU successfully onlined or offline udev events are sent. A udev rule like: :: + + SUBSYSTEM=="cpu", DRIVERS=="processor", DEVPATH=="/devices/system/cpu/*", RUN+="the_hotplug_receiver.sh" + +will receive all events. A script like: :: + + #!/bin/sh + + if [ "${ACTION}" = "offline" ] + then + echo "CPU ${DEVPATH##*/} offline" + + elif [ "${ACTION}" = "online" ] + then + echo "CPU ${DEVPATH##*/} online" + + fi + +can process the event further. + +Kernel Inline Documentations Reference +====================================== + +.. kernel-doc:: include/linux/cpuhotplug.h diff --git a/Documentation/core-api/index.rst b/Documentation/core-api/index.rst index 2872ca1a52f1..0d93d8089136 100644 --- a/Documentation/core-api/index.rst +++ b/Documentation/core-api/index.rst @@ -13,6 +13,7 @@ Core utilities assoc_array atomic_ops + cpu_hotplug local_ops workqueue diff --git a/Documentation/cpu-freq/user-guide.txt b/Documentation/cpu-freq/user-guide.txt index 107f6fdd7d14..391da64e9492 100644 --- a/Documentation/cpu-freq/user-guide.txt +++ b/Documentation/cpu-freq/user-guide.txt @@ -82,7 +82,9 @@ UltraSPARC-III ------- Several "PowerBook" and "iBook2" notebooks are supported. - +The following POWER processors are supported in powernv mode: +POWER8 +POWER9 1.5 SuperH ---------- diff --git a/Documentation/cpu-hotplug.txt b/Documentation/cpu-hotplug.txt deleted file mode 100644 index d02e8a451872..000000000000 --- a/Documentation/cpu-hotplug.txt +++ /dev/null @@ -1,452 +0,0 @@ - CPU hotplug Support in Linux(tm) Kernel - - Maintainers: - CPU Hotplug Core: - Rusty Russell <rusty@rustcorp.com.au> - Srivatsa Vaddagiri <vatsa@in.ibm.com> - i386: - Zwane Mwaikambo <zwanem@gmail.com> - ppc64: - Nathan Lynch <nathanl@austin.ibm.com> - Joel Schopp <jschopp@austin.ibm.com> - ia64/x86_64: - Ashok Raj <ashok.raj@intel.com> - s390: - Heiko Carstens <heiko.carstens@de.ibm.com> - -Authors: Ashok Raj <ashok.raj@intel.com> -Lots of feedback: Nathan Lynch <nathanl@austin.ibm.com>, - Joel Schopp <jschopp@austin.ibm.com> - -Introduction - -Modern advances in system architectures have introduced advanced error -reporting and correction capabilities in processors. CPU architectures permit -partitioning support, where compute resources of a single CPU could be made -available to virtual machine environments. There are couple OEMS that -support NUMA hardware which are hot pluggable as well, where physical -node insertion and removal require support for CPU hotplug. - -Such advances require CPUs available to a kernel to be removed either for -provisioning reasons, or for RAS purposes to keep an offending CPU off -system execution path. Hence the need for CPU hotplug support in the -Linux kernel. - -A more novel use of CPU-hotplug support is its use today in suspend -resume support for SMP. Dual-core and HT support makes even -a laptop run SMP kernels which didn't support these methods. SMP support -for suspend/resume is a work in progress. - -General Stuff about CPU Hotplug --------------------------------- - -Command Line Switches ---------------------- -maxcpus=n Restrict boot time cpus to n. Say if you have 4 cpus, using - maxcpus=2 will only boot 2. You can choose to bring the - other cpus later online, read FAQ's for more info. - -additional_cpus=n (*) Use this to limit hotpluggable cpus. This option sets - cpu_possible_mask = cpu_present_mask + additional_cpus - -cede_offline={"off","on"} Use this option to disable/enable putting offlined - processors to an extended H_CEDE state on - supported pseries platforms. - If nothing is specified, - cede_offline is set to "on". - -(*) Option valid only for following architectures -- ia64 - -ia64 uses the number of disabled local apics in ACPI tables MADT to -determine the number of potentially hot-pluggable cpus. The implementation -should only rely on this to count the # of cpus, but *MUST* not rely -on the apicid values in those tables for disabled apics. In the event -BIOS doesn't mark such hot-pluggable cpus as disabled entries, one could -use this parameter "additional_cpus=x" to represent those cpus in the -cpu_possible_mask. - -possible_cpus=n [s390,x86_64] use this to set hotpluggable cpus. - This option sets possible_cpus bits in - cpu_possible_mask. Thus keeping the numbers of bits set - constant even if the machine gets rebooted. - -CPU maps and such ------------------ -[More on cpumaps and primitive to manipulate, please check -include/linux/cpumask.h that has more descriptive text.] - -cpu_possible_mask: Bitmap of possible CPUs that can ever be available in the -system. This is used to allocate some boot time memory for per_cpu variables -that aren't designed to grow/shrink as CPUs are made available or removed. -Once set during boot time discovery phase, the map is static, i.e no bits -are added or removed anytime. Trimming it accurately for your system needs -upfront can save some boot time memory. See below for how we use heuristics -in x86_64 case to keep this under check. - -cpu_online_mask: Bitmap of all CPUs currently online. It's set in __cpu_up() -after a CPU is available for kernel scheduling and ready to receive -interrupts from devices. It's cleared when a CPU is brought down using -__cpu_disable(), before which all OS services including interrupts are -migrated to another target CPU. - -cpu_present_mask: Bitmap of CPUs currently present in the system. Not all -of them may be online. When physical hotplug is processed by the relevant -subsystem (e.g ACPI) can change and new bit either be added or removed -from the map depending on the event is hot-add/hot-remove. There are currently -no locking rules as of now. Typical usage is to init topology during boot, -at which time hotplug is disabled. - -You really dont need to manipulate any of the system cpu maps. They should -be read-only for most use. When setting up per-cpu resources almost always use -cpu_possible_mask/for_each_possible_cpu() to iterate. - -Never use anything other than cpumask_t to represent bitmap of CPUs. - - #include <linux/cpumask.h> - - for_each_possible_cpu - Iterate over cpu_possible_mask - for_each_online_cpu - Iterate over cpu_online_mask - for_each_present_cpu - Iterate over cpu_present_mask - for_each_cpu(x,mask) - Iterate over some random collection of cpu mask. - - #include <linux/cpu.h> - get_online_cpus() and put_online_cpus(): - -The above calls are used to inhibit cpu hotplug operations. While the -cpu_hotplug.refcount is non zero, the cpu_online_mask will not change. -If you merely need to avoid cpus going away, you could also use -preempt_disable() and preempt_enable() for those sections. -Just remember the critical section cannot call any -function that can sleep or schedule this process away. The preempt_disable() -will work as long as stop_machine_run() is used to take a cpu down. - -CPU Hotplug - Frequently Asked Questions. - -Q: How to enable my kernel to support CPU hotplug? -A: When doing make defconfig, Enable CPU hotplug support - - "Processor type and Features" -> Support for Hotpluggable CPUs - -Make sure that you have CONFIG_SMP turned on as well. - -You would need to enable CONFIG_HOTPLUG_CPU for SMP suspend/resume support -as well. - -Q: What architectures support CPU hotplug? -A: As of 2.6.14, the following architectures support CPU hotplug. - -i386 (Intel), ppc, ppc64, parisc, s390, ia64 and x86_64 - -Q: How to test if hotplug is supported on the newly built kernel? -A: You should now notice an entry in sysfs. - -Check if sysfs is mounted, using the "mount" command. You should notice -an entry as shown below in the output. - - .... - none on /sys type sysfs (rw) - .... - -If this is not mounted, do the following. - - #mkdir /sys - #mount -t sysfs sys /sys - -Now you should see entries for all present cpu, the following is an example -in a 8-way system. - - #pwd - #/sys/devices/system/cpu - #ls -l - total 0 - drwxr-xr-x 10 root root 0 Sep 19 07:44 . - drwxr-xr-x 13 root root 0 Sep 19 07:45 .. - drwxr-xr-x 3 root root 0 Sep 19 07:44 cpu0 - drwxr-xr-x 3 root root 0 Sep 19 07:44 cpu1 - drwxr-xr-x 3 root root 0 Sep 19 07:44 cpu2 - drwxr-xr-x 3 root root 0 Sep 19 07:44 cpu3 - drwxr-xr-x 3 root root 0 Sep 19 07:44 cpu4 - drwxr-xr-x 3 root root 0 Sep 19 07:44 cpu5 - drwxr-xr-x 3 root root 0 Sep 19 07:44 cpu6 - drwxr-xr-x 3 root root 0 Sep 19 07:48 cpu7 - -Under each directory you would find an "online" file which is the control -file to logically online/offline a processor. - -Q: Does hot-add/hot-remove refer to physical add/remove of cpus? -A: The usage of hot-add/remove may not be very consistently used in the code. -CONFIG_HOTPLUG_CPU enables logical online/offline capability in the kernel. -To support physical addition/removal, one would need some BIOS hooks and -the platform should have something like an attention button in PCI hotplug. -CONFIG_ACPI_HOTPLUG_CPU enables ACPI support for physical add/remove of CPUs. - -Q: How do I logically offline a CPU? -A: Do the following. - - #echo 0 > /sys/devices/system/cpu/cpuX/online - -Once the logical offline is successful, check - - #cat /proc/interrupts - -You should now not see the CPU that you removed. Also online file will report -the state as 0 when a CPU is offline and 1 when it's online. - - #To display the current cpu state. - #cat /sys/devices/system/cpu/cpuX/online - -Q: Why can't I remove CPU0 on some systems? -A: Some architectures may have some special dependency on a certain CPU. - -For e.g in IA64 platforms we have ability to send platform interrupts to the -OS. a.k.a Corrected Platform Error Interrupts (CPEI). In current ACPI -specifications, we didn't have a way to change the target CPU. Hence if the -current ACPI version doesn't support such re-direction, we disable that CPU -by making it not-removable. - -In such cases you will also notice that the online file is missing under cpu0. - -Q: Is CPU0 removable on X86? -A: Yes. If kernel is compiled with CONFIG_BOOTPARAM_HOTPLUG_CPU0=y, CPU0 is -removable by default. Otherwise, CPU0 is also removable by kernel option -cpu0_hotplug. - -But some features depend on CPU0. Two known dependencies are: - -1. Resume from hibernate/suspend depends on CPU0. Hibernate/suspend will fail if -CPU0 is offline and you need to online CPU0 before hibernate/suspend can -continue. -2. PIC interrupts also depend on CPU0. CPU0 can't be removed if a PIC interrupt -is detected. - -It's said poweroff/reboot may depend on CPU0 on some machines although I haven't -seen any poweroff/reboot failure so far after CPU0 is offline on a few tested -machines. - -Please let me know if you know or see any other dependencies of CPU0. - -If the dependencies are under your control, you can turn on CPU0 hotplug feature -either by CONFIG_BOOTPARAM_HOTPLUG_CPU0 or by kernel parameter cpu0_hotplug. - ---Fenghua Yu <fenghua.yu@intel.com> - -Q: How do I find out if a particular CPU is not removable? -A: Depending on the implementation, some architectures may show this by the -absence of the "online" file. This is done if it can be determined ahead of -time that this CPU cannot be removed. - -In some situations, this can be a run time check, i.e if you try to remove the -last CPU, this will not be permitted. You can find such failures by -investigating the return value of the "echo" command. - -Q: What happens when a CPU is being logically offlined? -A: The following happen, listed in no particular order :-) - -- A notification is sent to in-kernel registered modules by sending an event - CPU_DOWN_PREPARE or CPU_DOWN_PREPARE_FROZEN, depending on whether or not the - CPU is being offlined while tasks are frozen due to a suspend operation in - progress -- All processes are migrated away from this outgoing CPU to new CPUs. - The new CPU is chosen from each process' current cpuset, which may be - a subset of all online CPUs. -- All interrupts targeted to this CPU are migrated to a new CPU -- timers/bottom half/task lets are also migrated to a new CPU -- Once all services are migrated, kernel calls an arch specific routine - __cpu_disable() to perform arch specific cleanup. -- Once this is successful, an event for successful cleanup is sent by an event - CPU_DEAD (or CPU_DEAD_FROZEN if tasks are frozen due to a suspend while the - CPU is being offlined). - - "It is expected that each service cleans up when the CPU_DOWN_PREPARE - notifier is called, when CPU_DEAD is called it's expected there is nothing - running on behalf of this CPU that was offlined" - -Q: If I have some kernel code that needs to be aware of CPU arrival and - departure, how to i arrange for proper notification? -A: This is what you would need in your kernel code to receive notifications. - - #include <linux/cpu.h> - static int foobar_cpu_callback(struct notifier_block *nfb, - unsigned long action, void *hcpu) - { - unsigned int cpu = (unsigned long)hcpu; - - switch (action) { - case CPU_ONLINE: - case CPU_ONLINE_FROZEN: - foobar_online_action(cpu); - break; - case CPU_DEAD: - case CPU_DEAD_FROZEN: - foobar_dead_action(cpu); - break; - } - return NOTIFY_OK; - } - - static struct notifier_block foobar_cpu_notifier = - { - .notifier_call = foobar_cpu_callback, - }; - -You need to call register_cpu_notifier() from your init function. -Init functions could be of two types: -1. early init (init function called when only the boot processor is online). -2. late init (init function called _after_ all the CPUs are online). - -For the first case, you should add the following to your init function - - register_cpu_notifier(&foobar_cpu_notifier); - -For the second case, you should add the following to your init function - - register_hotcpu_notifier(&foobar_cpu_notifier); - -You can fail PREPARE notifiers if something doesn't work to prepare resources. -This will stop the activity and send a following CANCELED event back. - -CPU_DEAD should not be failed, its just a goodness indication, but bad -things will happen if a notifier in path sent a BAD notify code. - -Q: I don't see my action being called for all CPUs already up and running? -A: Yes, CPU notifiers are called only when new CPUs are on-lined or offlined. - If you need to perform some action for each CPU already in the system, then - do this: - - for_each_online_cpu(i) { - foobar_cpu_callback(&foobar_cpu_notifier, CPU_UP_PREPARE, i); - foobar_cpu_callback(&foobar_cpu_notifier, CPU_ONLINE, i); - } - - However, if you want to register a hotplug callback, as well as perform - some initialization for CPUs that are already online, then do this: - - Version 1: (Correct) - --------- - - cpu_notifier_register_begin(); - - for_each_online_cpu(i) { - foobar_cpu_callback(&foobar_cpu_notifier, - CPU_UP_PREPARE, i); - foobar_cpu_callback(&foobar_cpu_notifier, - CPU_ONLINE, i); - } - - /* Note the use of the double underscored version of the API */ - __register_cpu_notifier(&foobar_cpu_notifier); - - cpu_notifier_register_done(); - - Note that the following code is *NOT* the right way to achieve this, - because it is prone to an ABBA deadlock between the cpu_add_remove_lock - and the cpu_hotplug.lock. - - Version 2: (Wrong!) - --------- - - get_online_cpus(); - - for_each_online_cpu(i) { - foobar_cpu_callback(&foobar_cpu_notifier, - CPU_UP_PREPARE, i); - foobar_cpu_callback(&foobar_cpu_notifier, - CPU_ONLINE, i); - } - - register_cpu_notifier(&foobar_cpu_notifier); - - put_online_cpus(); - - So always use the first version shown above when you want to register - callbacks as well as initialize the already online CPUs. - - -Q: If I would like to develop CPU hotplug support for a new architecture, - what do I need at a minimum? -A: The following are what is required for CPU hotplug infrastructure to work - correctly. - - - Make sure you have an entry in Kconfig to enable CONFIG_HOTPLUG_CPU - - __cpu_up() - Arch interface to bring up a CPU - - __cpu_disable() - Arch interface to shutdown a CPU, no more interrupts - can be handled by the kernel after the routine - returns. Including local APIC timers etc are - shutdown. - - __cpu_die() - This actually supposed to ensure death of the CPU. - Actually look at some example code in other arch - that implement CPU hotplug. The processor is taken - down from the idle() loop for that specific - architecture. __cpu_die() typically waits for some - per_cpu state to be set, to ensure the processor - dead routine is called to be sure positively. - -Q: I need to ensure that a particular CPU is not removed when there is some - work specific to this CPU in progress. -A: There are two ways. If your code can be run in interrupt context, use - smp_call_function_single(), otherwise use work_on_cpu(). Note that - work_on_cpu() is slow, and can fail due to out of memory: - - int my_func_on_cpu(int cpu) - { - int err; - get_online_cpus(); - if (!cpu_online(cpu)) - err = -EINVAL; - else -#if NEEDS_BLOCKING - err = work_on_cpu(cpu, __my_func_on_cpu, NULL); -#else - smp_call_function_single(cpu, __my_func_on_cpu, &err, - true); -#endif - put_online_cpus(); - return err; - } - -Q: How do we determine how many CPUs are available for hotplug. -A: There is no clear spec defined way from ACPI that can give us that - information today. Based on some input from Natalie of Unisys, - that the ACPI MADT (Multiple APIC Description Tables) marks those possible - CPUs in a system with disabled status. - - Andi implemented some simple heuristics that count the number of disabled - CPUs in MADT as hotpluggable CPUS. In the case there are no disabled CPUS - we assume 1/2 the number of CPUs currently present can be hotplugged. - - Caveat: ACPI MADT can only provide 256 entries in systems with only ACPI 2.0c - or earlier ACPI version supported, because the apicid field in MADT is only - 8 bits. From ACPI 3.0, this limitation was removed since the apicid field - was extended to 32 bits with x2APIC introduced. - -User Space Notification - -Hotplug support for devices is common in Linux today. Its being used today to -support automatic configuration of network, usb and pci devices. A hotplug -event can be used to invoke an agent script to perform the configuration task. - -You can add /etc/hotplug/cpu.agent to handle hotplug notification user space -scripts. - - #!/bin/bash - # $Id: cpu.agent - # Kernel hotplug params include: - #ACTION=%s [online or offline] - #DEVPATH=%s - # - cd /etc/hotplug - . ./hotplug.functions - - case $ACTION in - online) - echo `date` ":cpu.agent" add cpu >> /tmp/hotplug.txt - ;; - offline) - echo `date` ":cpu.agent" remove cpu >>/tmp/hotplug.txt - ;; - *) - debug_mesg CPU $ACTION event not supported - exit 1 - ;; - esac diff --git a/Documentation/dev-tools/sparse.rst b/Documentation/dev-tools/sparse.rst index 78aa00a604a0..ffdcc97f6f5a 100644 --- a/Documentation/dev-tools/sparse.rst +++ b/Documentation/dev-tools/sparse.rst @@ -103,3 +103,9 @@ have already built it. The optional make variable CF can be used to pass arguments to sparse. The build system passes -Wbitwise to sparse automatically. + +Checking RCU annotations +~~~~~~~~~~~~~~~~~~~~~~~~ + +RCU annotations are not checked by default. To enable RCU annotation +checks, include -DCONFIG_SPARSE_RCU_POINTER in your CF flags. diff --git a/Documentation/devicetree/bindings/bus/qcom,ebi2.txt b/Documentation/devicetree/bindings/bus/qcom,ebi2.txt index 920681f552db..5a7d567f6833 100644 --- a/Documentation/devicetree/bindings/bus/qcom,ebi2.txt +++ b/Documentation/devicetree/bindings/bus/qcom,ebi2.txt @@ -51,7 +51,7 @@ Required properties: - compatible: should be one of: "qcom,msm8660-ebi2" "qcom,apq8060-ebi2" -- #address-cells: shoule be <2>: the first cell is the chipselect, +- #address-cells: should be <2>: the first cell is the chipselect, the second cell is the offset inside the memory range - #size-cells: should be <1> - ranges: should be set to: @@ -64,7 +64,7 @@ Required properties: - reg: two ranges of registers: EBI2 config and XMEM config areas - reg-names: should be "ebi2", "xmem" - clocks: two clocks, EBI_2X and EBI -- clock-names: shoule be "ebi2x", "ebi2" +- clock-names: should be "ebi2x", "ebi2" Optional subnodes: - Nodes inside the EBI2 will be considered device nodes. @@ -100,7 +100,7 @@ Optional properties arrays for FAST chip selects: assertion, with respect to the cycle where ADV (address valid) is asserted. 2 means 2 cycles between ADV and OE. Valid values 0, 1, 2 or 3. - qcom,xmem-read-hold-cycles: the length in cycles of the first segment of a - read transfer. For a single read trandfer this will be the time from CS + read transfer. For a single read transfer this will be the time from CS assertion to OE assertion. Valid values 0 thru 15. diff --git a/Documentation/devicetree/bindings/clock/mvebu-gated-clock.txt b/Documentation/devicetree/bindings/clock/mvebu-gated-clock.txt index cb8542d910b3..5142efc8099d 100644 --- a/Documentation/devicetree/bindings/clock/mvebu-gated-clock.txt +++ b/Documentation/devicetree/bindings/clock/mvebu-gated-clock.txt @@ -117,7 +117,7 @@ ID Clock Peripheral 25 tdm Time Division Mplx 28 xor1 XOR DMA 1 29 sata1lnk -30 sata1 SATA Host 0 +30 sata1 SATA Host 1 The following is a list of provided IDs for Dove: ID Clock Peripheral diff --git a/Documentation/devicetree/bindings/display/arm,pl11x.txt b/Documentation/devicetree/bindings/display/arm,pl11x.txt index 3e3039a8a253..ef89ab46b2c9 100644 --- a/Documentation/devicetree/bindings/display/arm,pl11x.txt +++ b/Documentation/devicetree/bindings/display/arm,pl11x.txt @@ -22,7 +22,7 @@ Required properties: - clocks: contains phandle and clock specifier pairs for the entries in the clock-names property. See - Documentation/devicetree/binding/clock/clock-bindings.txt + Documentation/devicetree/bindings/clock/clock-bindings.txt Optional properties: diff --git a/Documentation/devicetree/bindings/display/bridge/analogix_dp.txt b/Documentation/devicetree/bindings/display/bridge/analogix_dp.txt index 4a0f4f7682ad..0c7473dd0e51 100644 --- a/Documentation/devicetree/bindings/display/bridge/analogix_dp.txt +++ b/Documentation/devicetree/bindings/display/bridge/analogix_dp.txt @@ -33,7 +33,7 @@ Optional properties for dp-controller: in Documentation/devicetree/bindings/media/video-interfaces.txt, please refer to the SoC specific binding document: * Documentation/devicetree/bindings/display/exynos/exynos_dp.txt - * Documentation/devicetree/bindings/video/analogix_dp-rockchip.txt + * Documentation/devicetree/bindings/display/rockchip/analogix_dp-rockchip.txt [1]: Documentation/devicetree/bindings/media/video-interfaces.txt ------------------------------------------------------------------------------- diff --git a/Documentation/devicetree/bindings/video/bridge/anx7814.txt b/Documentation/devicetree/bindings/display/bridge/anx7814.txt index b2a22c28c9b3..b2a22c28c9b3 100644 --- a/Documentation/devicetree/bindings/video/bridge/anx7814.txt +++ b/Documentation/devicetree/bindings/display/bridge/anx7814.txt diff --git a/Documentation/devicetree/bindings/video/bridge/sil-sii8620.txt b/Documentation/devicetree/bindings/display/bridge/sil-sii8620.txt index 9409d9c6a260..9409d9c6a260 100644 --- a/Documentation/devicetree/bindings/video/bridge/sil-sii8620.txt +++ b/Documentation/devicetree/bindings/display/bridge/sil-sii8620.txt diff --git a/Documentation/devicetree/bindings/display/cirrus,clps711x-fb.txt b/Documentation/devicetree/bindings/display/cirrus,clps711x-fb.txt index e9c65746e2f1..b0e506610400 100644 --- a/Documentation/devicetree/bindings/display/cirrus,clps711x-fb.txt +++ b/Documentation/devicetree/bindings/display/cirrus,clps711x-fb.txt @@ -6,7 +6,7 @@ Required properties: location and size of the framebuffer memory. - clocks : phandle + clock specifier pair of the FB reference clock. - display : phandle to a display node as described in - Documentation/devicetree/bindings/display/display-timing.txt. + Documentation/devicetree/bindings/display/panel/display-timing.txt. Additionally, the display node has to define properties: - bits-per-pixel: Bits per pixel. - ac-prescale : LCD AC bias frequency. This frequency is the required diff --git a/Documentation/devicetree/bindings/display/exynos/exynos7-decon.txt b/Documentation/devicetree/bindings/display/exynos/exynos7-decon.txt index 3938caacf11c..027d6c210f7e 100644 --- a/Documentation/devicetree/bindings/display/exynos/exynos7-decon.txt +++ b/Documentation/devicetree/bindings/display/exynos/exynos7-decon.txt @@ -38,7 +38,7 @@ Optional Properties: Can be used in case timings cannot be provided otherwise or to override timings provided by the panel. -[1]: Documentation/devicetree/bindings/display/display-timing.txt +[1]: Documentation/devicetree/bindings/display/panel/display-timing.txt Example: diff --git a/Documentation/devicetree/bindings/display/exynos/samsung-fimd.txt b/Documentation/devicetree/bindings/display/exynos/samsung-fimd.txt index c7c6b9af87ac..18645e0228b0 100644 --- a/Documentation/devicetree/bindings/display/exynos/samsung-fimd.txt +++ b/Documentation/devicetree/bindings/display/exynos/samsung-fimd.txt @@ -83,7 +83,7 @@ in [2]. The following are properties specific to those nodes: 3 - for parallel output, 4 - for write-back interface -[1]: Documentation/devicetree/bindings/display/display-timing.txt +[1]: Documentation/devicetree/bindings/display/panel/display-timing.txt [2]: Documentation/devicetree/bindings/media/video-interfaces.txt Example: diff --git a/Documentation/devicetree/bindings/display/imx/fsl,imx-fb.txt b/Documentation/devicetree/bindings/display/imx/fsl,imx-fb.txt index 00d5f8ea7ec6..7a5c0e204c8e 100644 --- a/Documentation/devicetree/bindings/display/imx/fsl,imx-fb.txt +++ b/Documentation/devicetree/bindings/display/imx/fsl,imx-fb.txt @@ -9,7 +9,7 @@ Required properties: Required nodes: - display: Phandle to a display node as described in - Documentation/devicetree/bindings/display/display-timing.txt + Documentation/devicetree/bindings/display/panel/display-timing.txt Additional, the display node has to define properties: - bits-per-pixel: Bits per pixel - fsl,pcr: LCDC PCR value diff --git a/Documentation/devicetree/bindings/display/imx/ldb.txt b/Documentation/devicetree/bindings/display/imx/ldb.txt index a407462c885e..38c637fa39dd 100644 --- a/Documentation/devicetree/bindings/display/imx/ldb.txt +++ b/Documentation/devicetree/bindings/display/imx/ldb.txt @@ -64,7 +64,7 @@ Required properties: Optional properties (required if display-timings are used): - ddc-i2c-bus: phandle of an I2C controller used for DDC EDID probing - display-timings : A node that describes the display timings as defined in - Documentation/devicetree/bindings/display/display-timing.txt. + Documentation/devicetree/bindings/display/panel/display-timing.txt. - fsl,data-mapping : should be "spwg" or "jeida" This describes how the color bits are laid out in the serialized LVDS signal. diff --git a/Documentation/devicetree/bindings/display/mediatek/mediatek,disp.txt b/Documentation/devicetree/bindings/display/mediatek/mediatek,disp.txt index db6e77edbea8..708f5664a316 100644 --- a/Documentation/devicetree/bindings/display/mediatek/mediatek,disp.txt +++ b/Documentation/devicetree/bindings/display/mediatek/mediatek,disp.txt @@ -55,7 +55,7 @@ Required properties (DMA function blocks): "mediatek,<chip>-disp-rdma" "mediatek,<chip>-disp-wdma" - larb: Should contain a phandle pointing to the local arbiter device as defined - in Documentation/devicetree/bindings/soc/mediatek/mediatek,smi-larb.txt + in Documentation/devicetree/bindings/memory-controllers/mediatek,smi-larb.txt - iommus: Should point to the respective IOMMU block with master port as argument, see Documentation/devicetree/bindings/iommu/mediatek,iommu.txt for details. diff --git a/Documentation/devicetree/bindings/display/msm/dsi.txt b/Documentation/devicetree/bindings/display/msm/dsi.txt index 6b1cab17f52d..fa00e62e1cf6 100644 --- a/Documentation/devicetree/bindings/display/msm/dsi.txt +++ b/Documentation/devicetree/bindings/display/msm/dsi.txt @@ -108,7 +108,7 @@ Optional properties: - qcom,dsi-phy-regulator-ldo-mode: Boolean value indicating if the LDO mode PHY regulator is wanted. -[1] Documentation/devicetree/bindings/clocks/clock-bindings.txt +[1] Documentation/devicetree/bindings/clock/clock-bindings.txt [2] Documentation/devicetree/bindings/graph.txt [3] Documentation/devicetree/bindings/media/video-interfaces.txt [4] Documentation/devicetree/bindings/display/panel/ diff --git a/Documentation/devicetree/bindings/display/msm/edp.txt b/Documentation/devicetree/bindings/display/msm/edp.txt index 3a20f6ea5898..e63032be5401 100644 --- a/Documentation/devicetree/bindings/display/msm/edp.txt +++ b/Documentation/devicetree/bindings/display/msm/edp.txt @@ -10,7 +10,7 @@ Required properties: - interrupts: The interrupt signal from the eDP block. - power-domains: Should be <&mmcc MDSS_GDSC>. - clocks: device clocks - See Documentation/devicetree/bindings/clocks/clock-bindings.txt for details. + See Documentation/devicetree/bindings/clock/clock-bindings.txt for details. - clock-names: the following clocks are required: * "core_clk" * "iface_clk" diff --git a/Documentation/devicetree/bindings/display/msm/hdmi.txt b/Documentation/devicetree/bindings/display/msm/hdmi.txt index 2ad578984fcf..2d306f402d18 100644 --- a/Documentation/devicetree/bindings/display/msm/hdmi.txt +++ b/Documentation/devicetree/bindings/display/msm/hdmi.txt @@ -49,7 +49,7 @@ Required properties: * "hdmi_tx_l4" - power-domains: Should be <&mmcc MDSS_GDSC>. - clocks: device clocks - See Documentation/devicetree/bindings/clocks/clock-bindings.txt for details. + See Documentation/devicetree/bindings/clock/clock-bindings.txt for details. - core-vdda-supply: phandle to vdda regulator device node Example: diff --git a/Documentation/devicetree/bindings/display/panel/panel-dpi.txt b/Documentation/devicetree/bindings/display/panel/panel-dpi.txt index b52ac52757df..d4add13e592d 100644 --- a/Documentation/devicetree/bindings/display/panel/panel-dpi.txt +++ b/Documentation/devicetree/bindings/display/panel/panel-dpi.txt @@ -12,7 +12,7 @@ Optional properties: Required nodes: - "panel-timing" containing video timings - (Documentation/devicetree/bindings/display/display-timing.txt) + (Documentation/devicetree/bindings/display/panel/display-timing.txt) - Video port for DPI input Example diff --git a/Documentation/devicetree/bindings/display/panel/samsung,ld9040.txt b/Documentation/devicetree/bindings/display/panel/samsung,ld9040.txt index fc595d9b985b..354d4d1df4ff 100644 --- a/Documentation/devicetree/bindings/display/panel/samsung,ld9040.txt +++ b/Documentation/devicetree/bindings/display/panel/samsung,ld9040.txt @@ -20,7 +20,7 @@ The device node can contain one 'port' child node with one child 'endpoint' node, according to the bindings defined in [3]. This node should describe panel's video bus. -[1]: Documentation/devicetree/bindings/display/display-timing.txt +[1]: Documentation/devicetree/bindings/display/panel/display-timing.txt [2]: Documentation/devicetree/bindings/spi/spi-bus.txt [3]: Documentation/devicetree/bindings/media/video-interfaces.txt diff --git a/Documentation/devicetree/bindings/display/panel/samsung,s6e8aa0.txt b/Documentation/devicetree/bindings/display/panel/samsung,s6e8aa0.txt index 25701c81b5e0..9e766c5f86da 100644 --- a/Documentation/devicetree/bindings/display/panel/samsung,s6e8aa0.txt +++ b/Documentation/devicetree/bindings/display/panel/samsung,s6e8aa0.txt @@ -21,7 +21,7 @@ The device node can contain one 'port' child node with one child 'endpoint' node, according to the bindings defined in [2]. This node should describe panel's video bus. -[1]: Documentation/devicetree/bindings/display/display-timing.txt +[1]: Documentation/devicetree/bindings/display/panel/display-timing.txt [2]: Documentation/devicetree/bindings/media/video-interfaces.txt Example: diff --git a/Documentation/devicetree/bindings/display/rockchip/analogix_dp-rockchip.txt b/Documentation/devicetree/bindings/display/rockchip/analogix_dp-rockchip.txt index 01cced1c2a18..47665a12786f 100644 --- a/Documentation/devicetree/bindings/display/rockchip/analogix_dp-rockchip.txt +++ b/Documentation/devicetree/bindings/display/rockchip/analogix_dp-rockchip.txt @@ -35,7 +35,7 @@ Optional property for different chips: Required elements: "grf" For the below properties, please refer to Analogix DP binding document: - * Documentation/devicetree/bindings/drm/bridge/analogix_dp.txt + * Documentation/devicetree/bindings/display/bridge/analogix_dp.txt - phys (required) - phy-names (required) - hpd-gpios (optional) diff --git a/Documentation/devicetree/bindings/display/tilcdc/panel.txt b/Documentation/devicetree/bindings/display/tilcdc/panel.txt index f20b31cdc59a..808216310ea2 100644 --- a/Documentation/devicetree/bindings/display/tilcdc/panel.txt +++ b/Documentation/devicetree/bindings/display/tilcdc/panel.txt @@ -15,7 +15,7 @@ Required properties: - display-timings: typical videomode of lcd panel. Multiple video modes can be listed if the panel supports multiple timings, but the 'native-mode' should be the preferred/default resolution. Refer to - Documentation/devicetree/bindings/display/display-timing.txt for display + Documentation/devicetree/bindings/display/panel/display-timing.txt for display timing binding details. Optional properties: diff --git a/Documentation/devicetree/bindings/iommu/arm,smmu.txt b/Documentation/devicetree/bindings/iommu/arm,smmu.txt index e862d1485205..6cdf32d037fc 100644 --- a/Documentation/devicetree/bindings/iommu/arm,smmu.txt +++ b/Documentation/devicetree/bindings/iommu/arm,smmu.txt @@ -36,15 +36,15 @@ conditions. combined interrupt, it must be listed multiple times. - #iommu-cells : See Documentation/devicetree/bindings/iommu/iommu.txt - for details. With a value of 1, each "iommus" entry + for details. With a value of 1, each IOMMU specifier represents a distinct stream ID emitted by that device into the relevant SMMU. SMMUs with stream matching support and complex masters - may use a value of 2, where the second cell represents - an SMR mask to combine with the ID in the first cell. - Care must be taken to ensure the set of matched IDs - does not result in conflicts. + may use a value of 2, where the second cell of the + IOMMU specifier represents an SMR mask to combine with + the ID in the first cell. Care must be taken to ensure + the set of matched IDs does not result in conflicts. ** System MMU optional properties: diff --git a/Documentation/devicetree/bindings/mfd/as3722.txt b/Documentation/devicetree/bindings/mfd/as3722.txt index 4f64b2a73169..0b2a6099aa20 100644 --- a/Documentation/devicetree/bindings/mfd/as3722.txt +++ b/Documentation/devicetree/bindings/mfd/as3722.txt @@ -122,8 +122,7 @@ Following are properties of regulator subnode. Power-off: ========= -AS3722 supports the system power off by turning off all its rail. This -is provided through pm_power_off. +AS3722 supports the system power off by turning off all its rails. The device node should have the following properties to enable this functionality ams,system-power-controller: Boolean, to enable the power off functionality diff --git a/Documentation/devicetree/bindings/mfd/omap-usb-host.txt b/Documentation/devicetree/bindings/mfd/omap-usb-host.txt index 4721b2d521e4..aa1eaa59581b 100644 --- a/Documentation/devicetree/bindings/mfd/omap-usb-host.txt +++ b/Documentation/devicetree/bindings/mfd/omap-usb-host.txt @@ -64,8 +64,8 @@ Required properties if child node exists: Properties for children: The OMAP HS USB Host subsystem contains EHCI and OHCI controllers. -See Documentation/devicetree/bindings/usb/omap-ehci.txt and -omap3-ohci.txt +See Documentation/devicetree/bindings/usb/ehci-omap.txt and +Documentation/devicetree/bindings/usb/ohci-omap3.txt. Example for OMAP4: diff --git a/Documentation/devicetree/bindings/net/marvell-pp2.txt b/Documentation/devicetree/bindings/net/marvell-pp2.txt index aa4f4230bfd7..4754364df4c6 100644 --- a/Documentation/devicetree/bindings/net/marvell-pp2.txt +++ b/Documentation/devicetree/bindings/net/marvell-pp2.txt @@ -27,9 +27,7 @@ Optional properties (port): - marvell,loopback: port is loopback mode - phy: a phandle to a phy node defining the PHY address (as the reg - property, a single integer). Note: if this property isn't present, - then fixed link is assumed, and the 'fixed-link' property is - mandatory. + property, a single integer). Example: diff --git a/Documentation/devicetree/bindings/pci/pci-iommu.txt b/Documentation/devicetree/bindings/pci/pci-iommu.txt index 56c829621b9a..0def586fdcdf 100644 --- a/Documentation/devicetree/bindings/pci/pci-iommu.txt +++ b/Documentation/devicetree/bindings/pci/pci-iommu.txt @@ -32,17 +32,17 @@ PCI root complex Optional properties ------------------- -- iommu-map: Maps a Requester ID to an IOMMU and associated iommu-specifier +- iommu-map: Maps a Requester ID to an IOMMU and associated IOMMU specifier data. The property is an arbitrary number of tuples of (rid-base,iommu,iommu-base,length). Any RID r in the interval [rid-base, rid-base + length) is associated with - the listed IOMMU, with the iommu-specifier (r - rid-base + iommu-base). + the listed IOMMU, with the IOMMU specifier (r - rid-base + iommu-base). - iommu-map-mask: A mask to be applied to each Requester ID prior to being - mapped to an iommu-specifier per the iommu-map property. + mapped to an IOMMU specifier per the iommu-map property. Example (1) diff --git a/Documentation/devicetree/bindings/power/reset/gpio-poweroff.txt b/Documentation/devicetree/bindings/power/reset/gpio-poweroff.txt index d4eab9227ea4..e62d53d844cc 100644 --- a/Documentation/devicetree/bindings/power/reset/gpio-poweroff.txt +++ b/Documentation/devicetree/bindings/power/reset/gpio-poweroff.txt @@ -2,12 +2,12 @@ Driver a GPIO line that can be used to turn the power off. The driver supports both level triggered and edge triggered power off. At driver load time, the driver will request the given gpio line and -install a pm_power_off handler. If the optional properties 'input' is -not found, the GPIO line will be driven in the inactive +install a handler to power off the system. If the optional properties +'input' is not found, the GPIO line will be driven in the inactive state. Otherwise its configured as an input. -When the pm_power_off is called, the gpio is configured as an output, -and drive active, so triggering a level triggered power off +When the power-off handler is called, the gpio is configured as an +output, and drive active, so triggering a level triggered power off condition. This will also cause an inactive->active edge condition, so triggering positive edge triggered power off. After a delay of 100ms, the GPIO is set to inactive, thus causing an active->inactive edge, @@ -24,7 +24,7 @@ Required properties: Optional properties: - input : Initially configure the GPIO line as an input. Only reconfigure - it to an output when the pm_power_off function is called. If this optional + it to an output when the power-off handler is called. If this optional property is not specified, the GPIO is initialized as an output in its inactive state. diff --git a/Documentation/devicetree/bindings/power/reset/qnap-poweroff.txt b/Documentation/devicetree/bindings/power/reset/qnap-poweroff.txt index af25e77c0e0c..c363d7173129 100644 --- a/Documentation/devicetree/bindings/power/reset/qnap-poweroff.txt +++ b/Documentation/devicetree/bindings/power/reset/qnap-poweroff.txt @@ -3,8 +3,7 @@ QNAP NAS devices have a microcontroller controlling the main power supply. This microcontroller is connected to UART1 of the Kirkwood and Orion5x SoCs. Sending the character 'A', at 19200 baud, tells the -microcontroller to turn the power off. This driver adds a handler to -pm_power_off which is called to turn the power off. +microcontroller to turn the power off. Synology NAS devices use a similar scheme, but a different baud rate, 9600, and a different character, '1'. diff --git a/Documentation/devicetree/bindings/serial/fsl-imx-uart.txt b/Documentation/devicetree/bindings/serial/fsl-imx-uart.txt index 1e82802d8e32..574c3a2c77d5 100644 --- a/Documentation/devicetree/bindings/serial/fsl-imx-uart.txt +++ b/Documentation/devicetree/bindings/serial/fsl-imx-uart.txt @@ -6,11 +6,13 @@ Required properties: - interrupts : Should contain uart interrupt Optional properties: -- uart-has-rtscts : Indicate the uart has rts and cts - fsl,irda-mode : Indicate the uart supports irda mode - fsl,dte-mode : Indicate the uart works in DTE mode. The uart works in DCE mode by default. +Please check Documentation/devicetree/bindings/serial/serial.txt +for the complete list of generic properties. + Note: Each uart controller should have an alias correctly numbered in "aliases" node. diff --git a/Documentation/devicetree/bindings/soc/fsl/qman-portals.txt b/Documentation/devicetree/bindings/soc/fsl/qman-portals.txt index 47e46ccbc170..5a34f3ab7bea 100644 --- a/Documentation/devicetree/bindings/soc/fsl/qman-portals.txt +++ b/Documentation/devicetree/bindings/soc/fsl/qman-portals.txt @@ -5,7 +5,6 @@ Copyright (C) 2008 - 2014 Freescale Semiconductor Inc. CONTENTS - QMan Portal - - QMan Pool Channel - Example QMan Portal Node @@ -82,25 +81,6 @@ These subnodes should have the following properties: Definition: The phandle to the particular hardware device that this portal is connected to. -DPAA QMan Pool Channel Nodes - -Pool Channels are defined with the following properties. - -PROPERTIES - -- compatible - Usage: Required - Value type: <stringlist> - Definition: Must include "fsl,qman-pool-channel" - May include "fsl,<SoC>-qman-pool-channel" - -- fsl,qman-channel-id - Usage: Required - Value type: <u32> - Definition: The hardware index of the channel. This can also be - determined by dividing any of the channel's 8 work queue - IDs by 8 - EXAMPLE The example below shows a (P4080) QMan portals container/bus node with two portals diff --git a/Documentation/devicetree/bindings/usb/ehci-omap.txt b/Documentation/devicetree/bindings/usb/ehci-omap.txt index 3dc231c832b0..d77e11a975a2 100644 --- a/Documentation/devicetree/bindings/usb/ehci-omap.txt +++ b/Documentation/devicetree/bindings/usb/ehci-omap.txt @@ -29,4 +29,3 @@ usbhsehci: ehci@4a064c00 { &usbhsehci { phys = <&hsusb1_phy 0 &hsusb3_phy>; }; - diff --git a/Documentation/devicetree/bindings/vendor-prefixes.txt b/Documentation/devicetree/bindings/vendor-prefixes.txt index cf1c36616507..1d0bb2c9def4 100644 --- a/Documentation/devicetree/bindings/vendor-prefixes.txt +++ b/Documentation/devicetree/bindings/vendor-prefixes.txt @@ -104,11 +104,13 @@ everest Everest Semiconductor Co. Ltd. everspin Everspin Technologies, Inc. excito Excito ezchip EZchip Semiconductor +faraday Faraday Technology Corporation fcs Fairchild Semiconductor firefly Firefly focaltech FocalTech Systems Co.,Ltd friendlyarm Guangzhou FriendlyARM Computer Tech Co., Ltd fsl Freescale Semiconductor +fujitsu Fujitsu Ltd. ge General Electric Company geekbuying GeekBuying gef GE Fanuc Intelligent Platforms Embedded Systems, Inc. diff --git a/Documentation/dontdiff b/Documentation/dontdiff index a23edccd2059..77b92221f951 100644 --- a/Documentation/dontdiff +++ b/Documentation/dontdiff @@ -116,9 +116,11 @@ crc32table.h* cscope.* defkeymap.c devlist.h* +devicetable-offsets.h dnotify_test docproc dslm +dtc elf2ecoff elfconfig.h* evergreen_reg_safe.h @@ -153,8 +155,8 @@ keywords.c ksym.c* ksym.h* kxgettext -lex.c -lex.*.c +*lex.c +*lex.*.c linux logo_*.c logo_*_clut224.c @@ -215,6 +217,7 @@ series setup setup.bin setup.elf +sortextable sImage sm_tbl* split-include diff --git a/Documentation/driver-api/device-io.rst b/Documentation/driver-api/device-io.rst new file mode 100644 index 000000000000..b00b23903078 --- /dev/null +++ b/Documentation/driver-api/device-io.rst @@ -0,0 +1,201 @@ +.. Copyright 2001 Matthew Wilcox +.. +.. This documentation is free software; you can redistribute +.. it and/or modify it under the terms of the GNU General Public +.. License as published by the Free Software Foundation; either +.. version 2 of the License, or (at your option) any later +.. version. + +=============================== +Bus-Independent Device Accesses +=============================== + +:Author: Matthew Wilcox +:Author: Alan Cox + +Introduction +============ + +Linux provides an API which abstracts performing IO across all busses +and devices, allowing device drivers to be written independently of bus +type. + +Memory Mapped IO +================ + +Getting Access to the Device +---------------------------- + +The most widely supported form of IO is memory mapped IO. That is, a +part of the CPU's address space is interpreted not as accesses to +memory, but as accesses to a device. Some architectures define devices +to be at a fixed address, but most have some method of discovering +devices. The PCI bus walk is a good example of such a scheme. This +document does not cover how to receive such an address, but assumes you +are starting with one. Physical addresses are of type unsigned long. + +This address should not be used directly. Instead, to get an address +suitable for passing to the accessor functions described below, you +should call :c:func:`ioremap()`. An address suitable for accessing +the device will be returned to you. + +After you've finished using the device (say, in your module's exit +routine), call :c:func:`iounmap()` in order to return the address +space to the kernel. Most architectures allocate new address space each +time you call :c:func:`ioremap()`, and they can run out unless you +call :c:func:`iounmap()`. + +Accessing the device +-------------------- + +The part of the interface most used by drivers is reading and writing +memory-mapped registers on the device. Linux provides interfaces to read +and write 8-bit, 16-bit, 32-bit and 64-bit quantities. Due to a +historical accident, these are named byte, word, long and quad accesses. +Both read and write accesses are supported; there is no prefetch support +at this time. + +The functions are named readb(), readw(), readl(), readq(), +readb_relaxed(), readw_relaxed(), readl_relaxed(), readq_relaxed(), +writeb(), writew(), writel() and writeq(). + +Some devices (such as framebuffers) would like to use larger transfers than +8 bytes at a time. For these devices, the :c:func:`memcpy_toio()`, +:c:func:`memcpy_fromio()` and :c:func:`memset_io()` functions are +provided. Do not use memset or memcpy on IO addresses; they are not +guaranteed to copy data in order. + +The read and write functions are defined to be ordered. That is the +compiler is not permitted to reorder the I/O sequence. When the ordering +can be compiler optimised, you can use __readb() and friends to +indicate the relaxed ordering. Use this with care. + +While the basic functions are defined to be synchronous with respect to +each other and ordered with respect to each other the busses the devices +sit on may themselves have asynchronicity. In particular many authors +are burned by the fact that PCI bus writes are posted asynchronously. A +driver author must issue a read from the same device to ensure that +writes have occurred in the specific cases the author cares. This kind +of property cannot be hidden from driver writers in the API. In some +cases, the read used to flush the device may be expected to fail (if the +card is resetting, for example). In that case, the read should be done +from config space, which is guaranteed to soft-fail if the card doesn't +respond. + +The following is an example of flushing a write to a device when the +driver would like to ensure the write's effects are visible prior to +continuing execution:: + + static inline void + qla1280_disable_intrs(struct scsi_qla_host *ha) + { + struct device_reg *reg; + + reg = ha->iobase; + /* disable risc and host interrupts */ + WRT_REG_WORD(®->ictrl, 0); + /* + * The following read will ensure that the above write + * has been received by the device before we return from this + * function. + */ + RD_REG_WORD(®->ictrl); + ha->flags.ints_enabled = 0; + } + +In addition to write posting, on some large multiprocessing systems +(e.g. SGI Challenge, Origin and Altix machines) posted writes won't be +strongly ordered coming from different CPUs. Thus it's important to +properly protect parts of your driver that do memory-mapped writes with +locks and use the :c:func:`mmiowb()` to make sure they arrive in the +order intended. Issuing a regular readX() will also ensure write ordering, +but should only be used when the +driver has to be sure that the write has actually arrived at the device +(not that it's simply ordered with respect to other writes), since a +full readX() is a relatively expensive operation. + +Generally, one should use :c:func:`mmiowb()` prior to releasing a spinlock +that protects regions using :c:func:`writeb()` or similar functions that +aren't surrounded by readb() calls, which will ensure ordering +and flushing. The following pseudocode illustrates what might occur if +write ordering isn't guaranteed via :c:func:`mmiowb()` or one of the +readX() functions:: + + CPU A: spin_lock_irqsave(&dev_lock, flags) + CPU A: ... + CPU A: writel(newval, ring_ptr); + CPU A: spin_unlock_irqrestore(&dev_lock, flags) + ... + CPU B: spin_lock_irqsave(&dev_lock, flags) + CPU B: writel(newval2, ring_ptr); + CPU B: ... + CPU B: spin_unlock_irqrestore(&dev_lock, flags) + +In the case above, newval2 could be written to ring_ptr before newval. +Fixing it is easy though:: + + CPU A: spin_lock_irqsave(&dev_lock, flags) + CPU A: ... + CPU A: writel(newval, ring_ptr); + CPU A: mmiowb(); /* ensure no other writes beat us to the device */ + CPU A: spin_unlock_irqrestore(&dev_lock, flags) + ... + CPU B: spin_lock_irqsave(&dev_lock, flags) + CPU B: writel(newval2, ring_ptr); + CPU B: ... + CPU B: mmiowb(); + CPU B: spin_unlock_irqrestore(&dev_lock, flags) + +See tg3.c for a real world example of how to use :c:func:`mmiowb()` + +PCI ordering rules also guarantee that PIO read responses arrive after any +outstanding DMA writes from that bus, since for some devices the result of +a readb() call may signal to the driver that a DMA transaction is +complete. In many cases, however, the driver may want to indicate that the +next readb() call has no relation to any previous DMA writes +performed by the device. The driver can use readb_relaxed() for +these cases, although only some platforms will honor the relaxed +semantics. Using the relaxed read functions will provide significant +performance benefits on platforms that support it. The qla2xxx driver +provides examples of how to use readX_relaxed(). In many cases, a majority +of the driver's readX() calls can safely be converted to readX_relaxed() +calls, since only a few will indicate or depend on DMA completion. + +Port Space Accesses +=================== + +Port Space Explained +-------------------- + +Another form of IO commonly supported is Port Space. This is a range of +addresses separate to the normal memory address space. Access to these +addresses is generally not as fast as accesses to the memory mapped +addresses, and it also has a potentially smaller address space. + +Unlike memory mapped IO, no preparation is required to access port +space. + +Accessing Port Space +-------------------- + +Accesses to this space are provided through a set of functions which +allow 8-bit, 16-bit and 32-bit accesses; also known as byte, word and +long. These functions are :c:func:`inb()`, :c:func:`inw()`, +:c:func:`inl()`, :c:func:`outb()`, :c:func:`outw()` and +:c:func:`outl()`. + +Some variants are provided for these functions. Some devices require +that accesses to their ports are slowed down. This functionality is +provided by appending a ``_p`` to the end of the function. +There are also equivalents to memcpy. The :c:func:`ins()` and +:c:func:`outs()` functions copy bytes, words or longs to the given +port. + +Public Functions Provided +========================= + +.. kernel-doc:: arch/x86/include/asm/io.h + :internal: + +.. kernel-doc:: lib/pci_iomap.c + :export: diff --git a/Documentation/driver-api/device_link.rst b/Documentation/driver-api/device_link.rst index 5f5713448703..70e328e16aad 100644 --- a/Documentation/driver-api/device_link.rst +++ b/Documentation/driver-api/device_link.rst @@ -1,3 +1,6 @@ +.. |struct dev_pm_domain| replace:: :c:type:`struct dev_pm_domain <dev_pm_domain>` +.. |struct generic_pm_domain| replace:: :c:type:`struct generic_pm_domain <generic_pm_domain>` + ============ Device links ============ @@ -120,12 +123,11 @@ Examples is the same as if the MMU was the parent of the master device. The fact that both devices share the same power domain would normally - suggest usage of a :c:type:`struct dev_pm_domain` or :c:type:`struct - generic_pm_domain`, however these are not independent devices that - happen to share a power switch, but rather the MMU device serves the - busmaster device and is useless without it. A device link creates a - synthetic hierarchical relationship between the devices and is thus - more apt. + suggest usage of a |struct dev_pm_domain| or |struct generic_pm_domain|, + however these are not independent devices that happen to share a power + switch, but rather the MMU device serves the busmaster device and is + useless without it. A device link creates a synthetic hierarchical + relationship between the devices and is thus more apt. * A Thunderbolt host controller comprises a number of PCIe hotplug ports and an NHI device to manage the PCIe switch. On resume from system sleep, @@ -157,7 +159,7 @@ Examples Alternatives ============ -* A :c:type:`struct dev_pm_domain` can be used to override the bus, +* A |struct dev_pm_domain| can be used to override the bus, class or device type callbacks. It is intended for devices sharing a single on/off switch, however it does not guarantee a specific suspend/resume ordering, this needs to be implemented separately. @@ -166,7 +168,7 @@ Alternatives suspended. Furthermore it cannot be used to enforce a specific shutdown ordering or a driver presence dependency. -* A :c:type:`struct generic_pm_domain` is a lot more heavyweight than a +* A |struct generic_pm_domain| is a lot more heavyweight than a device link and does not allow for shutdown ordering or driver presence dependencies. It also cannot be used on ACPI systems. diff --git a/Documentation/driver-api/iio/buffers.rst b/Documentation/driver-api/iio/buffers.rst new file mode 100644 index 000000000000..02c99a6bee18 --- /dev/null +++ b/Documentation/driver-api/iio/buffers.rst @@ -0,0 +1,125 @@ +======= +Buffers +======= + +* struct :c:type:`iio_buffer` — general buffer structure +* :c:func:`iio_validate_scan_mask_onehot` — Validates that exactly one channel + is selected +* :c:func:`iio_buffer_get` — Grab a reference to the buffer +* :c:func:`iio_buffer_put` — Release the reference to the buffer + +The Industrial I/O core offers a way for continuous data capture based on a +trigger source. Multiple data channels can be read at once from +:file:`/dev/iio:device{X}` character device node, thus reducing the CPU load. + +IIO buffer sysfs interface +========================== +An IIO buffer has an associated attributes directory under +:file:`/sys/bus/iio/iio:device{X}/buffer/*`. Here are some of the existing +attributes: + +* :file:`length`, the total number of data samples (capacity) that can be + stored by the buffer. +* :file:`enable`, activate buffer capture. + +IIO buffer setup +================ + +The meta information associated with a channel reading placed in a buffer is +called a scan element . The important bits configuring scan elements are +exposed to userspace applications via the +:file:`/sys/bus/iio/iio:device{X}/scan_elements/*` directory. This file contains +attributes of the following form: + +* :file:`enable`, used for enabling a channel. If and only if its attribute + is non *zero*, then a triggered capture will contain data samples for this + channel. +* :file:`type`, description of the scan element data storage within the buffer + and hence the form in which it is read from user space. + Format is [be|le]:[s|u]bits/storagebitsXrepeat[>>shift] . + * *be* or *le*, specifies big or little endian. + * *s* or *u*, specifies if signed (2's complement) or unsigned. + * *bits*, is the number of valid data bits. + * *storagebits*, is the number of bits (after padding) that it occupies in the + buffer. + * *shift*, if specified, is the shift that needs to be applied prior to + masking out unused bits. + * *repeat*, specifies the number of bits/storagebits repetitions. When the + repeat element is 0 or 1, then the repeat value is omitted. + +For example, a driver for a 3-axis accelerometer with 12 bit resolution where +data is stored in two 8-bits registers as follows:: + + 7 6 5 4 3 2 1 0 + +---+---+---+---+---+---+---+---+ + |D3 |D2 |D1 |D0 | X | X | X | X | (LOW byte, address 0x06) + +---+---+---+---+---+---+---+---+ + + 7 6 5 4 3 2 1 0 + +---+---+---+---+---+---+---+---+ + |D11|D10|D9 |D8 |D7 |D6 |D5 |D4 | (HIGH byte, address 0x07) + +---+---+---+---+---+---+---+---+ + +will have the following scan element type for each axis:: + + $ cat /sys/bus/iio/devices/iio:device0/scan_elements/in_accel_y_type + le:s12/16>>4 + +A user space application will interpret data samples read from the buffer as +two byte little endian signed data, that needs a 4 bits right shift before +masking out the 12 valid bits of data. + +For implementing buffer support a driver should initialize the following +fields in iio_chan_spec definition:: + + struct iio_chan_spec { + /* other members */ + int scan_index + struct { + char sign; + u8 realbits; + u8 storagebits; + u8 shift; + u8 repeat; + enum iio_endian endianness; + } scan_type; + }; + +The driver implementing the accelerometer described above will have the +following channel definition:: + + struct struct iio_chan_spec accel_channels[] = { + { + .type = IIO_ACCEL, + .modified = 1, + .channel2 = IIO_MOD_X, + /* other stuff here */ + .scan_index = 0, + .scan_type = { + .sign = 's', + .realbits = 12, + .storagebits = 16, + .shift = 4, + .endianness = IIO_LE, + }, + } + /* similar for Y (with channel2 = IIO_MOD_Y, scan_index = 1) + * and Z (with channel2 = IIO_MOD_Z, scan_index = 2) axis + */ + } + +Here **scan_index** defines the order in which the enabled channels are placed +inside the buffer. Channels with a lower **scan_index** will be placed before +channels with a higher index. Each channel needs to have a unique +**scan_index**. + +Setting **scan_index** to -1 can be used to indicate that the specific channel +does not support buffered capture. In this case no entries will be created for +the channel in the scan_elements directory. + +More details +============ +.. kernel-doc:: include/linux/iio/buffer.h +.. kernel-doc:: drivers/iio/industrialio-buffer.c + :export: + diff --git a/Documentation/driver-api/iio/core.rst b/Documentation/driver-api/iio/core.rst new file mode 100644 index 000000000000..9a34ae03b679 --- /dev/null +++ b/Documentation/driver-api/iio/core.rst @@ -0,0 +1,182 @@ +============= +Core elements +============= + +The Industrial I/O core offers a unified framework for writing drivers for +many different types of embedded sensors. a standard interface to user space +applications manipulating sensors. The implementation can be found under +:file:`drivers/iio/industrialio-*` + +Industrial I/O Devices +---------------------- + +* struct :c:type:`iio_dev` - industrial I/O device +* :c:func:`iio_device_alloc()` - alocate an :c:type:`iio_dev` from a driver +* :c:func:`iio_device_free()` - free an :c:type:`iio_dev` from a driver +* :c:func:`iio_device_register()` - register a device with the IIO subsystem +* :c:func:`iio_device_unregister()` - unregister a device from the IIO + subsystem + +An IIO device usually corresponds to a single hardware sensor and it +provides all the information needed by a driver handling a device. +Let's first have a look at the functionality embedded in an IIO device +then we will show how a device driver makes use of an IIO device. + +There are two ways for a user space application to interact with an IIO driver. + +1. :file:`/sys/bus/iio/iio:device{X}/`, this represents a hardware sensor + and groups together the data channels of the same chip. +2. :file:`/dev/iio:device{X}`, character device node interface used for + buffered data transfer and for events information retrieval. + +A typical IIO driver will register itself as an :doc:`I2C <../i2c>` or +:doc:`SPI <../spi>` driver and will create two routines, probe and remove. + +At probe: + +1. Call :c:func:`iio_device_alloc()`, which allocates memory for an IIO device. +2. Initialize IIO device fields with driver specific information (e.g. + device name, device channels). +3. Call :c:func:`iio_device_register()`, this registers the device with the + IIO core. After this call the device is ready to accept requests from user + space applications. + +At remove, we free the resources allocated in probe in reverse order: + +1. :c:func:`iio_device_unregister()`, unregister the device from the IIO core. +2. :c:func:`iio_device_free()`, free the memory allocated for the IIO device. + +IIO device sysfs interface +========================== + +Attributes are sysfs files used to expose chip info and also allowing +applications to set various configuration parameters. For device with +index X, attributes can be found under /sys/bus/iio/iio:deviceX/ directory. +Common attributes are: + +* :file:`name`, description of the physical chip. +* :file:`dev`, shows the major:minor pair associated with + :file:`/dev/iio:deviceX` node. +* :file:`sampling_frequency_available`, available discrete set of sampling + frequency values for device. +* Available standard attributes for IIO devices are described in the + :file:`Documentation/ABI/testing/sysfs-bus-iio` file in the Linux kernel + sources. + +IIO device channels +=================== + +struct :c:type:`iio_chan_spec` - specification of a single channel + +An IIO device channel is a representation of a data channel. An IIO device can +have one or multiple channels. For example: + +* a thermometer sensor has one channel representing the temperature measurement. +* a light sensor with two channels indicating the measurements in the visible + and infrared spectrum. +* an accelerometer can have up to 3 channels representing acceleration on X, Y + and Z axes. + +An IIO channel is described by the struct :c:type:`iio_chan_spec`. +A thermometer driver for the temperature sensor in the example above would +have to describe its channel as follows:: + + static const struct iio_chan_spec temp_channel[] = { + { + .type = IIO_TEMP, + .info_mask_separate = BIT(IIO_CHAN_INFO_PROCESSED), + }, + }; + +Channel sysfs attributes exposed to userspace are specified in the form of +bitmasks. Depending on their shared info, attributes can be set in one of the +following masks: + +* **info_mask_separate**, attributes will be specific to + this channel +* **info_mask_shared_by_type**, attributes are shared by all channels of the + same type +* **info_mask_shared_by_dir**, attributes are shared by all channels of the same + direction +* **info_mask_shared_by_all**, attributes are shared by all channels + +When there are multiple data channels per channel type we have two ways to +distinguish between them: + +* set **.modified** field of :c:type:`iio_chan_spec` to 1. Modifiers are + specified using **.channel2** field of the same :c:type:`iio_chan_spec` + structure and are used to indicate a physically unique characteristic of the + channel such as its direction or spectral response. For example, a light + sensor can have two channels, one for infrared light and one for both + infrared and visible light. +* set **.indexed** field of :c:type:`iio_chan_spec` to 1. In this case the + channel is simply another instance with an index specified by the **.channel** + field. + +Here is how we can make use of the channel's modifiers:: + + static const struct iio_chan_spec light_channels[] = { + { + .type = IIO_INTENSITY, + .modified = 1, + .channel2 = IIO_MOD_LIGHT_IR, + .info_mask_separate = BIT(IIO_CHAN_INFO_RAW), + .info_mask_shared = BIT(IIO_CHAN_INFO_SAMP_FREQ), + }, + { + .type = IIO_INTENSITY, + .modified = 1, + .channel2 = IIO_MOD_LIGHT_BOTH, + .info_mask_separate = BIT(IIO_CHAN_INFO_RAW), + .info_mask_shared = BIT(IIO_CHAN_INFO_SAMP_FREQ), + }, + { + .type = IIO_LIGHT, + .info_mask_separate = BIT(IIO_CHAN_INFO_PROCESSED), + .info_mask_shared = BIT(IIO_CHAN_INFO_SAMP_FREQ), + }, + } + +This channel's definition will generate two separate sysfs files for raw data +retrieval: + +* :file:`/sys/bus/iio/iio:device{X}/in_intensity_ir_raw` +* :file:`/sys/bus/iio/iio:device{X}/in_intensity_both_raw` + +one file for processed data: + +* :file:`/sys/bus/iio/iio:device{X}/in_illuminance_input` + +and one shared sysfs file for sampling frequency: + +* :file:`/sys/bus/iio/iio:device{X}/sampling_frequency`. + +Here is how we can make use of the channel's indexing:: + + static const struct iio_chan_spec light_channels[] = { + { + .type = IIO_VOLTAGE, + .indexed = 1, + .channel = 0, + .info_mask_separate = BIT(IIO_CHAN_INFO_RAW), + }, + { + .type = IIO_VOLTAGE, + .indexed = 1, + .channel = 1, + .info_mask_separate = BIT(IIO_CHAN_INFO_RAW), + }, + } + +This will generate two separate attributes files for raw data retrieval: + +* :file:`/sys/bus/iio/devices/iio:device{X}/in_voltage0_raw`, representing + voltage measurement for channel 0. +* :file:`/sys/bus/iio/devices/iio:device{X}/in_voltage1_raw`, representing + voltage measurement for channel 1. + +More details +============ +.. kernel-doc:: include/linux/iio/iio.h +.. kernel-doc:: drivers/iio/industrialio-core.c + :export: diff --git a/Documentation/driver-api/iio/index.rst b/Documentation/driver-api/iio/index.rst new file mode 100644 index 000000000000..e5c3922d1b6f --- /dev/null +++ b/Documentation/driver-api/iio/index.rst @@ -0,0 +1,17 @@ +.. include:: <isonum.txt> + +Industrial I/O +============== + +**Copyright** |copy| 2015 Intel Corporation + +Contents: + +.. toctree:: + :maxdepth: 2 + + intro + core + buffers + triggers + triggered-buffers diff --git a/Documentation/driver-api/iio/intro.rst b/Documentation/driver-api/iio/intro.rst new file mode 100644 index 000000000000..3653fbd57069 --- /dev/null +++ b/Documentation/driver-api/iio/intro.rst @@ -0,0 +1,33 @@ +.. include:: <isonum.txt> + +============ +Introduction +============ + +The main purpose of the Industrial I/O subsystem (IIO) is to provide support +for devices that in some sense perform either +analog-to-digital conversion (ADC) or digital-to-analog conversion (DAC) +or both. The aim is to fill the gap between the somewhat similar hwmon and +:doc:`input <../input>` subsystems. Hwmon is directed at low sample rate +sensors used to monitor and control the system itself, like fan speed control +or temperature measurement. :doc:`Input <../input>` is, as its name suggests, +focused on human interaction input devices (keyboard, mouse, touchscreen). +In some cases there is considerable overlap between these and IIO. + +Devices that fall into this category include: + +* analog to digital converters (ADCs) +* accelerometers +* capacitance to digital converters (CDCs) +* digital to analog converters (DACs) +* gyroscopes +* inertial measurement units (IMUs) +* color and light sensors +* magnetometers +* pressure sensors +* proximity sensors +* temperature sensors + +Usually these sensors are connected via :doc:`SPI <../spi>` or +:doc:`I2C <../i2c>`. A common use case of the sensors devices is to have +combined functionality (e.g. light plus proximity sensor). diff --git a/Documentation/driver-api/iio/triggered-buffers.rst b/Documentation/driver-api/iio/triggered-buffers.rst new file mode 100644 index 000000000000..0db12660cc90 --- /dev/null +++ b/Documentation/driver-api/iio/triggered-buffers.rst @@ -0,0 +1,69 @@ +================= +Triggered Buffers +================= + +Now that we know what buffers and triggers are let's see how they work together. + +IIO triggered buffer setup +========================== + +* :c:func:`iio_triggered_buffer_setup` — Setup triggered buffer and pollfunc +* :c:func:`iio_triggered_buffer_cleanup` — Free resources allocated by + :c:func:`iio_triggered_buffer_setup` +* struct :c:type:`iio_buffer_setup_ops` — buffer setup related callbacks + +A typical triggered buffer setup looks like this:: + + const struct iio_buffer_setup_ops sensor_buffer_setup_ops = { + .preenable = sensor_buffer_preenable, + .postenable = sensor_buffer_postenable, + .postdisable = sensor_buffer_postdisable, + .predisable = sensor_buffer_predisable, + }; + + irqreturn_t sensor_iio_pollfunc(int irq, void *p) + { + pf->timestamp = iio_get_time_ns((struct indio_dev *)p); + return IRQ_WAKE_THREAD; + } + + irqreturn_t sensor_trigger_handler(int irq, void *p) + { + u16 buf[8]; + int i = 0; + + /* read data for each active channel */ + for_each_set_bit(bit, active_scan_mask, masklength) + buf[i++] = sensor_get_data(bit) + + iio_push_to_buffers_with_timestamp(indio_dev, buf, timestamp); + + iio_trigger_notify_done(trigger); + return IRQ_HANDLED; + } + + /* setup triggered buffer, usually in probe function */ + iio_triggered_buffer_setup(indio_dev, sensor_iio_polfunc, + sensor_trigger_handler, + sensor_buffer_setup_ops); + +The important things to notice here are: + +* :c:type:`iio_buffer_setup_ops`, the buffer setup functions to be called at + predefined points in the buffer configuration sequence (e.g. before enable, + after disable). If not specified, the IIO core uses the default + iio_triggered_buffer_setup_ops. +* **sensor_iio_pollfunc**, the function that will be used as top half of poll + function. It should do as little processing as possible, because it runs in + interrupt context. The most common operation is recording of the current + timestamp and for this reason one can use the IIO core defined + :c:func:`iio_pollfunc_store_time` function. +* **sensor_trigger_handler**, the function that will be used as bottom half of + the poll function. This runs in the context of a kernel thread and all the + processing takes place here. It usually reads data from the device and + stores it in the internal buffer together with the timestamp recorded in the + top half. + +More details +============ +.. kernel-doc:: drivers/iio/buffer/industrialio-triggered-buffer.c diff --git a/Documentation/driver-api/iio/triggers.rst b/Documentation/driver-api/iio/triggers.rst new file mode 100644 index 000000000000..f89d37e7dd82 --- /dev/null +++ b/Documentation/driver-api/iio/triggers.rst @@ -0,0 +1,80 @@ +======== +Triggers +======== + +* struct :c:type:`iio_trigger` — industrial I/O trigger device +* :c:func:`devm_iio_trigger_alloc` — Resource-managed iio_trigger_alloc +* :c:func:`devm_iio_trigger_free` — Resource-managed iio_trigger_free +* :c:func:`devm_iio_trigger_register` — Resource-managed iio_trigger_register +* :c:func:`devm_iio_trigger_unregister` — Resource-managed + iio_trigger_unregister +* :c:func:`iio_trigger_validate_own_device` — Check if a trigger and IIO + device belong to the same device + +In many situations it is useful for a driver to be able to capture data based +on some external event (trigger) as opposed to periodically polling for data. +An IIO trigger can be provided by a device driver that also has an IIO device +based on hardware generated events (e.g. data ready or threshold exceeded) or +provided by a separate driver from an independent interrupt source (e.g. GPIO +line connected to some external system, timer interrupt or user space writing +a specific file in sysfs). A trigger may initiate data capture for a number of +sensors and also it may be completely unrelated to the sensor itself. + +IIO trigger sysfs interface +=========================== + +There are two locations in sysfs related to triggers: + +* :file:`/sys/bus/iio/devices/trigger{Y}/*`, this file is created once an + IIO trigger is registered with the IIO core and corresponds to trigger + with index Y. + Because triggers can be very different depending on type there are few + standard attributes that we can describe here: + + * :file:`name`, trigger name that can be later used for association with a + device. + * :file:`sampling_frequency`, some timer based triggers use this attribute to + specify the frequency for trigger calls. + +* :file:`/sys/bus/iio/devices/iio:device{X}/trigger/*`, this directory is + created once the device supports a triggered buffer. We can associate a + trigger with our device by writing the trigger's name in the + :file:`current_trigger` file. + +IIO trigger setup +================= + +Let's see a simple example of how to setup a trigger to be used by a driver:: + + struct iio_trigger_ops trigger_ops = { + .set_trigger_state = sample_trigger_state, + .validate_device = sample_validate_device, + } + + struct iio_trigger *trig; + + /* first, allocate memory for our trigger */ + trig = iio_trigger_alloc(dev, "trig-%s-%d", name, idx); + + /* setup trigger operations field */ + trig->ops = &trigger_ops; + + /* now register the trigger with the IIO core */ + iio_trigger_register(trig); + +IIO trigger ops +=============== + +* struct :c:type:`iio_trigger_ops` — operations structure for an iio_trigger. + +Notice that a trigger has a set of operations attached: + +* :file:`set_trigger_state`, switch the trigger on/off on demand. +* :file:`validate_device`, function to validate the device when the current + trigger gets changed. + +More details +============ +.. kernel-doc:: include/linux/iio/trigger.h +.. kernel-doc:: drivers/iio/industrialio-trigger.c + :export: diff --git a/Documentation/driver-api/index.rst b/Documentation/driver-api/index.rst index dbd34c9c1d93..60db00d1532b 100644 --- a/Documentation/driver-api/index.rst +++ b/Documentation/driver-api/index.rst @@ -16,11 +16,15 @@ available subsections can be seen below. basics infrastructure + pm/index + device-io dma-buf device_link message-based sound frame-buffer + regulator + iio/index input usb spi diff --git a/Documentation/driver-api/pm/conf.py b/Documentation/driver-api/pm/conf.py new file mode 100644 index 000000000000..a89fac11272f --- /dev/null +++ b/Documentation/driver-api/pm/conf.py @@ -0,0 +1,10 @@ +# -*- coding: utf-8; mode: python -*- + +project = "Device Power Management" + +tags.add("subproject") + +latex_documents = [ + ('index', 'pm.tex', project, + 'The kernel development community', 'manual'), +] diff --git a/Documentation/driver-api/pm/devices.rst b/Documentation/driver-api/pm/devices.rst new file mode 100644 index 000000000000..bedd32388dac --- /dev/null +++ b/Documentation/driver-api/pm/devices.rst @@ -0,0 +1,736 @@ +.. |struct dev_pm_ops| replace:: :c:type:`struct dev_pm_ops <dev_pm_ops>` +.. |struct dev_pm_domain| replace:: :c:type:`struct dev_pm_domain <dev_pm_domain>` +.. |struct bus_type| replace:: :c:type:`struct bus_type <bus_type>` +.. |struct device_type| replace:: :c:type:`struct device_type <device_type>` +.. |struct class| replace:: :c:type:`struct class <class>` +.. |struct wakeup_source| replace:: :c:type:`struct wakeup_source <wakeup_source>` +.. |struct device| replace:: :c:type:`struct device <device>` + +============================== +Device Power Management Basics +============================== + +:: + + Copyright (c) 2010-2011 Rafael J. Wysocki <rjw@sisk.pl>, Novell Inc. + Copyright (c) 2010 Alan Stern <stern@rowland.harvard.edu> + Copyright (c) 2016 Intel Corp., Rafael J. Wysocki <rafael.j.wysocki@intel.com> + +Most of the code in Linux is device drivers, so most of the Linux power +management (PM) code is also driver-specific. Most drivers will do very +little; others, especially for platforms with small batteries (like cell +phones), will do a lot. + +This writeup gives an overview of how drivers interact with system-wide +power management goals, emphasizing the models and interfaces that are +shared by everything that hooks up to the driver model core. Read it as +background for the domain-specific work you'd do with any specific driver. + + +Two Models for Device Power Management +====================================== + +Drivers will use one or both of these models to put devices into low-power +states: + + System Sleep model: + + Drivers can enter low-power states as part of entering system-wide + low-power states like "suspend" (also known as "suspend-to-RAM"), or + (mostly for systems with disks) "hibernation" (also known as + "suspend-to-disk"). + + This is something that device, bus, and class drivers collaborate on + by implementing various role-specific suspend and resume methods to + cleanly power down hardware and software subsystems, then reactivate + them without loss of data. + + Some drivers can manage hardware wakeup events, which make the system + leave the low-power state. This feature may be enabled or disabled + using the relevant :file:`/sys/devices/.../power/wakeup` file (for + Ethernet drivers the ioctl interface used by ethtool may also be used + for this purpose); enabling it may cost some power usage, but let the + whole system enter low-power states more often. + + Runtime Power Management model: + + Devices may also be put into low-power states while the system is + running, independently of other power management activity in principle. + However, devices are not generally independent of each other (for + example, a parent device cannot be suspended unless all of its child + devices have been suspended). Moreover, depending on the bus type the + device is on, it may be necessary to carry out some bus-specific + operations on the device for this purpose. Devices put into low power + states at run time may require special handling during system-wide power + transitions (suspend or hibernation). + + For these reasons not only the device driver itself, but also the + appropriate subsystem (bus type, device type or device class) driver and + the PM core are involved in runtime power management. As in the system + sleep power management case, they need to collaborate by implementing + various role-specific suspend and resume methods, so that the hardware + is cleanly powered down and reactivated without data or service loss. + +There's not a lot to be said about those low-power states except that they are +very system-specific, and often device-specific. Also, that if enough devices +have been put into low-power states (at runtime), the effect may be very similar +to entering some system-wide low-power state (system sleep) ... and that +synergies exist, so that several drivers using runtime PM might put the system +into a state where even deeper power saving options are available. + +Most suspended devices will have quiesced all I/O: no more DMA or IRQs (except +for wakeup events), no more data read or written, and requests from upstream +drivers are no longer accepted. A given bus or platform may have different +requirements though. + +Examples of hardware wakeup events include an alarm from a real time clock, +network wake-on-LAN packets, keyboard or mouse activity, and media insertion +or removal (for PCMCIA, MMC/SD, USB, and so on). + +Interfaces for Entering System Sleep States +=========================================== + +There are programming interfaces provided for subsystems (bus type, device type, +device class) and device drivers to allow them to participate in the power +management of devices they are concerned with. These interfaces cover both +system sleep and runtime power management. + + +Device Power Management Operations +---------------------------------- + +Device power management operations, at the subsystem level as well as at the +device driver level, are implemented by defining and populating objects of type +|struct dev_pm_ops| defined in :file:`include/linux/pm.h`. The roles of the +methods included in it will be explained in what follows. For now, it should be +sufficient to remember that the last three methods are specific to runtime power +management while the remaining ones are used during system-wide power +transitions. + +There also is a deprecated "old" or "legacy" interface for power management +operations available at least for some subsystems. This approach does not use +|struct dev_pm_ops| objects and it is suitable only for implementing system +sleep power management methods in a limited way. Therefore it is not described +in this document, so please refer directly to the source code for more +information about it. + + +Subsystem-Level Methods +----------------------- + +The core methods to suspend and resume devices reside in +|struct dev_pm_ops| pointed to by the :c:member:`ops` member of +|struct dev_pm_domain|, or by the :c:member:`pm` member of |struct bus_type|, +|struct device_type| and |struct class|. They are mostly of interest to the +people writing infrastructure for platforms and buses, like PCI or USB, or +device type and device class drivers. They also are relevant to the writers of +device drivers whose subsystems (PM domains, device types, device classes and +bus types) don't provide all power management methods. + +Bus drivers implement these methods as appropriate for the hardware and the +drivers using it; PCI works differently from USB, and so on. Not many people +write subsystem-level drivers; most driver code is a "device driver" that builds +on top of bus-specific framework code. + +For more information on these driver calls, see the description later; +they are called in phases for every device, respecting the parent-child +sequencing in the driver model tree. + + +:file:`/sys/devices/.../power/wakeup` files +------------------------------------------- + +All device objects in the driver model contain fields that control the handling +of system wakeup events (hardware signals that can force the system out of a +sleep state). These fields are initialized by bus or device driver code using +:c:func:`device_set_wakeup_capable()` and :c:func:`device_set_wakeup_enable()`, +defined in :file:`include/linux/pm_wakeup.h`. + +The :c:member:`power.can_wakeup` flag just records whether the device (and its +driver) can physically support wakeup events. The +:c:func:`device_set_wakeup_capable()` routine affects this flag. The +:c:member:`power.wakeup` field is a pointer to an object of type +|struct wakeup_source| used for controlling whether or not the device should use +its system wakeup mechanism and for notifying the PM core of system wakeup +events signaled by the device. This object is only present for wakeup-capable +devices (i.e. devices whose :c:member:`can_wakeup` flags are set) and is created +(or removed) by :c:func:`device_set_wakeup_capable()`. + +Whether or not a device is capable of issuing wakeup events is a hardware +matter, and the kernel is responsible for keeping track of it. By contrast, +whether or not a wakeup-capable device should issue wakeup events is a policy +decision, and it is managed by user space through a sysfs attribute: the +:file:`power/wakeup` file. User space can write the "enabled" or "disabled" +strings to it to indicate whether or not, respectively, the device is supposed +to signal system wakeup. This file is only present if the +:c:member:`power.wakeup` object exists for the given device and is created (or +removed) along with that object, by :c:func:`device_set_wakeup_capable()`. +Reads from the file will return the corresponding string. + +The initial value in the :file:`power/wakeup` file is "disabled" for the +majority of devices; the major exceptions are power buttons, keyboards, and +Ethernet adapters whose WoL (wake-on-LAN) feature has been set up with ethtool. +It should also default to "enabled" for devices that don't generate wakeup +requests on their own but merely forward wakeup requests from one bus to another +(like PCI Express ports). + +The :c:func:`device_may_wakeup()` routine returns true only if the +:c:member:`power.wakeup` object exists and the corresponding :file:`power/wakeup` +file contains the "enabled" string. This information is used by subsystems, +like the PCI bus type code, to see whether or not to enable the devices' wakeup +mechanisms. If device wakeup mechanisms are enabled or disabled directly by +drivers, they also should use :c:func:`device_may_wakeup()` to decide what to do +during a system sleep transition. Device drivers, however, are not expected to +call :c:func:`device_set_wakeup_enable()` directly in any case. + +It ought to be noted that system wakeup is conceptually different from "remote +wakeup" used by runtime power management, although it may be supported by the +same physical mechanism. Remote wakeup is a feature allowing devices in +low-power states to trigger specific interrupts to signal conditions in which +they should be put into the full-power state. Those interrupts may or may not +be used to signal system wakeup events, depending on the hardware design. On +some systems it is impossible to trigger them from system sleep states. In any +case, remote wakeup should always be enabled for runtime power management for +all devices and drivers that support it. + + +:file:`/sys/devices/.../power/control` files +-------------------------------------------- + +Each device in the driver model has a flag to control whether it is subject to +runtime power management. This flag, :c:member:`runtime_auto`, is initialized +by the bus type (or generally subsystem) code using :c:func:`pm_runtime_allow()` +or :c:func:`pm_runtime_forbid()`; the default is to allow runtime power +management. + +The setting can be adjusted by user space by writing either "on" or "auto" to +the device's :file:`power/control` sysfs file. Writing "auto" calls +:c:func:`pm_runtime_allow()`, setting the flag and allowing the device to be +runtime power-managed by its driver. Writing "on" calls +:c:func:`pm_runtime_forbid()`, clearing the flag, returning the device to full +power if it was in a low-power state, and preventing the +device from being runtime power-managed. User space can check the current value +of the :c:member:`runtime_auto` flag by reading that file. + +The device's :c:member:`runtime_auto` flag has no effect on the handling of +system-wide power transitions. In particular, the device can (and in the +majority of cases should and will) be put into a low-power state during a +system-wide transition to a sleep state even though its :c:member:`runtime_auto` +flag is clear. + +For more information about the runtime power management framework, refer to +:file:`Documentation/power/runtime_pm.txt`. + + +Calling Drivers to Enter and Leave System Sleep States +====================================================== + +When the system goes into a sleep state, each device's driver is asked to +suspend the device by putting it into a state compatible with the target +system state. That's usually some version of "off", but the details are +system-specific. Also, wakeup-enabled devices will usually stay partly +functional in order to wake the system. + +When the system leaves that low-power state, the device's driver is asked to +resume it by returning it to full power. The suspend and resume operations +always go together, and both are multi-phase operations. + +For simple drivers, suspend might quiesce the device using class code +and then turn its hardware as "off" as possible during suspend_noirq. The +matching resume calls would then completely reinitialize the hardware +before reactivating its class I/O queues. + +More power-aware drivers might prepare the devices for triggering system wakeup +events. + + +Call Sequence Guarantees +------------------------ + +To ensure that bridges and similar links needing to talk to a device are +available when the device is suspended or resumed, the device hierarchy is +walked in a bottom-up order to suspend devices. A top-down order is +used to resume those devices. + +The ordering of the device hierarchy is defined by the order in which devices +get registered: a child can never be registered, probed or resumed before +its parent; and can't be removed or suspended after that parent. + +The policy is that the device hierarchy should match hardware bus topology. +[Or at least the control bus, for devices which use multiple busses.] +In particular, this means that a device registration may fail if the parent of +the device is suspending (i.e. has been chosen by the PM core as the next +device to suspend) or has already suspended, as well as after all of the other +devices have been suspended. Device drivers must be prepared to cope with such +situations. + + +System Power Management Phases +------------------------------ + +Suspending or resuming the system is done in several phases. Different phases +are used for suspend-to-idle, shallow (standby), and deep ("suspend-to-RAM") +sleep states and the hibernation state ("suspend-to-disk"). Each phase involves +executing callbacks for every device before the next phase begins. Not all +buses or classes support all these callbacks and not all drivers use all the +callbacks. The various phases always run after tasks have been frozen and +before they are unfrozen. Furthermore, the ``*_noirq phases`` run at a time +when IRQ handlers have been disabled (except for those marked with the +IRQF_NO_SUSPEND flag). + +All phases use PM domain, bus, type, class or driver callbacks (that is, methods +defined in ``dev->pm_domain->ops``, ``dev->bus->pm``, ``dev->type->pm``, +``dev->class->pm`` or ``dev->driver->pm``). These callbacks are regarded by the +PM core as mutually exclusive. Moreover, PM domain callbacks always take +precedence over all of the other callbacks and, for example, type callbacks take +precedence over bus, class and driver callbacks. To be precise, the following +rules are used to determine which callback to execute in the given phase: + + 1. If ``dev->pm_domain`` is present, the PM core will choose the callback + provided by ``dev->pm_domain->ops`` for execution. + + 2. Otherwise, if both ``dev->type`` and ``dev->type->pm`` are present, the + callback provided by ``dev->type->pm`` will be chosen for execution. + + 3. Otherwise, if both ``dev->class`` and ``dev->class->pm`` are present, + the callback provided by ``dev->class->pm`` will be chosen for + execution. + + 4. Otherwise, if both ``dev->bus`` and ``dev->bus->pm`` are present, the + callback provided by ``dev->bus->pm`` will be chosen for execution. + +This allows PM domains and device types to override callbacks provided by bus +types or device classes if necessary. + +The PM domain, type, class and bus callbacks may in turn invoke device- or +driver-specific methods stored in ``dev->driver->pm``, but they don't have to do +that. + +If the subsystem callback chosen for execution is not present, the PM core will +execute the corresponding method from the ``dev->driver->pm`` set instead if +there is one. + + +Entering System Suspend +----------------------- + +When the system goes into the freeze, standby or memory sleep state, +the phases are: ``prepare``, ``suspend``, ``suspend_late``, ``suspend_noirq``. + + 1. The ``prepare`` phase is meant to prevent races by preventing new + devices from being registered; the PM core would never know that all the + children of a device had been suspended if new children could be + registered at will. [By contrast, from the PM core's perspective, + devices may be unregistered at any time.] Unlike the other + suspend-related phases, during the ``prepare`` phase the device + hierarchy is traversed top-down. + + After the ``->prepare`` callback method returns, no new children may be + registered below the device. The method may also prepare the device or + driver in some way for the upcoming system power transition, but it + should not put the device into a low-power state. + + For devices supporting runtime power management, the return value of the + prepare callback can be used to indicate to the PM core that it may + safely leave the device in runtime suspend (if runtime-suspended + already), provided that all of the device's descendants are also left in + runtime suspend. Namely, if the prepare callback returns a positive + number and that happens for all of the descendants of the device too, + and all of them (including the device itself) are runtime-suspended, the + PM core will skip the ``suspend``, ``suspend_late`` and + ``suspend_noirq`` phases as well as all of the corresponding phases of + the subsequent device resume for all of these devices. In that case, + the ``->complete`` callback will be invoked directly after the + ``->prepare`` callback and is entirely responsible for putting the + device into a consistent state as appropriate. + + Note that this direct-complete procedure applies even if the device is + disabled for runtime PM; only the runtime-PM status matters. It follows + that if a device has system-sleep callbacks but does not support runtime + PM, then its prepare callback must never return a positive value. This + is because all such devices are initially set to runtime-suspended with + runtime PM disabled. + + 2. The ``->suspend`` methods should quiesce the device to stop it from + performing I/O. They also may save the device registers and put it into + the appropriate low-power state, depending on the bus type the device is + on, and they may enable wakeup events. + + 3. For a number of devices it is convenient to split suspend into the + "quiesce device" and "save device state" phases, in which cases + ``suspend_late`` is meant to do the latter. It is always executed after + runtime power management has been disabled for the device in question. + + 4. The ``suspend_noirq`` phase occurs after IRQ handlers have been disabled, + which means that the driver's interrupt handler will not be called while + the callback method is running. The ``->suspend_noirq`` methods should + save the values of the device's registers that weren't saved previously + and finally put the device into the appropriate low-power state. + + The majority of subsystems and device drivers need not implement this + callback. However, bus types allowing devices to share interrupt + vectors, like PCI, generally need it; otherwise a driver might encounter + an error during the suspend phase by fielding a shared interrupt + generated by some other device after its own device had been set to low + power. + +At the end of these phases, drivers should have stopped all I/O transactions +(DMA, IRQs), saved enough state that they can re-initialize or restore previous +state (as needed by the hardware), and placed the device into a low-power state. +On many platforms they will gate off one or more clock sources; sometimes they +will also switch off power supplies or reduce voltages. [Drivers supporting +runtime PM may already have performed some or all of these steps.] + +If :c:func:`device_may_wakeup(dev)` returns ``true``, the device should be +prepared for generating hardware wakeup signals to trigger a system wakeup event +when the system is in the sleep state. For example, :c:func:`enable_irq_wake()` +might identify GPIO signals hooked up to a switch or other external hardware, +and :c:func:`pci_enable_wake()` does something similar for the PCI PME signal. + +If any of these callbacks returns an error, the system won't enter the desired +low-power state. Instead, the PM core will unwind its actions by resuming all +the devices that were suspended. + + +Leaving System Suspend +---------------------- + +When resuming from freeze, standby or memory sleep, the phases are: +``resume_noirq``, ``resume_early``, ``resume``, ``complete``. + + 1. The ``->resume_noirq`` callback methods should perform any actions + needed before the driver's interrupt handlers are invoked. This + generally means undoing the actions of the ``suspend_noirq`` phase. If + the bus type permits devices to share interrupt vectors, like PCI, the + method should bring the device and its driver into a state in which the + driver can recognize if the device is the source of incoming interrupts, + if any, and handle them correctly. + + For example, the PCI bus type's ``->pm.resume_noirq()`` puts the device + into the full-power state (D0 in the PCI terminology) and restores the + standard configuration registers of the device. Then it calls the + device driver's ``->pm.resume_noirq()`` method to perform device-specific + actions. + + 2. The ``->resume_early`` methods should prepare devices for the execution + of the resume methods. This generally involves undoing the actions of + the preceding ``suspend_late`` phase. + + 3. The ``->resume`` methods should bring the device back to its operating + state, so that it can perform normal I/O. This generally involves + undoing the actions of the ``suspend`` phase. + + 4. The ``complete`` phase should undo the actions of the ``prepare`` phase. + For this reason, unlike the other resume-related phases, during the + ``complete`` phase the device hierarchy is traversed bottom-up. + + Note, however, that new children may be registered below the device as + soon as the ``->resume`` callbacks occur; it's not necessary to wait + until the ``complete`` phase with that. + + Moreover, if the preceding ``->prepare`` callback returned a positive + number, the device may have been left in runtime suspend throughout the + whole system suspend and resume (the ``suspend``, ``suspend_late``, + ``suspend_noirq`` phases of system suspend and the ``resume_noirq``, + ``resume_early``, ``resume`` phases of system resume may have been + skipped for it). In that case, the ``->complete`` callback is entirely + responsible for putting the device into a consistent state after system + suspend if necessary. [For example, it may need to queue up a runtime + resume request for the device for this purpose.] To check if that is + the case, the ``->complete`` callback can consult the device's + ``power.direct_complete`` flag. Namely, if that flag is set when the + ``->complete`` callback is being run, it has been called directly after + the preceding ``->prepare`` and special actions may be required + to make the device work correctly afterward. + +At the end of these phases, drivers should be as functional as they were before +suspending: I/O can be performed using DMA and IRQs, and the relevant clocks are +gated on. + +However, the details here may again be platform-specific. For example, +some systems support multiple "run" states, and the mode in effect at +the end of resume might not be the one which preceded suspension. +That means availability of certain clocks or power supplies changed, +which could easily affect how a driver works. + +Drivers need to be able to handle hardware which has been reset since all of the +suspend methods were called, for example by complete reinitialization. +This may be the hardest part, and the one most protected by NDA'd documents +and chip errata. It's simplest if the hardware state hasn't changed since +the suspend was carried out, but that can only be guaranteed if the target +system sleep entered was suspend-to-idle. For the other system sleep states +that may not be the case (and usually isn't for ACPI-defined system sleep +states, like S3). + +Drivers must also be prepared to notice that the device has been removed +while the system was powered down, whenever that's physically possible. +PCMCIA, MMC, USB, Firewire, SCSI, and even IDE are common examples of busses +where common Linux platforms will see such removal. Details of how drivers +will notice and handle such removals are currently bus-specific, and often +involve a separate thread. + +These callbacks may return an error value, but the PM core will ignore such +errors since there's nothing it can do about them other than printing them in +the system log. + + +Entering Hibernation +-------------------- + +Hibernating the system is more complicated than putting it into sleep states, +because it involves creating and saving a system image. Therefore there are +more phases for hibernation, with a different set of callbacks. These phases +always run after tasks have been frozen and enough memory has been freed. + +The general procedure for hibernation is to quiesce all devices ("freeze"), +create an image of the system memory while everything is stable, reactivate all +devices ("thaw"), write the image to permanent storage, and finally shut down +the system ("power off"). The phases used to accomplish this are: ``prepare``, +``freeze``, ``freeze_late``, ``freeze_noirq``, ``thaw_noirq``, ``thaw_early``, +``thaw``, ``complete``, ``prepare``, ``poweroff``, ``poweroff_late``, +``poweroff_noirq``. + + 1. The ``prepare`` phase is discussed in the "Entering System Suspend" + section above. + + 2. The ``->freeze`` methods should quiesce the device so that it doesn't + generate IRQs or DMA, and they may need to save the values of device + registers. However the device does not have to be put in a low-power + state, and to save time it's best not to do so. Also, the device should + not be prepared to generate wakeup events. + + 3. The ``freeze_late`` phase is analogous to the ``suspend_late`` phase + described earlier, except that the device should not be put into a + low-power state and should not be allowed to generate wakeup events. + + 4. The ``freeze_noirq`` phase is analogous to the ``suspend_noirq`` phase + discussed earlier, except again that the device should not be put into + a low-power state and should not be allowed to generate wakeup events. + +At this point the system image is created. All devices should be inactive and +the contents of memory should remain undisturbed while this happens, so that the +image forms an atomic snapshot of the system state. + + 5. The ``thaw_noirq`` phase is analogous to the ``resume_noirq`` phase + discussed earlier. The main difference is that its methods can assume + the device is in the same state as at the end of the ``freeze_noirq`` + phase. + + 6. The ``thaw_early`` phase is analogous to the ``resume_early`` phase + described above. Its methods should undo the actions of the preceding + ``freeze_late``, if necessary. + + 7. The ``thaw`` phase is analogous to the ``resume`` phase discussed + earlier. Its methods should bring the device back to an operating + state, so that it can be used for saving the image if necessary. + + 8. The ``complete`` phase is discussed in the "Leaving System Suspend" + section above. + +At this point the system image is saved, and the devices then need to be +prepared for the upcoming system shutdown. This is much like suspending them +before putting the system into the suspend-to-idle, shallow or deep sleep state, +and the phases are similar. + + 9. The ``prepare`` phase is discussed above. + + 10. The ``poweroff`` phase is analogous to the ``suspend`` phase. + + 11. The ``poweroff_late`` phase is analogous to the ``suspend_late`` phase. + + 12. The ``poweroff_noirq`` phase is analogous to the ``suspend_noirq`` phase. + +The ``->poweroff``, ``->poweroff_late`` and ``->poweroff_noirq`` callbacks +should do essentially the same things as the ``->suspend``, ``->suspend_late`` +and ``->suspend_noirq`` callbacks, respectively. The only notable difference is +that they need not store the device register values, because the registers +should already have been stored during the ``freeze``, ``freeze_late`` or +``freeze_noirq`` phases. + + +Leaving Hibernation +------------------- + +Resuming from hibernation is, again, more complicated than resuming from a sleep +state in which the contents of main memory are preserved, because it requires +a system image to be loaded into memory and the pre-hibernation memory contents +to be restored before control can be passed back to the image kernel. + +Although in principle the image might be loaded into memory and the +pre-hibernation memory contents restored by the boot loader, in practice this +can't be done because boot loaders aren't smart enough and there is no +established protocol for passing the necessary information. So instead, the +boot loader loads a fresh instance of the kernel, called "the restore kernel", +into memory and passes control to it in the usual way. Then the restore kernel +reads the system image, restores the pre-hibernation memory contents, and passes +control to the image kernel. Thus two different kernel instances are involved +in resuming from hibernation. In fact, the restore kernel may be completely +different from the image kernel: a different configuration and even a different +version. This has important consequences for device drivers and their +subsystems. + +To be able to load the system image into memory, the restore kernel needs to +include at least a subset of device drivers allowing it to access the storage +medium containing the image, although it doesn't need to include all of the +drivers present in the image kernel. After the image has been loaded, the +devices managed by the boot kernel need to be prepared for passing control back +to the image kernel. This is very similar to the initial steps involved in +creating a system image, and it is accomplished in the same way, using +``prepare``, ``freeze``, and ``freeze_noirq`` phases. However, the devices +affected by these phases are only those having drivers in the restore kernel; +other devices will still be in whatever state the boot loader left them. + +Should the restoration of the pre-hibernation memory contents fail, the restore +kernel would go through the "thawing" procedure described above, using the +``thaw_noirq``, ``thaw_early``, ``thaw``, and ``complete`` phases, and then +continue running normally. This happens only rarely. Most often the +pre-hibernation memory contents are restored successfully and control is passed +to the image kernel, which then becomes responsible for bringing the system back +to the working state. + +To achieve this, the image kernel must restore the devices' pre-hibernation +functionality. The operation is much like waking up from a sleep state (with +the memory contents preserved), although it involves different phases: +``restore_noirq``, ``restore_early``, ``restore``, ``complete``. + + 1. The ``restore_noirq`` phase is analogous to the ``resume_noirq`` phase. + + 2. The ``restore_early`` phase is analogous to the ``resume_early`` phase. + + 3. The ``restore`` phase is analogous to the ``resume`` phase. + + 4. The ``complete`` phase is discussed above. + +The main difference from ``resume[_early|_noirq]`` is that +``restore[_early|_noirq]`` must assume the device has been accessed and +reconfigured by the boot loader or the restore kernel. Consequently, the state +of the device may be different from the state remembered from the ``freeze``, +``freeze_late`` and ``freeze_noirq`` phases. The device may even need to be +reset and completely re-initialized. In many cases this difference doesn't +matter, so the ``->resume[_early|_noirq]`` and ``->restore[_early|_norq]`` +method pointers can be set to the same routines. Nevertheless, different +callback pointers are used in case there is a situation where it actually does +matter. + + +Power Management Notifiers +========================== + +There are some operations that cannot be carried out by the power management +callbacks discussed above, because the callbacks occur too late or too early. +To handle these cases, subsystems and device drivers may register power +management notifiers that are called before tasks are frozen and after they have +been thawed. Generally speaking, the PM notifiers are suitable for performing +actions that either require user space to be available, or at least won't +interfere with user space. + +For details refer to :doc:`notifiers`. + + +Device Low-Power (suspend) States +================================= + +Device low-power states aren't standard. One device might only handle +"on" and "off", while another might support a dozen different versions of +"on" (how many engines are active?), plus a state that gets back to "on" +faster than from a full "off". + +Some buses define rules about what different suspend states mean. PCI +gives one example: after the suspend sequence completes, a non-legacy +PCI device may not perform DMA or issue IRQs, and any wakeup events it +issues would be issued through the PME# bus signal. Plus, there are +several PCI-standard device states, some of which are optional. + +In contrast, integrated system-on-chip processors often use IRQs as the +wakeup event sources (so drivers would call :c:func:`enable_irq_wake`) and +might be able to treat DMA completion as a wakeup event (sometimes DMA can stay +active too, it'd only be the CPU and some peripherals that sleep). + +Some details here may be platform-specific. Systems may have devices that +can be fully active in certain sleep states, such as an LCD display that's +refreshed using DMA while most of the system is sleeping lightly ... and +its frame buffer might even be updated by a DSP or other non-Linux CPU while +the Linux control processor stays idle. + +Moreover, the specific actions taken may depend on the target system state. +One target system state might allow a given device to be very operational; +another might require a hard shut down with re-initialization on resume. +And two different target systems might use the same device in different +ways; the aforementioned LCD might be active in one product's "standby", +but a different product using the same SOC might work differently. + + +Device Power Management Domains +=============================== + +Sometimes devices share reference clocks or other power resources. In those +cases it generally is not possible to put devices into low-power states +individually. Instead, a set of devices sharing a power resource can be put +into a low-power state together at the same time by turning off the shared +power resource. Of course, they also need to be put into the full-power state +together, by turning the shared power resource on. A set of devices with this +property is often referred to as a power domain. A power domain may also be +nested inside another power domain. The nested domain is referred to as the +sub-domain of the parent domain. + +Support for power domains is provided through the :c:member:`pm_domain` field of +|struct device|. This field is a pointer to an object of type +|struct dev_pm_domain|, defined in :file:`include/linux/pm.h``, providing a set +of power management callbacks analogous to the subsystem-level and device driver +callbacks that are executed for the given device during all power transitions, +instead of the respective subsystem-level callbacks. Specifically, if a +device's :c:member:`pm_domain` pointer is not NULL, the ``->suspend()`` callback +from the object pointed to by it will be executed instead of its subsystem's +(e.g. bus type's) ``->suspend()`` callback and analogously for all of the +remaining callbacks. In other words, power management domain callbacks, if +defined for the given device, always take precedence over the callbacks provided +by the device's subsystem (e.g. bus type). + +The support for device power management domains is only relevant to platforms +needing to use the same device driver power management callbacks in many +different power domain configurations and wanting to avoid incorporating the +support for power domains into subsystem-level callbacks, for example by +modifying the platform bus type. Other platforms need not implement it or take +it into account in any way. + +Devices may be defined as IRQ-safe which indicates to the PM core that their +runtime PM callbacks may be invoked with disabled interrupts (see +:file:`Documentation/power/runtime_pm.txt` for more information). If an +IRQ-safe device belongs to a PM domain, the runtime PM of the domain will be +disallowed, unless the domain itself is defined as IRQ-safe. However, it +makes sense to define a PM domain as IRQ-safe only if all the devices in it +are IRQ-safe. Moreover, if an IRQ-safe domain has a parent domain, the runtime +PM of the parent is only allowed if the parent itself is IRQ-safe too with the +additional restriction that all child domains of an IRQ-safe parent must also +be IRQ-safe. + + +Runtime Power Management +======================== + +Many devices are able to dynamically power down while the system is still +running. This feature is useful for devices that are not being used, and +can offer significant power savings on a running system. These devices +often support a range of runtime power states, which might use names such +as "off", "sleep", "idle", "active", and so on. Those states will in some +cases (like PCI) be partially constrained by the bus the device uses, and will +usually include hardware states that are also used in system sleep states. + +A system-wide power transition can be started while some devices are in low +power states due to runtime power management. The system sleep PM callbacks +should recognize such situations and react to them appropriately, but the +necessary actions are subsystem-specific. + +In some cases the decision may be made at the subsystem level while in other +cases the device driver may be left to decide. In some cases it may be +desirable to leave a suspended device in that state during a system-wide power +transition, but in other cases the device must be put back into the full-power +state temporarily, for example so that its system wakeup capability can be +disabled. This all depends on the hardware and the design of the subsystem and +device driver in question. + +During system-wide resume from a sleep state it's easiest to put devices into +the full-power state, as explained in :file:`Documentation/power/runtime_pm.txt`. +Refer to that document for more information regarding this particular issue as +well as for information on the device runtime power management framework in +general. diff --git a/Documentation/driver-api/pm/index.rst b/Documentation/driver-api/pm/index.rst new file mode 100644 index 000000000000..2f6d0e9cf6b7 --- /dev/null +++ b/Documentation/driver-api/pm/index.rst @@ -0,0 +1,16 @@ +======================= +Device Power Management +======================= + +.. toctree:: + + devices + notifiers + types + +.. only:: subproject and html + + Indices + ======= + + * :ref:`genindex` diff --git a/Documentation/driver-api/pm/notifiers.rst b/Documentation/driver-api/pm/notifiers.rst new file mode 100644 index 000000000000..62f860026992 --- /dev/null +++ b/Documentation/driver-api/pm/notifiers.rst @@ -0,0 +1,70 @@ +============================= +Suspend/Hibernation Notifiers +============================= + +:: + + Copyright (c) 2016 Intel Corp., Rafael J. Wysocki <rafael.j.wysocki@intel.com> + +There are some operations that subsystems or drivers may want to carry out +before hibernation/suspend or after restore/resume, but they require the system +to be fully functional, so the drivers' and subsystems' ``->suspend()`` and +``->resume()`` or even ``->prepare()`` and ``->complete()`` callbacks are not +suitable for this purpose. + +For example, device drivers may want to upload firmware to their devices after +resume/restore, but they cannot do it by calling :c:func:`request_firmware()` +from their ``->resume()`` or ``->complete()`` callback routines (user land +processes are frozen at these points). The solution may be to load the firmware +into memory before processes are frozen and upload it from there in the +``->resume()`` routine. A suspend/hibernation notifier may be used for that. + +Subsystems or drivers having such needs can register suspend notifiers that +will be called upon the following events by the PM core: + +``PM_HIBERNATION_PREPARE`` + The system is going to hibernate, tasks will be frozen immediately. This + is different from ``PM_SUSPEND_PREPARE`` below, because in this case + additional work is done between the notifiers and the invocation of PM + callbacks for the "freeze" transition. + +``PM_POST_HIBERNATION`` + The system memory state has been restored from a hibernation image or an + error occurred during hibernation. Device restore callbacks have been + executed and tasks have been thawed. + +``PM_RESTORE_PREPARE`` + The system is going to restore a hibernation image. If all goes well, + the restored image kernel will issue a ``PM_POST_HIBERNATION`` + notification. + +``PM_POST_RESTORE`` + An error occurred during restore from hibernation. Device restore + callbacks have been executed and tasks have been thawed. + +``PM_SUSPEND_PREPARE`` + The system is preparing for suspend. + +``PM_POST_SUSPEND`` + The system has just resumed or an error occurred during suspend. Device + resume callbacks have been executed and tasks have been thawed. + +It is generally assumed that whatever the notifiers do for +``PM_HIBERNATION_PREPARE``, should be undone for ``PM_POST_HIBERNATION``. +Analogously, operations carried out for ``PM_SUSPEND_PREPARE`` should be +reversed for ``PM_POST_SUSPEND``. + +Moreover, if one of the notifiers fails for the ``PM_HIBERNATION_PREPARE`` or +``PM_SUSPEND_PREPARE`` event, the notifiers that have already succeeded for that +event will be called for ``PM_POST_HIBERNATION`` or ``PM_POST_SUSPEND``, +respectively. + +The hibernation and suspend notifiers are called with :c:data:`pm_mutex` held. +They are defined in the usual way, but their last argument is meaningless (it is +always NULL). + +To register and/or unregister a suspend notifier use +:c:func:`register_pm_notifier()` and :c:func:`unregister_pm_notifier()`, +respectively (both defined in :file:`include/linux/suspend.h`). If you don't +need to unregister the notifier, you can also use the :c:func:`pm_notifier()` +macro defined in :file:`include/linux/suspend.h`. diff --git a/Documentation/driver-api/pm/types.rst b/Documentation/driver-api/pm/types.rst new file mode 100644 index 000000000000..3ebdecc54104 --- /dev/null +++ b/Documentation/driver-api/pm/types.rst @@ -0,0 +1,5 @@ +================================== +Device Power Management Data Types +================================== + +.. kernel-doc:: include/linux/pm.h diff --git a/Documentation/driver-api/regulator.rst b/Documentation/driver-api/regulator.rst new file mode 100644 index 000000000000..520da0a5251d --- /dev/null +++ b/Documentation/driver-api/regulator.rst @@ -0,0 +1,170 @@ +.. Copyright 2007-2008 Wolfson Microelectronics + +.. This documentation is free software; you can redistribute +.. it and/or modify it under the terms of the GNU General Public +.. License version 2 as published by the Free Software Foundation. + +================================= +Voltage and current regulator API +================================= + +:Author: Liam Girdwood +:Author: Mark Brown + +Introduction +============ + +This framework is designed to provide a standard kernel interface to +control voltage and current regulators. + +The intention is to allow systems to dynamically control regulator power +output in order to save power and prolong battery life. This applies to +both voltage regulators (where voltage output is controllable) and +current sinks (where current limit is controllable). + +Note that additional (and currently more complete) documentation is +available in the Linux kernel source under +``Documentation/power/regulator``. + +Glossary +-------- + +The regulator API uses a number of terms which may not be familiar: + +Regulator + + Electronic device that supplies power to other devices. Most regulators + can enable and disable their output and some can also control their + output voltage or current. + +Consumer + + Electronic device which consumes power provided by a regulator. These + may either be static, requiring only a fixed supply, or dynamic, + requiring active management of the regulator at runtime. + +Power Domain + + The electronic circuit supplied by a given regulator, including the + regulator and all consumer devices. The configuration of the regulator + is shared between all the components in the circuit. + +Power Management Integrated Circuit (PMIC) + + An IC which contains numerous regulators and often also other + subsystems. In an embedded system the primary PMIC is often equivalent + to a combination of the PSU and southbridge in a desktop system. + +Consumer driver interface +========================= + +This offers a similar API to the kernel clock framework. Consumer +drivers use `get <#API-regulator-get>`__ and +`put <#API-regulator-put>`__ operations to acquire and release +regulators. Functions are provided to `enable <#API-regulator-enable>`__ +and `disable <#API-regulator-disable>`__ the regulator and to get and +set the runtime parameters of the regulator. + +When requesting regulators consumers use symbolic names for their +supplies, such as "Vcc", which are mapped into actual regulator devices +by the machine interface. + +A stub version of this API is provided when the regulator framework is +not in use in order to minimise the need to use ifdefs. + +Enabling and disabling +---------------------- + +The regulator API provides reference counted enabling and disabling of +regulators. Consumer devices use the :c:func:`regulator_enable()` and +:c:func:`regulator_disable()` functions to enable and disable +regulators. Calls to the two functions must be balanced. + +Note that since multiple consumers may be using a regulator and machine +constraints may not allow the regulator to be disabled there is no +guarantee that calling :c:func:`regulator_disable()` will actually +cause the supply provided by the regulator to be disabled. Consumer +drivers should assume that the regulator may be enabled at all times. + +Configuration +------------- + +Some consumer devices may need to be able to dynamically configure their +supplies. For example, MMC drivers may need to select the correct +operating voltage for their cards. This may be done while the regulator +is enabled or disabled. + +The :c:func:`regulator_set_voltage()` and +:c:func:`regulator_set_current_limit()` functions provide the primary +interface for this. Both take ranges of voltages and currents, supporting +drivers that do not require a specific value (eg, CPU frequency scaling +normally permits the CPU to use a wider range of supply voltages at lower +frequencies but does not require that the supply voltage be lowered). Where +an exact value is required both minimum and maximum values should be +identical. + +Callbacks +--------- + +Callbacks may also be registered for events such as regulation failures. + +Regulator driver interface +========================== + +Drivers for regulator chips register the regulators with the regulator +core, providing operations structures to the core. A notifier interface +allows error conditions to be reported to the core. + +Registration should be triggered by explicit setup done by the platform, +supplying a struct :c:type:`regulator_init_data` for the regulator +containing constraint and supply information. + +Machine interface +================= + +This interface provides a way to define how regulators are connected to +consumers on a given system and what the valid operating parameters are +for the system. + +Supplies +-------- + +Regulator supplies are specified using struct +:c:type:`regulator_consumer_supply`. This is done at driver registration +time as part of the machine constraints. + +Constraints +----------- + +As well as defining the connections the machine interface also provides +constraints defining the operations that clients are allowed to perform +and the parameters that may be set. This is required since generally +regulator devices will offer more flexibility than it is safe to use on +a given system, for example supporting higher supply voltages than the +consumers are rated for. + +This is done at driver registration time` by providing a +struct :c:type:`regulation_constraints`. + +The constraints may also specify an initial configuration for the +regulator in the constraints, which is particularly useful for use with +static consumers. + +API reference +============= + +Due to limitations of the kernel documentation framework and the +existing layout of the source code the entire regulator API is +documented here. + +.. kernel-doc:: include/linux/regulator/consumer.h + :internal: + +.. kernel-doc:: include/linux/regulator/machine.h + :internal: + +.. kernel-doc:: include/linux/regulator/driver.h + :internal: + +.. kernel-doc:: drivers/regulator/core.c + :export: diff --git a/Documentation/hwmon/ds1621 b/Documentation/hwmon/ds1621 index f775e612f582..fa3407997795 100644 --- a/Documentation/hwmon/ds1621 +++ b/Documentation/hwmon/ds1621 @@ -117,10 +117,10 @@ support, which is achieved via the R0 and R1 config register bits, where: R0..R1 ------ - 0 0 => 9 bits, 0.5 degrees Celcius - 1 0 => 10 bits, 0.25 degrees Celcius - 0 1 => 11 bits, 0.125 degrees Celcius - 1 1 => 12 bits, 0.0625 degrees Celcius + 0 0 => 9 bits, 0.5 degrees Celsius + 1 0 => 10 bits, 0.25 degrees Celsius + 0 1 => 11 bits, 0.125 degrees Celsius + 1 1 => 12 bits, 0.0625 degrees Celsius Note: At initial device power-on, the default resolution is set to 12-bits. diff --git a/Documentation/index.rst b/Documentation/index.rst index cb5d77699c60..f6e641a54bbc 100644 --- a/Documentation/index.rst +++ b/Documentation/index.rst @@ -47,7 +47,7 @@ These books get into the details of how specific kernel subsystems work from the point of view of a kernel developer. Much of the information here is taken directly from the kernel source, with supplemental material added as needed (or at least as we managed to add it — probably *not* all that is -needed). +needed). .. toctree:: :maxdepth: 2 @@ -68,6 +68,14 @@ Korean translations translations/ko_KR/index +Chinese translations +-------------------- + +.. toctree:: + :maxdepth: 1 + + translations/zh_CN/index + Indices and tables ================== diff --git a/Documentation/input/input.txt b/Documentation/input/input.txt index 0acfddbe2028..7ebce100fe90 100644 --- a/Documentation/input/input.txt +++ b/Documentation/input/input.txt @@ -279,10 +279,10 @@ struct input_event { 'time' is the timestamp, it returns the time at which the event happened. Type is for example EV_REL for relative moment, EV_KEY for a keypress or -release. More types are defined in include/linux/input.h. +release. More types are defined in include/uapi/linux/input-event-codes.h. 'code' is event code, for example REL_X or KEY_BACKSPACE, again a complete -list is in include/linux/input.h. +list is in include/uapi/linux/input-event-codes.h. 'value' is the value the event carries. Either a relative change for EV_REL, absolute new value for EV_ABS (joysticks ...), or 0 for EV_KEY for diff --git a/Documentation/ioctl/botching-up-ioctls.txt b/Documentation/ioctl/botching-up-ioctls.txt index 36138c632f7a..d02cfb48901c 100644 --- a/Documentation/ioctl/botching-up-ioctls.txt +++ b/Documentation/ioctl/botching-up-ioctls.txt @@ -24,7 +24,7 @@ Prerequisites ------------- First the prerequisites. Without these you have already failed, because you -will need to add a a 32-bit compat layer: +will need to add a 32-bit compat layer: * Only use fixed sized integers. To avoid conflicts with typedefs in userspace the kernel has special types like __u32, __s64. Use them. diff --git a/Documentation/livepatch/livepatch.txt b/Documentation/livepatch/livepatch.txt index 7f04e13ec53d..9d2096c7160d 100644 --- a/Documentation/livepatch/livepatch.txt +++ b/Documentation/livepatch/livepatch.txt @@ -358,7 +358,7 @@ The current Livepatch implementation has several limitations: Each function has to handle TOC and save LR before it could call the ftrace handler. This operation has to be reverted on return. Fortunately, the generic ftrace code has the same problem and all - this is is handled on the ftrace level. + this is handled on the ftrace level. + Kretprobes using the ftrace framework conflict with the patched diff --git a/Documentation/media/Makefile b/Documentation/media/Makefile index 32663602ff25..9b3e70b2cab2 100644 --- a/Documentation/media/Makefile +++ b/Documentation/media/Makefile @@ -36,7 +36,7 @@ quiet_cmd_genpdf = GENPDF $2 cmd_genpdf = convert $2 $3 quiet_cmd_gendot = DOT $2 - cmd_gendot = dot -Tsvg $2 > $3 + cmd_gendot = dot -Tsvg $2 > $3 || { rm -f $3; exit 1; } %.pdf: %.svg @$(call cmd,genpdf,$<,$@) @@ -103,6 +103,7 @@ html: all epub: all xml: all latex: $(IMGPDF) all +linkcheck: clean: -rm -f $(DOTTGT) $(IMGTGT) ${TARGETS} 2>/dev/null diff --git a/Documentation/networking/kcm.txt b/Documentation/networking/kcm.txt index 3476ede5bc2c..9a513295b07c 100644 --- a/Documentation/networking/kcm.txt +++ b/Documentation/networking/kcm.txt @@ -272,7 +272,7 @@ on the socket thus waking up the application thread. When the application sees the error (which may just be a disconnect) it should unattach the socket from KCM and then close it. It is assumed that once an error is posted on the TCP socket the data stream is unrecoverable (i.e. an error -may have occurred in in the middle of receiving a messssge). +may have occurred in the middle of receiving a messssge). TCP connection monitoring ------------------------- diff --git a/Documentation/power/00-INDEX b/Documentation/power/00-INDEX index 7cb6085839f3..7f3c2def2cac 100644 --- a/Documentation/power/00-INDEX +++ b/Documentation/power/00-INDEX @@ -14,8 +14,6 @@ freezing-of-tasks.txt - How processes and controlled during suspend interface.txt - Power management user interface in /sys/power -notifiers.txt - - Registering suspend notifiers in device drivers opp.txt - Operating Performance Point library pci.txt diff --git a/Documentation/power/devices.txt b/Documentation/power/devices.txt deleted file mode 100644 index 73ddea39a9ce..000000000000 --- a/Documentation/power/devices.txt +++ /dev/null @@ -1,716 +0,0 @@ -Device Power Management - -Copyright (c) 2010-2011 Rafael J. Wysocki <rjw@sisk.pl>, Novell Inc. -Copyright (c) 2010 Alan Stern <stern@rowland.harvard.edu> -Copyright (c) 2014 Intel Corp., Rafael J. Wysocki <rafael.j.wysocki@intel.com> - - -Most of the code in Linux is device drivers, so most of the Linux power -management (PM) code is also driver-specific. Most drivers will do very -little; others, especially for platforms with small batteries (like cell -phones), will do a lot. - -This writeup gives an overview of how drivers interact with system-wide -power management goals, emphasizing the models and interfaces that are -shared by everything that hooks up to the driver model core. Read it as -background for the domain-specific work you'd do with any specific driver. - - -Two Models for Device Power Management -====================================== -Drivers will use one or both of these models to put devices into low-power -states: - - System Sleep model: - Drivers can enter low-power states as part of entering system-wide - low-power states like "suspend" (also known as "suspend-to-RAM"), or - (mostly for systems with disks) "hibernation" (also known as - "suspend-to-disk"). - - This is something that device, bus, and class drivers collaborate on - by implementing various role-specific suspend and resume methods to - cleanly power down hardware and software subsystems, then reactivate - them without loss of data. - - Some drivers can manage hardware wakeup events, which make the system - leave the low-power state. This feature may be enabled or disabled - using the relevant /sys/devices/.../power/wakeup file (for Ethernet - drivers the ioctl interface used by ethtool may also be used for this - purpose); enabling it may cost some power usage, but let the whole - system enter low-power states more often. - - Runtime Power Management model: - Devices may also be put into low-power states while the system is - running, independently of other power management activity in principle. - However, devices are not generally independent of each other (for - example, a parent device cannot be suspended unless all of its child - devices have been suspended). Moreover, depending on the bus type the - device is on, it may be necessary to carry out some bus-specific - operations on the device for this purpose. Devices put into low power - states at run time may require special handling during system-wide power - transitions (suspend or hibernation). - - For these reasons not only the device driver itself, but also the - appropriate subsystem (bus type, device type or device class) driver and - the PM core are involved in runtime power management. As in the system - sleep power management case, they need to collaborate by implementing - various role-specific suspend and resume methods, so that the hardware - is cleanly powered down and reactivated without data or service loss. - -There's not a lot to be said about those low-power states except that they are -very system-specific, and often device-specific. Also, that if enough devices -have been put into low-power states (at runtime), the effect may be very similar -to entering some system-wide low-power state (system sleep) ... and that -synergies exist, so that several drivers using runtime PM might put the system -into a state where even deeper power saving options are available. - -Most suspended devices will have quiesced all I/O: no more DMA or IRQs (except -for wakeup events), no more data read or written, and requests from upstream -drivers are no longer accepted. A given bus or platform may have different -requirements though. - -Examples of hardware wakeup events include an alarm from a real time clock, -network wake-on-LAN packets, keyboard or mouse activity, and media insertion -or removal (for PCMCIA, MMC/SD, USB, and so on). - - -Interfaces for Entering System Sleep States -=========================================== -There are programming interfaces provided for subsystems (bus type, device type, -device class) and device drivers to allow them to participate in the power -management of devices they are concerned with. These interfaces cover both -system sleep and runtime power management. - - -Device Power Management Operations ----------------------------------- -Device power management operations, at the subsystem level as well as at the -device driver level, are implemented by defining and populating objects of type -struct dev_pm_ops: - -struct dev_pm_ops { - int (*prepare)(struct device *dev); - void (*complete)(struct device *dev); - int (*suspend)(struct device *dev); - int (*resume)(struct device *dev); - int (*freeze)(struct device *dev); - int (*thaw)(struct device *dev); - int (*poweroff)(struct device *dev); - int (*restore)(struct device *dev); - int (*suspend_late)(struct device *dev); - int (*resume_early)(struct device *dev); - int (*freeze_late)(struct device *dev); - int (*thaw_early)(struct device *dev); - int (*poweroff_late)(struct device *dev); - int (*restore_early)(struct device *dev); - int (*suspend_noirq)(struct device *dev); - int (*resume_noirq)(struct device *dev); - int (*freeze_noirq)(struct device *dev); - int (*thaw_noirq)(struct device *dev); - int (*poweroff_noirq)(struct device *dev); - int (*restore_noirq)(struct device *dev); - int (*runtime_suspend)(struct device *dev); - int (*runtime_resume)(struct device *dev); - int (*runtime_idle)(struct device *dev); -}; - -This structure is defined in include/linux/pm.h and the methods included in it -are also described in that file. Their roles will be explained in what follows. -For now, it should be sufficient to remember that the last three methods are -specific to runtime power management while the remaining ones are used during -system-wide power transitions. - -There also is a deprecated "old" or "legacy" interface for power management -operations available at least for some subsystems. This approach does not use -struct dev_pm_ops objects and it is suitable only for implementing system sleep -power management methods. Therefore it is not described in this document, so -please refer directly to the source code for more information about it. - - -Subsystem-Level Methods ------------------------ -The core methods to suspend and resume devices reside in struct dev_pm_ops -pointed to by the ops member of struct dev_pm_domain, or by the pm member of -struct bus_type, struct device_type and struct class. They are mostly of -interest to the people writing infrastructure for platforms and buses, like PCI -or USB, or device type and device class drivers. They also are relevant to the -writers of device drivers whose subsystems (PM domains, device types, device -classes and bus types) don't provide all power management methods. - -Bus drivers implement these methods as appropriate for the hardware and the -drivers using it; PCI works differently from USB, and so on. Not many people -write subsystem-level drivers; most driver code is a "device driver" that builds -on top of bus-specific framework code. - -For more information on these driver calls, see the description later; -they are called in phases for every device, respecting the parent-child -sequencing in the driver model tree. - - -/sys/devices/.../power/wakeup files ------------------------------------ -All device objects in the driver model contain fields that control the handling -of system wakeup events (hardware signals that can force the system out of a -sleep state). These fields are initialized by bus or device driver code using -device_set_wakeup_capable() and device_set_wakeup_enable(), defined in -include/linux/pm_wakeup.h. - -The "power.can_wakeup" flag just records whether the device (and its driver) can -physically support wakeup events. The device_set_wakeup_capable() routine -affects this flag. The "power.wakeup" field is a pointer to an object of type -struct wakeup_source used for controlling whether or not the device should use -its system wakeup mechanism and for notifying the PM core of system wakeup -events signaled by the device. This object is only present for wakeup-capable -devices (i.e. devices whose "can_wakeup" flags are set) and is created (or -removed) by device_set_wakeup_capable(). - -Whether or not a device is capable of issuing wakeup events is a hardware -matter, and the kernel is responsible for keeping track of it. By contrast, -whether or not a wakeup-capable device should issue wakeup events is a policy -decision, and it is managed by user space through a sysfs attribute: the -"power/wakeup" file. User space can write the strings "enabled" or "disabled" -to it to indicate whether or not, respectively, the device is supposed to signal -system wakeup. This file is only present if the "power.wakeup" object exists -for the given device and is created (or removed) along with that object, by -device_set_wakeup_capable(). Reads from the file will return the corresponding -string. - -The "power/wakeup" file is supposed to contain the "disabled" string initially -for the majority of devices; the major exceptions are power buttons, keyboards, -and Ethernet adapters whose WoL (wake-on-LAN) feature has been set up with -ethtool. It should also default to "enabled" for devices that don't generate -wakeup requests on their own but merely forward wakeup requests from one bus to -another (like PCI Express ports). - -The device_may_wakeup() routine returns true only if the "power.wakeup" object -exists and the corresponding "power/wakeup" file contains the string "enabled". -This information is used by subsystems, like the PCI bus type code, to see -whether or not to enable the devices' wakeup mechanisms. If device wakeup -mechanisms are enabled or disabled directly by drivers, they also should use -device_may_wakeup() to decide what to do during a system sleep transition. -Device drivers, however, are not supposed to call device_set_wakeup_enable() -directly in any case. - -It ought to be noted that system wakeup is conceptually different from "remote -wakeup" used by runtime power management, although it may be supported by the -same physical mechanism. Remote wakeup is a feature allowing devices in -low-power states to trigger specific interrupts to signal conditions in which -they should be put into the full-power state. Those interrupts may or may not -be used to signal system wakeup events, depending on the hardware design. On -some systems it is impossible to trigger them from system sleep states. In any -case, remote wakeup should always be enabled for runtime power management for -all devices and drivers that support it. - -/sys/devices/.../power/control files ------------------------------------- -Each device in the driver model has a flag to control whether it is subject to -runtime power management. This flag, called runtime_auto, is initialized by the -bus type (or generally subsystem) code using pm_runtime_allow() or -pm_runtime_forbid(); the default is to allow runtime power management. - -The setting can be adjusted by user space by writing either "on" or "auto" to -the device's power/control sysfs file. Writing "auto" calls pm_runtime_allow(), -setting the flag and allowing the device to be runtime power-managed by its -driver. Writing "on" calls pm_runtime_forbid(), clearing the flag, returning -the device to full power if it was in a low-power state, and preventing the -device from being runtime power-managed. User space can check the current value -of the runtime_auto flag by reading the file. - -The device's runtime_auto flag has no effect on the handling of system-wide -power transitions. In particular, the device can (and in the majority of cases -should and will) be put into a low-power state during a system-wide transition -to a sleep state even though its runtime_auto flag is clear. - -For more information about the runtime power management framework, refer to -Documentation/power/runtime_pm.txt. - - -Calling Drivers to Enter and Leave System Sleep States -====================================================== -When the system goes into a sleep state, each device's driver is asked to -suspend the device by putting it into a state compatible with the target -system state. That's usually some version of "off", but the details are -system-specific. Also, wakeup-enabled devices will usually stay partly -functional in order to wake the system. - -When the system leaves that low-power state, the device's driver is asked to -resume it by returning it to full power. The suspend and resume operations -always go together, and both are multi-phase operations. - -For simple drivers, suspend might quiesce the device using class code -and then turn its hardware as "off" as possible during suspend_noirq. The -matching resume calls would then completely reinitialize the hardware -before reactivating its class I/O queues. - -More power-aware drivers might prepare the devices for triggering system wakeup -events. - - -Call Sequence Guarantees ------------------------- -To ensure that bridges and similar links needing to talk to a device are -available when the device is suspended or resumed, the device tree is -walked in a bottom-up order to suspend devices. A top-down order is -used to resume those devices. - -The ordering of the device tree is defined by the order in which devices -get registered: a child can never be registered, probed or resumed before -its parent; and can't be removed or suspended after that parent. - -The policy is that the device tree should match hardware bus topology. -(Or at least the control bus, for devices which use multiple busses.) -In particular, this means that a device registration may fail if the parent of -the device is suspending (i.e. has been chosen by the PM core as the next -device to suspend) or has already suspended, as well as after all of the other -devices have been suspended. Device drivers must be prepared to cope with such -situations. - - -System Power Management Phases ------------------------------- -Suspending or resuming the system is done in several phases. Different phases -are used for freeze, standby, and memory sleep states ("suspend-to-RAM") and the -hibernation state ("suspend-to-disk"). Each phase involves executing callbacks -for every device before the next phase begins. Not all busses or classes -support all these callbacks and not all drivers use all the callbacks. The -various phases always run after tasks have been frozen and before they are -unfrozen. Furthermore, the *_noirq phases run at a time when IRQ handlers have -been disabled (except for those marked with the IRQF_NO_SUSPEND flag). - -All phases use PM domain, bus, type, class or driver callbacks (that is, methods -defined in dev->pm_domain->ops, dev->bus->pm, dev->type->pm, dev->class->pm or -dev->driver->pm). These callbacks are regarded by the PM core as mutually -exclusive. Moreover, PM domain callbacks always take precedence over all of the -other callbacks and, for example, type callbacks take precedence over bus, class -and driver callbacks. To be precise, the following rules are used to determine -which callback to execute in the given phase: - - 1. If dev->pm_domain is present, the PM core will choose the callback - included in dev->pm_domain->ops for execution - - 2. Otherwise, if both dev->type and dev->type->pm are present, the callback - included in dev->type->pm will be chosen for execution. - - 3. Otherwise, if both dev->class and dev->class->pm are present, the - callback included in dev->class->pm will be chosen for execution. - - 4. Otherwise, if both dev->bus and dev->bus->pm are present, the callback - included in dev->bus->pm will be chosen for execution. - -This allows PM domains and device types to override callbacks provided by bus -types or device classes if necessary. - -The PM domain, type, class and bus callbacks may in turn invoke device- or -driver-specific methods stored in dev->driver->pm, but they don't have to do -that. - -If the subsystem callback chosen for execution is not present, the PM core will -execute the corresponding method from dev->driver->pm instead if there is one. - - -Entering System Suspend ------------------------ -When the system goes into the freeze, standby or memory sleep state, -the phases are: - - prepare, suspend, suspend_late, suspend_noirq. - - 1. The prepare phase is meant to prevent races by preventing new devices - from being registered; the PM core would never know that all the - children of a device had been suspended if new children could be - registered at will. (By contrast, devices may be unregistered at any - time.) Unlike the other suspend-related phases, during the prepare - phase the device tree is traversed top-down. - - After the prepare callback method returns, no new children may be - registered below the device. The method may also prepare the device or - driver in some way for the upcoming system power transition, but it - should not put the device into a low-power state. - - For devices supporting runtime power management, the return value of the - prepare callback can be used to indicate to the PM core that it may - safely leave the device in runtime suspend (if runtime-suspended - already), provided that all of the device's descendants are also left in - runtime suspend. Namely, if the prepare callback returns a positive - number and that happens for all of the descendants of the device too, - and all of them (including the device itself) are runtime-suspended, the - PM core will skip the suspend, suspend_late and suspend_noirq suspend - phases as well as the resume_noirq, resume_early and resume phases of - the following system resume for all of these devices. In that case, - the complete callback will be called directly after the prepare callback - and is entirely responsible for bringing the device back to the - functional state as appropriate. - - Note that this direct-complete procedure applies even if the device is - disabled for runtime PM; only the runtime-PM status matters. It follows - that if a device has system-sleep callbacks but does not support runtime - PM, then its prepare callback must never return a positive value. This - is because all devices are initially set to runtime-suspended with - runtime PM disabled. - - 2. The suspend methods should quiesce the device to stop it from performing - I/O. They also may save the device registers and put it into the - appropriate low-power state, depending on the bus type the device is on, - and they may enable wakeup events. - - 3 For a number of devices it is convenient to split suspend into the - "quiesce device" and "save device state" phases, in which cases - suspend_late is meant to do the latter. It is always executed after - runtime power management has been disabled for all devices. - - 4. The suspend_noirq phase occurs after IRQ handlers have been disabled, - which means that the driver's interrupt handler will not be called while - the callback method is running. The methods should save the values of - the device's registers that weren't saved previously and finally put the - device into the appropriate low-power state. - - The majority of subsystems and device drivers need not implement this - callback. However, bus types allowing devices to share interrupt - vectors, like PCI, generally need it; otherwise a driver might encounter - an error during the suspend phase by fielding a shared interrupt - generated by some other device after its own device had been set to low - power. - -At the end of these phases, drivers should have stopped all I/O transactions -(DMA, IRQs), saved enough state that they can re-initialize or restore previous -state (as needed by the hardware), and placed the device into a low-power state. -On many platforms they will gate off one or more clock sources; sometimes they -will also switch off power supplies or reduce voltages. (Drivers supporting -runtime PM may already have performed some or all of these steps.) - -If device_may_wakeup(dev) returns true, the device should be prepared for -generating hardware wakeup signals to trigger a system wakeup event when the -system is in the sleep state. For example, enable_irq_wake() might identify -GPIO signals hooked up to a switch or other external hardware, and -pci_enable_wake() does something similar for the PCI PME signal. - -If any of these callbacks returns an error, the system won't enter the desired -low-power state. Instead the PM core will unwind its actions by resuming all -the devices that were suspended. - - -Leaving System Suspend ----------------------- -When resuming from freeze, standby or memory sleep, the phases are: - - resume_noirq, resume_early, resume, complete. - - 1. The resume_noirq callback methods should perform any actions needed - before the driver's interrupt handlers are invoked. This generally - means undoing the actions of the suspend_noirq phase. If the bus type - permits devices to share interrupt vectors, like PCI, the method should - bring the device and its driver into a state in which the driver can - recognize if the device is the source of incoming interrupts, if any, - and handle them correctly. - - For example, the PCI bus type's ->pm.resume_noirq() puts the device into - the full-power state (D0 in the PCI terminology) and restores the - standard configuration registers of the device. Then it calls the - device driver's ->pm.resume_noirq() method to perform device-specific - actions. - - 2. The resume_early methods should prepare devices for the execution of - the resume methods. This generally involves undoing the actions of the - preceding suspend_late phase. - - 3 The resume methods should bring the device back to its operating - state, so that it can perform normal I/O. This generally involves - undoing the actions of the suspend phase. - - 4. The complete phase should undo the actions of the prepare phase. Note, - however, that new children may be registered below the device as soon as - the resume callbacks occur; it's not necessary to wait until the - complete phase. - - Moreover, if the preceding prepare callback returned a positive number, - the device may have been left in runtime suspend throughout the whole - system suspend and resume (the suspend, suspend_late, suspend_noirq - phases of system suspend and the resume_noirq, resume_early, resume - phases of system resume may have been skipped for it). In that case, - the complete callback is entirely responsible for bringing the device - back to the functional state after system suspend if necessary. [For - example, it may need to queue up a runtime resume request for the device - for this purpose.] To check if that is the case, the complete callback - can consult the device's power.direct_complete flag. Namely, if that - flag is set when the complete callback is being run, it has been called - directly after the preceding prepare and special action may be required - to make the device work correctly afterward. - -At the end of these phases, drivers should be as functional as they were before -suspending: I/O can be performed using DMA and IRQs, and the relevant clocks are -gated on. - -However, the details here may again be platform-specific. For example, -some systems support multiple "run" states, and the mode in effect at -the end of resume might not be the one which preceded suspension. -That means availability of certain clocks or power supplies changed, -which could easily affect how a driver works. - -Drivers need to be able to handle hardware which has been reset since the -suspend methods were called, for example by complete reinitialization. -This may be the hardest part, and the one most protected by NDA'd documents -and chip errata. It's simplest if the hardware state hasn't changed since -the suspend was carried out, but that can't be guaranteed (in fact, it usually -is not the case). - -Drivers must also be prepared to notice that the device has been removed -while the system was powered down, whenever that's physically possible. -PCMCIA, MMC, USB, Firewire, SCSI, and even IDE are common examples of busses -where common Linux platforms will see such removal. Details of how drivers -will notice and handle such removals are currently bus-specific, and often -involve a separate thread. - -These callbacks may return an error value, but the PM core will ignore such -errors since there's nothing it can do about them other than printing them in -the system log. - - -Entering Hibernation --------------------- -Hibernating the system is more complicated than putting it into the other -sleep states, because it involves creating and saving a system image. -Therefore there are more phases for hibernation, with a different set of -callbacks. These phases always run after tasks have been frozen and memory has -been freed. - -The general procedure for hibernation is to quiesce all devices (freeze), create -an image of the system memory while everything is stable, reactivate all -devices (thaw), write the image to permanent storage, and finally shut down the -system (poweroff). The phases used to accomplish this are: - - prepare, freeze, freeze_late, freeze_noirq, thaw_noirq, thaw_early, - thaw, complete, prepare, poweroff, poweroff_late, poweroff_noirq - - 1. The prepare phase is discussed in the "Entering System Suspend" section - above. - - 2. The freeze methods should quiesce the device so that it doesn't generate - IRQs or DMA, and they may need to save the values of device registers. - However the device does not have to be put in a low-power state, and to - save time it's best not to do so. Also, the device should not be - prepared to generate wakeup events. - - 3. The freeze_late phase is analogous to the suspend_late phase described - above, except that the device should not be put in a low-power state and - should not be allowed to generate wakeup events by it. - - 4. The freeze_noirq phase is analogous to the suspend_noirq phase discussed - above, except again that the device should not be put in a low-power - state and should not be allowed to generate wakeup events. - -At this point the system image is created. All devices should be inactive and -the contents of memory should remain undisturbed while this happens, so that the -image forms an atomic snapshot of the system state. - - 5. The thaw_noirq phase is analogous to the resume_noirq phase discussed - above. The main difference is that its methods can assume the device is - in the same state as at the end of the freeze_noirq phase. - - 6. The thaw_early phase is analogous to the resume_early phase described - above. Its methods should undo the actions of the preceding - freeze_late, if necessary. - - 7. The thaw phase is analogous to the resume phase discussed above. Its - methods should bring the device back to an operating state, so that it - can be used for saving the image if necessary. - - 8. The complete phase is discussed in the "Leaving System Suspend" section - above. - -At this point the system image is saved, and the devices then need to be -prepared for the upcoming system shutdown. This is much like suspending them -before putting the system into the freeze, standby or memory sleep state, -and the phases are similar. - - 9. The prepare phase is discussed above. - - 10. The poweroff phase is analogous to the suspend phase. - - 11. The poweroff_late phase is analogous to the suspend_late phase. - - 12. The poweroff_noirq phase is analogous to the suspend_noirq phase. - -The poweroff, poweroff_late and poweroff_noirq callbacks should do essentially -the same things as the suspend, suspend_late and suspend_noirq callbacks, -respectively. The only notable difference is that they need not store the -device register values, because the registers should already have been stored -during the freeze, freeze_late or freeze_noirq phases. - - -Leaving Hibernation -------------------- -Resuming from hibernation is, again, more complicated than resuming from a sleep -state in which the contents of main memory are preserved, because it requires -a system image to be loaded into memory and the pre-hibernation memory contents -to be restored before control can be passed back to the image kernel. - -Although in principle, the image might be loaded into memory and the -pre-hibernation memory contents restored by the boot loader, in practice this -can't be done because boot loaders aren't smart enough and there is no -established protocol for passing the necessary information. So instead, the -boot loader loads a fresh instance of the kernel, called the boot kernel, into -memory and passes control to it in the usual way. Then the boot kernel reads -the system image, restores the pre-hibernation memory contents, and passes -control to the image kernel. Thus two different kernels are involved in -resuming from hibernation. In fact, the boot kernel may be completely different -from the image kernel: a different configuration and even a different version. -This has important consequences for device drivers and their subsystems. - -To be able to load the system image into memory, the boot kernel needs to -include at least a subset of device drivers allowing it to access the storage -medium containing the image, although it doesn't need to include all of the -drivers present in the image kernel. After the image has been loaded, the -devices managed by the boot kernel need to be prepared for passing control back -to the image kernel. This is very similar to the initial steps involved in -creating a system image, and it is accomplished in the same way, using prepare, -freeze, and freeze_noirq phases. However the devices affected by these phases -are only those having drivers in the boot kernel; other devices will still be in -whatever state the boot loader left them. - -Should the restoration of the pre-hibernation memory contents fail, the boot -kernel would go through the "thawing" procedure described above, using the -thaw_noirq, thaw, and complete phases, and then continue running normally. This -happens only rarely. Most often the pre-hibernation memory contents are -restored successfully and control is passed to the image kernel, which then -becomes responsible for bringing the system back to the working state. - -To achieve this, the image kernel must restore the devices' pre-hibernation -functionality. The operation is much like waking up from the memory sleep -state, although it involves different phases: - - restore_noirq, restore_early, restore, complete - - 1. The restore_noirq phase is analogous to the resume_noirq phase. - - 2. The restore_early phase is analogous to the resume_early phase. - - 3. The restore phase is analogous to the resume phase. - - 4. The complete phase is discussed above. - -The main difference from resume[_early|_noirq] is that restore[_early|_noirq] -must assume the device has been accessed and reconfigured by the boot loader or -the boot kernel. Consequently the state of the device may be different from the -state remembered from the freeze, freeze_late and freeze_noirq phases. The -device may even need to be reset and completely re-initialized. In many cases -this difference doesn't matter, so the resume[_early|_noirq] and -restore[_early|_norq] method pointers can be set to the same routines. -Nevertheless, different callback pointers are used in case there is a situation -where it actually does matter. - - -Device Power Management Domains -------------------------------- -Sometimes devices share reference clocks or other power resources. In those -cases it generally is not possible to put devices into low-power states -individually. Instead, a set of devices sharing a power resource can be put -into a low-power state together at the same time by turning off the shared -power resource. Of course, they also need to be put into the full-power state -together, by turning the shared power resource on. A set of devices with this -property is often referred to as a power domain. A power domain may also be -nested inside another power domain. The nested domain is referred to as the -sub-domain of the parent domain. - -Support for power domains is provided through the pm_domain field of struct -device. This field is a pointer to an object of type struct dev_pm_domain, -defined in include/linux/pm.h, providing a set of power management callbacks -analogous to the subsystem-level and device driver callbacks that are executed -for the given device during all power transitions, instead of the respective -subsystem-level callbacks. Specifically, if a device's pm_domain pointer is -not NULL, the ->suspend() callback from the object pointed to by it will be -executed instead of its subsystem's (e.g. bus type's) ->suspend() callback and -analogously for all of the remaining callbacks. In other words, power -management domain callbacks, if defined for the given device, always take -precedence over the callbacks provided by the device's subsystem (e.g. bus -type). - -The support for device power management domains is only relevant to platforms -needing to use the same device driver power management callbacks in many -different power domain configurations and wanting to avoid incorporating the -support for power domains into subsystem-level callbacks, for example by -modifying the platform bus type. Other platforms need not implement it or take -it into account in any way. - -Devices may be defined as IRQ-safe which indicates to the PM core that their -runtime PM callbacks may be invoked with disabled interrupts (see -Documentation/power/runtime_pm.txt for more information). If an IRQ-safe -device belongs to a PM domain, the runtime PM of the domain will be -disallowed, unless the domain itself is defined as IRQ-safe. However, it -makes sense to define a PM domain as IRQ-safe only if all the devices in it -are IRQ-safe. Moreover, if an IRQ-safe domain has a parent domain, the runtime -PM of the parent is only allowed if the parent itself is IRQ-safe too with the -additional restriction that all child domains of an IRQ-safe parent must also -be IRQ-safe. - -Device Low Power (suspend) States ---------------------------------- -Device low-power states aren't standard. One device might only handle -"on" and "off", while another might support a dozen different versions of -"on" (how many engines are active?), plus a state that gets back to "on" -faster than from a full "off". - -Some busses define rules about what different suspend states mean. PCI -gives one example: after the suspend sequence completes, a non-legacy -PCI device may not perform DMA or issue IRQs, and any wakeup events it -issues would be issued through the PME# bus signal. Plus, there are -several PCI-standard device states, some of which are optional. - -In contrast, integrated system-on-chip processors often use IRQs as the -wakeup event sources (so drivers would call enable_irq_wake) and might -be able to treat DMA completion as a wakeup event (sometimes DMA can stay -active too, it'd only be the CPU and some peripherals that sleep). - -Some details here may be platform-specific. Systems may have devices that -can be fully active in certain sleep states, such as an LCD display that's -refreshed using DMA while most of the system is sleeping lightly ... and -its frame buffer might even be updated by a DSP or other non-Linux CPU while -the Linux control processor stays idle. - -Moreover, the specific actions taken may depend on the target system state. -One target system state might allow a given device to be very operational; -another might require a hard shut down with re-initialization on resume. -And two different target systems might use the same device in different -ways; the aforementioned LCD might be active in one product's "standby", -but a different product using the same SOC might work differently. - - -Power Management Notifiers --------------------------- -There are some operations that cannot be carried out by the power management -callbacks discussed above, because the callbacks occur too late or too early. -To handle these cases, subsystems and device drivers may register power -management notifiers that are called before tasks are frozen and after they have -been thawed. Generally speaking, the PM notifiers are suitable for performing -actions that either require user space to be available, or at least won't -interfere with user space. - -For details refer to Documentation/power/notifiers.txt. - - -Runtime Power Management -======================== -Many devices are able to dynamically power down while the system is still -running. This feature is useful for devices that are not being used, and -can offer significant power savings on a running system. These devices -often support a range of runtime power states, which might use names such -as "off", "sleep", "idle", "active", and so on. Those states will in some -cases (like PCI) be partially constrained by the bus the device uses, and will -usually include hardware states that are also used in system sleep states. - -A system-wide power transition can be started while some devices are in low -power states due to runtime power management. The system sleep PM callbacks -should recognize such situations and react to them appropriately, but the -necessary actions are subsystem-specific. - -In some cases the decision may be made at the subsystem level while in other -cases the device driver may be left to decide. In some cases it may be -desirable to leave a suspended device in that state during a system-wide power -transition, but in other cases the device must be put back into the full-power -state temporarily, for example so that its system wakeup capability can be -disabled. This all depends on the hardware and the design of the subsystem and -device driver in question. - -During system-wide resume from a sleep state it's easiest to put devices into -the full-power state, as explained in Documentation/power/runtime_pm.txt. Refer -to that document for more information regarding this particular issue as well as -for information on the device runtime power management framework in general. diff --git a/Documentation/power/freezing-of-tasks.txt b/Documentation/power/freezing-of-tasks.txt index 85894d83b352..af005770e767 100644 --- a/Documentation/power/freezing-of-tasks.txt +++ b/Documentation/power/freezing-of-tasks.txt @@ -197,7 +197,8 @@ tasks, since it generally exists anyway. A driver must have all firmwares it may need in RAM before suspend() is called. If keeping them is not practical, for example due to their size, they must be -requested early enough using the suspend notifier API described in notifiers.txt. +requested early enough using the suspend notifier API described in +Documentation/driver-api/pm/notifiers.rst. VI. Are there any precautions to be taken to prevent freezing failures? diff --git a/Documentation/power/notifiers.txt b/Documentation/power/notifiers.txt deleted file mode 100644 index a81fa254303d..000000000000 --- a/Documentation/power/notifiers.txt +++ /dev/null @@ -1,55 +0,0 @@ -Suspend notifiers - (C) 2007-2011 Rafael J. Wysocki <rjw@sisk.pl>, GPL - -There are some operations that subsystems or drivers may want to carry out -before hibernation/suspend or after restore/resume, but they require the system -to be fully functional, so the drivers' and subsystems' .suspend() and .resume() -or even .prepare() and .complete() callbacks are not suitable for this purpose. -For example, device drivers may want to upload firmware to their devices after -resume/restore, but they cannot do it by calling request_firmware() from their -.resume() or .complete() routines (user land processes are frozen at these -points). The solution may be to load the firmware into memory before processes -are frozen and upload it from there in the .resume() routine. -A suspend/hibernation notifier may be used for this purpose. - -The subsystems or drivers having such needs can register suspend notifiers that -will be called upon the following events by the PM core: - -PM_HIBERNATION_PREPARE The system is going to hibernate, tasks will be frozen - immediately. This is different from PM_SUSPEND_PREPARE - below because here we do additional work between notifiers - and drivers freezing. - -PM_POST_HIBERNATION The system memory state has been restored from a - hibernation image or an error occurred during - hibernation. Device drivers' restore callbacks have - been executed and tasks have been thawed. - -PM_RESTORE_PREPARE The system is going to restore a hibernation image. - If all goes well, the restored kernel will issue a - PM_POST_HIBERNATION notification. - -PM_POST_RESTORE An error occurred during restore from hibernation. - Device drivers' restore callbacks have been executed - and tasks have been thawed. - -PM_SUSPEND_PREPARE The system is preparing for suspend. - -PM_POST_SUSPEND The system has just resumed or an error occurred during - suspend. Device drivers' resume callbacks have been - executed and tasks have been thawed. - -It is generally assumed that whatever the notifiers do for -PM_HIBERNATION_PREPARE, should be undone for PM_POST_HIBERNATION. Analogously, -operations performed for PM_SUSPEND_PREPARE should be reversed for -PM_POST_SUSPEND. Additionally, all of the notifiers are called for -PM_POST_HIBERNATION if one of them fails for PM_HIBERNATION_PREPARE, and -all of the notifiers are called for PM_POST_SUSPEND if one of them fails for -PM_SUSPEND_PREPARE. - -The hibernation and suspend notifiers are called with pm_mutex held. They are -defined in the usual way, but their last argument is meaningless (it is always -NULL). To register and/or unregister a suspend notifier use the functions -register_pm_notifier() and unregister_pm_notifier(), respectively, defined in -include/linux/suspend.h . If you don't need to unregister the notifier, you can -also use the pm_notifier() macro defined in include/linux/suspend.h . diff --git a/Documentation/power/pci.txt b/Documentation/power/pci.txt index 85c746cbab2c..a1b7f7158930 100644 --- a/Documentation/power/pci.txt +++ b/Documentation/power/pci.txt @@ -713,7 +713,7 @@ In addition to that the prepare() callback may carry out some operations preparing the device to be suspended, although it should not allocate memory (if additional memory is required to suspend the device, it has to be preallocated earlier, for example in a suspend/hibernate notifier as described -in Documentation/power/notifiers.txt). +in Documentation/driver-api/pm/notifiers.rst). 3.1.2. suspend() diff --git a/Documentation/pps/pps.txt b/Documentation/pps/pps.txt index 50022b3c8ebf..1fdbd5447216 100644 --- a/Documentation/pps/pps.txt +++ b/Documentation/pps/pps.txt @@ -63,7 +63,7 @@ for instance) is a PPS source too, and if not they should provide the possibility to open another device as PPS source. In LinuxPPS the PPS sources are simply char devices usually mapped -into files /dev/pps0, /dev/pps1, etc.. +into files /dev/pps0, /dev/pps1, etc. PPS with USB to serial devices @@ -71,9 +71,12 @@ PPS with USB to serial devices It is possible to grab the PPS from an USB to serial device. However, you should take into account the latencies and jitter introduced by -the USB stack. Users has reported clock instability around +-1ms when -synchronized with PPS through USB. This isn't suited for time server -synchronization. +the USB stack. Users have reported clock instability around +-1ms when +synchronized with PPS through USB. With USB 2.0, jitter may decrease +down to the order of 125 microseconds. + +This may be suitable for time server synchronization with NTP because +of its undersampling and algorithms. If your device doesn't report PPS, you can check that the feature is supported by its driver. Most of the time, you only need to add a call @@ -166,7 +169,8 @@ Testing the PPS support In order to test the PPS support even without specific hardware you can use the ktimer driver (see the client subsection in the PPS configuration menu) -and the userland tools provided in the Documentation/pps/ directory. +and the userland tools available in your distribution's pps-tools package, +http://linuxpps.org , or https://github.com/ago/pps-tools . Once you have enabled the compilation of ktimer just modprobe it (if not statically compiled): @@ -183,8 +187,8 @@ and the run ppstest as follow: source 0 - assert 1186592700.388931295, sequence: 365 - clear 0.000000000, sequence: 0 source 0 - assert 1186592701.389032765, sequence: 366 - clear 0.000000000, sequence: 0 -Please, note that to compile userland programs you need the file timepps.h -(see Documentation/pps/). +Please, note that to compile userland programs you need the file timepps.h . +This is available in the pps-tools repository mentioned above. Generators diff --git a/Documentation/thermal/nouveau_thermal b/Documentation/thermal/nouveau_thermal index 60bc29357ac3..6e17a11efcb0 100644 --- a/Documentation/thermal/nouveau_thermal +++ b/Documentation/thermal/nouveau_thermal @@ -42,7 +42,7 @@ thresholds can be configured thanks to the following HWMON attributes: * Critical: temp1_crit and temp1_crit_hyst; * Shutdown: temp1_emergency and temp1_emergency_hyst. -NOTE: Remember that the values are stored as milli degrees Celcius. Don't forget +NOTE: Remember that the values are stored as milli degrees Celsius. Don't forget to multiply! Fan management diff --git a/Documentation/translations/ja_JP/HOWTO b/Documentation/translations/ja_JP/HOWTO index b03fc8047f03..4ebd20750ef1 100644 --- a/Documentation/translations/ja_JP/HOWTO +++ b/Documentation/translations/ja_JP/HOWTO @@ -111,7 +111,7 @@ Linux カーãƒãƒ«ã‚½ãƒ¼ã‚¹ãƒ„リーã¯å¹…広ã„範囲ã®ãƒ‰ã‚ュメントをå カーãƒãƒ«ã®å¤‰æ›´ãŒã€ã‚«ãƒ¼ãƒãƒ«ãŒãƒ¦ãƒ¼ã‚¶ç©ºé–“ã«å…¬é–‹ã—ã¦ã„るインターフェイス㮠変更を引ãèµ·ã“ã™å ´åˆã€ãã®å¤‰æ›´ã‚’説明ã™ã‚‹ãƒžãƒ‹ãƒ¥ã‚¢ãƒ«ãƒšãƒ¼ã‚¸ã®ãƒ‘ッãƒã‚„æƒ…å ± をマニュアルページã®ãƒ¡ãƒ³ãƒ†ãƒŠ mtk.manpages@gmail.com ã«é€ã‚Šã€CC ã‚’ -linux-api@ver.kernel.org ã«é€ã‚‹ã“ã¨ã‚’勧ã‚ã¾ã™ã€‚ +linux-api@vger.kernel.org ã«é€ã‚‹ã“ã¨ã‚’勧ã‚ã¾ã™ã€‚ 以下ã¯ã‚«ãƒ¼ãƒãƒ«ã‚½ãƒ¼ã‚¹ãƒ„リーã«å«ã¾ã‚Œã¦ã„ã‚‹èªã‚“ã§ãŠãã¹ãファイルã®ä¸€è¦§ã§ ã™- diff --git a/Documentation/translations/ko_KR/howto.rst b/Documentation/translations/ko_KR/howto.rst index 3b0c15b277e0..2333697251dd 100644 --- a/Documentation/translations/ko_KR/howto.rst +++ b/Documentation/translations/ko_KR/howto.rst @@ -289,8 +289,8 @@ pub/linux/kernel/v4.x/ ë””ë ‰í† ë¦¬ì—ì„œ 참조ë 수 있다.개발 프로세ì Andrew Mortonì˜ ê¸€ì´ ìžˆë‹¤. *"커ë„ì´ ì–¸ì œ ë°°í¬ë 지는 ì•„ë¬´ë„ ëª¨ë¥¸ë‹¤. 왜ëƒí•˜ë©´ ë°°í¬ëŠ” ì•Œë ¤ì§„ - ë²„ê·¸ì˜ ìƒí™©ì— ë”°ë¼ ë°°í¬ë˜ëŠ” 것ì´ì§€ ë¯¸ë¦¬ì •í•´ ë†“ì€ ì‹œê°„ì— ë”°ë¼ - ë°°í¬ë˜ëŠ” ê²ƒì€ ì•„ë‹ˆê¸° 때문ì´ë‹¤."* + ë²„ê·¸ì˜ ìƒí™©ì— ë”°ë¼ ë°°í¬ë˜ëŠ” 것ì´ì§€ ë¯¸ë¦¬ì •í•´ ë†“ì€ ì‹œê°„ì— ë”°ë¼ + ë°°í¬ë˜ëŠ” ê²ƒì€ ì•„ë‹ˆê¸° 때문ì´ë‹¤."* 4.x.y - ì•ˆì • ì»¤ë„ íŠ¸ë¦¬ ~~~~~~~~~~~~~~~~~~~~~~ diff --git a/Documentation/translations/zh_CN/CodingStyle b/Documentation/translations/zh_CN/CodingStyle deleted file mode 100644 index dc101f48e713..000000000000 --- a/Documentation/translations/zh_CN/CodingStyle +++ /dev/null @@ -1,813 +0,0 @@ -Chinese translated version of Documentation/process/coding-style.rst - -If you have any comment or update to the content, please post to LKML directly. -However, if you have problem communicating in English you can also ask the -Chinese maintainer for help. Contact the Chinese maintainer, if this -translation is outdated or there is problem with translation. - -Chinese maintainer: Zhang Le <r0bertz@gentoo.org> ---------------------------------------------------------------------- -Documentation/process/coding-style.rstçš„ä¸æ–‡ç¿»è¯‘ - -如果想评论或更新本文的内容，请直接å‘信到LKMLã€‚å¦‚æžœä½ ä½¿ç”¨è‹±æ–‡äº¤æµæœ‰å›°éš¾çš„è¯ï¼Œä¹Ÿå¯ -以å‘ä¸æ–‡ç‰ˆç»´æŠ¤è€…求助。如果本翻译更新ä¸åŠæ—¶æˆ–者翻译å˜åœ¨é—®é¢˜ï¼Œè¯·è”ç³»ä¸æ–‡ç‰ˆç»´æŠ¤è€…。 - -ä¸æ–‡ç‰ˆç»´æŠ¤è€…： å¼ ä¹ Zhang Le <r0bertz@gentoo.org> -ä¸æ–‡ç‰ˆç¿»è¯‘者： å¼ ä¹ Zhang Le <r0bertz@gentoo.org> -ä¸æ–‡ç‰ˆæ ¡è¯‘者： çŽ‹èª Wang Cong <xiyou.wangcong@gmail.com> - wheelz <kernel.zeng@gmail.com> - 管æ—东 Xudong Guan <xudong.guan@gmail.com> - Li Zefan <lizf@cn.fujitsu.com> - Wang Chen <wangchen@cn.fujitsu.com> -以下为æ£æ–‡ ---------------------------------------------------------------------- - - Linuxå†…æ ¸ä»£ç é£Žæ ¼ - -这是一个简çŸçš„文档，æ述了 linux å†…æ ¸çš„é¦–é€‰ä»£ç é£Žæ ¼ã€‚ä»£ç é£Žæ ¼æ˜¯å› äººè€Œå¼‚çš„ï¼Œè€Œä¸”æˆ‘ -ä¸æ„¿æ„æŠŠè‡ªå·±çš„è§‚ç‚¹å¼ºåŠ ç»™ä»»ä½•äººï¼Œä½†è¿™å°±åƒæˆ‘去åšä»»ä½•äº‹æƒ…都必须éµå¾ªçš„åŽŸåˆ™é‚£æ ·ï¼Œæˆ‘ä¹Ÿ -希望在ç»å¤§å¤šæ•°äº‹ä¸Šä¿æŒè¿™ç§çš„æ€åº¦ã€‚请（在写代ç 时）至少考虑一下这里的代ç é£Žæ ¼ã€‚ - -é¦–å…ˆï¼Œæˆ‘å»ºè®®ä½ æ‰“å°ä¸€ä»½ GNU 代ç 规范，然åŽä¸è¦è¯»ã€‚烧了它，这是一个具有é‡å¤§è±¡å¾æ€§æ„义 -的动作。 - -ä¸ç®¡æ€Žæ ·ï¼ŒçŽ°åœ¨æˆ‘们开始： - - - ç¬¬ä¸€ç« ï¼šç¼©è¿› - -制表符是 8 个å—符，所以缩进也是 8 个å—符。有些异端è¿åŠ¨è¯•å›¾å°†ç¼©è¿›å˜ä¸º 4（甚至 2ï¼ï¼‰ -个å—ç¬¦æ·±ï¼Œè¿™å‡ ä¹Žç›¸å½“äºŽå°è¯•å°†åœ†å‘¨çŽ‡çš„值定义为 3。 - -ç†ç”±ï¼šç¼©è¿›çš„全部æ„义就在于清楚的定义一个控制å—èµ·æ¢äºŽä½•å¤„ã€‚å°¤å…¶æ˜¯å½“ä½ ç›¯ç€ä½ çš„å±å¹• -è¿žç»çœ‹äº† 20 å°æ—¶ä¹‹åŽï¼Œä½ 将会å‘çŽ°å¤§ä¸€ç‚¹çš„ç¼©è¿›ä¼šä½¿ä½ æ›´å®¹æ˜“åˆ†è¾¨ç¼©è¿›ã€‚ - -现在，有些人会抱怨 8 个å—符的缩进会使代ç å‘å³è¾¹ç§»åŠ¨çš„太远，在 80 个å—符的终端å±å¹•ä¸Š -å°±å¾ˆéš¾è¯»è¿™æ ·çš„ä»£ç 。这个问题的ç”æ¡ˆæ˜¯ï¼Œå¦‚æžœä½ éœ€è¦ 3 级以上的缩进，ä¸ç®¡ç”¨ä½•ç§æ–¹å¼ä½ -的代ç å·²ç»æœ‰é—®é¢˜äº†ï¼Œåº”该修æ£ä½ 的程åºã€‚ - -简而言之，8 个å—符的缩进å¯ä»¥è®©ä»£ç æ›´å®¹æ˜“é˜…è¯»ï¼Œè¿˜æœ‰ä¸€ä¸ªå¥½å¤„æ˜¯å½“ä½ çš„å‡½æ•°åµŒå¥—å¤ªæ·±çš„ -时候å¯ä»¥ç»™ä½ è¦å‘Šã€‚留心这个è¦å‘Šã€‚ - -在 switch è¯å¥ä¸æ¶ˆé™¤å¤šçº§ç¼©è¿›çš„首选的方å¼æ˜¯è®© “switch†和从属于它的 “caseâ€ æ ‡ç¾ -对é½äºŽåŒä¸€åˆ—，而ä¸è¦ “两次缩进†“caseâ€ æ ‡ç¾ã€‚比如： - - switch (suffix) { - case 'G': - case 'g': - mem <<= 30; - break; - case 'M': - case 'm': - mem <<= 20; - break; - case 'K': - case 'k': - mem <<= 10; - /* fall through */ - default: - break; - } - -ä¸è¦æŠŠå¤šä¸ªè¯å¥æ”¾åœ¨ä¸€è¡Œé‡Œï¼Œé™¤éžä½ 有什么东西è¦éšè—： - - if (condition) do_this; - do_something_everytime; - -也ä¸è¦åœ¨ä¸€è¡Œé‡Œæ”¾å¤šä¸ªèµ‹å€¼è¯å¥ã€‚å†…æ ¸ä»£ç é£Žæ ¼è¶…çº§ç®€å•ã€‚就是é¿å…å¯èƒ½å¯¼è‡´åˆ«äººè¯¯è¯»çš„表 -è¾¾å¼ã€‚ - -除了注释ã€æ–‡æ¡£å’Œ Kconfig 之外，ä¸è¦ä½¿ç”¨ç©ºæ ¼æ¥ç¼©è¿›ï¼Œå‰é¢çš„例å是例外，是有æ„为之。 - -选用一个好的编辑器，ä¸è¦åœ¨è¡Œå°¾ç•™ç©ºæ ¼ã€‚ - - - ç¬¬äºŒç« ï¼šæŠŠé•¿çš„è¡Œå’Œå—符串打散 - -代ç é£Žæ ¼çš„æ„义就在于使用平常使用的工具æ¥ç»´æŒä»£ç çš„å¯è¯»æ€§å’Œå¯ç»´æŠ¤æ€§ã€‚ - -æ¯ä¸€è¡Œçš„长度的é™åˆ¶æ˜¯ 80 列，我们强烈建议您éµå®ˆè¿™ä¸ªæƒ¯ä¾‹ã€‚ - -长于 80 列的è¯å¥è¦æ‰“æ•£æˆæœ‰æ„义的片段。除éžè¶…过 80 åˆ—èƒ½æ˜¾è‘—å¢žåŠ å¯è¯»æ€§ï¼Œå¹¶ä¸”ä¸ä¼šéšè— -ä¿¡æ¯ã€‚å片段è¦æ˜Žæ˜¾çŸäºŽæ¯ç‰‡æ®µï¼Œå¹¶æ˜Žæ˜¾é å³ã€‚è¿™åŒæ ·é€‚用于有ç€å¾ˆé•¿å‚数列表的函数头。 -然而，ç»å¯¹ä¸è¦æ‰“散对用户å¯è§çš„å—符串，例如 printk ä¿¡æ¯ï¼Œå› ä¸ºè¿™å°†å¯¼è‡´æ— æ³• grep 这些 -ä¿¡æ¯ã€‚ - - ç¬¬ä¸‰ç« ï¼šå¤§æ‹¬å·å’Œç©ºæ ¼çš„放置 - -Cè¯è¨€é£Žæ ¼ä¸å¦å¤–一个常è§é—®é¢˜æ˜¯å¤§æ‹¬å·çš„放置。和缩进大å°ä¸åŒï¼Œé€‰æ‹©æˆ–弃用æŸç§æ”¾ç½®ç– -ç•¥å¹¶æ²¡æœ‰å¤šå°‘æŠ€æœ¯ä¸Šçš„åŽŸå› ï¼Œä¸è¿‡é¦–选的方å¼ï¼Œå°±åƒ Kernighan å’Œ Ritchie 展示给我们的， -是把起始大括å·æ”¾åœ¨è¡Œå°¾ï¼Œè€ŒæŠŠç»“æŸå¤§æ‹¬å·æ”¾åœ¨è¡Œé¦–，所以： - - if (x is true) { - we do y - } - -这适用于所有的éžå‡½æ•°è¯å¥å—（ifã€switchã€forã€whileã€do）。比如： - - switch (action) { - case KOBJ_ADD: - return "add"; - case KOBJ_REMOVE: - return "remove"; - case KOBJ_CHANGE: - return "change"; - default: - return NULL; - } - -ä¸è¿‡ï¼Œæœ‰ä¸€ä¸ªä¾‹å¤–，那就是函数：函数的起始大括å·æ”¾ç½®äºŽä¸‹ä¸€è¡Œçš„开头，所以： - - int function(int x) - { - body of function - } - -全世界的异端å¯èƒ½ä¼šæŠ±æ€¨è¿™ä¸ªä¸ä¸€è‡´æ€§æ˜¯â€¦â€¦å‘ƒâ€¦â€¦ä¸ä¸€è‡´çš„，ä¸è¿‡æ‰€æœ‰æ€ç»´å¥å…¨çš„äººéƒ½çŸ¥é“ -(a) K&R 是 _æ£ç¡®çš„_，并且 (b) K&R 是æ£ç¡®çš„。æ¤å¤–，ä¸ç®¡æ€Žæ ·å‡½æ•°éƒ½æ˜¯ç‰¹æ®Šçš„（C -函数是ä¸èƒ½åµŒå¥—的）。 - -注æ„结æŸå¤§æ‹¬å·ç‹¬è‡ªå æ®ä¸€è¡Œï¼Œé™¤éžå®ƒåŽé¢è·Ÿç€åŒä¸€ä¸ªè¯å¥çš„剩余部分，也就是 do è¯å¥ä¸çš„ -“while†或者 if è¯å¥ä¸çš„ “elseâ€ï¼Œåƒè¿™æ ·ï¼š - - do { - body of do-loop - } while (condition); - -å’Œ - - if (x == y) { - .. - } else if (x > y) { - ... - } else { - .... - } - -ç†ç”±ï¼šK&R。 - -也请注æ„è¿™ç§å¤§æ‹¬å·çš„放置方å¼ä¹Ÿèƒ½ä½¿ç©ºï¼ˆæˆ–者差ä¸å¤šç©ºçš„）行的数é‡æœ€å°åŒ–，åŒæ—¶ä¸å¤±å¯ -è¯»æ€§ã€‚å› æ¤ï¼Œç”±äºŽä½ çš„å±å¹•ä¸Šçš„新行是ä¸å¯å†ç”Ÿèµ„æºï¼ˆæƒ³æƒ³ 25 行的终端å±å¹•ï¼‰ï¼Œä½ 将会有更 -多的空行æ¥æ”¾ç½®æ³¨é‡Šã€‚ - -当åªæœ‰ä¸€ä¸ªå•ç‹¬çš„è¯å¥çš„时候，ä¸ç”¨åŠ ä¸å¿…è¦çš„大括å·ã€‚ - - if (condition) - action(); - -å’Œ - - if (condition) - do_this(); - else - do_that(); - -这并ä¸é€‚用于åªæœ‰ä¸€ä¸ªæ¡ä»¶åˆ†æ”¯æ˜¯å•è¯å¥çš„情况；这时所有分支都è¦ä½¿ç”¨å¤§æ‹¬å·ï¼š - - if (condition) { - do_this(); - do_that(); - } else { - otherwise(); - } - - 3.1ï¼šç©ºæ ¼ - -Linux å†…æ ¸çš„ç©ºæ ¼ä½¿ç”¨æ–¹å¼ï¼ˆä¸»è¦ï¼‰å–决于它是用于函数还是关键å—。（大多数）关键å—åŽ -è¦åŠ ä¸€ä¸ªç©ºæ ¼ã€‚å€¼å¾—æ³¨æ„的例外是 sizeofã€typeofã€alignof å’Œ __attribute__，这些 -关键å—æŸäº›ç¨‹åº¦ä¸Šçœ‹èµ·æ¥æ›´åƒå‡½æ•°ï¼ˆå®ƒä»¬åœ¨ Linux 里也常常伴éšå°æ‹¬å·è€Œä½¿ç”¨ï¼Œå°½ç®¡åœ¨ C 里 -è¿™æ ·çš„å°æ‹¬å·ä¸æ˜¯å¿…éœ€çš„ï¼Œå°±åƒ â€œstruct fileinfo info†声明过åŽçš„ “sizeof infoâ€ï¼‰ã€‚ - -所以在这些关键å—之åŽæ”¾ä¸€ä¸ªç©ºæ ¼ï¼š - - if, switch, case, for, do, while - -但是ä¸è¦åœ¨ sizeofã€typeofã€alignof 或者 __attribute__ 这些关键å—之åŽæ”¾ç©ºæ ¼ã€‚例如， - - s = sizeof(struct file); - -ä¸è¦åœ¨å°æ‹¬å·é‡Œçš„表达å¼ä¸¤ä¾§åŠ ç©ºæ ¼ã€‚è¿™æ˜¯ä¸€ä¸ªå例： - - s = sizeof( struct file ); - -当声明指针类型或者返回指针类型的函数时，“*†的首选使用方å¼æ˜¯ä½¿ä¹‹é è¿‘å˜é‡å或者函 -æ•°å，而ä¸æ˜¯é 近类型å。例å： - - char *linux_banner; - unsigned long long memparse(char *ptr, char **retptr); - char *match_strdup(substring_t *s); - -在大多数二元和三元æ“ä½œç¬¦ä¸¤ä¾§ä½¿ç”¨ä¸€ä¸ªç©ºæ ¼ï¼Œä¾‹å¦‚ä¸‹é¢æ‰€æœ‰è¿™äº›æ“作符： - - = + - < > * / % | & ^ <= >= == != ? : - -但是一元æ“作符åŽä¸è¦åŠ ç©ºæ ¼ï¼š - - & * + - ~ ! sizeof typeof alignof __attribute__ defined - -åŽç¼€è‡ªåŠ 和自å‡ä¸€å…ƒæ“作符å‰ä¸åŠ ç©ºæ ¼ï¼š - - ++ -- - -å‰ç¼€è‡ªåŠ 和自å‡ä¸€å…ƒæ“作符åŽä¸åŠ ç©ºæ ¼ï¼š - - ++ -- - -‘.’ å’Œ “->†结构体æˆå‘˜æ“作符å‰åŽä¸åŠ ç©ºæ ¼ã€‚ - -ä¸è¦åœ¨è¡Œå°¾ç•™ç©ºç™½ã€‚有些å¯ä»¥è‡ªåŠ¨ç¼©è¿›çš„ç¼–è¾‘å™¨ä¼šåœ¨æ–°è¡Œçš„è¡Œé¦–åŠ å…¥é€‚é‡çš„空白，然åŽä½ -å°±å¯ä»¥ç›´æŽ¥åœ¨é‚£ä¸€è¡Œè¾“入代ç 。ä¸è¿‡å‡å¦‚ä½ æœ€åŽæ²¡æœ‰åœ¨é‚£ä¸€è¡Œè¾“入代ç ï¼Œæœ‰äº›ç¼–è¾‘å™¨å°±ä¸ -会移除已ç»åŠ 入的空白，就åƒä½ æ•…æ„留下一个åªæœ‰ç©ºç™½çš„行。包å«è¡Œå°¾ç©ºç™½çš„è¡Œå°±è¿™æ ·äº§ -生了。 - -当gitå‘现补ä¸åŒ…å«äº†è¡Œå°¾ç©ºç™½çš„时候会è¦å‘Šä½ ，并且å¯ä»¥åº”ä½ çš„è¦æ±‚去掉行尾空白；ä¸è¿‡ -å¦‚æžœä½ æ˜¯æ£åœ¨æ‰“一系列补ä¸ï¼Œè¿™æ ·åšä¼šå¯¼è‡´åŽé¢çš„è¡¥ä¸å¤±è´¥ï¼Œå› ä¸ºä½ æ”¹å˜äº†è¡¥ä¸çš„上下文。 - - - ç¬¬å››ç« ï¼šå‘½å - -C是一个简朴的è¯è¨€ï¼Œä½ 的命åä¹Ÿåº”è¯¥è¿™æ ·ã€‚å’Œ Modula-2 å’Œ Pascal 程åºå‘˜ä¸åŒï¼ŒC 程åºå‘˜ -ä¸ä½¿ç”¨ç±»ä¼¼ ThisVariableIsATemporaryCounter è¿™æ ·åŽä¸½çš„åå—。C 程åºå‘˜ä¼šç§°é‚£ä¸ªå˜é‡ -为 “tmpâ€ï¼Œè¿™æ ·å†™èµ·æ¥ä¼šæ›´å®¹æ˜“，而且至少ä¸ä¼šä»¤å…¶éš¾äºŽç†è§£ã€‚ - -ä¸è¿‡ï¼Œè™½ç„¶æ··ç”¨å¤§å°å†™çš„åå—是ä¸æ倡使用的，但是全局å˜é‡è¿˜æ˜¯éœ€è¦ä¸€ä¸ªå…·æ述性的åå— -。称一个全局函数为 “foo†是一个难以饶æ•çš„错误。 - -全局å˜é‡ï¼ˆåªæœ‰å½“ä½ çœŸæ£éœ€è¦å®ƒä»¬çš„时候å†ç”¨å®ƒï¼‰éœ€è¦æœ‰ä¸€ä¸ªå…·æ述性的åå—，就åƒå…¨å±€å‡½ -æ•°ã€‚å¦‚æžœä½ æœ‰ä¸€ä¸ªå¯ä»¥è®¡ç®—活动用户数é‡çš„å‡½æ•°ï¼Œä½ åº”è¯¥å«å®ƒ “count_active_users()†-或者类似的åå—ï¼Œä½ ä¸åº”该å«å®ƒ “cntuser()â€ã€‚ - -在函数åä¸åŒ…å«å‡½æ•°ç±»åž‹ï¼ˆæ‰€è°“的匈牙利命å法）是脑å出了问题——编译器知é“那些类型而 -ä¸”èƒ½å¤Ÿæ£€æŸ¥é‚£äº›ç±»åž‹ï¼Œè¿™æ ·åšåªèƒ½æŠŠç¨‹åºå‘˜å¼„ç³Šæ¶‚äº†ã€‚éš¾æ€ªå¾®è½¯æ€»æ˜¯åˆ¶é€ å‡ºæœ‰é—®é¢˜çš„ç¨‹åºã€‚ - -本地å˜é‡å应该简çŸï¼Œè€Œä¸”能够表达相关的å«ä¹‰ã€‚å¦‚æžœä½ æœ‰ä¸€äº›éšæœºçš„整数型的循环计数器 -，它应该被称为 “iâ€ã€‚å«å®ƒ “loop_counterâ€ å¹¶æ— ç›Šå¤„ï¼Œå¦‚æžœå®ƒæ²¡æœ‰è¢«è¯¯è§£çš„å¯èƒ½çš„è¯ã€‚ -类似的，“tmp†å¯ä»¥ç”¨æ¥ç§°å‘¼ä»»æ„类型的临时å˜é‡ã€‚ - -å¦‚æžœä½ æ€•æ··æ·†äº†ä½ çš„æœ¬åœ°å˜é‡åï¼Œä½ å°±é‡åˆ°å¦ä¸€ä¸ªé—®é¢˜äº†ï¼Œå«åšå‡½æ•°å¢žé•¿è·å°”蒙失衡综åˆç—‡ -。请看第å…ç« ï¼ˆå‡½æ•°ï¼‰ã€‚ - - - ç¬¬äº”ç« ï¼šTypedef - -ä¸è¦ä½¿ç”¨ç±»ä¼¼ “vps_t†之类的东西。 - -对结构体和指针使用 typedef æ˜¯ä¸€ä¸ªé”™è¯¯ã€‚å½“ä½ åœ¨ä»£ç 里看到： - - vps_t a; - -这代表什么æ„æ€å‘¢ï¼Ÿ - -相åï¼Œå¦‚æžœæ˜¯è¿™æ · - - struct virtual_container *a; - -ä½ å°±çŸ¥é“ â€œa†是什么了。 - -很多人认为 typedef “能æ高å¯è¯»æ€§â€ã€‚实际ä¸æ˜¯è¿™æ ·çš„。它们åªåœ¨ä¸‹åˆ—情况下有用： - - (a) 完全ä¸é€æ˜Žçš„对象（这ç§æƒ…况下è¦ä¸»åŠ¨ä½¿ç”¨ typedef æ¥éšè—这个对象实际上是什么）。 - - 例如：“pte_t†ç‰ä¸é€æ˜Žå¯¹è±¡ï¼Œä½ åªèƒ½ç”¨åˆé€‚的访问函数æ¥è®¿é—®å®ƒä»¬ã€‚ - - 注æ„ï¼ä¸é€æ˜Žæ€§å’Œâ€œè®¿é—®å‡½æ•°â€æœ¬èº«æ˜¯ä¸å¥½çš„。我们使用 pte_t ç‰ç±»åž‹çš„åŽŸå› åœ¨äºŽçœŸçš„æ˜¯ - 完全没有任何共用的å¯è®¿é—®ä¿¡æ¯ã€‚ - - (b) 清楚的整数类型，如æ¤ï¼Œè¿™å±‚抽象就å¯ä»¥å¸®åŠ©æ¶ˆé™¤åˆ°åº•æ˜¯ “int†还是 “long†的混淆。 - - u8/u16/u32 是完全没有问题的 typedef，ä¸è¿‡å®ƒä»¬æ›´ç¬¦åˆç±»åˆ« (d) 而ä¸æ˜¯è¿™é‡Œã€‚ - - å†æ¬¡æ³¨æ„ï¼è¦è¿™æ ·åšï¼Œå¿…é¡»äº‹å‡ºæœ‰å› ã€‚å¦‚æžœæŸä¸ªå˜é‡æ˜¯ “unsigned longâ€œï¼Œé‚£ä¹ˆæ²¡æœ‰å¿…è¦ - - typedef unsigned long myflags_t; - - ä¸è¿‡å¦‚æžœæœ‰ä¸€ä¸ªæ˜Žç¡®çš„åŽŸå› ï¼Œæ¯”å¦‚å®ƒåœ¨æŸç§æƒ…况下å¯èƒ½ä¼šæ˜¯ä¸€ä¸ª “unsigned int†而在 - 其他情况下å¯èƒ½ä¸º “unsigned longâ€ï¼Œé‚£ä¹ˆå°±ä¸è¦çŠ¹è±«ï¼Œè¯·åŠ¡å¿…使用 typedef。 - - (c) å½“ä½ ä½¿ç”¨sparse按å—é¢çš„创建一个新类型æ¥åšç±»åž‹æ£€æŸ¥çš„时候。 - - (d) å’Œæ ‡å‡†C99类型相åŒçš„类型，在æŸäº›ä¾‹å¤–的情况下。 - - 虽然让眼ç›å’Œè„‘ç‹æ¥é€‚åº”æ–°çš„æ ‡å‡†ç±»åž‹æ¯”å¦‚ “uint32_t†ä¸éœ€è¦èŠ±å¾ˆå¤šæ—¶é—´ï¼Œå¯æ˜¯æœ‰äº› - 人ä»ç„¶æ‹’ç»ä½¿ç”¨å®ƒä»¬ã€‚ - - å› æ¤ï¼ŒLinux 特有的ç‰åŒäºŽæ ‡å‡†ç±»åž‹çš„ “u8/u16/u32/u64†类型和它们的有符å·ç±»åž‹æ˜¯è¢« - å…è®¸çš„â€”â€”å°½ç®¡åœ¨ä½ è‡ªå·±çš„æ–°ä»£ç ä¸ï¼Œå®ƒä»¬ä¸æ˜¯å¼ºåˆ¶è¦æ±‚è¦ä½¿ç”¨çš„。 - - 当编辑已ç»ä½¿ç”¨äº†æŸä¸ªç±»åž‹é›†çš„已有代ç æ—¶ï¼Œä½ åº”è¯¥éµå¾ªé‚£äº›ä»£ç ä¸å·²ç»åšå‡ºçš„选择。 - - (e) å¯ä»¥åœ¨ç”¨æˆ·ç©ºé—´å®‰å…¨ä½¿ç”¨çš„类型。 - - 在æŸäº›ç”¨æˆ·ç©ºé—´å¯è§çš„结构体里，我们ä¸èƒ½è¦æ±‚C99类型而且ä¸èƒ½ç”¨ä¸Šé¢æ到的 “u32†- ç±»åž‹ã€‚å› æ¤ï¼Œæˆ‘们在与用户空间共享的所有结构体ä¸ä½¿ç”¨ __u32 和类似的类型。 - -å¯èƒ½è¿˜æœ‰å…¶ä»–的情况，ä¸è¿‡åŸºæœ¬çš„规则是永远ä¸è¦ä½¿ç”¨ typedef，除éžä½ å¯ä»¥æ˜Žç¡®çš„应用上 -è¿°æŸä¸ªè§„则ä¸çš„一个。 - -总的æ¥è¯´ï¼Œå¦‚æžœä¸€ä¸ªæŒ‡é’ˆæˆ–è€…ä¸€ä¸ªç»“æž„ä½“é‡Œçš„å…ƒç´ å¯ä»¥åˆç†çš„è¢«ç›´æŽ¥è®¿é—®åˆ°ï¼Œé‚£ä¹ˆå®ƒä»¬å°±ä¸ -应该是一个 typedef。 - - - 第å…ç« ï¼šå‡½æ•° - -函数应该简çŸè€Œæ¼‚亮，并且åªå®Œæˆä¸€ä»¶äº‹æƒ…。函数应该å¯ä»¥ä¸€å±æˆ–者两å±æ˜¾ç¤ºå®Œï¼ˆæˆ‘们都知 -é“ ISO/ANSI å±å¹•å¤§å°æ˜¯ 80x24），åªåšä¸€ä»¶äº‹æƒ…，而且把它åšå¥½ã€‚ - -一个函数的最大长度是和该函数的å¤æ‚度和缩进级数æˆåæ¯”çš„ã€‚æ‰€ä»¥ï¼Œå¦‚æžœä½ æœ‰ä¸€ä¸ªç†è®ºä¸Š -很简å•çš„åªæœ‰ä¸€ä¸ªå¾ˆé•¿ï¼ˆä½†æ˜¯ç®€å•ï¼‰çš„ case è¯å¥çš„å‡½æ•°ï¼Œè€Œä¸”ä½ éœ€è¦åœ¨æ¯ä¸ª case é‡Œåš -很多很å°çš„äº‹æƒ…ï¼Œè¿™æ ·çš„å‡½æ•°å°½ç®¡å¾ˆé•¿ï¼Œä½†ä¹Ÿæ˜¯å¯ä»¥çš„。 - -ä¸è¿‡ï¼Œå¦‚æžœä½ æœ‰ä¸€ä¸ªå¤æ‚çš„å‡½æ•°ï¼Œè€Œä¸”ä½ æ€€ç–‘ä¸€ä¸ªå¤©åˆ†ä¸æ˜¯å¾ˆé«˜çš„高ä¸ä¸€å¹´çº§å¦ç”Ÿå¯èƒ½ç”šè‡³ -æžä¸æ¸…æ¥šè¿™ä¸ªå‡½æ•°çš„ç›®çš„ï¼Œä½ åº”è¯¥ä¸¥æ ¼çš„éµå®ˆå‰é¢æ到的长度é™åˆ¶ã€‚使用辅助函数，并为之 -å–个具æ述性的åå—ï¼ˆå¦‚æžœä½ è§‰å¾—å®ƒä»¬çš„æ€§èƒ½å¾ˆé‡è¦çš„è¯ï¼Œå¯ä»¥è®©ç¼–译器内è”å®ƒä»¬ï¼Œè¿™æ ·çš„ -æ•ˆæžœå¾€å¾€ä¼šæ¯”ä½ å†™ä¸€ä¸ªå¤æ‚函数的效果è¦å¥½ã€‚） - -函数的å¦å¤–一个衡é‡æ ‡å‡†æ˜¯æœ¬åœ°å˜é‡çš„æ•°é‡ã€‚æ¤æ•°é‡ä¸åº”超过 5ï¼10 个，å¦åˆ™ä½ 的函数就有 -问题了。é‡æ–°è€ƒè™‘ä¸€ä¸‹ä½ çš„å‡½æ•°ï¼ŒæŠŠå®ƒåˆ†æ‹†æˆæ›´å°çš„函数。人的大脑一般å¯ä»¥è½»æ¾çš„åŒæ—¶è·Ÿ -踪 7 个ä¸åŒçš„事物，如果å†å¢žå¤šçš„è¯ï¼Œå°±ä¼šç³Šæ¶‚了。å³ä¾¿ä½ èªé¢–è¿‡äººï¼Œä½ ä¹Ÿå¯èƒ½ä¼šè®°ä¸æ¸…ä½ -2 个星期å‰åšè¿‡çš„事情。 - -在æºæ–‡ä»¶é‡Œï¼Œä½¿ç”¨ç©ºè¡Œéš”å¼€ä¸åŒçš„函数。如果该函数需è¦è¢«å¯¼å‡ºï¼Œå®ƒçš„ EXPORT* å®åº”该紧贴 -在它的结æŸå¤§æ‹¬å·ä¹‹ä¸‹ã€‚比如： - - int system_is_up(void) - { - return system_state == SYSTEM_RUNNING; - } - EXPORT_SYMBOL(system_is_up); - -在函数原型ä¸ï¼ŒåŒ…å«å‡½æ•°å和它们的数æ®ç±»åž‹ã€‚虽然Cè¯è¨€é‡Œæ²¡æœ‰è¿™æ ·çš„è¦æ±‚，在 Linux 里这 -是æ倡的åšæ³•ï¼Œå› ä¸ºè¿™æ ·å¯ä»¥å¾ˆç®€å•çš„给读者æ供更多的有价值的信æ¯ã€‚ - - - ç¬¬ä¸ƒç« ï¼šé›†ä¸çš„函数退出途径 - -虽然被æŸäº›äººå£°ç§°å·²ç»è¿‡æ—¶ï¼Œä½†æ˜¯ goto è¯å¥çš„ç‰ä»·ç‰©è¿˜æ˜¯ç»å¸¸è¢«ç¼–译器所使用，具体形å¼æ˜¯ -æ— æ¡ä»¶è·³è½¬æŒ‡ä»¤ã€‚ - -当一个函数从多个ä½ç½®é€€å‡ºï¼Œå¹¶ä¸”需è¦åšä¸€äº›ç±»ä¼¼æ¸…ç†çš„常è§æ“作时，goto è¯å¥å°±å¾ˆæ–¹ä¾¿äº†ã€‚ -如果并ä¸éœ€è¦æ¸…ç†æ“作，那么直接 return å³å¯ã€‚ - -ç†ç”±æ˜¯ï¼š - -- æ— æ¡ä»¶è¯å¥å®¹æ˜“ç†è§£å’Œè·Ÿè¸ª -- 嵌套程度å‡å° -- å¯ä»¥é¿å…由于修改时忘记更新æŸä¸ªå•ç‹¬çš„退出点而导致的错误 -- å‡è½»äº†ç¼–è¯‘å™¨çš„å·¥ä½œï¼Œæ— éœ€åˆ é™¤å†—ä½™ä»£ç ;) - - int fun(int a) - { - int result = 0; - char *buffer; - - buffer = kmalloc(SIZE, GFP_KERNEL); - if (!buffer) - return -ENOMEM; - - if (condition1) { - while (loop1) { - ... - } - result = 1; - goto out_buffer; - } - ... - out_buffer: - kfree(buffer); - return result; - } - -一个需è¦æ³¨æ„的常è§é”™è¯¯æ˜¯â€œä¸€ä¸ª err 错误â€ï¼Œå°±åƒè¿™æ ·ï¼š - - err: - kfree(foo->bar); - kfree(foo); - return ret; - -这段代ç 的错误是，在æŸäº›é€€å‡ºè·¯å¾„上 “foo†是 NULL。通常情况下，通过把它分离æˆä¸¤ä¸ª -é”™è¯¯æ ‡ç¾ â€œerr_bar:†和 “err_foo:†æ¥ä¿®å¤è¿™ä¸ªé”™è¯¯ã€‚ - - ç¬¬å…«ç« ï¼šæ³¨é‡Š - -注释是好的，ä¸è¿‡æœ‰è¿‡åº¦æ³¨é‡Šçš„å±é™©ã€‚永远ä¸è¦åœ¨æ³¨é‡Šé‡Œè§£é‡Šä½ 的代ç 是如何è¿ä½œçš„：更好 -çš„åšæ³•æ˜¯è®©åˆ«äººä¸€çœ‹ä½ 的代ç å°±å¯ä»¥æ˜Žç™½ï¼Œè§£é‡Šå†™çš„很差的代ç 是浪费时间。 - -ä¸€èˆ¬çš„ï¼Œä½ æƒ³è¦ä½ çš„æ³¨é‡Šå‘Šè¯‰åˆ«äººä½ çš„ä»£ç åšäº†ä»€ä¹ˆï¼Œè€Œä¸æ˜¯æ€Žä¹ˆåšçš„ã€‚ä¹Ÿè¯·ä½ ä¸è¦æŠŠæ³¨é‡Š -放在一个函数体内部：如果函数å¤æ‚åˆ°ä½ éœ€è¦ç‹¬ç«‹çš„注释其ä¸çš„ä¸€éƒ¨åˆ†ï¼Œä½ å¾ˆå¯èƒ½éœ€è¦å›žåˆ° -第å…ç« çœ‹ä¸€çœ‹ã€‚ä½ å¯ä»¥åšä¸€äº›å°æ³¨é‡Šæ¥æ³¨æ˜Žæˆ–è¦å‘ŠæŸäº›å¾ˆèªæ˜Žï¼ˆæˆ–者槽糕）的åšæ³•ï¼Œä½†ä¸è¦ -åŠ å¤ªå¤šã€‚ä½ åº”è¯¥åšçš„，是把注释放在函数的头部，告诉人们它åšäº†ä»€ä¹ˆï¼Œä¹Ÿå¯ä»¥åŠ 上它åšè¿™ -äº›äº‹æƒ…çš„åŽŸå› ã€‚ - -å½“æ³¨é‡Šå†…æ ¸API函数时，请使用 kernel-doc æ ¼å¼ã€‚请看 -Documentation/doc-guide/å’Œscripts/kernel-doc 以获得详细信æ¯ã€‚ - -Linuxçš„æ³¨é‡Šé£Žæ ¼æ˜¯ C89 “/* ... */â€ é£Žæ ¼ã€‚ä¸è¦ä½¿ç”¨ C99 é£Žæ ¼ “// ...†注释。 - -é•¿ï¼ˆå¤šè¡Œï¼‰çš„é¦–é€‰æ³¨é‡Šé£Žæ ¼æ˜¯ï¼š - - /* - * This is the preferred style for multi-line - * comments in the Linux kernel source code. - * Please use it consistently. - * - * Description: A column of asterisks on the left side, - * with beginning and ending almost-blank lines. - */ - -对于在 net/ å’Œ drivers/net/ çš„æ–‡ä»¶ï¼Œé¦–é€‰çš„é•¿ï¼ˆå¤šè¡Œï¼‰æ³¨é‡Šé£Žæ ¼æœ‰äº›ä¸åŒã€‚ - - /* The preferred comment style for files in net/ and drivers/net - * looks like this. - * - * It is nearly the same as the generally preferred comment style, - * but there is no initial almost-blank line. - */ - -注释数æ®ä¹Ÿæ˜¯å¾ˆé‡è¦çš„，ä¸ç®¡æ˜¯åŸºæœ¬ç±»åž‹è¿˜æ˜¯è¡ç”Ÿç±»åž‹ã€‚为了方便实现这一点，æ¯ä¸€è¡Œåº”åª -声明一个数æ®ï¼ˆä¸è¦ä½¿ç”¨é€—å·æ¥ä¸€æ¬¡å£°æ˜Žå¤šä¸ªæ•°æ®ï¼‰ã€‚è¿™æ ·ä½ å°±æœ‰ç©ºé—´æ¥ä¸ºæ¯ä¸ªæ•°æ®å†™ä¸€æ®µ -å°æ³¨é‡Šæ¥è§£é‡Šå®ƒä»¬çš„用途了。 - - - 第ä¹ç« ï¼šä½ å·²ç»æŠŠäº‹æƒ…弄糟了 - -è¿™æ²¡ä»€ä¹ˆï¼Œæˆ‘ä»¬éƒ½æ˜¯è¿™æ ·ã€‚å¯èƒ½ä½ 的使用了很长时间 Unix 的朋å‹å·²ç»å‘Šè¯‰ä½ “GNU emacs†能 -è‡ªåŠ¨å¸®ä½ æ ¼å¼åŒ– C æºä»£ç ï¼Œè€Œä¸”ä½ ä¹Ÿæ³¨æ„åˆ°äº†ï¼Œç¡®å®žæ˜¯è¿™æ ·ï¼Œä¸è¿‡å®ƒæ‰€ä½¿ç”¨çš„默认值和我们 -想è¦çš„相去甚远（实际上，甚至比éšæœºæ‰“的还è¦å·®â€”â€”æ— æ•°ä¸ªçŒ´å在 GNU emacs 里打å—æ°¸è¿œä¸ -ä¼šåˆ›é€ å‡ºä¸€ä¸ªå¥½ç¨‹åºï¼‰ï¼ˆè¯‘注：请å‚考 Infinite Monkey Theorem） - -æ‰€ä»¥ä½ è¦ä¹ˆæ”¾å¼ƒ GNU emacs，è¦ä¹ˆæ”¹å˜å®ƒè®©å®ƒä½¿ç”¨æ›´åˆç†çš„设定。è¦é‡‡ç”¨åŽä¸€ä¸ªæ–¹æ¡ˆï¼Œä½ å¯ -以把下é¢è¿™æ®µç²˜è´´åˆ°ä½ çš„ .emacs 文件里。 - -(defun c-lineup-arglist-tabs-only (ignored) - "Line up argument lists by tabs, not spaces" - (let* ((anchor (c-langelem-pos c-syntactic-element)) - (column (c-langelem-2nd-pos c-syntactic-element)) - (offset (- (1+ column) anchor)) - (steps (floor offset c-basic-offset))) - (* (max steps 1) - c-basic-offset))) - -(add-hook 'c-mode-common-hook - (lambda () - ;; Add kernel style - (c-add-style - "linux-tabs-only" - '("linux" (c-offsets-alist - (arglist-cont-nonempty - c-lineup-gcc-asm-reg - c-lineup-arglist-tabs-only)))))) - -(add-hook 'c-mode-hook - (lambda () - (let ((filename (buffer-file-name))) - ;; Enable kernel mode for the appropriate files - (when (and filename - (string-match (expand-file-name "~/src/linux-trees") - filename)) - (setq indent-tabs-mode t) - (setq show-trailing-whitespace t) - (c-set-style "linux-tabs-only"))))) - -这会让 emacs 在 ~/src/linux-trees 目录下的 C æºæ–‡ä»¶èŽ·å¾—æ›´å¥½çš„å†…æ ¸ä»£ç é£Žæ ¼ã€‚ - -ä¸è¿‡å°±ç®—ä½ å°è¯•è®© emacs æ£ç¡®çš„æ ¼å¼åŒ–代ç 失败了，也并ä¸æ„味ç€ä½ 失去了一切：还å¯ä»¥ç”¨ -“indentâ€ã€‚ - -ä¸è¿‡ï¼ŒGNU indent 也有和 GNU emacs ä¸€æ ·æœ‰é—®é¢˜çš„è®¾å®šï¼Œæ‰€ä»¥ä½ éœ€è¦ç»™å®ƒä¸€äº›å‘½ä»¤é€‰é¡¹ã€‚ä¸ -过，这还ä¸ç®—å¤ªç³Ÿç³•ï¼Œå› ä¸ºå°±ç®—æ˜¯ GNU indent çš„ä½œè€…ä¹Ÿè®¤åŒ K&R çš„æƒå¨æ€§ï¼ˆGNU 的人并ä¸æ˜¯ -å人，他们åªæ˜¯åœ¨è¿™ä¸ªé—®é¢˜ä¸Šè¢«ä¸¥é‡çš„è¯¯å¯¼äº†ï¼‰ï¼Œæ‰€ä»¥ä½ åªè¦ç»™ indent 指定选项 “-kr -i8†-（代表 “K&R，8 个å—符缩进â€ï¼‰ï¼Œæˆ–者使用 “scripts/Lindentâ€ï¼Œè¿™æ ·å°±å¯ä»¥ä»¥æœ€æ—¶é«¦çš„æ–¹å¼ -缩进æºä»£ç 。 - -“indent†有很多选项，特别是é‡æ–°æ ¼å¼åŒ–æ³¨é‡Šçš„æ—¶å€™ï¼Œä½ å¯èƒ½éœ€è¦çœ‹ä¸€ä¸‹å®ƒçš„手册页。ä¸è¿‡ -è®°ä½ï¼šâ€œindent†ä¸èƒ½ä¿®æ£åçš„ç¼–ç¨‹ä¹ æƒ¯ã€‚ - - - 第åç« ï¼šKconfig é…置文件 - -对于é布æºç æ ‘çš„æ‰€æœ‰ Kconfig* é…置文件æ¥è¯´ï¼Œå®ƒä»¬ç¼©è¿›æ–¹å¼ä¸Ž C 代ç 相比有所ä¸åŒã€‚紧挨 -在 “config†定义下é¢çš„行缩进一个制表符，帮助信æ¯åˆ™å†å¤šç¼©è¿› 2 ä¸ªç©ºæ ¼ã€‚æ¯”å¦‚ï¼š - -config AUDIT - bool "Auditing support" - depends on NET - help - Enable auditing infrastructure that can be used with another - kernel subsystem, such as SELinux (which requires this for - logging of avc messages output). Does not do system-call - auditing without CONFIG_AUDITSYSCALL. - -而那些å±é™©çš„功能（比如æŸäº›æ–‡ä»¶ç³»ç»Ÿçš„写支æŒï¼‰åº”该在它们的æ示å—符串里显著的声明这 -一点： - -config ADFS_FS_RW - bool "ADFS write support (DANGEROUS)" - depends on ADFS_FS - ... - -è¦æŸ¥çœ‹é…置文件的完整文档，请看 Documentation/kbuild/kconfig-language.txt。 - - - 第åä¸€ç« ï¼šæ•°æ®ç»“æž„ - -如果一个数æ®ç»“构，在创建和销æ¯å®ƒçš„å•çº¿æ‰§è¡ŒçŽ¯å¢ƒä¹‹å¤–å¯è§ï¼Œé‚£ä¹ˆå®ƒå¿…é¡»è¦æœ‰ä¸€ä¸ªå¼•ç”¨è®¡ -æ•°å™¨ã€‚å†…æ ¸é‡Œæ²¡æœ‰åžƒåœ¾æ”¶é›†ï¼ˆå¹¶ä¸”å†…æ ¸ä¹‹å¤–çš„åžƒåœ¾æ”¶é›†æ…¢ä¸”æ•ˆçŽ‡ä½Žä¸‹ï¼‰ï¼Œè¿™æ„味ç€ä½ ç»å¯¹éœ€ -è¦è®°å½•ä½ 对这ç§æ•°æ®ç»“构的使用情况。 - -引用计数æ„味ç€ä½ 能够é¿å…上é”，并且å…许多个用户并行访问这个数æ®ç»“构——而ä¸éœ€è¦æ‹…心 -这个数æ®ç»“æž„ä»…ä»…å› ä¸ºæš‚æ—¶ä¸è¢«ä½¿ç”¨å°±æ¶ˆå¤±äº†ï¼Œé‚£äº›ç”¨æˆ·å¯èƒ½ä¸è¿‡æ˜¯æ²‰ç¡äº†ä¸€é˜µæˆ–者åšäº†ä¸€ -些其他事情而已。 - -注æ„上é”ä¸èƒ½å–代引用计数。上é”是为了ä¿æŒæ•°æ®ç»“构的一致性，而引用计数是一个内å˜ç®¡ -ç†æŠ€å·§ã€‚通常二者都需è¦ï¼Œä¸è¦æŠŠä¸¤ä¸ªæžæ··äº†ã€‚ - -很多数æ®ç»“构实际上有2级引用计数，它们通常有ä¸åŒâ€œç±»â€çš„用户。å类计数器统计å类用 -户的数é‡ï¼Œæ¯å½“å类计数器å‡è‡³é›¶æ—¶ï¼Œå…¨å±€è®¡æ•°å™¨å‡ä¸€ã€‚ - -è¿™ç§â€œå¤šçº§å¼•ç”¨è®¡æ•°â€çš„例åå¯ä»¥åœ¨å†…å˜ç®¡ç†ï¼ˆâ€œstruct mm_structâ€ï¼šmm_users å’Œ mm_count） -和文件系统（“struct super_blockâ€ï¼šs_countå’Œs_active）ä¸æ‰¾åˆ°ã€‚ - -è®°ä½ï¼šå¦‚æžœå¦ä¸€ä¸ªæ‰§è¡Œçº¿ç´¢å¯ä»¥æ‰¾åˆ°ä½ çš„æ•°æ®ç»“构，但是这个数æ®ç»“构没有引用计数器，这 -é‡Œå‡ ä¹Žè‚¯å®šæ˜¯ä¸€ä¸ª bug。 - - - 第åäºŒç« ï¼šå®ï¼Œæžšä¸¾å’ŒRTL - -用于定义常é‡çš„å®çš„åå—åŠæžšä¸¾é‡Œçš„æ ‡ç¾éœ€è¦å¤§å†™ã€‚ - -#define CONSTANT 0x12345 - -åœ¨å®šä¹‰å‡ ä¸ªç›¸å…³çš„å¸¸é‡æ—¶ï¼Œæœ€å¥½ç”¨æžšä¸¾ã€‚ - -å®çš„åå—请用大写å—æ¯ï¼Œä¸è¿‡å½¢å¦‚函数的å®çš„åå—å¯ä»¥ç”¨å°å†™å—æ¯ã€‚ - -一般的，如果能写æˆå†…è”函数就ä¸è¦å†™æˆåƒå‡½æ•°çš„å®ã€‚ - -å«æœ‰å¤šä¸ªè¯å¥çš„å®åº”该被包å«åœ¨ä¸€ä¸ª do-while 代ç å—里： - - #define macrofun(a, b, c) \ - do { \ - if (a == 5) \ - do_this(b, c); \ - } while (0) - -使用å®çš„时候应é¿å…的事情： - -1) å½±å“控制æµç¨‹çš„å®ï¼š - - #define FOO(x) \ - do { \ - if (blah(x) < 0) \ - return -EBUGGERED; \ - } while (0) - -éžå¸¸ä¸å¥½ã€‚它看起æ¥åƒä¸€ä¸ªå‡½æ•°ï¼Œä¸è¿‡å´èƒ½å¯¼è‡´â€œè°ƒç”¨â€å®ƒçš„函数退出；ä¸è¦æ‰“乱读者大脑里 -çš„è¯æ³•åˆ†æžå™¨ã€‚ - -2) ä¾èµ–于一个固定åå—的本地å˜é‡çš„å®ï¼š - - #define FOO(val) bar(index, val) - -å¯èƒ½çœ‹èµ·æ¥åƒæ˜¯ä¸ªä¸é”™çš„东西，ä¸è¿‡å®ƒéžå¸¸å®¹æ˜“把读代ç 的人æžç³Šæ¶‚ï¼Œè€Œä¸”å®¹æ˜“å¯¼è‡´çœ‹èµ·æ¥ -ä¸ç›¸å…³çš„改动带æ¥é”™è¯¯ã€‚ - -3) 作为左值的带å‚æ•°çš„å®ï¼š FOO(x) = y；如果有人把 FOO å˜æˆä¸€ä¸ªå†…è”函数的è¯ï¼Œè¿™ç§ç”¨ -法就会出错了。 - -4) 忘记了优先级：使用表达å¼å®šä¹‰å¸¸é‡çš„å®å¿…须将表达å¼ç½®äºŽä¸€å¯¹å°æ‹¬å·ä¹‹å†…。带å‚æ•°çš„ -å®ä¹Ÿè¦æ³¨æ„æ¤ç±»é—®é¢˜ã€‚ - - #define CONSTANT 0x4000 - #define CONSTEXP (CONSTANT | 3) - -5) 在å®é‡Œå®šä¹‰ç±»ä¼¼å‡½æ•°çš„本地å˜é‡æ—¶å‘½å冲çªï¼š - - #define FOO(x) \ - ({ \ - typeof(x) ret; \ - ret = calc_ret(x); \ - (ret); \ - }) - -ret 是本地å˜é‡çš„通用åå— - __foo_ret æ›´ä¸å®¹æ˜“与一个已å˜åœ¨çš„å˜é‡å†²çªã€‚ - -cpp 手册对å®çš„讲解很详细。gcc internals 手册也详细讲解了 RTL（译注：register -transfer languageï¼‰ï¼Œå†…æ ¸é‡Œçš„æ±‡ç¼–è¯è¨€ç»å¸¸ç”¨åˆ°å®ƒã€‚ - - - 第åä¸‰ç« ï¼šæ‰“å°å†…æ ¸æ¶ˆæ¯ - -å†…æ ¸å¼€å‘者应该是å—过良好教育的。请一定注æ„å†…æ ¸ä¿¡æ¯çš„拼写，以给人以好的å°è±¡ã€‚ä¸è¦ -用ä¸è§„范的å•è¯æ¯”如 “dontâ€ï¼Œè€Œè¦ç”¨ “do notâ€æˆ–者 “don'tâ€ã€‚ä¿è¯è¿™äº›ä¿¡æ¯ç®€å•ã€æ˜Žäº†ã€ -æ— æ§ä¹‰ã€‚ - -å†…æ ¸ä¿¡æ¯ä¸å¿…以å¥å·ï¼ˆè¯‘注：英文å¥å·ï¼Œå³ç‚¹ï¼‰ç»“æŸã€‚ - -在å°æ‹¬å·é‡Œæ‰“å°æ•°å— (%d) 没有任何价值，应该é¿å…è¿™æ ·åšã€‚ - -<linux/device.h> 里有一些驱动模型诊æ–å®ï¼Œä½ 应该使用它们，以确ä¿ä¿¡æ¯å¯¹åº”于æ£ç¡®çš„ -è®¾å¤‡å’Œé©±åŠ¨ï¼Œå¹¶ä¸”è¢«æ ‡è®°äº†æ£ç¡®çš„消æ¯çº§åˆ«ã€‚这些å®æœ‰ï¼šdev_err()，dev_warn()， -dev_info() ç‰ç‰ã€‚对于那些ä¸å’ŒæŸä¸ªç‰¹å®šè®¾å¤‡ç›¸å…³è¿žçš„ä¿¡æ¯ï¼Œ<linux/printk.h> 定义了 -pr_notice()，pr_info()，pr_warn()，pr_err() 和其他。 - -写出好的调试信æ¯å¯ä»¥æ˜¯ä¸€ä¸ªå¾ˆå¤§çš„æŒ‘æˆ˜ï¼›ä¸€æ—¦ä½ å†™å‡ºåŽï¼Œè¿™äº›ä¿¡æ¯åœ¨è¿œç¨‹é™¤é”™æ—¶èƒ½æä¾›æžå¤§ -的帮助。然而打å°è°ƒè¯•ä¿¡æ¯çš„处ç†æ–¹å¼åŒæ‰“å°éžè°ƒè¯•ä¿¡æ¯ä¸åŒã€‚其他 pr_XXX() å‡½æ•°èƒ½æ— æ¡ä»¶åœ° -打å°ï¼Œpr_debug() å´ä¸ï¼›é»˜è®¤æƒ…况下它ä¸ä¼šè¢«ç¼–译，除éžå®šä¹‰äº† DEBUG 或设定了 -CONFIG_DYNAMIC_DEBUG。实际这åŒæ ·æ˜¯ä¸ºäº† dev_dbg()，一个相关约定是在一个已ç»å¼€å¯äº† -DEBUG 时，使用 VERBOSE_DEBUG æ¥æ·»åŠ dev_vdbg()。 - -许多å系统拥有 Kconfig 调试选项æ¥å¼€å¯ -DDEBUG 在对应的 Makefile 里é¢ï¼›åœ¨å…¶ä»– -情况下，特殊文件使用 #define DEBUG。当一æ¡è°ƒè¯•ä¿¡æ¯éœ€è¦è¢«æ— æ¡ä»¶æ‰“å°æ—¶ï¼Œä¾‹å¦‚，如果 -å·²ç»åŒ…å«ä¸€ä¸ªè°ƒè¯•ç›¸å…³çš„ #ifdef æ¡ä»¶ï¼Œprintk(KERN_DEBUG ...) å°±å¯è¢«ä½¿ç”¨ã€‚ - - - 第åå››ç« ï¼šåˆ†é…å†…å˜ - -å†…æ ¸æ供了下é¢çš„一般用途的内å˜åˆ†é…函数： -kmalloc()，kzalloc()，kmalloc_array()，kcalloc()，vmalloc() å’Œ vzalloc()。 -请å‚考 API 文档以获å–有关它们的详细信æ¯ã€‚ - -ä¼ é€’ç»“æž„ä½“å¤§å°çš„首选形å¼æ˜¯è¿™æ ·çš„： - - p = kmalloc(sizeof(*p), ...); - -å¦å¤–一ç§ä¼ 递方å¼ä¸ï¼Œsizeof çš„æ“作数是结构体的åå—ï¼Œè¿™æ ·ä¼šé™ä½Žå¯è¯»æ€§ï¼Œå¹¶ä¸”å¯èƒ½ä¼šå¼• -å…¥ bug。有å¯èƒ½æŒ‡é’ˆå˜é‡ç±»åž‹è¢«æ”¹å˜æ—¶ï¼Œè€Œå¯¹åº”çš„ä¼ é€’ç»™å†…å˜åˆ†é…函数的 sizeof 的结果ä¸å˜ã€‚ - -强制转æ¢ä¸€ä¸ª void 指针返回值是多余的。C è¯è¨€æœ¬èº«ä¿è¯äº†ä»Ž void 指针到其他任何指针类型 -的转æ¢æ˜¯æ²¡æœ‰é—®é¢˜çš„。 - -分é…一个数组的首选形å¼æ˜¯è¿™æ ·çš„： - - p = kmalloc_array(n, sizeof(...), ...); - -分é…一个零长数组的首选形å¼æ˜¯è¿™æ ·çš„： - - p = kcalloc(n, sizeof(...), ...); - -两ç§å½¢å¼æ£€æŸ¥åˆ†é…å¤§å° n * sizeof(...) 的溢出，如果溢出返回 NULL。 - - - 第åäº”ç« ï¼šå†…è”弊病 - -有一个常è§çš„误解是内è”函数是 gcc æ供的å¯ä»¥è®©ä»£ç è¿è¡Œæ›´å¿«çš„ä¸€ä¸ªé€‰é¡¹ã€‚è™½ç„¶ä½¿ç”¨å†…è” -函数有时候是æ°å½“的（比如作为一ç§æ›¿ä»£å®çš„æ–¹å¼ï¼Œè¯·çœ‹ç¬¬åäºŒç« ï¼‰ï¼Œä¸è¿‡å¾ˆå¤šæƒ…况下ä¸æ˜¯ -è¿™æ ·ã€‚inline 关键å—çš„è¿‡åº¦ä½¿ç”¨ä¼šä½¿å†…æ ¸å˜å¤§ï¼Œä»Žè€Œä½¿æ•´ä¸ªç³»ç»Ÿè¿è¡Œé€Ÿåº¦å˜æ…¢ã€‚å› ä¸ºå¤§å†…æ ¸ -会å 用更多的指令高速缓å˜ï¼ˆè¯‘注：一级缓å˜é€šå¸¸æ˜¯æŒ‡ä»¤ç¼“å˜å’Œæ•°æ®ç¼“å˜åˆ†å¼€çš„）而且会导 -致 pagecache çš„å¯ç”¨å†…å˜å‡å°‘。想象一下，一次pagecache未命ä¸å°±ä¼šå¯¼è‡´ä¸€æ¬¡ç£ç›˜å¯»å€ï¼Œ -将耗时 5 毫秒。5 毫秒的时间内 CPU 能执行很多很多指令。 - -一个基本的原则是如果一个函数有 3 行以上，就ä¸è¦æŠŠå®ƒå˜æˆå†…è”函数。这个原则的一个例 -å¤–æ˜¯ï¼Œå¦‚æžœä½ çŸ¥é“æŸä¸ªå‚数是一个编译时常é‡ï¼Œè€Œä¸”å› ä¸ºè¿™ä¸ªå¸¸é‡ä½ 确定编译器在编译时能 -ä¼˜åŒ–æŽ‰ä½ çš„å‡½æ•°çš„å¤§éƒ¨åˆ†ä»£ç ，那ä»ç„¶å¯ä»¥ç»™å®ƒåŠ 上 inline 关键å—。kmalloc() 内è”函数就 -是一个很好的例å。 - -人们ç»å¸¸ä¸»å¼ ç»™ static 的而且åªç”¨äº†ä¸€æ¬¡çš„å‡½æ•°åŠ ä¸Š inline，如æ¤ä¸ä¼šæœ‰ä»»ä½•æŸå¤±ï¼Œå› 为没 -有什么好æƒè¡¡çš„。虽然从技术上说这是æ£ç¡®çš„，但是实际上这ç§æƒ…况下å³ä½¿ä¸åŠ inline gcc -也å¯ä»¥è‡ªåŠ¨ä½¿å…¶å†…è”。而且其他用户å¯èƒ½ä¼šè¦æ±‚移除 inline，由æ¤è€Œæ¥çš„争论会抵消 inline -自身的潜在价值，得ä¸å¿å¤±ã€‚ - - - 第åå…ç« ï¼šå‡½æ•°è¿”å›žå€¼åŠå‘½å - -函数å¯ä»¥è¿”回很多ç§ä¸åŒç±»åž‹çš„值，最常è§çš„一ç§æ˜¯è¡¨æ˜Žå‡½æ•°æ‰§è¡ŒæˆåŠŸæˆ–è€…å¤±è´¥çš„å€¼ã€‚è¿™æ · -的一个值å¯ä»¥è¡¨ç¤ºä¸ºä¸€ä¸ªé”™è¯¯ä»£ç 整数（-Exxxï¼å¤±è´¥ï¼Œ0ï¼æˆåŠŸï¼‰æˆ–者一个“æˆåŠŸâ€å¸ƒå°”值（ -0ï¼å¤±è´¥ï¼Œéž0ï¼æˆåŠŸï¼‰ã€‚ - -æ··åˆä½¿ç”¨è¿™ä¸¤ç§è¡¨è¾¾æ–¹å¼æ˜¯éš¾äºŽå‘现的 bug çš„æ¥æºã€‚如果 C è¯è¨€æœ¬èº«ä¸¥æ ¼åŒºåˆ†æ•´å½¢å’Œå¸ƒå°”åž‹å˜ -é‡ï¼Œé‚£ä¹ˆç¼–译器就能够帮我们å‘现这些错误……ä¸è¿‡ C è¯è¨€ä¸åŒºåˆ†ã€‚为了é¿å…äº§ç”Ÿè¿™ç§ bug，请 -éµå¾ªä¸‹é¢çš„惯例： - - 如果函数的åå—是一个动作或者强制性的命令，那么这个函数应该返回错误代ç æ•´ - 数。如果是一个判æ–，那么函数应该返回一个“æˆåŠŸâ€å¸ƒå°”值。 - -比如，“add work†是一个命令，所以 add_work() 函数在æˆåŠŸæ—¶è¿”回 0，在失败时返回 -EBUSY。 -ç±»ä¼¼çš„ï¼Œå› ä¸º “PCI device present†是一个判æ–，所以 pci_dev_present() 函数在æˆåŠŸæ‰¾åˆ° -一个匹é…的设备时应该返回 1，如果找ä¸åˆ°æ—¶åº”该返回 0。 - -所有导出（译注：EXPORT）的函数都必须éµå®ˆè¿™ä¸ªæƒ¯ä¾‹ï¼Œæ‰€æœ‰çš„公共函数也都应该如æ¤ã€‚ç§ -有（static）函数ä¸éœ€è¦å¦‚æ¤ï¼Œä½†æ˜¯æˆ‘们也推èè¿™æ ·åšã€‚ - -返回值是实际计算结果而ä¸æ˜¯è®¡ç®—是å¦æˆåŠŸçš„æ ‡å¿—çš„å‡½æ•°ä¸å—æ¤æƒ¯ä¾‹çš„é™åˆ¶ã€‚一般的，他们 -通过返回一些æ£å¸¸å€¼èŒƒå›´ä¹‹å¤–的结果æ¥è¡¨ç¤ºå‡ºé”™ã€‚典型的例å是返回指针的函数，他们使用 -NULL 或者 ERR_PTR 机制æ¥æŠ¥å‘Šé”™è¯¯ã€‚ - - - 第åä¸ƒç« ï¼šä¸è¦é‡æ–°å‘æ˜Žå†…æ ¸å® - -头文件 include/linux/kernel.h 包å«äº†ä¸€äº›å®ï¼Œä½ 应该使用它们，而ä¸è¦è‡ªå·±å†™ä¸€äº›å®ƒä»¬çš„ -å˜ç§ã€‚æ¯”å¦‚ï¼Œå¦‚æžœä½ éœ€è¦è®¡ç®—ä¸€ä¸ªæ•°ç»„çš„é•¿åº¦ï¼Œä½¿ç”¨è¿™ä¸ªå® - - #define ARRAY_SIZE(x) (sizeof(x) / sizeof((x)[0])) - -ç±»ä¼¼çš„ï¼Œå¦‚æžœä½ è¦è®¡ç®—æŸç»“构体æˆå‘˜çš„大å°ï¼Œä½¿ç”¨ - - #define FIELD_SIZEOF(t, f) (sizeof(((t*)0)->f)) - -还有å¯ä»¥åšä¸¥æ ¼çš„类型检查的 min() å’Œ max() å®ï¼Œå¦‚æžœä½ éœ€è¦å¯ä»¥ä½¿ç”¨å®ƒä»¬ã€‚ä½ å¯ä»¥è‡ªå·±çœ‹çœ‹ -é‚£ä¸ªå¤´æ–‡ä»¶é‡Œè¿˜å®šä¹‰äº†ä»€ä¹ˆä½ å¯ä»¥æ‹¿æ¥ç”¨çš„东西，如果有定义的è¯ï¼Œä½ å°±ä¸åº”åœ¨ä½ çš„ä»£ç 里 -自己é‡æ–°å®šä¹‰ã€‚ - - - 第åå…«ç« ï¼šç¼–è¾‘å™¨æ¨¡å¼è¡Œå’Œå…¶ä»–需è¦ç½—嗦的事情 - -有一些编辑器å¯ä»¥è§£é‡ŠåµŒå…¥åœ¨æºæ–‡ä»¶é‡Œçš„ç”±ä¸€äº›ç‰¹æ®Šæ ‡è®°æ ‡æ˜Žçš„é…置信æ¯ã€‚比如，emacs -èƒ½å¤Ÿè§£é‡Šè¢«æ ‡è®°æˆè¿™æ ·çš„行： - - -*- mode: c -*- - -æˆ–è€…è¿™æ ·çš„ï¼š - - /* - Local Variables: - compile-command: "gcc -DMAGIC_DEBUG_FLAG foo.c" - End: - */ - -Vim èƒ½å¤Ÿè§£é‡Šè¿™æ ·çš„æ ‡è®°ï¼š - - /* vim:set sw=8 noet */ - -ä¸è¦åœ¨æºä»£ç ä¸åŒ…å«ä»»ä½•è¿™æ ·çš„内容。æ¯ä¸ªäººéƒ½æœ‰ä»–自己的编辑器é…ç½®ï¼Œä½ çš„æºæ–‡ä»¶ä¸åº” -该覆盖别人的é…置。这包括有关缩进和模å¼é…ç½®çš„æ ‡è®°ã€‚äººä»¬å¯ä»¥ä½¿ç”¨ä»–们自己定制的模 -å¼ï¼Œæˆ–者使用其他å¯ä»¥äº§ç”Ÿæ£ç¡®çš„缩进的巧妙方法。 - - - 第åä¹ç« ：内è”汇编 - -在特定架构的代ç ä¸ï¼Œä½ 也许需è¦å†…è”汇编æ¥ä½¿ç”¨ CPU 接å£å’Œå¹³å°ç›¸å…³åŠŸèƒ½ã€‚åœ¨éœ€è¦ -这么åšæ—¶ï¼Œä¸è¦çŠ¹è±«ã€‚然而，当 C å¯ä»¥å®Œæˆå·¥ä½œæ—¶ï¼Œä¸è¦æ— 端地使用内è”汇编。如果 -å¯èƒ½ï¼Œä½ å¯ä»¥å¹¶ä¸”应该用 C 和硬件交互。 - -考虑去写通用一点的内è”汇编作为简明的辅助函数，而ä¸æ˜¯é‡å¤å†™ä¸‹å®ƒä»¬çš„ç»†èŠ‚ã€‚è®°ä½ -内è”汇编å¯ä»¥ä½¿ç”¨ C å‚数。 - -大而特殊的汇编函数应该放在 .S 文件ä¸ï¼Œå¯¹åº” C 的原型定义在 C 头文件ä¸ã€‚汇编 -函数的 C 原型应该使用 “asmlinkageâ€ã€‚ - -ä½ å¯èƒ½éœ€è¦å°†ä½ 的汇编è¯å¥æ ‡è®°ä¸º volatile，æ¥é˜»æ¢ GCC 在没å‘现任何副作用åŽå°± -ç§»é™¤äº†å®ƒã€‚ä½ ä¸å¿…æ€»æ˜¯è¿™æ ·åšï¼Œè™½ç„¶ï¼Œè¿™æ ·å¯ä»¥é™åˆ¶ä¸å¿…è¦çš„优化。 - -在写一个包å«å¤šæ¡æŒ‡ä»¤çš„å•ä¸ªå†…è”汇编è¯å¥æ—¶ï¼ŒæŠŠæ¯æ¡æŒ‡ä»¤ç”¨å¼•å·å—符串分离，并写在 -å•ç‹¬ä¸€è¡Œï¼Œåœ¨æ¯ä¸ªå—符串结尾，除了 \n\t 结尾之外，在汇编输出ä¸é€‚当地缩进下 -一æ¡æŒ‡ä»¤ï¼š - - asm ("magic %reg1, #42\n\t" - "more_magic %reg2, %reg3" - : /* outputs */ : /* inputs */ : /* clobbers */); - - - 第二åç« ï¼šæ¡ä»¶ç¼–译 - -åªè¦å¯èƒ½ï¼Œå°±ä¸è¦åœ¨ .c 文件里é¢ä½¿ç”¨é¢„处ç†æ¡ä»¶ï¼›è¿™æ ·åšè®©ä»£ç 更难阅读并且逻辑难以 -跟踪。替代方案是，在头文件定义函数在这些 .c 文件ä¸ä½¿ç”¨è¿™ç±»çš„æ¡ä»¶è¡¨è¾¾å¼ï¼Œæ供空 -æ“作的桩版本（译注：桩程åºï¼Œæ˜¯æŒ‡ç”¨æ¥æ›¿æ¢ä¸€éƒ¨åˆ†åŠŸèƒ½çš„程åºæ®µï¼‰åœ¨ #else 情况下， -å†ä»Ž .c 文件ä¸æ— æ¡ä»¶åœ°è°ƒç”¨è¿™äº›å‡½æ•°ã€‚编译器会é¿å…生æˆä»»ä½•æ¡©è°ƒç”¨çš„代ç ，产生一致 -çš„ç»“æžœï¼Œä½†é€»è¾‘å°†æ›´åŠ æ¸…æ™°ã€‚ - -å®å¯ç¼–译整个函数，而ä¸æ˜¯éƒ¨åˆ†å‡½æ•°æˆ–部分表达å¼ã€‚而ä¸æ˜¯åœ¨ä¸€ä¸ªè¡¨è¾¾å¼æ·»åŠ ifdef， -解æžéƒ¨åˆ†æˆ–全部表达å¼åˆ°ä¸€ä¸ªå•ç‹¬çš„辅助函数，并应用æ¡ä»¶åˆ°è¯¥å‡½æ•°å†…。 - -å¦‚æžœä½ æœ‰ä¸€ä¸ªåœ¨ç‰¹å®šé…ç½®ä¸å¯èƒ½æ˜¯æœªä½¿ç”¨çš„函数或å˜é‡ï¼Œç¼–译器将è¦å‘Šå®ƒå®šä¹‰äº†ä½†æœªä½¿ç”¨ï¼Œ -æ ‡è®°è¿™ä¸ªå®šä¹‰ä¸º __maybe_unused 而ä¸æ˜¯å°†å®ƒåŒ…å«åœ¨ä¸€ä¸ªé¢„处ç†æ¡ä»¶ä¸ã€‚（然而，如果 -一个函数或å˜é‡æ€»æ˜¯æœªä½¿ç”¨çš„ï¼Œå°±ç›´æŽ¥åˆ é™¤å®ƒã€‚ï¼‰ - -在代ç ä¸ï¼Œå¯èƒ½çš„情况下，使用 IS_ENABLED å®æ¥è½¬åŒ–æŸä¸ª Kconfig æ ‡è®°ä¸º C 的布尔 -表达å¼ï¼Œå¹¶åœ¨æ£å¸¸çš„ C æ¡ä»¶ä¸ä½¿ç”¨å®ƒï¼š - - if (IS_ENABLED(CONFIG_SOMETHING)) { - ... - } - -ç¼–è¯‘å™¨ä¼šæ— æ¡ä»¶åœ°åšå¸¸æ•°åˆå¹¶ï¼Œå°±åƒä½¿ç”¨ #ifdef é‚£æ ·ï¼ŒåŒ…å«æˆ–排除代ç å—，所以这ä¸ä¼š -带æ¥ä»»ä½•è¿è¡Œæ—¶å¼€é”€ã€‚然而，这ç§æ–¹æ³•ä¾æ—§å…许 C 编译器查看å—内的代ç ，并检查它的æ£ç¡® -性（è¯æ³•ï¼Œç±»åž‹ï¼Œç¬¦å·å¼•ç”¨ï¼Œç‰ç‰ï¼‰ã€‚å› æ¤ï¼Œå¦‚æžœæ¡ä»¶ä¸æ»¡è¶³ï¼Œä»£ç å—内的引用符å·å°†ä¸å˜åœ¨ï¼Œ -ä½ å¿…é¡»ç»§ç»ä½¿ç”¨ #ifdef。 - -在任何有æ„义的 #if 或 #ifdef å—çš„æœ«å°¾ï¼ˆè¶…è¿‡å‡ è¡Œï¼‰ï¼Œåœ¨ #endif åŒä¸€è¡Œçš„åŽé¢å†™ä¸‹ -注释，指出该æ¡ä»¶è¡¨è¾¾å¼è¢«ä½¿ç”¨ã€‚例如： - - #ifdef CONFIG_SOMETHING - ... - #endif /* CONFIG_SOMETHING */ - - - 附录 I：å‚考 - -The C Programming Language, 第二版 -作者：Brian W. Kernighan å’Œ Denni M. Ritchie. -Prentice Hall, Inc., 1988. -ISBN 0-13-110362-8 (软皮), 0-13-110370-9 (硬皮). - -The Practice of Programming -作者：Brian W. Kernighan å’Œ Rob Pike. -Addison-Wesley, Inc., 1999. -ISBN 0-201-61586-X. - -GNU 手册 - éµå¾ª K&R æ ‡å‡†å’Œæ¤æ–‡æœ¬ - cpp, gcc, gcc internals and indent, -都å¯ä»¥ä»Ž http://www.gnu.org/manual/ 找到 - -WG14是Cè¯è¨€çš„å›½é™…æ ‡å‡†åŒ–å·¥ä½œç»„ï¼ŒURL: http://www.open-std.org/JTC1/SC22/WG14/ - -Kernel process/coding-style.rst，作者 greg@kroah.com å‘表于OLS 2002： -http://www.kroah.com/linux/talks/ols_2002_kernel_codingstyle_talk/html/ diff --git a/Documentation/translations/zh_CN/coding-style.rst b/Documentation/translations/zh_CN/coding-style.rst new file mode 100644 index 000000000000..1466aa64b8b4 --- /dev/null +++ b/Documentation/translations/zh_CN/coding-style.rst @@ -0,0 +1,950 @@ +Chinese translated version of Documentation/process/coding-style.rst + +If you have any comment or update to the content, please post to LKML directly. +However, if you have problem communicating in English you can also ask the +Chinese maintainer for help. Contact the Chinese maintainer, if this +translation is outdated or there is problem with translation. + +Chinese maintainer: Zhang Le <r0bertz@gentoo.org> + +--------------------------------------------------------------------- + +Documentation/process/coding-style.rst çš„ä¸æ–‡ç¿»è¯‘ + +如果想评论或更新本文的内容，请直接å‘信到LKMLã€‚å¦‚æžœä½ ä½¿ç”¨è‹±æ–‡äº¤æµæœ‰å›°éš¾çš„è¯ï¼Œ +也å¯ä»¥å‘ä¸æ–‡ç‰ˆç»´æŠ¤è€…求助。如果本翻译更新ä¸åŠæ—¶æˆ–者翻译å˜åœ¨é—®é¢˜ï¼Œè¯·è”ç³»ä¸æ–‡ç‰ˆ +维护者:: + + ä¸æ–‡ç‰ˆç»´æŠ¤è€…： å¼ ä¹ Zhang Le <r0bertz@gentoo.org> + ä¸æ–‡ç‰ˆç¿»è¯‘者： å¼ ä¹ Zhang Le <r0bertz@gentoo.org> + ä¸æ–‡ç‰ˆæ ¡è¯‘者： çŽ‹èª Wang Cong <xiyou.wangcong@gmail.com> + wheelz <kernel.zeng@gmail.com> + 管æ—东 Xudong Guan <xudong.guan@gmail.com> + Li Zefan <lizf@cn.fujitsu.com> + Wang Chen <wangchen@cn.fujitsu.com> + +以下为æ£æ–‡ + +--------------------------------------------------------------------- + +Linux å†…æ ¸ä»£ç é£Žæ ¼ +========================= + +这是一个简çŸçš„文档，æ述了 linux å†…æ ¸çš„é¦–é€‰ä»£ç é£Žæ ¼ã€‚ä»£ç é£Žæ ¼æ˜¯å› äººè€Œå¼‚çš„ï¼Œ +而且我ä¸æ„¿æ„æŠŠè‡ªå·±çš„è§‚ç‚¹å¼ºåŠ ç»™ä»»ä½•äººï¼Œä½†è¿™å°±åƒæˆ‘去åšä»»ä½•äº‹æƒ…都必须éµå¾ªçš„原则 +é‚£æ ·ï¼Œæˆ‘ä¹Ÿå¸Œæœ›åœ¨ç»å¤§å¤šæ•°äº‹ä¸Šä¿æŒè¿™ç§çš„æ€åº¦ã€‚请 (在写代ç æ—¶) 至少考虑一下这里 +的代ç é£Žæ ¼ã€‚ + +é¦–å…ˆï¼Œæˆ‘å»ºè®®ä½ æ‰“å°ä¸€ä»½ GNU 代ç 规范，然åŽä¸è¦è¯»ã€‚烧了它，这是一个具有é‡å¤§è±¡å¾ +性æ„义的动作。 + +ä¸ç®¡æ€Žæ ·ï¼ŒçŽ°åœ¨æˆ‘们开始： + + +1) 缩进 +-------------- + +制表符是 8 个å—符，所以缩进也是 8 个å—符。有些异端è¿åŠ¨è¯•å›¾å°†ç¼©è¿›å˜ä¸º 4 (甚至 +2ï¼) å—ç¬¦æ·±ï¼Œè¿™å‡ ä¹Žç›¸å½“äºŽå°è¯•å°†åœ†å‘¨çŽ‡çš„值定义为 3。 + +ç†ç”±ï¼šç¼©è¿›çš„全部æ„义就在于清楚的定义一个控制å—èµ·æ¢äºŽä½•å¤„ã€‚å°¤å…¶æ˜¯å½“ä½ ç›¯ç€ä½ çš„ +å±å¹•è¿žç»çœ‹äº† 20 å°æ—¶ä¹‹åŽï¼Œä½ 将会å‘çŽ°å¤§ä¸€ç‚¹çš„ç¼©è¿›ä¼šä½¿ä½ æ›´å®¹æ˜“åˆ†è¾¨ç¼©è¿›ã€‚ + +现在，有些人会抱怨 8 个å—符的缩进会使代ç å‘å³è¾¹ç§»åŠ¨çš„太远，在 80 个å—符的终端 +å±å¹•ä¸Šå°±å¾ˆéš¾è¯»è¿™æ ·çš„代ç 。这个问题的ç”æ¡ˆæ˜¯ï¼Œå¦‚æžœä½ éœ€è¦ 3 级以上的缩进，ä¸ç®¡ç”¨ +何ç§æ–¹å¼ä½ 的代ç å·²ç»æœ‰é—®é¢˜äº†ï¼Œåº”该修æ£ä½ 的程åºã€‚ + +简而言之，8 个å—符的缩进å¯ä»¥è®©ä»£ç æ›´å®¹æ˜“é˜…è¯»ï¼Œè¿˜æœ‰ä¸€ä¸ªå¥½å¤„æ˜¯å½“ä½ çš„å‡½æ•°åµŒå¥—å¤ª +深的时候å¯ä»¥ç»™ä½ è¦å‘Šã€‚留心这个è¦å‘Šã€‚ + +在 switch è¯å¥ä¸æ¶ˆé™¤å¤šçº§ç¼©è¿›çš„首选的方å¼æ˜¯è®© ``switch`` 和从属于它的 ``case`` +æ ‡ç¾å¯¹é½äºŽåŒä¸€åˆ—，而ä¸è¦ ``两次缩进`` ``case`` æ ‡ç¾ã€‚比如： + +.. code-block:: c + + switch (suffix) { + case 'G': + case 'g': + mem <<= 30; + break; + case 'M': + case 'm': + mem <<= 20; + break; + case 'K': + case 'k': + mem <<= 10; + /* fall through */ + default: + break; + } + +ä¸è¦æŠŠå¤šä¸ªè¯å¥æ”¾åœ¨ä¸€è¡Œé‡Œï¼Œé™¤éžä½ 有什么东西è¦éšè—： + +.. code-block:: c + + if (condition) do_this; + do_something_everytime; + +也ä¸è¦åœ¨ä¸€è¡Œé‡Œæ”¾å¤šä¸ªèµ‹å€¼è¯å¥ã€‚å†…æ ¸ä»£ç é£Žæ ¼è¶…çº§ç®€å•ã€‚就是é¿å…å¯èƒ½å¯¼è‡´åˆ«äººè¯¯è¯» +的表达å¼ã€‚ + +除了注释ã€æ–‡æ¡£å’Œ Kconfig 之外，ä¸è¦ä½¿ç”¨ç©ºæ ¼æ¥ç¼©è¿›ï¼Œå‰é¢çš„例å是例外，是有æ„为 +之。 + +选用一个好的编辑器，ä¸è¦åœ¨è¡Œå°¾ç•™ç©ºæ ¼ã€‚ + + +2) 把长的行和å—符串打散 +------------------------------ + +代ç é£Žæ ¼çš„æ„义就在于使用平常使用的工具æ¥ç»´æŒä»£ç çš„å¯è¯»æ€§å’Œå¯ç»´æŠ¤æ€§ã€‚ + +æ¯ä¸€è¡Œçš„长度的é™åˆ¶æ˜¯ 80 列，我们强烈建议您éµå®ˆè¿™ä¸ªæƒ¯ä¾‹ã€‚ + +长于 80 列的è¯å¥è¦æ‰“æ•£æˆæœ‰æ„义的片段。除éžè¶…过 80 åˆ—èƒ½æ˜¾è‘—å¢žåŠ å¯è¯»æ€§ï¼Œå¹¶ä¸”ä¸ +会éšè—ä¿¡æ¯ã€‚å片段è¦æ˜Žæ˜¾çŸäºŽæ¯ç‰‡æ®µï¼Œå¹¶æ˜Žæ˜¾é å³ã€‚è¿™åŒæ ·é€‚用于有ç€å¾ˆé•¿å‚数列表 +的函数头。然而，ç»å¯¹ä¸è¦æ‰“散对用户å¯è§çš„å—符串，例如 printk ä¿¡æ¯ï¼Œå› ä¸ºè¿™æ ·å°± +很难对它们 grep。 + + +3) 大括å·å’Œç©ºæ ¼çš„放置 +------------------------------ + +C è¯è¨€é£Žæ ¼ä¸å¦å¤–一个常è§é—®é¢˜æ˜¯å¤§æ‹¬å·çš„放置。和缩进大å°ä¸åŒï¼Œé€‰æ‹©æˆ–弃用æŸç§æ”¾ +ç½®ç–ç•¥å¹¶æ²¡æœ‰å¤šå°‘æŠ€æœ¯ä¸Šçš„åŽŸå› ï¼Œä¸è¿‡é¦–选的方å¼ï¼Œå°±åƒ Kernighan å’Œ Ritchie 展示 +给我们的，是把起始大括å·æ”¾åœ¨è¡Œå°¾ï¼Œè€ŒæŠŠç»“æŸå¤§æ‹¬å·æ”¾åœ¨è¡Œé¦–，所以： + +.. code-block:: c + + if (x is true) { + we do y + } + +这适用于所有的éžå‡½æ•°è¯å¥å— (if, switch, for, while, do)。比如： + +.. code-block:: c + + switch (action) { + case KOBJ_ADD: + return "add"; + case KOBJ_REMOVE: + return "remove"; + case KOBJ_CHANGE: + return "change"; + default: + return NULL; + } + +ä¸è¿‡ï¼Œæœ‰ä¸€ä¸ªä¾‹å¤–，那就是函数：函数的起始大括å·æ”¾ç½®äºŽä¸‹ä¸€è¡Œçš„开头，所以： + +.. code-block:: c + + int function(int x) + { + body of function + } + +全世界的异端å¯èƒ½ä¼šæŠ±æ€¨è¿™ä¸ªä¸ä¸€è‡´æ€§æ˜¯... 呃... ä¸ä¸€è‡´çš„，ä¸è¿‡æ‰€æœ‰æ€ç»´å¥å…¨çš„人 +éƒ½çŸ¥é“ (a) K&R 是 **æ£ç¡®çš„** 并且 (b) K&R 是æ£ç¡®çš„。æ¤å¤–，ä¸ç®¡æ€Žæ ·å‡½æ•°éƒ½æ˜¯ç‰¹ +殊的 (C 函数是ä¸èƒ½åµŒå¥—çš„)。 + +注æ„结æŸå¤§æ‹¬å·ç‹¬è‡ªå æ®ä¸€è¡Œï¼Œé™¤éžå®ƒåŽé¢è·Ÿç€åŒä¸€ä¸ªè¯å¥çš„剩余部分，也就是 do è¯ +å¥ä¸çš„ "while" 或者 if è¯å¥ä¸çš„ "else"，åƒè¿™æ ·ï¼š + +.. code-block:: c + + do { + body of do-loop + } while (condition); + +å’Œ + +.. code-block:: c + + if (x == y) { + .. + } else if (x > y) { + ... + } else { + .... + } + +ç†ç”±ï¼šK&R。 + +也请注æ„è¿™ç§å¤§æ‹¬å·çš„放置方å¼ä¹Ÿèƒ½ä½¿ç©º (或者差ä¸å¤šç©ºçš„) 行的数é‡æœ€å°åŒ–，åŒæ—¶ä¸ +失å¯è¯»æ€§ã€‚å› æ¤ï¼Œç”±äºŽä½ çš„å±å¹•ä¸Šçš„新行是ä¸å¯å†ç”Ÿèµ„æº (想想 25 行的终端å±å¹•)ï¼Œä½ +将会有更多的空行æ¥æ”¾ç½®æ³¨é‡Šã€‚ + +当åªæœ‰ä¸€ä¸ªå•ç‹¬çš„è¯å¥çš„时候，ä¸ç”¨åŠ ä¸å¿…è¦çš„大括å·ã€‚ + +.. code-block:: c + + if (condition) + action(); + +å’Œ + +.. code-block:: c + + if (condition) + do_this(); + else + do_that(); + +这并ä¸é€‚用于åªæœ‰ä¸€ä¸ªæ¡ä»¶åˆ†æ”¯æ˜¯å•è¯å¥çš„情况；这时所有分支都è¦ä½¿ç”¨å¤§æ‹¬å·ï¼š + +.. code-block:: c + + if (condition) { + do_this(); + do_that(); + } else { + otherwise(); + } + +3.1) ç©ºæ ¼ +******************** + +Linux å†…æ ¸çš„ç©ºæ ¼ä½¿ç”¨æ–¹å¼ (主è¦) å–决于它是用于函数还是关键å—。(大多数) å…³é”®å— +åŽè¦åŠ ä¸€ä¸ªç©ºæ ¼ã€‚å€¼å¾—æ³¨æ„的例外是 sizeof, typeof, alignof å’Œ __attribute__，这 +些关键å—æŸäº›ç¨‹åº¦ä¸Šçœ‹èµ·æ¥æ›´åƒå‡½æ•° (它们在 Linux 里也常常伴éšå°æ‹¬å·è€Œä½¿ç”¨ï¼Œå°½ç®¡ +在 C é‡Œè¿™æ ·çš„å°æ‹¬å·ä¸æ˜¯å¿…éœ€çš„ï¼Œå°±åƒ ``struct fileinfo info;`` 声明过åŽçš„ +``sizeof info``)。 + +所以在这些关键å—之åŽæ”¾ä¸€ä¸ªç©ºæ ¼:: + + if, switch, case, for, do, while + +但是ä¸è¦åœ¨ sizeof, typeof, alignof 或者 __attribute__ 这些关键å—之åŽæ”¾ç©ºæ ¼ã€‚ +例如， + +.. code-block:: c + + s = sizeof(struct file); + +ä¸è¦åœ¨å°æ‹¬å·é‡Œçš„表达å¼ä¸¤ä¾§åŠ ç©ºæ ¼ã€‚è¿™æ˜¯ä¸€ä¸ª **å例** ： + +.. code-block:: c + + s = sizeof( struct file ); + +当声明指针类型或者返回指针类型的函数时， ``*`` 的首选使用方å¼æ˜¯ä½¿ä¹‹é è¿‘å˜é‡å +或者函数å，而ä¸æ˜¯é 近类型å。例å： + +.. code-block:: c + + char *linux_banner; + unsigned long long memparse(char *ptr, char **retptr); + char *match_strdup(substring_t *s); + +在大多数二元和三元æ“ä½œç¬¦ä¸¤ä¾§ä½¿ç”¨ä¸€ä¸ªç©ºæ ¼ï¼Œä¾‹å¦‚ä¸‹é¢æ‰€æœ‰è¿™äº›æ“作符:: + + = + - < > * / % | & ^ <= >= == != ? : + +但是一元æ“作符åŽä¸è¦åŠ ç©ºæ ¼:: + + & * + - ~ ! sizeof typeof alignof __attribute__ defined + +åŽç¼€è‡ªåŠ 和自å‡ä¸€å…ƒæ“作符å‰ä¸åŠ ç©ºæ ¼:: + + ++ -- + +å‰ç¼€è‡ªåŠ 和自å‡ä¸€å…ƒæ“作符åŽä¸åŠ ç©ºæ ¼:: + + ++ -- + +``.`` å’Œ ``->`` 结构体æˆå‘˜æ“作符å‰åŽä¸åŠ ç©ºæ ¼ã€‚ + +ä¸è¦åœ¨è¡Œå°¾ç•™ç©ºç™½ã€‚有些å¯ä»¥è‡ªåŠ¨ç¼©è¿›çš„ç¼–è¾‘å™¨ä¼šåœ¨æ–°è¡Œçš„è¡Œé¦–åŠ å…¥é€‚é‡çš„ç©ºç™½ï¼Œç„¶åŽ +ä½ å°±å¯ä»¥ç›´æŽ¥åœ¨é‚£ä¸€è¡Œè¾“入代ç 。ä¸è¿‡å‡å¦‚ä½ æœ€åŽæ²¡æœ‰åœ¨é‚£ä¸€è¡Œè¾“入代ç ，有些编辑器 +å°±ä¸ä¼šç§»é™¤å·²ç»åŠ 入的空白，就åƒä½ æ•…æ„留下一个åªæœ‰ç©ºç™½çš„行。包å«è¡Œå°¾ç©ºç™½çš„行就 +è¿™æ ·äº§ç”Ÿäº†ã€‚ + +当 git å‘现补ä¸åŒ…å«äº†è¡Œå°¾ç©ºç™½çš„时候会è¦å‘Šä½ ，并且å¯ä»¥åº”ä½ çš„è¦æ±‚去掉行尾空白； +ä¸è¿‡å¦‚æžœä½ æ˜¯æ£åœ¨æ‰“一系列补ä¸ï¼Œè¿™æ ·åšä¼šå¯¼è‡´åŽé¢çš„è¡¥ä¸å¤±è´¥ï¼Œå› ä¸ºä½ æ”¹å˜äº†è¡¥ä¸çš„ +上下文。 + + +4) 命å +------------------------------ + +C 是一个简朴的è¯è¨€ï¼Œä½ 的命åä¹Ÿåº”è¯¥è¿™æ ·ã€‚å’Œ Modula-2 å’Œ Pascal 程åºå‘˜ä¸åŒï¼Œ +C 程åºå‘˜ä¸ä½¿ç”¨ç±»ä¼¼ ThisVariableIsATemporaryCounter è¿™æ ·åŽä¸½çš„åå—。C 程åºå‘˜ä¼š +称那个å˜é‡ä¸º ``tmp`` ï¼Œè¿™æ ·å†™èµ·æ¥ä¼šæ›´å®¹æ˜“，而且至少ä¸ä¼šä»¤å…¶éš¾äºŽç†è§£ã€‚ + +ä¸è¿‡ï¼Œè™½ç„¶æ··ç”¨å¤§å°å†™çš„åå—是ä¸æ倡使用的，但是全局å˜é‡è¿˜æ˜¯éœ€è¦ä¸€ä¸ªå…·æ述性的 +åå—。称一个全局函数为 ``foo`` 是一个难以饶æ•çš„错误。 + +全局å˜é‡ (åªæœ‰å½“ä½ **真æ£** 需è¦å®ƒä»¬çš„时候å†ç”¨å®ƒ) 需è¦æœ‰ä¸€ä¸ªå…·æ述性的åå—，就 +åƒå…¨å±€å‡½æ•°ã€‚å¦‚æžœä½ æœ‰ä¸€ä¸ªå¯ä»¥è®¡ç®—活动用户数é‡çš„å‡½æ•°ï¼Œä½ åº”è¯¥å«å®ƒ +``count_active_users()`` 或者类似的åå—ï¼Œä½ ä¸åº”该å«å®ƒ ``cntuser()`` 。 + +在函数åä¸åŒ…å«å‡½æ•°ç±»åž‹ (所谓的匈牙利命å法) 是脑å出了问题——编译器知é“那些类 +åž‹è€Œä¸”èƒ½å¤Ÿæ£€æŸ¥é‚£äº›ç±»åž‹ï¼Œè¿™æ ·åšåªèƒ½æŠŠç¨‹åºå‘˜å¼„ç³Šæ¶‚äº†ã€‚éš¾æ€ªå¾®è½¯æ€»æ˜¯åˆ¶é€ å‡ºæœ‰é—®é¢˜ +的程åºã€‚ + +本地å˜é‡å应该简çŸï¼Œè€Œä¸”能够表达相关的å«ä¹‰ã€‚å¦‚æžœä½ æœ‰ä¸€äº›éšæœºçš„整数型的循环计 +数器，它应该被称为 ``i`` 。å«å®ƒ ``loop_counter`` å¹¶æ— ç›Šå¤„ï¼Œå¦‚æžœå®ƒæ²¡æœ‰è¢«è¯¯è§£çš„ +å¯èƒ½çš„è¯ã€‚类似的， ``tmp`` å¯ä»¥ç”¨æ¥ç§°å‘¼ä»»æ„类型的临时å˜é‡ã€‚ + +å¦‚æžœä½ æ€•æ··æ·†äº†ä½ çš„æœ¬åœ°å˜é‡åï¼Œä½ å°±é‡åˆ°å¦ä¸€ä¸ªé—®é¢˜äº†ï¼Œå«åšå‡½æ•°å¢žé•¿è·å°”蒙失衡综 +åˆç—‡ã€‚请看第å…ç« (函数)。 + + +5) Typedef +----------- + +ä¸è¦ä½¿ç”¨ç±»ä¼¼ ``vps_t`` 之类的东西。 + +对结构体和指针使用 typedef 是一个 **错误** ã€‚å½“ä½ åœ¨ä»£ç 里看到： + +.. code-block:: c + + vps_t a; + +这代表什么æ„æ€å‘¢ï¼Ÿ + +相åï¼Œå¦‚æžœæ˜¯è¿™æ · + +.. code-block:: c + + struct virtual_container *a; + +ä½ å°±çŸ¥é“ ``a`` 是什么了。 + +很多人认为 typedef ``能æ高å¯è¯»æ€§`` 。实际ä¸æ˜¯è¿™æ ·çš„。它们åªåœ¨ä¸‹åˆ—情况下有用： + + (a) 完全ä¸é€æ˜Žçš„对象 (è¿™ç§æƒ…况下è¦ä¸»åŠ¨ä½¿ç”¨ typedef æ¥ **éšè—** 这个对象实际上 + 是什么)。 + + 例如： ``pte_t`` ç‰ä¸é€æ˜Žå¯¹è±¡ï¼Œä½ åªèƒ½ç”¨åˆé€‚的访问函数æ¥è®¿é—®å®ƒä»¬ã€‚ + + .. note:: + + ä¸é€æ˜Žæ€§å’Œ "访问函数" 本身是ä¸å¥½çš„。我们使用 pte_t ç‰ç±»åž‹çš„åŽŸå› åœ¨äºŽçœŸ + 的是完全没有任何共用的å¯è®¿é—®ä¿¡æ¯ã€‚ + + (b) 清楚的整数类型，如æ¤ï¼Œè¿™å±‚抽象就å¯ä»¥ **帮助** 消除到底是 ``int`` 还是 + ``long`` 的混淆。 + + u8/u16/u32 是完全没有问题的 typedef，ä¸è¿‡å®ƒä»¬æ›´ç¬¦åˆç±»åˆ« (d) 而ä¸æ˜¯è¿™é‡Œã€‚ + + .. note:: + + è¦è¿™æ ·åšï¼Œå¿…é¡»äº‹å‡ºæœ‰å› ã€‚å¦‚æžœæŸä¸ªå˜é‡æ˜¯ ``unsigned long`` ï¼Œé‚£ä¹ˆæ²¡æœ‰å¿…è¦ + + typedef unsigned long myflags_t; + + ä¸è¿‡å¦‚æžœæœ‰ä¸€ä¸ªæ˜Žç¡®çš„åŽŸå› ï¼Œæ¯”å¦‚å®ƒåœ¨æŸç§æƒ…况下å¯èƒ½ä¼šæ˜¯ä¸€ä¸ª ``unsigned int`` + 而在其他情况下å¯èƒ½ä¸º ``unsigned long`` ，那么就ä¸è¦çŠ¹è±«ï¼Œè¯·åŠ¡å¿…使用 + typedef。 + + (c) å½“ä½ ä½¿ç”¨ sparse 按å—é¢çš„创建一个 **æ–°** 类型æ¥åšç±»åž‹æ£€æŸ¥çš„时候。 + + (d) å’Œæ ‡å‡† C99 类型相åŒçš„类型，在æŸäº›ä¾‹å¤–的情况下。 + + 虽然让眼ç›å’Œè„‘ç‹æ¥é€‚åº”æ–°çš„æ ‡å‡†ç±»åž‹æ¯”å¦‚ ``uint32_t`` ä¸éœ€è¦èŠ±å¾ˆå¤šæ—¶é—´ï¼Œå¯ + 是有些人ä»ç„¶æ‹’ç»ä½¿ç”¨å®ƒä»¬ã€‚ + + å› æ¤ï¼ŒLinux 特有的ç‰åŒäºŽæ ‡å‡†ç±»åž‹çš„ ``u8/u16/u32/u64`` ç±»åž‹å’Œå®ƒä»¬çš„æœ‰ç¬¦å· + 类型是被å…è®¸çš„â€”â€”å°½ç®¡åœ¨ä½ è‡ªå·±çš„æ–°ä»£ç ä¸ï¼Œå®ƒä»¬ä¸æ˜¯å¼ºåˆ¶è¦æ±‚è¦ä½¿ç”¨çš„。 + + 当编辑已ç»ä½¿ç”¨äº†æŸä¸ªç±»åž‹é›†çš„已有代ç æ—¶ï¼Œä½ åº”è¯¥éµå¾ªé‚£äº›ä»£ç ä¸å·²ç»åšå‡ºçš„选 + 择。 + + (e) å¯ä»¥åœ¨ç”¨æˆ·ç©ºé—´å®‰å…¨ä½¿ç”¨çš„类型。 + + 在æŸäº›ç”¨æˆ·ç©ºé—´å¯è§çš„结构体里，我们ä¸èƒ½è¦æ±‚ C99 类型而且ä¸èƒ½ç”¨ä¸Šé¢æ到的 + ``u32`` ç±»åž‹ã€‚å› æ¤ï¼Œæˆ‘们在与用户空间共享的所有结构体ä¸ä½¿ç”¨ __u32 和类似 + 的类型。 + +å¯èƒ½è¿˜æœ‰å…¶ä»–的情况，ä¸è¿‡åŸºæœ¬çš„规则是 **永远ä¸è¦** 使用 typedef，除éžä½ å¯ä»¥æ˜Ž +确的应用上述æŸä¸ªè§„则ä¸çš„一个。 + +总的æ¥è¯´ï¼Œå¦‚æžœä¸€ä¸ªæŒ‡é’ˆæˆ–è€…ä¸€ä¸ªç»“æž„ä½“é‡Œçš„å…ƒç´ å¯ä»¥åˆç†çš„被直接访问到，那么它们 +å°±ä¸åº”该是一个 typedef。 + + +6) 函数 +------------------------------ + +函数应该简çŸè€Œæ¼‚亮，并且åªå®Œæˆä¸€ä»¶äº‹æƒ…。函数应该å¯ä»¥ä¸€å±æˆ–者两å±æ˜¾ç¤ºå®Œ (我们 +éƒ½çŸ¥é“ ISO/ANSI å±å¹•å¤§å°æ˜¯ 80x24)，åªåšä¸€ä»¶äº‹æƒ…，而且把它åšå¥½ã€‚ + +一个函数的最大长度是和该函数的å¤æ‚度和缩进级数æˆåæ¯”çš„ã€‚æ‰€ä»¥ï¼Œå¦‚æžœä½ æœ‰ä¸€ä¸ªç† +论上很简å•çš„åªæœ‰ä¸€ä¸ªå¾ˆé•¿ (但是简å•) çš„ case è¯å¥çš„å‡½æ•°ï¼Œè€Œä¸”ä½ éœ€è¦åœ¨æ¯ä¸ª case +里åšå¾ˆå¤šå¾ˆå°çš„äº‹æƒ…ï¼Œè¿™æ ·çš„å‡½æ•°å°½ç®¡å¾ˆé•¿ï¼Œä½†ä¹Ÿæ˜¯å¯ä»¥çš„。 + +ä¸è¿‡ï¼Œå¦‚æžœä½ æœ‰ä¸€ä¸ªå¤æ‚çš„å‡½æ•°ï¼Œè€Œä¸”ä½ æ€€ç–‘ä¸€ä¸ªå¤©åˆ†ä¸æ˜¯å¾ˆé«˜çš„高ä¸ä¸€å¹´çº§å¦ç”Ÿå¯èƒ½ +甚至æžä¸æ¸…æ¥šè¿™ä¸ªå‡½æ•°çš„ç›®çš„ï¼Œä½ åº”è¯¥ä¸¥æ ¼éµå®ˆå‰é¢æ到的长度é™åˆ¶ã€‚使用辅助函数， +并为之å–个具æ述性的åå— (å¦‚æžœä½ è§‰å¾—å®ƒä»¬çš„æ€§èƒ½å¾ˆé‡è¦çš„è¯ï¼Œå¯ä»¥è®©ç¼–译器内è”它 +ä»¬ï¼Œè¿™æ ·çš„æ•ˆæžœå¾€å¾€ä¼šæ¯”ä½ å†™ä¸€ä¸ªå¤æ‚函数的效果è¦å¥½ã€‚) + +函数的å¦å¤–一个衡é‡æ ‡å‡†æ˜¯æœ¬åœ°å˜é‡çš„æ•°é‡ã€‚æ¤æ•°é‡ä¸åº”超过 5ï¼10 个，å¦åˆ™ä½ 的函数 +就有问题了。é‡æ–°è€ƒè™‘ä¸€ä¸‹ä½ çš„å‡½æ•°ï¼ŒæŠŠå®ƒåˆ†æ‹†æˆæ›´å°çš„函数。人的大脑一般å¯ä»¥è½»æ¾ +çš„åŒæ—¶è·Ÿè¸ª 7 个ä¸åŒçš„事物，如果å†å¢žå¤šçš„è¯ï¼Œå°±ä¼šç³Šæ¶‚了。å³ä¾¿ä½ èªé¢–è¿‡äººï¼Œä½ ä¹Ÿå¯ +能会记ä¸æ¸…ä½ 2 个星期å‰åšè¿‡çš„事情。 + +在æºæ–‡ä»¶é‡Œï¼Œä½¿ç”¨ç©ºè¡Œéš”å¼€ä¸åŒçš„函数。如果该函数需è¦è¢«å¯¼å‡ºï¼Œå®ƒçš„ **EXPORT** å® +应该紧贴在它的结æŸå¤§æ‹¬å·ä¹‹ä¸‹ã€‚比如： + +.. code-block:: c + + int system_is_up(void) + { + return system_state == SYSTEM_RUNNING; + } + EXPORT_SYMBOL(system_is_up); + +在函数原型ä¸ï¼ŒåŒ…å«å‡½æ•°å和它们的数æ®ç±»åž‹ã€‚虽然 C è¯è¨€é‡Œæ²¡æœ‰è¿™æ ·çš„è¦æ±‚，在 +Linux 里这是æ倡的åšæ³•ï¼Œå› ä¸ºè¿™æ ·å¯ä»¥å¾ˆç®€å•çš„给读者æ供更多的有价值的信æ¯ã€‚ + + +7) 集ä¸çš„函数退出途径 +------------------------------ + +虽然被æŸäº›äººå£°ç§°å·²ç»è¿‡æ—¶ï¼Œä½†æ˜¯ goto è¯å¥çš„ç‰ä»·ç‰©è¿˜æ˜¯ç»å¸¸è¢«ç¼–译器所使用，具体 +å½¢å¼æ˜¯æ— æ¡ä»¶è·³è½¬æŒ‡ä»¤ã€‚ + +当一个函数从多个ä½ç½®é€€å‡ºï¼Œå¹¶ä¸”需è¦åšä¸€äº›ç±»ä¼¼æ¸…ç†çš„常è§æ“作时，goto è¯å¥å°±å¾ˆæ–¹ +便了。如果并ä¸éœ€è¦æ¸…ç†æ“作，那么直接 return å³å¯ã€‚ + +选择一个能够说明 goto 行为或它为何å˜åœ¨çš„æ ‡ç¾å。如果 goto è¦é‡Šæ”¾ ``buffer``, +一个ä¸é”™çš„åå—å¯ä»¥æ˜¯ ``out_free_buffer:`` ã€‚åˆ«åŽ»ä½¿ç”¨åƒ ``err1:`` å’Œ ``err2:`` +è¿™æ ·çš„GW_BASIC åç§°ï¼Œå› ä¸ºä¸€æ—¦ä½ æ·»åŠ æˆ–åˆ é™¤äº† (函数的) é€€å‡ºè·¯å¾„ï¼Œä½ å°±å¿…é¡»å¯¹å®ƒä»¬ +é‡æ–°ç¼–å·ï¼Œè¿™æ ·ä¼šéš¾ä»¥åŽ»æ£€éªŒæ£ç¡®æ€§ã€‚ + +使用 goto çš„ç†ç”±æ˜¯ï¼š + +- æ— æ¡ä»¶è¯å¥å®¹æ˜“ç†è§£å’Œè·Ÿè¸ª +- 嵌套程度å‡å° +- å¯ä»¥é¿å…由于修改时忘记更新个别的退出点而导致错误 +- 让编译器çœåŽ»åˆ 除冗余代ç 的工作 ;) + +.. code-block:: c + + int fun(int a) + { + int result = 0; + char *buffer; + + buffer = kmalloc(SIZE, GFP_KERNEL); + if (!buffer) + return -ENOMEM; + + if (condition1) { + while (loop1) { + ... + } + result = 1; + goto out_free_buffer; + } + ... + out_free_buffer: + kfree(buffer); + return result; + } + +一个需è¦æ³¨æ„的常è§é”™è¯¯æ˜¯ ``一个 err 错误`` ，就åƒè¿™æ ·ï¼š + +.. code-block:: c + + err: + kfree(foo->bar); + kfree(foo); + return ret; + +这段代ç 的错误是，在æŸäº›é€€å‡ºè·¯å¾„上 ``foo`` 是 NULL。通常情况下，通过把它分离 +æˆä¸¤ä¸ªé”™è¯¯æ ‡ç¾ ``err_free_bar:`` å’Œ ``err_free_foo:`` æ¥ä¿®å¤è¿™ä¸ªé”™è¯¯ï¼š + +.. code-block:: c + + err_free_bar: + kfree(foo->bar); + err_free_foo: + kfree(foo); + return ret; + +ç†æƒ³æƒ…å†µä¸‹ï¼Œä½ åº”è¯¥æ¨¡æ‹Ÿé”™è¯¯æ¥æµ‹è¯•æ‰€æœ‰é€€å‡ºè·¯å¾„。 + + +8) 注释 +------------------------------ + +注释是好的，ä¸è¿‡æœ‰è¿‡åº¦æ³¨é‡Šçš„å±é™©ã€‚永远ä¸è¦åœ¨æ³¨é‡Šé‡Œè§£é‡Šä½ 的代ç 是如何è¿ä½œçš„： +更好的åšæ³•æ˜¯è®©åˆ«äººä¸€çœ‹ä½ 的代ç å°±å¯ä»¥æ˜Žç™½ï¼Œè§£é‡Šå†™çš„很差的代ç 是浪费时间。 + +ä¸€èˆ¬çš„ï¼Œä½ æƒ³è¦ä½ çš„æ³¨é‡Šå‘Šè¯‰åˆ«äººä½ çš„ä»£ç åšäº†ä»€ä¹ˆï¼Œè€Œä¸æ˜¯æ€Žä¹ˆåšçš„ã€‚ä¹Ÿè¯·ä½ ä¸è¦æŠŠ +注释放在一个函数体内部：如果函数å¤æ‚åˆ°ä½ éœ€è¦ç‹¬ç«‹çš„注释其ä¸çš„ä¸€éƒ¨åˆ†ï¼Œä½ å¾ˆå¯èƒ½ +需è¦å›žåˆ°ç¬¬å…ç« çœ‹ä¸€çœ‹ã€‚ä½ å¯ä»¥åšä¸€äº›å°æ³¨é‡Šæ¥æ³¨æ˜Žæˆ–è¦å‘ŠæŸäº›å¾ˆèªæ˜Ž (或者槽糕) çš„ +åšæ³•ï¼Œä½†ä¸è¦åŠ å¤ªå¤šã€‚ä½ åº”è¯¥åšçš„，是把注释放在函数的头部，告诉人们它åšäº†ä»€ä¹ˆï¼Œ +也å¯ä»¥åŠ 上它åšè¿™äº›äº‹æƒ…çš„åŽŸå› ã€‚ + +å½“æ³¨é‡Šå†…æ ¸ API 函数时，请使用 kernel-doc æ ¼å¼ã€‚请看 +Documentation/doc-guide/ å’Œ scripts/kernel-doc 以获得详细信æ¯ã€‚ + +é•¿ (多行) æ³¨é‡Šçš„é¦–é€‰é£Žæ ¼æ˜¯ï¼š + +.. code-block:: c + + /* + * This is the preferred style for multi-line + * comments in the Linux kernel source code. + * Please use it consistently. + * + * Description: A column of asterisks on the left side, + * with beginning and ending almost-blank lines. + */ + +对于在 net/ å’Œ drivers/net/ 的文件，首选的长 (多行) æ³¨é‡Šé£Žæ ¼æœ‰äº›ä¸åŒã€‚ + +.. code-block:: c + + /* The preferred comment style for files in net/ and drivers/net + * looks like this. + * + * It is nearly the same as the generally preferred comment style, + * but there is no initial almost-blank line. + */ + +注释数æ®ä¹Ÿæ˜¯å¾ˆé‡è¦çš„，ä¸ç®¡æ˜¯åŸºæœ¬ç±»åž‹è¿˜æ˜¯è¡ç”Ÿç±»åž‹ã€‚为了方便实现这一点，æ¯ä¸€è¡Œ +应åªå£°æ˜Žä¸€ä¸ªæ•°æ® (ä¸è¦ä½¿ç”¨é€—å·æ¥ä¸€æ¬¡å£°æ˜Žå¤šä¸ªæ•°æ®)ã€‚è¿™æ ·ä½ å°±æœ‰ç©ºé—´æ¥ä¸ºæ¯ä¸ªæ•°æ® +写一段å°æ³¨é‡Šæ¥è§£é‡Šå®ƒä»¬çš„用途了。 + + +9) ä½ å·²ç»æŠŠäº‹æƒ…弄糟了 +------------------------------ + +è¿™æ²¡ä»€ä¹ˆï¼Œæˆ‘ä»¬éƒ½æ˜¯è¿™æ ·ã€‚å¯èƒ½ä½ 的使用了很长时间 Unix 的朋å‹å·²ç»å‘Šè¯‰ä½ +``GNU emacs`` èƒ½è‡ªåŠ¨å¸®ä½ æ ¼å¼åŒ– C æºä»£ç ï¼Œè€Œä¸”ä½ ä¹Ÿæ³¨æ„åˆ°äº†ï¼Œç¡®å®žæ˜¯è¿™æ ·ï¼Œä¸è¿‡å®ƒ +所使用的默认值和我们想è¦çš„相去甚远 (实际上，甚至比éšæœºæ‰“的还è¦å·®â€”â€”æ— æ•°ä¸ªçŒ´å +在 GNU emacs 里打å—永远ä¸ä¼šåˆ›é€ 出一个好程åº) (译注：Infinite Monkey Theorem) + +æ‰€ä»¥ä½ è¦ä¹ˆæ”¾å¼ƒ GNU emacs，è¦ä¹ˆæ”¹å˜å®ƒè®©å®ƒä½¿ç”¨æ›´åˆç†çš„设定。è¦é‡‡ç”¨åŽä¸€ä¸ªæ–¹æ¡ˆï¼Œ +ä½ å¯ä»¥æŠŠä¸‹é¢è¿™æ®µç²˜è´´åˆ°ä½ çš„ .emacs 文件里。 + +.. code-block:: none + + (defun c-lineup-arglist-tabs-only (ignored) + "Line up argument lists by tabs, not spaces" + (let* ((anchor (c-langelem-pos c-syntactic-element)) + (column (c-langelem-2nd-pos c-syntactic-element)) + (offset (- (1+ column) anchor)) + (steps (floor offset c-basic-offset))) + (* (max steps 1) + c-basic-offset))) + + (add-hook 'c-mode-common-hook + (lambda () + ;; Add kernel style + (c-add-style + "linux-tabs-only" + '("linux" (c-offsets-alist + (arglist-cont-nonempty + c-lineup-gcc-asm-reg + c-lineup-arglist-tabs-only)))))) + + (add-hook 'c-mode-hook + (lambda () + (let ((filename (buffer-file-name))) + ;; Enable kernel mode for the appropriate files + (when (and filename + (string-match (expand-file-name "~/src/linux-trees") + filename)) + (setq indent-tabs-mode t) + (setq show-trailing-whitespace t) + (c-set-style "linux-tabs-only"))))) + +这会让 emacs 在 ``~/src/linux-trees`` 下的 C æºæ–‡ä»¶èŽ·å¾—æ›´å¥½çš„å†…æ ¸ä»£ç é£Žæ ¼ã€‚ + +ä¸è¿‡å°±ç®—ä½ å°è¯•è®© emacs æ£ç¡®çš„æ ¼å¼åŒ–代ç 失败了，也并ä¸æ„味ç€ä½ å¤±åŽ»äº†ä¸€åˆ‡ï¼šè¿˜å¯ +以用 ``indent`` 。 + +ä¸è¿‡ï¼ŒGNU indent 也有和 GNU emacs ä¸€æ ·æœ‰é—®é¢˜çš„è®¾å®šï¼Œæ‰€ä»¥ä½ éœ€è¦ç»™å®ƒä¸€äº›å‘½ä»¤é€‰ +项。ä¸è¿‡ï¼Œè¿™è¿˜ä¸ç®—å¤ªç³Ÿç³•ï¼Œå› ä¸ºå°±ç®—æ˜¯ GNU indent çš„ä½œè€…ä¹Ÿè®¤åŒ K&R çš„æƒå¨æ€§ +(GNU 的人并ä¸æ˜¯å人，他们åªæ˜¯åœ¨è¿™ä¸ªé—®é¢˜ä¸Šè¢«ä¸¥é‡çš„误导了)ï¼Œæ‰€ä»¥ä½ åªè¦ç»™ indent +指定选项 ``-kr -i8`` (代表 ``K&R，8 å—符缩进``)，或使用 ``scripts/Lindent`` +è¿™æ ·å°±å¯ä»¥ä»¥æœ€æ—¶é«¦çš„æ–¹å¼ç¼©è¿›æºä»£ç 。 + +``indent`` 有很多选项，特别是é‡æ–°æ ¼å¼åŒ–æ³¨é‡Šçš„æ—¶å€™ï¼Œä½ å¯èƒ½éœ€è¦çœ‹ä¸€ä¸‹å®ƒçš„手册。 +ä¸è¿‡è®°ä½ï¼š ``indent`` ä¸èƒ½ä¿®æ£åçš„ç¼–ç¨‹ä¹ æƒ¯ã€‚ + + +10) Kconfig é…置文件 +------------------------------ + +对于é布æºç æ ‘çš„æ‰€æœ‰ Kconfig* é…置文件æ¥è¯´ï¼Œå®ƒä»¬ç¼©è¿›æ–¹å¼æœ‰æ‰€ä¸åŒã€‚ç´§æŒ¨ç€ +``config`` 定义的行，用一个制表符缩进，然而 help ä¿¡æ¯çš„缩进则é¢å¤–å¢žåŠ 2 个空 +æ ¼ã€‚ä¸¾ä¸ªä¾‹å:: + + config AUDIT + bool "Auditing support" + depends on NET + help + Enable auditing infrastructure that can be used with another + kernel subsystem, such as SELinux (which requires this for + logging of avc messages output). Does not do system-call + auditing without CONFIG_AUDITSYSCALL. + +而那些å±é™©çš„功能 (比如æŸäº›æ–‡ä»¶ç³»ç»Ÿçš„写支æŒ) 应该在它们的æ示å—符串里显著的声 +明这一点:: + + config ADFS_FS_RW + bool "ADFS write support (DANGEROUS)" + depends on ADFS_FS + ... + +è¦æŸ¥çœ‹é…置文件的完整文档，请看 Documentation/kbuild/kconfig-language.txt。 + + +11) æ•°æ®ç»“æž„ +------------------------------ + +如果一个数æ®ç»“构，在创建和销æ¯å®ƒçš„å•çº¿æ‰§è¡ŒçŽ¯å¢ƒä¹‹å¤–å¯è§ï¼Œé‚£ä¹ˆå®ƒå¿…é¡»è¦æœ‰ä¸€ä¸ªå¼• +ç”¨è®¡æ•°å™¨ã€‚å†…æ ¸é‡Œæ²¡æœ‰åžƒåœ¾æ”¶é›† (å¹¶ä¸”å†…æ ¸ä¹‹å¤–çš„åžƒåœ¾æ”¶é›†æ…¢ä¸”æ•ˆçŽ‡ä½Žä¸‹)，这æ„味ç€ä½ +ç»å¯¹éœ€è¦è®°å½•ä½ 对这ç§æ•°æ®ç»“构的使用情况。 + +引用计数æ„味ç€ä½ 能够é¿å…上é”，并且å…许多个用户并行访问这个数æ®ç»“构——而ä¸éœ€è¦ +担心这个数æ®ç»“æž„ä»…ä»…å› ä¸ºæš‚æ—¶ä¸è¢«ä½¿ç”¨å°±æ¶ˆå¤±äº†ï¼Œé‚£äº›ç”¨æˆ·å¯èƒ½ä¸è¿‡æ˜¯æ²‰ç¡äº†ä¸€é˜µæˆ– +者åšäº†ä¸€äº›å…¶ä»–事情而已。 + +注æ„ä¸Šé” **ä¸èƒ½** å–代引用计数。上é”是为了ä¿æŒæ•°æ®ç»“构的一致性，而引用计数是一 +个内å˜ç®¡ç†æŠ€å·§ã€‚通常二者都需è¦ï¼Œä¸è¦æŠŠä¸¤ä¸ªæžæ··äº†ã€‚ + +很多数æ®ç»“构实际上有 2 级引用计数，它们通常有ä¸åŒ ``ç±»`` 的用户。å类计数器统 +计å类用户的数é‡ï¼Œæ¯å½“å类计数器å‡è‡³é›¶æ—¶ï¼Œå…¨å±€è®¡æ•°å™¨å‡ä¸€ã€‚ + +è¿™ç§ ``多级引用计数`` 的例åå¯ä»¥åœ¨å†…å˜ç®¡ç† (``struct mm_struct``: mm_users å’Œ +mm_count)，和文件系统 (``struct super_block``: s_count å’Œ s_active) ä¸æ‰¾åˆ°ã€‚ + +è®°ä½ï¼šå¦‚æžœå¦ä¸€ä¸ªæ‰§è¡Œçº¿ç´¢å¯ä»¥æ‰¾åˆ°ä½ çš„æ•°æ®ç»“构，但这个数æ®ç»“构没有引用计数器， +è¿™é‡Œå‡ ä¹Žè‚¯å®šæ˜¯ä¸€ä¸ª bug。 + + +12) å®ï¼Œæžšä¸¾å’ŒRTL +------------------------------ + +用于定义常é‡çš„å®çš„åå—åŠæžšä¸¾é‡Œçš„æ ‡ç¾éœ€è¦å¤§å†™ã€‚ + +.. code-block:: c + + #define CONSTANT 0x12345 + +åœ¨å®šä¹‰å‡ ä¸ªç›¸å…³çš„å¸¸é‡æ—¶ï¼Œæœ€å¥½ç”¨æžšä¸¾ã€‚ + +å®çš„åå—请用大写å—æ¯ï¼Œä¸è¿‡å½¢å¦‚函数的å®çš„åå—å¯ä»¥ç”¨å°å†™å—æ¯ã€‚ + +一般的，如果能写æˆå†…è”函数就ä¸è¦å†™æˆåƒå‡½æ•°çš„å®ã€‚ + +å«æœ‰å¤šä¸ªè¯å¥çš„å®åº”该被包å«åœ¨ä¸€ä¸ª do-while 代ç å—里： + +.. code-block:: c + + #define macrofun(a, b, c) \ + do { \ + if (a == 5) \ + do_this(b, c); \ + } while (0) + +使用å®çš„时候应é¿å…的事情： + +1) å½±å“控制æµç¨‹çš„å®ï¼š + +.. code-block:: c + + #define FOO(x) \ + do { \ + if (blah(x) < 0) \ + return -EBUGGERED; \ + } while (0) + +**éžå¸¸** ä¸å¥½ã€‚它看起æ¥åƒä¸€ä¸ªå‡½æ•°ï¼Œä¸è¿‡å´èƒ½å¯¼è‡´ ``调用`` 它的函数退出；ä¸è¦æ‰“ +乱读者大脑里的è¯æ³•åˆ†æžå™¨ã€‚ + +2) ä¾èµ–于一个固定åå—的本地å˜é‡çš„å®ï¼š + +.. code-block:: c + + #define FOO(val) bar(index, val) + +å¯èƒ½çœ‹èµ·æ¥åƒæ˜¯ä¸ªä¸é”™çš„东西，ä¸è¿‡å®ƒéžå¸¸å®¹æ˜“把读代ç 的人æžç³Šæ¶‚，而且容易导致看起 +æ¥ä¸ç›¸å…³çš„改动带æ¥é”™è¯¯ã€‚ + +3) 作为左值的带å‚æ•°çš„å®ï¼š FOO(x) = y；如果有人把 FOO å˜æˆä¸€ä¸ªå†…è”函数的è¯ï¼Œè¿™ + ç§ç”¨æ³•å°±ä¼šå‡ºé”™äº†ã€‚ + +4) 忘记了优先级：使用表达å¼å®šä¹‰å¸¸é‡çš„å®å¿…须将表达å¼ç½®äºŽä¸€å¯¹å°æ‹¬å·ä¹‹å†…。带å‚æ•° + çš„å®ä¹Ÿè¦æ³¨æ„æ¤ç±»é—®é¢˜ã€‚ + +.. code-block:: c + + #define CONSTANT 0x4000 + #define CONSTEXP (CONSTANT | 3) + +5) 在å®é‡Œå®šä¹‰ç±»ä¼¼å‡½æ•°çš„本地å˜é‡æ—¶å‘½å冲çªï¼š + +.. code-block:: c + + #define FOO(x) \ + ({ \ + typeof(x) ret; \ + ret = calc_ret(x); \ + (ret); \ + }) + +ret 是本地å˜é‡çš„通用åå— - __foo_ret æ›´ä¸å®¹æ˜“与一个已å˜åœ¨çš„å˜é‡å†²çªã€‚ + +cpp 手册对å®çš„讲解很详细。gcc internals 手册也详细讲解了 RTLï¼Œå†…æ ¸é‡Œçš„æ±‡ç¼–è¯ +言ç»å¸¸ç”¨åˆ°å®ƒã€‚ + + +13) 打å°å†…æ ¸æ¶ˆæ¯ +------------------------------ + +å†…æ ¸å¼€å‘者应该是å—过良好教育的。请一定注æ„å†…æ ¸ä¿¡æ¯çš„拼写，以给人以好的å°è±¡ã€‚ +ä¸è¦ç”¨ä¸è§„范的å•è¯æ¯”如 ``dont``，而è¦ç”¨ ``do not`` 或者 ``don't`` 。ä¿è¯è¿™äº›ä¿¡ +æ¯ç®€å•æ˜Žäº†,æ— æ§ä¹‰ã€‚ + +å†…æ ¸ä¿¡æ¯ä¸å¿…以英文å¥å·ç»“æŸã€‚ + +在å°æ‹¬å·é‡Œæ‰“å°æ•°å— (%d) 没有任何价值，应该é¿å…è¿™æ ·åšã€‚ + +<linux/device.h> 里有一些驱动模型诊æ–å®ï¼Œä½ 应该使用它们，以确ä¿ä¿¡æ¯å¯¹åº”于æ£ç¡® +çš„è®¾å¤‡å’Œé©±åŠ¨ï¼Œå¹¶ä¸”è¢«æ ‡è®°äº†æ£ç¡®çš„消æ¯çº§åˆ«ã€‚这些å®æœ‰ï¼šdev_err(), dev_warn(), +dev_info() ç‰ç‰ã€‚对于那些ä¸å’ŒæŸä¸ªç‰¹å®šè®¾å¤‡ç›¸å…³è¿žçš„ä¿¡æ¯ï¼Œ<linux/printk.h> 定义 +了 pr_notice(), pr_info(), pr_warn(), pr_err() 和其他。 + +写出好的调试信æ¯å¯ä»¥æ˜¯ä¸€ä¸ªå¾ˆå¤§çš„æŒ‘æˆ˜ï¼›ä¸€æ—¦ä½ å†™å‡ºåŽï¼Œè¿™äº›ä¿¡æ¯åœ¨è¿œç¨‹é™¤é”™æ—¶èƒ½æ +ä¾›æžå¤§çš„帮助。然而打å°è°ƒè¯•ä¿¡æ¯çš„处ç†æ–¹å¼åŒæ‰“å°éžè°ƒè¯•ä¿¡æ¯ä¸åŒã€‚其他 pr_XXX() +å‡½æ•°èƒ½æ— æ¡ä»¶åœ°æ‰“å°ï¼Œpr_debug() å´ä¸ï¼›é»˜è®¤æƒ…况下它ä¸ä¼šè¢«ç¼–译，除éžå®šä¹‰äº† DEBUG +或设定了 CONFIG_DYNAMIC_DEBUG。实际这åŒæ ·æ˜¯ä¸ºäº† dev_dbg()，一个相关约定是在一 +个已ç»å¼€å¯äº† DEBUG 时，使用 VERBOSE_DEBUG æ¥æ·»åŠ dev_vdbg()。 + +许多å系统拥有 Kconfig 调试选项æ¥å¼€å¯ -DDEBUG 在对应的 Makefile 里é¢ï¼›åœ¨å…¶ä»– +情况下，特殊文件使用 #define DEBUG。当一æ¡è°ƒè¯•ä¿¡æ¯éœ€è¦è¢«æ— æ¡ä»¶æ‰“å°æ—¶ï¼Œä¾‹å¦‚， +如果已ç»åŒ…å«ä¸€ä¸ªè°ƒè¯•ç›¸å…³çš„ #ifdef æ¡ä»¶ï¼Œprintk(KERN_DEBUG ...) å°±å¯è¢«ä½¿ç”¨ã€‚ + + +14) 分é…å†…å˜ +------------------------------ + +å†…æ ¸æ供了下é¢çš„一般用途的内å˜åˆ†é…函数： +kmalloc(), kzalloc(), kmalloc_array(), kcalloc(), vmalloc() å’Œ vzalloc()。 +请å‚考 API 文档以获å–有关它们的详细信æ¯ã€‚ + +ä¼ é€’ç»“æž„ä½“å¤§å°çš„首选形å¼æ˜¯è¿™æ ·çš„： + +.. code-block:: c + + p = kmalloc(sizeof(*p), ...); + +å¦å¤–一ç§ä¼ 递方å¼ä¸ï¼Œsizeof çš„æ“作数是结构体的åå—ï¼Œè¿™æ ·ä¼šé™ä½Žå¯è¯»æ€§ï¼Œå¹¶ä¸”å¯èƒ½ +会引入 bug。有å¯èƒ½æŒ‡é’ˆå˜é‡ç±»åž‹è¢«æ”¹å˜æ—¶ï¼Œè€Œå¯¹åº”çš„ä¼ é€’ç»™å†…å˜åˆ†é…函数的 sizeof +的结果ä¸å˜ã€‚ + +强制转æ¢ä¸€ä¸ª void 指针返回值是多余的。C è¯è¨€æœ¬èº«ä¿è¯äº†ä»Ž void 指针到其他任何 +指针类型的转æ¢æ˜¯æ²¡æœ‰é—®é¢˜çš„。 + +分é…一个数组的首选形å¼æ˜¯è¿™æ ·çš„： + +.. code-block:: c + + p = kmalloc_array(n, sizeof(...), ...); + +分é…一个零长数组的首选形å¼æ˜¯è¿™æ ·çš„： + +.. code-block:: c + + p = kcalloc(n, sizeof(...), ...); + +两ç§å½¢å¼æ£€æŸ¥åˆ†é…å¤§å° n * sizeof(...) 的溢出，如果溢出返回 NULL。 + + +15) 内è”弊病 +------------------------------ + +有一个常è§çš„误解是 ``内è”`` 是 gcc æ供的å¯ä»¥è®©ä»£ç è¿è¡Œæ›´å¿«çš„一个选项。虽然使 +用内è”函数有时候是æ°å½“çš„ (比如作为一ç§æ›¿ä»£å®çš„æ–¹å¼ï¼Œè¯·çœ‹ç¬¬åäºŒç« )，ä¸è¿‡å¾ˆå¤šæƒ… +况下ä¸æ˜¯è¿™æ ·ã€‚inline çš„è¿‡åº¦ä½¿ç”¨ä¼šä½¿å†…æ ¸å˜å¤§ï¼Œä»Žè€Œä½¿æ•´ä¸ªç³»ç»Ÿè¿è¡Œé€Ÿåº¦å˜æ…¢ã€‚ +å› ä¸ºä½“ç§¯å¤§å†…æ ¸ä¼šå 用更多的指令高速缓å˜ï¼Œè€Œä¸”会导致 pagecache çš„å¯ç”¨å†…å˜å‡å°‘。 +想象一下，一次 pagecache 未命ä¸å°±ä¼šå¯¼è‡´ä¸€æ¬¡ç£ç›˜å¯»å€ï¼Œå°†è€—æ—¶ 5 毫秒。5 毫秒的 +时间内 CPU 能执行很多很多指令。 + +一个基本的原则是如果一个函数有 3 行以上，就ä¸è¦æŠŠå®ƒå˜æˆå†…è”函数。这个原则的一 +ä¸ªä¾‹å¤–æ˜¯ï¼Œå¦‚æžœä½ çŸ¥é“æŸä¸ªå‚数是一个编译时常é‡ï¼Œè€Œä¸”å› ä¸ºè¿™ä¸ªå¸¸é‡ä½ 确定编译器在 +ç¼–è¯‘æ—¶èƒ½ä¼˜åŒ–æŽ‰ä½ çš„å‡½æ•°çš„å¤§éƒ¨åˆ†ä»£ç ，那ä»ç„¶å¯ä»¥ç»™å®ƒåŠ 上 inline 关键å—。 +kmalloc() 内è”函数就是一个很好的例å。 + +人们ç»å¸¸ä¸»å¼ ç»™ static 的而且åªç”¨äº†ä¸€æ¬¡çš„å‡½æ•°åŠ ä¸Š inline，如æ¤ä¸ä¼šæœ‰ä»»ä½•æŸå¤±ï¼Œ +å› ä¸ºæ²¡æœ‰ä»€ä¹ˆå¥½æƒè¡¡çš„。虽然从技术上说这是æ£ç¡®çš„，但是实际上这ç§æƒ…况下å³ä½¿ä¸åŠ +inline gcc 也å¯ä»¥è‡ªåŠ¨ä½¿å…¶å†…è”。而且其他用户å¯èƒ½ä¼šè¦æ±‚移除 inline，由æ¤è€Œæ¥çš„ +争论会抵消 inline 自身的潜在价值，得ä¸å¿å¤±ã€‚ + + +16) 函数返回值åŠå‘½å +------------------------------ + +函数å¯ä»¥è¿”回多ç§ä¸åŒç±»åž‹çš„值，最常è§çš„一ç§æ˜¯è¡¨æ˜Žå‡½æ•°æ‰§è¡ŒæˆåŠŸæˆ–è€…å¤±è´¥çš„å€¼ã€‚è¿™æ · +的一个值å¯ä»¥è¡¨ç¤ºä¸ºä¸€ä¸ªé”™è¯¯ä»£ç æ•´æ•° (-Exxxï¼å¤±è´¥ï¼Œ0ï¼æˆåŠŸ) 或者一个 ``æˆåŠŸ`` +布尔值 (0ï¼å¤±è´¥ï¼Œéž0ï¼æˆåŠŸ)。 + +æ··åˆä½¿ç”¨è¿™ä¸¤ç§è¡¨è¾¾æ–¹å¼æ˜¯éš¾äºŽå‘现的 bug çš„æ¥æºã€‚如果 C è¯è¨€æœ¬èº«ä¸¥æ ¼åŒºåˆ†æ•´å½¢å’Œ +布尔型å˜é‡ï¼Œé‚£ä¹ˆç¼–译器就能够帮我们å‘现这些错误... ä¸è¿‡ C è¯è¨€ä¸åŒºåˆ†ã€‚为了é¿å… +äº§ç”Ÿè¿™ç§ bug，请éµå¾ªä¸‹é¢çš„惯例:: + + 如果函数的åå—是一个动作或者强制性的命令，那么这个函数应该返回错误代 + ç 整数。如果是一个判æ–，那么函数应该返回一个 "æˆåŠŸ" 布尔值。 + +比如， ``add work`` 是一个命令，所以 add_work() 在æˆåŠŸæ—¶è¿”回 0，在失败时返回 +-EBUSYã€‚ç±»ä¼¼çš„ï¼Œå› ä¸º ``PCI device present`` 是一个判æ–，所以 pci_dev_present() +在æˆåŠŸæ‰¾åˆ°ä¸€ä¸ªåŒ¹é…的设备时应该返回 1，如果找ä¸åˆ°æ—¶åº”该返回 0。 + +所有 EXPORTed 函数都必须éµå®ˆè¿™ä¸ªæƒ¯ä¾‹ï¼Œæ‰€æœ‰çš„公共函数也都应该如æ¤ã€‚ç§æœ‰ +(static) 函数ä¸éœ€è¦å¦‚æ¤ï¼Œä½†æ˜¯æˆ‘们也推èè¿™æ ·åšã€‚ + +返回值是实际计算结果而ä¸æ˜¯è®¡ç®—是å¦æˆåŠŸçš„æ ‡å¿—çš„å‡½æ•°ä¸å—æ¤æƒ¯ä¾‹çš„é™åˆ¶ã€‚一般的， +他们通过返回一些æ£å¸¸å€¼èŒƒå›´ä¹‹å¤–的结果æ¥è¡¨ç¤ºå‡ºé”™ã€‚典型的例å是返回指针的函数， +他们使用 NULL 或者 ERR_PTR 机制æ¥æŠ¥å‘Šé”™è¯¯ã€‚ + + +17) ä¸è¦é‡æ–°å‘æ˜Žå†…æ ¸å® +------------------------------ + +头文件 include/linux/kernel.h 包å«äº†ä¸€äº›å®ï¼Œä½ 应该使用它们，而ä¸è¦è‡ªå·±å†™ä¸€äº› +它们的å˜ç§ã€‚æ¯”å¦‚ï¼Œå¦‚æžœä½ éœ€è¦è®¡ç®—ä¸€ä¸ªæ•°ç»„çš„é•¿åº¦ï¼Œä½¿ç”¨è¿™ä¸ªå® + +.. code-block:: c + + #define ARRAY_SIZE(x) (sizeof(x) / sizeof((x)[0])) + +ç±»ä¼¼çš„ï¼Œå¦‚æžœä½ è¦è®¡ç®—æŸç»“构体æˆå‘˜çš„大å°ï¼Œä½¿ç”¨ + +.. code-block:: c + + #define FIELD_SIZEOF(t, f) (sizeof(((t*)0)->f)) + +还有å¯ä»¥åšä¸¥æ ¼çš„类型检查的 min() å’Œ max() å®ï¼Œå¦‚æžœä½ éœ€è¦å¯ä»¥ä½¿ç”¨å®ƒä»¬ã€‚ä½ å¯ä»¥ +è‡ªå·±çœ‹çœ‹é‚£ä¸ªå¤´æ–‡ä»¶é‡Œè¿˜å®šä¹‰äº†ä»€ä¹ˆä½ å¯ä»¥æ‹¿æ¥ç”¨çš„东西，如果有定义的è¯ï¼Œä½ å°±ä¸åº” +åœ¨ä½ çš„ä»£ç 里自己é‡æ–°å®šä¹‰ã€‚ + + +18) 编辑器模å¼è¡Œå’Œå…¶ä»–需è¦ç½—嗦的事情 +-------------------------------------------------- + +有一些编辑器å¯ä»¥è§£é‡ŠåµŒå…¥åœ¨æºæ–‡ä»¶é‡Œçš„ç”±ä¸€äº›ç‰¹æ®Šæ ‡è®°æ ‡æ˜Žçš„é…置信æ¯ã€‚比如，emacs +èƒ½å¤Ÿè§£é‡Šè¢«æ ‡è®°æˆè¿™æ ·çš„行： + +.. code-block:: c + + -*- mode: c -*- + +æˆ–è€…è¿™æ ·çš„ï¼š + +.. code-block:: c + + /* + Local Variables: + compile-command: "gcc -DMAGIC_DEBUG_FLAG foo.c" + End: + */ + +Vim èƒ½å¤Ÿè§£é‡Šè¿™æ ·çš„æ ‡è®°ï¼š + +.. code-block:: c + + /* vim:set sw=8 noet */ + +ä¸è¦åœ¨æºä»£ç ä¸åŒ…å«ä»»ä½•è¿™æ ·çš„内容。æ¯ä¸ªäººéƒ½æœ‰ä»–自己的编辑器é…ç½®ï¼Œä½ çš„æºæ–‡ä»¶ä¸ +应该覆盖别人的é…置。这包括有关缩进和模å¼é…ç½®çš„æ ‡è®°ã€‚äººä»¬å¯ä»¥ä½¿ç”¨ä»–们自己定制 +的模å¼ï¼Œæˆ–者使用其他å¯ä»¥äº§ç”Ÿæ£ç¡®çš„缩进的巧妙方法。 + + +19) 内è”汇编 +------------------------------ + +在特定架构的代ç ä¸ï¼Œä½ å¯èƒ½éœ€è¦å†…è”汇编与 CPU 和平å°ç›¸å…³åŠŸèƒ½è¿žæŽ¥ã€‚需è¦è¿™ä¹ˆåšæ—¶ +å°±ä¸è¦çŠ¹è±«ã€‚然而，当 C å¯ä»¥å®Œæˆå·¥ä½œæ—¶ï¼Œä¸è¦å¹³ç™½æ— 故地使用内è”汇编。在å¯èƒ½çš„情 +å†µä¸‹ï¼Œä½ å¯ä»¥å¹¶ä¸”应该用 C 和硬件沟通。 + +请考虑去写æ†ç»‘通用ä½å…ƒ (wrap common bits) 的内è”汇编的简å•è¾…助函数，别去é‡å¤ +地写下åªæœ‰ç»†å¾®å·®å¼‚内è”汇编。记ä½å†…è”汇编å¯ä»¥ä½¿ç”¨ C å‚数。 + +大型，有一定å¤æ‚度的汇编函数应该放在 .S 文件内，用相应的 C 原型定义在 C 头文 +件ä¸ã€‚汇编函数的 C 原型应该使用 ``asmlinkage`` 。 + +ä½ å¯èƒ½éœ€è¦æŠŠæ±‡ç¼–è¯å¥æ ‡è®°ä¸º volatile，用æ¥é˜»æ¢ GCC 在没å‘现任何副作用åŽå°±æŠŠå®ƒ +ç§»é™¤äº†ã€‚ä½ ä¸å¿…æ€»æ˜¯è¿™æ ·åšï¼Œå°½ç®¡ï¼Œè¿™ä¸å¿…è¦çš„举动会é™åˆ¶ä¼˜åŒ–。 + +在写一个包å«å¤šæ¡æŒ‡ä»¤çš„å•ä¸ªå†…è”汇编è¯å¥æ—¶ï¼ŒæŠŠæ¯æ¡æŒ‡ä»¤ç”¨å¼•å·åˆ†å‰²è€Œä¸”å„å 一行， +除了最åŽä¸€æ¡æŒ‡ä»¤å¤–，在æ¯ä¸ªæŒ‡ä»¤ç»“å°¾åŠ ä¸Š \n\t，让汇编输出时å¯ä»¥æ£ç¡®åœ°ç¼©è¿›ä¸‹ä¸€æ¡ +指令： + +.. code-block:: c + + asm ("magic %reg1, #42\n\t" + "more_magic %reg2, %reg3" + : /* outputs */ : /* inputs */ : /* clobbers */); + + +20) æ¡ä»¶ç¼–译 +------------------------------ + +åªè¦å¯èƒ½ï¼Œå°±ä¸è¦åœ¨ .c 文件里é¢ä½¿ç”¨é¢„处ç†æ¡ä»¶ (#if, #ifdef)ï¼›è¿™æ ·åšè®©ä»£ç æ›´éš¾ +阅读并且更难去跟踪逻辑。替代方案是，在头文件ä¸ç”¨é¢„处ç†æ¡ä»¶æ供给那些 .c 文件 +使用，å†ç»™ #else æ供一个空桩 (no-op stub) 版本，然åŽåœ¨ .c æ–‡ä»¶å†…æ— æ¡ä»¶åœ°è°ƒç”¨ +那些 (定义在头文件内的) å‡½æ•°ã€‚è¿™æ ·åšï¼Œç¼–译器会é¿å…为桩函数 (stub) çš„è°ƒç”¨ç”Ÿæˆ +任何代ç ，产生的结果是相åŒçš„ï¼Œä½†é€»è¾‘å°†æ›´åŠ æ¸…æ™°ã€‚ + +最好倾å‘于编译整个函数，而ä¸æ˜¯å‡½æ•°çš„一部分或表达å¼çš„一部分。与其放一个 ifdef +在表达å¼å†…，ä¸å¦‚分解出部分或全部表达å¼ï¼Œæ”¾è¿›ä¸€ä¸ªå•ç‹¬çš„è¾…åŠ©å‡½æ•°ï¼Œå¹¶åº”ç”¨é¢„å¤„ç† +æ¡ä»¶åˆ°è¿™ä¸ªè¾…助函数内。 + +å¦‚æžœä½ æœ‰ä¸€ä¸ªåœ¨ç‰¹å®šé…ç½®ä¸ï¼Œå¯èƒ½å˜æˆæœªä½¿ç”¨çš„函数或å˜é‡ï¼Œç¼–译器会è¦å‘Šå®ƒå®šä¹‰äº†ä½† +æœªä½¿ç”¨ï¼ŒæŠŠå®ƒæ ‡è®°ä¸º __maybe_unused 而ä¸æ˜¯å°†å®ƒåŒ…å«åœ¨ä¸€ä¸ªé¢„处ç†æ¡ä»¶ä¸ã€‚(然而，如 +果一个函数或å˜é‡æ€»æ˜¯æœªä½¿ç”¨ï¼Œå°±ç›´æŽ¥åˆ 除它。) + +在代ç ä¸ï¼Œå°½å¯èƒ½åœ°ä½¿ç”¨ IS_ENABLED å®æ¥è½¬åŒ–æŸä¸ª Kconfig æ ‡è®°ä¸º C 的布尔 +表达å¼ï¼Œå¹¶åœ¨ä¸€èˆ¬çš„ C æ¡ä»¶ä¸ä½¿ç”¨å®ƒï¼š + +.. code-block:: c + + if (IS_ENABLED(CONFIG_SOMETHING)) { + ... + } + +编译器会åšå¸¸é‡æŠ˜å ，然åŽå°±åƒä½¿ç”¨ #ifdef é‚£æ ·åŽ»åŒ…å«æˆ–排除代ç å—，所以这ä¸ä¼šå¸¦ +æ¥ä»»ä½•è¿è¡Œæ—¶å¼€é”€ã€‚然而，这ç§æ–¹æ³•ä¾æ—§å…许 C 编译器查看å—内的代ç ï¼Œå¹¶æ£€æŸ¥å®ƒçš„æ£ +确性 (è¯æ³•ï¼Œç±»åž‹ï¼Œç¬¦å·å¼•ç”¨ï¼Œç‰ç‰)ã€‚å› æ¤ï¼Œå¦‚æžœæ¡ä»¶ä¸æ»¡è¶³ï¼Œä»£ç å—内的引用符å·å°± +ä¸å˜åœ¨æ—¶ï¼Œä½ 还是必须去用 #ifdef。 + +在任何有æ„义的 #if 或 #ifdef å—的末尾 (è¶…è¿‡å‡ è¡Œçš„)，在 #endif åŒä¸€è¡Œçš„åŽé¢å†™ä¸‹ +注解，注释这个æ¡ä»¶è¡¨è¾¾å¼ã€‚例如： + +.. code-block:: c + + #ifdef CONFIG_SOMETHING + ... + #endif /* CONFIG_SOMETHING */ + + +附录 I) å‚考 +------------------- + +The C Programming Language, 第二版 +作者：Brian W. Kernighan å’Œ Denni M. Ritchie. +Prentice Hall, Inc., 1988. +ISBN 0-13-110362-8 (软皮), 0-13-110370-9 (硬皮). + +The Practice of Programming +作者：Brian W. Kernighan å’Œ Rob Pike. +Addison-Wesley, Inc., 1999. +ISBN 0-201-61586-X. + +GNU 手册 - éµå¾ª K&R æ ‡å‡†å’Œæ¤æ–‡æœ¬ - cpp, gcc, gcc internals and indent, +都å¯ä»¥ä»Ž http://www.gnu.org/manual/ 找到 + +WG14 是 C è¯è¨€çš„å›½é™…æ ‡å‡†åŒ–å·¥ä½œç»„ï¼ŒURL: http://www.open-std.org/JTC1/SC22/WG14/ + +Kernel process/coding-style.rst，作者 greg@kroah.com å‘表于 OLS 2002： +http://www.kroah.com/linux/talks/ols_2002_kernel_codingstyle_talk/html/ diff --git a/Documentation/translations/zh_CN/index.rst b/Documentation/translations/zh_CN/index.rst new file mode 100644 index 000000000000..75956d669962 --- /dev/null +++ b/Documentation/translations/zh_CN/index.rst @@ -0,0 +1,12 @@ +.. raw:: latex + + \renewcommand\thesection* + \renewcommand\thesubsection* + +Chinese translations +==================== + +.. toctree:: + :maxdepth: 1 + + coding-style diff --git a/Documentation/usb/power-management.txt b/Documentation/usb/power-management.txt index 0a94ffe17ab6..00e706997130 100644 --- a/Documentation/usb/power-management.txt +++ b/Documentation/usb/power-management.txt @@ -543,7 +543,7 @@ relevant attribute files are usb2_hardware_lpm and usb3_hardware_lpm. When a USB 3.0 lpm-capable device is plugged in to a xHCI host which supports link PM, it will check if U1 and U2 exit latencies have been set in the BOS - descriptor; if the check is is passed and the host + descriptor; if the check is passed and the host supports USB3 hardware LPM, USB3 hardware LPM will be enabled for the device and these files will be created. The files hold a string value (enable or disable) diff --git a/Documentation/virtual/kvm/api.txt b/Documentation/virtual/kvm/api.txt index 4470671b0c26..069450938b79 100644 --- a/Documentation/virtual/kvm/api.txt +++ b/Documentation/virtual/kvm/api.txt @@ -2061,6 +2061,8 @@ registers, find a list below: MIPS | KVM_REG_MIPS_LO | 64 MIPS | KVM_REG_MIPS_PC | 64 MIPS | KVM_REG_MIPS_CP0_INDEX | 32 + MIPS | KVM_REG_MIPS_CP0_ENTRYLO0 | 64 + MIPS | KVM_REG_MIPS_CP0_ENTRYLO1 | 64 MIPS | KVM_REG_MIPS_CP0_CONTEXT | 64 MIPS | KVM_REG_MIPS_CP0_USERLOCAL | 64 MIPS | KVM_REG_MIPS_CP0_PAGEMASK | 32 @@ -2071,9 +2073,11 @@ registers, find a list below: MIPS | KVM_REG_MIPS_CP0_ENTRYHI | 64 MIPS | KVM_REG_MIPS_CP0_COMPARE | 32 MIPS | KVM_REG_MIPS_CP0_STATUS | 32 + MIPS | KVM_REG_MIPS_CP0_INTCTL | 32 MIPS | KVM_REG_MIPS_CP0_CAUSE | 32 MIPS | KVM_REG_MIPS_CP0_EPC | 64 MIPS | KVM_REG_MIPS_CP0_PRID | 32 + MIPS | KVM_REG_MIPS_CP0_EBASE | 64 MIPS | KVM_REG_MIPS_CP0_CONFIG | 32 MIPS | KVM_REG_MIPS_CP0_CONFIG1 | 32 MIPS | KVM_REG_MIPS_CP0_CONFIG2 | 32 @@ -2148,6 +2152,12 @@ patterns depending on whether they're 32-bit or 64-bit registers: 0x7020 0000 0001 00 <reg:5> <sel:3> (32-bit) 0x7030 0000 0001 00 <reg:5> <sel:3> (64-bit) +Note: KVM_REG_MIPS_CP0_ENTRYLO0 and KVM_REG_MIPS_CP0_ENTRYLO1 are the MIPS64 +versions of the EntryLo registers regardless of the word size of the host +hardware, host kernel, guest, and whether XPA is present in the guest, i.e. +with the RI and XI bits (if they exist) in bits 63 and 62 respectively, and +the PFNX field starting at bit 30. + MIPS KVM control registers (see above) have the following id bit patterns: 0x7030 0000 0002 <reg:16> @@ -2443,18 +2453,20 @@ are, it will do nothing and return an EBUSY error. The parameter is a pointer to a 32-bit unsigned integer variable containing the order (log base 2) of the desired size of the hash table, which must be between 18 and 46. On successful return from the -ioctl, it will have been updated with the order of the hash table that -was allocated. +ioctl, the value will not be changed by the kernel. If no hash table has been allocated when any vcpu is asked to run (with the KVM_RUN ioctl), the host kernel will allocate a default-sized hash table (16 MB). If this ioctl is called when a hash table has already been allocated, -the kernel will clear out the existing hash table (zero all HPTEs) and -return the hash table order in the parameter. (If the guest is using -the virtualized real-mode area (VRMA) facility, the kernel will -re-create the VMRA HPTEs on the next KVM_RUN of any vcpu.) +with a different order from the existing hash table, the existing hash +table will be freed and a new one allocated. If this is ioctl is +called when a hash table has already been allocated of the same order +as specified, the kernel will clear out the existing hash table (zero +all HPTEs). In either case, if the guest is using the virtualized +real-mode area (VRMA) facility, the kernel will re-create the VMRA +HPTEs on the next KVM_RUN of any vcpu. 4.77 KVM_S390_INTERRUPT @@ -3177,7 +3189,7 @@ of IOMMU pages. The rest of functionality is identical to KVM_CREATE_SPAPR_TCE. -4.98 KVM_REINJECT_CONTROL +4.99 KVM_REINJECT_CONTROL Capability: KVM_CAP_REINJECT_CONTROL Architectures: x86 @@ -3201,7 +3213,7 @@ struct kvm_reinject_control { pit_reinject = 0 (!reinject mode) is recommended, unless running an old operating system that uses the PIT for timing (e.g. Linux 2.4.x). -4.99 KVM_PPC_CONFIGURE_V3_MMU +4.100 KVM_PPC_CONFIGURE_V3_MMU Capability: KVM_CAP_PPC_RADIX_MMU or KVM_CAP_PPC_HASH_MMU_V3 Architectures: ppc @@ -3232,7 +3244,7 @@ process table, which is in the guest's space. This field is formatted as the second doubleword of the partition table entry, as defined in the Power ISA V3.00, Book III section 5.7.6.1. -4.100 KVM_PPC_GET_RMMU_INFO +4.101 KVM_PPC_GET_RMMU_INFO Capability: KVM_CAP_PPC_RADIX_MMU Architectures: ppc @@ -3266,6 +3278,101 @@ The ap_encodings gives the supported page sizes and their AP field encodings, encoded with the AP value in the top 3 bits and the log base 2 of the page size in the bottom 6 bits. +4.102 KVM_PPC_RESIZE_HPT_PREPARE + +Capability: KVM_CAP_SPAPR_RESIZE_HPT +Architectures: powerpc +Type: vm ioctl +Parameters: struct kvm_ppc_resize_hpt (in) +Returns: 0 on successful completion, + >0 if a new HPT is being prepared, the value is an estimated + number of milliseconds until preparation is complete + -EFAULT if struct kvm_reinject_control cannot be read, + -EINVAL if the supplied shift or flags are invalid + -ENOMEM if unable to allocate the new HPT + -ENOSPC if there was a hash collision when moving existing + HPT entries to the new HPT + -EIO on other error conditions + +Used to implement the PAPR extension for runtime resizing of a guest's +Hashed Page Table (HPT). Specifically this starts, stops or monitors +the preparation of a new potential HPT for the guest, essentially +implementing the H_RESIZE_HPT_PREPARE hypercall. + +If called with shift > 0 when there is no pending HPT for the guest, +this begins preparation of a new pending HPT of size 2^(shift) bytes. +It then returns a positive integer with the estimated number of +milliseconds until preparation is complete. + +If called when there is a pending HPT whose size does not match that +requested in the parameters, discards the existing pending HPT and +creates a new one as above. + +If called when there is a pending HPT of the size requested, will: + * If preparation of the pending HPT is already complete, return 0 + * If preparation of the pending HPT has failed, return an error + code, then discard the pending HPT. + * If preparation of the pending HPT is still in progress, return an + estimated number of milliseconds until preparation is complete. + +If called with shift == 0, discards any currently pending HPT and +returns 0 (i.e. cancels any in-progress preparation). + +flags is reserved for future expansion, currently setting any bits in +flags will result in an -EINVAL. + +Normally this will be called repeatedly with the same parameters until +it returns <= 0. The first call will initiate preparation, subsequent +ones will monitor preparation until it completes or fails. + +struct kvm_ppc_resize_hpt { + __u64 flags; + __u32 shift; + __u32 pad; +}; + +4.103 KVM_PPC_RESIZE_HPT_COMMIT + +Capability: KVM_CAP_SPAPR_RESIZE_HPT +Architectures: powerpc +Type: vm ioctl +Parameters: struct kvm_ppc_resize_hpt (in) +Returns: 0 on successful completion, + -EFAULT if struct kvm_reinject_control cannot be read, + -EINVAL if the supplied shift or flags are invalid + -ENXIO is there is no pending HPT, or the pending HPT doesn't + have the requested size + -EBUSY if the pending HPT is not fully prepared + -ENOSPC if there was a hash collision when moving existing + HPT entries to the new HPT + -EIO on other error conditions + +Used to implement the PAPR extension for runtime resizing of a guest's +Hashed Page Table (HPT). Specifically this requests that the guest be +transferred to working with the new HPT, essentially implementing the +H_RESIZE_HPT_COMMIT hypercall. + +This should only be called after KVM_PPC_RESIZE_HPT_PREPARE has +returned 0 with the same parameters. In other cases +KVM_PPC_RESIZE_HPT_COMMIT will return an error (usually -ENXIO or +-EBUSY, though others may be possible if the preparation was started, +but failed). + +This will have undefined effects on the guest if it has not already +placed itself in a quiescent state where no vcpu will make MMU enabled +memory accesses. + +On succsful completion, the pending HPT will become the guest's active +HPT and the previous HPT will be discarded. + +On failure, the guest will still be operating on its previous HPT. + +struct kvm_ppc_resize_hpt { + __u64 flags; + __u32 shift; + __u32 pad; +}; + 5. The kvm_run structure ------------------------ @@ -3282,7 +3389,18 @@ struct kvm_run { Request that KVM_RUN return when it becomes possible to inject external interrupts into the guest. Useful in conjunction with KVM_INTERRUPT. - __u8 padding1[7]; + __u8 immediate_exit; + +This field is polled once when KVM_RUN starts; if non-zero, KVM_RUN +exits immediately, returning -EINTR. In the common scenario where a +signal is used to "kick" a VCPU out of KVM_RUN, this field can be used +to avoid usage of KVM_SET_SIGNAL_MASK, which has worse scalability. +Rather than blocking the signal outside KVM_RUN, userspace can set up +a signal handler that sets run->immediate_exit to a non-zero value. + +This field is ignored if KVM_CAP_IMMEDIATE_EXIT is not available. + + __u8 padding1[6]; /* out */ __u32 exit_reason; diff --git a/Documentation/virtual/kvm/devices/arm-vgic-v3.txt b/Documentation/virtual/kvm/devices/arm-vgic-v3.txt index 9348b3caccd7..c1a24612c198 100644 --- a/Documentation/virtual/kvm/devices/arm-vgic-v3.txt +++ b/Documentation/virtual/kvm/devices/arm-vgic-v3.txt @@ -118,7 +118,7 @@ Groups: -EBUSY: One or more VCPUs are running - KVM_DEV_ARM_VGIC_CPU_SYSREGS + KVM_DEV_ARM_VGIC_GRP_CPU_SYSREGS Attributes: The attr field of kvm_device_attr encodes two values: bits: | 63 .... 32 | 31 .... 16 | 15 .... 0 | @@ -139,13 +139,15 @@ Groups: All system regs accessed through this API are (rw, 64-bit) and kvm_device_attr.addr points to a __u64 value. - KVM_DEV_ARM_VGIC_CPU_SYSREGS accesses the CPU interface registers for the + KVM_DEV_ARM_VGIC_GRP_CPU_SYSREGS accesses the CPU interface registers for the CPU specified by the mpidr field. + CPU interface registers access is not implemented for AArch32 mode. + Error -ENXIO is returned when accessed in AArch32 mode. Errors: -ENXIO: Getting or setting this register is not yet supported -EBUSY: VCPU is running - -EINVAL: Invalid mpidr supplied + -EINVAL: Invalid mpidr or register value supplied KVM_DEV_ARM_VGIC_GRP_NR_IRQS @@ -204,3 +206,6 @@ Groups: architecture defined MPIDR, and the field is encoded as follows: | 63 .... 56 | 55 .... 48 | 47 .... 40 | 39 .... 32 | | Aff3 | Aff2 | Aff1 | Aff0 | + Errors: + -EINVAL: vINTID is not multiple of 32 or + info field is not VGIC_LEVEL_INFO_LINE_LEVEL diff --git a/Documentation/virtual/kvm/hypercalls.txt b/Documentation/virtual/kvm/hypercalls.txt index c8d040e27046..feaaa634f154 100644 --- a/Documentation/virtual/kvm/hypercalls.txt +++ b/Documentation/virtual/kvm/hypercalls.txt @@ -81,3 +81,38 @@ the vcpu to sleep until occurrence of an appropriate event. Another vcpu of the same guest can wakeup the sleeping vcpu by issuing KVM_HC_KICK_CPU hypercall, specifying APIC ID (a1) of the vcpu to be woken up. An additional argument (a0) is used in the hypercall for future use. + + +6. KVM_HC_CLOCK_PAIRING +------------------------ +Architecture: x86 +Status: active +Purpose: Hypercall used to synchronize host and guest clocks. +Usage: + +a0: guest physical address where host copies +"struct kvm_clock_offset" structure. + +a1: clock_type, ATM only KVM_CLOCK_PAIRING_WALLCLOCK (0) +is supported (corresponding to the host's CLOCK_REALTIME clock). + + struct kvm_clock_pairing { + __s64 sec; + __s64 nsec; + __u64 tsc; + __u32 flags; + __u32 pad[9]; + }; + + Where: + * sec: seconds from clock_type clock. + * nsec: nanoseconds from clock_type clock. + * tsc: guest TSC value used to calculate sec/nsec pair + * flags: flags, unused (0) at the moment. + +The hypercall lets a guest compute a precise timestamp across +host and guest. The guest can use the returned TSC value to +compute the CLOCK_REALTIME for its clock, at the same instant. + +Returns KVM_EOPNOTSUPP if the host does not use TSC clocksource, +or if clock type is different than KVM_CLOCK_PAIRING_WALLCLOCK. diff --git a/Documentation/virtual/kvm/locking.txt b/Documentation/virtual/kvm/locking.txt index fd013bf4115b..1bb8bcaf8497 100644 --- a/Documentation/virtual/kvm/locking.txt +++ b/Documentation/virtual/kvm/locking.txt @@ -26,9 +26,16 @@ sections. Fast page fault: Fast page fault is the fast path which fixes the guest page fault out of -the mmu-lock on x86. Currently, the page fault can be fast only if the -shadow page table is present and it is caused by write-protect, that means -we just need change the W bit of the spte. +the mmu-lock on x86. Currently, the page fault can be fast in one of the +following two cases: + +1. Access Tracking: The SPTE is not present, but it is marked for access +tracking i.e. the SPTE_SPECIAL_MASK is set. That means we need to +restore the saved R/X bits. This is described in more detail later below. + +2. Write-Protection: The SPTE is present and the fault is +caused by write-protect. That means we just need to change the W bit of the +spte. What we use to avoid all the race is the SPTE_HOST_WRITEABLE bit and SPTE_MMU_WRITEABLE bit on the spte: @@ -38,7 +45,8 @@ SPTE_MMU_WRITEABLE bit on the spte: page write-protection. On fast page fault path, we will use cmpxchg to atomically set the spte W -bit if spte.SPTE_HOST_WRITEABLE = 1 and spte.SPTE_WRITE_PROTECT = 1, this +bit if spte.SPTE_HOST_WRITEABLE = 1 and spte.SPTE_WRITE_PROTECT = 1, or +restore the saved R/X bits if VMX_EPT_TRACK_ACCESS mask is set, or both. This is safe because whenever changing these bits can be detected by cmpxchg. But we need carefully check these cases: @@ -142,6 +150,21 @@ Since the spte is "volatile" if it can be updated out of mmu-lock, we always atomically update the spte, the race caused by fast page fault can be avoided, See the comments in spte_has_volatile_bits() and mmu_spte_update(). +Lockless Access Tracking: + +This is used for Intel CPUs that are using EPT but do not support the EPT A/D +bits. In this case, when the KVM MMU notifier is called to track accesses to a +page (via kvm_mmu_notifier_clear_flush_young), it marks the PTE as not-present +by clearing the RWX bits in the PTE and storing the original R & X bits in +some unused/ignored bits. In addition, the SPTE_SPECIAL_MASK is also set on the +PTE (using the ignored bit 62). When the VM tries to access the page later on, +a fault is generated and the fast page fault mechanism described above is used +to atomically restore the PTE to a Present state. The W bit is not saved when +the PTE is marked for access tracking and during restoration to the Present +state, the W bit is set depending on whether or not it was a write access. If +it wasn't, then the W bit will remain clear until a write access happens, at +which time it will be set using the Dirty tracking mechanism described above. + 3. Reference ------------ diff --git a/Documentation/vm/transhuge.txt b/Documentation/vm/transhuge.txt index 8fda5b7b24e9..cd28d5ee5273 100644 --- a/Documentation/vm/transhuge.txt +++ b/Documentation/vm/transhuge.txt @@ -302,7 +302,7 @@ thp_split_page is incremented every time a huge page is split into base reason is that a huge page is old and is being reclaimed. This action implies splitting all PMD the page mapped with. -thp_split_page_failed is is incremented if kernel fails to split huge +thp_split_page_failed is incremented if kernel fails to split huge page. This can happen if the page was pinned by somebody. thp_deferred_split_page is incremented when a huge page is put onto split diff --git a/MAINTAINERS b/MAINTAINERS index 5ca9a7e163d8..ad4f2ce991ca 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -8367,6 +8367,7 @@ F: drivers/media/dvb-frontends/mn88473* MODULE SUPPORT M: Jessica Yu <jeyu@redhat.com> M: Rusty Russell <rusty@rustcorp.com.au> +T: git git://git.kernel.org/pub/scm/linux/kernel/git/jeyu/linux.git modules-next S: Maintained F: include/linux/module.h F: kernel/module.c @@ -9999,6 +10000,14 @@ S: Supported F: Documentation/preempt-locking.txt F: include/linux/preempt.h +PRINTK +M: Petr Mladek <pmladek@suse.com> +M: Sergey Senozhatsky <sergey.senozhatsky@gmail.com> +R: Steven Rostedt <rostedt@goodmis.org> +S: Maintained +F: kernel/printk/ +F: include/linux/printk.h + PRISM54 WIRELESS DRIVER M: "Luis R. Rodriguez" <mcgrof@gmail.com> L: linux-wireless@vger.kernel.org @@ -1446,7 +1446,7 @@ $(help-board-dirs): help-%: # Documentation targets # --------------------------------------------------------------------------- -DOC_TARGETS := xmldocs sgmldocs psdocs latexdocs pdfdocs htmldocs mandocs installmandocs epubdocs cleandocs +DOC_TARGETS := xmldocs sgmldocs psdocs latexdocs pdfdocs htmldocs mandocs installmandocs epubdocs cleandocs linkcheckdocs PHONY += $(DOC_TARGETS) $(DOC_TARGETS): scripts_basic FORCE $(Q)$(MAKE) $(build)=scripts build_docproc build_check-lc_ctype diff --git a/arch/arm/include/asm/kvm_host.h b/arch/arm/include/asm/kvm_host.h index d5423ab15ed5..cc495d799c67 100644 --- a/arch/arm/include/asm/kvm_host.h +++ b/arch/arm/include/asm/kvm_host.h @@ -60,9 +60,6 @@ struct kvm_arch { /* The last vcpu id that ran on each physical CPU */ int __percpu *last_vcpu_ran; - /* Timer */ - struct arch_timer_kvm timer; - /* * Anything that is not used directly from assembly code goes * here. diff --git a/arch/arm/include/asm/kvm_mmu.h b/arch/arm/include/asm/kvm_mmu.h index 74a44727f8e1..95f38dcd611d 100644 --- a/arch/arm/include/asm/kvm_mmu.h +++ b/arch/arm/include/asm/kvm_mmu.h @@ -129,8 +129,7 @@ static inline bool vcpu_has_cache_enabled(struct kvm_vcpu *vcpu) static inline void __coherent_cache_guest_page(struct kvm_vcpu *vcpu, kvm_pfn_t pfn, - unsigned long size, - bool ipa_uncached) + unsigned long size) { /* * If we are going to insert an instruction page and the icache is @@ -150,18 +149,12 @@ static inline void __coherent_cache_guest_page(struct kvm_vcpu *vcpu, * and iterate over the range. */ - bool need_flush = !vcpu_has_cache_enabled(vcpu) || ipa_uncached; - VM_BUG_ON(size & ~PAGE_MASK); - if (!need_flush && !icache_is_pipt()) - goto vipt_cache; - while (size) { void *va = kmap_atomic_pfn(pfn); - if (need_flush) - kvm_flush_dcache_to_poc(va, PAGE_SIZE); + kvm_flush_dcache_to_poc(va, PAGE_SIZE); if (icache_is_pipt()) __cpuc_coherent_user_range((unsigned long)va, @@ -173,7 +166,6 @@ static inline void __coherent_cache_guest_page(struct kvm_vcpu *vcpu, kunmap_atomic(va); } -vipt_cache: if (!icache_is_pipt() && !icache_is_vivt_asid_tagged()) { /* any kind of VIPT cache */ __flush_icache_all(); diff --git a/arch/arm/include/uapi/asm/kvm.h b/arch/arm/include/uapi/asm/kvm.h index af05f8e0903e..6ebd3e6a1fd1 100644 --- a/arch/arm/include/uapi/asm/kvm.h +++ b/arch/arm/include/uapi/asm/kvm.h @@ -181,10 +181,23 @@ struct kvm_arch_memory_slot { #define KVM_DEV_ARM_VGIC_GRP_CPU_REGS 2 #define KVM_DEV_ARM_VGIC_CPUID_SHIFT 32 #define KVM_DEV_ARM_VGIC_CPUID_MASK (0xffULL << KVM_DEV_ARM_VGIC_CPUID_SHIFT) +#define KVM_DEV_ARM_VGIC_V3_MPIDR_SHIFT 32 +#define KVM_DEV_ARM_VGIC_V3_MPIDR_MASK \ + (0xffffffffULL << KVM_DEV_ARM_VGIC_V3_MPIDR_SHIFT) #define KVM_DEV_ARM_VGIC_OFFSET_SHIFT 0 #define KVM_DEV_ARM_VGIC_OFFSET_MASK (0xffffffffULL << KVM_DEV_ARM_VGIC_OFFSET_SHIFT) +#define KVM_DEV_ARM_VGIC_SYSREG_INSTR_MASK (0xffff) #define KVM_DEV_ARM_VGIC_GRP_NR_IRQS 3 #define KVM_DEV_ARM_VGIC_GRP_CTRL 4 +#define KVM_DEV_ARM_VGIC_GRP_REDIST_REGS 5 +#define KVM_DEV_ARM_VGIC_GRP_CPU_SYSREGS 6 +#define KVM_DEV_ARM_VGIC_GRP_LEVEL_INFO 7 +#define KVM_DEV_ARM_VGIC_LINE_LEVEL_INFO_SHIFT 10 +#define KVM_DEV_ARM_VGIC_LINE_LEVEL_INFO_MASK \ + (0x3fffffULL << KVM_DEV_ARM_VGIC_LINE_LEVEL_INFO_SHIFT) +#define KVM_DEV_ARM_VGIC_LINE_LEVEL_INTID_MASK 0x3ff +#define VGIC_LEVEL_INFO_LINE_LEVEL 0 + #define KVM_DEV_ARM_VGIC_CTRL_INIT 0 /* KVM_IRQ_LINE irq field index values */ diff --git a/arch/arm/kvm/Makefile b/arch/arm/kvm/Makefile index d571243ab4d1..7b3670c2ae7b 100644 --- a/arch/arm/kvm/Makefile +++ b/arch/arm/kvm/Makefile @@ -7,7 +7,7 @@ ifeq ($(plus_virt),+virt) plus_virt_def := -DREQUIRES_VIRT=1 endif -ccflags-y += -Iarch/arm/kvm +ccflags-y += -Iarch/arm/kvm -Ivirt/kvm/arm/vgic CFLAGS_arm.o := -I. $(plus_virt_def) CFLAGS_mmu.o := -I. @@ -20,7 +20,7 @@ kvm-arm-y = $(KVM)/kvm_main.o $(KVM)/coalesced_mmio.o $(KVM)/eventfd.o $(KVM)/vf obj-$(CONFIG_KVM_ARM_HOST) += hyp/ obj-y += kvm-arm.o init.o interrupts.o obj-y += arm.o handle_exit.o guest.o mmu.o emulate.o reset.o -obj-y += coproc.o coproc_a15.o coproc_a7.o mmio.o psci.o perf.o +obj-y += coproc.o coproc_a15.o coproc_a7.o mmio.o psci.o perf.o vgic-v3-coproc.o obj-y += $(KVM)/arm/aarch32.o obj-y += $(KVM)/arm/vgic/vgic.o @@ -33,5 +33,6 @@ obj-y += $(KVM)/arm/vgic/vgic-mmio-v2.o obj-y += $(KVM)/arm/vgic/vgic-mmio-v3.o obj-y += $(KVM)/arm/vgic/vgic-kvm-device.o obj-y += $(KVM)/arm/vgic/vgic-its.o +obj-y += $(KVM)/arm/vgic/vgic-debug.o obj-y += $(KVM)/irqchip.o obj-y += $(KVM)/arm/arch_timer.o diff --git a/arch/arm/kvm/arm.c b/arch/arm/kvm/arm.c index 9d7446456e0c..c9a2103faeb9 100644 --- a/arch/arm/kvm/arm.c +++ b/arch/arm/kvm/arm.c @@ -135,7 +135,6 @@ int kvm_arch_init_vm(struct kvm *kvm, unsigned long type) goto out_free_stage2_pgd; kvm_vgic_early_init(kvm); - kvm_timer_init(kvm); /* Mark the initial VMID generation invalid */ kvm->arch.vmid_gen = 0; @@ -207,6 +206,7 @@ int kvm_vm_ioctl_check_extension(struct kvm *kvm, long ext) case KVM_CAP_ARM_PSCI_0_2: case KVM_CAP_READONLY_MEM: case KVM_CAP_MP_STATE: + case KVM_CAP_IMMEDIATE_EXIT: r = 1; break; case KVM_CAP_COALESCED_MMIO: @@ -301,7 +301,8 @@ void kvm_arch_vcpu_destroy(struct kvm_vcpu *vcpu) int kvm_cpu_has_pending_timer(struct kvm_vcpu *vcpu) { - return kvm_timer_should_fire(vcpu); + return kvm_timer_should_fire(vcpu_vtimer(vcpu)) || + kvm_timer_should_fire(vcpu_ptimer(vcpu)); } void kvm_arch_vcpu_blocking(struct kvm_vcpu *vcpu) @@ -604,6 +605,9 @@ int kvm_arch_vcpu_ioctl_run(struct kvm_vcpu *vcpu, struct kvm_run *run) return ret; } + if (run->immediate_exit) + return -EINTR; + if (vcpu->sigset_active) sigprocmask(SIG_SETMASK, &vcpu->sigset, &sigsaved); diff --git a/arch/arm/kvm/mmu.c b/arch/arm/kvm/mmu.c index a5265edbeeab..962616fd4ddd 100644 --- a/arch/arm/kvm/mmu.c +++ b/arch/arm/kvm/mmu.c @@ -1232,9 +1232,9 @@ void kvm_arch_mmu_enable_log_dirty_pt_masked(struct kvm *kvm, } static void coherent_cache_guest_page(struct kvm_vcpu *vcpu, kvm_pfn_t pfn, - unsigned long size, bool uncached) + unsigned long size) { - __coherent_cache_guest_page(vcpu, pfn, size, uncached); + __coherent_cache_guest_page(vcpu, pfn, size); } static int user_mem_abort(struct kvm_vcpu *vcpu, phys_addr_t fault_ipa, @@ -1250,7 +1250,6 @@ static int user_mem_abort(struct kvm_vcpu *vcpu, phys_addr_t fault_ipa, struct vm_area_struct *vma; kvm_pfn_t pfn; pgprot_t mem_type = PAGE_S2; - bool fault_ipa_uncached; bool logging_active = memslot_is_logging(memslot); unsigned long flags = 0; @@ -1337,8 +1336,6 @@ static int user_mem_abort(struct kvm_vcpu *vcpu, phys_addr_t fault_ipa, if (!hugetlb && !force_pte) hugetlb = transparent_hugepage_adjust(&pfn, &fault_ipa); - fault_ipa_uncached = memslot->flags & KVM_MEMSLOT_INCOHERENT; - if (hugetlb) { pmd_t new_pmd = pfn_pmd(pfn, mem_type); new_pmd = pmd_mkhuge(new_pmd); @@ -1346,7 +1343,7 @@ static int user_mem_abort(struct kvm_vcpu *vcpu, phys_addr_t fault_ipa, new_pmd = kvm_s2pmd_mkwrite(new_pmd); kvm_set_pfn_dirty(pfn); } - coherent_cache_guest_page(vcpu, pfn, PMD_SIZE, fault_ipa_uncached); + coherent_cache_guest_page(vcpu, pfn, PMD_SIZE); ret = stage2_set_pmd_huge(kvm, memcache, fault_ipa, &new_pmd); } else { pte_t new_pte = pfn_pte(pfn, mem_type); @@ -1356,7 +1353,7 @@ static int user_mem_abort(struct kvm_vcpu *vcpu, phys_addr_t fault_ipa, kvm_set_pfn_dirty(pfn); mark_page_dirty(kvm, gfn); } - coherent_cache_guest_page(vcpu, pfn, PAGE_SIZE, fault_ipa_uncached); + coherent_cache_guest_page(vcpu, pfn, PAGE_SIZE); ret = stage2_set_pte(kvm, memcache, fault_ipa, &new_pte, flags); } @@ -1879,15 +1876,6 @@ void kvm_arch_free_memslot(struct kvm *kvm, struct kvm_memory_slot *free, int kvm_arch_create_memslot(struct kvm *kvm, struct kvm_memory_slot *slot, unsigned long npages) { - /* - * Readonly memslots are not incoherent with the caches by definition, - * but in practice, they are used mostly to emulate ROMs or NOR flashes - * that the guest may consider devices and hence map as uncached. - * To prevent incoherency issues in these cases, tag all readonly - * regions as incoherent. - */ - if (slot->flags & KVM_MEM_READONLY) - slot->flags |= KVM_MEMSLOT_INCOHERENT; return 0; } diff --git a/arch/arm/kvm/reset.c b/arch/arm/kvm/reset.c index 4b5e802e57d1..1da8b2d14550 100644 --- a/arch/arm/kvm/reset.c +++ b/arch/arm/kvm/reset.c @@ -37,6 +37,11 @@ static struct kvm_regs cortexa_regs_reset = { .usr_regs.ARM_cpsr = SVC_MODE | PSR_A_BIT | PSR_I_BIT | PSR_F_BIT, }; +static const struct kvm_irq_level cortexa_ptimer_irq = { + { .irq = 30 }, + .level = 1, +}; + static const struct kvm_irq_level cortexa_vtimer_irq = { { .irq = 27 }, .level = 1, @@ -58,6 +63,7 @@ int kvm_reset_vcpu(struct kvm_vcpu *vcpu) { struct kvm_regs *reset_regs; const struct kvm_irq_level *cpu_vtimer_irq; + const struct kvm_irq_level *cpu_ptimer_irq; switch (vcpu->arch.target) { case KVM_ARM_TARGET_CORTEX_A7: @@ -65,6 +71,7 @@ int kvm_reset_vcpu(struct kvm_vcpu *vcpu) reset_regs = &cortexa_regs_reset; vcpu->arch.midr = read_cpuid_id(); cpu_vtimer_irq = &cortexa_vtimer_irq; + cpu_ptimer_irq = &cortexa_ptimer_irq; break; default: return -ENODEV; @@ -77,5 +84,5 @@ int kvm_reset_vcpu(struct kvm_vcpu *vcpu) kvm_reset_coprocs(vcpu); /* Reset arch_timer context */ - return kvm_timer_vcpu_reset(vcpu, cpu_vtimer_irq); + return kvm_timer_vcpu_reset(vcpu, cpu_vtimer_irq, cpu_ptimer_irq); } diff --git a/arch/arm/kvm/vgic-v3-coproc.c b/arch/arm/kvm/vgic-v3-coproc.c new file mode 100644 index 000000000000..f41abf76366f --- /dev/null +++ b/arch/arm/kvm/vgic-v3-coproc.c @@ -0,0 +1,35 @@ +/* + * VGIC system registers handling functions for AArch32 mode + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + */ + +#include <linux/kvm.h> +#include <linux/kvm_host.h> +#include <asm/kvm_emulate.h> +#include "vgic.h" + +int vgic_v3_has_cpu_sysregs_attr(struct kvm_vcpu *vcpu, bool is_write, u64 id, + u64 *reg) +{ + /* + * TODO: Implement for AArch32 + */ + return -ENXIO; +} + +int vgic_v3_cpu_sysregs_uaccess(struct kvm_vcpu *vcpu, bool is_write, u64 id, + u64 *reg) +{ + /* + * TODO: Implement for AArch32 + */ + return -ENXIO; +} diff --git a/arch/arm64/include/asm/kvm_host.h b/arch/arm64/include/asm/kvm_host.h index 443b387021f2..f21fd3894370 100644 --- a/arch/arm64/include/asm/kvm_host.h +++ b/arch/arm64/include/asm/kvm_host.h @@ -70,9 +70,6 @@ struct kvm_arch { /* Interrupt controller */ struct vgic_dist vgic; - - /* Timer */ - struct arch_timer_kvm timer; }; #define KVM_NR_MEM_OBJS 40 diff --git a/arch/arm64/include/asm/kvm_mmu.h b/arch/arm64/include/asm/kvm_mmu.h index 55772c13a375..ed1246014901 100644 --- a/arch/arm64/include/asm/kvm_mmu.h +++ b/arch/arm64/include/asm/kvm_mmu.h @@ -236,13 +236,11 @@ static inline bool vcpu_has_cache_enabled(struct kvm_vcpu *vcpu) static inline void __coherent_cache_guest_page(struct kvm_vcpu *vcpu, kvm_pfn_t pfn, - unsigned long size, - bool ipa_uncached) + unsigned long size) { void *va = page_address(pfn_to_page(pfn)); - if (!vcpu_has_cache_enabled(vcpu) || ipa_uncached) - kvm_flush_dcache_to_poc(va, size); + kvm_flush_dcache_to_poc(va, size); if (!icache_is_aliasing()) { /* PIPT */ flush_icache_range((unsigned long)va, diff --git a/arch/arm64/include/uapi/asm/kvm.h b/arch/arm64/include/uapi/asm/kvm.h index 3051f86a9b5f..c2860358ae3e 100644 --- a/arch/arm64/include/uapi/asm/kvm.h +++ b/arch/arm64/include/uapi/asm/kvm.h @@ -201,10 +201,23 @@ struct kvm_arch_memory_slot { #define KVM_DEV_ARM_VGIC_GRP_CPU_REGS 2 #define KVM_DEV_ARM_VGIC_CPUID_SHIFT 32 #define KVM_DEV_ARM_VGIC_CPUID_MASK (0xffULL << KVM_DEV_ARM_VGIC_CPUID_SHIFT) +#define KVM_DEV_ARM_VGIC_V3_MPIDR_SHIFT 32 +#define KVM_DEV_ARM_VGIC_V3_MPIDR_MASK \ + (0xffffffffULL << KVM_DEV_ARM_VGIC_V3_MPIDR_SHIFT) #define KVM_DEV_ARM_VGIC_OFFSET_SHIFT 0 #define KVM_DEV_ARM_VGIC_OFFSET_MASK (0xffffffffULL << KVM_DEV_ARM_VGIC_OFFSET_SHIFT) +#define KVM_DEV_ARM_VGIC_SYSREG_INSTR_MASK (0xffff) #define KVM_DEV_ARM_VGIC_GRP_NR_IRQS 3 #define KVM_DEV_ARM_VGIC_GRP_CTRL 4 +#define KVM_DEV_ARM_VGIC_GRP_REDIST_REGS 5 +#define KVM_DEV_ARM_VGIC_GRP_CPU_SYSREGS 6 +#define KVM_DEV_ARM_VGIC_GRP_LEVEL_INFO 7 +#define KVM_DEV_ARM_VGIC_LINE_LEVEL_INFO_SHIFT 10 +#define KVM_DEV_ARM_VGIC_LINE_LEVEL_INFO_MASK \ + (0x3fffffULL << KVM_DEV_ARM_VGIC_LINE_LEVEL_INFO_SHIFT) +#define KVM_DEV_ARM_VGIC_LINE_LEVEL_INTID_MASK 0x3ff +#define VGIC_LEVEL_INFO_LINE_LEVEL 0 + #define KVM_DEV_ARM_VGIC_CTRL_INIT 0 /* Device Control API on vcpu fd */ diff --git a/arch/arm64/kvm/Makefile b/arch/arm64/kvm/Makefile index d50a82a16ff6..afd51bebb9c5 100644 --- a/arch/arm64/kvm/Makefile +++ b/arch/arm64/kvm/Makefile @@ -2,7 +2,7 @@ # Makefile for Kernel-based Virtual Machine module # -ccflags-y += -Iarch/arm64/kvm +ccflags-y += -Iarch/arm64/kvm -Ivirt/kvm/arm/vgic CFLAGS_arm.o := -I. CFLAGS_mmu.o := -I. @@ -19,6 +19,7 @@ kvm-$(CONFIG_KVM_ARM_HOST) += $(ARM)/psci.o $(ARM)/perf.o kvm-$(CONFIG_KVM_ARM_HOST) += inject_fault.o regmap.o kvm-$(CONFIG_KVM_ARM_HOST) += hyp.o hyp-init.o handle_exit.o kvm-$(CONFIG_KVM_ARM_HOST) += guest.o debug.o reset.o sys_regs.o sys_regs_generic_v8.o +kvm-$(CONFIG_KVM_ARM_HOST) += vgic-sys-reg-v3.o kvm-$(CONFIG_KVM_ARM_HOST) += $(KVM)/arm/aarch32.o kvm-$(CONFIG_KVM_ARM_HOST) += $(KVM)/arm/vgic/vgic.o @@ -31,6 +32,7 @@ kvm-$(CONFIG_KVM_ARM_HOST) += $(KVM)/arm/vgic/vgic-mmio-v2.o kvm-$(CONFIG_KVM_ARM_HOST) += $(KVM)/arm/vgic/vgic-mmio-v3.o kvm-$(CONFIG_KVM_ARM_HOST) += $(KVM)/arm/vgic/vgic-kvm-device.o kvm-$(CONFIG_KVM_ARM_HOST) += $(KVM)/arm/vgic/vgic-its.o +kvm-$(CONFIG_KVM_ARM_HOST) += $(KVM)/arm/vgic/vgic-debug.o kvm-$(CONFIG_KVM_ARM_HOST) += $(KVM)/irqchip.o kvm-$(CONFIG_KVM_ARM_HOST) += $(KVM)/arm/arch_timer.o kvm-$(CONFIG_KVM_ARM_PMU) += $(KVM)/arm/pmu.o diff --git a/arch/arm64/kvm/reset.c b/arch/arm64/kvm/reset.c index e95d4f68bf54..d9e9697de1b2 100644 --- a/arch/arm64/kvm/reset.c +++ b/arch/arm64/kvm/reset.c @@ -46,6 +46,11 @@ static const struct kvm_regs default_regs_reset32 = { COMPAT_PSR_I_BIT | COMPAT_PSR_F_BIT), }; +static const struct kvm_irq_level default_ptimer_irq = { + .irq = 30, + .level = 1, +}; + static const struct kvm_irq_level default_vtimer_irq = { .irq = 27, .level = 1, @@ -104,6 +109,7 @@ int kvm_arch_dev_ioctl_check_extension(struct kvm *kvm, long ext) int kvm_reset_vcpu(struct kvm_vcpu *vcpu) { const struct kvm_irq_level *cpu_vtimer_irq; + const struct kvm_irq_level *cpu_ptimer_irq; const struct kvm_regs *cpu_reset; switch (vcpu->arch.target) { @@ -117,6 +123,7 @@ int kvm_reset_vcpu(struct kvm_vcpu *vcpu) } cpu_vtimer_irq = &default_vtimer_irq; + cpu_ptimer_irq = &default_ptimer_irq; break; } @@ -130,5 +137,5 @@ int kvm_reset_vcpu(struct kvm_vcpu *vcpu) kvm_pmu_vcpu_reset(vcpu); /* Reset timer */ - return kvm_timer_vcpu_reset(vcpu, cpu_vtimer_irq); + return kvm_timer_vcpu_reset(vcpu, cpu_vtimer_irq, cpu_ptimer_irq); } diff --git a/arch/arm64/kvm/sys_regs.c b/arch/arm64/kvm/sys_regs.c index 87e7e6608cd8..0e26f8c2b56f 100644 --- a/arch/arm64/kvm/sys_regs.c +++ b/arch/arm64/kvm/sys_regs.c @@ -820,6 +820,61 @@ static bool access_pmuserenr(struct kvm_vcpu *vcpu, struct sys_reg_params *p, CRm((0b1100 | (((n) >> 3) & 0x3))), Op2(((n) & 0x7)), \ access_pmu_evtyper, reset_unknown, (PMEVTYPER0_EL0 + n), } +static bool access_cntp_tval(struct kvm_vcpu *vcpu, + struct sys_reg_params *p, + const struct sys_reg_desc *r) +{ + struct arch_timer_context *ptimer = vcpu_ptimer(vcpu); + u64 now = kvm_phys_timer_read(); + + if (p->is_write) + ptimer->cnt_cval = p->regval + now; + else + p->regval = ptimer->cnt_cval - now; + + return true; +} + +static bool access_cntp_ctl(struct kvm_vcpu *vcpu, + struct sys_reg_params *p, + const struct sys_reg_desc *r) +{ + struct arch_timer_context *ptimer = vcpu_ptimer(vcpu); + + if (p->is_write) { + /* ISTATUS bit is read-only */ + ptimer->cnt_ctl = p->regval & ~ARCH_TIMER_CTRL_IT_STAT; + } else { + u64 now = kvm_phys_timer_read(); + + p->regval = ptimer->cnt_ctl; + /* + * Set ISTATUS bit if it's expired. + * Note that according to ARMv8 ARM Issue A.k, ISTATUS bit is + * UNKNOWN when ENABLE bit is 0, so we chose to set ISTATUS bit + * regardless of ENABLE bit for our implementation convenience. + */ + if (ptimer->cnt_cval <= now) + p->regval |= ARCH_TIMER_CTRL_IT_STAT; + } + + return true; +} + +static bool access_cntp_cval(struct kvm_vcpu *vcpu, + struct sys_reg_params *p, + const struct sys_reg_desc *r) +{ + struct arch_timer_context *ptimer = vcpu_ptimer(vcpu); + + if (p->is_write) + ptimer->cnt_cval = p->regval; + else + p->regval = ptimer->cnt_cval; + + return true; +} + /* * Architected system registers. * Important: Must be sorted ascending by Op0, Op1, CRn, CRm, Op2 @@ -1029,6 +1084,16 @@ static const struct sys_reg_desc sys_reg_descs[] = { { Op0(0b11), Op1(0b011), CRn(0b1101), CRm(0b0000), Op2(0b011), NULL, reset_unknown, TPIDRRO_EL0 }, + /* CNTP_TVAL_EL0 */ + { Op0(0b11), Op1(0b011), CRn(0b1110), CRm(0b0010), Op2(0b000), + access_cntp_tval }, + /* CNTP_CTL_EL0 */ + { Op0(0b11), Op1(0b011), CRn(0b1110), CRm(0b0010), Op2(0b001), + access_cntp_ctl }, + /* CNTP_CVAL_EL0 */ + { Op0(0b11), Op1(0b011), CRn(0b1110), CRm(0b0010), Op2(0b010), + access_cntp_cval }, + /* PMEVCNTRn_EL0 */ PMU_PMEVCNTR_EL0(0), PMU_PMEVCNTR_EL0(1), @@ -1795,6 +1860,17 @@ static bool index_to_params(u64 id, struct sys_reg_params *params) } } +const struct sys_reg_desc *find_reg_by_id(u64 id, + struct sys_reg_params *params, + const struct sys_reg_desc table[], + unsigned int num) +{ + if (!index_to_params(id, params)) + return NULL; + + return find_reg(params, table, num); +} + /* Decode an index value, and find the sys_reg_desc entry. */ static const struct sys_reg_desc *index_to_sys_reg_desc(struct kvm_vcpu *vcpu, u64 id) @@ -1807,11 +1883,8 @@ static const struct sys_reg_desc *index_to_sys_reg_desc(struct kvm_vcpu *vcpu, if ((id & KVM_REG_ARM_COPROC_MASK) != KVM_REG_ARM64_SYSREG) return NULL; - if (!index_to_params(id, ¶ms)) - return NULL; - table = get_target_table(vcpu->arch.target, true, &num); - r = find_reg(¶ms, table, num); + r = find_reg_by_id(id, ¶ms, table, num); if (!r) r = find_reg(¶ms, sys_reg_descs, ARRAY_SIZE(sys_reg_descs)); @@ -1918,10 +1991,8 @@ static int get_invariant_sys_reg(u64 id, void __user *uaddr) struct sys_reg_params params; const struct sys_reg_desc *r; - if (!index_to_params(id, ¶ms)) - return -ENOENT; - - r = find_reg(¶ms, invariant_sys_regs, ARRAY_SIZE(invariant_sys_regs)); + r = find_reg_by_id(id, ¶ms, invariant_sys_regs, + ARRAY_SIZE(invariant_sys_regs)); if (!r) return -ENOENT; @@ -1935,9 +2006,8 @@ static int set_invariant_sys_reg(u64 id, void __user *uaddr) int err; u64 val = 0; /* Make sure high bits are 0 for 32-bit regs */ - if (!index_to_params(id, ¶ms)) - return -ENOENT; - r = find_reg(¶ms, invariant_sys_regs, ARRAY_SIZE(invariant_sys_regs)); + r = find_reg_by_id(id, ¶ms, invariant_sys_regs, + ARRAY_SIZE(invariant_sys_regs)); if (!r) return -ENOENT; diff --git a/arch/arm64/kvm/sys_regs.h b/arch/arm64/kvm/sys_regs.h index dbbb01cfbee9..9c6ffd0f0196 100644 --- a/arch/arm64/kvm/sys_regs.h +++ b/arch/arm64/kvm/sys_regs.h @@ -136,6 +136,10 @@ static inline int cmp_sys_reg(const struct sys_reg_desc *i1, return i1->Op2 - i2->Op2; } +const struct sys_reg_desc *find_reg_by_id(u64 id, + struct sys_reg_params *params, + const struct sys_reg_desc table[], + unsigned int num); #define Op0(_x) .Op0 = _x #define Op1(_x) .Op1 = _x diff --git a/arch/arm64/kvm/vgic-sys-reg-v3.c b/arch/arm64/kvm/vgic-sys-reg-v3.c new file mode 100644 index 000000000000..79f37e37d367 --- /dev/null +++ b/arch/arm64/kvm/vgic-sys-reg-v3.c @@ -0,0 +1,346 @@ +/* + * VGIC system registers handling functions for AArch64 mode + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + */ + +#include <linux/irqchip/arm-gic-v3.h> +#include <linux/kvm.h> +#include <linux/kvm_host.h> +#include <asm/kvm_emulate.h> +#include "vgic.h" +#include "sys_regs.h" + +static bool access_gic_ctlr(struct kvm_vcpu *vcpu, struct sys_reg_params *p, + const struct sys_reg_desc *r) +{ + u32 host_pri_bits, host_id_bits, host_seis, host_a3v, seis, a3v; + struct vgic_cpu *vgic_v3_cpu = &vcpu->arch.vgic_cpu; + struct vgic_vmcr vmcr; + u64 val; + + vgic_get_vmcr(vcpu, &vmcr); + if (p->is_write) { + val = p->regval; + + /* + * Disallow restoring VM state if not supported by this + * hardware. + */ + host_pri_bits = ((val & ICC_CTLR_EL1_PRI_BITS_MASK) >> + ICC_CTLR_EL1_PRI_BITS_SHIFT) + 1; + if (host_pri_bits > vgic_v3_cpu->num_pri_bits) + return false; + + vgic_v3_cpu->num_pri_bits = host_pri_bits; + + host_id_bits = (val & ICC_CTLR_EL1_ID_BITS_MASK) >> + ICC_CTLR_EL1_ID_BITS_SHIFT; + if (host_id_bits > vgic_v3_cpu->num_id_bits) + return false; + + vgic_v3_cpu->num_id_bits = host_id_bits; + + host_seis = ((kvm_vgic_global_state.ich_vtr_el2 & + ICH_VTR_SEIS_MASK) >> ICH_VTR_SEIS_SHIFT); + seis = (val & ICC_CTLR_EL1_SEIS_MASK) >> + ICC_CTLR_EL1_SEIS_SHIFT; + if (host_seis != seis) + return false; + + host_a3v = ((kvm_vgic_global_state.ich_vtr_el2 & + ICH_VTR_A3V_MASK) >> ICH_VTR_A3V_SHIFT); + a3v = (val & ICC_CTLR_EL1_A3V_MASK) >> ICC_CTLR_EL1_A3V_SHIFT; + if (host_a3v != a3v) + return false; + + /* + * Here set VMCR.CTLR in ICC_CTLR_EL1 layout. + * The vgic_set_vmcr() will convert to ICH_VMCR layout. + */ + vmcr.ctlr = val & ICC_CTLR_EL1_CBPR_MASK; + vmcr.ctlr |= val & ICC_CTLR_EL1_EOImode_MASK; + vgic_set_vmcr(vcpu, &vmcr); + } else { + val = 0; + val |= (vgic_v3_cpu->num_pri_bits - 1) << + ICC_CTLR_EL1_PRI_BITS_SHIFT; + val |= vgic_v3_cpu->num_id_bits << ICC_CTLR_EL1_ID_BITS_SHIFT; + val |= ((kvm_vgic_global_state.ich_vtr_el2 & + ICH_VTR_SEIS_MASK) >> ICH_VTR_SEIS_SHIFT) << + ICC_CTLR_EL1_SEIS_SHIFT; + val |= ((kvm_vgic_global_state.ich_vtr_el2 & + ICH_VTR_A3V_MASK) >> ICH_VTR_A3V_SHIFT) << + ICC_CTLR_EL1_A3V_SHIFT; + /* + * The VMCR.CTLR value is in ICC_CTLR_EL1 layout. + * Extract it directly using ICC_CTLR_EL1 reg definitions. + */ + val |= vmcr.ctlr & ICC_CTLR_EL1_CBPR_MASK; + val |= vmcr.ctlr & ICC_CTLR_EL1_EOImode_MASK; + + p->regval = val; + } + + return true; +} + +static bool access_gic_pmr(struct kvm_vcpu *vcpu, struct sys_reg_params *p, + const struct sys_reg_desc *r) +{ + struct vgic_vmcr vmcr; + + vgic_get_vmcr(vcpu, &vmcr); + if (p->is_write) { + vmcr.pmr = (p->regval & ICC_PMR_EL1_MASK) >> ICC_PMR_EL1_SHIFT; + vgic_set_vmcr(vcpu, &vmcr); + } else { + p->regval = (vmcr.pmr << ICC_PMR_EL1_SHIFT) & ICC_PMR_EL1_MASK; + } + + return true; +} + +static bool access_gic_bpr0(struct kvm_vcpu *vcpu, struct sys_reg_params *p, + const struct sys_reg_desc *r) +{ + struct vgic_vmcr vmcr; + + vgic_get_vmcr(vcpu, &vmcr); + if (p->is_write) { + vmcr.bpr = (p->regval & ICC_BPR0_EL1_MASK) >> + ICC_BPR0_EL1_SHIFT; + vgic_set_vmcr(vcpu, &vmcr); + } else { + p->regval = (vmcr.bpr << ICC_BPR0_EL1_SHIFT) & + ICC_BPR0_EL1_MASK; + } + + return true; +} + +static bool access_gic_bpr1(struct kvm_vcpu *vcpu, struct sys_reg_params *p, + const struct sys_reg_desc *r) +{ + struct vgic_vmcr vmcr; + + if (!p->is_write) + p->regval = 0; + + vgic_get_vmcr(vcpu, &vmcr); + if (!((vmcr.ctlr & ICH_VMCR_CBPR_MASK) >> ICH_VMCR_CBPR_SHIFT)) { + if (p->is_write) { + vmcr.abpr = (p->regval & ICC_BPR1_EL1_MASK) >> + ICC_BPR1_EL1_SHIFT; + vgic_set_vmcr(vcpu, &vmcr); + } else { + p->regval = (vmcr.abpr << ICC_BPR1_EL1_SHIFT) & + ICC_BPR1_EL1_MASK; + } + } else { + if (!p->is_write) + p->regval = min((vmcr.bpr + 1), 7U); + } + + return true; +} + +static bool access_gic_grpen0(struct kvm_vcpu *vcpu, struct sys_reg_params *p, + const struct sys_reg_desc *r) +{ + struct vgic_vmcr vmcr; + + vgic_get_vmcr(vcpu, &vmcr); + if (p->is_write) { + vmcr.grpen0 = (p->regval & ICC_IGRPEN0_EL1_MASK) >> + ICC_IGRPEN0_EL1_SHIFT; + vgic_set_vmcr(vcpu, &vmcr); + } else { + p->regval = (vmcr.grpen0 << ICC_IGRPEN0_EL1_SHIFT) & + ICC_IGRPEN0_EL1_MASK; + } + + return true; +} + +static bool access_gic_grpen1(struct kvm_vcpu *vcpu, struct sys_reg_params *p, + const struct sys_reg_desc *r) +{ + struct vgic_vmcr vmcr; + + vgic_get_vmcr(vcpu, &vmcr); + if (p->is_write) { + vmcr.grpen1 = (p->regval & ICC_IGRPEN1_EL1_MASK) >> + ICC_IGRPEN1_EL1_SHIFT; + vgic_set_vmcr(vcpu, &vmcr); + } else { + p->regval = (vmcr.grpen1 << ICC_IGRPEN1_EL1_SHIFT) & + ICC_IGRPEN1_EL1_MASK; + } + + return true; +} + +static void vgic_v3_access_apr_reg(struct kvm_vcpu *vcpu, + struct sys_reg_params *p, u8 apr, u8 idx) +{ + struct vgic_v3_cpu_if *vgicv3 = &vcpu->arch.vgic_cpu.vgic_v3; + uint32_t *ap_reg; + + if (apr) + ap_reg = &vgicv3->vgic_ap1r[idx]; + else + ap_reg = &vgicv3->vgic_ap0r[idx]; + + if (p->is_write) + *ap_reg = p->regval; + else + p->regval = *ap_reg; +} + +static bool access_gic_aprn(struct kvm_vcpu *vcpu, struct sys_reg_params *p, + const struct sys_reg_desc *r, u8 apr) +{ + struct vgic_cpu *vgic_v3_cpu = &vcpu->arch.vgic_cpu; + u8 idx = r->Op2 & 3; + + /* + * num_pri_bits are initialized with HW supported values. + * We can rely safely on num_pri_bits even if VM has not + * restored ICC_CTLR_EL1 before restoring APnR registers. + */ + switch (vgic_v3_cpu->num_pri_bits) { + case 7: + vgic_v3_access_apr_reg(vcpu, p, apr, idx); + break; + case 6: + if (idx > 1) + goto err; + vgic_v3_access_apr_reg(vcpu, p, apr, idx); + break; + default: + if (idx > 0) + goto err; + vgic_v3_access_apr_reg(vcpu, p, apr, idx); + } + + return true; +err: + if (!p->is_write) + p->regval = 0; + + return false; +} + +static bool access_gic_ap0r(struct kvm_vcpu *vcpu, struct sys_reg_params *p, + const struct sys_reg_desc *r) + +{ + return access_gic_aprn(vcpu, p, r, 0); +} + +static bool access_gic_ap1r(struct kvm_vcpu *vcpu, struct sys_reg_params *p, + const struct sys_reg_desc *r) +{ + return access_gic_aprn(vcpu, p, r, 1); +} + +static bool access_gic_sre(struct kvm_vcpu *vcpu, struct sys_reg_params *p, + const struct sys_reg_desc *r) +{ + struct vgic_v3_cpu_if *vgicv3 = &vcpu->arch.vgic_cpu.vgic_v3; + + /* Validate SRE bit */ + if (p->is_write) { + if (!(p->regval & ICC_SRE_EL1_SRE)) + return false; + } else { + p->regval = vgicv3->vgic_sre; + } + + return true; +} +static const struct sys_reg_desc gic_v3_icc_reg_descs[] = { + /* ICC_PMR_EL1 */ + { Op0(3), Op1(0), CRn(4), CRm(6), Op2(0), access_gic_pmr }, + /* ICC_BPR0_EL1 */ + { Op0(3), Op1(0), CRn(12), CRm(8), Op2(3), access_gic_bpr0 }, + /* ICC_AP0R0_EL1 */ + { Op0(3), Op1(0), CRn(12), CRm(8), Op2(4), access_gic_ap0r }, + /* ICC_AP0R1_EL1 */ + { Op0(3), Op1(0), CRn(12), CRm(8), Op2(5), access_gic_ap0r }, + /* ICC_AP0R2_EL1 */ + { Op0(3), Op1(0), CRn(12), CRm(8), Op2(6), access_gic_ap0r }, + /* ICC_AP0R3_EL1 */ + { Op0(3), Op1(0), CRn(12), CRm(8), Op2(7), access_gic_ap0r }, + /* ICC_AP1R0_EL1 */ + { Op0(3), Op1(0), CRn(12), CRm(9), Op2(0), access_gic_ap1r }, + /* ICC_AP1R1_EL1 */ + { Op0(3), Op1(0), CRn(12), CRm(9), Op2(1), access_gic_ap1r }, + /* ICC_AP1R2_EL1 */ + { Op0(3), Op1(0), CRn(12), CRm(9), Op2(2), access_gic_ap1r }, + /* ICC_AP1R3_EL1 */ + { Op0(3), Op1(0), CRn(12), CRm(9), Op2(3), access_gic_ap1r }, + /* ICC_BPR1_EL1 */ + { Op0(3), Op1(0), CRn(12), CRm(12), Op2(3), access_gic_bpr1 }, + /* ICC_CTLR_EL1 */ + { Op0(3), Op1(0), CRn(12), CRm(12), Op2(4), access_gic_ctlr }, + /* ICC_SRE_EL1 */ + { Op0(3), Op1(0), CRn(12), CRm(12), Op2(5), access_gic_sre }, + /* ICC_IGRPEN0_EL1 */ + { Op0(3), Op1(0), CRn(12), CRm(12), Op2(6), access_gic_grpen0 }, + /* ICC_GRPEN1_EL1 */ + { Op0(3), Op1(0), CRn(12), CRm(12), Op2(7), access_gic_grpen1 }, +}; + +int vgic_v3_has_cpu_sysregs_attr(struct kvm_vcpu *vcpu, bool is_write, u64 id, + u64 *reg) +{ + struct sys_reg_params params; + u64 sysreg = (id & KVM_DEV_ARM_VGIC_SYSREG_MASK) | KVM_REG_SIZE_U64; + + params.regval = *reg; + params.is_write = is_write; + params.is_aarch32 = false; + params.is_32bit = false; + + if (find_reg_by_id(sysreg, ¶ms, gic_v3_icc_reg_descs, + ARRAY_SIZE(gic_v3_icc_reg_descs))) + return 0; + + return -ENXIO; +} + +int vgic_v3_cpu_sysregs_uaccess(struct kvm_vcpu *vcpu, bool is_write, u64 id, + u64 *reg) +{ + struct sys_reg_params params; + const struct sys_reg_desc *r; + u64 sysreg = (id & KVM_DEV_ARM_VGIC_SYSREG_MASK) | KVM_REG_SIZE_U64; + + if (is_write) + params.regval = *reg; + params.is_write = is_write; + params.is_aarch32 = false; + params.is_32bit = false; + + r = find_reg_by_id(sysreg, ¶ms, gic_v3_icc_reg_descs, + ARRAY_SIZE(gic_v3_icc_reg_descs)); + if (!r) + return -ENXIO; + + if (!r->access(vcpu, ¶ms, r)) + return -EINVAL; + + if (!is_write) + *reg = params.regval; + + return 0; +} diff --git a/arch/mips/include/asm/kvm_host.h b/arch/mips/include/asm/kvm_host.h index bebec370324f..05e785fc061d 100644 --- a/arch/mips/include/asm/kvm_host.h +++ b/arch/mips/include/asm/kvm_host.h @@ -43,6 +43,7 @@ #define KVM_REG_MIPS_CP0_ENTRYHI MIPS_CP0_64(10, 0) #define KVM_REG_MIPS_CP0_COMPARE MIPS_CP0_32(11, 0) #define KVM_REG_MIPS_CP0_STATUS MIPS_CP0_32(12, 0) +#define KVM_REG_MIPS_CP0_INTCTL MIPS_CP0_32(12, 1) #define KVM_REG_MIPS_CP0_CAUSE MIPS_CP0_32(13, 0) #define KVM_REG_MIPS_CP0_EPC MIPS_CP0_64(14, 0) #define KVM_REG_MIPS_CP0_PRID MIPS_CP0_32(15, 0) @@ -64,7 +65,7 @@ #define KVM_REG_MIPS_CP0_KSCRATCH6 MIPS_CP0_64(31, 7) -#define KVM_MAX_VCPUS 1 +#define KVM_MAX_VCPUS 8 #define KVM_USER_MEM_SLOTS 8 /* memory slots that does not exposed to userspace */ #define KVM_PRIVATE_MEM_SLOTS 0 @@ -88,6 +89,7 @@ #define KVM_GUEST_KUSEG 0x00000000UL #define KVM_GUEST_KSEG0 0x40000000UL +#define KVM_GUEST_KSEG1 0x40000000UL #define KVM_GUEST_KSEG23 0x60000000UL #define KVM_GUEST_KSEGX(a) ((_ACAST32_(a)) & 0xe0000000) #define KVM_GUEST_CPHYSADDR(a) ((_ACAST32_(a)) & 0x1fffffff) @@ -104,7 +106,6 @@ #define KVM_GUEST_KSEG23ADDR(a) (KVM_GUEST_CPHYSADDR(a) | KVM_GUEST_KSEG23) #define KVM_INVALID_PAGE 0xdeadbeef -#define KVM_INVALID_INST 0xdeadbeef #define KVM_INVALID_ADDR 0xdeadbeef /* @@ -121,8 +122,6 @@ static inline bool kvm_is_error_hva(unsigned long addr) return IS_ERR_VALUE(addr); } -extern atomic_t kvm_mips_instance; - struct kvm_vm_stat { ulong remote_tlb_flush; }; @@ -156,12 +155,8 @@ struct kvm_arch_memory_slot { }; struct kvm_arch { - /* Guest GVA->HPA page table */ - unsigned long *guest_pmap; - unsigned long guest_pmap_npages; - - /* Wired host TLB used for the commpage */ - int commpage_tlb; + /* Guest physical mm */ + struct mm_struct gpa_mm; }; #define N_MIPS_COPROC_REGS 32 @@ -233,6 +228,7 @@ enum emulation_result { EMULATE_FAIL, /* can't emulate this instruction */ EMULATE_WAIT, /* WAIT instruction */ EMULATE_PRIV_FAIL, + EMULATE_EXCEPT, /* A guest exception has been generated */ }; #define mips3_paddr_to_tlbpfn(x) \ @@ -250,6 +246,7 @@ enum emulation_result { #define TLB_ASID(x) ((x).tlb_hi & KVM_ENTRYHI_ASID) #define TLB_LO_IDX(x, va) (((va) >> PAGE_SHIFT) & 1) #define TLB_IS_VALID(x, va) ((x).tlb_lo[TLB_LO_IDX(x, va)] & ENTRYLO_V) +#define TLB_IS_DIRTY(x, va) ((x).tlb_lo[TLB_LO_IDX(x, va)] & ENTRYLO_D) #define TLB_HI_VPN2_HIT(x, y) ((TLB_VPN2(x) & ~(x).tlb_mask) == \ ((y) & VPN2_MASK & ~(x).tlb_mask)) #define TLB_HI_ASID_HIT(x, y) (TLB_IS_GLOBAL(x) || \ @@ -261,6 +258,17 @@ struct kvm_mips_tlb { long tlb_lo[2]; }; +#define KVM_NR_MEM_OBJS 4 + +/* + * We don't want allocation failures within the mmu code, so we preallocate + * enough memory for a single page fault in a cache. + */ +struct kvm_mmu_memory_cache { + int nobjs; + void *objects[KVM_NR_MEM_OBJS]; +}; + #define KVM_MIPS_AUX_FPU 0x1 #define KVM_MIPS_AUX_MSA 0x2 @@ -275,6 +283,8 @@ struct kvm_vcpu_arch { unsigned long host_cp0_badvaddr; unsigned long host_cp0_epc; u32 host_cp0_cause; + u32 host_cp0_badinstr; + u32 host_cp0_badinstrp; /* GPRS */ unsigned long gprs[32]; @@ -318,20 +328,18 @@ struct kvm_vcpu_arch { /* Bitmask of pending exceptions to be cleared */ unsigned long pending_exceptions_clr; - /* Save/Restore the entryhi register when are are preempted/scheduled back in */ - unsigned long preempt_entryhi; - /* S/W Based TLB for guest */ struct kvm_mips_tlb guest_tlb[KVM_MIPS_GUEST_TLB_SIZE]; - /* Cached guest kernel/user ASIDs */ - u32 guest_user_asid[NR_CPUS]; - u32 guest_kernel_asid[NR_CPUS]; + /* Guest kernel/user [partial] mm */ struct mm_struct guest_kernel_mm, guest_user_mm; /* Guest ASID of last user mode execution */ unsigned int last_user_gasid; + /* Cache some mmu pages needed inside spinlock regions */ + struct kvm_mmu_memory_cache mmu_page_cache; + int last_sched_cpu; /* WAIT executed */ @@ -339,14 +347,15 @@ struct kvm_vcpu_arch { u8 fpu_enabled; u8 msa_enabled; - u8 kscratch_enabled; }; #define kvm_read_c0_guest_index(cop0) (cop0->reg[MIPS_CP0_TLB_INDEX][0]) #define kvm_write_c0_guest_index(cop0, val) (cop0->reg[MIPS_CP0_TLB_INDEX][0] = val) #define kvm_read_c0_guest_entrylo0(cop0) (cop0->reg[MIPS_CP0_TLB_LO0][0]) +#define kvm_write_c0_guest_entrylo0(cop0, val) (cop0->reg[MIPS_CP0_TLB_LO0][0] = (val)) #define kvm_read_c0_guest_entrylo1(cop0) (cop0->reg[MIPS_CP0_TLB_LO1][0]) +#define kvm_write_c0_guest_entrylo1(cop0, val) (cop0->reg[MIPS_CP0_TLB_LO1][0] = (val)) #define kvm_read_c0_guest_context(cop0) (cop0->reg[MIPS_CP0_TLB_CONTEXT][0]) #define kvm_write_c0_guest_context(cop0, val) (cop0->reg[MIPS_CP0_TLB_CONTEXT][0] = (val)) #define kvm_read_c0_guest_userlocal(cop0) (cop0->reg[MIPS_CP0_TLB_CONTEXT][2]) @@ -522,9 +531,17 @@ struct kvm_mips_callbacks { int (*handle_msa_fpe)(struct kvm_vcpu *vcpu); int (*handle_fpe)(struct kvm_vcpu *vcpu); int (*handle_msa_disabled)(struct kvm_vcpu *vcpu); - int (*vm_init)(struct kvm *kvm); int (*vcpu_init)(struct kvm_vcpu *vcpu); + void (*vcpu_uninit)(struct kvm_vcpu *vcpu); int (*vcpu_setup)(struct kvm_vcpu *vcpu); + void (*flush_shadow_all)(struct kvm *kvm); + /* + * Must take care of flushing any cached GPA PTEs (e.g. guest entries in + * VZ root TLB, or T&E GVA page tables and corresponding root TLB + * mappings). + */ + void (*flush_shadow_memslot)(struct kvm *kvm, + const struct kvm_memory_slot *slot); gpa_t (*gva_to_gpa)(gva_t gva); void (*queue_timer_int)(struct kvm_vcpu *vcpu); void (*dequeue_timer_int)(struct kvm_vcpu *vcpu); @@ -542,8 +559,10 @@ struct kvm_mips_callbacks { const struct kvm_one_reg *reg, s64 *v); int (*set_one_reg)(struct kvm_vcpu *vcpu, const struct kvm_one_reg *reg, s64 v); - int (*vcpu_get_regs)(struct kvm_vcpu *vcpu); - int (*vcpu_set_regs)(struct kvm_vcpu *vcpu); + int (*vcpu_load)(struct kvm_vcpu *vcpu, int cpu); + int (*vcpu_put)(struct kvm_vcpu *vcpu, int cpu); + int (*vcpu_run)(struct kvm_run *run, struct kvm_vcpu *vcpu); + void (*vcpu_reenter)(struct kvm_run *run, struct kvm_vcpu *vcpu); }; extern struct kvm_mips_callbacks *kvm_mips_callbacks; int kvm_mips_emulation_init(struct kvm_mips_callbacks **install_callbacks); @@ -556,6 +575,7 @@ extern int kvm_mips_handle_exit(struct kvm_run *run, struct kvm_vcpu *vcpu); /* Building of entry/exception code */ int kvm_mips_entry_setup(void); void *kvm_mips_build_vcpu_run(void *addr); +void *kvm_mips_build_tlb_refill_exception(void *addr, void *handler); void *kvm_mips_build_exception(void *addr, void *handler); void *kvm_mips_build_exit(void *addr); @@ -580,54 +600,125 @@ u32 kvm_get_user_asid(struct kvm_vcpu *vcpu); u32 kvm_get_commpage_asid (struct kvm_vcpu *vcpu); extern int kvm_mips_handle_kseg0_tlb_fault(unsigned long badbaddr, - struct kvm_vcpu *vcpu); + struct kvm_vcpu *vcpu, + bool write_fault); extern int kvm_mips_handle_commpage_tlb_fault(unsigned long badvaddr, struct kvm_vcpu *vcpu); extern int kvm_mips_handle_mapped_seg_tlb_fault(struct kvm_vcpu *vcpu, - struct kvm_mips_tlb *tlb); + struct kvm_mips_tlb *tlb, + unsigned long gva, + bool write_fault); extern enum emulation_result kvm_mips_handle_tlbmiss(u32 cause, u32 *opc, struct kvm_run *run, - struct kvm_vcpu *vcpu); - -extern enum emulation_result kvm_mips_handle_tlbmod(u32 cause, - u32 *opc, - struct kvm_run *run, - struct kvm_vcpu *vcpu); + struct kvm_vcpu *vcpu, + bool write_fault); extern void kvm_mips_dump_host_tlbs(void); extern void kvm_mips_dump_guest_tlbs(struct kvm_vcpu *vcpu); -extern int kvm_mips_host_tlb_write(struct kvm_vcpu *vcpu, unsigned long entryhi, - unsigned long entrylo0, - unsigned long entrylo1, - int flush_dcache_mask); -extern void kvm_mips_flush_host_tlb(int skip_kseg0); -extern int kvm_mips_host_tlb_inv(struct kvm_vcpu *vcpu, unsigned long entryhi); +extern int kvm_mips_host_tlb_inv(struct kvm_vcpu *vcpu, unsigned long entryhi, + bool user, bool kernel); extern int kvm_mips_guest_tlb_lookup(struct kvm_vcpu *vcpu, unsigned long entryhi); -extern int kvm_mips_host_tlb_lookup(struct kvm_vcpu *vcpu, unsigned long vaddr); -extern unsigned long kvm_mips_translate_guest_kseg0_to_hpa(struct kvm_vcpu *vcpu, - unsigned long gva); -extern void kvm_get_new_mmu_context(struct mm_struct *mm, unsigned long cpu, - struct kvm_vcpu *vcpu); -extern void kvm_local_flush_tlb_all(void); -extern void kvm_mips_alloc_new_mmu_context(struct kvm_vcpu *vcpu); -extern void kvm_mips_vcpu_load(struct kvm_vcpu *vcpu, int cpu); -extern void kvm_mips_vcpu_put(struct kvm_vcpu *vcpu); + +void kvm_mips_suspend_mm(int cpu); +void kvm_mips_resume_mm(int cpu); + +/* MMU handling */ + +/** + * enum kvm_mips_flush - Types of MMU flushes. + * @KMF_USER: Flush guest user virtual memory mappings. + * Guest USeg only. + * @KMF_KERN: Flush guest kernel virtual memory mappings. + * Guest USeg and KSeg2/3. + * @KMF_GPA: Flush guest physical memory mappings. + * Also includes KSeg0 if KMF_KERN is set. + */ +enum kvm_mips_flush { + KMF_USER = 0x0, + KMF_KERN = 0x1, + KMF_GPA = 0x2, +}; +void kvm_mips_flush_gva_pt(pgd_t *pgd, enum kvm_mips_flush flags); +bool kvm_mips_flush_gpa_pt(struct kvm *kvm, gfn_t start_gfn, gfn_t end_gfn); +int kvm_mips_mkclean_gpa_pt(struct kvm *kvm, gfn_t start_gfn, gfn_t end_gfn); +pgd_t *kvm_pgd_alloc(void); +void kvm_mmu_free_memory_caches(struct kvm_vcpu *vcpu); +void kvm_trap_emul_invalidate_gva(struct kvm_vcpu *vcpu, unsigned long addr, + bool user); +void kvm_trap_emul_gva_lockless_begin(struct kvm_vcpu *vcpu); +void kvm_trap_emul_gva_lockless_end(struct kvm_vcpu *vcpu); + +enum kvm_mips_fault_result { + KVM_MIPS_MAPPED = 0, + KVM_MIPS_GVA, + KVM_MIPS_GPA, + KVM_MIPS_TLB, + KVM_MIPS_TLBINV, + KVM_MIPS_TLBMOD, +}; +enum kvm_mips_fault_result kvm_trap_emul_gva_fault(struct kvm_vcpu *vcpu, + unsigned long gva, + bool write); + +#define KVM_ARCH_WANT_MMU_NOTIFIER +int kvm_unmap_hva(struct kvm *kvm, unsigned long hva); +int kvm_unmap_hva_range(struct kvm *kvm, + unsigned long start, unsigned long end); +void kvm_set_spte_hva(struct kvm *kvm, unsigned long hva, pte_t pte); +int kvm_age_hva(struct kvm *kvm, unsigned long start, unsigned long end); +int kvm_test_age_hva(struct kvm *kvm, unsigned long hva); + +static inline void kvm_arch_mmu_notifier_invalidate_page(struct kvm *kvm, + unsigned long address) +{ +} /* Emulation */ -u32 kvm_get_inst(u32 *opc, struct kvm_vcpu *vcpu); +int kvm_get_inst(u32 *opc, struct kvm_vcpu *vcpu, u32 *out); enum emulation_result update_pc(struct kvm_vcpu *vcpu, u32 cause); +int kvm_get_badinstr(u32 *opc, struct kvm_vcpu *vcpu, u32 *out); +int kvm_get_badinstrp(u32 *opc, struct kvm_vcpu *vcpu, u32 *out); + +/** + * kvm_is_ifetch_fault() - Find whether a TLBL exception is due to ifetch fault. + * @vcpu: Virtual CPU. + * + * Returns: Whether the TLBL exception was likely due to an instruction + * fetch fault rather than a data load fault. + */ +static inline bool kvm_is_ifetch_fault(struct kvm_vcpu_arch *vcpu) +{ + unsigned long badvaddr = vcpu->host_cp0_badvaddr; + unsigned long epc = msk_isa16_mode(vcpu->pc); + u32 cause = vcpu->host_cp0_cause; + + if (epc == badvaddr) + return true; + + /* + * Branches may be 32-bit or 16-bit instructions. + * This isn't exact, but we don't really support MIPS16 or microMIPS yet + * in KVM anyway. + */ + if ((cause & CAUSEF_BD) && badvaddr - epc <= 4) + return true; + + return false; +} extern enum emulation_result kvm_mips_emulate_inst(u32 cause, u32 *opc, struct kvm_run *run, struct kvm_vcpu *vcpu); +long kvm_mips_guest_exception_base(struct kvm_vcpu *vcpu); + extern enum emulation_result kvm_mips_emulate_syscall(u32 cause, u32 *opc, struct kvm_run *run, @@ -761,10 +852,6 @@ static inline void kvm_arch_sync_events(struct kvm *kvm) {} static inline void kvm_arch_free_memslot(struct kvm *kvm, struct kvm_memory_slot *free, struct kvm_memory_slot *dont) {} static inline void kvm_arch_memslots_updated(struct kvm *kvm, struct kvm_memslots *slots) {} -static inline void kvm_arch_flush_shadow_all(struct kvm *kvm) {} -static inline void kvm_arch_flush_shadow_memslot(struct kvm *kvm, - struct kvm_memory_slot *slot) {} -static inline void kvm_arch_vcpu_uninit(struct kvm_vcpu *vcpu) {} static inline void kvm_arch_sched_in(struct kvm_vcpu *vcpu, int cpu) {} static inline void kvm_arch_vcpu_blocking(struct kvm_vcpu *vcpu) {} static inline void kvm_arch_vcpu_unblocking(struct kvm_vcpu *vcpu) {} diff --git a/arch/mips/include/asm/mmu_context.h b/arch/mips/include/asm/mmu_context.h index ddd57ade1aa8..2abf94f72c0a 100644 --- a/arch/mips/include/asm/mmu_context.h +++ b/arch/mips/include/asm/mmu_context.h @@ -29,9 +29,11 @@ do { \ } \ } while (0) +extern void tlbmiss_handler_setup_pgd(unsigned long); + +/* Note: This is also implemented with uasm in arch/mips/kvm/entry.c */ #define TLBMISS_HANDLER_SETUP_PGD(pgd) \ do { \ - extern void tlbmiss_handler_setup_pgd(unsigned long); \ tlbmiss_handler_setup_pgd((unsigned long)(pgd)); \ htw_set_pwbase((unsigned long)pgd); \ } while (0) @@ -97,17 +99,12 @@ static inline void enter_lazy_tlb(struct mm_struct *mm, struct task_struct *tsk) static inline void get_new_mmu_context(struct mm_struct *mm, unsigned long cpu) { - extern void kvm_local_flush_tlb_all(void); unsigned long asid = asid_cache(cpu); if (!((asid += cpu_asid_inc()) & cpu_asid_mask(&cpu_data[cpu]))) { if (cpu_has_vtag_icache) flush_icache_all(); -#ifdef CONFIG_KVM - kvm_local_flush_tlb_all(); /* start new asid cycle */ -#else local_flush_tlb_all(); /* start new asid cycle */ -#endif if (!asid) /* fix version if needed */ asid = asid_first_version(cpu); } diff --git a/arch/mips/include/uapi/asm/kvm.h b/arch/mips/include/uapi/asm/kvm.h index 6985eb59b085..a8a0199bf760 100644 --- a/arch/mips/include/uapi/asm/kvm.h +++ b/arch/mips/include/uapi/asm/kvm.h @@ -19,6 +19,8 @@ * Some parts derived from the x86 version of this file. */ +#define __KVM_HAVE_READONLY_MEM + /* * for KVM_GET_REGS and KVM_SET_REGS * diff --git a/arch/mips/kvm/Kconfig b/arch/mips/kvm/Kconfig index 7c56d6b124d1..65067327db12 100644 --- a/arch/mips/kvm/Kconfig +++ b/arch/mips/kvm/Kconfig @@ -20,7 +20,9 @@ config KVM select EXPORT_UASM select PREEMPT_NOTIFIERS select ANON_INODES + select KVM_GENERIC_DIRTYLOG_READ_PROTECT select KVM_MMIO + select MMU_NOTIFIER select SRCU ---help--- Support for hosting Guest kernels. diff --git a/arch/mips/kvm/dyntrans.c b/arch/mips/kvm/dyntrans.c index 010cef240688..f8e772564d74 100644 --- a/arch/mips/kvm/dyntrans.c +++ b/arch/mips/kvm/dyntrans.c @@ -13,6 +13,7 @@ #include <linux/err.h> #include <linux/highmem.h> #include <linux/kvm_host.h> +#include <linux/uaccess.h> #include <linux/vmalloc.h> #include <linux/fs.h> #include <linux/bootmem.h> @@ -29,28 +30,37 @@ static int kvm_mips_trans_replace(struct kvm_vcpu *vcpu, u32 *opc, union mips_instruction replace) { - unsigned long paddr, flags; - void *vaddr; - - if (KVM_GUEST_KSEGX((unsigned long)opc) == KVM_GUEST_KSEG0) { - paddr = kvm_mips_translate_guest_kseg0_to_hpa(vcpu, - (unsigned long)opc); - vaddr = kmap_atomic(pfn_to_page(PHYS_PFN(paddr))); - vaddr += paddr & ~PAGE_MASK; - memcpy(vaddr, (void *)&replace, sizeof(u32)); - local_flush_icache_range((unsigned long)vaddr, - (unsigned long)vaddr + 32); - kunmap_atomic(vaddr); - } else if (KVM_GUEST_KSEGX((unsigned long) opc) == KVM_GUEST_KSEG23) { - local_irq_save(flags); - memcpy((void *)opc, (void *)&replace, sizeof(u32)); - __local_flush_icache_user_range((unsigned long)opc, - (unsigned long)opc + 32); - local_irq_restore(flags); - } else { - kvm_err("%s: Invalid address: %p\n", __func__, opc); - return -EFAULT; + unsigned long vaddr = (unsigned long)opc; + int err; + +retry: + /* The GVA page table is still active so use the Linux TLB handlers */ + kvm_trap_emul_gva_lockless_begin(vcpu); + err = put_user(replace.word, opc); + kvm_trap_emul_gva_lockless_end(vcpu); + + if (unlikely(err)) { + /* + * We write protect clean pages in GVA page table so normal + * Linux TLB mod handler doesn't silently dirty the page. + * Its also possible we raced with a GVA invalidation. + * Try to force the page to become dirty. + */ + err = kvm_trap_emul_gva_fault(vcpu, vaddr, true); + if (unlikely(err)) { + kvm_info("%s: Address unwriteable: %p\n", + __func__, opc); + return -EFAULT; + } + + /* + * Try again. This will likely trigger a TLB refill, which will + * fetch the new dirty entry from the GVA page table, which + * should then succeed. + */ + goto retry; } + __local_flush_icache_user_range(vaddr, vaddr + 4); return 0; } diff --git a/arch/mips/kvm/emulate.c b/arch/mips/kvm/emulate.c index aa0937423e28..d40cfaad4529 100644 --- a/arch/mips/kvm/emulate.c +++ b/arch/mips/kvm/emulate.c @@ -38,23 +38,25 @@ * Compute the return address and do emulate branch simulation, if required. * This function should be called only in branch delay slot active. */ -unsigned long kvm_compute_return_epc(struct kvm_vcpu *vcpu, - unsigned long instpc) +static int kvm_compute_return_epc(struct kvm_vcpu *vcpu, unsigned long instpc, + unsigned long *out) { unsigned int dspcontrol; union mips_instruction insn; struct kvm_vcpu_arch *arch = &vcpu->arch; long epc = instpc; - long nextpc = KVM_INVALID_INST; + long nextpc; + int err; - if (epc & 3) - goto unaligned; + if (epc & 3) { + kvm_err("%s: unaligned epc\n", __func__); + return -EINVAL; + } /* Read the instruction */ - insn.word = kvm_get_inst((u32 *) epc, vcpu); - - if (insn.word == KVM_INVALID_INST) - return KVM_INVALID_INST; + err = kvm_get_badinstrp((u32 *)epc, vcpu, &insn.word); + if (err) + return err; switch (insn.i_format.opcode) { /* jr and jalr are in r_format format. */ @@ -66,6 +68,8 @@ unsigned long kvm_compute_return_epc(struct kvm_vcpu *vcpu, case jr_op: nextpc = arch->gprs[insn.r_format.rs]; break; + default: + return -EINVAL; } break; @@ -114,8 +118,11 @@ unsigned long kvm_compute_return_epc(struct kvm_vcpu *vcpu, nextpc = epc; break; case bposge32_op: - if (!cpu_has_dsp) - goto sigill; + if (!cpu_has_dsp) { + kvm_err("%s: DSP branch but not DSP ASE\n", + __func__); + return -EINVAL; + } dspcontrol = rddsp(0x01); @@ -125,6 +132,8 @@ unsigned long kvm_compute_return_epc(struct kvm_vcpu *vcpu, epc += 8; nextpc = epc; break; + default: + return -EINVAL; } break; @@ -189,7 +198,7 @@ unsigned long kvm_compute_return_epc(struct kvm_vcpu *vcpu, /* And now the FPA/cp1 branch instructions. */ case cop1_op: kvm_err("%s: unsupported cop1_op\n", __func__); - break; + return -EINVAL; #ifdef CONFIG_CPU_MIPSR6 /* R6 added the following compact branches with forbidden slots */ @@ -198,19 +207,19 @@ unsigned long kvm_compute_return_epc(struct kvm_vcpu *vcpu, /* only rt == 0 isn't compact branch */ if (insn.i_format.rt != 0) goto compact_branch; - break; + return -EINVAL; case pop10_op: case pop30_op: /* only rs == rt == 0 is reserved, rest are compact branches */ if (insn.i_format.rs != 0 || insn.i_format.rt != 0) goto compact_branch; - break; + return -EINVAL; case pop66_op: case pop76_op: /* only rs == 0 isn't compact branch */ if (insn.i_format.rs != 0) goto compact_branch; - break; + return -EINVAL; compact_branch: /* * If we've hit an exception on the forbidden slot, then @@ -221,42 +230,74 @@ compact_branch: break; #else compact_branch: - /* Compact branches not supported before R6 */ - break; + /* Fall through - Compact branches not supported before R6 */ #endif + default: + return -EINVAL; } - return nextpc; - -unaligned: - kvm_err("%s: unaligned epc\n", __func__); - return nextpc; - -sigill: - kvm_err("%s: DSP branch but not DSP ASE\n", __func__); - return nextpc; + *out = nextpc; + return 0; } enum emulation_result update_pc(struct kvm_vcpu *vcpu, u32 cause) { - unsigned long branch_pc; - enum emulation_result er = EMULATE_DONE; + int err; if (cause & CAUSEF_BD) { - branch_pc = kvm_compute_return_epc(vcpu, vcpu->arch.pc); - if (branch_pc == KVM_INVALID_INST) { - er = EMULATE_FAIL; - } else { - vcpu->arch.pc = branch_pc; - kvm_debug("BD update_pc(): New PC: %#lx\n", - vcpu->arch.pc); - } - } else + err = kvm_compute_return_epc(vcpu, vcpu->arch.pc, + &vcpu->arch.pc); + if (err) + return EMULATE_FAIL; + } else { vcpu->arch.pc += 4; + } kvm_debug("update_pc(): New PC: %#lx\n", vcpu->arch.pc); - return er; + return EMULATE_DONE; +} + +/** + * kvm_get_badinstr() - Get bad instruction encoding. + * @opc: Guest pointer to faulting instruction. + * @vcpu: KVM VCPU information. + * + * Gets the instruction encoding of the faulting instruction, using the saved + * BadInstr register value if it exists, otherwise falling back to reading guest + * memory at @opc. + * + * Returns: The instruction encoding of the faulting instruction. + */ +int kvm_get_badinstr(u32 *opc, struct kvm_vcpu *vcpu, u32 *out) +{ + if (cpu_has_badinstr) { + *out = vcpu->arch.host_cp0_badinstr; + return 0; + } else { + return kvm_get_inst(opc, vcpu, out); + } +} + +/** + * kvm_get_badinstrp() - Get bad prior instruction encoding. + * @opc: Guest pointer to prior faulting instruction. + * @vcpu: KVM VCPU information. + * + * Gets the instruction encoding of the prior faulting instruction (the branch + * containing the delay slot which faulted), using the saved BadInstrP register + * value if it exists, otherwise falling back to reading guest memory at @opc. + * + * Returns: The instruction encoding of the prior faulting instruction. + */ +int kvm_get_badinstrp(u32 *opc, struct kvm_vcpu *vcpu, u32 *out) +{ + if (cpu_has_badinstrp) { + *out = vcpu->arch.host_cp0_badinstrp; + return 0; + } else { + return kvm_get_inst(opc, vcpu, out); + } } /** @@ -856,22 +897,30 @@ enum emulation_result kvm_mips_emul_tlbr(struct kvm_vcpu *vcpu) static void kvm_mips_invalidate_guest_tlb(struct kvm_vcpu *vcpu, struct kvm_mips_tlb *tlb) { + struct mm_struct *kern_mm = &vcpu->arch.guest_kernel_mm; + struct mm_struct *user_mm = &vcpu->arch.guest_user_mm; int cpu, i; bool user; /* No need to flush for entries which are already invalid */ if (!((tlb->tlb_lo[0] | tlb->tlb_lo[1]) & ENTRYLO_V)) return; + /* Don't touch host kernel page tables or TLB mappings */ + if ((unsigned long)tlb->tlb_hi > 0x7fffffff) + return; /* User address space doesn't need flushing for KSeg2/3 changes */ user = tlb->tlb_hi < KVM_GUEST_KSEG0; preempt_disable(); + /* Invalidate page table entries */ + kvm_trap_emul_invalidate_gva(vcpu, tlb->tlb_hi & VPN2_MASK, user); + /* * Probe the shadow host TLB for the entry being overwritten, if one * matches, invalidate it */ - kvm_mips_host_tlb_inv(vcpu, tlb->tlb_hi); + kvm_mips_host_tlb_inv(vcpu, tlb->tlb_hi, user, true); /* Invalidate the whole ASID on other CPUs */ cpu = smp_processor_id(); @@ -879,8 +928,8 @@ static void kvm_mips_invalidate_guest_tlb(struct kvm_vcpu *vcpu, if (i == cpu) continue; if (user) - vcpu->arch.guest_user_asid[i] = 0; - vcpu->arch.guest_kernel_asid[i] = 0; + cpu_context(i, user_mm) = 0; + cpu_context(i, kern_mm) = 0; } preempt_enable(); @@ -1017,7 +1066,7 @@ unsigned int kvm_mips_config4_wrmask(struct kvm_vcpu *vcpu) unsigned int mask = MIPS_CONF_M; /* KScrExist */ - mask |= (unsigned int)vcpu->arch.kscratch_enabled << 16; + mask |= 0xfc << MIPS_CONF4_KSCREXIST_SHIFT; return mask; } @@ -1056,6 +1105,7 @@ enum emulation_result kvm_mips_emulate_CP0(union mips_instruction inst, struct kvm_vcpu *vcpu) { struct mips_coproc *cop0 = vcpu->arch.cop0; + struct mm_struct *kern_mm = &vcpu->arch.guest_kernel_mm; enum emulation_result er = EMULATE_DONE; u32 rt, rd, sel; unsigned long curr_pc; @@ -1150,14 +1200,13 @@ enum emulation_result kvm_mips_emulate_CP0(union mips_instruction inst, er = EMULATE_FAIL; break; } -#define C0_EBASE_CORE_MASK 0xff if ((rd == MIPS_CP0_PRID) && (sel == 1)) { - /* Preserve CORE number */ - kvm_change_c0_guest_ebase(cop0, - ~(C0_EBASE_CORE_MASK), + /* + * Preserve core number, and keep the exception + * base in guest KSeg0. + */ + kvm_change_c0_guest_ebase(cop0, 0x1ffff000, vcpu->arch.gprs[rt]); - kvm_err("MTCz, cop0->reg[EBASE]: %#lx\n", - kvm_read_c0_guest_ebase(cop0)); } else if (rd == MIPS_CP0_TLB_HI && sel == 0) { u32 nasid = vcpu->arch.gprs[rt] & KVM_ENTRYHI_ASID; @@ -1169,6 +1218,17 @@ enum emulation_result kvm_mips_emulate_CP0(union mips_instruction inst, nasid); /* + * Flush entries from the GVA page + * tables. + * Guest user page table will get + * flushed lazily on re-entry to guest + * user if the guest ASID actually + * changes. + */ + kvm_mips_flush_gva_pt(kern_mm->pgd, + KMF_KERN); + + /* * Regenerate/invalidate kernel MMU * context. * The user MMU context will be @@ -1178,13 +1238,10 @@ enum emulation_result kvm_mips_emulate_CP0(union mips_instruction inst, */ preempt_disable(); cpu = smp_processor_id(); - kvm_get_new_mmu_context(&vcpu->arch.guest_kernel_mm, - cpu, vcpu); - vcpu->arch.guest_kernel_asid[cpu] = - vcpu->arch.guest_kernel_mm.context.asid[cpu]; + get_new_mmu_context(kern_mm, cpu); for_each_possible_cpu(i) if (i != cpu) - vcpu->arch.guest_kernel_asid[i] = 0; + cpu_context(i, kern_mm) = 0; preempt_enable(); } kvm_write_c0_guest_entryhi(cop0, @@ -1639,12 +1696,56 @@ enum emulation_result kvm_mips_emulate_load(union mips_instruction inst, return er; } +static enum emulation_result kvm_mips_guest_cache_op(int (*fn)(unsigned long), + unsigned long curr_pc, + unsigned long addr, + struct kvm_run *run, + struct kvm_vcpu *vcpu, + u32 cause) +{ + int err; + + for (;;) { + /* Carefully attempt the cache operation */ + kvm_trap_emul_gva_lockless_begin(vcpu); + err = fn(addr); + kvm_trap_emul_gva_lockless_end(vcpu); + + if (likely(!err)) + return EMULATE_DONE; + + /* + * Try to handle the fault and retry, maybe we just raced with a + * GVA invalidation. + */ + switch (kvm_trap_emul_gva_fault(vcpu, addr, false)) { + case KVM_MIPS_GVA: + case KVM_MIPS_GPA: + /* bad virtual or physical address */ + return EMULATE_FAIL; + case KVM_MIPS_TLB: + /* no matching guest TLB */ + vcpu->arch.host_cp0_badvaddr = addr; + vcpu->arch.pc = curr_pc; + kvm_mips_emulate_tlbmiss_ld(cause, NULL, run, vcpu); + return EMULATE_EXCEPT; + case KVM_MIPS_TLBINV: + /* invalid matching guest TLB */ + vcpu->arch.host_cp0_badvaddr = addr; + vcpu->arch.pc = curr_pc; + kvm_mips_emulate_tlbinv_ld(cause, NULL, run, vcpu); + return EMULATE_EXCEPT; + default: + break; + }; + } +} + enum emulation_result kvm_mips_emulate_cache(union mips_instruction inst, u32 *opc, u32 cause, struct kvm_run *run, struct kvm_vcpu *vcpu) { - struct mips_coproc *cop0 = vcpu->arch.cop0; enum emulation_result er = EMULATE_DONE; u32 cache, op_inst, op, base; s16 offset; @@ -1701,80 +1802,16 @@ enum emulation_result kvm_mips_emulate_cache(union mips_instruction inst, goto done; } - preempt_disable(); - if (KVM_GUEST_KSEGX(va) == KVM_GUEST_KSEG0) { - if (kvm_mips_host_tlb_lookup(vcpu, va) < 0 && - kvm_mips_handle_kseg0_tlb_fault(va, vcpu)) { - kvm_err("%s: handling mapped kseg0 tlb fault for %lx, vcpu: %p, ASID: %#lx\n", - __func__, va, vcpu, read_c0_entryhi()); - er = EMULATE_FAIL; - preempt_enable(); - goto done; - } - } else if ((KVM_GUEST_KSEGX(va) < KVM_GUEST_KSEG0) || - KVM_GUEST_KSEGX(va) == KVM_GUEST_KSEG23) { - int index; - - /* If an entry already exists then skip */ - if (kvm_mips_host_tlb_lookup(vcpu, va) >= 0) - goto skip_fault; - - /* - * If address not in the guest TLB, then give the guest a fault, - * the resulting handler will do the right thing - */ - index = kvm_mips_guest_tlb_lookup(vcpu, (va & VPN2_MASK) | - (kvm_read_c0_guest_entryhi - (cop0) & KVM_ENTRYHI_ASID)); - - if (index < 0) { - vcpu->arch.host_cp0_badvaddr = va; - vcpu->arch.pc = curr_pc; - er = kvm_mips_emulate_tlbmiss_ld(cause, NULL, run, - vcpu); - preempt_enable(); - goto dont_update_pc; - } else { - struct kvm_mips_tlb *tlb = &vcpu->arch.guest_tlb[index]; - /* - * Check if the entry is valid, if not then setup a TLB - * invalid exception to the guest - */ - if (!TLB_IS_VALID(*tlb, va)) { - vcpu->arch.host_cp0_badvaddr = va; - vcpu->arch.pc = curr_pc; - er = kvm_mips_emulate_tlbinv_ld(cause, NULL, - run, vcpu); - preempt_enable(); - goto dont_update_pc; - } - /* - * We fault an entry from the guest tlb to the - * shadow host TLB - */ - if (kvm_mips_handle_mapped_seg_tlb_fault(vcpu, tlb)) { - kvm_err("%s: handling mapped seg tlb fault for %lx, index: %u, vcpu: %p, ASID: %#lx\n", - __func__, va, index, vcpu, - read_c0_entryhi()); - er = EMULATE_FAIL; - preempt_enable(); - goto done; - } - } - } else { - kvm_err("INVALID CACHE INDEX/ADDRESS (cache: %#x, op: %#x, base[%d]: %#lx, offset: %#x\n", - cache, op, base, arch->gprs[base], offset); - er = EMULATE_FAIL; - preempt_enable(); - goto done; - - } - -skip_fault: /* XXXKYMA: Only a subset of cache ops are supported, used by Linux */ if (op_inst == Hit_Writeback_Inv_D || op_inst == Hit_Invalidate_D) { - flush_dcache_line(va); - + /* + * Perform the dcache part of icache synchronisation on the + * guest's behalf. + */ + er = kvm_mips_guest_cache_op(protected_writeback_dcache_line, + curr_pc, va, run, vcpu, cause); + if (er != EMULATE_DONE) + goto done; #ifdef CONFIG_KVM_MIPS_DYN_TRANS /* * Replace the CACHE instruction, with a SYNCI, not the same, @@ -1783,8 +1820,15 @@ skip_fault: kvm_mips_trans_cache_va(inst, opc, vcpu); #endif } else if (op_inst == Hit_Invalidate_I) { - flush_dcache_line(va); - flush_icache_line(va); + /* Perform the icache synchronisation on the guest's behalf */ + er = kvm_mips_guest_cache_op(protected_writeback_dcache_line, + curr_pc, va, run, vcpu, cause); + if (er != EMULATE_DONE) + goto done; + er = kvm_mips_guest_cache_op(protected_flush_icache_line, + curr_pc, va, run, vcpu, cause); + if (er != EMULATE_DONE) + goto done; #ifdef CONFIG_KVM_MIPS_DYN_TRANS /* Replace the CACHE instruction, with a SYNCI */ @@ -1796,17 +1840,13 @@ skip_fault: er = EMULATE_FAIL; } - preempt_enable(); done: /* Rollback PC only if emulation was unsuccessful */ if (er == EMULATE_FAIL) vcpu->arch.pc = curr_pc; - -dont_update_pc: - /* - * This is for exceptions whose emulation updates the PC, so do not - * overwrite the PC under any circumstances - */ + /* Guest exception needs guest to resume */ + if (er == EMULATE_EXCEPT) + er = EMULATE_DONE; return er; } @@ -1817,12 +1857,14 @@ enum emulation_result kvm_mips_emulate_inst(u32 cause, u32 *opc, { union mips_instruction inst; enum emulation_result er = EMULATE_DONE; + int err; /* Fetch the instruction. */ if (cause & CAUSEF_BD) opc += 1; - - inst.word = kvm_get_inst(opc, vcpu); + err = kvm_get_badinstr(opc, vcpu, &inst.word); + if (err) + return EMULATE_FAIL; switch (inst.r_format.opcode) { case cop0_op: @@ -1874,6 +1916,22 @@ unknown: return er; } +/** + * kvm_mips_guest_exception_base() - Find guest exception vector base address. + * + * Returns: The base address of the current guest exception vector, taking + * both Guest.CP0_Status.BEV and Guest.CP0_EBase into account. + */ +long kvm_mips_guest_exception_base(struct kvm_vcpu *vcpu) +{ + struct mips_coproc *cop0 = vcpu->arch.cop0; + + if (kvm_read_c0_guest_status(cop0) & ST0_BEV) + return KVM_GUEST_CKSEG1ADDR(0x1fc00200); + else + return kvm_read_c0_guest_ebase(cop0) & MIPS_EBASE_BASE; +} + enum emulation_result kvm_mips_emulate_syscall(u32 cause, u32 *opc, struct kvm_run *run, @@ -1899,7 +1957,7 @@ enum emulation_result kvm_mips_emulate_syscall(u32 cause, (EXCCODE_SYS << CAUSEB_EXCCODE)); /* Set PC to the exception entry point */ - arch->pc = KVM_GUEST_KSEG0 + 0x180; + arch->pc = kvm_mips_guest_exception_base(vcpu) + 0x180; } else { kvm_err("Trying to deliver SYSCALL when EXL is already set\n"); @@ -1933,13 +1991,13 @@ enum emulation_result kvm_mips_emulate_tlbmiss_ld(u32 cause, arch->pc); /* set pc to the exception entry point */ - arch->pc = KVM_GUEST_KSEG0 + 0x0; + arch->pc = kvm_mips_guest_exception_base(vcpu) + 0x0; } else { kvm_debug("[EXL == 1] delivering TLB MISS @ pc %#lx\n", arch->pc); - arch->pc = KVM_GUEST_KSEG0 + 0x180; + arch->pc = kvm_mips_guest_exception_base(vcpu) + 0x180; } kvm_change_c0_guest_cause(cop0, (0xff), @@ -1949,8 +2007,6 @@ enum emulation_result kvm_mips_emulate_tlbmiss_ld(u32 cause, kvm_write_c0_guest_badvaddr(cop0, vcpu->arch.host_cp0_badvaddr); /* XXXKYMA: is the context register used by linux??? */ kvm_write_c0_guest_entryhi(cop0, entryhi); - /* Blow away the shadow host TLBs */ - kvm_mips_flush_host_tlb(1); return EMULATE_DONE; } @@ -1978,16 +2034,14 @@ enum emulation_result kvm_mips_emulate_tlbinv_ld(u32 cause, kvm_debug("[EXL == 0] delivering TLB INV @ pc %#lx\n", arch->pc); - - /* set pc to the exception entry point */ - arch->pc = KVM_GUEST_KSEG0 + 0x180; - } else { kvm_debug("[EXL == 1] delivering TLB MISS @ pc %#lx\n", arch->pc); - arch->pc = KVM_GUEST_KSEG0 + 0x180; } + /* set pc to the exception entry point */ + arch->pc = kvm_mips_guest_exception_base(vcpu) + 0x180; + kvm_change_c0_guest_cause(cop0, (0xff), (EXCCODE_TLBL << CAUSEB_EXCCODE)); @@ -1995,8 +2049,6 @@ enum emulation_result kvm_mips_emulate_tlbinv_ld(u32 cause, kvm_write_c0_guest_badvaddr(cop0, vcpu->arch.host_cp0_badvaddr); /* XXXKYMA: is the context register used by linux??? */ kvm_write_c0_guest_entryhi(cop0, entryhi); - /* Blow away the shadow host TLBs */ - kvm_mips_flush_host_tlb(1); return EMULATE_DONE; } @@ -2025,11 +2077,11 @@ enum emulation_result kvm_mips_emulate_tlbmiss_st(u32 cause, arch->pc); /* Set PC to the exception entry point */ - arch->pc = KVM_GUEST_KSEG0 + 0x0; + arch->pc = kvm_mips_guest_exception_base(vcpu) + 0x0; } else { kvm_debug("[EXL == 1] Delivering TLB MISS @ pc %#lx\n", arch->pc); - arch->pc = KVM_GUEST_KSEG0 + 0x180; + arch->pc = kvm_mips_guest_exception_base(vcpu) + 0x180; } kvm_change_c0_guest_cause(cop0, (0xff), @@ -2039,8 +2091,6 @@ enum emulation_result kvm_mips_emulate_tlbmiss_st(u32 cause, kvm_write_c0_guest_badvaddr(cop0, vcpu->arch.host_cp0_badvaddr); /* XXXKYMA: is the context register used by linux??? */ kvm_write_c0_guest_entryhi(cop0, entryhi); - /* Blow away the shadow host TLBs */ - kvm_mips_flush_host_tlb(1); return EMULATE_DONE; } @@ -2067,15 +2117,14 @@ enum emulation_result kvm_mips_emulate_tlbinv_st(u32 cause, kvm_debug("[EXL == 0] Delivering TLB MISS @ pc %#lx\n", arch->pc); - - /* Set PC to the exception entry point */ - arch->pc = KVM_GUEST_KSEG0 + 0x180; } else { kvm_debug("[EXL == 1] Delivering TLB MISS @ pc %#lx\n", arch->pc); - arch->pc = KVM_GUEST_KSEG0 + 0x180; } + /* Set PC to the exception entry point */ + arch->pc = kvm_mips_guest_exception_base(vcpu) + 0x180; + kvm_change_c0_guest_cause(cop0, (0xff), (EXCCODE_TLBS << CAUSEB_EXCCODE)); @@ -2083,41 +2132,10 @@ enum emulation_result kvm_mips_emulate_tlbinv_st(u32 cause, kvm_write_c0_guest_badvaddr(cop0, vcpu->arch.host_cp0_badvaddr); /* XXXKYMA: is the context register used by linux??? */ kvm_write_c0_guest_entryhi(cop0, entryhi); - /* Blow away the shadow host TLBs */ - kvm_mips_flush_host_tlb(1); return EMULATE_DONE; } -/* TLBMOD: store into address matching TLB with Dirty bit off */ -enum emulation_result kvm_mips_handle_tlbmod(u32 cause, u32 *opc, - struct kvm_run *run, - struct kvm_vcpu *vcpu) -{ - enum emulation_result er = EMULATE_DONE; -#ifdef DEBUG - struct mips_coproc *cop0 = vcpu->arch.cop0; - unsigned long entryhi = (vcpu->arch.host_cp0_badvaddr & VPN2_MASK) | - (kvm_read_c0_guest_entryhi(cop0) & KVM_ENTRYHI_ASID); - int index; - - /* If address not in the guest TLB, then we are in trouble */ - index = kvm_mips_guest_tlb_lookup(vcpu, entryhi); - if (index < 0) { - /* XXXKYMA Invalidate and retry */ - kvm_mips_host_tlb_inv(vcpu, vcpu->arch.host_cp0_badvaddr); - kvm_err("%s: host got TLBMOD for %#lx but entry not present in Guest TLB\n", - __func__, entryhi); - kvm_mips_dump_guest_tlbs(vcpu); - kvm_mips_dump_host_tlbs(); - return EMULATE_FAIL; - } -#endif - - er = kvm_mips_emulate_tlbmod(cause, opc, run, vcpu); - return er; -} - enum emulation_result kvm_mips_emulate_tlbmod(u32 cause, u32 *opc, struct kvm_run *run, @@ -2140,14 +2158,13 @@ enum emulation_result kvm_mips_emulate_tlbmod(u32 cause, kvm_debug("[EXL == 0] Delivering TLB MOD @ pc %#lx\n", arch->pc); - - arch->pc = KVM_GUEST_KSEG0 + 0x180; } else { kvm_debug("[EXL == 1] Delivering TLB MOD @ pc %#lx\n", arch->pc); - arch->pc = KVM_GUEST_KSEG0 + 0x180; } + arch->pc = kvm_mips_guest_exception_base(vcpu) + 0x180; + kvm_change_c0_guest_cause(cop0, (0xff), (EXCCODE_MOD << CAUSEB_EXCCODE)); @@ -2155,8 +2172,6 @@ enum emulation_result kvm_mips_emulate_tlbmod(u32 cause, kvm_write_c0_guest_badvaddr(cop0, vcpu->arch.host_cp0_badvaddr); /* XXXKYMA: is the context register used by linux??? */ kvm_write_c0_guest_entryhi(cop0, entryhi); - /* Blow away the shadow host TLBs */ - kvm_mips_flush_host_tlb(1); return EMULATE_DONE; } @@ -2181,7 +2196,7 @@ enum emulation_result kvm_mips_emulate_fpu_exc(u32 cause, } - arch->pc = KVM_GUEST_KSEG0 + 0x180; + arch->pc = kvm_mips_guest_exception_base(vcpu) + 0x180; kvm_change_c0_guest_cause(cop0, (0xff), (EXCCODE_CPU << CAUSEB_EXCCODE)); @@ -2215,7 +2230,7 @@ enum emulation_result kvm_mips_emulate_ri_exc(u32 cause, (EXCCODE_RI << CAUSEB_EXCCODE)); /* Set PC to the exception entry point */ - arch->pc = KVM_GUEST_KSEG0 + 0x180; + arch->pc = kvm_mips_guest_exception_base(vcpu) + 0x180; } else { kvm_err("Trying to deliver RI when EXL is already set\n"); @@ -2250,7 +2265,7 @@ enum emulation_result kvm_mips_emulate_bp_exc(u32 cause, (EXCCODE_BP << CAUSEB_EXCCODE)); /* Set PC to the exception entry point */ - arch->pc = KVM_GUEST_KSEG0 + 0x180; + arch->pc = kvm_mips_guest_exception_base(vcpu) + 0x180; } else { kvm_err("Trying to deliver BP when EXL is already set\n"); @@ -2285,7 +2300,7 @@ enum emulation_result kvm_mips_emulate_trap_exc(u32 cause, (EXCCODE_TR << CAUSEB_EXCCODE)); /* Set PC to the exception entry point */ - arch->pc = KVM_GUEST_KSEG0 + 0x180; + arch->pc = kvm_mips_guest_exception_base(vcpu) + 0x180; } else { kvm_err("Trying to deliver TRAP when EXL is already set\n"); @@ -2320,7 +2335,7 @@ enum emulation_result kvm_mips_emulate_msafpe_exc(u32 cause, (EXCCODE_MSAFPE << CAUSEB_EXCCODE)); /* Set PC to the exception entry point */ - arch->pc = KVM_GUEST_KSEG0 + 0x180; + arch->pc = kvm_mips_guest_exception_base(vcpu) + 0x180; } else { kvm_err("Trying to deliver MSAFPE when EXL is already set\n"); @@ -2355,7 +2370,7 @@ enum emulation_result kvm_mips_emulate_fpe_exc(u32 cause, (EXCCODE_FPE << CAUSEB_EXCCODE)); /* Set PC to the exception entry point */ - arch->pc = KVM_GUEST_KSEG0 + 0x180; + arch->pc = kvm_mips_guest_exception_base(vcpu) + 0x180; } else { kvm_err("Trying to deliver FPE when EXL is already set\n"); @@ -2390,7 +2405,7 @@ enum emulation_result kvm_mips_emulate_msadis_exc(u32 cause, (EXCCODE_MSADIS << CAUSEB_EXCCODE)); /* Set PC to the exception entry point */ - arch->pc = KVM_GUEST_KSEG0 + 0x180; + arch->pc = kvm_mips_guest_exception_base(vcpu) + 0x180; } else { kvm_err("Trying to deliver MSADIS when EXL is already set\n"); @@ -2409,6 +2424,7 @@ enum emulation_result kvm_mips_handle_ri(u32 cause, u32 *opc, enum emulation_result er = EMULATE_DONE; unsigned long curr_pc; union mips_instruction inst; + int err; /* * Update PC and hold onto current PC in case there is @@ -2422,11 +2438,9 @@ enum emulation_result kvm_mips_handle_ri(u32 cause, u32 *opc, /* Fetch the instruction. */ if (cause & CAUSEF_BD) opc += 1; - - inst.word = kvm_get_inst(opc, vcpu); - - if (inst.word == KVM_INVALID_INST) { - kvm_err("%s: Cannot get inst @ %p\n", __func__, opc); + err = kvm_get_badinstr(opc, vcpu, &inst.word); + if (err) { + kvm_err("%s: Cannot get inst @ %p (%d)\n", __func__, opc, err); return EMULATE_FAIL; } @@ -2557,7 +2571,7 @@ static enum emulation_result kvm_mips_emulate_exc(u32 cause, (exccode << CAUSEB_EXCCODE)); /* Set PC to the exception entry point */ - arch->pc = KVM_GUEST_KSEG0 + 0x180; + arch->pc = kvm_mips_guest_exception_base(vcpu) + 0x180; kvm_write_c0_guest_badvaddr(cop0, vcpu->arch.host_cp0_badvaddr); kvm_debug("Delivering EXC %d @ pc %#lx, badVaddr: %#lx\n", @@ -2670,7 +2684,8 @@ enum emulation_result kvm_mips_check_privilege(u32 cause, enum emulation_result kvm_mips_handle_tlbmiss(u32 cause, u32 *opc, struct kvm_run *run, - struct kvm_vcpu *vcpu) + struct kvm_vcpu *vcpu, + bool write_fault) { enum emulation_result er = EMULATE_DONE; u32 exccode = (cause >> CAUSEB_EXCCODE) & 0x1f; @@ -2726,7 +2741,8 @@ enum emulation_result kvm_mips_handle_tlbmiss(u32 cause, * OK we have a Guest TLB entry, now inject it into the * shadow host TLB */ - if (kvm_mips_handle_mapped_seg_tlb_fault(vcpu, tlb)) { + if (kvm_mips_handle_mapped_seg_tlb_fault(vcpu, tlb, va, + write_fault)) { kvm_err("%s: handling mapped seg tlb fault for %lx, index: %u, vcpu: %p, ASID: %#lx\n", __func__, va, index, vcpu, read_c0_entryhi()); diff --git a/arch/mips/kvm/entry.c b/arch/mips/kvm/entry.c index e92fb190e2d6..c5b254c4d0da 100644 --- a/arch/mips/kvm/entry.c +++ b/arch/mips/kvm/entry.c @@ -12,8 +12,11 @@ */ #include <linux/kvm_host.h> +#include <linux/log2.h> +#include <asm/mmu_context.h> #include <asm/msa.h> #include <asm/setup.h> +#include <asm/tlbex.h> #include <asm/uasm.h> /* Register names */ @@ -50,6 +53,8 @@ /* Some CP0 registers */ #define C0_HWRENA 7, 0 #define C0_BADVADDR 8, 0 +#define C0_BADINSTR 8, 1 +#define C0_BADINSTRP 8, 2 #define C0_ENTRYHI 10, 0 #define C0_STATUS 12, 0 #define C0_CAUSE 13, 0 @@ -89,6 +94,21 @@ static void *kvm_mips_build_ret_from_exit(void *addr); static void *kvm_mips_build_ret_to_guest(void *addr); static void *kvm_mips_build_ret_to_host(void *addr); +/* + * The version of this function in tlbex.c uses current_cpu_type(), but for KVM + * we assume symmetry. + */ +static int c0_kscratch(void) +{ + switch (boot_cpu_type()) { + case CPU_XLP: + case CPU_XLR: + return 22; + default: + return 31; + } +} + /** * kvm_mips_entry_setup() - Perform global setup for entry code. * @@ -103,18 +123,21 @@ int kvm_mips_entry_setup(void) * We prefer to use KScratchN registers if they are available over the * defaults above, which may not work on all cores. */ - unsigned int kscratch_mask = cpu_data[0].kscratch_mask & 0xfc; + unsigned int kscratch_mask = cpu_data[0].kscratch_mask; + + if (pgd_reg != -1) + kscratch_mask &= ~BIT(pgd_reg); /* Pick a scratch register for storing VCPU */ if (kscratch_mask) { - scratch_vcpu[0] = 31; + scratch_vcpu[0] = c0_kscratch(); scratch_vcpu[1] = ffs(kscratch_mask) - 1; kscratch_mask &= ~BIT(scratch_vcpu[1]); } /* Pick a scratch register to use as a temp for saving state */ if (kscratch_mask) { - scratch_tmp[0] = 31; + scratch_tmp[0] = c0_kscratch(); scratch_tmp[1] = ffs(kscratch_mask) - 1; kscratch_mask &= ~BIT(scratch_tmp[1]); } @@ -130,7 +153,7 @@ static void kvm_mips_build_save_scratch(u32 **p, unsigned int tmp, UASM_i_SW(p, tmp, offsetof(struct pt_regs, cp0_epc), frame); /* Save the temp scratch register value in cp0_cause of stack frame */ - if (scratch_tmp[0] == 31) { + if (scratch_tmp[0] == c0_kscratch()) { UASM_i_MFC0(p, tmp, scratch_tmp[0], scratch_tmp[1]); UASM_i_SW(p, tmp, offsetof(struct pt_regs, cp0_cause), frame); } @@ -146,7 +169,7 @@ static void kvm_mips_build_restore_scratch(u32 **p, unsigned int tmp, UASM_i_LW(p, tmp, offsetof(struct pt_regs, cp0_epc), frame); UASM_i_MTC0(p, tmp, scratch_vcpu[0], scratch_vcpu[1]); - if (scratch_tmp[0] == 31) { + if (scratch_tmp[0] == c0_kscratch()) { UASM_i_LW(p, tmp, offsetof(struct pt_regs, cp0_cause), frame); UASM_i_MTC0(p, tmp, scratch_tmp[0], scratch_tmp[1]); } @@ -286,23 +309,26 @@ static void *kvm_mips_build_enter_guest(void *addr) uasm_i_andi(&p, T0, T0, KSU_USER | ST0_ERL | ST0_EXL); uasm_i_xori(&p, T0, T0, KSU_USER); uasm_il_bnez(&p, &r, T0, label_kernel_asid); - UASM_i_ADDIU(&p, T1, K1, - offsetof(struct kvm_vcpu_arch, guest_kernel_asid)); + UASM_i_ADDIU(&p, T1, K1, offsetof(struct kvm_vcpu_arch, + guest_kernel_mm.context.asid)); /* else user */ - UASM_i_ADDIU(&p, T1, K1, - offsetof(struct kvm_vcpu_arch, guest_user_asid)); + UASM_i_ADDIU(&p, T1, K1, offsetof(struct kvm_vcpu_arch, + guest_user_mm.context.asid)); uasm_l_kernel_asid(&l, p); /* t1: contains the base of the ASID array, need to get the cpu id */ /* smp_processor_id */ uasm_i_lw(&p, T2, offsetof(struct thread_info, cpu), GP); - /* x4 */ - uasm_i_sll(&p, T2, T2, 2); + /* index the ASID array */ + uasm_i_sll(&p, T2, T2, ilog2(sizeof(long))); UASM_i_ADDU(&p, T3, T1, T2); - uasm_i_lw(&p, K0, 0, T3); + UASM_i_LW(&p, K0, 0, T3); #ifdef CONFIG_MIPS_ASID_BITS_VARIABLE - /* x sizeof(struct cpuinfo_mips)/4 */ - uasm_i_addiu(&p, T3, ZERO, sizeof(struct cpuinfo_mips)/4); + /* + * reuse ASID array offset + * cpuinfo_mips is a multiple of sizeof(long) + */ + uasm_i_addiu(&p, T3, ZERO, sizeof(struct cpuinfo_mips)/sizeof(long)); uasm_i_mul(&p, T2, T2, T3); UASM_i_LA_mostly(&p, AT, (long)&cpu_data[0].asid_mask); @@ -312,7 +338,20 @@ static void *kvm_mips_build_enter_guest(void *addr) #else uasm_i_andi(&p, K0, K0, MIPS_ENTRYHI_ASID); #endif - uasm_i_mtc0(&p, K0, C0_ENTRYHI); + + /* + * Set up KVM T&E GVA pgd. + * This does roughly the same as TLBMISS_HANDLER_SETUP_PGD(): + * - call tlbmiss_handler_setup_pgd(mm->pgd) + * - but skips write into CP0_PWBase for now + */ + UASM_i_LW(&p, A0, (int)offsetof(struct mm_struct, pgd) - + (int)offsetof(struct mm_struct, context.asid), T1); + + UASM_i_LA(&p, T9, (unsigned long)tlbmiss_handler_setup_pgd); + uasm_i_jalr(&p, RA, T9); + uasm_i_mtc0(&p, K0, C0_ENTRYHI); + uasm_i_ehb(&p); /* Disable RDHWR access */ @@ -348,6 +387,80 @@ static void *kvm_mips_build_enter_guest(void *addr) } /** + * kvm_mips_build_tlb_refill_exception() - Assemble TLB refill handler. + * @addr: Address to start writing code. + * @handler: Address of common handler (within range of @addr). + * + * Assemble TLB refill exception fast path handler for guest execution. + * + * Returns: Next address after end of written function. + */ +void *kvm_mips_build_tlb_refill_exception(void *addr, void *handler) +{ + u32 *p = addr; + struct uasm_label labels[2]; + struct uasm_reloc relocs[2]; + struct uasm_label *l = labels; + struct uasm_reloc *r = relocs; + + memset(labels, 0, sizeof(labels)); + memset(relocs, 0, sizeof(relocs)); + + /* Save guest k1 into scratch register */ + UASM_i_MTC0(&p, K1, scratch_tmp[0], scratch_tmp[1]); + + /* Get the VCPU pointer from the VCPU scratch register */ + UASM_i_MFC0(&p, K1, scratch_vcpu[0], scratch_vcpu[1]); + + /* Save guest k0 into VCPU structure */ + UASM_i_SW(&p, K0, offsetof(struct kvm_vcpu, arch.gprs[K0]), K1); + + /* + * Some of the common tlbex code uses current_cpu_type(). For KVM we + * assume symmetry and just disable preemption to silence the warning. + */ + preempt_disable(); + + /* + * Now for the actual refill bit. A lot of this can be common with the + * Linux TLB refill handler, however we don't need to handle so many + * cases. We only need to handle user mode refills, and user mode runs + * with 32-bit addressing. + * + * Therefore the branch to label_vmalloc generated by build_get_pmde64() + * that isn't resolved should never actually get taken and is harmless + * to leave in place for now. + */ + +#ifdef CONFIG_64BIT + build_get_pmde64(&p, &l, &r, K0, K1); /* get pmd in K1 */ +#else + build_get_pgde32(&p, K0, K1); /* get pgd in K1 */ +#endif + + /* we don't support huge pages yet */ + + build_get_ptep(&p, K0, K1); + build_update_entries(&p, K0, K1); + build_tlb_write_entry(&p, &l, &r, tlb_random); + + preempt_enable(); + + /* Get the VCPU pointer from the VCPU scratch register again */ + UASM_i_MFC0(&p, K1, scratch_vcpu[0], scratch_vcpu[1]); + + /* Restore the guest's k0/k1 registers */ + UASM_i_LW(&p, K0, offsetof(struct kvm_vcpu, arch.gprs[K0]), K1); + uasm_i_ehb(&p); + UASM_i_MFC0(&p, K1, scratch_tmp[0], scratch_tmp[1]); + + /* Jump to guest */ + uasm_i_eret(&p); + + return p; +} + +/** * kvm_mips_build_exception() - Assemble first level guest exception handler. * @addr: Address to start writing code. * @handler: Address of common handler (within range of @addr). @@ -468,6 +581,18 @@ void *kvm_mips_build_exit(void *addr) uasm_i_mfc0(&p, K0, C0_CAUSE); uasm_i_sw(&p, K0, offsetof(struct kvm_vcpu_arch, host_cp0_cause), K1); + if (cpu_has_badinstr) { + uasm_i_mfc0(&p, K0, C0_BADINSTR); + uasm_i_sw(&p, K0, offsetof(struct kvm_vcpu_arch, + host_cp0_badinstr), K1); + } + + if (cpu_has_badinstrp) { + uasm_i_mfc0(&p, K0, C0_BADINSTRP); + uasm_i_sw(&p, K0, offsetof(struct kvm_vcpu_arch, + host_cp0_badinstrp), K1); + } + /* Now restore the host state just enough to run the handlers */ /* Switch EBASE to the one used by Linux */ diff --git a/arch/mips/kvm/interrupt.c b/arch/mips/kvm/interrupt.c index e88403b3dcdd..aa0a1a00faf6 100644 --- a/arch/mips/kvm/interrupt.c +++ b/arch/mips/kvm/interrupt.c @@ -183,10 +183,11 @@ int kvm_mips_irq_deliver_cb(struct kvm_vcpu *vcpu, unsigned int priority, (exccode << CAUSEB_EXCCODE)); /* XXXSL Set PC to the interrupt exception entry point */ + arch->pc = kvm_mips_guest_exception_base(vcpu); if (kvm_read_c0_guest_cause(cop0) & CAUSEF_IV) - arch->pc = KVM_GUEST_KSEG0 + 0x200; + arch->pc += 0x200; else - arch->pc = KVM_GUEST_KSEG0 + 0x180; + arch->pc += 0x180; clear_bit(priority, &vcpu->arch.pending_exceptions); } diff --git a/arch/mips/kvm/mips.c b/arch/mips/kvm/mips.c index 29ec9ab3fd55..ed81e5ac1426 100644 --- a/arch/mips/kvm/mips.c +++ b/arch/mips/kvm/mips.c @@ -22,6 +22,7 @@ #include <asm/page.h> #include <asm/cacheflush.h> #include <asm/mmu_context.h> +#include <asm/pgalloc.h> #include <asm/pgtable.h> #include <linux/kvm_host.h> @@ -63,18 +64,6 @@ struct kvm_stats_debugfs_item debugfs_entries[] = { {NULL} }; -static int kvm_mips_reset_vcpu(struct kvm_vcpu *vcpu) -{ - int i; - - for_each_possible_cpu(i) { - vcpu->arch.guest_kernel_asid[i] = 0; - vcpu->arch.guest_user_asid[i] = 0; - } - - return 0; -} - /* * XXXKYMA: We are simulatoring a processor that has the WII bit set in * Config7, so we are "runnable" if interrupts are pending @@ -104,39 +93,12 @@ void kvm_arch_check_processor_compat(void *rtn) *(int *)rtn = 0; } -static void kvm_mips_init_tlbs(struct kvm *kvm) -{ - unsigned long wired; - - /* - * Add a wired entry to the TLB, it is used to map the commpage to - * the Guest kernel - */ - wired = read_c0_wired(); - write_c0_wired(wired + 1); - mtc0_tlbw_hazard(); - kvm->arch.commpage_tlb = wired; - - kvm_debug("[%d] commpage TLB: %d\n", smp_processor_id(), - kvm->arch.commpage_tlb); -} - -static void kvm_mips_init_vm_percpu(void *arg) -{ - struct kvm *kvm = (struct kvm *)arg; - - kvm_mips_init_tlbs(kvm); - kvm_mips_callbacks->vm_init(kvm); - -} - int kvm_arch_init_vm(struct kvm *kvm, unsigned long type) { - if (atomic_inc_return(&kvm_mips_instance) == 1) { - kvm_debug("%s: 1st KVM instance, setup host TLB parameters\n", - __func__); - on_each_cpu(kvm_mips_init_vm_percpu, kvm, 1); - } + /* Allocate page table to map GPA -> RPA */ + kvm->arch.gpa_mm.pgd = kvm_pgd_alloc(); + if (!kvm->arch.gpa_mm.pgd) + return -ENOMEM; return 0; } @@ -156,13 +118,6 @@ void kvm_mips_free_vcpus(struct kvm *kvm) unsigned int i; struct kvm_vcpu *vcpu; - /* Put the pages we reserved for the guest pmap */ - for (i = 0; i < kvm->arch.guest_pmap_npages; i++) { - if (kvm->arch.guest_pmap[i] != KVM_INVALID_PAGE) - kvm_release_pfn_clean(kvm->arch.guest_pmap[i]); - } - kfree(kvm->arch.guest_pmap); - kvm_for_each_vcpu(i, vcpu, kvm) { kvm_arch_vcpu_free(vcpu); } @@ -177,25 +132,17 @@ void kvm_mips_free_vcpus(struct kvm *kvm) mutex_unlock(&kvm->lock); } -static void kvm_mips_uninit_tlbs(void *arg) +static void kvm_mips_free_gpa_pt(struct kvm *kvm) { - /* Restore wired count */ - write_c0_wired(0); - mtc0_tlbw_hazard(); - /* Clear out all the TLBs */ - kvm_local_flush_tlb_all(); + /* It should always be safe to remove after flushing the whole range */ + WARN_ON(!kvm_mips_flush_gpa_pt(kvm, 0, ~0)); + pgd_free(NULL, kvm->arch.gpa_mm.pgd); } void kvm_arch_destroy_vm(struct kvm *kvm) { kvm_mips_free_vcpus(kvm); - - /* If this is the last instance, restore wired count */ - if (atomic_dec_return(&kvm_mips_instance) == 0) { - kvm_debug("%s: last KVM instance, restoring TLB parameters\n", - __func__); - on_each_cpu(kvm_mips_uninit_tlbs, NULL, 1); - } + kvm_mips_free_gpa_pt(kvm); } long kvm_arch_dev_ioctl(struct file *filp, unsigned int ioctl, @@ -210,6 +157,32 @@ int kvm_arch_create_memslot(struct kvm *kvm, struct kvm_memory_slot *slot, return 0; } +void kvm_arch_flush_shadow_all(struct kvm *kvm) +{ + /* Flush whole GPA */ + kvm_mips_flush_gpa_pt(kvm, 0, ~0); + + /* Let implementation do the rest */ + kvm_mips_callbacks->flush_shadow_all(kvm); +} + +void kvm_arch_flush_shadow_memslot(struct kvm *kvm, + struct kvm_memory_slot *slot) +{ + /* + * The slot has been made invalid (ready for moving or deletion), so we + * need to ensure that it can no longer be accessed by any guest VCPUs. + */ + + spin_lock(&kvm->mmu_lock); + /* Flush slot from GPA */ + kvm_mips_flush_gpa_pt(kvm, slot->base_gfn, + slot->base_gfn + slot->npages - 1); + /* Let implementation do the rest */ + kvm_mips_callbacks->flush_shadow_memslot(kvm, slot); + spin_unlock(&kvm->mmu_lock); +} + int kvm_arch_prepare_memory_region(struct kvm *kvm, struct kvm_memory_slot *memslot, const struct kvm_userspace_memory_region *mem, @@ -224,35 +197,32 @@ void kvm_arch_commit_memory_region(struct kvm *kvm, const struct kvm_memory_slot *new, enum kvm_mr_change change) { - unsigned long npages = 0; - int i; + int needs_flush; kvm_debug("%s: kvm: %p slot: %d, GPA: %llx, size: %llx, QVA: %llx\n", __func__, kvm, mem->slot, mem->guest_phys_addr, mem->memory_size, mem->userspace_addr); - /* Setup Guest PMAP table */ - if (!kvm->arch.guest_pmap) { - if (mem->slot == 0) - npages = mem->memory_size >> PAGE_SHIFT; - - if (npages) { - kvm->arch.guest_pmap_npages = npages; - kvm->arch.guest_pmap = - kzalloc(npages * sizeof(unsigned long), GFP_KERNEL); - - if (!kvm->arch.guest_pmap) { - kvm_err("Failed to allocate guest PMAP\n"); - return; - } - - kvm_debug("Allocated space for Guest PMAP Table (%ld pages) @ %p\n", - npages, kvm->arch.guest_pmap); - - /* Now setup the page table */ - for (i = 0; i < npages; i++) - kvm->arch.guest_pmap[i] = KVM_INVALID_PAGE; - } + /* + * If dirty page logging is enabled, write protect all pages in the slot + * ready for dirty logging. + * + * There is no need to do this in any of the following cases: + * CREATE: No dirty mappings will already exist. + * MOVE/DELETE: The old mappings will already have been cleaned up by + * kvm_arch_flush_shadow_memslot() + */ + if (change == KVM_MR_FLAGS_ONLY && + (!(old->flags & KVM_MEM_LOG_DIRTY_PAGES) && + new->flags & KVM_MEM_LOG_DIRTY_PAGES)) { + spin_lock(&kvm->mmu_lock); + /* Write protect GPA page table entries */ + needs_flush = kvm_mips_mkclean_gpa_pt(kvm, new->base_gfn, + new->base_gfn + new->npages - 1); + /* Let implementation do the rest */ + if (needs_flush) + kvm_mips_callbacks->flush_shadow_memslot(kvm, new); + spin_unlock(&kvm->mmu_lock); } } @@ -276,7 +246,7 @@ static inline void dump_handler(const char *symbol, void *start, void *end) struct kvm_vcpu *kvm_arch_vcpu_create(struct kvm *kvm, unsigned int id) { int err, size; - void *gebase, *p, *handler; + void *gebase, *p, *handler, *refill_start, *refill_end; int i; struct kvm_vcpu *vcpu = kzalloc(sizeof(struct kvm_vcpu), GFP_KERNEL); @@ -329,8 +299,9 @@ struct kvm_vcpu *kvm_arch_vcpu_create(struct kvm *kvm, unsigned int id) /* Build guest exception vectors dynamically in unmapped memory */ handler = gebase + 0x2000; - /* TLB Refill, EXL = 0 */ - kvm_mips_build_exception(gebase, handler); + /* TLB refill */ + refill_start = gebase; + refill_end = kvm_mips_build_tlb_refill_exception(refill_start, handler); /* General Exception Entry point */ kvm_mips_build_exception(gebase + 0x180, handler); @@ -356,6 +327,7 @@ struct kvm_vcpu *kvm_arch_vcpu_create(struct kvm *kvm, unsigned int id) pr_debug("#include <asm/regdef.h>\n"); pr_debug("\n"); dump_handler("kvm_vcpu_run", vcpu->arch.vcpu_run, p); + dump_handler("kvm_tlb_refill", refill_start, refill_end); dump_handler("kvm_gen_exc", gebase + 0x180, gebase + 0x200); dump_handler("kvm_exit", gebase + 0x2000, vcpu->arch.vcpu_run); @@ -406,6 +378,7 @@ void kvm_arch_vcpu_free(struct kvm_vcpu *vcpu) kvm_mips_dump_stats(vcpu); + kvm_mmu_free_memory_caches(vcpu); kfree(vcpu->arch.guest_ebase); kfree(vcpu->arch.kseg0_commpage); kfree(vcpu); @@ -422,37 +395,9 @@ int kvm_arch_vcpu_ioctl_set_guest_debug(struct kvm_vcpu *vcpu, return -ENOIOCTLCMD; } -/* Must be called with preemption disabled, just before entering guest */ -static void kvm_mips_check_asids(struct kvm_vcpu *vcpu) -{ - struct mips_coproc *cop0 = vcpu->arch.cop0; - int i, cpu = smp_processor_id(); - unsigned int gasid; - - /* - * Lazy host ASID regeneration for guest user mode. - * If the guest ASID has changed since the last guest usermode - * execution, regenerate the host ASID so as to invalidate stale TLB - * entries. - */ - if (!KVM_GUEST_KERNEL_MODE(vcpu)) { - gasid = kvm_read_c0_guest_entryhi(cop0) & KVM_ENTRYHI_ASID; - if (gasid != vcpu->arch.last_user_gasid) { - kvm_get_new_mmu_context(&vcpu->arch.guest_user_mm, cpu, - vcpu); - vcpu->arch.guest_user_asid[cpu] = - vcpu->arch.guest_user_mm.context.asid[cpu]; - for_each_possible_cpu(i) - if (i != cpu) - vcpu->arch.guest_user_asid[cpu] = 0; - vcpu->arch.last_user_gasid = gasid; - } - } -} - int kvm_arch_vcpu_ioctl_run(struct kvm_vcpu *vcpu, struct kvm_run *run) { - int r = 0; + int r = -EINTR; sigset_t sigsaved; if (vcpu->sigset_active) @@ -464,31 +409,30 @@ int kvm_arch_vcpu_ioctl_run(struct kvm_vcpu *vcpu, struct kvm_run *run) vcpu->mmio_needed = 0; } + if (run->immediate_exit) + goto out; + lose_fpu(1); local_irq_disable(); - /* Check if we have any exceptions/interrupts pending */ - kvm_mips_deliver_interrupts(vcpu, - kvm_read_c0_guest_cause(vcpu->arch.cop0)); - guest_enter_irqoff(); - - /* Disable hardware page table walking while in guest */ - htw_stop(); - trace_kvm_enter(vcpu); - kvm_mips_check_asids(vcpu); - - r = vcpu->arch.vcpu_run(run, vcpu); - trace_kvm_out(vcpu); + /* + * Make sure the read of VCPU requests in vcpu_run() callback is not + * reordered ahead of the write to vcpu->mode, or we could miss a TLB + * flush request while the requester sees the VCPU as outside of guest + * mode and not needing an IPI. + */ + smp_store_mb(vcpu->mode, IN_GUEST_MODE); - /* Re-enable HTW before enabling interrupts */ - htw_start(); + r = kvm_mips_callbacks->vcpu_run(run, vcpu); + trace_kvm_out(vcpu); guest_exit_irqoff(); local_irq_enable(); +out: if (vcpu->sigset_active) sigprocmask(SIG_SETMASK, &sigsaved, NULL); @@ -580,33 +524,6 @@ static u64 kvm_mips_get_one_regs[] = { KVM_REG_MIPS_LO, #endif KVM_REG_MIPS_PC, - - KVM_REG_MIPS_CP0_INDEX, - KVM_REG_MIPS_CP0_CONTEXT, - KVM_REG_MIPS_CP0_USERLOCAL, - KVM_REG_MIPS_CP0_PAGEMASK, - KVM_REG_MIPS_CP0_WIRED, - KVM_REG_MIPS_CP0_HWRENA, - KVM_REG_MIPS_CP0_BADVADDR, - KVM_REG_MIPS_CP0_COUNT, - KVM_REG_MIPS_CP0_ENTRYHI, - KVM_REG_MIPS_CP0_COMPARE, - KVM_REG_MIPS_CP0_STATUS, - KVM_REG_MIPS_CP0_CAUSE, - KVM_REG_MIPS_CP0_EPC, - KVM_REG_MIPS_CP0_PRID, - KVM_REG_MIPS_CP0_CONFIG, - KVM_REG_MIPS_CP0_CONFIG1, - KVM_REG_MIPS_CP0_CONFIG2, - KVM_REG_MIPS_CP0_CONFIG3, - KVM_REG_MIPS_CP0_CONFIG4, - KVM_REG_MIPS_CP0_CONFIG5, - KVM_REG_MIPS_CP0_CONFIG7, - KVM_REG_MIPS_CP0_ERROREPC, - - KVM_REG_MIPS_COUNT_CTL, - KVM_REG_MIPS_COUNT_RESUME, - KVM_REG_MIPS_COUNT_HZ, }; static u64 kvm_mips_get_one_regs_fpu[] = { @@ -619,15 +536,6 @@ static u64 kvm_mips_get_one_regs_msa[] = { KVM_REG_MIPS_MSA_CSR, }; -static u64 kvm_mips_get_one_regs_kscratch[] = { - KVM_REG_MIPS_CP0_KSCRATCH1, - KVM_REG_MIPS_CP0_KSCRATCH2, - KVM_REG_MIPS_CP0_KSCRATCH3, - KVM_REG_MIPS_CP0_KSCRATCH4, - KVM_REG_MIPS_CP0_KSCRATCH5, - KVM_REG_MIPS_CP0_KSCRATCH6, -}; - static unsigned long kvm_mips_num_regs(struct kvm_vcpu *vcpu) { unsigned long ret; @@ -641,7 +549,6 @@ static unsigned long kvm_mips_num_regs(struct kvm_vcpu *vcpu) } if (kvm_mips_guest_can_have_msa(&vcpu->arch)) ret += ARRAY_SIZE(kvm_mips_get_one_regs_msa) + 32; - ret += __arch_hweight8(vcpu->arch.kscratch_enabled); ret += kvm_mips_callbacks->num_regs(vcpu); return ret; @@ -694,16 +601,6 @@ static int kvm_mips_copy_reg_indices(struct kvm_vcpu *vcpu, u64 __user *indices) } } - for (i = 0; i < 6; ++i) { - if (!(vcpu->arch.kscratch_enabled & BIT(i + 2))) - continue; - - if (copy_to_user(indices, &kvm_mips_get_one_regs_kscratch[i], - sizeof(kvm_mips_get_one_regs_kscratch[i]))) - return -EFAULT; - ++indices; - } - return kvm_mips_callbacks->copy_reg_indices(vcpu, indices); } @@ -794,95 +691,6 @@ static int kvm_mips_get_reg(struct kvm_vcpu *vcpu, v = fpu->msacsr; break; - /* Co-processor 0 registers */ - case KVM_REG_MIPS_CP0_INDEX: - v = (long)kvm_read_c0_guest_index(cop0); - break; - case KVM_REG_MIPS_CP0_CONTEXT: - v = (long)kvm_read_c0_guest_context(cop0); - break; - case KVM_REG_MIPS_CP0_USERLOCAL: - v = (long)kvm_read_c0_guest_userlocal(cop0); - break; - case KVM_REG_MIPS_CP0_PAGEMASK: - v = (long)kvm_read_c0_guest_pagemask(cop0); - break; - case KVM_REG_MIPS_CP0_WIRED: - v = (long)kvm_read_c0_guest_wired(cop0); - break; - case KVM_REG_MIPS_CP0_HWRENA: - v = (long)kvm_read_c0_guest_hwrena(cop0); - break; - case KVM_REG_MIPS_CP0_BADVADDR: - v = (long)kvm_read_c0_guest_badvaddr(cop0); - break; - case KVM_REG_MIPS_CP0_ENTRYHI: - v = (long)kvm_read_c0_guest_entryhi(cop0); - break; - case KVM_REG_MIPS_CP0_COMPARE: - v = (long)kvm_read_c0_guest_compare(cop0); - break; - case KVM_REG_MIPS_CP0_STATUS: - v = (long)kvm_read_c0_guest_status(cop0); - break; - case KVM_REG_MIPS_CP0_CAUSE: - v = (long)kvm_read_c0_guest_cause(cop0); - break; - case KVM_REG_MIPS_CP0_EPC: - v = (long)kvm_read_c0_guest_epc(cop0); - break; - case KVM_REG_MIPS_CP0_PRID: - v = (long)kvm_read_c0_guest_prid(cop0); - break; - case KVM_REG_MIPS_CP0_CONFIG: - v = (long)kvm_read_c0_guest_config(cop0); - break; - case KVM_REG_MIPS_CP0_CONFIG1: - v = (long)kvm_read_c0_guest_config1(cop0); - break; - case KVM_REG_MIPS_CP0_CONFIG2: - v = (long)kvm_read_c0_guest_config2(cop0); - break; - case KVM_REG_MIPS_CP0_CONFIG3: - v = (long)kvm_read_c0_guest_config3(cop0); - break; - case KVM_REG_MIPS_CP0_CONFIG4: - v = (long)kvm_read_c0_guest_config4(cop0); - break; - case KVM_REG_MIPS_CP0_CONFIG5: - v = (long)kvm_read_c0_guest_config5(cop0); - break; - case KVM_REG_MIPS_CP0_CONFIG7: - v = (long)kvm_read_c0_guest_config7(cop0); - break; - case KVM_REG_MIPS_CP0_ERROREPC: - v = (long)kvm_read_c0_guest_errorepc(cop0); - break; - case KVM_REG_MIPS_CP0_KSCRATCH1 ... KVM_REG_MIPS_CP0_KSCRATCH6: - idx = reg->id - KVM_REG_MIPS_CP0_KSCRATCH1 + 2; - if (!(vcpu->arch.kscratch_enabled & BIT(idx))) - return -EINVAL; - switch (idx) { - case 2: - v = (long)kvm_read_c0_guest_kscratch1(cop0); - break; - case 3: - v = (long)kvm_read_c0_guest_kscratch2(cop0); - break; - case 4: - v = (long)kvm_read_c0_guest_kscratch3(cop0); - break; - case 5: - v = (long)kvm_read_c0_guest_kscratch4(cop0); - break; - case 6: - v = (long)kvm_read_c0_guest_kscratch5(cop0); - break; - case 7: - v = (long)kvm_read_c0_guest_kscratch6(cop0); - break; - } - break; /* registers to be handled specially */ default: ret = kvm_mips_callbacks->get_one_reg(vcpu, reg, &v); @@ -1014,68 +822,6 @@ static int kvm_mips_set_reg(struct kvm_vcpu *vcpu, fpu->msacsr = v; break; - /* Co-processor 0 registers */ - case KVM_REG_MIPS_CP0_INDEX: - kvm_write_c0_guest_index(cop0, v); - break; - case KVM_REG_MIPS_CP0_CONTEXT: - kvm_write_c0_guest_context(cop0, v); - break; - case KVM_REG_MIPS_CP0_USERLOCAL: - kvm_write_c0_guest_userlocal(cop0, v); - break; - case KVM_REG_MIPS_CP0_PAGEMASK: - kvm_write_c0_guest_pagemask(cop0, v); - break; - case KVM_REG_MIPS_CP0_WIRED: - kvm_write_c0_guest_wired(cop0, v); - break; - case KVM_REG_MIPS_CP0_HWRENA: - kvm_write_c0_guest_hwrena(cop0, v); - break; - case KVM_REG_MIPS_CP0_BADVADDR: - kvm_write_c0_guest_badvaddr(cop0, v); - break; - case KVM_REG_MIPS_CP0_ENTRYHI: - kvm_write_c0_guest_entryhi(cop0, v); - break; - case KVM_REG_MIPS_CP0_STATUS: - kvm_write_c0_guest_status(cop0, v); - break; - case KVM_REG_MIPS_CP0_EPC: - kvm_write_c0_guest_epc(cop0, v); - break; - case KVM_REG_MIPS_CP0_PRID: - kvm_write_c0_guest_prid(cop0, v); - break; - case KVM_REG_MIPS_CP0_ERROREPC: - kvm_write_c0_guest_errorepc(cop0, v); - break; - case KVM_REG_MIPS_CP0_KSCRATCH1 ... KVM_REG_MIPS_CP0_KSCRATCH6: - idx = reg->id - KVM_REG_MIPS_CP0_KSCRATCH1 + 2; - if (!(vcpu->arch.kscratch_enabled & BIT(idx))) - return -EINVAL; - switch (idx) { - case 2: - kvm_write_c0_guest_kscratch1(cop0, v); - break; - case 3: - kvm_write_c0_guest_kscratch2(cop0, v); - break; - case 4: - kvm_write_c0_guest_kscratch3(cop0, v); - break; - case 5: - kvm_write_c0_guest_kscratch4(cop0, v); - break; - case 6: - kvm_write_c0_guest_kscratch5(cop0, v); - break; - case 7: - kvm_write_c0_guest_kscratch6(cop0, v); - break; - } - break; /* registers to be handled specially */ default: return kvm_mips_callbacks->set_one_reg(vcpu, reg, v); @@ -1144,18 +890,12 @@ long kvm_arch_vcpu_ioctl(struct file *filp, unsigned int ioctl, return -E2BIG; return kvm_mips_copy_reg_indices(vcpu, user_list->reg); } - case KVM_NMI: - /* Treat the NMI as a CPU reset */ - r = kvm_mips_reset_vcpu(vcpu); - break; case KVM_INTERRUPT: { struct kvm_mips_interrupt irq; - r = -EFAULT; if (copy_from_user(&irq, argp, sizeof(irq))) - goto out; - + return -EFAULT; kvm_debug("[%d] %s: irq: %d\n", vcpu->vcpu_id, __func__, irq.irq); @@ -1165,56 +905,57 @@ long kvm_arch_vcpu_ioctl(struct file *filp, unsigned int ioctl, case KVM_ENABLE_CAP: { struct kvm_enable_cap cap; - r = -EFAULT; if (copy_from_user(&cap, argp, sizeof(cap))) - goto out; + return -EFAULT; r = kvm_vcpu_ioctl_enable_cap(vcpu, &cap); break; } default: r = -ENOIOCTLCMD; } - -out: return r; } -/* Get (and clear) the dirty memory log for a memory slot. */ +/** + * kvm_vm_ioctl_get_dirty_log - get and clear the log of dirty pages in a slot + * @kvm: kvm instance + * @log: slot id and address to which we copy the log + * + * Steps 1-4 below provide general overview of dirty page logging. See + * kvm_get_dirty_log_protect() function description for additional details. + * + * We call kvm_get_dirty_log_protect() to handle steps 1-3, upon return we + * always flush the TLB (step 4) even if previous step failed and the dirty + * bitmap may be corrupt. Regardless of previous outcome the KVM logging API + * does not preclude user space subsequent dirty log read. Flushing TLB ensures + * writes will be marked dirty for next log read. + * + * 1. Take a snapshot of the bit and clear it if needed. + * 2. Write protect the corresponding page. + * 3. Copy the snapshot to the userspace. + * 4. Flush TLB's if needed. + */ int kvm_vm_ioctl_get_dirty_log(struct kvm *kvm, struct kvm_dirty_log *log) { struct kvm_memslots *slots; struct kvm_memory_slot *memslot; - unsigned long ga, ga_end; - int is_dirty = 0; + bool is_dirty = false; int r; - unsigned long n; mutex_lock(&kvm->slots_lock); - r = kvm_get_dirty_log(kvm, log, &is_dirty); - if (r) - goto out; + r = kvm_get_dirty_log_protect(kvm, log, &is_dirty); - /* If nothing is dirty, don't bother messing with page tables. */ if (is_dirty) { slots = kvm_memslots(kvm); memslot = id_to_memslot(slots, log->slot); - ga = memslot->base_gfn << PAGE_SHIFT; - ga_end = ga + (memslot->npages << PAGE_SHIFT); - - kvm_info("%s: dirty, ga: %#lx, ga_end %#lx\n", __func__, ga, - ga_end); - - n = kvm_dirty_bitmap_bytes(memslot); - memset(memslot->dirty_bitmap, 0, n); + /* Let implementation handle TLB/GVA invalidation */ + kvm_mips_callbacks->flush_shadow_memslot(kvm, memslot); } - r = 0; -out: mutex_unlock(&kvm->slots_lock); return r; - } long kvm_arch_vm_ioctl(struct file *filp, unsigned int ioctl, unsigned long arg) @@ -1282,11 +1023,20 @@ int kvm_vm_ioctl_check_extension(struct kvm *kvm, long ext) switch (ext) { case KVM_CAP_ONE_REG: case KVM_CAP_ENABLE_CAP: + case KVM_CAP_READONLY_MEM: + case KVM_CAP_SYNC_MMU: + case KVM_CAP_IMMEDIATE_EXIT: r = 1; break; case KVM_CAP_COALESCED_MMIO: r = KVM_COALESCED_MMIO_PAGE_OFFSET; break; + case KVM_CAP_NR_VCPUS: + r = num_online_cpus(); + break; + case KVM_CAP_MAX_VCPUS: + r = KVM_MAX_VCPUS; + break; case KVM_CAP_MIPS_FPU: /* We don't handle systems with inconsistent cpu_has_fpu */ r = !!raw_cpu_has_fpu; @@ -1400,13 +1150,23 @@ static enum hrtimer_restart kvm_mips_comparecount_wakeup(struct hrtimer *timer) int kvm_arch_vcpu_init(struct kvm_vcpu *vcpu) { - kvm_mips_callbacks->vcpu_init(vcpu); + int err; + + err = kvm_mips_callbacks->vcpu_init(vcpu); + if (err) + return err; + hrtimer_init(&vcpu->arch.comparecount_timer, CLOCK_MONOTONIC, HRTIMER_MODE_REL); vcpu->arch.comparecount_timer.function = kvm_mips_comparecount_wakeup; return 0; } +void kvm_arch_vcpu_uninit(struct kvm_vcpu *vcpu) +{ + kvm_mips_callbacks->vcpu_uninit(vcpu); +} + int kvm_arch_vcpu_ioctl_translate(struct kvm_vcpu *vcpu, struct kvm_translation *tr) { @@ -1440,8 +1200,11 @@ int kvm_mips_handle_exit(struct kvm_run *run, struct kvm_vcpu *vcpu) u32 __user *opc = (u32 __user *) vcpu->arch.pc; unsigned long badvaddr = vcpu->arch.host_cp0_badvaddr; enum emulation_result er = EMULATE_DONE; + u32 inst; int ret = RESUME_GUEST; + vcpu->mode = OUTSIDE_GUEST_MODE; + /* re-enable HTW before enabling interrupts */ htw_start(); @@ -1564,8 +1327,12 @@ int kvm_mips_handle_exit(struct kvm_run *run, struct kvm_vcpu *vcpu) break; default: + if (cause & CAUSEF_BD) + opc += 1; + inst = 0; + kvm_get_badinstr(opc, vcpu, &inst); kvm_err("Exception Code: %d, not yet handled, @ PC: %p, inst: 0x%08x BadVaddr: %#lx Status: %#lx\n", - exccode, opc, kvm_get_inst(opc, vcpu), badvaddr, + exccode, opc, inst, badvaddr, kvm_read_c0_guest_status(vcpu->arch.cop0)); kvm_arch_vcpu_dump_regs(vcpu); run->exit_reason = KVM_EXIT_INTERNAL_ERROR; @@ -1593,7 +1360,15 @@ skip_emul: if (ret == RESUME_GUEST) { trace_kvm_reenter(vcpu); - kvm_mips_check_asids(vcpu); + /* + * Make sure the read of VCPU requests in vcpu_reenter() + * callback is not reordered ahead of the write to vcpu->mode, + * or we could miss a TLB flush request while the requester sees + * the VCPU as outside of guest mode and not needing an IPI. + */ + smp_store_mb(vcpu->mode, IN_GUEST_MODE); + + kvm_mips_callbacks->vcpu_reenter(run, vcpu); /* * If FPU / MSA are enabled (i.e. the guest's FPU / MSA context diff --git a/arch/mips/kvm/mmu.c b/arch/mips/kvm/mmu.c index 3b677c851be0..cb0faade311e 100644 --- a/arch/mips/kvm/mmu.c +++ b/arch/mips/kvm/mmu.c @@ -11,86 +11,995 @@ #include <linux/highmem.h> #include <linux/kvm_host.h> +#include <linux/uaccess.h> #include <asm/mmu_context.h> +#include <asm/pgalloc.h> -static u32 kvm_mips_get_kernel_asid(struct kvm_vcpu *vcpu) +/* + * KVM_MMU_CACHE_MIN_PAGES is the number of GPA page table translation levels + * for which pages need to be cached. + */ +#if defined(__PAGETABLE_PMD_FOLDED) +#define KVM_MMU_CACHE_MIN_PAGES 1 +#else +#define KVM_MMU_CACHE_MIN_PAGES 2 +#endif + +static int mmu_topup_memory_cache(struct kvm_mmu_memory_cache *cache, + int min, int max) { - int cpu = smp_processor_id(); + void *page; + + BUG_ON(max > KVM_NR_MEM_OBJS); + if (cache->nobjs >= min) + return 0; + while (cache->nobjs < max) { + page = (void *)__get_free_page(GFP_KERNEL); + if (!page) + return -ENOMEM; + cache->objects[cache->nobjs++] = page; + } + return 0; +} - return vcpu->arch.guest_kernel_asid[cpu] & - cpu_asid_mask(&cpu_data[cpu]); +static void mmu_free_memory_cache(struct kvm_mmu_memory_cache *mc) +{ + while (mc->nobjs) + free_page((unsigned long)mc->objects[--mc->nobjs]); } -static u32 kvm_mips_get_user_asid(struct kvm_vcpu *vcpu) +static void *mmu_memory_cache_alloc(struct kvm_mmu_memory_cache *mc) { - int cpu = smp_processor_id(); + void *p; - return vcpu->arch.guest_user_asid[cpu] & - cpu_asid_mask(&cpu_data[cpu]); + BUG_ON(!mc || !mc->nobjs); + p = mc->objects[--mc->nobjs]; + return p; } -static int kvm_mips_map_page(struct kvm *kvm, gfn_t gfn) +void kvm_mmu_free_memory_caches(struct kvm_vcpu *vcpu) { - int srcu_idx, err = 0; - kvm_pfn_t pfn; + mmu_free_memory_cache(&vcpu->arch.mmu_page_cache); +} + +/** + * kvm_pgd_init() - Initialise KVM GPA page directory. + * @page: Pointer to page directory (PGD) for KVM GPA. + * + * Initialise a KVM GPA page directory with pointers to the invalid table, i.e. + * representing no mappings. This is similar to pgd_init(), however it + * initialises all the page directory pointers, not just the ones corresponding + * to the userland address space (since it is for the guest physical address + * space rather than a virtual address space). + */ +static void kvm_pgd_init(void *page) +{ + unsigned long *p, *end; + unsigned long entry; + +#ifdef __PAGETABLE_PMD_FOLDED + entry = (unsigned long)invalid_pte_table; +#else + entry = (unsigned long)invalid_pmd_table; +#endif + + p = (unsigned long *)page; + end = p + PTRS_PER_PGD; + + do { + p[0] = entry; + p[1] = entry; + p[2] = entry; + p[3] = entry; + p[4] = entry; + p += 8; + p[-3] = entry; + p[-2] = entry; + p[-1] = entry; + } while (p != end); +} + +/** + * kvm_pgd_alloc() - Allocate and initialise a KVM GPA page directory. + * + * Allocate a blank KVM GPA page directory (PGD) for representing guest physical + * to host physical page mappings. + * + * Returns: Pointer to new KVM GPA page directory. + * NULL on allocation failure. + */ +pgd_t *kvm_pgd_alloc(void) +{ + pgd_t *ret; + + ret = (pgd_t *)__get_free_pages(GFP_KERNEL, PGD_ORDER); + if (ret) + kvm_pgd_init(ret); + + return ret; +} + +/** + * kvm_mips_walk_pgd() - Walk page table with optional allocation. + * @pgd: Page directory pointer. + * @addr: Address to index page table using. + * @cache: MMU page cache to allocate new page tables from, or NULL. + * + * Walk the page tables pointed to by @pgd to find the PTE corresponding to the + * address @addr. If page tables don't exist for @addr, they will be created + * from the MMU cache if @cache is not NULL. + * + * Returns: Pointer to pte_t corresponding to @addr. + * NULL if a page table doesn't exist for @addr and !@cache. + * NULL if a page table allocation failed. + */ +static pte_t *kvm_mips_walk_pgd(pgd_t *pgd, struct kvm_mmu_memory_cache *cache, + unsigned long addr) +{ + pud_t *pud; + pmd_t *pmd; + + pgd += pgd_index(addr); + if (pgd_none(*pgd)) { + /* Not used on MIPS yet */ + BUG(); + return NULL; + } + pud = pud_offset(pgd, addr); + if (pud_none(*pud)) { + pmd_t *new_pmd; + + if (!cache) + return NULL; + new_pmd = mmu_memory_cache_alloc(cache); + pmd_init((unsigned long)new_pmd, + (unsigned long)invalid_pte_table); + pud_populate(NULL, pud, new_pmd); + } + pmd = pmd_offset(pud, addr); + if (pmd_none(*pmd)) { + pte_t *new_pte; + + if (!cache) + return NULL; + new_pte = mmu_memory_cache_alloc(cache); + clear_page(new_pte); + pmd_populate_kernel(NULL, pmd, new_pte); + } + return pte_offset(pmd, addr); +} + +/* Caller must hold kvm->mm_lock */ +static pte_t *kvm_mips_pte_for_gpa(struct kvm *kvm, + struct kvm_mmu_memory_cache *cache, + unsigned long addr) +{ + return kvm_mips_walk_pgd(kvm->arch.gpa_mm.pgd, cache, addr); +} + +/* + * kvm_mips_flush_gpa_{pte,pmd,pud,pgd,pt}. + * Flush a range of guest physical address space from the VM's GPA page tables. + */ + +static bool kvm_mips_flush_gpa_pte(pte_t *pte, unsigned long start_gpa, + unsigned long end_gpa) +{ + int i_min = __pte_offset(start_gpa); + int i_max = __pte_offset(end_gpa); + bool safe_to_remove = (i_min == 0 && i_max == PTRS_PER_PTE - 1); + int i; + + for (i = i_min; i <= i_max; ++i) { + if (!pte_present(pte[i])) + continue; + + set_pte(pte + i, __pte(0)); + } + return safe_to_remove; +} + +static bool kvm_mips_flush_gpa_pmd(pmd_t *pmd, unsigned long start_gpa, + unsigned long end_gpa) +{ + pte_t *pte; + unsigned long end = ~0ul; + int i_min = __pmd_offset(start_gpa); + int i_max = __pmd_offset(end_gpa); + bool safe_to_remove = (i_min == 0 && i_max == PTRS_PER_PMD - 1); + int i; + + for (i = i_min; i <= i_max; ++i, start_gpa = 0) { + if (!pmd_present(pmd[i])) + continue; + + pte = pte_offset(pmd + i, 0); + if (i == i_max) + end = end_gpa; + + if (kvm_mips_flush_gpa_pte(pte, start_gpa, end)) { + pmd_clear(pmd + i); + pte_free_kernel(NULL, pte); + } else { + safe_to_remove = false; + } + } + return safe_to_remove; +} + +static bool kvm_mips_flush_gpa_pud(pud_t *pud, unsigned long start_gpa, + unsigned long end_gpa) +{ + pmd_t *pmd; + unsigned long end = ~0ul; + int i_min = __pud_offset(start_gpa); + int i_max = __pud_offset(end_gpa); + bool safe_to_remove = (i_min == 0 && i_max == PTRS_PER_PUD - 1); + int i; + + for (i = i_min; i <= i_max; ++i, start_gpa = 0) { + if (!pud_present(pud[i])) + continue; + + pmd = pmd_offset(pud + i, 0); + if (i == i_max) + end = end_gpa; + + if (kvm_mips_flush_gpa_pmd(pmd, start_gpa, end)) { + pud_clear(pud + i); + pmd_free(NULL, pmd); + } else { + safe_to_remove = false; + } + } + return safe_to_remove; +} + +static bool kvm_mips_flush_gpa_pgd(pgd_t *pgd, unsigned long start_gpa, + unsigned long end_gpa) +{ + pud_t *pud; + unsigned long end = ~0ul; + int i_min = pgd_index(start_gpa); + int i_max = pgd_index(end_gpa); + bool safe_to_remove = (i_min == 0 && i_max == PTRS_PER_PGD - 1); + int i; + + for (i = i_min; i <= i_max; ++i, start_gpa = 0) { + if (!pgd_present(pgd[i])) + continue; + + pud = pud_offset(pgd + i, 0); + if (i == i_max) + end = end_gpa; + + if (kvm_mips_flush_gpa_pud(pud, start_gpa, end)) { + pgd_clear(pgd + i); + pud_free(NULL, pud); + } else { + safe_to_remove = false; + } + } + return safe_to_remove; +} + +/** + * kvm_mips_flush_gpa_pt() - Flush a range of guest physical addresses. + * @kvm: KVM pointer. + * @start_gfn: Guest frame number of first page in GPA range to flush. + * @end_gfn: Guest frame number of last page in GPA range to flush. + * + * Flushes a range of GPA mappings from the GPA page tables. + * + * The caller must hold the @kvm->mmu_lock spinlock. + * + * Returns: Whether its safe to remove the top level page directory because + * all lower levels have been removed. + */ +bool kvm_mips_flush_gpa_pt(struct kvm *kvm, gfn_t start_gfn, gfn_t end_gfn) +{ + return kvm_mips_flush_gpa_pgd(kvm->arch.gpa_mm.pgd, + start_gfn << PAGE_SHIFT, + end_gfn << PAGE_SHIFT); +} + +#define BUILD_PTE_RANGE_OP(name, op) \ +static int kvm_mips_##name##_pte(pte_t *pte, unsigned long start, \ + unsigned long end) \ +{ \ + int ret = 0; \ + int i_min = __pte_offset(start); \ + int i_max = __pte_offset(end); \ + int i; \ + pte_t old, new; \ + \ + for (i = i_min; i <= i_max; ++i) { \ + if (!pte_present(pte[i])) \ + continue; \ + \ + old = pte[i]; \ + new = op(old); \ + if (pte_val(new) == pte_val(old)) \ + continue; \ + set_pte(pte + i, new); \ + ret = 1; \ + } \ + return ret; \ +} \ + \ +/* returns true if anything was done */ \ +static int kvm_mips_##name##_pmd(pmd_t *pmd, unsigned long start, \ + unsigned long end) \ +{ \ + int ret = 0; \ + pte_t *pte; \ + unsigned long cur_end = ~0ul; \ + int i_min = __pmd_offset(start); \ + int i_max = __pmd_offset(end); \ + int i; \ + \ + for (i = i_min; i <= i_max; ++i, start = 0) { \ + if (!pmd_present(pmd[i])) \ + continue; \ + \ + pte = pte_offset(pmd + i, 0); \ + if (i == i_max) \ + cur_end = end; \ + \ + ret |= kvm_mips_##name##_pte(pte, start, cur_end); \ + } \ + return ret; \ +} \ + \ +static int kvm_mips_##name##_pud(pud_t *pud, unsigned long start, \ + unsigned long end) \ +{ \ + int ret = 0; \ + pmd_t *pmd; \ + unsigned long cur_end = ~0ul; \ + int i_min = __pud_offset(start); \ + int i_max = __pud_offset(end); \ + int i; \ + \ + for (i = i_min; i <= i_max; ++i, start = 0) { \ + if (!pud_present(pud[i])) \ + continue; \ + \ + pmd = pmd_offset(pud + i, 0); \ + if (i == i_max) \ + cur_end = end; \ + \ + ret |= kvm_mips_##name##_pmd(pmd, start, cur_end); \ + } \ + return ret; \ +} \ + \ +static int kvm_mips_##name##_pgd(pgd_t *pgd, unsigned long start, \ + unsigned long end) \ +{ \ + int ret = 0; \ + pud_t *pud; \ + unsigned long cur_end = ~0ul; \ + int i_min = pgd_index(start); \ + int i_max = pgd_index(end); \ + int i; \ + \ + for (i = i_min; i <= i_max; ++i, start = 0) { \ + if (!pgd_present(pgd[i])) \ + continue; \ + \ + pud = pud_offset(pgd + i, 0); \ + if (i == i_max) \ + cur_end = end; \ + \ + ret |= kvm_mips_##name##_pud(pud, start, cur_end); \ + } \ + return ret; \ +} + +/* + * kvm_mips_mkclean_gpa_pt. + * Mark a range of guest physical address space clean (writes fault) in the VM's + * GPA page table to allow dirty page tracking. + */ - if (kvm->arch.guest_pmap[gfn] != KVM_INVALID_PAGE) +BUILD_PTE_RANGE_OP(mkclean, pte_mkclean) + +/** + * kvm_mips_mkclean_gpa_pt() - Make a range of guest physical addresses clean. + * @kvm: KVM pointer. + * @start_gfn: Guest frame number of first page in GPA range to flush. + * @end_gfn: Guest frame number of last page in GPA range to flush. + * + * Make a range of GPA mappings clean so that guest writes will fault and + * trigger dirty page logging. + * + * The caller must hold the @kvm->mmu_lock spinlock. + * + * Returns: Whether any GPA mappings were modified, which would require + * derived mappings (GVA page tables & TLB enties) to be + * invalidated. + */ +int kvm_mips_mkclean_gpa_pt(struct kvm *kvm, gfn_t start_gfn, gfn_t end_gfn) +{ + return kvm_mips_mkclean_pgd(kvm->arch.gpa_mm.pgd, + start_gfn << PAGE_SHIFT, + end_gfn << PAGE_SHIFT); +} + +/** + * kvm_arch_mmu_enable_log_dirty_pt_masked() - write protect dirty pages + * @kvm: The KVM pointer + * @slot: The memory slot associated with mask + * @gfn_offset: The gfn offset in memory slot + * @mask: The mask of dirty pages at offset 'gfn_offset' in this memory + * slot to be write protected + * + * Walks bits set in mask write protects the associated pte's. Caller must + * acquire @kvm->mmu_lock. + */ +void kvm_arch_mmu_enable_log_dirty_pt_masked(struct kvm *kvm, + struct kvm_memory_slot *slot, + gfn_t gfn_offset, unsigned long mask) +{ + gfn_t base_gfn = slot->base_gfn + gfn_offset; + gfn_t start = base_gfn + __ffs(mask); + gfn_t end = base_gfn + __fls(mask); + + kvm_mips_mkclean_gpa_pt(kvm, start, end); +} + +/* + * kvm_mips_mkold_gpa_pt. + * Mark a range of guest physical address space old (all accesses fault) in the + * VM's GPA page table to allow detection of commonly used pages. + */ + +BUILD_PTE_RANGE_OP(mkold, pte_mkold) + +static int kvm_mips_mkold_gpa_pt(struct kvm *kvm, gfn_t start_gfn, + gfn_t end_gfn) +{ + return kvm_mips_mkold_pgd(kvm->arch.gpa_mm.pgd, + start_gfn << PAGE_SHIFT, + end_gfn << PAGE_SHIFT); +} + +static int handle_hva_to_gpa(struct kvm *kvm, + unsigned long start, + unsigned long end, + int (*handler)(struct kvm *kvm, gfn_t gfn, + gpa_t gfn_end, + struct kvm_memory_slot *memslot, + void *data), + void *data) +{ + struct kvm_memslots *slots; + struct kvm_memory_slot *memslot; + int ret = 0; + + slots = kvm_memslots(kvm); + + /* we only care about the pages that the guest sees */ + kvm_for_each_memslot(memslot, slots) { + unsigned long hva_start, hva_end; + gfn_t gfn, gfn_end; + + hva_start = max(start, memslot->userspace_addr); + hva_end = min(end, memslot->userspace_addr + + (memslot->npages << PAGE_SHIFT)); + if (hva_start >= hva_end) + continue; + + /* + * {gfn(page) | page intersects with [hva_start, hva_end)} = + * {gfn_start, gfn_start+1, ..., gfn_end-1}. + */ + gfn = hva_to_gfn_memslot(hva_start, memslot); + gfn_end = hva_to_gfn_memslot(hva_end + PAGE_SIZE - 1, memslot); + + ret |= handler(kvm, gfn, gfn_end, memslot, data); + } + + return ret; +} + + +static int kvm_unmap_hva_handler(struct kvm *kvm, gfn_t gfn, gfn_t gfn_end, + struct kvm_memory_slot *memslot, void *data) +{ + kvm_mips_flush_gpa_pt(kvm, gfn, gfn_end); + return 1; +} + +int kvm_unmap_hva(struct kvm *kvm, unsigned long hva) +{ + unsigned long end = hva + PAGE_SIZE; + + handle_hva_to_gpa(kvm, hva, end, &kvm_unmap_hva_handler, NULL); + + kvm_mips_callbacks->flush_shadow_all(kvm); + return 0; +} + +int kvm_unmap_hva_range(struct kvm *kvm, unsigned long start, unsigned long end) +{ + handle_hva_to_gpa(kvm, start, end, &kvm_unmap_hva_handler, NULL); + + kvm_mips_callbacks->flush_shadow_all(kvm); + return 0; +} + +static int kvm_set_spte_handler(struct kvm *kvm, gfn_t gfn, gfn_t gfn_end, + struct kvm_memory_slot *memslot, void *data) +{ + gpa_t gpa = gfn << PAGE_SHIFT; + pte_t hva_pte = *(pte_t *)data; + pte_t *gpa_pte = kvm_mips_pte_for_gpa(kvm, NULL, gpa); + pte_t old_pte; + + if (!gpa_pte) + return 0; + + /* Mapping may need adjusting depending on memslot flags */ + old_pte = *gpa_pte; + if (memslot->flags & KVM_MEM_LOG_DIRTY_PAGES && !pte_dirty(old_pte)) + hva_pte = pte_mkclean(hva_pte); + else if (memslot->flags & KVM_MEM_READONLY) + hva_pte = pte_wrprotect(hva_pte); + + set_pte(gpa_pte, hva_pte); + + /* Replacing an absent or old page doesn't need flushes */ + if (!pte_present(old_pte) || !pte_young(old_pte)) return 0; + /* Pages swapped, aged, moved, or cleaned require flushes */ + return !pte_present(hva_pte) || + !pte_young(hva_pte) || + pte_pfn(old_pte) != pte_pfn(hva_pte) || + (pte_dirty(old_pte) && !pte_dirty(hva_pte)); +} + +void kvm_set_spte_hva(struct kvm *kvm, unsigned long hva, pte_t pte) +{ + unsigned long end = hva + PAGE_SIZE; + int ret; + + ret = handle_hva_to_gpa(kvm, hva, end, &kvm_set_spte_handler, &pte); + if (ret) + kvm_mips_callbacks->flush_shadow_all(kvm); +} + +static int kvm_age_hva_handler(struct kvm *kvm, gfn_t gfn, gfn_t gfn_end, + struct kvm_memory_slot *memslot, void *data) +{ + return kvm_mips_mkold_gpa_pt(kvm, gfn, gfn_end); +} + +static int kvm_test_age_hva_handler(struct kvm *kvm, gfn_t gfn, gfn_t gfn_end, + struct kvm_memory_slot *memslot, void *data) +{ + gpa_t gpa = gfn << PAGE_SHIFT; + pte_t *gpa_pte = kvm_mips_pte_for_gpa(kvm, NULL, gpa); + + if (!gpa_pte) + return 0; + return pte_young(*gpa_pte); +} + +int kvm_age_hva(struct kvm *kvm, unsigned long start, unsigned long end) +{ + return handle_hva_to_gpa(kvm, start, end, kvm_age_hva_handler, NULL); +} + +int kvm_test_age_hva(struct kvm *kvm, unsigned long hva) +{ + return handle_hva_to_gpa(kvm, hva, hva, kvm_test_age_hva_handler, NULL); +} + +/** + * _kvm_mips_map_page_fast() - Fast path GPA fault handler. + * @vcpu: VCPU pointer. + * @gpa: Guest physical address of fault. + * @write_fault: Whether the fault was due to a write. + * @out_entry: New PTE for @gpa (written on success unless NULL). + * @out_buddy: New PTE for @gpa's buddy (written on success unless + * NULL). + * + * Perform fast path GPA fault handling, doing all that can be done without + * calling into KVM. This handles marking old pages young (for idle page + * tracking), and dirtying of clean pages (for dirty page logging). + * + * Returns: 0 on success, in which case we can update derived mappings and + * resume guest execution. + * -EFAULT on failure due to absent GPA mapping or write to + * read-only page, in which case KVM must be consulted. + */ +static int _kvm_mips_map_page_fast(struct kvm_vcpu *vcpu, unsigned long gpa, + bool write_fault, + pte_t *out_entry, pte_t *out_buddy) +{ + struct kvm *kvm = vcpu->kvm; + gfn_t gfn = gpa >> PAGE_SHIFT; + pte_t *ptep; + kvm_pfn_t pfn = 0; /* silence bogus GCC warning */ + bool pfn_valid = false; + int ret = 0; + + spin_lock(&kvm->mmu_lock); + + /* Fast path - just check GPA page table for an existing entry */ + ptep = kvm_mips_pte_for_gpa(kvm, NULL, gpa); + if (!ptep || !pte_present(*ptep)) { + ret = -EFAULT; + goto out; + } + + /* Track access to pages marked old */ + if (!pte_young(*ptep)) { + set_pte(ptep, pte_mkyoung(*ptep)); + pfn = pte_pfn(*ptep); + pfn_valid = true; + /* call kvm_set_pfn_accessed() after unlock */ + } + if (write_fault && !pte_dirty(*ptep)) { + if (!pte_write(*ptep)) { + ret = -EFAULT; + goto out; + } + + /* Track dirtying of writeable pages */ + set_pte(ptep, pte_mkdirty(*ptep)); + pfn = pte_pfn(*ptep); + mark_page_dirty(kvm, gfn); + kvm_set_pfn_dirty(pfn); + } + + if (out_entry) + *out_entry = *ptep; + if (out_buddy) + *out_buddy = *ptep_buddy(ptep); + +out: + spin_unlock(&kvm->mmu_lock); + if (pfn_valid) + kvm_set_pfn_accessed(pfn); + return ret; +} + +/** + * kvm_mips_map_page() - Map a guest physical page. + * @vcpu: VCPU pointer. + * @gpa: Guest physical address of fault. + * @write_fault: Whether the fault was due to a write. + * @out_entry: New PTE for @gpa (written on success unless NULL). + * @out_buddy: New PTE for @gpa's buddy (written on success unless + * NULL). + * + * Handle GPA faults by creating a new GPA mapping (or updating an existing + * one). + * + * This takes care of marking pages young or dirty (idle/dirty page tracking), + * asking KVM for the corresponding PFN, and creating a mapping in the GPA page + * tables. Derived mappings (GVA page tables and TLBs) must be handled by the + * caller. + * + * Returns: 0 on success, in which case the caller may use the @out_entry + * and @out_buddy PTEs to update derived mappings and resume guest + * execution. + * -EFAULT if there is no memory region at @gpa or a write was + * attempted to a read-only memory region. This is usually handled + * as an MMIO access. + */ +static int kvm_mips_map_page(struct kvm_vcpu *vcpu, unsigned long gpa, + bool write_fault, + pte_t *out_entry, pte_t *out_buddy) +{ + struct kvm *kvm = vcpu->kvm; + struct kvm_mmu_memory_cache *memcache = &vcpu->arch.mmu_page_cache; + gfn_t gfn = gpa >> PAGE_SHIFT; + int srcu_idx, err; + kvm_pfn_t pfn; + pte_t *ptep, entry, old_pte; + bool writeable; + unsigned long prot_bits; + unsigned long mmu_seq; + + /* Try the fast path to handle old / clean pages */ srcu_idx = srcu_read_lock(&kvm->srcu); - pfn = gfn_to_pfn(kvm, gfn); + err = _kvm_mips_map_page_fast(vcpu, gpa, write_fault, out_entry, + out_buddy); + if (!err) + goto out; + /* We need a minimum of cached pages ready for page table creation */ + err = mmu_topup_memory_cache(memcache, KVM_MMU_CACHE_MIN_PAGES, + KVM_NR_MEM_OBJS); + if (err) + goto out; + +retry: + /* + * Used to check for invalidations in progress, of the pfn that is + * returned by pfn_to_pfn_prot below. + */ + mmu_seq = kvm->mmu_notifier_seq; + /* + * Ensure the read of mmu_notifier_seq isn't reordered with PTE reads in + * gfn_to_pfn_prot() (which calls get_user_pages()), so that we don't + * risk the page we get a reference to getting unmapped before we have a + * chance to grab the mmu_lock without mmu_notifier_retry() noticing. + * + * This smp_rmb() pairs with the effective smp_wmb() of the combination + * of the pte_unmap_unlock() after the PTE is zapped, and the + * spin_lock() in kvm_mmu_notifier_invalidate_<page|range_end>() before + * mmu_notifier_seq is incremented. + */ + smp_rmb(); + + /* Slow path - ask KVM core whether we can access this GPA */ + pfn = gfn_to_pfn_prot(kvm, gfn, write_fault, &writeable); if (is_error_noslot_pfn(pfn)) { - kvm_err("Couldn't get pfn for gfn %#llx!\n", gfn); err = -EFAULT; goto out; } - kvm->arch.guest_pmap[gfn] = pfn; + spin_lock(&kvm->mmu_lock); + /* Check if an invalidation has taken place since we got pfn */ + if (mmu_notifier_retry(kvm, mmu_seq)) { + /* + * This can happen when mappings are changed asynchronously, but + * also synchronously if a COW is triggered by + * gfn_to_pfn_prot(). + */ + spin_unlock(&kvm->mmu_lock); + kvm_release_pfn_clean(pfn); + goto retry; + } + + /* Ensure page tables are allocated */ + ptep = kvm_mips_pte_for_gpa(kvm, memcache, gpa); + + /* Set up the PTE */ + prot_bits = _PAGE_PRESENT | __READABLE | _page_cachable_default; + if (writeable) { + prot_bits |= _PAGE_WRITE; + if (write_fault) { + prot_bits |= __WRITEABLE; + mark_page_dirty(kvm, gfn); + kvm_set_pfn_dirty(pfn); + } + } + entry = pfn_pte(pfn, __pgprot(prot_bits)); + + /* Write the PTE */ + old_pte = *ptep; + set_pte(ptep, entry); + + err = 0; + if (out_entry) + *out_entry = *ptep; + if (out_buddy) + *out_buddy = *ptep_buddy(ptep); + + spin_unlock(&kvm->mmu_lock); + kvm_release_pfn_clean(pfn); + kvm_set_pfn_accessed(pfn); out: srcu_read_unlock(&kvm->srcu, srcu_idx); return err; } -/* Translate guest KSEG0 addresses to Host PA */ -unsigned long kvm_mips_translate_guest_kseg0_to_hpa(struct kvm_vcpu *vcpu, - unsigned long gva) +static pte_t *kvm_trap_emul_pte_for_gva(struct kvm_vcpu *vcpu, + unsigned long addr) { - gfn_t gfn; - unsigned long offset = gva & ~PAGE_MASK; - struct kvm *kvm = vcpu->kvm; + struct kvm_mmu_memory_cache *memcache = &vcpu->arch.mmu_page_cache; + pgd_t *pgdp; + int ret; + + /* We need a minimum of cached pages ready for page table creation */ + ret = mmu_topup_memory_cache(memcache, KVM_MMU_CACHE_MIN_PAGES, + KVM_NR_MEM_OBJS); + if (ret) + return NULL; + + if (KVM_GUEST_KERNEL_MODE(vcpu)) + pgdp = vcpu->arch.guest_kernel_mm.pgd; + else + pgdp = vcpu->arch.guest_user_mm.pgd; + + return kvm_mips_walk_pgd(pgdp, memcache, addr); +} - if (KVM_GUEST_KSEGX(gva) != KVM_GUEST_KSEG0) { - kvm_err("%s/%p: Invalid gva: %#lx\n", __func__, - __builtin_return_address(0), gva); - return KVM_INVALID_PAGE; +void kvm_trap_emul_invalidate_gva(struct kvm_vcpu *vcpu, unsigned long addr, + bool user) +{ + pgd_t *pgdp; + pte_t *ptep; + + addr &= PAGE_MASK << 1; + + pgdp = vcpu->arch.guest_kernel_mm.pgd; + ptep = kvm_mips_walk_pgd(pgdp, NULL, addr); + if (ptep) { + ptep[0] = pfn_pte(0, __pgprot(0)); + ptep[1] = pfn_pte(0, __pgprot(0)); + } + + if (user) { + pgdp = vcpu->arch.guest_user_mm.pgd; + ptep = kvm_mips_walk_pgd(pgdp, NULL, addr); + if (ptep) { + ptep[0] = pfn_pte(0, __pgprot(0)); + ptep[1] = pfn_pte(0, __pgprot(0)); + } } +} - gfn = (KVM_GUEST_CPHYSADDR(gva) >> PAGE_SHIFT); +/* + * kvm_mips_flush_gva_{pte,pmd,pud,pgd,pt}. + * Flush a range of guest physical address space from the VM's GPA page tables. + */ - if (gfn >= kvm->arch.guest_pmap_npages) { - kvm_err("%s: Invalid gfn: %#llx, GVA: %#lx\n", __func__, gfn, - gva); - return KVM_INVALID_PAGE; +static bool kvm_mips_flush_gva_pte(pte_t *pte, unsigned long start_gva, + unsigned long end_gva) +{ + int i_min = __pte_offset(start_gva); + int i_max = __pte_offset(end_gva); + bool safe_to_remove = (i_min == 0 && i_max == PTRS_PER_PTE - 1); + int i; + + /* + * There's no freeing to do, so there's no point clearing individual + * entries unless only part of the last level page table needs flushing. + */ + if (safe_to_remove) + return true; + + for (i = i_min; i <= i_max; ++i) { + if (!pte_present(pte[i])) + continue; + + set_pte(pte + i, __pte(0)); } + return false; +} - if (kvm_mips_map_page(vcpu->kvm, gfn) < 0) - return KVM_INVALID_ADDR; +static bool kvm_mips_flush_gva_pmd(pmd_t *pmd, unsigned long start_gva, + unsigned long end_gva) +{ + pte_t *pte; + unsigned long end = ~0ul; + int i_min = __pmd_offset(start_gva); + int i_max = __pmd_offset(end_gva); + bool safe_to_remove = (i_min == 0 && i_max == PTRS_PER_PMD - 1); + int i; + + for (i = i_min; i <= i_max; ++i, start_gva = 0) { + if (!pmd_present(pmd[i])) + continue; + + pte = pte_offset(pmd + i, 0); + if (i == i_max) + end = end_gva; + + if (kvm_mips_flush_gva_pte(pte, start_gva, end)) { + pmd_clear(pmd + i); + pte_free_kernel(NULL, pte); + } else { + safe_to_remove = false; + } + } + return safe_to_remove; +} - return (kvm->arch.guest_pmap[gfn] << PAGE_SHIFT) + offset; +static bool kvm_mips_flush_gva_pud(pud_t *pud, unsigned long start_gva, + unsigned long end_gva) +{ + pmd_t *pmd; + unsigned long end = ~0ul; + int i_min = __pud_offset(start_gva); + int i_max = __pud_offset(end_gva); + bool safe_to_remove = (i_min == 0 && i_max == PTRS_PER_PUD - 1); + int i; + + for (i = i_min; i <= i_max; ++i, start_gva = 0) { + if (!pud_present(pud[i])) + continue; + + pmd = pmd_offset(pud + i, 0); + if (i == i_max) + end = end_gva; + + if (kvm_mips_flush_gva_pmd(pmd, start_gva, end)) { + pud_clear(pud + i); + pmd_free(NULL, pmd); + } else { + safe_to_remove = false; + } + } + return safe_to_remove; +} + +static bool kvm_mips_flush_gva_pgd(pgd_t *pgd, unsigned long start_gva, + unsigned long end_gva) +{ + pud_t *pud; + unsigned long end = ~0ul; + int i_min = pgd_index(start_gva); + int i_max = pgd_index(end_gva); + bool safe_to_remove = (i_min == 0 && i_max == PTRS_PER_PGD - 1); + int i; + + for (i = i_min; i <= i_max; ++i, start_gva = 0) { + if (!pgd_present(pgd[i])) + continue; + + pud = pud_offset(pgd + i, 0); + if (i == i_max) + end = end_gva; + + if (kvm_mips_flush_gva_pud(pud, start_gva, end)) { + pgd_clear(pgd + i); + pud_free(NULL, pud); + } else { + safe_to_remove = false; + } + } + return safe_to_remove; +} + +void kvm_mips_flush_gva_pt(pgd_t *pgd, enum kvm_mips_flush flags) +{ + if (flags & KMF_GPA) { + /* all of guest virtual address space could be affected */ + if (flags & KMF_KERN) + /* useg, kseg0, seg2/3 */ + kvm_mips_flush_gva_pgd(pgd, 0, 0x7fffffff); + else + /* useg */ + kvm_mips_flush_gva_pgd(pgd, 0, 0x3fffffff); + } else { + /* useg */ + kvm_mips_flush_gva_pgd(pgd, 0, 0x3fffffff); + + /* kseg2/3 */ + if (flags & KMF_KERN) + kvm_mips_flush_gva_pgd(pgd, 0x60000000, 0x7fffffff); + } +} + +static pte_t kvm_mips_gpa_pte_to_gva_unmapped(pte_t pte) +{ + /* + * Don't leak writeable but clean entries from GPA page tables. We don't + * want the normal Linux tlbmod handler to handle dirtying when KVM + * accesses guest memory. + */ + if (!pte_dirty(pte)) + pte = pte_wrprotect(pte); + + return pte; +} + +static pte_t kvm_mips_gpa_pte_to_gva_mapped(pte_t pte, long entrylo) +{ + /* Guest EntryLo overrides host EntryLo */ + if (!(entrylo & ENTRYLO_D)) + pte = pte_mkclean(pte); + + return kvm_mips_gpa_pte_to_gva_unmapped(pte); } /* XXXKYMA: Must be called with interrupts disabled */ int kvm_mips_handle_kseg0_tlb_fault(unsigned long badvaddr, - struct kvm_vcpu *vcpu) + struct kvm_vcpu *vcpu, + bool write_fault) { - gfn_t gfn; - kvm_pfn_t pfn0, pfn1; - unsigned long vaddr = 0; - unsigned long entryhi = 0, entrylo0 = 0, entrylo1 = 0; - struct kvm *kvm = vcpu->kvm; - const int flush_dcache_mask = 0; - int ret; + unsigned long gpa; + pte_t pte_gpa[2], *ptep_gva; + int idx; if (KVM_GUEST_KSEGX(badvaddr) != KVM_GUEST_KSEG0) { kvm_err("%s: Invalid BadVaddr: %#lx\n", __func__, badvaddr); @@ -98,49 +1007,39 @@ int kvm_mips_handle_kseg0_tlb_fault(unsigned long badvaddr, return -1; } - gfn = (KVM_GUEST_CPHYSADDR(badvaddr) >> PAGE_SHIFT); - if ((gfn | 1) >= kvm->arch.guest_pmap_npages) { - kvm_err("%s: Invalid gfn: %#llx, BadVaddr: %#lx\n", __func__, - gfn, badvaddr); - kvm_mips_dump_host_tlbs(); + /* Get the GPA page table entry */ + gpa = KVM_GUEST_CPHYSADDR(badvaddr); + idx = (badvaddr >> PAGE_SHIFT) & 1; + if (kvm_mips_map_page(vcpu, gpa, write_fault, &pte_gpa[idx], + &pte_gpa[!idx]) < 0) return -1; - } - vaddr = badvaddr & (PAGE_MASK << 1); - if (kvm_mips_map_page(vcpu->kvm, gfn) < 0) + /* Get the GVA page table entry */ + ptep_gva = kvm_trap_emul_pte_for_gva(vcpu, badvaddr & ~PAGE_SIZE); + if (!ptep_gva) { + kvm_err("No ptep for gva %lx\n", badvaddr); return -1; + } - if (kvm_mips_map_page(vcpu->kvm, gfn ^ 0x1) < 0) - return -1; - - pfn0 = kvm->arch.guest_pmap[gfn & ~0x1]; - pfn1 = kvm->arch.guest_pmap[gfn | 0x1]; - - entrylo0 = mips3_paddr_to_tlbpfn(pfn0 << PAGE_SHIFT) | - ((_page_cachable_default >> _CACHE_SHIFT) << ENTRYLO_C_SHIFT) | - ENTRYLO_D | ENTRYLO_V; - entrylo1 = mips3_paddr_to_tlbpfn(pfn1 << PAGE_SHIFT) | - ((_page_cachable_default >> _CACHE_SHIFT) << ENTRYLO_C_SHIFT) | - ENTRYLO_D | ENTRYLO_V; - - preempt_disable(); - entryhi = (vaddr | kvm_mips_get_kernel_asid(vcpu)); - ret = kvm_mips_host_tlb_write(vcpu, entryhi, entrylo0, entrylo1, - flush_dcache_mask); - preempt_enable(); + /* Copy a pair of entries from GPA page table to GVA page table */ + ptep_gva[0] = kvm_mips_gpa_pte_to_gva_unmapped(pte_gpa[0]); + ptep_gva[1] = kvm_mips_gpa_pte_to_gva_unmapped(pte_gpa[1]); - return ret; + /* Invalidate this entry in the TLB, guest kernel ASID only */ + kvm_mips_host_tlb_inv(vcpu, badvaddr, false, true); + return 0; } int kvm_mips_handle_mapped_seg_tlb_fault(struct kvm_vcpu *vcpu, - struct kvm_mips_tlb *tlb) + struct kvm_mips_tlb *tlb, + unsigned long gva, + bool write_fault) { - unsigned long entryhi = 0, entrylo0 = 0, entrylo1 = 0; struct kvm *kvm = vcpu->kvm; - kvm_pfn_t pfn0, pfn1; - gfn_t gfn0, gfn1; long tlb_lo[2]; - int ret; + pte_t pte_gpa[2], *ptep_buddy, *ptep_gva; + unsigned int idx = TLB_LO_IDX(*tlb, gva); + bool kernel = KVM_GUEST_KERNEL_MODE(vcpu); tlb_lo[0] = tlb->tlb_lo[0]; tlb_lo[1] = tlb->tlb_lo[1]; @@ -149,70 +1048,64 @@ int kvm_mips_handle_mapped_seg_tlb_fault(struct kvm_vcpu *vcpu, * The commpage address must not be mapped to anything else if the guest * TLB contains entries nearby, or commpage accesses will break. */ - if (!((tlb->tlb_hi ^ KVM_GUEST_COMMPAGE_ADDR) & - VPN2_MASK & (PAGE_MASK << 1))) - tlb_lo[(KVM_GUEST_COMMPAGE_ADDR >> PAGE_SHIFT) & 1] = 0; - - gfn0 = mips3_tlbpfn_to_paddr(tlb_lo[0]) >> PAGE_SHIFT; - gfn1 = mips3_tlbpfn_to_paddr(tlb_lo[1]) >> PAGE_SHIFT; - if (gfn0 >= kvm->arch.guest_pmap_npages || - gfn1 >= kvm->arch.guest_pmap_npages) { - kvm_err("%s: Invalid gfn: [%#llx, %#llx], EHi: %#lx\n", - __func__, gfn0, gfn1, tlb->tlb_hi); - kvm_mips_dump_guest_tlbs(vcpu); - return -1; - } + if (!((gva ^ KVM_GUEST_COMMPAGE_ADDR) & VPN2_MASK & (PAGE_MASK << 1))) + tlb_lo[TLB_LO_IDX(*tlb, KVM_GUEST_COMMPAGE_ADDR)] = 0; - if (kvm_mips_map_page(kvm, gfn0) < 0) + /* Get the GPA page table entry */ + if (kvm_mips_map_page(vcpu, mips3_tlbpfn_to_paddr(tlb_lo[idx]), + write_fault, &pte_gpa[idx], NULL) < 0) return -1; - if (kvm_mips_map_page(kvm, gfn1) < 0) + /* And its GVA buddy's GPA page table entry if it also exists */ + pte_gpa[!idx] = pfn_pte(0, __pgprot(0)); + if (tlb_lo[!idx] & ENTRYLO_V) { + spin_lock(&kvm->mmu_lock); + ptep_buddy = kvm_mips_pte_for_gpa(kvm, NULL, + mips3_tlbpfn_to_paddr(tlb_lo[!idx])); + if (ptep_buddy) + pte_gpa[!idx] = *ptep_buddy; + spin_unlock(&kvm->mmu_lock); + } + + /* Get the GVA page table entry pair */ + ptep_gva = kvm_trap_emul_pte_for_gva(vcpu, gva & ~PAGE_SIZE); + if (!ptep_gva) { + kvm_err("No ptep for gva %lx\n", gva); return -1; + } - pfn0 = kvm->arch.guest_pmap[gfn0]; - pfn1 = kvm->arch.guest_pmap[gfn1]; + /* Copy a pair of entries from GPA page table to GVA page table */ + ptep_gva[0] = kvm_mips_gpa_pte_to_gva_mapped(pte_gpa[0], tlb_lo[0]); + ptep_gva[1] = kvm_mips_gpa_pte_to_gva_mapped(pte_gpa[1], tlb_lo[1]); - /* Get attributes from the Guest TLB */ - entrylo0 = mips3_paddr_to_tlbpfn(pfn0 << PAGE_SHIFT) | - ((_page_cachable_default >> _CACHE_SHIFT) << ENTRYLO_C_SHIFT) | - (tlb_lo[0] & ENTRYLO_D) | - (tlb_lo[0] & ENTRYLO_V); - entrylo1 = mips3_paddr_to_tlbpfn(pfn1 << PAGE_SHIFT) | - ((_page_cachable_default >> _CACHE_SHIFT) << ENTRYLO_C_SHIFT) | - (tlb_lo[1] & ENTRYLO_D) | - (tlb_lo[1] & ENTRYLO_V); + /* Invalidate this entry in the TLB, current guest mode ASID only */ + kvm_mips_host_tlb_inv(vcpu, gva, !kernel, kernel); kvm_debug("@ %#lx tlb_lo0: 0x%08lx tlb_lo1: 0x%08lx\n", vcpu->arch.pc, tlb->tlb_lo[0], tlb->tlb_lo[1]); - preempt_disable(); - entryhi = (tlb->tlb_hi & VPN2_MASK) | (KVM_GUEST_KERNEL_MODE(vcpu) ? - kvm_mips_get_kernel_asid(vcpu) : - kvm_mips_get_user_asid(vcpu)); - ret = kvm_mips_host_tlb_write(vcpu, entryhi, entrylo0, entrylo1, - tlb->tlb_mask); - preempt_enable(); - - return ret; + return 0; } -void kvm_get_new_mmu_context(struct mm_struct *mm, unsigned long cpu, - struct kvm_vcpu *vcpu) +int kvm_mips_handle_commpage_tlb_fault(unsigned long badvaddr, + struct kvm_vcpu *vcpu) { - unsigned long asid = asid_cache(cpu); - - asid += cpu_asid_inc(); - if (!(asid & cpu_asid_mask(&cpu_data[cpu]))) { - if (cpu_has_vtag_icache) - flush_icache_all(); - - kvm_local_flush_tlb_all(); /* start new asid cycle */ + kvm_pfn_t pfn; + pte_t *ptep; - if (!asid) /* fix version if needed */ - asid = asid_first_version(cpu); + ptep = kvm_trap_emul_pte_for_gva(vcpu, badvaddr); + if (!ptep) { + kvm_err("No ptep for commpage %lx\n", badvaddr); + return -1; } - cpu_context(cpu, mm) = asid_cache(cpu) = asid; + pfn = PFN_DOWN(virt_to_phys(vcpu->arch.kseg0_commpage)); + /* Also set valid and dirty, so refill handler doesn't have to */ + *ptep = pte_mkyoung(pte_mkdirty(pfn_pte(pfn, PAGE_SHARED))); + + /* Invalidate this entry in the TLB, guest kernel ASID only */ + kvm_mips_host_tlb_inv(vcpu, badvaddr, false, true); + return 0; } /** @@ -235,42 +1128,13 @@ static void kvm_mips_migrate_count(struct kvm_vcpu *vcpu) /* Restore ASID once we are scheduled back after preemption */ void kvm_arch_vcpu_load(struct kvm_vcpu *vcpu, int cpu) { - unsigned long asid_mask = cpu_asid_mask(&cpu_data[cpu]); unsigned long flags; - int newasid = 0; kvm_debug("%s: vcpu %p, cpu: %d\n", __func__, vcpu, cpu); - /* Allocate new kernel and user ASIDs if needed */ - local_irq_save(flags); - if ((vcpu->arch.guest_kernel_asid[cpu] ^ asid_cache(cpu)) & - asid_version_mask(cpu)) { - kvm_get_new_mmu_context(&vcpu->arch.guest_kernel_mm, cpu, vcpu); - vcpu->arch.guest_kernel_asid[cpu] = - vcpu->arch.guest_kernel_mm.context.asid[cpu]; - newasid++; - - kvm_debug("[%d]: cpu_context: %#lx\n", cpu, - cpu_context(cpu, current->mm)); - kvm_debug("[%d]: Allocated new ASID for Guest Kernel: %#x\n", - cpu, vcpu->arch.guest_kernel_asid[cpu]); - } - - if ((vcpu->arch.guest_user_asid[cpu] ^ asid_cache(cpu)) & - asid_version_mask(cpu)) { - kvm_get_new_mmu_context(&vcpu->arch.guest_user_mm, cpu, vcpu); - vcpu->arch.guest_user_asid[cpu] = - vcpu->arch.guest_user_mm.context.asid[cpu]; - newasid++; - - kvm_debug("[%d]: cpu_context: %#lx\n", cpu, - cpu_context(cpu, current->mm)); - kvm_debug("[%d]: Allocated new ASID for Guest User: %#x\n", cpu, - vcpu->arch.guest_user_asid[cpu]); - } - + vcpu->cpu = cpu; if (vcpu->arch.last_sched_cpu != cpu) { kvm_debug("[%d->%d]KVM VCPU[%d] switch\n", vcpu->arch.last_sched_cpu, cpu, vcpu->vcpu_id); @@ -282,42 +1146,10 @@ void kvm_arch_vcpu_load(struct kvm_vcpu *vcpu, int cpu) kvm_mips_migrate_count(vcpu); } - if (!newasid) { - /* - * If we preempted while the guest was executing, then reload - * the pre-empted ASID - */ - if (current->flags & PF_VCPU) { - write_c0_entryhi(vcpu->arch. - preempt_entryhi & asid_mask); - ehb(); - } - } else { - /* New ASIDs were allocated for the VM */ - - /* - * Were we in guest context? If so then the pre-empted ASID is - * no longer valid, we need to set it to what it should be based - * on the mode of the Guest (Kernel/User) - */ - if (current->flags & PF_VCPU) { - if (KVM_GUEST_KERNEL_MODE(vcpu)) - write_c0_entryhi(vcpu->arch. - guest_kernel_asid[cpu] & - asid_mask); - else - write_c0_entryhi(vcpu->arch. - guest_user_asid[cpu] & - asid_mask); - ehb(); - } - } - /* restore guest state to registers */ - kvm_mips_callbacks->vcpu_set_regs(vcpu); + kvm_mips_callbacks->vcpu_load(vcpu, cpu); local_irq_restore(flags); - } /* ASID can change if another task is scheduled during preemption */ @@ -329,75 +1161,90 @@ void kvm_arch_vcpu_put(struct kvm_vcpu *vcpu) local_irq_save(flags); cpu = smp_processor_id(); - - vcpu->arch.preempt_entryhi = read_c0_entryhi(); vcpu->arch.last_sched_cpu = cpu; + vcpu->cpu = -1; /* save guest state in registers */ - kvm_mips_callbacks->vcpu_get_regs(vcpu); - - if (((cpu_context(cpu, current->mm) ^ asid_cache(cpu)) & - asid_version_mask(cpu))) { - kvm_debug("%s: Dropping MMU Context: %#lx\n", __func__, - cpu_context(cpu, current->mm)); - drop_mmu_context(current->mm, cpu); - } - write_c0_entryhi(cpu_asid(cpu, current->mm)); - ehb(); + kvm_mips_callbacks->vcpu_put(vcpu, cpu); local_irq_restore(flags); } -u32 kvm_get_inst(u32 *opc, struct kvm_vcpu *vcpu) +/** + * kvm_trap_emul_gva_fault() - Safely attempt to handle a GVA access fault. + * @vcpu: Virtual CPU. + * @gva: Guest virtual address to be accessed. + * @write: True if write attempted (must be dirtied and made writable). + * + * Safely attempt to handle a GVA fault, mapping GVA pages if necessary, and + * dirtying the page if @write so that guest instructions can be modified. + * + * Returns: KVM_MIPS_MAPPED on success. + * KVM_MIPS_GVA if bad guest virtual address. + * KVM_MIPS_GPA if bad guest physical address. + * KVM_MIPS_TLB if guest TLB not present. + * KVM_MIPS_TLBINV if guest TLB present but not valid. + * KVM_MIPS_TLBMOD if guest TLB read only. + */ +enum kvm_mips_fault_result kvm_trap_emul_gva_fault(struct kvm_vcpu *vcpu, + unsigned long gva, + bool write) { struct mips_coproc *cop0 = vcpu->arch.cop0; - unsigned long paddr, flags, vpn2, asid; - unsigned long va = (unsigned long)opc; - void *vaddr; - u32 inst; + struct kvm_mips_tlb *tlb; int index; - if (KVM_GUEST_KSEGX(va) < KVM_GUEST_KSEG0 || - KVM_GUEST_KSEGX(va) == KVM_GUEST_KSEG23) { - local_irq_save(flags); - index = kvm_mips_host_tlb_lookup(vcpu, va); - if (index >= 0) { - inst = *(opc); - } else { - vpn2 = va & VPN2_MASK; - asid = kvm_read_c0_guest_entryhi(cop0) & - KVM_ENTRYHI_ASID; - index = kvm_mips_guest_tlb_lookup(vcpu, vpn2 | asid); - if (index < 0) { - kvm_err("%s: get_user_failed for %p, vcpu: %p, ASID: %#lx\n", - __func__, opc, vcpu, read_c0_entryhi()); - kvm_mips_dump_host_tlbs(); - kvm_mips_dump_guest_tlbs(vcpu); - local_irq_restore(flags); - return KVM_INVALID_INST; - } - if (kvm_mips_handle_mapped_seg_tlb_fault(vcpu, - &vcpu->arch.guest_tlb[index])) { - kvm_err("%s: handling mapped seg tlb fault failed for %p, index: %u, vcpu: %p, ASID: %#lx\n", - __func__, opc, index, vcpu, - read_c0_entryhi()); - kvm_mips_dump_guest_tlbs(vcpu); - local_irq_restore(flags); - return KVM_INVALID_INST; - } - inst = *(opc); - } - local_irq_restore(flags); - } else if (KVM_GUEST_KSEGX(va) == KVM_GUEST_KSEG0) { - paddr = kvm_mips_translate_guest_kseg0_to_hpa(vcpu, va); - vaddr = kmap_atomic(pfn_to_page(PHYS_PFN(paddr))); - vaddr += paddr & ~PAGE_MASK; - inst = *(u32 *)vaddr; - kunmap_atomic(vaddr); + if (KVM_GUEST_KSEGX(gva) == KVM_GUEST_KSEG0) { + if (kvm_mips_handle_kseg0_tlb_fault(gva, vcpu, write) < 0) + return KVM_MIPS_GPA; + } else if ((KVM_GUEST_KSEGX(gva) < KVM_GUEST_KSEG0) || + KVM_GUEST_KSEGX(gva) == KVM_GUEST_KSEG23) { + /* Address should be in the guest TLB */ + index = kvm_mips_guest_tlb_lookup(vcpu, (gva & VPN2_MASK) | + (kvm_read_c0_guest_entryhi(cop0) & KVM_ENTRYHI_ASID)); + if (index < 0) + return KVM_MIPS_TLB; + tlb = &vcpu->arch.guest_tlb[index]; + + /* Entry should be valid, and dirty for writes */ + if (!TLB_IS_VALID(*tlb, gva)) + return KVM_MIPS_TLBINV; + if (write && !TLB_IS_DIRTY(*tlb, gva)) + return KVM_MIPS_TLBMOD; + + if (kvm_mips_handle_mapped_seg_tlb_fault(vcpu, tlb, gva, write)) + return KVM_MIPS_GPA; } else { - kvm_err("%s: illegal address: %p\n", __func__, opc); - return KVM_INVALID_INST; + return KVM_MIPS_GVA; } - return inst; + return KVM_MIPS_MAPPED; +} + +int kvm_get_inst(u32 *opc, struct kvm_vcpu *vcpu, u32 *out) +{ + int err; + +retry: + kvm_trap_emul_gva_lockless_begin(vcpu); + err = get_user(*out, opc); + kvm_trap_emul_gva_lockless_end(vcpu); + + if (unlikely(err)) { + /* + * Try to handle the fault, maybe we just raced with a GVA + * invalidation. + */ + err = kvm_trap_emul_gva_fault(vcpu, (unsigned long)opc, + false); + if (unlikely(err)) { + kvm_err("%s: illegal address: %p\n", + __func__, opc); + return -EFAULT; + } + + /* Hopefully it'll work now */ + goto retry; + } + return 0; } diff --git a/arch/mips/kvm/tlb.c b/arch/mips/kvm/tlb.c index 254377d8e0b9..2819eb793345 100644 --- a/arch/mips/kvm/tlb.c +++ b/arch/mips/kvm/tlb.c @@ -33,28 +33,20 @@ #define KVM_GUEST_PC_TLB 0 #define KVM_GUEST_SP_TLB 1 -atomic_t kvm_mips_instance; -EXPORT_SYMBOL_GPL(kvm_mips_instance); - static u32 kvm_mips_get_kernel_asid(struct kvm_vcpu *vcpu) { + struct mm_struct *kern_mm = &vcpu->arch.guest_kernel_mm; int cpu = smp_processor_id(); - return vcpu->arch.guest_kernel_asid[cpu] & - cpu_asid_mask(&cpu_data[cpu]); + return cpu_asid(cpu, kern_mm); } static u32 kvm_mips_get_user_asid(struct kvm_vcpu *vcpu) { + struct mm_struct *user_mm = &vcpu->arch.guest_user_mm; int cpu = smp_processor_id(); - return vcpu->arch.guest_user_asid[cpu] & - cpu_asid_mask(&cpu_data[cpu]); -} - -inline u32 kvm_mips_get_commpage_asid(struct kvm_vcpu *vcpu) -{ - return vcpu->kvm->arch.commpage_tlb; + return cpu_asid(cpu, user_mm); } /* Structure defining an tlb entry data set. */ @@ -104,109 +96,6 @@ void kvm_mips_dump_guest_tlbs(struct kvm_vcpu *vcpu) } EXPORT_SYMBOL_GPL(kvm_mips_dump_guest_tlbs); -/* XXXKYMA: Must be called with interrupts disabled */ -/* set flush_dcache_mask == 0 if no dcache flush required */ -int kvm_mips_host_tlb_write(struct kvm_vcpu *vcpu, unsigned long entryhi, - unsigned long entrylo0, unsigned long entrylo1, - int flush_dcache_mask) -{ - unsigned long flags; - unsigned long old_entryhi; - int idx; - - local_irq_save(flags); - - old_entryhi = read_c0_entryhi(); - write_c0_entryhi(entryhi); - mtc0_tlbw_hazard(); - - tlb_probe(); - tlb_probe_hazard(); - idx = read_c0_index(); - - if (idx > current_cpu_data.tlbsize) { - kvm_err("%s: Invalid Index: %d\n", __func__, idx); - kvm_mips_dump_host_tlbs(); - local_irq_restore(flags); - return -1; - } - - write_c0_entrylo0(entrylo0); - write_c0_entrylo1(entrylo1); - mtc0_tlbw_hazard(); - - if (idx < 0) - tlb_write_random(); - else - tlb_write_indexed(); - tlbw_use_hazard(); - - kvm_debug("@ %#lx idx: %2d [entryhi(R): %#lx] entrylo0(R): 0x%08lx, entrylo1(R): 0x%08lx\n", - vcpu->arch.pc, idx, read_c0_entryhi(), - read_c0_entrylo0(), read_c0_entrylo1()); - - /* Flush D-cache */ - if (flush_dcache_mask) { - if (entrylo0 & ENTRYLO_V) { - ++vcpu->stat.flush_dcache_exits; - flush_data_cache_page((entryhi & VPN2_MASK) & - ~flush_dcache_mask); - } - if (entrylo1 & ENTRYLO_V) { - ++vcpu->stat.flush_dcache_exits; - flush_data_cache_page(((entryhi & VPN2_MASK) & - ~flush_dcache_mask) | - (0x1 << PAGE_SHIFT)); - } - } - - /* Restore old ASID */ - write_c0_entryhi(old_entryhi); - mtc0_tlbw_hazard(); - local_irq_restore(flags); - return 0; -} -EXPORT_SYMBOL_GPL(kvm_mips_host_tlb_write); - -int kvm_mips_handle_commpage_tlb_fault(unsigned long badvaddr, - struct kvm_vcpu *vcpu) -{ - kvm_pfn_t pfn; - unsigned long flags, old_entryhi = 0, vaddr = 0; - unsigned long entrylo[2] = { 0, 0 }; - unsigned int pair_idx; - - pfn = PFN_DOWN(virt_to_phys(vcpu->arch.kseg0_commpage)); - pair_idx = (badvaddr >> PAGE_SHIFT) & 1; - entrylo[pair_idx] = mips3_paddr_to_tlbpfn(pfn << PAGE_SHIFT) | - ((_page_cachable_default >> _CACHE_SHIFT) << ENTRYLO_C_SHIFT) | - ENTRYLO_D | ENTRYLO_V; - - local_irq_save(flags); - - old_entryhi = read_c0_entryhi(); - vaddr = badvaddr & (PAGE_MASK << 1); - write_c0_entryhi(vaddr | kvm_mips_get_kernel_asid(vcpu)); - write_c0_entrylo0(entrylo[0]); - write_c0_entrylo1(entrylo[1]); - write_c0_index(kvm_mips_get_commpage_asid(vcpu)); - mtc0_tlbw_hazard(); - tlb_write_indexed(); - tlbw_use_hazard(); - - kvm_debug("@ %#lx idx: %2d [entryhi(R): %#lx] entrylo0 (R): 0x%08lx, entrylo1(R): 0x%08lx\n", - vcpu->arch.pc, read_c0_index(), read_c0_entryhi(), - read_c0_entrylo0(), read_c0_entrylo1()); - - /* Restore old ASID */ - write_c0_entryhi(old_entryhi); - mtc0_tlbw_hazard(); - local_irq_restore(flags); - - return 0; -} -EXPORT_SYMBOL_GPL(kvm_mips_handle_commpage_tlb_fault); - int kvm_mips_guest_tlb_lookup(struct kvm_vcpu *vcpu, unsigned long entryhi) { int i; @@ -228,51 +117,11 @@ int kvm_mips_guest_tlb_lookup(struct kvm_vcpu *vcpu, unsigned long entryhi) } EXPORT_SYMBOL_GPL(kvm_mips_guest_tlb_lookup); -int kvm_mips_host_tlb_lookup(struct kvm_vcpu *vcpu, unsigned long vaddr) -{ - unsigned long old_entryhi, flags; - int idx; - - local_irq_save(flags); - - old_entryhi = read_c0_entryhi(); - - if (KVM_GUEST_KERNEL_MODE(vcpu)) - write_c0_entryhi((vaddr & VPN2_MASK) | - kvm_mips_get_kernel_asid(vcpu)); - else { - write_c0_entryhi((vaddr & VPN2_MASK) | - kvm_mips_get_user_asid(vcpu)); - } - - mtc0_tlbw_hazard(); - - tlb_probe(); - tlb_probe_hazard(); - idx = read_c0_index(); - - /* Restore old ASID */ - write_c0_entryhi(old_entryhi); - mtc0_tlbw_hazard(); - - local_irq_restore(flags); - - kvm_debug("Host TLB lookup, %#lx, idx: %2d\n", vaddr, idx); - - return idx; -} -EXPORT_SYMBOL_GPL(kvm_mips_host_tlb_lookup); - -int kvm_mips_host_tlb_inv(struct kvm_vcpu *vcpu, unsigned long va) +static int _kvm_mips_host_tlb_inv(unsigned long entryhi) { int idx; - unsigned long flags, old_entryhi; - - local_irq_save(flags); - - old_entryhi = read_c0_entryhi(); - write_c0_entryhi((va & VPN2_MASK) | kvm_mips_get_user_asid(vcpu)); + write_c0_entryhi(entryhi); mtc0_tlbw_hazard(); tlb_probe(); @@ -282,7 +131,7 @@ int kvm_mips_host_tlb_inv(struct kvm_vcpu *vcpu, unsigned long va) if (idx >= current_cpu_data.tlbsize) BUG(); - if (idx > 0) { + if (idx >= 0) { write_c0_entryhi(UNIQUE_ENTRYHI(idx)); write_c0_entrylo0(0); write_c0_entrylo1(0); @@ -292,93 +141,75 @@ int kvm_mips_host_tlb_inv(struct kvm_vcpu *vcpu, unsigned long va) tlbw_use_hazard(); } - write_c0_entryhi(old_entryhi); - mtc0_tlbw_hazard(); - - local_irq_restore(flags); - - if (idx > 0) - kvm_debug("%s: Invalidated entryhi %#lx @ idx %d\n", __func__, - (va & VPN2_MASK) | kvm_mips_get_user_asid(vcpu), idx); - - return 0; + return idx; } -EXPORT_SYMBOL_GPL(kvm_mips_host_tlb_inv); -void kvm_mips_flush_host_tlb(int skip_kseg0) +int kvm_mips_host_tlb_inv(struct kvm_vcpu *vcpu, unsigned long va, + bool user, bool kernel) { - unsigned long flags; - unsigned long old_entryhi, entryhi; - unsigned long old_pagemask; - int entry = 0; - int maxentry = current_cpu_data.tlbsize; + int idx_user, idx_kernel; + unsigned long flags, old_entryhi; local_irq_save(flags); old_entryhi = read_c0_entryhi(); - old_pagemask = read_c0_pagemask(); - - /* Blast 'em all away. */ - for (entry = 0; entry < maxentry; entry++) { - write_c0_index(entry); - - if (skip_kseg0) { - mtc0_tlbr_hazard(); - tlb_read(); - tlb_read_hazard(); - - entryhi = read_c0_entryhi(); - /* Don't blow away guest kernel entries */ - if (KVM_GUEST_KSEGX(entryhi) == KVM_GUEST_KSEG0) - continue; - - write_c0_pagemask(old_pagemask); - } - - /* Make sure all entries differ. */ - write_c0_entryhi(UNIQUE_ENTRYHI(entry)); - write_c0_entrylo0(0); - write_c0_entrylo1(0); - mtc0_tlbw_hazard(); - - tlb_write_indexed(); - tlbw_use_hazard(); - } + if (user) + idx_user = _kvm_mips_host_tlb_inv((va & VPN2_MASK) | + kvm_mips_get_user_asid(vcpu)); + if (kernel) + idx_kernel = _kvm_mips_host_tlb_inv((va & VPN2_MASK) | + kvm_mips_get_kernel_asid(vcpu)); write_c0_entryhi(old_entryhi); - write_c0_pagemask(old_pagemask); mtc0_tlbw_hazard(); local_irq_restore(flags); + + if (user && idx_user >= 0) + kvm_debug("%s: Invalidated guest user entryhi %#lx @ idx %d\n", + __func__, (va & VPN2_MASK) | + kvm_mips_get_user_asid(vcpu), idx_user); + if (kernel && idx_kernel >= 0) + kvm_debug("%s: Invalidated guest kernel entryhi %#lx @ idx %d\n", + __func__, (va & VPN2_MASK) | + kvm_mips_get_kernel_asid(vcpu), idx_kernel); + + return 0; } -EXPORT_SYMBOL_GPL(kvm_mips_flush_host_tlb); +EXPORT_SYMBOL_GPL(kvm_mips_host_tlb_inv); -void kvm_local_flush_tlb_all(void) +/** + * kvm_mips_suspend_mm() - Suspend the active mm. + * @cpu The CPU we're running on. + * + * Suspend the active_mm, ready for a switch to a KVM guest virtual address + * space. This is left active for the duration of guest context, including time + * with interrupts enabled, so we need to be careful not to confuse e.g. cache + * management IPIs. + * + * kvm_mips_resume_mm() should be called before context switching to a different + * process so we don't need to worry about reference counting. + * + * This needs to be in static kernel code to avoid exporting init_mm. + */ +void kvm_mips_suspend_mm(int cpu) { - unsigned long flags; - unsigned long old_ctx; - int entry = 0; - - local_irq_save(flags); - /* Save old context and create impossible VPN2 value */ - old_ctx = read_c0_entryhi(); - write_c0_entrylo0(0); - write_c0_entrylo1(0); - - /* Blast 'em all away. */ - while (entry < current_cpu_data.tlbsize) { - /* Make sure all entries differ. */ - write_c0_entryhi(UNIQUE_ENTRYHI(entry)); - write_c0_index(entry); - mtc0_tlbw_hazard(); - tlb_write_indexed(); - tlbw_use_hazard(); - entry++; - } - write_c0_entryhi(old_ctx); - mtc0_tlbw_hazard(); + cpumask_clear_cpu(cpu, mm_cpumask(current->active_mm)); + current->active_mm = &init_mm; +} +EXPORT_SYMBOL_GPL(kvm_mips_suspend_mm); - local_irq_restore(flags); +/** + * kvm_mips_resume_mm() - Resume the current process mm. + * @cpu The CPU we're running on. + * + * Resume the mm of the current process, after a switch back from a KVM guest + * virtual address space (see kvm_mips_suspend_mm()). + */ +void kvm_mips_resume_mm(int cpu) +{ + cpumask_set_cpu(cpu, mm_cpumask(current->mm)); + current->active_mm = current->mm; } -EXPORT_SYMBOL_GPL(kvm_local_flush_tlb_all); +EXPORT_SYMBOL_GPL(kvm_mips_resume_mm); diff --git a/arch/mips/kvm/trap_emul.c b/arch/mips/kvm/trap_emul.c index 3b20441f2beb..b1fa53b252ea 100644 --- a/arch/mips/kvm/trap_emul.c +++ b/arch/mips/kvm/trap_emul.c @@ -11,9 +11,11 @@ #include <linux/errno.h> #include <linux/err.h> -#include <linux/vmalloc.h> - #include <linux/kvm_host.h> +#include <linux/uaccess.h> +#include <linux/vmalloc.h> +#include <asm/mmu_context.h> +#include <asm/pgalloc.h> #include "interrupt.h" @@ -21,9 +23,12 @@ static gpa_t kvm_trap_emul_gva_to_gpa_cb(gva_t gva) { gpa_t gpa; gva_t kseg = KSEGX(gva); + gva_t gkseg = KVM_GUEST_KSEGX(gva); if ((kseg == CKSEG0) || (kseg == CKSEG1)) gpa = CPHYSADDR(gva); + else if (gkseg == KVM_GUEST_KSEG0) + gpa = KVM_GUEST_CPHYSADDR(gva); else { kvm_err("%s: cannot find GPA for GVA: %#lx\n", __func__, gva); kvm_mips_dump_host_tlbs(); @@ -83,48 +88,134 @@ static int kvm_trap_emul_handle_cop_unusable(struct kvm_vcpu *vcpu) return ret; } +static int kvm_mips_bad_load(u32 cause, u32 *opc, struct kvm_run *run, + struct kvm_vcpu *vcpu) +{ + enum emulation_result er; + union mips_instruction inst; + int err; + + /* A code fetch fault doesn't count as an MMIO */ + if (kvm_is_ifetch_fault(&vcpu->arch)) { + run->exit_reason = KVM_EXIT_INTERNAL_ERROR; + return RESUME_HOST; + } + + /* Fetch the instruction. */ + if (cause & CAUSEF_BD) + opc += 1; + err = kvm_get_badinstr(opc, vcpu, &inst.word); + if (err) { + run->exit_reason = KVM_EXIT_INTERNAL_ERROR; + return RESUME_HOST; + } + + /* Emulate the load */ + er = kvm_mips_emulate_load(inst, cause, run, vcpu); + if (er == EMULATE_FAIL) { + kvm_err("Emulate load from MMIO space failed\n"); + run->exit_reason = KVM_EXIT_INTERNAL_ERROR; + } else { + run->exit_reason = KVM_EXIT_MMIO; + } + return RESUME_HOST; +} + +static int kvm_mips_bad_store(u32 cause, u32 *opc, struct kvm_run *run, + struct kvm_vcpu *vcpu) +{ + enum emulation_result er; + union mips_instruction inst; + int err; + + /* Fetch the instruction. */ + if (cause & CAUSEF_BD) + opc += 1; + err = kvm_get_badinstr(opc, vcpu, &inst.word); + if (err) { + run->exit_reason = KVM_EXIT_INTERNAL_ERROR; + return RESUME_HOST; + } + + /* Emulate the store */ + er = kvm_mips_emulate_store(inst, cause, run, vcpu); + if (er == EMULATE_FAIL) { + kvm_err("Emulate store to MMIO space failed\n"); + run->exit_reason = KVM_EXIT_INTERNAL_ERROR; + } else { + run->exit_reason = KVM_EXIT_MMIO; + } + return RESUME_HOST; +} + +static int kvm_mips_bad_access(u32 cause, u32 *opc, struct kvm_run *run, + struct kvm_vcpu *vcpu, bool store) +{ + if (store) + return kvm_mips_bad_store(cause, opc, run, vcpu); + else + return kvm_mips_bad_load(cause, opc, run, vcpu); +} + static int kvm_trap_emul_handle_tlb_mod(struct kvm_vcpu *vcpu) { + struct mips_coproc *cop0 = vcpu->arch.cop0; struct kvm_run *run = vcpu->run; u32 __user *opc = (u32 __user *) vcpu->arch.pc; unsigned long badvaddr = vcpu->arch.host_cp0_badvaddr; u32 cause = vcpu->arch.host_cp0_cause; - enum emulation_result er = EMULATE_DONE; - int ret = RESUME_GUEST; + struct kvm_mips_tlb *tlb; + unsigned long entryhi; + int index; if (KVM_GUEST_KSEGX(badvaddr) < KVM_GUEST_KSEG0 || KVM_GUEST_KSEGX(badvaddr) == KVM_GUEST_KSEG23) { - kvm_debug("USER/KSEG23 ADDR TLB MOD fault: cause %#x, PC: %p, BadVaddr: %#lx\n", - cause, opc, badvaddr); - er = kvm_mips_handle_tlbmod(cause, opc, run, vcpu); + /* + * First find the mapping in the guest TLB. If the failure to + * write was due to the guest TLB, it should be up to the guest + * to handle it. + */ + entryhi = (badvaddr & VPN2_MASK) | + (kvm_read_c0_guest_entryhi(cop0) & KVM_ENTRYHI_ASID); + index = kvm_mips_guest_tlb_lookup(vcpu, entryhi); - if (er == EMULATE_DONE) - ret = RESUME_GUEST; - else { + /* + * These should never happen. + * They would indicate stale host TLB entries. + */ + if (unlikely(index < 0)) { run->exit_reason = KVM_EXIT_INTERNAL_ERROR; - ret = RESUME_HOST; + return RESUME_HOST; } - } else if (KVM_GUEST_KSEGX(badvaddr) == KVM_GUEST_KSEG0) { + tlb = vcpu->arch.guest_tlb + index; + if (unlikely(!TLB_IS_VALID(*tlb, badvaddr))) { + run->exit_reason = KVM_EXIT_INTERNAL_ERROR; + return RESUME_HOST; + } + /* - * XXXKYMA: The guest kernel does not expect to get this fault - * when we are not using HIGHMEM. Need to address this in a - * HIGHMEM kernel + * Guest entry not dirty? That would explain the TLB modified + * exception. Relay that on to the guest so it can handle it. */ - kvm_err("TLB MOD fault not handled, cause %#x, PC: %p, BadVaddr: %#lx\n", - cause, opc, badvaddr); - kvm_mips_dump_host_tlbs(); - kvm_arch_vcpu_dump_regs(vcpu); - run->exit_reason = KVM_EXIT_INTERNAL_ERROR; - ret = RESUME_HOST; + if (!TLB_IS_DIRTY(*tlb, badvaddr)) { + kvm_mips_emulate_tlbmod(cause, opc, run, vcpu); + return RESUME_GUEST; + } + + if (kvm_mips_handle_mapped_seg_tlb_fault(vcpu, tlb, badvaddr, + true)) + /* Not writable, needs handling as MMIO */ + return kvm_mips_bad_store(cause, opc, run, vcpu); + return RESUME_GUEST; + } else if (KVM_GUEST_KSEGX(badvaddr) == KVM_GUEST_KSEG0) { + if (kvm_mips_handle_kseg0_tlb_fault(badvaddr, vcpu, true) < 0) + /* Not writable, needs handling as MMIO */ + return kvm_mips_bad_store(cause, opc, run, vcpu); + return RESUME_GUEST; } else { - kvm_err("Illegal TLB Mod fault address , cause %#x, PC: %p, BadVaddr: %#lx\n", - cause, opc, badvaddr); - kvm_mips_dump_host_tlbs(); - kvm_arch_vcpu_dump_regs(vcpu); - run->exit_reason = KVM_EXIT_INTERNAL_ERROR; - ret = RESUME_HOST; + /* host kernel addresses are all handled as MMIO */ + return kvm_mips_bad_store(cause, opc, run, vcpu); } - return ret; } static int kvm_trap_emul_handle_tlb_miss(struct kvm_vcpu *vcpu, bool store) @@ -157,7 +248,7 @@ static int kvm_trap_emul_handle_tlb_miss(struct kvm_vcpu *vcpu, bool store) * into the shadow host TLB */ - er = kvm_mips_handle_tlbmiss(cause, opc, run, vcpu); + er = kvm_mips_handle_tlbmiss(cause, opc, run, vcpu, store); if (er == EMULATE_DONE) ret = RESUME_GUEST; else { @@ -169,29 +260,15 @@ static int kvm_trap_emul_handle_tlb_miss(struct kvm_vcpu *vcpu, bool store) * All KSEG0 faults are handled by KVM, as the guest kernel does * not expect to ever get them */ - if (kvm_mips_handle_kseg0_tlb_fault - (vcpu->arch.host_cp0_badvaddr, vcpu) < 0) { - run->exit_reason = KVM_EXIT_INTERNAL_ERROR; - ret = RESUME_HOST; - } + if (kvm_mips_handle_kseg0_tlb_fault(badvaddr, vcpu, store) < 0) + ret = kvm_mips_bad_access(cause, opc, run, vcpu, store); } else if (KVM_GUEST_KERNEL_MODE(vcpu) && (KSEGX(badvaddr) == CKSEG0 || KSEGX(badvaddr) == CKSEG1)) { /* * With EVA we may get a TLB exception instead of an address * error when the guest performs MMIO to KSeg1 addresses. */ - kvm_debug("Emulate %s MMIO space\n", - store ? "Store to" : "Load from"); - er = kvm_mips_emulate_inst(cause, opc, run, vcpu); - if (er == EMULATE_FAIL) { - kvm_err("Emulate %s MMIO space failed\n", - store ? "Store to" : "Load from"); - run->exit_reason = KVM_EXIT_INTERNAL_ERROR; - ret = RESUME_HOST; - } else { - run->exit_reason = KVM_EXIT_MMIO; - ret = RESUME_HOST; - } + ret = kvm_mips_bad_access(cause, opc, run, vcpu, store); } else { kvm_err("Illegal TLB %s fault address , cause %#x, PC: %p, BadVaddr: %#lx\n", store ? "ST" : "LD", cause, opc, badvaddr); @@ -219,21 +296,11 @@ static int kvm_trap_emul_handle_addr_err_st(struct kvm_vcpu *vcpu) u32 __user *opc = (u32 __user *) vcpu->arch.pc; unsigned long badvaddr = vcpu->arch.host_cp0_badvaddr; u32 cause = vcpu->arch.host_cp0_cause; - enum emulation_result er = EMULATE_DONE; int ret = RESUME_GUEST; if (KVM_GUEST_KERNEL_MODE(vcpu) && (KSEGX(badvaddr) == CKSEG0 || KSEGX(badvaddr) == CKSEG1)) { - kvm_debug("Emulate Store to MMIO space\n"); - er = kvm_mips_emulate_inst(cause, opc, run, vcpu); - if (er == EMULATE_FAIL) { - kvm_err("Emulate Store to MMIO space failed\n"); - run->exit_reason = KVM_EXIT_INTERNAL_ERROR; - ret = RESUME_HOST; - } else { - run->exit_reason = KVM_EXIT_MMIO; - ret = RESUME_HOST; - } + ret = kvm_mips_bad_store(cause, opc, run, vcpu); } else { kvm_err("Address Error (STORE): cause %#x, PC: %p, BadVaddr: %#lx\n", cause, opc, badvaddr); @@ -249,26 +316,15 @@ static int kvm_trap_emul_handle_addr_err_ld(struct kvm_vcpu *vcpu) u32 __user *opc = (u32 __user *) vcpu->arch.pc; unsigned long badvaddr = vcpu->arch.host_cp0_badvaddr; u32 cause = vcpu->arch.host_cp0_cause; - enum emulation_result er = EMULATE_DONE; int ret = RESUME_GUEST; if (KSEGX(badvaddr) == CKSEG0 || KSEGX(badvaddr) == CKSEG1) { - kvm_debug("Emulate Load from MMIO space @ %#lx\n", badvaddr); - er = kvm_mips_emulate_inst(cause, opc, run, vcpu); - if (er == EMULATE_FAIL) { - kvm_err("Emulate Load from MMIO space failed\n"); - run->exit_reason = KVM_EXIT_INTERNAL_ERROR; - ret = RESUME_HOST; - } else { - run->exit_reason = KVM_EXIT_MMIO; - ret = RESUME_HOST; - } + ret = kvm_mips_bad_load(cause, opc, run, vcpu); } else { kvm_err("Address Error (LOAD): cause %#x, PC: %p, BadVaddr: %#lx\n", cause, opc, badvaddr); run->exit_reason = KVM_EXIT_INTERNAL_ERROR; ret = RESUME_HOST; - er = EMULATE_FAIL; } return ret; } @@ -428,16 +484,75 @@ static int kvm_trap_emul_handle_msa_disabled(struct kvm_vcpu *vcpu) return ret; } -static int kvm_trap_emul_vm_init(struct kvm *kvm) +static int kvm_trap_emul_vcpu_init(struct kvm_vcpu *vcpu) { + struct mm_struct *kern_mm = &vcpu->arch.guest_kernel_mm; + struct mm_struct *user_mm = &vcpu->arch.guest_user_mm; + + /* + * Allocate GVA -> HPA page tables. + * MIPS doesn't use the mm_struct pointer argument. + */ + kern_mm->pgd = pgd_alloc(kern_mm); + if (!kern_mm->pgd) + return -ENOMEM; + + user_mm->pgd = pgd_alloc(user_mm); + if (!user_mm->pgd) { + pgd_free(kern_mm, kern_mm->pgd); + return -ENOMEM; + } + return 0; } -static int kvm_trap_emul_vcpu_init(struct kvm_vcpu *vcpu) +static void kvm_mips_emul_free_gva_pt(pgd_t *pgd) { - vcpu->arch.kscratch_enabled = 0xfc; + /* Don't free host kernel page tables copied from init_mm.pgd */ + const unsigned long end = 0x80000000; + unsigned long pgd_va, pud_va, pmd_va; + pud_t *pud; + pmd_t *pmd; + pte_t *pte; + int i, j, k; + + for (i = 0; i < USER_PTRS_PER_PGD; i++) { + if (pgd_none(pgd[i])) + continue; + + pgd_va = (unsigned long)i << PGDIR_SHIFT; + if (pgd_va >= end) + break; + pud = pud_offset(pgd + i, 0); + for (j = 0; j < PTRS_PER_PUD; j++) { + if (pud_none(pud[j])) + continue; + + pud_va = pgd_va | ((unsigned long)j << PUD_SHIFT); + if (pud_va >= end) + break; + pmd = pmd_offset(pud + j, 0); + for (k = 0; k < PTRS_PER_PMD; k++) { + if (pmd_none(pmd[k])) + continue; + + pmd_va = pud_va | (k << PMD_SHIFT); + if (pmd_va >= end) + break; + pte = pte_offset(pmd + k, 0); + pte_free_kernel(NULL, pte); + } + pmd_free(NULL, pmd); + } + pud_free(NULL, pud); + } + pgd_free(NULL, pgd); +} - return 0; +static void kvm_trap_emul_vcpu_uninit(struct kvm_vcpu *vcpu) +{ + kvm_mips_emul_free_gva_pt(vcpu->arch.guest_kernel_mm.pgd); + kvm_mips_emul_free_gva_pt(vcpu->arch.guest_user_mm.pgd); } static int kvm_trap_emul_vcpu_setup(struct kvm_vcpu *vcpu) @@ -499,6 +614,9 @@ static int kvm_trap_emul_vcpu_setup(struct kvm_vcpu *vcpu) /* Set Wait IE/IXMT Ignore in Config7, IAR, AR */ kvm_write_c0_guest_config7(cop0, (MIPS_CONF7_WII) | (1 << 10)); + /* Status */ + kvm_write_c0_guest_status(cop0, ST0_BEV | ST0_ERL); + /* * Setup IntCtl defaults, compatibility mode for timer interrupts (HW5) */ @@ -508,17 +626,76 @@ static int kvm_trap_emul_vcpu_setup(struct kvm_vcpu *vcpu) kvm_write_c0_guest_ebase(cop0, KVM_GUEST_KSEG0 | (vcpu_id & MIPS_EBASE_CPUNUM)); + /* Put PC at guest reset vector */ + vcpu->arch.pc = KVM_GUEST_CKSEG1ADDR(0x1fc00000); + return 0; } +static void kvm_trap_emul_flush_shadow_all(struct kvm *kvm) +{ + /* Flush GVA page tables and invalidate GVA ASIDs on all VCPUs */ + kvm_flush_remote_tlbs(kvm); +} + +static void kvm_trap_emul_flush_shadow_memslot(struct kvm *kvm, + const struct kvm_memory_slot *slot) +{ + kvm_trap_emul_flush_shadow_all(kvm); +} + +static u64 kvm_trap_emul_get_one_regs[] = { + KVM_REG_MIPS_CP0_INDEX, + KVM_REG_MIPS_CP0_ENTRYLO0, + KVM_REG_MIPS_CP0_ENTRYLO1, + KVM_REG_MIPS_CP0_CONTEXT, + KVM_REG_MIPS_CP0_USERLOCAL, + KVM_REG_MIPS_CP0_PAGEMASK, + KVM_REG_MIPS_CP0_WIRED, + KVM_REG_MIPS_CP0_HWRENA, + KVM_REG_MIPS_CP0_BADVADDR, + KVM_REG_MIPS_CP0_COUNT, + KVM_REG_MIPS_CP0_ENTRYHI, + KVM_REG_MIPS_CP0_COMPARE, + KVM_REG_MIPS_CP0_STATUS, + KVM_REG_MIPS_CP0_INTCTL, + KVM_REG_MIPS_CP0_CAUSE, + KVM_REG_MIPS_CP0_EPC, + KVM_REG_MIPS_CP0_PRID, + KVM_REG_MIPS_CP0_EBASE, + KVM_REG_MIPS_CP0_CONFIG, + KVM_REG_MIPS_CP0_CONFIG1, + KVM_REG_MIPS_CP0_CONFIG2, + KVM_REG_MIPS_CP0_CONFIG3, + KVM_REG_MIPS_CP0_CONFIG4, + KVM_REG_MIPS_CP0_CONFIG5, + KVM_REG_MIPS_CP0_CONFIG7, + KVM_REG_MIPS_CP0_ERROREPC, + KVM_REG_MIPS_CP0_KSCRATCH1, + KVM_REG_MIPS_CP0_KSCRATCH2, + KVM_REG_MIPS_CP0_KSCRATCH3, + KVM_REG_MIPS_CP0_KSCRATCH4, + KVM_REG_MIPS_CP0_KSCRATCH5, + KVM_REG_MIPS_CP0_KSCRATCH6, + + KVM_REG_MIPS_COUNT_CTL, + KVM_REG_MIPS_COUNT_RESUME, + KVM_REG_MIPS_COUNT_HZ, +}; + static unsigned long kvm_trap_emul_num_regs(struct kvm_vcpu *vcpu) { - return 0; + return ARRAY_SIZE(kvm_trap_emul_get_one_regs); } static int kvm_trap_emul_copy_reg_indices(struct kvm_vcpu *vcpu, u64 __user *indices) { + if (copy_to_user(indices, kvm_trap_emul_get_one_regs, + sizeof(kvm_trap_emul_get_one_regs))) + return -EFAULT; + indices += ARRAY_SIZE(kvm_trap_emul_get_one_regs); + return 0; } @@ -526,7 +703,81 @@ static int kvm_trap_emul_get_one_reg(struct kvm_vcpu *vcpu, const struct kvm_one_reg *reg, s64 *v) { + struct mips_coproc *cop0 = vcpu->arch.cop0; + switch (reg->id) { + case KVM_REG_MIPS_CP0_INDEX: + *v = (long)kvm_read_c0_guest_index(cop0); + break; + case KVM_REG_MIPS_CP0_ENTRYLO0: + *v = kvm_read_c0_guest_entrylo0(cop0); + break; + case KVM_REG_MIPS_CP0_ENTRYLO1: + *v = kvm_read_c0_guest_entrylo1(cop0); + break; + case KVM_REG_MIPS_CP0_CONTEXT: + *v = (long)kvm_read_c0_guest_context(cop0); + break; + case KVM_REG_MIPS_CP0_USERLOCAL: + *v = (long)kvm_read_c0_guest_userlocal(cop0); + break; + case KVM_REG_MIPS_CP0_PAGEMASK: + *v = (long)kvm_read_c0_guest_pagemask(cop0); + break; + case KVM_REG_MIPS_CP0_WIRED: + *v = (long)kvm_read_c0_guest_wired(cop0); + break; + case KVM_REG_MIPS_CP0_HWRENA: + *v = (long)kvm_read_c0_guest_hwrena(cop0); + break; + case KVM_REG_MIPS_CP0_BADVADDR: + *v = (long)kvm_read_c0_guest_badvaddr(cop0); + break; + case KVM_REG_MIPS_CP0_ENTRYHI: + *v = (long)kvm_read_c0_guest_entryhi(cop0); + break; + case KVM_REG_MIPS_CP0_COMPARE: + *v = (long)kvm_read_c0_guest_compare(cop0); + break; + case KVM_REG_MIPS_CP0_STATUS: + *v = (long)kvm_read_c0_guest_status(cop0); + break; + case KVM_REG_MIPS_CP0_INTCTL: + *v = (long)kvm_read_c0_guest_intctl(cop0); + break; + case KVM_REG_MIPS_CP0_CAUSE: + *v = (long)kvm_read_c0_guest_cause(cop0); + break; + case KVM_REG_MIPS_CP0_EPC: + *v = (long)kvm_read_c0_guest_epc(cop0); + break; + case KVM_REG_MIPS_CP0_PRID: + *v = (long)kvm_read_c0_guest_prid(cop0); + break; + case KVM_REG_MIPS_CP0_EBASE: + *v = (long)kvm_read_c0_guest_ebase(cop0); + break; + case KVM_REG_MIPS_CP0_CONFIG: + *v = (long)kvm_read_c0_guest_config(cop0); + break; + case KVM_REG_MIPS_CP0_CONFIG1: + *v = (long)kvm_read_c0_guest_config1(cop0); + break; + case KVM_REG_MIPS_CP0_CONFIG2: + *v = (long)kvm_read_c0_guest_config2(cop0); + break; + case KVM_REG_MIPS_CP0_CONFIG3: + *v = (long)kvm_read_c0_guest_config3(cop0); + break; + case KVM_REG_MIPS_CP0_CONFIG4: + *v = (long)kvm_read_c0_guest_config4(cop0); + break; + case KVM_REG_MIPS_CP0_CONFIG5: + *v = (long)kvm_read_c0_guest_config5(cop0); + break; + case KVM_REG_MIPS_CP0_CONFIG7: + *v = (long)kvm_read_c0_guest_config7(cop0); + break; case KVM_REG_MIPS_CP0_COUNT: *v = kvm_mips_read_count(vcpu); break; @@ -539,6 +790,27 @@ static int kvm_trap_emul_get_one_reg(struct kvm_vcpu *vcpu, case KVM_REG_MIPS_COUNT_HZ: *v = vcpu->arch.count_hz; break; + case KVM_REG_MIPS_CP0_ERROREPC: + *v = (long)kvm_read_c0_guest_errorepc(cop0); + break; + case KVM_REG_MIPS_CP0_KSCRATCH1: + *v = (long)kvm_read_c0_guest_kscratch1(cop0); + break; + case KVM_REG_MIPS_CP0_KSCRATCH2: + *v = (long)kvm_read_c0_guest_kscratch2(cop0); + break; + case KVM_REG_MIPS_CP0_KSCRATCH3: + *v = (long)kvm_read_c0_guest_kscratch3(cop0); + break; + case KVM_REG_MIPS_CP0_KSCRATCH4: + *v = (long)kvm_read_c0_guest_kscratch4(cop0); + break; + case KVM_REG_MIPS_CP0_KSCRATCH5: + *v = (long)kvm_read_c0_guest_kscratch5(cop0); + break; + case KVM_REG_MIPS_CP0_KSCRATCH6: + *v = (long)kvm_read_c0_guest_kscratch6(cop0); + break; default: return -EINVAL; } @@ -554,6 +826,56 @@ static int kvm_trap_emul_set_one_reg(struct kvm_vcpu *vcpu, unsigned int cur, change; switch (reg->id) { + case KVM_REG_MIPS_CP0_INDEX: + kvm_write_c0_guest_index(cop0, v); + break; + case KVM_REG_MIPS_CP0_ENTRYLO0: + kvm_write_c0_guest_entrylo0(cop0, v); + break; + case KVM_REG_MIPS_CP0_ENTRYLO1: + kvm_write_c0_guest_entrylo1(cop0, v); + break; + case KVM_REG_MIPS_CP0_CONTEXT: + kvm_write_c0_guest_context(cop0, v); + break; + case KVM_REG_MIPS_CP0_USERLOCAL: + kvm_write_c0_guest_userlocal(cop0, v); + break; + case KVM_REG_MIPS_CP0_PAGEMASK: + kvm_write_c0_guest_pagemask(cop0, v); + break; + case KVM_REG_MIPS_CP0_WIRED: + kvm_write_c0_guest_wired(cop0, v); + break; + case KVM_REG_MIPS_CP0_HWRENA: + kvm_write_c0_guest_hwrena(cop0, v); + break; + case KVM_REG_MIPS_CP0_BADVADDR: + kvm_write_c0_guest_badvaddr(cop0, v); + break; + case KVM_REG_MIPS_CP0_ENTRYHI: + kvm_write_c0_guest_entryhi(cop0, v); + break; + case KVM_REG_MIPS_CP0_STATUS: + kvm_write_c0_guest_status(cop0, v); + break; + case KVM_REG_MIPS_CP0_INTCTL: + /* No VInt, so no VS, read-only for now */ + break; + case KVM_REG_MIPS_CP0_EPC: + kvm_write_c0_guest_epc(cop0, v); + break; + case KVM_REG_MIPS_CP0_PRID: + kvm_write_c0_guest_prid(cop0, v); + break; + case KVM_REG_MIPS_CP0_EBASE: + /* + * Allow core number to be written, but the exception base must + * remain in guest KSeg0. + */ + kvm_change_c0_guest_ebase(cop0, 0x1ffff000 | MIPS_EBASE_CPUNUM, + v); + break; case KVM_REG_MIPS_CP0_COUNT: kvm_mips_write_count(vcpu, v); break; @@ -618,6 +940,9 @@ static int kvm_trap_emul_set_one_reg(struct kvm_vcpu *vcpu, kvm_write_c0_guest_config5(cop0, v); } break; + case KVM_REG_MIPS_CP0_CONFIG7: + /* writes ignored */ + break; case KVM_REG_MIPS_COUNT_CTL: ret = kvm_mips_set_count_ctl(vcpu, v); break; @@ -627,24 +952,269 @@ static int kvm_trap_emul_set_one_reg(struct kvm_vcpu *vcpu, case KVM_REG_MIPS_COUNT_HZ: ret = kvm_mips_set_count_hz(vcpu, v); break; + case KVM_REG_MIPS_CP0_ERROREPC: + kvm_write_c0_guest_errorepc(cop0, v); + break; + case KVM_REG_MIPS_CP0_KSCRATCH1: + kvm_write_c0_guest_kscratch1(cop0, v); + break; + case KVM_REG_MIPS_CP0_KSCRATCH2: + kvm_write_c0_guest_kscratch2(cop0, v); + break; + case KVM_REG_MIPS_CP0_KSCRATCH3: + kvm_write_c0_guest_kscratch3(cop0, v); + break; + case KVM_REG_MIPS_CP0_KSCRATCH4: + kvm_write_c0_guest_kscratch4(cop0, v); + break; + case KVM_REG_MIPS_CP0_KSCRATCH5: + kvm_write_c0_guest_kscratch5(cop0, v); + break; + case KVM_REG_MIPS_CP0_KSCRATCH6: + kvm_write_c0_guest_kscratch6(cop0, v); + break; default: return -EINVAL; } return ret; } -static int kvm_trap_emul_vcpu_get_regs(struct kvm_vcpu *vcpu) +static int kvm_trap_emul_vcpu_load(struct kvm_vcpu *vcpu, int cpu) { - kvm_lose_fpu(vcpu); + struct mm_struct *kern_mm = &vcpu->arch.guest_kernel_mm; + struct mm_struct *user_mm = &vcpu->arch.guest_user_mm; + struct mm_struct *mm; + + /* + * Were we in guest context? If so, restore the appropriate ASID based + * on the mode of the Guest (Kernel/User). + */ + if (current->flags & PF_VCPU) { + mm = KVM_GUEST_KERNEL_MODE(vcpu) ? kern_mm : user_mm; + if ((cpu_context(cpu, mm) ^ asid_cache(cpu)) & + asid_version_mask(cpu)) + get_new_mmu_context(mm, cpu); + write_c0_entryhi(cpu_asid(cpu, mm)); + TLBMISS_HANDLER_SETUP_PGD(mm->pgd); + kvm_mips_suspend_mm(cpu); + ehb(); + } return 0; } -static int kvm_trap_emul_vcpu_set_regs(struct kvm_vcpu *vcpu) +static int kvm_trap_emul_vcpu_put(struct kvm_vcpu *vcpu, int cpu) { + kvm_lose_fpu(vcpu); + + if (current->flags & PF_VCPU) { + /* Restore normal Linux process memory map */ + if (((cpu_context(cpu, current->mm) ^ asid_cache(cpu)) & + asid_version_mask(cpu))) + get_new_mmu_context(current->mm, cpu); + write_c0_entryhi(cpu_asid(cpu, current->mm)); + TLBMISS_HANDLER_SETUP_PGD(current->mm->pgd); + kvm_mips_resume_mm(cpu); + ehb(); + } + return 0; } +static void kvm_trap_emul_check_requests(struct kvm_vcpu *vcpu, int cpu, + bool reload_asid) +{ + struct mm_struct *kern_mm = &vcpu->arch.guest_kernel_mm; + struct mm_struct *user_mm = &vcpu->arch.guest_user_mm; + struct mm_struct *mm; + int i; + + if (likely(!vcpu->requests)) + return; + + if (kvm_check_request(KVM_REQ_TLB_FLUSH, vcpu)) { + /* + * Both kernel & user GVA mappings must be invalidated. The + * caller is just about to check whether the ASID is stale + * anyway so no need to reload it here. + */ + kvm_mips_flush_gva_pt(kern_mm->pgd, KMF_GPA | KMF_KERN); + kvm_mips_flush_gva_pt(user_mm->pgd, KMF_GPA | KMF_USER); + for_each_possible_cpu(i) { + cpu_context(i, kern_mm) = 0; + cpu_context(i, user_mm) = 0; + } + + /* Generate new ASID for current mode */ + if (reload_asid) { + mm = KVM_GUEST_KERNEL_MODE(vcpu) ? kern_mm : user_mm; + get_new_mmu_context(mm, cpu); + htw_stop(); + write_c0_entryhi(cpu_asid(cpu, mm)); + TLBMISS_HANDLER_SETUP_PGD(mm->pgd); + htw_start(); + } + } +} + +/** + * kvm_trap_emul_gva_lockless_begin() - Begin lockless access to GVA space. + * @vcpu: VCPU pointer. + * + * Call before a GVA space access outside of guest mode, to ensure that + * asynchronous TLB flush requests are handled or delayed until completion of + * the GVA access (as indicated by a matching kvm_trap_emul_gva_lockless_end()). + * + * Should be called with IRQs already enabled. + */ +void kvm_trap_emul_gva_lockless_begin(struct kvm_vcpu *vcpu) +{ + /* We re-enable IRQs in kvm_trap_emul_gva_lockless_end() */ + WARN_ON_ONCE(irqs_disabled()); + + /* + * The caller is about to access the GVA space, so we set the mode to + * force TLB flush requests to send an IPI, and also disable IRQs to + * delay IPI handling until kvm_trap_emul_gva_lockless_end(). + */ + local_irq_disable(); + + /* + * Make sure the read of VCPU requests is not reordered ahead of the + * write to vcpu->mode, or we could miss a TLB flush request while + * the requester sees the VCPU as outside of guest mode and not needing + * an IPI. + */ + smp_store_mb(vcpu->mode, READING_SHADOW_PAGE_TABLES); + + /* + * If a TLB flush has been requested (potentially while + * OUTSIDE_GUEST_MODE and assumed immediately effective), perform it + * before accessing the GVA space, and be sure to reload the ASID if + * necessary as it'll be immediately used. + * + * TLB flush requests after this check will trigger an IPI due to the + * mode change above, which will be delayed due to IRQs disabled. + */ + kvm_trap_emul_check_requests(vcpu, smp_processor_id(), true); +} + +/** + * kvm_trap_emul_gva_lockless_end() - End lockless access to GVA space. + * @vcpu: VCPU pointer. + * + * Called after a GVA space access outside of guest mode. Should have a matching + * call to kvm_trap_emul_gva_lockless_begin(). + */ +void kvm_trap_emul_gva_lockless_end(struct kvm_vcpu *vcpu) +{ + /* + * Make sure the write to vcpu->mode is not reordered in front of GVA + * accesses, or a TLB flush requester may not think it necessary to send + * an IPI. + */ + smp_store_release(&vcpu->mode, OUTSIDE_GUEST_MODE); + + /* + * Now that the access to GVA space is complete, its safe for pending + * TLB flush request IPIs to be handled (which indicates completion). + */ + local_irq_enable(); +} + +static void kvm_trap_emul_vcpu_reenter(struct kvm_run *run, + struct kvm_vcpu *vcpu) +{ + struct mm_struct *kern_mm = &vcpu->arch.guest_kernel_mm; + struct mm_struct *user_mm = &vcpu->arch.guest_user_mm; + struct mm_struct *mm; + struct mips_coproc *cop0 = vcpu->arch.cop0; + int i, cpu = smp_processor_id(); + unsigned int gasid; + + /* + * No need to reload ASID, IRQs are disabled already so there's no rush, + * and we'll check if we need to regenerate below anyway before + * re-entering the guest. + */ + kvm_trap_emul_check_requests(vcpu, cpu, false); + + if (KVM_GUEST_KERNEL_MODE(vcpu)) { + mm = kern_mm; + } else { + mm = user_mm; + + /* + * Lazy host ASID regeneration / PT flush for guest user mode. + * If the guest ASID has changed since the last guest usermode + * execution, invalidate the stale TLB entries and flush GVA PT + * entries too. + */ + gasid = kvm_read_c0_guest_entryhi(cop0) & KVM_ENTRYHI_ASID; + if (gasid != vcpu->arch.last_user_gasid) { + kvm_mips_flush_gva_pt(user_mm->pgd, KMF_USER); + for_each_possible_cpu(i) + cpu_context(i, user_mm) = 0; + vcpu->arch.last_user_gasid = gasid; + } + } + + /* + * Check if ASID is stale. This may happen due to a TLB flush request or + * a lazy user MM invalidation. + */ + if ((cpu_context(cpu, mm) ^ asid_cache(cpu)) & + asid_version_mask(cpu)) + get_new_mmu_context(mm, cpu); +} + +static int kvm_trap_emul_vcpu_run(struct kvm_run *run, struct kvm_vcpu *vcpu) +{ + int cpu = smp_processor_id(); + int r; + + /* Check if we have any exceptions/interrupts pending */ + kvm_mips_deliver_interrupts(vcpu, + kvm_read_c0_guest_cause(vcpu->arch.cop0)); + + kvm_trap_emul_vcpu_reenter(run, vcpu); + + /* + * We use user accessors to access guest memory, but we don't want to + * invoke Linux page faulting. + */ + pagefault_disable(); + + /* Disable hardware page table walking while in guest */ + htw_stop(); + + /* + * While in guest context we're in the guest's address space, not the + * host process address space, so we need to be careful not to confuse + * e.g. cache management IPIs. + */ + kvm_mips_suspend_mm(cpu); + + r = vcpu->arch.vcpu_run(run, vcpu); + + /* We may have migrated while handling guest exits */ + cpu = smp_processor_id(); + + /* Restore normal Linux process memory map */ + if (((cpu_context(cpu, current->mm) ^ asid_cache(cpu)) & + asid_version_mask(cpu))) + get_new_mmu_context(current->mm, cpu); + write_c0_entryhi(cpu_asid(cpu, current->mm)); + TLBMISS_HANDLER_SETUP_PGD(current->mm->pgd); + kvm_mips_resume_mm(cpu); + + htw_start(); + + pagefault_enable(); + + return r; +} + static struct kvm_mips_callbacks kvm_trap_emul_callbacks = { /* exit handlers */ .handle_cop_unusable = kvm_trap_emul_handle_cop_unusable, @@ -661,9 +1231,11 @@ static struct kvm_mips_callbacks kvm_trap_emul_callbacks = { .handle_fpe = kvm_trap_emul_handle_fpe, .handle_msa_disabled = kvm_trap_emul_handle_msa_disabled, - .vm_init = kvm_trap_emul_vm_init, .vcpu_init = kvm_trap_emul_vcpu_init, + .vcpu_uninit = kvm_trap_emul_vcpu_uninit, .vcpu_setup = kvm_trap_emul_vcpu_setup, + .flush_shadow_all = kvm_trap_emul_flush_shadow_all, + .flush_shadow_memslot = kvm_trap_emul_flush_shadow_memslot, .gva_to_gpa = kvm_trap_emul_gva_to_gpa_cb, .queue_timer_int = kvm_mips_queue_timer_int_cb, .dequeue_timer_int = kvm_mips_dequeue_timer_int_cb, @@ -675,8 +1247,10 @@ static struct kvm_mips_callbacks kvm_trap_emul_callbacks = { .copy_reg_indices = kvm_trap_emul_copy_reg_indices, .get_one_reg = kvm_trap_emul_get_one_reg, .set_one_reg = kvm_trap_emul_set_one_reg, - .vcpu_get_regs = kvm_trap_emul_vcpu_get_regs, - .vcpu_set_regs = kvm_trap_emul_vcpu_set_regs, + .vcpu_load = kvm_trap_emul_vcpu_load, + .vcpu_put = kvm_trap_emul_vcpu_put, + .vcpu_run = kvm_trap_emul_vcpu_run, + .vcpu_reenter = kvm_trap_emul_vcpu_reenter, }; int kvm_mips_emulation_init(struct kvm_mips_callbacks **install_callbacks) diff --git a/arch/powerpc/include/asm/kvm_book3s_64.h b/arch/powerpc/include/asm/kvm_book3s_64.h index 0db010cc4e65..d9b48f5bb606 100644 --- a/arch/powerpc/include/asm/kvm_book3s_64.h +++ b/arch/powerpc/include/asm/kvm_book3s_64.h @@ -22,6 +22,10 @@ #include <asm/book3s/64/mmu-hash.h> +/* Power architecture requires HPT is at least 256kiB, at most 64TiB */ +#define PPC_MIN_HPT_ORDER 18 +#define PPC_MAX_HPT_ORDER 46 + #ifdef CONFIG_KVM_BOOK3S_PR_POSSIBLE static inline struct kvmppc_book3s_shadow_vcpu *svcpu_get(struct kvm_vcpu *vcpu) { @@ -356,6 +360,18 @@ extern void kvmppc_mmu_debugfs_init(struct kvm *kvm); extern void kvmhv_rm_send_ipi(int cpu); +static inline unsigned long kvmppc_hpt_npte(struct kvm_hpt_info *hpt) +{ + /* HPTEs are 2**4 bytes long */ + return 1UL << (hpt->order - 4); +} + +static inline unsigned long kvmppc_hpt_mask(struct kvm_hpt_info *hpt) +{ + /* 128 (2**7) bytes in each HPTEG */ + return (1UL << (hpt->order - 7)) - 1; +} + #endif /* CONFIG_KVM_BOOK3S_HV_POSSIBLE */ #endif /* __ASM_KVM_BOOK3S_64_H__ */ diff --git a/arch/powerpc/include/asm/kvm_host.h b/arch/powerpc/include/asm/kvm_host.h index b2dbeac3f450..7bba8f415627 100644 --- a/arch/powerpc/include/asm/kvm_host.h +++ b/arch/powerpc/include/asm/kvm_host.h @@ -241,12 +241,24 @@ struct kvm_arch_memory_slot { #endif /* CONFIG_KVM_BOOK3S_HV_POSSIBLE */ }; +struct kvm_hpt_info { + /* Host virtual (linear mapping) address of guest HPT */ + unsigned long virt; + /* Array of reverse mapping entries for each guest HPTE */ + struct revmap_entry *rev; + /* Guest HPT size is 2**(order) bytes */ + u32 order; + /* 1 if HPT allocated with CMA, 0 otherwise */ + int cma; +}; + +struct kvm_resize_hpt; + struct kvm_arch { unsigned int lpid; #ifdef CONFIG_KVM_BOOK3S_HV_POSSIBLE unsigned int tlb_sets; - unsigned long hpt_virt; - struct revmap_entry *revmap; + struct kvm_hpt_info hpt; atomic64_t mmio_update; unsigned int host_lpid; unsigned long host_lpcr; @@ -256,20 +268,17 @@ struct kvm_arch { unsigned long lpcr; unsigned long vrma_slb_v; int hpte_setup_done; - u32 hpt_order; atomic_t vcpus_running; u32 online_vcores; - unsigned long hpt_npte; - unsigned long hpt_mask; atomic_t hpte_mod_interest; cpumask_t need_tlb_flush; cpumask_t cpu_in_guest; - int hpt_cma_alloc; u8 radix; pgd_t *pgtable; u64 process_table; struct dentry *debugfs_dir; struct dentry *htab_dentry; + struct kvm_resize_hpt *resize_hpt; /* protected by kvm->lock */ #endif /* CONFIG_KVM_BOOK3S_HV_POSSIBLE */ #ifdef CONFIG_KVM_BOOK3S_PR_POSSIBLE struct mutex hpt_mutex; diff --git a/arch/powerpc/include/asm/kvm_ppc.h b/arch/powerpc/include/asm/kvm_ppc.h index 48c760f89590..dd11c4c8c56a 100644 --- a/arch/powerpc/include/asm/kvm_ppc.h +++ b/arch/powerpc/include/asm/kvm_ppc.h @@ -155,9 +155,10 @@ extern void kvmppc_core_destroy_mmu(struct kvm_vcpu *vcpu); extern int kvmppc_kvm_pv(struct kvm_vcpu *vcpu); extern void kvmppc_map_magic(struct kvm_vcpu *vcpu); -extern long kvmppc_alloc_hpt(struct kvm *kvm, u32 *htab_orderp); -extern long kvmppc_alloc_reset_hpt(struct kvm *kvm, u32 *htab_orderp); -extern void kvmppc_free_hpt(struct kvm *kvm); +extern int kvmppc_allocate_hpt(struct kvm_hpt_info *info, u32 order); +extern void kvmppc_set_hpt(struct kvm *kvm, struct kvm_hpt_info *info); +extern long kvmppc_alloc_reset_hpt(struct kvm *kvm, int order); +extern void kvmppc_free_hpt(struct kvm_hpt_info *info); extern long kvmppc_prepare_vrma(struct kvm *kvm, struct kvm_userspace_memory_region *mem); extern void kvmppc_map_vrma(struct kvm_vcpu *vcpu, @@ -186,8 +187,8 @@ extern long kvmppc_h_stuff_tce(struct kvm_vcpu *vcpu, unsigned long tce_value, unsigned long npages); extern long kvmppc_h_get_tce(struct kvm_vcpu *vcpu, unsigned long liobn, unsigned long ioba); -extern struct page *kvm_alloc_hpt(unsigned long nr_pages); -extern void kvm_release_hpt(struct page *page, unsigned long nr_pages); +extern struct page *kvm_alloc_hpt_cma(unsigned long nr_pages); +extern void kvm_free_hpt_cma(struct page *page, unsigned long nr_pages); extern int kvmppc_core_init_vm(struct kvm *kvm); extern void kvmppc_core_destroy_vm(struct kvm *kvm); extern void kvmppc_core_free_memslot(struct kvm *kvm, @@ -214,6 +215,10 @@ extern void kvmppc_bookehv_exit(void); extern int kvmppc_prepare_to_enter(struct kvm_vcpu *vcpu); extern int kvm_vm_ioctl_get_htab_fd(struct kvm *kvm, struct kvm_get_htab_fd *); +extern long kvm_vm_ioctl_resize_hpt_prepare(struct kvm *kvm, + struct kvm_ppc_resize_hpt *rhpt); +extern long kvm_vm_ioctl_resize_hpt_commit(struct kvm *kvm, + struct kvm_ppc_resize_hpt *rhpt); int kvm_vcpu_ioctl_interrupt(struct kvm_vcpu *vcpu, struct kvm_interrupt *irq); diff --git a/arch/powerpc/include/uapi/asm/kvm.h b/arch/powerpc/include/uapi/asm/kvm.h index cc0908b6c2a0..4edbe4bb0e8b 100644 --- a/arch/powerpc/include/uapi/asm/kvm.h +++ b/arch/powerpc/include/uapi/asm/kvm.h @@ -633,5 +633,7 @@ struct kvm_ppc_rmmu_info { #define KVM_XICS_LEVEL_SENSITIVE (1ULL << 40) #define KVM_XICS_MASKED (1ULL << 41) #define KVM_XICS_PENDING (1ULL << 42) +#define KVM_XICS_PRESENTED (1ULL << 43) +#define KVM_XICS_QUEUED (1ULL << 44) #endif /* __LINUX_KVM_POWERPC_H */ diff --git a/arch/powerpc/kvm/book3s_32_mmu.c b/arch/powerpc/kvm/book3s_32_mmu.c index a2eb6d354a57..1992676c7a94 100644 --- a/arch/powerpc/kvm/book3s_32_mmu.c +++ b/arch/powerpc/kvm/book3s_32_mmu.c @@ -224,7 +224,8 @@ static int kvmppc_mmu_book3s_32_xlate_pte(struct kvm_vcpu *vcpu, gva_t eaddr, ptem = kvmppc_mmu_book3s_32_get_ptem(sre, eaddr, primary); if(copy_from_user(pteg, (void __user *)ptegp, sizeof(pteg))) { - printk(KERN_ERR "KVM: Can't copy data from 0x%lx!\n", ptegp); + printk_ratelimited(KERN_ERR + "KVM: Can't copy data from 0x%lx!\n", ptegp); goto no_page_found; } diff --git a/arch/powerpc/kvm/book3s_64_mmu.c b/arch/powerpc/kvm/book3s_64_mmu.c index b9131aa1aedf..70153578131a 100644 --- a/arch/powerpc/kvm/book3s_64_mmu.c +++ b/arch/powerpc/kvm/book3s_64_mmu.c @@ -265,7 +265,8 @@ do_second: goto no_page_found; if(copy_from_user(pteg, (void __user *)ptegp, sizeof(pteg))) { - printk(KERN_ERR "KVM can't copy data from 0x%lx!\n", ptegp); + printk_ratelimited(KERN_ERR + "KVM: Can't copy data from 0x%lx!\n", ptegp); goto no_page_found; } diff --git a/arch/powerpc/kvm/book3s_64_mmu_hv.c b/arch/powerpc/kvm/book3s_64_mmu_hv.c index 9df3d940acec..f3158fb16de3 100644 --- a/arch/powerpc/kvm/book3s_64_mmu_hv.c +++ b/arch/powerpc/kvm/book3s_64_mmu_hv.c @@ -40,84 +40,101 @@ #include "trace_hv.h" -/* Power architecture requires HPT is at least 256kB */ -#define PPC_MIN_HPT_ORDER 18 +//#define DEBUG_RESIZE_HPT 1 + +#ifdef DEBUG_RESIZE_HPT +#define resize_hpt_debug(resize, ...) \ + do { \ + printk(KERN_DEBUG "RESIZE HPT %p: ", resize); \ + printk(__VA_ARGS__); \ + } while (0) +#else +#define resize_hpt_debug(resize, ...) \ + do { } while (0) +#endif static long kvmppc_virtmode_do_h_enter(struct kvm *kvm, unsigned long flags, long pte_index, unsigned long pteh, unsigned long ptel, unsigned long *pte_idx_ret); + +struct kvm_resize_hpt { + /* These fields read-only after init */ + struct kvm *kvm; + struct work_struct work; + u32 order; + + /* These fields protected by kvm->lock */ + int error; + bool prepare_done; + + /* Private to the work thread, until prepare_done is true, + * then protected by kvm->resize_hpt_sem */ + struct kvm_hpt_info hpt; +}; + static void kvmppc_rmap_reset(struct kvm *kvm); -long kvmppc_alloc_hpt(struct kvm *kvm, u32 *htab_orderp) +int kvmppc_allocate_hpt(struct kvm_hpt_info *info, u32 order) { unsigned long hpt = 0; - struct revmap_entry *rev; + int cma = 0; struct page *page = NULL; - long order = KVM_DEFAULT_HPT_ORDER; + struct revmap_entry *rev; + unsigned long npte; - if (htab_orderp) { - order = *htab_orderp; - if (order < PPC_MIN_HPT_ORDER) - order = PPC_MIN_HPT_ORDER; - } + if ((order < PPC_MIN_HPT_ORDER) || (order > PPC_MAX_HPT_ORDER)) + return -EINVAL; - kvm->arch.hpt_cma_alloc = 0; - page = kvm_alloc_hpt(1ul << (order - PAGE_SHIFT)); + page = kvm_alloc_hpt_cma(1ul << (order - PAGE_SHIFT)); if (page) { hpt = (unsigned long)pfn_to_kaddr(page_to_pfn(page)); memset((void *)hpt, 0, (1ul << order)); - kvm->arch.hpt_cma_alloc = 1; + cma = 1; } - /* Lastly try successively smaller sizes from the page allocator */ - /* Only do this if userspace didn't specify a size via ioctl */ - while (!hpt && order > PPC_MIN_HPT_ORDER && !htab_orderp) { - hpt = __get_free_pages(GFP_KERNEL|__GFP_ZERO|__GFP_REPEAT| - __GFP_NOWARN, order - PAGE_SHIFT); - if (!hpt) - --order; - } + if (!hpt) + hpt = __get_free_pages(GFP_KERNEL|__GFP_ZERO|__GFP_REPEAT + |__GFP_NOWARN, order - PAGE_SHIFT); if (!hpt) return -ENOMEM; - kvm->arch.hpt_virt = hpt; - kvm->arch.hpt_order = order; /* HPTEs are 2**4 bytes long */ - kvm->arch.hpt_npte = 1ul << (order - 4); - /* 128 (2**7) bytes in each HPTEG */ - kvm->arch.hpt_mask = (1ul << (order - 7)) - 1; - - atomic64_set(&kvm->arch.mmio_update, 0); + npte = 1ul << (order - 4); /* Allocate reverse map array */ - rev = vmalloc(sizeof(struct revmap_entry) * kvm->arch.hpt_npte); + rev = vmalloc(sizeof(struct revmap_entry) * npte); if (!rev) { - pr_err("kvmppc_alloc_hpt: Couldn't alloc reverse map array\n"); - goto out_freehpt; + pr_err("kvmppc_allocate_hpt: Couldn't alloc reverse map array\n"); + if (cma) + kvm_free_hpt_cma(page, 1 << (order - PAGE_SHIFT)); + else + free_pages(hpt, order - PAGE_SHIFT); + return -ENOMEM; } - kvm->arch.revmap = rev; - kvm->arch.sdr1 = __pa(hpt) | (order - 18); - pr_info("KVM guest htab at %lx (order %ld), LPID %x\n", - hpt, order, kvm->arch.lpid); + info->order = order; + info->virt = hpt; + info->cma = cma; + info->rev = rev; - if (htab_orderp) - *htab_orderp = order; return 0; +} - out_freehpt: - if (kvm->arch.hpt_cma_alloc) - kvm_release_hpt(page, 1 << (order - PAGE_SHIFT)); - else - free_pages(hpt, order - PAGE_SHIFT); - return -ENOMEM; +void kvmppc_set_hpt(struct kvm *kvm, struct kvm_hpt_info *info) +{ + atomic64_set(&kvm->arch.mmio_update, 0); + kvm->arch.hpt = *info; + kvm->arch.sdr1 = __pa(info->virt) | (info->order - 18); + + pr_debug("KVM guest htab at %lx (order %ld), LPID %x\n", + info->virt, (long)info->order, kvm->arch.lpid); } -long kvmppc_alloc_reset_hpt(struct kvm *kvm, u32 *htab_orderp) +long kvmppc_alloc_reset_hpt(struct kvm *kvm, int order) { long err = -EBUSY; - long order; + struct kvm_hpt_info info; if (kvm_is_radix(kvm)) return -EINVAL; @@ -132,36 +149,44 @@ long kvmppc_alloc_reset_hpt(struct kvm *kvm, u32 *htab_orderp) goto out; } } - if (kvm->arch.hpt_virt) { - order = kvm->arch.hpt_order; + if (kvm->arch.hpt.order == order) { + /* We already have a suitable HPT */ + /* Set the entire HPT to 0, i.e. invalid HPTEs */ - memset((void *)kvm->arch.hpt_virt, 0, 1ul << order); + memset((void *)kvm->arch.hpt.virt, 0, 1ul << order); /* * Reset all the reverse-mapping chains for all memslots */ kvmppc_rmap_reset(kvm); /* Ensure that each vcpu will flush its TLB on next entry. */ cpumask_setall(&kvm->arch.need_tlb_flush); - *htab_orderp = order; err = 0; - } else { - err = kvmppc_alloc_hpt(kvm, htab_orderp); - order = *htab_orderp; + goto out; } - out: + + if (kvm->arch.hpt.virt) + kvmppc_free_hpt(&kvm->arch.hpt); + + err = kvmppc_allocate_hpt(&info, order); + if (err < 0) + goto out; + kvmppc_set_hpt(kvm, &info); + +out: mutex_unlock(&kvm->lock); return err; } -void kvmppc_free_hpt(struct kvm *kvm) +void kvmppc_free_hpt(struct kvm_hpt_info *info) { - vfree(kvm->arch.revmap); - if (kvm->arch.hpt_cma_alloc) - kvm_release_hpt(virt_to_page(kvm->arch.hpt_virt), - 1 << (kvm->arch.hpt_order - PAGE_SHIFT)); - else if (kvm->arch.hpt_virt) - free_pages(kvm->arch.hpt_virt, - kvm->arch.hpt_order - PAGE_SHIFT); + vfree(info->rev); + if (info->cma) + kvm_free_hpt_cma(virt_to_page(info->virt), + 1 << (info->order - PAGE_SHIFT)); + else if (info->virt) + free_pages(info->virt, info->order - PAGE_SHIFT); + info->virt = 0; + info->order = 0; } /* Bits in first HPTE dword for pagesize 4k, 64k or 16M */ @@ -196,8 +221,8 @@ void kvmppc_map_vrma(struct kvm_vcpu *vcpu, struct kvm_memory_slot *memslot, if (npages > 1ul << (40 - porder)) npages = 1ul << (40 - porder); /* Can't use more than 1 HPTE per HPTEG */ - if (npages > kvm->arch.hpt_mask + 1) - npages = kvm->arch.hpt_mask + 1; + if (npages > kvmppc_hpt_mask(&kvm->arch.hpt) + 1) + npages = kvmppc_hpt_mask(&kvm->arch.hpt) + 1; hp0 = HPTE_V_1TB_SEG | (VRMA_VSID << (40 - 16)) | HPTE_V_BOLTED | hpte0_pgsize_encoding(psize); @@ -207,7 +232,8 @@ void kvmppc_map_vrma(struct kvm_vcpu *vcpu, struct kvm_memory_slot *memslot, for (i = 0; i < npages; ++i) { addr = i << porder; /* can't use hpt_hash since va > 64 bits */ - hash = (i ^ (VRMA_VSID ^ (VRMA_VSID << 25))) & kvm->arch.hpt_mask; + hash = (i ^ (VRMA_VSID ^ (VRMA_VSID << 25))) + & kvmppc_hpt_mask(&kvm->arch.hpt); /* * We assume that the hash table is empty and no * vcpus are using it at this stage. Since we create @@ -340,11 +366,11 @@ static int kvmppc_mmu_book3s_64_hv_xlate(struct kvm_vcpu *vcpu, gva_t eaddr, preempt_enable(); return -ENOENT; } - hptep = (__be64 *)(kvm->arch.hpt_virt + (index << 4)); + hptep = (__be64 *)(kvm->arch.hpt.virt + (index << 4)); v = orig_v = be64_to_cpu(hptep[0]) & ~HPTE_V_HVLOCK; if (cpu_has_feature(CPU_FTR_ARCH_300)) v = hpte_new_to_old_v(v, be64_to_cpu(hptep[1])); - gr = kvm->arch.revmap[index].guest_rpte; + gr = kvm->arch.hpt.rev[index].guest_rpte; unlock_hpte(hptep, orig_v); preempt_enable(); @@ -485,8 +511,8 @@ int kvmppc_book3s_hv_page_fault(struct kvm_run *run, struct kvm_vcpu *vcpu, } } index = vcpu->arch.pgfault_index; - hptep = (__be64 *)(kvm->arch.hpt_virt + (index << 4)); - rev = &kvm->arch.revmap[index]; + hptep = (__be64 *)(kvm->arch.hpt.virt + (index << 4)); + rev = &kvm->arch.hpt.rev[index]; preempt_disable(); while (!try_lock_hpte(hptep, HPTE_V_HVLOCK)) cpu_relax(); @@ -745,13 +771,53 @@ static int kvm_handle_hva(struct kvm *kvm, unsigned long hva, return kvm_handle_hva_range(kvm, hva, hva + 1, handler); } +/* Must be called with both HPTE and rmap locked */ +static void kvmppc_unmap_hpte(struct kvm *kvm, unsigned long i, + unsigned long *rmapp, unsigned long gfn) +{ + __be64 *hptep = (__be64 *) (kvm->arch.hpt.virt + (i << 4)); + struct revmap_entry *rev = kvm->arch.hpt.rev; + unsigned long j, h; + unsigned long ptel, psize, rcbits; + + j = rev[i].forw; + if (j == i) { + /* chain is now empty */ + *rmapp &= ~(KVMPPC_RMAP_PRESENT | KVMPPC_RMAP_INDEX); + } else { + /* remove i from chain */ + h = rev[i].back; + rev[h].forw = j; + rev[j].back = h; + rev[i].forw = rev[i].back = i; + *rmapp = (*rmapp & ~KVMPPC_RMAP_INDEX) | j; + } + + /* Now check and modify the HPTE */ + ptel = rev[i].guest_rpte; + psize = hpte_page_size(be64_to_cpu(hptep[0]), ptel); + if ((be64_to_cpu(hptep[0]) & HPTE_V_VALID) && + hpte_rpn(ptel, psize) == gfn) { + hptep[0] |= cpu_to_be64(HPTE_V_ABSENT); + kvmppc_invalidate_hpte(kvm, hptep, i); + hptep[1] &= ~cpu_to_be64(HPTE_R_KEY_HI | HPTE_R_KEY_LO); + /* Harvest R and C */ + rcbits = be64_to_cpu(hptep[1]) & (HPTE_R_R | HPTE_R_C); + *rmapp |= rcbits << KVMPPC_RMAP_RC_SHIFT; + if (rcbits & HPTE_R_C) + kvmppc_update_rmap_change(rmapp, psize); + if (rcbits & ~rev[i].guest_rpte) { + rev[i].guest_rpte = ptel | rcbits; + note_hpte_modification(kvm, &rev[i]); + } + } +} + static int kvm_unmap_rmapp(struct kvm *kvm, struct kvm_memory_slot *memslot, unsigned long gfn) { - struct revmap_entry *rev = kvm->arch.revmap; - unsigned long h, i, j; + unsigned long i; __be64 *hptep; - unsigned long ptel, psize, rcbits; unsigned long *rmapp; rmapp = &memslot->arch.rmap[gfn - memslot->base_gfn]; @@ -768,7 +834,7 @@ static int kvm_unmap_rmapp(struct kvm *kvm, struct kvm_memory_slot *memslot, * rmap chain lock. */ i = *rmapp & KVMPPC_RMAP_INDEX; - hptep = (__be64 *) (kvm->arch.hpt_virt + (i << 4)); + hptep = (__be64 *) (kvm->arch.hpt.virt + (i << 4)); if (!try_lock_hpte(hptep, HPTE_V_HVLOCK)) { /* unlock rmap before spinning on the HPTE lock */ unlock_rmap(rmapp); @@ -776,37 +842,8 @@ static int kvm_unmap_rmapp(struct kvm *kvm, struct kvm_memory_slot *memslot, cpu_relax(); continue; } - j = rev[i].forw; - if (j == i) { - /* chain is now empty */ - *rmapp &= ~(KVMPPC_RMAP_PRESENT | KVMPPC_RMAP_INDEX); - } else { - /* remove i from chain */ - h = rev[i].back; - rev[h].forw = j; - rev[j].back = h; - rev[i].forw = rev[i].back = i; - *rmapp = (*rmapp & ~KVMPPC_RMAP_INDEX) | j; - } - /* Now check and modify the HPTE */ - ptel = rev[i].guest_rpte; - psize = hpte_page_size(be64_to_cpu(hptep[0]), ptel); - if ((be64_to_cpu(hptep[0]) & HPTE_V_VALID) && - hpte_rpn(ptel, psize) == gfn) { - hptep[0] |= cpu_to_be64(HPTE_V_ABSENT); - kvmppc_invalidate_hpte(kvm, hptep, i); - hptep[1] &= ~cpu_to_be64(HPTE_R_KEY_HI | HPTE_R_KEY_LO); - /* Harvest R and C */ - rcbits = be64_to_cpu(hptep[1]) & (HPTE_R_R | HPTE_R_C); - *rmapp |= rcbits << KVMPPC_RMAP_RC_SHIFT; - if (rcbits & HPTE_R_C) - kvmppc_update_rmap_change(rmapp, psize); - if (rcbits & ~rev[i].guest_rpte) { - rev[i].guest_rpte = ptel | rcbits; - note_hpte_modification(kvm, &rev[i]); - } - } + kvmppc_unmap_hpte(kvm, i, rmapp, gfn); unlock_rmap(rmapp); __unlock_hpte(hptep, be64_to_cpu(hptep[0])); } @@ -860,7 +897,7 @@ void kvmppc_core_flush_memslot_hv(struct kvm *kvm, static int kvm_age_rmapp(struct kvm *kvm, struct kvm_memory_slot *memslot, unsigned long gfn) { - struct revmap_entry *rev = kvm->arch.revmap; + struct revmap_entry *rev = kvm->arch.hpt.rev; unsigned long head, i, j; __be64 *hptep; int ret = 0; @@ -880,7 +917,7 @@ static int kvm_age_rmapp(struct kvm *kvm, struct kvm_memory_slot *memslot, i = head = *rmapp & KVMPPC_RMAP_INDEX; do { - hptep = (__be64 *) (kvm->arch.hpt_virt + (i << 4)); + hptep = (__be64 *) (kvm->arch.hpt.virt + (i << 4)); j = rev[i].forw; /* If this HPTE isn't referenced, ignore it */ @@ -923,7 +960,7 @@ int kvm_age_hva_hv(struct kvm *kvm, unsigned long start, unsigned long end) static int kvm_test_age_rmapp(struct kvm *kvm, struct kvm_memory_slot *memslot, unsigned long gfn) { - struct revmap_entry *rev = kvm->arch.revmap; + struct revmap_entry *rev = kvm->arch.hpt.rev; unsigned long head, i, j; unsigned long *hp; int ret = 1; @@ -940,7 +977,7 @@ static int kvm_test_age_rmapp(struct kvm *kvm, struct kvm_memory_slot *memslot, if (*rmapp & KVMPPC_RMAP_PRESENT) { i = head = *rmapp & KVMPPC_RMAP_INDEX; do { - hp = (unsigned long *)(kvm->arch.hpt_virt + (i << 4)); + hp = (unsigned long *)(kvm->arch.hpt.virt + (i << 4)); j = rev[i].forw; if (be64_to_cpu(hp[1]) & HPTE_R_R) goto out; @@ -980,7 +1017,7 @@ static int vcpus_running(struct kvm *kvm) */ static int kvm_test_clear_dirty_npages(struct kvm *kvm, unsigned long *rmapp) { - struct revmap_entry *rev = kvm->arch.revmap; + struct revmap_entry *rev = kvm->arch.hpt.rev; unsigned long head, i, j; unsigned long n; unsigned long v, r; @@ -1005,7 +1042,7 @@ static int kvm_test_clear_dirty_npages(struct kvm *kvm, unsigned long *rmapp) i = head = *rmapp & KVMPPC_RMAP_INDEX; do { unsigned long hptep1; - hptep = (__be64 *) (kvm->arch.hpt_virt + (i << 4)); + hptep = (__be64 *) (kvm->arch.hpt.virt + (i << 4)); j = rev[i].forw; /* @@ -1172,6 +1209,363 @@ void kvmppc_unpin_guest_page(struct kvm *kvm, void *va, unsigned long gpa, } /* + * HPT resizing + */ +static int resize_hpt_allocate(struct kvm_resize_hpt *resize) +{ + int rc; + + rc = kvmppc_allocate_hpt(&resize->hpt, resize->order); + if (rc < 0) + return rc; + + resize_hpt_debug(resize, "resize_hpt_allocate(): HPT @ 0x%lx\n", + resize->hpt.virt); + + return 0; +} + +static unsigned long resize_hpt_rehash_hpte(struct kvm_resize_hpt *resize, + unsigned long idx) +{ + struct kvm *kvm = resize->kvm; + struct kvm_hpt_info *old = &kvm->arch.hpt; + struct kvm_hpt_info *new = &resize->hpt; + unsigned long old_hash_mask = (1ULL << (old->order - 7)) - 1; + unsigned long new_hash_mask = (1ULL << (new->order - 7)) - 1; + __be64 *hptep, *new_hptep; + unsigned long vpte, rpte, guest_rpte; + int ret; + struct revmap_entry *rev; + unsigned long apsize, psize, avpn, pteg, hash; + unsigned long new_idx, new_pteg, replace_vpte; + + hptep = (__be64 *)(old->virt + (idx << 4)); + + /* Guest is stopped, so new HPTEs can't be added or faulted + * in, only unmapped or altered by host actions. So, it's + * safe to check this before we take the HPTE lock */ + vpte = be64_to_cpu(hptep[0]); + if (!(vpte & HPTE_V_VALID) && !(vpte & HPTE_V_ABSENT)) + return 0; /* nothing to do */ + + while (!try_lock_hpte(hptep, HPTE_V_HVLOCK)) + cpu_relax(); + + vpte = be64_to_cpu(hptep[0]); + + ret = 0; + if (!(vpte & HPTE_V_VALID) && !(vpte & HPTE_V_ABSENT)) + /* Nothing to do */ + goto out; + + /* Unmap */ + rev = &old->rev[idx]; + guest_rpte = rev->guest_rpte; + + ret = -EIO; + apsize = hpte_page_size(vpte, guest_rpte); + if (!apsize) + goto out; + + if (vpte & HPTE_V_VALID) { + unsigned long gfn = hpte_rpn(guest_rpte, apsize); + int srcu_idx = srcu_read_lock(&kvm->srcu); + struct kvm_memory_slot *memslot = + __gfn_to_memslot(kvm_memslots(kvm), gfn); + + if (memslot) { + unsigned long *rmapp; + rmapp = &memslot->arch.rmap[gfn - memslot->base_gfn]; + + lock_rmap(rmapp); + kvmppc_unmap_hpte(kvm, idx, rmapp, gfn); + unlock_rmap(rmapp); + } + + srcu_read_unlock(&kvm->srcu, srcu_idx); + } + + /* Reload PTE after unmap */ + vpte = be64_to_cpu(hptep[0]); + + BUG_ON(vpte & HPTE_V_VALID); + BUG_ON(!(vpte & HPTE_V_ABSENT)); + + ret = 0; + if (!(vpte & HPTE_V_BOLTED)) + goto out; + + rpte = be64_to_cpu(hptep[1]); + psize = hpte_base_page_size(vpte, rpte); + avpn = HPTE_V_AVPN_VAL(vpte) & ~((psize - 1) >> 23); + pteg = idx / HPTES_PER_GROUP; + if (vpte & HPTE_V_SECONDARY) + pteg = ~pteg; + + if (!(vpte & HPTE_V_1TB_SEG)) { + unsigned long offset, vsid; + + /* We only have 28 - 23 bits of offset in avpn */ + offset = (avpn & 0x1f) << 23; + vsid = avpn >> 5; + /* We can find more bits from the pteg value */ + if (psize < (1ULL << 23)) + offset |= ((vsid ^ pteg) & old_hash_mask) * psize; + + hash = vsid ^ (offset / psize); + } else { + unsigned long offset, vsid; + + /* We only have 40 - 23 bits of seg_off in avpn */ + offset = (avpn & 0x1ffff) << 23; + vsid = avpn >> 17; + if (psize < (1ULL << 23)) + offset |= ((vsid ^ (vsid << 25) ^ pteg) & old_hash_mask) * psize; + + hash = vsid ^ (vsid << 25) ^ (offset / psize); + } + + new_pteg = hash & new_hash_mask; + if (vpte & HPTE_V_SECONDARY) { + BUG_ON(~pteg != (hash & old_hash_mask)); + new_pteg = ~new_pteg; + } else { + BUG_ON(pteg != (hash & old_hash_mask)); + } + + new_idx = new_pteg * HPTES_PER_GROUP + (idx % HPTES_PER_GROUP); + new_hptep = (__be64 *)(new->virt + (new_idx << 4)); + + replace_vpte = be64_to_cpu(new_hptep[0]); + + if (replace_vpte & (HPTE_V_VALID | HPTE_V_ABSENT)) { + BUG_ON(new->order >= old->order); + + if (replace_vpte & HPTE_V_BOLTED) { + if (vpte & HPTE_V_BOLTED) + /* Bolted collision, nothing we can do */ + ret = -ENOSPC; + /* Discard the new HPTE */ + goto out; + } + + /* Discard the previous HPTE */ + } + + new_hptep[1] = cpu_to_be64(rpte); + new->rev[new_idx].guest_rpte = guest_rpte; + /* No need for a barrier, since new HPT isn't active */ + new_hptep[0] = cpu_to_be64(vpte); + unlock_hpte(new_hptep, vpte); + +out: + unlock_hpte(hptep, vpte); + return ret; +} + +static int resize_hpt_rehash(struct kvm_resize_hpt *resize) +{ + struct kvm *kvm = resize->kvm; + unsigned long i; + int rc; + + /* + * resize_hpt_rehash_hpte() doesn't handle the new-format HPTEs + * that POWER9 uses, and could well hit a BUG_ON on POWER9. + */ + if (cpu_has_feature(CPU_FTR_ARCH_300)) + return -EIO; + for (i = 0; i < kvmppc_hpt_npte(&kvm->arch.hpt); i++) { + rc = resize_hpt_rehash_hpte(resize, i); + if (rc != 0) + return rc; + } + + return 0; +} + +static void resize_hpt_pivot(struct kvm_resize_hpt *resize) +{ + struct kvm *kvm = resize->kvm; + struct kvm_hpt_info hpt_tmp; + + /* Exchange the pending tables in the resize structure with + * the active tables */ + + resize_hpt_debug(resize, "resize_hpt_pivot()\n"); + + spin_lock(&kvm->mmu_lock); + asm volatile("ptesync" : : : "memory"); + + hpt_tmp = kvm->arch.hpt; + kvmppc_set_hpt(kvm, &resize->hpt); + resize->hpt = hpt_tmp; + + spin_unlock(&kvm->mmu_lock); + + synchronize_srcu_expedited(&kvm->srcu); + + resize_hpt_debug(resize, "resize_hpt_pivot() done\n"); +} + +static void resize_hpt_release(struct kvm *kvm, struct kvm_resize_hpt *resize) +{ + BUG_ON(kvm->arch.resize_hpt != resize); + + if (!resize) + return; + + if (resize->hpt.virt) + kvmppc_free_hpt(&resize->hpt); + + kvm->arch.resize_hpt = NULL; + kfree(resize); +} + +static void resize_hpt_prepare_work(struct work_struct *work) +{ + struct kvm_resize_hpt *resize = container_of(work, + struct kvm_resize_hpt, + work); + struct kvm *kvm = resize->kvm; + int err; + + resize_hpt_debug(resize, "resize_hpt_prepare_work(): order = %d\n", + resize->order); + + err = resize_hpt_allocate(resize); + + mutex_lock(&kvm->lock); + + resize->error = err; + resize->prepare_done = true; + + mutex_unlock(&kvm->lock); +} + +long kvm_vm_ioctl_resize_hpt_prepare(struct kvm *kvm, + struct kvm_ppc_resize_hpt *rhpt) +{ + unsigned long flags = rhpt->flags; + unsigned long shift = rhpt->shift; + struct kvm_resize_hpt *resize; + int ret; + + if (flags != 0) + return -EINVAL; + + if (shift && ((shift < 18) || (shift > 46))) + return -EINVAL; + + mutex_lock(&kvm->lock); + + resize = kvm->arch.resize_hpt; + + if (resize) { + if (resize->order == shift) { + /* Suitable resize in progress */ + if (resize->prepare_done) { + ret = resize->error; + if (ret != 0) + resize_hpt_release(kvm, resize); + } else { + ret = 100; /* estimated time in ms */ + } + + goto out; + } + + /* not suitable, cancel it */ + resize_hpt_release(kvm, resize); + } + + ret = 0; + if (!shift) + goto out; /* nothing to do */ + + /* start new resize */ + + resize = kzalloc(sizeof(*resize), GFP_KERNEL); + resize->order = shift; + resize->kvm = kvm; + INIT_WORK(&resize->work, resize_hpt_prepare_work); + kvm->arch.resize_hpt = resize; + + schedule_work(&resize->work); + + ret = 100; /* estimated time in ms */ + +out: + mutex_unlock(&kvm->lock); + return ret; +} + +static void resize_hpt_boot_vcpu(void *opaque) +{ + /* Nothing to do, just force a KVM exit */ +} + +long kvm_vm_ioctl_resize_hpt_commit(struct kvm *kvm, + struct kvm_ppc_resize_hpt *rhpt) +{ + unsigned long flags = rhpt->flags; + unsigned long shift = rhpt->shift; + struct kvm_resize_hpt *resize; + long ret; + + if (flags != 0) + return -EINVAL; + + if (shift && ((shift < 18) || (shift > 46))) + return -EINVAL; + + mutex_lock(&kvm->lock); + + resize = kvm->arch.resize_hpt; + + /* This shouldn't be possible */ + ret = -EIO; + if (WARN_ON(!kvm->arch.hpte_setup_done)) + goto out_no_hpt; + + /* Stop VCPUs from running while we mess with the HPT */ + kvm->arch.hpte_setup_done = 0; + smp_mb(); + + /* Boot all CPUs out of the guest so they re-read + * hpte_setup_done */ + on_each_cpu(resize_hpt_boot_vcpu, NULL, 1); + + ret = -ENXIO; + if (!resize || (resize->order != shift)) + goto out; + + ret = -EBUSY; + if (!resize->prepare_done) + goto out; + + ret = resize->error; + if (ret != 0) + goto out; + + ret = resize_hpt_rehash(resize); + if (ret != 0) + goto out; + + resize_hpt_pivot(resize); + +out: + /* Let VCPUs run again */ + kvm->arch.hpte_setup_done = 1; + smp_mb(); +out_no_hpt: + resize_hpt_release(kvm, resize); + mutex_unlock(&kvm->lock); + return ret; +} + +/* * Functions for reading and writing the hash table via reads and * writes on a file descriptor. * @@ -1311,8 +1705,8 @@ static ssize_t kvm_htab_read(struct file *file, char __user *buf, flags = ctx->flags; i = ctx->index; - hptp = (__be64 *)(kvm->arch.hpt_virt + (i * HPTE_SIZE)); - revp = kvm->arch.revmap + i; + hptp = (__be64 *)(kvm->arch.hpt.virt + (i * HPTE_SIZE)); + revp = kvm->arch.hpt.rev + i; lbuf = (unsigned long __user *)buf; nb = 0; @@ -1327,7 +1721,7 @@ static ssize_t kvm_htab_read(struct file *file, char __user *buf, /* Skip uninteresting entries, i.e. clean on not-first pass */ if (!first_pass) { - while (i < kvm->arch.hpt_npte && + while (i < kvmppc_hpt_npte(&kvm->arch.hpt) && !hpte_dirty(revp, hptp)) { ++i; hptp += 2; @@ -1337,7 +1731,7 @@ static ssize_t kvm_htab_read(struct file *file, char __user *buf, hdr.index = i; /* Grab a series of valid entries */ - while (i < kvm->arch.hpt_npte && + while (i < kvmppc_hpt_npte(&kvm->arch.hpt) && hdr.n_valid < 0xffff && nb + HPTE_SIZE < count && record_hpte(flags, hptp, hpte, revp, 1, first_pass)) { @@ -1353,7 +1747,7 @@ static ssize_t kvm_htab_read(struct file *file, char __user *buf, ++revp; } /* Now skip invalid entries while we can */ - while (i < kvm->arch.hpt_npte && + while (i < kvmppc_hpt_npte(&kvm->arch.hpt) && hdr.n_invalid < 0xffff && record_hpte(flags, hptp, hpte, revp, 0, first_pass)) { /* found an invalid entry */ @@ -1374,7 +1768,7 @@ static ssize_t kvm_htab_read(struct file *file, char __user *buf, } /* Check if we've wrapped around the hash table */ - if (i >= kvm->arch.hpt_npte) { + if (i >= kvmppc_hpt_npte(&kvm->arch.hpt)) { i = 0; ctx->first_pass = 0; break; @@ -1433,11 +1827,11 @@ static ssize_t kvm_htab_write(struct file *file, const char __user *buf, err = -EINVAL; i = hdr.index; - if (i >= kvm->arch.hpt_npte || - i + hdr.n_valid + hdr.n_invalid > kvm->arch.hpt_npte) + if (i >= kvmppc_hpt_npte(&kvm->arch.hpt) || + i + hdr.n_valid + hdr.n_invalid > kvmppc_hpt_npte(&kvm->arch.hpt)) break; - hptp = (__be64 *)(kvm->arch.hpt_virt + (i * HPTE_SIZE)); + hptp = (__be64 *)(kvm->arch.hpt.virt + (i * HPTE_SIZE)); lbuf = (unsigned long __user *)buf; for (j = 0; j < hdr.n_valid; ++j) { __be64 hpte_v; @@ -1624,8 +2018,9 @@ static ssize_t debugfs_htab_read(struct file *file, char __user *buf, kvm = p->kvm; i = p->hpt_index; - hptp = (__be64 *)(kvm->arch.hpt_virt + (i * HPTE_SIZE)); - for (; len != 0 && i < kvm->arch.hpt_npte; ++i, hptp += 2) { + hptp = (__be64 *)(kvm->arch.hpt.virt + (i * HPTE_SIZE)); + for (; len != 0 && i < kvmppc_hpt_npte(&kvm->arch.hpt); + ++i, hptp += 2) { if (!(be64_to_cpu(hptp[0]) & (HPTE_V_VALID | HPTE_V_ABSENT))) continue; @@ -1635,7 +2030,7 @@ static ssize_t debugfs_htab_read(struct file *file, char __user *buf, cpu_relax(); v = be64_to_cpu(hptp[0]) & ~HPTE_V_HVLOCK; hr = be64_to_cpu(hptp[1]); - gr = kvm->arch.revmap[i].guest_rpte; + gr = kvm->arch.hpt.rev[i].guest_rpte; unlock_hpte(hptp, v); preempt_enable(); diff --git a/arch/powerpc/kvm/book3s_64_vio.c b/arch/powerpc/kvm/book3s_64_vio.c index c379ff5a4438..491c5d8120f7 100644 --- a/arch/powerpc/kvm/book3s_64_vio.c +++ b/arch/powerpc/kvm/book3s_64_vio.c @@ -171,6 +171,7 @@ long kvm_vm_ioctl_create_spapr_tce(struct kvm *kvm, goto fail; } + ret = -ENOMEM; stt = kzalloc(sizeof(*stt) + npages * sizeof(struct page *), GFP_KERNEL); if (!stt) diff --git a/arch/powerpc/kvm/book3s_hv.c b/arch/powerpc/kvm/book3s_hv.c index e4a79679342e..1e107ece4e37 100644 --- a/arch/powerpc/kvm/book3s_hv.c +++ b/arch/powerpc/kvm/book3s_hv.c @@ -182,7 +182,8 @@ static void kvmppc_fast_vcpu_kick_hv(struct kvm_vcpu *vcpu) ++vcpu->stat.halt_wakeup; } - if (kvmppc_ipi_thread(vcpu->arch.thread_cpu)) + cpu = READ_ONCE(vcpu->arch.thread_cpu); + if (cpu >= 0 && kvmppc_ipi_thread(cpu)) return; /* CPU points to the first thread of the core */ @@ -773,12 +774,8 @@ int kvmppc_pseries_do_hcall(struct kvm_vcpu *vcpu) } tvcpu->arch.prodded = 1; smp_mb(); - if (vcpu->arch.ceded) { - if (swait_active(&vcpu->wq)) { - swake_up(&vcpu->wq); - vcpu->stat.halt_wakeup++; - } - } + if (tvcpu->arch.ceded) + kvmppc_fast_vcpu_kick_hv(tvcpu); break; case H_CONFER: target = kvmppc_get_gpr(vcpu, 4); @@ -2665,7 +2662,8 @@ static int kvmppc_vcore_check_block(struct kvmppc_vcore *vc) int i; for_each_runnable_thread(i, vcpu, vc) { - if (vcpu->arch.pending_exceptions || !vcpu->arch.ceded) + if (vcpu->arch.pending_exceptions || !vcpu->arch.ceded || + vcpu->arch.prodded) return 1; } @@ -2851,7 +2849,7 @@ static int kvmppc_run_vcpu(struct kvm_run *kvm_run, struct kvm_vcpu *vcpu) break; n_ceded = 0; for_each_runnable_thread(i, v, vc) { - if (!v->arch.pending_exceptions) + if (!v->arch.pending_exceptions && !v->arch.prodded) n_ceded += v->arch.ceded; else v->arch.ceded = 0; @@ -3199,12 +3197,23 @@ static int kvmppc_hv_setup_htab_rma(struct kvm_vcpu *vcpu) goto out; /* another vcpu beat us to it */ /* Allocate hashed page table (if not done already) and reset it */ - if (!kvm->arch.hpt_virt) { - err = kvmppc_alloc_hpt(kvm, NULL); - if (err) { + if (!kvm->arch.hpt.virt) { + int order = KVM_DEFAULT_HPT_ORDER; + struct kvm_hpt_info info; + + err = kvmppc_allocate_hpt(&info, order); + /* If we get here, it means userspace didn't specify a + * size explicitly. So, try successively smaller + * sizes if the default failed. */ + while ((err == -ENOMEM) && --order >= PPC_MIN_HPT_ORDER) + err = kvmppc_allocate_hpt(&info, order); + + if (err < 0) { pr_err("KVM: Couldn't alloc HPT\n"); goto out; } + + kvmppc_set_hpt(kvm, &info); } /* Look up the memslot for guest physical address 0 */ @@ -3413,6 +3422,9 @@ static int kvmppc_core_init_vm_hv(struct kvm *kvm) kvm->arch.lpcr = lpcr; + /* Initialization for future HPT resizes */ + kvm->arch.resize_hpt = NULL; + /* * Work out how many sets the TLB has, for the use of * the TLB invalidation loop in book3s_hv_rmhandlers.S. @@ -3469,7 +3481,7 @@ static void kvmppc_core_destroy_vm_hv(struct kvm *kvm) if (kvm_is_radix(kvm)) kvmppc_free_radix(kvm); else - kvmppc_free_hpt(kvm); + kvmppc_free_hpt(&kvm->arch.hpt); kvmppc_free_pimap(kvm); } @@ -3695,12 +3707,9 @@ static long kvm_arch_vm_ioctl_hv(struct file *filp, r = -EFAULT; if (get_user(htab_order, (u32 __user *)argp)) break; - r = kvmppc_alloc_reset_hpt(kvm, &htab_order); + r = kvmppc_alloc_reset_hpt(kvm, htab_order); if (r) break; - r = -EFAULT; - if (put_user(htab_order, (u32 __user *)argp)) - break; r = 0; break; } @@ -3715,6 +3724,28 @@ static long kvm_arch_vm_ioctl_hv(struct file *filp, break; } + case KVM_PPC_RESIZE_HPT_PREPARE: { + struct kvm_ppc_resize_hpt rhpt; + + r = -EFAULT; + if (copy_from_user(&rhpt, argp, sizeof(rhpt))) + break; + + r = kvm_vm_ioctl_resize_hpt_prepare(kvm, &rhpt); + break; + } + + case KVM_PPC_RESIZE_HPT_COMMIT: { + struct kvm_ppc_resize_hpt rhpt; + + r = -EFAULT; + if (copy_from_user(&rhpt, argp, sizeof(rhpt))) + break; + + r = kvm_vm_ioctl_resize_hpt_commit(kvm, &rhpt); + break; + } + default: r = -ENOTTY; } diff --git a/arch/powerpc/kvm/book3s_hv_builtin.c b/arch/powerpc/kvm/book3s_hv_builtin.c index 2f69fbc19bb0..c42a7e63b39e 100644 --- a/arch/powerpc/kvm/book3s_hv_builtin.c +++ b/arch/powerpc/kvm/book3s_hv_builtin.c @@ -52,19 +52,19 @@ static int __init early_parse_kvm_cma_resv(char *p) } early_param("kvm_cma_resv_ratio", early_parse_kvm_cma_resv); -struct page *kvm_alloc_hpt(unsigned long nr_pages) +struct page *kvm_alloc_hpt_cma(unsigned long nr_pages) { VM_BUG_ON(order_base_2(nr_pages) < KVM_CMA_CHUNK_ORDER - PAGE_SHIFT); return cma_alloc(kvm_cma, nr_pages, order_base_2(HPT_ALIGN_PAGES)); } -EXPORT_SYMBOL_GPL(kvm_alloc_hpt); +EXPORT_SYMBOL_GPL(kvm_alloc_hpt_cma); -void kvm_release_hpt(struct page *page, unsigned long nr_pages) +void kvm_free_hpt_cma(struct page *page, unsigned long nr_pages) { cma_release(kvm_cma, page, nr_pages); } -EXPORT_SYMBOL_GPL(kvm_release_hpt); +EXPORT_SYMBOL_GPL(kvm_free_hpt_cma); /** * kvm_cma_reserve() - reserve area for kvm hash pagetable diff --git a/arch/powerpc/kvm/book3s_hv_rm_mmu.c b/arch/powerpc/kvm/book3s_hv_rm_mmu.c index b095afcd4309..6fca970373ee 100644 --- a/arch/powerpc/kvm/book3s_hv_rm_mmu.c +++ b/arch/powerpc/kvm/book3s_hv_rm_mmu.c @@ -86,10 +86,10 @@ void kvmppc_add_revmap_chain(struct kvm *kvm, struct revmap_entry *rev, if (*rmap & KVMPPC_RMAP_PRESENT) { i = *rmap & KVMPPC_RMAP_INDEX; - head = &kvm->arch.revmap[i]; + head = &kvm->arch.hpt.rev[i]; if (realmode) head = real_vmalloc_addr(head); - tail = &kvm->arch.revmap[head->back]; + tail = &kvm->arch.hpt.rev[head->back]; if (realmode) tail = real_vmalloc_addr(tail); rev->forw = i; @@ -154,8 +154,8 @@ static void remove_revmap_chain(struct kvm *kvm, long pte_index, lock_rmap(rmap); head = *rmap & KVMPPC_RMAP_INDEX; - next = real_vmalloc_addr(&kvm->arch.revmap[rev->forw]); - prev = real_vmalloc_addr(&kvm->arch.revmap[rev->back]); + next = real_vmalloc_addr(&kvm->arch.hpt.rev[rev->forw]); + prev = real_vmalloc_addr(&kvm->arch.hpt.rev[rev->back]); next->back = rev->back; prev->forw = rev->forw; if (head == pte_index) { @@ -292,11 +292,11 @@ long kvmppc_do_h_enter(struct kvm *kvm, unsigned long flags, /* Find and lock the HPTEG slot to use */ do_insert: - if (pte_index >= kvm->arch.hpt_npte) + if (pte_index >= kvmppc_hpt_npte(&kvm->arch.hpt)) return H_PARAMETER; if (likely((flags & H_EXACT) == 0)) { pte_index &= ~7UL; - hpte = (__be64 *)(kvm->arch.hpt_virt + (pte_index << 4)); + hpte = (__be64 *)(kvm->arch.hpt.virt + (pte_index << 4)); for (i = 0; i < 8; ++i) { if ((be64_to_cpu(*hpte) & HPTE_V_VALID) == 0 && try_lock_hpte(hpte, HPTE_V_HVLOCK | HPTE_V_VALID | @@ -327,7 +327,7 @@ long kvmppc_do_h_enter(struct kvm *kvm, unsigned long flags, } pte_index += i; } else { - hpte = (__be64 *)(kvm->arch.hpt_virt + (pte_index << 4)); + hpte = (__be64 *)(kvm->arch.hpt.virt + (pte_index << 4)); if (!try_lock_hpte(hpte, HPTE_V_HVLOCK | HPTE_V_VALID | HPTE_V_ABSENT)) { /* Lock the slot and check again */ @@ -344,7 +344,7 @@ long kvmppc_do_h_enter(struct kvm *kvm, unsigned long flags, } /* Save away the guest's idea of the second HPTE dword */ - rev = &kvm->arch.revmap[pte_index]; + rev = &kvm->arch.hpt.rev[pte_index]; if (realmode) rev = real_vmalloc_addr(rev); if (rev) { @@ -469,9 +469,9 @@ long kvmppc_do_h_remove(struct kvm *kvm, unsigned long flags, if (kvm_is_radix(kvm)) return H_FUNCTION; - if (pte_index >= kvm->arch.hpt_npte) + if (pte_index >= kvmppc_hpt_npte(&kvm->arch.hpt)) return H_PARAMETER; - hpte = (__be64 *)(kvm->arch.hpt_virt + (pte_index << 4)); + hpte = (__be64 *)(kvm->arch.hpt.virt + (pte_index << 4)); while (!try_lock_hpte(hpte, HPTE_V_HVLOCK)) cpu_relax(); pte = orig_pte = be64_to_cpu(hpte[0]); @@ -487,7 +487,7 @@ long kvmppc_do_h_remove(struct kvm *kvm, unsigned long flags, return H_NOT_FOUND; } - rev = real_vmalloc_addr(&kvm->arch.revmap[pte_index]); + rev = real_vmalloc_addr(&kvm->arch.hpt.rev[pte_index]); v = pte & ~HPTE_V_HVLOCK; if (v & HPTE_V_VALID) { hpte[0] &= ~cpu_to_be64(HPTE_V_VALID); @@ -557,13 +557,13 @@ long kvmppc_h_bulk_remove(struct kvm_vcpu *vcpu) break; } if (req != 1 || flags == 3 || - pte_index >= kvm->arch.hpt_npte) { + pte_index >= kvmppc_hpt_npte(&kvm->arch.hpt)) { /* parameter error */ args[j] = ((0xa0 | flags) << 56) + pte_index; ret = H_PARAMETER; break; } - hp = (__be64 *) (kvm->arch.hpt_virt + (pte_index << 4)); + hp = (__be64 *) (kvm->arch.hpt.virt + (pte_index << 4)); /* to avoid deadlock, don't spin except for first */ if (!try_lock_hpte(hp, HPTE_V_HVLOCK)) { if (n) @@ -600,7 +600,7 @@ long kvmppc_h_bulk_remove(struct kvm_vcpu *vcpu) } args[j] = ((0x80 | flags) << 56) + pte_index; - rev = real_vmalloc_addr(&kvm->arch.revmap[pte_index]); + rev = real_vmalloc_addr(&kvm->arch.hpt.rev[pte_index]); note_hpte_modification(kvm, rev); if (!(hp0 & HPTE_V_VALID)) { @@ -657,10 +657,10 @@ long kvmppc_h_protect(struct kvm_vcpu *vcpu, unsigned long flags, if (kvm_is_radix(kvm)) return H_FUNCTION; - if (pte_index >= kvm->arch.hpt_npte) + if (pte_index >= kvmppc_hpt_npte(&kvm->arch.hpt)) return H_PARAMETER; - hpte = (__be64 *)(kvm->arch.hpt_virt + (pte_index << 4)); + hpte = (__be64 *)(kvm->arch.hpt.virt + (pte_index << 4)); while (!try_lock_hpte(hpte, HPTE_V_HVLOCK)) cpu_relax(); v = pte_v = be64_to_cpu(hpte[0]); @@ -680,7 +680,7 @@ long kvmppc_h_protect(struct kvm_vcpu *vcpu, unsigned long flags, /* Update guest view of 2nd HPTE dword */ mask = HPTE_R_PP0 | HPTE_R_PP | HPTE_R_N | HPTE_R_KEY_HI | HPTE_R_KEY_LO; - rev = real_vmalloc_addr(&kvm->arch.revmap[pte_index]); + rev = real_vmalloc_addr(&kvm->arch.hpt.rev[pte_index]); if (rev) { r = (rev->guest_rpte & ~mask) | bits; rev->guest_rpte = r; @@ -728,15 +728,15 @@ long kvmppc_h_read(struct kvm_vcpu *vcpu, unsigned long flags, if (kvm_is_radix(kvm)) return H_FUNCTION; - if (pte_index >= kvm->arch.hpt_npte) + if (pte_index >= kvmppc_hpt_npte(&kvm->arch.hpt)) return H_PARAMETER; if (flags & H_READ_4) { pte_index &= ~3; n = 4; } - rev = real_vmalloc_addr(&kvm->arch.revmap[pte_index]); + rev = real_vmalloc_addr(&kvm->arch.hpt.rev[pte_index]); for (i = 0; i < n; ++i, ++pte_index) { - hpte = (__be64 *)(kvm->arch.hpt_virt + (pte_index << 4)); + hpte = (__be64 *)(kvm->arch.hpt.virt + (pte_index << 4)); v = be64_to_cpu(hpte[0]) & ~HPTE_V_HVLOCK; r = be64_to_cpu(hpte[1]); if (cpu_has_feature(CPU_FTR_ARCH_300)) { @@ -769,11 +769,11 @@ long kvmppc_h_clear_ref(struct kvm_vcpu *vcpu, unsigned long flags, if (kvm_is_radix(kvm)) return H_FUNCTION; - if (pte_index >= kvm->arch.hpt_npte) + if (pte_index >= kvmppc_hpt_npte(&kvm->arch.hpt)) return H_PARAMETER; - rev = real_vmalloc_addr(&kvm->arch.revmap[pte_index]); - hpte = (__be64 *)(kvm->arch.hpt_virt + (pte_index << 4)); + rev = real_vmalloc_addr(&kvm->arch.hpt.rev[pte_index]); + hpte = (__be64 *)(kvm->arch.hpt.virt + (pte_index << 4)); while (!try_lock_hpte(hpte, HPTE_V_HVLOCK)) cpu_relax(); v = be64_to_cpu(hpte[0]); @@ -817,11 +817,11 @@ long kvmppc_h_clear_mod(struct kvm_vcpu *vcpu, unsigned long flags, if (kvm_is_radix(kvm)) return H_FUNCTION; - if (pte_index >= kvm->arch.hpt_npte) + if (pte_index >= kvmppc_hpt_npte(&kvm->arch.hpt)) return H_PARAMETER; - rev = real_vmalloc_addr(&kvm->arch.revmap[pte_index]); - hpte = (__be64 *)(kvm->arch.hpt_virt + (pte_index << 4)); + rev = real_vmalloc_addr(&kvm->arch.hpt.rev[pte_index]); + hpte = (__be64 *)(kvm->arch.hpt.virt + (pte_index << 4)); while (!try_lock_hpte(hpte, HPTE_V_HVLOCK)) cpu_relax(); v = be64_to_cpu(hpte[0]); @@ -970,7 +970,7 @@ long kvmppc_hv_find_lock_hpte(struct kvm *kvm, gva_t eaddr, unsigned long slb_v, somask = (1UL << 28) - 1; vsid = (slb_v & ~SLB_VSID_B) >> SLB_VSID_SHIFT; } - hash = (vsid ^ ((eaddr & somask) >> pshift)) & kvm->arch.hpt_mask; + hash = (vsid ^ ((eaddr & somask) >> pshift)) & kvmppc_hpt_mask(&kvm->arch.hpt); avpn = slb_v & ~(somask >> 16); /* also includes B */ avpn |= (eaddr & somask) >> 16; @@ -981,7 +981,7 @@ long kvmppc_hv_find_lock_hpte(struct kvm *kvm, gva_t eaddr, unsigned long slb_v, val |= avpn; for (;;) { - hpte = (__be64 *)(kvm->arch.hpt_virt + (hash << 7)); + hpte = (__be64 *)(kvm->arch.hpt.virt + (hash << 7)); for (i = 0; i < 16; i += 2) { /* Read the PTE racily */ @@ -1017,7 +1017,7 @@ long kvmppc_hv_find_lock_hpte(struct kvm *kvm, gva_t eaddr, unsigned long slb_v, if (val & HPTE_V_SECONDARY) break; val |= HPTE_V_SECONDARY; - hash = hash ^ kvm->arch.hpt_mask; + hash = hash ^ kvmppc_hpt_mask(&kvm->arch.hpt); } return -1; } @@ -1066,14 +1066,14 @@ long kvmppc_hpte_hv_fault(struct kvm_vcpu *vcpu, unsigned long addr, return status; /* there really was no HPTE */ return 0; /* for prot fault, HPTE disappeared */ } - hpte = (__be64 *)(kvm->arch.hpt_virt + (index << 4)); + hpte = (__be64 *)(kvm->arch.hpt.virt + (index << 4)); v = orig_v = be64_to_cpu(hpte[0]) & ~HPTE_V_HVLOCK; r = be64_to_cpu(hpte[1]); if (cpu_has_feature(CPU_FTR_ARCH_300)) { v = hpte_new_to_old_v(v, r); r = hpte_new_to_old_r(r); } - rev = real_vmalloc_addr(&kvm->arch.revmap[index]); + rev = real_vmalloc_addr(&kvm->arch.hpt.rev[index]); gr = rev->guest_rpte; unlock_hpte(hpte, orig_v); diff --git a/arch/powerpc/kvm/book3s_hv_rm_xics.c b/arch/powerpc/kvm/book3s_hv_rm_xics.c index 29f43ed6d5eb..e78542d99cd6 100644 --- a/arch/powerpc/kvm/book3s_hv_rm_xics.c +++ b/arch/powerpc/kvm/book3s_hv_rm_xics.c @@ -35,7 +35,7 @@ int kvm_irq_bypass = 1; EXPORT_SYMBOL(kvm_irq_bypass); static void icp_rm_deliver_irq(struct kvmppc_xics *xics, struct kvmppc_icp *icp, - u32 new_irq); + u32 new_irq, bool check_resend); static int xics_opal_set_server(unsigned int hw_irq, int server_cpu); /* -- ICS routines -- */ @@ -44,20 +44,12 @@ static void ics_rm_check_resend(struct kvmppc_xics *xics, { int i; - arch_spin_lock(&ics->lock); - for (i = 0; i < KVMPPC_XICS_IRQ_PER_ICS; i++) { struct ics_irq_state *state = &ics->irq_state[i]; - - if (!state->resend) - continue; - - arch_spin_unlock(&ics->lock); - icp_rm_deliver_irq(xics, icp, state->number); - arch_spin_lock(&ics->lock); + if (state->resend) + icp_rm_deliver_irq(xics, icp, state->number, true); } - arch_spin_unlock(&ics->lock); } /* -- ICP routines -- */ @@ -288,7 +280,7 @@ static bool icp_rm_try_to_deliver(struct kvmppc_icp *icp, u32 irq, u8 priority, } static void icp_rm_deliver_irq(struct kvmppc_xics *xics, struct kvmppc_icp *icp, - u32 new_irq) + u32 new_irq, bool check_resend) { struct ics_irq_state *state; struct kvmppc_ics *ics; @@ -333,6 +325,10 @@ static void icp_rm_deliver_irq(struct kvmppc_xics *xics, struct kvmppc_icp *icp, } } + if (check_resend) + if (!state->resend) + goto out; + /* Clear the resend bit of that interrupt */ state->resend = 0; @@ -378,7 +374,9 @@ static void icp_rm_deliver_irq(struct kvmppc_xics *xics, struct kvmppc_icp *icp, */ if (reject && reject != XICS_IPI) { arch_spin_unlock(&ics->lock); + icp->n_reject++; new_irq = reject; + check_resend = 0; goto again; } } else { @@ -386,10 +384,16 @@ static void icp_rm_deliver_irq(struct kvmppc_xics *xics, struct kvmppc_icp *icp, * We failed to deliver the interrupt we need to set the * resend map bit and mark the ICS state as needing a resend */ - set_bit(ics->icsid, icp->resend_map); state->resend = 1; /* + * Make sure when checking resend, we don't miss the resend + * if resend_map bit is seen and cleared. + */ + smp_wmb(); + set_bit(ics->icsid, icp->resend_map); + + /* * If the need_resend flag got cleared in the ICP some time * between icp_rm_try_to_deliver() atomic update and now, then * we know it might have missed the resend_map bit. So we @@ -397,7 +401,9 @@ static void icp_rm_deliver_irq(struct kvmppc_xics *xics, struct kvmppc_icp *icp, */ smp_mb(); if (!icp->state.need_resend) { + state->resend = 0; arch_spin_unlock(&ics->lock); + check_resend = 0; goto again; } } @@ -592,7 +598,7 @@ int kvmppc_rm_h_ipi(struct kvm_vcpu *vcpu, unsigned long server, /* Handle reject in real mode */ if (reject && reject != XICS_IPI) { this_icp->n_reject++; - icp_rm_deliver_irq(xics, icp, reject); + icp_rm_deliver_irq(xics, icp, reject, false); } /* Handle resends in real mode */ @@ -660,59 +666,45 @@ int kvmppc_rm_h_cppr(struct kvm_vcpu *vcpu, unsigned long cppr) */ if (reject && reject != XICS_IPI) { icp->n_reject++; - icp_rm_deliver_irq(xics, icp, reject); + icp_rm_deliver_irq(xics, icp, reject, false); } bail: return check_too_hard(xics, icp); } -int kvmppc_rm_h_eoi(struct kvm_vcpu *vcpu, unsigned long xirr) +static int ics_rm_eoi(struct kvm_vcpu *vcpu, u32 irq) { struct kvmppc_xics *xics = vcpu->kvm->arch.xics; struct kvmppc_icp *icp = vcpu->arch.icp; struct kvmppc_ics *ics; struct ics_irq_state *state; - u32 irq = xirr & 0x00ffffff; u16 src; - - if (!xics || !xics->real_mode) - return H_TOO_HARD; + u32 pq_old, pq_new; /* - * ICP State: EOI + * ICS EOI handling: For LSI, if P bit is still set, we need to + * resend it. * - * Note: If EOI is incorrectly used by SW to lower the CPPR - * value (ie more favored), we do not check for rejection of - * a pending interrupt, this is a SW error and PAPR sepcifies - * that we don't have to deal with it. - * - * The sending of an EOI to the ICS is handled after the - * CPPR update - * - * ICP State: Down_CPPR which we handle - * in a separate function as it's shared with H_CPPR. + * For MSI, we move Q bit into P (and clear Q). If it is set, + * resend it. */ - icp_rm_down_cppr(xics, icp, xirr >> 24); - /* IPIs have no EOI */ - if (irq == XICS_IPI) - goto bail; - /* - * EOI handling: If the interrupt is still asserted, we need to - * resend it. We can take a lockless "peek" at the ICS state here. - * - * "Message" interrupts will never have "asserted" set - */ ics = kvmppc_xics_find_ics(xics, irq, &src); if (!ics) goto bail; + state = &ics->irq_state[src]; - /* Still asserted, resend it */ - if (state->asserted) { - icp->n_reject++; - icp_rm_deliver_irq(xics, icp, irq); - } + if (state->lsi) + pq_new = state->pq_state; + else + do { + pq_old = state->pq_state; + pq_new = pq_old >> 1; + } while (cmpxchg(&state->pq_state, pq_old, pq_new) != pq_old); + + if (pq_new & PQ_PRESENTED) + icp_rm_deliver_irq(xics, NULL, irq, false); if (!hlist_empty(&vcpu->kvm->irq_ack_notifier_list)) { icp->rm_action |= XICS_RM_NOTIFY_EOI; @@ -733,10 +725,43 @@ int kvmppc_rm_h_eoi(struct kvm_vcpu *vcpu, unsigned long xirr) state->intr_cpu = -1; } } + bail: return check_too_hard(xics, icp); } +int kvmppc_rm_h_eoi(struct kvm_vcpu *vcpu, unsigned long xirr) +{ + struct kvmppc_xics *xics = vcpu->kvm->arch.xics; + struct kvmppc_icp *icp = vcpu->arch.icp; + u32 irq = xirr & 0x00ffffff; + + if (!xics || !xics->real_mode) + return H_TOO_HARD; + + /* + * ICP State: EOI + * + * Note: If EOI is incorrectly used by SW to lower the CPPR + * value (ie more favored), we do not check for rejection of + * a pending interrupt, this is a SW error and PAPR specifies + * that we don't have to deal with it. + * + * The sending of an EOI to the ICS is handled after the + * CPPR update + * + * ICP State: Down_CPPR which we handle + * in a separate function as it's shared with H_CPPR. + */ + icp_rm_down_cppr(xics, icp, xirr >> 24); + + /* IPIs have no EOI */ + if (irq == XICS_IPI) + return check_too_hard(xics, icp); + + return ics_rm_eoi(vcpu, irq); +} + unsigned long eoi_rc; static void icp_eoi(struct irq_chip *c, u32 hwirq, __be32 xirr, bool *again) @@ -823,14 +848,33 @@ long kvmppc_deliver_irq_passthru(struct kvm_vcpu *vcpu, { struct kvmppc_xics *xics; struct kvmppc_icp *icp; + struct kvmppc_ics *ics; + struct ics_irq_state *state; u32 irq; + u16 src; + u32 pq_old, pq_new; irq = irq_map->v_hwirq; xics = vcpu->kvm->arch.xics; icp = vcpu->arch.icp; kvmppc_rm_handle_irq_desc(irq_map->desc); - icp_rm_deliver_irq(xics, icp, irq); + + ics = kvmppc_xics_find_ics(xics, irq, &src); + if (!ics) + return 2; + + state = &ics->irq_state[src]; + + /* only MSIs register bypass producers, so it must be MSI here */ + do { + pq_old = state->pq_state; + pq_new = ((pq_old << 1) & 3) | PQ_PRESENTED; + } while (cmpxchg(&state->pq_state, pq_old, pq_new) != pq_old); + + /* Test P=1, Q=0, this is the only case where we present */ + if (pq_new == PQ_PRESENTED) + icp_rm_deliver_irq(xics, icp, irq, false); /* EOI the interrupt */ icp_eoi(irq_desc_get_chip(irq_map->desc), irq_map->r_hwirq, xirr, diff --git a/arch/powerpc/kvm/book3s_pr.c b/arch/powerpc/kvm/book3s_pr.c index 1482961ceb4d..d4dfc0ca2a44 100644 --- a/arch/powerpc/kvm/book3s_pr.c +++ b/arch/powerpc/kvm/book3s_pr.c @@ -902,6 +902,69 @@ static void kvmppc_clear_debug(struct kvm_vcpu *vcpu) } } +static int kvmppc_exit_pr_progint(struct kvm_run *run, struct kvm_vcpu *vcpu, + unsigned int exit_nr) +{ + enum emulation_result er; + ulong flags; + u32 last_inst; + int emul, r; + + /* + * shadow_srr1 only contains valid flags if we came here via a program + * exception. The other exceptions (emulation assist, FP unavailable, + * etc.) do not provide flags in SRR1, so use an illegal-instruction + * exception when injecting a program interrupt into the guest. + */ + if (exit_nr == BOOK3S_INTERRUPT_PROGRAM) + flags = vcpu->arch.shadow_srr1 & 0x1f0000ull; + else + flags = SRR1_PROGILL; + + emul = kvmppc_get_last_inst(vcpu, INST_GENERIC, &last_inst); + if (emul != EMULATE_DONE) + return RESUME_GUEST; + + if (kvmppc_get_msr(vcpu) & MSR_PR) { +#ifdef EXIT_DEBUG + pr_info("Userspace triggered 0x700 exception at\n 0x%lx (0x%x)\n", + kvmppc_get_pc(vcpu), last_inst); +#endif + if ((last_inst & 0xff0007ff) != (INS_DCBZ & 0xfffffff7)) { + kvmppc_core_queue_program(vcpu, flags); + return RESUME_GUEST; + } + } + + vcpu->stat.emulated_inst_exits++; + er = kvmppc_emulate_instruction(run, vcpu); + switch (er) { + case EMULATE_DONE: + r = RESUME_GUEST_NV; + break; + case EMULATE_AGAIN: + r = RESUME_GUEST; + break; + case EMULATE_FAIL: + pr_crit("%s: emulation at %lx failed (%08x)\n", + __func__, kvmppc_get_pc(vcpu), last_inst); + kvmppc_core_queue_program(vcpu, flags); + r = RESUME_GUEST; + break; + case EMULATE_DO_MMIO: + run->exit_reason = KVM_EXIT_MMIO; + r = RESUME_HOST_NV; + break; + case EMULATE_EXIT_USER: + r = RESUME_HOST_NV; + break; + default: + BUG(); + } + + return r; +} + int kvmppc_handle_exit_pr(struct kvm_run *run, struct kvm_vcpu *vcpu, unsigned int exit_nr) { @@ -1044,71 +1107,8 @@ int kvmppc_handle_exit_pr(struct kvm_run *run, struct kvm_vcpu *vcpu, break; case BOOK3S_INTERRUPT_PROGRAM: case BOOK3S_INTERRUPT_H_EMUL_ASSIST: - { - enum emulation_result er; - ulong flags; - u32 last_inst; - int emul; - -program_interrupt: - /* - * shadow_srr1 only contains valid flags if we came here via - * a program exception. The other exceptions (emulation assist, - * FP unavailable, etc.) do not provide flags in SRR1, so use - * an illegal-instruction exception when injecting a program - * interrupt into the guest. - */ - if (exit_nr == BOOK3S_INTERRUPT_PROGRAM) - flags = vcpu->arch.shadow_srr1 & 0x1f0000ull; - else - flags = SRR1_PROGILL; - - emul = kvmppc_get_last_inst(vcpu, INST_GENERIC, &last_inst); - if (emul != EMULATE_DONE) { - r = RESUME_GUEST; - break; - } - - if (kvmppc_get_msr(vcpu) & MSR_PR) { -#ifdef EXIT_DEBUG - pr_info("Userspace triggered 0x700 exception at\n 0x%lx (0x%x)\n", - kvmppc_get_pc(vcpu), last_inst); -#endif - if ((last_inst & 0xff0007ff) != - (INS_DCBZ & 0xfffffff7)) { - kvmppc_core_queue_program(vcpu, flags); - r = RESUME_GUEST; - break; - } - } - - vcpu->stat.emulated_inst_exits++; - er = kvmppc_emulate_instruction(run, vcpu); - switch (er) { - case EMULATE_DONE: - r = RESUME_GUEST_NV; - break; - case EMULATE_AGAIN: - r = RESUME_GUEST; - break; - case EMULATE_FAIL: - printk(KERN_CRIT "%s: emulation at %lx failed (%08x)\n", - __func__, kvmppc_get_pc(vcpu), last_inst); - kvmppc_core_queue_program(vcpu, flags); - r = RESUME_GUEST; - break; - case EMULATE_DO_MMIO: - run->exit_reason = KVM_EXIT_MMIO; - r = RESUME_HOST_NV; - break; - case EMULATE_EXIT_USER: - r = RESUME_HOST_NV; - break; - default: - BUG(); - } + r = kvmppc_exit_pr_progint(run, vcpu, exit_nr); break; - } case BOOK3S_INTERRUPT_SYSCALL: { u32 last_sc; @@ -1185,7 +1185,7 @@ program_interrupt: emul = kvmppc_get_last_inst(vcpu, INST_GENERIC, &last_inst); if (emul == EMULATE_DONE) - goto program_interrupt; + r = kvmppc_exit_pr_progint(run, vcpu, exit_nr); else r = RESUME_GUEST; diff --git a/arch/powerpc/kvm/book3s_xics.c b/arch/powerpc/kvm/book3s_xics.c index 20dff102a06f..e48803e2918d 100644 --- a/arch/powerpc/kvm/book3s_xics.c +++ b/arch/powerpc/kvm/book3s_xics.c @@ -63,7 +63,7 @@ /* -- ICS routines -- */ static void icp_deliver_irq(struct kvmppc_xics *xics, struct kvmppc_icp *icp, - u32 new_irq); + u32 new_irq, bool check_resend); /* * Return value ideally indicates how the interrupt was handled, but no @@ -75,6 +75,7 @@ static int ics_deliver_irq(struct kvmppc_xics *xics, u32 irq, u32 level) struct ics_irq_state *state; struct kvmppc_ics *ics; u16 src; + u32 pq_old, pq_new; XICS_DBG("ics deliver %#x (level: %d)\n", irq, level); @@ -87,25 +88,41 @@ static int ics_deliver_irq(struct kvmppc_xics *xics, u32 irq, u32 level) if (!state->exists) return -EINVAL; + if (level == KVM_INTERRUPT_SET_LEVEL || level == KVM_INTERRUPT_SET) + level = 1; + else if (level == KVM_INTERRUPT_UNSET) + level = 0; /* - * We set state->asserted locklessly. This should be fine as - * we are the only setter, thus concurrent access is undefined - * to begin with. + * Take other values the same as 1, consistent with original code. + * maybe WARN here? */ - if ((level == 1 && state->lsi) || level == KVM_INTERRUPT_SET_LEVEL) - state->asserted = 1; - else if (level == 0 || level == KVM_INTERRUPT_UNSET) { - state->asserted = 0; + + if (!state->lsi && level == 0) /* noop for MSI */ return 0; - } + + do { + pq_old = state->pq_state; + if (state->lsi) { + if (level) { + if (pq_old & PQ_PRESENTED) + /* Setting already set LSI ... */ + return 0; + + pq_new = PQ_PRESENTED; + } else + pq_new = 0; + } else + pq_new = ((pq_old << 1) & 3) | PQ_PRESENTED; + } while (cmpxchg(&state->pq_state, pq_old, pq_new) != pq_old); + + /* Test P=1, Q=0, this is the only case where we present */ + if (pq_new == PQ_PRESENTED) + icp_deliver_irq(xics, NULL, irq, false); /* Record which CPU this arrived on for passed-through interrupts */ if (state->host_irq) state->intr_cpu = raw_smp_processor_id(); - /* Attempt delivery */ - icp_deliver_irq(xics, NULL, irq); - return 0; } @@ -114,29 +131,14 @@ static void ics_check_resend(struct kvmppc_xics *xics, struct kvmppc_ics *ics, { int i; - unsigned long flags; - - local_irq_save(flags); - arch_spin_lock(&ics->lock); - for (i = 0; i < KVMPPC_XICS_IRQ_PER_ICS; i++) { struct ics_irq_state *state = &ics->irq_state[i]; - - if (!state->resend) - continue; - - XICS_DBG("resend %#x prio %#x\n", state->number, - state->priority); - - arch_spin_unlock(&ics->lock); - local_irq_restore(flags); - icp_deliver_irq(xics, icp, state->number); - local_irq_save(flags); - arch_spin_lock(&ics->lock); + if (state->resend) { + XICS_DBG("resend %#x prio %#x\n", state->number, + state->priority); + icp_deliver_irq(xics, icp, state->number, true); + } } - - arch_spin_unlock(&ics->lock); - local_irq_restore(flags); } static bool write_xive(struct kvmppc_xics *xics, struct kvmppc_ics *ics, @@ -155,6 +157,7 @@ static bool write_xive(struct kvmppc_xics *xics, struct kvmppc_ics *ics, deliver = false; if ((state->masked_pending || state->resend) && priority != MASKED) { state->masked_pending = 0; + state->resend = 0; deliver = true; } @@ -189,7 +192,7 @@ int kvmppc_xics_set_xive(struct kvm *kvm, u32 irq, u32 server, u32 priority) state->masked_pending, state->resend); if (write_xive(xics, ics, state, server, priority, priority)) - icp_deliver_irq(xics, icp, irq); + icp_deliver_irq(xics, icp, irq, false); return 0; } @@ -242,7 +245,7 @@ int kvmppc_xics_int_on(struct kvm *kvm, u32 irq) if (write_xive(xics, ics, state, state->server, state->saved_priority, state->saved_priority)) - icp_deliver_irq(xics, icp, irq); + icp_deliver_irq(xics, icp, irq, false); return 0; } @@ -376,7 +379,7 @@ static bool icp_try_to_deliver(struct kvmppc_icp *icp, u32 irq, u8 priority, } static void icp_deliver_irq(struct kvmppc_xics *xics, struct kvmppc_icp *icp, - u32 new_irq) + u32 new_irq, bool check_resend) { struct ics_irq_state *state; struct kvmppc_ics *ics; @@ -422,6 +425,10 @@ static void icp_deliver_irq(struct kvmppc_xics *xics, struct kvmppc_icp *icp, } } + if (check_resend) + if (!state->resend) + goto out; + /* Clear the resend bit of that interrupt */ state->resend = 0; @@ -470,6 +477,7 @@ static void icp_deliver_irq(struct kvmppc_xics *xics, struct kvmppc_icp *icp, arch_spin_unlock(&ics->lock); local_irq_restore(flags); new_irq = reject; + check_resend = 0; goto again; } } else { @@ -477,10 +485,16 @@ static void icp_deliver_irq(struct kvmppc_xics *xics, struct kvmppc_icp *icp, * We failed to deliver the interrupt we need to set the * resend map bit and mark the ICS state as needing a resend */ - set_bit(ics->icsid, icp->resend_map); state->resend = 1; /* + * Make sure when checking resend, we don't miss the resend + * if resend_map bit is seen and cleared. + */ + smp_wmb(); + set_bit(ics->icsid, icp->resend_map); + + /* * If the need_resend flag got cleared in the ICP some time * between icp_try_to_deliver() atomic update and now, then * we know it might have missed the resend_map bit. So we @@ -488,8 +502,10 @@ static void icp_deliver_irq(struct kvmppc_xics *xics, struct kvmppc_icp *icp, */ smp_mb(); if (!icp->state.need_resend) { + state->resend = 0; arch_spin_unlock(&ics->lock); local_irq_restore(flags); + check_resend = 0; goto again; } } @@ -681,7 +697,7 @@ static noinline int kvmppc_h_ipi(struct kvm_vcpu *vcpu, unsigned long server, /* Handle reject */ if (reject && reject != XICS_IPI) - icp_deliver_irq(xics, icp, reject); + icp_deliver_irq(xics, icp, reject, false); /* Handle resend */ if (resend) @@ -761,17 +777,54 @@ static noinline void kvmppc_h_cppr(struct kvm_vcpu *vcpu, unsigned long cppr) * attempt (see comments in icp_deliver_irq). */ if (reject && reject != XICS_IPI) - icp_deliver_irq(xics, icp, reject); + icp_deliver_irq(xics, icp, reject, false); } -static noinline int kvmppc_h_eoi(struct kvm_vcpu *vcpu, unsigned long xirr) +static int ics_eoi(struct kvm_vcpu *vcpu, u32 irq) { struct kvmppc_xics *xics = vcpu->kvm->arch.xics; struct kvmppc_icp *icp = vcpu->arch.icp; struct kvmppc_ics *ics; struct ics_irq_state *state; - u32 irq = xirr & 0x00ffffff; u16 src; + u32 pq_old, pq_new; + + /* + * ICS EOI handling: For LSI, if P bit is still set, we need to + * resend it. + * + * For MSI, we move Q bit into P (and clear Q). If it is set, + * resend it. + */ + + ics = kvmppc_xics_find_ics(xics, irq, &src); + if (!ics) { + XICS_DBG("ios_eoi: IRQ 0x%06x not found !\n", irq); + return H_PARAMETER; + } + state = &ics->irq_state[src]; + + if (state->lsi) + pq_new = state->pq_state; + else + do { + pq_old = state->pq_state; + pq_new = pq_old >> 1; + } while (cmpxchg(&state->pq_state, pq_old, pq_new) != pq_old); + + if (pq_new & PQ_PRESENTED) + icp_deliver_irq(xics, icp, irq, false); + + kvm_notify_acked_irq(vcpu->kvm, 0, irq); + + return H_SUCCESS; +} + +static noinline int kvmppc_h_eoi(struct kvm_vcpu *vcpu, unsigned long xirr) +{ + struct kvmppc_xics *xics = vcpu->kvm->arch.xics; + struct kvmppc_icp *icp = vcpu->arch.icp; + u32 irq = xirr & 0x00ffffff; XICS_DBG("h_eoi vcpu %d eoi %#lx\n", vcpu->vcpu_id, xirr); @@ -794,26 +847,8 @@ static noinline int kvmppc_h_eoi(struct kvm_vcpu *vcpu, unsigned long xirr) /* IPIs have no EOI */ if (irq == XICS_IPI) return H_SUCCESS; - /* - * EOI handling: If the interrupt is still asserted, we need to - * resend it. We can take a lockless "peek" at the ICS state here. - * - * "Message" interrupts will never have "asserted" set - */ - ics = kvmppc_xics_find_ics(xics, irq, &src); - if (!ics) { - XICS_DBG("h_eoi: IRQ 0x%06x not found !\n", irq); - return H_PARAMETER; - } - state = &ics->irq_state[src]; - /* Still asserted, resend it */ - if (state->asserted) - icp_deliver_irq(xics, icp, irq); - - kvm_notify_acked_irq(vcpu->kvm, 0, irq); - - return H_SUCCESS; + return ics_eoi(vcpu, irq); } int kvmppc_xics_rm_complete(struct kvm_vcpu *vcpu, u32 hcall) @@ -832,10 +867,6 @@ int kvmppc_xics_rm_complete(struct kvm_vcpu *vcpu, u32 hcall) icp->n_rm_check_resend++; icp_check_resend(xics, icp->rm_resend_icp); } - if (icp->rm_action & XICS_RM_REJECT) { - icp->n_rm_reject++; - icp_deliver_irq(xics, icp, icp->rm_reject); - } if (icp->rm_action & XICS_RM_NOTIFY_EOI) { icp->n_rm_notify_eoi++; kvm_notify_acked_irq(vcpu->kvm, 0, icp->rm_eoied_irq); @@ -920,7 +951,7 @@ static int xics_debug_show(struct seq_file *m, void *private) int icsid, i; unsigned long flags; unsigned long t_rm_kick_vcpu, t_rm_check_resend; - unsigned long t_rm_reject, t_rm_notify_eoi; + unsigned long t_rm_notify_eoi; unsigned long t_reject, t_check_resend; if (!kvm) @@ -929,7 +960,6 @@ static int xics_debug_show(struct seq_file *m, void *private) t_rm_kick_vcpu = 0; t_rm_notify_eoi = 0; t_rm_check_resend = 0; - t_rm_reject = 0; t_check_resend = 0; t_reject = 0; @@ -952,14 +982,13 @@ static int xics_debug_show(struct seq_file *m, void *private) t_rm_kick_vcpu += icp->n_rm_kick_vcpu; t_rm_notify_eoi += icp->n_rm_notify_eoi; t_rm_check_resend += icp->n_rm_check_resend; - t_rm_reject += icp->n_rm_reject; t_check_resend += icp->n_check_resend; t_reject += icp->n_reject; } - seq_printf(m, "ICP Guest->Host totals: kick_vcpu=%lu check_resend=%lu reject=%lu notify_eoi=%lu\n", + seq_printf(m, "ICP Guest->Host totals: kick_vcpu=%lu check_resend=%lu notify_eoi=%lu\n", t_rm_kick_vcpu, t_rm_check_resend, - t_rm_reject, t_rm_notify_eoi); + t_rm_notify_eoi); seq_printf(m, "ICP Real Mode totals: check_resend=%lu resend=%lu\n", t_check_resend, t_reject); for (icsid = 0; icsid <= KVMPPC_XICS_MAX_ICS_ID; icsid++) { @@ -977,9 +1006,9 @@ static int xics_debug_show(struct seq_file *m, void *private) for (i = 0; i < KVMPPC_XICS_IRQ_PER_ICS; i++) { struct ics_irq_state *irq = &ics->irq_state[i]; - seq_printf(m, "irq 0x%06x: server %#x prio %#x save prio %#x asserted %d resend %d masked pending %d\n", + seq_printf(m, "irq 0x%06x: server %#x prio %#x save prio %#x pq_state %d resend %d masked pending %d\n", irq->number, irq->server, irq->priority, - irq->saved_priority, irq->asserted, + irq->saved_priority, irq->pq_state, irq->resend, irq->masked_pending); } @@ -1198,10 +1227,17 @@ static int xics_get_source(struct kvmppc_xics *xics, long irq, u64 addr) val |= prio << KVM_XICS_PRIORITY_SHIFT; if (irqp->lsi) { val |= KVM_XICS_LEVEL_SENSITIVE; - if (irqp->asserted) + if (irqp->pq_state & PQ_PRESENTED) val |= KVM_XICS_PENDING; } else if (irqp->masked_pending || irqp->resend) val |= KVM_XICS_PENDING; + + if (irqp->pq_state & PQ_PRESENTED) + val |= KVM_XICS_PRESENTED; + + if (irqp->pq_state & PQ_QUEUED) + val |= KVM_XICS_QUEUED; + ret = 0; } arch_spin_unlock(&ics->lock); @@ -1253,18 +1289,20 @@ static int xics_set_source(struct kvmppc_xics *xics, long irq, u64 addr) irqp->resend = 0; irqp->masked_pending = 0; irqp->lsi = 0; - irqp->asserted = 0; - if (val & KVM_XICS_LEVEL_SENSITIVE) { + irqp->pq_state = 0; + if (val & KVM_XICS_LEVEL_SENSITIVE) irqp->lsi = 1; - if (val & KVM_XICS_PENDING) - irqp->asserted = 1; - } + /* If PENDING, set P in case P is not saved because of old code */ + if (val & KVM_XICS_PRESENTED || val & KVM_XICS_PENDING) + irqp->pq_state |= PQ_PRESENTED; + if (val & KVM_XICS_QUEUED) + irqp->pq_state |= PQ_QUEUED; irqp->exists = 1; arch_spin_unlock(&ics->lock); local_irq_restore(flags); if (val & KVM_XICS_PENDING) - icp_deliver_irq(xics, NULL, irqp->number); + icp_deliver_irq(xics, NULL, irqp->number, false); return 0; } diff --git a/arch/powerpc/kvm/book3s_xics.h b/arch/powerpc/kvm/book3s_xics.h index 2a50320b55ca..ec5474cf70c6 100644 --- a/arch/powerpc/kvm/book3s_xics.h +++ b/arch/powerpc/kvm/book3s_xics.h @@ -31,16 +31,19 @@ /* Priority value to use for disabling an interrupt */ #define MASKED 0xff +#define PQ_PRESENTED 1 +#define PQ_QUEUED 2 + /* State for one irq source */ struct ics_irq_state { u32 number; u32 server; + u32 pq_state; u8 priority; u8 saved_priority; u8 resend; u8 masked_pending; u8 lsi; /* level-sensitive interrupt */ - u8 asserted; /* Only for LSI */ u8 exists; int intr_cpu; u32 host_irq; @@ -73,7 +76,6 @@ struct kvmppc_icp { */ #define XICS_RM_KICK_VCPU 0x1 #define XICS_RM_CHECK_RESEND 0x2 -#define XICS_RM_REJECT 0x4 #define XICS_RM_NOTIFY_EOI 0x8 u32 rm_action; struct kvm_vcpu *rm_kick_target; @@ -84,7 +86,6 @@ struct kvmppc_icp { /* Counters for each reason we exited real mode */ unsigned long n_rm_kick_vcpu; unsigned long n_rm_check_resend; - unsigned long n_rm_reject; unsigned long n_rm_notify_eoi; /* Counters for handling ICP processing in real mode */ unsigned long n_check_resend; diff --git a/arch/powerpc/kvm/powerpc.c b/arch/powerpc/kvm/powerpc.c index 40a5b2d75ed1..2b38d824e9e5 100644 --- a/arch/powerpc/kvm/powerpc.c +++ b/arch/powerpc/kvm/powerpc.c @@ -511,6 +511,7 @@ int kvm_vm_ioctl_check_extension(struct kvm *kvm, long ext) case KVM_CAP_ONE_REG: case KVM_CAP_IOEVENTFD: case KVM_CAP_DEVICE_CTRL: + case KVM_CAP_IMMEDIATE_EXIT: r = 1; break; case KVM_CAP_PPC_PAIRED_SINGLES: @@ -612,6 +613,10 @@ int kvm_vm_ioctl_check_extension(struct kvm *kvm, long ext) case KVM_CAP_SPAPR_MULTITCE: r = 1; break; + case KVM_CAP_SPAPR_RESIZE_HPT: + /* Disable this on POWER9 until code handles new HPTE format */ + r = !!hv_enabled && !cpu_has_feature(CPU_FTR_ARCH_300); + break; #endif case KVM_CAP_PPC_HTM: r = cpu_has_feature(CPU_FTR_TM_COMP) && @@ -1114,7 +1119,10 @@ int kvm_arch_vcpu_ioctl_run(struct kvm_vcpu *vcpu, struct kvm_run *run) #endif } - r = kvmppc_vcpu_run(run, vcpu); + if (run->immediate_exit) + r = -EINTR; + else + r = kvmppc_vcpu_run(run, vcpu); if (vcpu->sigset_active) sigprocmask(SIG_SETMASK, &sigsaved, NULL); diff --git a/arch/s390/kvm/gaccess.c b/arch/s390/kvm/gaccess.c index 4aa8a7e2a1da..4492c9363178 100644 --- a/arch/s390/kvm/gaccess.c +++ b/arch/s390/kvm/gaccess.c @@ -373,7 +373,7 @@ void ipte_unlock(struct kvm_vcpu *vcpu) ipte_unlock_simple(vcpu); } -static int ar_translation(struct kvm_vcpu *vcpu, union asce *asce, ar_t ar, +static int ar_translation(struct kvm_vcpu *vcpu, union asce *asce, u8 ar, enum gacc_mode mode) { union alet alet; @@ -465,7 +465,9 @@ static int ar_translation(struct kvm_vcpu *vcpu, union asce *asce, ar_t ar, struct trans_exc_code_bits { unsigned long addr : 52; /* Translation-exception Address */ unsigned long fsi : 2; /* Access Exception Fetch/Store Indication */ - unsigned long : 6; + unsigned long : 2; + unsigned long b56 : 1; + unsigned long : 3; unsigned long b60 : 1; unsigned long b61 : 1; unsigned long as : 2; /* ASCE Identifier */ @@ -485,7 +487,7 @@ enum prot_type { }; static int trans_exc(struct kvm_vcpu *vcpu, int code, unsigned long gva, - ar_t ar, enum gacc_mode mode, enum prot_type prot) + u8 ar, enum gacc_mode mode, enum prot_type prot) { struct kvm_s390_pgm_info *pgm = &vcpu->arch.pgm; struct trans_exc_code_bits *tec; @@ -497,14 +499,18 @@ static int trans_exc(struct kvm_vcpu *vcpu, int code, unsigned long gva, switch (code) { case PGM_PROTECTION: switch (prot) { + case PROT_TYPE_LA: + tec->b56 = 1; + break; + case PROT_TYPE_KEYC: + tec->b60 = 1; + break; case PROT_TYPE_ALC: tec->b60 = 1; /* FALL THROUGH */ case PROT_TYPE_DAT: tec->b61 = 1; break; - default: /* LA and KEYC set b61 to 0, other params undefined */ - return code; } /* FALL THROUGH */ case PGM_ASCE_TYPE: @@ -539,7 +545,7 @@ static int trans_exc(struct kvm_vcpu *vcpu, int code, unsigned long gva, } static int get_vcpu_asce(struct kvm_vcpu *vcpu, union asce *asce, - unsigned long ga, ar_t ar, enum gacc_mode mode) + unsigned long ga, u8 ar, enum gacc_mode mode) { int rc; struct psw_bits psw = psw_bits(vcpu->arch.sie_block->gpsw); @@ -771,7 +777,7 @@ static int low_address_protection_enabled(struct kvm_vcpu *vcpu, return 1; } -static int guest_page_range(struct kvm_vcpu *vcpu, unsigned long ga, ar_t ar, +static int guest_page_range(struct kvm_vcpu *vcpu, unsigned long ga, u8 ar, unsigned long *pages, unsigned long nr_pages, const union asce asce, enum gacc_mode mode) { @@ -803,7 +809,7 @@ static int guest_page_range(struct kvm_vcpu *vcpu, unsigned long ga, ar_t ar, return 0; } -int access_guest(struct kvm_vcpu *vcpu, unsigned long ga, ar_t ar, void *data, +int access_guest(struct kvm_vcpu *vcpu, unsigned long ga, u8 ar, void *data, unsigned long len, enum gacc_mode mode) { psw_t *psw = &vcpu->arch.sie_block->gpsw; @@ -877,7 +883,7 @@ int access_guest_real(struct kvm_vcpu *vcpu, unsigned long gra, * Note: The IPTE lock is not taken during this function, so the caller * has to take care of this. */ -int guest_translate_address(struct kvm_vcpu *vcpu, unsigned long gva, ar_t ar, +int guest_translate_address(struct kvm_vcpu *vcpu, unsigned long gva, u8 ar, unsigned long *gpa, enum gacc_mode mode) { psw_t *psw = &vcpu->arch.sie_block->gpsw; @@ -910,7 +916,7 @@ int guest_translate_address(struct kvm_vcpu *vcpu, unsigned long gva, ar_t ar, /** * check_gva_range - test a range of guest virtual addresses for accessibility */ -int check_gva_range(struct kvm_vcpu *vcpu, unsigned long gva, ar_t ar, +int check_gva_range(struct kvm_vcpu *vcpu, unsigned long gva, u8 ar, unsigned long length, enum gacc_mode mode) { unsigned long gpa; diff --git a/arch/s390/kvm/gaccess.h b/arch/s390/kvm/gaccess.h index 8756569ad938..7ce47fd36f28 100644 --- a/arch/s390/kvm/gaccess.h +++ b/arch/s390/kvm/gaccess.h @@ -162,11 +162,11 @@ enum gacc_mode { }; int guest_translate_address(struct kvm_vcpu *vcpu, unsigned long gva, - ar_t ar, unsigned long *gpa, enum gacc_mode mode); -int check_gva_range(struct kvm_vcpu *vcpu, unsigned long gva, ar_t ar, + u8 ar, unsigned long *gpa, enum gacc_mode mode); +int check_gva_range(struct kvm_vcpu *vcpu, unsigned long gva, u8 ar, unsigned long length, enum gacc_mode mode); -int access_guest(struct kvm_vcpu *vcpu, unsigned long ga, ar_t ar, void *data, +int access_guest(struct kvm_vcpu *vcpu, unsigned long ga, u8 ar, void *data, unsigned long len, enum gacc_mode mode); int access_guest_real(struct kvm_vcpu *vcpu, unsigned long gra, @@ -218,7 +218,7 @@ int access_guest_real(struct kvm_vcpu *vcpu, unsigned long gra, * if data has been changed in guest space in case of an exception. */ static inline __must_check -int write_guest(struct kvm_vcpu *vcpu, unsigned long ga, ar_t ar, void *data, +int write_guest(struct kvm_vcpu *vcpu, unsigned long ga, u8 ar, void *data, unsigned long len) { return access_guest(vcpu, ga, ar, data, len, GACC_STORE); @@ -238,7 +238,7 @@ int write_guest(struct kvm_vcpu *vcpu, unsigned long ga, ar_t ar, void *data, * data will be copied from guest space to kernel space. */ static inline __must_check -int read_guest(struct kvm_vcpu *vcpu, unsigned long ga, ar_t ar, void *data, +int read_guest(struct kvm_vcpu *vcpu, unsigned long ga, u8 ar, void *data, unsigned long len) { return access_guest(vcpu, ga, ar, data, len, GACC_FETCH); @@ -247,10 +247,11 @@ int read_guest(struct kvm_vcpu *vcpu, unsigned long ga, ar_t ar, void *data, /** * read_guest_instr - copy instruction data from guest space to kernel space * @vcpu: virtual cpu + * @ga: guest address * @data: destination address in kernel space * @len: number of bytes to copy * - * Copy @len bytes from the current psw address (guest space) to @data (kernel + * Copy @len bytes from the given address (guest space) to @data (kernel * space). * * The behaviour of read_guest_instr is identical to read_guest, except that @@ -258,10 +259,10 @@ int read_guest(struct kvm_vcpu *vcpu, unsigned long ga, ar_t ar, void *data, * address-space mode. */ static inline __must_check -int read_guest_instr(struct kvm_vcpu *vcpu, void *data, unsigned long len) +int read_guest_instr(struct kvm_vcpu *vcpu, unsigned long ga, void *data, + unsigned long len) { - return access_guest(vcpu, vcpu->arch.sie_block->gpsw.addr, 0, data, len, - GACC_IFETCH); + return access_guest(vcpu, ga, 0, data, len, GACC_IFETCH); } /** diff --git a/arch/s390/kvm/guestdbg.c b/arch/s390/kvm/guestdbg.c index d7c6a7f53ced..23d9a4e12da1 100644 --- a/arch/s390/kvm/guestdbg.c +++ b/arch/s390/kvm/guestdbg.c @@ -388,14 +388,13 @@ void kvm_s390_prepare_debug_exit(struct kvm_vcpu *vcpu) #define per_write_wp_event(code) \ (code & (PER_CODE_STORE | PER_CODE_STORE_REAL)) -static int debug_exit_required(struct kvm_vcpu *vcpu) +static int debug_exit_required(struct kvm_vcpu *vcpu, u8 perc, + unsigned long peraddr) { - u8 perc = vcpu->arch.sie_block->perc; struct kvm_debug_exit_arch *debug_exit = &vcpu->run->debug.arch; struct kvm_hw_wp_info_arch *wp_info = NULL; struct kvm_hw_bp_info_arch *bp_info = NULL; unsigned long addr = vcpu->arch.sie_block->gpsw.addr; - unsigned long peraddr = vcpu->arch.sie_block->peraddr; if (guestdbg_hw_bp_enabled(vcpu)) { if (per_write_wp_event(perc) && @@ -437,36 +436,118 @@ exit_required: return 1; } +static int per_fetched_addr(struct kvm_vcpu *vcpu, unsigned long *addr) +{ + u8 exec_ilen = 0; + u16 opcode[3]; + int rc; + + if (vcpu->arch.sie_block->icptcode == ICPT_PROGI) { + /* PER address references the fetched or the execute instr */ + *addr = vcpu->arch.sie_block->peraddr; + /* + * Manually detect if we have an EXECUTE instruction. As + * instructions are always 2 byte aligned we can read the + * first two bytes unconditionally + */ + rc = read_guest_instr(vcpu, *addr, &opcode, 2); + if (rc) + return rc; + if (opcode[0] >> 8 == 0x44) + exec_ilen = 4; + if ((opcode[0] & 0xff0f) == 0xc600) + exec_ilen = 6; + } else { + /* instr was suppressed, calculate the responsible instr */ + *addr = __rewind_psw(vcpu->arch.sie_block->gpsw, + kvm_s390_get_ilen(vcpu)); + if (vcpu->arch.sie_block->icptstatus & 0x01) { + exec_ilen = (vcpu->arch.sie_block->icptstatus & 0x60) >> 4; + if (!exec_ilen) + exec_ilen = 4; + } + } + + if (exec_ilen) { + /* read the complete EXECUTE instr to detect the fetched addr */ + rc = read_guest_instr(vcpu, *addr, &opcode, exec_ilen); + if (rc) + return rc; + if (exec_ilen == 6) { + /* EXECUTE RELATIVE LONG - RIL-b format */ + s32 rl = *((s32 *) (opcode + 1)); + + /* rl is a _signed_ 32 bit value specifying halfwords */ + *addr += (u64)(s64) rl * 2; + } else { + /* EXECUTE - RX-a format */ + u32 base = (opcode[1] & 0xf000) >> 12; + u32 disp = opcode[1] & 0x0fff; + u32 index = opcode[0] & 0x000f; + + *addr = base ? vcpu->run->s.regs.gprs[base] : 0; + *addr += index ? vcpu->run->s.regs.gprs[index] : 0; + *addr += disp; + } + *addr = kvm_s390_logical_to_effective(vcpu, *addr); + } + return 0; +} + #define guest_per_enabled(vcpu) \ (vcpu->arch.sie_block->gpsw.mask & PSW_MASK_PER) int kvm_s390_handle_per_ifetch_icpt(struct kvm_vcpu *vcpu) { + const u64 cr10 = vcpu->arch.sie_block->gcr[10]; + const u64 cr11 = vcpu->arch.sie_block->gcr[11]; const u8 ilen = kvm_s390_get_ilen(vcpu); struct kvm_s390_pgm_info pgm_info = { .code = PGM_PER, .per_code = PER_CODE_IFETCH, .per_address = __rewind_psw(vcpu->arch.sie_block->gpsw, ilen), }; + unsigned long fetched_addr; + int rc; /* * The PSW points to the next instruction, therefore the intercepted * instruction generated a PER i-fetch event. PER address therefore * points at the previous PSW address (could be an EXECUTE function). */ - return kvm_s390_inject_prog_irq(vcpu, &pgm_info); + if (!guestdbg_enabled(vcpu)) + return kvm_s390_inject_prog_irq(vcpu, &pgm_info); + + if (debug_exit_required(vcpu, pgm_info.per_code, pgm_info.per_address)) + vcpu->guest_debug |= KVM_GUESTDBG_EXIT_PENDING; + + if (!guest_per_enabled(vcpu) || + !(vcpu->arch.sie_block->gcr[9] & PER_EVENT_IFETCH)) + return 0; + + rc = per_fetched_addr(vcpu, &fetched_addr); + if (rc < 0) + return rc; + if (rc) + /* instruction-fetching exceptions */ + return kvm_s390_inject_program_int(vcpu, PGM_ADDRESSING); + + if (in_addr_range(fetched_addr, cr10, cr11)) + return kvm_s390_inject_prog_irq(vcpu, &pgm_info); + return 0; } -static void filter_guest_per_event(struct kvm_vcpu *vcpu) +static int filter_guest_per_event(struct kvm_vcpu *vcpu) { const u8 perc = vcpu->arch.sie_block->perc; - u64 peraddr = vcpu->arch.sie_block->peraddr; u64 addr = vcpu->arch.sie_block->gpsw.addr; u64 cr9 = vcpu->arch.sie_block->gcr[9]; u64 cr10 = vcpu->arch.sie_block->gcr[10]; u64 cr11 = vcpu->arch.sie_block->gcr[11]; /* filter all events, demanded by the guest */ u8 guest_perc = perc & (cr9 >> 24) & PER_CODE_MASK; + unsigned long fetched_addr; + int rc; if (!guest_per_enabled(vcpu)) guest_perc = 0; @@ -478,9 +559,17 @@ static void filter_guest_per_event(struct kvm_vcpu *vcpu) guest_perc &= ~PER_CODE_BRANCH; /* filter "instruction-fetching" events */ - if (guest_perc & PER_CODE_IFETCH && - !in_addr_range(peraddr, cr10, cr11)) - guest_perc &= ~PER_CODE_IFETCH; + if (guest_perc & PER_CODE_IFETCH) { + rc = per_fetched_addr(vcpu, &fetched_addr); + if (rc < 0) + return rc; + /* + * Don't inject an irq on exceptions. This would make handling + * on icpt code 8 very complex (as PSW was already rewound). + */ + if (rc || !in_addr_range(fetched_addr, cr10, cr11)) + guest_perc &= ~PER_CODE_IFETCH; + } /* All other PER events will be given to the guest */ /* TODO: Check altered address/address space */ @@ -489,6 +578,7 @@ static void filter_guest_per_event(struct kvm_vcpu *vcpu) if (!guest_perc) vcpu->arch.sie_block->iprcc &= ~PGM_PER; + return 0; } #define pssec(vcpu) (vcpu->arch.sie_block->gcr[1] & _ASCE_SPACE_SWITCH) @@ -496,14 +586,17 @@ static void filter_guest_per_event(struct kvm_vcpu *vcpu) #define old_ssec(vcpu) ((vcpu->arch.sie_block->tecmc >> 31) & 0x1) #define old_as_is_home(vcpu) !(vcpu->arch.sie_block->tecmc & 0xffff) -void kvm_s390_handle_per_event(struct kvm_vcpu *vcpu) +int kvm_s390_handle_per_event(struct kvm_vcpu *vcpu) { - int new_as; + int rc, new_as; - if (debug_exit_required(vcpu)) + if (debug_exit_required(vcpu, vcpu->arch.sie_block->perc, + vcpu->arch.sie_block->peraddr)) vcpu->guest_debug |= KVM_GUESTDBG_EXIT_PENDING; - filter_guest_per_event(vcpu); + rc = filter_guest_per_event(vcpu); + if (rc) + return rc; /* * Only RP, SAC, SACF, PT, PTI, PR, PC instructions can trigger @@ -532,4 +625,5 @@ void kvm_s390_handle_per_event(struct kvm_vcpu *vcpu) (pssec(vcpu) || old_ssec(vcpu))) vcpu->arch.sie_block->iprcc = PGM_SPACE_SWITCH; } + return 0; } diff --git a/arch/s390/kvm/intercept.c b/arch/s390/kvm/intercept.c index 7a27eebab28a..59920f96ebc0 100644 --- a/arch/s390/kvm/intercept.c +++ b/arch/s390/kvm/intercept.c @@ -238,7 +238,9 @@ static int handle_prog(struct kvm_vcpu *vcpu) vcpu->stat.exit_program_interruption++; if (guestdbg_enabled(vcpu) && per_event(vcpu)) { - kvm_s390_handle_per_event(vcpu); + rc = kvm_s390_handle_per_event(vcpu); + if (rc) + return rc; /* the interrupt might have been filtered out completely */ if (vcpu->arch.sie_block->iprcc == 0) return 0; @@ -359,6 +361,9 @@ static int handle_partial_execution(struct kvm_vcpu *vcpu) static int handle_operexc(struct kvm_vcpu *vcpu) { + psw_t oldpsw, newpsw; + int rc; + vcpu->stat.exit_operation_exception++; trace_kvm_s390_handle_operexc(vcpu, vcpu->arch.sie_block->ipa, vcpu->arch.sie_block->ipb); @@ -369,6 +374,24 @@ static int handle_operexc(struct kvm_vcpu *vcpu) if (vcpu->arch.sie_block->ipa == 0 && vcpu->kvm->arch.user_instr0) return -EOPNOTSUPP; + rc = read_guest_lc(vcpu, __LC_PGM_NEW_PSW, &newpsw, sizeof(psw_t)); + if (rc) + return rc; + /* + * Avoid endless loops of operation exceptions, if the pgm new + * PSW will cause a new operation exception. + * The heuristic checks if the pgm new psw is within 6 bytes before + * the faulting psw address (with same DAT, AS settings) and the + * new psw is not a wait psw and the fault was not triggered by + * problem state. + */ + oldpsw = vcpu->arch.sie_block->gpsw; + if (oldpsw.addr - newpsw.addr <= 6 && + !(newpsw.mask & PSW_MASK_WAIT) && + !(oldpsw.mask & PSW_MASK_PSTATE) && + (newpsw.mask & PSW_MASK_ASC) == (oldpsw.mask & PSW_MASK_ASC) && + (newpsw.mask & PSW_MASK_DAT) == (oldpsw.mask & PSW_MASK_DAT)) + return -EOPNOTSUPP; return kvm_s390_inject_program_int(vcpu, PGM_OPERATION); } diff --git a/arch/s390/kvm/kvm-s390.c b/arch/s390/kvm/kvm-s390.c index b604854df02c..f5694838234d 100644 --- a/arch/s390/kvm/kvm-s390.c +++ b/arch/s390/kvm/kvm-s390.c @@ -218,7 +218,7 @@ static void allow_cpu_feat(unsigned long nr) static inline int plo_test_bit(unsigned char nr) { register unsigned long r0 asm("0") = (unsigned long) nr | 0x100; - int cc = 3; /* subfunction not available */ + int cc; asm volatile( /* Parameter registers are ignored for "test bit" */ @@ -371,6 +371,7 @@ int kvm_vm_ioctl_check_extension(struct kvm *kvm, long ext) case KVM_CAP_S390_IRQCHIP: case KVM_CAP_VM_ATTRIBUTES: case KVM_CAP_MP_STATE: + case KVM_CAP_IMMEDIATE_EXIT: case KVM_CAP_S390_INJECT_IRQ: case KVM_CAP_S390_USER_SIGP: case KVM_CAP_S390_USER_STSI: @@ -443,6 +444,9 @@ int kvm_vm_ioctl_get_dirty_log(struct kvm *kvm, struct kvm_memory_slot *memslot; int is_dirty = 0; + if (kvm_is_ucontrol(kvm)) + return -EINVAL; + mutex_lock(&kvm->slots_lock); r = -EINVAL; @@ -506,6 +510,14 @@ static int kvm_vm_ioctl_enable_cap(struct kvm *kvm, struct kvm_enable_cap *cap) } else if (MACHINE_HAS_VX) { set_kvm_facility(kvm->arch.model.fac_mask, 129); set_kvm_facility(kvm->arch.model.fac_list, 129); + if (test_facility(134)) { + set_kvm_facility(kvm->arch.model.fac_mask, 134); + set_kvm_facility(kvm->arch.model.fac_list, 134); + } + if (test_facility(135)) { + set_kvm_facility(kvm->arch.model.fac_mask, 135); + set_kvm_facility(kvm->arch.model.fac_list, 135); + } r = 0; } else r = -EINVAL; @@ -822,6 +834,13 @@ static int kvm_s390_set_processor(struct kvm *kvm, struct kvm_device_attr *attr) } memcpy(kvm->arch.model.fac_list, proc->fac_list, S390_ARCH_FAC_LIST_SIZE_BYTE); + VM_EVENT(kvm, 3, "SET: guest ibc: 0x%4.4x, guest cpuid: 0x%16.16llx", + kvm->arch.model.ibc, + kvm->arch.model.cpuid); + VM_EVENT(kvm, 3, "SET: guest faclist: 0x%16.16llx.%16.16llx.%16.16llx", + kvm->arch.model.fac_list[0], + kvm->arch.model.fac_list[1], + kvm->arch.model.fac_list[2]); } else ret = -EFAULT; kfree(proc); @@ -895,6 +914,13 @@ static int kvm_s390_get_processor(struct kvm *kvm, struct kvm_device_attr *attr) proc->ibc = kvm->arch.model.ibc; memcpy(&proc->fac_list, kvm->arch.model.fac_list, S390_ARCH_FAC_LIST_SIZE_BYTE); + VM_EVENT(kvm, 3, "GET: guest ibc: 0x%4.4x, guest cpuid: 0x%16.16llx", + kvm->arch.model.ibc, + kvm->arch.model.cpuid); + VM_EVENT(kvm, 3, "GET: guest faclist: 0x%16.16llx.%16.16llx.%16.16llx", + kvm->arch.model.fac_list[0], + kvm->arch.model.fac_list[1], + kvm->arch.model.fac_list[2]); if (copy_to_user((void __user *)attr->addr, proc, sizeof(*proc))) ret = -EFAULT; kfree(proc); @@ -918,6 +944,17 @@ static int kvm_s390_get_machine(struct kvm *kvm, struct kvm_device_attr *attr) S390_ARCH_FAC_LIST_SIZE_BYTE); memcpy((unsigned long *)&mach->fac_list, S390_lowcore.stfle_fac_list, sizeof(S390_lowcore.stfle_fac_list)); + VM_EVENT(kvm, 3, "GET: host ibc: 0x%4.4x, host cpuid: 0x%16.16llx", + kvm->arch.model.ibc, + kvm->arch.model.cpuid); + VM_EVENT(kvm, 3, "GET: host facmask: 0x%16.16llx.%16.16llx.%16.16llx", + mach->fac_mask[0], + mach->fac_mask[1], + mach->fac_mask[2]); + VM_EVENT(kvm, 3, "GET: host faclist: 0x%16.16llx.%16.16llx.%16.16llx", + mach->fac_list[0], + mach->fac_list[1], + mach->fac_list[2]); if (copy_to_user((void __user *)attr->addr, mach, sizeof(*mach))) ret = -EFAULT; kfree(mach); @@ -1939,6 +1976,8 @@ int kvm_arch_vcpu_setup(struct kvm_vcpu *vcpu) if (test_kvm_facility(vcpu->kvm, 8) && sclp.has_pfmfi) vcpu->arch.sie_block->ecb2 |= 0x08; + if (test_kvm_facility(vcpu->kvm, 130)) + vcpu->arch.sie_block->ecb2 |= 0x20; vcpu->arch.sie_block->eca = 0x1002000U; if (sclp.has_cei) vcpu->arch.sie_block->eca |= 0x80000000U; @@ -2579,7 +2618,7 @@ static int vcpu_post_run_fault_in_sie(struct kvm_vcpu *vcpu) * to look up the current opcode to get the length of the instruction * to be able to forward the PSW. */ - rc = read_guest_instr(vcpu, &opcode, 1); + rc = read_guest_instr(vcpu, vcpu->arch.sie_block->gpsw.addr, &opcode, 1); ilen = insn_length(opcode); if (rc < 0) { return rc; @@ -2761,6 +2800,9 @@ int kvm_arch_vcpu_ioctl_run(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run) int rc; sigset_t sigsaved; + if (kvm_run->immediate_exit) + return -EINTR; + if (guestdbg_exit_pending(vcpu)) { kvm_s390_prepare_debug_exit(vcpu); return 0; diff --git a/arch/s390/kvm/kvm-s390.h b/arch/s390/kvm/kvm-s390.h index 3a4e97f1a9e6..af9fa91a0c91 100644 --- a/arch/s390/kvm/kvm-s390.h +++ b/arch/s390/kvm/kvm-s390.h @@ -86,9 +86,7 @@ static inline void kvm_s390_set_prefix(struct kvm_vcpu *vcpu, u32 prefix) kvm_make_request(KVM_REQ_MMU_RELOAD, vcpu); } -typedef u8 __bitwise ar_t; - -static inline u64 kvm_s390_get_base_disp_s(struct kvm_vcpu *vcpu, ar_t *ar) +static inline u64 kvm_s390_get_base_disp_s(struct kvm_vcpu *vcpu, u8 *ar) { u32 base2 = vcpu->arch.sie_block->ipb >> 28; u32 disp2 = ((vcpu->arch.sie_block->ipb & 0x0fff0000) >> 16); @@ -101,7 +99,7 @@ static inline u64 kvm_s390_get_base_disp_s(struct kvm_vcpu *vcpu, ar_t *ar) static inline void kvm_s390_get_base_disp_sse(struct kvm_vcpu *vcpu, u64 *address1, u64 *address2, - ar_t *ar_b1, ar_t *ar_b2) + u8 *ar_b1, u8 *ar_b2) { u32 base1 = (vcpu->arch.sie_block->ipb & 0xf0000000) >> 28; u32 disp1 = (vcpu->arch.sie_block->ipb & 0x0fff0000) >> 16; @@ -125,7 +123,7 @@ static inline void kvm_s390_get_regs_rre(struct kvm_vcpu *vcpu, int *r1, int *r2 *r2 = (vcpu->arch.sie_block->ipb & 0x000f0000) >> 16; } -static inline u64 kvm_s390_get_base_disp_rsy(struct kvm_vcpu *vcpu, ar_t *ar) +static inline u64 kvm_s390_get_base_disp_rsy(struct kvm_vcpu *vcpu, u8 *ar) { u32 base2 = vcpu->arch.sie_block->ipb >> 28; u32 disp2 = ((vcpu->arch.sie_block->ipb & 0x0fff0000) >> 16) + @@ -140,7 +138,7 @@ static inline u64 kvm_s390_get_base_disp_rsy(struct kvm_vcpu *vcpu, ar_t *ar) return (base2 ? vcpu->run->s.regs.gprs[base2] : 0) + (long)(int)disp2; } -static inline u64 kvm_s390_get_base_disp_rs(struct kvm_vcpu *vcpu, ar_t *ar) +static inline u64 kvm_s390_get_base_disp_rs(struct kvm_vcpu *vcpu, u8 *ar) { u32 base2 = vcpu->arch.sie_block->ipb >> 28; u32 disp2 = ((vcpu->arch.sie_block->ipb & 0x0fff0000) >> 16); @@ -379,7 +377,7 @@ int kvm_s390_import_bp_data(struct kvm_vcpu *vcpu, void kvm_s390_clear_bp_data(struct kvm_vcpu *vcpu); void kvm_s390_prepare_debug_exit(struct kvm_vcpu *vcpu); int kvm_s390_handle_per_ifetch_icpt(struct kvm_vcpu *vcpu); -void kvm_s390_handle_per_event(struct kvm_vcpu *vcpu); +int kvm_s390_handle_per_event(struct kvm_vcpu *vcpu); /* support for Basic/Extended SCA handling */ static inline union ipte_control *kvm_s390_get_ipte_control(struct kvm *kvm) diff --git a/arch/s390/kvm/priv.c b/arch/s390/kvm/priv.c index 794503516bd4..fb4b494cde9b 100644 --- a/arch/s390/kvm/priv.c +++ b/arch/s390/kvm/priv.c @@ -54,7 +54,7 @@ int kvm_s390_handle_aa(struct kvm_vcpu *vcpu) static int handle_set_clock(struct kvm_vcpu *vcpu) { int rc; - ar_t ar; + u8 ar; u64 op2, val; if (vcpu->arch.sie_block->gpsw.mask & PSW_MASK_PSTATE) @@ -79,7 +79,7 @@ static int handle_set_prefix(struct kvm_vcpu *vcpu) u64 operand2; u32 address; int rc; - ar_t ar; + u8 ar; vcpu->stat.instruction_spx++; @@ -117,7 +117,7 @@ static int handle_store_prefix(struct kvm_vcpu *vcpu) u64 operand2; u32 address; int rc; - ar_t ar; + u8 ar; vcpu->stat.instruction_stpx++; @@ -147,7 +147,7 @@ static int handle_store_cpu_address(struct kvm_vcpu *vcpu) u16 vcpu_id = vcpu->vcpu_id; u64 ga; int rc; - ar_t ar; + u8 ar; vcpu->stat.instruction_stap++; @@ -380,7 +380,7 @@ static int handle_tpi(struct kvm_vcpu *vcpu) u32 tpi_data[3]; int rc; u64 addr; - ar_t ar; + u8 ar; addr = kvm_s390_get_base_disp_s(vcpu, &ar); if (addr & 3) @@ -548,7 +548,7 @@ int kvm_s390_handle_lpsw(struct kvm_vcpu *vcpu) psw_compat_t new_psw; u64 addr; int rc; - ar_t ar; + u8 ar; if (gpsw->mask & PSW_MASK_PSTATE) return kvm_s390_inject_program_int(vcpu, PGM_PRIVILEGED_OP); @@ -575,7 +575,7 @@ static int handle_lpswe(struct kvm_vcpu *vcpu) psw_t new_psw; u64 addr; int rc; - ar_t ar; + u8 ar; if (vcpu->arch.sie_block->gpsw.mask & PSW_MASK_PSTATE) return kvm_s390_inject_program_int(vcpu, PGM_PRIVILEGED_OP); @@ -597,7 +597,7 @@ static int handle_stidp(struct kvm_vcpu *vcpu) u64 stidp_data = vcpu->kvm->arch.model.cpuid; u64 operand2; int rc; - ar_t ar; + u8 ar; vcpu->stat.instruction_stidp++; @@ -644,7 +644,7 @@ static void handle_stsi_3_2_2(struct kvm_vcpu *vcpu, struct sysinfo_3_2_2 *mem) ASCEBC(mem->vm[0].cpi, 16); } -static void insert_stsi_usr_data(struct kvm_vcpu *vcpu, u64 addr, ar_t ar, +static void insert_stsi_usr_data(struct kvm_vcpu *vcpu, u64 addr, u8 ar, u8 fc, u8 sel1, u16 sel2) { vcpu->run->exit_reason = KVM_EXIT_S390_STSI; @@ -663,7 +663,7 @@ static int handle_stsi(struct kvm_vcpu *vcpu) unsigned long mem = 0; u64 operand2; int rc = 0; - ar_t ar; + u8 ar; vcpu->stat.instruction_stsi++; VCPU_EVENT(vcpu, 3, "STSI: fc: %u sel1: %u sel2: %u", fc, sel1, sel2); @@ -970,7 +970,7 @@ int kvm_s390_handle_lctl(struct kvm_vcpu *vcpu) int reg, rc, nr_regs; u32 ctl_array[16]; u64 ga; - ar_t ar; + u8 ar; vcpu->stat.instruction_lctl++; @@ -1009,7 +1009,7 @@ int kvm_s390_handle_stctl(struct kvm_vcpu *vcpu) int reg, rc, nr_regs; u32 ctl_array[16]; u64 ga; - ar_t ar; + u8 ar; vcpu->stat.instruction_stctl++; @@ -1043,7 +1043,7 @@ static int handle_lctlg(struct kvm_vcpu *vcpu) int reg, rc, nr_regs; u64 ctl_array[16]; u64 ga; - ar_t ar; + u8 ar; vcpu->stat.instruction_lctlg++; @@ -1081,7 +1081,7 @@ static int handle_stctg(struct kvm_vcpu *vcpu) int reg, rc, nr_regs; u64 ctl_array[16]; u64 ga; - ar_t ar; + u8 ar; vcpu->stat.instruction_stctg++; @@ -1132,7 +1132,7 @@ static int handle_tprot(struct kvm_vcpu *vcpu) unsigned long hva, gpa; int ret = 0, cc = 0; bool writable; - ar_t ar; + u8 ar; vcpu->stat.instruction_tprot++; diff --git a/arch/s390/kvm/vsie.c b/arch/s390/kvm/vsie.c index a9a9d974d9a4..38556e395915 100644 --- a/arch/s390/kvm/vsie.c +++ b/arch/s390/kvm/vsie.c @@ -324,6 +324,9 @@ static int shadow_scb(struct kvm_vcpu *vcpu, struct vsie_page *vsie_page) /* Run-time-Instrumentation */ if (test_kvm_facility(vcpu->kvm, 64)) scb_s->ecb3 |= scb_o->ecb3 & 0x01U; + /* Instruction Execution Prevention */ + if (test_kvm_facility(vcpu->kvm, 130)) + scb_s->ecb2 |= scb_o->ecb2 & 0x20U; if (test_kvm_cpu_feat(vcpu->kvm, KVM_S390_VM_CPU_FEAT_SIIF)) scb_s->eca |= scb_o->eca & 0x00000001U; if (test_kvm_cpu_feat(vcpu->kvm, KVM_S390_VM_CPU_FEAT_IB)) diff --git a/arch/s390/mm/pgtable.c b/arch/s390/mm/pgtable.c index beb90f3993e6..b48dc5f1900b 100644 --- a/arch/s390/mm/pgtable.c +++ b/arch/s390/mm/pgtable.c @@ -744,7 +744,7 @@ int reset_guest_reference_bit(struct mm_struct *mm, unsigned long addr) pgste_set_unlock(ptep, new); pte_unmap_unlock(ptep, ptl); - return 0; + return cc; } EXPORT_SYMBOL(reset_guest_reference_bit); diff --git a/arch/s390/tools/gen_facilities.c b/arch/s390/tools/gen_facilities.c index 8cc53b1e6d03..0cf802de52a1 100644 --- a/arch/s390/tools/gen_facilities.c +++ b/arch/s390/tools/gen_facilities.c @@ -80,6 +80,8 @@ static struct facility_def facility_defs[] = { 76, /* msa extension 3 */ 77, /* msa extension 4 */ 78, /* enhanced-DAT 2 */ + 130, /* instruction-execution-protection */ + 131, /* enhanced-SOP 2 and side-effect */ -1 /* END */ } }, diff --git a/arch/x86/include/asm/desc.h b/arch/x86/include/asm/desc.h index 12080d87da3b..cb8f9149f6c8 100644 --- a/arch/x86/include/asm/desc.h +++ b/arch/x86/include/asm/desc.h @@ -177,16 +177,8 @@ static inline void __set_tss_desc(unsigned cpu, unsigned int entry, void *addr) struct desc_struct *d = get_cpu_gdt_table(cpu); tss_desc tss; - /* - * sizeof(unsigned long) coming from an extra "long" at the end - * of the iobitmap. See tss_struct definition in processor.h - * - * -1? seg base+limit should be pointing to the address of the - * last valid byte - */ set_tssldt_descriptor(&tss, (unsigned long)addr, DESC_TSS, - IO_BITMAP_OFFSET + IO_BITMAP_BYTES + - sizeof(unsigned long) - 1); + __KERNEL_TSS_LIMIT); write_gdt_entry(d, entry, &tss, DESC_TSS); } @@ -213,6 +205,54 @@ static inline void native_load_tr_desc(void) asm volatile("ltr %w0"::"q" (GDT_ENTRY_TSS*8)); } +static inline void force_reload_TR(void) +{ + struct desc_struct *d = get_cpu_gdt_table(smp_processor_id()); + tss_desc tss; + + memcpy(&tss, &d[GDT_ENTRY_TSS], sizeof(tss_desc)); + + /* + * LTR requires an available TSS, and the TSS is currently + * busy. Make it be available so that LTR will work. + */ + tss.type = DESC_TSS; + write_gdt_entry(d, GDT_ENTRY_TSS, &tss, DESC_TSS); + + load_TR_desc(); +} + +DECLARE_PER_CPU(bool, need_tr_refresh); + +static inline void refresh_TR(void) +{ + DEBUG_LOCKS_WARN_ON(preemptible()); + + if (unlikely(this_cpu_read(need_tr_refresh))) { + force_reload_TR(); + this_cpu_write(need_tr_refresh, false); + } +} + +/* + * If you do something evil that corrupts the cached TSS limit (I'm looking + * at you, VMX exits), call this function. + * + * The optimization here is that the TSS limit only matters for Linux if the + * IO bitmap is in use. If the TSS limit gets forced to its minimum value, + * everything works except that IO bitmap will be ignored and all CPL 3 IO + * instructions will #GP, which is exactly what we want for normal tasks. + */ +static inline void invalidate_tss_limit(void) +{ + DEBUG_LOCKS_WARN_ON(preemptible()); + + if (unlikely(test_thread_flag(TIF_IO_BITMAP))) + force_reload_TR(); + else + this_cpu_write(need_tr_refresh, true); +} + static inline void native_load_gdt(const struct desc_ptr *dtr) { asm volatile("lgdt %0"::"m" (*dtr)); diff --git a/arch/x86/include/asm/kvm_emulate.h b/arch/x86/include/asm/kvm_emulate.h index e9cd7befcb76..3e8c287090e4 100644 --- a/arch/x86/include/asm/kvm_emulate.h +++ b/arch/x86/include/asm/kvm_emulate.h @@ -441,5 +441,6 @@ int emulator_task_switch(struct x86_emulate_ctxt *ctxt, int emulate_int_real(struct x86_emulate_ctxt *ctxt, int irq); void emulator_invalidate_register_cache(struct x86_emulate_ctxt *ctxt); void emulator_writeback_register_cache(struct x86_emulate_ctxt *ctxt); +bool emulator_can_use_gpa(struct x86_emulate_ctxt *ctxt); #endif /* _ASM_X86_KVM_X86_EMULATE_H */ diff --git a/arch/x86/include/asm/kvm_host.h b/arch/x86/include/asm/kvm_host.h index a7066dc1a7e9..74ef58c8ff53 100644 --- a/arch/x86/include/asm/kvm_host.h +++ b/arch/x86/include/asm/kvm_host.h @@ -55,7 +55,6 @@ #define KVM_REQ_TRIPLE_FAULT 10 #define KVM_REQ_MMU_SYNC 11 #define KVM_REQ_CLOCK_UPDATE 12 -#define KVM_REQ_DEACTIVATE_FPU 13 #define KVM_REQ_EVENT 14 #define KVM_REQ_APF_HALT 15 #define KVM_REQ_STEAL_UPDATE 16 @@ -115,7 +114,7 @@ static inline gfn_t gfn_to_index(gfn_t gfn, gfn_t base_gfn, int level) #define KVM_PERMILLE_MMU_PAGES 20 #define KVM_MIN_ALLOC_MMU_PAGES 64 -#define KVM_MMU_HASH_SHIFT 10 +#define KVM_MMU_HASH_SHIFT 12 #define KVM_NUM_MMU_PAGES (1 << KVM_MMU_HASH_SHIFT) #define KVM_MIN_FREE_MMU_PAGES 5 #define KVM_REFILL_PAGES 25 @@ -208,6 +207,13 @@ enum { PFERR_WRITE_MASK | \ PFERR_PRESENT_MASK) +/* + * The mask used to denote special SPTEs, which can be either MMIO SPTEs or + * Access Tracking SPTEs. We use bit 62 instead of bit 63 to avoid conflicting + * with the SVE bit in EPT PTEs. + */ +#define SPTE_SPECIAL_MASK (1ULL << 62) + /* apic attention bits */ #define KVM_APIC_CHECK_VAPIC 0 /* @@ -668,6 +674,9 @@ struct kvm_vcpu_arch { int pending_ioapic_eoi; int pending_external_vector; + + /* GPA available (AMD only) */ + bool gpa_available; }; struct kvm_lpage_info { @@ -716,6 +725,12 @@ struct kvm_hv { HV_REFERENCE_TSC_PAGE tsc_ref; }; +enum kvm_irqchip_mode { + KVM_IRQCHIP_NONE, + KVM_IRQCHIP_KERNEL, /* created with KVM_CREATE_IRQCHIP */ + KVM_IRQCHIP_SPLIT, /* created with KVM_CAP_SPLIT_IRQCHIP */ +}; + struct kvm_arch { unsigned int n_used_mmu_pages; unsigned int n_requested_mmu_pages; @@ -788,7 +803,7 @@ struct kvm_arch { u64 disabled_quirks; - bool irqchip_split; + enum kvm_irqchip_mode irqchip_mode; u8 nr_reserved_ioapic_pins; bool disabled_lapic_found; @@ -815,6 +830,7 @@ struct kvm_vm_stat { ulong mmu_unsync; ulong remote_tlb_flush; ulong lpages; + ulong max_mmu_page_hash_collisions; }; struct kvm_vcpu_stat { @@ -844,6 +860,7 @@ struct kvm_vcpu_stat { u64 hypercalls; u64 irq_injections; u64 nmi_injections; + u64 req_event; }; struct x86_instruction_info; @@ -918,8 +935,6 @@ struct kvm_x86_ops { unsigned long (*get_rflags)(struct kvm_vcpu *vcpu); void (*set_rflags)(struct kvm_vcpu *vcpu, unsigned long rflags); u32 (*get_pkru)(struct kvm_vcpu *vcpu); - void (*fpu_activate)(struct kvm_vcpu *vcpu); - void (*fpu_deactivate)(struct kvm_vcpu *vcpu); void (*tlb_flush)(struct kvm_vcpu *vcpu); @@ -951,7 +966,7 @@ struct kvm_x86_ops { void (*set_virtual_x2apic_mode)(struct kvm_vcpu *vcpu, bool set); void (*set_apic_access_page_addr)(struct kvm_vcpu *vcpu, hpa_t hpa); void (*deliver_posted_interrupt)(struct kvm_vcpu *vcpu, int vector); - void (*sync_pir_to_irr)(struct kvm_vcpu *vcpu); + int (*sync_pir_to_irr)(struct kvm_vcpu *vcpu); int (*set_tss_addr)(struct kvm *kvm, unsigned int addr); int (*get_tdp_level)(void); u64 (*get_mt_mask)(struct kvm_vcpu *vcpu, gfn_t gfn, bool is_mmio); @@ -1050,7 +1065,8 @@ void kvm_mmu_setup(struct kvm_vcpu *vcpu); void kvm_mmu_init_vm(struct kvm *kvm); void kvm_mmu_uninit_vm(struct kvm *kvm); void kvm_mmu_set_mask_ptes(u64 user_mask, u64 accessed_mask, - u64 dirty_mask, u64 nx_mask, u64 x_mask, u64 p_mask); + u64 dirty_mask, u64 nx_mask, u64 x_mask, u64 p_mask, + u64 acc_track_mask); void kvm_mmu_reset_context(struct kvm_vcpu *vcpu); void kvm_mmu_slot_remove_write_access(struct kvm *kvm, diff --git a/arch/x86/include/asm/kvmclock.h b/arch/x86/include/asm/kvmclock.h new file mode 100644 index 000000000000..f260bef63591 --- /dev/null +++ b/arch/x86/include/asm/kvmclock.h @@ -0,0 +1,6 @@ +#ifndef _ASM_X86_KVM_CLOCK_H +#define _ASM_X86_KVM_CLOCK_H + +extern struct clocksource kvm_clock; + +#endif /* _ASM_X86_KVM_CLOCK_H */ diff --git a/arch/x86/include/asm/paravirt.h b/arch/x86/include/asm/paravirt.h index 1eea6ca40694..f75fbfe550f2 100644 --- a/arch/x86/include/asm/paravirt.h +++ b/arch/x86/include/asm/paravirt.h @@ -673,7 +673,7 @@ static __always_inline void pv_kick(int cpu) PVOP_VCALL1(pv_lock_ops.kick, cpu); } -static __always_inline bool pv_vcpu_is_preempted(int cpu) +static __always_inline bool pv_vcpu_is_preempted(long cpu) { return PVOP_CALLEE1(bool, pv_lock_ops.vcpu_is_preempted, cpu); } diff --git a/arch/x86/include/asm/processor.h b/arch/x86/include/asm/processor.h index e6cfe7ba2d65..f385eca5407a 100644 --- a/arch/x86/include/asm/processor.h +++ b/arch/x86/include/asm/processor.h @@ -304,7 +304,7 @@ struct x86_hw_tss { u16 reserved5; u16 io_bitmap_base; -} __attribute__((packed)) ____cacheline_aligned; +} __attribute__((packed)); #endif /* @@ -342,6 +342,16 @@ struct tss_struct { DECLARE_PER_CPU_SHARED_ALIGNED(struct tss_struct, cpu_tss); +/* + * sizeof(unsigned long) coming from an extra "long" at the end + * of the iobitmap. + * + * -1? seg base+limit should be pointing to the address of the + * last valid byte + */ +#define __KERNEL_TSS_LIMIT \ + (IO_BITMAP_OFFSET + IO_BITMAP_BYTES + sizeof(unsigned long) - 1) + #ifdef CONFIG_X86_32 DECLARE_PER_CPU(unsigned long, cpu_current_top_of_stack); #endif diff --git a/arch/x86/include/asm/qspinlock.h b/arch/x86/include/asm/qspinlock.h index c343ab52579f..48a706f641f2 100644 --- a/arch/x86/include/asm/qspinlock.h +++ b/arch/x86/include/asm/qspinlock.h @@ -34,7 +34,7 @@ static inline void queued_spin_unlock(struct qspinlock *lock) } #define vcpu_is_preempted vcpu_is_preempted -static inline bool vcpu_is_preempted(int cpu) +static inline bool vcpu_is_preempted(long cpu) { return pv_vcpu_is_preempted(cpu); } diff --git a/arch/x86/include/asm/vmx.h b/arch/x86/include/asm/vmx.h index 2b5b2d4b924e..cc54b7026567 100644 --- a/arch/x86/include/asm/vmx.h +++ b/arch/x86/include/asm/vmx.h @@ -467,8 +467,16 @@ enum vmcs_field { #define VMX_EPT_WRITABLE_MASK 0x2ull #define VMX_EPT_EXECUTABLE_MASK 0x4ull #define VMX_EPT_IPAT_BIT (1ull << 6) -#define VMX_EPT_ACCESS_BIT (1ull << 8) -#define VMX_EPT_DIRTY_BIT (1ull << 9) +#define VMX_EPT_ACCESS_BIT (1ull << 8) +#define VMX_EPT_DIRTY_BIT (1ull << 9) +#define VMX_EPT_RWX_MASK (VMX_EPT_READABLE_MASK | \ + VMX_EPT_WRITABLE_MASK | \ + VMX_EPT_EXECUTABLE_MASK) +#define VMX_EPT_MT_MASK (7ull << VMX_EPT_MT_EPTE_SHIFT) + +/* The mask to use to trigger an EPT Misconfiguration in order to track MMIO */ +#define VMX_EPT_MISCONFIG_WX_VALUE (VMX_EPT_WRITABLE_MASK | \ + VMX_EPT_EXECUTABLE_MASK) #define VMX_EPT_IDENTITY_PAGETABLE_ADDR 0xfffbc000ul @@ -500,6 +508,22 @@ struct vmx_msr_entry { #define ENTRY_FAIL_VMCS_LINK_PTR 4 /* + * Exit Qualifications for EPT Violations + */ +#define EPT_VIOLATION_ACC_READ_BIT 0 +#define EPT_VIOLATION_ACC_WRITE_BIT 1 +#define EPT_VIOLATION_ACC_INSTR_BIT 2 +#define EPT_VIOLATION_READABLE_BIT 3 +#define EPT_VIOLATION_WRITABLE_BIT 4 +#define EPT_VIOLATION_EXECUTABLE_BIT 5 +#define EPT_VIOLATION_ACC_READ (1 << EPT_VIOLATION_ACC_READ_BIT) +#define EPT_VIOLATION_ACC_WRITE (1 << EPT_VIOLATION_ACC_WRITE_BIT) +#define EPT_VIOLATION_ACC_INSTR (1 << EPT_VIOLATION_ACC_INSTR_BIT) +#define EPT_VIOLATION_READABLE (1 << EPT_VIOLATION_READABLE_BIT) +#define EPT_VIOLATION_WRITABLE (1 << EPT_VIOLATION_WRITABLE_BIT) +#define EPT_VIOLATION_EXECUTABLE (1 << EPT_VIOLATION_EXECUTABLE_BIT) + +/* * VM-instruction error numbers */ enum vm_instruction_error_number { diff --git a/arch/x86/include/uapi/asm/kvm_para.h b/arch/x86/include/uapi/asm/kvm_para.h index 1421a6585126..cff0bb6556f8 100644 --- a/arch/x86/include/uapi/asm/kvm_para.h +++ b/arch/x86/include/uapi/asm/kvm_para.h @@ -50,6 +50,15 @@ struct kvm_steal_time { __u32 pad[11]; }; +#define KVM_CLOCK_PAIRING_WALLCLOCK 0 +struct kvm_clock_pairing { + __s64 sec; + __s64 nsec; + __u64 tsc; + __u32 flags; + __u32 pad[9]; +}; + #define KVM_STEAL_ALIGNMENT_BITS 5 #define KVM_STEAL_VALID_BITS ((-1ULL << (KVM_STEAL_ALIGNMENT_BITS + 1))) #define KVM_STEAL_RESERVED_MASK (((1 << KVM_STEAL_ALIGNMENT_BITS) - 1 ) << 1) diff --git a/arch/x86/kernel/asm-offsets_64.c b/arch/x86/kernel/asm-offsets_64.c index 210927ee2e74..99332f550c48 100644 --- a/arch/x86/kernel/asm-offsets_64.c +++ b/arch/x86/kernel/asm-offsets_64.c @@ -13,6 +13,10 @@ static char syscalls_ia32[] = { #include <asm/syscalls_32.h> }; +#if defined(CONFIG_KVM_GUEST) && defined(CONFIG_PARAVIRT_SPINLOCKS) +#include <asm/kvm_para.h> +#endif + int main(void) { #ifdef CONFIG_PARAVIRT @@ -22,6 +26,11 @@ int main(void) BLANK(); #endif +#if defined(CONFIG_KVM_GUEST) && defined(CONFIG_PARAVIRT_SPINLOCKS) + OFFSET(KVM_STEAL_TIME_preempted, kvm_steal_time, preempted); + BLANK(); +#endif + #define ENTRY(entry) OFFSET(pt_regs_ ## entry, pt_regs, entry) ENTRY(bx); ENTRY(cx); diff --git a/arch/x86/kernel/ioport.c b/arch/x86/kernel/ioport.c index 589b3193f102..b01bc8517450 100644 --- a/arch/x86/kernel/ioport.c +++ b/arch/x86/kernel/ioport.c @@ -16,6 +16,7 @@ #include <linux/syscalls.h> #include <linux/bitmap.h> #include <asm/syscalls.h> +#include <asm/desc.h> /* * this changes the io permissions bitmap in the current task. @@ -45,6 +46,10 @@ asmlinkage long sys_ioperm(unsigned long from, unsigned long num, int turn_on) memset(bitmap, 0xff, IO_BITMAP_BYTES); t->io_bitmap_ptr = bitmap; set_thread_flag(TIF_IO_BITMAP); + + preempt_disable(); + refresh_TR(); + preempt_enable(); } /* diff --git a/arch/x86/kernel/kvm.c b/arch/x86/kernel/kvm.c index 099fcba4981d..14f65a5f938e 100644 --- a/arch/x86/kernel/kvm.c +++ b/arch/x86/kernel/kvm.c @@ -589,7 +589,8 @@ out: local_irq_restore(flags); } -__visible bool __kvm_vcpu_is_preempted(int cpu) +#ifdef CONFIG_X86_32 +__visible bool __kvm_vcpu_is_preempted(long cpu) { struct kvm_steal_time *src = &per_cpu(steal_time, cpu); @@ -597,6 +598,29 @@ __visible bool __kvm_vcpu_is_preempted(int cpu) } PV_CALLEE_SAVE_REGS_THUNK(__kvm_vcpu_is_preempted); +#else + +#include <asm/asm-offsets.h> + +extern bool __raw_callee_save___kvm_vcpu_is_preempted(long); + +/* + * Hand-optimize version for x86-64 to avoid 8 64-bit register saving and + * restoring to/from the stack. + */ +asm( +".pushsection .text;" +".global __raw_callee_save___kvm_vcpu_is_preempted;" +".type __raw_callee_save___kvm_vcpu_is_preempted, @function;" +"__raw_callee_save___kvm_vcpu_is_preempted:" +"movq __per_cpu_offset(,%rdi,8), %rax;" +"cmpb $0, " __stringify(KVM_STEAL_TIME_preempted) "+steal_time(%rax);" +"setne %al;" +"ret;" +".popsection"); + +#endif + /* * Setup pv_lock_ops to exploit KVM_FEATURE_PV_UNHALT if present. */ diff --git a/arch/x86/kernel/kvmclock.c b/arch/x86/kernel/kvmclock.c index 542710b99f52..bae6ea6cfb94 100644 --- a/arch/x86/kernel/kvmclock.c +++ b/arch/x86/kernel/kvmclock.c @@ -28,6 +28,7 @@ #include <asm/x86_init.h> #include <asm/reboot.h> +#include <asm/kvmclock.h> static int kvmclock __ro_after_init = 1; static int msr_kvm_system_time = MSR_KVM_SYSTEM_TIME; @@ -49,6 +50,7 @@ struct pvclock_vsyscall_time_info *pvclock_pvti_cpu0_va(void) { return hv_clock; } +EXPORT_SYMBOL_GPL(pvclock_pvti_cpu0_va); /* * The wallclock is the time of day when we booted. Since then, some time may @@ -174,13 +176,14 @@ bool kvm_check_and_clear_guest_paused(void) return ret; } -static struct clocksource kvm_clock = { +struct clocksource kvm_clock = { .name = "kvm-clock", .read = kvm_clock_get_cycles, .rating = 400, .mask = CLOCKSOURCE_MASK(64), .flags = CLOCK_SOURCE_IS_CONTINUOUS, }; +EXPORT_SYMBOL_GPL(kvm_clock); int kvm_register_clock(char *txt) { diff --git a/arch/x86/kernel/paravirt-spinlocks.c b/arch/x86/kernel/paravirt-spinlocks.c index 6259327f3454..8f2d1c9d43a8 100644 --- a/arch/x86/kernel/paravirt-spinlocks.c +++ b/arch/x86/kernel/paravirt-spinlocks.c @@ -20,7 +20,7 @@ bool pv_is_native_spin_unlock(void) __raw_callee_save___native_queued_spin_unlock; } -__visible bool __native_vcpu_is_preempted(int cpu) +__visible bool __native_vcpu_is_preempted(long cpu) { return false; } diff --git a/arch/x86/kernel/process.c b/arch/x86/kernel/process.c index b615a1113f58..7780efa635b9 100644 --- a/arch/x86/kernel/process.c +++ b/arch/x86/kernel/process.c @@ -32,6 +32,7 @@ #include <asm/mce.h> #include <asm/vm86.h> #include <asm/switch_to.h> +#include <asm/desc.h> /* * per-CPU TSS segments. Threads are completely 'soft' on Linux, @@ -64,6 +65,9 @@ __visible DEFINE_PER_CPU_SHARED_ALIGNED(struct tss_struct, cpu_tss) = { }; EXPORT_PER_CPU_SYMBOL(cpu_tss); +DEFINE_PER_CPU(bool, need_tr_refresh); +EXPORT_PER_CPU_SYMBOL_GPL(need_tr_refresh); + /* * this gets called so that we can store lazy state into memory and copy the * current task into the new thread. @@ -209,6 +213,12 @@ void __switch_to_xtra(struct task_struct *prev_p, struct task_struct *next_p, */ memcpy(tss->io_bitmap, next->io_bitmap_ptr, max(prev->io_bitmap_max, next->io_bitmap_max)); + + /* + * Make sure that the TSS limit is correct for the CPU + * to notice the IO bitmap. + */ + refresh_TR(); } else if (test_tsk_thread_flag(prev_p, TIF_IO_BITMAP)) { /* * Clear any possible leftover bits: diff --git a/arch/x86/kvm/cpuid.c b/arch/x86/kvm/cpuid.c index e85f6bd7b9d5..1d155cc56629 100644 --- a/arch/x86/kvm/cpuid.c +++ b/arch/x86/kvm/cpuid.c @@ -123,8 +123,6 @@ int kvm_update_cpuid(struct kvm_vcpu *vcpu) if (best && (best->eax & (F(XSAVES) | F(XSAVEC)))) best->ebx = xstate_required_size(vcpu->arch.xcr0, true); - kvm_x86_ops->fpu_activate(vcpu); - /* * The existing code assumes virtual address is 48-bit in the canonical * address checks; exit if it is ever changed. @@ -383,7 +381,7 @@ static inline int __do_cpuid_ent(struct kvm_cpuid_entry2 *entry, u32 function, /* cpuid 7.0.ecx*/ const u32 kvm_cpuid_7_0_ecx_x86_features = - F(AVX512VBMI) | F(PKU) | 0 /*OSPKE*/; + F(AVX512VBMI) | F(PKU) | 0 /*OSPKE*/ | F(AVX512_VPOPCNTDQ); /* cpuid 7.0.edx*/ const u32 kvm_cpuid_7_0_edx_x86_features = @@ -861,12 +859,6 @@ void kvm_cpuid(struct kvm_vcpu *vcpu, u32 *eax, u32 *ebx, u32 *ecx, u32 *edx) if (!best) best = check_cpuid_limit(vcpu, function, index); - /* - * Perfmon not yet supported for L2 guest. - */ - if (is_guest_mode(vcpu) && function == 0xa) - best = NULL; - if (best) { *eax = best->eax; *ebx = best->ebx; diff --git a/arch/x86/kvm/emulate.c b/arch/x86/kvm/emulate.c index cedbba0f3402..45c7306c8780 100644 --- a/arch/x86/kvm/emulate.c +++ b/arch/x86/kvm/emulate.c @@ -173,6 +173,7 @@ #define NearBranch ((u64)1 << 52) /* Near branches */ #define No16 ((u64)1 << 53) /* No 16 bit operand */ #define IncSP ((u64)1 << 54) /* SP is incremented before ModRM calc */ +#define TwoMemOp ((u64)1 << 55) /* Instruction has two memory operand */ #define DstXacc (DstAccLo | SrcAccHi | SrcWrite) @@ -4298,7 +4299,7 @@ static const struct opcode group1[] = { }; static const struct opcode group1A[] = { - I(DstMem | SrcNone | Mov | Stack | IncSP, em_pop), N, N, N, N, N, N, N, + I(DstMem | SrcNone | Mov | Stack | IncSP | TwoMemOp, em_pop), N, N, N, N, N, N, N, }; static const struct opcode group2[] = { @@ -4336,7 +4337,7 @@ static const struct opcode group5[] = { I(SrcMemFAddr | ImplicitOps, em_call_far), I(SrcMem | NearBranch, em_jmp_abs), I(SrcMemFAddr | ImplicitOps, em_jmp_far), - I(SrcMem | Stack, em_push), D(Undefined), + I(SrcMem | Stack | TwoMemOp, em_push), D(Undefined), }; static const struct opcode group6[] = { @@ -4556,8 +4557,8 @@ static const struct opcode opcode_table[256] = { /* 0xA0 - 0xA7 */ I2bv(DstAcc | SrcMem | Mov | MemAbs, em_mov), I2bv(DstMem | SrcAcc | Mov | MemAbs | PageTable, em_mov), - I2bv(SrcSI | DstDI | Mov | String, em_mov), - F2bv(SrcSI | DstDI | String | NoWrite, em_cmp_r), + I2bv(SrcSI | DstDI | Mov | String | TwoMemOp, em_mov), + F2bv(SrcSI | DstDI | String | NoWrite | TwoMemOp, em_cmp_r), /* 0xA8 - 0xAF */ F2bv(DstAcc | SrcImm | NoWrite, em_test), I2bv(SrcAcc | DstDI | Mov | String, em_mov), @@ -5671,3 +5672,14 @@ void emulator_writeback_register_cache(struct x86_emulate_ctxt *ctxt) { writeback_registers(ctxt); } + +bool emulator_can_use_gpa(struct x86_emulate_ctxt *ctxt) +{ + if (ctxt->rep_prefix && (ctxt->d & String)) + return false; + + if (ctxt->d & TwoMemOp) + return false; + + return true; +} diff --git a/arch/x86/kvm/hyperv.c b/arch/x86/kvm/hyperv.c index 2ecd7dab4631..f701d4430727 100644 --- a/arch/x86/kvm/hyperv.c +++ b/arch/x86/kvm/hyperv.c @@ -305,13 +305,13 @@ static int synic_set_irq(struct kvm_vcpu_hv_synic *synic, u32 sint) return -ENOENT; memset(&irq, 0, sizeof(irq)); - irq.dest_id = kvm_apic_id(vcpu->arch.apic); + irq.shorthand = APIC_DEST_SELF; irq.dest_mode = APIC_DEST_PHYSICAL; irq.delivery_mode = APIC_DM_FIXED; irq.vector = vector; irq.level = 1; - ret = kvm_irq_delivery_to_apic(vcpu->kvm, NULL, &irq, NULL); + ret = kvm_irq_delivery_to_apic(vcpu->kvm, vcpu->arch.apic, &irq, NULL); trace_kvm_hv_synic_set_irq(vcpu->vcpu_id, sint, irq.vector, ret); return ret; } diff --git a/arch/x86/kvm/i8259.c b/arch/x86/kvm/i8259.c index 7cc2360f1848..73ea24d4f119 100644 --- a/arch/x86/kvm/i8259.c +++ b/arch/x86/kvm/i8259.c @@ -598,14 +598,14 @@ static const struct kvm_io_device_ops picdev_eclr_ops = { .write = picdev_eclr_write, }; -struct kvm_pic *kvm_create_pic(struct kvm *kvm) +int kvm_pic_init(struct kvm *kvm) { struct kvm_pic *s; int ret; s = kzalloc(sizeof(struct kvm_pic), GFP_KERNEL); if (!s) - return NULL; + return -ENOMEM; spin_lock_init(&s->lock); s->kvm = kvm; s->pics[0].elcr_mask = 0xf8; @@ -635,7 +635,9 @@ struct kvm_pic *kvm_create_pic(struct kvm *kvm) mutex_unlock(&kvm->slots_lock); - return s; + kvm->arch.vpic = s; + + return 0; fail_unreg_1: kvm_io_bus_unregister_dev(kvm, KVM_PIO_BUS, &s->dev_slave); @@ -648,13 +650,17 @@ fail_unlock: kfree(s); - return NULL; + return ret; } -void kvm_destroy_pic(struct kvm_pic *vpic) +void kvm_pic_destroy(struct kvm *kvm) { + struct kvm_pic *vpic = kvm->arch.vpic; + kvm_io_bus_unregister_dev(vpic->kvm, KVM_PIO_BUS, &vpic->dev_master); kvm_io_bus_unregister_dev(vpic->kvm, KVM_PIO_BUS, &vpic->dev_slave); kvm_io_bus_unregister_dev(vpic->kvm, KVM_PIO_BUS, &vpic->dev_eclr); + + kvm->arch.vpic = NULL; kfree(vpic); } diff --git a/arch/x86/kvm/irq.h b/arch/x86/kvm/irq.h index 035731eb3897..40d5b2cf6061 100644 --- a/arch/x86/kvm/irq.h +++ b/arch/x86/kvm/irq.h @@ -73,8 +73,8 @@ struct kvm_pic { unsigned long irq_states[PIC_NUM_PINS]; }; -struct kvm_pic *kvm_create_pic(struct kvm *kvm); -void kvm_destroy_pic(struct kvm_pic *vpic); +int kvm_pic_init(struct kvm *kvm); +void kvm_pic_destroy(struct kvm *kvm); int kvm_pic_read_irq(struct kvm *kvm); void kvm_pic_update_irq(struct kvm_pic *s); @@ -93,18 +93,19 @@ static inline int pic_in_kernel(struct kvm *kvm) static inline int irqchip_split(struct kvm *kvm) { - return kvm->arch.irqchip_split; + return kvm->arch.irqchip_mode == KVM_IRQCHIP_SPLIT; } -static inline int irqchip_in_kernel(struct kvm *kvm) +static inline int irqchip_kernel(struct kvm *kvm) { - struct kvm_pic *vpic = pic_irqchip(kvm); - bool ret; + return kvm->arch.irqchip_mode == KVM_IRQCHIP_KERNEL; +} - ret = (vpic != NULL); - ret |= irqchip_split(kvm); +static inline int irqchip_in_kernel(struct kvm *kvm) +{ + bool ret = kvm->arch.irqchip_mode != KVM_IRQCHIP_NONE; - /* Read vpic before kvm->irq_routing. */ + /* Matches with wmb after initializing kvm->irq_routing. */ smp_rmb(); return ret; } diff --git a/arch/x86/kvm/irq_comm.c b/arch/x86/kvm/irq_comm.c index 6c0191615f23..b96d3893f121 100644 --- a/arch/x86/kvm/irq_comm.c +++ b/arch/x86/kvm/irq_comm.c @@ -41,15 +41,6 @@ static int kvm_set_pic_irq(struct kvm_kernel_irq_routing_entry *e, bool line_status) { struct kvm_pic *pic = pic_irqchip(kvm); - - /* - * XXX: rejecting pic routes when pic isn't in use would be better, - * but the default routing table is installed while kvm->arch.vpic is - * NULL and KVM_CREATE_IRQCHIP can race with KVM_IRQ_LINE. - */ - if (!pic) - return -1; - return kvm_pic_set_irq(pic, e->irqchip.pin, irq_source_id, level); } @@ -58,10 +49,6 @@ static int kvm_set_ioapic_irq(struct kvm_kernel_irq_routing_entry *e, bool line_status) { struct kvm_ioapic *ioapic = kvm->arch.vioapic; - - if (!ioapic) - return -1; - return kvm_ioapic_set_irq(ioapic, e->irqchip.pin, irq_source_id, level, line_status); } @@ -297,16 +284,20 @@ int kvm_set_routing_entry(struct kvm *kvm, case KVM_IRQ_ROUTING_IRQCHIP: delta = 0; switch (ue->u.irqchip.irqchip) { - case KVM_IRQCHIP_PIC_MASTER: - e->set = kvm_set_pic_irq; - max_pin = PIC_NUM_PINS; - break; case KVM_IRQCHIP_PIC_SLAVE: + delta = 8; + /* fall through */ + case KVM_IRQCHIP_PIC_MASTER: + if (!pic_in_kernel(kvm)) + goto out; + e->set = kvm_set_pic_irq; max_pin = PIC_NUM_PINS; - delta = 8; break; case KVM_IRQCHIP_IOAPIC: + if (!ioapic_in_kernel(kvm)) + goto out; + max_pin = KVM_IOAPIC_NUM_PINS; e->set = kvm_set_ioapic_irq; break; @@ -409,7 +400,7 @@ int kvm_setup_empty_irq_routing(struct kvm *kvm) void kvm_arch_post_irq_routing_update(struct kvm *kvm) { - if (ioapic_in_kernel(kvm) || !irqchip_in_kernel(kvm)) + if (!irqchip_split(kvm)) return; kvm_make_scan_ioapic_request(kvm); } diff --git a/arch/x86/kvm/lapic.c b/arch/x86/kvm/lapic.c index 2f6ef5121a4c..bad6a25067bc 100644 --- a/arch/x86/kvm/lapic.c +++ b/arch/x86/kvm/lapic.c @@ -115,6 +115,16 @@ static inline int apic_enabled(struct kvm_lapic *apic) (LVT_MASK | APIC_MODE_MASK | APIC_INPUT_POLARITY | \ APIC_LVT_REMOTE_IRR | APIC_LVT_LEVEL_TRIGGER) +static inline u8 kvm_xapic_id(struct kvm_lapic *apic) +{ + return kvm_lapic_get_reg(apic, APIC_ID) >> 24; +} + +static inline u32 kvm_x2apic_id(struct kvm_lapic *apic) +{ + return apic->vcpu->vcpu_id; +} + static inline bool kvm_apic_map_get_logical_dest(struct kvm_apic_map *map, u32 dest_id, struct kvm_lapic ***cluster, u16 *mask) { switch (map->mode) { @@ -159,13 +169,13 @@ static void recalculate_apic_map(struct kvm *kvm) struct kvm_apic_map *new, *old = NULL; struct kvm_vcpu *vcpu; int i; - u32 max_id = 255; + u32 max_id = 255; /* enough space for any xAPIC ID */ mutex_lock(&kvm->arch.apic_map_lock); kvm_for_each_vcpu(i, vcpu, kvm) if (kvm_apic_present(vcpu)) - max_id = max(max_id, kvm_apic_id(vcpu->arch.apic)); + max_id = max(max_id, kvm_x2apic_id(vcpu->arch.apic)); new = kvm_kvzalloc(sizeof(struct kvm_apic_map) + sizeof(struct kvm_lapic *) * ((u64)max_id + 1)); @@ -179,16 +189,28 @@ static void recalculate_apic_map(struct kvm *kvm) struct kvm_lapic *apic = vcpu->arch.apic; struct kvm_lapic **cluster; u16 mask; - u32 ldr, aid; + u32 ldr; + u8 xapic_id; + u32 x2apic_id; if (!kvm_apic_present(vcpu)) continue; - aid = kvm_apic_id(apic); - ldr = kvm_lapic_get_reg(apic, APIC_LDR); + xapic_id = kvm_xapic_id(apic); + x2apic_id = kvm_x2apic_id(apic); - if (aid <= new->max_apic_id) - new->phys_map[aid] = apic; + /* Hotplug hack: see kvm_apic_match_physical_addr(), ... */ + if ((apic_x2apic_mode(apic) || x2apic_id > 0xff) && + x2apic_id <= new->max_apic_id) + new->phys_map[x2apic_id] = apic; + /* + * ... xAPIC ID of VCPUs with APIC ID > 0xff will wrap-around, + * prevent them from masking VCPUs with APIC ID <= 0xff. + */ + if (!apic_x2apic_mode(apic) && !new->phys_map[xapic_id]) + new->phys_map[xapic_id] = apic; + + ldr = kvm_lapic_get_reg(apic, APIC_LDR); if (apic_x2apic_mode(apic)) { new->mode |= KVM_APIC_MODE_X2APIC; @@ -250,6 +272,8 @@ static inline void kvm_apic_set_x2apic_id(struct kvm_lapic *apic, u32 id) { u32 ldr = ((id >> 4) << 16) | (1 << (id & 0xf)); + WARN_ON_ONCE(id != apic->vcpu->vcpu_id); + kvm_lapic_set_reg(apic, APIC_ID, id); kvm_lapic_set_reg(apic, APIC_LDR, ldr); recalculate_apic_map(apic->vcpu->kvm); @@ -317,7 +341,7 @@ static int find_highest_vector(void *bitmap) vec >= 0; vec -= APIC_VECTORS_PER_REG) { reg = bitmap + REG_POS(vec); if (*reg) - return fls(*reg) - 1 + vec; + return __fls(*reg) + vec; } return -1; @@ -337,27 +361,32 @@ static u8 count_vectors(void *bitmap) return count; } -void __kvm_apic_update_irr(u32 *pir, void *regs) +int __kvm_apic_update_irr(u32 *pir, void *regs) { - u32 i, pir_val; + u32 i, vec; + u32 pir_val, irr_val; + int max_irr = -1; - for (i = 0; i <= 7; i++) { + for (i = vec = 0; i <= 7; i++, vec += 32) { pir_val = READ_ONCE(pir[i]); + irr_val = *((u32 *)(regs + APIC_IRR + i * 0x10)); if (pir_val) { - pir_val = xchg(&pir[i], 0); - *((u32 *)(regs + APIC_IRR + i * 0x10)) |= pir_val; + irr_val |= xchg(&pir[i], 0); + *((u32 *)(regs + APIC_IRR + i * 0x10)) = irr_val; } + if (irr_val) + max_irr = __fls(irr_val) + vec; } + + return max_irr; } EXPORT_SYMBOL_GPL(__kvm_apic_update_irr); -void kvm_apic_update_irr(struct kvm_vcpu *vcpu, u32 *pir) +int kvm_apic_update_irr(struct kvm_vcpu *vcpu, u32 *pir) { struct kvm_lapic *apic = vcpu->arch.apic; - __kvm_apic_update_irr(pir, apic->regs); - - kvm_make_request(KVM_REQ_EVENT, vcpu); + return __kvm_apic_update_irr(pir, apic->regs); } EXPORT_SYMBOL_GPL(kvm_apic_update_irr); @@ -377,8 +406,6 @@ static inline int apic_find_highest_irr(struct kvm_lapic *apic) if (!apic->irr_pending) return -1; - if (apic->vcpu->arch.apicv_active) - kvm_x86_ops->sync_pir_to_irr(apic->vcpu); result = apic_search_irr(apic); ASSERT(result == -1 || result >= 16); @@ -392,9 +419,10 @@ static inline void apic_clear_irr(int vec, struct kvm_lapic *apic) vcpu = apic->vcpu; if (unlikely(vcpu->arch.apicv_active)) { - /* try to update RVI */ + /* need to update RVI */ apic_clear_vector(vec, apic->regs + APIC_IRR); - kvm_make_request(KVM_REQ_EVENT, vcpu); + kvm_x86_ops->hwapic_irr_update(vcpu, + apic_find_highest_irr(apic)); } else { apic->irr_pending = false; apic_clear_vector(vec, apic->regs + APIC_IRR); @@ -484,6 +512,7 @@ int kvm_lapic_find_highest_irr(struct kvm_vcpu *vcpu) */ return apic_find_highest_irr(vcpu->arch.apic); } +EXPORT_SYMBOL_GPL(kvm_lapic_find_highest_irr); static int __apic_accept_irq(struct kvm_lapic *apic, int delivery_mode, int vector, int level, int trig_mode, @@ -500,16 +529,14 @@ int kvm_apic_set_irq(struct kvm_vcpu *vcpu, struct kvm_lapic_irq *irq, static int pv_eoi_put_user(struct kvm_vcpu *vcpu, u8 val) { - - return kvm_write_guest_cached(vcpu->kvm, &vcpu->arch.pv_eoi.data, &val, - sizeof(val)); + return kvm_vcpu_write_guest_cached(vcpu, &vcpu->arch.pv_eoi.data, &val, + sizeof(val)); } static int pv_eoi_get_user(struct kvm_vcpu *vcpu, u8 *val) { - - return kvm_read_guest_cached(vcpu->kvm, &vcpu->arch.pv_eoi.data, val, - sizeof(*val)); + return kvm_vcpu_read_guest_cached(vcpu, &vcpu->arch.pv_eoi.data, val, + sizeof(*val)); } static inline bool pv_eoi_enabled(struct kvm_vcpu *vcpu) @@ -546,7 +573,19 @@ static void pv_eoi_clr_pending(struct kvm_vcpu *vcpu) __clear_bit(KVM_APIC_PV_EOI_PENDING, &vcpu->arch.apic_attention); } -static void apic_update_ppr(struct kvm_lapic *apic) +static int apic_has_interrupt_for_ppr(struct kvm_lapic *apic, u32 ppr) +{ + int highest_irr; + if (kvm_x86_ops->sync_pir_to_irr && apic->vcpu->arch.apicv_active) + highest_irr = kvm_x86_ops->sync_pir_to_irr(apic->vcpu); + else + highest_irr = apic_find_highest_irr(apic); + if (highest_irr == -1 || (highest_irr & 0xF0) <= ppr) + return -1; + return highest_irr; +} + +static bool __apic_update_ppr(struct kvm_lapic *apic, u32 *new_ppr) { u32 tpr, isrv, ppr, old_ppr; int isr; @@ -564,13 +603,28 @@ static void apic_update_ppr(struct kvm_lapic *apic) apic_debug("vlapic %p, ppr 0x%x, isr 0x%x, isrv 0x%x", apic, ppr, isr, isrv); - if (old_ppr != ppr) { + *new_ppr = ppr; + if (old_ppr != ppr) kvm_lapic_set_reg(apic, APIC_PROCPRI, ppr); - if (ppr < old_ppr) - kvm_make_request(KVM_REQ_EVENT, apic->vcpu); - } + + return ppr < old_ppr; +} + +static void apic_update_ppr(struct kvm_lapic *apic) +{ + u32 ppr; + + if (__apic_update_ppr(apic, &ppr) && + apic_has_interrupt_for_ppr(apic, ppr) != -1) + kvm_make_request(KVM_REQ_EVENT, apic->vcpu); } +void kvm_apic_update_ppr(struct kvm_vcpu *vcpu) +{ + apic_update_ppr(vcpu->arch.apic); +} +EXPORT_SYMBOL_GPL(kvm_apic_update_ppr); + static void apic_set_tpr(struct kvm_lapic *apic, u32 tpr) { kvm_lapic_set_reg(apic, APIC_TASKPRI, tpr); @@ -579,10 +633,8 @@ static void apic_set_tpr(struct kvm_lapic *apic, u32 tpr) static bool kvm_apic_broadcast(struct kvm_lapic *apic, u32 mda) { - if (apic_x2apic_mode(apic)) - return mda == X2APIC_BROADCAST; - - return GET_APIC_DEST_FIELD(mda) == APIC_BROADCAST; + return mda == (apic_x2apic_mode(apic) ? + X2APIC_BROADCAST : APIC_BROADCAST); } static bool kvm_apic_match_physical_addr(struct kvm_lapic *apic, u32 mda) @@ -591,9 +643,18 @@ static bool kvm_apic_match_physical_addr(struct kvm_lapic *apic, u32 mda) return true; if (apic_x2apic_mode(apic)) - return mda == kvm_apic_id(apic); + return mda == kvm_x2apic_id(apic); - return mda == SET_APIC_DEST_FIELD(kvm_apic_id(apic)); + /* + * Hotplug hack: Make LAPIC in xAPIC mode also accept interrupts as if + * it were in x2APIC mode. Hotplugged VCPUs start in xAPIC mode and + * this allows unique addressing of VCPUs with APIC ID over 0xff. + * The 0xff condition is needed because writeable xAPIC ID. + */ + if (kvm_x2apic_id(apic) > 0xff && mda == kvm_x2apic_id(apic)) + return true; + + return mda == kvm_xapic_id(apic); } static bool kvm_apic_match_logical_addr(struct kvm_lapic *apic, u32 mda) @@ -610,7 +671,6 @@ static bool kvm_apic_match_logical_addr(struct kvm_lapic *apic, u32 mda) && (logical_id & mda & 0xffff) != 0; logical_id = GET_APIC_LOGICAL_ID(logical_id); - mda = GET_APIC_DEST_FIELD(mda); switch (kvm_lapic_get_reg(apic, APIC_DFR)) { case APIC_DFR_FLAT: @@ -627,9 +687,9 @@ static bool kvm_apic_match_logical_addr(struct kvm_lapic *apic, u32 mda) /* The KVM local APIC implementation has two quirks: * - * - the xAPIC MDA stores the destination at bits 24-31, while this - * is not true of struct kvm_lapic_irq's dest_id field. This is - * just a quirk in the API and is not problematic. + * - Real hardware delivers interrupts destined to x2APIC ID > 0xff to LAPICs + * in xAPIC mode if the "destination & 0xff" matches its xAPIC ID. + * KVM doesn't do that aliasing. * * - in-kernel IOAPIC messages have to be delivered directly to * x2APIC, because the kernel does not support interrupt remapping. @@ -645,13 +705,12 @@ static u32 kvm_apic_mda(struct kvm_vcpu *vcpu, unsigned int dest_id, struct kvm_lapic *source, struct kvm_lapic *target) { bool ipi = source != NULL; - bool x2apic_mda = apic_x2apic_mode(ipi ? source : target); if (!vcpu->kvm->arch.x2apic_broadcast_quirk_disabled && - !ipi && dest_id == APIC_BROADCAST && x2apic_mda) + !ipi && dest_id == APIC_BROADCAST && apic_x2apic_mode(target)) return X2APIC_BROADCAST; - return x2apic_mda ? dest_id : SET_APIC_DEST_FIELD(dest_id); + return dest_id; } bool kvm_apic_match_dest(struct kvm_vcpu *vcpu, struct kvm_lapic *source, @@ -1907,9 +1966,9 @@ void kvm_lapic_reset(struct kvm_vcpu *vcpu, bool init_event) vcpu->arch.apic_arb_prio = 0; vcpu->arch.apic_attention = 0; - apic_debug("%s: vcpu=%p, id=%d, base_msr=" + apic_debug("%s: vcpu=%p, id=0x%x, base_msr=" "0x%016" PRIx64 ", base_address=0x%0lx.\n", __func__, - vcpu, kvm_apic_id(apic), + vcpu, kvm_lapic_get_reg(apic, APIC_ID), vcpu->arch.apic_base, apic->base_address); } @@ -2021,17 +2080,13 @@ nomem: int kvm_apic_has_interrupt(struct kvm_vcpu *vcpu) { struct kvm_lapic *apic = vcpu->arch.apic; - int highest_irr; + u32 ppr; if (!apic_enabled(apic)) return -1; - apic_update_ppr(apic); - highest_irr = apic_find_highest_irr(apic); - if ((highest_irr == -1) || - ((highest_irr & 0xF0) <= kvm_lapic_get_reg(apic, APIC_PROCPRI))) - return -1; - return highest_irr; + __apic_update_ppr(apic, &ppr); + return apic_has_interrupt_for_ppr(apic, ppr); } int kvm_apic_accept_pic_intr(struct kvm_vcpu *vcpu) @@ -2067,6 +2122,7 @@ int kvm_get_apic_interrupt(struct kvm_vcpu *vcpu) { int vector = kvm_apic_has_interrupt(vcpu); struct kvm_lapic *apic = vcpu->arch.apic; + u32 ppr; if (vector == -1) return -1; @@ -2078,13 +2134,23 @@ int kvm_get_apic_interrupt(struct kvm_vcpu *vcpu) * because the process would deliver it through the IDT. */ - apic_set_isr(vector, apic); - apic_update_ppr(apic); apic_clear_irr(vector, apic); - if (test_bit(vector, vcpu_to_synic(vcpu)->auto_eoi_bitmap)) { - apic_clear_isr(vector, apic); + /* + * For auto-EOI interrupts, there might be another pending + * interrupt above PPR, so check whether to raise another + * KVM_REQ_EVENT. + */ apic_update_ppr(apic); + } else { + /* + * For normal interrupts, PPR has been raised and there cannot + * be a higher-priority pending interrupt---except if there was + * a concurrent interrupt injection, but that would have + * triggered KVM_REQ_EVENT already. + */ + apic_set_isr(vector, apic); + __apic_update_ppr(apic, &ppr); } return vector; @@ -2145,8 +2211,7 @@ int kvm_apic_set_state(struct kvm_vcpu *vcpu, struct kvm_lapic_state *s) 1 : count_vectors(apic->regs + APIC_ISR); apic->highest_isr_cache = -1; if (vcpu->arch.apicv_active) { - if (kvm_x86_ops->apicv_post_state_restore) - kvm_x86_ops->apicv_post_state_restore(vcpu); + kvm_x86_ops->apicv_post_state_restore(vcpu); kvm_x86_ops->hwapic_irr_update(vcpu, apic_find_highest_irr(apic)); kvm_x86_ops->hwapic_isr_update(vcpu, @@ -2220,8 +2285,8 @@ void kvm_lapic_sync_from_vapic(struct kvm_vcpu *vcpu) if (!test_bit(KVM_APIC_CHECK_VAPIC, &vcpu->arch.apic_attention)) return; - if (kvm_read_guest_cached(vcpu->kvm, &vcpu->arch.apic->vapic_cache, &data, - sizeof(u32))) + if (kvm_vcpu_read_guest_cached(vcpu, &vcpu->arch.apic->vapic_cache, &data, + sizeof(u32))) return; apic_set_tpr(vcpu->arch.apic, data & 0xff); @@ -2273,14 +2338,14 @@ void kvm_lapic_sync_to_vapic(struct kvm_vcpu *vcpu) max_isr = 0; data = (tpr & 0xff) | ((max_isr & 0xf0) << 8) | (max_irr << 24); - kvm_write_guest_cached(vcpu->kvm, &vcpu->arch.apic->vapic_cache, &data, - sizeof(u32)); + kvm_vcpu_write_guest_cached(vcpu, &vcpu->arch.apic->vapic_cache, &data, + sizeof(u32)); } int kvm_lapic_set_vapic_addr(struct kvm_vcpu *vcpu, gpa_t vapic_addr) { if (vapic_addr) { - if (kvm_gfn_to_hva_cache_init(vcpu->kvm, + if (kvm_vcpu_gfn_to_hva_cache_init(vcpu, &vcpu->arch.apic->vapic_cache, vapic_addr, sizeof(u32))) return -EINVAL; @@ -2374,7 +2439,7 @@ int kvm_lapic_enable_pv_eoi(struct kvm_vcpu *vcpu, u64 data) vcpu->arch.pv_eoi.msr_val = data; if (!pv_eoi_enabled(vcpu)) return 0; - return kvm_gfn_to_hva_cache_init(vcpu->kvm, &vcpu->arch.pv_eoi.data, + return kvm_vcpu_gfn_to_hva_cache_init(vcpu, &vcpu->arch.pv_eoi.data, addr, sizeof(u8)); } diff --git a/arch/x86/kvm/lapic.h b/arch/x86/kvm/lapic.h index ff8039d61672..bcbe811f3b97 100644 --- a/arch/x86/kvm/lapic.h +++ b/arch/x86/kvm/lapic.h @@ -71,8 +71,9 @@ int kvm_lapic_reg_read(struct kvm_lapic *apic, u32 offset, int len, bool kvm_apic_match_dest(struct kvm_vcpu *vcpu, struct kvm_lapic *source, int short_hand, unsigned int dest, int dest_mode); -void __kvm_apic_update_irr(u32 *pir, void *regs); -void kvm_apic_update_irr(struct kvm_vcpu *vcpu, u32 *pir); +int __kvm_apic_update_irr(u32 *pir, void *regs); +int kvm_apic_update_irr(struct kvm_vcpu *vcpu, u32 *pir); +void kvm_apic_update_ppr(struct kvm_vcpu *vcpu); int kvm_apic_set_irq(struct kvm_vcpu *vcpu, struct kvm_lapic_irq *irq, struct dest_map *dest_map); int kvm_apic_local_deliver(struct kvm_lapic *apic, int lvt_type); @@ -203,17 +204,6 @@ static inline int kvm_lapic_latched_init(struct kvm_vcpu *vcpu) return lapic_in_kernel(vcpu) && test_bit(KVM_APIC_INIT, &vcpu->arch.apic->pending_events); } -static inline u32 kvm_apic_id(struct kvm_lapic *apic) -{ - /* To avoid a race between apic_base and following APIC_ID update when - * switching to x2apic_mode, the x2apic mode returns initial x2apic id. - */ - if (apic_x2apic_mode(apic)) - return apic->vcpu->vcpu_id; - - return kvm_lapic_get_reg(apic, APIC_ID) >> 24; -} - bool kvm_apic_pending_eoi(struct kvm_vcpu *vcpu, int vector); void wait_lapic_expire(struct kvm_vcpu *vcpu); diff --git a/arch/x86/kvm/mmu.c b/arch/x86/kvm/mmu.c index 7012de4a1fed..2fd7586aad4d 100644 --- a/arch/x86/kvm/mmu.c +++ b/arch/x86/kvm/mmu.c @@ -37,6 +37,8 @@ #include <linux/srcu.h> #include <linux/slab.h> #include <linux/uaccess.h> +#include <linux/hash.h> +#include <linux/kern_levels.h> #include <asm/page.h> #include <asm/cmpxchg.h> @@ -129,6 +131,10 @@ module_param(dbg, bool, 0644); #define ACC_USER_MASK PT_USER_MASK #define ACC_ALL (ACC_EXEC_MASK | ACC_WRITE_MASK | ACC_USER_MASK) +/* The mask for the R/X bits in EPT PTEs */ +#define PT64_EPT_READABLE_MASK 0x1ull +#define PT64_EPT_EXECUTABLE_MASK 0x4ull + #include <trace/events/kvm.h> #define CREATE_TRACE_POINTS @@ -178,15 +184,40 @@ static u64 __read_mostly shadow_dirty_mask; static u64 __read_mostly shadow_mmio_mask; static u64 __read_mostly shadow_present_mask; +/* + * The mask/value to distinguish a PTE that has been marked not-present for + * access tracking purposes. + * The mask would be either 0 if access tracking is disabled, or + * SPTE_SPECIAL_MASK|VMX_EPT_RWX_MASK if access tracking is enabled. + */ +static u64 __read_mostly shadow_acc_track_mask; +static const u64 shadow_acc_track_value = SPTE_SPECIAL_MASK; + +/* + * The mask/shift to use for saving the original R/X bits when marking the PTE + * as not-present for access tracking purposes. We do not save the W bit as the + * PTEs being access tracked also need to be dirty tracked, so the W bit will be + * restored only when a write is attempted to the page. + */ +static const u64 shadow_acc_track_saved_bits_mask = PT64_EPT_READABLE_MASK | + PT64_EPT_EXECUTABLE_MASK; +static const u64 shadow_acc_track_saved_bits_shift = PT64_SECOND_AVAIL_BITS_SHIFT; + static void mmu_spte_set(u64 *sptep, u64 spte); static void mmu_free_roots(struct kvm_vcpu *vcpu); void kvm_mmu_set_mmio_spte_mask(u64 mmio_mask) { - shadow_mmio_mask = mmio_mask; + shadow_mmio_mask = mmio_mask | SPTE_SPECIAL_MASK; } EXPORT_SYMBOL_GPL(kvm_mmu_set_mmio_spte_mask); +static inline bool is_access_track_spte(u64 spte) +{ + /* Always false if shadow_acc_track_mask is zero. */ + return (spte & shadow_acc_track_mask) == shadow_acc_track_value; +} + /* * the low bit of the generation number is always presumed to be zero. * This disables mmio caching during memslot updates. The concept is @@ -284,17 +315,35 @@ static bool check_mmio_spte(struct kvm_vcpu *vcpu, u64 spte) } void kvm_mmu_set_mask_ptes(u64 user_mask, u64 accessed_mask, - u64 dirty_mask, u64 nx_mask, u64 x_mask, u64 p_mask) + u64 dirty_mask, u64 nx_mask, u64 x_mask, u64 p_mask, + u64 acc_track_mask) { + if (acc_track_mask != 0) + acc_track_mask |= SPTE_SPECIAL_MASK; + shadow_user_mask = user_mask; shadow_accessed_mask = accessed_mask; shadow_dirty_mask = dirty_mask; shadow_nx_mask = nx_mask; shadow_x_mask = x_mask; shadow_present_mask = p_mask; + shadow_acc_track_mask = acc_track_mask; + WARN_ON(shadow_accessed_mask != 0 && shadow_acc_track_mask != 0); } EXPORT_SYMBOL_GPL(kvm_mmu_set_mask_ptes); +void kvm_mmu_clear_all_pte_masks(void) +{ + shadow_user_mask = 0; + shadow_accessed_mask = 0; + shadow_dirty_mask = 0; + shadow_nx_mask = 0; + shadow_x_mask = 0; + shadow_mmio_mask = 0; + shadow_present_mask = 0; + shadow_acc_track_mask = 0; +} + static int is_cpuid_PSE36(void) { return 1; @@ -307,7 +356,7 @@ static int is_nx(struct kvm_vcpu *vcpu) static int is_shadow_present_pte(u64 pte) { - return (pte & 0xFFFFFFFFull) && !is_mmio_spte(pte); + return (pte != 0) && !is_mmio_spte(pte); } static int is_large_pte(u64 pte) @@ -324,6 +373,11 @@ static int is_last_spte(u64 pte, int level) return 0; } +static bool is_executable_pte(u64 spte) +{ + return (spte & (shadow_x_mask | shadow_nx_mask)) == shadow_x_mask; +} + static kvm_pfn_t spte_to_pfn(u64 pte) { return (pte & PT64_BASE_ADDR_MASK) >> PAGE_SHIFT; @@ -473,7 +527,7 @@ retry: } #endif -static bool spte_is_locklessly_modifiable(u64 spte) +static bool spte_can_locklessly_be_made_writable(u64 spte) { return (spte & (SPTE_HOST_WRITEABLE | SPTE_MMU_WRITEABLE)) == (SPTE_HOST_WRITEABLE | SPTE_MMU_WRITEABLE); @@ -481,36 +535,38 @@ static bool spte_is_locklessly_modifiable(u64 spte) static bool spte_has_volatile_bits(u64 spte) { + if (!is_shadow_present_pte(spte)) + return false; + /* * Always atomically update spte if it can be updated * out of mmu-lock, it can ensure dirty bit is not lost, * also, it can help us to get a stable is_writable_pte() * to ensure tlb flush is not missed. */ - if (spte_is_locklessly_modifiable(spte)) + if (spte_can_locklessly_be_made_writable(spte) || + is_access_track_spte(spte)) return true; - if (!shadow_accessed_mask) - return false; - - if (!is_shadow_present_pte(spte)) - return false; - - if ((spte & shadow_accessed_mask) && - (!is_writable_pte(spte) || (spte & shadow_dirty_mask))) - return false; + if (shadow_accessed_mask) { + if ((spte & shadow_accessed_mask) == 0 || + (is_writable_pte(spte) && (spte & shadow_dirty_mask) == 0)) + return true; + } - return true; + return false; } -static bool spte_is_bit_cleared(u64 old_spte, u64 new_spte, u64 bit_mask) +static bool is_accessed_spte(u64 spte) { - return (old_spte & bit_mask) && !(new_spte & bit_mask); + return shadow_accessed_mask ? spte & shadow_accessed_mask + : !is_access_track_spte(spte); } -static bool spte_is_bit_changed(u64 old_spte, u64 new_spte, u64 bit_mask) +static bool is_dirty_spte(u64 spte) { - return (old_spte & bit_mask) != (new_spte & bit_mask); + return shadow_dirty_mask ? spte & shadow_dirty_mask + : spte & PT_WRITABLE_MASK; } /* Rules for using mmu_spte_set: @@ -525,25 +581,19 @@ static void mmu_spte_set(u64 *sptep, u64 new_spte) __set_spte(sptep, new_spte); } -/* Rules for using mmu_spte_update: - * Update the state bits, it means the mapped pfn is not changed. - * - * Whenever we overwrite a writable spte with a read-only one we - * should flush remote TLBs. Otherwise rmap_write_protect - * will find a read-only spte, even though the writable spte - * might be cached on a CPU's TLB, the return value indicates this - * case. +/* + * Update the SPTE (excluding the PFN), but do not track changes in its + * accessed/dirty status. */ -static bool mmu_spte_update(u64 *sptep, u64 new_spte) +static u64 mmu_spte_update_no_track(u64 *sptep, u64 new_spte) { u64 old_spte = *sptep; - bool ret = false; WARN_ON(!is_shadow_present_pte(new_spte)); if (!is_shadow_present_pte(old_spte)) { mmu_spte_set(sptep, new_spte); - return ret; + return old_spte; } if (!spte_has_volatile_bits(old_spte)) @@ -551,45 +601,62 @@ static bool mmu_spte_update(u64 *sptep, u64 new_spte) else old_spte = __update_clear_spte_slow(sptep, new_spte); + WARN_ON(spte_to_pfn(old_spte) != spte_to_pfn(new_spte)); + + return old_spte; +} + +/* Rules for using mmu_spte_update: + * Update the state bits, it means the mapped pfn is not changed. + * + * Whenever we overwrite a writable spte with a read-only one we + * should flush remote TLBs. Otherwise rmap_write_protect + * will find a read-only spte, even though the writable spte + * might be cached on a CPU's TLB, the return value indicates this + * case. + * + * Returns true if the TLB needs to be flushed + */ +static bool mmu_spte_update(u64 *sptep, u64 new_spte) +{ + bool flush = false; + u64 old_spte = mmu_spte_update_no_track(sptep, new_spte); + + if (!is_shadow_present_pte(old_spte)) + return false; + /* * For the spte updated out of mmu-lock is safe, since * we always atomically update it, see the comments in * spte_has_volatile_bits(). */ - if (spte_is_locklessly_modifiable(old_spte) && + if (spte_can_locklessly_be_made_writable(old_spte) && !is_writable_pte(new_spte)) - ret = true; - - if (!shadow_accessed_mask) { - /* - * We don't set page dirty when dropping non-writable spte. - * So do it now if the new spte is becoming non-writable. - */ - if (ret) - kvm_set_pfn_dirty(spte_to_pfn(old_spte)); - return ret; - } + flush = true; /* - * Flush TLB when accessed/dirty bits are changed in the page tables, + * Flush TLB when accessed/dirty states are changed in the page tables, * to guarantee consistency between TLB and page tables. */ - if (spte_is_bit_changed(old_spte, new_spte, - shadow_accessed_mask | shadow_dirty_mask)) - ret = true; - if (spte_is_bit_cleared(old_spte, new_spte, shadow_accessed_mask)) + if (is_accessed_spte(old_spte) && !is_accessed_spte(new_spte)) { + flush = true; kvm_set_pfn_accessed(spte_to_pfn(old_spte)); - if (spte_is_bit_cleared(old_spte, new_spte, shadow_dirty_mask)) + } + + if (is_dirty_spte(old_spte) && !is_dirty_spte(new_spte)) { + flush = true; kvm_set_pfn_dirty(spte_to_pfn(old_spte)); + } - return ret; + return flush; } /* * Rules for using mmu_spte_clear_track_bits: * It sets the sptep from present to nonpresent, and track the * state bits, it is used to clear the last level sptep. + * Returns non-zero if the PTE was previously valid. */ static int mmu_spte_clear_track_bits(u64 *sptep) { @@ -613,11 +680,12 @@ static int mmu_spte_clear_track_bits(u64 *sptep) */ WARN_ON(!kvm_is_reserved_pfn(pfn) && !page_count(pfn_to_page(pfn))); - if (!shadow_accessed_mask || old_spte & shadow_accessed_mask) + if (is_accessed_spte(old_spte)) kvm_set_pfn_accessed(pfn); - if (old_spte & (shadow_dirty_mask ? shadow_dirty_mask : - PT_WRITABLE_MASK)) + + if (is_dirty_spte(old_spte)) kvm_set_pfn_dirty(pfn); + return 1; } @@ -636,6 +704,78 @@ static u64 mmu_spte_get_lockless(u64 *sptep) return __get_spte_lockless(sptep); } +static u64 mark_spte_for_access_track(u64 spte) +{ + if (shadow_accessed_mask != 0) + return spte & ~shadow_accessed_mask; + + if (shadow_acc_track_mask == 0 || is_access_track_spte(spte)) + return spte; + + /* + * Making an Access Tracking PTE will result in removal of write access + * from the PTE. So, verify that we will be able to restore the write + * access in the fast page fault path later on. + */ + WARN_ONCE((spte & PT_WRITABLE_MASK) && + !spte_can_locklessly_be_made_writable(spte), + "kvm: Writable SPTE is not locklessly dirty-trackable\n"); + + WARN_ONCE(spte & (shadow_acc_track_saved_bits_mask << + shadow_acc_track_saved_bits_shift), + "kvm: Access Tracking saved bit locations are not zero\n"); + + spte |= (spte & shadow_acc_track_saved_bits_mask) << + shadow_acc_track_saved_bits_shift; + spte &= ~shadow_acc_track_mask; + spte |= shadow_acc_track_value; + + return spte; +} + +/* Restore an acc-track PTE back to a regular PTE */ +static u64 restore_acc_track_spte(u64 spte) +{ + u64 new_spte = spte; + u64 saved_bits = (spte >> shadow_acc_track_saved_bits_shift) + & shadow_acc_track_saved_bits_mask; + + WARN_ON_ONCE(!is_access_track_spte(spte)); + + new_spte &= ~shadow_acc_track_mask; + new_spte &= ~(shadow_acc_track_saved_bits_mask << + shadow_acc_track_saved_bits_shift); + new_spte |= saved_bits; + + return new_spte; +} + +/* Returns the Accessed status of the PTE and resets it at the same time. */ +static bool mmu_spte_age(u64 *sptep) +{ + u64 spte = mmu_spte_get_lockless(sptep); + + if (!is_accessed_spte(spte)) + return false; + + if (shadow_accessed_mask) { + clear_bit((ffs(shadow_accessed_mask) - 1), + (unsigned long *)sptep); + } else { + /* + * Capture the dirty status of the page, so that it doesn't get + * lost when the SPTE is marked for access tracking. + */ + if (is_writable_pte(spte)) + kvm_set_pfn_dirty(spte_to_pfn(spte)); + + spte = mark_spte_for_access_track(spte); + mmu_spte_update_no_track(sptep, spte); + } + + return true; +} + static void walk_shadow_page_lockless_begin(struct kvm_vcpu *vcpu) { /* @@ -1212,7 +1352,7 @@ static bool spte_write_protect(u64 *sptep, bool pt_protect) u64 spte = *sptep; if (!is_writable_pte(spte) && - !(pt_protect && spte_is_locklessly_modifiable(spte))) + !(pt_protect && spte_can_locklessly_be_made_writable(spte))) return false; rmap_printk("rmap_write_protect: spte %p %llx\n", sptep, *sptep); @@ -1420,7 +1560,7 @@ static int kvm_set_pte_rmapp(struct kvm *kvm, struct kvm_rmap_head *rmap_head, restart: for_each_rmap_spte(rmap_head, &iter, sptep) { rmap_printk("kvm_set_pte_rmapp: spte %p %llx gfn %llx (%d)\n", - sptep, *sptep, gfn, level); + sptep, *sptep, gfn, level); need_flush = 1; @@ -1433,7 +1573,8 @@ restart: new_spte &= ~PT_WRITABLE_MASK; new_spte &= ~SPTE_HOST_WRITEABLE; - new_spte &= ~shadow_accessed_mask; + + new_spte = mark_spte_for_access_track(new_spte); mmu_spte_clear_track_bits(sptep); mmu_spte_set(sptep, new_spte); @@ -1595,15 +1736,8 @@ static int kvm_age_rmapp(struct kvm *kvm, struct kvm_rmap_head *rmap_head, struct rmap_iterator uninitialized_var(iter); int young = 0; - BUG_ON(!shadow_accessed_mask); - - for_each_rmap_spte(rmap_head, &iter, sptep) { - if (*sptep & shadow_accessed_mask) { - young = 1; - clear_bit((ffs(shadow_accessed_mask) - 1), - (unsigned long *)sptep); - } - } + for_each_rmap_spte(rmap_head, &iter, sptep) + young |= mmu_spte_age(sptep); trace_kvm_age_page(gfn, level, slot, young); return young; @@ -1615,24 +1749,20 @@ static int kvm_test_age_rmapp(struct kvm *kvm, struct kvm_rmap_head *rmap_head, { u64 *sptep; struct rmap_iterator iter; - int young = 0; /* - * If there's no access bit in the secondary pte set by the - * hardware it's up to gup-fast/gup to set the access bit in - * the primary pte or in the page structure. + * If there's no access bit in the secondary pte set by the hardware and + * fast access tracking is also not enabled, it's up to gup-fast/gup to + * set the access bit in the primary pte or in the page structure. */ - if (!shadow_accessed_mask) + if (!shadow_accessed_mask && !shadow_acc_track_mask) goto out; - for_each_rmap_spte(rmap_head, &iter, sptep) { - if (*sptep & shadow_accessed_mask) { - young = 1; - break; - } - } + for_each_rmap_spte(rmap_head, &iter, sptep) + if (is_accessed_spte(*sptep)) + return 1; out: - return young; + return 0; } #define RMAP_RECYCLE_THRESHOLD 1000 @@ -1660,7 +1790,7 @@ int kvm_age_hva(struct kvm *kvm, unsigned long start, unsigned long end) * This has some overhead, but not as much as the cost of swapping * out actively used pages or breaking up actively used hugepages. */ - if (!shadow_accessed_mask) + if (!shadow_accessed_mask && !shadow_acc_track_mask) return kvm_handle_hva_range(kvm, start, end, 0, kvm_unmap_rmapp); @@ -1713,7 +1843,7 @@ static void kvm_mmu_free_page(struct kvm_mmu_page *sp) static unsigned kvm_page_table_hashfn(gfn_t gfn) { - return gfn & ((1 << KVM_MMU_HASH_SHIFT) - 1); + return hash_64(gfn, KVM_MMU_HASH_SHIFT); } static void mmu_page_add_parent_pte(struct kvm_vcpu *vcpu, @@ -1904,17 +2034,17 @@ static void kvm_mmu_commit_zap_page(struct kvm *kvm, * since it has been deleted from active_mmu_pages but still can be found * at hast list. * - * for_each_gfn_valid_sp() has skipped that kind of pages. + * for_each_valid_sp() has skipped that kind of pages. */ -#define for_each_gfn_valid_sp(_kvm, _sp, _gfn) \ +#define for_each_valid_sp(_kvm, _sp, _gfn) \ hlist_for_each_entry(_sp, \ &(_kvm)->arch.mmu_page_hash[kvm_page_table_hashfn(_gfn)], hash_link) \ - if ((_sp)->gfn != (_gfn) || is_obsolete_sp((_kvm), (_sp)) \ - || (_sp)->role.invalid) {} else + if (is_obsolete_sp((_kvm), (_sp)) || (_sp)->role.invalid) { \ + } else #define for_each_gfn_indirect_valid_sp(_kvm, _sp, _gfn) \ - for_each_gfn_valid_sp(_kvm, _sp, _gfn) \ - if ((_sp)->role.direct) {} else + for_each_valid_sp(_kvm, _sp, _gfn) \ + if ((_sp)->gfn != (_gfn) || (_sp)->role.direct) {} else /* @sp->gfn should be write-protected at the call site */ static bool __kvm_sync_page(struct kvm_vcpu *vcpu, struct kvm_mmu_page *sp, @@ -2116,6 +2246,7 @@ static struct kvm_mmu_page *kvm_mmu_get_page(struct kvm_vcpu *vcpu, struct kvm_mmu_page *sp; bool need_sync = false; bool flush = false; + int collisions = 0; LIST_HEAD(invalid_list); role = vcpu->arch.mmu.base_role; @@ -2130,7 +2261,12 @@ static struct kvm_mmu_page *kvm_mmu_get_page(struct kvm_vcpu *vcpu, quadrant &= (1 << ((PT32_PT_BITS - PT64_PT_BITS) * level)) - 1; role.quadrant = quadrant; } - for_each_gfn_valid_sp(vcpu->kvm, sp, gfn) { + for_each_valid_sp(vcpu->kvm, sp, gfn) { + if (sp->gfn != gfn) { + collisions++; + continue; + } + if (!need_sync && sp->unsync) need_sync = true; @@ -2153,7 +2289,7 @@ static struct kvm_mmu_page *kvm_mmu_get_page(struct kvm_vcpu *vcpu, __clear_sp_write_flooding_count(sp); trace_kvm_mmu_get_page(sp, false); - return sp; + goto out; } ++vcpu->kvm->stat.mmu_cache_miss; @@ -2183,6 +2319,9 @@ static struct kvm_mmu_page *kvm_mmu_get_page(struct kvm_vcpu *vcpu, trace_kvm_mmu_get_page(sp, true); kvm_mmu_flush_or_zap(vcpu, &invalid_list, false, flush); +out: + if (collisions > vcpu->kvm->stat.max_mmu_page_hash_collisions) + vcpu->kvm->stat.max_mmu_page_hash_collisions = collisions; return sp; } @@ -2583,6 +2722,9 @@ static int set_spte(struct kvm_vcpu *vcpu, u64 *sptep, spte |= shadow_dirty_mask; } + if (speculative) + spte = mark_spte_for_access_track(spte); + set_pte: if (mmu_spte_update(sptep, spte)) kvm_flush_remote_tlbs(vcpu->kvm); @@ -2636,7 +2778,7 @@ static bool mmu_set_spte(struct kvm_vcpu *vcpu, u64 *sptep, unsigned pte_access, pgprintk("%s: setting spte %llx\n", __func__, *sptep); pgprintk("instantiating %s PTE (%s) at %llx (%llx) addr %p\n", is_large_pte(*sptep)? "2MB" : "4kB", - *sptep & PT_PRESENT_MASK ?"RW":"R", gfn, + *sptep & PT_WRITABLE_MASK ? "RW" : "R", gfn, *sptep, sptep); if (!was_rmapped && is_large_pte(*sptep)) ++vcpu->kvm->stat.lpages; @@ -2869,33 +3011,43 @@ static bool page_fault_can_be_fast(u32 error_code) if (unlikely(error_code & PFERR_RSVD_MASK)) return false; + /* See if the page fault is due to an NX violation */ + if (unlikely(((error_code & (PFERR_FETCH_MASK | PFERR_PRESENT_MASK)) + == (PFERR_FETCH_MASK | PFERR_PRESENT_MASK)))) + return false; + /* - * #PF can be fast only if the shadow page table is present and it - * is caused by write-protect, that means we just need change the - * W bit of the spte which can be done out of mmu-lock. + * #PF can be fast if: + * 1. The shadow page table entry is not present, which could mean that + * the fault is potentially caused by access tracking (if enabled). + * 2. The shadow page table entry is present and the fault + * is caused by write-protect, that means we just need change the W + * bit of the spte which can be done out of mmu-lock. + * + * However, if access tracking is disabled we know that a non-present + * page must be a genuine page fault where we have to create a new SPTE. + * So, if access tracking is disabled, we return true only for write + * accesses to a present page. */ - if (!(error_code & PFERR_PRESENT_MASK) || - !(error_code & PFERR_WRITE_MASK)) - return false; - return true; + return shadow_acc_track_mask != 0 || + ((error_code & (PFERR_WRITE_MASK | PFERR_PRESENT_MASK)) + == (PFERR_WRITE_MASK | PFERR_PRESENT_MASK)); } +/* + * Returns true if the SPTE was fixed successfully. Otherwise, + * someone else modified the SPTE from its original value. + */ static bool fast_pf_fix_direct_spte(struct kvm_vcpu *vcpu, struct kvm_mmu_page *sp, - u64 *sptep, u64 spte) + u64 *sptep, u64 old_spte, u64 new_spte) { gfn_t gfn; WARN_ON(!sp->role.direct); /* - * The gfn of direct spte is stable since it is calculated - * by sp->gfn. - */ - gfn = kvm_mmu_page_get_gfn(sp, sptep - sp->spt); - - /* * Theoretically we could also set dirty bit (and flush TLB) here in * order to eliminate unnecessary PML logging. See comments in * set_spte. But fast_page_fault is very unlikely to happen with PML @@ -2907,12 +3059,33 @@ fast_pf_fix_direct_spte(struct kvm_vcpu *vcpu, struct kvm_mmu_page *sp, * * Compare with set_spte where instead shadow_dirty_mask is set. */ - if (cmpxchg64(sptep, spte, spte | PT_WRITABLE_MASK) == spte) + if (cmpxchg64(sptep, old_spte, new_spte) != old_spte) + return false; + + if (is_writable_pte(new_spte) && !is_writable_pte(old_spte)) { + /* + * The gfn of direct spte is stable since it is + * calculated by sp->gfn. + */ + gfn = kvm_mmu_page_get_gfn(sp, sptep - sp->spt); kvm_vcpu_mark_page_dirty(vcpu, gfn); + } return true; } +static bool is_access_allowed(u32 fault_err_code, u64 spte) +{ + if (fault_err_code & PFERR_FETCH_MASK) + return is_executable_pte(spte); + + if (fault_err_code & PFERR_WRITE_MASK) + return is_writable_pte(spte); + + /* Fault was on Read access */ + return spte & PT_PRESENT_MASK; +} + /* * Return value: * - true: let the vcpu to access on the same address again. @@ -2923,8 +3096,9 @@ static bool fast_page_fault(struct kvm_vcpu *vcpu, gva_t gva, int level, { struct kvm_shadow_walk_iterator iterator; struct kvm_mmu_page *sp; - bool ret = false; + bool fault_handled = false; u64 spte = 0ull; + uint retry_count = 0; if (!VALID_PAGE(vcpu->arch.mmu.root_hpa)) return false; @@ -2933,66 +3107,93 @@ static bool fast_page_fault(struct kvm_vcpu *vcpu, gva_t gva, int level, return false; walk_shadow_page_lockless_begin(vcpu); - for_each_shadow_entry_lockless(vcpu, gva, iterator, spte) - if (!is_shadow_present_pte(spte) || iterator.level < level) + + do { + u64 new_spte; + + for_each_shadow_entry_lockless(vcpu, gva, iterator, spte) + if (!is_shadow_present_pte(spte) || + iterator.level < level) + break; + + sp = page_header(__pa(iterator.sptep)); + if (!is_last_spte(spte, sp->role.level)) break; - /* - * If the mapping has been changed, let the vcpu fault on the - * same address again. - */ - if (!is_shadow_present_pte(spte)) { - ret = true; - goto exit; - } + /* + * Check whether the memory access that caused the fault would + * still cause it if it were to be performed right now. If not, + * then this is a spurious fault caused by TLB lazily flushed, + * or some other CPU has already fixed the PTE after the + * current CPU took the fault. + * + * Need not check the access of upper level table entries since + * they are always ACC_ALL. + */ + if (is_access_allowed(error_code, spte)) { + fault_handled = true; + break; + } - sp = page_header(__pa(iterator.sptep)); - if (!is_last_spte(spte, sp->role.level)) - goto exit; + new_spte = spte; - /* - * Check if it is a spurious fault caused by TLB lazily flushed. - * - * Need not check the access of upper level table entries since - * they are always ACC_ALL. - */ - if (is_writable_pte(spte)) { - ret = true; - goto exit; - } + if (is_access_track_spte(spte)) + new_spte = restore_acc_track_spte(new_spte); - /* - * Currently, to simplify the code, only the spte write-protected - * by dirty-log can be fast fixed. - */ - if (!spte_is_locklessly_modifiable(spte)) - goto exit; + /* + * Currently, to simplify the code, write-protection can + * be removed in the fast path only if the SPTE was + * write-protected for dirty-logging or access tracking. + */ + if ((error_code & PFERR_WRITE_MASK) && + spte_can_locklessly_be_made_writable(spte)) + { + new_spte |= PT_WRITABLE_MASK; - /* - * Do not fix write-permission on the large spte since we only dirty - * the first page into the dirty-bitmap in fast_pf_fix_direct_spte() - * that means other pages are missed if its slot is dirty-logged. - * - * Instead, we let the slow page fault path create a normal spte to - * fix the access. - * - * See the comments in kvm_arch_commit_memory_region(). - */ - if (sp->role.level > PT_PAGE_TABLE_LEVEL) - goto exit; + /* + * Do not fix write-permission on the large spte. Since + * we only dirty the first page into the dirty-bitmap in + * fast_pf_fix_direct_spte(), other pages are missed + * if its slot has dirty logging enabled. + * + * Instead, we let the slow page fault path create a + * normal spte to fix the access. + * + * See the comments in kvm_arch_commit_memory_region(). + */ + if (sp->role.level > PT_PAGE_TABLE_LEVEL) + break; + } + + /* Verify that the fault can be handled in the fast path */ + if (new_spte == spte || + !is_access_allowed(error_code, new_spte)) + break; + + /* + * Currently, fast page fault only works for direct mapping + * since the gfn is not stable for indirect shadow page. See + * Documentation/virtual/kvm/locking.txt to get more detail. + */ + fault_handled = fast_pf_fix_direct_spte(vcpu, sp, + iterator.sptep, spte, + new_spte); + if (fault_handled) + break; + + if (++retry_count > 4) { + printk_once(KERN_WARNING + "kvm: Fast #PF retrying more than 4 times.\n"); + break; + } + + } while (true); - /* - * Currently, fast page fault only works for direct mapping since - * the gfn is not stable for indirect shadow page. - * See Documentation/virtual/kvm/locking.txt to get more detail. - */ - ret = fast_pf_fix_direct_spte(vcpu, sp, iterator.sptep, spte); -exit: trace_fast_page_fault(vcpu, gva, error_code, iterator.sptep, - spte, ret); + spte, fault_handled); walk_shadow_page_lockless_end(vcpu); - return ret; + return fault_handled; } static bool try_async_pf(struct kvm_vcpu *vcpu, bool prefault, gfn_t gfn, @@ -5063,6 +5264,8 @@ static void mmu_destroy_caches(void) int kvm_mmu_module_init(void) { + kvm_mmu_clear_all_pte_masks(); + pte_list_desc_cache = kmem_cache_create("pte_list_desc", sizeof(struct pte_list_desc), 0, 0, NULL); diff --git a/arch/x86/kvm/svm.c b/arch/x86/kvm/svm.c index 08a4d3ab3455..d1efe2c62b3f 100644 --- a/arch/x86/kvm/svm.c +++ b/arch/x86/kvm/svm.c @@ -971,8 +971,8 @@ static void svm_disable_lbrv(struct vcpu_svm *svm) * a particular vCPU. */ #define SVM_VM_DATA_HASH_BITS 8 -DECLARE_HASHTABLE(svm_vm_data_hash, SVM_VM_DATA_HASH_BITS); -static spinlock_t svm_vm_data_hash_lock; +static DEFINE_HASHTABLE(svm_vm_data_hash, SVM_VM_DATA_HASH_BITS); +static DEFINE_SPINLOCK(svm_vm_data_hash_lock); /* Note: * This function is called from IOMMU driver to notify @@ -1077,8 +1077,6 @@ static __init int svm_hardware_setup(void) } else { pr_info("AVIC enabled\n"); - hash_init(svm_vm_data_hash); - spin_lock_init(&svm_vm_data_hash_lock); amd_iommu_register_ga_log_notifier(&avic_ga_log_notifier); } } @@ -1159,7 +1157,6 @@ static void init_vmcb(struct vcpu_svm *svm) struct vmcb_control_area *control = &svm->vmcb->control; struct vmcb_save_area *save = &svm->vmcb->save; - svm->vcpu.fpu_active = 1; svm->vcpu.arch.hflags = 0; set_cr_intercept(svm, INTERCEPT_CR0_READ); @@ -1901,15 +1898,12 @@ static void update_cr0_intercept(struct vcpu_svm *svm) ulong gcr0 = svm->vcpu.arch.cr0; u64 *hcr0 = &svm->vmcb->save.cr0; - if (!svm->vcpu.fpu_active) - *hcr0 |= SVM_CR0_SELECTIVE_MASK; - else - *hcr0 = (*hcr0 & ~SVM_CR0_SELECTIVE_MASK) - | (gcr0 & SVM_CR0_SELECTIVE_MASK); + *hcr0 = (*hcr0 & ~SVM_CR0_SELECTIVE_MASK) + | (gcr0 & SVM_CR0_SELECTIVE_MASK); mark_dirty(svm->vmcb, VMCB_CR); - if (gcr0 == *hcr0 && svm->vcpu.fpu_active) { + if (gcr0 == *hcr0) { clr_cr_intercept(svm, INTERCEPT_CR0_READ); clr_cr_intercept(svm, INTERCEPT_CR0_WRITE); } else { @@ -1940,8 +1934,6 @@ static void svm_set_cr0(struct kvm_vcpu *vcpu, unsigned long cr0) if (!npt_enabled) cr0 |= X86_CR0_PG | X86_CR0_WP; - if (!vcpu->fpu_active) - cr0 |= X86_CR0_TS; /* * re-enable caching here because the QEMU bios * does not do it - this results in some delay at @@ -2160,22 +2152,6 @@ static int ac_interception(struct vcpu_svm *svm) return 1; } -static void svm_fpu_activate(struct kvm_vcpu *vcpu) -{ - struct vcpu_svm *svm = to_svm(vcpu); - - clr_exception_intercept(svm, NM_VECTOR); - - svm->vcpu.fpu_active = 1; - update_cr0_intercept(svm); -} - -static int nm_interception(struct vcpu_svm *svm) -{ - svm_fpu_activate(&svm->vcpu); - return 1; -} - static bool is_erratum_383(void) { int err, i; @@ -2573,9 +2549,6 @@ static int nested_svm_exit_special(struct vcpu_svm *svm) if (!npt_enabled && svm->apf_reason == 0) return NESTED_EXIT_HOST; break; - case SVM_EXIT_EXCP_BASE + NM_VECTOR: - nm_interception(svm); - break; default: break; } @@ -4020,7 +3993,6 @@ static int (*const svm_exit_handlers[])(struct vcpu_svm *svm) = { [SVM_EXIT_EXCP_BASE + BP_VECTOR] = bp_interception, [SVM_EXIT_EXCP_BASE + UD_VECTOR] = ud_interception, [SVM_EXIT_EXCP_BASE + PF_VECTOR] = pf_interception, - [SVM_EXIT_EXCP_BASE + NM_VECTOR] = nm_interception, [SVM_EXIT_EXCP_BASE + MC_VECTOR] = mc_interception, [SVM_EXIT_EXCP_BASE + AC_VECTOR] = ac_interception, [SVM_EXIT_INTR] = intr_interception, @@ -4182,6 +4154,8 @@ static int handle_exit(struct kvm_vcpu *vcpu) trace_kvm_exit(exit_code, vcpu, KVM_ISA_SVM); + vcpu->arch.gpa_available = (exit_code == SVM_EXIT_NPF); + if (!is_cr_intercept(svm, INTERCEPT_CR0_WRITE)) vcpu->arch.cr0 = svm->vmcb->save.cr0; if (npt_enabled) @@ -4357,11 +4331,6 @@ static void svm_load_eoi_exitmap(struct kvm_vcpu *vcpu, u64 *eoi_exit_bitmap) return; } -static void svm_sync_pir_to_irr(struct kvm_vcpu *vcpu) -{ - return; -} - static void svm_deliver_avic_intr(struct kvm_vcpu *vcpu, int vec) { kvm_lapic_set_irr(vec, vcpu->arch.apic); @@ -5077,14 +5046,6 @@ static bool svm_has_wbinvd_exit(void) return true; } -static void svm_fpu_deactivate(struct kvm_vcpu *vcpu) -{ - struct vcpu_svm *svm = to_svm(vcpu); - - set_exception_intercept(svm, NM_VECTOR); - update_cr0_intercept(svm); -} - #define PRE_EX(exit) { .exit_code = (exit), \ .stage = X86_ICPT_PRE_EXCEPT, } #define POST_EX(exit) { .exit_code = (exit), \ @@ -5345,9 +5306,6 @@ static struct kvm_x86_ops svm_x86_ops __ro_after_init = { .get_pkru = svm_get_pkru, - .fpu_activate = svm_fpu_activate, - .fpu_deactivate = svm_fpu_deactivate, - .tlb_flush = svm_flush_tlb, .run = svm_vcpu_run, @@ -5371,7 +5329,6 @@ static struct kvm_x86_ops svm_x86_ops __ro_after_init = { .get_enable_apicv = svm_get_enable_apicv, .refresh_apicv_exec_ctrl = svm_refresh_apicv_exec_ctrl, .load_eoi_exitmap = svm_load_eoi_exitmap, - .sync_pir_to_irr = svm_sync_pir_to_irr, .hwapic_irr_update = svm_hwapic_irr_update, .hwapic_isr_update = svm_hwapic_isr_update, .apicv_post_state_restore = avic_post_state_restore, diff --git a/arch/x86/kvm/vmx.c b/arch/x86/kvm/vmx.c index a236decb81e4..ef4ba71dbb66 100644 --- a/arch/x86/kvm/vmx.c +++ b/arch/x86/kvm/vmx.c @@ -1856,7 +1856,7 @@ static void update_exception_bitmap(struct kvm_vcpu *vcpu) u32 eb; eb = (1u << PF_VECTOR) | (1u << UD_VECTOR) | (1u << MC_VECTOR) | - (1u << NM_VECTOR) | (1u << DB_VECTOR) | (1u << AC_VECTOR); + (1u << DB_VECTOR) | (1u << AC_VECTOR); if ((vcpu->guest_debug & (KVM_GUESTDBG_ENABLE | KVM_GUESTDBG_USE_SW_BP)) == (KVM_GUESTDBG_ENABLE | KVM_GUESTDBG_USE_SW_BP)) @@ -1865,8 +1865,6 @@ static void update_exception_bitmap(struct kvm_vcpu *vcpu) eb = ~0; if (enable_ept) eb &= ~(1u << PF_VECTOR); /* bypass_guest_pf = 0 */ - if (vcpu->fpu_active) - eb &= ~(1u << NM_VECTOR); /* When we are running a nested L2 guest and L1 specified for it a * certain exception bitmap, we must trap the same exceptions and pass @@ -1992,19 +1990,6 @@ static void add_atomic_switch_msr(struct vcpu_vmx *vmx, unsigned msr, m->host[i].value = host_val; } -static void reload_tss(void) -{ - /* - * VT restores TR but not its size. Useless. - */ - struct desc_ptr *gdt = this_cpu_ptr(&host_gdt); - struct desc_struct *descs; - - descs = (void *)gdt->address; - descs[GDT_ENTRY_TSS].type = 9; /* available TSS */ - load_TR_desc(); -} - static bool update_transition_efer(struct vcpu_vmx *vmx, int efer_offset) { u64 guest_efer = vmx->vcpu.arch.efer; @@ -2059,41 +2044,36 @@ static bool update_transition_efer(struct vcpu_vmx *vmx, int efer_offset) } } +#ifdef CONFIG_X86_32 +/* + * On 32-bit kernels, VM exits still load the FS and GS bases from the + * VMCS rather than the segment table. KVM uses this helper to figure + * out the current bases to poke them into the VMCS before entry. + */ static unsigned long segment_base(u16 selector) { struct desc_ptr *gdt = this_cpu_ptr(&host_gdt); struct desc_struct *d; - unsigned long table_base; + struct desc_struct *table; unsigned long v; - if (!(selector & ~3)) + if (!(selector & ~SEGMENT_RPL_MASK)) return 0; - table_base = gdt->address; + table = (struct desc_struct *)gdt->address; - if (selector & 4) { /* from ldt */ + if ((selector & SEGMENT_TI_MASK) == SEGMENT_LDT) { u16 ldt_selector = kvm_read_ldt(); - if (!(ldt_selector & ~3)) + if (!(ldt_selector & ~SEGMENT_RPL_MASK)) return 0; - table_base = segment_base(ldt_selector); + table = (struct desc_struct *)segment_base(ldt_selector); } - d = (struct desc_struct *)(table_base + (selector & ~7)); - v = get_desc_base(d); -#ifdef CONFIG_X86_64 - if (d->s == 0 && (d->type == 2 || d->type == 9 || d->type == 11)) - v |= ((unsigned long)((struct ldttss_desc64 *)d)->base3) << 32; -#endif + v = get_desc_base(&table[selector >> 3]); return v; } - -static inline unsigned long kvm_read_tr_base(void) -{ - u16 tr; - asm("str %0" : "=g"(tr)); - return segment_base(tr); -} +#endif static void vmx_save_host_state(struct kvm_vcpu *vcpu) { @@ -2179,7 +2159,7 @@ static void __vmx_load_host_state(struct vcpu_vmx *vmx) loadsegment(es, vmx->host_state.es_sel); } #endif - reload_tss(); + invalidate_tss_limit(); #ifdef CONFIG_X86_64 wrmsrl(MSR_KERNEL_GS_BASE, vmx->msr_host_kernel_gs_base); #endif @@ -2294,10 +2274,19 @@ static void vmx_vcpu_load(struct kvm_vcpu *vcpu, int cpu) /* * Linux uses per-cpu TSS and GDT, so set these when switching - * processors. + * processors. See 22.2.4. */ - vmcs_writel(HOST_TR_BASE, kvm_read_tr_base()); /* 22.2.4 */ - vmcs_writel(HOST_GDTR_BASE, gdt->address); /* 22.2.4 */ + vmcs_writel(HOST_TR_BASE, + (unsigned long)this_cpu_ptr(&cpu_tss)); + vmcs_writel(HOST_GDTR_BASE, gdt->address); + + /* + * VM exits change the host TR limit to 0x67 after a VM + * exit. This is okay, since 0x67 covers everything except + * the IO bitmap and have have code to handle the IO bitmap + * being lost after a VM exit. + */ + BUILD_BUG_ON(IO_BITMAP_OFFSET - 1 != 0x67); rdmsrl(MSR_IA32_SYSENTER_ESP, sysenter_esp); vmcs_writel(HOST_IA32_SYSENTER_ESP, sysenter_esp); /* 22.2.3 */ @@ -2340,25 +2329,6 @@ static void vmx_vcpu_put(struct kvm_vcpu *vcpu) } } -static void vmx_fpu_activate(struct kvm_vcpu *vcpu) -{ - ulong cr0; - - if (vcpu->fpu_active) - return; - vcpu->fpu_active = 1; - cr0 = vmcs_readl(GUEST_CR0); - cr0 &= ~(X86_CR0_TS | X86_CR0_MP); - cr0 |= kvm_read_cr0_bits(vcpu, X86_CR0_TS | X86_CR0_MP); - vmcs_writel(GUEST_CR0, cr0); - update_exception_bitmap(vcpu); - vcpu->arch.cr0_guest_owned_bits = X86_CR0_TS; - if (is_guest_mode(vcpu)) - vcpu->arch.cr0_guest_owned_bits &= - ~get_vmcs12(vcpu)->cr0_guest_host_mask; - vmcs_writel(CR0_GUEST_HOST_MASK, ~vcpu->arch.cr0_guest_owned_bits); -} - static void vmx_decache_cr0_guest_bits(struct kvm_vcpu *vcpu); /* @@ -2377,33 +2347,6 @@ static inline unsigned long nested_read_cr4(struct vmcs12 *fields) (fields->cr4_read_shadow & fields->cr4_guest_host_mask); } -static void vmx_fpu_deactivate(struct kvm_vcpu *vcpu) -{ - /* Note that there is no vcpu->fpu_active = 0 here. The caller must - * set this *before* calling this function. - */ - vmx_decache_cr0_guest_bits(vcpu); - vmcs_set_bits(GUEST_CR0, X86_CR0_TS | X86_CR0_MP); - update_exception_bitmap(vcpu); - vcpu->arch.cr0_guest_owned_bits = 0; - vmcs_writel(CR0_GUEST_HOST_MASK, ~vcpu->arch.cr0_guest_owned_bits); - if (is_guest_mode(vcpu)) { - /* - * L1's specified read shadow might not contain the TS bit, - * so now that we turned on shadowing of this bit, we need to - * set this bit of the shadow. Like in nested_vmx_run we need - * nested_read_cr0(vmcs12), but vmcs12->guest_cr0 is not yet - * up-to-date here because we just decached cr0.TS (and we'll - * only update vmcs12->guest_cr0 on nested exit). - */ - struct vmcs12 *vmcs12 = get_vmcs12(vcpu); - vmcs12->guest_cr0 = (vmcs12->guest_cr0 & ~X86_CR0_TS) | - (vcpu->arch.cr0 & X86_CR0_TS); - vmcs_writel(CR0_READ_SHADOW, nested_read_cr0(vmcs12)); - } else - vmcs_writel(CR0_READ_SHADOW, vcpu->arch.cr0); -} - static unsigned long vmx_get_rflags(struct kvm_vcpu *vcpu) { unsigned long rflags, save_rflags; @@ -3962,7 +3905,7 @@ static void fix_rmode_seg(int seg, struct kvm_segment *save) } vmcs_write16(sf->selector, var.selector); - vmcs_write32(sf->base, var.base); + vmcs_writel(sf->base, var.base); vmcs_write32(sf->limit, var.limit); vmcs_write32(sf->ar_bytes, vmx_segment_access_rights(&var)); } @@ -4232,9 +4175,6 @@ static void vmx_set_cr0(struct kvm_vcpu *vcpu, unsigned long cr0) if (enable_ept) ept_update_paging_mode_cr0(&hw_cr0, cr0, vcpu); - if (!vcpu->fpu_active) - hw_cr0 |= X86_CR0_TS | X86_CR0_MP; - vmcs_writel(CR0_READ_SHADOW, cr0); vmcs_writel(GUEST_CR0, hw_cr0); vcpu->arch.cr0 = cr0; @@ -4953,7 +4893,7 @@ static bool vmx_get_enable_apicv(void) return enable_apicv; } -static int vmx_complete_nested_posted_interrupt(struct kvm_vcpu *vcpu) +static void vmx_complete_nested_posted_interrupt(struct kvm_vcpu *vcpu) { struct vcpu_vmx *vmx = to_vmx(vcpu); int max_irr; @@ -4964,19 +4904,15 @@ static int vmx_complete_nested_posted_interrupt(struct kvm_vcpu *vcpu) vmx->nested.pi_pending) { vmx->nested.pi_pending = false; if (!pi_test_and_clear_on(vmx->nested.pi_desc)) - return 0; + return; max_irr = find_last_bit( (unsigned long *)vmx->nested.pi_desc->pir, 256); if (max_irr == 256) - return 0; + return; vapic_page = kmap(vmx->nested.virtual_apic_page); - if (!vapic_page) { - WARN_ON(1); - return -ENOMEM; - } __kvm_apic_update_irr(vmx->nested.pi_desc->pir, vapic_page); kunmap(vmx->nested.virtual_apic_page); @@ -4987,7 +4923,6 @@ static int vmx_complete_nested_posted_interrupt(struct kvm_vcpu *vcpu) vmcs_write16(GUEST_INTR_STATUS, status); } } - return 0; } static inline bool kvm_vcpu_trigger_posted_interrupt(struct kvm_vcpu *vcpu) @@ -5056,26 +4991,12 @@ static void vmx_deliver_posted_interrupt(struct kvm_vcpu *vcpu, int vector) if (pi_test_and_set_pir(vector, &vmx->pi_desc)) return; - r = pi_test_and_set_on(&vmx->pi_desc); - kvm_make_request(KVM_REQ_EVENT, vcpu); - if (r || !kvm_vcpu_trigger_posted_interrupt(vcpu)) - kvm_vcpu_kick(vcpu); -} - -static void vmx_sync_pir_to_irr(struct kvm_vcpu *vcpu) -{ - struct vcpu_vmx *vmx = to_vmx(vcpu); - - if (!pi_test_on(&vmx->pi_desc)) + /* If a previous notification has sent the IPI, nothing to do. */ + if (pi_test_and_set_on(&vmx->pi_desc)) return; - pi_clear_on(&vmx->pi_desc); - /* - * IOMMU can write to PIR.ON, so the barrier matters even on UP. - * But on x86 this is just a compiler barrier anyway. - */ - smp_mb__after_atomic(); - kvm_apic_update_irr(vcpu, vmx->pi_desc.pir); + if (!kvm_vcpu_trigger_posted_interrupt(vcpu)) + kvm_vcpu_kick(vcpu); } /* @@ -5236,10 +5157,8 @@ static void ept_set_mmio_spte_mask(void) /* * EPT Misconfigurations can be generated if the value of bits 2:0 * of an EPT paging-structure entry is 110b (write/execute). - * Also, magic bits (0x3ull << 62) is set to quickly identify mmio - * spte. */ - kvm_mmu_set_mmio_spte_mask((0x3ull << 62) | 0x6ull); + kvm_mmu_set_mmio_spte_mask(VMX_EPT_MISCONFIG_WX_VALUE); } #define VMX_XSS_EXIT_BITMAP 0 @@ -5342,7 +5261,9 @@ static int vmx_vcpu_setup(struct vcpu_vmx *vmx) /* 22.2.1, 20.8.1 */ vm_entry_controls_init(vmx, vmcs_config.vmentry_ctrl); - vmcs_writel(CR0_GUEST_HOST_MASK, ~0UL); + vmx->vcpu.arch.cr0_guest_owned_bits = X86_CR0_TS; + vmcs_writel(CR0_GUEST_HOST_MASK, ~X86_CR0_TS); + set_cr4_guest_host_mask(vmx); if (vmx_xsaves_supported()) @@ -5446,7 +5367,7 @@ static void vmx_vcpu_reset(struct kvm_vcpu *vcpu, bool init_event) vmx_set_cr0(vcpu, cr0); /* enter rmode */ vmx_set_cr4(vcpu, 0); vmx_set_efer(vcpu, 0); - vmx_fpu_activate(vcpu); + update_exception_bitmap(vcpu); vpid_sync_context(vmx->vpid); @@ -5480,26 +5401,20 @@ static bool nested_exit_on_nmi(struct kvm_vcpu *vcpu) static void enable_irq_window(struct kvm_vcpu *vcpu) { - u32 cpu_based_vm_exec_control; - - cpu_based_vm_exec_control = vmcs_read32(CPU_BASED_VM_EXEC_CONTROL); - cpu_based_vm_exec_control |= CPU_BASED_VIRTUAL_INTR_PENDING; - vmcs_write32(CPU_BASED_VM_EXEC_CONTROL, cpu_based_vm_exec_control); + vmcs_set_bits(CPU_BASED_VM_EXEC_CONTROL, + CPU_BASED_VIRTUAL_INTR_PENDING); } static void enable_nmi_window(struct kvm_vcpu *vcpu) { - u32 cpu_based_vm_exec_control; - if (!cpu_has_virtual_nmis() || vmcs_read32(GUEST_INTERRUPTIBILITY_INFO) & GUEST_INTR_STATE_STI) { enable_irq_window(vcpu); return; } - cpu_based_vm_exec_control = vmcs_read32(CPU_BASED_VM_EXEC_CONTROL); - cpu_based_vm_exec_control |= CPU_BASED_VIRTUAL_NMI_PENDING; - vmcs_write32(CPU_BASED_VM_EXEC_CONTROL, cpu_based_vm_exec_control); + vmcs_set_bits(CPU_BASED_VM_EXEC_CONTROL, + CPU_BASED_VIRTUAL_NMI_PENDING); } static void vmx_inject_irq(struct kvm_vcpu *vcpu) @@ -5725,11 +5640,6 @@ static int handle_exception(struct kvm_vcpu *vcpu) if (is_nmi(intr_info)) return 1; /* already handled by vmx_vcpu_run() */ - if (is_no_device(intr_info)) { - vmx_fpu_activate(vcpu); - return 1; - } - if (is_invalid_opcode(intr_info)) { if (is_guest_mode(vcpu)) { kvm_queue_exception(vcpu, UD_VECTOR); @@ -5919,22 +5829,6 @@ static int handle_set_cr4(struct kvm_vcpu *vcpu, unsigned long val) return kvm_set_cr4(vcpu, val); } -/* called to set cr0 as appropriate for clts instruction exit. */ -static void handle_clts(struct kvm_vcpu *vcpu) -{ - if (is_guest_mode(vcpu)) { - /* - * We get here when L2 did CLTS, and L1 didn't shadow CR0.TS - * but we did (!fpu_active). We need to keep GUEST_CR0.TS on, - * just pretend it's off (also in arch.cr0 for fpu_activate). - */ - vmcs_writel(CR0_READ_SHADOW, - vmcs_readl(CR0_READ_SHADOW) & ~X86_CR0_TS); - vcpu->arch.cr0 &= ~X86_CR0_TS; - } else - vmx_set_cr0(vcpu, kvm_read_cr0_bits(vcpu, ~X86_CR0_TS)); -} - static int handle_cr(struct kvm_vcpu *vcpu) { unsigned long exit_qualification, val; @@ -5980,9 +5874,9 @@ static int handle_cr(struct kvm_vcpu *vcpu) } break; case 2: /* clts */ - handle_clts(vcpu); + WARN_ONCE(1, "Guest should always own CR0.TS"); + vmx_set_cr0(vcpu, kvm_read_cr0_bits(vcpu, ~X86_CR0_TS)); trace_kvm_cr_write(0, kvm_read_cr0(vcpu)); - vmx_fpu_activate(vcpu); return kvm_skip_emulated_instruction(vcpu); case 1: /*mov from cr*/ switch (cr) { @@ -6152,18 +6046,14 @@ static int handle_wrmsr(struct kvm_vcpu *vcpu) static int handle_tpr_below_threshold(struct kvm_vcpu *vcpu) { - kvm_make_request(KVM_REQ_EVENT, vcpu); + kvm_apic_update_ppr(vcpu); return 1; } static int handle_interrupt_window(struct kvm_vcpu *vcpu) { - u32 cpu_based_vm_exec_control; - - /* clear pending irq */ - cpu_based_vm_exec_control = vmcs_read32(CPU_BASED_VM_EXEC_CONTROL); - cpu_based_vm_exec_control &= ~CPU_BASED_VIRTUAL_INTR_PENDING; - vmcs_write32(CPU_BASED_VM_EXEC_CONTROL, cpu_based_vm_exec_control); + vmcs_clear_bits(CPU_BASED_VM_EXEC_CONTROL, + CPU_BASED_VIRTUAL_INTR_PENDING); kvm_make_request(KVM_REQ_EVENT, vcpu); @@ -6374,15 +6264,22 @@ static int handle_ept_violation(struct kvm_vcpu *vcpu) gpa = vmcs_read64(GUEST_PHYSICAL_ADDRESS); trace_kvm_page_fault(gpa, exit_qualification); - /* it is a read fault? */ - error_code = (exit_qualification << 2) & PFERR_USER_MASK; - /* it is a write fault? */ - error_code |= exit_qualification & PFERR_WRITE_MASK; - /* It is a fetch fault? */ - error_code |= (exit_qualification << 2) & PFERR_FETCH_MASK; - /* ept page table is present? */ - error_code |= (exit_qualification & 0x38) != 0; - + /* Is it a read fault? */ + error_code = (exit_qualification & EPT_VIOLATION_ACC_READ) + ? PFERR_USER_MASK : 0; + /* Is it a write fault? */ + error_code |= (exit_qualification & EPT_VIOLATION_ACC_WRITE) + ? PFERR_WRITE_MASK : 0; + /* Is it a fetch fault? */ + error_code |= (exit_qualification & EPT_VIOLATION_ACC_INSTR) + ? PFERR_FETCH_MASK : 0; + /* ept page table entry is present? */ + error_code |= (exit_qualification & + (EPT_VIOLATION_READABLE | EPT_VIOLATION_WRITABLE | + EPT_VIOLATION_EXECUTABLE)) + ? PFERR_PRESENT_MASK : 0; + + vcpu->arch.gpa_available = true; vcpu->arch.exit_qualification = exit_qualification; return kvm_mmu_page_fault(vcpu, gpa, error_code, NULL, 0); @@ -6400,6 +6297,7 @@ static int handle_ept_misconfig(struct kvm_vcpu *vcpu) } ret = handle_mmio_page_fault(vcpu, gpa, true); + vcpu->arch.gpa_available = true; if (likely(ret == RET_MMIO_PF_EMULATE)) return x86_emulate_instruction(vcpu, gpa, 0, NULL, 0) == EMULATE_DONE; @@ -6421,12 +6319,8 @@ static int handle_ept_misconfig(struct kvm_vcpu *vcpu) static int handle_nmi_window(struct kvm_vcpu *vcpu) { - u32 cpu_based_vm_exec_control; - - /* clear pending NMI */ - cpu_based_vm_exec_control = vmcs_read32(CPU_BASED_VM_EXEC_CONTROL); - cpu_based_vm_exec_control &= ~CPU_BASED_VIRTUAL_NMI_PENDING; - vmcs_write32(CPU_BASED_VM_EXEC_CONTROL, cpu_based_vm_exec_control); + vmcs_clear_bits(CPU_BASED_VM_EXEC_CONTROL, + CPU_BASED_VIRTUAL_NMI_PENDING); ++vcpu->stat.nmi_window_exits; kvm_make_request(KVM_REQ_EVENT, vcpu); @@ -6572,6 +6466,19 @@ static void wakeup_handler(void) spin_unlock(&per_cpu(blocked_vcpu_on_cpu_lock, cpu)); } +void vmx_enable_tdp(void) +{ + kvm_mmu_set_mask_ptes(VMX_EPT_READABLE_MASK, + enable_ept_ad_bits ? VMX_EPT_ACCESS_BIT : 0ull, + enable_ept_ad_bits ? VMX_EPT_DIRTY_BIT : 0ull, + 0ull, VMX_EPT_EXECUTABLE_MASK, + cpu_has_vmx_ept_execute_only() ? 0ull : VMX_EPT_READABLE_MASK, + enable_ept_ad_bits ? 0ull : VMX_EPT_RWX_MASK); + + ept_set_mmio_spte_mask(); + kvm_enable_tdp(); +} + static __init int hardware_setup(void) { int r = -ENOMEM, i, msr; @@ -6651,8 +6558,10 @@ static __init int hardware_setup(void) if (!cpu_has_vmx_ple()) ple_gap = 0; - if (!cpu_has_vmx_apicv()) + if (!cpu_has_vmx_apicv()) { enable_apicv = 0; + kvm_x86_ops->sync_pir_to_irr = NULL; + } if (cpu_has_vmx_tsc_scaling()) { kvm_has_tsc_control = true; @@ -6697,16 +6606,9 @@ static __init int hardware_setup(void) /* SELF-IPI */ vmx_disable_intercept_msr_x2apic(0x83f, MSR_TYPE_W, true); - if (enable_ept) { - kvm_mmu_set_mask_ptes(VMX_EPT_READABLE_MASK, - (enable_ept_ad_bits) ? VMX_EPT_ACCESS_BIT : 0ull, - (enable_ept_ad_bits) ? VMX_EPT_DIRTY_BIT : 0ull, - 0ull, VMX_EPT_EXECUTABLE_MASK, - cpu_has_vmx_ept_execute_only() ? - 0ull : VMX_EPT_READABLE_MASK); - ept_set_mmio_spte_mask(); - kvm_enable_tdp(); - } else + if (enable_ept) + vmx_enable_tdp(); + else kvm_disable_tdp(); update_ple_window_actual_max(); @@ -7085,13 +6987,18 @@ static int nested_vmx_check_vmptr(struct kvm_vcpu *vcpu, int exit_reason, } page = nested_get_page(vcpu, vmptr); - if (page == NULL || - *(u32 *)kmap(page) != VMCS12_REVISION) { + if (page == NULL) { nested_vmx_failInvalid(vcpu); + return kvm_skip_emulated_instruction(vcpu); + } + if (*(u32 *)kmap(page) != VMCS12_REVISION) { kunmap(page); + nested_release_page_clean(page); + nested_vmx_failInvalid(vcpu); return kvm_skip_emulated_instruction(vcpu); } kunmap(page); + nested_release_page_clean(page); vmx->nested.vmxon_ptr = vmptr; break; case EXIT_REASON_VMCLEAR: @@ -7129,6 +7036,53 @@ static int nested_vmx_check_vmptr(struct kvm_vcpu *vcpu, int exit_reason, return 0; } +static int enter_vmx_operation(struct kvm_vcpu *vcpu) +{ + struct vcpu_vmx *vmx = to_vmx(vcpu); + struct vmcs *shadow_vmcs; + + if (cpu_has_vmx_msr_bitmap()) { + vmx->nested.msr_bitmap = + (unsigned long *)__get_free_page(GFP_KERNEL); + if (!vmx->nested.msr_bitmap) + goto out_msr_bitmap; + } + + vmx->nested.cached_vmcs12 = kmalloc(VMCS12_SIZE, GFP_KERNEL); + if (!vmx->nested.cached_vmcs12) + goto out_cached_vmcs12; + + if (enable_shadow_vmcs) { + shadow_vmcs = alloc_vmcs(); + if (!shadow_vmcs) + goto out_shadow_vmcs; + /* mark vmcs as shadow */ + shadow_vmcs->revision_id |= (1u << 31); + /* init shadow vmcs */ + vmcs_clear(shadow_vmcs); + vmx->vmcs01.shadow_vmcs = shadow_vmcs; + } + + INIT_LIST_HEAD(&(vmx->nested.vmcs02_pool)); + vmx->nested.vmcs02_num = 0; + + hrtimer_init(&vmx->nested.preemption_timer, CLOCK_MONOTONIC, + HRTIMER_MODE_REL_PINNED); + vmx->nested.preemption_timer.function = vmx_preemption_timer_fn; + + vmx->nested.vmxon = true; + return 0; + +out_shadow_vmcs: + kfree(vmx->nested.cached_vmcs12); + +out_cached_vmcs12: + free_page((unsigned long)vmx->nested.msr_bitmap); + +out_msr_bitmap: + return -ENOMEM; +} + /* * Emulate the VMXON instruction. * Currently, we just remember that VMX is active, and do not save or even @@ -7139,9 +7093,9 @@ static int nested_vmx_check_vmptr(struct kvm_vcpu *vcpu, int exit_reason, */ static int handle_vmon(struct kvm_vcpu *vcpu) { + int ret; struct kvm_segment cs; struct vcpu_vmx *vmx = to_vmx(vcpu); - struct vmcs *shadow_vmcs; const u64 VMXON_NEEDED_FEATURES = FEATURE_CONTROL_LOCKED | FEATURE_CONTROL_VMXON_ENABLED_OUTSIDE_SMX; @@ -7168,9 +7122,6 @@ static int handle_vmon(struct kvm_vcpu *vcpu) return 1; } - if (nested_vmx_check_vmptr(vcpu, EXIT_REASON_VMON, NULL)) - return 1; - if (vmx->nested.vmxon) { nested_vmx_failValid(vcpu, VMXERR_VMXON_IN_VMX_ROOT_OPERATION); return kvm_skip_emulated_instruction(vcpu); @@ -7182,48 +7133,15 @@ static int handle_vmon(struct kvm_vcpu *vcpu) return 1; } - if (cpu_has_vmx_msr_bitmap()) { - vmx->nested.msr_bitmap = - (unsigned long *)__get_free_page(GFP_KERNEL); - if (!vmx->nested.msr_bitmap) - goto out_msr_bitmap; - } - - vmx->nested.cached_vmcs12 = kmalloc(VMCS12_SIZE, GFP_KERNEL); - if (!vmx->nested.cached_vmcs12) - goto out_cached_vmcs12; - - if (enable_shadow_vmcs) { - shadow_vmcs = alloc_vmcs(); - if (!shadow_vmcs) - goto out_shadow_vmcs; - /* mark vmcs as shadow */ - shadow_vmcs->revision_id |= (1u << 31); - /* init shadow vmcs */ - vmcs_clear(shadow_vmcs); - vmx->vmcs01.shadow_vmcs = shadow_vmcs; - } - - INIT_LIST_HEAD(&(vmx->nested.vmcs02_pool)); - vmx->nested.vmcs02_num = 0; - - hrtimer_init(&vmx->nested.preemption_timer, CLOCK_MONOTONIC, - HRTIMER_MODE_REL_PINNED); - vmx->nested.preemption_timer.function = vmx_preemption_timer_fn; - - vmx->nested.vmxon = true; + if (nested_vmx_check_vmptr(vcpu, EXIT_REASON_VMON, NULL)) + return 1; + + ret = enter_vmx_operation(vcpu); + if (ret) + return ret; nested_vmx_succeed(vcpu); return kvm_skip_emulated_instruction(vcpu); - -out_shadow_vmcs: - kfree(vmx->nested.cached_vmcs12); - -out_cached_vmcs12: - free_page((unsigned long)vmx->nested.msr_bitmap); - -out_msr_bitmap: - return -ENOMEM; } /* @@ -7672,6 +7590,18 @@ static int handle_vmwrite(struct kvm_vcpu *vcpu) return kvm_skip_emulated_instruction(vcpu); } +static void set_current_vmptr(struct vcpu_vmx *vmx, gpa_t vmptr) +{ + vmx->nested.current_vmptr = vmptr; + if (enable_shadow_vmcs) { + vmcs_set_bits(SECONDARY_VM_EXEC_CONTROL, + SECONDARY_EXEC_SHADOW_VMCS); + vmcs_write64(VMCS_LINK_POINTER, + __pa(vmx->vmcs01.shadow_vmcs)); + vmx->nested.sync_shadow_vmcs = true; + } +} + /* Emulate the VMPTRLD instruction */ static int handle_vmptrld(struct kvm_vcpu *vcpu) { @@ -7702,7 +7632,6 @@ static int handle_vmptrld(struct kvm_vcpu *vcpu) } nested_release_vmcs12(vmx); - vmx->nested.current_vmptr = vmptr; vmx->nested.current_vmcs12 = new_vmcs12; vmx->nested.current_vmcs12_page = page; /* @@ -7711,14 +7640,7 @@ static int handle_vmptrld(struct kvm_vcpu *vcpu) */ memcpy(vmx->nested.cached_vmcs12, vmx->nested.current_vmcs12, VMCS12_SIZE); - - if (enable_shadow_vmcs) { - vmcs_set_bits(SECONDARY_VM_EXEC_CONTROL, - SECONDARY_EXEC_SHADOW_VMCS); - vmcs_write64(VMCS_LINK_POINTER, - __pa(vmx->vmcs01.shadow_vmcs)); - vmx->nested.sync_shadow_vmcs = true; - } + set_current_vmptr(vmx, vmptr); } nested_vmx_succeed(vcpu); @@ -8191,8 +8113,6 @@ static bool nested_vmx_exit_handled(struct kvm_vcpu *vcpu) case EXIT_REASON_TASK_SWITCH: return true; case EXIT_REASON_CPUID: - if (kvm_register_read(vcpu, VCPU_REGS_RAX) == 0xa) - return false; return true; case EXIT_REASON_HLT: return nested_cpu_has(vmcs12, CPU_BASED_HLT_EXITING); @@ -8350,7 +8270,7 @@ static void kvm_flush_pml_buffers(struct kvm *kvm) static void vmx_dump_sel(char *name, uint32_t sel) { pr_err("%s sel=0x%04x, attr=0x%05x, limit=0x%08x, base=0x%016lx\n", - name, vmcs_read32(sel), + name, vmcs_read16(sel), vmcs_read32(sel + GUEST_ES_AR_BYTES - GUEST_ES_SELECTOR), vmcs_read32(sel + GUEST_ES_LIMIT - GUEST_ES_SELECTOR), vmcs_readl(sel + GUEST_ES_BASE - GUEST_ES_SELECTOR)); @@ -8514,6 +8434,7 @@ static int vmx_handle_exit(struct kvm_vcpu *vcpu) u32 vectoring_info = vmx->idt_vectoring_info; trace_kvm_exit(exit_reason, vcpu, KVM_ISA_VMX); + vcpu->arch.gpa_available = false; /* * Flush logged GPAs PML buffer, this will make dirty_bitmap more @@ -8732,6 +8653,27 @@ static void vmx_hwapic_irr_update(struct kvm_vcpu *vcpu, int max_irr) } } +static int vmx_sync_pir_to_irr(struct kvm_vcpu *vcpu) +{ + struct vcpu_vmx *vmx = to_vmx(vcpu); + int max_irr; + + WARN_ON(!vcpu->arch.apicv_active); + if (pi_test_on(&vmx->pi_desc)) { + pi_clear_on(&vmx->pi_desc); + /* + * IOMMU can write to PIR.ON, so the barrier matters even on UP. + * But on x86 this is just a compiler barrier anyway. + */ + smp_mb__after_atomic(); + max_irr = kvm_apic_update_irr(vcpu, vmx->pi_desc.pir); + } else { + max_irr = kvm_lapic_find_highest_irr(vcpu); + } + vmx_hwapic_irr_update(vcpu, max_irr); + return max_irr; +} + static void vmx_load_eoi_exitmap(struct kvm_vcpu *vcpu, u64 *eoi_exit_bitmap) { if (!kvm_vcpu_apicv_active(vcpu)) @@ -8743,6 +8685,14 @@ static void vmx_load_eoi_exitmap(struct kvm_vcpu *vcpu, u64 *eoi_exit_bitmap) vmcs_write64(EOI_EXIT_BITMAP3, eoi_exit_bitmap[3]); } +static void vmx_apicv_post_state_restore(struct kvm_vcpu *vcpu) +{ + struct vcpu_vmx *vmx = to_vmx(vcpu); + + pi_clear_on(&vmx->pi_desc); + memset(vmx->pi_desc.pir, 0, sizeof(vmx->pi_desc.pir)); +} + static void vmx_complete_atomic_exit(struct vcpu_vmx *vmx) { u32 exit_intr_info; @@ -9588,17 +9538,16 @@ static void vmx_inject_page_fault_nested(struct kvm_vcpu *vcpu, kvm_inject_page_fault(vcpu, fault); } -static bool nested_get_vmcs12_pages(struct kvm_vcpu *vcpu, +static inline bool nested_vmx_merge_msr_bitmap(struct kvm_vcpu *vcpu, + struct vmcs12 *vmcs12); + +static void nested_get_vmcs12_pages(struct kvm_vcpu *vcpu, struct vmcs12 *vmcs12) { struct vcpu_vmx *vmx = to_vmx(vcpu); - int maxphyaddr = cpuid_maxphyaddr(vcpu); + u64 hpa; if (nested_cpu_has2(vmcs12, SECONDARY_EXEC_VIRTUALIZE_APIC_ACCESSES)) { - if (!PAGE_ALIGNED(vmcs12->apic_access_addr) || - vmcs12->apic_access_addr >> maxphyaddr) - return false; - /* * Translate L1 physical address to host physical * address for vmcs02. Keep the page pinned, so this @@ -9609,59 +9558,80 @@ static bool nested_get_vmcs12_pages(struct kvm_vcpu *vcpu, nested_release_page(vmx->nested.apic_access_page); vmx->nested.apic_access_page = nested_get_page(vcpu, vmcs12->apic_access_addr); + /* + * If translation failed, no matter: This feature asks + * to exit when accessing the given address, and if it + * can never be accessed, this feature won't do + * anything anyway. + */ + if (vmx->nested.apic_access_page) { + hpa = page_to_phys(vmx->nested.apic_access_page); + vmcs_write64(APIC_ACCESS_ADDR, hpa); + } else { + vmcs_clear_bits(SECONDARY_VM_EXEC_CONTROL, + SECONDARY_EXEC_VIRTUALIZE_APIC_ACCESSES); + } + } else if (!(nested_cpu_has_virt_x2apic_mode(vmcs12)) && + cpu_need_virtualize_apic_accesses(&vmx->vcpu)) { + vmcs_set_bits(SECONDARY_VM_EXEC_CONTROL, + SECONDARY_EXEC_VIRTUALIZE_APIC_ACCESSES); + kvm_vcpu_reload_apic_access_page(vcpu); } if (nested_cpu_has(vmcs12, CPU_BASED_TPR_SHADOW)) { - if (!PAGE_ALIGNED(vmcs12->virtual_apic_page_addr) || - vmcs12->virtual_apic_page_addr >> maxphyaddr) - return false; - if (vmx->nested.virtual_apic_page) /* shouldn't happen */ nested_release_page(vmx->nested.virtual_apic_page); vmx->nested.virtual_apic_page = nested_get_page(vcpu, vmcs12->virtual_apic_page_addr); /* - * Failing the vm entry is _not_ what the processor does - * but it's basically the only possibility we have. - * We could still enter the guest if CR8 load exits are - * enabled, CR8 store exits are enabled, and virtualize APIC - * access is disabled; in this case the processor would never - * use the TPR shadow and we could simply clear the bit from - * the execution control. But such a configuration is useless, - * so let's keep the code simple. + * If translation failed, VM entry will fail because + * prepare_vmcs02 set VIRTUAL_APIC_PAGE_ADDR to -1ull. + * Failing the vm entry is _not_ what the processor + * does but it's basically the only possibility we + * have. We could still enter the guest if CR8 load + * exits are enabled, CR8 store exits are enabled, and + * virtualize APIC access is disabled; in this case + * the processor would never use the TPR shadow and we + * could simply clear the bit from the execution + * control. But such a configuration is useless, so + * let's keep the code simple. */ - if (!vmx->nested.virtual_apic_page) - return false; + if (vmx->nested.virtual_apic_page) { + hpa = page_to_phys(vmx->nested.virtual_apic_page); + vmcs_write64(VIRTUAL_APIC_PAGE_ADDR, hpa); + } } if (nested_cpu_has_posted_intr(vmcs12)) { - if (!IS_ALIGNED(vmcs12->posted_intr_desc_addr, 64) || - vmcs12->posted_intr_desc_addr >> maxphyaddr) - return false; - if (vmx->nested.pi_desc_page) { /* shouldn't happen */ kunmap(vmx->nested.pi_desc_page); nested_release_page(vmx->nested.pi_desc_page); } vmx->nested.pi_desc_page = nested_get_page(vcpu, vmcs12->posted_intr_desc_addr); - if (!vmx->nested.pi_desc_page) - return false; - vmx->nested.pi_desc = (struct pi_desc *)kmap(vmx->nested.pi_desc_page); if (!vmx->nested.pi_desc) { nested_release_page_clean(vmx->nested.pi_desc_page); - return false; + return; } vmx->nested.pi_desc = (struct pi_desc *)((void *)vmx->nested.pi_desc + (unsigned long)(vmcs12->posted_intr_desc_addr & (PAGE_SIZE - 1))); + vmcs_write64(POSTED_INTR_DESC_ADDR, + page_to_phys(vmx->nested.pi_desc_page) + + (unsigned long)(vmcs12->posted_intr_desc_addr & + (PAGE_SIZE - 1))); } - - return true; + if (cpu_has_vmx_msr_bitmap() && + nested_cpu_has(vmcs12, CPU_BASED_USE_MSR_BITMAPS) && + nested_vmx_merge_msr_bitmap(vcpu, vmcs12)) + ; + else + vmcs_clear_bits(CPU_BASED_VM_EXEC_CONTROL, + CPU_BASED_USE_MSR_BITMAPS); } static void vmx_start_preemption_timer(struct kvm_vcpu *vcpu) @@ -9730,11 +9700,6 @@ static inline bool nested_vmx_merge_msr_bitmap(struct kvm_vcpu *vcpu, return false; } msr_bitmap_l1 = (unsigned long *)kmap(page); - if (!msr_bitmap_l1) { - nested_release_page_clean(page); - WARN_ON(1); - return false; - } memset(msr_bitmap_l0, 0xff, PAGE_SIZE); @@ -9982,7 +9947,7 @@ static bool nested_cr3_valid(struct kvm_vcpu *vcpu, unsigned long val) * is assigned to entry_failure_code on failure. */ static int nested_vmx_load_cr3(struct kvm_vcpu *vcpu, unsigned long cr3, bool nested_ept, - unsigned long *entry_failure_code) + u32 *entry_failure_code) { if (cr3 != kvm_read_cr3(vcpu) || (!nested_ept && pdptrs_changed(vcpu))) { if (!nested_cr3_valid(vcpu, cr3)) { @@ -10022,7 +9987,7 @@ static int nested_vmx_load_cr3(struct kvm_vcpu *vcpu, unsigned long cr3, bool ne * is assigned to entry_failure_code on failure. */ static int prepare_vmcs02(struct kvm_vcpu *vcpu, struct vmcs12 *vmcs12, - unsigned long *entry_failure_code) + bool from_vmentry, u32 *entry_failure_code) { struct vcpu_vmx *vmx = to_vmx(vcpu); u32 exec_control; @@ -10065,21 +10030,26 @@ static int prepare_vmcs02(struct kvm_vcpu *vcpu, struct vmcs12 *vmcs12, vmcs_writel(GUEST_GDTR_BASE, vmcs12->guest_gdtr_base); vmcs_writel(GUEST_IDTR_BASE, vmcs12->guest_idtr_base); - if (vmcs12->vm_entry_controls & VM_ENTRY_LOAD_DEBUG_CONTROLS) { + if (from_vmentry && + (vmcs12->vm_entry_controls & VM_ENTRY_LOAD_DEBUG_CONTROLS)) { kvm_set_dr(vcpu, 7, vmcs12->guest_dr7); vmcs_write64(GUEST_IA32_DEBUGCTL, vmcs12->guest_ia32_debugctl); } else { kvm_set_dr(vcpu, 7, vcpu->arch.dr7); vmcs_write64(GUEST_IA32_DEBUGCTL, vmx->nested.vmcs01_debugctl); } - vmcs_write32(VM_ENTRY_INTR_INFO_FIELD, - vmcs12->vm_entry_intr_info_field); - vmcs_write32(VM_ENTRY_EXCEPTION_ERROR_CODE, - vmcs12->vm_entry_exception_error_code); - vmcs_write32(VM_ENTRY_INSTRUCTION_LEN, - vmcs12->vm_entry_instruction_len); - vmcs_write32(GUEST_INTERRUPTIBILITY_INFO, - vmcs12->guest_interruptibility_info); + if (from_vmentry) { + vmcs_write32(VM_ENTRY_INTR_INFO_FIELD, + vmcs12->vm_entry_intr_info_field); + vmcs_write32(VM_ENTRY_EXCEPTION_ERROR_CODE, + vmcs12->vm_entry_exception_error_code); + vmcs_write32(VM_ENTRY_INSTRUCTION_LEN, + vmcs12->vm_entry_instruction_len); + vmcs_write32(GUEST_INTERRUPTIBILITY_INFO, + vmcs12->guest_interruptibility_info); + } else { + vmcs_write32(VM_ENTRY_INTR_INFO_FIELD, 0); + } vmcs_write32(GUEST_SYSENTER_CS, vmcs12->guest_sysenter_cs); vmx_set_rflags(vcpu, vmcs12->guest_rflags); vmcs_writel(GUEST_PENDING_DBG_EXCEPTIONS, @@ -10108,12 +10078,9 @@ static int prepare_vmcs02(struct kvm_vcpu *vcpu, struct vmcs12 *vmcs12, vmx->nested.posted_intr_nv = vmcs12->posted_intr_nv; vmx->nested.pi_pending = false; vmcs_write16(POSTED_INTR_NV, POSTED_INTR_VECTOR); - vmcs_write64(POSTED_INTR_DESC_ADDR, - page_to_phys(vmx->nested.pi_desc_page) + - (unsigned long)(vmcs12->posted_intr_desc_addr & - (PAGE_SIZE - 1))); - } else + } else { exec_control &= ~PIN_BASED_POSTED_INTR; + } vmcs_write32(PIN_BASED_VM_EXEC_CONTROL, exec_control); @@ -10158,26 +10125,6 @@ static int prepare_vmcs02(struct kvm_vcpu *vcpu, struct vmcs12 *vmcs12, CPU_BASED_ACTIVATE_SECONDARY_CONTROLS)) exec_control |= vmcs12->secondary_vm_exec_control; - if (exec_control & SECONDARY_EXEC_VIRTUALIZE_APIC_ACCESSES) { - /* - * If translation failed, no matter: This feature asks - * to exit when accessing the given address, and if it - * can never be accessed, this feature won't do - * anything anyway. - */ - if (!vmx->nested.apic_access_page) - exec_control &= - ~SECONDARY_EXEC_VIRTUALIZE_APIC_ACCESSES; - else - vmcs_write64(APIC_ACCESS_ADDR, - page_to_phys(vmx->nested.apic_access_page)); - } else if (!(nested_cpu_has_virt_x2apic_mode(vmcs12)) && - cpu_need_virtualize_apic_accesses(&vmx->vcpu)) { - exec_control |= - SECONDARY_EXEC_VIRTUALIZE_APIC_ACCESSES; - kvm_vcpu_reload_apic_access_page(vcpu); - } - if (exec_control & SECONDARY_EXEC_VIRTUAL_INTR_DELIVERY) { vmcs_write64(EOI_EXIT_BITMAP0, vmcs12->eoi_exit_bitmap0); @@ -10192,6 +10139,15 @@ static int prepare_vmcs02(struct kvm_vcpu *vcpu, struct vmcs12 *vmcs12, } nested_ept_enabled = (exec_control & SECONDARY_EXEC_ENABLE_EPT) != 0; + + /* + * Write an illegal value to APIC_ACCESS_ADDR. Later, + * nested_get_vmcs12_pages will either fix it up or + * remove the VM execution control. + */ + if (exec_control & SECONDARY_EXEC_VIRTUALIZE_APIC_ACCESSES) + vmcs_write64(APIC_ACCESS_ADDR, -1ull); + vmcs_write32(SECONDARY_VM_EXEC_CONTROL, exec_control); } @@ -10228,19 +10184,16 @@ static int prepare_vmcs02(struct kvm_vcpu *vcpu, struct vmcs12 *vmcs12, exec_control &= ~CPU_BASED_TPR_SHADOW; exec_control |= vmcs12->cpu_based_vm_exec_control; + /* + * Write an illegal value to VIRTUAL_APIC_PAGE_ADDR. Later, if + * nested_get_vmcs12_pages can't fix it up, the illegal value + * will result in a VM entry failure. + */ if (exec_control & CPU_BASED_TPR_SHADOW) { - vmcs_write64(VIRTUAL_APIC_PAGE_ADDR, - page_to_phys(vmx->nested.virtual_apic_page)); + vmcs_write64(VIRTUAL_APIC_PAGE_ADDR, -1ull); vmcs_write32(TPR_THRESHOLD, vmcs12->tpr_threshold); } - if (cpu_has_vmx_msr_bitmap() && - exec_control & CPU_BASED_USE_MSR_BITMAPS && - nested_vmx_merge_msr_bitmap(vcpu, vmcs12)) - ; /* MSR_BITMAP will be set by following vmx_set_efer. */ - else - exec_control &= ~CPU_BASED_USE_MSR_BITMAPS; - /* * Merging of IO bitmap not currently supported. * Rather, exit every time. @@ -10272,16 +10225,18 @@ static int prepare_vmcs02(struct kvm_vcpu *vcpu, struct vmcs12 *vmcs12, ~VM_ENTRY_IA32E_MODE) | (vmcs_config.vmentry_ctrl & ~VM_ENTRY_IA32E_MODE)); - if (vmcs12->vm_entry_controls & VM_ENTRY_LOAD_IA32_PAT) { + if (from_vmentry && + (vmcs12->vm_entry_controls & VM_ENTRY_LOAD_IA32_PAT)) { vmcs_write64(GUEST_IA32_PAT, vmcs12->guest_ia32_pat); vcpu->arch.pat = vmcs12->guest_ia32_pat; - } else if (vmcs_config.vmentry_ctrl & VM_ENTRY_LOAD_IA32_PAT) + } else if (vmcs_config.vmentry_ctrl & VM_ENTRY_LOAD_IA32_PAT) { vmcs_write64(GUEST_IA32_PAT, vmx->vcpu.arch.pat); - + } set_cr4_guest_host_mask(vmx); - if (vmcs12->vm_entry_controls & VM_ENTRY_LOAD_BNDCFGS) + if (from_vmentry && + vmcs12->vm_entry_controls & VM_ENTRY_LOAD_BNDCFGS) vmcs_write64(GUEST_BNDCFGS, vmcs12->guest_bndcfgs); if (vmcs12->cpu_based_vm_exec_control & CPU_BASED_USE_TSC_OFFSETING) @@ -10320,8 +10275,8 @@ static int prepare_vmcs02(struct kvm_vcpu *vcpu, struct vmcs12 *vmcs12, } /* - * This sets GUEST_CR0 to vmcs12->guest_cr0, with possibly a modified - * TS bit (for lazy fpu) and bits which we consider mandatory enabled. + * This sets GUEST_CR0 to vmcs12->guest_cr0, possibly modifying those + * bits which we consider mandatory enabled. * The CR0_READ_SHADOW is what L2 should have expected to read given * the specifications by L1; It's not enough to take * vmcs12->cr0_read_shadow because on our cr0_guest_host_mask we we @@ -10333,7 +10288,8 @@ static int prepare_vmcs02(struct kvm_vcpu *vcpu, struct vmcs12 *vmcs12, vmx_set_cr4(vcpu, vmcs12->guest_cr4); vmcs_writel(CR4_READ_SHADOW, nested_read_cr4(vmcs12)); - if (vmcs12->vm_entry_controls & VM_ENTRY_LOAD_IA32_EFER) + if (from_vmentry && + (vmcs12->vm_entry_controls & VM_ENTRY_LOAD_IA32_EFER)) vcpu->arch.efer = vmcs12->guest_ia32_efer; else if (vmcs12->vm_entry_controls & VM_ENTRY_IA32E_MODE) vcpu->arch.efer |= (EFER_LMA | EFER_LME); @@ -10367,73 +10323,22 @@ static int prepare_vmcs02(struct kvm_vcpu *vcpu, struct vmcs12 *vmcs12, return 0; } -/* - * nested_vmx_run() handles a nested entry, i.e., a VMLAUNCH or VMRESUME on L1 - * for running an L2 nested guest. - */ -static int nested_vmx_run(struct kvm_vcpu *vcpu, bool launch) +static int check_vmentry_prereqs(struct kvm_vcpu *vcpu, struct vmcs12 *vmcs12) { - struct vmcs12 *vmcs12; struct vcpu_vmx *vmx = to_vmx(vcpu); - int cpu; - struct loaded_vmcs *vmcs02; - bool ia32e; - u32 msr_entry_idx; - unsigned long exit_qualification; - - if (!nested_vmx_check_permission(vcpu)) - return 1; - - if (!nested_vmx_check_vmcs12(vcpu)) - goto out; - - vmcs12 = get_vmcs12(vcpu); - - if (enable_shadow_vmcs) - copy_shadow_to_vmcs12(vmx); - - /* - * The nested entry process starts with enforcing various prerequisites - * on vmcs12 as required by the Intel SDM, and act appropriately when - * they fail: As the SDM explains, some conditions should cause the - * instruction to fail, while others will cause the instruction to seem - * to succeed, but return an EXIT_REASON_INVALID_STATE. - * To speed up the normal (success) code path, we should avoid checking - * for misconfigurations which will anyway be caught by the processor - * when using the merged vmcs02. - */ - if (vmcs12->launch_state == launch) { - nested_vmx_failValid(vcpu, - launch ? VMXERR_VMLAUNCH_NONCLEAR_VMCS - : VMXERR_VMRESUME_NONLAUNCHED_VMCS); - goto out; - } if (vmcs12->guest_activity_state != GUEST_ACTIVITY_ACTIVE && - vmcs12->guest_activity_state != GUEST_ACTIVITY_HLT) { - nested_vmx_failValid(vcpu, VMXERR_ENTRY_INVALID_CONTROL_FIELD); - goto out; - } + vmcs12->guest_activity_state != GUEST_ACTIVITY_HLT) + return VMXERR_ENTRY_INVALID_CONTROL_FIELD; - if (!nested_get_vmcs12_pages(vcpu, vmcs12)) { - nested_vmx_failValid(vcpu, VMXERR_ENTRY_INVALID_CONTROL_FIELD); - goto out; - } + if (nested_vmx_check_msr_bitmap_controls(vcpu, vmcs12)) + return VMXERR_ENTRY_INVALID_CONTROL_FIELD; - if (nested_vmx_check_msr_bitmap_controls(vcpu, vmcs12)) { - nested_vmx_failValid(vcpu, VMXERR_ENTRY_INVALID_CONTROL_FIELD); - goto out; - } + if (nested_vmx_check_apicv_controls(vcpu, vmcs12)) + return VMXERR_ENTRY_INVALID_CONTROL_FIELD; - if (nested_vmx_check_apicv_controls(vcpu, vmcs12)) { - nested_vmx_failValid(vcpu, VMXERR_ENTRY_INVALID_CONTROL_FIELD); - goto out; - } - - if (nested_vmx_check_msr_switch_controls(vcpu, vmcs12)) { - nested_vmx_failValid(vcpu, VMXERR_ENTRY_INVALID_CONTROL_FIELD); - goto out; - } + if (nested_vmx_check_msr_switch_controls(vcpu, vmcs12)) + return VMXERR_ENTRY_INVALID_CONTROL_FIELD; if (!vmx_control_verify(vmcs12->cpu_based_vm_exec_control, vmx->nested.nested_vmx_procbased_ctls_low, @@ -10450,28 +10355,30 @@ static int nested_vmx_run(struct kvm_vcpu *vcpu, bool launch) !vmx_control_verify(vmcs12->vm_entry_controls, vmx->nested.nested_vmx_entry_ctls_low, vmx->nested.nested_vmx_entry_ctls_high)) - { - nested_vmx_failValid(vcpu, VMXERR_ENTRY_INVALID_CONTROL_FIELD); - goto out; - } + return VMXERR_ENTRY_INVALID_CONTROL_FIELD; if (!nested_host_cr0_valid(vcpu, vmcs12->host_cr0) || !nested_host_cr4_valid(vcpu, vmcs12->host_cr4) || - !nested_cr3_valid(vcpu, vmcs12->host_cr3)) { - nested_vmx_failValid(vcpu, - VMXERR_ENTRY_INVALID_HOST_STATE_FIELD); - goto out; - } + !nested_cr3_valid(vcpu, vmcs12->host_cr3)) + return VMXERR_ENTRY_INVALID_HOST_STATE_FIELD; + + return 0; +} + +static int check_vmentry_postreqs(struct kvm_vcpu *vcpu, struct vmcs12 *vmcs12, + u32 *exit_qual) +{ + bool ia32e; + + *exit_qual = ENTRY_FAIL_DEFAULT; if (!nested_guest_cr0_valid(vcpu, vmcs12->guest_cr0) || - !nested_guest_cr4_valid(vcpu, vmcs12->guest_cr4)) { - nested_vmx_entry_failure(vcpu, vmcs12, - EXIT_REASON_INVALID_STATE, ENTRY_FAIL_DEFAULT); + !nested_guest_cr4_valid(vcpu, vmcs12->guest_cr4)) return 1; - } - if (vmcs12->vmcs_link_pointer != -1ull) { - nested_vmx_entry_failure(vcpu, vmcs12, - EXIT_REASON_INVALID_STATE, ENTRY_FAIL_VMCS_LINK_PTR); + + if (!nested_cpu_has2(vmcs12, SECONDARY_EXEC_SHADOW_VMCS) && + vmcs12->vmcs_link_pointer != -1ull) { + *exit_qual = ENTRY_FAIL_VMCS_LINK_PTR; return 1; } @@ -10484,16 +10391,14 @@ static int nested_vmx_run(struct kvm_vcpu *vcpu, bool launch) * to bit 8 (LME) if bit 31 in the CR0 field (corresponding to * CR0.PG) is 1. */ - if (vmcs12->vm_entry_controls & VM_ENTRY_LOAD_IA32_EFER) { + if (to_vmx(vcpu)->nested.nested_run_pending && + (vmcs12->vm_entry_controls & VM_ENTRY_LOAD_IA32_EFER)) { ia32e = (vmcs12->vm_entry_controls & VM_ENTRY_IA32E_MODE) != 0; if (!kvm_valid_efer(vcpu, vmcs12->guest_ia32_efer) || ia32e != !!(vmcs12->guest_ia32_efer & EFER_LMA) || ((vmcs12->guest_cr0 & X86_CR0_PG) && - ia32e != !!(vmcs12->guest_ia32_efer & EFER_LME))) { - nested_vmx_entry_failure(vcpu, vmcs12, - EXIT_REASON_INVALID_STATE, ENTRY_FAIL_DEFAULT); + ia32e != !!(vmcs12->guest_ia32_efer & EFER_LME))) return 1; - } } /* @@ -10507,28 +10412,26 @@ static int nested_vmx_run(struct kvm_vcpu *vcpu, bool launch) VM_EXIT_HOST_ADDR_SPACE_SIZE) != 0; if (!kvm_valid_efer(vcpu, vmcs12->host_ia32_efer) || ia32e != !!(vmcs12->host_ia32_efer & EFER_LMA) || - ia32e != !!(vmcs12->host_ia32_efer & EFER_LME)) { - nested_vmx_entry_failure(vcpu, vmcs12, - EXIT_REASON_INVALID_STATE, ENTRY_FAIL_DEFAULT); + ia32e != !!(vmcs12->host_ia32_efer & EFER_LME)) return 1; - } } - /* - * We're finally done with prerequisite checking, and can start with - * the nested entry. - */ + return 0; +} + +static int enter_vmx_non_root_mode(struct kvm_vcpu *vcpu, bool from_vmentry) +{ + struct vcpu_vmx *vmx = to_vmx(vcpu); + struct vmcs12 *vmcs12 = get_vmcs12(vcpu); + struct loaded_vmcs *vmcs02; + int cpu; + u32 msr_entry_idx; + u32 exit_qual; vmcs02 = nested_get_current_vmcs02(vmx); if (!vmcs02) return -ENOMEM; - /* - * After this point, the trap flag no longer triggers a singlestep trap - * on the vm entry instructions. Don't call - * kvm_skip_emulated_instruction. - */ - skip_emulated_instruction(vcpu); enter_guest_mode(vcpu); if (!(vmcs12->vm_entry_controls & VM_ENTRY_LOAD_DEBUG_CONTROLS)) @@ -10543,14 +10446,16 @@ static int nested_vmx_run(struct kvm_vcpu *vcpu, bool launch) vmx_segment_cache_clear(vmx); - if (prepare_vmcs02(vcpu, vmcs12, &exit_qualification)) { + if (prepare_vmcs02(vcpu, vmcs12, from_vmentry, &exit_qual)) { leave_guest_mode(vcpu); vmx_load_vmcs01(vcpu); nested_vmx_entry_failure(vcpu, vmcs12, - EXIT_REASON_INVALID_STATE, exit_qualification); + EXIT_REASON_INVALID_STATE, exit_qual); return 1; } + nested_get_vmcs12_pages(vcpu, vmcs12); + msr_entry_idx = nested_vmx_load_msr(vcpu, vmcs12->vm_entry_msr_load_addr, vmcs12->vm_entry_msr_load_count); @@ -10564,17 +10469,90 @@ static int nested_vmx_run(struct kvm_vcpu *vcpu, bool launch) vmcs12->launch_state = 1; - if (vmcs12->guest_activity_state == GUEST_ACTIVITY_HLT) - return kvm_vcpu_halt(vcpu); - - vmx->nested.nested_run_pending = 1; - /* * Note no nested_vmx_succeed or nested_vmx_fail here. At this point * we are no longer running L1, and VMLAUNCH/VMRESUME has not yet * returned as far as L1 is concerned. It will only return (and set * the success flag) when L2 exits (see nested_vmx_vmexit()). */ + return 0; +} + +/* + * nested_vmx_run() handles a nested entry, i.e., a VMLAUNCH or VMRESUME on L1 + * for running an L2 nested guest. + */ +static int nested_vmx_run(struct kvm_vcpu *vcpu, bool launch) +{ + struct vmcs12 *vmcs12; + struct vcpu_vmx *vmx = to_vmx(vcpu); + u32 exit_qual; + int ret; + + if (!nested_vmx_check_permission(vcpu)) + return 1; + + if (!nested_vmx_check_vmcs12(vcpu)) + goto out; + + vmcs12 = get_vmcs12(vcpu); + + if (enable_shadow_vmcs) + copy_shadow_to_vmcs12(vmx); + + /* + * The nested entry process starts with enforcing various prerequisites + * on vmcs12 as required by the Intel SDM, and act appropriately when + * they fail: As the SDM explains, some conditions should cause the + * instruction to fail, while others will cause the instruction to seem + * to succeed, but return an EXIT_REASON_INVALID_STATE. + * To speed up the normal (success) code path, we should avoid checking + * for misconfigurations which will anyway be caught by the processor + * when using the merged vmcs02. + */ + if (vmcs12->launch_state == launch) { + nested_vmx_failValid(vcpu, + launch ? VMXERR_VMLAUNCH_NONCLEAR_VMCS + : VMXERR_VMRESUME_NONLAUNCHED_VMCS); + goto out; + } + + ret = check_vmentry_prereqs(vcpu, vmcs12); + if (ret) { + nested_vmx_failValid(vcpu, ret); + goto out; + } + + /* + * After this point, the trap flag no longer triggers a singlestep trap + * on the vm entry instructions; don't call kvm_skip_emulated_instruction. + * This is not 100% correct; for performance reasons, we delegate most + * of the checks on host state to the processor. If those fail, + * the singlestep trap is missed. + */ + skip_emulated_instruction(vcpu); + + ret = check_vmentry_postreqs(vcpu, vmcs12, &exit_qual); + if (ret) { + nested_vmx_entry_failure(vcpu, vmcs12, + EXIT_REASON_INVALID_STATE, exit_qual); + return 1; + } + + /* + * We're finally done with prerequisite checking, and can start with + * the nested entry. + */ + + ret = enter_vmx_non_root_mode(vcpu, true); + if (ret) + return ret; + + if (vmcs12->guest_activity_state == GUEST_ACTIVITY_HLT) + return kvm_vcpu_halt(vcpu); + + vmx->nested.nested_run_pending = 1; + return 1; out: @@ -10696,7 +10674,8 @@ static int vmx_check_nested_events(struct kvm_vcpu *vcpu, bool external_intr) return 0; } - return vmx_complete_nested_posted_interrupt(vcpu); + vmx_complete_nested_posted_interrupt(vcpu); + return 0; } static u32 vmx_get_preemption_timer_value(struct kvm_vcpu *vcpu) @@ -10714,21 +10693,13 @@ static u32 vmx_get_preemption_timer_value(struct kvm_vcpu *vcpu) } /* - * prepare_vmcs12 is part of what we need to do when the nested L2 guest exits - * and we want to prepare to run its L1 parent. L1 keeps a vmcs for L2 (vmcs12), - * and this function updates it to reflect the changes to the guest state while - * L2 was running (and perhaps made some exits which were handled directly by L0 - * without going back to L1), and to reflect the exit reason. - * Note that we do not have to copy here all VMCS fields, just those that - * could have changed by the L2 guest or the exit - i.e., the guest-state and - * exit-information fields only. Other fields are modified by L1 with VMWRITE, - * which already writes to vmcs12 directly. + * Update the guest state fields of vmcs12 to reflect changes that + * occurred while L2 was running. (The "IA-32e mode guest" bit of the + * VM-entry controls is also updated, since this is really a guest + * state bit.) */ -static void prepare_vmcs12(struct kvm_vcpu *vcpu, struct vmcs12 *vmcs12, - u32 exit_reason, u32 exit_intr_info, - unsigned long exit_qualification) +static void sync_vmcs12(struct kvm_vcpu *vcpu, struct vmcs12 *vmcs12) { - /* update guest state fields: */ vmcs12->guest_cr0 = vmcs12_guest_cr0(vcpu, vmcs12); vmcs12->guest_cr4 = vmcs12_guest_cr4(vcpu, vmcs12); @@ -10834,6 +10805,25 @@ static void prepare_vmcs12(struct kvm_vcpu *vcpu, struct vmcs12 *vmcs12, vmcs12->guest_bndcfgs = vmcs_read64(GUEST_BNDCFGS); if (nested_cpu_has_xsaves(vmcs12)) vmcs12->xss_exit_bitmap = vmcs_read64(XSS_EXIT_BITMAP); +} + +/* + * prepare_vmcs12 is part of what we need to do when the nested L2 guest exits + * and we want to prepare to run its L1 parent. L1 keeps a vmcs for L2 (vmcs12), + * and this function updates it to reflect the changes to the guest state while + * L2 was running (and perhaps made some exits which were handled directly by L0 + * without going back to L1), and to reflect the exit reason. + * Note that we do not have to copy here all VMCS fields, just those that + * could have changed by the L2 guest or the exit - i.e., the guest-state and + * exit-information fields only. Other fields are modified by L1 with VMWRITE, + * which already writes to vmcs12 directly. + */ +static void prepare_vmcs12(struct kvm_vcpu *vcpu, struct vmcs12 *vmcs12, + u32 exit_reason, u32 exit_intr_info, + unsigned long exit_qualification) +{ + /* update guest state fields: */ + sync_vmcs12(vcpu, vmcs12); /* update exit information fields: */ @@ -10884,7 +10874,7 @@ static void load_vmcs12_host_state(struct kvm_vcpu *vcpu, struct vmcs12 *vmcs12) { struct kvm_segment seg; - unsigned long entry_failure_code; + u32 entry_failure_code; if (vmcs12->vm_exit_controls & VM_EXIT_LOAD_IA32_EFER) vcpu->arch.efer = vmcs12->host_ia32_efer; @@ -10899,24 +10889,15 @@ static void load_vmcs12_host_state(struct kvm_vcpu *vcpu, vmx_set_rflags(vcpu, X86_EFLAGS_FIXED); /* * Note that calling vmx_set_cr0 is important, even if cr0 hasn't - * actually changed, because it depends on the current state of - * fpu_active (which may have changed). - * Note that vmx_set_cr0 refers to efer set above. + * actually changed, because vmx_set_cr0 refers to efer set above. + * + * CR0_GUEST_HOST_MASK is already set in the original vmcs01 + * (KVM doesn't change it); */ + vcpu->arch.cr0_guest_owned_bits = X86_CR0_TS; vmx_set_cr0(vcpu, vmcs12->host_cr0); - /* - * If we did fpu_activate()/fpu_deactivate() during L2's run, we need - * to apply the same changes to L1's vmcs. We just set cr0 correctly, - * but we also need to update cr0_guest_host_mask and exception_bitmap. - */ - update_exception_bitmap(vcpu); - vcpu->arch.cr0_guest_owned_bits = (vcpu->fpu_active ? X86_CR0_TS : 0); - vmcs_writel(CR0_GUEST_HOST_MASK, ~vcpu->arch.cr0_guest_owned_bits); - /* - * Note that CR4_GUEST_HOST_MASK is already set in the original vmcs01 - * (KVM doesn't change it)- no reason to call set_cr4_guest_host_mask(); - */ + /* Same as above - no reason to call set_cr4_guest_host_mask(). */ vcpu->arch.cr4_guest_owned_bits = ~vmcs_readl(CR4_GUEST_HOST_MASK); kvm_set_cr4(vcpu, vmcs12->host_cr4); @@ -11545,9 +11526,6 @@ static struct kvm_x86_ops vmx_x86_ops __ro_after_init = { .get_pkru = vmx_get_pkru, - .fpu_activate = vmx_fpu_activate, - .fpu_deactivate = vmx_fpu_deactivate, - .tlb_flush = vmx_flush_tlb, .run = vmx_vcpu_run, @@ -11572,6 +11550,7 @@ static struct kvm_x86_ops vmx_x86_ops __ro_after_init = { .get_enable_apicv = vmx_get_enable_apicv, .refresh_apicv_exec_ctrl = vmx_refresh_apicv_exec_ctrl, .load_eoi_exitmap = vmx_load_eoi_exitmap, + .apicv_post_state_restore = vmx_apicv_post_state_restore, .hwapic_irr_update = vmx_hwapic_irr_update, .hwapic_isr_update = vmx_hwapic_isr_update, .sync_pir_to_irr = vmx_sync_pir_to_irr, diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c index e52c9088660f..b2a4b11274b0 100644 --- a/arch/x86/kvm/x86.c +++ b/arch/x86/kvm/x86.c @@ -180,6 +180,7 @@ struct kvm_stats_debugfs_item debugfs_entries[] = { { "insn_emulation_fail", VCPU_STAT(insn_emulation_fail) }, { "irq_injections", VCPU_STAT(irq_injections) }, { "nmi_injections", VCPU_STAT(nmi_injections) }, + { "req_event", VCPU_STAT(req_event) }, { "mmu_shadow_zapped", VM_STAT(mmu_shadow_zapped) }, { "mmu_pte_write", VM_STAT(mmu_pte_write) }, { "mmu_pte_updated", VM_STAT(mmu_pte_updated) }, @@ -190,6 +191,8 @@ struct kvm_stats_debugfs_item debugfs_entries[] = { { "mmu_unsync", VM_STAT(mmu_unsync) }, { "remote_tlb_flush", VM_STAT(remote_tlb_flush) }, { "largepages", VM_STAT(lpages) }, + { "max_mmu_page_hash_collisions", + VM_STAT(max_mmu_page_hash_collisions) }, { NULL } }; @@ -1139,6 +1142,7 @@ struct pvclock_gtod_data { u64 boot_ns; u64 nsec_base; + u64 wall_time_sec; }; static struct pvclock_gtod_data pvclock_gtod_data; @@ -1162,6 +1166,8 @@ static void update_pvclock_gtod(struct timekeeper *tk) vdata->boot_ns = boot_ns; vdata->nsec_base = tk->tkr_mono.xtime_nsec; + vdata->wall_time_sec = tk->xtime_sec; + write_seqcount_end(&vdata->seq); } #endif @@ -1623,6 +1629,28 @@ static int do_monotonic_boot(s64 *t, u64 *cycle_now) return mode; } +static int do_realtime(struct timespec *ts, u64 *cycle_now) +{ + struct pvclock_gtod_data *gtod = &pvclock_gtod_data; + unsigned long seq; + int mode; + u64 ns; + + do { + seq = read_seqcount_begin(>od->seq); + mode = gtod->clock.vclock_mode; + ts->tv_sec = gtod->wall_time_sec; + ns = gtod->nsec_base; + ns += vgettsc(cycle_now); + ns >>= gtod->clock.shift; + } while (unlikely(read_seqcount_retry(>od->seq, seq))); + + ts->tv_sec += __iter_div_u64_rem(ns, NSEC_PER_SEC, &ns); + ts->tv_nsec = ns; + + return mode; +} + /* returns true if host is using tsc clocksource */ static bool kvm_get_time_and_clockread(s64 *kernel_ns, u64 *cycle_now) { @@ -1632,6 +1660,17 @@ static bool kvm_get_time_and_clockread(s64 *kernel_ns, u64 *cycle_now) return do_monotonic_boot(kernel_ns, cycle_now) == VCLOCK_TSC; } + +/* returns true if host is using tsc clocksource */ +static bool kvm_get_walltime_and_clockread(struct timespec *ts, + u64 *cycle_now) +{ + /* checked again under seqlock below */ + if (pvclock_gtod_data.clock.vclock_mode != VCLOCK_TSC) + return false; + + return do_realtime(ts, cycle_now) == VCLOCK_TSC; +} #endif /* @@ -1772,7 +1811,7 @@ static void kvm_setup_pvclock_page(struct kvm_vcpu *v) struct kvm_vcpu_arch *vcpu = &v->arch; struct pvclock_vcpu_time_info guest_hv_clock; - if (unlikely(kvm_read_guest_cached(v->kvm, &vcpu->pv_time, + if (unlikely(kvm_vcpu_read_guest_cached(v, &vcpu->pv_time, &guest_hv_clock, sizeof(guest_hv_clock)))) return; @@ -1793,9 +1832,9 @@ static void kvm_setup_pvclock_page(struct kvm_vcpu *v) BUILD_BUG_ON(offsetof(struct pvclock_vcpu_time_info, version) != 0); vcpu->hv_clock.version = guest_hv_clock.version + 1; - kvm_write_guest_cached(v->kvm, &vcpu->pv_time, - &vcpu->hv_clock, - sizeof(vcpu->hv_clock.version)); + kvm_vcpu_write_guest_cached(v, &vcpu->pv_time, + &vcpu->hv_clock, + sizeof(vcpu->hv_clock.version)); smp_wmb(); @@ -1809,16 +1848,16 @@ static void kvm_setup_pvclock_page(struct kvm_vcpu *v) trace_kvm_pvclock_update(v->vcpu_id, &vcpu->hv_clock); - kvm_write_guest_cached(v->kvm, &vcpu->pv_time, - &vcpu->hv_clock, - sizeof(vcpu->hv_clock)); + kvm_vcpu_write_guest_cached(v, &vcpu->pv_time, + &vcpu->hv_clock, + sizeof(vcpu->hv_clock)); smp_wmb(); vcpu->hv_clock.version++; - kvm_write_guest_cached(v->kvm, &vcpu->pv_time, - &vcpu->hv_clock, - sizeof(vcpu->hv_clock.version)); + kvm_vcpu_write_guest_cached(v, &vcpu->pv_time, + &vcpu->hv_clock, + sizeof(vcpu->hv_clock.version)); } static int kvm_guest_time_update(struct kvm_vcpu *v) @@ -2051,7 +2090,7 @@ static int kvm_pv_enable_async_pf(struct kvm_vcpu *vcpu, u64 data) return 0; } - if (kvm_gfn_to_hva_cache_init(vcpu->kvm, &vcpu->arch.apf.data, gpa, + if (kvm_vcpu_gfn_to_hva_cache_init(vcpu, &vcpu->arch.apf.data, gpa, sizeof(u32))) return 1; @@ -2070,7 +2109,7 @@ static void record_steal_time(struct kvm_vcpu *vcpu) if (!(vcpu->arch.st.msr_val & KVM_MSR_ENABLED)) return; - if (unlikely(kvm_read_guest_cached(vcpu->kvm, &vcpu->arch.st.stime, + if (unlikely(kvm_vcpu_read_guest_cached(vcpu, &vcpu->arch.st.stime, &vcpu->arch.st.steal, sizeof(struct kvm_steal_time)))) return; @@ -2081,7 +2120,7 @@ static void record_steal_time(struct kvm_vcpu *vcpu) vcpu->arch.st.steal.version += 1; - kvm_write_guest_cached(vcpu->kvm, &vcpu->arch.st.stime, + kvm_vcpu_write_guest_cached(vcpu, &vcpu->arch.st.stime, &vcpu->arch.st.steal, sizeof(struct kvm_steal_time)); smp_wmb(); @@ -2090,14 +2129,14 @@ static void record_steal_time(struct kvm_vcpu *vcpu) vcpu->arch.st.last_steal; vcpu->arch.st.last_steal = current->sched_info.run_delay; - kvm_write_guest_cached(vcpu->kvm, &vcpu->arch.st.stime, + kvm_vcpu_write_guest_cached(vcpu, &vcpu->arch.st.stime, &vcpu->arch.st.steal, sizeof(struct kvm_steal_time)); smp_wmb(); vcpu->arch.st.steal.version += 1; - kvm_write_guest_cached(vcpu->kvm, &vcpu->arch.st.stime, + kvm_vcpu_write_guest_cached(vcpu, &vcpu->arch.st.stime, &vcpu->arch.st.steal, sizeof(struct kvm_steal_time)); } @@ -2202,7 +2241,7 @@ int kvm_set_msr_common(struct kvm_vcpu *vcpu, struct msr_data *msr_info) if (!(data & 1)) break; - if (kvm_gfn_to_hva_cache_init(vcpu->kvm, + if (kvm_vcpu_gfn_to_hva_cache_init(vcpu, &vcpu->arch.pv_time, data & ~1ULL, sizeof(struct pvclock_vcpu_time_info))) vcpu->arch.pv_time_enabled = false; @@ -2223,7 +2262,7 @@ int kvm_set_msr_common(struct kvm_vcpu *vcpu, struct msr_data *msr_info) if (data & KVM_STEAL_RESERVED_MASK) return 1; - if (kvm_gfn_to_hva_cache_init(vcpu->kvm, &vcpu->arch.st.stime, + if (kvm_vcpu_gfn_to_hva_cache_init(vcpu, &vcpu->arch.st.stime, data & KVM_STEAL_VALID_BITS, sizeof(struct kvm_steal_time))) return 1; @@ -2633,6 +2672,7 @@ int kvm_vm_ioctl_check_extension(struct kvm *kvm, long ext) case KVM_CAP_DISABLE_QUIRKS: case KVM_CAP_SET_BOOT_CPU_ID: case KVM_CAP_SPLIT_IRQCHIP: + case KVM_CAP_IMMEDIATE_EXIT: #ifdef CONFIG_KVM_DEVICE_ASSIGNMENT case KVM_CAP_ASSIGN_DEV_IRQ: case KVM_CAP_PCI_2_3: @@ -2836,7 +2876,7 @@ static void kvm_steal_time_set_preempted(struct kvm_vcpu *vcpu) vcpu->arch.st.steal.preempted = 1; - kvm_write_guest_offset_cached(vcpu->kvm, &vcpu->arch.st.stime, + kvm_vcpu_write_guest_offset_cached(vcpu, &vcpu->arch.st.stime, &vcpu->arch.st.steal.preempted, offsetof(struct kvm_steal_time, preempted), sizeof(vcpu->arch.st.steal.preempted)); @@ -2870,7 +2910,7 @@ void kvm_arch_vcpu_put(struct kvm_vcpu *vcpu) static int kvm_vcpu_ioctl_get_lapic(struct kvm_vcpu *vcpu, struct kvm_lapic_state *s) { - if (vcpu->arch.apicv_active) + if (kvm_x86_ops->sync_pir_to_irr && vcpu->arch.apicv_active) kvm_x86_ops->sync_pir_to_irr(vcpu); return kvm_apic_get_state(vcpu, s); @@ -3897,7 +3937,7 @@ static int kvm_vm_ioctl_enable_cap(struct kvm *kvm, goto split_irqchip_unlock; /* Pairs with irqchip_in_kernel. */ smp_wmb(); - kvm->arch.irqchip_split = true; + kvm->arch.irqchip_mode = KVM_IRQCHIP_SPLIT; kvm->arch.nr_reserved_ioapic_pins = cap->args[0]; r = 0; split_irqchip_unlock: @@ -3960,40 +4000,41 @@ long kvm_arch_vm_ioctl(struct file *filp, r = kvm_vm_ioctl_get_nr_mmu_pages(kvm); break; case KVM_CREATE_IRQCHIP: { - struct kvm_pic *vpic; - mutex_lock(&kvm->lock); + r = -EEXIST; - if (kvm->arch.vpic) + if (irqchip_in_kernel(kvm)) goto create_irqchip_unlock; + r = -EINVAL; if (kvm->created_vcpus) goto create_irqchip_unlock; - r = -ENOMEM; - vpic = kvm_create_pic(kvm); - if (vpic) { - r = kvm_ioapic_init(kvm); - if (r) { - mutex_lock(&kvm->slots_lock); - kvm_destroy_pic(vpic); - mutex_unlock(&kvm->slots_lock); - goto create_irqchip_unlock; - } - } else + + r = kvm_pic_init(kvm); + if (r) + goto create_irqchip_unlock; + + r = kvm_ioapic_init(kvm); + if (r) { + mutex_lock(&kvm->slots_lock); + kvm_pic_destroy(kvm); + mutex_unlock(&kvm->slots_lock); goto create_irqchip_unlock; + } + r = kvm_setup_default_irq_routing(kvm); if (r) { mutex_lock(&kvm->slots_lock); mutex_lock(&kvm->irq_lock); kvm_ioapic_destroy(kvm); - kvm_destroy_pic(vpic); + kvm_pic_destroy(kvm); mutex_unlock(&kvm->irq_lock); mutex_unlock(&kvm->slots_lock); goto create_irqchip_unlock; } - /* Write kvm->irq_routing before kvm->arch.vpic. */ + /* Write kvm->irq_routing before enabling irqchip_in_kernel. */ smp_wmb(); - kvm->arch.vpic = vpic; + kvm->arch.irqchip_mode = KVM_IRQCHIP_KERNEL; create_irqchip_unlock: mutex_unlock(&kvm->lock); break; @@ -4029,7 +4070,7 @@ long kvm_arch_vm_ioctl(struct file *filp, } r = -ENXIO; - if (!irqchip_in_kernel(kvm) || irqchip_split(kvm)) + if (!irqchip_kernel(kvm)) goto get_irqchip_out; r = kvm_vm_ioctl_get_irqchip(kvm, chip); if (r) @@ -4053,7 +4094,7 @@ long kvm_arch_vm_ioctl(struct file *filp, } r = -ENXIO; - if (!irqchip_in_kernel(kvm) || irqchip_split(kvm)) + if (!irqchip_kernel(kvm)) goto set_irqchip_out; r = kvm_vm_ioctl_set_irqchip(kvm, chip); if (r) @@ -4462,6 +4503,21 @@ out: } EXPORT_SYMBOL_GPL(kvm_write_guest_virt_system); +static int vcpu_is_mmio_gpa(struct kvm_vcpu *vcpu, unsigned long gva, + gpa_t gpa, bool write) +{ + /* For APIC access vmexit */ + if ((gpa & PAGE_MASK) == APIC_DEFAULT_PHYS_BASE) + return 1; + + if (vcpu_match_mmio_gpa(vcpu, gpa)) { + trace_vcpu_match_mmio(gva, gpa, write, true); + return 1; + } + + return 0; +} + static int vcpu_mmio_gva_to_gpa(struct kvm_vcpu *vcpu, unsigned long gva, gpa_t *gpa, struct x86_exception *exception, bool write) @@ -4488,16 +4544,7 @@ static int vcpu_mmio_gva_to_gpa(struct kvm_vcpu *vcpu, unsigned long gva, if (*gpa == UNMAPPED_GVA) return -1; - /* For APIC access vmexit */ - if ((*gpa & PAGE_MASK) == APIC_DEFAULT_PHYS_BASE) - return 1; - - if (vcpu_match_mmio_gpa(vcpu, *gpa)) { - trace_vcpu_match_mmio(gva, *gpa, write, true); - return 1; - } - - return 0; + return vcpu_is_mmio_gpa(vcpu, gva, *gpa, write); } int emulator_write_phys(struct kvm_vcpu *vcpu, gpa_t gpa, @@ -4594,6 +4641,22 @@ static int emulator_read_write_onepage(unsigned long addr, void *val, int handled, ret; bool write = ops->write; struct kvm_mmio_fragment *frag; + struct x86_emulate_ctxt *ctxt = &vcpu->arch.emulate_ctxt; + + /* + * If the exit was due to a NPF we may already have a GPA. + * If the GPA is present, use it to avoid the GVA to GPA table walk. + * Note, this cannot be used on string operations since string + * operation using rep will only have the initial GPA from the NPF + * occurred. + */ + if (vcpu->arch.gpa_available && + emulator_can_use_gpa(ctxt) && + vcpu_is_mmio_gpa(vcpu, addr, exception->address, write) && + (addr & ~PAGE_MASK) == (exception->address & ~PAGE_MASK)) { + gpa = exception->address; + goto mmio; + } ret = vcpu_mmio_gva_to_gpa(vcpu, addr, &gpa, exception, write); @@ -5610,6 +5673,9 @@ int x86_emulate_instruction(struct kvm_vcpu *vcpu, } restart: + /* Save the faulting GPA (cr2) in the address field */ + ctxt->exception.address = cr2; + r = x86_emulate_insn(ctxt); if (r == EMULATION_INTERCEPTED) @@ -5924,9 +5990,6 @@ static void kvm_set_mmio_spte_mask(void) /* Mask the reserved physical address bits. */ mask = rsvd_bits(maxphyaddr, 51); - /* Bit 62 is always reserved for 32bit host. */ - mask |= 0x3ull << 62; - /* Set the present bit. */ mask |= 1ull; @@ -6025,7 +6088,7 @@ int kvm_arch_init(void *opaque) kvm_mmu_set_mask_ptes(PT_USER_MASK, PT_ACCESSED_MASK, PT_DIRTY_MASK, PT64_NX_MASK, 0, - PT_PRESENT_MASK); + PT_PRESENT_MASK, 0); kvm_timer_init(); perf_register_guest_info_callbacks(&kvm_guest_cbs); @@ -6087,6 +6150,35 @@ int kvm_emulate_halt(struct kvm_vcpu *vcpu) } EXPORT_SYMBOL_GPL(kvm_emulate_halt); +#ifdef CONFIG_X86_64 +static int kvm_pv_clock_pairing(struct kvm_vcpu *vcpu, gpa_t paddr, + unsigned long clock_type) +{ + struct kvm_clock_pairing clock_pairing; + struct timespec ts; + u64 cycle; + int ret; + + if (clock_type != KVM_CLOCK_PAIRING_WALLCLOCK) + return -KVM_EOPNOTSUPP; + + if (kvm_get_walltime_and_clockread(&ts, &cycle) == false) + return -KVM_EOPNOTSUPP; + + clock_pairing.sec = ts.tv_sec; + clock_pairing.nsec = ts.tv_nsec; + clock_pairing.tsc = kvm_read_l1_tsc(vcpu, cycle); + clock_pairing.flags = 0; + + ret = 0; + if (kvm_write_guest(vcpu->kvm, paddr, &clock_pairing, + sizeof(struct kvm_clock_pairing))) + ret = -KVM_EFAULT; + + return ret; +} +#endif + /* * kvm_pv_kick_cpu_op: Kick a vcpu. * @@ -6151,6 +6243,11 @@ int kvm_emulate_hypercall(struct kvm_vcpu *vcpu) kvm_pv_kick_cpu_op(vcpu->kvm, a0, a1); ret = 0; break; +#ifdef CONFIG_X86_64 + case KVM_HC_CLOCK_PAIRING: + ret = kvm_pv_clock_pairing(vcpu, a0, a1); + break; +#endif default: ret = -KVM_ENOSYS; break; @@ -6564,7 +6661,7 @@ static void vcpu_scan_ioapic(struct kvm_vcpu *vcpu) if (irqchip_split(vcpu->kvm)) kvm_scan_ioapic_routes(vcpu, vcpu->arch.ioapic_handled_vectors); else { - if (vcpu->arch.apicv_active) + if (kvm_x86_ops->sync_pir_to_irr && vcpu->arch.apicv_active) kvm_x86_ops->sync_pir_to_irr(vcpu); kvm_ioapic_scan_entry(vcpu, vcpu->arch.ioapic_handled_vectors); } @@ -6655,10 +6752,6 @@ static int vcpu_enter_guest(struct kvm_vcpu *vcpu) r = 0; goto out; } - if (kvm_check_request(KVM_REQ_DEACTIVATE_FPU, vcpu)) { - vcpu->fpu_active = 0; - kvm_x86_ops->fpu_deactivate(vcpu); - } if (kvm_check_request(KVM_REQ_APF_HALT, vcpu)) { /* Page is swapped out. Do synthetic halt */ vcpu->arch.apf.halted = true; @@ -6718,21 +6811,8 @@ static int vcpu_enter_guest(struct kvm_vcpu *vcpu) kvm_hv_process_stimers(vcpu); } - /* - * KVM_REQ_EVENT is not set when posted interrupts are set by - * VT-d hardware, so we have to update RVI unconditionally. - */ - if (kvm_lapic_enabled(vcpu)) { - /* - * Update architecture specific hints for APIC - * virtual interrupt delivery. - */ - if (vcpu->arch.apicv_active) - kvm_x86_ops->hwapic_irr_update(vcpu, - kvm_lapic_find_highest_irr(vcpu)); - } - if (kvm_check_request(KVM_REQ_EVENT, vcpu) || req_int_win) { + ++vcpu->stat.req_event; kvm_apic_accept_events(vcpu); if (vcpu->arch.mp_state == KVM_MP_STATE_INIT_RECEIVED) { r = 1; @@ -6773,22 +6853,40 @@ static int vcpu_enter_guest(struct kvm_vcpu *vcpu) preempt_disable(); kvm_x86_ops->prepare_guest_switch(vcpu); - if (vcpu->fpu_active) - kvm_load_guest_fpu(vcpu); + kvm_load_guest_fpu(vcpu); + + /* + * Disable IRQs before setting IN_GUEST_MODE. Posted interrupt + * IPI are then delayed after guest entry, which ensures that they + * result in virtual interrupt delivery. + */ + local_irq_disable(); vcpu->mode = IN_GUEST_MODE; srcu_read_unlock(&vcpu->kvm->srcu, vcpu->srcu_idx); /* - * We should set ->mode before check ->requests, - * Please see the comment in kvm_make_all_cpus_request. - * This also orders the write to mode from any reads - * to the page tables done while the VCPU is running. - * Please see the comment in kvm_flush_remote_tlbs. + * 1) We should set ->mode before checking ->requests. Please see + * the comment in kvm_make_all_cpus_request. + * + * 2) For APICv, we should set ->mode before checking PIR.ON. This + * pairs with the memory barrier implicit in pi_test_and_set_on + * (see vmx_deliver_posted_interrupt). + * + * 3) This also orders the write to mode from any reads to the page + * tables done while the VCPU is running. Please see the comment + * in kvm_flush_remote_tlbs. */ smp_mb__after_srcu_read_unlock(); - local_irq_disable(); + /* + * This handles the case where a posted interrupt was + * notified with kvm_vcpu_kick. + */ + if (kvm_lapic_enabled(vcpu)) { + if (kvm_x86_ops->sync_pir_to_irr && vcpu->arch.apicv_active) + kvm_x86_ops->sync_pir_to_irr(vcpu); + } if (vcpu->mode == EXITING_GUEST_MODE || vcpu->requests || need_resched() || signal_pending(current)) { @@ -6927,6 +7025,9 @@ static inline int vcpu_block(struct kvm *kvm, struct kvm_vcpu *vcpu) static inline bool kvm_vcpu_running(struct kvm_vcpu *vcpu) { + if (is_guest_mode(vcpu) && kvm_x86_ops->check_nested_events) + kvm_x86_ops->check_nested_events(vcpu, false); + return (vcpu->arch.mp_state == KVM_MP_STATE_RUNNABLE && !vcpu->arch.apf.halted); } @@ -7098,7 +7199,10 @@ int kvm_arch_vcpu_ioctl_run(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run) } else WARN_ON(vcpu->arch.pio.count || vcpu->mmio_needed); - r = vcpu_run(vcpu); + if (kvm_run->immediate_exit) + r = -EINTR; + else + r = vcpu_run(vcpu); out: post_kvm_run_save(vcpu); @@ -8293,9 +8397,6 @@ static inline bool kvm_vcpu_has_events(struct kvm_vcpu *vcpu) int kvm_arch_vcpu_runnable(struct kvm_vcpu *vcpu) { - if (is_guest_mode(vcpu) && kvm_x86_ops->check_nested_events) - kvm_x86_ops->check_nested_events(vcpu, false); - return kvm_vcpu_running(vcpu) || kvm_vcpu_has_events(vcpu); } @@ -8432,9 +8533,8 @@ static void kvm_del_async_pf_gfn(struct kvm_vcpu *vcpu, gfn_t gfn) static int apf_put_user(struct kvm_vcpu *vcpu, u32 val) { - - return kvm_write_guest_cached(vcpu->kvm, &vcpu->arch.apf.data, &val, - sizeof(val)); + return kvm_vcpu_write_guest_cached(vcpu, &vcpu->arch.apf.data, &val, + sizeof(val)); } void kvm_arch_async_page_not_present(struct kvm_vcpu *vcpu, diff --git a/drivers/iommu/dmar.c b/drivers/iommu/dmar.c index d9c0decfc91a..36e3f430d265 100644 --- a/drivers/iommu/dmar.c +++ b/drivers/iommu/dmar.c @@ -1108,8 +1108,10 @@ error: static void free_iommu(struct intel_iommu *iommu) { - iommu_device_sysfs_remove(&iommu->iommu); - iommu_device_unregister(&iommu->iommu); + if (intel_iommu_enabled) { + iommu_device_unregister(&iommu->iommu); + iommu_device_sysfs_remove(&iommu->iommu); + } if (iommu->irq) { if (iommu->pr_irq) { diff --git a/drivers/of/base.c b/drivers/of/base.c index a88387bc0ac1..d7c4629a3a2d 100644 --- a/drivers/of/base.c +++ b/drivers/of/base.c @@ -843,8 +843,11 @@ struct device_node *of_find_node_opts_by_path(const char *path, const char **opt if (!np) np = of_node_get(of_root); while (np && *path == '/') { + struct device_node *tmp = np; + path++; /* Increment past '/' delimiter */ np = __of_find_node_by_path(np, path); + of_node_put(tmp); path = strchrnul(path, '/'); if (separator && separator < path) break; @@ -2495,3 +2498,40 @@ struct device_node *of_graph_get_remote_port(const struct device_node *node) return of_get_next_parent(np); } EXPORT_SYMBOL(of_graph_get_remote_port); + +/** + * of_graph_get_remote_node() - get remote parent device_node for given port/endpoint + * @node: pointer to parent device_node containing graph port/endpoint + * @port: identifier (value of reg property) of the parent port node + * @endpoint: identifier (value of reg property) of the endpoint node + * + * Return: Remote device node associated with remote endpoint node linked + * to @node. Use of_node_put() on it when done. + */ +struct device_node *of_graph_get_remote_node(const struct device_node *node, + u32 port, u32 endpoint) +{ + struct device_node *endpoint_node, *remote; + + endpoint_node = of_graph_get_endpoint_by_regs(node, port, endpoint); + if (!endpoint_node) { + pr_debug("no valid endpoint (%d, %d) for node %s\n", + port, endpoint, node->full_name); + return NULL; + } + + remote = of_graph_get_remote_port_parent(endpoint_node); + of_node_put(endpoint_node); + if (!remote) { + pr_debug("no valid remote node\n"); + return NULL; + } + + if (!of_device_is_available(remote)) { + pr_debug("not available for remote node\n"); + return NULL; + } + + return remote; +} +EXPORT_SYMBOL(of_graph_get_remote_node); diff --git a/drivers/of/fdt.c b/drivers/of/fdt.c index 82967b07f7be..e5ce4b59e162 100644 --- a/drivers/of/fdt.c +++ b/drivers/of/fdt.c @@ -9,7 +9,7 @@ * version 2 as published by the Free Software Foundation. */ -#define pr_fmt(fmt) "OF: fdt:" fmt +#define pr_fmt(fmt) "OF: fdt: " fmt #include <linux/crc32.h> #include <linux/kernel.h> diff --git a/drivers/of/irq.c b/drivers/of/irq.c index 3fda9a32defb..7c56b72d1dc6 100644 --- a/drivers/of/irq.c +++ b/drivers/of/irq.c @@ -104,7 +104,7 @@ int of_irq_parse_raw(const __be32 *addr, struct of_phandle_args *out_irq) const __be32 *match_array = initial_match_array; const __be32 *tmp, *imap, *imask, dummy_imask[] = { [0 ... MAX_PHANDLE_ARGS] = ~0 }; u32 intsize = 1, addrsize, newintsize = 0, newaddrsize = 0; - int imaplen, match, i; + int imaplen, match, i, rc = -EINVAL; #ifdef DEBUG of_print_phandle_args("of_irq_parse_raw: ", out_irq); @@ -134,7 +134,7 @@ int of_irq_parse_raw(const __be32 *addr, struct of_phandle_args *out_irq) pr_debug("of_irq_parse_raw: ipar=%s, size=%d\n", of_node_full_name(ipar), intsize); if (out_irq->args_count != intsize) - return -EINVAL; + goto fail; /* Look for this #address-cells. We have to implement the old linux * trick of looking for the parent here as some device-trees rely on it @@ -153,8 +153,10 @@ int of_irq_parse_raw(const __be32 *addr, struct of_phandle_args *out_irq) pr_debug(" -> addrsize=%d\n", addrsize); /* Range check so that the temporary buffer doesn't overflow */ - if (WARN_ON(addrsize + intsize > MAX_PHANDLE_ARGS)) + if (WARN_ON(addrsize + intsize > MAX_PHANDLE_ARGS)) { + rc = -EFAULT; goto fail; + } /* Precalculate the match array - this simplifies match loop */ for (i = 0; i < addrsize; i++) @@ -240,10 +242,11 @@ int of_irq_parse_raw(const __be32 *addr, struct of_phandle_args *out_irq) newintsize, newaddrsize); /* Check for malformed properties */ - if (WARN_ON(newaddrsize + newintsize > MAX_PHANDLE_ARGS)) - goto fail; - if (imaplen < (newaddrsize + newintsize)) + if (WARN_ON(newaddrsize + newintsize > MAX_PHANDLE_ARGS) + || (imaplen < (newaddrsize + newintsize))) { + rc = -EFAULT; goto fail; + } imap += newaddrsize + newintsize; imaplen -= newaddrsize + newintsize; @@ -271,11 +274,13 @@ int of_irq_parse_raw(const __be32 *addr, struct of_phandle_args *out_irq) ipar = newpar; newpar = NULL; } + rc = -ENOENT; /* No interrupt-map found */ + fail: of_node_put(ipar); of_node_put(newpar); - return -EINVAL; + return rc; } EXPORT_SYMBOL_GPL(of_irq_parse_raw); diff --git a/drivers/of/of_pci_irq.c b/drivers/of/of_pci_irq.c index 2306313c0029..c175d9cd0bb5 100644 --- a/drivers/of/of_pci_irq.c +++ b/drivers/of/of_pci_irq.c @@ -93,7 +93,15 @@ int of_irq_parse_pci(const struct pci_dev *pdev, struct of_phandle_args *out_irq goto err; return 0; err: - dev_err(&pdev->dev, "of_irq_parse_pci() failed with rc=%d\n", rc); + if (rc == -ENOENT) { + dev_warn(&pdev->dev, + "%s: no interrupt-map found, INTx interrupts not available\n", + __func__); + pr_warn_once("%s: possibly some PCI slots don't have level triggered interrupts capability\n", + __func__); + } else { + dev_err(&pdev->dev, "%s: failed with rc=%d\n", __func__, rc); + } return rc; } EXPORT_SYMBOL_GPL(of_irq_parse_pci); diff --git a/drivers/of/of_reserved_mem.c b/drivers/of/of_reserved_mem.c index 366d8c3c7989..d507c3569a88 100644 --- a/drivers/of/of_reserved_mem.c +++ b/drivers/of/of_reserved_mem.c @@ -354,6 +354,10 @@ int of_reserved_mem_device_init_by_idx(struct device *dev, mutex_lock(&of_rmem_assigned_device_mutex); list_add(&rd->list, &of_rmem_assigned_device_list); mutex_unlock(&of_rmem_assigned_device_mutex); + /* ensure that dma_ops is set for virtual devices + * using reserved memory + */ + of_dma_configure(dev, np); dev_info(dev, "assigned reserved memory node %s\n", rmem->name); } else { diff --git a/drivers/of/overlay.c b/drivers/of/overlay.c index 0d4cda7050e0..7827786718d8 100644 --- a/drivers/of/overlay.c +++ b/drivers/of/overlay.c @@ -18,7 +18,6 @@ #include <linux/string.h> #include <linux/ctype.h> #include <linux/errno.h> -#include <linux/string.h> #include <linux/slab.h> #include <linux/err.h> #include <linux/idr.h> @@ -314,7 +313,6 @@ static int of_build_overlay_info(struct of_overlay *ov, cnt = 0; for_each_child_of_node(tree, node) { - memset(&ovinfo[cnt], 0, sizeof(*ovinfo)); err = of_fill_overlay_info(ov, node, &ovinfo[cnt]); if (err == 0) cnt++; diff --git a/drivers/of/platform.c b/drivers/of/platform.c index b8064bc2b6eb..5dfcc967dd05 100644 --- a/drivers/of/platform.c +++ b/drivers/of/platform.c @@ -76,7 +76,7 @@ EXPORT_SYMBOL(of_find_device_by_node); * derive a unique name. If it cannot, then it will prepend names from * parent nodes until a unique name can be derived. */ -void of_device_make_bus_id(struct device *dev) +static void of_device_make_bus_id(struct device *dev) { struct device_node *node = dev->of_node; const __be32 *reg; diff --git a/drivers/of/resolver.c b/drivers/of/resolver.c index 8bf12e904fd2..7ae9863cb0a4 100644 --- a/drivers/of/resolver.c +++ b/drivers/of/resolver.c @@ -18,7 +18,6 @@ #include <linux/string.h> #include <linux/ctype.h> #include <linux/errno.h> -#include <linux/string.h> #include <linux/slab.h> /* illegal phandle value (set when unresolved) */ diff --git a/drivers/of/unittest.c b/drivers/of/unittest.c index 53c83d66eb7e..62db55b97c10 100644 --- a/drivers/of/unittest.c +++ b/drivers/of/unittest.c @@ -17,7 +17,6 @@ #include <linux/slab.h> #include <linux/device.h> #include <linux/platform_device.h> -#include <linux/of_platform.h> #include <linux/i2c.h> #include <linux/i2c-mux.h> @@ -1181,7 +1180,7 @@ static void of_unittest_destroy_tracked_overlays(void) } while (defers > 0); } -static int of_unittest_apply_overlay(int unittest_nr, int overlay_nr, +static int of_unittest_apply_overlay(int overlay_nr, int unittest_nr, int *overlay_id) { struct device_node *np = NULL; @@ -1840,7 +1839,7 @@ static void of_unittest_overlay_i2c_15(void) int ret; /* device should enable */ - ret = of_unittest_apply_overlay_check(16, 15, 0, 1, I2C_OVERLAY); + ret = of_unittest_apply_overlay_check(15, 15, 0, 1, I2C_OVERLAY); if (ret != 0) return; diff --git a/drivers/ptp/Kconfig b/drivers/ptp/Kconfig index bdce33291161..384f661a6496 100644 --- a/drivers/ptp/Kconfig +++ b/drivers/ptp/Kconfig @@ -90,4 +90,16 @@ config PTP_1588_CLOCK_PCH To compile this driver as a module, choose M here: the module will be called ptp_pch. +config PTP_1588_CLOCK_KVM + tristate "KVM virtual PTP clock" + depends on PTP_1588_CLOCK + depends on KVM_GUEST && X86 + default y + help + This driver adds support for using kvm infrastructure as a PTP + clock. This clock is only useful if you are using KVM guests. + + To compile this driver as a module, choose M here: the module + will be called ptp_kvm. + endmenu diff --git a/drivers/ptp/Makefile b/drivers/ptp/Makefile index 8b58597298de..530736161a8b 100644 --- a/drivers/ptp/Makefile +++ b/drivers/ptp/Makefile @@ -6,3 +6,4 @@ ptp-y := ptp_clock.o ptp_chardev.o ptp_sysfs.o obj-$(CONFIG_PTP_1588_CLOCK) += ptp.o obj-$(CONFIG_PTP_1588_CLOCK_IXP46X) += ptp_ixp46x.o obj-$(CONFIG_PTP_1588_CLOCK_PCH) += ptp_pch.o +obj-$(CONFIG_PTP_1588_CLOCK_KVM) += ptp_kvm.o diff --git a/drivers/ptp/ptp_kvm.c b/drivers/ptp/ptp_kvm.c new file mode 100644 index 000000000000..09b4df74291e --- /dev/null +++ b/drivers/ptp/ptp_kvm.c @@ -0,0 +1,207 @@ +/* + * Virtual PTP 1588 clock for use with KVM guests + * + * Copyright (C) 2017 Red Hat Inc. + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + */ +#include <linux/device.h> +#include <linux/err.h> +#include <linux/init.h> +#include <linux/kernel.h> +#include <linux/module.h> +#include <uapi/linux/kvm_para.h> +#include <asm/kvm_para.h> +#include <asm/pvclock.h> +#include <asm/kvmclock.h> +#include <uapi/asm/kvm_para.h> + +#include <linux/ptp_clock_kernel.h> + +struct kvm_ptp_clock { + struct ptp_clock *ptp_clock; + struct ptp_clock_info caps; +}; + +DEFINE_SPINLOCK(kvm_ptp_lock); + +static struct pvclock_vsyscall_time_info *hv_clock; + +static struct kvm_clock_pairing clock_pair; +static phys_addr_t clock_pair_gpa; + +static int ptp_kvm_get_time_fn(ktime_t *device_time, + struct system_counterval_t *system_counter, + void *ctx) +{ + unsigned long ret; + struct timespec64 tspec; + unsigned version; + int cpu; + struct pvclock_vcpu_time_info *src; + + spin_lock(&kvm_ptp_lock); + + preempt_disable_notrace(); + cpu = smp_processor_id(); + src = &hv_clock[cpu].pvti; + + do { + /* + * We are using a TSC value read in the hosts + * kvm_hc_clock_pairing handling. + * So any changes to tsc_to_system_mul + * and tsc_shift or any other pvclock + * data invalidate that measurement. + */ + version = pvclock_read_begin(src); + + ret = kvm_hypercall2(KVM_HC_CLOCK_PAIRING, + clock_pair_gpa, + KVM_CLOCK_PAIRING_WALLCLOCK); + if (ret != 0) { + pr_err_ratelimited("clock pairing hypercall ret %lu\n", ret); + spin_unlock(&kvm_ptp_lock); + preempt_enable_notrace(); + return -EOPNOTSUPP; + } + + tspec.tv_sec = clock_pair.sec; + tspec.tv_nsec = clock_pair.nsec; + ret = __pvclock_read_cycles(src, clock_pair.tsc); + } while (pvclock_read_retry(src, version)); + + preempt_enable_notrace(); + + system_counter->cycles = ret; + system_counter->cs = &kvm_clock; + + *device_time = timespec64_to_ktime(tspec); + + spin_unlock(&kvm_ptp_lock); + + return 0; +} + +static int ptp_kvm_getcrosststamp(struct ptp_clock_info *ptp, + struct system_device_crosststamp *xtstamp) +{ + return get_device_system_crosststamp(ptp_kvm_get_time_fn, NULL, + NULL, xtstamp); +} + +/* + * PTP clock operations + */ + +static int ptp_kvm_adjfreq(struct ptp_clock_info *ptp, s32 ppb) +{ + return -EOPNOTSUPP; +} + +static int ptp_kvm_adjtime(struct ptp_clock_info *ptp, s64 delta) +{ + return -EOPNOTSUPP; +} + +static int ptp_kvm_settime(struct ptp_clock_info *ptp, + const struct timespec64 *ts) +{ + return -EOPNOTSUPP; +} + +static int ptp_kvm_gettime(struct ptp_clock_info *ptp, struct timespec64 *ts) +{ + unsigned long ret; + struct timespec64 tspec; + + spin_lock(&kvm_ptp_lock); + + ret = kvm_hypercall2(KVM_HC_CLOCK_PAIRING, + clock_pair_gpa, + KVM_CLOCK_PAIRING_WALLCLOCK); + if (ret != 0) { + pr_err_ratelimited("clock offset hypercall ret %lu\n", ret); + spin_unlock(&kvm_ptp_lock); + return -EOPNOTSUPP; + } + + tspec.tv_sec = clock_pair.sec; + tspec.tv_nsec = clock_pair.nsec; + spin_unlock(&kvm_ptp_lock); + + memcpy(ts, &tspec, sizeof(struct timespec64)); + + return 0; +} + +static int ptp_kvm_enable(struct ptp_clock_info *ptp, + struct ptp_clock_request *rq, int on) +{ + return -EOPNOTSUPP; +} + +static struct ptp_clock_info ptp_kvm_caps = { + .owner = THIS_MODULE, + .name = "KVM virtual PTP", + .max_adj = 0, + .n_ext_ts = 0, + .n_pins = 0, + .pps = 0, + .adjfreq = ptp_kvm_adjfreq, + .adjtime = ptp_kvm_adjtime, + .gettime64 = ptp_kvm_gettime, + .settime64 = ptp_kvm_settime, + .enable = ptp_kvm_enable, + .getcrosststamp = ptp_kvm_getcrosststamp, +}; + +/* module operations */ + +static struct kvm_ptp_clock kvm_ptp_clock; + +static void __exit ptp_kvm_exit(void) +{ + ptp_clock_unregister(kvm_ptp_clock.ptp_clock); +} + +static int __init ptp_kvm_init(void) +{ + long ret; + + clock_pair_gpa = slow_virt_to_phys(&clock_pair); + hv_clock = pvclock_pvti_cpu0_va(); + + if (!hv_clock) + return -ENODEV; + + ret = kvm_hypercall2(KVM_HC_CLOCK_PAIRING, clock_pair_gpa, + KVM_CLOCK_PAIRING_WALLCLOCK); + if (ret == -KVM_ENOSYS || ret == -KVM_EOPNOTSUPP) + return -ENODEV; + + kvm_ptp_clock.caps = ptp_kvm_caps; + + kvm_ptp_clock.ptp_clock = ptp_clock_register(&kvm_ptp_clock.caps, NULL); + + if (IS_ERR(kvm_ptp_clock.ptp_clock)) + return PTR_ERR(kvm_ptp_clock.ptp_clock); + + return 0; +} + +module_init(ptp_kvm_init); +module_exit(ptp_kvm_exit); + +MODULE_AUTHOR("Marcelo Tosatti <mtosatti@redhat.com>"); +MODULE_DESCRIPTION("PTP clock using KVMCLOCK"); +MODULE_LICENSE("GPL"); @@ -1082,7 +1082,7 @@ dax_iomap_actor(struct inode *inode, loff_t pos, loff_t length, void *data, */ ssize_t dax_iomap_rw(struct kiocb *iocb, struct iov_iter *iter, - struct iomap_ops *ops) + const struct iomap_ops *ops) { struct address_space *mapping = iocb->ki_filp->f_mapping; struct inode *inode = mapping->host; @@ -1130,7 +1130,7 @@ static int dax_fault_return(int error) * necessary locking for the page fault to proceed successfully. */ int dax_iomap_fault(struct vm_area_struct *vma, struct vm_fault *vmf, - struct iomap_ops *ops) + const struct iomap_ops *ops) { struct address_space *mapping = vma->vm_file->f_mapping; struct inode *inode = mapping->host; @@ -1338,7 +1338,7 @@ fallback: return VM_FAULT_FALLBACK; } -int dax_iomap_pmd_fault(struct vm_fault *vmf, struct iomap_ops *ops) +int dax_iomap_pmd_fault(struct vm_fault *vmf, const struct iomap_ops *ops) { struct vm_area_struct *vma = vmf->vma; struct address_space *mapping = vma->vm_file->f_mapping; diff --git a/fs/ext2/ext2.h b/fs/ext2/ext2.h index 37e2be784ac7..5e64de9c5093 100644 --- a/fs/ext2/ext2.h +++ b/fs/ext2/ext2.h @@ -814,7 +814,7 @@ extern const struct file_operations ext2_file_operations; /* inode.c */ extern const struct address_space_operations ext2_aops; extern const struct address_space_operations ext2_nobh_aops; -extern struct iomap_ops ext2_iomap_ops; +extern const struct iomap_ops ext2_iomap_ops; /* namei.c */ extern const struct inode_operations ext2_dir_inode_operations; diff --git a/fs/ext2/inode.c b/fs/ext2/inode.c index f073bfca694b..128cce540645 100644 --- a/fs/ext2/inode.c +++ b/fs/ext2/inode.c @@ -842,13 +842,13 @@ ext2_iomap_end(struct inode *inode, loff_t offset, loff_t length, return 0; } -struct iomap_ops ext2_iomap_ops = { +const struct iomap_ops ext2_iomap_ops = { .iomap_begin = ext2_iomap_begin, .iomap_end = ext2_iomap_end, }; #else /* Define empty ops for !CONFIG_FS_DAX case to avoid ugly ifdefs */ -struct iomap_ops ext2_iomap_ops; +const struct iomap_ops ext2_iomap_ops; #endif /* CONFIG_FS_DAX */ int ext2_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo, diff --git a/fs/ext4/ext4.h b/fs/ext4/ext4.h index 01d52b98f9a7..cee23b684f47 100644 --- a/fs/ext4/ext4.h +++ b/fs/ext4/ext4.h @@ -3244,7 +3244,7 @@ static inline void ext4_clear_io_unwritten_flag(ext4_io_end_t *io_end) } } -extern struct iomap_ops ext4_iomap_ops; +extern const struct iomap_ops ext4_iomap_ops; #endif /* __KERNEL__ */ diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c index f622d4a577e3..75212a6e69f8 100644 --- a/fs/ext4/inode.c +++ b/fs/ext4/inode.c @@ -3450,7 +3450,7 @@ orphan_del: return ret; } -struct iomap_ops ext4_iomap_ops = { +const struct iomap_ops ext4_iomap_ops = { .iomap_begin = ext4_iomap_begin, .iomap_end = ext4_iomap_end, }; diff --git a/fs/internal.h b/fs/internal.h index b63cf3af2dc2..11c6d89dce9c 100644 --- a/fs/internal.h +++ b/fs/internal.h @@ -182,7 +182,7 @@ typedef loff_t (*iomap_actor_t)(struct inode *inode, loff_t pos, loff_t len, void *data, struct iomap *iomap); loff_t iomap_apply(struct inode *inode, loff_t pos, loff_t length, - unsigned flags, struct iomap_ops *ops, void *data, + unsigned flags, const struct iomap_ops *ops, void *data, iomap_actor_t actor); /* direct-io.c: */ diff --git a/fs/iomap.c b/fs/iomap.c index a51cb4c07d4d..d89f70bbb952 100644 --- a/fs/iomap.c +++ b/fs/iomap.c @@ -41,7 +41,7 @@ */ loff_t iomap_apply(struct inode *inode, loff_t pos, loff_t length, unsigned flags, - struct iomap_ops *ops, void *data, iomap_actor_t actor) + const struct iomap_ops *ops, void *data, iomap_actor_t actor) { struct iomap iomap = { 0 }; loff_t written = 0, ret; @@ -235,7 +235,7 @@ again: ssize_t iomap_file_buffered_write(struct kiocb *iocb, struct iov_iter *iter, - struct iomap_ops *ops) + const struct iomap_ops *ops) { struct inode *inode = iocb->ki_filp->f_mapping->host; loff_t pos = iocb->ki_pos, ret = 0, written = 0; @@ -318,7 +318,7 @@ iomap_dirty_actor(struct inode *inode, loff_t pos, loff_t length, void *data, int iomap_file_dirty(struct inode *inode, loff_t pos, loff_t len, - struct iomap_ops *ops) + const struct iomap_ops *ops) { loff_t ret; @@ -398,7 +398,7 @@ iomap_zero_range_actor(struct inode *inode, loff_t pos, loff_t count, int iomap_zero_range(struct inode *inode, loff_t pos, loff_t len, bool *did_zero, - struct iomap_ops *ops) + const struct iomap_ops *ops) { loff_t ret; @@ -418,7 +418,7 @@ EXPORT_SYMBOL_GPL(iomap_zero_range); int iomap_truncate_page(struct inode *inode, loff_t pos, bool *did_zero, - struct iomap_ops *ops) + const struct iomap_ops *ops) { unsigned blocksize = (1 << inode->i_blkbits); unsigned off = pos & (blocksize - 1); @@ -446,7 +446,7 @@ iomap_page_mkwrite_actor(struct inode *inode, loff_t pos, loff_t length, } int iomap_page_mkwrite(struct vm_area_struct *vma, struct vm_fault *vmf, - struct iomap_ops *ops) + const struct iomap_ops *ops) { struct page *page = vmf->page; struct inode *inode = file_inode(vma->vm_file); @@ -545,7 +545,7 @@ iomap_fiemap_actor(struct inode *inode, loff_t pos, loff_t length, void *data, } int iomap_fiemap(struct inode *inode, struct fiemap_extent_info *fi, - loff_t start, loff_t len, struct iomap_ops *ops) + loff_t start, loff_t len, const struct iomap_ops *ops) { struct fiemap_ctx ctx; loff_t ret; @@ -839,8 +839,8 @@ iomap_dio_actor(struct inode *inode, loff_t pos, loff_t length, } ssize_t -iomap_dio_rw(struct kiocb *iocb, struct iov_iter *iter, struct iomap_ops *ops, - iomap_dio_end_io_t end_io) +iomap_dio_rw(struct kiocb *iocb, struct iov_iter *iter, + const struct iomap_ops *ops, iomap_dio_end_io_t end_io) { struct address_space *mapping = iocb->ki_filp->f_mapping; struct inode *inode = file_inode(iocb->ki_filp); diff --git a/fs/xfs/libxfs/xfs_alloc.c b/fs/xfs/libxfs/xfs_alloc.c index 9f06a211e157..369adcc18c02 100644 --- a/fs/xfs/libxfs/xfs_alloc.c +++ b/fs/xfs/libxfs/xfs_alloc.c @@ -221,20 +221,22 @@ xfs_alloc_get_rec( * Compute aligned version of the found extent. * Takes alignment and min length into account. */ -STATIC void +STATIC bool xfs_alloc_compute_aligned( xfs_alloc_arg_t *args, /* allocation argument structure */ xfs_agblock_t foundbno, /* starting block in found extent */ xfs_extlen_t foundlen, /* length in found extent */ xfs_agblock_t *resbno, /* result block number */ - xfs_extlen_t *reslen) /* result length */ + xfs_extlen_t *reslen, /* result length */ + unsigned *busy_gen) { - xfs_agblock_t bno; - xfs_extlen_t len; + xfs_agblock_t bno = foundbno; + xfs_extlen_t len = foundlen; xfs_extlen_t diff; + bool busy; /* Trim busy sections out of found extent */ - xfs_extent_busy_trim(args, foundbno, foundlen, &bno, &len); + busy = xfs_extent_busy_trim(args, &bno, &len, busy_gen); /* * If we have a largish extent that happens to start before min_agbno, @@ -259,6 +261,8 @@ xfs_alloc_compute_aligned( *resbno = bno; *reslen = len; } + + return busy; } /* @@ -737,10 +741,11 @@ xfs_alloc_ag_vextent_exact( int error; xfs_agblock_t fbno; /* start block of found extent */ xfs_extlen_t flen; /* length of found extent */ - xfs_agblock_t tbno; /* start block of trimmed extent */ - xfs_extlen_t tlen; /* length of trimmed extent */ - xfs_agblock_t tend; /* end block of trimmed extent */ + xfs_agblock_t tbno; /* start block of busy extent */ + xfs_extlen_t tlen; /* length of busy extent */ + xfs_agblock_t tend; /* end block of busy extent */ int i; /* success/failure of operation */ + unsigned busy_gen; ASSERT(args->alignment == 1); @@ -773,7 +778,9 @@ xfs_alloc_ag_vextent_exact( /* * Check for overlapping busy extents. */ - xfs_extent_busy_trim(args, fbno, flen, &tbno, &tlen); + tbno = fbno; + tlen = flen; + xfs_extent_busy_trim(args, &tbno, &tlen, &busy_gen); /* * Give up if the start of the extent is busy, or the freespace isn't @@ -853,6 +860,7 @@ xfs_alloc_find_best_extent( xfs_agblock_t sdiff; int error; int i; + unsigned busy_gen; /* The good extent is perfect, no need to search. */ if (!gdiff) @@ -866,7 +874,8 @@ xfs_alloc_find_best_extent( if (error) goto error0; XFS_WANT_CORRUPTED_GOTO(args->mp, i == 1, error0); - xfs_alloc_compute_aligned(args, *sbno, *slen, sbnoa, slena); + xfs_alloc_compute_aligned(args, *sbno, *slen, + sbnoa, slena, &busy_gen); /* * The good extent is closer than this one. @@ -955,7 +964,8 @@ xfs_alloc_ag_vextent_near( xfs_extlen_t ltlena; /* aligned ... */ xfs_agblock_t ltnew; /* useful start bno of left side */ xfs_extlen_t rlen; /* length of returned extent */ - int forced = 0; + bool busy; + unsigned busy_gen; #ifdef DEBUG /* * Randomly don't execute the first algorithm. @@ -982,6 +992,7 @@ restart: ltlen = 0; gtlena = 0; ltlena = 0; + busy = false; /* * Get a cursor for the by-size btree. @@ -1064,8 +1075,8 @@ restart: if ((error = xfs_alloc_get_rec(cnt_cur, <bno, <len, &i))) goto error0; XFS_WANT_CORRUPTED_GOTO(args->mp, i == 1, error0); - xfs_alloc_compute_aligned(args, ltbno, ltlen, - <bnoa, <lena); + busy = xfs_alloc_compute_aligned(args, ltbno, ltlen, + <bnoa, <lena, &busy_gen); if (ltlena < args->minlen) continue; if (ltbnoa < args->min_agbno || ltbnoa > args->max_agbno) @@ -1183,8 +1194,8 @@ restart: if ((error = xfs_alloc_get_rec(bno_cur_lt, <bno, <len, &i))) goto error0; XFS_WANT_CORRUPTED_GOTO(args->mp, i == 1, error0); - xfs_alloc_compute_aligned(args, ltbno, ltlen, - <bnoa, <lena); + busy |= xfs_alloc_compute_aligned(args, ltbno, ltlen, + <bnoa, <lena, &busy_gen); if (ltlena >= args->minlen && ltbnoa >= args->min_agbno) break; if ((error = xfs_btree_decrement(bno_cur_lt, 0, &i))) @@ -1199,8 +1210,8 @@ restart: if ((error = xfs_alloc_get_rec(bno_cur_gt, >bno, >len, &i))) goto error0; XFS_WANT_CORRUPTED_GOTO(args->mp, i == 1, error0); - xfs_alloc_compute_aligned(args, gtbno, gtlen, - >bnoa, >lena); + busy |= xfs_alloc_compute_aligned(args, gtbno, gtlen, + >bnoa, >lena, &busy_gen); if (gtlena >= args->minlen && gtbnoa <= args->max_agbno) break; if ((error = xfs_btree_increment(bno_cur_gt, 0, &i))) @@ -1261,9 +1272,9 @@ restart: if (bno_cur_lt == NULL && bno_cur_gt == NULL) { xfs_btree_del_cursor(cnt_cur, XFS_BTREE_NOERROR); - if (!forced++) { + if (busy) { trace_xfs_alloc_near_busy(args); - xfs_log_force(args->mp, XFS_LOG_SYNC); + xfs_extent_busy_flush(args->mp, args->pag, busy_gen); goto restart; } trace_xfs_alloc_size_neither(args); @@ -1344,7 +1355,8 @@ xfs_alloc_ag_vextent_size( int i; /* temp status variable */ xfs_agblock_t rbno; /* returned block number */ xfs_extlen_t rlen; /* length of returned extent */ - int forced = 0; + bool busy; + unsigned busy_gen; restart: /* @@ -1353,6 +1365,7 @@ restart: cnt_cur = xfs_allocbt_init_cursor(args->mp, args->tp, args->agbp, args->agno, XFS_BTNUM_CNT); bno_cur = NULL; + busy = false; /* * Look for an entry >= maxlen+alignment-1 blocks. @@ -1362,14 +1375,13 @@ restart: goto error0; /* - * If none or we have busy extents that we cannot allocate from, then - * we have to settle for a smaller extent. In the case that there are - * no large extents, this will return the last entry in the tree unless - * the tree is empty. In the case that there are only busy large - * extents, this will return the largest small extent unless there + * If none then we have to settle for a smaller extent. In the case that + * there are no large extents, this will return the last entry in the + * tree unless the tree is empty. In the case that there are only busy + * large extents, this will return the largest small extent unless there * are no smaller extents available. */ - if (!i || forced > 1) { + if (!i) { error = xfs_alloc_ag_vextent_small(args, cnt_cur, &fbno, &flen, &i); if (error) @@ -1380,13 +1392,11 @@ restart: return 0; } ASSERT(i == 1); - xfs_alloc_compute_aligned(args, fbno, flen, &rbno, &rlen); + busy = xfs_alloc_compute_aligned(args, fbno, flen, &rbno, + &rlen, &busy_gen); } else { /* * Search for a non-busy extent that is large enough. - * If we are at low space, don't check, or if we fall of - * the end of the btree, turn off the busy check and - * restart. */ for (;;) { error = xfs_alloc_get_rec(cnt_cur, &fbno, &flen, &i); @@ -1394,8 +1404,8 @@ restart: goto error0; XFS_WANT_CORRUPTED_GOTO(args->mp, i == 1, error0); - xfs_alloc_compute_aligned(args, fbno, flen, - &rbno, &rlen); + busy = xfs_alloc_compute_aligned(args, fbno, flen, + &rbno, &rlen, &busy_gen); if (rlen >= args->maxlen) break; @@ -1407,18 +1417,13 @@ restart: /* * Our only valid extents must have been busy. * Make it unbusy by forcing the log out and - * retrying. If we've been here before, forcing - * the log isn't making the extents available, - * which means they have probably been freed in - * this transaction. In that case, we have to - * give up on them and we'll attempt a minlen - * allocation the next time around. + * retrying. */ xfs_btree_del_cursor(cnt_cur, XFS_BTREE_NOERROR); trace_xfs_alloc_size_busy(args); - if (!forced++) - xfs_log_force(args->mp, XFS_LOG_SYNC); + xfs_extent_busy_flush(args->mp, + args->pag, busy_gen); goto restart; } } @@ -1454,8 +1459,8 @@ restart: XFS_WANT_CORRUPTED_GOTO(args->mp, i == 1, error0); if (flen < bestrlen) break; - xfs_alloc_compute_aligned(args, fbno, flen, - &rbno, &rlen); + busy = xfs_alloc_compute_aligned(args, fbno, flen, + &rbno, &rlen, &busy_gen); rlen = XFS_EXTLEN_MIN(args->maxlen, rlen); XFS_WANT_CORRUPTED_GOTO(args->mp, rlen == 0 || (rlen <= flen && rbno + rlen <= fbno + flen), @@ -1484,10 +1489,10 @@ restart: */ args->len = rlen; if (rlen < args->minlen) { - if (!forced++) { + if (busy) { xfs_btree_del_cursor(cnt_cur, XFS_BTREE_NOERROR); trace_xfs_alloc_size_busy(args); - xfs_log_force(args->mp, XFS_LOG_SYNC); + xfs_extent_busy_flush(args->mp, args->pag, busy_gen); goto restart; } goto out_nominleft; @@ -2659,21 +2664,11 @@ xfs_alloc_vextent( args->agbno = XFS_FSB_TO_AGBNO(mp, args->fsbno); args->type = XFS_ALLOCTYPE_NEAR_BNO; /* FALLTHROUGH */ - case XFS_ALLOCTYPE_ANY_AG: - case XFS_ALLOCTYPE_START_AG: case XFS_ALLOCTYPE_FIRST_AG: /* * Rotate through the allocation groups looking for a winner. */ - if (type == XFS_ALLOCTYPE_ANY_AG) { - /* - * Start with the last place we left off. - */ - args->agno = sagno = (mp->m_agfrotor / rotorstep) % - mp->m_sb.sb_agcount; - args->type = XFS_ALLOCTYPE_THIS_AG; - flags = XFS_ALLOC_FLAG_TRYLOCK; - } else if (type == XFS_ALLOCTYPE_FIRST_AG) { + if (type == XFS_ALLOCTYPE_FIRST_AG) { /* * Start with allocation group given by bno. */ @@ -2682,8 +2677,6 @@ xfs_alloc_vextent( sagno = 0; flags = 0; } else { - if (type == XFS_ALLOCTYPE_START_AG) - args->type = XFS_ALLOCTYPE_THIS_AG; /* * Start with the given allocation group. */ @@ -2751,7 +2744,7 @@ xfs_alloc_vextent( } xfs_perag_put(args->pag); } - if (bump_rotor || (type == XFS_ALLOCTYPE_ANY_AG)) { + if (bump_rotor) { if (args->agno == sagno) mp->m_agfrotor = (mp->m_agfrotor + 1) % (mp->m_sb.sb_agcount * rotorstep); diff --git a/fs/xfs/libxfs/xfs_alloc.h b/fs/xfs/libxfs/xfs_alloc.h index 1d0f48a501a3..2a8d0fa6fbbe 100644 --- a/fs/xfs/libxfs/xfs_alloc.h +++ b/fs/xfs/libxfs/xfs_alloc.h @@ -29,9 +29,7 @@ extern struct workqueue_struct *xfs_alloc_wq; /* * Freespace allocation types. Argument to xfs_alloc_[v]extent. */ -#define XFS_ALLOCTYPE_ANY_AG 0x01 /* allocate anywhere, use rotor */ #define XFS_ALLOCTYPE_FIRST_AG 0x02 /* ... start at ag 0 */ -#define XFS_ALLOCTYPE_START_AG 0x04 /* anywhere, start in this a.g. */ #define XFS_ALLOCTYPE_THIS_AG 0x08 /* anywhere in this a.g. */ #define XFS_ALLOCTYPE_START_BNO 0x10 /* near this block else anywhere */ #define XFS_ALLOCTYPE_NEAR_BNO 0x20 /* in this a.g. and near this block */ @@ -41,9 +39,7 @@ extern struct workqueue_struct *xfs_alloc_wq; typedef unsigned int xfs_alloctype_t; #define XFS_ALLOC_TYPES \ - { XFS_ALLOCTYPE_ANY_AG, "ANY_AG" }, \ { XFS_ALLOCTYPE_FIRST_AG, "FIRST_AG" }, \ - { XFS_ALLOCTYPE_START_AG, "START_AG" }, \ { XFS_ALLOCTYPE_THIS_AG, "THIS_AG" }, \ { XFS_ALLOCTYPE_START_BNO, "START_BNO" }, \ { XFS_ALLOCTYPE_NEAR_BNO, "NEAR_BNO" }, \ diff --git a/fs/xfs/libxfs/xfs_bmap.c b/fs/xfs/libxfs/xfs_bmap.c index bfc00de5c6f1..a9c66d47757a 100644 --- a/fs/xfs/libxfs/xfs_bmap.c +++ b/fs/xfs/libxfs/xfs_bmap.c @@ -740,15 +740,9 @@ xfs_bmap_extents_to_btree( * Fill in the root. */ block = ifp->if_broot; - if (xfs_sb_version_hascrc(&mp->m_sb)) - xfs_btree_init_block_int(mp, block, XFS_BUF_DADDR_NULL, - XFS_BMAP_CRC_MAGIC, 1, 1, ip->i_ino, - XFS_BTREE_LONG_PTRS | XFS_BTREE_CRC_BLOCKS); - else - xfs_btree_init_block_int(mp, block, XFS_BUF_DADDR_NULL, - XFS_BMAP_MAGIC, 1, 1, ip->i_ino, + xfs_btree_init_block_int(mp, block, XFS_BUF_DADDR_NULL, + XFS_BTNUM_BMAP, 1, 1, ip->i_ino, XFS_BTREE_LONG_PTRS); - /* * Need a cursor. Can't allocate until bb_level is filled in. */ @@ -804,9 +798,7 @@ try_another_ag: */ ASSERT(args.fsbno != NULLFSBLOCK); ASSERT(*firstblock == NULLFSBLOCK || - args.agno == XFS_FSB_TO_AGNO(mp, *firstblock) || - (dfops->dop_low && - args.agno > XFS_FSB_TO_AGNO(mp, *firstblock))); + args.agno >= XFS_FSB_TO_AGNO(mp, *firstblock)); *firstblock = cur->bc_private.b.firstblock = args.fsbno; cur->bc_private.b.allocated++; ip->i_d.di_nblocks++; @@ -817,13 +809,8 @@ try_another_ag: */ abp->b_ops = &xfs_bmbt_buf_ops; ablock = XFS_BUF_TO_BLOCK(abp); - if (xfs_sb_version_hascrc(&mp->m_sb)) - xfs_btree_init_block_int(mp, ablock, abp->b_bn, - XFS_BMAP_CRC_MAGIC, 0, 0, ip->i_ino, - XFS_BTREE_LONG_PTRS | XFS_BTREE_CRC_BLOCKS); - else - xfs_btree_init_block_int(mp, ablock, abp->b_bn, - XFS_BMAP_MAGIC, 0, 0, ip->i_ino, + xfs_btree_init_block_int(mp, ablock, abp->b_bn, + XFS_BTNUM_BMAP, 0, 0, ip->i_ino, XFS_BTREE_LONG_PTRS); arp = XFS_BMBT_REC_ADDR(mp, ablock, 1); @@ -1278,7 +1265,6 @@ xfs_bmap_read_extents( /* REFERENCED */ xfs_extnum_t room; /* number of entries there's room for */ - bno = NULLFSBLOCK; mp = ip->i_mount; ifp = XFS_IFORK_PTR(ip, whichfork); exntf = (whichfork != XFS_DATA_FORK) ? XFS_EXTFMT_NOSTATE : @@ -1291,9 +1277,7 @@ xfs_bmap_read_extents( ASSERT(level > 0); pp = XFS_BMAP_BROOT_PTR_ADDR(mp, block, 1, ifp->if_broot_bytes); bno = be64_to_cpu(*pp); - ASSERT(bno != NULLFSBLOCK); - ASSERT(XFS_FSB_TO_AGNO(mp, bno) < mp->m_sb.sb_agcount); - ASSERT(XFS_FSB_TO_AGBNO(mp, bno) < mp->m_sb.sb_agblocks); + /* * Go down the tree until leaf level is reached, following the first * pointer (leftmost) at each level. @@ -1864,6 +1848,7 @@ xfs_bmap_add_extent_delay_real( */ trace_xfs_bmap_pre_update(bma->ip, bma->idx, state, _THIS_IP_); xfs_bmbt_set_startblock(ep, new->br_startblock); + xfs_bmbt_set_state(ep, new->br_state); trace_xfs_bmap_post_update(bma->ip, bma->idx, state, _THIS_IP_); (*nextents)++; @@ -2202,6 +2187,7 @@ STATIC int /* error */ xfs_bmap_add_extent_unwritten_real( struct xfs_trans *tp, xfs_inode_t *ip, /* incore inode pointer */ + int whichfork, xfs_extnum_t *idx, /* extent number to update/insert */ xfs_btree_cur_t **curp, /* if *curp is null, not a btree */ xfs_bmbt_irec_t *new, /* new data to add to file extents */ @@ -2221,12 +2207,14 @@ xfs_bmap_add_extent_unwritten_real( /* left is 0, right is 1, prev is 2 */ int rval=0; /* return value (logging flags) */ int state = 0;/* state bits, accessed thru macros */ - struct xfs_mount *mp = tp->t_mountp; + struct xfs_mount *mp = ip->i_mount; *logflagsp = 0; cur = *curp; - ifp = XFS_IFORK_PTR(ip, XFS_DATA_FORK); + ifp = XFS_IFORK_PTR(ip, whichfork); + if (whichfork == XFS_COW_FORK) + state |= BMAP_COWFORK; ASSERT(*idx >= 0); ASSERT(*idx <= xfs_iext_count(ifp)); @@ -2285,7 +2273,7 @@ xfs_bmap_add_extent_unwritten_real( * Don't set contiguous if the combined extent would be too large. * Also check for all-three-contiguous being too large. */ - if (*idx < xfs_iext_count(&ip->i_df) - 1) { + if (*idx < xfs_iext_count(ifp) - 1) { state |= BMAP_RIGHT_VALID; xfs_bmbt_get_all(xfs_iext_get_ext(ifp, *idx + 1), &RIGHT); if (isnullstartblock(RIGHT.br_startblock)) @@ -2325,7 +2313,8 @@ xfs_bmap_add_extent_unwritten_real( trace_xfs_bmap_post_update(ip, *idx, state, _THIS_IP_); xfs_iext_remove(ip, *idx + 1, 2, state); - ip->i_d.di_nextents -= 2; + XFS_IFORK_NEXT_SET(ip, whichfork, + XFS_IFORK_NEXTENTS(ip, whichfork) - 2); if (cur == NULL) rval = XFS_ILOG_CORE | XFS_ILOG_DEXT; else { @@ -2368,7 +2357,8 @@ xfs_bmap_add_extent_unwritten_real( trace_xfs_bmap_post_update(ip, *idx, state, _THIS_IP_); xfs_iext_remove(ip, *idx + 1, 1, state); - ip->i_d.di_nextents--; + XFS_IFORK_NEXT_SET(ip, whichfork, + XFS_IFORK_NEXTENTS(ip, whichfork) - 1); if (cur == NULL) rval = XFS_ILOG_CORE | XFS_ILOG_DEXT; else { @@ -2403,7 +2393,8 @@ xfs_bmap_add_extent_unwritten_real( xfs_bmbt_set_state(ep, newext); trace_xfs_bmap_post_update(ip, *idx, state, _THIS_IP_); xfs_iext_remove(ip, *idx + 1, 1, state); - ip->i_d.di_nextents--; + XFS_IFORK_NEXT_SET(ip, whichfork, + XFS_IFORK_NEXTENTS(ip, whichfork) - 1); if (cur == NULL) rval = XFS_ILOG_CORE | XFS_ILOG_DEXT; else { @@ -2515,7 +2506,8 @@ xfs_bmap_add_extent_unwritten_real( trace_xfs_bmap_post_update(ip, *idx, state, _THIS_IP_); xfs_iext_insert(ip, *idx, 1, new, state); - ip->i_d.di_nextents++; + XFS_IFORK_NEXT_SET(ip, whichfork, + XFS_IFORK_NEXTENTS(ip, whichfork) + 1); if (cur == NULL) rval = XFS_ILOG_CORE | XFS_ILOG_DEXT; else { @@ -2593,7 +2585,8 @@ xfs_bmap_add_extent_unwritten_real( ++*idx; xfs_iext_insert(ip, *idx, 1, new, state); - ip->i_d.di_nextents++; + XFS_IFORK_NEXT_SET(ip, whichfork, + XFS_IFORK_NEXTENTS(ip, whichfork) + 1); if (cur == NULL) rval = XFS_ILOG_CORE | XFS_ILOG_DEXT; else { @@ -2641,7 +2634,8 @@ xfs_bmap_add_extent_unwritten_real( ++*idx; xfs_iext_insert(ip, *idx, 2, &r[0], state); - ip->i_d.di_nextents += 2; + XFS_IFORK_NEXT_SET(ip, whichfork, + XFS_IFORK_NEXTENTS(ip, whichfork) + 2); if (cur == NULL) rval = XFS_ILOG_CORE | XFS_ILOG_DEXT; else { @@ -2695,17 +2689,17 @@ xfs_bmap_add_extent_unwritten_real( } /* update reverse mappings */ - error = xfs_rmap_convert_extent(mp, dfops, ip, XFS_DATA_FORK, new); + error = xfs_rmap_convert_extent(mp, dfops, ip, whichfork, new); if (error) goto done; /* convert to a btree if necessary */ - if (xfs_bmap_needs_btree(ip, XFS_DATA_FORK)) { + if (xfs_bmap_needs_btree(ip, whichfork)) { int tmp_logflags; /* partial log flag return val */ ASSERT(cur == NULL); error = xfs_bmap_extents_to_btree(tp, ip, first, dfops, &cur, - 0, &tmp_logflags, XFS_DATA_FORK); + 0, &tmp_logflags, whichfork); *logflagsp |= tmp_logflags; if (error) goto done; @@ -2717,7 +2711,7 @@ xfs_bmap_add_extent_unwritten_real( *curp = cur; } - xfs_bmap_check_leaf_extents(*curp, ip, XFS_DATA_FORK); + xfs_bmap_check_leaf_extents(*curp, ip, whichfork); done: *logflagsp |= rval; return error; @@ -2809,7 +2803,8 @@ xfs_bmap_add_extent_hole_delay( oldlen = startblockval(left.br_startblock) + startblockval(new->br_startblock) + startblockval(right.br_startblock); - newlen = xfs_bmap_worst_indlen(ip, temp); + newlen = XFS_FILBLKS_MIN(xfs_bmap_worst_indlen(ip, temp), + oldlen); xfs_bmbt_set_startblock(xfs_iext_get_ext(ifp, *idx), nullstartblock((int)newlen)); trace_xfs_bmap_post_update(ip, *idx, state, _THIS_IP_); @@ -2830,7 +2825,8 @@ xfs_bmap_add_extent_hole_delay( xfs_bmbt_set_blockcount(xfs_iext_get_ext(ifp, *idx), temp); oldlen = startblockval(left.br_startblock) + startblockval(new->br_startblock); - newlen = xfs_bmap_worst_indlen(ip, temp); + newlen = XFS_FILBLKS_MIN(xfs_bmap_worst_indlen(ip, temp), + oldlen); xfs_bmbt_set_startblock(xfs_iext_get_ext(ifp, *idx), nullstartblock((int)newlen)); trace_xfs_bmap_post_update(ip, *idx, state, _THIS_IP_); @@ -2846,7 +2842,8 @@ xfs_bmap_add_extent_hole_delay( temp = new->br_blockcount + right.br_blockcount; oldlen = startblockval(new->br_startblock) + startblockval(right.br_startblock); - newlen = xfs_bmap_worst_indlen(ip, temp); + newlen = XFS_FILBLKS_MIN(xfs_bmap_worst_indlen(ip, temp), + oldlen); xfs_bmbt_set_allf(xfs_iext_get_ext(ifp, *idx), new->br_startoff, nullstartblock((int)newlen), temp, right.br_state); @@ -2899,13 +2896,14 @@ xfs_bmap_add_extent_hole_real( ASSERT(!isnullstartblock(new->br_startblock)); ASSERT(!bma->cur || !(bma->cur->bc_private.b.flags & XFS_BTCUR_BPRV_WASDEL)); - ASSERT(whichfork != XFS_COW_FORK); XFS_STATS_INC(mp, xs_add_exlist); state = 0; if (whichfork == XFS_ATTR_FORK) state |= BMAP_ATTRFORK; + if (whichfork == XFS_COW_FORK) + state |= BMAP_COWFORK; /* * Check and set flags if this segment has a left neighbor. @@ -3822,17 +3820,13 @@ xfs_bmap_btalloc( * the first block that was allocated. */ ASSERT(*ap->firstblock == NULLFSBLOCK || - XFS_FSB_TO_AGNO(mp, *ap->firstblock) == - XFS_FSB_TO_AGNO(mp, args.fsbno) || - (ap->dfops->dop_low && - XFS_FSB_TO_AGNO(mp, *ap->firstblock) < - XFS_FSB_TO_AGNO(mp, args.fsbno))); + XFS_FSB_TO_AGNO(mp, *ap->firstblock) <= + XFS_FSB_TO_AGNO(mp, args.fsbno)); ap->blkno = args.fsbno; if (*ap->firstblock == NULLFSBLOCK) *ap->firstblock = args.fsbno; - ASSERT(nullfb || fb_agno == args.agno || - (ap->dfops->dop_low && fb_agno < args.agno)); + ASSERT(nullfb || fb_agno <= args.agno); ap->length = args.len; if (!(ap->flags & XFS_BMAPI_COWFORK)) ap->ip->i_d.di_nblocks += args.len; @@ -4368,10 +4362,16 @@ xfs_bmapi_allocate( bma->got.br_state = XFS_EXT_NORM; /* - * A wasdelay extent has been initialized, so shouldn't be flagged - * as unwritten. + * In the data fork, a wasdelay extent has been initialized, so + * shouldn't be flagged as unwritten. + * + * For the cow fork, however, we convert delalloc reservations + * (extents allocated for speculative preallocation) to + * allocated unwritten extents, and only convert the unwritten + * extents to real extents when we're about to write the data. */ - if (!bma->wasdel && (bma->flags & XFS_BMAPI_PREALLOC) && + if ((!bma->wasdel || (bma->flags & XFS_BMAPI_COWFORK)) && + (bma->flags & XFS_BMAPI_PREALLOC) && xfs_sb_version_hasextflgbit(&mp->m_sb)) bma->got.br_state = XFS_EXT_UNWRITTEN; @@ -4422,8 +4422,6 @@ xfs_bmapi_convert_unwritten( (XFS_BMAPI_PREALLOC | XFS_BMAPI_CONVERT)) return 0; - ASSERT(whichfork != XFS_COW_FORK); - /* * Modify (by adding) the state flag, if writing. */ @@ -4448,8 +4446,8 @@ xfs_bmapi_convert_unwritten( return error; } - error = xfs_bmap_add_extent_unwritten_real(bma->tp, bma->ip, &bma->idx, - &bma->cur, mval, bma->firstblock, bma->dfops, + error = xfs_bmap_add_extent_unwritten_real(bma->tp, bma->ip, whichfork, + &bma->idx, &bma->cur, mval, bma->firstblock, bma->dfops, &tmp_logflags); /* * Log the inode core unconditionally in the unwritten extent conversion @@ -4458,8 +4456,12 @@ xfs_bmapi_convert_unwritten( * in the transaction for the sake of fsync(), even if nothing has * changed, because fsync() will not force the log for this transaction * unless it sees the inode pinned. + * + * Note: If we're only converting cow fork extents, there aren't + * any on-disk updates to make, so we don't need to log anything. */ - bma->logflags |= tmp_logflags | XFS_ILOG_CORE; + if (whichfork != XFS_COW_FORK) + bma->logflags |= tmp_logflags | XFS_ILOG_CORE; if (error) return error; @@ -4533,15 +4535,15 @@ xfs_bmapi_write( ASSERT(*nmap >= 1); ASSERT(*nmap <= XFS_BMAP_MAX_NMAP); ASSERT(!(flags & XFS_BMAPI_IGSTATE)); - ASSERT(tp != NULL); + ASSERT(tp != NULL || + (flags & (XFS_BMAPI_CONVERT | XFS_BMAPI_COWFORK)) == + (XFS_BMAPI_CONVERT | XFS_BMAPI_COWFORK)); ASSERT(len > 0); ASSERT(XFS_IFORK_FORMAT(ip, whichfork) != XFS_DINODE_FMT_LOCAL); ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL)); ASSERT(!(flags & XFS_BMAPI_REMAP) || whichfork == XFS_DATA_FORK); ASSERT(!(flags & XFS_BMAPI_PREALLOC) || !(flags & XFS_BMAPI_REMAP)); ASSERT(!(flags & XFS_BMAPI_CONVERT) || !(flags & XFS_BMAPI_REMAP)); - ASSERT(!(flags & XFS_BMAPI_PREALLOC) || whichfork != XFS_COW_FORK); - ASSERT(!(flags & XFS_BMAPI_CONVERT) || whichfork != XFS_COW_FORK); /* zeroing is for currently only for data extents, not metadata */ ASSERT((flags & (XFS_BMAPI_METADATA | XFS_BMAPI_ZERO)) != @@ -4746,13 +4748,9 @@ error0: if (bma.cur) { if (!error) { ASSERT(*firstblock == NULLFSBLOCK || - XFS_FSB_TO_AGNO(mp, *firstblock) == + XFS_FSB_TO_AGNO(mp, *firstblock) <= XFS_FSB_TO_AGNO(mp, - bma.cur->bc_private.b.firstblock) || - (dfops->dop_low && - XFS_FSB_TO_AGNO(mp, *firstblock) < - XFS_FSB_TO_AGNO(mp, - bma.cur->bc_private.b.firstblock))); + bma.cur->bc_private.b.firstblock)); *firstblock = bma.cur->bc_private.b.firstblock; } xfs_btree_del_cursor(bma.cur, @@ -4787,34 +4785,59 @@ xfs_bmap_split_indlen( xfs_filblks_t len2 = *indlen2; xfs_filblks_t nres = len1 + len2; /* new total res. */ xfs_filblks_t stolen = 0; + xfs_filblks_t resfactor; /* * Steal as many blocks as we can to try and satisfy the worst case * indlen for both new extents. */ - while (nres > ores && avail) { - nres--; - avail--; - stolen++; - } + if (ores < nres && avail) + stolen = XFS_FILBLKS_MIN(nres - ores, avail); + ores += stolen; + + /* nothing else to do if we've satisfied the new reservation */ + if (ores >= nres) + return stolen; + + /* + * We can't meet the total required reservation for the two extents. + * Calculate the percent of the overall shortage between both extents + * and apply this percentage to each of the requested indlen values. + * This distributes the shortage fairly and reduces the chances that one + * of the two extents is left with nothing when extents are repeatedly + * split. + */ + resfactor = (ores * 100); + do_div(resfactor, nres); + len1 *= resfactor; + do_div(len1, 100); + len2 *= resfactor; + do_div(len2, 100); + ASSERT(len1 + len2 <= ores); + ASSERT(len1 < *indlen1 && len2 < *indlen2); /* - * The only blocks available are those reserved for the original - * extent and what we can steal from the extent being removed. - * If this still isn't enough to satisfy the combined - * requirements for the two new extents, skim blocks off of each - * of the new reservations until they match what is available. + * Hand out the remainder to each extent. If one of the two reservations + * is zero, we want to make sure that one gets a block first. The loop + * below starts with len1, so hand len2 a block right off the bat if it + * is zero. */ - while (nres > ores) { - if (len1) { - len1--; - nres--; + ores -= (len1 + len2); + ASSERT((*indlen1 - len1) + (*indlen2 - len2) >= ores); + if (ores && !len2 && *indlen2) { + len2++; + ores--; + } + while (ores) { + if (len1 < *indlen1) { + len1++; + ores--; } - if (nres == ores) + if (!ores) break; - if (len2) { - len2--; - nres--; + if (len2 < *indlen2) { + len2++; + ores--; } } @@ -5556,8 +5579,8 @@ __xfs_bunmapi( } del.br_state = XFS_EXT_UNWRITTEN; error = xfs_bmap_add_extent_unwritten_real(tp, ip, - &lastx, &cur, &del, firstblock, dfops, - &logflags); + whichfork, &lastx, &cur, &del, + firstblock, dfops, &logflags); if (error) goto error0; goto nodelete; @@ -5610,8 +5633,9 @@ __xfs_bunmapi( prev.br_state = XFS_EXT_UNWRITTEN; lastx--; error = xfs_bmap_add_extent_unwritten_real(tp, - ip, &lastx, &cur, &prev, - firstblock, dfops, &logflags); + ip, whichfork, &lastx, &cur, + &prev, firstblock, dfops, + &logflags); if (error) goto error0; goto nodelete; @@ -5619,8 +5643,9 @@ __xfs_bunmapi( ASSERT(del.br_state == XFS_EXT_NORM); del.br_state = XFS_EXT_UNWRITTEN; error = xfs_bmap_add_extent_unwritten_real(tp, - ip, &lastx, &cur, &del, - firstblock, dfops, &logflags); + ip, whichfork, &lastx, &cur, + &del, firstblock, dfops, + &logflags); if (error) goto error0; goto nodelete; diff --git a/fs/xfs/libxfs/xfs_bmap_btree.c b/fs/xfs/libxfs/xfs_bmap_btree.c index d9be241fc86f..f93072b58a58 100644 --- a/fs/xfs/libxfs/xfs_bmap_btree.c +++ b/fs/xfs/libxfs/xfs_bmap_btree.c @@ -71,15 +71,9 @@ xfs_bmdr_to_bmbt( xfs_bmbt_key_t *tkp; __be64 *tpp; - if (xfs_sb_version_hascrc(&mp->m_sb)) - xfs_btree_init_block_int(mp, rblock, XFS_BUF_DADDR_NULL, - XFS_BMAP_CRC_MAGIC, 0, 0, ip->i_ino, - XFS_BTREE_LONG_PTRS | XFS_BTREE_CRC_BLOCKS); - else - xfs_btree_init_block_int(mp, rblock, XFS_BUF_DADDR_NULL, - XFS_BMAP_MAGIC, 0, 0, ip->i_ino, + xfs_btree_init_block_int(mp, rblock, XFS_BUF_DADDR_NULL, + XFS_BTNUM_BMAP, 0, 0, ip->i_ino, XFS_BTREE_LONG_PTRS); - rblock->bb_level = dblock->bb_level; ASSERT(be16_to_cpu(rblock->bb_level) > 0); rblock->bb_numrecs = dblock->bb_numrecs; diff --git a/fs/xfs/libxfs/xfs_btree.c b/fs/xfs/libxfs/xfs_btree.c index 21e6a6ab6b9a..c3decedc9455 100644 --- a/fs/xfs/libxfs/xfs_btree.c +++ b/fs/xfs/libxfs/xfs_btree.c @@ -50,8 +50,18 @@ static const __uint32_t xfs_magics[2][XFS_BTNUM_MAX] = { XFS_BMAP_CRC_MAGIC, XFS_IBT_CRC_MAGIC, XFS_FIBT_CRC_MAGIC, XFS_REFC_CRC_MAGIC } }; -#define xfs_btree_magic(cur) \ - xfs_magics[!!((cur)->bc_flags & XFS_BTREE_CRC_BLOCKS)][cur->bc_btnum] + +__uint32_t +xfs_btree_magic( + int crc, + xfs_btnum_t btnum) +{ + __uint32_t magic = xfs_magics[crc][btnum]; + + /* Ensure we asked for crc for crc-only magics. */ + ASSERT(magic != 0); + return magic; +} STATIC int /* error (0 or EFSCORRUPTED) */ xfs_btree_check_lblock( @@ -62,10 +72,13 @@ xfs_btree_check_lblock( { int lblock_ok = 1; /* block passes checks */ struct xfs_mount *mp; /* file system mount point */ + xfs_btnum_t btnum = cur->bc_btnum; + int crc; mp = cur->bc_mp; + crc = xfs_sb_version_hascrc(&mp->m_sb); - if (xfs_sb_version_hascrc(&mp->m_sb)) { + if (crc) { lblock_ok = lblock_ok && uuid_equal(&block->bb_u.l.bb_uuid, &mp->m_sb.sb_meta_uuid) && @@ -74,7 +87,7 @@ xfs_btree_check_lblock( } lblock_ok = lblock_ok && - be32_to_cpu(block->bb_magic) == xfs_btree_magic(cur) && + be32_to_cpu(block->bb_magic) == xfs_btree_magic(crc, btnum) && be16_to_cpu(block->bb_level) == level && be16_to_cpu(block->bb_numrecs) <= cur->bc_ops->get_maxrecs(cur, level) && @@ -110,13 +123,16 @@ xfs_btree_check_sblock( struct xfs_agf *agf; /* ag. freespace structure */ xfs_agblock_t agflen; /* native ag. freespace length */ int sblock_ok = 1; /* block passes checks */ + xfs_btnum_t btnum = cur->bc_btnum; + int crc; mp = cur->bc_mp; + crc = xfs_sb_version_hascrc(&mp->m_sb); agbp = cur->bc_private.a.agbp; agf = XFS_BUF_TO_AGF(agbp); agflen = be32_to_cpu(agf->agf_length); - if (xfs_sb_version_hascrc(&mp->m_sb)) { + if (crc) { sblock_ok = sblock_ok && uuid_equal(&block->bb_u.s.bb_uuid, &mp->m_sb.sb_meta_uuid) && @@ -125,7 +141,7 @@ xfs_btree_check_sblock( } sblock_ok = sblock_ok && - be32_to_cpu(block->bb_magic) == xfs_btree_magic(cur) && + be32_to_cpu(block->bb_magic) == xfs_btree_magic(crc, btnum) && be16_to_cpu(block->bb_level) == level && be16_to_cpu(block->bb_numrecs) <= cur->bc_ops->get_maxrecs(cur, level) && @@ -810,7 +826,8 @@ xfs_btree_read_bufl( xfs_daddr_t d; /* real disk block address */ int error; - ASSERT(fsbno != NULLFSBLOCK); + if (!XFS_FSB_SANITY_CHECK(mp, fsbno)) + return -EFSCORRUPTED; d = XFS_FSB_TO_DADDR(mp, fsbno); error = xfs_trans_read_buf(mp, tp, mp->m_ddev_targp, d, mp->m_bsize, lock, &bp, ops); @@ -1084,12 +1101,15 @@ xfs_btree_init_block_int( struct xfs_mount *mp, struct xfs_btree_block *buf, xfs_daddr_t blkno, - __u32 magic, + xfs_btnum_t btnum, __u16 level, __u16 numrecs, __u64 owner, unsigned int flags) { + int crc = xfs_sb_version_hascrc(&mp->m_sb); + __u32 magic = xfs_btree_magic(crc, btnum); + buf->bb_magic = cpu_to_be32(magic); buf->bb_level = cpu_to_be16(level); buf->bb_numrecs = cpu_to_be16(numrecs); @@ -1097,7 +1117,7 @@ xfs_btree_init_block_int( if (flags & XFS_BTREE_LONG_PTRS) { buf->bb_u.l.bb_leftsib = cpu_to_be64(NULLFSBLOCK); buf->bb_u.l.bb_rightsib = cpu_to_be64(NULLFSBLOCK); - if (flags & XFS_BTREE_CRC_BLOCKS) { + if (crc) { buf->bb_u.l.bb_blkno = cpu_to_be64(blkno); buf->bb_u.l.bb_owner = cpu_to_be64(owner); uuid_copy(&buf->bb_u.l.bb_uuid, &mp->m_sb.sb_meta_uuid); @@ -1110,7 +1130,7 @@ xfs_btree_init_block_int( buf->bb_u.s.bb_leftsib = cpu_to_be32(NULLAGBLOCK); buf->bb_u.s.bb_rightsib = cpu_to_be32(NULLAGBLOCK); - if (flags & XFS_BTREE_CRC_BLOCKS) { + if (crc) { buf->bb_u.s.bb_blkno = cpu_to_be64(blkno); buf->bb_u.s.bb_owner = cpu_to_be32(__owner); uuid_copy(&buf->bb_u.s.bb_uuid, &mp->m_sb.sb_meta_uuid); @@ -1123,14 +1143,14 @@ void xfs_btree_init_block( struct xfs_mount *mp, struct xfs_buf *bp, - __u32 magic, + xfs_btnum_t btnum, __u16 level, __u16 numrecs, __u64 owner, unsigned int flags) { xfs_btree_init_block_int(mp, XFS_BUF_TO_BLOCK(bp), bp->b_bn, - magic, level, numrecs, owner, flags); + btnum, level, numrecs, owner, flags); } STATIC void @@ -1140,7 +1160,7 @@ xfs_btree_init_block_cur( int level, int numrecs) { - __u64 owner; + __u64 owner; /* * we can pull the owner from the cursor right now as the different @@ -1154,7 +1174,7 @@ xfs_btree_init_block_cur( owner = cur->bc_private.a.agno; xfs_btree_init_block_int(cur->bc_mp, XFS_BUF_TO_BLOCK(bp), bp->b_bn, - xfs_btree_magic(cur), level, numrecs, + cur->bc_btnum, level, numrecs, owner, cur->bc_flags); } diff --git a/fs/xfs/libxfs/xfs_btree.h b/fs/xfs/libxfs/xfs_btree.h index b69b947c4c1b..4bb62580a7fd 100644 --- a/fs/xfs/libxfs/xfs_btree.h +++ b/fs/xfs/libxfs/xfs_btree.h @@ -76,6 +76,8 @@ union xfs_btree_rec { #define XFS_BTNUM_RMAP ((xfs_btnum_t)XFS_BTNUM_RMAPi) #define XFS_BTNUM_REFC ((xfs_btnum_t)XFS_BTNUM_REFCi) +__uint32_t xfs_btree_magic(int crc, xfs_btnum_t btnum); + /* * For logging record fields. */ @@ -378,7 +380,7 @@ void xfs_btree_init_block( struct xfs_mount *mp, struct xfs_buf *bp, - __u32 magic, + xfs_btnum_t btnum, __u16 level, __u16 numrecs, __u64 owner, @@ -389,7 +391,7 @@ xfs_btree_init_block_int( struct xfs_mount *mp, struct xfs_btree_block *buf, xfs_daddr_t blkno, - __u32 magic, + xfs_btnum_t btnum, __u16 level, __u16 numrecs, __u64 owner, @@ -456,7 +458,7 @@ static inline int xfs_btree_get_level(struct xfs_btree_block *block) #define XFS_FILBLKS_MAX(a,b) max_t(xfs_filblks_t, (a), (b)) #define XFS_FSB_SANITY_CHECK(mp,fsb) \ - (XFS_FSB_TO_AGNO(mp, fsb) < mp->m_sb.sb_agcount && \ + (fsb && XFS_FSB_TO_AGNO(mp, fsb) < mp->m_sb.sb_agcount && \ XFS_FSB_TO_AGBNO(mp, fsb) < mp->m_sb.sb_agblocks) /* diff --git a/fs/xfs/libxfs/xfs_da_btree.c b/fs/xfs/libxfs/xfs_da_btree.c index f2dc1a950c85..1bdf2888295b 100644 --- a/fs/xfs/libxfs/xfs_da_btree.c +++ b/fs/xfs/libxfs/xfs_da_btree.c @@ -2633,7 +2633,7 @@ out_free: /* * Readahead the dir/attr block. */ -xfs_daddr_t +int xfs_da_reada_buf( struct xfs_inode *dp, xfs_dablk_t bno, @@ -2664,7 +2664,5 @@ out_free: if (mapp != &map) kmem_free(mapp); - if (error) - return -1; - return mappedbno; + return error; } diff --git a/fs/xfs/libxfs/xfs_da_btree.h b/fs/xfs/libxfs/xfs_da_btree.h index 98c75cbe6ac2..4e29cb6a3627 100644 --- a/fs/xfs/libxfs/xfs_da_btree.h +++ b/fs/xfs/libxfs/xfs_da_btree.h @@ -201,7 +201,7 @@ int xfs_da_read_buf(struct xfs_trans *trans, struct xfs_inode *dp, xfs_dablk_t bno, xfs_daddr_t mappedbno, struct xfs_buf **bpp, int whichfork, const struct xfs_buf_ops *ops); -xfs_daddr_t xfs_da_reada_buf(struct xfs_inode *dp, xfs_dablk_t bno, +int xfs_da_reada_buf(struct xfs_inode *dp, xfs_dablk_t bno, xfs_daddr_t mapped_bno, int whichfork, const struct xfs_buf_ops *ops); int xfs_da_shrink_inode(xfs_da_args_t *args, xfs_dablk_t dead_blkno, diff --git a/fs/xfs/libxfs/xfs_dir2_node.c b/fs/xfs/libxfs/xfs_dir2_node.c index 75a557432d0f..bbd1238852b3 100644 --- a/fs/xfs/libxfs/xfs_dir2_node.c +++ b/fs/xfs/libxfs/xfs_dir2_node.c @@ -155,6 +155,42 @@ const struct xfs_buf_ops xfs_dir3_free_buf_ops = { .verify_write = xfs_dir3_free_write_verify, }; +/* Everything ok in the free block header? */ +static bool +xfs_dir3_free_header_check( + struct xfs_inode *dp, + xfs_dablk_t fbno, + struct xfs_buf *bp) +{ + struct xfs_mount *mp = dp->i_mount; + unsigned int firstdb; + int maxbests; + + maxbests = dp->d_ops->free_max_bests(mp->m_dir_geo); + firstdb = (xfs_dir2_da_to_db(mp->m_dir_geo, fbno) - + xfs_dir2_byte_to_db(mp->m_dir_geo, XFS_DIR2_FREE_OFFSET)) * + maxbests; + if (xfs_sb_version_hascrc(&mp->m_sb)) { + struct xfs_dir3_free_hdr *hdr3 = bp->b_addr; + + if (be32_to_cpu(hdr3->firstdb) != firstdb) + return false; + if (be32_to_cpu(hdr3->nvalid) > maxbests) + return false; + if (be32_to_cpu(hdr3->nvalid) < be32_to_cpu(hdr3->nused)) + return false; + } else { + struct xfs_dir2_free_hdr *hdr = bp->b_addr; + + if (be32_to_cpu(hdr->firstdb) != firstdb) + return false; + if (be32_to_cpu(hdr->nvalid) > maxbests) + return false; + if (be32_to_cpu(hdr->nvalid) < be32_to_cpu(hdr->nused)) + return false; + } + return true; +} static int __xfs_dir3_free_read( @@ -168,11 +204,22 @@ __xfs_dir3_free_read( err = xfs_da_read_buf(tp, dp, fbno, mappedbno, bpp, XFS_DATA_FORK, &xfs_dir3_free_buf_ops); + if (err || !*bpp) + return err; + + /* Check things that we can't do in the verifier. */ + if (!xfs_dir3_free_header_check(dp, fbno, *bpp)) { + xfs_buf_ioerror(*bpp, -EFSCORRUPTED); + xfs_verifier_error(*bpp); + xfs_trans_brelse(tp, *bpp); + return -EFSCORRUPTED; + } /* try read returns without an error or *bpp if it lands in a hole */ - if (!err && tp && *bpp) + if (tp) xfs_trans_buf_set_type(tp, *bpp, XFS_BLFT_DIR_FREE_BUF); - return err; + + return 0; } int diff --git a/fs/xfs/libxfs/xfs_ialloc.c b/fs/xfs/libxfs/xfs_ialloc.c index f272abff11e1..d41ade5d293e 100644 --- a/fs/xfs/libxfs/xfs_ialloc.c +++ b/fs/xfs/libxfs/xfs_ialloc.c @@ -51,8 +51,7 @@ xfs_ialloc_cluster_alignment( struct xfs_mount *mp) { if (xfs_sb_version_hasalign(&mp->m_sb) && - mp->m_sb.sb_inoalignmt >= - XFS_B_TO_FSBT(mp, mp->m_inode_cluster_size)) + mp->m_sb.sb_inoalignmt >= xfs_icluster_size_fsb(mp)) return mp->m_sb.sb_inoalignmt; return 1; } diff --git a/fs/xfs/libxfs/xfs_inode_fork.c b/fs/xfs/libxfs/xfs_inode_fork.c index 222e103356c6..25c1e078aef6 100644 --- a/fs/xfs/libxfs/xfs_inode_fork.c +++ b/fs/xfs/libxfs/xfs_inode_fork.c @@ -26,6 +26,7 @@ #include "xfs_inode.h" #include "xfs_trans.h" #include "xfs_inode_item.h" +#include "xfs_btree.h" #include "xfs_bmap_btree.h" #include "xfs_bmap.h" #include "xfs_error.h" @@ -429,11 +430,13 @@ xfs_iformat_btree( /* REFERENCED */ int nrecs; int size; + int level; ifp = XFS_IFORK_PTR(ip, whichfork); dfp = (xfs_bmdr_block_t *)XFS_DFORK_PTR(dip, whichfork); size = XFS_BMAP_BROOT_SPACE(mp, dfp); nrecs = be16_to_cpu(dfp->bb_numrecs); + level = be16_to_cpu(dfp->bb_level); /* * blow out if -- fork has less extents than can fit in @@ -446,7 +449,8 @@ xfs_iformat_btree( XFS_IFORK_MAXEXT(ip, whichfork) || XFS_BMDR_SPACE_CALC(nrecs) > XFS_DFORK_SIZE(dip, mp, whichfork) || - XFS_IFORK_NEXTENTS(ip, whichfork) > ip->i_d.di_nblocks)) { + XFS_IFORK_NEXTENTS(ip, whichfork) > ip->i_d.di_nblocks) || + level == 0 || level > XFS_BTREE_MAXLEVELS) { xfs_warn(mp, "corrupt inode %Lu (btree).", (unsigned long long) ip->i_ino); XFS_CORRUPTION_ERROR("xfs_iformat_btree", XFS_ERRLEVEL_LOW, @@ -497,15 +501,14 @@ xfs_iread_extents( * We know that the size is valid (it's checked in iformat_btree) */ ifp->if_bytes = ifp->if_real_bytes = 0; - ifp->if_flags |= XFS_IFEXTENTS; xfs_iext_add(ifp, 0, nextents); error = xfs_bmap_read_extents(tp, ip, whichfork); if (error) { xfs_iext_destroy(ifp); - ifp->if_flags &= ~XFS_IFEXTENTS; return error; } xfs_validate_extents(ifp, nextents, XFS_EXTFMT_INODE(ip)); + ifp->if_flags |= XFS_IFEXTENTS; return 0; } /* diff --git a/fs/xfs/libxfs/xfs_log_recover.h b/fs/xfs/libxfs/xfs_log_recover.h index d9f65e2d5cc8..29a01ec89dd0 100644 --- a/fs/xfs/libxfs/xfs_log_recover.h +++ b/fs/xfs/libxfs/xfs_log_recover.h @@ -42,7 +42,6 @@ typedef struct xlog_recover_item { xfs_log_iovec_t *ri_buf; /* ptr to regions buffer */ } xlog_recover_item_t; -struct xlog_tid; typedef struct xlog_recover { struct hlist_node r_list; xlog_tid_t r_log_tid; /* log's transaction id */ diff --git a/fs/xfs/xfs_aops.c b/fs/xfs/xfs_aops.c index 631e7c0e0a29..1ff9df7a3ce8 100644 --- a/fs/xfs/xfs_aops.c +++ b/fs/xfs/xfs_aops.c @@ -481,6 +481,12 @@ xfs_submit_ioend( struct xfs_ioend *ioend, int status) { + /* Convert CoW extents to regular */ + if (!status && ioend->io_type == XFS_IO_COW) { + status = xfs_reflink_convert_cow(XFS_I(ioend->io_inode), + ioend->io_offset, ioend->io_size); + } + /* Reserve log space if we might write beyond the on-disk inode size. */ if (!status && ioend->io_type != XFS_IO_UNWRITTEN && diff --git a/fs/xfs/xfs_bmap_util.c b/fs/xfs/xfs_bmap_util.c index c1417919ab0a..8b75dcea5966 100644 --- a/fs/xfs/xfs_bmap_util.c +++ b/fs/xfs/xfs_bmap_util.c @@ -88,7 +88,6 @@ int xfs_bmap_rtalloc( struct xfs_bmalloca *ap) /* bmap alloc argument struct */ { - xfs_alloctype_t atype = 0; /* type for allocation routines */ int error; /* error return value */ xfs_mount_t *mp; /* mount point structure */ xfs_extlen_t prod = 0; /* product factor for allocators */ @@ -155,18 +154,14 @@ xfs_bmap_rtalloc( /* * Realtime allocation, done through xfs_rtallocate_extent. */ - atype = ap->blkno == 0 ? XFS_ALLOCTYPE_ANY_AG : XFS_ALLOCTYPE_NEAR_BNO; do_div(ap->blkno, mp->m_sb.sb_rextsize); rtb = ap->blkno; ap->length = ralen; - if ((error = xfs_rtallocate_extent(ap->tp, ap->blkno, 1, ap->length, - &ralen, atype, ap->wasdel, prod, &rtb))) - return error; - if (rtb == NULLFSBLOCK && prod > 1 && - (error = xfs_rtallocate_extent(ap->tp, ap->blkno, 1, - ap->length, &ralen, atype, - ap->wasdel, 1, &rtb))) + error = xfs_rtallocate_extent(ap->tp, ap->blkno, 1, ap->length, + &ralen, ap->wasdel, prod, &rtb); + if (error) return error; + ap->blkno = rtb; if (ap->blkno != NULLFSBLOCK) { ap->blkno *= mp->m_sb.sb_rextsize; @@ -787,11 +782,9 @@ xfs_getbmap( xfs_iunlock(ip, XFS_IOLOCK_SHARED); for (i = 0; i < cur_ext; i++) { - int full = 0; /* user array is full */ - /* format results & advance arg */ - error = formatter(&arg, &out[i], &full); - if (error || full) + error = formatter(&arg, &out[i]); + if (error) break; } @@ -917,17 +910,18 @@ xfs_can_free_eofblocks(struct xfs_inode *ip, bool force) */ int xfs_free_eofblocks( - xfs_mount_t *mp, - xfs_inode_t *ip, - bool need_iolock) + struct xfs_inode *ip) { - xfs_trans_t *tp; - int error; - xfs_fileoff_t end_fsb; - xfs_fileoff_t last_fsb; - xfs_filblks_t map_len; - int nimaps; - xfs_bmbt_irec_t imap; + struct xfs_trans *tp; + int error; + xfs_fileoff_t end_fsb; + xfs_fileoff_t last_fsb; + xfs_filblks_t map_len; + int nimaps; + struct xfs_bmbt_irec imap; + struct xfs_mount *mp = ip->i_mount; + + ASSERT(xfs_isilocked(ip, XFS_IOLOCK_EXCL)); /* * Figure out if there are any blocks beyond the end @@ -944,6 +938,10 @@ xfs_free_eofblocks( error = xfs_bmapi_read(ip, end_fsb, map_len, &imap, &nimaps, 0); xfs_iunlock(ip, XFS_ILOCK_SHARED); + /* + * If there are blocks after the end of file, truncate the file to its + * current size to free them up. + */ if (!error && (nimaps != 0) && (imap.br_startblock != HOLESTARTBLOCK || ip->i_delayed_blks)) { @@ -954,22 +952,13 @@ xfs_free_eofblocks( if (error) return error; - /* - * There are blocks after the end of file. - * Free them up now by truncating the file to - * its current size. - */ - if (need_iolock) { - if (!xfs_ilock_nowait(ip, XFS_IOLOCK_EXCL)) - return -EAGAIN; - } + /* wait on dio to ensure i_size has settled */ + inode_dio_wait(VFS_I(ip)); error = xfs_trans_alloc(mp, &M_RES(mp)->tr_itruncate, 0, 0, 0, &tp); if (error) { ASSERT(XFS_FORCED_SHUTDOWN(mp)); - if (need_iolock) - xfs_iunlock(ip, XFS_IOLOCK_EXCL); return error; } @@ -997,8 +986,6 @@ xfs_free_eofblocks( } xfs_iunlock(ip, XFS_ILOCK_EXCL); - if (need_iolock) - xfs_iunlock(ip, XFS_IOLOCK_EXCL); } return error; } @@ -1393,10 +1380,16 @@ xfs_shift_file_space( xfs_fileoff_t stop_fsb; xfs_fileoff_t next_fsb; xfs_fileoff_t shift_fsb; + uint resblks; ASSERT(direction == SHIFT_LEFT || direction == SHIFT_RIGHT); if (direction == SHIFT_LEFT) { + /* + * Reserve blocks to cover potential extent merges after left + * shift operations. + */ + resblks = XFS_DIOSTRAT_SPACE_RES(mp, 0); next_fsb = XFS_B_TO_FSB(mp, offset + len); stop_fsb = XFS_B_TO_FSB(mp, VFS_I(ip)->i_size); } else { @@ -1404,6 +1397,7 @@ xfs_shift_file_space( * If right shift, delegate the work of initialization of * next_fsb to xfs_bmap_shift_extent as it has ilock held. */ + resblks = 0; next_fsb = NULLFSBLOCK; stop_fsb = XFS_B_TO_FSB(mp, offset); } @@ -1415,7 +1409,7 @@ xfs_shift_file_space( * into the accessible region of the file. */ if (xfs_can_free_eofblocks(ip, true)) { - error = xfs_free_eofblocks(mp, ip, false); + error = xfs_free_eofblocks(ip); if (error) return error; } @@ -1445,21 +1439,14 @@ xfs_shift_file_space( } while (!error && !done) { - /* - * We would need to reserve permanent block for transaction. - * This will come into picture when after shifting extent into - * hole we found that adjacent extents can be merged which - * may lead to freeing of a block during record update. - */ - error = xfs_trans_alloc(mp, &M_RES(mp)->tr_write, - XFS_DIOSTRAT_SPACE_RES(mp, 0), 0, 0, &tp); + error = xfs_trans_alloc(mp, &M_RES(mp)->tr_write, resblks, 0, 0, + &tp); if (error) break; xfs_ilock(ip, XFS_ILOCK_EXCL); error = xfs_trans_reserve_quota(tp, mp, ip->i_udquot, - ip->i_gdquot, ip->i_pdquot, - XFS_DIOSTRAT_SPACE_RES(mp, 0), 0, + ip->i_gdquot, ip->i_pdquot, resblks, 0, XFS_QMOPT_RES_REGBLKS); if (error) goto out_trans_cancel; diff --git a/fs/xfs/xfs_bmap_util.h b/fs/xfs/xfs_bmap_util.h index 68a621a8e0c0..135d8267e284 100644 --- a/fs/xfs/xfs_bmap_util.h +++ b/fs/xfs/xfs_bmap_util.h @@ -35,7 +35,7 @@ int xfs_bmap_punch_delalloc_range(struct xfs_inode *ip, xfs_fileoff_t start_fsb, xfs_fileoff_t length); /* bmap to userspace formatter - copy to user & advance pointer */ -typedef int (*xfs_bmap_format_t)(void **, struct getbmapx *, int *); +typedef int (*xfs_bmap_format_t)(void **, struct getbmapx *); int xfs_getbmap(struct xfs_inode *ip, struct getbmapx *bmv, xfs_bmap_format_t formatter, void *arg); @@ -63,8 +63,7 @@ int xfs_insert_file_space(struct xfs_inode *, xfs_off_t offset, /* EOF block manipulation functions */ bool xfs_can_free_eofblocks(struct xfs_inode *ip, bool force); -int xfs_free_eofblocks(struct xfs_mount *mp, struct xfs_inode *ip, - bool need_iolock); +int xfs_free_eofblocks(struct xfs_inode *ip); int xfs_swap_extents(struct xfs_inode *ip, struct xfs_inode *tip, struct xfs_swapext *sx); diff --git a/fs/xfs/xfs_buf_item.c b/fs/xfs/xfs_buf_item.c index 2975cb2319f4..0306168af332 100644 --- a/fs/xfs/xfs_buf_item.c +++ b/fs/xfs/xfs_buf_item.c @@ -1162,6 +1162,7 @@ xfs_buf_iodone_callbacks( */ bp->b_last_error = 0; bp->b_retries = 0; + bp->b_first_retry_time = 0; xfs_buf_do_callbacks(bp); bp->b_fspriv = NULL; diff --git a/fs/xfs/xfs_discard.c b/fs/xfs/xfs_discard.c index 4ff499aa7338..d796ffac7296 100644 --- a/fs/xfs/xfs_discard.c +++ b/fs/xfs/xfs_discard.c @@ -208,32 +208,3 @@ xfs_ioc_trim( return -EFAULT; return 0; } - -int -xfs_discard_extents( - struct xfs_mount *mp, - struct list_head *list) -{ - struct xfs_extent_busy *busyp; - int error = 0; - - list_for_each_entry(busyp, list, list) { - trace_xfs_discard_extent(mp, busyp->agno, busyp->bno, - busyp->length); - - error = blkdev_issue_discard(mp->m_ddev_targp->bt_bdev, - XFS_AGB_TO_DADDR(mp, busyp->agno, busyp->bno), - XFS_FSB_TO_BB(mp, busyp->length), - GFP_NOFS, 0); - if (error && error != -EOPNOTSUPP) { - xfs_info(mp, - "discard failed for extent [0x%llx,%u], error %d", - (unsigned long long)busyp->bno, - busyp->length, - error); - return error; - } - } - - return 0; -} diff --git a/fs/xfs/xfs_discard.h b/fs/xfs/xfs_discard.h index 344879aea646..0f070f9e44e1 100644 --- a/fs/xfs/xfs_discard.h +++ b/fs/xfs/xfs_discard.h @@ -5,6 +5,5 @@ struct fstrim_range; struct list_head; extern int xfs_ioc_trim(struct xfs_mount *, struct fstrim_range __user *); -extern int xfs_discard_extents(struct xfs_mount *, struct list_head *); #endif /* XFS_DISCARD_H */ diff --git a/fs/xfs/xfs_extent_busy.c b/fs/xfs/xfs_extent_busy.c index 162dc186cf04..77760dbf0242 100644 --- a/fs/xfs/xfs_extent_busy.c +++ b/fs/xfs/xfs_extent_busy.c @@ -45,18 +45,7 @@ xfs_extent_busy_insert( struct rb_node **rbp; struct rb_node *parent = NULL; - new = kmem_zalloc(sizeof(struct xfs_extent_busy), KM_MAYFAIL); - if (!new) { - /* - * No Memory! Since it is now not possible to track the free - * block, make this a synchronous transaction to insure that - * the block is not reused before this transaction commits. - */ - trace_xfs_extent_busy_enomem(tp->t_mountp, agno, bno, len); - xfs_trans_set_sync(tp); - return; - } - + new = kmem_zalloc(sizeof(struct xfs_extent_busy), KM_SLEEP); new->agno = agno; new->bno = bno; new->length = len; @@ -345,25 +334,31 @@ restart: * subset of the extent that is not busy. If *rlen is smaller than * args->minlen no suitable extent could be found, and the higher level * code needs to force out the log and retry the allocation. + * + * Return the current busy generation for the AG if the extent is busy. This + * value can be used to wait for at least one of the currently busy extents + * to be cleared. Note that the busy list is not guaranteed to be empty after + * the gen is woken. The state of a specific extent must always be confirmed + * with another call to xfs_extent_busy_trim() before it can be used. */ -void +bool xfs_extent_busy_trim( struct xfs_alloc_arg *args, - xfs_agblock_t bno, - xfs_extlen_t len, - xfs_agblock_t *rbno, - xfs_extlen_t *rlen) + xfs_agblock_t *bno, + xfs_extlen_t *len, + unsigned *busy_gen) { xfs_agblock_t fbno; xfs_extlen_t flen; struct rb_node *rbp; + bool ret = false; - ASSERT(len > 0); + ASSERT(*len > 0); spin_lock(&args->pag->pagb_lock); restart: - fbno = bno; - flen = len; + fbno = *bno; + flen = *len; rbp = args->pag->pagb_tree.rb_node; while (rbp && flen >= args->minlen) { struct xfs_extent_busy *busyp = @@ -515,24 +510,25 @@ restart: flen = fend - fbno; } - spin_unlock(&args->pag->pagb_lock); +out: - if (fbno != bno || flen != len) { - trace_xfs_extent_busy_trim(args->mp, args->agno, bno, len, + if (fbno != *bno || flen != *len) { + trace_xfs_extent_busy_trim(args->mp, args->agno, *bno, *len, fbno, flen); + *bno = fbno; + *len = flen; + *busy_gen = args->pag->pagb_gen; + ret = true; } - *rbno = fbno; - *rlen = flen; - return; + spin_unlock(&args->pag->pagb_lock); + return ret; fail: /* * Return a zero extent length as failure indications. All callers * re-check if the trimmed extent satisfies the minlen requirement. */ - spin_unlock(&args->pag->pagb_lock); - trace_xfs_extent_busy_trim(args->mp, args->agno, bno, len, fbno, 0); - *rbno = fbno; - *rlen = 0; + flen = 0; + goto out; } STATIC void @@ -551,6 +547,21 @@ xfs_extent_busy_clear_one( kmem_free(busyp); } +static void +xfs_extent_busy_put_pag( + struct xfs_perag *pag, + bool wakeup) + __releases(pag->pagb_lock) +{ + if (wakeup) { + pag->pagb_gen++; + wake_up_all(&pag->pagb_wait); + } + + spin_unlock(&pag->pagb_lock); + xfs_perag_put(pag); +} + /* * Remove all extents on the passed in list from the busy extents tree. * If do_discard is set skip extents that need to be discarded, and mark @@ -565,27 +576,76 @@ xfs_extent_busy_clear( struct xfs_extent_busy *busyp, *n; struct xfs_perag *pag = NULL; xfs_agnumber_t agno = NULLAGNUMBER; + bool wakeup = false; list_for_each_entry_safe(busyp, n, list, list) { if (busyp->agno != agno) { - if (pag) { - spin_unlock(&pag->pagb_lock); - xfs_perag_put(pag); - } - pag = xfs_perag_get(mp, busyp->agno); - spin_lock(&pag->pagb_lock); + if (pag) + xfs_extent_busy_put_pag(pag, wakeup); agno = busyp->agno; + pag = xfs_perag_get(mp, agno); + spin_lock(&pag->pagb_lock); + wakeup = false; } if (do_discard && busyp->length && - !(busyp->flags & XFS_EXTENT_BUSY_SKIP_DISCARD)) + !(busyp->flags & XFS_EXTENT_BUSY_SKIP_DISCARD)) { busyp->flags = XFS_EXTENT_BUSY_DISCARDED; - else + } else { xfs_extent_busy_clear_one(mp, pag, busyp); + wakeup = true; + } } - if (pag) { - spin_unlock(&pag->pagb_lock); + if (pag) + xfs_extent_busy_put_pag(pag, wakeup); +} + +/* + * Flush out all busy extents for this AG. + */ +void +xfs_extent_busy_flush( + struct xfs_mount *mp, + struct xfs_perag *pag, + unsigned busy_gen) +{ + DEFINE_WAIT (wait); + int log_flushed = 0, error; + + trace_xfs_log_force(mp, 0, _THIS_IP_); + error = _xfs_log_force(mp, XFS_LOG_SYNC, &log_flushed); + if (error) + return; + + do { + prepare_to_wait(&pag->pagb_wait, &wait, TASK_KILLABLE); + if (busy_gen != READ_ONCE(pag->pagb_gen)) + break; + schedule(); + } while (1); + + finish_wait(&pag->pagb_wait, &wait); +} + +void +xfs_extent_busy_wait_all( + struct xfs_mount *mp) +{ + DEFINE_WAIT (wait); + xfs_agnumber_t agno; + + for (agno = 0; agno < mp->m_sb.sb_agcount; agno++) { + struct xfs_perag *pag = xfs_perag_get(mp, agno); + + do { + prepare_to_wait(&pag->pagb_wait, &wait, TASK_KILLABLE); + if (RB_EMPTY_ROOT(&pag->pagb_tree)) + break; + schedule(); + } while (1); + finish_wait(&pag->pagb_wait, &wait); + xfs_perag_put(pag); } } @@ -596,9 +656,17 @@ xfs_extent_busy_clear( int xfs_extent_busy_ag_cmp( void *priv, - struct list_head *a, - struct list_head *b) + struct list_head *l1, + struct list_head *l2) { - return container_of(a, struct xfs_extent_busy, list)->agno - - container_of(b, struct xfs_extent_busy, list)->agno; + struct xfs_extent_busy *b1 = + container_of(l1, struct xfs_extent_busy, list); + struct xfs_extent_busy *b2 = + container_of(l2, struct xfs_extent_busy, list); + s32 diff; + + diff = b1->agno - b2->agno; + if (!diff) + diff = b1->bno - b2->bno; + return diff; } diff --git a/fs/xfs/xfs_extent_busy.h b/fs/xfs/xfs_extent_busy.h index bfff284d2dcc..60195ea1b84a 100644 --- a/fs/xfs/xfs_extent_busy.h +++ b/fs/xfs/xfs_extent_busy.h @@ -58,9 +58,16 @@ void xfs_extent_busy_reuse(struct xfs_mount *mp, xfs_agnumber_t agno, xfs_agblock_t fbno, xfs_extlen_t flen, bool userdata); +bool +xfs_extent_busy_trim(struct xfs_alloc_arg *args, xfs_agblock_t *bno, + xfs_extlen_t *len, unsigned *busy_gen); + +void +xfs_extent_busy_flush(struct xfs_mount *mp, struct xfs_perag *pag, + unsigned busy_gen); + void -xfs_extent_busy_trim(struct xfs_alloc_arg *args, xfs_agblock_t bno, - xfs_extlen_t len, xfs_agblock_t *rbno, xfs_extlen_t *rlen); +xfs_extent_busy_wait_all(struct xfs_mount *mp); int xfs_extent_busy_ag_cmp(void *priv, struct list_head *a, struct list_head *b); diff --git a/fs/xfs/xfs_file.c b/fs/xfs/xfs_file.c index 9d8440b07b53..022014016d80 100644 --- a/fs/xfs/xfs_file.c +++ b/fs/xfs/xfs_file.c @@ -527,6 +527,15 @@ xfs_file_dio_aio_write( if ((iocb->ki_pos & mp->m_blockmask) || ((iocb->ki_pos + count) & mp->m_blockmask)) { unaligned_io = 1; + + /* + * We can't properly handle unaligned direct I/O to reflink + * files yet, as we can't unshare a partial block. + */ + if (xfs_is_reflink_inode(ip)) { + trace_xfs_reflink_bounce_dio_write(ip, iocb->ki_pos, count); + return -EREMCHG; + } iolock = XFS_IOLOCK_EXCL; } else { iolock = XFS_IOLOCK_SHARED; @@ -552,14 +561,6 @@ xfs_file_dio_aio_write( } trace_xfs_file_direct_write(ip, count, iocb->ki_pos); - - /* If this is a block-aligned directio CoW, remap immediately. */ - if (xfs_is_reflink_inode(ip) && !unaligned_io) { - ret = xfs_reflink_allocate_cow_range(ip, iocb->ki_pos, count); - if (ret) - goto out; - } - ret = iomap_dio_rw(iocb, from, &xfs_iomap_ops, xfs_dio_write_end_io); out: xfs_iunlock(ip, iolock); @@ -614,8 +615,10 @@ xfs_file_buffered_aio_write( struct xfs_inode *ip = XFS_I(inode); ssize_t ret; int enospc = 0; - int iolock = XFS_IOLOCK_EXCL; + int iolock; +write_retry: + iolock = XFS_IOLOCK_EXCL; xfs_ilock(ip, iolock); ret = xfs_file_aio_write_checks(iocb, from, &iolock); @@ -625,7 +628,6 @@ xfs_file_buffered_aio_write( /* We can write back this queue in page reclaim */ current->backing_dev_info = inode_to_bdi(inode); -write_retry: trace_xfs_file_buffered_write(ip, iov_iter_count(from), iocb->ki_pos); ret = iomap_file_buffered_write(iocb, from, &xfs_iomap_ops); if (likely(ret >= 0)) @@ -641,18 +643,21 @@ write_retry: * running at the same time. */ if (ret == -EDQUOT && !enospc) { + xfs_iunlock(ip, iolock); enospc = xfs_inode_free_quota_eofblocks(ip); if (enospc) goto write_retry; enospc = xfs_inode_free_quota_cowblocks(ip); if (enospc) goto write_retry; + iolock = 0; } else if (ret == -ENOSPC && !enospc) { struct xfs_eofblocks eofb = {0}; enospc = 1; xfs_flush_inodes(ip->i_mount); - eofb.eof_scan_owner = ip->i_ino; /* for locking */ + + xfs_iunlock(ip, iolock); eofb.eof_flags = XFS_EOF_FLAGS_SYNC; xfs_icache_free_eofblocks(ip->i_mount, &eofb); goto write_retry; @@ -660,7 +665,8 @@ write_retry: current->backing_dev_info = NULL; out: - xfs_iunlock(ip, iolock); + if (iolock) + xfs_iunlock(ip, iolock); return ret; } @@ -908,9 +914,9 @@ xfs_dir_open( */ mode = xfs_ilock_data_map_shared(ip); if (ip->i_d.di_nextents > 0) - xfs_dir3_data_readahead(ip, 0, -1); + error = xfs_dir3_data_readahead(ip, 0, -1); xfs_iunlock(ip, mode); - return 0; + return error; } STATIC int diff --git a/fs/xfs/xfs_fsops.c b/fs/xfs/xfs_fsops.c index 242e8091296d..6ccaae9eb0ee 100644 --- a/fs/xfs/xfs_fsops.c +++ b/fs/xfs/xfs_fsops.c @@ -352,12 +352,7 @@ xfs_growfs_data_private( goto error0; } - if (xfs_sb_version_hascrc(&mp->m_sb)) - xfs_btree_init_block(mp, bp, XFS_ABTB_CRC_MAGIC, 0, 1, - agno, XFS_BTREE_CRC_BLOCKS); - else - xfs_btree_init_block(mp, bp, XFS_ABTB_MAGIC, 0, 1, - agno, 0); + xfs_btree_init_block(mp, bp, XFS_BTNUM_BNO, 0, 1, agno, 0); arec = XFS_ALLOC_REC_ADDR(mp, XFS_BUF_TO_BLOCK(bp), 1); arec->ar_startblock = cpu_to_be32(mp->m_ag_prealloc_blocks); @@ -381,12 +376,7 @@ xfs_growfs_data_private( goto error0; } - if (xfs_sb_version_hascrc(&mp->m_sb)) - xfs_btree_init_block(mp, bp, XFS_ABTC_CRC_MAGIC, 0, 1, - agno, XFS_BTREE_CRC_BLOCKS); - else - xfs_btree_init_block(mp, bp, XFS_ABTC_MAGIC, 0, 1, - agno, 0); + xfs_btree_init_block(mp, bp, XFS_BTNUM_CNT, 0, 1, agno, 0); arec = XFS_ALLOC_REC_ADDR(mp, XFS_BUF_TO_BLOCK(bp), 1); arec->ar_startblock = cpu_to_be32(mp->m_ag_prealloc_blocks); @@ -413,8 +403,8 @@ xfs_growfs_data_private( goto error0; } - xfs_btree_init_block(mp, bp, XFS_RMAP_CRC_MAGIC, 0, 0, - agno, XFS_BTREE_CRC_BLOCKS); + xfs_btree_init_block(mp, bp, XFS_BTNUM_RMAP, 0, 0, + agno, 0); block = XFS_BUF_TO_BLOCK(bp); @@ -488,12 +478,7 @@ xfs_growfs_data_private( goto error0; } - if (xfs_sb_version_hascrc(&mp->m_sb)) - xfs_btree_init_block(mp, bp, XFS_IBT_CRC_MAGIC, 0, 0, - agno, XFS_BTREE_CRC_BLOCKS); - else - xfs_btree_init_block(mp, bp, XFS_IBT_MAGIC, 0, 0, - agno, 0); + xfs_btree_init_block(mp, bp, XFS_BTNUM_INO , 0, 0, agno, 0); error = xfs_bwrite(bp); xfs_buf_relse(bp); @@ -513,13 +498,8 @@ xfs_growfs_data_private( goto error0; } - if (xfs_sb_version_hascrc(&mp->m_sb)) - xfs_btree_init_block(mp, bp, XFS_FIBT_CRC_MAGIC, - 0, 0, agno, - XFS_BTREE_CRC_BLOCKS); - else - xfs_btree_init_block(mp, bp, XFS_FIBT_MAGIC, 0, - 0, agno, 0); + xfs_btree_init_block(mp, bp, XFS_BTNUM_FINO, + 0, 0, agno, 0); error = xfs_bwrite(bp); xfs_buf_relse(bp); @@ -540,9 +520,8 @@ xfs_growfs_data_private( goto error0; } - xfs_btree_init_block(mp, bp, XFS_REFC_CRC_MAGIC, - 0, 0, agno, - XFS_BTREE_CRC_BLOCKS); + xfs_btree_init_block(mp, bp, XFS_BTNUM_REFC, + 0, 0, agno, 0); error = xfs_bwrite(bp); xfs_buf_relse(bp); diff --git a/fs/xfs/xfs_icache.c b/fs/xfs/xfs_icache.c index 70ca4f608321..7234b9748c36 100644 --- a/fs/xfs/xfs_icache.c +++ b/fs/xfs/xfs_icache.c @@ -1322,13 +1322,10 @@ xfs_inode_free_eofblocks( int flags, void *args) { - int ret; + int ret = 0; struct xfs_eofblocks *eofb = args; - bool need_iolock = true; int match; - ASSERT(!eofb || (eofb && eofb->eof_scan_owner != 0)); - if (!xfs_can_free_eofblocks(ip, false)) { /* inode could be preallocated or append-only */ trace_xfs_inode_free_eofblocks_invalid(ip); @@ -1356,21 +1353,19 @@ xfs_inode_free_eofblocks( if (eofb->eof_flags & XFS_EOF_FLAGS_MINFILESIZE && XFS_ISIZE(ip) < eofb->eof_min_file_size) return 0; - - /* - * A scan owner implies we already hold the iolock. Skip it in - * xfs_free_eofblocks() to avoid deadlock. This also eliminates - * the possibility of EAGAIN being returned. - */ - if (eofb->eof_scan_owner == ip->i_ino) - need_iolock = false; } - ret = xfs_free_eofblocks(ip->i_mount, ip, need_iolock); - - /* don't revisit the inode if we're not waiting */ - if (ret == -EAGAIN && !(flags & SYNC_WAIT)) - ret = 0; + /* + * If the caller is waiting, return -EAGAIN to keep the background + * scanner moving and revisit the inode in a subsequent pass. + */ + if (!xfs_ilock_nowait(ip, XFS_IOLOCK_EXCL)) { + if (flags & SYNC_WAIT) + ret = -EAGAIN; + return ret; + } + ret = xfs_free_eofblocks(ip); + xfs_iunlock(ip, XFS_IOLOCK_EXCL); return ret; } @@ -1417,15 +1412,10 @@ __xfs_inode_free_quota_eofblocks( struct xfs_eofblocks eofb = {0}; struct xfs_dquot *dq; - ASSERT(xfs_isilocked(ip, XFS_IOLOCK_EXCL)); - /* - * Set the scan owner to avoid a potential livelock. Otherwise, the scan - * can repeatedly trylock on the inode we're currently processing. We - * run a sync scan to increase effectiveness and use the union filter to + * Run a sync scan to increase effectiveness and use the union filter to * cover all applicable quotas in a single scan. */ - eofb.eof_scan_owner = ip->i_ino; eofb.eof_flags = XFS_EOF_FLAGS_UNION|XFS_EOF_FLAGS_SYNC; if (XFS_IS_UQUOTA_ENFORCED(ip->i_mount)) { @@ -1577,12 +1567,9 @@ xfs_inode_free_cowblocks( { int ret; struct xfs_eofblocks *eofb = args; - bool need_iolock = true; int match; struct xfs_ifork *ifp = XFS_IFORK_PTR(ip, XFS_COW_FORK); - ASSERT(!eofb || (eofb && eofb->eof_scan_owner != 0)); - /* * Just clear the tag if we have an empty cow fork or none at all. It's * possible the inode was fully unshared since it was originally tagged. @@ -1615,28 +1602,16 @@ xfs_inode_free_cowblocks( if (eofb->eof_flags & XFS_EOF_FLAGS_MINFILESIZE && XFS_ISIZE(ip) < eofb->eof_min_file_size) return 0; - - /* - * A scan owner implies we already hold the iolock. Skip it in - * xfs_free_eofblocks() to avoid deadlock. This also eliminates - * the possibility of EAGAIN being returned. - */ - if (eofb->eof_scan_owner == ip->i_ino) - need_iolock = false; } /* Free the CoW blocks */ - if (need_iolock) { - xfs_ilock(ip, XFS_IOLOCK_EXCL); - xfs_ilock(ip, XFS_MMAPLOCK_EXCL); - } + xfs_ilock(ip, XFS_IOLOCK_EXCL); + xfs_ilock(ip, XFS_MMAPLOCK_EXCL); ret = xfs_reflink_cancel_cow_range(ip, 0, NULLFILEOFF); - if (need_iolock) { - xfs_iunlock(ip, XFS_MMAPLOCK_EXCL); - xfs_iunlock(ip, XFS_IOLOCK_EXCL); - } + xfs_iunlock(ip, XFS_MMAPLOCK_EXCL); + xfs_iunlock(ip, XFS_IOLOCK_EXCL); return ret; } diff --git a/fs/xfs/xfs_icache.h b/fs/xfs/xfs_icache.h index a1e02f4708ab..8a7c849b4dea 100644 --- a/fs/xfs/xfs_icache.h +++ b/fs/xfs/xfs_icache.h @@ -27,7 +27,6 @@ struct xfs_eofblocks { kgid_t eof_gid; prid_t eof_prid; __u64 eof_min_file_size; - xfs_ino_t eof_scan_owner; }; #define SYNC_WAIT 0x0001 /* wait for i/o to complete */ @@ -102,7 +101,6 @@ xfs_fs_eofblocks_from_user( dst->eof_flags = src->eof_flags; dst->eof_prid = src->eof_prid; dst->eof_min_file_size = src->eof_min_file_size; - dst->eof_scan_owner = NULLFSINO; dst->eof_uid = INVALID_UID; if (src->eof_flags & XFS_EOF_FLAGS_UID) { diff --git a/fs/xfs/xfs_inode.c b/fs/xfs/xfs_inode.c index de32f0fe47c8..edfa6a55b064 100644 --- a/fs/xfs/xfs_inode.c +++ b/fs/xfs/xfs_inode.c @@ -1692,32 +1692,34 @@ xfs_release( if (xfs_can_free_eofblocks(ip, false)) { /* + * Check if the inode is being opened, written and closed + * frequently and we have delayed allocation blocks outstanding + * (e.g. streaming writes from the NFS server), truncating the + * blocks past EOF will cause fragmentation to occur. + * + * In this case don't do the truncation, but we have to be + * careful how we detect this case. Blocks beyond EOF show up as + * i_delayed_blks even when the inode is clean, so we need to + * truncate them away first before checking for a dirty release. + * Hence on the first dirty close we will still remove the + * speculative allocation, but after that we will leave it in + * place. + */ + if (xfs_iflags_test(ip, XFS_IDIRTY_RELEASE)) + return 0; + /* * If we can't get the iolock just skip truncating the blocks * past EOF because we could deadlock with the mmap_sem - * otherwise. We'll get another chance to drop them once the + * otherwise. We'll get another chance to drop them once the * last reference to the inode is dropped, so we'll never leak * blocks permanently. - * - * Further, check if the inode is being opened, written and - * closed frequently and we have delayed allocation blocks - * outstanding (e.g. streaming writes from the NFS server), - * truncating the blocks past EOF will cause fragmentation to - * occur. - * - * In this case don't do the truncation, either, but we have to - * be careful how we detect this case. Blocks beyond EOF show - * up as i_delayed_blks even when the inode is clean, so we - * need to truncate them away first before checking for a dirty - * release. Hence on the first dirty close we will still remove - * the speculative allocation, but after that we will leave it - * in place. */ - if (xfs_iflags_test(ip, XFS_IDIRTY_RELEASE)) - return 0; - - error = xfs_free_eofblocks(mp, ip, true); - if (error && error != -EAGAIN) - return error; + if (xfs_ilock_nowait(ip, XFS_IOLOCK_EXCL)) { + error = xfs_free_eofblocks(ip); + xfs_iunlock(ip, XFS_IOLOCK_EXCL); + if (error) + return error; + } /* delalloc blocks after truncation means it really is dirty */ if (ip->i_delayed_blks) @@ -1904,8 +1906,11 @@ xfs_inactive( * cache. Post-eof blocks must be freed, lest we end up with * broken free space accounting. */ - if (xfs_can_free_eofblocks(ip, true)) - xfs_free_eofblocks(mp, ip, false); + if (xfs_can_free_eofblocks(ip, true)) { + xfs_ilock(ip, XFS_IOLOCK_EXCL); + xfs_free_eofblocks(ip); + xfs_iunlock(ip, XFS_IOLOCK_EXCL); + } return; } diff --git a/fs/xfs/xfs_ioctl.c b/fs/xfs/xfs_ioctl.c index c67cfb451fd3..cf1363dbf32b 100644 --- a/fs/xfs/xfs_ioctl.c +++ b/fs/xfs/xfs_ioctl.c @@ -1524,7 +1524,7 @@ out_drop_write: } STATIC int -xfs_getbmap_format(void **ap, struct getbmapx *bmv, int *full) +xfs_getbmap_format(void **ap, struct getbmapx *bmv) { struct getbmap __user *base = (struct getbmap __user *)*ap; @@ -1567,7 +1567,7 @@ xfs_ioc_getbmap( } STATIC int -xfs_getbmapx_format(void **ap, struct getbmapx *bmv, int *full) +xfs_getbmapx_format(void **ap, struct getbmapx *bmv) { struct getbmapx __user *base = (struct getbmapx __user *)*ap; diff --git a/fs/xfs/xfs_iomap.c b/fs/xfs/xfs_iomap.c index 1aa3abd67b36..41662fb14e87 100644 --- a/fs/xfs/xfs_iomap.c +++ b/fs/xfs/xfs_iomap.c @@ -162,7 +162,7 @@ xfs_iomap_write_direct( xfs_fileoff_t last_fsb; xfs_filblks_t count_fsb, resaligned; xfs_fsblock_t firstfsb; - xfs_extlen_t extsz, temp; + xfs_extlen_t extsz; int nimaps; int quota_flag; int rt; @@ -203,14 +203,7 @@ xfs_iomap_write_direct( } count_fsb = last_fsb - offset_fsb; ASSERT(count_fsb > 0); - - resaligned = count_fsb; - if (unlikely(extsz)) { - if ((temp = do_mod(offset_fsb, extsz))) - resaligned += temp; - if ((temp = do_mod(resaligned, extsz))) - resaligned += extsz - temp; - } + resaligned = xfs_aligned_fsb_count(offset_fsb, count_fsb, extsz); if (unlikely(rt)) { resrtextents = qblocks = resaligned; @@ -685,7 +678,7 @@ xfs_iomap_write_allocate( int nres; if (whichfork == XFS_COW_FORK) - flags |= XFS_BMAPI_COWFORK; + flags |= XFS_BMAPI_COWFORK | XFS_BMAPI_PREALLOC; /* * Make sure that the dquots are there. @@ -1002,47 +995,31 @@ xfs_file_iomap_begin( offset_fsb = XFS_B_TO_FSBT(mp, offset); end_fsb = XFS_B_TO_FSB(mp, offset + length); - if (xfs_is_reflink_inode(ip) && - (flags & IOMAP_WRITE) && (flags & IOMAP_DIRECT)) { - shared = xfs_reflink_find_cow_mapping(ip, offset, &imap); - if (shared) { - xfs_iunlock(ip, lockmode); - goto alloc_done; - } - ASSERT(!isnullstartblock(imap.br_startblock)); - } - error = xfs_bmapi_read(ip, offset_fsb, end_fsb - offset_fsb, &imap, &nimaps, 0); if (error) goto out_unlock; - if ((flags & IOMAP_REPORT) || - (xfs_is_reflink_inode(ip) && - (flags & IOMAP_WRITE) && (flags & IOMAP_DIRECT))) { + if (flags & IOMAP_REPORT) { /* Trim the mapping to the nearest shared extent boundary. */ error = xfs_reflink_trim_around_shared(ip, &imap, &shared, &trimmed); if (error) goto out_unlock; - - /* - * We're here because we're trying to do a directio write to a - * region that isn't aligned to a filesystem block. If the - * extent is shared, fall back to buffered mode to handle the - * RMW. - */ - if (!(flags & IOMAP_REPORT) && shared) { - trace_xfs_reflink_bounce_dio_write(ip, &imap); - error = -EREMCHG; - goto out_unlock; - } } if ((flags & (IOMAP_WRITE | IOMAP_ZERO)) && xfs_is_reflink_inode(ip)) { - error = xfs_reflink_reserve_cow(ip, &imap, &shared); - if (error) - goto out_unlock; + if (flags & IOMAP_DIRECT) { + /* may drop and re-acquire the ilock */ + error = xfs_reflink_allocate_cow(ip, &imap, &shared, + &lockmode); + if (error) + goto out_unlock; + } else { + error = xfs_reflink_reserve_cow(ip, &imap, &shared); + if (error) + goto out_unlock; + } end_fsb = imap.br_startoff + imap.br_blockcount; length = XFS_FSB_TO_B(mp, end_fsb) - offset; @@ -1071,7 +1048,6 @@ xfs_file_iomap_begin( if (error) return error; -alloc_done: iomap->flags = IOMAP_F_NEW; trace_xfs_iomap_alloc(ip, offset, length, 0, &imap); } else { @@ -1102,7 +1078,19 @@ xfs_file_iomap_end_delalloc( xfs_fileoff_t end_fsb; int error = 0; - start_fsb = XFS_B_TO_FSB(mp, offset + written); + /* behave as if the write failed if drop writes is enabled */ + if (xfs_mp_drop_writes(mp)) + written = 0; + + /* + * start_fsb refers to the first unused block after a short write. If + * nothing was written, round offset down to point at the first block in + * the range. + */ + if (unlikely(!written)) + start_fsb = XFS_B_TO_FSBT(mp, offset); + else + start_fsb = XFS_B_TO_FSB(mp, offset + written); end_fsb = XFS_B_TO_FSB(mp, offset + length); /* @@ -1114,6 +1102,9 @@ xfs_file_iomap_end_delalloc( * blocks in the range, they are ours. */ if (start_fsb < end_fsb) { + truncate_pagecache_range(VFS_I(ip), XFS_FSB_TO_B(mp, start_fsb), + XFS_FSB_TO_B(mp, end_fsb) - 1); + xfs_ilock(ip, XFS_ILOCK_EXCL); error = xfs_bmap_punch_delalloc_range(ip, start_fsb, end_fsb - start_fsb); @@ -1144,7 +1135,7 @@ xfs_file_iomap_end( return 0; } -struct iomap_ops xfs_iomap_ops = { +const struct iomap_ops xfs_iomap_ops = { .iomap_begin = xfs_file_iomap_begin, .iomap_end = xfs_file_iomap_end, }; @@ -1190,6 +1181,6 @@ out_unlock: return error; } -struct iomap_ops xfs_xattr_iomap_ops = { +const struct iomap_ops xfs_xattr_iomap_ops = { .iomap_begin = xfs_xattr_iomap_begin, }; diff --git a/fs/xfs/xfs_iomap.h b/fs/xfs/xfs_iomap.h index 6d45cf01fcff..00db3ecea084 100644 --- a/fs/xfs/xfs_iomap.h +++ b/fs/xfs/xfs_iomap.h @@ -33,7 +33,27 @@ void xfs_bmbt_to_iomap(struct xfs_inode *, struct iomap *, struct xfs_bmbt_irec *); xfs_extlen_t xfs_eof_alignment(struct xfs_inode *ip, xfs_extlen_t extsize); -extern struct iomap_ops xfs_iomap_ops; -extern struct iomap_ops xfs_xattr_iomap_ops; +static inline xfs_filblks_t +xfs_aligned_fsb_count( + xfs_fileoff_t offset_fsb, + xfs_filblks_t count_fsb, + xfs_extlen_t extsz) +{ + if (extsz) { + xfs_extlen_t align; + + align = do_mod(offset_fsb, extsz); + if (align) + count_fsb += align; + align = do_mod(count_fsb, extsz); + if (align) + count_fsb += extsz - align; + } + + return count_fsb; +} + +extern const struct iomap_ops xfs_iomap_ops; +extern const struct iomap_ops xfs_xattr_iomap_ops; #endif /* __XFS_IOMAP_H__*/ diff --git a/fs/xfs/xfs_log.h b/fs/xfs/xfs_log.h index b5e71072fde5..cc5a9f1574e7 100644 --- a/fs/xfs/xfs_log.h +++ b/fs/xfs/xfs_log.h @@ -124,7 +124,6 @@ struct xlog_ticket; struct xfs_log_item; struct xfs_item_ops; struct xfs_trans; -struct xfs_log_callback; xfs_lsn_t xfs_log_done(struct xfs_mount *mp, struct xlog_ticket *ticket, diff --git a/fs/xfs/xfs_log_cil.c b/fs/xfs/xfs_log_cil.c index a4ab192e1792..82f1cbcc4de1 100644 --- a/fs/xfs/xfs_log_cil.c +++ b/fs/xfs/xfs_log_cil.c @@ -30,6 +30,9 @@ #include "xfs_trans_priv.h" #include "xfs_log.h" #include "xfs_log_priv.h" +#include "xfs_trace.h" + +struct workqueue_struct *xfs_discard_wq; /* * Allocate a new ticket. Failing to get a new ticket makes it really hard to @@ -491,6 +494,75 @@ xlog_cil_free_logvec( } } +static void +xlog_discard_endio_work( + struct work_struct *work) +{ + struct xfs_cil_ctx *ctx = + container_of(work, struct xfs_cil_ctx, discard_endio_work); + struct xfs_mount *mp = ctx->cil->xc_log->l_mp; + + xfs_extent_busy_clear(mp, &ctx->busy_extents, false); + kmem_free(ctx); +} + +/* + * Queue up the actual completion to a thread to avoid IRQ-safe locking for + * pagb_lock. Note that we need a unbounded workqueue, otherwise we might + * get the execution delayed up to 30 seconds for weird reasons. + */ +static void +xlog_discard_endio( + struct bio *bio) +{ + struct xfs_cil_ctx *ctx = bio->bi_private; + + INIT_WORK(&ctx->discard_endio_work, xlog_discard_endio_work); + queue_work(xfs_discard_wq, &ctx->discard_endio_work); +} + +static void +xlog_discard_busy_extents( + struct xfs_mount *mp, + struct xfs_cil_ctx *ctx) +{ + struct list_head *list = &ctx->busy_extents; + struct xfs_extent_busy *busyp; + struct bio *bio = NULL; + struct blk_plug plug; + int error = 0; + + ASSERT(mp->m_flags & XFS_MOUNT_DISCARD); + + blk_start_plug(&plug); + list_for_each_entry(busyp, list, list) { + trace_xfs_discard_extent(mp, busyp->agno, busyp->bno, + busyp->length); + + error = __blkdev_issue_discard(mp->m_ddev_targp->bt_bdev, + XFS_AGB_TO_DADDR(mp, busyp->agno, busyp->bno), + XFS_FSB_TO_BB(mp, busyp->length), + GFP_NOFS, 0, &bio); + if (error && error != -EOPNOTSUPP) { + xfs_info(mp, + "discard failed for extent [0x%llx,%u], error %d", + (unsigned long long)busyp->bno, + busyp->length, + error); + break; + } + } + + if (bio) { + bio->bi_private = ctx; + bio->bi_end_io = xlog_discard_endio; + submit_bio(bio); + } else { + xlog_discard_endio_work(&ctx->discard_endio_work); + } + blk_finish_plug(&plug); +} + /* * Mark all items committed and clear busy extents. We free the log vector * chains in a separate pass so that we unpin the log items as quickly as @@ -525,14 +597,10 @@ xlog_cil_committed( xlog_cil_free_logvec(ctx->lv_chain); - if (!list_empty(&ctx->busy_extents)) { - ASSERT(mp->m_flags & XFS_MOUNT_DISCARD); - - xfs_discard_extents(mp, &ctx->busy_extents); - xfs_extent_busy_clear(mp, &ctx->busy_extents, false); - } - - kmem_free(ctx); + if (!list_empty(&ctx->busy_extents)) + xlog_discard_busy_extents(mp, ctx); + else + kmem_free(ctx); } /* diff --git a/fs/xfs/xfs_log_priv.h b/fs/xfs/xfs_log_priv.h index 2b6eec52178e..c2604a5366f2 100644 --- a/fs/xfs/xfs_log_priv.h +++ b/fs/xfs/xfs_log_priv.h @@ -257,6 +257,7 @@ struct xfs_cil_ctx { struct xfs_log_vec *lv_chain; /* logvecs being pushed */ struct xfs_log_callback log_cb; /* completion callback hook. */ struct list_head committing; /* ctx committing list */ + struct work_struct discard_endio_work; }; /* diff --git a/fs/xfs/xfs_mount.c b/fs/xfs/xfs_mount.c index 9b9540db17a6..450bde68bb75 100644 --- a/fs/xfs/xfs_mount.c +++ b/fs/xfs/xfs_mount.c @@ -45,6 +45,7 @@ #include "xfs_rmap_btree.h" #include "xfs_refcount_btree.h" #include "xfs_reflink.h" +#include "xfs_extent_busy.h" static DEFINE_MUTEX(xfs_uuid_table_mutex); @@ -187,7 +188,7 @@ xfs_initialize_perag( xfs_agnumber_t *maxagi) { xfs_agnumber_t index; - xfs_agnumber_t first_initialised = 0; + xfs_agnumber_t first_initialised = NULLAGNUMBER; xfs_perag_t *pag; int error = -ENOMEM; @@ -202,22 +203,21 @@ xfs_initialize_perag( xfs_perag_put(pag); continue; } - if (!first_initialised) - first_initialised = index; pag = kmem_zalloc(sizeof(*pag), KM_MAYFAIL); if (!pag) - goto out_unwind; + goto out_unwind_new_pags; pag->pag_agno = index; pag->pag_mount = mp; spin_lock_init(&pag->pag_ici_lock); mutex_init(&pag->pag_ici_reclaim_lock); INIT_RADIX_TREE(&pag->pag_ici_root, GFP_ATOMIC); if (xfs_buf_hash_init(pag)) - goto out_unwind; + goto out_free_pag; + init_waitqueue_head(&pag->pagb_wait); if (radix_tree_preload(GFP_NOFS)) - goto out_unwind; + goto out_hash_destroy; spin_lock(&mp->m_perag_lock); if (radix_tree_insert(&mp->m_perag_tree, index, pag)) { @@ -225,10 +225,13 @@ xfs_initialize_perag( spin_unlock(&mp->m_perag_lock); radix_tree_preload_end(); error = -EEXIST; - goto out_unwind; + goto out_hash_destroy; } spin_unlock(&mp->m_perag_lock); radix_tree_preload_end(); + /* first new pag is fully initialized */ + if (first_initialised == NULLAGNUMBER) + first_initialised = index; } index = xfs_set_inode_alloc(mp, agcount); @@ -239,11 +242,16 @@ xfs_initialize_perag( mp->m_ag_prealloc_blocks = xfs_prealloc_blocks(mp); return 0; -out_unwind: +out_hash_destroy: xfs_buf_hash_destroy(pag); +out_free_pag: kmem_free(pag); - for (; index > first_initialised; index--) { +out_unwind_new_pags: + /* unwind any prior newly initialized pags */ + for (index = first_initialised; index < agcount; index++) { pag = radix_tree_delete(&mp->m_perag_tree, index); + if (!pag) + break; xfs_buf_hash_destroy(pag); kmem_free(pag); } @@ -1073,6 +1081,13 @@ xfs_unmountfs( xfs_log_force(mp, XFS_LOG_SYNC); /* + * Wait for all busy extents to be freed, including completion of + * any discard operation. + */ + xfs_extent_busy_wait_all(mp); + flush_workqueue(xfs_discard_wq); + + /* * We now need to tell the world we are unmounting. This will allow * us to detect that the filesystem is going away and we should error * out anything that we have been retrying in the background. This will diff --git a/fs/xfs/xfs_mount.h b/fs/xfs/xfs_mount.h index 7f351f706b7a..6db6fd6b82b0 100644 --- a/fs/xfs/xfs_mount.h +++ b/fs/xfs/xfs_mount.h @@ -200,11 +200,12 @@ typedef struct xfs_mount { /* * DEBUG mode instrumentation to test and/or trigger delayed allocation * block killing in the event of failed writes. When enabled, all - * buffered writes are forced to fail. All delalloc blocks in the range - * of the write (including pre-existing delalloc blocks!) are tossed as - * part of the write failure error handling sequence. + * buffered writes are silenty dropped and handled as if they failed. + * All delalloc blocks in the range of the write (including pre-existing + * delalloc blocks!) are tossed as part of the write failure error + * handling sequence. */ - bool m_fail_writes; + bool m_drop_writes; #endif } xfs_mount_t; @@ -325,13 +326,13 @@ xfs_daddr_to_agbno(struct xfs_mount *mp, xfs_daddr_t d) #ifdef DEBUG static inline bool -xfs_mp_fail_writes(struct xfs_mount *mp) +xfs_mp_drop_writes(struct xfs_mount *mp) { - return mp->m_fail_writes; + return mp->m_drop_writes; } #else static inline bool -xfs_mp_fail_writes(struct xfs_mount *mp) +xfs_mp_drop_writes(struct xfs_mount *mp) { return 0; } @@ -384,6 +385,8 @@ typedef struct xfs_perag { xfs_agino_t pagl_rightrec; spinlock_t pagb_lock; /* lock for pagb_tree */ struct rb_root pagb_tree; /* ordered tree of busy extents */ + unsigned int pagb_gen; /* generation count for pagb_tree */ + wait_queue_head_t pagb_wait; /* woken when pagb_gen changes */ atomic_t pagf_fstrms; /* # of filestreams active in this AG */ diff --git a/fs/xfs/xfs_reflink.c b/fs/xfs/xfs_reflink.c index 07593a362cd0..da6d08fb359c 100644 --- a/fs/xfs/xfs_reflink.c +++ b/fs/xfs/xfs_reflink.c @@ -82,11 +82,22 @@ * mappings are a reservation against the free space in the filesystem; * adjacent mappings can also be combined into fewer larger mappings. * + * As an optimization, the CoW extent size hint (cowextsz) creates + * outsized aligned delalloc reservations in the hope of landing out of + * order nearby CoW writes in a single extent on disk, thereby reducing + * fragmentation and improving future performance. + * + * D: --RRRRRRSSSRRRRRRRR--- (data fork) + * C: ------DDDDDDD--------- (CoW fork) + * * When dirty pages are being written out (typically in writepage), the - * delalloc reservations are converted into real mappings by allocating - * blocks and replacing the delalloc mapping with real ones. A delalloc - * mapping can be replaced by several real ones if the free space is - * fragmented. + * delalloc reservations are converted into unwritten mappings by + * allocating blocks and replacing the delalloc mapping with real ones. + * A delalloc mapping can be replaced by several unwritten ones if the + * free space is fragmented. + * + * D: --RRRRRRSSSRRRRRRRR--- + * C: ------UUUUUUU--------- * * We want to adapt the delalloc mechanism for copy-on-write, since the * write paths are similar. The first two steps (creating the reservation @@ -101,13 +112,29 @@ * Block-aligned directio writes will use the same mechanism as buffered * writes. * + * Just prior to submitting the actual disk write requests, we convert + * the extents representing the range of the file actually being written + * (as opposed to extra pieces created for the cowextsize hint) to real + * extents. This will become important in the next step: + * + * D: --RRRRRRSSSRRRRRRRR--- + * C: ------UUrrUUU--------- + * * CoW remapping must be done after the data block write completes, * because we don't want to destroy the old data fork map until we're sure * the new block has been written. Since the new mappings are kept in a * separate fork, we can simply iterate these mappings to find the ones * that cover the file blocks that we just CoW'd. For each extent, simply * unmap the corresponding range in the data fork, map the new range into - * the data fork, and remove the extent from the CoW fork. + * the data fork, and remove the extent from the CoW fork. Because of + * the presence of the cowextsize hint, however, we must be careful + * only to remap the blocks that we've actually written out -- we must + * never remap delalloc reservations nor CoW staging blocks that have + * yet to be written. This corresponds exactly to the real extents in + * the CoW fork: + * + * D: --RRRRRRrrSRRRRRRRR--- + * C: ------UU--UUU--------- * * Since the remapping operation can be applied to an arbitrary file * range, we record the need for the remap step as a flag in the ioend @@ -296,103 +323,165 @@ xfs_reflink_reserve_cow( return 0; } -/* Allocate all CoW reservations covering a range of blocks in a file. */ -static int -__xfs_reflink_allocate_cow( - struct xfs_inode *ip, - xfs_fileoff_t *offset_fsb, - xfs_fileoff_t end_fsb) +/* Convert part of an unwritten CoW extent to a real one. */ +STATIC int +xfs_reflink_convert_cow_extent( + struct xfs_inode *ip, + struct xfs_bmbt_irec *imap, + xfs_fileoff_t offset_fsb, + xfs_filblks_t count_fsb, + struct xfs_defer_ops *dfops) { - struct xfs_mount *mp = ip->i_mount; - struct xfs_bmbt_irec imap; - struct xfs_defer_ops dfops; - struct xfs_trans *tp; - xfs_fsblock_t first_block; - int nimaps = 1, error; - bool shared; - - xfs_defer_init(&dfops, &first_block); + xfs_fsblock_t first_block; + int nimaps = 1; - error = xfs_trans_alloc(mp, &M_RES(mp)->tr_write, 0, 0, - XFS_TRANS_RESERVE, &tp); - if (error) - return error; + if (imap->br_state == XFS_EXT_NORM) + return 0; - xfs_ilock(ip, XFS_ILOCK_EXCL); + xfs_trim_extent(imap, offset_fsb, count_fsb); + trace_xfs_reflink_convert_cow(ip, imap); + if (imap->br_blockcount == 0) + return 0; + return xfs_bmapi_write(NULL, ip, imap->br_startoff, imap->br_blockcount, + XFS_BMAPI_COWFORK | XFS_BMAPI_CONVERT, &first_block, + 0, imap, &nimaps, dfops); +} - /* Read extent from the source file. */ - nimaps = 1; - error = xfs_bmapi_read(ip, *offset_fsb, end_fsb - *offset_fsb, - &imap, &nimaps, 0); - if (error) - goto out_unlock; - ASSERT(nimaps == 1); +/* Convert all of the unwritten CoW extents in a file's range to real ones. */ +int +xfs_reflink_convert_cow( + struct xfs_inode *ip, + xfs_off_t offset, + xfs_off_t count) +{ + struct xfs_bmbt_irec got; + struct xfs_defer_ops dfops; + struct xfs_mount *mp = ip->i_mount; + struct xfs_ifork *ifp = XFS_IFORK_PTR(ip, XFS_COW_FORK); + xfs_fileoff_t offset_fsb = XFS_B_TO_FSBT(mp, offset); + xfs_fileoff_t end_fsb = XFS_B_TO_FSB(mp, offset + count); + xfs_extnum_t idx; + bool found; + int error = 0; - error = xfs_reflink_reserve_cow(ip, &imap, &shared); - if (error) - goto out_trans_cancel; + xfs_ilock(ip, XFS_ILOCK_EXCL); - if (!shared) { - *offset_fsb = imap.br_startoff + imap.br_blockcount; - goto out_trans_cancel; + /* Convert all the extents to real from unwritten. */ + for (found = xfs_iext_lookup_extent(ip, ifp, offset_fsb, &idx, &got); + found && got.br_startoff < end_fsb; + found = xfs_iext_get_extent(ifp, ++idx, &got)) { + error = xfs_reflink_convert_cow_extent(ip, &got, offset_fsb, + end_fsb - offset_fsb, &dfops); + if (error) + break; } - xfs_trans_ijoin(tp, ip, 0); - error = xfs_bmapi_write(tp, ip, imap.br_startoff, imap.br_blockcount, - XFS_BMAPI_COWFORK, &first_block, - XFS_EXTENTADD_SPACE_RES(mp, XFS_DATA_FORK), - &imap, &nimaps, &dfops); - if (error) - goto out_trans_cancel; - - error = xfs_defer_finish(&tp, &dfops, NULL); - if (error) - goto out_trans_cancel; - - error = xfs_trans_commit(tp); - - *offset_fsb = imap.br_startoff + imap.br_blockcount; -out_unlock: + /* Finish up. */ xfs_iunlock(ip, XFS_ILOCK_EXCL); return error; -out_trans_cancel: - xfs_defer_cancel(&dfops); - xfs_trans_cancel(tp); - goto out_unlock; } -/* Allocate all CoW reservations covering a part of a file. */ +/* Allocate all CoW reservations covering a range of blocks in a file. */ int -xfs_reflink_allocate_cow_range( +xfs_reflink_allocate_cow( struct xfs_inode *ip, - xfs_off_t offset, - xfs_off_t count) + struct xfs_bmbt_irec *imap, + bool *shared, + uint *lockmode) { struct xfs_mount *mp = ip->i_mount; - xfs_fileoff_t offset_fsb = XFS_B_TO_FSBT(mp, offset); - xfs_fileoff_t end_fsb = XFS_B_TO_FSB(mp, offset + count); - int error; + xfs_fileoff_t offset_fsb = imap->br_startoff; + xfs_filblks_t count_fsb = imap->br_blockcount; + struct xfs_bmbt_irec got; + struct xfs_defer_ops dfops; + struct xfs_trans *tp = NULL; + xfs_fsblock_t first_block; + int nimaps, error = 0; + bool trimmed; + xfs_filblks_t resaligned; + xfs_extlen_t resblks = 0; + xfs_extnum_t idx; +retry: ASSERT(xfs_is_reflink_inode(ip)); - - trace_xfs_reflink_allocate_cow_range(ip, offset, count); + ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL | XFS_ILOCK_SHARED)); /* - * Make sure that the dquots are there. + * Even if the extent is not shared we might have a preallocation for + * it in the COW fork. If so use it. */ - error = xfs_qm_dqattach(ip, 0); - if (error) - return error; + if (xfs_iext_lookup_extent(ip, ip->i_cowfp, offset_fsb, &idx, &got) && + got.br_startoff <= offset_fsb) { + *shared = true; - while (offset_fsb < end_fsb) { - error = __xfs_reflink_allocate_cow(ip, &offset_fsb, end_fsb); - if (error) { - trace_xfs_reflink_allocate_cow_range_error(ip, error, - _RET_IP_); - break; + /* If we have a real allocation in the COW fork we're done. */ + if (!isnullstartblock(got.br_startblock)) { + xfs_trim_extent(&got, offset_fsb, count_fsb); + *imap = got; + goto convert; } + + xfs_trim_extent(imap, got.br_startoff, got.br_blockcount); + } else { + error = xfs_reflink_trim_around_shared(ip, imap, shared, &trimmed); + if (error || !*shared) + goto out; + } + + if (!tp) { + resaligned = xfs_aligned_fsb_count(imap->br_startoff, + imap->br_blockcount, xfs_get_cowextsz_hint(ip)); + resblks = XFS_DIOSTRAT_SPACE_RES(mp, resaligned); + + xfs_iunlock(ip, *lockmode); + error = xfs_trans_alloc(mp, &M_RES(mp)->tr_write, resblks, 0, 0, &tp); + *lockmode = XFS_ILOCK_EXCL; + xfs_ilock(ip, *lockmode); + + if (error) + return error; + + error = xfs_qm_dqattach_locked(ip, 0); + if (error) + goto out; + goto retry; } + error = xfs_trans_reserve_quota_nblks(tp, ip, resblks, 0, + XFS_QMOPT_RES_REGBLKS); + if (error) + goto out; + + xfs_trans_ijoin(tp, ip, 0); + + xfs_defer_init(&dfops, &first_block); + nimaps = 1; + + /* Allocate the entire reservation as unwritten blocks. */ + error = xfs_bmapi_write(tp, ip, imap->br_startoff, imap->br_blockcount, + XFS_BMAPI_COWFORK | XFS_BMAPI_PREALLOC, &first_block, + resblks, imap, &nimaps, &dfops); + if (error) + goto out_bmap_cancel; + + /* Finish up. */ + error = xfs_defer_finish(&tp, &dfops, NULL); + if (error) + goto out_bmap_cancel; + + error = xfs_trans_commit(tp); + if (error) + return error; +convert: + return xfs_reflink_convert_cow_extent(ip, imap, offset_fsb, count_fsb, + &dfops); +out_bmap_cancel: + xfs_defer_cancel(&dfops); + xfs_trans_unreserve_quota_nblks(tp, ip, (long)resblks, 0, + XFS_QMOPT_RES_REGBLKS); +out: + if (tp) + xfs_trans_cancel(tp); return error; } @@ -641,6 +730,16 @@ xfs_reflink_end_cow( ASSERT(!isnullstartblock(got.br_startblock)); + /* + * Don't remap unwritten extents; these are + * speculatively preallocated CoW extents that have been + * allocated but have not yet been involved in a write. + */ + if (got.br_state == XFS_EXT_UNWRITTEN) { + idx--; + goto next_extent; + } + /* Unmap the old blocks in the data fork. */ xfs_defer_init(&dfops, &firstfsb); rlen = del.br_blockcount; @@ -855,13 +954,14 @@ STATIC int xfs_reflink_update_dest( struct xfs_inode *dest, xfs_off_t newlen, - xfs_extlen_t cowextsize) + xfs_extlen_t cowextsize, + bool is_dedupe) { struct xfs_mount *mp = dest->i_mount; struct xfs_trans *tp; int error; - if (newlen <= i_size_read(VFS_I(dest)) && cowextsize == 0) + if (is_dedupe && newlen <= i_size_read(VFS_I(dest)) && cowextsize == 0) return 0; error = xfs_trans_alloc(mp, &M_RES(mp)->tr_ichange, 0, 0, 0, &tp); @@ -882,6 +982,10 @@ xfs_reflink_update_dest( dest->i_d.di_flags2 |= XFS_DIFLAG2_COWEXTSIZE; } + if (!is_dedupe) { + xfs_trans_ichgtime(tp, dest, + XFS_ICHGTIME_MOD | XFS_ICHGTIME_CHG); + } xfs_trans_log_inode(tp, dest, XFS_ILOG_CORE); error = xfs_trans_commit(tp); @@ -1195,7 +1299,8 @@ xfs_reflink_remap_range( !(dest->i_d.di_flags2 & XFS_DIFLAG2_COWEXTSIZE)) cowextsize = src->i_d.di_cowextsize; - ret = xfs_reflink_update_dest(dest, pos_out + len, cowextsize); + ret = xfs_reflink_update_dest(dest, pos_out + len, cowextsize, + is_dedupe); out_unlock: xfs_iunlock(src, XFS_MMAPLOCK_EXCL); diff --git a/fs/xfs/xfs_reflink.h b/fs/xfs/xfs_reflink.h index aa6a4d64bd35..33ac9b8db683 100644 --- a/fs/xfs/xfs_reflink.h +++ b/fs/xfs/xfs_reflink.h @@ -28,8 +28,10 @@ extern int xfs_reflink_trim_around_shared(struct xfs_inode *ip, extern int xfs_reflink_reserve_cow(struct xfs_inode *ip, struct xfs_bmbt_irec *imap, bool *shared); -extern int xfs_reflink_allocate_cow_range(struct xfs_inode *ip, - xfs_off_t offset, xfs_off_t count); +extern int xfs_reflink_allocate_cow(struct xfs_inode *ip, + struct xfs_bmbt_irec *imap, bool *shared, uint *lockmode); +extern int xfs_reflink_convert_cow(struct xfs_inode *ip, xfs_off_t offset, + xfs_off_t count); extern bool xfs_reflink_find_cow_mapping(struct xfs_inode *ip, xfs_off_t offset, struct xfs_bmbt_irec *imap); extern void xfs_reflink_trim_irec_to_next_cow(struct xfs_inode *ip, diff --git a/fs/xfs/xfs_rtalloc.c b/fs/xfs/xfs_rtalloc.c index 802bcc326d9f..c57aa7f18087 100644 --- a/fs/xfs/xfs_rtalloc.c +++ b/fs/xfs/xfs_rtalloc.c @@ -1093,7 +1093,6 @@ xfs_rtallocate_extent( xfs_extlen_t minlen, /* minimum length to allocate */ xfs_extlen_t maxlen, /* maximum length to allocate */ xfs_extlen_t *len, /* out: actual length allocated */ - xfs_alloctype_t type, /* allocation type XFS_ALLOCTYPE... */ int wasdel, /* was a delayed allocation extent */ xfs_extlen_t prod, /* extent product factor */ xfs_rtblock_t *rtblock) /* out: start block allocated */ @@ -1123,27 +1122,16 @@ xfs_rtallocate_extent( } } +retry: sumbp = NULL; - /* - * Allocate by size, or near another block, or exactly at some block. - */ - switch (type) { - case XFS_ALLOCTYPE_ANY_AG: + if (bno == 0) { error = xfs_rtallocate_extent_size(mp, tp, minlen, maxlen, len, &sumbp, &sb, prod, &r); - break; - case XFS_ALLOCTYPE_NEAR_BNO: + } else { error = xfs_rtallocate_extent_near(mp, tp, bno, minlen, maxlen, len, &sumbp, &sb, prod, &r); - break; - case XFS_ALLOCTYPE_THIS_BNO: - error = xfs_rtallocate_extent_exact(mp, tp, bno, minlen, maxlen, - len, &sumbp, &sb, prod, &r); - break; - default: - error = -EIO; - ASSERT(0); } + if (error) return error; @@ -1158,7 +1146,11 @@ xfs_rtallocate_extent( xfs_trans_mod_sb(tp, XFS_TRANS_SB_RES_FREXTENTS, -slen); else xfs_trans_mod_sb(tp, XFS_TRANS_SB_FREXTENTS, -slen); + } else if (prod > 1) { + prod = 1; + goto retry; } + *rtblock = r; return 0; } diff --git a/fs/xfs/xfs_rtalloc.h b/fs/xfs/xfs_rtalloc.h index 355dd9e1cb64..51dd3c726608 100644 --- a/fs/xfs/xfs_rtalloc.h +++ b/fs/xfs/xfs_rtalloc.h @@ -40,7 +40,6 @@ xfs_rtallocate_extent( xfs_extlen_t minlen, /* minimum length to allocate */ xfs_extlen_t maxlen, /* maximum length to allocate */ xfs_extlen_t *len, /* out: actual length allocated */ - xfs_alloctype_t type, /* allocation type XFS_ALLOCTYPE... */ int wasdel, /* was a delayed allocation extent */ xfs_extlen_t prod, /* extent product factor */ xfs_rtblock_t *rtblock); /* out: start block allocated */ @@ -122,7 +121,7 @@ int xfs_rtfree_range(struct xfs_mount *mp, struct xfs_trans *tp, #else -# define xfs_rtallocate_extent(t,b,min,max,l,a,f,p,rb) (ENOSYS) +# define xfs_rtallocate_extent(t,b,min,max,l,f,p,rb) (ENOSYS) # define xfs_rtfree_extent(t,b,l) (ENOSYS) # define xfs_rtpick_extent(m,t,l,rb) (ENOSYS) # define xfs_growfs_rt(mp,in) (ENOSYS) diff --git a/fs/xfs/xfs_super.c b/fs/xfs/xfs_super.c index eecbaac08eba..890862f2447c 100644 --- a/fs/xfs/xfs_super.c +++ b/fs/xfs/xfs_super.c @@ -1956,12 +1956,20 @@ xfs_init_workqueues(void) if (!xfs_alloc_wq) return -ENOMEM; + xfs_discard_wq = alloc_workqueue("xfsdiscard", WQ_UNBOUND, 0); + if (!xfs_discard_wq) + goto out_free_alloc_wq; + return 0; +out_free_alloc_wq: + destroy_workqueue(xfs_alloc_wq); + return -ENOMEM; } STATIC void xfs_destroy_workqueues(void) { + destroy_workqueue(xfs_discard_wq); destroy_workqueue(xfs_alloc_wq); } diff --git a/fs/xfs/xfs_super.h b/fs/xfs/xfs_super.h index b6418abd85ad..5f2f32408011 100644 --- a/fs/xfs/xfs_super.h +++ b/fs/xfs/xfs_super.h @@ -73,6 +73,8 @@ extern const struct quotactl_ops xfs_quotactl_operations; extern void xfs_reinit_percpu_counters(struct xfs_mount *mp); +extern struct workqueue_struct *xfs_discard_wq; + #define XFS_M(sb) ((struct xfs_mount *)((sb)->s_fs_info)) #endif /* __XFS_SUPER_H__ */ diff --git a/fs/xfs/xfs_sysfs.c b/fs/xfs/xfs_sysfs.c index de6195e38910..80ac15fb9638 100644 --- a/fs/xfs/xfs_sysfs.c +++ b/fs/xfs/xfs_sysfs.c @@ -93,7 +93,7 @@ to_mp(struct kobject *kobject) #ifdef DEBUG STATIC ssize_t -fail_writes_store( +drop_writes_store( struct kobject *kobject, const char *buf, size_t count) @@ -107,9 +107,9 @@ fail_writes_store( return ret; if (val == 1) - mp->m_fail_writes = true; + mp->m_drop_writes = true; else if (val == 0) - mp->m_fail_writes = false; + mp->m_drop_writes = false; else return -EINVAL; @@ -117,21 +117,21 @@ fail_writes_store( } STATIC ssize_t -fail_writes_show( +drop_writes_show( struct kobject *kobject, char *buf) { struct xfs_mount *mp = to_mp(kobject); - return snprintf(buf, PAGE_SIZE, "%d\n", mp->m_fail_writes ? 1 : 0); + return snprintf(buf, PAGE_SIZE, "%d\n", mp->m_drop_writes ? 1 : 0); } -XFS_SYSFS_ATTR_RW(fail_writes); +XFS_SYSFS_ATTR_RW(drop_writes); #endif /* DEBUG */ static struct attribute *xfs_mp_attrs[] = { #ifdef DEBUG - ATTR_LIST(fail_writes), + ATTR_LIST(drop_writes), #endif NULL, }; diff --git a/fs/xfs/xfs_trace.h b/fs/xfs/xfs_trace.h index 69c5bcd9a51b..fb7555e73a62 100644 --- a/fs/xfs/xfs_trace.h +++ b/fs/xfs/xfs_trace.h @@ -2245,7 +2245,6 @@ DEFINE_BTREE_CUR_EVENT(xfs_btree_overlapped_query_range); /* deferred ops */ struct xfs_defer_pending; -struct xfs_defer_intake; struct xfs_defer_ops; DECLARE_EVENT_CLASS(xfs_defer_class, @@ -3089,6 +3088,7 @@ DECLARE_EVENT_CLASS(xfs_inode_irec_class, __field(xfs_fileoff_t, lblk) __field(xfs_extlen_t, len) __field(xfs_fsblock_t, pblk) + __field(int, state) ), TP_fast_assign( __entry->dev = VFS_I(ip)->i_sb->s_dev; @@ -3096,13 +3096,15 @@ DECLARE_EVENT_CLASS(xfs_inode_irec_class, __entry->lblk = irec->br_startoff; __entry->len = irec->br_blockcount; __entry->pblk = irec->br_startblock; + __entry->state = irec->br_state; ), - TP_printk("dev %d:%d ino 0x%llx lblk 0x%llx len 0x%x pblk %llu", + TP_printk("dev %d:%d ino 0x%llx lblk 0x%llx len 0x%x pblk %llu st %d", MAJOR(__entry->dev), MINOR(__entry->dev), __entry->ino, __entry->lblk, __entry->len, - __entry->pblk) + __entry->pblk, + __entry->state) ); #define DEFINE_INODE_IREC_EVENT(name) \ DEFINE_EVENT(xfs_inode_irec_class, name, \ @@ -3242,11 +3244,11 @@ DEFINE_INODE_IREC_EVENT(xfs_reflink_trim_around_shared); DEFINE_INODE_IREC_EVENT(xfs_reflink_cow_alloc); DEFINE_INODE_IREC_EVENT(xfs_reflink_cow_found); DEFINE_INODE_IREC_EVENT(xfs_reflink_cow_enospc); +DEFINE_INODE_IREC_EVENT(xfs_reflink_convert_cow); DEFINE_RW_EVENT(xfs_reflink_reserve_cow); -DEFINE_RW_EVENT(xfs_reflink_allocate_cow_range); -DEFINE_INODE_IREC_EVENT(xfs_reflink_bounce_dio_write); +DEFINE_SIMPLE_IO_EVENT(xfs_reflink_bounce_dio_write); DEFINE_IOMAP_EVENT(xfs_reflink_find_cow_mapping); DEFINE_INODE_IREC_EVENT(xfs_reflink_trim_irec); @@ -3254,7 +3256,6 @@ DEFINE_SIMPLE_IO_EVENT(xfs_reflink_cancel_cow_range); DEFINE_SIMPLE_IO_EVENT(xfs_reflink_end_cow); DEFINE_INODE_IREC_EVENT(xfs_reflink_cow_remap); -DEFINE_INODE_ERROR_EVENT(xfs_reflink_allocate_cow_range_error); DEFINE_INODE_ERROR_EVENT(xfs_reflink_cancel_cow_range_error); DEFINE_INODE_ERROR_EVENT(xfs_reflink_end_cow_error); diff --git a/fs/xfs/xfs_trans.h b/fs/xfs/xfs_trans.h index 61b7fbdd3ebd..1646f659b60f 100644 --- a/fs/xfs/xfs_trans.h +++ b/fs/xfs/xfs_trans.h @@ -32,7 +32,6 @@ struct xfs_mount; struct xfs_trans; struct xfs_trans_res; struct xfs_dquot_acct; -struct xfs_busy_extent; struct xfs_rud_log_item; struct xfs_rui_log_item; struct xfs_btree_cur; diff --git a/include/kvm/arm_arch_timer.h b/include/kvm/arm_arch_timer.h index 5c970ce67949..fe797d6ef89d 100644 --- a/include/kvm/arm_arch_timer.h +++ b/include/kvm/arm_arch_timer.h @@ -23,20 +23,24 @@ #include <linux/hrtimer.h> #include <linux/workqueue.h> -struct arch_timer_kvm { +struct arch_timer_context { + /* Registers: control register, timer value */ + u32 cnt_ctl; + u64 cnt_cval; + + /* Timer IRQ */ + struct kvm_irq_level irq; + + /* Active IRQ state caching */ + bool active_cleared_last; + /* Virtual offset */ u64 cntvoff; }; struct arch_timer_cpu { - /* Registers: control register, timer value */ - u32 cntv_ctl; /* Saved/restored */ - u64 cntv_cval; /* Saved/restored */ - - /* - * Anything that is not used directly from assembly code goes - * here. - */ + struct arch_timer_context vtimer; + struct arch_timer_context ptimer; /* Background timer used when the guest is not running */ struct hrtimer timer; @@ -47,21 +51,15 @@ struct arch_timer_cpu { /* Background timer active */ bool armed; - /* Timer IRQ */ - struct kvm_irq_level irq; - - /* Active IRQ state caching */ - bool active_cleared_last; - /* Is the timer enabled */ bool enabled; }; int kvm_timer_hyp_init(void); int kvm_timer_enable(struct kvm_vcpu *vcpu); -void kvm_timer_init(struct kvm *kvm); int kvm_timer_vcpu_reset(struct kvm_vcpu *vcpu, - const struct kvm_irq_level *irq); + const struct kvm_irq_level *virt_irq, + const struct kvm_irq_level *phys_irq); void kvm_timer_vcpu_init(struct kvm_vcpu *vcpu); void kvm_timer_flush_hwstate(struct kvm_vcpu *vcpu); void kvm_timer_sync_hwstate(struct kvm_vcpu *vcpu); @@ -70,11 +68,16 @@ void kvm_timer_vcpu_terminate(struct kvm_vcpu *vcpu); u64 kvm_arm_timer_get_reg(struct kvm_vcpu *, u64 regid); int kvm_arm_timer_set_reg(struct kvm_vcpu *, u64 regid, u64 value); -bool kvm_timer_should_fire(struct kvm_vcpu *vcpu); +bool kvm_timer_should_fire(struct arch_timer_context *timer_ctx); void kvm_timer_schedule(struct kvm_vcpu *vcpu); void kvm_timer_unschedule(struct kvm_vcpu *vcpu); +u64 kvm_phys_timer_read(void); + void kvm_timer_vcpu_put(struct kvm_vcpu *vcpu); void kvm_timer_init_vhe(void); + +#define vcpu_vtimer(v) (&(v)->arch.timer_cpu.vtimer) +#define vcpu_ptimer(v) (&(v)->arch.timer_cpu.ptimer) #endif diff --git a/include/kvm/arm_vgic.h b/include/kvm/arm_vgic.h index 002f0922cd92..b72dd2ad5f44 100644 --- a/include/kvm/arm_vgic.h +++ b/include/kvm/arm_vgic.h @@ -71,6 +71,8 @@ struct vgic_global { /* GIC system register CPU interface */ struct static_key_false gicv3_cpuif; + + u32 ich_vtr_el2; }; extern struct vgic_global kvm_vgic_global_state; @@ -101,9 +103,10 @@ struct vgic_irq { */ u32 intid; /* Guest visible INTID */ - bool pending; bool line_level; /* Level only */ - bool soft_pending; /* Level only */ + bool pending_latch; /* The pending latch state used to calculate + * the pending state for both level + * and edge triggered IRQs. */ bool active; /* not used for LPIs */ bool enabled; bool hw; /* Tied to HW IRQ */ @@ -165,6 +168,8 @@ struct vgic_its { struct list_head collection_list; }; +struct vgic_state_iter; + struct vgic_dist { bool in_kernel; bool ready; @@ -212,6 +217,9 @@ struct vgic_dist { spinlock_t lpi_list_lock; struct list_head lpi_list_head; int lpi_list_count; + + /* used by vgic-debug */ + struct vgic_state_iter *iter; }; struct vgic_v2_cpu_if { @@ -269,6 +277,12 @@ struct vgic_cpu { u64 pendbaser; bool lpis_enabled; + + /* Cache guest priority bits */ + u32 num_pri_bits; + + /* Cache guest interrupt ID bits */ + u32 num_id_bits; }; extern struct static_key_false vgic_v2_cpuif_trap; diff --git a/include/linux/dax.h b/include/linux/dax.h index c1bd6ab5e974..1e77ff5818f1 100644 --- a/include/linux/dax.h +++ b/include/linux/dax.h @@ -37,9 +37,9 @@ static inline void *dax_radix_locked_entry(sector_t sector, unsigned long flags) } ssize_t dax_iomap_rw(struct kiocb *iocb, struct iov_iter *iter, - struct iomap_ops *ops); + const struct iomap_ops *ops); int dax_iomap_fault(struct vm_area_struct *vma, struct vm_fault *vmf, - struct iomap_ops *ops); + const struct iomap_ops *ops); int dax_delete_mapping_entry(struct address_space *mapping, pgoff_t index); int dax_invalidate_mapping_entry(struct address_space *mapping, pgoff_t index); int dax_invalidate_mapping_entry_sync(struct address_space *mapping, @@ -71,14 +71,14 @@ static inline unsigned int dax_radix_order(void *entry) return PMD_SHIFT - PAGE_SHIFT; return 0; } -int dax_iomap_pmd_fault(struct vm_fault *vmf, struct iomap_ops *ops); +int dax_iomap_pmd_fault(struct vm_fault *vmf, const struct iomap_ops *ops); #else static inline unsigned int dax_radix_order(void *entry) { return 0; } static inline int dax_iomap_pmd_fault(struct vm_fault *vmf, - struct iomap_ops *ops) + const struct iomap_ops *ops) { return VM_FAULT_FALLBACK; } diff --git a/include/linux/iomap.h b/include/linux/iomap.h index a4c94b86401e..891459caa278 100644 --- a/include/linux/iomap.h +++ b/include/linux/iomap.h @@ -72,17 +72,17 @@ struct iomap_ops { }; ssize_t iomap_file_buffered_write(struct kiocb *iocb, struct iov_iter *from, - struct iomap_ops *ops); + const struct iomap_ops *ops); int iomap_file_dirty(struct inode *inode, loff_t pos, loff_t len, - struct iomap_ops *ops); + const struct iomap_ops *ops); int iomap_zero_range(struct inode *inode, loff_t pos, loff_t len, - bool *did_zero, struct iomap_ops *ops); + bool *did_zero, const struct iomap_ops *ops); int iomap_truncate_page(struct inode *inode, loff_t pos, bool *did_zero, - struct iomap_ops *ops); + const struct iomap_ops *ops); int iomap_page_mkwrite(struct vm_area_struct *vma, struct vm_fault *vmf, - struct iomap_ops *ops); + const struct iomap_ops *ops); int iomap_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo, - loff_t start, loff_t len, struct iomap_ops *ops); + loff_t start, loff_t len, const struct iomap_ops *ops); /* * Flags for direct I/O ->end_io: @@ -92,6 +92,6 @@ int iomap_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo, typedef int (iomap_dio_end_io_t)(struct kiocb *iocb, ssize_t ret, unsigned flags); ssize_t iomap_dio_rw(struct kiocb *iocb, struct iov_iter *iter, - struct iomap_ops *ops, iomap_dio_end_io_t end_io); + const struct iomap_ops *ops, iomap_dio_end_io_t end_io); #endif /* LINUX_IOMAP_H */ diff --git a/include/linux/irqchip/arm-gic-v3.h b/include/linux/irqchip/arm-gic-v3.h index 725e86b506f3..672cfef72fc8 100644 --- a/include/linux/irqchip/arm-gic-v3.h +++ b/include/linux/irqchip/arm-gic-v3.h @@ -349,8 +349,30 @@ /* * CPU interface registers */ -#define ICC_CTLR_EL1_EOImode_drop_dir (0U << 1) -#define ICC_CTLR_EL1_EOImode_drop (1U << 1) +#define ICC_CTLR_EL1_EOImode_SHIFT (1) +#define ICC_CTLR_EL1_EOImode_drop_dir (0U << ICC_CTLR_EL1_EOImode_SHIFT) +#define ICC_CTLR_EL1_EOImode_drop (1U << ICC_CTLR_EL1_EOImode_SHIFT) +#define ICC_CTLR_EL1_EOImode_MASK (1 << ICC_CTLR_EL1_EOImode_SHIFT) +#define ICC_CTLR_EL1_CBPR_SHIFT 0 +#define ICC_CTLR_EL1_CBPR_MASK (1 << ICC_CTLR_EL1_CBPR_SHIFT) +#define ICC_CTLR_EL1_PRI_BITS_SHIFT 8 +#define ICC_CTLR_EL1_PRI_BITS_MASK (0x7 << ICC_CTLR_EL1_PRI_BITS_SHIFT) +#define ICC_CTLR_EL1_ID_BITS_SHIFT 11 +#define ICC_CTLR_EL1_ID_BITS_MASK (0x7 << ICC_CTLR_EL1_ID_BITS_SHIFT) +#define ICC_CTLR_EL1_SEIS_SHIFT 14 +#define ICC_CTLR_EL1_SEIS_MASK (0x1 << ICC_CTLR_EL1_SEIS_SHIFT) +#define ICC_CTLR_EL1_A3V_SHIFT 15 +#define ICC_CTLR_EL1_A3V_MASK (0x1 << ICC_CTLR_EL1_A3V_SHIFT) +#define ICC_PMR_EL1_SHIFT 0 +#define ICC_PMR_EL1_MASK (0xff << ICC_PMR_EL1_SHIFT) +#define ICC_BPR0_EL1_SHIFT 0 +#define ICC_BPR0_EL1_MASK (0x7 << ICC_BPR0_EL1_SHIFT) +#define ICC_BPR1_EL1_SHIFT 0 +#define ICC_BPR1_EL1_MASK (0x7 << ICC_BPR1_EL1_SHIFT) +#define ICC_IGRPEN0_EL1_SHIFT 0 +#define ICC_IGRPEN0_EL1_MASK (1 << ICC_IGRPEN0_EL1_SHIFT) +#define ICC_IGRPEN1_EL1_SHIFT 0 +#define ICC_IGRPEN1_EL1_MASK (1 << ICC_IGRPEN1_EL1_SHIFT) #define ICC_SRE_EL1_SRE (1U << 0) /* @@ -379,14 +401,29 @@ #define ICH_HCR_EN (1 << 0) #define ICH_HCR_UIE (1 << 1) -#define ICH_VMCR_CTLR_SHIFT 0 -#define ICH_VMCR_CTLR_MASK (0x21f << ICH_VMCR_CTLR_SHIFT) +#define ICH_VMCR_CBPR_SHIFT 4 +#define ICH_VMCR_CBPR_MASK (1 << ICH_VMCR_CBPR_SHIFT) +#define ICH_VMCR_EOIM_SHIFT 9 +#define ICH_VMCR_EOIM_MASK (1 << ICH_VMCR_EOIM_SHIFT) #define ICH_VMCR_BPR1_SHIFT 18 #define ICH_VMCR_BPR1_MASK (7 << ICH_VMCR_BPR1_SHIFT) #define ICH_VMCR_BPR0_SHIFT 21 #define ICH_VMCR_BPR0_MASK (7 << ICH_VMCR_BPR0_SHIFT) #define ICH_VMCR_PMR_SHIFT 24 #define ICH_VMCR_PMR_MASK (0xffUL << ICH_VMCR_PMR_SHIFT) +#define ICH_VMCR_ENG0_SHIFT 0 +#define ICH_VMCR_ENG0_MASK (1 << ICH_VMCR_ENG0_SHIFT) +#define ICH_VMCR_ENG1_SHIFT 1 +#define ICH_VMCR_ENG1_MASK (1 << ICH_VMCR_ENG1_SHIFT) + +#define ICH_VTR_PRI_BITS_SHIFT 29 +#define ICH_VTR_PRI_BITS_MASK (7 << ICH_VTR_PRI_BITS_SHIFT) +#define ICH_VTR_ID_BITS_SHIFT 23 +#define ICH_VTR_ID_BITS_MASK (7 << ICH_VTR_ID_BITS_SHIFT) +#define ICH_VTR_SEIS_SHIFT 22 +#define ICH_VTR_SEIS_MASK (1 << ICH_VTR_SEIS_SHIFT) +#define ICH_VTR_A3V_SHIFT 21 +#define ICH_VTR_A3V_MASK (1 << ICH_VTR_A3V_SHIFT) #define ICC_IAR1_EL1_SPURIOUS 0x3ff diff --git a/include/linux/kvm_host.h b/include/linux/kvm_host.h index 1c5190dab2c1..8d69d5150748 100644 --- a/include/linux/kvm_host.h +++ b/include/linux/kvm_host.h @@ -45,7 +45,6 @@ * include/linux/kvm_h. */ #define KVM_MEMSLOT_INVALID (1UL << 16) -#define KVM_MEMSLOT_INCOHERENT (1UL << 17) /* Two fragments for cross MMIO pages. */ #define KVM_MAX_MMIO_FRAGMENTS 2 @@ -222,7 +221,6 @@ struct kvm_vcpu { struct mutex mutex; struct kvm_run *run; - int fpu_active; int guest_fpu_loaded, guest_xcr0_loaded; struct swait_queue_head wq; struct pid *pid; @@ -642,18 +640,18 @@ int kvm_read_guest_page(struct kvm *kvm, gfn_t gfn, void *data, int offset, int kvm_read_guest_atomic(struct kvm *kvm, gpa_t gpa, void *data, unsigned long len); int kvm_read_guest(struct kvm *kvm, gpa_t gpa, void *data, unsigned long len); -int kvm_read_guest_cached(struct kvm *kvm, struct gfn_to_hva_cache *ghc, - void *data, unsigned long len); +int kvm_vcpu_read_guest_cached(struct kvm_vcpu *vcpu, struct gfn_to_hva_cache *ghc, + void *data, unsigned long len); int kvm_write_guest_page(struct kvm *kvm, gfn_t gfn, const void *data, int offset, int len); int kvm_write_guest(struct kvm *kvm, gpa_t gpa, const void *data, unsigned long len); -int kvm_write_guest_cached(struct kvm *kvm, struct gfn_to_hva_cache *ghc, - void *data, unsigned long len); -int kvm_write_guest_offset_cached(struct kvm *kvm, struct gfn_to_hva_cache *ghc, - void *data, int offset, unsigned long len); -int kvm_gfn_to_hva_cache_init(struct kvm *kvm, struct gfn_to_hva_cache *ghc, - gpa_t gpa, unsigned long len); +int kvm_vcpu_write_guest_cached(struct kvm_vcpu *v, struct gfn_to_hva_cache *ghc, + void *data, unsigned long len); +int kvm_vcpu_write_guest_offset_cached(struct kvm_vcpu *v, struct gfn_to_hva_cache *ghc, + void *data, int offset, unsigned long len); +int kvm_vcpu_gfn_to_hva_cache_init(struct kvm_vcpu *v, struct gfn_to_hva_cache *ghc, + gpa_t gpa, unsigned long len); int kvm_clear_guest_page(struct kvm *kvm, gfn_t gfn, int offset, int len); int kvm_clear_guest(struct kvm *kvm, gpa_t gpa, unsigned long len); struct kvm_memory_slot *gfn_to_memslot(struct kvm *kvm, gfn_t gfn); diff --git a/include/linux/module.h b/include/linux/module.h index f4f542ed3d92..0297c5cd7cdf 100644 --- a/include/linux/module.h +++ b/include/linux/module.h @@ -126,13 +126,13 @@ extern void cleanup_module(void); /* Each module must use one module_init(). */ #define module_init(initfn) \ - static inline initcall_t __inittest(void) \ + static inline initcall_t __maybe_unused __inittest(void) \ { return initfn; } \ int init_module(void) __attribute__((alias(#initfn))); /* This is only required if you want to be unloadable. */ #define module_exit(exitfn) \ - static inline exitcall_t __exittest(void) \ + static inline exitcall_t __maybe_unused __exittest(void) \ { return exitfn; } \ void cleanup_module(void) __attribute__((alias(#exitfn))); @@ -281,8 +281,6 @@ enum module_state { MODULE_STATE_UNFORMED, /* Still setting it up. */ }; -struct module; - struct mod_tree_node { struct module *mod; struct latch_tree_node node; diff --git a/include/linux/of_device.h b/include/linux/of_device.h index e9afbcc8de12..c12dace043f3 100644 --- a/include/linux/of_device.h +++ b/include/linux/of_device.h @@ -13,7 +13,6 @@ struct device; #ifdef CONFIG_OF extern const struct of_device_id *of_match_device( const struct of_device_id *matches, const struct device *dev); -extern void of_device_make_bus_id(struct device *dev); /** * of_driver_match_device - Tell if a driver's of_match_table matches a device. diff --git a/include/linux/of_graph.h b/include/linux/of_graph.h index bb3a5a2cd570..abdb02eaef06 100644 --- a/include/linux/of_graph.h +++ b/include/linux/of_graph.h @@ -51,6 +51,8 @@ struct device_node *of_graph_get_endpoint_by_regs( struct device_node *of_graph_get_remote_port_parent( const struct device_node *node); struct device_node *of_graph_get_remote_port(const struct device_node *node); +struct device_node *of_graph_get_remote_node(const struct device_node *node, + u32 port, u32 endpoint); #else static inline int of_graph_parse_endpoint(const struct device_node *node, @@ -89,6 +91,12 @@ static inline struct device_node *of_graph_get_remote_port( { return NULL; } +static inline struct device_node *of_graph_get_remote_node( + const struct device_node *node, + u32 port, u32 endpoint) +{ + return NULL; +} #endif /* CONFIG_OF */ diff --git a/include/linux/pm.h b/include/linux/pm.h index f926af41e122..a0894bc52bb4 100644 --- a/include/linux/pm.h +++ b/include/linux/pm.h @@ -64,24 +64,7 @@ typedef struct pm_message { } pm_message_t; /** - * struct dev_pm_ops - device PM callbacks - * - * Several device power state transitions are externally visible, affecting - * the state of pending I/O queues and (for drivers that touch hardware) - * interrupts, wakeups, DMA, and other hardware state. There may also be - * internal transitions to various low-power modes which are transparent - * to the rest of the driver stack (such as a driver that's ON gating off - * clocks which are not in active use). - * - * The externally visible transitions are handled with the help of callbacks - * included in this structure in such a way that two levels of callbacks are - * involved. First, the PM core executes callbacks provided by PM domains, - * device types, classes and bus types. They are the subsystem-level callbacks - * supposed to execute callbacks provided by device drivers, although they may - * choose not to do that. If the driver callbacks are executed, they have to - * collaborate with the subsystem-level callbacks to achieve the goals - * appropriate for the given system transition, given transition phase and the - * subsystem the device belongs to. + * struct dev_pm_ops - device PM callbacks. * * @prepare: The principal role of this callback is to prevent new children of * the device from being registered after it has returned (the driver's @@ -240,34 +223,6 @@ typedef struct pm_message { * driver's interrupt handler, which is guaranteed not to run while * @restore_noirq() is being executed. Analogous to @resume_noirq(). * - * All of the above callbacks, except for @complete(), return error codes. - * However, the error codes returned by the resume operations, @resume(), - * @thaw(), @restore(), @resume_noirq(), @thaw_noirq(), and @restore_noirq(), do - * not cause the PM core to abort the resume transition during which they are - * returned. The error codes returned in those cases are only printed by the PM - * core to the system logs for debugging purposes. Still, it is recommended - * that drivers only return error codes from their resume methods in case of an - * unrecoverable failure (i.e. when the device being handled refuses to resume - * and becomes unusable) to allow us to modify the PM core in the future, so - * that it can avoid attempting to handle devices that failed to resume and - * their children. - * - * It is allowed to unregister devices while the above callbacks are being - * executed. However, a callback routine must NOT try to unregister the device - * it was called for, although it may unregister children of that device (for - * example, if it detects that a child was unplugged while the system was - * asleep). - * - * Refer to Documentation/power/admin-guide/devices.rst for more information about the role - * of the above callbacks in the system suspend process. - * - * There also are callbacks related to runtime power management of devices. - * Again, these callbacks are executed by the PM core only for subsystems - * (PM domains, device types, classes and bus types) and the subsystem-level - * callbacks are supposed to invoke the driver callbacks. Moreover, the exact - * actions to be performed by a device driver's callbacks generally depend on - * the platform and subsystem the device belongs to. - * * @runtime_suspend: Prepare the device for a condition in which it won't be * able to communicate with the CPU(s) and RAM due to power management. * This need not mean that the device should be put into a low-power state. @@ -287,11 +242,51 @@ typedef struct pm_message { * Check these conditions, and return 0 if it's appropriate to let the PM * core queue a suspend request for the device. * - * Refer to Documentation/power/runtime_pm.txt for more information about the - * role of the above callbacks in device runtime power management. + * Several device power state transitions are externally visible, affecting + * the state of pending I/O queues and (for drivers that touch hardware) + * interrupts, wakeups, DMA, and other hardware state. There may also be + * internal transitions to various low-power modes which are transparent + * to the rest of the driver stack (such as a driver that's ON gating off + * clocks which are not in active use). * + * The externally visible transitions are handled with the help of callbacks + * included in this structure in such a way that, typically, two levels of + * callbacks are involved. First, the PM core executes callbacks provided by PM + * domains, device types, classes and bus types. They are the subsystem-level + * callbacks expected to execute callbacks provided by device drivers, although + * they may choose not to do that. If the driver callbacks are executed, they + * have to collaborate with the subsystem-level callbacks to achieve the goals + * appropriate for the given system transition, given transition phase and the + * subsystem the device belongs to. + * + * All of the above callbacks, except for @complete(), return error codes. + * However, the error codes returned by @resume(), @thaw(), @restore(), + * @resume_noirq(), @thaw_noirq(), and @restore_noirq(), do not cause the PM + * core to abort the resume transition during which they are returned. The + * error codes returned in those cases are only printed to the system logs for + * debugging purposes. Still, it is recommended that drivers only return error + * codes from their resume methods in case of an unrecoverable failure (i.e. + * when the device being handled refuses to resume and becomes unusable) to + * allow the PM core to be modified in the future, so that it can avoid + * attempting to handle devices that failed to resume and their children. + * + * It is allowed to unregister devices while the above callbacks are being + * executed. However, a callback routine MUST NOT try to unregister the device + * it was called for, although it may unregister children of that device (for + * example, if it detects that a child was unplugged while the system was + * asleep). + * + * There also are callbacks related to runtime power management of devices. + * Again, as a rule these callbacks are executed by the PM core for subsystems + * (PM domains, device types, classes and bus types) and the subsystem-level + * callbacks are expected to invoke the driver callbacks. Moreover, the exact + * actions to be performed by a device driver's callbacks generally depend on + * the platform and subsystem the device belongs to. + * + * Refer to Documentation/power/runtime_pm.txt for more information about the + * role of the @runtime_suspend(), @runtime_resume() and @runtime_idle() + * callbacks in device runtime power management. */ - struct dev_pm_ops { int (*prepare)(struct device *dev); void (*complete)(struct device *dev); @@ -391,7 +386,7 @@ const struct dev_pm_ops name = { \ SET_RUNTIME_PM_OPS(suspend_fn, resume_fn, idle_fn) \ } -/** +/* * PM_EVENT_ messages * * The following PM_EVENT_ messages are defined for the internal use of the PM @@ -487,7 +482,7 @@ const struct dev_pm_ops name = { \ #define PMSG_IS_AUTO(msg) (((msg).event & PM_EVENT_AUTO) != 0) -/** +/* * Device run-time power management status. * * These status labels are used internally by the PM core to indicate the @@ -517,7 +512,7 @@ enum rpm_status { RPM_SUSPENDING, }; -/** +/* * Device run-time power management request types. * * RPM_REQ_NONE Do nothing. @@ -616,15 +611,18 @@ extern void update_pm_runtime_accounting(struct device *dev); extern int dev_pm_get_subsys_data(struct device *dev); extern void dev_pm_put_subsys_data(struct device *dev); -/* - * Power domains provide callbacks that are executed during system suspend, - * hibernation, system resume and during runtime PM transitions along with - * subsystem-level and driver-level callbacks. +/** + * struct dev_pm_domain - power management domain representation. * + * @ops: Power management operations associated with this domain. * @detach: Called when removing a device from the domain. * @activate: Called before executing probe routines for bus types and drivers. * @sync: Called after successful driver probe. * @dismiss: Called after unsuccessful driver probe and after driver removal. + * + * Power domains provide callbacks that are executed during system suspend, + * hibernation, system resume and during runtime PM transitions instead of + * subsystem-level and driver-level callbacks. */ struct dev_pm_domain { struct dev_pm_ops ops; diff --git a/include/linux/printk.h b/include/linux/printk.h index 3472cc6b7a60..571257e0f53d 100644 --- a/include/linux/printk.h +++ b/include/linux/printk.h @@ -147,17 +147,11 @@ void early_printk(const char *s, ...) { } #endif #ifdef CONFIG_PRINTK_NMI -extern void printk_nmi_init(void); extern void printk_nmi_enter(void); extern void printk_nmi_exit(void); -extern void printk_nmi_flush(void); -extern void printk_nmi_flush_on_panic(void); #else -static inline void printk_nmi_init(void) { } static inline void printk_nmi_enter(void) { } static inline void printk_nmi_exit(void) { } -static inline void printk_nmi_flush(void) { } -static inline void printk_nmi_flush_on_panic(void) { } #endif /* PRINTK_NMI */ #ifdef CONFIG_PRINTK @@ -209,6 +203,9 @@ void __init setup_log_buf(int early); __printf(1, 2) void dump_stack_set_arch_desc(const char *fmt, ...); void dump_stack_print_info(const char *log_lvl); void show_regs_print_info(const char *log_lvl); +extern void printk_safe_init(void); +extern void printk_safe_flush(void); +extern void printk_safe_flush_on_panic(void); #else static inline __printf(1, 0) int vprintk(const char *s, va_list args) @@ -268,6 +265,18 @@ static inline void dump_stack_print_info(const char *log_lvl) static inline void show_regs_print_info(const char *log_lvl) { } + +static inline void printk_safe_init(void) +{ +} + +static inline void printk_safe_flush(void) +{ +} + +static inline void printk_safe_flush_on_panic(void) +{ +} #endif extern asmlinkage void dump_stack(void) __cold; diff --git a/include/uapi/linux/kvm.h b/include/uapi/linux/kvm.h index e0035808c814..f51d5082a377 100644 --- a/include/uapi/linux/kvm.h +++ b/include/uapi/linux/kvm.h @@ -218,7 +218,8 @@ struct kvm_hyperv_exit { struct kvm_run { /* in */ __u8 request_interrupt_window; - __u8 padding1[7]; + __u8 immediate_exit; + __u8 padding1[6]; /* out */ __u32 exit_reason; @@ -685,6 +686,13 @@ struct kvm_ppc_smmu_info { struct kvm_ppc_one_seg_page_size sps[KVM_PPC_PAGE_SIZES_MAX_SZ]; }; +/* for KVM_PPC_RESIZE_HPT_{PREPARE,COMMIT} */ +struct kvm_ppc_resize_hpt { + __u64 flags; + __u32 shift; + __u32 pad; +}; + #define KVMIO 0xAE /* machine type bits, to be used as argument to KVM_CREATE_VM */ @@ -871,8 +879,10 @@ struct kvm_ppc_smmu_info { #define KVM_CAP_S390_USER_INSTR0 130 #define KVM_CAP_MSI_DEVID 131 #define KVM_CAP_PPC_HTM 132 +#define KVM_CAP_SPAPR_RESIZE_HPT 133 #define KVM_CAP_PPC_MMU_RADIX 134 #define KVM_CAP_PPC_MMU_HASH_V3 135 +#define KVM_CAP_IMMEDIATE_EXIT 136 #ifdef KVM_CAP_IRQ_ROUTING @@ -1189,6 +1199,9 @@ struct kvm_s390_ucas_mapping { #define KVM_ARM_SET_DEVICE_ADDR _IOW(KVMIO, 0xab, struct kvm_arm_device_addr) /* Available with KVM_CAP_PPC_RTAS */ #define KVM_PPC_RTAS_DEFINE_TOKEN _IOW(KVMIO, 0xac, struct kvm_rtas_token_args) +/* Available with KVM_CAP_SPAPR_RESIZE_HPT */ +#define KVM_PPC_RESIZE_HPT_PREPARE _IOR(KVMIO, 0xad, struct kvm_ppc_resize_hpt) +#define KVM_PPC_RESIZE_HPT_COMMIT _IOR(KVMIO, 0xae, struct kvm_ppc_resize_hpt) /* Available with KVM_CAP_PPC_RADIX_MMU or KVM_CAP_PPC_HASH_MMU_V3 */ #define KVM_PPC_CONFIGURE_V3_MMU _IOW(KVMIO, 0xaf, struct kvm_ppc_mmuv3_cfg) /* Available with KVM_CAP_PPC_RADIX_MMU */ diff --git a/include/uapi/linux/kvm_para.h b/include/uapi/linux/kvm_para.h index bf6cd7d5cac2..fed506aeff62 100644 --- a/include/uapi/linux/kvm_para.h +++ b/include/uapi/linux/kvm_para.h @@ -14,6 +14,7 @@ #define KVM_EFAULT EFAULT #define KVM_E2BIG E2BIG #define KVM_EPERM EPERM +#define KVM_EOPNOTSUPP 95 #define KVM_HC_VAPIC_POLL_IRQ 1 #define KVM_HC_MMU_OP 2 @@ -23,6 +24,7 @@ #define KVM_HC_MIPS_GET_CLOCK_FREQ 6 #define KVM_HC_MIPS_EXIT_VM 7 #define KVM_HC_MIPS_CONSOLE_OUTPUT 8 +#define KVM_HC_CLOCK_PAIRING 9 /* * hypercalls use architecture specific diff --git a/init/Kconfig b/init/Kconfig index 22f437ff65e5..8c39615165b7 100644 --- a/init/Kconfig +++ b/init/Kconfig @@ -861,17 +861,19 @@ config LOG_CPU_MAX_BUF_SHIFT 13 => 8 KB for each CPU 12 => 4 KB for each CPU -config NMI_LOG_BUF_SHIFT - int "Temporary per-CPU NMI log buffer size (12 => 4KB, 13 => 8KB)" +config PRINTK_SAFE_LOG_BUF_SHIFT + int "Temporary per-CPU printk log buffer size (12 => 4KB, 13 => 8KB)" range 10 21 default 13 - depends on PRINTK_NMI + depends on PRINTK help - Select the size of a per-CPU buffer where NMI messages are temporary - stored. They are copied to the main log buffer in a safe context - to avoid a deadlock. The value defines the size as a power of 2. + Select the size of an alternate printk per-CPU buffer where messages + printed from usafe contexts are temporary stored. One example would + be NMI messages, another one - printk recursion. The messages are + copied to the main log buffer in a safe context to avoid a deadlock. + The value defines the size as a power of 2. - NMI messages are rare and limited. The largest one is when + Those messages are rare and limited. The largest one is when a backtrace is printed. It usually fits into 4KB. Select 8KB if you want to be on the safe side. diff --git a/init/main.c b/init/main.c index c8a00f0f10ff..24ea48745061 100644 --- a/init/main.c +++ b/init/main.c @@ -581,7 +581,7 @@ asmlinkage __visible void __init start_kernel(void) timekeeping_init(); time_init(); sched_clock_postinit(); - printk_nmi_init(); + printk_safe_init(); perf_event_init(); profile_init(); call_function_init(); diff --git a/kernel/kexec_core.c b/kernel/kexec_core.c index a01974e1bf6b..bfe62d5b3872 100644 --- a/kernel/kexec_core.c +++ b/kernel/kexec_core.c @@ -916,7 +916,7 @@ void crash_kexec(struct pt_regs *regs) old_cpu = atomic_cmpxchg(&panic_cpu, PANIC_CPU_INVALID, this_cpu); if (old_cpu == PANIC_CPU_INVALID) { /* This is the 1st CPU which comes here, so go ahead. */ - printk_nmi_flush_on_panic(); + printk_safe_flush_on_panic(); __crash_kexec(regs); /* diff --git a/kernel/module.c b/kernel/module.c index a3889169a3ae..7eba6dea4f41 100644 --- a/kernel/module.c +++ b/kernel/module.c @@ -2811,6 +2811,8 @@ static int check_modinfo_livepatch(struct module *mod, struct load_info *info) if (get_modinfo(info, "livepatch")) { mod->klp = true; add_taint_module(mod, TAINT_LIVEPATCH, LOCKDEP_STILL_OK); + pr_notice_once("%s: tainting kernel with TAINT_LIVEPATCH\n", + mod->name); } return 0; @@ -3723,6 +3725,7 @@ static int load_module(struct load_info *info, const char __user *uargs, mod_sysfs_teardown(mod); coming_cleanup: mod->state = MODULE_STATE_GOING; + destroy_params(mod->kp, mod->num_kp); blocking_notifier_call_chain(&module_notify_list, MODULE_STATE_GOING, mod); klp_module_going(mod); @@ -4169,22 +4172,23 @@ const struct exception_table_entry *search_module_extables(unsigned long addr) struct module *mod; preempt_disable(); - list_for_each_entry_rcu(mod, &modules, list) { - if (mod->state == MODULE_STATE_UNFORMED) - continue; - if (mod->num_exentries == 0) - continue; + mod = __module_address(addr); + if (!mod) + goto out; - e = search_extable(mod->extable, - mod->extable + mod->num_exentries - 1, - addr); - if (e) - break; - } + if (!mod->num_exentries) + goto out; + + e = search_extable(mod->extable, + mod->extable + mod->num_exentries - 1, + addr); +out: preempt_enable(); - /* Now, if we found one, we are running inside it now, hence - we cannot unload the module, hence no refcnt needed. */ + /* + * Now, if we found one, we are running inside it now, hence + * we cannot unload the module, hence no refcnt needed. + */ return e; } diff --git a/kernel/panic.c b/kernel/panic.c index 08aa88dde7de..b95959733ce0 100644 --- a/kernel/panic.c +++ b/kernel/panic.c @@ -188,7 +188,7 @@ void panic(const char *fmt, ...) * Bypass the panic_cpu check and call __crash_kexec directly. */ if (!_crash_kexec_post_notifiers) { - printk_nmi_flush_on_panic(); + printk_safe_flush_on_panic(); __crash_kexec(NULL); /* @@ -213,7 +213,7 @@ void panic(const char *fmt, ...) atomic_notifier_call_chain(&panic_notifier_list, 0, buf); /* Call flush even twice. It tries harder with a single online CPU */ - printk_nmi_flush_on_panic(); + printk_safe_flush_on_panic(); kmsg_dump(KMSG_DUMP_PANIC); /* diff --git a/kernel/printk/Makefile b/kernel/printk/Makefile index abb0042a427b..4a2ffc39eb95 100644 --- a/kernel/printk/Makefile +++ b/kernel/printk/Makefile @@ -1,3 +1,3 @@ obj-y = printk.o -obj-$(CONFIG_PRINTK_NMI) += nmi.o +obj-$(CONFIG_PRINTK) += printk_safe.o obj-$(CONFIG_A11Y_BRAILLE_CONSOLE) += braille.o diff --git a/kernel/printk/internal.h b/kernel/printk/internal.h index 7fd2838fa417..1db044f808b7 100644 --- a/kernel/printk/internal.h +++ b/kernel/printk/internal.h @@ -16,42 +16,55 @@ */ #include <linux/percpu.h> -typedef __printf(1, 0) int (*printk_func_t)(const char *fmt, va_list args); +#ifdef CONFIG_PRINTK -int __printf(1, 0) vprintk_default(const char *fmt, va_list args); - -#ifdef CONFIG_PRINTK_NMI +#define PRINTK_SAFE_CONTEXT_MASK 0x7fffffff +#define PRINTK_NMI_CONTEXT_MASK 0x80000000 extern raw_spinlock_t logbuf_lock; +__printf(1, 0) int vprintk_default(const char *fmt, va_list args); +__printf(1, 0) int vprintk_func(const char *fmt, va_list args); +void __printk_safe_enter(void); +void __printk_safe_exit(void); + +#define printk_safe_enter_irqsave(flags) \ + do { \ + local_irq_save(flags); \ + __printk_safe_enter(); \ + } while (0) + +#define printk_safe_exit_irqrestore(flags) \ + do { \ + __printk_safe_exit(); \ + local_irq_restore(flags); \ + } while (0) + +#define printk_safe_enter_irq() \ + do { \ + local_irq_disable(); \ + __printk_safe_enter(); \ + } while (0) + +#define printk_safe_exit_irq() \ + do { \ + __printk_safe_exit(); \ + local_irq_enable(); \ + } while (0) + +#else + +__printf(1, 0) int vprintk_func(const char *fmt, va_list args) { return 0; } + /* - * printk() could not take logbuf_lock in NMI context. Instead, - * it temporary stores the strings into a per-CPU buffer. - * The alternative implementation is chosen transparently - * via per-CPU variable. + * In !PRINTK builds we still export logbuf_lock spin_lock, console_sem + * semaphore and some of console functions (console_unlock()/etc.), so + * printk-safe must preserve the existing local IRQ guarantees. */ -DECLARE_PER_CPU(printk_func_t, printk_func); -static inline __printf(1, 0) int vprintk_func(const char *fmt, va_list args) -{ - return this_cpu_read(printk_func)(fmt, args); -} - -extern atomic_t nmi_message_lost; -static inline int get_nmi_message_lost(void) -{ - return atomic_xchg(&nmi_message_lost, 0); -} - -#else /* CONFIG_PRINTK_NMI */ - -static inline __printf(1, 0) int vprintk_func(const char *fmt, va_list args) -{ - return vprintk_default(fmt, args); -} - -static inline int get_nmi_message_lost(void) -{ - return 0; -} - -#endif /* CONFIG_PRINTK_NMI */ +#define printk_safe_enter_irqsave(flags) local_irq_save(flags) +#define printk_safe_exit_irqrestore(flags) local_irq_restore(flags) + +#define printk_safe_enter_irq() local_irq_disable() +#define printk_safe_exit_irq() local_irq_enable() + +#endif /* CONFIG_PRINTK */ diff --git a/kernel/printk/printk.c b/kernel/printk/printk.c index 4ba3d34938c0..34da86e73d00 100644 --- a/kernel/printk/printk.c +++ b/kernel/printk/printk.c @@ -213,17 +213,36 @@ static int nr_ext_console_drivers; static int __down_trylock_console_sem(unsigned long ip) { - if (down_trylock(&console_sem)) + int lock_failed; + unsigned long flags; + + /* + * Here and in __up_console_sem() we need to be in safe mode, + * because spindump/WARN/etc from under console ->lock will + * deadlock in printk()->down_trylock_console_sem() otherwise. + */ + printk_safe_enter_irqsave(flags); + lock_failed = down_trylock(&console_sem); + printk_safe_exit_irqrestore(flags); + + if (lock_failed) return 1; mutex_acquire(&console_lock_dep_map, 0, 1, ip); return 0; } #define down_trylock_console_sem() __down_trylock_console_sem(_RET_IP_) -#define up_console_sem() do { \ - mutex_release(&console_lock_dep_map, 1, _RET_IP_);\ - up(&console_sem);\ -} while (0) +static void __up_console_sem(unsigned long ip) +{ + unsigned long flags; + + mutex_release(&console_lock_dep_map, 1, ip); + + printk_safe_enter_irqsave(flags); + up(&console_sem); + printk_safe_exit_irqrestore(flags); +} +#define up_console_sem() __up_console_sem(_RET_IP_) /* * This is used for debugging the mess that is the VT code by @@ -351,6 +370,34 @@ __packed __aligned(4) */ DEFINE_RAW_SPINLOCK(logbuf_lock); +/* + * Helper macros to lock/unlock logbuf_lock and switch between + * printk-safe/unsafe modes. + */ +#define logbuf_lock_irq() \ + do { \ + printk_safe_enter_irq(); \ + raw_spin_lock(&logbuf_lock); \ + } while (0) + +#define logbuf_unlock_irq() \ + do { \ + raw_spin_unlock(&logbuf_lock); \ + printk_safe_exit_irq(); \ + } while (0) + +#define logbuf_lock_irqsave(flags) \ + do { \ + printk_safe_enter_irqsave(flags); \ + raw_spin_lock(&logbuf_lock); \ + } while (0) + +#define logbuf_unlock_irqrestore(flags) \ + do { \ + raw_spin_unlock(&logbuf_lock); \ + printk_safe_exit_irqrestore(flags); \ + } while (0) + #ifdef CONFIG_PRINTK DECLARE_WAIT_QUEUE_HEAD(log_wait); /* the next printk record to read by syslog(READ) or /proc/kmsg */ @@ -782,20 +829,21 @@ static ssize_t devkmsg_read(struct file *file, char __user *buf, ret = mutex_lock_interruptible(&user->lock); if (ret) return ret; - raw_spin_lock_irq(&logbuf_lock); + + logbuf_lock_irq(); while (user->seq == log_next_seq) { if (file->f_flags & O_NONBLOCK) { ret = -EAGAIN; - raw_spin_unlock_irq(&logbuf_lock); + logbuf_unlock_irq(); goto out; } - raw_spin_unlock_irq(&logbuf_lock); + logbuf_unlock_irq(); ret = wait_event_interruptible(log_wait, user->seq != log_next_seq); if (ret) goto out; - raw_spin_lock_irq(&logbuf_lock); + logbuf_lock_irq(); } if (user->seq < log_first_seq) { @@ -803,7 +851,7 @@ static ssize_t devkmsg_read(struct file *file, char __user *buf, user->idx = log_first_idx; user->seq = log_first_seq; ret = -EPIPE; - raw_spin_unlock_irq(&logbuf_lock); + logbuf_unlock_irq(); goto out; } @@ -816,7 +864,7 @@ static ssize_t devkmsg_read(struct file *file, char __user *buf, user->idx = log_next(user->idx); user->seq++; - raw_spin_unlock_irq(&logbuf_lock); + logbuf_unlock_irq(); if (len > count) { ret = -EINVAL; @@ -843,7 +891,7 @@ static loff_t devkmsg_llseek(struct file *file, loff_t offset, int whence) if (offset) return -ESPIPE; - raw_spin_lock_irq(&logbuf_lock); + logbuf_lock_irq(); switch (whence) { case SEEK_SET: /* the first record */ @@ -867,7 +915,7 @@ static loff_t devkmsg_llseek(struct file *file, loff_t offset, int whence) default: ret = -EINVAL; } - raw_spin_unlock_irq(&logbuf_lock); + logbuf_unlock_irq(); return ret; } @@ -881,7 +929,7 @@ static unsigned int devkmsg_poll(struct file *file, poll_table *wait) poll_wait(file, &log_wait, wait); - raw_spin_lock_irq(&logbuf_lock); + logbuf_lock_irq(); if (user->seq < log_next_seq) { /* return error when data has vanished underneath us */ if (user->seq < log_first_seq) @@ -889,7 +937,7 @@ static unsigned int devkmsg_poll(struct file *file, poll_table *wait) else ret = POLLIN|POLLRDNORM; } - raw_spin_unlock_irq(&logbuf_lock); + logbuf_unlock_irq(); return ret; } @@ -919,10 +967,10 @@ static int devkmsg_open(struct inode *inode, struct file *file) mutex_init(&user->lock); - raw_spin_lock_irq(&logbuf_lock); + logbuf_lock_irq(); user->idx = log_first_idx; user->seq = log_first_seq; - raw_spin_unlock_irq(&logbuf_lock); + logbuf_unlock_irq(); file->private_data = user; return 0; @@ -1064,13 +1112,13 @@ void __init setup_log_buf(int early) return; } - raw_spin_lock_irqsave(&logbuf_lock, flags); + logbuf_lock_irqsave(flags); log_buf_len = new_log_buf_len; log_buf = new_log_buf; new_log_buf_len = 0; free = __LOG_BUF_LEN - log_next_idx; memcpy(log_buf, __log_buf, __LOG_BUF_LEN); - raw_spin_unlock_irqrestore(&logbuf_lock, flags); + logbuf_unlock_irqrestore(flags); pr_info("log_buf_len: %d bytes\n", log_buf_len); pr_info("early log buf free: %d(%d%%)\n", @@ -1248,7 +1296,7 @@ static int syslog_print(char __user *buf, int size) size_t n; size_t skip; - raw_spin_lock_irq(&logbuf_lock); + logbuf_lock_irq(); if (syslog_seq < log_first_seq) { /* messages are gone, move to first one */ syslog_seq = log_first_seq; @@ -1256,7 +1304,7 @@ static int syslog_print(char __user *buf, int size) syslog_partial = 0; } if (syslog_seq == log_next_seq) { - raw_spin_unlock_irq(&logbuf_lock); + logbuf_unlock_irq(); break; } @@ -1275,7 +1323,7 @@ static int syslog_print(char __user *buf, int size) syslog_partial += n; } else n = 0; - raw_spin_unlock_irq(&logbuf_lock); + logbuf_unlock_irq(); if (!n) break; @@ -1304,7 +1352,7 @@ static int syslog_print_all(char __user *buf, int size, bool clear) if (!text) return -ENOMEM; - raw_spin_lock_irq(&logbuf_lock); + logbuf_lock_irq(); if (buf) { u64 next_seq; u64 seq; @@ -1352,12 +1400,12 @@ static int syslog_print_all(char __user *buf, int size, bool clear) idx = log_next(idx); seq++; - raw_spin_unlock_irq(&logbuf_lock); + logbuf_unlock_irq(); if (copy_to_user(buf + len, text, textlen)) len = -EFAULT; else len += textlen; - raw_spin_lock_irq(&logbuf_lock); + logbuf_lock_irq(); if (seq < log_first_seq) { /* messages are gone, move to next one */ @@ -1371,7 +1419,7 @@ static int syslog_print_all(char __user *buf, int size, bool clear) clear_seq = log_next_seq; clear_idx = log_next_idx; } - raw_spin_unlock_irq(&logbuf_lock); + logbuf_unlock_irq(); kfree(text); return len; @@ -1458,7 +1506,7 @@ int do_syslog(int type, char __user *buf, int len, int source) break; /* Number of chars in the log buffer */ case SYSLOG_ACTION_SIZE_UNREAD: - raw_spin_lock_irq(&logbuf_lock); + logbuf_lock_irq(); if (syslog_seq < log_first_seq) { /* messages are gone, move to first one */ syslog_seq = log_first_seq; @@ -1486,7 +1534,7 @@ int do_syslog(int type, char __user *buf, int len, int source) } error -= syslog_partial; } - raw_spin_unlock_irq(&logbuf_lock); + logbuf_unlock_irq(); break; /* Size of the log buffer */ case SYSLOG_ACTION_SIZE_BUFFER: @@ -1510,8 +1558,7 @@ SYSCALL_DEFINE3(syslog, int, type, char __user *, buf, int, len) * log_buf[start] to log_buf[end - 1]. * The console_lock must be held. */ -static void call_console_drivers(int level, - const char *ext_text, size_t ext_len, +static void call_console_drivers(const char *ext_text, size_t ext_len, const char *text, size_t len) { struct console *con; @@ -1538,28 +1585,6 @@ static void call_console_drivers(int level, } } -/* - * Zap console related locks when oopsing. - * To leave time for slow consoles to print a full oops, - * only zap at most once every 30 seconds. - */ -static void zap_locks(void) -{ - static unsigned long oops_timestamp; - - if (time_after_eq(jiffies, oops_timestamp) && - !time_after(jiffies, oops_timestamp + 30 * HZ)) - return; - - oops_timestamp = jiffies; - - debug_locks_off(); - /* If a crash is occurring, make sure we can't deadlock */ - raw_spin_lock_init(&logbuf_lock); - /* And make sure that we print immediately */ - sema_init(&console_sem, 1); -} - int printk_delay_msec __read_mostly; static inline void printk_delay(void) @@ -1669,18 +1694,13 @@ asmlinkage int vprintk_emit(int facility, int level, const char *dict, size_t dictlen, const char *fmt, va_list args) { - static bool recursion_bug; static char textbuf[LOG_LINE_MAX]; char *text = textbuf; size_t text_len = 0; enum log_flags lflags = 0; unsigned long flags; - int this_cpu; int printed_len = 0; - int nmi_message_lost; bool in_sched = false; - /* cpu currently holding logbuf_lock in this function */ - static unsigned int logbuf_cpu = UINT_MAX; if (level == LOGLEVEL_SCHED) { level = LOGLEVEL_DEFAULT; @@ -1690,53 +1710,8 @@ asmlinkage int vprintk_emit(int facility, int level, boot_delay_msec(level); printk_delay(); - local_irq_save(flags); - this_cpu = smp_processor_id(); - - /* - * Ouch, printk recursed into itself! - */ - if (unlikely(logbuf_cpu == this_cpu)) { - /* - * If a crash is occurring during printk() on this CPU, - * then try to get the crash message out but make sure - * we can't deadlock. Otherwise just return to avoid the - * recursion and return - but flag the recursion so that - * it can be printed at the next appropriate moment: - */ - if (!oops_in_progress && !lockdep_recursing(current)) { - recursion_bug = true; - local_irq_restore(flags); - return 0; - } - zap_locks(); - } - - lockdep_off(); /* This stops the holder of console_sem just where we want him */ - raw_spin_lock(&logbuf_lock); - logbuf_cpu = this_cpu; - - if (unlikely(recursion_bug)) { - static const char recursion_msg[] = - "BUG: recent printk recursion!"; - - recursion_bug = false; - /* emit KERN_CRIT message */ - printed_len += log_store(0, 2, LOG_PREFIX|LOG_NEWLINE, 0, - NULL, 0, recursion_msg, - strlen(recursion_msg)); - } - - nmi_message_lost = get_nmi_message_lost(); - if (unlikely(nmi_message_lost)) { - text_len = scnprintf(textbuf, sizeof(textbuf), - "BAD LUCK: lost %d message(s) from NMI context!", - nmi_message_lost); - printed_len += log_store(0, 2, LOG_PREFIX|LOG_NEWLINE, 0, - NULL, 0, textbuf, text_len); - } - + logbuf_lock_irqsave(flags); /* * The printf needs to come first; we need the syslog * prefix which might be passed-in as a parameter. @@ -1779,14 +1754,10 @@ asmlinkage int vprintk_emit(int facility, int level, printed_len += log_output(facility, level, lflags, dict, dictlen, text, text_len); - logbuf_cpu = UINT_MAX; - raw_spin_unlock(&logbuf_lock); - lockdep_on(); - local_irq_restore(flags); + logbuf_unlock_irqrestore(flags); /* If called from the scheduler, we can not call up(). */ if (!in_sched) { - lockdep_off(); /* * Try to acquire and then immediately release the console * semaphore. The release will print out buffers and wake up @@ -1794,7 +1765,6 @@ asmlinkage int vprintk_emit(int facility, int level, */ if (console_trylock()) console_unlock(); - lockdep_on(); } return printed_len; @@ -1803,7 +1773,7 @@ EXPORT_SYMBOL(vprintk_emit); asmlinkage int vprintk(const char *fmt, va_list args) { - return vprintk_emit(0, LOGLEVEL_DEFAULT, NULL, 0, fmt, args); + return vprintk_func(fmt, args); } EXPORT_SYMBOL(vprintk); @@ -1895,16 +1865,12 @@ static ssize_t msg_print_ext_header(char *buf, size_t size, static ssize_t msg_print_ext_body(char *buf, size_t size, char *dict, size_t dict_len, char *text, size_t text_len) { return 0; } -static void call_console_drivers(int level, - const char *ext_text, size_t ext_len, +static void call_console_drivers(const char *ext_text, size_t ext_len, const char *text, size_t len) {} static size_t msg_print_text(const struct printk_log *msg, bool syslog, char *buf, size_t size) { return 0; } static bool suppress_message_printing(int level) { return false; } -/* Still needs to be defined for users */ -DEFINE_PER_CPU(printk_func_t, printk_func); - #endif /* CONFIG_PRINTK */ #ifdef CONFIG_EARLY_PRINTK @@ -2220,9 +2186,9 @@ again: struct printk_log *msg; size_t ext_len = 0; size_t len; - int level; - raw_spin_lock_irqsave(&logbuf_lock, flags); + printk_safe_enter_irqsave(flags); + raw_spin_lock(&logbuf_lock); if (seen_seq != log_next_seq) { wake_klogd = true; seen_seq = log_next_seq; @@ -2243,8 +2209,7 @@ skip: break; msg = log_from_idx(console_idx); - level = msg->level; - if (suppress_message_printing(level)) { + if (suppress_message_printing(msg->level)) { /* * Skip record we have buffered and already printed * directly to the console when we received it, and @@ -2270,9 +2235,9 @@ skip: raw_spin_unlock(&logbuf_lock); stop_critical_timings(); /* don't trace print latency */ - call_console_drivers(level, ext_text, ext_len, text, len); + call_console_drivers(ext_text, ext_len, text, len); start_critical_timings(); - local_irq_restore(flags); + printk_safe_exit_irqrestore(flags); if (do_cond_resched) cond_resched(); @@ -2295,7 +2260,8 @@ skip: */ raw_spin_lock(&logbuf_lock); retry = console_seq != log_next_seq; - raw_spin_unlock_irqrestore(&logbuf_lock, flags); + raw_spin_unlock(&logbuf_lock); + printk_safe_exit_irqrestore(flags); if (retry && console_trylock()) goto again; @@ -2558,10 +2524,10 @@ void register_console(struct console *newcon) * console_unlock(); will print out the buffered messages * for us. */ - raw_spin_lock_irqsave(&logbuf_lock, flags); + logbuf_lock_irqsave(flags); console_seq = syslog_seq; console_idx = syslog_idx; - raw_spin_unlock_irqrestore(&logbuf_lock, flags); + logbuf_unlock_irqrestore(flags); /* * We're about to replay the log buffer. Only do this to the * just-registered console to avoid excessive message spam to @@ -2860,12 +2826,12 @@ void kmsg_dump(enum kmsg_dump_reason reason) /* initialize iterator with data about the stored records */ dumper->active = true; - raw_spin_lock_irqsave(&logbuf_lock, flags); + logbuf_lock_irqsave(flags); dumper->cur_seq = clear_seq; dumper->cur_idx = clear_idx; dumper->next_seq = log_next_seq; dumper->next_idx = log_next_idx; - raw_spin_unlock_irqrestore(&logbuf_lock, flags); + logbuf_unlock_irqrestore(flags); /* invoke dumper which will iterate over records */ dumper->dump(dumper, reason); @@ -2950,9 +2916,9 @@ bool kmsg_dump_get_line(struct kmsg_dumper *dumper, bool syslog, unsigned long flags; bool ret; - raw_spin_lock_irqsave(&logbuf_lock, flags); + logbuf_lock_irqsave(flags); ret = kmsg_dump_get_line_nolock(dumper, syslog, line, size, len); - raw_spin_unlock_irqrestore(&logbuf_lock, flags); + logbuf_unlock_irqrestore(flags); return ret; } @@ -2991,7 +2957,7 @@ bool kmsg_dump_get_buffer(struct kmsg_dumper *dumper, bool syslog, if (!dumper->active) goto out; - raw_spin_lock_irqsave(&logbuf_lock, flags); + logbuf_lock_irqsave(flags); if (dumper->cur_seq < log_first_seq) { /* messages are gone, move to first available one */ dumper->cur_seq = log_first_seq; @@ -3000,7 +2966,7 @@ bool kmsg_dump_get_buffer(struct kmsg_dumper *dumper, bool syslog, /* last entry */ if (dumper->cur_seq >= dumper->next_seq) { - raw_spin_unlock_irqrestore(&logbuf_lock, flags); + logbuf_unlock_irqrestore(flags); goto out; } @@ -3042,7 +3008,7 @@ bool kmsg_dump_get_buffer(struct kmsg_dumper *dumper, bool syslog, dumper->next_seq = next_seq; dumper->next_idx = next_idx; ret = true; - raw_spin_unlock_irqrestore(&logbuf_lock, flags); + logbuf_unlock_irqrestore(flags); out: if (len) *len = l; @@ -3080,9 +3046,9 @@ void kmsg_dump_rewind(struct kmsg_dumper *dumper) { unsigned long flags; - raw_spin_lock_irqsave(&logbuf_lock, flags); + logbuf_lock_irqsave(flags); kmsg_dump_rewind_nolock(dumper); - raw_spin_unlock_irqrestore(&logbuf_lock, flags); + logbuf_unlock_irqrestore(flags); } EXPORT_SYMBOL_GPL(kmsg_dump_rewind); diff --git a/kernel/printk/nmi.c b/kernel/printk/printk_safe.c index f011aaef583c..033e50a7d706 100644 --- a/kernel/printk/nmi.c +++ b/kernel/printk/printk_safe.c @@ -1,5 +1,5 @@ /* - * nmi.c - Safe printk in NMI context + * printk_safe.c - Safe printk for printk-deadlock-prone contexts * * This program is free software; you can redistribute it and/or * modify it under the terms of the GNU General Public License @@ -32,36 +32,58 @@ * is later flushed into the main ring buffer via IRQ work. * * The alternative implementation is chosen transparently - * via @printk_func per-CPU variable. + * by examinig current printk() context mask stored in @printk_context + * per-CPU variable. * * The implementation allows to flush the strings also from another CPU. * There are situations when we want to make sure that all buffers * were handled or when IRQs are blocked. */ -DEFINE_PER_CPU(printk_func_t, printk_func) = vprintk_default; -static int printk_nmi_irq_ready; -atomic_t nmi_message_lost; +static int printk_safe_irq_ready; -#define NMI_LOG_BUF_LEN ((1 << CONFIG_NMI_LOG_BUF_SHIFT) - \ - sizeof(atomic_t) - sizeof(struct irq_work)) +#define SAFE_LOG_BUF_LEN ((1 << CONFIG_PRINTK_SAFE_LOG_BUF_SHIFT) - \ + sizeof(atomic_t) - \ + sizeof(atomic_t) - \ + sizeof(struct irq_work)) -struct nmi_seq_buf { +struct printk_safe_seq_buf { atomic_t len; /* length of written data */ + atomic_t message_lost; struct irq_work work; /* IRQ work that flushes the buffer */ - unsigned char buffer[NMI_LOG_BUF_LEN]; + unsigned char buffer[SAFE_LOG_BUF_LEN]; }; -static DEFINE_PER_CPU(struct nmi_seq_buf, nmi_print_seq); + +static DEFINE_PER_CPU(struct printk_safe_seq_buf, safe_print_seq); +static DEFINE_PER_CPU(int, printk_context); + +#ifdef CONFIG_PRINTK_NMI +static DEFINE_PER_CPU(struct printk_safe_seq_buf, nmi_print_seq); +#endif + +/* Get flushed in a more safe context. */ +static void queue_flush_work(struct printk_safe_seq_buf *s) +{ + if (printk_safe_irq_ready) { + /* Make sure that IRQ work is really initialized. */ + smp_rmb(); + irq_work_queue(&s->work); + } +} /* - * Safe printk() for NMI context. It uses a per-CPU buffer to - * store the message. NMIs are not nested, so there is always only - * one writer running. But the buffer might get flushed from another - * CPU, so we need to be careful. + * Add a message to per-CPU context-dependent buffer. NMI and printk-safe + * have dedicated buffers, because otherwise printk-safe preempted by + * NMI-printk would have overwritten the NMI messages. + * + * The messages are fushed from irq work (or from panic()), possibly, + * from other CPU, concurrently with printk_safe_log_store(). Should this + * happen, printk_safe_log_store() will notice the buffer->len mismatch + * and repeat the write. */ -static int vprintk_nmi(const char *fmt, va_list args) +static int printk_safe_log_store(struct printk_safe_seq_buf *s, + const char *fmt, va_list args) { - struct nmi_seq_buf *s = this_cpu_ptr(&nmi_print_seq); - int add = 0; + int add; size_t len; again: @@ -69,18 +91,21 @@ again: /* The trailing '\0' is not counted into len. */ if (len >= sizeof(s->buffer) - 1) { - atomic_inc(&nmi_message_lost); + atomic_inc(&s->message_lost); + queue_flush_work(s); return 0; } /* - * Make sure that all old data have been read before the buffer was - * reseted. This is not needed when we just append data. + * Make sure that all old data have been read before the buffer + * was reset. This is not needed when we just append data. */ if (!len) smp_rmb(); add = vscnprintf(s->buffer + len, sizeof(s->buffer) - len, fmt, args); + if (!add) + return 0; /* * Do it once again if the buffer has been flushed in the meantime. @@ -90,32 +115,23 @@ again: if (atomic_cmpxchg(&s->len, len, len + add) != len) goto again; - /* Get flushed in a more safe context. */ - if (add && printk_nmi_irq_ready) { - /* Make sure that IRQ work is really initialized. */ - smp_rmb(); - irq_work_queue(&s->work); - } - + queue_flush_work(s); return add; } -static void printk_nmi_flush_line(const char *text, int len) +static inline void printk_safe_flush_line(const char *text, int len) { /* - * The buffers are flushed in NMI only on panic. The messages must - * go only into the ring buffer at this stage. Consoles will get - * explicitly called later when a crashdump is not generated. + * Avoid any console drivers calls from here, because we may be + * in NMI or printk_safe context (when in panic). The messages + * must go only into the ring buffer at this stage. Consoles will + * get explicitly called later when a crashdump is not generated. */ - if (in_nmi()) - printk_deferred("%.*s", len, text); - else - printk("%.*s", len, text); - + printk_deferred("%.*s", len, text); } /* printk part of the temporary buffer line by line */ -static int printk_nmi_flush_buffer(const char *start, size_t len) +static int printk_safe_flush_buffer(const char *start, size_t len) { const char *c, *end; bool header; @@ -127,7 +143,7 @@ static int printk_nmi_flush_buffer(const char *start, size_t len) /* Print line by line. */ while (c < end) { if (*c == '\n') { - printk_nmi_flush_line(start, c - start + 1); + printk_safe_flush_line(start, c - start + 1); start = ++c; header = true; continue; @@ -140,7 +156,7 @@ static int printk_nmi_flush_buffer(const char *start, size_t len) continue; } - printk_nmi_flush_line(start, c - start); + printk_safe_flush_line(start, c - start); start = c++; header = true; continue; @@ -154,22 +170,31 @@ static int printk_nmi_flush_buffer(const char *start, size_t len) if (start < end && !header) { static const char newline[] = KERN_CONT "\n"; - printk_nmi_flush_line(start, end - start); - printk_nmi_flush_line(newline, strlen(newline)); + printk_safe_flush_line(start, end - start); + printk_safe_flush_line(newline, strlen(newline)); } return len; } +static void report_message_lost(struct printk_safe_seq_buf *s) +{ + int lost = atomic_xchg(&s->message_lost, 0); + + if (lost) + printk_deferred("Lost %d message(s)!\n", lost); +} + /* - * Flush data from the associated per_CPU buffer. The function + * Flush data from the associated per-CPU buffer. The function * can be called either via IRQ work or independently. */ -static void __printk_nmi_flush(struct irq_work *work) +static void __printk_safe_flush(struct irq_work *work) { static raw_spinlock_t read_lock = __RAW_SPIN_LOCK_INITIALIZER(read_lock); - struct nmi_seq_buf *s = container_of(work, struct nmi_seq_buf, work); + struct printk_safe_seq_buf *s = + container_of(work, struct printk_safe_seq_buf, work); unsigned long flags; size_t len; int i; @@ -194,9 +219,9 @@ more: * buffer size. */ if ((i && i >= len) || len > sizeof(s->buffer)) { - const char *msg = "printk_nmi_flush: internal error\n"; + const char *msg = "printk_safe_flush: internal error\n"; - printk_nmi_flush_line(msg, strlen(msg)); + printk_safe_flush_line(msg, strlen(msg)); len = 0; } @@ -205,7 +230,7 @@ more: /* Make sure that data has been written up to the @len */ smp_rmb(); - i += printk_nmi_flush_buffer(s->buffer + i, len - i); + i += printk_safe_flush_buffer(s->buffer + i, len - i); /* * Check that nothing has got added in the meantime and truncate @@ -217,35 +242,40 @@ more: goto more; out: + report_message_lost(s); raw_spin_unlock_irqrestore(&read_lock, flags); } /** - * printk_nmi_flush - flush all per-cpu nmi buffers. + * printk_safe_flush - flush all per-cpu nmi buffers. * * The buffers are flushed automatically via IRQ work. This function * is useful only when someone wants to be sure that all buffers have * been flushed at some point. */ -void printk_nmi_flush(void) +void printk_safe_flush(void) { int cpu; - for_each_possible_cpu(cpu) - __printk_nmi_flush(&per_cpu(nmi_print_seq, cpu).work); + for_each_possible_cpu(cpu) { +#ifdef CONFIG_PRINTK_NMI + __printk_safe_flush(&per_cpu(nmi_print_seq, cpu).work); +#endif + __printk_safe_flush(&per_cpu(safe_print_seq, cpu).work); + } } /** - * printk_nmi_flush_on_panic - flush all per-cpu nmi buffers when the system + * printk_safe_flush_on_panic - flush all per-cpu nmi buffers when the system * goes down. * - * Similar to printk_nmi_flush() but it can be called even in NMI context when + * Similar to printk_safe_flush() but it can be called even in NMI context when * the system goes down. It does the best effort to get NMI messages into * the main ring buffer. * * Note that it could try harder when there is only one CPU online. */ -void printk_nmi_flush_on_panic(void) +void printk_safe_flush_on_panic(void) { /* * Make sure that we could access the main ring buffer. @@ -259,33 +289,97 @@ void printk_nmi_flush_on_panic(void) raw_spin_lock_init(&logbuf_lock); } - printk_nmi_flush(); + printk_safe_flush(); } -void __init printk_nmi_init(void) +#ifdef CONFIG_PRINTK_NMI +/* + * Safe printk() for NMI context. It uses a per-CPU buffer to + * store the message. NMIs are not nested, so there is always only + * one writer running. But the buffer might get flushed from another + * CPU, so we need to be careful. + */ +static int vprintk_nmi(const char *fmt, va_list args) { - int cpu; + struct printk_safe_seq_buf *s = this_cpu_ptr(&nmi_print_seq); - for_each_possible_cpu(cpu) { - struct nmi_seq_buf *s = &per_cpu(nmi_print_seq, cpu); + return printk_safe_log_store(s, fmt, args); +} - init_irq_work(&s->work, __printk_nmi_flush); - } +void printk_nmi_enter(void) +{ + this_cpu_or(printk_context, PRINTK_NMI_CONTEXT_MASK); +} - /* Make sure that IRQ works are initialized before enabling. */ - smp_wmb(); - printk_nmi_irq_ready = 1; +void printk_nmi_exit(void) +{ + this_cpu_and(printk_context, ~PRINTK_NMI_CONTEXT_MASK); +} - /* Flush pending messages that did not have scheduled IRQ works. */ - printk_nmi_flush(); +#else + +static int vprintk_nmi(const char *fmt, va_list args) +{ + return 0; } -void printk_nmi_enter(void) +#endif /* CONFIG_PRINTK_NMI */ + +/* + * Lock-less printk(), to avoid deadlocks should the printk() recurse + * into itself. It uses a per-CPU buffer to store the message, just like + * NMI. + */ +static int vprintk_safe(const char *fmt, va_list args) { - this_cpu_write(printk_func, vprintk_nmi); + struct printk_safe_seq_buf *s = this_cpu_ptr(&safe_print_seq); + + return printk_safe_log_store(s, fmt, args); } -void printk_nmi_exit(void) +/* Can be preempted by NMI. */ +void __printk_safe_enter(void) +{ + this_cpu_inc(printk_context); +} + +/* Can be preempted by NMI. */ +void __printk_safe_exit(void) { - this_cpu_write(printk_func, vprintk_default); + this_cpu_dec(printk_context); +} + +__printf(1, 0) int vprintk_func(const char *fmt, va_list args) +{ + if (this_cpu_read(printk_context) & PRINTK_NMI_CONTEXT_MASK) + return vprintk_nmi(fmt, args); + + if (this_cpu_read(printk_context) & PRINTK_SAFE_CONTEXT_MASK) + return vprintk_safe(fmt, args); + + return vprintk_default(fmt, args); +} + +void __init printk_safe_init(void) +{ + int cpu; + + for_each_possible_cpu(cpu) { + struct printk_safe_seq_buf *s; + + s = &per_cpu(safe_print_seq, cpu); + init_irq_work(&s->work, __printk_safe_flush); + +#ifdef CONFIG_PRINTK_NMI + s = &per_cpu(nmi_print_seq, cpu); + init_irq_work(&s->work, __printk_safe_flush); +#endif + } + + /* Make sure that IRQ works are initialized before enabling. */ + smp_wmb(); + printk_safe_irq_ready = 1; + + /* Flush pending messages that did not have scheduled IRQ works. */ + printk_safe_flush(); } diff --git a/kernel/seccomp.c b/kernel/seccomp.c index f8f88ebcb3ba..e15185c28de5 100644 --- a/kernel/seccomp.c +++ b/kernel/seccomp.c @@ -643,11 +643,14 @@ static int __seccomp_filter(int this_syscall, const struct seccomp_data *sd, default: { siginfo_t info; audit_seccomp(this_syscall, SIGSYS, action); - /* Show the original registers in the dump. */ - syscall_rollback(current, task_pt_regs(current)); - /* Trigger a manual coredump since do_exit skips it. */ - seccomp_init_siginfo(&info, this_syscall, data); - do_coredump(&info); + /* Dump core only if this is the last remaining thread. */ + if (get_nr_threads(current) == 1) { + /* Show the original registers in the dump. */ + syscall_rollback(current, task_pt_regs(current)); + /* Trigger a manual coredump since do_exit skips it. */ + seccomp_init_siginfo(&info, this_syscall, data); + do_coredump(&info); + } do_exit(SIGSYS); } } diff --git a/lib/nmi_backtrace.c b/lib/nmi_backtrace.c index 75554754eadf..5f7999eacad5 100644 --- a/lib/nmi_backtrace.c +++ b/lib/nmi_backtrace.c @@ -77,7 +77,7 @@ void nmi_trigger_cpumask_backtrace(const cpumask_t *mask, * Force flush any remote buffers that might be stuck in IRQ context * and therefore could not run their irq_work. */ - printk_nmi_flush(); + printk_safe_flush(); clear_bit_unlock(0, &backtrace_flag); put_cpu(); diff --git a/scripts/dtc/checks.c b/scripts/dtc/checks.c index 386f9563313f..3d18e45374c8 100644 --- a/scripts/dtc/checks.c +++ b/scripts/dtc/checks.c @@ -40,16 +40,11 @@ enum checkstatus { struct check; -typedef void (*tree_check_fn)(struct check *c, struct node *dt); -typedef void (*node_check_fn)(struct check *c, struct node *dt, struct node *node); -typedef void (*prop_check_fn)(struct check *c, struct node *dt, - struct node *node, struct property *prop); +typedef void (*check_fn)(struct check *c, struct dt_info *dti, struct node *node); struct check { const char *name; - tree_check_fn tree_fn; - node_check_fn node_fn; - prop_check_fn prop_fn; + check_fn fn; void *data; bool warn, error; enum checkstatus status; @@ -58,45 +53,24 @@ struct check { struct check **prereq; }; -#define CHECK_ENTRY(nm, tfn, nfn, pfn, d, w, e, ...) \ - static struct check *nm##_prereqs[] = { __VA_ARGS__ }; \ - static struct check nm = { \ - .name = #nm, \ - .tree_fn = (tfn), \ - .node_fn = (nfn), \ - .prop_fn = (pfn), \ - .data = (d), \ - .warn = (w), \ - .error = (e), \ +#define CHECK_ENTRY(_nm, _fn, _d, _w, _e, ...) \ + static struct check *_nm##_prereqs[] = { __VA_ARGS__ }; \ + static struct check _nm = { \ + .name = #_nm, \ + .fn = (_fn), \ + .data = (_d), \ + .warn = (_w), \ + .error = (_e), \ .status = UNCHECKED, \ - .num_prereqs = ARRAY_SIZE(nm##_prereqs), \ - .prereq = nm##_prereqs, \ + .num_prereqs = ARRAY_SIZE(_nm##_prereqs), \ + .prereq = _nm##_prereqs, \ }; -#define WARNING(nm, tfn, nfn, pfn, d, ...) \ - CHECK_ENTRY(nm, tfn, nfn, pfn, d, true, false, __VA_ARGS__) -#define ERROR(nm, tfn, nfn, pfn, d, ...) \ - CHECK_ENTRY(nm, tfn, nfn, pfn, d, false, true, __VA_ARGS__) -#define CHECK(nm, tfn, nfn, pfn, d, ...) \ - CHECK_ENTRY(nm, tfn, nfn, pfn, d, false, false, __VA_ARGS__) - -#define TREE_WARNING(nm, d, ...) \ - WARNING(nm, check_##nm, NULL, NULL, d, __VA_ARGS__) -#define TREE_ERROR(nm, d, ...) \ - ERROR(nm, check_##nm, NULL, NULL, d, __VA_ARGS__) -#define TREE_CHECK(nm, d, ...) \ - CHECK(nm, check_##nm, NULL, NULL, d, __VA_ARGS__) -#define NODE_WARNING(nm, d, ...) \ - WARNING(nm, NULL, check_##nm, NULL, d, __VA_ARGS__) -#define NODE_ERROR(nm, d, ...) \ - ERROR(nm, NULL, check_##nm, NULL, d, __VA_ARGS__) -#define NODE_CHECK(nm, d, ...) \ - CHECK(nm, NULL, check_##nm, NULL, d, __VA_ARGS__) -#define PROP_WARNING(nm, d, ...) \ - WARNING(nm, NULL, NULL, check_##nm, d, __VA_ARGS__) -#define PROP_ERROR(nm, d, ...) \ - ERROR(nm, NULL, NULL, check_##nm, d, __VA_ARGS__) -#define PROP_CHECK(nm, d, ...) \ - CHECK(nm, NULL, NULL, check_##nm, d, __VA_ARGS__) +#define WARNING(_nm, _fn, _d, ...) \ + CHECK_ENTRY(_nm, _fn, _d, true, false, __VA_ARGS__) +#define ERROR(_nm, _fn, _d, ...) \ + CHECK_ENTRY(_nm, _fn, _d, false, true, __VA_ARGS__) +#define CHECK(_nm, _fn, _d, ...) \ + CHECK_ENTRY(_nm, _fn, _d, false, false, __VA_ARGS__) #ifdef __GNUC__ static inline void check_msg(struct check *c, const char *fmt, ...) __attribute__((format (printf, 2, 3))); @@ -123,27 +97,21 @@ static inline void check_msg(struct check *c, const char *fmt, ...) check_msg((c), __VA_ARGS__); \ } while (0) -static void check_nodes_props(struct check *c, struct node *dt, struct node *node) +static void check_nodes_props(struct check *c, struct dt_info *dti, struct node *node) { struct node *child; - struct property *prop; TRACE(c, "%s", node->fullpath); - if (c->node_fn) - c->node_fn(c, dt, node); - - if (c->prop_fn) - for_each_property(node, prop) { - TRACE(c, "%s\t'%s'", node->fullpath, prop->name); - c->prop_fn(c, dt, node, prop); - } + if (c->fn) + c->fn(c, dti, node); for_each_child(node, child) - check_nodes_props(c, dt, child); + check_nodes_props(c, dti, child); } -static bool run_check(struct check *c, struct node *dt) +static bool run_check(struct check *c, struct dt_info *dti) { + struct node *dt = dti->dt; bool error = false; int i; @@ -156,7 +124,7 @@ static bool run_check(struct check *c, struct node *dt) for (i = 0; i < c->num_prereqs; i++) { struct check *prq = c->prereq[i]; - error = error || run_check(prq, dt); + error = error || run_check(prq, dti); if (prq->status != PASSED) { c->status = PREREQ; check_msg(c, "Failed prerequisite '%s'", @@ -167,11 +135,8 @@ static bool run_check(struct check *c, struct node *dt) if (c->status != UNCHECKED) goto out; - if (c->node_fn || c->prop_fn) - check_nodes_props(c, dt, dt); + check_nodes_props(c, dti, dt); - if (c->tree_fn) - c->tree_fn(c, dt); if (c->status == UNCHECKED) c->status = PASSED; @@ -189,13 +154,14 @@ out: */ /* A check which always fails, for testing purposes only */ -static inline void check_always_fail(struct check *c, struct node *dt) +static inline void check_always_fail(struct check *c, struct dt_info *dti, + struct node *node) { FAIL(c, "always_fail check"); } -TREE_CHECK(always_fail, NULL); +CHECK(always_fail, check_always_fail, NULL); -static void check_is_string(struct check *c, struct node *root, +static void check_is_string(struct check *c, struct dt_info *dti, struct node *node) { struct property *prop; @@ -210,11 +176,11 @@ static void check_is_string(struct check *c, struct node *root, propname, node->fullpath); } #define WARNING_IF_NOT_STRING(nm, propname) \ - WARNING(nm, NULL, check_is_string, NULL, (propname)) + WARNING(nm, check_is_string, (propname)) #define ERROR_IF_NOT_STRING(nm, propname) \ - ERROR(nm, NULL, check_is_string, NULL, (propname)) + ERROR(nm, check_is_string, (propname)) -static void check_is_cell(struct check *c, struct node *root, +static void check_is_cell(struct check *c, struct dt_info *dti, struct node *node) { struct property *prop; @@ -229,15 +195,15 @@ static void check_is_cell(struct check *c, struct node *root, propname, node->fullpath); } #define WARNING_IF_NOT_CELL(nm, propname) \ - WARNING(nm, NULL, check_is_cell, NULL, (propname)) + WARNING(nm, check_is_cell, (propname)) #define ERROR_IF_NOT_CELL(nm, propname) \ - ERROR(nm, NULL, check_is_cell, NULL, (propname)) + ERROR(nm, check_is_cell, (propname)) /* * Structural check functions */ -static void check_duplicate_node_names(struct check *c, struct node *dt, +static void check_duplicate_node_names(struct check *c, struct dt_info *dti, struct node *node) { struct node *child, *child2; @@ -250,9 +216,9 @@ static void check_duplicate_node_names(struct check *c, struct node *dt, FAIL(c, "Duplicate node name %s", child->fullpath); } -NODE_ERROR(duplicate_node_names, NULL); +ERROR(duplicate_node_names, check_duplicate_node_names, NULL); -static void check_duplicate_property_names(struct check *c, struct node *dt, +static void check_duplicate_property_names(struct check *c, struct dt_info *dti, struct node *node) { struct property *prop, *prop2; @@ -267,14 +233,14 @@ static void check_duplicate_property_names(struct check *c, struct node *dt, } } } -NODE_ERROR(duplicate_property_names, NULL); +ERROR(duplicate_property_names, check_duplicate_property_names, NULL); #define LOWERCASE "abcdefghijklmnopqrstuvwxyz" #define UPPERCASE "ABCDEFGHIJKLMNOPQRSTUVWXYZ" #define DIGITS "0123456789" #define PROPNODECHARS LOWERCASE UPPERCASE DIGITS ",._+*#?-" -static void check_node_name_chars(struct check *c, struct node *dt, +static void check_node_name_chars(struct check *c, struct dt_info *dti, struct node *node) { int n = strspn(node->name, c->data); @@ -283,19 +249,19 @@ static void check_node_name_chars(struct check *c, struct node *dt, FAIL(c, "Bad character '%c' in node %s", node->name[n], node->fullpath); } -NODE_ERROR(node_name_chars, PROPNODECHARS "@"); +ERROR(node_name_chars, check_node_name_chars, PROPNODECHARS "@"); -static void check_node_name_format(struct check *c, struct node *dt, +static void check_node_name_format(struct check *c, struct dt_info *dti, struct node *node) { if (strchr(get_unitname(node), '@')) FAIL(c, "Node %s has multiple '@' characters in name", node->fullpath); } -NODE_ERROR(node_name_format, NULL, &node_name_chars); +ERROR(node_name_format, check_node_name_format, NULL, &node_name_chars); -static void check_unit_address_vs_reg(struct check *c, struct node *dt, - struct node *node) +static void check_unit_address_vs_reg(struct check *c, struct dt_info *dti, + struct node *node) { const char *unitname = get_unitname(node); struct property *prop = get_property(node, "reg"); @@ -316,18 +282,22 @@ static void check_unit_address_vs_reg(struct check *c, struct node *dt, node->fullpath); } } -NODE_WARNING(unit_address_vs_reg, NULL); +WARNING(unit_address_vs_reg, check_unit_address_vs_reg, NULL); -static void check_property_name_chars(struct check *c, struct node *dt, - struct node *node, struct property *prop) +static void check_property_name_chars(struct check *c, struct dt_info *dti, + struct node *node) { - int n = strspn(prop->name, c->data); + struct property *prop; + + for_each_property(node, prop) { + int n = strspn(prop->name, c->data); - if (n < strlen(prop->name)) - FAIL(c, "Bad character '%c' in property name \"%s\", node %s", - prop->name[n], prop->name, node->fullpath); + if (n < strlen(prop->name)) + FAIL(c, "Bad character '%c' in property name \"%s\", node %s", + prop->name[n], prop->name, node->fullpath); + } } -PROP_ERROR(property_name_chars, PROPNODECHARS); +ERROR(property_name_chars, check_property_name_chars, PROPNODECHARS); #define DESCLABEL_FMT "%s%s%s%s%s" #define DESCLABEL_ARGS(node,prop,mark) \ @@ -336,10 +306,11 @@ PROP_ERROR(property_name_chars, PROPNODECHARS); ((prop) ? (prop)->name : ""), \ ((prop) ? "' in " : ""), (node)->fullpath -static void check_duplicate_label(struct check *c, struct node *dt, +static void check_duplicate_label(struct check *c, struct dt_info *dti, const char *label, struct node *node, struct property *prop, struct marker *mark) { + struct node *dt = dti->dt; struct node *othernode = NULL; struct property *otherprop = NULL; struct marker *othermark = NULL; @@ -362,44 +333,43 @@ static void check_duplicate_label(struct check *c, struct node *dt, DESCLABEL_ARGS(othernode, otherprop, othermark)); } -static void check_duplicate_label_node(struct check *c, struct node *dt, +static void check_duplicate_label_node(struct check *c, struct dt_info *dti, struct node *node) { struct label *l; + struct property *prop; for_each_label(node->labels, l) - check_duplicate_label(c, dt, l->label, node, NULL, NULL); -} -static void check_duplicate_label_prop(struct check *c, struct node *dt, - struct node *node, struct property *prop) -{ - struct marker *m = prop->val.markers; - struct label *l; + check_duplicate_label(c, dti, l->label, node, NULL, NULL); + + for_each_property(node, prop) { + struct marker *m = prop->val.markers; - for_each_label(prop->labels, l) - check_duplicate_label(c, dt, l->label, node, prop, NULL); + for_each_label(prop->labels, l) + check_duplicate_label(c, dti, l->label, node, prop, NULL); - for_each_marker_of_type(m, LABEL) - check_duplicate_label(c, dt, m->ref, node, prop, m); + for_each_marker_of_type(m, LABEL) + check_duplicate_label(c, dti, m->ref, node, prop, m); + } } -ERROR(duplicate_label, NULL, check_duplicate_label_node, - check_duplicate_label_prop, NULL); +ERROR(duplicate_label, check_duplicate_label_node, NULL); -static void check_explicit_phandles(struct check *c, struct node *root, - struct node *node, struct property *prop) +static cell_t check_phandle_prop(struct check *c, struct dt_info *dti, + struct node *node, const char *propname) { + struct node *root = dti->dt; + struct property *prop; struct marker *m; - struct node *other; cell_t phandle; - if (!streq(prop->name, "phandle") - && !streq(prop->name, "linux,phandle")) - return; + prop = get_property(node, propname); + if (!prop) + return 0; if (prop->val.len != sizeof(cell_t)) { FAIL(c, "%s has bad length (%d) %s property", node->fullpath, prop->val.len, prop->name); - return; + return 0; } m = prop->val.markers; @@ -411,14 +381,13 @@ static void check_explicit_phandles(struct check *c, struct node *root, * by construction. */ { FAIL(c, "%s in %s is a reference to another node", prop->name, node->fullpath); - return; } /* But setting this node's phandle equal to its own * phandle is allowed - that means allocate a unique * phandle for this node, even if it's not otherwise * referenced. The value will be filled in later, so - * no further checking for now. */ - return; + * we treat it as having no phandle data for now. */ + return 0; } phandle = propval_cell(prop); @@ -426,12 +395,36 @@ static void check_explicit_phandles(struct check *c, struct node *root, if ((phandle == 0) || (phandle == -1)) { FAIL(c, "%s has bad value (0x%x) in %s property", node->fullpath, phandle, prop->name); - return; + return 0; } - if (node->phandle && (node->phandle != phandle)) - FAIL(c, "%s has %s property which replaces existing phandle information", - node->fullpath, prop->name); + return phandle; +} + +static void check_explicit_phandles(struct check *c, struct dt_info *dti, + struct node *node) +{ + struct node *root = dti->dt; + struct node *other; + cell_t phandle, linux_phandle; + + /* Nothing should have assigned phandles yet */ + assert(!node->phandle); + + phandle = check_phandle_prop(c, dti, node, "phandle"); + + linux_phandle = check_phandle_prop(c, dti, node, "linux,phandle"); + + if (!phandle && !linux_phandle) + /* No valid phandles; nothing further to check */ + return; + + if (linux_phandle && phandle && (phandle != linux_phandle)) + FAIL(c, "%s has mismatching 'phandle' and 'linux,phandle'" + " properties", node->fullpath); + + if (linux_phandle && !phandle) + phandle = linux_phandle; other = get_node_by_phandle(root, phandle); if (other && (other != node)) { @@ -442,9 +435,9 @@ static void check_explicit_phandles(struct check *c, struct node *root, node->phandle = phandle; } -PROP_ERROR(explicit_phandles, NULL); +ERROR(explicit_phandles, check_explicit_phandles, NULL); -static void check_name_properties(struct check *c, struct node *root, +static void check_name_properties(struct check *c, struct dt_info *dti, struct node *node) { struct property **pp, *prop = NULL; @@ -472,60 +465,73 @@ static void check_name_properties(struct check *c, struct node *root, } } ERROR_IF_NOT_STRING(name_is_string, "name"); -NODE_ERROR(name_properties, NULL, &name_is_string); +ERROR(name_properties, check_name_properties, NULL, &name_is_string); /* * Reference fixup functions */ -static void fixup_phandle_references(struct check *c, struct node *dt, - struct node *node, struct property *prop) +static void fixup_phandle_references(struct check *c, struct dt_info *dti, + struct node *node) { - struct marker *m = prop->val.markers; - struct node *refnode; - cell_t phandle; + struct node *dt = dti->dt; + struct property *prop; - for_each_marker_of_type(m, REF_PHANDLE) { - assert(m->offset + sizeof(cell_t) <= prop->val.len); + for_each_property(node, prop) { + struct marker *m = prop->val.markers; + struct node *refnode; + cell_t phandle; + + for_each_marker_of_type(m, REF_PHANDLE) { + assert(m->offset + sizeof(cell_t) <= prop->val.len); + + refnode = get_node_by_ref(dt, m->ref); + if (! refnode) { + if (!(dti->dtsflags & DTSF_PLUGIN)) + FAIL(c, "Reference to non-existent node or " + "label \"%s\"\n", m->ref); + else /* mark the entry as unresolved */ + *((cell_t *)(prop->val.val + m->offset)) = + cpu_to_fdt32(0xffffffff); + continue; + } - refnode = get_node_by_ref(dt, m->ref); - if (! refnode) { - FAIL(c, "Reference to non-existent node or label \"%s\"\n", - m->ref); - continue; + phandle = get_node_phandle(dt, refnode); + *((cell_t *)(prop->val.val + m->offset)) = cpu_to_fdt32(phandle); } - - phandle = get_node_phandle(dt, refnode); - *((cell_t *)(prop->val.val + m->offset)) = cpu_to_fdt32(phandle); } } -ERROR(phandle_references, NULL, NULL, fixup_phandle_references, NULL, +ERROR(phandle_references, fixup_phandle_references, NULL, &duplicate_node_names, &explicit_phandles); -static void fixup_path_references(struct check *c, struct node *dt, - struct node *node, struct property *prop) +static void fixup_path_references(struct check *c, struct dt_info *dti, + struct node *node) { - struct marker *m = prop->val.markers; - struct node *refnode; - char *path; - - for_each_marker_of_type(m, REF_PATH) { - assert(m->offset <= prop->val.len); - - refnode = get_node_by_ref(dt, m->ref); - if (!refnode) { - FAIL(c, "Reference to non-existent node or label \"%s\"\n", - m->ref); - continue; - } + struct node *dt = dti->dt; + struct property *prop; + + for_each_property(node, prop) { + struct marker *m = prop->val.markers; + struct node *refnode; + char *path; + + for_each_marker_of_type(m, REF_PATH) { + assert(m->offset <= prop->val.len); - path = refnode->fullpath; - prop->val = data_insert_at_marker(prop->val, m, path, - strlen(path) + 1); + refnode = get_node_by_ref(dt, m->ref); + if (!refnode) { + FAIL(c, "Reference to non-existent node or label \"%s\"\n", + m->ref); + continue; + } + + path = refnode->fullpath; + prop->val = data_insert_at_marker(prop->val, m, path, + strlen(path) + 1); + } } } -ERROR(path_references, NULL, NULL, fixup_path_references, NULL, - &duplicate_node_names); +ERROR(path_references, fixup_path_references, NULL, &duplicate_node_names); /* * Semantic checks @@ -538,7 +544,7 @@ WARNING_IF_NOT_STRING(device_type_is_string, "device_type"); WARNING_IF_NOT_STRING(model_is_string, "model"); WARNING_IF_NOT_STRING(status_is_string, "status"); -static void fixup_addr_size_cells(struct check *c, struct node *dt, +static void fixup_addr_size_cells(struct check *c, struct dt_info *dti, struct node *node) { struct property *prop; @@ -554,7 +560,7 @@ static void fixup_addr_size_cells(struct check *c, struct node *dt, if (prop) node->size_cells = propval_cell(prop); } -WARNING(addr_size_cells, NULL, fixup_addr_size_cells, NULL, NULL, +WARNING(addr_size_cells, fixup_addr_size_cells, NULL, &address_cells_is_cell, &size_cells_is_cell); #define node_addr_cells(n) \ @@ -562,7 +568,7 @@ WARNING(addr_size_cells, NULL, fixup_addr_size_cells, NULL, NULL, #define node_size_cells(n) \ (((n)->size_cells == -1) ? 1 : (n)->size_cells) -static void check_reg_format(struct check *c, struct node *dt, +static void check_reg_format(struct check *c, struct dt_info *dti, struct node *node) { struct property *prop; @@ -589,9 +595,9 @@ static void check_reg_format(struct check *c, struct node *dt, "(#address-cells == %d, #size-cells == %d)", node->fullpath, prop->val.len, addr_cells, size_cells); } -NODE_WARNING(reg_format, NULL, &addr_size_cells); +WARNING(reg_format, check_reg_format, NULL, &addr_size_cells); -static void check_ranges_format(struct check *c, struct node *dt, +static void check_ranges_format(struct check *c, struct dt_info *dti, struct node *node) { struct property *prop; @@ -630,12 +636,12 @@ static void check_ranges_format(struct check *c, struct node *dt, p_addr_cells, c_addr_cells, c_size_cells); } } -NODE_WARNING(ranges_format, NULL, &addr_size_cells); +WARNING(ranges_format, check_ranges_format, NULL, &addr_size_cells); /* * Style checks */ -static void check_avoid_default_addr_size(struct check *c, struct node *dt, +static void check_avoid_default_addr_size(struct check *c, struct dt_info *dti, struct node *node) { struct property *reg, *ranges; @@ -657,14 +663,21 @@ static void check_avoid_default_addr_size(struct check *c, struct node *dt, FAIL(c, "Relying on default #size-cells value for %s", node->fullpath); } -NODE_WARNING(avoid_default_addr_size, NULL, &addr_size_cells); +WARNING(avoid_default_addr_size, check_avoid_default_addr_size, NULL, + &addr_size_cells); static void check_obsolete_chosen_interrupt_controller(struct check *c, - struct node *dt) + struct dt_info *dti, + struct node *node) { + struct node *dt = dti->dt; struct node *chosen; struct property *prop; + if (node != dt) + return; + + chosen = get_node_by_path(dt, "/chosen"); if (!chosen) return; @@ -674,7 +687,8 @@ static void check_obsolete_chosen_interrupt_controller(struct check *c, FAIL(c, "/chosen has obsolete \"interrupt-controller\" " "property"); } -TREE_WARNING(obsolete_chosen_interrupt_controller, NULL); +WARNING(obsolete_chosen_interrupt_controller, + check_obsolete_chosen_interrupt_controller, NULL); static struct check *check_table[] = { &duplicate_node_names, &duplicate_property_names, @@ -760,9 +774,8 @@ void parse_checks_option(bool warn, bool error, const char *arg) die("Unrecognized check name \"%s\"\n", name); } -void process_checks(bool force, struct boot_info *bi) +void process_checks(bool force, struct dt_info *dti) { - struct node *dt = bi->dt; int i; int error = 0; @@ -770,7 +783,7 @@ void process_checks(bool force, struct boot_info *bi) struct check *c = check_table[i]; if (c->warn || c->error) - error = error || run_check(c, dt); + error = error || run_check(c, dti); } if (error) { diff --git a/scripts/dtc/dtc-lexer.l b/scripts/dtc/dtc-lexer.l index 790fbf6cf2d7..c600603044f3 100644 --- a/scripts/dtc/dtc-lexer.l +++ b/scripts/dtc/dtc-lexer.l @@ -121,6 +121,11 @@ static void lexical_error(const char *fmt, ...); return DT_V1; } +<*>"/plugin/" { + DPRINT("Keyword: /plugin/\n"); + return DT_PLUGIN; + } + <*>"/memreserve/" { DPRINT("Keyword: /memreserve/\n"); BEGIN_DEFAULT(); @@ -184,16 +189,16 @@ static void lexical_error(const char *fmt, ...); if (d.len == 1) { lexical_error("Empty character literal"); yylval.integer = 0; - return DT_CHAR_LITERAL; - } + } else { + yylval.integer = (unsigned char)d.val[0]; - yylval.integer = (unsigned char)d.val[0]; - - if (d.len > 2) - lexical_error("Character literal has %d" - " characters instead of 1", - d.len - 1); + if (d.len > 2) + lexical_error("Character literal has %d" + " characters instead of 1", + d.len - 1); + } + data_free(d); return DT_CHAR_LITERAL; } diff --git a/scripts/dtc/dtc-lexer.lex.c_shipped b/scripts/dtc/dtc-lexer.lex.c_shipped index ba525c2f9fc2..2c862bc86ad0 100644 --- a/scripts/dtc/dtc-lexer.lex.c_shipped +++ b/scripts/dtc/dtc-lexer.lex.c_shipped @@ -8,8 +8,8 @@ #define FLEX_SCANNER #define YY_FLEX_MAJOR_VERSION 2 -#define YY_FLEX_MINOR_VERSION 5 -#define YY_FLEX_SUBMINOR_VERSION 39 +#define YY_FLEX_MINOR_VERSION 6 +#define YY_FLEX_SUBMINOR_VERSION 1 #if YY_FLEX_SUBMINOR_VERSION > 0 #define FLEX_BETA #endif @@ -88,25 +88,13 @@ typedef unsigned int flex_uint32_t; #endif /* ! FLEXINT_H */ -#ifdef __cplusplus - -/* The "const" storage-class-modifier is valid. */ -#define YY_USE_CONST - -#else /* ! __cplusplus */ - -/* C99 requires __STDC__ to be defined as 1. */ -#if defined (__STDC__) - -#define YY_USE_CONST - -#endif /* defined (__STDC__) */ -#endif /* ! __cplusplus */ - -#ifdef YY_USE_CONST +/* TODO: this is always defined, so inline it */ #define yyconst const + +#if defined(__GNUC__) && __GNUC__ >= 3 +#define yynoreturn __attribute__((__noreturn__)) #else -#define yyconst +#define yynoreturn #endif /* Returned upon end-of-file. */ @@ -167,7 +155,7 @@ typedef struct yy_buffer_state *YY_BUFFER_STATE; typedef size_t yy_size_t; #endif -extern yy_size_t yyleng; +extern int yyleng; extern FILE *yyin, *yyout; @@ -206,12 +194,12 @@ struct yy_buffer_state /* Size of input buffer in bytes, not including room for EOB * characters. */ - yy_size_t yy_buf_size; + int yy_buf_size; /* Number of characters read into yy_ch_buf, not including EOB * characters. */ - yy_size_t yy_n_chars; + int yy_n_chars; /* Whether we "own" the buffer - i.e., we know we created it, * and can realloc() it to grow it, and should free() it to @@ -234,7 +222,7 @@ struct yy_buffer_state int yy_bs_lineno; /**< The line count. */ int yy_bs_column; /**< The column count. */ - + /* Whether to try to fill the input buffer when we reach the * end of it. */ @@ -262,7 +250,7 @@ struct yy_buffer_state /* Stack of input buffers. */ static size_t yy_buffer_stack_top = 0; /**< index of top of stack. */ static size_t yy_buffer_stack_max = 0; /**< capacity of stack. */ -static YY_BUFFER_STATE * yy_buffer_stack = 0; /**< Stack as an array. */ +static YY_BUFFER_STATE * yy_buffer_stack = NULL; /**< Stack as an array. */ /* We provide macros for accessing buffer states in case in the * future we want to put the buffer states in a more general @@ -281,11 +269,11 @@ static YY_BUFFER_STATE * yy_buffer_stack = 0; /**< Stack as an array. */ /* yy_hold_char holds the character lost when yytext is formed. */ static char yy_hold_char; -static yy_size_t yy_n_chars; /* number of characters read into yy_ch_buf */ -yy_size_t yyleng; +static int yy_n_chars; /* number of characters read into yy_ch_buf */ +int yyleng; /* Points to current character in buffer. */ -static char *yy_c_buf_p = (char *) 0; +static char *yy_c_buf_p = NULL; static int yy_init = 0; /* whether we need to initialize */ static int yy_start = 0; /* start state number */ @@ -310,7 +298,7 @@ static void yy_init_buffer (YY_BUFFER_STATE b,FILE *file ); YY_BUFFER_STATE yy_scan_buffer (char *base,yy_size_t size ); YY_BUFFER_STATE yy_scan_string (yyconst char *yy_str ); -YY_BUFFER_STATE yy_scan_bytes (yyconst char *bytes,yy_size_t len ); +YY_BUFFER_STATE yy_scan_bytes (yyconst char *bytes,int len ); void *yyalloc (yy_size_t ); void *yyrealloc (void *,yy_size_t ); @@ -342,12 +330,12 @@ void yyfree (void * ); /* Begin user sect3 */ -#define yywrap() 1 +#define yywrap() (/*CONSTCOND*/1) #define YY_SKIP_YYWRAP typedef unsigned char YY_CHAR; -FILE *yyin = (FILE *) 0, *yyout = (FILE *) 0; +FILE *yyin = NULL, *yyout = NULL; typedef int yy_state_type; @@ -356,25 +344,28 @@ extern int yylineno; int yylineno = 1; extern char *yytext; +#ifdef yytext_ptr +#undef yytext_ptr +#endif #define yytext_ptr yytext static yy_state_type yy_get_previous_state (void ); static yy_state_type yy_try_NUL_trans (yy_state_type current_state ); static int yy_get_next_buffer (void ); -static void yy_fatal_error (yyconst char msg[] ); +static void yynoreturn yy_fatal_error (yyconst char* msg ); /* Done after the current pattern has been matched and before the * corresponding action - sets up yytext. */ #define YY_DO_BEFORE_ACTION \ (yytext_ptr) = yy_bp; \ - yyleng = (size_t) (yy_cp - yy_bp); \ + yyleng = (int) (yy_cp - yy_bp); \ (yy_hold_char) = *yy_cp; \ *yy_cp = '\0'; \ (yy_c_buf_p) = yy_cp; -#define YY_NUM_RULES 30 -#define YY_END_OF_BUFFER 31 +#define YY_NUM_RULES 31 +#define YY_END_OF_BUFFER 32 /* This struct is not used in this scanner, but its presence is necessary. */ struct yy_trans_info @@ -382,28 +373,29 @@ struct yy_trans_info flex_int32_t yy_verify; flex_int32_t yy_nxt; }; -static yyconst flex_int16_t yy_accept[159] = +static yyconst flex_int16_t yy_accept[166] = { 0, - 0, 0, 0, 0, 0, 0, 0, 0, 31, 29, - 18, 18, 29, 29, 29, 29, 29, 29, 29, 29, - 29, 29, 29, 29, 29, 29, 15, 16, 16, 29, - 16, 10, 10, 18, 26, 0, 3, 0, 27, 12, - 0, 0, 11, 0, 0, 0, 0, 0, 0, 0, - 21, 23, 25, 24, 22, 0, 9, 28, 0, 0, - 0, 14, 14, 16, 16, 16, 10, 10, 10, 0, - 12, 0, 11, 0, 0, 0, 20, 0, 0, 0, - 0, 0, 0, 0, 0, 16, 10, 10, 10, 0, - 13, 19, 0, 0, 0, 0, 0, 0, 0, 0, - - 0, 16, 0, 0, 0, 0, 0, 0, 0, 0, - 0, 16, 6, 0, 0, 0, 0, 0, 0, 2, - 0, 0, 0, 0, 0, 0, 0, 0, 4, 17, - 0, 0, 2, 0, 0, 0, 0, 0, 0, 0, - 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, - 5, 8, 0, 0, 0, 0, 7, 0 + 0, 0, 0, 0, 0, 0, 0, 0, 32, 30, + 19, 19, 30, 30, 30, 30, 30, 30, 30, 30, + 30, 30, 30, 30, 30, 30, 16, 17, 17, 30, + 17, 11, 11, 19, 27, 0, 3, 0, 28, 13, + 0, 0, 12, 0, 0, 0, 0, 0, 0, 0, + 0, 22, 24, 26, 25, 23, 0, 10, 29, 0, + 0, 0, 15, 15, 17, 17, 17, 11, 11, 11, + 0, 13, 0, 12, 0, 0, 0, 21, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 17, 11, 11, + 11, 0, 14, 20, 0, 0, 0, 0, 0, 0, + + 0, 0, 0, 0, 17, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 17, 7, 0, 0, 0, + 0, 0, 0, 0, 2, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 4, 18, 0, 0, 5, 2, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 1, 0, 0, 0, 0, 6, 9, 0, + 0, 0, 0, 8, 0 } ; -static yyconst flex_int32_t yy_ec[256] = +static yyconst YY_CHAR yy_ec[256] = { 0, 1, 1, 1, 1, 1, 1, 1, 1, 2, 3, 4, 4, 4, 1, 1, 1, 1, 1, 1, 1, @@ -416,9 +408,9 @@ static yyconst flex_int32_t yy_ec[256] = 22, 22, 22, 22, 24, 22, 22, 25, 22, 22, 1, 26, 27, 1, 22, 1, 21, 28, 29, 30, - 31, 21, 22, 22, 32, 22, 22, 33, 34, 35, - 36, 37, 22, 38, 39, 40, 41, 42, 22, 25, - 43, 22, 44, 45, 46, 1, 1, 1, 1, 1, + 31, 21, 32, 22, 33, 22, 22, 34, 35, 36, + 37, 38, 22, 39, 40, 41, 42, 43, 22, 25, + 44, 22, 45, 46, 47, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, @@ -435,163 +427,165 @@ static yyconst flex_int32_t yy_ec[256] = 1, 1, 1, 1, 1 } ; -static yyconst flex_int32_t yy_meta[47] = +static yyconst YY_CHAR yy_meta[48] = { 0, 1, 1, 1, 1, 1, 1, 2, 3, 1, 2, 2, 2, 4, 5, 5, 5, 6, 1, 1, 1, 7, 8, 8, 8, 8, 1, 1, 7, 7, 7, 7, 8, 8, 8, 8, 8, 8, 8, 8, 8, - 8, 8, 8, 3, 1, 4 + 8, 8, 8, 8, 3, 1, 4 } ; -static yyconst flex_int16_t yy_base[173] = +static yyconst flex_uint16_t yy_base[180] = { 0, - 0, 383, 34, 382, 65, 381, 37, 105, 387, 391, - 54, 111, 367, 110, 109, 109, 112, 41, 366, 104, - 367, 338, 124, 117, 0, 144, 391, 0, 121, 0, - 135, 155, 140, 179, 391, 160, 391, 379, 391, 0, - 368, 141, 391, 167, 370, 376, 346, 103, 342, 345, - 391, 391, 391, 391, 391, 358, 391, 391, 175, 342, - 338, 391, 355, 0, 185, 339, 184, 347, 346, 0, - 0, 322, 175, 357, 175, 363, 352, 324, 330, 323, - 332, 326, 201, 324, 329, 322, 391, 333, 181, 309, - 391, 341, 340, 313, 320, 338, 178, 311, 146, 317, - - 314, 315, 335, 331, 303, 300, 309, 299, 308, 188, - 336, 335, 391, 305, 320, 281, 283, 271, 203, 288, - 281, 271, 266, 264, 245, 242, 208, 104, 391, 391, - 244, 218, 204, 219, 206, 224, 201, 212, 204, 229, - 215, 208, 207, 200, 219, 391, 233, 221, 200, 181, - 391, 391, 149, 122, 86, 41, 391, 391, 245, 251, - 259, 263, 267, 273, 280, 284, 292, 300, 304, 310, - 318, 326 + 0, 393, 35, 392, 66, 391, 38, 107, 397, 401, + 55, 113, 377, 112, 111, 111, 114, 42, 376, 106, + 377, 347, 126, 120, 0, 147, 401, 0, 124, 0, + 137, 158, 170, 163, 401, 153, 401, 389, 401, 0, + 378, 120, 401, 131, 380, 386, 355, 139, 351, 355, + 351, 401, 401, 401, 401, 401, 367, 401, 401, 185, + 350, 346, 401, 364, 0, 185, 347, 189, 356, 355, + 0, 0, 330, 180, 366, 141, 372, 361, 332, 338, + 331, 341, 334, 326, 205, 331, 337, 329, 401, 341, + 167, 316, 401, 349, 348, 320, 328, 346, 180, 318, + + 324, 209, 324, 320, 322, 342, 338, 309, 306, 315, + 305, 315, 312, 192, 342, 341, 401, 293, 306, 282, + 268, 252, 255, 203, 285, 282, 272, 268, 252, 233, + 232, 239, 208, 107, 401, 401, 238, 211, 401, 211, + 212, 208, 228, 203, 215, 207, 233, 222, 212, 211, + 203, 227, 401, 237, 225, 204, 185, 401, 401, 149, + 128, 88, 42, 401, 401, 253, 259, 267, 271, 275, + 281, 288, 292, 300, 308, 312, 318, 326, 334 } ; -static yyconst flex_int16_t yy_def[173] = +static yyconst flex_int16_t yy_def[180] = { 0, - 158, 1, 1, 3, 158, 5, 1, 1, 158, 158, - 158, 158, 158, 159, 160, 161, 158, 158, 158, 158, - 162, 158, 158, 158, 163, 162, 158, 164, 165, 164, - 164, 158, 158, 158, 158, 159, 158, 159, 158, 166, - 158, 161, 158, 161, 167, 168, 158, 158, 158, 158, - 158, 158, 158, 158, 158, 162, 158, 158, 158, 158, - 158, 158, 162, 164, 165, 164, 158, 158, 158, 169, - 166, 170, 161, 167, 167, 168, 158, 158, 158, 158, - 158, 158, 158, 158, 158, 164, 158, 158, 169, 170, - 158, 158, 158, 158, 158, 158, 158, 158, 158, 158, - - 158, 164, 158, 158, 158, 158, 158, 158, 158, 171, - 158, 164, 158, 158, 158, 158, 158, 158, 171, 158, - 171, 158, 158, 158, 158, 158, 158, 158, 158, 158, - 158, 158, 158, 158, 158, 158, 158, 158, 158, 158, - 172, 158, 158, 158, 172, 158, 172, 158, 158, 158, - 158, 158, 158, 158, 158, 158, 158, 0, 158, 158, - 158, 158, 158, 158, 158, 158, 158, 158, 158, 158, - 158, 158 + 165, 1, 1, 3, 165, 5, 1, 1, 165, 165, + 165, 165, 165, 166, 167, 168, 165, 165, 165, 165, + 169, 165, 165, 165, 170, 169, 165, 171, 172, 171, + 171, 165, 165, 165, 165, 166, 165, 166, 165, 173, + 165, 168, 165, 168, 174, 175, 165, 165, 165, 165, + 165, 165, 165, 165, 165, 165, 169, 165, 165, 165, + 165, 165, 165, 169, 171, 172, 171, 165, 165, 165, + 176, 173, 177, 168, 174, 174, 175, 165, 165, 165, + 165, 165, 165, 165, 165, 165, 165, 171, 165, 165, + 176, 177, 165, 165, 165, 165, 165, 165, 165, 165, + + 165, 165, 165, 165, 171, 165, 165, 165, 165, 165, + 165, 165, 165, 178, 165, 171, 165, 165, 165, 165, + 165, 165, 165, 178, 165, 178, 165, 165, 165, 165, + 165, 165, 165, 165, 165, 165, 165, 165, 165, 165, + 165, 165, 165, 165, 165, 165, 165, 179, 165, 165, + 165, 179, 165, 179, 165, 165, 165, 165, 165, 165, + 165, 165, 165, 165, 0, 165, 165, 165, 165, 165, + 165, 165, 165, 165, 165, 165, 165, 165, 165 } ; -static yyconst flex_int16_t yy_nxt[438] = +static yyconst flex_uint16_t yy_nxt[449] = { 0, 10, 11, 12, 11, 13, 14, 10, 15, 16, 10, 10, 10, 17, 10, 10, 10, 10, 18, 19, 20, 21, 21, 21, 21, 21, 10, 10, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, - 21, 21, 21, 10, 22, 10, 24, 25, 25, 25, - 32, 33, 33, 157, 26, 34, 34, 34, 51, 52, - 27, 26, 26, 26, 26, 10, 11, 12, 11, 13, - 14, 28, 15, 16, 28, 28, 28, 24, 28, 28, - 28, 10, 18, 19, 20, 29, 29, 29, 29, 29, - 30, 10, 29, 29, 29, 29, 29, 29, 29, 29, - - 29, 29, 29, 29, 29, 29, 29, 29, 10, 22, - 10, 23, 34, 34, 34, 37, 39, 43, 32, 33, - 33, 45, 54, 55, 46, 59, 45, 64, 156, 46, - 64, 64, 64, 79, 44, 38, 59, 57, 134, 47, - 135, 48, 80, 49, 47, 50, 48, 99, 61, 43, - 50, 110, 41, 67, 67, 67, 60, 63, 63, 63, - 57, 155, 68, 69, 63, 37, 44, 66, 67, 67, - 67, 63, 63, 63, 63, 73, 59, 68, 69, 70, - 34, 34, 34, 43, 75, 38, 154, 92, 83, 83, - 83, 64, 44, 120, 64, 64, 64, 67, 67, 67, - - 44, 57, 99, 68, 69, 107, 68, 69, 120, 127, - 108, 153, 152, 121, 83, 83, 83, 133, 133, 133, - 146, 133, 133, 133, 146, 140, 140, 140, 121, 141, - 140, 140, 140, 151, 141, 158, 150, 149, 148, 144, - 147, 143, 142, 139, 147, 36, 36, 36, 36, 36, - 36, 36, 36, 40, 138, 137, 136, 40, 40, 42, - 42, 42, 42, 42, 42, 42, 42, 56, 56, 56, - 56, 62, 132, 62, 64, 131, 130, 64, 129, 64, - 64, 65, 128, 158, 65, 65, 65, 65, 71, 127, - 71, 71, 74, 74, 74, 74, 74, 74, 74, 74, - - 76, 76, 76, 76, 76, 76, 76, 76, 89, 126, - 89, 90, 125, 90, 90, 124, 90, 90, 119, 119, - 119, 119, 119, 119, 119, 119, 145, 145, 145, 145, - 145, 145, 145, 145, 123, 122, 59, 59, 118, 117, - 116, 115, 114, 113, 45, 112, 108, 111, 109, 106, - 105, 104, 46, 103, 91, 87, 102, 101, 100, 98, - 97, 96, 95, 94, 93, 77, 75, 91, 88, 87, - 86, 57, 85, 84, 57, 82, 81, 78, 77, 75, - 72, 158, 58, 57, 53, 35, 158, 31, 23, 23, - 9, 158, 158, 158, 158, 158, 158, 158, 158, 158, - - 158, 158, 158, 158, 158, 158, 158, 158, 158, 158, - 158, 158, 158, 158, 158, 158, 158, 158, 158, 158, - 158, 158, 158, 158, 158, 158, 158, 158, 158, 158, - 158, 158, 158, 158, 158, 158, 158 + 21, 21, 21, 21, 10, 22, 10, 24, 25, 25, + 25, 32, 33, 33, 164, 26, 34, 34, 34, 52, + 53, 27, 26, 26, 26, 26, 10, 11, 12, 11, + 13, 14, 28, 15, 16, 28, 28, 28, 24, 28, + 28, 28, 10, 18, 19, 20, 29, 29, 29, 29, + 29, 30, 10, 29, 29, 29, 29, 29, 29, 29, + + 29, 29, 29, 29, 29, 29, 29, 29, 29, 29, + 10, 22, 10, 23, 34, 34, 34, 37, 39, 43, + 32, 33, 33, 45, 55, 56, 46, 60, 43, 45, + 65, 163, 46, 65, 65, 65, 44, 38, 60, 74, + 58, 47, 141, 48, 142, 44, 49, 47, 50, 48, + 76, 51, 62, 94, 50, 41, 44, 51, 37, 61, + 64, 64, 64, 58, 34, 34, 34, 64, 162, 80, + 67, 68, 68, 68, 64, 64, 64, 64, 38, 81, + 69, 70, 71, 68, 68, 68, 60, 161, 43, 69, + 70, 65, 69, 70, 65, 65, 65, 125, 85, 85, + + 85, 58, 68, 68, 68, 44, 102, 110, 125, 133, + 102, 69, 70, 111, 114, 160, 159, 126, 85, 85, + 85, 140, 140, 140, 140, 140, 140, 153, 126, 147, + 147, 147, 153, 148, 147, 147, 147, 158, 148, 165, + 157, 156, 155, 151, 150, 149, 146, 154, 145, 144, + 143, 139, 154, 36, 36, 36, 36, 36, 36, 36, + 36, 40, 138, 137, 136, 40, 40, 42, 42, 42, + 42, 42, 42, 42, 42, 57, 57, 57, 57, 63, + 135, 63, 65, 134, 165, 65, 133, 65, 65, 66, + 132, 131, 66, 66, 66, 66, 72, 130, 72, 72, + + 75, 75, 75, 75, 75, 75, 75, 75, 77, 77, + 77, 77, 77, 77, 77, 77, 91, 129, 91, 92, + 128, 92, 92, 127, 92, 92, 124, 124, 124, 124, + 124, 124, 124, 124, 152, 152, 152, 152, 152, 152, + 152, 152, 60, 60, 123, 122, 121, 120, 119, 118, + 117, 45, 116, 111, 115, 113, 112, 109, 108, 107, + 46, 106, 93, 89, 105, 104, 103, 101, 100, 99, + 98, 97, 96, 95, 78, 76, 93, 90, 89, 88, + 58, 87, 86, 58, 84, 83, 82, 79, 78, 76, + 73, 165, 59, 58, 54, 35, 165, 31, 23, 23, + + 9, 165, 165, 165, 165, 165, 165, 165, 165, 165, + 165, 165, 165, 165, 165, 165, 165, 165, 165, 165, + 165, 165, 165, 165, 165, 165, 165, 165, 165, 165, + 165, 165, 165, 165, 165, 165, 165, 165, 165, 165, + 165, 165, 165, 165, 165, 165, 165, 165 } ; -static yyconst flex_int16_t yy_chk[438] = +static yyconst flex_int16_t yy_chk[449] = { 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, - 1, 1, 1, 1, 1, 1, 3, 3, 3, 3, - 7, 7, 7, 156, 3, 11, 11, 11, 18, 18, - 3, 3, 3, 3, 3, 5, 5, 5, 5, 5, + 1, 1, 1, 1, 1, 1, 1, 3, 3, 3, + 3, 7, 7, 7, 163, 3, 11, 11, 11, 18, + 18, 3, 3, 3, 3, 3, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, - 5, 8, 12, 12, 12, 14, 15, 16, 8, 8, - 8, 17, 20, 20, 17, 23, 24, 29, 155, 24, - 29, 29, 29, 48, 16, 14, 31, 29, 128, 17, - 128, 17, 48, 17, 24, 17, 24, 99, 24, 42, - 24, 99, 15, 33, 33, 33, 23, 26, 26, 26, - 26, 154, 33, 33, 26, 36, 42, 31, 32, 32, - 32, 26, 26, 26, 26, 44, 59, 32, 32, 32, - 34, 34, 34, 73, 75, 36, 153, 75, 59, 59, - 59, 65, 44, 110, 65, 65, 65, 67, 67, 67, - - 73, 65, 83, 89, 89, 97, 67, 67, 119, 127, - 97, 150, 149, 110, 83, 83, 83, 133, 133, 133, - 141, 127, 127, 127, 145, 136, 136, 136, 119, 136, - 140, 140, 140, 148, 140, 147, 144, 143, 142, 139, - 141, 138, 137, 135, 145, 159, 159, 159, 159, 159, - 159, 159, 159, 160, 134, 132, 131, 160, 160, 161, - 161, 161, 161, 161, 161, 161, 161, 162, 162, 162, - 162, 163, 126, 163, 164, 125, 124, 164, 123, 164, - 164, 165, 122, 121, 165, 165, 165, 165, 166, 120, - 166, 166, 167, 167, 167, 167, 167, 167, 167, 167, - - 168, 168, 168, 168, 168, 168, 168, 168, 169, 118, - 169, 170, 117, 170, 170, 116, 170, 170, 171, 171, - 171, 171, 171, 171, 171, 171, 172, 172, 172, 172, - 172, 172, 172, 172, 115, 114, 112, 111, 109, 108, - 107, 106, 105, 104, 103, 102, 101, 100, 98, 96, - 95, 94, 93, 92, 90, 88, 86, 85, 84, 82, - 81, 80, 79, 78, 77, 76, 74, 72, 69, 68, - 66, 63, 61, 60, 56, 50, 49, 47, 46, 45, + 5, 5, 5, 8, 12, 12, 12, 14, 15, 16, + 8, 8, 8, 17, 20, 20, 17, 23, 42, 24, + 29, 162, 24, 29, 29, 29, 16, 14, 31, 44, + 29, 17, 134, 17, 134, 42, 17, 24, 17, 24, + 76, 17, 24, 76, 24, 15, 44, 24, 36, 23, + 26, 26, 26, 26, 34, 34, 34, 26, 161, 48, + 31, 32, 32, 32, 26, 26, 26, 26, 36, 48, + 32, 32, 32, 33, 33, 33, 60, 160, 74, 91, + 91, 66, 33, 33, 66, 66, 66, 114, 60, 60, + + 60, 66, 68, 68, 68, 74, 85, 99, 124, 133, + 102, 68, 68, 99, 102, 157, 156, 114, 85, 85, + 85, 133, 133, 133, 140, 140, 140, 148, 124, 143, + 143, 143, 152, 143, 147, 147, 147, 155, 147, 154, + 151, 150, 149, 146, 145, 144, 142, 148, 141, 138, + 137, 132, 152, 166, 166, 166, 166, 166, 166, 166, + 166, 167, 131, 130, 129, 167, 167, 168, 168, 168, + 168, 168, 168, 168, 168, 169, 169, 169, 169, 170, + 128, 170, 171, 127, 126, 171, 125, 171, 171, 172, + 123, 122, 172, 172, 172, 172, 173, 121, 173, 173, + + 174, 174, 174, 174, 174, 174, 174, 174, 175, 175, + 175, 175, 175, 175, 175, 175, 176, 120, 176, 177, + 119, 177, 177, 118, 177, 177, 178, 178, 178, 178, + 178, 178, 178, 178, 179, 179, 179, 179, 179, 179, + 179, 179, 116, 115, 113, 112, 111, 110, 109, 108, + 107, 106, 105, 104, 103, 101, 100, 98, 97, 96, + 95, 94, 92, 90, 88, 87, 86, 84, 83, 82, + 81, 80, 79, 78, 77, 75, 73, 70, 69, 67, + 64, 62, 61, 57, 51, 50, 49, 47, 46, 45, 41, 38, 22, 21, 19, 13, 9, 6, 4, 2, - 158, 158, 158, 158, 158, 158, 158, 158, 158, 158, - 158, 158, 158, 158, 158, 158, 158, 158, 158, 158, - 158, 158, 158, 158, 158, 158, 158, 158, 158, 158, - 158, 158, 158, 158, 158, 158, 158, 158, 158, 158, - 158, 158, 158, 158, 158, 158, 158 + 165, 165, 165, 165, 165, 165, 165, 165, 165, 165, + 165, 165, 165, 165, 165, 165, 165, 165, 165, 165, + 165, 165, 165, 165, 165, 165, 165, 165, 165, 165, + 165, 165, 165, 165, 165, 165, 165, 165, 165, 165, + 165, 165, 165, 165, 165, 165, 165, 165 } ; static yy_state_type yy_last_accepting_state; @@ -662,7 +656,7 @@ static int dts_version = 1; static void push_input_file(const char *filename); static bool pop_input_file(void); static void lexical_error(const char *fmt, ...); -#line 666 "dtc-lexer.lex.c" +#line 660 "dtc-lexer.lex.c" #define INITIAL 0 #define BYTESTRING 1 @@ -698,19 +692,19 @@ void yyset_extra (YY_EXTRA_TYPE user_defined ); FILE *yyget_in (void ); -void yyset_in (FILE * in_str ); +void yyset_in (FILE * _in_str ); FILE *yyget_out (void ); -void yyset_out (FILE * out_str ); +void yyset_out (FILE * _out_str ); -yy_size_t yyget_leng (void ); + int yyget_leng (void ); char *yyget_text (void ); int yyget_lineno (void ); -void yyset_lineno (int line_number ); +void yyset_lineno (int _line_number ); /* Macros after this point can all be overridden by user definitions in * section 1. @@ -724,6 +718,10 @@ extern int yywrap (void ); #endif #endif +#ifndef YY_NO_UNPUT + +#endif + #ifndef yytext_ptr static void yy_flex_strncpy (char *,yyconst char *,int ); #endif @@ -757,7 +755,7 @@ static int input (void ); /* This used to be an fputs(), but since the string might contain NUL's, * we now use fwrite(). */ -#define ECHO do { if (fwrite( yytext, yyleng, 1, yyout )) {} } while (0) +#define ECHO do { if (fwrite( yytext, (size_t) yyleng, 1, yyout )) {} } while (0) #endif /* Gets input and stuffs it into "buf". number of characters read, or YY_NULL, @@ -781,7 +779,7 @@ static int input (void ); else \ { \ errno=0; \ - while ( (result = fread(buf, 1, max_size, yyin))==0 && ferror(yyin)) \ + while ( (result = (int) fread(buf, 1, max_size, yyin))==0 && ferror(yyin)) \ { \ if( errno != EINTR) \ { \ @@ -836,7 +834,7 @@ extern int yylex (void); /* Code executed at the end of each rule. */ #ifndef YY_BREAK -#define YY_BREAK break; +#define YY_BREAK /*LINTED*/break; #endif #define YY_RULE_SETUP \ @@ -849,9 +847,9 @@ extern int yylex (void); */ YY_DECL { - register yy_state_type yy_current_state; - register char *yy_cp, *yy_bp; - register int yy_act; + yy_state_type yy_current_state; + char *yy_cp, *yy_bp; + int yy_act; if ( !(yy_init) ) { @@ -882,9 +880,9 @@ YY_DECL { #line 68 "dtc-lexer.l" -#line 886 "dtc-lexer.lex.c" +#line 884 "dtc-lexer.lex.c" - while ( 1 ) /* loops until end-of-file is reached */ + while ( /*CONSTCOND*/1 ) /* loops until end-of-file is reached */ { yy_cp = (yy_c_buf_p); @@ -901,7 +899,7 @@ YY_DECL yy_match: do { - register YY_CHAR yy_c = yy_ec[YY_SC_TO_UI(*yy_cp)] ; + YY_CHAR yy_c = yy_ec[YY_SC_TO_UI(*yy_cp)] ; if ( yy_accept[yy_current_state] ) { (yy_last_accepting_state) = yy_current_state; @@ -910,13 +908,13 @@ yy_match: while ( yy_chk[yy_base[yy_current_state] + yy_c] != yy_current_state ) { yy_current_state = (int) yy_def[yy_current_state]; - if ( yy_current_state >= 159 ) + if ( yy_current_state >= 166 ) yy_c = yy_meta[(unsigned int) yy_c]; } - yy_current_state = yy_nxt[yy_base[yy_current_state] + (unsigned int) yy_c]; + yy_current_state = yy_nxt[yy_base[yy_current_state] + (flex_int16_t) yy_c]; ++yy_cp; } - while ( yy_current_state != 158 ); + while ( yy_current_state != 165 ); yy_cp = (yy_last_accepting_cpos); yy_current_state = (yy_last_accepting_state); @@ -1015,23 +1013,31 @@ case 5: YY_RULE_SETUP #line 124 "dtc-lexer.l" { + DPRINT("Keyword: /plugin/\n"); + return DT_PLUGIN; + } + YY_BREAK +case 6: +YY_RULE_SETUP +#line 129 "dtc-lexer.l" +{ DPRINT("Keyword: /memreserve/\n"); BEGIN_DEFAULT(); return DT_MEMRESERVE; } YY_BREAK -case 6: +case 7: YY_RULE_SETUP -#line 130 "dtc-lexer.l" +#line 135 "dtc-lexer.l" { DPRINT("Keyword: /bits/\n"); BEGIN_DEFAULT(); return DT_BITS; } YY_BREAK -case 7: +case 8: YY_RULE_SETUP -#line 136 "dtc-lexer.l" +#line 141 "dtc-lexer.l" { DPRINT("Keyword: /delete-property/\n"); DPRINT("<PROPNODENAME>\n"); @@ -1039,9 +1045,9 @@ YY_RULE_SETUP return DT_DEL_PROP; } YY_BREAK -case 8: +case 9: YY_RULE_SETUP -#line 143 "dtc-lexer.l" +#line 148 "dtc-lexer.l" { DPRINT("Keyword: /delete-node/\n"); DPRINT("<PROPNODENAME>\n"); @@ -1049,9 +1055,9 @@ YY_RULE_SETUP return DT_DEL_NODE; } YY_BREAK -case 9: +case 10: YY_RULE_SETUP -#line 150 "dtc-lexer.l" +#line 155 "dtc-lexer.l" { DPRINT("Label: %s\n", yytext); yylval.labelref = xstrdup(yytext); @@ -1059,9 +1065,9 @@ YY_RULE_SETUP return DT_LABEL; } YY_BREAK -case 10: +case 11: YY_RULE_SETUP -#line 157 "dtc-lexer.l" +#line 162 "dtc-lexer.l" { char *e; DPRINT("Integer Literal: '%s'\n", yytext); @@ -1084,10 +1090,10 @@ YY_RULE_SETUP return DT_LITERAL; } YY_BREAK -case 11: -/* rule 11 can match eol */ +case 12: +/* rule 12 can match eol */ YY_RULE_SETUP -#line 179 "dtc-lexer.l" +#line 184 "dtc-lexer.l" { struct data d; DPRINT("Character literal: %s\n", yytext); @@ -1096,31 +1102,31 @@ YY_RULE_SETUP if (d.len == 1) { lexical_error("Empty character literal"); yylval.integer = 0; - return DT_CHAR_LITERAL; - } - - yylval.integer = (unsigned char)d.val[0]; + } else { + yylval.integer = (unsigned char)d.val[0]; - if (d.len > 2) - lexical_error("Character literal has %d" - " characters instead of 1", - d.len - 1); + if (d.len > 2) + lexical_error("Character literal has %d" + " characters instead of 1", + d.len - 1); + } + data_free(d); return DT_CHAR_LITERAL; } YY_BREAK -case 12: +case 13: YY_RULE_SETUP -#line 200 "dtc-lexer.l" +#line 205 "dtc-lexer.l" { /* label reference */ DPRINT("Ref: %s\n", yytext+1); yylval.labelref = xstrdup(yytext+1); return DT_REF; } YY_BREAK -case 13: +case 14: YY_RULE_SETUP -#line 206 "dtc-lexer.l" +#line 211 "dtc-lexer.l" { /* new-style path reference */ yytext[yyleng-1] = '\0'; DPRINT("Ref: %s\n", yytext+2); @@ -1128,27 +1134,27 @@ YY_RULE_SETUP return DT_REF; } YY_BREAK -case 14: +case 15: YY_RULE_SETUP -#line 213 "dtc-lexer.l" +#line 218 "dtc-lexer.l" { yylval.byte = strtol(yytext, NULL, 16); DPRINT("Byte: %02x\n", (int)yylval.byte); return DT_BYTE; } YY_BREAK -case 15: +case 16: YY_RULE_SETUP -#line 219 "dtc-lexer.l" +#line 224 "dtc-lexer.l" { DPRINT("/BYTESTRING\n"); BEGIN_DEFAULT(); return ']'; } YY_BREAK -case 16: +case 17: YY_RULE_SETUP -#line 225 "dtc-lexer.l" +#line 230 "dtc-lexer.l" { DPRINT("PropNodeName: %s\n", yytext); yylval.propnodename = xstrdup((yytext[0] == '\\') ? @@ -1157,75 +1163,75 @@ YY_RULE_SETUP return DT_PROPNODENAME; } YY_BREAK -case 17: +case 18: YY_RULE_SETUP -#line 233 "dtc-lexer.l" +#line 238 "dtc-lexer.l" { DPRINT("Binary Include\n"); return DT_INCBIN; } YY_BREAK -case 18: -/* rule 18 can match eol */ -YY_RULE_SETUP -#line 238 "dtc-lexer.l" -/* eat whitespace */ - YY_BREAK case 19: /* rule 19 can match eol */ YY_RULE_SETUP -#line 239 "dtc-lexer.l" -/* eat C-style comments */ +#line 243 "dtc-lexer.l" +/* eat whitespace */ YY_BREAK case 20: /* rule 20 can match eol */ YY_RULE_SETUP -#line 240 "dtc-lexer.l" -/* eat C++-style comments */ +#line 244 "dtc-lexer.l" +/* eat C-style comments */ YY_BREAK case 21: +/* rule 21 can match eol */ YY_RULE_SETUP -#line 242 "dtc-lexer.l" -{ return DT_LSHIFT; }; +#line 245 "dtc-lexer.l" +/* eat C++-style comments */ YY_BREAK case 22: YY_RULE_SETUP -#line 243 "dtc-lexer.l" -{ return DT_RSHIFT; }; +#line 247 "dtc-lexer.l" +{ return DT_LSHIFT; }; YY_BREAK case 23: YY_RULE_SETUP -#line 244 "dtc-lexer.l" -{ return DT_LE; }; +#line 248 "dtc-lexer.l" +{ return DT_RSHIFT; }; YY_BREAK case 24: YY_RULE_SETUP -#line 245 "dtc-lexer.l" -{ return DT_GE; }; +#line 249 "dtc-lexer.l" +{ return DT_LE; }; YY_BREAK case 25: YY_RULE_SETUP -#line 246 "dtc-lexer.l" -{ return DT_EQ; }; +#line 250 "dtc-lexer.l" +{ return DT_GE; }; YY_BREAK case 26: YY_RULE_SETUP -#line 247 "dtc-lexer.l" -{ return DT_NE; }; +#line 251 "dtc-lexer.l" +{ return DT_EQ; }; YY_BREAK case 27: YY_RULE_SETUP -#line 248 "dtc-lexer.l" -{ return DT_AND; }; +#line 252 "dtc-lexer.l" +{ return DT_NE; }; YY_BREAK case 28: YY_RULE_SETUP -#line 249 "dtc-lexer.l" -{ return DT_OR; }; +#line 253 "dtc-lexer.l" +{ return DT_AND; }; YY_BREAK case 29: YY_RULE_SETUP -#line 251 "dtc-lexer.l" +#line 254 "dtc-lexer.l" +{ return DT_OR; }; + YY_BREAK +case 30: +YY_RULE_SETUP +#line 256 "dtc-lexer.l" { DPRINT("Char: %c (\\x%02x)\n", yytext[0], (unsigned)yytext[0]); @@ -1241,12 +1247,12 @@ YY_RULE_SETUP return yytext[0]; } YY_BREAK -case 30: +case 31: YY_RULE_SETUP -#line 266 "dtc-lexer.l" +#line 271 "dtc-lexer.l" ECHO; YY_BREAK -#line 1250 "dtc-lexer.lex.c" +#line 1256 "dtc-lexer.lex.c" case YY_END_OF_BUFFER: { @@ -1388,9 +1394,9 @@ ECHO; */ static int yy_get_next_buffer (void) { - register char *dest = YY_CURRENT_BUFFER_LVALUE->yy_ch_buf; - register char *source = (yytext_ptr); - register int number_to_move, i; + char *dest = YY_CURRENT_BUFFER_LVALUE->yy_ch_buf; + char *source = (yytext_ptr); + yy_size_t number_to_move, i; int ret_val; if ( (yy_c_buf_p) > &YY_CURRENT_BUFFER_LVALUE->yy_ch_buf[(yy_n_chars) + 1] ) @@ -1419,7 +1425,7 @@ static int yy_get_next_buffer (void) /* Try to read more data. */ /* First move last chars to start of buffer. */ - number_to_move = (int) ((yy_c_buf_p) - (yytext_ptr)) - 1; + number_to_move = (yy_size_t) ((yy_c_buf_p) - (yytext_ptr)) - 1; for ( i = 0; i < number_to_move; ++i ) *(dest++) = *(source++); @@ -1432,7 +1438,7 @@ static int yy_get_next_buffer (void) else { - yy_size_t num_to_read = + int num_to_read = YY_CURRENT_BUFFER_LVALUE->yy_buf_size - number_to_move - 1; while ( num_to_read <= 0 ) @@ -1446,7 +1452,7 @@ static int yy_get_next_buffer (void) if ( b->yy_is_our_buffer ) { - yy_size_t new_size = b->yy_buf_size * 2; + int new_size = b->yy_buf_size * 2; if ( new_size <= 0 ) b->yy_buf_size += b->yy_buf_size / 8; @@ -1459,7 +1465,7 @@ static int yy_get_next_buffer (void) } else /* Can't grow it, we don't own it. */ - b->yy_ch_buf = 0; + b->yy_ch_buf = NULL; if ( ! b->yy_ch_buf ) YY_FATAL_ERROR( @@ -1501,9 +1507,9 @@ static int yy_get_next_buffer (void) else ret_val = EOB_ACT_CONTINUE_SCAN; - if ((yy_size_t) ((yy_n_chars) + number_to_move) > YY_CURRENT_BUFFER_LVALUE->yy_buf_size) { + if ((int) ((yy_n_chars) + number_to_move) > YY_CURRENT_BUFFER_LVALUE->yy_buf_size) { /* Extend the array by 50%, plus the number we really need. */ - yy_size_t new_size = (yy_n_chars) + number_to_move + ((yy_n_chars) >> 1); + int new_size = (yy_n_chars) + number_to_move + ((yy_n_chars) >> 1); YY_CURRENT_BUFFER_LVALUE->yy_ch_buf = (char *) yyrealloc((void *) YY_CURRENT_BUFFER_LVALUE->yy_ch_buf,new_size ); if ( ! YY_CURRENT_BUFFER_LVALUE->yy_ch_buf ) YY_FATAL_ERROR( "out of dynamic memory in yy_get_next_buffer()" ); @@ -1522,15 +1528,15 @@ static int yy_get_next_buffer (void) static yy_state_type yy_get_previous_state (void) { - register yy_state_type yy_current_state; - register char *yy_cp; + yy_state_type yy_current_state; + char *yy_cp; yy_current_state = (yy_start); yy_current_state += YY_AT_BOL(); for ( yy_cp = (yytext_ptr) + YY_MORE_ADJ; yy_cp < (yy_c_buf_p); ++yy_cp ) { - register YY_CHAR yy_c = (*yy_cp ? yy_ec[YY_SC_TO_UI(*yy_cp)] : 1); + YY_CHAR yy_c = (*yy_cp ? yy_ec[YY_SC_TO_UI(*yy_cp)] : 1); if ( yy_accept[yy_current_state] ) { (yy_last_accepting_state) = yy_current_state; @@ -1539,10 +1545,10 @@ static int yy_get_next_buffer (void) while ( yy_chk[yy_base[yy_current_state] + yy_c] != yy_current_state ) { yy_current_state = (int) yy_def[yy_current_state]; - if ( yy_current_state >= 159 ) + if ( yy_current_state >= 166 ) yy_c = yy_meta[(unsigned int) yy_c]; } - yy_current_state = yy_nxt[yy_base[yy_current_state] + (unsigned int) yy_c]; + yy_current_state = yy_nxt[yy_base[yy_current_state] + (flex_int16_t) yy_c]; } return yy_current_state; @@ -1555,10 +1561,10 @@ static int yy_get_next_buffer (void) */ static yy_state_type yy_try_NUL_trans (yy_state_type yy_current_state ) { - register int yy_is_jam; - register char *yy_cp = (yy_c_buf_p); + int yy_is_jam; + char *yy_cp = (yy_c_buf_p); - register YY_CHAR yy_c = 1; + YY_CHAR yy_c = 1; if ( yy_accept[yy_current_state] ) { (yy_last_accepting_state) = yy_current_state; @@ -1567,15 +1573,19 @@ static int yy_get_next_buffer (void) while ( yy_chk[yy_base[yy_current_state] + yy_c] != yy_current_state ) { yy_current_state = (int) yy_def[yy_current_state]; - if ( yy_current_state >= 159 ) + if ( yy_current_state >= 166 ) yy_c = yy_meta[(unsigned int) yy_c]; } - yy_current_state = yy_nxt[yy_base[yy_current_state] + (unsigned int) yy_c]; - yy_is_jam = (yy_current_state == 158); + yy_current_state = yy_nxt[yy_base[yy_current_state] + (flex_int16_t) yy_c]; + yy_is_jam = (yy_current_state == 165); return yy_is_jam ? 0 : yy_current_state; } +#ifndef YY_NO_UNPUT + +#endif + #ifndef YY_NO_INPUT #ifdef __cplusplus static int yyinput (void) @@ -1600,7 +1610,7 @@ static int yy_get_next_buffer (void) else { /* need more input */ - yy_size_t offset = (yy_c_buf_p) - (yytext_ptr); + int offset = (yy_c_buf_p) - (yytext_ptr); ++(yy_c_buf_p); switch ( yy_get_next_buffer( ) ) @@ -1624,7 +1634,7 @@ static int yy_get_next_buffer (void) case EOB_ACT_END_OF_FILE: { if ( yywrap( ) ) - return EOF; + return 0; if ( ! (yy_did_buffer_switch_on_eof) ) YY_NEW_FILE; @@ -1727,7 +1737,7 @@ static void yy_load_buffer_state (void) if ( ! b ) YY_FATAL_ERROR( "out of dynamic memory in yy_create_buffer()" ); - b->yy_buf_size = size; + b->yy_buf_size = (yy_size_t)size; /* yy_ch_buf has to be 2 characters longer than the size given because * we need to put in 2 end-of-buffer characters. @@ -1874,7 +1884,7 @@ void yypop_buffer_state (void) */ static void yyensure_buffer_stack (void) { - yy_size_t num_to_alloc; + int num_to_alloc; if (!(yy_buffer_stack)) { @@ -1882,15 +1892,15 @@ static void yyensure_buffer_stack (void) * scanner will even need a stack. We use 2 instead of 1 to avoid an * immediate realloc on the next call. */ - num_to_alloc = 1; + num_to_alloc = 1; /* After all that talk, this was set to 1 anyways... */ (yy_buffer_stack) = (struct yy_buffer_state**)yyalloc (num_to_alloc * sizeof(struct yy_buffer_state*) ); if ( ! (yy_buffer_stack) ) YY_FATAL_ERROR( "out of dynamic memory in yyensure_buffer_stack()" ); - + memset((yy_buffer_stack), 0, num_to_alloc * sizeof(struct yy_buffer_state*)); - + (yy_buffer_stack_max) = num_to_alloc; (yy_buffer_stack_top) = 0; return; @@ -1899,7 +1909,7 @@ static void yyensure_buffer_stack (void) if ((yy_buffer_stack_top) >= ((yy_buffer_stack_max)) - 1){ /* Increase the buffer to prepare for a possible push. */ - int grow_size = 8 /* arbitrary grow size */; + yy_size_t grow_size = 8 /* arbitrary grow size */; num_to_alloc = (yy_buffer_stack_max) + grow_size; (yy_buffer_stack) = (struct yy_buffer_state**)yyrealloc @@ -1919,7 +1929,7 @@ static void yyensure_buffer_stack (void) * @param base the character buffer * @param size the size in bytes of the character buffer * - * @return the newly allocated buffer state object. + * @return the newly allocated buffer state object. */ YY_BUFFER_STATE yy_scan_buffer (char * base, yy_size_t size ) { @@ -1929,7 +1939,7 @@ YY_BUFFER_STATE yy_scan_buffer (char * base, yy_size_t size ) base[size-2] != YY_END_OF_BUFFER_CHAR || base[size-1] != YY_END_OF_BUFFER_CHAR ) /* They forgot to leave room for the EOB's. */ - return 0; + return NULL; b = (YY_BUFFER_STATE) yyalloc(sizeof( struct yy_buffer_state ) ); if ( ! b ) @@ -1938,7 +1948,7 @@ YY_BUFFER_STATE yy_scan_buffer (char * base, yy_size_t size ) b->yy_buf_size = size - 2; /* "- 2" to take care of EOB's */ b->yy_buf_pos = b->yy_ch_buf = base; b->yy_is_our_buffer = 0; - b->yy_input_file = 0; + b->yy_input_file = NULL; b->yy_n_chars = b->yy_buf_size; b->yy_is_interactive = 0; b->yy_at_bol = 1; @@ -1961,7 +1971,7 @@ YY_BUFFER_STATE yy_scan_buffer (char * base, yy_size_t size ) YY_BUFFER_STATE yy_scan_string (yyconst char * yystr ) { - return yy_scan_bytes(yystr,strlen(yystr) ); + return yy_scan_bytes(yystr,(int) strlen(yystr) ); } /** Setup the input buffer state to scan the given bytes. The next call to yylex() will @@ -1971,7 +1981,7 @@ YY_BUFFER_STATE yy_scan_string (yyconst char * yystr ) * * @return the newly allocated buffer state object. */ -YY_BUFFER_STATE yy_scan_bytes (yyconst char * yybytes, yy_size_t _yybytes_len ) +YY_BUFFER_STATE yy_scan_bytes (yyconst char * yybytes, int _yybytes_len ) { YY_BUFFER_STATE b; char *buf; @@ -1979,7 +1989,7 @@ YY_BUFFER_STATE yy_scan_bytes (yyconst char * yybytes, yy_size_t _yybytes_len yy_size_t i; /* Get memory for full buffer, including space for trailing EOB's. */ - n = _yybytes_len + 2; + n = (yy_size_t) _yybytes_len + 2; buf = (char *) yyalloc(n ); if ( ! buf ) YY_FATAL_ERROR( "out of dynamic memory in yy_scan_bytes()" ); @@ -2005,9 +2015,9 @@ YY_BUFFER_STATE yy_scan_bytes (yyconst char * yybytes, yy_size_t _yybytes_len #define YY_EXIT_FAILURE 2 #endif -static void yy_fatal_error (yyconst char* msg ) +static void yynoreturn yy_fatal_error (yyconst char* msg ) { - (void) fprintf( stderr, "%s\n", msg ); + (void) fprintf( stderr, "%s\n", msg ); exit( YY_EXIT_FAILURE ); } @@ -2035,7 +2045,7 @@ static void yy_fatal_error (yyconst char* msg ) */ int yyget_lineno (void) { - + return yylineno; } @@ -2058,7 +2068,7 @@ FILE *yyget_out (void) /** Get the length of the current token. * */ -yy_size_t yyget_leng (void) +int yyget_leng (void) { return yyleng; } @@ -2073,29 +2083,29 @@ char *yyget_text (void) } /** Set the current line number. - * @param line_number + * @param _line_number line number * */ -void yyset_lineno (int line_number ) +void yyset_lineno (int _line_number ) { - yylineno = line_number; + yylineno = _line_number; } /** Set the input stream. This does not discard the current * input buffer. - * @param in_str A readable stream. + * @param _in_str A readable stream. * * @see yy_switch_to_buffer */ -void yyset_in (FILE * in_str ) +void yyset_in (FILE * _in_str ) { - yyin = in_str ; + yyin = _in_str ; } -void yyset_out (FILE * out_str ) +void yyset_out (FILE * _out_str ) { - yyout = out_str ; + yyout = _out_str ; } int yyget_debug (void) @@ -2103,9 +2113,9 @@ int yyget_debug (void) return yy_flex_debug; } -void yyset_debug (int bdebug ) +void yyset_debug (int _bdebug ) { - yy_flex_debug = bdebug ; + yy_flex_debug = _bdebug ; } static int yy_init_globals (void) @@ -2114,10 +2124,10 @@ static int yy_init_globals (void) * This function is called from yylex_destroy(), so don't allocate here. */ - (yy_buffer_stack) = 0; + (yy_buffer_stack) = NULL; (yy_buffer_stack_top) = 0; (yy_buffer_stack_max) = 0; - (yy_c_buf_p) = (char *) 0; + (yy_c_buf_p) = NULL; (yy_init) = 0; (yy_start) = 0; @@ -2126,8 +2136,8 @@ static int yy_init_globals (void) yyin = stdin; yyout = stdout; #else - yyin = (FILE *) 0; - yyout = (FILE *) 0; + yyin = NULL; + yyout = NULL; #endif /* For future reference: Set errno on error, since we are called by @@ -2165,7 +2175,8 @@ int yylex_destroy (void) #ifndef yytext_ptr static void yy_flex_strncpy (char* s1, yyconst char * s2, int n ) { - register int i; + + int i; for ( i = 0; i < n; ++i ) s1[i] = s2[i]; } @@ -2174,7 +2185,7 @@ static void yy_flex_strncpy (char* s1, yyconst char * s2, int n ) #ifdef YY_NEED_STRLEN static int yy_flex_strlen (yyconst char * s ) { - register int n; + int n; for ( n = 0; s[n]; ++n ) ; @@ -2184,11 +2195,12 @@ static int yy_flex_strlen (yyconst char * s ) void *yyalloc (yy_size_t size ) { - return (void *) malloc( size ); + return malloc(size); } void *yyrealloc (void * ptr, yy_size_t size ) { + /* The cast to (char *) in the following accommodates both * implementations that use char* generic pointers, and those * that use void* generic pointers. It works with the latter @@ -2196,17 +2208,17 @@ void *yyrealloc (void * ptr, yy_size_t size ) * any pointer type to void*, and deal with argument conversions * as though doing an assignment. */ - return (void *) realloc( (char *) ptr, size ); + return realloc(ptr, size); } void yyfree (void * ptr ) { - free( (char *) ptr ); /* see yyrealloc() for (char *) cast */ + free( (char *) ptr ); /* see yyrealloc() for (char *) cast */ } #define YYTABLES_NAME "yytables" -#line 265 "dtc-lexer.l" +#line 271 "dtc-lexer.l" diff --git a/scripts/dtc/dtc-parser.tab.c_shipped b/scripts/dtc/dtc-parser.tab.c_shipped index 31cec50a1265..2965227a1b4a 100644 --- a/scripts/dtc/dtc-parser.tab.c_shipped +++ b/scripts/dtc/dtc-parser.tab.c_shipped @@ -1,8 +1,8 @@ -/* A Bison parser, made by GNU Bison 3.0.2. */ +/* A Bison parser, made by GNU Bison 3.0.4. */ /* Bison implementation for Yacc-like parsers in C - Copyright (C) 1984, 1989-1990, 2000-2013 Free Software Foundation, Inc. + Copyright (C) 1984, 1989-1990, 2000-2015 Free Software Foundation, Inc. This program is free software: you can redistribute it and/or modify it under the terms of the GNU General Public License as published by @@ -44,7 +44,7 @@ #define YYBISON 1 /* Bison version. */ -#define YYBISON_VERSION "3.0.2" +#define YYBISON_VERSION "3.0.4" /* Skeleton name. */ #define YYSKELETON_NAME "yacc.c" @@ -65,6 +65,7 @@ #line 20 "dtc-parser.y" /* yacc.c:339 */ #include <stdio.h> +#include <inttypes.h> #include "dtc.h" #include "srcpos.h" @@ -77,10 +78,10 @@ extern void yyerror(char const *s); treesource_error = true; \ } while (0) -extern struct boot_info *the_boot_info; +extern struct dt_info *parser_output; extern bool treesource_error; -#line 84 "dtc-parser.tab.c" /* yacc.c:339 */ +#line 85 "dtc-parser.tab.c" /* yacc.c:339 */ # ifndef YY_NULLPTR # if defined __cplusplus && 201103L <= __cplusplus @@ -116,35 +117,36 @@ extern int yydebug; enum yytokentype { DT_V1 = 258, - DT_MEMRESERVE = 259, - DT_LSHIFT = 260, - DT_RSHIFT = 261, - DT_LE = 262, - DT_GE = 263, - DT_EQ = 264, - DT_NE = 265, - DT_AND = 266, - DT_OR = 267, - DT_BITS = 268, - DT_DEL_PROP = 269, - DT_DEL_NODE = 270, - DT_PROPNODENAME = 271, - DT_LITERAL = 272, - DT_CHAR_LITERAL = 273, - DT_BYTE = 274, - DT_STRING = 275, - DT_LABEL = 276, - DT_REF = 277, - DT_INCBIN = 278 + DT_PLUGIN = 259, + DT_MEMRESERVE = 260, + DT_LSHIFT = 261, + DT_RSHIFT = 262, + DT_LE = 263, + DT_GE = 264, + DT_EQ = 265, + DT_NE = 266, + DT_AND = 267, + DT_OR = 268, + DT_BITS = 269, + DT_DEL_PROP = 270, + DT_DEL_NODE = 271, + DT_PROPNODENAME = 272, + DT_LITERAL = 273, + DT_CHAR_LITERAL = 274, + DT_BYTE = 275, + DT_STRING = 276, + DT_LABEL = 277, + DT_REF = 278, + DT_INCBIN = 279 }; #endif /* Value type. */ #if ! defined YYSTYPE && ! defined YYSTYPE_IS_DECLARED -typedef union YYSTYPE YYSTYPE; + union YYSTYPE { -#line 38 "dtc-parser.y" /* yacc.c:355 */ +#line 39 "dtc-parser.y" /* yacc.c:355 */ char *propnodename; char *labelref; @@ -162,9 +164,12 @@ union YYSTYPE struct node *nodelist; struct reserve_info *re; uint64_t integer; + unsigned int flags; -#line 167 "dtc-parser.tab.c" /* yacc.c:355 */ +#line 170 "dtc-parser.tab.c" /* yacc.c:355 */ }; + +typedef union YYSTYPE YYSTYPE; # define YYSTYPE_IS_TRIVIAL 1 # define YYSTYPE_IS_DECLARED 1 #endif @@ -192,7 +197,7 @@ int yyparse (void); /* Copy the second part of user declarations. */ -#line 196 "dtc-parser.tab.c" /* yacc.c:358 */ +#line 201 "dtc-parser.tab.c" /* yacc.c:358 */ #ifdef short # undef short @@ -434,23 +439,23 @@ union yyalloc #endif /* !YYCOPY_NEEDED */ /* YYFINAL -- State number of the termination state. */ -#define YYFINAL 4 +#define YYFINAL 6 /* YYLAST -- Last index in YYTABLE. */ -#define YYLAST 136 +#define YYLAST 138 /* YYNTOKENS -- Number of terminals. */ -#define YYNTOKENS 47 +#define YYNTOKENS 48 /* YYNNTS -- Number of nonterminals. */ -#define YYNNTS 28 +#define YYNNTS 30 /* YYNRULES -- Number of rules. */ -#define YYNRULES 80 +#define YYNRULES 84 /* YYNSTATES -- Number of states. */ -#define YYNSTATES 144 +#define YYNSTATES 149 /* YYTRANSLATE[YYX] -- Symbol number corresponding to YYX as returned by yylex, with out-of-bounds checking. */ #define YYUNDEFTOK 2 -#define YYMAXUTOK 278 +#define YYMAXUTOK 279 #define YYTRANSLATE(YYX) \ ((unsigned int) (YYX) <= YYMAXUTOK ? yytranslate[YYX] : YYUNDEFTOK) @@ -462,16 +467,16 @@ static const yytype_uint8 yytranslate[] = 0, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, - 2, 2, 2, 46, 2, 2, 2, 44, 40, 2, - 32, 34, 43, 41, 33, 42, 2, 25, 2, 2, - 2, 2, 2, 2, 2, 2, 2, 2, 37, 24, - 35, 28, 29, 36, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 47, 2, 2, 2, 45, 41, 2, + 33, 35, 44, 42, 34, 43, 2, 26, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 38, 25, + 36, 29, 30, 37, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, - 2, 30, 2, 31, 39, 2, 2, 2, 2, 2, + 2, 31, 2, 32, 40, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, - 2, 2, 2, 26, 38, 27, 45, 2, 2, 2, + 2, 2, 2, 27, 39, 28, 46, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, @@ -486,22 +491,22 @@ static const yytype_uint8 yytranslate[] = 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, - 15, 16, 17, 18, 19, 20, 21, 22, 23 + 15, 16, 17, 18, 19, 20, 21, 22, 23, 24 }; #if YYDEBUG /* YYRLINE[YYN] -- Source line where rule number YYN was defined. */ static const yytype_uint16 yyrline[] = { - 0, 104, 104, 113, 116, 123, 127, 135, 139, 144, - 155, 165, 180, 188, 191, 198, 202, 206, 210, 218, - 222, 226, 230, 234, 250, 260, 268, 271, 275, 282, - 298, 303, 322, 336, 343, 344, 345, 352, 356, 357, - 361, 362, 366, 367, 371, 372, 376, 377, 381, 382, - 386, 387, 388, 392, 393, 394, 395, 396, 400, 401, - 402, 406, 407, 408, 412, 413, 422, 431, 435, 436, - 437, 438, 443, 446, 450, 458, 461, 465, 473, 477, - 481 + 0, 109, 109, 117, 121, 128, 129, 139, 142, 149, + 153, 161, 165, 170, 181, 191, 206, 214, 217, 224, + 228, 232, 236, 244, 248, 252, 256, 260, 276, 286, + 294, 297, 301, 308, 324, 329, 348, 362, 369, 370, + 371, 378, 382, 383, 387, 388, 392, 393, 397, 398, + 402, 403, 407, 408, 412, 413, 414, 418, 419, 420, + 421, 422, 426, 427, 428, 432, 433, 434, 438, 439, + 448, 457, 461, 462, 463, 464, 469, 472, 476, 484, + 487, 491, 499, 503, 507 }; #endif @@ -510,19 +515,20 @@ static const yytype_uint16 yyrline[] = First, the terminals, then, starting at YYNTOKENS, nonterminals. */ static const char *const yytname[] = { - "$end", "error", "$undefined", "DT_V1", "DT_MEMRESERVE", "DT_LSHIFT", - "DT_RSHIFT", "DT_LE", "DT_GE", "DT_EQ", "DT_NE", "DT_AND", "DT_OR", - "DT_BITS", "DT_DEL_PROP", "DT_DEL_NODE", "DT_PROPNODENAME", "DT_LITERAL", - "DT_CHAR_LITERAL", "DT_BYTE", "DT_STRING", "DT_LABEL", "DT_REF", - "DT_INCBIN", "';'", "'/'", "'{'", "'}'", "'='", "'>'", "'['", "']'", - "'('", "','", "')'", "'<'", "'?'", "':'", "'|'", "'^'", "'&'", "'+'", - "'-'", "'*'", "'%'", "'~'", "'!'", "$accept", "sourcefile", - "memreserves", "memreserve", "devicetree", "nodedef", "proplist", - "propdef", "propdata", "propdataprefix", "arrayprefix", "integer_prim", - "integer_expr", "integer_trinary", "integer_or", "integer_and", - "integer_bitor", "integer_bitxor", "integer_bitand", "integer_eq", - "integer_rela", "integer_shift", "integer_add", "integer_mul", - "integer_unary", "bytestring", "subnodes", "subnode", YY_NULLPTR + "$end", "error", "$undefined", "DT_V1", "DT_PLUGIN", "DT_MEMRESERVE", + "DT_LSHIFT", "DT_RSHIFT", "DT_LE", "DT_GE", "DT_EQ", "DT_NE", "DT_AND", + "DT_OR", "DT_BITS", "DT_DEL_PROP", "DT_DEL_NODE", "DT_PROPNODENAME", + "DT_LITERAL", "DT_CHAR_LITERAL", "DT_BYTE", "DT_STRING", "DT_LABEL", + "DT_REF", "DT_INCBIN", "';'", "'/'", "'{'", "'}'", "'='", "'>'", "'['", + "']'", "'('", "','", "')'", "'<'", "'?'", "':'", "'|'", "'^'", "'&'", + "'+'", "'-'", "'*'", "'%'", "'~'", "'!'", "$accept", "sourcefile", + "header", "headers", "memreserves", "memreserve", "devicetree", + "nodedef", "proplist", "propdef", "propdata", "propdataprefix", + "arrayprefix", "integer_prim", "integer_expr", "integer_trinary", + "integer_or", "integer_and", "integer_bitor", "integer_bitxor", + "integer_bitand", "integer_eq", "integer_rela", "integer_shift", + "integer_add", "integer_mul", "integer_unary", "bytestring", "subnodes", + "subnode", YY_NULLPTR }; #endif @@ -533,16 +539,16 @@ static const yytype_uint16 yytoknum[] = { 0, 256, 257, 258, 259, 260, 261, 262, 263, 264, 265, 266, 267, 268, 269, 270, 271, 272, 273, 274, - 275, 276, 277, 278, 59, 47, 123, 125, 61, 62, - 91, 93, 40, 44, 41, 60, 63, 58, 124, 94, - 38, 43, 45, 42, 37, 126, 33 + 275, 276, 277, 278, 279, 59, 47, 123, 125, 61, + 62, 91, 93, 40, 44, 41, 60, 63, 58, 124, + 94, 38, 43, 45, 42, 37, 126, 33 }; # endif -#define YYPACT_NINF -81 +#define YYPACT_NINF -44 #define yypact_value_is_default(Yystate) \ - (!!((Yystate) == (-81))) + (!!((Yystate) == (-44))) #define YYTABLE_NINF -1 @@ -553,21 +559,21 @@ static const yytype_uint16 yytoknum[] = STATE-NUM. */ static const yytype_int8 yypact[] = { - 16, -11, 21, 10, -81, 25, 10, 19, 10, -81, - -81, -9, 25, -81, 2, 51, -81, -9, -9, -9, - -81, 1, -81, -6, 50, 14, 28, 29, 36, 3, - 58, 44, -3, -81, 47, -81, -81, 65, 68, 2, - 2, -81, -81, -81, -81, -9, -9, -9, -9, -9, - -9, -9, -9, -9, -9, -9, -9, -9, -9, -9, - -9, -9, -9, -9, -81, 63, 69, 2, -81, -81, - 50, 57, 14, 28, 29, 36, 3, 3, 58, 58, - 58, 58, 44, 44, -3, -3, -81, -81, -81, 79, - 80, -8, 63, -81, 72, 63, -81, -81, -9, 76, - 77, -81, -81, -81, -81, -81, 78, -81, -81, -81, - -81, -81, 35, 4, -81, -81, -81, -81, 86, -81, - -81, -81, 73, -81, -81, 33, 71, 84, 39, -81, - -81, -81, -81, -81, 41, -81, -81, -81, 25, -81, - 74, 25, 75, -81 + 14, 27, 61, 14, 8, 18, -44, -44, 37, 8, + 40, 8, 64, -44, -44, -12, 37, -44, 50, 52, + -44, -44, -12, -12, -12, -44, 51, -44, -4, 78, + 53, 54, 55, 17, 2, 30, 38, -3, -44, 66, + -44, -44, 70, 72, 50, 50, -44, -44, -44, -44, + -12, -12, -12, -12, -12, -12, -12, -12, -12, -12, + -12, -12, -12, -12, -12, -12, -12, -12, -12, -44, + 3, 73, 50, -44, -44, 78, 59, 53, 54, 55, + 17, 2, 2, 30, 30, 30, 30, 38, 38, -3, + -3, -44, -44, -44, 82, 83, 44, 3, -44, 74, + 3, -44, -44, -12, 76, 79, -44, -44, -44, -44, + -44, 80, -44, -44, -44, -44, -44, -10, 36, -44, + -44, -44, -44, 85, -44, -44, -44, 75, -44, -44, + 21, 71, 88, -6, -44, -44, -44, -44, -44, 11, + -44, -44, -44, 37, -44, 77, 37, 81, -44 }; /* YYDEFACT[STATE-NUM] -- Default reduction number in state STATE-NUM. @@ -575,37 +581,37 @@ static const yytype_int8 yypact[] = means the default is an error. */ static const yytype_uint8 yydefact[] = { - 0, 0, 0, 3, 1, 0, 0, 0, 3, 34, - 35, 0, 0, 6, 0, 2, 4, 0, 0, 0, - 68, 0, 37, 38, 40, 42, 44, 46, 48, 50, - 53, 60, 63, 67, 0, 13, 7, 0, 0, 0, - 0, 69, 70, 71, 36, 0, 0, 0, 0, 0, + 0, 0, 0, 5, 7, 3, 1, 6, 0, 0, + 0, 7, 0, 38, 39, 0, 0, 10, 0, 2, + 8, 4, 0, 0, 0, 72, 0, 41, 42, 44, + 46, 48, 50, 52, 54, 57, 64, 67, 71, 0, + 17, 11, 0, 0, 0, 0, 73, 74, 75, 40, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, - 0, 0, 0, 0, 5, 75, 0, 0, 10, 8, - 41, 0, 43, 45, 47, 49, 51, 52, 56, 57, - 55, 54, 58, 59, 61, 62, 65, 64, 66, 0, - 0, 0, 0, 14, 0, 75, 11, 9, 0, 0, - 0, 16, 26, 78, 18, 80, 0, 77, 76, 39, - 17, 79, 0, 0, 12, 25, 15, 27, 0, 19, - 28, 22, 0, 72, 30, 0, 0, 0, 0, 33, - 32, 20, 31, 29, 0, 73, 74, 21, 0, 24, - 0, 0, 0, 23 + 0, 0, 0, 0, 0, 0, 0, 0, 0, 9, + 79, 0, 0, 14, 12, 45, 0, 47, 49, 51, + 53, 55, 56, 60, 61, 59, 58, 62, 63, 65, + 66, 69, 68, 70, 0, 0, 0, 0, 18, 0, + 79, 15, 13, 0, 0, 0, 20, 30, 82, 22, + 84, 0, 81, 80, 43, 21, 83, 0, 0, 16, + 29, 19, 31, 0, 23, 32, 26, 0, 76, 34, + 0, 0, 0, 0, 37, 36, 24, 35, 33, 0, + 77, 78, 25, 0, 28, 0, 0, 0, 27 }; /* YYPGOTO[NTERM-NUM]. */ static const yytype_int8 yypgoto[] = { - -81, -81, 100, 104, -81, -38, -81, -80, -81, -81, - -81, -5, 66, 13, -81, 70, 67, 81, 64, 82, - 37, 27, 34, 38, -14, -81, 22, 24 + -44, -44, -44, 103, 99, 104, -44, -43, -44, -21, + -44, -44, -44, -8, 63, 9, -44, 65, 67, 68, + 69, 62, 26, 4, 22, 23, -19, -44, 20, 28 }; /* YYDEFGOTO[NTERM-NUM]. */ static const yytype_int16 yydefgoto[] = { - -1, 2, 7, 8, 15, 36, 65, 93, 112, 113, - 125, 20, 21, 22, 23, 24, 25, 26, 27, 28, - 29, 30, 31, 32, 33, 128, 94, 95 + -1, 2, 3, 4, 10, 11, 19, 41, 70, 98, + 117, 118, 130, 25, 26, 27, 28, 29, 30, 31, + 32, 33, 34, 35, 36, 37, 38, 133, 99, 100 }; /* YYTABLE[YYPACT[STATE-NUM]] -- What to do in state STATE-NUM. If @@ -613,87 +619,87 @@ static const yytype_int16 yydefgoto[] = number is the opposite. If YYTABLE_NINF, syntax error. */ static const yytype_uint8 yytable[] = { - 12, 68, 69, 41, 42, 43, 45, 34, 9, 10, - 53, 54, 104, 3, 5, 107, 101, 118, 35, 1, - 102, 4, 61, 11, 119, 120, 121, 122, 35, 97, - 46, 6, 55, 17, 123, 44, 18, 19, 56, 124, - 62, 63, 9, 10, 14, 51, 52, 86, 87, 88, - 9, 10, 48, 103, 129, 130, 115, 11, 135, 116, - 136, 47, 131, 57, 58, 11, 37, 49, 117, 50, - 137, 64, 38, 39, 138, 139, 40, 89, 90, 91, - 78, 79, 80, 81, 92, 59, 60, 66, 76, 77, - 67, 82, 83, 96, 98, 99, 100, 84, 85, 106, - 110, 111, 114, 126, 134, 127, 133, 141, 16, 143, - 13, 109, 71, 74, 72, 70, 105, 108, 0, 0, - 132, 0, 0, 0, 0, 0, 0, 0, 0, 73, - 0, 0, 75, 140, 0, 0, 142 + 16, 73, 74, 46, 47, 48, 13, 14, 39, 50, + 58, 59, 120, 8, 140, 121, 141, 1, 94, 95, + 96, 15, 12, 66, 122, 97, 142, 56, 57, 102, + 9, 22, 60, 51, 23, 24, 62, 63, 61, 13, + 14, 67, 68, 134, 135, 143, 144, 91, 92, 93, + 123, 136, 5, 108, 15, 13, 14, 124, 125, 126, + 127, 6, 83, 84, 85, 86, 18, 128, 42, 106, + 15, 40, 129, 107, 43, 44, 109, 40, 45, 112, + 64, 65, 81, 82, 87, 88, 49, 89, 90, 21, + 52, 69, 53, 71, 54, 72, 55, 103, 101, 104, + 105, 115, 111, 131, 116, 119, 7, 138, 132, 139, + 20, 146, 114, 17, 76, 75, 148, 80, 0, 77, + 113, 78, 137, 79, 0, 110, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 145, 0, 0, 147 }; static const yytype_int16 yycheck[] = { - 5, 39, 40, 17, 18, 19, 12, 12, 17, 18, - 7, 8, 92, 24, 4, 95, 24, 13, 26, 3, - 28, 0, 25, 32, 20, 21, 22, 23, 26, 67, - 36, 21, 29, 42, 30, 34, 45, 46, 35, 35, - 43, 44, 17, 18, 25, 9, 10, 61, 62, 63, - 17, 18, 38, 91, 21, 22, 21, 32, 19, 24, - 21, 11, 29, 5, 6, 32, 15, 39, 33, 40, - 31, 24, 21, 22, 33, 34, 25, 14, 15, 16, - 53, 54, 55, 56, 21, 41, 42, 22, 51, 52, - 22, 57, 58, 24, 37, 16, 16, 59, 60, 27, - 24, 24, 24, 17, 20, 32, 35, 33, 8, 34, - 6, 98, 46, 49, 47, 45, 92, 95, -1, -1, - 125, -1, -1, -1, -1, -1, -1, -1, -1, 48, - -1, -1, 50, 138, -1, -1, 141 + 8, 44, 45, 22, 23, 24, 18, 19, 16, 13, + 8, 9, 22, 5, 20, 25, 22, 3, 15, 16, + 17, 33, 4, 26, 34, 22, 32, 10, 11, 72, + 22, 43, 30, 37, 46, 47, 6, 7, 36, 18, + 19, 44, 45, 22, 23, 34, 35, 66, 67, 68, + 14, 30, 25, 96, 33, 18, 19, 21, 22, 23, + 24, 0, 58, 59, 60, 61, 26, 31, 16, 25, + 33, 27, 36, 29, 22, 23, 97, 27, 26, 100, + 42, 43, 56, 57, 62, 63, 35, 64, 65, 25, + 12, 25, 39, 23, 40, 23, 41, 38, 25, 17, + 17, 25, 28, 18, 25, 25, 3, 36, 33, 21, + 11, 34, 103, 9, 51, 50, 35, 55, -1, 52, + 100, 53, 130, 54, -1, 97, -1, -1, -1, -1, + -1, -1, -1, -1, -1, 143, -1, -1, 146 }; /* YYSTOS[STATE-NUM] -- The (internal number of the) accessing symbol of state STATE-NUM. */ static const yytype_uint8 yystos[] = { - 0, 3, 48, 24, 0, 4, 21, 49, 50, 17, - 18, 32, 58, 50, 25, 51, 49, 42, 45, 46, - 58, 59, 60, 61, 62, 63, 64, 65, 66, 67, - 68, 69, 70, 71, 58, 26, 52, 15, 21, 22, - 25, 71, 71, 71, 34, 12, 36, 11, 38, 39, - 40, 9, 10, 7, 8, 29, 35, 5, 6, 41, - 42, 25, 43, 44, 24, 53, 22, 22, 52, 52, - 62, 59, 63, 64, 65, 66, 67, 67, 68, 68, - 68, 68, 69, 69, 70, 70, 71, 71, 71, 14, - 15, 16, 21, 54, 73, 74, 24, 52, 37, 16, - 16, 24, 28, 52, 54, 74, 27, 54, 73, 60, - 24, 24, 55, 56, 24, 21, 24, 33, 13, 20, - 21, 22, 23, 30, 35, 57, 17, 32, 72, 21, - 22, 29, 58, 35, 20, 19, 21, 31, 33, 34, - 58, 33, 58, 34 + 0, 3, 49, 50, 51, 25, 0, 51, 5, 22, + 52, 53, 4, 18, 19, 33, 61, 53, 26, 54, + 52, 25, 43, 46, 47, 61, 62, 63, 64, 65, + 66, 67, 68, 69, 70, 71, 72, 73, 74, 61, + 27, 55, 16, 22, 23, 26, 74, 74, 74, 35, + 13, 37, 12, 39, 40, 41, 10, 11, 8, 9, + 30, 36, 6, 7, 42, 43, 26, 44, 45, 25, + 56, 23, 23, 55, 55, 65, 62, 66, 67, 68, + 69, 70, 70, 71, 71, 71, 71, 72, 72, 73, + 73, 74, 74, 74, 15, 16, 17, 22, 57, 76, + 77, 25, 55, 38, 17, 17, 25, 29, 55, 57, + 77, 28, 57, 76, 63, 25, 25, 58, 59, 25, + 22, 25, 34, 14, 21, 22, 23, 24, 31, 36, + 60, 18, 33, 75, 22, 23, 30, 61, 36, 21, + 20, 22, 32, 34, 35, 61, 34, 61, 35 }; /* YYR1[YYN] -- Symbol number of symbol that rule YYN derives. */ static const yytype_uint8 yyr1[] = { - 0, 47, 48, 49, 49, 50, 50, 51, 51, 51, - 51, 51, 52, 53, 53, 54, 54, 54, 54, 55, - 55, 55, 55, 55, 55, 55, 56, 56, 56, 57, - 57, 57, 57, 57, 58, 58, 58, 59, 60, 60, - 61, 61, 62, 62, 63, 63, 64, 64, 65, 65, - 66, 66, 66, 67, 67, 67, 67, 67, 68, 68, - 68, 69, 69, 69, 70, 70, 70, 70, 71, 71, - 71, 71, 72, 72, 72, 73, 73, 73, 74, 74, - 74 + 0, 48, 49, 50, 50, 51, 51, 52, 52, 53, + 53, 54, 54, 54, 54, 54, 55, 56, 56, 57, + 57, 57, 57, 58, 58, 58, 58, 58, 58, 58, + 59, 59, 59, 60, 60, 60, 60, 60, 61, 61, + 61, 62, 63, 63, 64, 64, 65, 65, 66, 66, + 67, 67, 68, 68, 69, 69, 69, 70, 70, 70, + 70, 70, 71, 71, 71, 72, 72, 72, 73, 73, + 73, 73, 74, 74, 74, 74, 75, 75, 75, 76, + 76, 76, 77, 77, 77 }; /* YYR2[YYN] -- Number of symbols on the right hand side of rule YYN. */ static const yytype_uint8 yyr2[] = { - 0, 2, 4, 0, 2, 4, 2, 2, 3, 4, - 3, 4, 5, 0, 2, 4, 2, 3, 2, 2, - 3, 4, 2, 9, 5, 2, 0, 2, 2, 3, - 1, 2, 2, 2, 1, 1, 3, 1, 1, 5, - 1, 3, 1, 3, 1, 3, 1, 3, 1, 3, - 1, 3, 3, 1, 3, 3, 3, 3, 3, 3, - 1, 3, 3, 1, 3, 3, 3, 1, 1, 2, - 2, 2, 0, 2, 2, 0, 2, 2, 2, 3, - 2 + 0, 2, 3, 2, 4, 1, 2, 0, 2, 4, + 2, 2, 3, 4, 3, 4, 5, 0, 2, 4, + 2, 3, 2, 2, 3, 4, 2, 9, 5, 2, + 0, 2, 2, 3, 1, 2, 2, 2, 1, 1, + 3, 1, 1, 5, 1, 3, 1, 3, 1, 3, + 1, 3, 1, 3, 1, 3, 3, 1, 3, 3, + 3, 3, 3, 3, 1, 3, 3, 1, 3, 3, + 3, 1, 1, 2, 2, 2, 0, 2, 2, 0, + 2, 2, 2, 3, 2 }; @@ -1463,65 +1469,91 @@ yyreduce: switch (yyn) { case 2: -#line 105 "dtc-parser.y" /* yacc.c:1646 */ +#line 110 "dtc-parser.y" /* yacc.c:1646 */ { - the_boot_info = build_boot_info((yyvsp[-1].re), (yyvsp[0].node), - guess_boot_cpuid((yyvsp[0].node))); + parser_output = build_dt_info((yyvsp[-2].flags), (yyvsp[-1].re), (yyvsp[0].node), + guess_boot_cpuid((yyvsp[0].node))); } -#line 1472 "dtc-parser.tab.c" /* yacc.c:1646 */ +#line 1478 "dtc-parser.tab.c" /* yacc.c:1646 */ break; case 3: -#line 113 "dtc-parser.y" /* yacc.c:1646 */ +#line 118 "dtc-parser.y" /* yacc.c:1646 */ { - (yyval.re) = NULL; + (yyval.flags) = DTSF_V1; } -#line 1480 "dtc-parser.tab.c" /* yacc.c:1646 */ +#line 1486 "dtc-parser.tab.c" /* yacc.c:1646 */ break; case 4: -#line 117 "dtc-parser.y" /* yacc.c:1646 */ +#line 122 "dtc-parser.y" /* yacc.c:1646 */ + { + (yyval.flags) = DTSF_V1 | DTSF_PLUGIN; + } +#line 1494 "dtc-parser.tab.c" /* yacc.c:1646 */ + break; + + case 6: +#line 130 "dtc-parser.y" /* yacc.c:1646 */ + { + if ((yyvsp[0].flags) != (yyvsp[-1].flags)) + ERROR(&(yylsp[0]), "Header flags don't match earlier ones"); + (yyval.flags) = (yyvsp[-1].flags); + } +#line 1504 "dtc-parser.tab.c" /* yacc.c:1646 */ + break; + + case 7: +#line 139 "dtc-parser.y" /* yacc.c:1646 */ + { + (yyval.re) = NULL; + } +#line 1512 "dtc-parser.tab.c" /* yacc.c:1646 */ + break; + + case 8: +#line 143 "dtc-parser.y" /* yacc.c:1646 */ { (yyval.re) = chain_reserve_entry((yyvsp[-1].re), (yyvsp[0].re)); } -#line 1488 "dtc-parser.tab.c" /* yacc.c:1646 */ +#line 1520 "dtc-parser.tab.c" /* yacc.c:1646 */ break; - case 5: -#line 124 "dtc-parser.y" /* yacc.c:1646 */ + case 9: +#line 150 "dtc-parser.y" /* yacc.c:1646 */ { (yyval.re) = build_reserve_entry((yyvsp[-2].integer), (yyvsp[-1].integer)); } -#line 1496 "dtc-parser.tab.c" /* yacc.c:1646 */ +#line 1528 "dtc-parser.tab.c" /* yacc.c:1646 */ break; - case 6: -#line 128 "dtc-parser.y" /* yacc.c:1646 */ + case 10: +#line 154 "dtc-parser.y" /* yacc.c:1646 */ { add_label(&(yyvsp[0].re)->labels, (yyvsp[-1].labelref)); (yyval.re) = (yyvsp[0].re); } -#line 1505 "dtc-parser.tab.c" /* yacc.c:1646 */ +#line 1537 "dtc-parser.tab.c" /* yacc.c:1646 */ break; - case 7: -#line 136 "dtc-parser.y" /* yacc.c:1646 */ + case 11: +#line 162 "dtc-parser.y" /* yacc.c:1646 */ { (yyval.node) = name_node((yyvsp[0].node), ""); } -#line 1513 "dtc-parser.tab.c" /* yacc.c:1646 */ +#line 1545 "dtc-parser.tab.c" /* yacc.c:1646 */ break; - case 8: -#line 140 "dtc-parser.y" /* yacc.c:1646 */ + case 12: +#line 166 "dtc-parser.y" /* yacc.c:1646 */ { (yyval.node) = merge_nodes((yyvsp[-2].node), (yyvsp[0].node)); } -#line 1521 "dtc-parser.tab.c" /* yacc.c:1646 */ +#line 1553 "dtc-parser.tab.c" /* yacc.c:1646 */ break; - case 9: -#line 145 "dtc-parser.y" /* yacc.c:1646 */ + case 13: +#line 171 "dtc-parser.y" /* yacc.c:1646 */ { struct node *target = get_node_by_ref((yyvsp[-3].node), (yyvsp[-1].labelref)); @@ -1532,11 +1564,11 @@ yyreduce: ERROR(&(yylsp[-1]), "Label or path %s not found", (yyvsp[-1].labelref)); (yyval.node) = (yyvsp[-3].node); } -#line 1536 "dtc-parser.tab.c" /* yacc.c:1646 */ +#line 1568 "dtc-parser.tab.c" /* yacc.c:1646 */ break; - case 10: -#line 156 "dtc-parser.y" /* yacc.c:1646 */ + case 14: +#line 182 "dtc-parser.y" /* yacc.c:1646 */ { struct node *target = get_node_by_ref((yyvsp[-2].node), (yyvsp[-1].labelref)); @@ -1546,11 +1578,11 @@ yyreduce: ERROR(&(yylsp[-1]), "Label or path %s not found", (yyvsp[-1].labelref)); (yyval.node) = (yyvsp[-2].node); } -#line 1550 "dtc-parser.tab.c" /* yacc.c:1646 */ +#line 1582 "dtc-parser.tab.c" /* yacc.c:1646 */ break; - case 11: -#line 166 "dtc-parser.y" /* yacc.c:1646 */ + case 15: +#line 192 "dtc-parser.y" /* yacc.c:1646 */ { struct node *target = get_node_by_ref((yyvsp[-3].node), (yyvsp[-1].labelref)); @@ -1562,100 +1594,100 @@ yyreduce: (yyval.node) = (yyvsp[-3].node); } -#line 1566 "dtc-parser.tab.c" /* yacc.c:1646 */ +#line 1598 "dtc-parser.tab.c" /* yacc.c:1646 */ break; - case 12: -#line 181 "dtc-parser.y" /* yacc.c:1646 */ + case 16: +#line 207 "dtc-parser.y" /* yacc.c:1646 */ { (yyval.node) = build_node((yyvsp[-3].proplist), (yyvsp[-2].nodelist)); } -#line 1574 "dtc-parser.tab.c" /* yacc.c:1646 */ +#line 1606 "dtc-parser.tab.c" /* yacc.c:1646 */ break; - case 13: -#line 188 "dtc-parser.y" /* yacc.c:1646 */ + case 17: +#line 214 "dtc-parser.y" /* yacc.c:1646 */ { (yyval.proplist) = NULL; } -#line 1582 "dtc-parser.tab.c" /* yacc.c:1646 */ +#line 1614 "dtc-parser.tab.c" /* yacc.c:1646 */ break; - case 14: -#line 192 "dtc-parser.y" /* yacc.c:1646 */ + case 18: +#line 218 "dtc-parser.y" /* yacc.c:1646 */ { (yyval.proplist) = chain_property((yyvsp[0].prop), (yyvsp[-1].proplist)); } -#line 1590 "dtc-parser.tab.c" /* yacc.c:1646 */ +#line 1622 "dtc-parser.tab.c" /* yacc.c:1646 */ break; - case 15: -#line 199 "dtc-parser.y" /* yacc.c:1646 */ + case 19: +#line 225 "dtc-parser.y" /* yacc.c:1646 */ { (yyval.prop) = build_property((yyvsp[-3].propnodename), (yyvsp[-1].data)); } -#line 1598 "dtc-parser.tab.c" /* yacc.c:1646 */ +#line 1630 "dtc-parser.tab.c" /* yacc.c:1646 */ break; - case 16: -#line 203 "dtc-parser.y" /* yacc.c:1646 */ + case 20: +#line 229 "dtc-parser.y" /* yacc.c:1646 */ { (yyval.prop) = build_property((yyvsp[-1].propnodename), empty_data); } -#line 1606 "dtc-parser.tab.c" /* yacc.c:1646 */ +#line 1638 "dtc-parser.tab.c" /* yacc.c:1646 */ break; - case 17: -#line 207 "dtc-parser.y" /* yacc.c:1646 */ + case 21: +#line 233 "dtc-parser.y" /* yacc.c:1646 */ { (yyval.prop) = build_property_delete((yyvsp[-1].propnodename)); } -#line 1614 "dtc-parser.tab.c" /* yacc.c:1646 */ +#line 1646 "dtc-parser.tab.c" /* yacc.c:1646 */ break; - case 18: -#line 211 "dtc-parser.y" /* yacc.c:1646 */ + case 22: +#line 237 "dtc-parser.y" /* yacc.c:1646 */ { add_label(&(yyvsp[0].prop)->labels, (yyvsp[-1].labelref)); (yyval.prop) = (yyvsp[0].prop); } -#line 1623 "dtc-parser.tab.c" /* yacc.c:1646 */ +#line 1655 "dtc-parser.tab.c" /* yacc.c:1646 */ break; - case 19: -#line 219 "dtc-parser.y" /* yacc.c:1646 */ + case 23: +#line 245 "dtc-parser.y" /* yacc.c:1646 */ { (yyval.data) = data_merge((yyvsp[-1].data), (yyvsp[0].data)); } -#line 1631 "dtc-parser.tab.c" /* yacc.c:1646 */ +#line 1663 "dtc-parser.tab.c" /* yacc.c:1646 */ break; - case 20: -#line 223 "dtc-parser.y" /* yacc.c:1646 */ + case 24: +#line 249 "dtc-parser.y" /* yacc.c:1646 */ { (yyval.data) = data_merge((yyvsp[-2].data), (yyvsp[-1].array).data); } -#line 1639 "dtc-parser.tab.c" /* yacc.c:1646 */ +#line 1671 "dtc-parser.tab.c" /* yacc.c:1646 */ break; - case 21: -#line 227 "dtc-parser.y" /* yacc.c:1646 */ + case 25: +#line 253 "dtc-parser.y" /* yacc.c:1646 */ { (yyval.data) = data_merge((yyvsp[-3].data), (yyvsp[-1].data)); } -#line 1647 "dtc-parser.tab.c" /* yacc.c:1646 */ +#line 1679 "dtc-parser.tab.c" /* yacc.c:1646 */ break; - case 22: -#line 231 "dtc-parser.y" /* yacc.c:1646 */ + case 26: +#line 257 "dtc-parser.y" /* yacc.c:1646 */ { (yyval.data) = data_add_marker((yyvsp[-1].data), REF_PATH, (yyvsp[0].labelref)); } -#line 1655 "dtc-parser.tab.c" /* yacc.c:1646 */ +#line 1687 "dtc-parser.tab.c" /* yacc.c:1646 */ break; - case 23: -#line 235 "dtc-parser.y" /* yacc.c:1646 */ + case 27: +#line 261 "dtc-parser.y" /* yacc.c:1646 */ { FILE *f = srcfile_relative_open((yyvsp[-5].data).val, NULL); struct data d; @@ -1671,11 +1703,11 @@ yyreduce: (yyval.data) = data_merge((yyvsp[-8].data), d); fclose(f); } -#line 1675 "dtc-parser.tab.c" /* yacc.c:1646 */ +#line 1707 "dtc-parser.tab.c" /* yacc.c:1646 */ break; - case 24: -#line 251 "dtc-parser.y" /* yacc.c:1646 */ + case 28: +#line 277 "dtc-parser.y" /* yacc.c:1646 */ { FILE *f = srcfile_relative_open((yyvsp[-1].data).val, NULL); struct data d = empty_data; @@ -1685,43 +1717,43 @@ yyreduce: (yyval.data) = data_merge((yyvsp[-4].data), d); fclose(f); } -#line 1689 "dtc-parser.tab.c" /* yacc.c:1646 */ +#line 1721 "dtc-parser.tab.c" /* yacc.c:1646 */ break; - case 25: -#line 261 "dtc-parser.y" /* yacc.c:1646 */ + case 29: +#line 287 "dtc-parser.y" /* yacc.c:1646 */ { (yyval.data) = data_add_marker((yyvsp[-1].data), LABEL, (yyvsp[0].labelref)); } -#line 1697 "dtc-parser.tab.c" /* yacc.c:1646 */ +#line 1729 "dtc-parser.tab.c" /* yacc.c:1646 */ break; - case 26: -#line 268 "dtc-parser.y" /* yacc.c:1646 */ + case 30: +#line 294 "dtc-parser.y" /* yacc.c:1646 */ { (yyval.data) = empty_data; } -#line 1705 "dtc-parser.tab.c" /* yacc.c:1646 */ +#line 1737 "dtc-parser.tab.c" /* yacc.c:1646 */ break; - case 27: -#line 272 "dtc-parser.y" /* yacc.c:1646 */ + case 31: +#line 298 "dtc-parser.y" /* yacc.c:1646 */ { (yyval.data) = (yyvsp[-1].data); } -#line 1713 "dtc-parser.tab.c" /* yacc.c:1646 */ +#line 1745 "dtc-parser.tab.c" /* yacc.c:1646 */ break; - case 28: -#line 276 "dtc-parser.y" /* yacc.c:1646 */ + case 32: +#line 302 "dtc-parser.y" /* yacc.c:1646 */ { (yyval.data) = data_add_marker((yyvsp[-1].data), LABEL, (yyvsp[0].labelref)); } -#line 1721 "dtc-parser.tab.c" /* yacc.c:1646 */ +#line 1753 "dtc-parser.tab.c" /* yacc.c:1646 */ break; - case 29: -#line 283 "dtc-parser.y" /* yacc.c:1646 */ + case 33: +#line 309 "dtc-parser.y" /* yacc.c:1646 */ { unsigned long long bits; @@ -1737,20 +1769,20 @@ yyreduce: (yyval.array).data = empty_data; (yyval.array).bits = bits; } -#line 1741 "dtc-parser.tab.c" /* yacc.c:1646 */ +#line 1773 "dtc-parser.tab.c" /* yacc.c:1646 */ break; - case 30: -#line 299 "dtc-parser.y" /* yacc.c:1646 */ + case 34: +#line 325 "dtc-parser.y" /* yacc.c:1646 */ { (yyval.array).data = empty_data; (yyval.array).bits = 32; } -#line 1750 "dtc-parser.tab.c" /* yacc.c:1646 */ +#line 1782 "dtc-parser.tab.c" /* yacc.c:1646 */ break; - case 31: -#line 304 "dtc-parser.y" /* yacc.c:1646 */ + case 35: +#line 330 "dtc-parser.y" /* yacc.c:1646 */ { if ((yyvsp[-1].array).bits < 64) { uint64_t mask = (1ULL << (yyvsp[-1].array).bits) - 1; @@ -1769,11 +1801,11 @@ yyreduce: (yyval.array).data = data_append_integer((yyvsp[-1].array).data, (yyvsp[0].integer), (yyvsp[-1].array).bits); } -#line 1773 "dtc-parser.tab.c" /* yacc.c:1646 */ +#line 1805 "dtc-parser.tab.c" /* yacc.c:1646 */ break; - case 32: -#line 323 "dtc-parser.y" /* yacc.c:1646 */ + case 36: +#line 349 "dtc-parser.y" /* yacc.c:1646 */ { uint64_t val = ~0ULL >> (64 - (yyvsp[-1].array).bits); @@ -1787,129 +1819,129 @@ yyreduce: (yyval.array).data = data_append_integer((yyvsp[-1].array).data, val, (yyvsp[-1].array).bits); } -#line 1791 "dtc-parser.tab.c" /* yacc.c:1646 */ +#line 1823 "dtc-parser.tab.c" /* yacc.c:1646 */ break; - case 33: -#line 337 "dtc-parser.y" /* yacc.c:1646 */ + case 37: +#line 363 "dtc-parser.y" /* yacc.c:1646 */ { (yyval.array).data = data_add_marker((yyvsp[-1].array).data, LABEL, (yyvsp[0].labelref)); } -#line 1799 "dtc-parser.tab.c" /* yacc.c:1646 */ +#line 1831 "dtc-parser.tab.c" /* yacc.c:1646 */ break; - case 36: -#line 346 "dtc-parser.y" /* yacc.c:1646 */ + case 40: +#line 372 "dtc-parser.y" /* yacc.c:1646 */ { (yyval.integer) = (yyvsp[-1].integer); } -#line 1807 "dtc-parser.tab.c" /* yacc.c:1646 */ +#line 1839 "dtc-parser.tab.c" /* yacc.c:1646 */ break; - case 39: -#line 357 "dtc-parser.y" /* yacc.c:1646 */ + case 43: +#line 383 "dtc-parser.y" /* yacc.c:1646 */ { (yyval.integer) = (yyvsp[-4].integer) ? (yyvsp[-2].integer) : (yyvsp[0].integer); } -#line 1813 "dtc-parser.tab.c" /* yacc.c:1646 */ +#line 1845 "dtc-parser.tab.c" /* yacc.c:1646 */ break; - case 41: -#line 362 "dtc-parser.y" /* yacc.c:1646 */ + case 45: +#line 388 "dtc-parser.y" /* yacc.c:1646 */ { (yyval.integer) = (yyvsp[-2].integer) || (yyvsp[0].integer); } -#line 1819 "dtc-parser.tab.c" /* yacc.c:1646 */ +#line 1851 "dtc-parser.tab.c" /* yacc.c:1646 */ break; - case 43: -#line 367 "dtc-parser.y" /* yacc.c:1646 */ + case 47: +#line 393 "dtc-parser.y" /* yacc.c:1646 */ { (yyval.integer) = (yyvsp[-2].integer) && (yyvsp[0].integer); } -#line 1825 "dtc-parser.tab.c" /* yacc.c:1646 */ +#line 1857 "dtc-parser.tab.c" /* yacc.c:1646 */ break; - case 45: -#line 372 "dtc-parser.y" /* yacc.c:1646 */ + case 49: +#line 398 "dtc-parser.y" /* yacc.c:1646 */ { (yyval.integer) = (yyvsp[-2].integer) | (yyvsp[0].integer); } -#line 1831 "dtc-parser.tab.c" /* yacc.c:1646 */ +#line 1863 "dtc-parser.tab.c" /* yacc.c:1646 */ break; - case 47: -#line 377 "dtc-parser.y" /* yacc.c:1646 */ + case 51: +#line 403 "dtc-parser.y" /* yacc.c:1646 */ { (yyval.integer) = (yyvsp[-2].integer) ^ (yyvsp[0].integer); } -#line 1837 "dtc-parser.tab.c" /* yacc.c:1646 */ +#line 1869 "dtc-parser.tab.c" /* yacc.c:1646 */ break; - case 49: -#line 382 "dtc-parser.y" /* yacc.c:1646 */ + case 53: +#line 408 "dtc-parser.y" /* yacc.c:1646 */ { (yyval.integer) = (yyvsp[-2].integer) & (yyvsp[0].integer); } -#line 1843 "dtc-parser.tab.c" /* yacc.c:1646 */ +#line 1875 "dtc-parser.tab.c" /* yacc.c:1646 */ break; - case 51: -#line 387 "dtc-parser.y" /* yacc.c:1646 */ + case 55: +#line 413 "dtc-parser.y" /* yacc.c:1646 */ { (yyval.integer) = (yyvsp[-2].integer) == (yyvsp[0].integer); } -#line 1849 "dtc-parser.tab.c" /* yacc.c:1646 */ +#line 1881 "dtc-parser.tab.c" /* yacc.c:1646 */ break; - case 52: -#line 388 "dtc-parser.y" /* yacc.c:1646 */ + case 56: +#line 414 "dtc-parser.y" /* yacc.c:1646 */ { (yyval.integer) = (yyvsp[-2].integer) != (yyvsp[0].integer); } -#line 1855 "dtc-parser.tab.c" /* yacc.c:1646 */ +#line 1887 "dtc-parser.tab.c" /* yacc.c:1646 */ break; - case 54: -#line 393 "dtc-parser.y" /* yacc.c:1646 */ + case 58: +#line 419 "dtc-parser.y" /* yacc.c:1646 */ { (yyval.integer) = (yyvsp[-2].integer) < (yyvsp[0].integer); } -#line 1861 "dtc-parser.tab.c" /* yacc.c:1646 */ +#line 1893 "dtc-parser.tab.c" /* yacc.c:1646 */ break; - case 55: -#line 394 "dtc-parser.y" /* yacc.c:1646 */ + case 59: +#line 420 "dtc-parser.y" /* yacc.c:1646 */ { (yyval.integer) = (yyvsp[-2].integer) > (yyvsp[0].integer); } -#line 1867 "dtc-parser.tab.c" /* yacc.c:1646 */ +#line 1899 "dtc-parser.tab.c" /* yacc.c:1646 */ break; - case 56: -#line 395 "dtc-parser.y" /* yacc.c:1646 */ + case 60: +#line 421 "dtc-parser.y" /* yacc.c:1646 */ { (yyval.integer) = (yyvsp[-2].integer) <= (yyvsp[0].integer); } -#line 1873 "dtc-parser.tab.c" /* yacc.c:1646 */ +#line 1905 "dtc-parser.tab.c" /* yacc.c:1646 */ break; - case 57: -#line 396 "dtc-parser.y" /* yacc.c:1646 */ + case 61: +#line 422 "dtc-parser.y" /* yacc.c:1646 */ { (yyval.integer) = (yyvsp[-2].integer) >= (yyvsp[0].integer); } -#line 1879 "dtc-parser.tab.c" /* yacc.c:1646 */ +#line 1911 "dtc-parser.tab.c" /* yacc.c:1646 */ break; - case 58: -#line 400 "dtc-parser.y" /* yacc.c:1646 */ + case 62: +#line 426 "dtc-parser.y" /* yacc.c:1646 */ { (yyval.integer) = (yyvsp[-2].integer) << (yyvsp[0].integer); } -#line 1885 "dtc-parser.tab.c" /* yacc.c:1646 */ +#line 1917 "dtc-parser.tab.c" /* yacc.c:1646 */ break; - case 59: -#line 401 "dtc-parser.y" /* yacc.c:1646 */ + case 63: +#line 427 "dtc-parser.y" /* yacc.c:1646 */ { (yyval.integer) = (yyvsp[-2].integer) >> (yyvsp[0].integer); } -#line 1891 "dtc-parser.tab.c" /* yacc.c:1646 */ +#line 1923 "dtc-parser.tab.c" /* yacc.c:1646 */ break; - case 61: -#line 406 "dtc-parser.y" /* yacc.c:1646 */ + case 65: +#line 432 "dtc-parser.y" /* yacc.c:1646 */ { (yyval.integer) = (yyvsp[-2].integer) + (yyvsp[0].integer); } -#line 1897 "dtc-parser.tab.c" /* yacc.c:1646 */ +#line 1929 "dtc-parser.tab.c" /* yacc.c:1646 */ break; - case 62: -#line 407 "dtc-parser.y" /* yacc.c:1646 */ + case 66: +#line 433 "dtc-parser.y" /* yacc.c:1646 */ { (yyval.integer) = (yyvsp[-2].integer) - (yyvsp[0].integer); } -#line 1903 "dtc-parser.tab.c" /* yacc.c:1646 */ +#line 1935 "dtc-parser.tab.c" /* yacc.c:1646 */ break; - case 64: -#line 412 "dtc-parser.y" /* yacc.c:1646 */ + case 68: +#line 438 "dtc-parser.y" /* yacc.c:1646 */ { (yyval.integer) = (yyvsp[-2].integer) * (yyvsp[0].integer); } -#line 1909 "dtc-parser.tab.c" /* yacc.c:1646 */ +#line 1941 "dtc-parser.tab.c" /* yacc.c:1646 */ break; - case 65: -#line 414 "dtc-parser.y" /* yacc.c:1646 */ + case 69: +#line 440 "dtc-parser.y" /* yacc.c:1646 */ { if ((yyvsp[0].integer) != 0) { (yyval.integer) = (yyvsp[-2].integer) / (yyvsp[0].integer); @@ -1918,11 +1950,11 @@ yyreduce: (yyval.integer) = 0; } } -#line 1922 "dtc-parser.tab.c" /* yacc.c:1646 */ +#line 1954 "dtc-parser.tab.c" /* yacc.c:1646 */ break; - case 66: -#line 423 "dtc-parser.y" /* yacc.c:1646 */ + case 70: +#line 449 "dtc-parser.y" /* yacc.c:1646 */ { if ((yyvsp[0].integer) != 0) { (yyval.integer) = (yyvsp[-2].integer) % (yyvsp[0].integer); @@ -1931,103 +1963,103 @@ yyreduce: (yyval.integer) = 0; } } -#line 1935 "dtc-parser.tab.c" /* yacc.c:1646 */ +#line 1967 "dtc-parser.tab.c" /* yacc.c:1646 */ break; - case 69: -#line 436 "dtc-parser.y" /* yacc.c:1646 */ + case 73: +#line 462 "dtc-parser.y" /* yacc.c:1646 */ { (yyval.integer) = -(yyvsp[0].integer); } -#line 1941 "dtc-parser.tab.c" /* yacc.c:1646 */ +#line 1973 "dtc-parser.tab.c" /* yacc.c:1646 */ break; - case 70: -#line 437 "dtc-parser.y" /* yacc.c:1646 */ + case 74: +#line 463 "dtc-parser.y" /* yacc.c:1646 */ { (yyval.integer) = ~(yyvsp[0].integer); } -#line 1947 "dtc-parser.tab.c" /* yacc.c:1646 */ +#line 1979 "dtc-parser.tab.c" /* yacc.c:1646 */ break; - case 71: -#line 438 "dtc-parser.y" /* yacc.c:1646 */ + case 75: +#line 464 "dtc-parser.y" /* yacc.c:1646 */ { (yyval.integer) = !(yyvsp[0].integer); } -#line 1953 "dtc-parser.tab.c" /* yacc.c:1646 */ +#line 1985 "dtc-parser.tab.c" /* yacc.c:1646 */ break; - case 72: -#line 443 "dtc-parser.y" /* yacc.c:1646 */ + case 76: +#line 469 "dtc-parser.y" /* yacc.c:1646 */ { (yyval.data) = empty_data; } -#line 1961 "dtc-parser.tab.c" /* yacc.c:1646 */ +#line 1993 "dtc-parser.tab.c" /* yacc.c:1646 */ break; - case 73: -#line 447 "dtc-parser.y" /* yacc.c:1646 */ + case 77: +#line 473 "dtc-parser.y" /* yacc.c:1646 */ { (yyval.data) = data_append_byte((yyvsp[-1].data), (yyvsp[0].byte)); } -#line 1969 "dtc-parser.tab.c" /* yacc.c:1646 */ +#line 2001 "dtc-parser.tab.c" /* yacc.c:1646 */ break; - case 74: -#line 451 "dtc-parser.y" /* yacc.c:1646 */ + case 78: +#line 477 "dtc-parser.y" /* yacc.c:1646 */ { (yyval.data) = data_add_marker((yyvsp[-1].data), LABEL, (yyvsp[0].labelref)); } -#line 1977 "dtc-parser.tab.c" /* yacc.c:1646 */ +#line 2009 "dtc-parser.tab.c" /* yacc.c:1646 */ break; - case 75: -#line 458 "dtc-parser.y" /* yacc.c:1646 */ + case 79: +#line 484 "dtc-parser.y" /* yacc.c:1646 */ { (yyval.nodelist) = NULL; } -#line 1985 "dtc-parser.tab.c" /* yacc.c:1646 */ +#line 2017 "dtc-parser.tab.c" /* yacc.c:1646 */ break; - case 76: -#line 462 "dtc-parser.y" /* yacc.c:1646 */ + case 80: +#line 488 "dtc-parser.y" /* yacc.c:1646 */ { (yyval.nodelist) = chain_node((yyvsp[-1].node), (yyvsp[0].nodelist)); } -#line 1993 "dtc-parser.tab.c" /* yacc.c:1646 */ +#line 2025 "dtc-parser.tab.c" /* yacc.c:1646 */ break; - case 77: -#line 466 "dtc-parser.y" /* yacc.c:1646 */ + case 81: +#line 492 "dtc-parser.y" /* yacc.c:1646 */ { ERROR(&(yylsp[0]), "Properties must precede subnodes"); YYERROR; } -#line 2002 "dtc-parser.tab.c" /* yacc.c:1646 */ +#line 2034 "dtc-parser.tab.c" /* yacc.c:1646 */ break; - case 78: -#line 474 "dtc-parser.y" /* yacc.c:1646 */ + case 82: +#line 500 "dtc-parser.y" /* yacc.c:1646 */ { (yyval.node) = name_node((yyvsp[0].node), (yyvsp[-1].propnodename)); } -#line 2010 "dtc-parser.tab.c" /* yacc.c:1646 */ +#line 2042 "dtc-parser.tab.c" /* yacc.c:1646 */ break; - case 79: -#line 478 "dtc-parser.y" /* yacc.c:1646 */ + case 83: +#line 504 "dtc-parser.y" /* yacc.c:1646 */ { (yyval.node) = name_node(build_node_delete(), (yyvsp[-1].propnodename)); } -#line 2018 "dtc-parser.tab.c" /* yacc.c:1646 */ +#line 2050 "dtc-parser.tab.c" /* yacc.c:1646 */ break; - case 80: -#line 482 "dtc-parser.y" /* yacc.c:1646 */ + case 84: +#line 508 "dtc-parser.y" /* yacc.c:1646 */ { add_label(&(yyvsp[0].node)->labels, (yyvsp[-1].labelref)); (yyval.node) = (yyvsp[0].node); } -#line 2027 "dtc-parser.tab.c" /* yacc.c:1646 */ +#line 2059 "dtc-parser.tab.c" /* yacc.c:1646 */ break; -#line 2031 "dtc-parser.tab.c" /* yacc.c:1646 */ +#line 2063 "dtc-parser.tab.c" /* yacc.c:1646 */ default: break; } /* User semantic actions sometimes alter yychar, and that requires @@ -2262,7 +2294,7 @@ yyreturn: #endif return yyresult; } -#line 488 "dtc-parser.y" /* yacc.c:1906 */ +#line 514 "dtc-parser.y" /* yacc.c:1906 */ void yyerror(char const *s) diff --git a/scripts/dtc/dtc-parser.tab.h_shipped b/scripts/dtc/dtc-parser.tab.h_shipped index 30867c688300..6aa512c1b337 100644 --- a/scripts/dtc/dtc-parser.tab.h_shipped +++ b/scripts/dtc/dtc-parser.tab.h_shipped @@ -1,8 +1,8 @@ -/* A Bison parser, made by GNU Bison 3.0.2. */ +/* A Bison parser, made by GNU Bison 3.0.4. */ /* Bison interface for Yacc-like parsers in C - Copyright (C) 1984, 1989-1990, 2000-2013 Free Software Foundation, Inc. + Copyright (C) 1984, 1989-1990, 2000-2015 Free Software Foundation, Inc. This program is free software: you can redistribute it and/or modify it under the terms of the GNU General Public License as published by @@ -46,35 +46,36 @@ extern int yydebug; enum yytokentype { DT_V1 = 258, - DT_MEMRESERVE = 259, - DT_LSHIFT = 260, - DT_RSHIFT = 261, - DT_LE = 262, - DT_GE = 263, - DT_EQ = 264, - DT_NE = 265, - DT_AND = 266, - DT_OR = 267, - DT_BITS = 268, - DT_DEL_PROP = 269, - DT_DEL_NODE = 270, - DT_PROPNODENAME = 271, - DT_LITERAL = 272, - DT_CHAR_LITERAL = 273, - DT_BYTE = 274, - DT_STRING = 275, - DT_LABEL = 276, - DT_REF = 277, - DT_INCBIN = 278 + DT_PLUGIN = 259, + DT_MEMRESERVE = 260, + DT_LSHIFT = 261, + DT_RSHIFT = 262, + DT_LE = 263, + DT_GE = 264, + DT_EQ = 265, + DT_NE = 266, + DT_AND = 267, + DT_OR = 268, + DT_BITS = 269, + DT_DEL_PROP = 270, + DT_DEL_NODE = 271, + DT_PROPNODENAME = 272, + DT_LITERAL = 273, + DT_CHAR_LITERAL = 274, + DT_BYTE = 275, + DT_STRING = 276, + DT_LABEL = 277, + DT_REF = 278, + DT_INCBIN = 279 }; #endif /* Value type. */ #if ! defined YYSTYPE && ! defined YYSTYPE_IS_DECLARED -typedef union YYSTYPE YYSTYPE; + union YYSTYPE { -#line 38 "dtc-parser.y" /* yacc.c:1909 */ +#line 39 "dtc-parser.y" /* yacc.c:1909 */ char *propnodename; char *labelref; @@ -92,9 +93,12 @@ union YYSTYPE struct node *nodelist; struct reserve_info *re; uint64_t integer; + unsigned int flags; -#line 97 "dtc-parser.tab.h" /* yacc.c:1909 */ +#line 99 "dtc-parser.tab.h" /* yacc.c:1909 */ }; + +typedef union YYSTYPE YYSTYPE; # define YYSTYPE_IS_TRIVIAL 1 # define YYSTYPE_IS_DECLARED 1 #endif diff --git a/scripts/dtc/dtc-parser.y b/scripts/dtc/dtc-parser.y index 000873f070fd..b2fd4d155839 100644 --- a/scripts/dtc/dtc-parser.y +++ b/scripts/dtc/dtc-parser.y @@ -19,6 +19,7 @@ */ %{ #include <stdio.h> +#include <inttypes.h> #include "dtc.h" #include "srcpos.h" @@ -31,7 +32,7 @@ extern void yyerror(char const *s); treesource_error = true; \ } while (0) -extern struct boot_info *the_boot_info; +extern struct dt_info *parser_output; extern bool treesource_error; %} @@ -52,9 +53,11 @@ extern bool treesource_error; struct node *nodelist; struct reserve_info *re; uint64_t integer; + unsigned int flags; } %token DT_V1 +%token DT_PLUGIN %token DT_MEMRESERVE %token DT_LSHIFT DT_RSHIFT DT_LE DT_GE DT_EQ DT_NE DT_AND DT_OR %token DT_BITS @@ -71,6 +74,8 @@ extern bool treesource_error; %type <data> propdata %type <data> propdataprefix +%type <flags> header +%type <flags> headers %type <re> memreserve %type <re> memreserves %type <array> arrayprefix @@ -101,10 +106,31 @@ extern bool treesource_error; %% sourcefile: - DT_V1 ';' memreserves devicetree + headers memreserves devicetree { - the_boot_info = build_boot_info($3, $4, - guess_boot_cpuid($4)); + parser_output = build_dt_info($1, $2, $3, + guess_boot_cpuid($3)); + } + ; + +header: + DT_V1 ';' + { + $$ = DTSF_V1; + } + | DT_V1 ';' DT_PLUGIN ';' + { + $$ = DTSF_V1 | DTSF_PLUGIN; + } + ; + +headers: + header + | header headers + { + if ($2 != $1) + ERROR(&@2, "Header flags don't match earlier ones"); + $$ = $1; } ; diff --git a/scripts/dtc/dtc.c b/scripts/dtc/dtc.c index 5fa23c406266..a4edf4c7aebf 100644 --- a/scripts/dtc/dtc.c +++ b/scripts/dtc/dtc.c @@ -30,7 +30,16 @@ int quiet; /* Level of quietness */ int reservenum; /* Number of memory reservation slots */ int minsize; /* Minimum blob size */ int padsize; /* Additional padding to blob */ +int alignsize; /* Additional padding to blob accroding to the alignsize */ int phandle_format = PHANDLE_BOTH; /* Use linux,phandle or phandle properties */ +int generate_symbols; /* enable symbols & fixup support */ +int generate_fixups; /* suppress generation of fixups on symbol support */ +int auto_label_aliases; /* auto generate labels -> aliases */ + +static int is_power_of_2(int x) +{ + return (x > 0) && ((x & (x - 1)) == 0); +} static void fill_fullpaths(struct node *tree, const char *prefix) { @@ -53,7 +62,7 @@ static void fill_fullpaths(struct node *tree, const char *prefix) #define FDT_VERSION(version) _FDT_VERSION(version) #define _FDT_VERSION(version) #version static const char usage_synopsis[] = "dtc [options] <input file>"; -static const char usage_short_opts[] = "qI:O:o:V:d:R:S:p:fb:i:H:sW:E:hv"; +static const char usage_short_opts[] = "qI:O:o:V:d:R:S:p:a:fb:i:H:sW:E:@Ahv"; static struct option const usage_long_opts[] = { {"quiet", no_argument, NULL, 'q'}, {"in-format", a_argument, NULL, 'I'}, @@ -64,6 +73,7 @@ static struct option const usage_long_opts[] = { {"reserve", a_argument, NULL, 'R'}, {"space", a_argument, NULL, 'S'}, {"pad", a_argument, NULL, 'p'}, + {"align", a_argument, NULL, 'a'}, {"boot-cpu", a_argument, NULL, 'b'}, {"force", no_argument, NULL, 'f'}, {"include", a_argument, NULL, 'i'}, @@ -71,6 +81,8 @@ static struct option const usage_long_opts[] = { {"phandle", a_argument, NULL, 'H'}, {"warning", a_argument, NULL, 'W'}, {"error", a_argument, NULL, 'E'}, + {"symbols", no_argument, NULL, '@'}, + {"auto-alias", no_argument, NULL, 'A'}, {"help", no_argument, NULL, 'h'}, {"version", no_argument, NULL, 'v'}, {NULL, no_argument, NULL, 0x0}, @@ -91,6 +103,7 @@ static const char * const usage_opts_help[] = { "\n\tMake space for <number> reserve map entries (for dtb and asm output)", "\n\tMake the blob at least <bytes> long (extra space)", "\n\tAdd padding to the blob of <bytes> long (extra space)", + "\n\tMake the blob align to the <bytes> (extra space)", "\n\tSet the physical boot cpu", "\n\tTry to produce output even if the input tree has errors", "\n\tAdd a path to search for include files", @@ -101,6 +114,8 @@ static const char * const usage_opts_help[] = { "\t\tboth - Both \"linux,phandle\" and \"phandle\" properties", "\n\tEnable/disable warnings (prefix with \"no-\")", "\n\tEnable/disable errors (prefix with \"no-\")", + "\n\tEnable generation of symbols", + "\n\tEnable auto-alias of labels", "\n\tPrint this help and exit", "\n\tPrint version and exit", NULL, @@ -153,7 +168,7 @@ static const char *guess_input_format(const char *fname, const char *fallback) int main(int argc, char *argv[]) { - struct boot_info *bi; + struct dt_info *dti; const char *inform = NULL; const char *outform = NULL; const char *outname = "-"; @@ -169,6 +184,7 @@ int main(int argc, char *argv[]) reservenum = 0; minsize = 0; padsize = 0; + alignsize = 0; while ((opt = util_getopt_long()) != EOF) { switch (opt) { @@ -196,6 +212,12 @@ int main(int argc, char *argv[]) case 'p': padsize = strtol(optarg, NULL, 0); break; + case 'a': + alignsize = strtol(optarg, NULL, 0); + if (!is_power_of_2(alignsize)) + die("Invalid argument \"%d\" to -a option\n", + optarg); + break; case 'f': force = true; break; @@ -234,6 +256,13 @@ int main(int argc, char *argv[]) parse_checks_option(false, true, optarg); break; + case '@': + generate_symbols = 1; + break; + case 'A': + auto_label_aliases = 1; + break; + case 'h': usage(NULL); default: @@ -272,11 +301,11 @@ int main(int argc, char *argv[]) } } if (streq(inform, "dts")) - bi = dt_from_source(arg); + dti = dt_from_source(arg); else if (streq(inform, "fs")) - bi = dt_from_fs(arg); + dti = dt_from_fs(arg); else if(streq(inform, "dtb")) - bi = dt_from_blob(arg); + dti = dt_from_blob(arg); else die("Unknown input format \"%s\"\n", inform); @@ -286,13 +315,29 @@ int main(int argc, char *argv[]) } if (cmdline_boot_cpuid != -1) - bi->boot_cpuid_phys = cmdline_boot_cpuid; + dti->boot_cpuid_phys = cmdline_boot_cpuid; + + fill_fullpaths(dti->dt, ""); + process_checks(force, dti); + + /* on a plugin, generate by default */ + if (dti->dtsflags & DTSF_PLUGIN) { + generate_fixups = 1; + } - fill_fullpaths(bi->dt, ""); - process_checks(force, bi); + if (auto_label_aliases) + generate_label_tree(dti, "aliases", false); + + if (generate_symbols) + generate_label_tree(dti, "__symbols__", true); + + if (generate_fixups) { + generate_fixups_tree(dti, "__fixups__"); + generate_local_fixups_tree(dti, "__local_fixups__"); + } if (sort) - sort_tree(bi); + sort_tree(dti); if (streq(outname, "-")) { outf = stdout; @@ -304,11 +349,11 @@ int main(int argc, char *argv[]) } if (streq(outform, "dts")) { - dt_to_source(outf, bi); + dt_to_source(outf, dti); } else if (streq(outform, "dtb")) { - dt_to_blob(outf, bi, outversion); + dt_to_blob(outf, dti, outversion); } else if (streq(outform, "asm")) { - dt_to_asm(outf, bi, outversion); + dt_to_asm(outf, dti, outversion); } else if (streq(outform, "null")) { /* do nothing */ } else { diff --git a/scripts/dtc/dtc.h b/scripts/dtc/dtc.h index 56212c8df660..c6f125c68ba8 100644 --- a/scripts/dtc/dtc.h +++ b/scripts/dtc/dtc.h @@ -53,7 +53,11 @@ extern int quiet; /* Level of quietness */ extern int reservenum; /* Number of memory reservation slots */ extern int minsize; /* Minimum blob size */ extern int padsize; /* Additional padding to blob */ +extern int alignsize; /* Additional padding to blob accroding to the alignsize */ extern int phandle_format; /* Use linux,phandle or phandle properties */ +extern int generate_symbols; /* generate symbols for nodes with labels */ +extern int generate_fixups; /* generate fixups */ +extern int auto_label_aliases; /* auto generate labels -> aliases */ #define PHANDLE_LEGACY 0x1 #define PHANDLE_EPAPR 0x2 @@ -201,6 +205,8 @@ void delete_property(struct property *prop); void add_child(struct node *parent, struct node *child); void delete_node_by_name(struct node *parent, char *name); void delete_node(struct node *node); +void append_to_property(struct node *node, + char *name, const void *data, int len); const char *get_unitname(struct node *node); struct property *get_property(struct node *node, const char *propname); @@ -235,35 +241,44 @@ struct reserve_info *add_reserve_entry(struct reserve_info *list, struct reserve_info *new); -struct boot_info { +struct dt_info { + unsigned int dtsflags; struct reserve_info *reservelist; - struct node *dt; /* the device tree */ uint32_t boot_cpuid_phys; + struct node *dt; /* the device tree */ }; -struct boot_info *build_boot_info(struct reserve_info *reservelist, - struct node *tree, uint32_t boot_cpuid_phys); -void sort_tree(struct boot_info *bi); +/* DTS version flags definitions */ +#define DTSF_V1 0x0001 /* /dts-v1/ */ +#define DTSF_PLUGIN 0x0002 /* /plugin/ */ + +struct dt_info *build_dt_info(unsigned int dtsflags, + struct reserve_info *reservelist, + struct node *tree, uint32_t boot_cpuid_phys); +void sort_tree(struct dt_info *dti); +void generate_label_tree(struct dt_info *dti, char *name, bool allocph); +void generate_fixups_tree(struct dt_info *dti, char *name); +void generate_local_fixups_tree(struct dt_info *dti, char *name); /* Checks */ void parse_checks_option(bool warn, bool error, const char *arg); -void process_checks(bool force, struct boot_info *bi); +void process_checks(bool force, struct dt_info *dti); /* Flattened trees */ -void dt_to_blob(FILE *f, struct boot_info *bi, int version); -void dt_to_asm(FILE *f, struct boot_info *bi, int version); +void dt_to_blob(FILE *f, struct dt_info *dti, int version); +void dt_to_asm(FILE *f, struct dt_info *dti, int version); -struct boot_info *dt_from_blob(const char *fname); +struct dt_info *dt_from_blob(const char *fname); /* Tree source */ -void dt_to_source(FILE *f, struct boot_info *bi); -struct boot_info *dt_from_source(const char *f); +void dt_to_source(FILE *f, struct dt_info *dti); +struct dt_info *dt_from_source(const char *f); /* FS trees */ -struct boot_info *dt_from_fs(const char *dirname); +struct dt_info *dt_from_fs(const char *dirname); #endif /* _DTC_H */ diff --git a/scripts/dtc/flattree.c b/scripts/dtc/flattree.c index ec14954f5810..ebac548b3fa8 100644 --- a/scripts/dtc/flattree.c +++ b/scripts/dtc/flattree.c @@ -366,7 +366,7 @@ static void make_fdt_header(struct fdt_header *fdt, fdt->size_dt_struct = cpu_to_fdt32(dtsize); } -void dt_to_blob(FILE *f, struct boot_info *bi, int version) +void dt_to_blob(FILE *f, struct dt_info *dti, int version) { struct version_info *vi = NULL; int i; @@ -384,29 +384,36 @@ void dt_to_blob(FILE *f, struct boot_info *bi, int version) if (!vi) die("Unknown device tree blob version %d\n", version); - flatten_tree(bi->dt, &bin_emitter, &dtbuf, &strbuf, vi); + flatten_tree(dti->dt, &bin_emitter, &dtbuf, &strbuf, vi); bin_emit_cell(&dtbuf, FDT_END); - reservebuf = flatten_reserve_list(bi->reservelist, vi); + reservebuf = flatten_reserve_list(dti->reservelist, vi); /* Make header */ make_fdt_header(&fdt, vi, reservebuf.len, dtbuf.len, strbuf.len, - bi->boot_cpuid_phys); + dti->boot_cpuid_phys); /* * If the user asked for more space than is used, adjust the totalsize. */ if (minsize > 0) { padlen = minsize - fdt32_to_cpu(fdt.totalsize); - if ((padlen < 0) && (quiet < 1)) - fprintf(stderr, - "Warning: blob size %d >= minimum size %d\n", - fdt32_to_cpu(fdt.totalsize), minsize); + if (padlen < 0) { + padlen = 0; + if (quiet < 1) + fprintf(stderr, + "Warning: blob size %d >= minimum size %d\n", + fdt32_to_cpu(fdt.totalsize), minsize); + } } if (padsize > 0) padlen = padsize; + if (alignsize > 0) + padlen = ALIGN(fdt32_to_cpu(fdt.totalsize) + padlen, alignsize) + - fdt32_to_cpu(fdt.totalsize); + if (padlen > 0) { int tsize = fdt32_to_cpu(fdt.totalsize); tsize += padlen; @@ -460,7 +467,7 @@ static void dump_stringtable_asm(FILE *f, struct data strbuf) } } -void dt_to_asm(FILE *f, struct boot_info *bi, int version) +void dt_to_asm(FILE *f, struct dt_info *dti, int version) { struct version_info *vi = NULL; int i; @@ -500,7 +507,7 @@ void dt_to_asm(FILE *f, struct boot_info *bi, int version) if (vi->flags & FTF_BOOTCPUID) { fprintf(f, "\t/* boot_cpuid_phys */\n"); - asm_emit_cell(f, bi->boot_cpuid_phys); + asm_emit_cell(f, dti->boot_cpuid_phys); } if (vi->flags & FTF_STRTABSIZE) { @@ -530,7 +537,7 @@ void dt_to_asm(FILE *f, struct boot_info *bi, int version) * Use .long on high and low halfs of u64s to avoid .quad * as it appears .quad isn't available in some assemblers. */ - for (re = bi->reservelist; re; re = re->next) { + for (re = dti->reservelist; re; re = re->next) { struct label *l; for_each_label(re->labels, l) { @@ -550,7 +557,7 @@ void dt_to_asm(FILE *f, struct boot_info *bi, int version) fprintf(f, "\t.long\t0, 0\n\t.long\t0, 0\n"); emit_label(f, symprefix, "struct_start"); - flatten_tree(bi->dt, &asm_emitter, f, &strbuf, vi); + flatten_tree(dti->dt, &asm_emitter, f, &strbuf, vi); fprintf(f, "\t/* FDT_END */\n"); asm_emit_cell(f, FDT_END); @@ -572,6 +579,8 @@ void dt_to_asm(FILE *f, struct boot_info *bi, int version) if (padsize > 0) { fprintf(f, "\t.space\t%d, 0\n", padsize); } + if (alignsize > 0) + asm_emit_align(f, alignsize); emit_label(f, symprefix, "blob_abs_end"); data_free(strbuf); @@ -797,11 +806,15 @@ static struct node *unflatten_tree(struct inbuf *dtbuf, } } while (val != FDT_END_NODE); + if (node->name != flatname) { + free(flatname); + } + return node; } -struct boot_info *dt_from_blob(const char *fname) +struct dt_info *dt_from_blob(const char *fname) { FILE *f; uint32_t magic, totalsize, version, size_dt, boot_cpuid_phys; @@ -929,5 +942,5 @@ struct boot_info *dt_from_blob(const char *fname) fclose(f); - return build_boot_info(reservelist, tree, boot_cpuid_phys); + return build_dt_info(DTSF_V1, reservelist, tree, boot_cpuid_phys); } diff --git a/scripts/dtc/fstree.c b/scripts/dtc/fstree.c index 6d1beec9581d..ae7d06c3c492 100644 --- a/scripts/dtc/fstree.c +++ b/scripts/dtc/fstree.c @@ -79,13 +79,12 @@ static struct node *read_fstree(const char *dirname) return tree; } -struct boot_info *dt_from_fs(const char *dirname) +struct dt_info *dt_from_fs(const char *dirname) { struct node *tree; tree = read_fstree(dirname); tree = name_node(tree, ""); - return build_boot_info(NULL, tree, guess_boot_cpuid(tree)); + return build_dt_info(DTSF_V1, NULL, tree, guess_boot_cpuid(tree)); } - diff --git a/scripts/dtc/libfdt/Makefile.libfdt b/scripts/dtc/libfdt/Makefile.libfdt index 09c322ed82ba..098b3f36e668 100644 --- a/scripts/dtc/libfdt/Makefile.libfdt +++ b/scripts/dtc/libfdt/Makefile.libfdt @@ -7,5 +7,5 @@ LIBFDT_soname = libfdt.$(SHAREDLIB_EXT).1 LIBFDT_INCLUDES = fdt.h libfdt.h libfdt_env.h LIBFDT_VERSION = version.lds LIBFDT_SRCS = fdt.c fdt_ro.c fdt_wip.c fdt_sw.c fdt_rw.c fdt_strerror.c fdt_empty_tree.c \ - fdt_addresses.c + fdt_addresses.c fdt_overlay.c LIBFDT_OBJS = $(LIBFDT_SRCS:%.c=%.o) diff --git a/scripts/dtc/libfdt/fdt_ro.c b/scripts/dtc/libfdt/fdt_ro.c index 50cce864283c..3d00d2eee0e3 100644 --- a/scripts/dtc/libfdt/fdt_ro.c +++ b/scripts/dtc/libfdt/fdt_ro.c @@ -88,6 +88,32 @@ static int _fdt_string_eq(const void *fdt, int stroffset, return (strlen(p) == len) && (memcmp(p, s, len) == 0); } +uint32_t fdt_get_max_phandle(const void *fdt) +{ + uint32_t max_phandle = 0; + int offset; + + for (offset = fdt_next_node(fdt, -1, NULL);; + offset = fdt_next_node(fdt, offset, NULL)) { + uint32_t phandle; + + if (offset == -FDT_ERR_NOTFOUND) + return max_phandle; + + if (offset < 0) + return (uint32_t)-1; + + phandle = fdt_get_phandle(fdt, offset); + if (phandle == (uint32_t)-1) + continue; + + if (phandle > max_phandle) + max_phandle = phandle; + } + + return 0; +} + int fdt_get_mem_rsv(const void *fdt, int n, uint64_t *address, uint64_t *size) { FDT_CHECK_HEADER(fdt); @@ -545,7 +571,7 @@ int fdt_stringlist_count(const void *fdt, int nodeoffset, const char *property) list = fdt_getprop(fdt, nodeoffset, property, &length); if (!list) - return -length; + return length; end = list + length; @@ -571,7 +597,7 @@ int fdt_stringlist_search(const void *fdt, int nodeoffset, const char *property, list = fdt_getprop(fdt, nodeoffset, property, &length); if (!list) - return -length; + return length; len = strlen(string) + 1; end = list + length; diff --git a/scripts/dtc/libfdt/fdt_rw.c b/scripts/dtc/libfdt/fdt_rw.c index 8be02b1f68f3..2eed4f58387c 100644 --- a/scripts/dtc/libfdt/fdt_rw.c +++ b/scripts/dtc/libfdt/fdt_rw.c @@ -191,17 +191,13 @@ int fdt_add_mem_rsv(void *fdt, uint64_t address, uint64_t size) int fdt_del_mem_rsv(void *fdt, int n) { struct fdt_reserve_entry *re = _fdt_mem_rsv_w(fdt, n); - int err; FDT_RW_CHECK_HEADER(fdt); if (n >= fdt_num_mem_rsv(fdt)) return -FDT_ERR_NOTFOUND; - err = _fdt_splice_mem_rsv(fdt, re, 1, 0); - if (err) - return err; - return 0; + return _fdt_splice_mem_rsv(fdt, re, 1, 0); } static int _fdt_resize_property(void *fdt, int nodeoffset, const char *name, diff --git a/scripts/dtc/libfdt/fdt_strerror.c b/scripts/dtc/libfdt/fdt_strerror.c index e6c3ceee8c58..9677a1887e57 100644 --- a/scripts/dtc/libfdt/fdt_strerror.c +++ b/scripts/dtc/libfdt/fdt_strerror.c @@ -69,6 +69,7 @@ static struct fdt_errtabent fdt_errtable[] = { FDT_ERRTABENT(FDT_ERR_BADOFFSET), FDT_ERRTABENT(FDT_ERR_BADPATH), + FDT_ERRTABENT(FDT_ERR_BADPHANDLE), FDT_ERRTABENT(FDT_ERR_BADSTATE), FDT_ERRTABENT(FDT_ERR_TRUNCATED), @@ -76,6 +77,11 @@ static struct fdt_errtabent fdt_errtable[] = { FDT_ERRTABENT(FDT_ERR_BADVERSION), FDT_ERRTABENT(FDT_ERR_BADSTRUCTURE), FDT_ERRTABENT(FDT_ERR_BADLAYOUT), + FDT_ERRTABENT(FDT_ERR_INTERNAL), + FDT_ERRTABENT(FDT_ERR_BADNCELLS), + FDT_ERRTABENT(FDT_ERR_BADVALUE), + FDT_ERRTABENT(FDT_ERR_BADOVERLAY), + FDT_ERRTABENT(FDT_ERR_NOPHANDLES), }; #define FDT_ERRTABSIZE (sizeof(fdt_errtable) / sizeof(fdt_errtable[0])) diff --git a/scripts/dtc/libfdt/fdt_wip.c b/scripts/dtc/libfdt/fdt_wip.c index c5bbb68d3273..6aaab399929c 100644 --- a/scripts/dtc/libfdt/fdt_wip.c +++ b/scripts/dtc/libfdt/fdt_wip.c @@ -55,21 +55,42 @@ #include "libfdt_internal.h" +int fdt_setprop_inplace_namelen_partial(void *fdt, int nodeoffset, + const char *name, int namelen, + uint32_t idx, const void *val, + int len) +{ + void *propval; + int proplen; + + propval = fdt_getprop_namelen_w(fdt, nodeoffset, name, namelen, + &proplen); + if (!propval) + return proplen; + + if (proplen < (len + idx)) + return -FDT_ERR_NOSPACE; + + memcpy((char *)propval + idx, val, len); + return 0; +} + int fdt_setprop_inplace(void *fdt, int nodeoffset, const char *name, const void *val, int len) { - void *propval; + const void *propval; int proplen; - propval = fdt_getprop_w(fdt, nodeoffset, name, &proplen); + propval = fdt_getprop(fdt, nodeoffset, name, &proplen); if (! propval) return proplen; if (proplen != len) return -FDT_ERR_NOSPACE; - memcpy(propval, val, len); - return 0; + return fdt_setprop_inplace_namelen_partial(fdt, nodeoffset, name, + strlen(name), 0, + val, len); } static void _fdt_nop_region(void *start, int len) diff --git a/scripts/dtc/libfdt/libfdt.h b/scripts/dtc/libfdt/libfdt.h index 59ca33976e56..b842b156fa17 100644 --- a/scripts/dtc/libfdt/libfdt.h +++ b/scripts/dtc/libfdt/libfdt.h @@ -61,7 +61,7 @@ #define FDT_ERR_NOTFOUND 1 /* FDT_ERR_NOTFOUND: The requested node or property does not exist */ #define FDT_ERR_EXISTS 2 - /* FDT_ERR_EXISTS: Attemped to create a node or property which + /* FDT_ERR_EXISTS: Attempted to create a node or property which * already exists */ #define FDT_ERR_NOSPACE 3 /* FDT_ERR_NOSPACE: Operation needed to expand the device @@ -79,8 +79,10 @@ * (e.g. missing a leading / for a function which requires an * absolute path) */ #define FDT_ERR_BADPHANDLE 6 - /* FDT_ERR_BADPHANDLE: Function was passed an invalid phandle - * value. phandle values of 0 and -1 are not permitted. */ + /* FDT_ERR_BADPHANDLE: Function was passed an invalid phandle. + * This can be caused either by an invalid phandle property + * length, or the phandle value was either 0 or -1, which are + * not permitted. */ #define FDT_ERR_BADSTATE 7 /* FDT_ERR_BADSTATE: Function was passed an incomplete device * tree created by the sequential-write functions, which is @@ -126,7 +128,16 @@ * value. For example: a property expected to contain a string list * is not NUL-terminated within the length of its value. */ -#define FDT_ERR_MAX 15 +#define FDT_ERR_BADOVERLAY 16 + /* FDT_ERR_BADOVERLAY: The device tree overlay, while + * correctly structured, cannot be applied due to some + * unexpected or missing value, property or node. */ + +#define FDT_ERR_NOPHANDLES 17 + /* FDT_ERR_NOPHANDLES: The device tree doesn't have any + * phandle available anymore without causing an overflow */ + +#define FDT_ERR_MAX 17 /**********************************************************************/ /* Low-level functions (you probably don't need these) */ @@ -168,27 +179,55 @@ int fdt_first_subnode(const void *fdt, int offset); */ int fdt_next_subnode(const void *fdt, int offset); +/** + * fdt_for_each_subnode - iterate over all subnodes of a parent + * + * @node: child node (int, lvalue) + * @fdt: FDT blob (const void *) + * @parent: parent node (int) + * + * This is actually a wrapper around a for loop and would be used like so: + * + * fdt_for_each_subnode(node, fdt, parent) { + * Use node + * ... + * } + * + * if ((node < 0) && (node != -FDT_ERR_NOT_FOUND)) { + * Error handling + * } + * + * Note that this is implemented as a macro and @node is used as + * iterator in the loop. The parent variable be constant or even a + * literal. + * + */ +#define fdt_for_each_subnode(node, fdt, parent) \ + for (node = fdt_first_subnode(fdt, parent); \ + node >= 0; \ + node = fdt_next_subnode(fdt, node)) + /**********************************************************************/ /* General functions */ /**********************************************************************/ #define fdt_get_header(fdt, field) \ (fdt32_to_cpu(((const struct fdt_header *)(fdt))->field)) -#define fdt_magic(fdt) (fdt_get_header(fdt, magic)) +#define fdt_magic(fdt) (fdt_get_header(fdt, magic)) #define fdt_totalsize(fdt) (fdt_get_header(fdt, totalsize)) #define fdt_off_dt_struct(fdt) (fdt_get_header(fdt, off_dt_struct)) #define fdt_off_dt_strings(fdt) (fdt_get_header(fdt, off_dt_strings)) #define fdt_off_mem_rsvmap(fdt) (fdt_get_header(fdt, off_mem_rsvmap)) #define fdt_version(fdt) (fdt_get_header(fdt, version)) -#define fdt_last_comp_version(fdt) (fdt_get_header(fdt, last_comp_version)) -#define fdt_boot_cpuid_phys(fdt) (fdt_get_header(fdt, boot_cpuid_phys)) -#define fdt_size_dt_strings(fdt) (fdt_get_header(fdt, size_dt_strings)) +#define fdt_last_comp_version(fdt) (fdt_get_header(fdt, last_comp_version)) +#define fdt_boot_cpuid_phys(fdt) (fdt_get_header(fdt, boot_cpuid_phys)) +#define fdt_size_dt_strings(fdt) (fdt_get_header(fdt, size_dt_strings)) #define fdt_size_dt_struct(fdt) (fdt_get_header(fdt, size_dt_struct)) #define __fdt_set_hdr(name) \ static inline void fdt_set_##name(void *fdt, uint32_t val) \ { \ - struct fdt_header *fdth = (struct fdt_header*)fdt; \ + struct fdt_header *fdth = (struct fdt_header *)fdt; \ fdth->name = cpu_to_fdt32(val); \ } __fdt_set_hdr(magic); @@ -259,6 +298,21 @@ int fdt_move(const void *fdt, void *buf, int bufsize); const char *fdt_string(const void *fdt, int stroffset); /** + * fdt_get_max_phandle - retrieves the highest phandle in a tree + * @fdt: pointer to the device tree blob + * + * fdt_get_max_phandle retrieves the highest phandle in the given + * device tree. This will ignore badly formatted phandles, or phandles + * with a value of 0 or -1. + * + * returns: + * the highest phandle on success + * 0, if no phandle was found in the device tree + * -1, if an error occurred + */ +uint32_t fdt_get_max_phandle(const void *fdt); + +/** * fdt_num_mem_rsv - retrieve the number of memory reserve map entries * @fdt: pointer to the device tree blob * @@ -318,8 +372,9 @@ int fdt_subnode_offset_namelen(const void *fdt, int parentoffset, * returns: * structure block offset of the requested subnode (>=0), on success * -FDT_ERR_NOTFOUND, if the requested subnode does not exist - * -FDT_ERR_BADOFFSET, if parentoffset did not point to an FDT_BEGIN_NODE tag - * -FDT_ERR_BADMAGIC, + * -FDT_ERR_BADOFFSET, if parentoffset did not point to an FDT_BEGIN_NODE + * tag + * -FDT_ERR_BADMAGIC, * -FDT_ERR_BADVERSION, * -FDT_ERR_BADSTATE, * -FDT_ERR_BADSTRUCTURE, @@ -351,7 +406,8 @@ int fdt_path_offset_namelen(const void *fdt, const char *path, int namelen); * address). * * returns: - * structure block offset of the node with the requested path (>=0), on success + * structure block offset of the node with the requested path (>=0), on + * success * -FDT_ERR_BADPATH, given path does not begin with '/' or is invalid * -FDT_ERR_NOTFOUND, if the requested node does not exist * -FDT_ERR_BADMAGIC, @@ -375,10 +431,12 @@ int fdt_path_offset(const void *fdt, const char *path); * * returns: * pointer to the node's name, on success - * If lenp is non-NULL, *lenp contains the length of that name (>=0) + * If lenp is non-NULL, *lenp contains the length of that name + * (>=0) * NULL, on error * if lenp is non-NULL *lenp contains an error code (<0): - * -FDT_ERR_BADOFFSET, nodeoffset did not point to FDT_BEGIN_NODE tag + * -FDT_ERR_BADOFFSET, nodeoffset did not point to FDT_BEGIN_NODE + * tag * -FDT_ERR_BADMAGIC, * -FDT_ERR_BADVERSION, * -FDT_ERR_BADSTATE, standard meanings @@ -427,6 +485,33 @@ int fdt_first_property_offset(const void *fdt, int nodeoffset); int fdt_next_property_offset(const void *fdt, int offset); /** + * fdt_for_each_property_offset - iterate over all properties of a node + * + * @property_offset: property offset (int, lvalue) + * @fdt: FDT blob (const void *) + * @node: node offset (int) + * + * This is actually a wrapper around a for loop and would be used like so: + * + * fdt_for_each_property_offset(property, fdt, node) { + * Use property + * ... + * } + * + * if ((property < 0) && (property != -FDT_ERR_NOT_FOUND)) { + * Error handling + * } + * + * Note that this is implemented as a macro and property is used as + * iterator in the loop. The node variable can be constant or even a + * literal. + */ +#define fdt_for_each_property_offset(property, fdt, node) \ + for (property = fdt_first_property_offset(fdt, node); \ + property >= 0; \ + property = fdt_next_property_offset(fdt, property)) + +/** * fdt_get_property_by_offset - retrieve the property at a given offset * @fdt: pointer to the device tree blob * @offset: offset of the property to retrieve @@ -490,7 +575,8 @@ const struct fdt_property *fdt_get_property_namelen(const void *fdt, * NULL, on error * if lenp is non-NULL, *lenp contains an error code (<0): * -FDT_ERR_NOTFOUND, node does not have named property - * -FDT_ERR_BADOFFSET, nodeoffset did not point to FDT_BEGIN_NODE tag + * -FDT_ERR_BADOFFSET, nodeoffset did not point to FDT_BEGIN_NODE + * tag * -FDT_ERR_BADMAGIC, * -FDT_ERR_BADVERSION, * -FDT_ERR_BADSTATE, @@ -554,6 +640,13 @@ const void *fdt_getprop_by_offset(const void *fdt, int offset, */ const void *fdt_getprop_namelen(const void *fdt, int nodeoffset, const char *name, int namelen, int *lenp); +static inline void *fdt_getprop_namelen_w(void *fdt, int nodeoffset, + const char *name, int namelen, + int *lenp) +{ + return (void *)(uintptr_t)fdt_getprop_namelen(fdt, nodeoffset, name, + namelen, lenp); +} /** * fdt_getprop - retrieve the value of a given property @@ -575,7 +668,8 @@ const void *fdt_getprop_namelen(const void *fdt, int nodeoffset, * NULL, on error * if lenp is non-NULL, *lenp contains an error code (<0): * -FDT_ERR_NOTFOUND, node does not have named property - * -FDT_ERR_BADOFFSET, nodeoffset did not point to FDT_BEGIN_NODE tag + * -FDT_ERR_BADOFFSET, nodeoffset did not point to FDT_BEGIN_NODE + * tag * -FDT_ERR_BADMAGIC, * -FDT_ERR_BADVERSION, * -FDT_ERR_BADSTATE, @@ -617,7 +711,7 @@ const char *fdt_get_alias_namelen(const void *fdt, const char *name, int namelen); /** - * fdt_get_alias - retreive the path referenced by a given alias + * fdt_get_alias - retrieve the path referenced by a given alias * @fdt: pointer to the device tree blob * @name: name of the alias th look up * @@ -647,7 +741,7 @@ const char *fdt_get_alias(const void *fdt, const char *name); * 0, on success * buf contains the absolute path of the node at * nodeoffset, as a NUL-terminated string. - * -FDT_ERR_BADOFFSET, nodeoffset does not refer to a BEGIN_NODE tag + * -FDT_ERR_BADOFFSET, nodeoffset does not refer to a BEGIN_NODE tag * -FDT_ERR_NOSPACE, the path of the given node is longer than (bufsize-1) * characters and will not fit in the given buffer. * -FDT_ERR_BADMAGIC, @@ -677,11 +771,11 @@ int fdt_get_path(const void *fdt, int nodeoffset, char *buf, int buflen); * structure from the start to nodeoffset. * * returns: - * structure block offset of the node at node offset's ancestor * of depth supernodedepth (>=0), on success - * -FDT_ERR_BADOFFSET, nodeoffset does not refer to a BEGIN_NODE tag -* -FDT_ERR_NOTFOUND, supernodedepth was greater than the depth of nodeoffset + * -FDT_ERR_BADOFFSET, nodeoffset does not refer to a BEGIN_NODE tag + * -FDT_ERR_NOTFOUND, supernodedepth was greater than the depth of + * nodeoffset * -FDT_ERR_BADMAGIC, * -FDT_ERR_BADVERSION, * -FDT_ERR_BADSTATE, @@ -703,7 +797,7 @@ int fdt_supernode_atdepth_offset(const void *fdt, int nodeoffset, * * returns: * depth of the node at nodeoffset (>=0), on success - * -FDT_ERR_BADOFFSET, nodeoffset does not refer to a BEGIN_NODE tag + * -FDT_ERR_BADOFFSET, nodeoffset does not refer to a BEGIN_NODE tag * -FDT_ERR_BADMAGIC, * -FDT_ERR_BADVERSION, * -FDT_ERR_BADSTATE, @@ -726,7 +820,7 @@ int fdt_node_depth(const void *fdt, int nodeoffset); * returns: * structure block offset of the parent of the node at nodeoffset * (>=0), on success - * -FDT_ERR_BADOFFSET, nodeoffset does not refer to a BEGIN_NODE tag + * -FDT_ERR_BADOFFSET, nodeoffset does not refer to a BEGIN_NODE tag * -FDT_ERR_BADMAGIC, * -FDT_ERR_BADVERSION, * -FDT_ERR_BADSTATE, @@ -766,7 +860,7 @@ int fdt_parent_offset(const void *fdt, int nodeoffset); * on success * -FDT_ERR_NOTFOUND, no node matching the criterion exists in the * tree after startoffset - * -FDT_ERR_BADOFFSET, nodeoffset does not refer to a BEGIN_NODE tag + * -FDT_ERR_BADOFFSET, nodeoffset does not refer to a BEGIN_NODE tag * -FDT_ERR_BADMAGIC, * -FDT_ERR_BADVERSION, * -FDT_ERR_BADSTATE, @@ -813,7 +907,7 @@ int fdt_node_offset_by_phandle(const void *fdt, uint32_t phandle); * 1, if the node has a 'compatible' property, but it does not list * the given string * -FDT_ERR_NOTFOUND, if the given node has no 'compatible' property - * -FDT_ERR_BADOFFSET, if nodeoffset does not refer to a BEGIN_NODE tag + * -FDT_ERR_BADOFFSET, if nodeoffset does not refer to a BEGIN_NODE tag * -FDT_ERR_BADMAGIC, * -FDT_ERR_BADVERSION, * -FDT_ERR_BADSTATE, @@ -850,7 +944,7 @@ int fdt_node_check_compatible(const void *fdt, int nodeoffset, * on success * -FDT_ERR_NOTFOUND, no node matching the criterion exists in the * tree after startoffset - * -FDT_ERR_BADOFFSET, nodeoffset does not refer to a BEGIN_NODE tag + * -FDT_ERR_BADOFFSET, nodeoffset does not refer to a BEGIN_NODE tag * -FDT_ERR_BADMAGIC, * -FDT_ERR_BADVERSION, * -FDT_ERR_BADSTATE, @@ -960,7 +1054,8 @@ const char *fdt_stringlist_get(const void *fdt, int nodeoffset, * returns: * 0 <= n < FDT_MAX_NCELLS, on success * 2, if the node has no #address-cells property - * -FDT_ERR_BADNCELLS, if the node has a badly formatted or invalid #address-cells property + * -FDT_ERR_BADNCELLS, if the node has a badly formatted or invalid + * #address-cells property * -FDT_ERR_BADMAGIC, * -FDT_ERR_BADVERSION, * -FDT_ERR_BADSTATE, @@ -980,7 +1075,8 @@ int fdt_address_cells(const void *fdt, int nodeoffset); * returns: * 0 <= n < FDT_MAX_NCELLS, on success * 2, if the node has no #address-cells property - * -FDT_ERR_BADNCELLS, if the node has a badly formatted or invalid #size-cells property + * -FDT_ERR_BADNCELLS, if the node has a badly formatted or invalid + * #size-cells property * -FDT_ERR_BADMAGIC, * -FDT_ERR_BADVERSION, * -FDT_ERR_BADSTATE, @@ -995,6 +1091,27 @@ int fdt_size_cells(const void *fdt, int nodeoffset); /**********************************************************************/ /** + * fdt_setprop_inplace_namelen_partial - change a property's value, + * but not its size + * @fdt: pointer to the device tree blob + * @nodeoffset: offset of the node whose property to change + * @name: name of the property to change + * @namelen: number of characters of name to consider + * @idx: index of the property to change in the array + * @val: pointer to data to replace the property value with + * @len: length of the property value + * + * Identical to fdt_setprop_inplace(), but modifies the given property + * starting from the given index, and using only the first characters + * of the name. It is useful when you want to manipulate only one value of + * an array and you have a string that doesn't end with \0. + */ +int fdt_setprop_inplace_namelen_partial(void *fdt, int nodeoffset, + const char *name, int namelen, + uint32_t idx, const void *val, + int len); + +/** * fdt_setprop_inplace - change a property's value, but not its size * @fdt: pointer to the device tree blob * @nodeoffset: offset of the node whose property to change @@ -1604,9 +1721,11 @@ int fdt_add_subnode_namelen(void *fdt, int parentoffset, * change the offsets of some existing nodes. * returns: - * structure block offset of the created nodeequested subnode (>=0), on success + * structure block offset of the created nodeequested subnode (>=0), on + * success * -FDT_ERR_NOTFOUND, if the requested subnode does not exist - * -FDT_ERR_BADOFFSET, if parentoffset did not point to an FDT_BEGIN_NODE tag + * -FDT_ERR_BADOFFSET, if parentoffset did not point to an FDT_BEGIN_NODE + * tag * -FDT_ERR_EXISTS, if the node at parentoffset already has a subnode of * the given name * -FDT_ERR_NOSPACE, if there is insufficient free space in the @@ -1644,6 +1763,37 @@ int fdt_add_subnode(void *fdt, int parentoffset, const char *name); */ int fdt_del_node(void *fdt, int nodeoffset); +/** + * fdt_overlay_apply - Applies a DT overlay on a base DT + * @fdt: pointer to the base device tree blob + * @fdto: pointer to the device tree overlay blob + * + * fdt_overlay_apply() will apply the given device tree overlay on the + * given base device tree. + * + * Expect the base device tree to be modified, even if the function + * returns an error. + * + * returns: + * 0, on success + * -FDT_ERR_NOSPACE, there's not enough space in the base device tree + * -FDT_ERR_NOTFOUND, the overlay points to some inexistant nodes or + * properties in the base DT + * -FDT_ERR_BADPHANDLE, + * -FDT_ERR_BADOVERLAY, + * -FDT_ERR_NOPHANDLES, + * -FDT_ERR_INTERNAL, + * -FDT_ERR_BADLAYOUT, + * -FDT_ERR_BADMAGIC, + * -FDT_ERR_BADOFFSET, + * -FDT_ERR_BADPATH, + * -FDT_ERR_BADVERSION, + * -FDT_ERR_BADSTRUCTURE, + * -FDT_ERR_BADSTATE, + * -FDT_ERR_TRUNCATED, standard meanings + */ +int fdt_overlay_apply(void *fdt, void *fdto); + /**********************************************************************/ /* Debugging / informational functions */ /**********************************************************************/ diff --git a/scripts/dtc/libfdt/libfdt_env.h b/scripts/dtc/libfdt/libfdt_env.h index 9dea97dfff81..99f936dacc60 100644 --- a/scripts/dtc/libfdt/libfdt_env.h +++ b/scripts/dtc/libfdt/libfdt_env.h @@ -54,6 +54,7 @@ #include <stddef.h> #include <stdint.h> +#include <stdlib.h> #include <string.h> #ifdef __CHECKER__ diff --git a/scripts/dtc/livetree.c b/scripts/dtc/livetree.c index e229b84432f9..afa2f67b142a 100644 --- a/scripts/dtc/livetree.c +++ b/scripts/dtc/livetree.c @@ -204,7 +204,7 @@ struct node *merge_nodes(struct node *old_node, struct node *new_node) } } - /* if no collision occured, add child to the old node. */ + /* if no collision occurred, add child to the old node. */ if (new_child) add_child(old_node, new_child); } @@ -296,6 +296,23 @@ void delete_node(struct node *node) delete_labels(&node->labels); } +void append_to_property(struct node *node, + char *name, const void *data, int len) +{ + struct data d; + struct property *p; + + p = get_property(node, name); + if (p) { + d = data_append_data(p->val, data, len); + p->val = d; + } else { + d = data_append_data(empty_data, data, len); + p = build_property(name, d); + add_property(node, p); + } +} + struct reserve_info *build_reserve_entry(uint64_t address, uint64_t size) { struct reserve_info *new = xmalloc(sizeof(*new)); @@ -335,17 +352,19 @@ struct reserve_info *add_reserve_entry(struct reserve_info *list, return list; } -struct boot_info *build_boot_info(struct reserve_info *reservelist, - struct node *tree, uint32_t boot_cpuid_phys) +struct dt_info *build_dt_info(unsigned int dtsflags, + struct reserve_info *reservelist, + struct node *tree, uint32_t boot_cpuid_phys) { - struct boot_info *bi; + struct dt_info *dti; - bi = xmalloc(sizeof(*bi)); - bi->reservelist = reservelist; - bi->dt = tree; - bi->boot_cpuid_phys = boot_cpuid_phys; + dti = xmalloc(sizeof(*dti)); + dti->dtsflags = dtsflags; + dti->reservelist = reservelist; + dti->dt = tree; + dti->boot_cpuid_phys = boot_cpuid_phys; - return bi; + return dti; } /* @@ -592,12 +611,12 @@ static int cmp_reserve_info(const void *ax, const void *bx) return 0; } -static void sort_reserve_entries(struct boot_info *bi) +static void sort_reserve_entries(struct dt_info *dti) { struct reserve_info *ri, **tbl; int n = 0, i = 0; - for (ri = bi->reservelist; + for (ri = dti->reservelist; ri; ri = ri->next) n++; @@ -607,14 +626,14 @@ static void sort_reserve_entries(struct boot_info *bi) tbl = xmalloc(n * sizeof(*tbl)); - for (ri = bi->reservelist; + for (ri = dti->reservelist; ri; ri = ri->next) tbl[i++] = ri; qsort(tbl, n, sizeof(*tbl), cmp_reserve_info); - bi->reservelist = tbl[0]; + dti->reservelist = tbl[0]; for (i = 0; i < (n-1); i++) tbl[i]->next = tbl[i+1]; tbl[n-1]->next = NULL; @@ -704,8 +723,256 @@ static void sort_node(struct node *node) sort_node(c); } -void sort_tree(struct boot_info *bi) +void sort_tree(struct dt_info *dti) +{ + sort_reserve_entries(dti); + sort_node(dti->dt); +} + +/* utility helper to avoid code duplication */ +static struct node *build_and_name_child_node(struct node *parent, char *name) +{ + struct node *node; + + node = build_node(NULL, NULL); + name_node(node, xstrdup(name)); + add_child(parent, node); + + return node; +} + +static struct node *build_root_node(struct node *dt, char *name) +{ + struct node *an; + + an = get_subnode(dt, name); + if (!an) + an = build_and_name_child_node(dt, name); + + if (!an) + die("Could not build root node /%s\n", name); + + return an; +} + +static bool any_label_tree(struct dt_info *dti, struct node *node) +{ + struct node *c; + + if (node->labels) + return true; + + for_each_child(node, c) + if (any_label_tree(dti, c)) + return true; + + return false; +} + +static void generate_label_tree_internal(struct dt_info *dti, + struct node *an, struct node *node, + bool allocph) +{ + struct node *dt = dti->dt; + struct node *c; + struct property *p; + struct label *l; + + /* if there are labels */ + if (node->labels) { + + /* now add the label in the node */ + for_each_label(node->labels, l) { + + /* check whether the label already exists */ + p = get_property(an, l->label); + if (p) { + fprintf(stderr, "WARNING: label %s already" + " exists in /%s", l->label, + an->name); + continue; + } + + /* insert it */ + p = build_property(l->label, + data_copy_mem(node->fullpath, + strlen(node->fullpath) + 1)); + add_property(an, p); + } + + /* force allocation of a phandle for this node */ + if (allocph) + (void)get_node_phandle(dt, node); + } + + for_each_child(node, c) + generate_label_tree_internal(dti, an, c, allocph); +} + +static bool any_fixup_tree(struct dt_info *dti, struct node *node) +{ + struct node *c; + struct property *prop; + struct marker *m; + + for_each_property(node, prop) { + m = prop->val.markers; + for_each_marker_of_type(m, REF_PHANDLE) { + if (!get_node_by_ref(dti->dt, m->ref)) + return true; + } + } + + for_each_child(node, c) { + if (any_fixup_tree(dti, c)) + return true; + } + + return false; +} + +static void add_fixup_entry(struct dt_info *dti, struct node *fn, + struct node *node, struct property *prop, + struct marker *m) { - sort_reserve_entries(bi); - sort_node(bi->dt); + char *entry; + + /* m->ref can only be a REF_PHANDLE, but check anyway */ + assert(m->type == REF_PHANDLE); + + /* there shouldn't be any ':' in the arguments */ + if (strchr(node->fullpath, ':') || strchr(prop->name, ':')) + die("arguments should not contain ':'\n"); + + xasprintf(&entry, "%s:%s:%u", + node->fullpath, prop->name, m->offset); + append_to_property(fn, m->ref, entry, strlen(entry) + 1); +} + +static void generate_fixups_tree_internal(struct dt_info *dti, + struct node *fn, + struct node *node) +{ + struct node *dt = dti->dt; + struct node *c; + struct property *prop; + struct marker *m; + struct node *refnode; + + for_each_property(node, prop) { + m = prop->val.markers; + for_each_marker_of_type(m, REF_PHANDLE) { + refnode = get_node_by_ref(dt, m->ref); + if (!refnode) + add_fixup_entry(dti, fn, node, prop, m); + } + } + + for_each_child(node, c) + generate_fixups_tree_internal(dti, fn, c); +} + +static bool any_local_fixup_tree(struct dt_info *dti, struct node *node) +{ + struct node *c; + struct property *prop; + struct marker *m; + + for_each_property(node, prop) { + m = prop->val.markers; + for_each_marker_of_type(m, REF_PHANDLE) { + if (get_node_by_ref(dti->dt, m->ref)) + return true; + } + } + + for_each_child(node, c) { + if (any_local_fixup_tree(dti, c)) + return true; + } + + return false; +} + +static void add_local_fixup_entry(struct dt_info *dti, + struct node *lfn, struct node *node, + struct property *prop, struct marker *m, + struct node *refnode) +{ + struct node *wn, *nwn; /* local fixup node, walk node, new */ + uint32_t value_32; + char **compp; + int i, depth; + + /* walk back retreiving depth */ + depth = 0; + for (wn = node; wn; wn = wn->parent) + depth++; + + /* allocate name array */ + compp = xmalloc(sizeof(*compp) * depth); + + /* store names in the array */ + for (wn = node, i = depth - 1; wn; wn = wn->parent, i--) + compp[i] = wn->name; + + /* walk the path components creating nodes if they don't exist */ + for (wn = lfn, i = 1; i < depth; i++, wn = nwn) { + /* if no node exists, create it */ + nwn = get_subnode(wn, compp[i]); + if (!nwn) + nwn = build_and_name_child_node(wn, compp[i]); + } + + free(compp); + + value_32 = cpu_to_fdt32(m->offset); + append_to_property(wn, prop->name, &value_32, sizeof(value_32)); +} + +static void generate_local_fixups_tree_internal(struct dt_info *dti, + struct node *lfn, + struct node *node) +{ + struct node *dt = dti->dt; + struct node *c; + struct property *prop; + struct marker *m; + struct node *refnode; + + for_each_property(node, prop) { + m = prop->val.markers; + for_each_marker_of_type(m, REF_PHANDLE) { + refnode = get_node_by_ref(dt, m->ref); + if (refnode) + add_local_fixup_entry(dti, lfn, node, prop, m, refnode); + } + } + + for_each_child(node, c) + generate_local_fixups_tree_internal(dti, lfn, c); +} + +void generate_label_tree(struct dt_info *dti, char *name, bool allocph) +{ + if (!any_label_tree(dti, dti->dt)) + return; + generate_label_tree_internal(dti, build_root_node(dti->dt, name), + dti->dt, allocph); +} + +void generate_fixups_tree(struct dt_info *dti, char *name) +{ + if (!any_fixup_tree(dti, dti->dt)) + return; + generate_fixups_tree_internal(dti, build_root_node(dti->dt, name), + dti->dt); +} + +void generate_local_fixups_tree(struct dt_info *dti, char *name) +{ + if (!any_local_fixup_tree(dti, dti->dt)) + return; + generate_local_fixups_tree_internal(dti, build_root_node(dti->dt, name), + dti->dt); } diff --git a/scripts/dtc/srcpos.c b/scripts/dtc/srcpos.c index f534c22a888d..aa3aad04cec4 100644 --- a/scripts/dtc/srcpos.c +++ b/scripts/dtc/srcpos.c @@ -246,46 +246,27 @@ srcpos_copy(struct srcpos *pos) return pos_new; } - - -void -srcpos_dump(struct srcpos *pos) -{ - printf("file : \"%s\"\n", - pos->file ? (char *) pos->file : "<no file>"); - printf("first_line : %d\n", pos->first_line); - printf("first_column: %d\n", pos->first_column); - printf("last_line : %d\n", pos->last_line); - printf("last_column : %d\n", pos->last_column); - printf("file : %s\n", pos->file->name); -} - - char * srcpos_string(struct srcpos *pos) { const char *fname = "<no-file>"; char *pos_str; - int rc; if (pos) fname = pos->file->name; if (pos->first_line != pos->last_line) - rc = asprintf(&pos_str, "%s:%d.%d-%d.%d", fname, - pos->first_line, pos->first_column, - pos->last_line, pos->last_column); + xasprintf(&pos_str, "%s:%d.%d-%d.%d", fname, + pos->first_line, pos->first_column, + pos->last_line, pos->last_column); else if (pos->first_column != pos->last_column) - rc = asprintf(&pos_str, "%s:%d.%d-%d", fname, - pos->first_line, pos->first_column, - pos->last_column); + xasprintf(&pos_str, "%s:%d.%d-%d", fname, + pos->first_line, pos->first_column, + pos->last_column); else - rc = asprintf(&pos_str, "%s:%d.%d", fname, - pos->first_line, pos->first_column); - - if (rc == -1) - die("Couldn't allocate in srcpos string"); + xasprintf(&pos_str, "%s:%d.%d", fname, + pos->first_line, pos->first_column); return pos_str; } diff --git a/scripts/dtc/srcpos.h b/scripts/dtc/srcpos.h index f81827bd684a..2cdfcd82e95e 100644 --- a/scripts/dtc/srcpos.h +++ b/scripts/dtc/srcpos.h @@ -105,7 +105,6 @@ extern struct srcpos srcpos_empty; extern void srcpos_update(struct srcpos *pos, const char *text, int len); extern struct srcpos *srcpos_copy(struct srcpos *pos); extern char *srcpos_string(struct srcpos *pos); -extern void srcpos_dump(struct srcpos *pos); extern void srcpos_verror(struct srcpos *pos, const char *prefix, const char *fmt, va_list va) diff --git a/scripts/dtc/treesource.c b/scripts/dtc/treesource.c index a55d1d128cce..c9d8967969f9 100644 --- a/scripts/dtc/treesource.c +++ b/scripts/dtc/treesource.c @@ -25,12 +25,12 @@ extern FILE *yyin; extern int yyparse(void); extern YYLTYPE yylloc; -struct boot_info *the_boot_info; +struct dt_info *parser_output; bool treesource_error; -struct boot_info *dt_from_source(const char *fname) +struct dt_info *dt_from_source(const char *fname) { - the_boot_info = NULL; + parser_output = NULL; treesource_error = false; srcfile_push(fname); @@ -43,7 +43,7 @@ struct boot_info *dt_from_source(const char *fname) if (treesource_error) die("Syntax error parsing input tree\n"); - return the_boot_info; + return parser_output; } static void write_prefix(FILE *f, int level) @@ -263,13 +263,13 @@ static void write_tree_source_node(FILE *f, struct node *tree, int level) } -void dt_to_source(FILE *f, struct boot_info *bi) +void dt_to_source(FILE *f, struct dt_info *dti) { struct reserve_info *re; fprintf(f, "/dts-v1/;\n\n"); - for (re = bi->reservelist; re; re = re->next) { + for (re = dti->reservelist; re; re = re->next) { struct label *l; for_each_label(re->labels, l) @@ -279,6 +279,6 @@ void dt_to_source(FILE *f, struct boot_info *bi) (unsigned long long)re->re.size); } - write_tree_source_node(f, bi->dt, 0); + write_tree_source_node(f, dti->dt, 0); } diff --git a/scripts/dtc/util.c b/scripts/dtc/util.c index fb124eea4919..3550f86bd6df 100644 --- a/scripts/dtc/util.c +++ b/scripts/dtc/util.c @@ -46,6 +46,36 @@ char *xstrdup(const char *s) return d; } +/* based in part from (3) vsnprintf */ +int xasprintf(char **strp, const char *fmt, ...) +{ + int n, size = 128; /* start with 128 bytes */ + char *p; + va_list ap; + + /* initial pointer is NULL making the fist realloc to be malloc */ + p = NULL; + while (1) { + p = xrealloc(p, size); + + /* Try to print in the allocated space. */ + va_start(ap, fmt); + n = vsnprintf(p, size, fmt, ap); + va_end(ap); + + /* If that worked, return the string. */ + if (n > -1 && n < size) + break; + /* Else try again with more space. */ + if (n > -1) /* glibc 2.1 */ + size = n + 1; /* precisely what is needed */ + else /* glibc 2.0 */ + size *= 2; /* twice the old size */ + } + *strp = p; + return strlen(p); +} + char *join_path(const char *path, const char *name) { int lenp = strlen(path); diff --git a/scripts/dtc/util.h b/scripts/dtc/util.h index f800b6011fb1..f5c4f1b50d30 100644 --- a/scripts/dtc/util.h +++ b/scripts/dtc/util.h @@ -59,6 +59,7 @@ static inline void *xrealloc(void *p, size_t len) } extern char *xstrdup(const char *s); +extern int xasprintf(char **strp, const char *fmt, ...); extern char *join_path(const char *path, const char *name); /** diff --git a/scripts/dtc/version_gen.h b/scripts/dtc/version_gen.h index ad9b05ae698b..16c2e53a85e3 100644 --- a/scripts/dtc/version_gen.h +++ b/scripts/dtc/version_gen.h @@ -1 +1 @@ -#define DTC_VERSION "DTC 1.4.1-g53bf130b" +#define DTC_VERSION "DTC 1.4.2-g0931cea3" diff --git a/scripts/kernel-doc b/scripts/kernel-doc index 030fc633acd4..33c85dfdfce9 100755 --- a/scripts/kernel-doc +++ b/scripts/kernel-doc @@ -199,12 +199,12 @@ EOF # 'funcname()' - function # '$ENVVAR' - environmental variable # '&struct_name' - name of a structure (up to two words including 'struct') +# '&struct_name.member' - name of a structure member # '@parameter' - name of a parameter # '%CONST' - name of a constant. ## init lots of data - my $errors = 0; my $warnings = 0; my $anon_struct_union = 0; @@ -214,14 +214,19 @@ my $type_constant = '\%([-_\w]+)'; my $type_func = '(\w+)\(\)'; my $type_param = '\@(\w+(\.\.\.)?)'; my $type_fp_param = '\@(\w+)\(\)'; # Special RST handling for func ptr params -my $type_struct = '\&((struct\s*)*[_\w]+)'; -my $type_struct_xml = '\\&((struct\s*)*[_\w]+)'; my $type_env = '(\$\w+)'; -my $type_enum_full = '\&(enum)\s*([_\w]+)'; -my $type_struct_full = '\&(struct)\s*([_\w]+)'; -my $type_typedef_full = '\&(typedef)\s*([_\w]+)'; -my $type_union_full = '\&(union)\s*([_\w]+)'; -my $type_member = '\&([_\w]+)((\.|->)[_\w]+)'; +my $type_enum = '\&(enum\s*([_\w]+))'; +my $type_struct = '\&(struct\s*([_\w]+))'; +my $type_typedef = '\&(typedef\s*([_\w]+))'; +my $type_union = '\&(union\s*([_\w]+))'; +my $type_member = '\&([_\w]+)(\.|->)([_\w]+)'; +my $type_fallback = '\&([_\w]+)'; +my $type_enum_xml = '\&(enum\s*([_\w]+))'; +my $type_struct_xml = '\&(struct\s*([_\w]+))'; +my $type_typedef_xml = '\&(typedef\s*([_\w]+))'; +my $type_union_xml = '\&(union\s*([_\w]+))'; +my $type_member_xml = '\&([_\w]+)(\.|-\>)([_\w]+)'; +my $type_fallback_xml = '\&([_\w]+)'; my $type_member_func = $type_member . '\(\)'; # Output conversion substitutions. @@ -231,9 +236,14 @@ my $type_member_func = $type_member . '\(\)'; my @highlights_html = ( [$type_constant, "<i>\$1</i>"], [$type_func, "<b>\$1</b>"], + [$type_enum_xml, "<i>\$1</i>"], [$type_struct_xml, "<i>\$1</i>"], + [$type_typedef_xml, "<i>\$1</i>"], + [$type_union_xml, "<i>\$1</i>"], [$type_env, "<b><i>\$1</i></b>"], - [$type_param, "<tt><b>\$1</b></tt>"] + [$type_param, "<tt><b>\$1</b></tt>"], + [$type_member_xml, "<tt><i>\$1</i>\$2\$3</tt>"], + [$type_fallback_xml, "<i>\$1</i>"] ); my $local_lt = "\\\\\\\\lt:"; my $local_gt = "\\\\\\\\gt:"; @@ -243,9 +253,14 @@ my $blankline_html = $local_lt . "p" . $local_gt; # was "<p>" my @highlights_html5 = ( [$type_constant, "<span class=\"const\">\$1</span>"], [$type_func, "<span class=\"func\">\$1</span>"], + [$type_enum_xml, "<span class=\"enum\">\$1</span>"], [$type_struct_xml, "<span class=\"struct\">\$1</span>"], + [$type_typedef_xml, "<span class=\"typedef\">\$1</span>"], + [$type_union_xml, "<span class=\"union\">\$1</span>"], [$type_env, "<span class=\"env\">\$1</span>"], - [$type_param, "<span class=\"param\">\$1</span>]"] + [$type_param, "<span class=\"param\">\$1</span>]"], + [$type_member_xml, "<span class=\"literal\"><span class=\"struct\">\$1</span>\$2<span class=\"member\">\$3</span></span>"], + [$type_fallback_xml, "<span class=\"struct\">\$1</span>"] ); my $blankline_html5 = $local_lt . "br /" . $local_gt; @@ -253,10 +268,15 @@ my $blankline_html5 = $local_lt . "br /" . $local_gt; my @highlights_xml = ( ["([^=])\\\"([^\\\"<]+)\\\"", "\$1<quote>\$2</quote>"], [$type_constant, "<constant>\$1</constant>"], + [$type_enum_xml, "<type>\$1</type>"], [$type_struct_xml, "<structname>\$1</structname>"], + [$type_typedef_xml, "<type>\$1</type>"], + [$type_union_xml, "<structname>\$1</structname>"], [$type_param, "<parameter>\$1</parameter>"], [$type_func, "<function>\$1</function>"], - [$type_env, "<envar>\$1</envar>"] + [$type_env, "<envar>\$1</envar>"], + [$type_member_xml, "<literal><structname>\$1</structname>\$2<structfield>\$3</structfield></literal>"], + [$type_fallback_xml, "<structname>\$1</structname>"] ); my $blankline_xml = $local_lt . "/para" . $local_gt . $local_lt . "para" . $local_gt . "\n"; @@ -264,9 +284,14 @@ my $blankline_xml = $local_lt . "/para" . $local_gt . $local_lt . "para" . $loca my @highlights_gnome = ( [$type_constant, "<replaceable class=\"option\">\$1</replaceable>"], [$type_func, "<function>\$1</function>"], + [$type_enum, "<type>\$1</type>"], [$type_struct, "<structname>\$1</structname>"], + [$type_typedef, "<type>\$1</type>"], + [$type_union, "<structname>\$1</structname>"], [$type_env, "<envar>\$1</envar>"], - [$type_param, "<parameter>\$1</parameter>" ] + [$type_param, "<parameter>\$1</parameter>" ], + [$type_member, "<literal><structname>\$1</structname>\$2<structfield>\$3</structfield></literal>"], + [$type_fallback, "<structname>\$1</structname>"] ); my $blankline_gnome = "</para><para>\n"; @@ -274,8 +299,13 @@ my $blankline_gnome = "</para><para>\n"; my @highlights_man = ( [$type_constant, "\$1"], [$type_func, "\\\\fB\$1\\\\fP"], + [$type_enum, "\\\\fI\$1\\\\fP"], [$type_struct, "\\\\fI\$1\\\\fP"], - [$type_param, "\\\\fI\$1\\\\fP"] + [$type_typedef, "\\\\fI\$1\\\\fP"], + [$type_union, "\\\\fI\$1\\\\fP"], + [$type_param, "\\\\fI\$1\\\\fP"], + [$type_member, "\\\\fI\$1\$2\$3\\\\fP"], + [$type_fallback, "\\\\fI\$1\\\\fP"] ); my $blankline_man = ""; @@ -283,8 +313,13 @@ my $blankline_man = ""; my @highlights_text = ( [$type_constant, "\$1"], [$type_func, "\$1"], + [$type_enum, "\$1"], [$type_struct, "\$1"], - [$type_param, "\$1"] + [$type_typedef, "\$1"], + [$type_union, "\$1"], + [$type_param, "\$1"], + [$type_member, "\$1\$2\$3"], + [$type_fallback, "\$1"] ); my $blankline_text = ""; @@ -292,16 +327,16 @@ my $blankline_text = ""; my @highlights_rst = ( [$type_constant, "``\$1``"], # Note: need to escape () to avoid func matching later - [$type_member_func, "\\:c\\:type\\:`\$1\$2\\\\(\\\\) <\$1>`"], - [$type_member, "\\:c\\:type\\:`\$1\$2 <\$1>`"], + [$type_member_func, "\\:c\\:type\\:`\$1\$2\$3\\\\(\\\\) <\$1>`"], + [$type_member, "\\:c\\:type\\:`\$1\$2\$3 <\$1>`"], [$type_fp_param, "**\$1\\\\(\\\\)**"], [$type_func, "\\:c\\:func\\:`\$1()`"], - [$type_struct_full, "\\:c\\:type\\:`\$1 \$2 <\$2>`"], - [$type_enum_full, "\\:c\\:type\\:`\$1 \$2 <\$2>`"], - [$type_typedef_full, "\\:c\\:type\\:`\$1 \$2 <\$2>`"], - [$type_union_full, "\\:c\\:type\\:`\$1 \$2 <\$2>`"], + [$type_enum, "\\:c\\:type\\:`\$1 <\$2>`"], + [$type_struct, "\\:c\\:type\\:`\$1 <\$2>`"], + [$type_typedef, "\\:c\\:type\\:`\$1 <\$2>`"], + [$type_union, "\\:c\\:type\\:`\$1 <\$2>`"], # in rst this can refer to any type - [$type_struct, "\\:c\\:type\\:`\$1`"], + [$type_fallback, "\\:c\\:type\\:`\$1`"], [$type_param, "**\$1**"] ); my $blankline_rst = "\n"; @@ -310,8 +345,13 @@ my $blankline_rst = "\n"; my @highlights_list = ( [$type_constant, "\$1"], [$type_func, "\$1"], + [$type_enum, "\$1"], [$type_struct, "\$1"], - [$type_param, "\$1"] + [$type_typedef, "\$1"], + [$type_union, "\$1"], + [$type_param, "\$1"], + [$type_member, "\$1"], + [$type_fallback, "\$1"] ); my $blankline_list = ""; @@ -1131,8 +1171,9 @@ sub output_function_xml(%) { foreach $parameter (@{$args{'parameterlist'}}) { my $parameter_name = $parameter; $parameter_name =~ s/\[.*//; + $type = $args{'parametertypes'}{$parameter}; - print " <varlistentry>\n <term><parameter>$parameter</parameter></term>\n"; + print " <varlistentry>\n <term><parameter>$type $parameter</parameter></term>\n"; print " <listitem>\n <para>\n"; $lineprefix=" "; output_highlight($args{'parameterdescs'}{$parameter_name}); @@ -1223,8 +1264,9 @@ sub output_struct_xml(%) { defined($args{'parameterdescs'}{$parameter_name}) || next; ($args{'parameterdescs'}{$parameter_name} ne $undescribed) || next; + $type = $args{'parametertypes'}{$parameter}; print " <varlistentry>"; - print " <term>$parameter</term>\n"; + print " <term><literal>$type $parameter</literal></term>\n"; print " <listitem><para>\n"; output_highlight($args{'parameterdescs'}{$parameter_name}); print " </para></listitem>\n"; @@ -1883,7 +1925,7 @@ sub output_function_rst(%) { $lineprefix = " "; foreach $parameter (@{$args{'parameterlist'}}) { my $parameter_name = $parameter; - #$parameter_name =~ s/\[.*//; + $parameter_name =~ s/\[.*//; $type = $args{'parametertypes'}{$parameter}; if ($type ne "") { @@ -2409,6 +2451,7 @@ sub push_parameter($$$) { # "[blah" in a parameter string; ###$param =~ s/\s*//g; push @parameterlist, $param; + $type =~ s/\s\s+/ /g; $parametertypes{$param} = $type; } @@ -2505,7 +2548,13 @@ sub dump_function($$) { $prototype =~ s/__must_check +//; $prototype =~ s/__weak +//; my $define = $prototype =~ s/^#\s*define\s+//; #ak added - $prototype =~ s/__attribute__\s*\(\([a-z,]*\)\)//; + $prototype =~ s/__attribute__\s*\(\( + (?: + [\w\s]++ # attribute name + (?:\([^)]*+\))? # attribute arguments + \s*+,? # optional comma at the end + )+ + \)\)\s+//x; # Yes, this truly is vile. We are looking for: # 1. Return type (may be nothing if we're looking at a macro) @@ -2533,21 +2582,21 @@ sub dump_function($$) { $noret = 1; } elsif ($prototype =~ m/^()([a-zA-Z0-9_~:]+)\s*\(([^\(]*)\)/ || $prototype =~ m/^(\w+)\s+([a-zA-Z0-9_~:]+)\s*\(([^\(]*)\)/ || - $prototype =~ m/^(\w+\s*\*)\s*([a-zA-Z0-9_~:]+)\s*\(([^\(]*)\)/ || + $prototype =~ m/^(\w+\s*\*+)\s*([a-zA-Z0-9_~:]+)\s*\(([^\(]*)\)/ || $prototype =~ m/^(\w+\s+\w+)\s+([a-zA-Z0-9_~:]+)\s*\(([^\(]*)\)/ || $prototype =~ m/^(\w+\s+\w+\s*\*+)\s*([a-zA-Z0-9_~:]+)\s*\(([^\(]*)\)/ || $prototype =~ m/^(\w+\s+\w+\s+\w+)\s+([a-zA-Z0-9_~:]+)\s*\(([^\(]*)\)/ || - $prototype =~ m/^(\w+\s+\w+\s+\w+\s*\*)\s*([a-zA-Z0-9_~:]+)\s*\(([^\(]*)\)/ || + $prototype =~ m/^(\w+\s+\w+\s+\w+\s*\*+)\s*([a-zA-Z0-9_~:]+)\s*\(([^\(]*)\)/ || $prototype =~ m/^()([a-zA-Z0-9_~:]+)\s*\(([^\{]*)\)/ || $prototype =~ m/^(\w+)\s+([a-zA-Z0-9_~:]+)\s*\(([^\{]*)\)/ || - $prototype =~ m/^(\w+\s*\*)\s*([a-zA-Z0-9_~:]+)\s*\(([^\{]*)\)/ || + $prototype =~ m/^(\w+\s*\*+)\s*([a-zA-Z0-9_~:]+)\s*\(([^\{]*)\)/ || $prototype =~ m/^(\w+\s+\w+)\s+([a-zA-Z0-9_~:]+)\s*\(([^\{]*)\)/ || - $prototype =~ m/^(\w+\s+\w+\s*\*)\s*([a-zA-Z0-9_~:]+)\s*\(([^\{]*)\)/ || + $prototype =~ m/^(\w+\s+\w+\s*\*+)\s*([a-zA-Z0-9_~:]+)\s*\(([^\{]*)\)/ || $prototype =~ m/^(\w+\s+\w+\s+\w+)\s+([a-zA-Z0-9_~:]+)\s*\(([^\{]*)\)/ || - $prototype =~ m/^(\w+\s+\w+\s+\w+\s*\*)\s*([a-zA-Z0-9_~:]+)\s*\(([^\{]*)\)/ || + $prototype =~ m/^(\w+\s+\w+\s+\w+\s*\*+)\s*([a-zA-Z0-9_~:]+)\s*\(([^\{]*)\)/ || $prototype =~ m/^(\w+\s+\w+\s+\w+\s+\w+)\s+([a-zA-Z0-9_~:]+)\s*\(([^\{]*)\)/ || - $prototype =~ m/^(\w+\s+\w+\s+\w+\s+\w+\s*\*)\s*([a-zA-Z0-9_~:]+)\s*\(([^\{]*)\)/ || - $prototype =~ m/^(\w+\s+\w+\s*\*\s*\w+\s*\*\s*)\s*([a-zA-Z0-9_~:]+)\s*\(([^\{]*)\)/) { + $prototype =~ m/^(\w+\s+\w+\s+\w+\s+\w+\s*\*+)\s*([a-zA-Z0-9_~:]+)\s*\(([^\{]*)\)/ || + $prototype =~ m/^(\w+\s+\w+\s*\*+\s*\w+\s*\*+\s*)\s*([a-zA-Z0-9_~:]+)\s*\(([^\{]*)\)/) { $return_type = $1; $declaration_name = $2; my $args = $3; diff --git a/virt/kvm/arm/arch_timer.c b/virt/kvm/arm/arch_timer.c index 6a084cd57b88..35d7100e0815 100644 --- a/virt/kvm/arm/arch_timer.c +++ b/virt/kvm/arm/arch_timer.c @@ -37,10 +37,10 @@ static u32 host_vtimer_irq_flags; void kvm_timer_vcpu_put(struct kvm_vcpu *vcpu) { - vcpu->arch.timer_cpu.active_cleared_last = false; + vcpu_vtimer(vcpu)->active_cleared_last = false; } -static u64 kvm_phys_timer_read(void) +u64 kvm_phys_timer_read(void) { return timecounter->cc->read(timecounter->cc); } @@ -98,12 +98,12 @@ static void kvm_timer_inject_irq_work(struct work_struct *work) kvm_vcpu_kick(vcpu); } -static u64 kvm_timer_compute_delta(struct kvm_vcpu *vcpu) +static u64 kvm_timer_compute_delta(struct arch_timer_context *timer_ctx) { u64 cval, now; - cval = vcpu->arch.timer_cpu.cntv_cval; - now = kvm_phys_timer_read() - vcpu->kvm->arch.timer.cntvoff; + cval = timer_ctx->cnt_cval; + now = kvm_phys_timer_read() - timer_ctx->cntvoff; if (now < cval) { u64 ns; @@ -118,6 +118,35 @@ static u64 kvm_timer_compute_delta(struct kvm_vcpu *vcpu) return 0; } +static bool kvm_timer_irq_can_fire(struct arch_timer_context *timer_ctx) +{ + return !(timer_ctx->cnt_ctl & ARCH_TIMER_CTRL_IT_MASK) && + (timer_ctx->cnt_ctl & ARCH_TIMER_CTRL_ENABLE); +} + +/* + * Returns the earliest expiration time in ns among guest timers. + * Note that it will return 0 if none of timers can fire. + */ +static u64 kvm_timer_earliest_exp(struct kvm_vcpu *vcpu) +{ + u64 min_virt = ULLONG_MAX, min_phys = ULLONG_MAX; + struct arch_timer_context *vtimer = vcpu_vtimer(vcpu); + struct arch_timer_context *ptimer = vcpu_ptimer(vcpu); + + if (kvm_timer_irq_can_fire(vtimer)) + min_virt = kvm_timer_compute_delta(vtimer); + + if (kvm_timer_irq_can_fire(ptimer)) + min_phys = kvm_timer_compute_delta(ptimer); + + /* If none of timers can fire, then return 0 */ + if ((min_virt == ULLONG_MAX) && (min_phys == ULLONG_MAX)) + return 0; + + return min(min_virt, min_phys); +} + static enum hrtimer_restart kvm_timer_expire(struct hrtimer *hrt) { struct arch_timer_cpu *timer; @@ -132,7 +161,7 @@ static enum hrtimer_restart kvm_timer_expire(struct hrtimer *hrt) * PoV (NTP on the host may have forced it to expire * early). If we should have slept longer, restart it. */ - ns = kvm_timer_compute_delta(vcpu); + ns = kvm_timer_earliest_exp(vcpu); if (unlikely(ns)) { hrtimer_forward_now(hrt, ns_to_ktime(ns)); return HRTIMER_RESTART; @@ -142,42 +171,33 @@ static enum hrtimer_restart kvm_timer_expire(struct hrtimer *hrt) return HRTIMER_NORESTART; } -static bool kvm_timer_irq_can_fire(struct kvm_vcpu *vcpu) -{ - struct arch_timer_cpu *timer = &vcpu->arch.timer_cpu; - - return !(timer->cntv_ctl & ARCH_TIMER_CTRL_IT_MASK) && - (timer->cntv_ctl & ARCH_TIMER_CTRL_ENABLE); -} - -bool kvm_timer_should_fire(struct kvm_vcpu *vcpu) +bool kvm_timer_should_fire(struct arch_timer_context *timer_ctx) { - struct arch_timer_cpu *timer = &vcpu->arch.timer_cpu; u64 cval, now; - if (!kvm_timer_irq_can_fire(vcpu)) + if (!kvm_timer_irq_can_fire(timer_ctx)) return false; - cval = timer->cntv_cval; - now = kvm_phys_timer_read() - vcpu->kvm->arch.timer.cntvoff; + cval = timer_ctx->cnt_cval; + now = kvm_phys_timer_read() - timer_ctx->cntvoff; return cval <= now; } -static void kvm_timer_update_irq(struct kvm_vcpu *vcpu, bool new_level) +static void kvm_timer_update_irq(struct kvm_vcpu *vcpu, bool new_level, + struct arch_timer_context *timer_ctx) { int ret; - struct arch_timer_cpu *timer = &vcpu->arch.timer_cpu; BUG_ON(!vgic_initialized(vcpu->kvm)); - timer->active_cleared_last = false; - timer->irq.level = new_level; - trace_kvm_timer_update_irq(vcpu->vcpu_id, timer->irq.irq, - timer->irq.level); - ret = kvm_vgic_inject_mapped_irq(vcpu->kvm, vcpu->vcpu_id, - timer->irq.irq, - timer->irq.level); + timer_ctx->active_cleared_last = false; + timer_ctx->irq.level = new_level; + trace_kvm_timer_update_irq(vcpu->vcpu_id, timer_ctx->irq.irq, + timer_ctx->irq.level); + + ret = kvm_vgic_inject_irq(vcpu->kvm, vcpu->vcpu_id, timer_ctx->irq.irq, + timer_ctx->irq.level); WARN_ON(ret); } @@ -188,22 +208,43 @@ static void kvm_timer_update_irq(struct kvm_vcpu *vcpu, bool new_level) static int kvm_timer_update_state(struct kvm_vcpu *vcpu) { struct arch_timer_cpu *timer = &vcpu->arch.timer_cpu; + struct arch_timer_context *vtimer = vcpu_vtimer(vcpu); + struct arch_timer_context *ptimer = vcpu_ptimer(vcpu); /* * If userspace modified the timer registers via SET_ONE_REG before - * the vgic was initialized, we mustn't set the timer->irq.level value + * the vgic was initialized, we mustn't set the vtimer->irq.level value * because the guest would never see the interrupt. Instead wait * until we call this function from kvm_timer_flush_hwstate. */ if (!vgic_initialized(vcpu->kvm) || !timer->enabled) return -ENODEV; - if (kvm_timer_should_fire(vcpu) != timer->irq.level) - kvm_timer_update_irq(vcpu, !timer->irq.level); + if (kvm_timer_should_fire(vtimer) != vtimer->irq.level) + kvm_timer_update_irq(vcpu, !vtimer->irq.level, vtimer); + + if (kvm_timer_should_fire(ptimer) != ptimer->irq.level) + kvm_timer_update_irq(vcpu, !ptimer->irq.level, ptimer); return 0; } +/* Schedule the background timer for the emulated timer. */ +static void kvm_timer_emulate(struct kvm_vcpu *vcpu, + struct arch_timer_context *timer_ctx) +{ + struct arch_timer_cpu *timer = &vcpu->arch.timer_cpu; + + if (kvm_timer_should_fire(timer_ctx)) + return; + + if (!kvm_timer_irq_can_fire(timer_ctx)) + return; + + /* The timer has not yet expired, schedule a background timer */ + timer_arm(timer, kvm_timer_compute_delta(timer_ctx)); +} + /* * Schedule the background timer before calling kvm_vcpu_block, so that this * thread is removed from its waitqueue and made runnable when there's a timer @@ -212,26 +253,31 @@ static int kvm_timer_update_state(struct kvm_vcpu *vcpu) void kvm_timer_schedule(struct kvm_vcpu *vcpu) { struct arch_timer_cpu *timer = &vcpu->arch.timer_cpu; + struct arch_timer_context *vtimer = vcpu_vtimer(vcpu); + struct arch_timer_context *ptimer = vcpu_ptimer(vcpu); BUG_ON(timer_is_armed(timer)); /* - * No need to schedule a background timer if the guest timer has + * No need to schedule a background timer if any guest timer has * already expired, because kvm_vcpu_block will return before putting * the thread to sleep. */ - if (kvm_timer_should_fire(vcpu)) + if (kvm_timer_should_fire(vtimer) || kvm_timer_should_fire(ptimer)) return; /* - * If the timer is not capable of raising interrupts (disabled or + * If both timers are not capable of raising interrupts (disabled or * masked), then there's no more work for us to do. */ - if (!kvm_timer_irq_can_fire(vcpu)) + if (!kvm_timer_irq_can_fire(vtimer) && !kvm_timer_irq_can_fire(ptimer)) return; - /* The timer has not yet expired, schedule a background timer */ - timer_arm(timer, kvm_timer_compute_delta(vcpu)); + /* + * The guest timers have not yet expired, schedule a background timer. + * Set the earliest expiration time among the guest timers. + */ + timer_arm(timer, kvm_timer_earliest_exp(vcpu)); } void kvm_timer_unschedule(struct kvm_vcpu *vcpu) @@ -249,13 +295,16 @@ void kvm_timer_unschedule(struct kvm_vcpu *vcpu) */ void kvm_timer_flush_hwstate(struct kvm_vcpu *vcpu) { - struct arch_timer_cpu *timer = &vcpu->arch.timer_cpu; + struct arch_timer_context *vtimer = vcpu_vtimer(vcpu); bool phys_active; int ret; if (kvm_timer_update_state(vcpu)) return; + /* Set the background timer for the physical timer emulation. */ + kvm_timer_emulate(vcpu, vcpu_ptimer(vcpu)); + /* * If we enter the guest with the virtual input level to the VGIC * asserted, then we have already told the VGIC what we need to, and @@ -273,8 +322,8 @@ void kvm_timer_flush_hwstate(struct kvm_vcpu *vcpu) * to ensure that hardware interrupts from the timer triggers a guest * exit. */ - phys_active = timer->irq.level || - kvm_vgic_map_is_active(vcpu, timer->irq.irq); + phys_active = vtimer->irq.level || + kvm_vgic_map_is_active(vcpu, vtimer->irq.irq); /* * We want to avoid hitting the (re)distributor as much as @@ -296,7 +345,7 @@ void kvm_timer_flush_hwstate(struct kvm_vcpu *vcpu) * - cached value is "active clear" * - value to be programmed is "active clear" */ - if (timer->active_cleared_last && !phys_active) + if (vtimer->active_cleared_last && !phys_active) return; ret = irq_set_irqchip_state(host_vtimer_irq, @@ -304,7 +353,7 @@ void kvm_timer_flush_hwstate(struct kvm_vcpu *vcpu) phys_active); WARN_ON(ret); - timer->active_cleared_last = !phys_active; + vtimer->active_cleared_last = !phys_active; } /** @@ -318,7 +367,11 @@ void kvm_timer_sync_hwstate(struct kvm_vcpu *vcpu) { struct arch_timer_cpu *timer = &vcpu->arch.timer_cpu; - BUG_ON(timer_is_armed(timer)); + /* + * This is to cancel the background timer for the physical timer + * emulation if it is set. + */ + timer_disarm(timer); /* * The guest could have modified the timer registers or the timer @@ -328,9 +381,11 @@ void kvm_timer_sync_hwstate(struct kvm_vcpu *vcpu) } int kvm_timer_vcpu_reset(struct kvm_vcpu *vcpu, - const struct kvm_irq_level *irq) + const struct kvm_irq_level *virt_irq, + const struct kvm_irq_level *phys_irq) { - struct arch_timer_cpu *timer = &vcpu->arch.timer_cpu; + struct arch_timer_context *vtimer = vcpu_vtimer(vcpu); + struct arch_timer_context *ptimer = vcpu_ptimer(vcpu); /* * The vcpu timer irq number cannot be determined in @@ -338,7 +393,8 @@ int kvm_timer_vcpu_reset(struct kvm_vcpu *vcpu, * kvm_vcpu_set_target(). To handle this, we determine * vcpu timer irq number when the vcpu is reset. */ - timer->irq.irq = irq->irq; + vtimer->irq.irq = virt_irq->irq; + ptimer->irq.irq = phys_irq->irq; /* * The bits in CNTV_CTL are architecturally reset to UNKNOWN for ARMv8 @@ -346,16 +402,40 @@ int kvm_timer_vcpu_reset(struct kvm_vcpu *vcpu, * resets the timer to be disabled and unmasked and is compliant with * the ARMv7 architecture. */ - timer->cntv_ctl = 0; + vtimer->cnt_ctl = 0; + ptimer->cnt_ctl = 0; kvm_timer_update_state(vcpu); return 0; } +/* Make the updates of cntvoff for all vtimer contexts atomic */ +static void update_vtimer_cntvoff(struct kvm_vcpu *vcpu, u64 cntvoff) +{ + int i; + struct kvm *kvm = vcpu->kvm; + struct kvm_vcpu *tmp; + + mutex_lock(&kvm->lock); + kvm_for_each_vcpu(i, tmp, kvm) + vcpu_vtimer(tmp)->cntvoff = cntvoff; + + /* + * When called from the vcpu create path, the CPU being created is not + * included in the loop above, so we just set it here as well. + */ + vcpu_vtimer(vcpu)->cntvoff = cntvoff; + mutex_unlock(&kvm->lock); +} + void kvm_timer_vcpu_init(struct kvm_vcpu *vcpu) { struct arch_timer_cpu *timer = &vcpu->arch.timer_cpu; + /* Synchronize cntvoff across all vtimers of a VM. */ + update_vtimer_cntvoff(vcpu, kvm_phys_timer_read()); + vcpu_ptimer(vcpu)->cntvoff = 0; + INIT_WORK(&timer->expired, kvm_timer_inject_irq_work); hrtimer_init(&timer->timer, CLOCK_MONOTONIC, HRTIMER_MODE_ABS); timer->timer.function = kvm_timer_expire; @@ -368,17 +448,17 @@ static void kvm_timer_init_interrupt(void *info) int kvm_arm_timer_set_reg(struct kvm_vcpu *vcpu, u64 regid, u64 value) { - struct arch_timer_cpu *timer = &vcpu->arch.timer_cpu; + struct arch_timer_context *vtimer = vcpu_vtimer(vcpu); switch (regid) { case KVM_REG_ARM_TIMER_CTL: - timer->cntv_ctl = value; + vtimer->cnt_ctl = value; break; case KVM_REG_ARM_TIMER_CNT: - vcpu->kvm->arch.timer.cntvoff = kvm_phys_timer_read() - value; + update_vtimer_cntvoff(vcpu, kvm_phys_timer_read() - value); break; case KVM_REG_ARM_TIMER_CVAL: - timer->cntv_cval = value; + vtimer->cnt_cval = value; break; default: return -1; @@ -390,15 +470,15 @@ int kvm_arm_timer_set_reg(struct kvm_vcpu *vcpu, u64 regid, u64 value) u64 kvm_arm_timer_get_reg(struct kvm_vcpu *vcpu, u64 regid) { - struct arch_timer_cpu *timer = &vcpu->arch.timer_cpu; + struct arch_timer_context *vtimer = vcpu_vtimer(vcpu); switch (regid) { case KVM_REG_ARM_TIMER_CTL: - return timer->cntv_ctl; + return vtimer->cnt_ctl; case KVM_REG_ARM_TIMER_CNT: - return kvm_phys_timer_read() - vcpu->kvm->arch.timer.cntvoff; + return kvm_phys_timer_read() - vtimer->cntvoff; case KVM_REG_ARM_TIMER_CVAL: - return timer->cntv_cval; + return vtimer->cnt_cval; } return (u64)-1; } @@ -462,14 +542,16 @@ int kvm_timer_hyp_init(void) void kvm_timer_vcpu_terminate(struct kvm_vcpu *vcpu) { struct arch_timer_cpu *timer = &vcpu->arch.timer_cpu; + struct arch_timer_context *vtimer = vcpu_vtimer(vcpu); timer_disarm(timer); - kvm_vgic_unmap_phys_irq(vcpu, timer->irq.irq); + kvm_vgic_unmap_phys_irq(vcpu, vtimer->irq.irq); } int kvm_timer_enable(struct kvm_vcpu *vcpu) { struct arch_timer_cpu *timer = &vcpu->arch.timer_cpu; + struct arch_timer_context *vtimer = vcpu_vtimer(vcpu); struct irq_desc *desc; struct irq_data *data; int phys_irq; @@ -497,7 +579,7 @@ int kvm_timer_enable(struct kvm_vcpu *vcpu) * Tell the VGIC that the virtual interrupt is tied to a * physical interrupt. We do that once per VCPU. */ - ret = kvm_vgic_map_phys_irq(vcpu, timer->irq.irq, phys_irq); + ret = kvm_vgic_map_phys_irq(vcpu, vtimer->irq.irq, phys_irq); if (ret) return ret; @@ -506,11 +588,6 @@ int kvm_timer_enable(struct kvm_vcpu *vcpu) return 0; } -void kvm_timer_init(struct kvm *kvm) -{ - kvm->arch.timer.cntvoff = kvm_phys_timer_read(); -} - /* * On VHE system, we only need to configure trap on physical timer and counter * accesses in EL0 and EL1 once, not for every world switch. diff --git a/virt/kvm/arm/hyp/timer-sr.c b/virt/kvm/arm/hyp/timer-sr.c index 63e28dd18bb0..4734915ab71f 100644 --- a/virt/kvm/arm/hyp/timer-sr.c +++ b/virt/kvm/arm/hyp/timer-sr.c @@ -25,11 +25,12 @@ void __hyp_text __timer_save_state(struct kvm_vcpu *vcpu) { struct arch_timer_cpu *timer = &vcpu->arch.timer_cpu; + struct arch_timer_context *vtimer = vcpu_vtimer(vcpu); u64 val; if (timer->enabled) { - timer->cntv_ctl = read_sysreg_el0(cntv_ctl); - timer->cntv_cval = read_sysreg_el0(cntv_cval); + vtimer->cnt_ctl = read_sysreg_el0(cntv_ctl); + vtimer->cnt_cval = read_sysreg_el0(cntv_cval); } /* Disable the virtual timer */ @@ -52,8 +53,8 @@ void __hyp_text __timer_save_state(struct kvm_vcpu *vcpu) void __hyp_text __timer_restore_state(struct kvm_vcpu *vcpu) { - struct kvm *kvm = kern_hyp_va(vcpu->kvm); struct arch_timer_cpu *timer = &vcpu->arch.timer_cpu; + struct arch_timer_context *vtimer = vcpu_vtimer(vcpu); u64 val; /* Those bits are already configured at boot on VHE-system */ @@ -69,9 +70,9 @@ void __hyp_text __timer_restore_state(struct kvm_vcpu *vcpu) } if (timer->enabled) { - write_sysreg(kvm->arch.timer.cntvoff, cntvoff_el2); - write_sysreg_el0(timer->cntv_cval, cntv_cval); + write_sysreg(vtimer->cntvoff, cntvoff_el2); + write_sysreg_el0(vtimer->cnt_cval, cntv_cval); isb(); - write_sysreg_el0(timer->cntv_ctl, cntv_ctl); + write_sysreg_el0(vtimer->cnt_ctl, cntv_ctl); } } diff --git a/virt/kvm/arm/vgic/vgic-debug.c b/virt/kvm/arm/vgic/vgic-debug.c new file mode 100644 index 000000000000..7072ab743332 --- /dev/null +++ b/virt/kvm/arm/vgic/vgic-debug.c @@ -0,0 +1,283 @@ +/* + * Copyright (C) 2016 Linaro + * Author: Christoffer Dall <christoffer.dall@linaro.org> + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program. If not, see <http://www.gnu.org/licenses/>. + */ + +#include <linux/cpu.h> +#include <linux/debugfs.h> +#include <linux/interrupt.h> +#include <linux/kvm_host.h> +#include <linux/seq_file.h> +#include <kvm/arm_vgic.h> +#include <asm/kvm_mmu.h> +#include "vgic.h" + +/* + * Structure to control looping through the entire vgic state. We start at + * zero for each field and move upwards. So, if dist_id is 0 we print the + * distributor info. When dist_id is 1, we have already printed it and move + * on. + * + * When vcpu_id < nr_cpus we print the vcpu info until vcpu_id == nr_cpus and + * so on. + */ +struct vgic_state_iter { + int nr_cpus; + int nr_spis; + int dist_id; + int vcpu_id; + int intid; +}; + +static void iter_next(struct vgic_state_iter *iter) +{ + if (iter->dist_id == 0) { + iter->dist_id++; + return; + } + + iter->intid++; + if (iter->intid == VGIC_NR_PRIVATE_IRQS && + ++iter->vcpu_id < iter->nr_cpus) + iter->intid = 0; +} + +static void iter_init(struct kvm *kvm, struct vgic_state_iter *iter, + loff_t pos) +{ + int nr_cpus = atomic_read(&kvm->online_vcpus); + + memset(iter, 0, sizeof(*iter)); + + iter->nr_cpus = nr_cpus; + iter->nr_spis = kvm->arch.vgic.nr_spis; + + /* Fast forward to the right position if needed */ + while (pos--) + iter_next(iter); +} + +static bool end_of_vgic(struct vgic_state_iter *iter) +{ + return iter->dist_id > 0 && + iter->vcpu_id == iter->nr_cpus && + (iter->intid - VGIC_NR_PRIVATE_IRQS) == iter->nr_spis; +} + +static void *vgic_debug_start(struct seq_file *s, loff_t *pos) +{ + struct kvm *kvm = (struct kvm *)s->private; + struct vgic_state_iter *iter; + + mutex_lock(&kvm->lock); + iter = kvm->arch.vgic.iter; + if (iter) { + iter = ERR_PTR(-EBUSY); + goto out; + } + + iter = kmalloc(sizeof(*iter), GFP_KERNEL); + if (!iter) { + iter = ERR_PTR(-ENOMEM); + goto out; + } + + iter_init(kvm, iter, *pos); + kvm->arch.vgic.iter = iter; + + if (end_of_vgic(iter)) + iter = NULL; +out: + mutex_unlock(&kvm->lock); + return iter; +} + +static void *vgic_debug_next(struct seq_file *s, void *v, loff_t *pos) +{ + struct kvm *kvm = (struct kvm *)s->private; + struct vgic_state_iter *iter = kvm->arch.vgic.iter; + + ++*pos; + iter_next(iter); + if (end_of_vgic(iter)) + iter = NULL; + return iter; +} + +static void vgic_debug_stop(struct seq_file *s, void *v) +{ + struct kvm *kvm = (struct kvm *)s->private; + struct vgic_state_iter *iter; + + /* + * If the seq file wasn't properly opened, there's nothing to clearn + * up. + */ + if (IS_ERR(v)) + return; + + mutex_lock(&kvm->lock); + iter = kvm->arch.vgic.iter; + kfree(iter); + kvm->arch.vgic.iter = NULL; + mutex_unlock(&kvm->lock); +} + +static void print_dist_state(struct seq_file *s, struct vgic_dist *dist) +{ + seq_printf(s, "Distributor\n"); + seq_printf(s, "===========\n"); + seq_printf(s, "vgic_model:\t%s\n", + (dist->vgic_model == KVM_DEV_TYPE_ARM_VGIC_V3) ? + "GICv3" : "GICv2"); + seq_printf(s, "nr_spis:\t%d\n", dist->nr_spis); + seq_printf(s, "enabled:\t%d\n", dist->enabled); + seq_printf(s, "\n"); + + seq_printf(s, "P=pending_latch, L=line_level, A=active\n"); + seq_printf(s, "E=enabled, H=hw, C=config (level=1, edge=0)\n"); +} + +static void print_header(struct seq_file *s, struct vgic_irq *irq, + struct kvm_vcpu *vcpu) +{ + int id = 0; + char *hdr = "SPI "; + + if (vcpu) { + hdr = "VCPU"; + id = vcpu->vcpu_id; + } + + seq_printf(s, "\n"); + seq_printf(s, "%s%2d TYP ID TGT_ID PLAEHC HWID TARGET SRC PRI VCPU_ID\n", hdr, id); + seq_printf(s, "---------------------------------------------------------------\n"); +} + +static void print_irq_state(struct seq_file *s, struct vgic_irq *irq, + struct kvm_vcpu *vcpu) +{ + char *type; + if (irq->intid < VGIC_NR_SGIS) + type = "SGI"; + else if (irq->intid < VGIC_NR_PRIVATE_IRQS) + type = "PPI"; + else + type = "SPI"; + + if (irq->intid ==0 || irq->intid == VGIC_NR_PRIVATE_IRQS) + print_header(s, irq, vcpu); + + seq_printf(s, " %s %4d " + " %2d " + "%d%d%d%d%d%d " + "%8d " + "%8x " + " %2x " + "%3d " + " %2d " + "\n", + type, irq->intid, + (irq->target_vcpu) ? irq->target_vcpu->vcpu_id : -1, + irq->pending_latch, + irq->line_level, + irq->active, + irq->enabled, + irq->hw, + irq->config == VGIC_CONFIG_LEVEL, + irq->hwintid, + irq->mpidr, + irq->source, + irq->priority, + (irq->vcpu) ? irq->vcpu->vcpu_id : -1); + +} + +static int vgic_debug_show(struct seq_file *s, void *v) +{ + struct kvm *kvm = (struct kvm *)s->private; + struct vgic_state_iter *iter = (struct vgic_state_iter *)v; + struct vgic_irq *irq; + struct kvm_vcpu *vcpu = NULL; + + if (iter->dist_id == 0) { + print_dist_state(s, &kvm->arch.vgic); + return 0; + } + + if (!kvm->arch.vgic.initialized) + return 0; + + if (iter->vcpu_id < iter->nr_cpus) { + vcpu = kvm_get_vcpu(kvm, iter->vcpu_id); + irq = &vcpu->arch.vgic_cpu.private_irqs[iter->intid]; + } else { + irq = &kvm->arch.vgic.spis[iter->intid - VGIC_NR_PRIVATE_IRQS]; + } + + spin_lock(&irq->irq_lock); + print_irq_state(s, irq, vcpu); + spin_unlock(&irq->irq_lock); + + return 0; +} + +static struct seq_operations vgic_debug_seq_ops = { + .start = vgic_debug_start, + .next = vgic_debug_next, + .stop = vgic_debug_stop, + .show = vgic_debug_show +}; + +static int debug_open(struct inode *inode, struct file *file) +{ + int ret; + ret = seq_open(file, &vgic_debug_seq_ops); + if (!ret) { + struct seq_file *seq; + /* seq_open will have modified file->private_data */ + seq = file->private_data; + seq->private = inode->i_private; + } + + return ret; +}; + +static struct file_operations vgic_debug_fops = { + .owner = THIS_MODULE, + .open = debug_open, + .read = seq_read, + .llseek = seq_lseek, + .release = seq_release +}; + +int vgic_debug_init(struct kvm *kvm) +{ + if (!kvm->debugfs_dentry) + return -ENOENT; + + if (!debugfs_create_file("vgic-state", 0444, + kvm->debugfs_dentry, + kvm, + &vgic_debug_fops)) + return -ENOMEM; + + return 0; +} + +int vgic_debug_destroy(struct kvm *kvm) +{ + return 0; +} diff --git a/virt/kvm/arm/vgic/vgic-init.c b/virt/kvm/arm/vgic/vgic-init.c index c737ea0a310a..276139a24e6f 100644 --- a/virt/kvm/arm/vgic/vgic-init.c +++ b/virt/kvm/arm/vgic/vgic-init.c @@ -259,6 +259,8 @@ int vgic_init(struct kvm *kvm) if (ret) goto out; + vgic_debug_init(kvm); + dist->initialized = true; out: return ret; @@ -288,6 +290,8 @@ static void __kvm_vgic_destroy(struct kvm *kvm) struct kvm_vcpu *vcpu; int i; + vgic_debug_destroy(kvm); + kvm_vgic_dist_destroy(kvm); kvm_for_each_vcpu(i, vcpu, kvm) diff --git a/virt/kvm/arm/vgic/vgic-irqfd.c b/virt/kvm/arm/vgic/vgic-irqfd.c index d918dcf26a5a..f138ed2e9c63 100644 --- a/virt/kvm/arm/vgic/vgic-irqfd.c +++ b/virt/kvm/arm/vgic/vgic-irqfd.c @@ -99,6 +99,9 @@ int kvm_set_msi(struct kvm_kernel_irq_routing_entry *e, if (!vgic_has_its(kvm)) return -ENODEV; + if (!level) + return -1; + return vgic_its_inject_msi(kvm, &msi); } diff --git a/virt/kvm/arm/vgic/vgic-its.c b/virt/kvm/arm/vgic/vgic-its.c index 8c2b3cdcb2c5..571b64a01c50 100644 --- a/virt/kvm/arm/vgic/vgic-its.c +++ b/virt/kvm/arm/vgic/vgic-its.c @@ -350,7 +350,7 @@ static int its_sync_lpi_pending_table(struct kvm_vcpu *vcpu) irq = vgic_get_irq(vcpu->kvm, NULL, intids[i]); spin_lock(&irq->irq_lock); - irq->pending = pendmask & (1U << bit_nr); + irq->pending_latch = pendmask & (1U << bit_nr); vgic_queue_irq_unlock(vcpu->kvm, irq); vgic_put_irq(vcpu->kvm, irq); } @@ -465,7 +465,7 @@ static int vgic_its_trigger_msi(struct kvm *kvm, struct vgic_its *its, return -EBUSY; spin_lock(&itte->irq->irq_lock); - itte->irq->pending = true; + itte->irq->pending_latch = true; vgic_queue_irq_unlock(kvm, itte->irq); return 0; @@ -913,7 +913,7 @@ static int vgic_its_cmd_handle_clear(struct kvm *kvm, struct vgic_its *its, if (!itte) return E_ITS_CLEAR_UNMAPPED_INTERRUPT; - itte->irq->pending = false; + itte->irq->pending_latch = false; return 0; } diff --git a/virt/kvm/arm/vgic/vgic-kvm-device.c b/virt/kvm/arm/vgic/vgic-kvm-device.c index fbe87a63d250..d181d2baee9c 100644 --- a/virt/kvm/arm/vgic/vgic-kvm-device.c +++ b/virt/kvm/arm/vgic/vgic-kvm-device.c @@ -17,6 +17,7 @@ #include <kvm/arm_vgic.h> #include <linux/uaccess.h> #include <asm/kvm_mmu.h> +#include <asm/cputype.h> #include "vgic.h" /* common helpers */ @@ -230,14 +231,8 @@ int kvm_register_vgic_device(unsigned long type) return ret; } -struct vgic_reg_attr { - struct kvm_vcpu *vcpu; - gpa_t addr; -}; - -static int parse_vgic_v2_attr(struct kvm_device *dev, - struct kvm_device_attr *attr, - struct vgic_reg_attr *reg_attr) +int vgic_v2_parse_attr(struct kvm_device *dev, struct kvm_device_attr *attr, + struct vgic_reg_attr *reg_attr) { int cpuid; @@ -292,14 +287,14 @@ static bool lock_all_vcpus(struct kvm *kvm) } /** - * vgic_attr_regs_access_v2 - allows user space to access VGIC v2 state + * vgic_v2_attr_regs_access - allows user space to access VGIC v2 state * * @dev: kvm device handle * @attr: kvm device attribute * @reg: address the value is read or written * @is_write: true if userspace is writing a register */ -static int vgic_attr_regs_access_v2(struct kvm_device *dev, +static int vgic_v2_attr_regs_access(struct kvm_device *dev, struct kvm_device_attr *attr, u32 *reg, bool is_write) { @@ -308,7 +303,7 @@ static int vgic_attr_regs_access_v2(struct kvm_device *dev, struct kvm_vcpu *vcpu; int ret; - ret = parse_vgic_v2_attr(dev, attr, ®_attr); + ret = vgic_v2_parse_attr(dev, attr, ®_attr); if (ret) return ret; @@ -362,7 +357,7 @@ static int vgic_v2_set_attr(struct kvm_device *dev, if (get_user(reg, uaddr)) return -EFAULT; - return vgic_attr_regs_access_v2(dev, attr, ®, true); + return vgic_v2_attr_regs_access(dev, attr, ®, true); } } @@ -384,7 +379,7 @@ static int vgic_v2_get_attr(struct kvm_device *dev, u32 __user *uaddr = (u32 __user *)(long)attr->addr; u32 reg = 0; - ret = vgic_attr_regs_access_v2(dev, attr, ®, false); + ret = vgic_v2_attr_regs_access(dev, attr, ®, false); if (ret) return ret; return put_user(reg, uaddr); @@ -428,16 +423,211 @@ struct kvm_device_ops kvm_arm_vgic_v2_ops = { .has_attr = vgic_v2_has_attr, }; +int vgic_v3_parse_attr(struct kvm_device *dev, struct kvm_device_attr *attr, + struct vgic_reg_attr *reg_attr) +{ + unsigned long vgic_mpidr, mpidr_reg; + + /* + * For KVM_DEV_ARM_VGIC_GRP_DIST_REGS group, + * attr might not hold MPIDR. Hence assume vcpu0. + */ + if (attr->group != KVM_DEV_ARM_VGIC_GRP_DIST_REGS) { + vgic_mpidr = (attr->attr & KVM_DEV_ARM_VGIC_V3_MPIDR_MASK) >> + KVM_DEV_ARM_VGIC_V3_MPIDR_SHIFT; + + mpidr_reg = VGIC_TO_MPIDR(vgic_mpidr); + reg_attr->vcpu = kvm_mpidr_to_vcpu(dev->kvm, mpidr_reg); + } else { + reg_attr->vcpu = kvm_get_vcpu(dev->kvm, 0); + } + + if (!reg_attr->vcpu) + return -EINVAL; + + reg_attr->addr = attr->attr & KVM_DEV_ARM_VGIC_OFFSET_MASK; + + return 0; +} + +/* + * vgic_v3_attr_regs_access - allows user space to access VGIC v3 state + * + * @dev: kvm device handle + * @attr: kvm device attribute + * @reg: address the value is read or written + * @is_write: true if userspace is writing a register + */ +static int vgic_v3_attr_regs_access(struct kvm_device *dev, + struct kvm_device_attr *attr, + u64 *reg, bool is_write) +{ + struct vgic_reg_attr reg_attr; + gpa_t addr; + struct kvm_vcpu *vcpu; + int ret; + u32 tmp32; + + ret = vgic_v3_parse_attr(dev, attr, ®_attr); + if (ret) + return ret; + + vcpu = reg_attr.vcpu; + addr = reg_attr.addr; + + mutex_lock(&dev->kvm->lock); + + if (unlikely(!vgic_initialized(dev->kvm))) { + ret = -EBUSY; + goto out; + } + + if (!lock_all_vcpus(dev->kvm)) { + ret = -EBUSY; + goto out; + } + + switch (attr->group) { + case KVM_DEV_ARM_VGIC_GRP_DIST_REGS: + if (is_write) + tmp32 = *reg; + + ret = vgic_v3_dist_uaccess(vcpu, is_write, addr, &tmp32); + if (!is_write) + *reg = tmp32; + break; + case KVM_DEV_ARM_VGIC_GRP_REDIST_REGS: + if (is_write) + tmp32 = *reg; + + ret = vgic_v3_redist_uaccess(vcpu, is_write, addr, &tmp32); + if (!is_write) + *reg = tmp32; + break; + case KVM_DEV_ARM_VGIC_GRP_CPU_SYSREGS: { + u64 regid; + + regid = (attr->attr & KVM_DEV_ARM_VGIC_SYSREG_INSTR_MASK); + ret = vgic_v3_cpu_sysregs_uaccess(vcpu, is_write, + regid, reg); + break; + } + case KVM_DEV_ARM_VGIC_GRP_LEVEL_INFO: { + unsigned int info, intid; + + info = (attr->attr & KVM_DEV_ARM_VGIC_LINE_LEVEL_INFO_MASK) >> + KVM_DEV_ARM_VGIC_LINE_LEVEL_INFO_SHIFT; + if (info == VGIC_LEVEL_INFO_LINE_LEVEL) { + intid = attr->attr & + KVM_DEV_ARM_VGIC_LINE_LEVEL_INTID_MASK; + ret = vgic_v3_line_level_info_uaccess(vcpu, is_write, + intid, reg); + } else { + ret = -EINVAL; + } + break; + } + default: + ret = -EINVAL; + break; + } + + unlock_all_vcpus(dev->kvm); +out: + mutex_unlock(&dev->kvm->lock); + return ret; +} + static int vgic_v3_set_attr(struct kvm_device *dev, struct kvm_device_attr *attr) { - return vgic_set_common_attr(dev, attr); + int ret; + + ret = vgic_set_common_attr(dev, attr); + if (ret != -ENXIO) + return ret; + + switch (attr->group) { + case KVM_DEV_ARM_VGIC_GRP_DIST_REGS: + case KVM_DEV_ARM_VGIC_GRP_REDIST_REGS: { + u32 __user *uaddr = (u32 __user *)(long)attr->addr; + u32 tmp32; + u64 reg; + + if (get_user(tmp32, uaddr)) + return -EFAULT; + + reg = tmp32; + return vgic_v3_attr_regs_access(dev, attr, ®, true); + } + case KVM_DEV_ARM_VGIC_GRP_CPU_SYSREGS: { + u64 __user *uaddr = (u64 __user *)(long)attr->addr; + u64 reg; + + if (get_user(reg, uaddr)) + return -EFAULT; + + return vgic_v3_attr_regs_access(dev, attr, ®, true); + } + case KVM_DEV_ARM_VGIC_GRP_LEVEL_INFO: { + u32 __user *uaddr = (u32 __user *)(long)attr->addr; + u64 reg; + u32 tmp32; + + if (get_user(tmp32, uaddr)) + return -EFAULT; + + reg = tmp32; + return vgic_v3_attr_regs_access(dev, attr, ®, true); + } + } + return -ENXIO; } static int vgic_v3_get_attr(struct kvm_device *dev, struct kvm_device_attr *attr) { - return vgic_get_common_attr(dev, attr); + int ret; + + ret = vgic_get_common_attr(dev, attr); + if (ret != -ENXIO) + return ret; + + switch (attr->group) { + case KVM_DEV_ARM_VGIC_GRP_DIST_REGS: + case KVM_DEV_ARM_VGIC_GRP_REDIST_REGS: { + u32 __user *uaddr = (u32 __user *)(long)attr->addr; + u64 reg; + u32 tmp32; + + ret = vgic_v3_attr_regs_access(dev, attr, ®, false); + if (ret) + return ret; + tmp32 = reg; + return put_user(tmp32, uaddr); + } + case KVM_DEV_ARM_VGIC_GRP_CPU_SYSREGS: { + u64 __user *uaddr = (u64 __user *)(long)attr->addr; + u64 reg; + + ret = vgic_v3_attr_regs_access(dev, attr, ®, false); + if (ret) + return ret; + return put_user(reg, uaddr); + } + case KVM_DEV_ARM_VGIC_GRP_LEVEL_INFO: { + u32 __user *uaddr = (u32 __user *)(long)attr->addr; + u64 reg; + u32 tmp32; + + ret = vgic_v3_attr_regs_access(dev, attr, ®, false); + if (ret) + return ret; + tmp32 = reg; + return put_user(tmp32, uaddr); + } + } + return -ENXIO; } static int vgic_v3_has_attr(struct kvm_device *dev, @@ -451,8 +641,19 @@ static int vgic_v3_has_attr(struct kvm_device *dev, return 0; } break; + case KVM_DEV_ARM_VGIC_GRP_DIST_REGS: + case KVM_DEV_ARM_VGIC_GRP_REDIST_REGS: + case KVM_DEV_ARM_VGIC_GRP_CPU_SYSREGS: + return vgic_v3_has_attr_regs(dev, attr); case KVM_DEV_ARM_VGIC_GRP_NR_IRQS: return 0; + case KVM_DEV_ARM_VGIC_GRP_LEVEL_INFO: { + if (((attr->attr & KVM_DEV_ARM_VGIC_LINE_LEVEL_INFO_MASK) >> + KVM_DEV_ARM_VGIC_LINE_LEVEL_INFO_SHIFT) == + VGIC_LEVEL_INFO_LINE_LEVEL) + return 0; + break; + } case KVM_DEV_ARM_VGIC_GRP_CTRL: switch (attr->attr) { case KVM_DEV_ARM_VGIC_CTRL_INIT: diff --git a/virt/kvm/arm/vgic/vgic-mmio-v2.c b/virt/kvm/arm/vgic/vgic-mmio-v2.c index 78e34bc4d89b..a3ad7ff95c9b 100644 --- a/virt/kvm/arm/vgic/vgic-mmio-v2.c +++ b/virt/kvm/arm/vgic/vgic-mmio-v2.c @@ -98,7 +98,7 @@ static void vgic_mmio_write_sgir(struct kvm_vcpu *source_vcpu, irq = vgic_get_irq(source_vcpu->kvm, vcpu, intid); spin_lock(&irq->irq_lock); - irq->pending = true; + irq->pending_latch = true; irq->source |= 1U << source_vcpu->vcpu_id; vgic_queue_irq_unlock(source_vcpu->kvm, irq); @@ -182,7 +182,7 @@ static void vgic_mmio_write_sgipendc(struct kvm_vcpu *vcpu, irq->source &= ~((val >> (i * 8)) & 0xff); if (!irq->source) - irq->pending = false; + irq->pending_latch = false; spin_unlock(&irq->irq_lock); vgic_put_irq(vcpu->kvm, irq); @@ -204,7 +204,7 @@ static void vgic_mmio_write_sgipends(struct kvm_vcpu *vcpu, irq->source |= (val >> (i * 8)) & 0xff; if (irq->source) { - irq->pending = true; + irq->pending_latch = true; vgic_queue_irq_unlock(vcpu->kvm, irq); } else { spin_unlock(&irq->irq_lock); @@ -213,22 +213,6 @@ static void vgic_mmio_write_sgipends(struct kvm_vcpu *vcpu, } } -static void vgic_set_vmcr(struct kvm_vcpu *vcpu, struct vgic_vmcr *vmcr) -{ - if (kvm_vgic_global_state.type == VGIC_V2) - vgic_v2_set_vmcr(vcpu, vmcr); - else - vgic_v3_set_vmcr(vcpu, vmcr); -} - -static void vgic_get_vmcr(struct kvm_vcpu *vcpu, struct vgic_vmcr *vmcr) -{ - if (kvm_vgic_global_state.type == VGIC_V2) - vgic_v2_get_vmcr(vcpu, vmcr); - else - vgic_v3_get_vmcr(vcpu, vmcr); -} - #define GICC_ARCH_VERSION_V2 0x2 /* These are for userland accesses only, there is no guest-facing emulation. */ @@ -369,21 +353,30 @@ unsigned int vgic_v2_init_dist_iodev(struct vgic_io_device *dev) int vgic_v2_has_attr_regs(struct kvm_device *dev, struct kvm_device_attr *attr) { - int nr_irqs = dev->kvm->arch.vgic.nr_spis + VGIC_NR_PRIVATE_IRQS; - const struct vgic_register_region *regions; + const struct vgic_register_region *region; + struct vgic_io_device iodev; + struct vgic_reg_attr reg_attr; + struct kvm_vcpu *vcpu; gpa_t addr; - int nr_regions, i, len; + int ret; + + ret = vgic_v2_parse_attr(dev, attr, ®_attr); + if (ret) + return ret; - addr = attr->attr & KVM_DEV_ARM_VGIC_OFFSET_MASK; + vcpu = reg_attr.vcpu; + addr = reg_attr.addr; switch (attr->group) { case KVM_DEV_ARM_VGIC_GRP_DIST_REGS: - regions = vgic_v2_dist_registers; - nr_regions = ARRAY_SIZE(vgic_v2_dist_registers); + iodev.regions = vgic_v2_dist_registers; + iodev.nr_regions = ARRAY_SIZE(vgic_v2_dist_registers); + iodev.base_addr = 0; break; case KVM_DEV_ARM_VGIC_GRP_CPU_REGS: - regions = vgic_v2_cpu_registers; - nr_regions = ARRAY_SIZE(vgic_v2_cpu_registers); + iodev.regions = vgic_v2_cpu_registers; + iodev.nr_regions = ARRAY_SIZE(vgic_v2_cpu_registers); + iodev.base_addr = 0; break; default: return -ENXIO; @@ -393,43 +386,11 @@ int vgic_v2_has_attr_regs(struct kvm_device *dev, struct kvm_device_attr *attr) if (addr & 3) return -ENXIO; - for (i = 0; i < nr_regions; i++) { - if (regions[i].bits_per_irq) - len = (regions[i].bits_per_irq * nr_irqs) / 8; - else - len = regions[i].len; - - if (regions[i].reg_offset <= addr && - regions[i].reg_offset + len > addr) - return 0; - } - - return -ENXIO; -} - -/* - * When userland tries to access the VGIC register handlers, we need to - * create a usable struct vgic_io_device to be passed to the handlers and we - * have to set up a buffer similar to what would have happened if a guest MMIO - * access occurred, including doing endian conversions on BE systems. - */ -static int vgic_uaccess(struct kvm_vcpu *vcpu, struct vgic_io_device *dev, - bool is_write, int offset, u32 *val) -{ - unsigned int len = 4; - u8 buf[4]; - int ret; - - if (is_write) { - vgic_data_host_to_mmio_bus(buf, len, *val); - ret = kvm_io_gic_ops.write(vcpu, &dev->dev, offset, len, buf); - } else { - ret = kvm_io_gic_ops.read(vcpu, &dev->dev, offset, len, buf); - if (!ret) - *val = vgic_data_mmio_bus_to_host(buf, len); - } + region = vgic_get_mmio_region(vcpu, &iodev, addr, sizeof(u32)); + if (!region) + return -ENXIO; - return ret; + return 0; } int vgic_v2_cpuif_uaccess(struct kvm_vcpu *vcpu, bool is_write, diff --git a/virt/kvm/arm/vgic/vgic-mmio-v3.c b/virt/kvm/arm/vgic/vgic-mmio-v3.c index 50f42f0f8c4f..6afb3b484886 100644 --- a/virt/kvm/arm/vgic/vgic-mmio-v3.c +++ b/virt/kvm/arm/vgic/vgic-mmio-v3.c @@ -18,6 +18,8 @@ #include <kvm/arm_vgic.h> #include <asm/kvm_emulate.h> +#include <asm/kvm_arm.h> +#include <asm/kvm_mmu.h> #include "vgic.h" #include "vgic-mmio.h" @@ -207,6 +209,60 @@ static unsigned long vgic_mmio_read_v3_idregs(struct kvm_vcpu *vcpu, return 0; } +static unsigned long vgic_v3_uaccess_read_pending(struct kvm_vcpu *vcpu, + gpa_t addr, unsigned int len) +{ + u32 intid = VGIC_ADDR_TO_INTID(addr, 1); + u32 value = 0; + int i; + + /* + * pending state of interrupt is latched in pending_latch variable. + * Userspace will save and restore pending state and line_level + * separately. + * Refer to Documentation/virtual/kvm/devices/arm-vgic-v3.txt + * for handling of ISPENDR and ICPENDR. + */ + for (i = 0; i < len * 8; i++) { + struct vgic_irq *irq = vgic_get_irq(vcpu->kvm, vcpu, intid + i); + + if (irq->pending_latch) + value |= (1U << i); + + vgic_put_irq(vcpu->kvm, irq); + } + + return value; +} + +static void vgic_v3_uaccess_write_pending(struct kvm_vcpu *vcpu, + gpa_t addr, unsigned int len, + unsigned long val) +{ + u32 intid = VGIC_ADDR_TO_INTID(addr, 1); + int i; + + for (i = 0; i < len * 8; i++) { + struct vgic_irq *irq = vgic_get_irq(vcpu->kvm, vcpu, intid + i); + + spin_lock(&irq->irq_lock); + if (test_bit(i, &val)) { + /* + * pending_latch is set irrespective of irq type + * (level or edge) to avoid dependency that VM should + * restore irq config before pending info. + */ + irq->pending_latch = true; + vgic_queue_irq_unlock(vcpu->kvm, irq); + } else { + irq->pending_latch = false; + spin_unlock(&irq->irq_lock); + } + + vgic_put_irq(vcpu->kvm, irq); + } +} + /* We want to avoid outer shareable. */ u64 vgic_sanitise_shareability(u64 field) { @@ -356,7 +412,7 @@ static void vgic_mmio_write_pendbase(struct kvm_vcpu *vcpu, * We take some special care here to fix the calculation of the register * offset. */ -#define REGISTER_DESC_WITH_BITS_PER_IRQ_SHARED(off, rd, wr, bpi, acc) \ +#define REGISTER_DESC_WITH_BITS_PER_IRQ_SHARED(off, rd, wr, ur, uw, bpi, acc) \ { \ .reg_offset = off, \ .bits_per_irq = bpi, \ @@ -371,47 +427,54 @@ static void vgic_mmio_write_pendbase(struct kvm_vcpu *vcpu, .access_flags = acc, \ .read = rd, \ .write = wr, \ + .uaccess_read = ur, \ + .uaccess_write = uw, \ } static const struct vgic_register_region vgic_v3_dist_registers[] = { REGISTER_DESC_WITH_LENGTH(GICD_CTLR, vgic_mmio_read_v3_misc, vgic_mmio_write_v3_misc, 16, VGIC_ACCESS_32bit), + REGISTER_DESC_WITH_LENGTH(GICD_STATUSR, + vgic_mmio_read_rao, vgic_mmio_write_wi, 4, + VGIC_ACCESS_32bit), REGISTER_DESC_WITH_BITS_PER_IRQ_SHARED(GICD_IGROUPR, - vgic_mmio_read_rao, vgic_mmio_write_wi, 1, + vgic_mmio_read_rao, vgic_mmio_write_wi, NULL, NULL, 1, VGIC_ACCESS_32bit), REGISTER_DESC_WITH_BITS_PER_IRQ_SHARED(GICD_ISENABLER, - vgic_mmio_read_enable, vgic_mmio_write_senable, 1, + vgic_mmio_read_enable, vgic_mmio_write_senable, NULL, NULL, 1, VGIC_ACCESS_32bit), REGISTER_DESC_WITH_BITS_PER_IRQ_SHARED(GICD_ICENABLER, - vgic_mmio_read_enable, vgic_mmio_write_cenable, 1, + vgic_mmio_read_enable, vgic_mmio_write_cenable, NULL, NULL, 1, VGIC_ACCESS_32bit), REGISTER_DESC_WITH_BITS_PER_IRQ_SHARED(GICD_ISPENDR, - vgic_mmio_read_pending, vgic_mmio_write_spending, 1, + vgic_mmio_read_pending, vgic_mmio_write_spending, + vgic_v3_uaccess_read_pending, vgic_v3_uaccess_write_pending, 1, VGIC_ACCESS_32bit), REGISTER_DESC_WITH_BITS_PER_IRQ_SHARED(GICD_ICPENDR, - vgic_mmio_read_pending, vgic_mmio_write_cpending, 1, + vgic_mmio_read_pending, vgic_mmio_write_cpending, + vgic_mmio_read_raz, vgic_mmio_write_wi, 1, VGIC_ACCESS_32bit), REGISTER_DESC_WITH_BITS_PER_IRQ_SHARED(GICD_ISACTIVER, - vgic_mmio_read_active, vgic_mmio_write_sactive, 1, + vgic_mmio_read_active, vgic_mmio_write_sactive, NULL, NULL, 1, VGIC_ACCESS_32bit), REGISTER_DESC_WITH_BITS_PER_IRQ_SHARED(GICD_ICACTIVER, - vgic_mmio_read_active, vgic_mmio_write_cactive, 1, + vgic_mmio_read_active, vgic_mmio_write_cactive, NULL, NULL, 1, VGIC_ACCESS_32bit), REGISTER_DESC_WITH_BITS_PER_IRQ_SHARED(GICD_IPRIORITYR, - vgic_mmio_read_priority, vgic_mmio_write_priority, 8, - VGIC_ACCESS_32bit | VGIC_ACCESS_8bit), + vgic_mmio_read_priority, vgic_mmio_write_priority, NULL, NULL, + 8, VGIC_ACCESS_32bit | VGIC_ACCESS_8bit), REGISTER_DESC_WITH_BITS_PER_IRQ_SHARED(GICD_ITARGETSR, - vgic_mmio_read_raz, vgic_mmio_write_wi, 8, + vgic_mmio_read_raz, vgic_mmio_write_wi, NULL, NULL, 8, VGIC_ACCESS_32bit | VGIC_ACCESS_8bit), REGISTER_DESC_WITH_BITS_PER_IRQ_SHARED(GICD_ICFGR, - vgic_mmio_read_config, vgic_mmio_write_config, 2, + vgic_mmio_read_config, vgic_mmio_write_config, NULL, NULL, 2, VGIC_ACCESS_32bit), REGISTER_DESC_WITH_BITS_PER_IRQ_SHARED(GICD_IGRPMODR, - vgic_mmio_read_raz, vgic_mmio_write_wi, 1, + vgic_mmio_read_raz, vgic_mmio_write_wi, NULL, NULL, 1, VGIC_ACCESS_32bit), REGISTER_DESC_WITH_BITS_PER_IRQ_SHARED(GICD_IROUTER, - vgic_mmio_read_irouter, vgic_mmio_write_irouter, 64, + vgic_mmio_read_irouter, vgic_mmio_write_irouter, NULL, NULL, 64, VGIC_ACCESS_64bit | VGIC_ACCESS_32bit), REGISTER_DESC_WITH_LENGTH(GICD_IDREGS, vgic_mmio_read_v3_idregs, vgic_mmio_write_wi, 48, @@ -422,12 +485,18 @@ static const struct vgic_register_region vgic_v3_rdbase_registers[] = { REGISTER_DESC_WITH_LENGTH(GICR_CTLR, vgic_mmio_read_v3r_ctlr, vgic_mmio_write_v3r_ctlr, 4, VGIC_ACCESS_32bit), + REGISTER_DESC_WITH_LENGTH(GICR_STATUSR, + vgic_mmio_read_raz, vgic_mmio_write_wi, 4, + VGIC_ACCESS_32bit), REGISTER_DESC_WITH_LENGTH(GICR_IIDR, vgic_mmio_read_v3r_iidr, vgic_mmio_write_wi, 4, VGIC_ACCESS_32bit), REGISTER_DESC_WITH_LENGTH(GICR_TYPER, vgic_mmio_read_v3r_typer, vgic_mmio_write_wi, 8, VGIC_ACCESS_64bit | VGIC_ACCESS_32bit), + REGISTER_DESC_WITH_LENGTH(GICR_WAKER, + vgic_mmio_read_raz, vgic_mmio_write_wi, 4, + VGIC_ACCESS_32bit), REGISTER_DESC_WITH_LENGTH(GICR_PROPBASER, vgic_mmio_read_propbase, vgic_mmio_write_propbase, 8, VGIC_ACCESS_64bit | VGIC_ACCESS_32bit), @@ -449,11 +518,13 @@ static const struct vgic_register_region vgic_v3_sgibase_registers[] = { REGISTER_DESC_WITH_LENGTH(GICR_ICENABLER0, vgic_mmio_read_enable, vgic_mmio_write_cenable, 4, VGIC_ACCESS_32bit), - REGISTER_DESC_WITH_LENGTH(GICR_ISPENDR0, - vgic_mmio_read_pending, vgic_mmio_write_spending, 4, + REGISTER_DESC_WITH_LENGTH_UACCESS(GICR_ISPENDR0, + vgic_mmio_read_pending, vgic_mmio_write_spending, + vgic_v3_uaccess_read_pending, vgic_v3_uaccess_write_pending, 4, VGIC_ACCESS_32bit), - REGISTER_DESC_WITH_LENGTH(GICR_ICPENDR0, - vgic_mmio_read_pending, vgic_mmio_write_cpending, 4, + REGISTER_DESC_WITH_LENGTH_UACCESS(GICR_ICPENDR0, + vgic_mmio_read_pending, vgic_mmio_write_cpending, + vgic_mmio_read_raz, vgic_mmio_write_wi, 4, VGIC_ACCESS_32bit), REGISTER_DESC_WITH_LENGTH(GICR_ISACTIVER0, vgic_mmio_read_active, vgic_mmio_write_sactive, 4, @@ -546,6 +617,54 @@ int vgic_register_redist_iodevs(struct kvm *kvm, gpa_t redist_base_address) return ret; } +int vgic_v3_has_attr_regs(struct kvm_device *dev, struct kvm_device_attr *attr) +{ + const struct vgic_register_region *region; + struct vgic_io_device iodev; + struct vgic_reg_attr reg_attr; + struct kvm_vcpu *vcpu; + gpa_t addr; + int ret; + + ret = vgic_v3_parse_attr(dev, attr, ®_attr); + if (ret) + return ret; + + vcpu = reg_attr.vcpu; + addr = reg_attr.addr; + + switch (attr->group) { + case KVM_DEV_ARM_VGIC_GRP_DIST_REGS: + iodev.regions = vgic_v3_dist_registers; + iodev.nr_regions = ARRAY_SIZE(vgic_v3_dist_registers); + iodev.base_addr = 0; + break; + case KVM_DEV_ARM_VGIC_GRP_REDIST_REGS:{ + iodev.regions = vgic_v3_rdbase_registers; + iodev.nr_regions = ARRAY_SIZE(vgic_v3_rdbase_registers); + iodev.base_addr = 0; + break; + } + case KVM_DEV_ARM_VGIC_GRP_CPU_SYSREGS: { + u64 reg, id; + + id = (attr->attr & KVM_DEV_ARM_VGIC_SYSREG_INSTR_MASK); + return vgic_v3_has_cpu_sysregs_attr(vcpu, 0, id, ®); + } + default: + return -ENXIO; + } + + /* We only support aligned 32-bit accesses. */ + if (addr & 3) + return -ENXIO; + + region = vgic_get_mmio_region(vcpu, &iodev, addr, sizeof(u32)); + if (!region) + return -ENXIO; + + return 0; +} /* * Compare a given affinity (level 1-3 and a level 0 mask, from the SGI * generation register ICC_SGI1R_EL1) with a given VCPU. @@ -646,9 +765,55 @@ void vgic_v3_dispatch_sgi(struct kvm_vcpu *vcpu, u64 reg) irq = vgic_get_irq(vcpu->kvm, c_vcpu, sgi); spin_lock(&irq->irq_lock); - irq->pending = true; + irq->pending_latch = true; vgic_queue_irq_unlock(vcpu->kvm, irq); vgic_put_irq(vcpu->kvm, irq); } } + +int vgic_v3_dist_uaccess(struct kvm_vcpu *vcpu, bool is_write, + int offset, u32 *val) +{ + struct vgic_io_device dev = { + .regions = vgic_v3_dist_registers, + .nr_regions = ARRAY_SIZE(vgic_v3_dist_registers), + }; + + return vgic_uaccess(vcpu, &dev, is_write, offset, val); +} + +int vgic_v3_redist_uaccess(struct kvm_vcpu *vcpu, bool is_write, + int offset, u32 *val) +{ + struct vgic_io_device rd_dev = { + .regions = vgic_v3_rdbase_registers, + .nr_regions = ARRAY_SIZE(vgic_v3_rdbase_registers), + }; + + struct vgic_io_device sgi_dev = { + .regions = vgic_v3_sgibase_registers, + .nr_regions = ARRAY_SIZE(vgic_v3_sgibase_registers), + }; + + /* SGI_base is the next 64K frame after RD_base */ + if (offset >= SZ_64K) + return vgic_uaccess(vcpu, &sgi_dev, is_write, offset - SZ_64K, + val); + else + return vgic_uaccess(vcpu, &rd_dev, is_write, offset, val); +} + +int vgic_v3_line_level_info_uaccess(struct kvm_vcpu *vcpu, bool is_write, + u32 intid, u64 *val) +{ + if (intid % 32) + return -EINVAL; + + if (is_write) + vgic_write_irq_line_level_info(vcpu, intid, *val); + else + *val = vgic_read_irq_line_level_info(vcpu, intid); + + return 0; +} diff --git a/virt/kvm/arm/vgic/vgic-mmio.c b/virt/kvm/arm/vgic/vgic-mmio.c index ebe1b9fa3c4d..3654b4c835ef 100644 --- a/virt/kvm/arm/vgic/vgic-mmio.c +++ b/virt/kvm/arm/vgic/vgic-mmio.c @@ -111,7 +111,7 @@ unsigned long vgic_mmio_read_pending(struct kvm_vcpu *vcpu, for (i = 0; i < len * 8; i++) { struct vgic_irq *irq = vgic_get_irq(vcpu->kvm, vcpu, intid + i); - if (irq->pending) + if (irq_is_pending(irq)) value |= (1U << i); vgic_put_irq(vcpu->kvm, irq); @@ -131,9 +131,7 @@ void vgic_mmio_write_spending(struct kvm_vcpu *vcpu, struct vgic_irq *irq = vgic_get_irq(vcpu->kvm, vcpu, intid + i); spin_lock(&irq->irq_lock); - irq->pending = true; - if (irq->config == VGIC_CONFIG_LEVEL) - irq->soft_pending = true; + irq->pending_latch = true; vgic_queue_irq_unlock(vcpu->kvm, irq); vgic_put_irq(vcpu->kvm, irq); @@ -152,12 +150,7 @@ void vgic_mmio_write_cpending(struct kvm_vcpu *vcpu, spin_lock(&irq->irq_lock); - if (irq->config == VGIC_CONFIG_LEVEL) { - irq->soft_pending = false; - irq->pending = irq->line_level; - } else { - irq->pending = false; - } + irq->pending_latch = false; spin_unlock(&irq->irq_lock); vgic_put_irq(vcpu->kvm, irq); @@ -359,18 +352,70 @@ void vgic_mmio_write_config(struct kvm_vcpu *vcpu, irq = vgic_get_irq(vcpu->kvm, vcpu, intid + i); spin_lock(&irq->irq_lock); - if (test_bit(i * 2 + 1, &val)) { + if (test_bit(i * 2 + 1, &val)) irq->config = VGIC_CONFIG_EDGE; - } else { + else irq->config = VGIC_CONFIG_LEVEL; - irq->pending = irq->line_level | irq->soft_pending; - } spin_unlock(&irq->irq_lock); vgic_put_irq(vcpu->kvm, irq); } } +u64 vgic_read_irq_line_level_info(struct kvm_vcpu *vcpu, u32 intid) +{ + int i; + u64 val = 0; + int nr_irqs = vcpu->kvm->arch.vgic.nr_spis + VGIC_NR_PRIVATE_IRQS; + + for (i = 0; i < 32; i++) { + struct vgic_irq *irq; + + if ((intid + i) < VGIC_NR_SGIS || (intid + i) >= nr_irqs) + continue; + + irq = vgic_get_irq(vcpu->kvm, vcpu, intid + i); + if (irq->config == VGIC_CONFIG_LEVEL && irq->line_level) + val |= (1U << i); + + vgic_put_irq(vcpu->kvm, irq); + } + + return val; +} + +void vgic_write_irq_line_level_info(struct kvm_vcpu *vcpu, u32 intid, + const u64 val) +{ + int i; + int nr_irqs = vcpu->kvm->arch.vgic.nr_spis + VGIC_NR_PRIVATE_IRQS; + + for (i = 0; i < 32; i++) { + struct vgic_irq *irq; + bool new_level; + + if ((intid + i) < VGIC_NR_SGIS || (intid + i) >= nr_irqs) + continue; + + irq = vgic_get_irq(vcpu->kvm, vcpu, intid + i); + + /* + * Line level is set irrespective of irq type + * (level or edge) to avoid dependency that VM should + * restore irq config before line level. + */ + new_level = !!(val & (1U << i)); + spin_lock(&irq->irq_lock); + irq->line_level = new_level; + if (new_level) + vgic_queue_irq_unlock(vcpu->kvm, irq); + else + spin_unlock(&irq->irq_lock); + + vgic_put_irq(vcpu->kvm, irq); + } +} + static int match_region(const void *key, const void *elt) { const unsigned int offset = (unsigned long)key; @@ -394,6 +439,22 @@ vgic_find_mmio_region(const struct vgic_register_region *region, int nr_regions, sizeof(region[0]), match_region); } +void vgic_set_vmcr(struct kvm_vcpu *vcpu, struct vgic_vmcr *vmcr) +{ + if (kvm_vgic_global_state.type == VGIC_V2) + vgic_v2_set_vmcr(vcpu, vmcr); + else + vgic_v3_set_vmcr(vcpu, vmcr); +} + +void vgic_get_vmcr(struct kvm_vcpu *vcpu, struct vgic_vmcr *vmcr) +{ + if (kvm_vgic_global_state.type == VGIC_V2) + vgic_v2_get_vmcr(vcpu, vmcr); + else + vgic_v3_get_vmcr(vcpu, vmcr); +} + /* * kvm_mmio_read_buf() returns a value in a format where it can be converted * to a byte array and be directly observed as the guest wanted it to appear @@ -484,6 +545,74 @@ static bool check_region(const struct kvm *kvm, return false; } +const struct vgic_register_region * +vgic_get_mmio_region(struct kvm_vcpu *vcpu, struct vgic_io_device *iodev, + gpa_t addr, int len) +{ + const struct vgic_register_region *region; + + region = vgic_find_mmio_region(iodev->regions, iodev->nr_regions, + addr - iodev->base_addr); + if (!region || !check_region(vcpu->kvm, region, addr, len)) + return NULL; + + return region; +} + +static int vgic_uaccess_read(struct kvm_vcpu *vcpu, struct kvm_io_device *dev, + gpa_t addr, u32 *val) +{ + struct vgic_io_device *iodev = kvm_to_vgic_iodev(dev); + const struct vgic_register_region *region; + struct kvm_vcpu *r_vcpu; + + region = vgic_get_mmio_region(vcpu, iodev, addr, sizeof(u32)); + if (!region) { + *val = 0; + return 0; + } + + r_vcpu = iodev->redist_vcpu ? iodev->redist_vcpu : vcpu; + if (region->uaccess_read) + *val = region->uaccess_read(r_vcpu, addr, sizeof(u32)); + else + *val = region->read(r_vcpu, addr, sizeof(u32)); + + return 0; +} + +static int vgic_uaccess_write(struct kvm_vcpu *vcpu, struct kvm_io_device *dev, + gpa_t addr, const u32 *val) +{ + struct vgic_io_device *iodev = kvm_to_vgic_iodev(dev); + const struct vgic_register_region *region; + struct kvm_vcpu *r_vcpu; + + region = vgic_get_mmio_region(vcpu, iodev, addr, sizeof(u32)); + if (!region) + return 0; + + r_vcpu = iodev->redist_vcpu ? iodev->redist_vcpu : vcpu; + if (region->uaccess_write) + region->uaccess_write(r_vcpu, addr, sizeof(u32), *val); + else + region->write(r_vcpu, addr, sizeof(u32), *val); + + return 0; +} + +/* + * Userland access to VGIC registers. + */ +int vgic_uaccess(struct kvm_vcpu *vcpu, struct vgic_io_device *dev, + bool is_write, int offset, u32 *val) +{ + if (is_write) + return vgic_uaccess_write(vcpu, &dev->dev, offset, val); + else + return vgic_uaccess_read(vcpu, &dev->dev, offset, val); +} + static int dispatch_mmio_read(struct kvm_vcpu *vcpu, struct kvm_io_device *dev, gpa_t addr, int len, void *val) { @@ -491,9 +620,8 @@ static int dispatch_mmio_read(struct kvm_vcpu *vcpu, struct kvm_io_device *dev, const struct vgic_register_region *region; unsigned long data = 0; - region = vgic_find_mmio_region(iodev->regions, iodev->nr_regions, - addr - iodev->base_addr); - if (!region || !check_region(vcpu->kvm, region, addr, len)) { + region = vgic_get_mmio_region(vcpu, iodev, addr, len); + if (!region) { memset(val, 0, len); return 0; } @@ -524,9 +652,8 @@ static int dispatch_mmio_write(struct kvm_vcpu *vcpu, struct kvm_io_device *dev, const struct vgic_register_region *region; unsigned long data = vgic_data_mmio_bus_to_host(val, len); - region = vgic_find_mmio_region(iodev->regions, iodev->nr_regions, - addr - iodev->base_addr); - if (!region || !check_region(vcpu->kvm, region, addr, len)) + region = vgic_get_mmio_region(vcpu, iodev, addr, len); + if (!region) return 0; switch (iodev->iodev_type) { diff --git a/virt/kvm/arm/vgic/vgic-mmio.h b/virt/kvm/arm/vgic/vgic-mmio.h index 84961b4e4422..98bb566b660a 100644 --- a/virt/kvm/arm/vgic/vgic-mmio.h +++ b/virt/kvm/arm/vgic/vgic-mmio.h @@ -34,6 +34,10 @@ struct vgic_register_region { gpa_t addr, unsigned int len, unsigned long val); }; + unsigned long (*uaccess_read)(struct kvm_vcpu *vcpu, gpa_t addr, + unsigned int len); + void (*uaccess_write)(struct kvm_vcpu *vcpu, gpa_t addr, + unsigned int len, unsigned long val); }; extern struct kvm_io_device_ops kvm_io_gic_ops; @@ -86,6 +90,18 @@ extern struct kvm_io_device_ops kvm_io_gic_ops; .write = wr, \ } +#define REGISTER_DESC_WITH_LENGTH_UACCESS(off, rd, wr, urd, uwr, length, acc) \ + { \ + .reg_offset = off, \ + .bits_per_irq = 0, \ + .len = length, \ + .access_flags = acc, \ + .read = rd, \ + .write = wr, \ + .uaccess_read = urd, \ + .uaccess_write = uwr, \ + } + int kvm_vgic_register_mmio_region(struct kvm *kvm, struct kvm_vcpu *vcpu, struct vgic_register_region *reg_desc, struct vgic_io_device *region, @@ -158,6 +174,14 @@ void vgic_mmio_write_config(struct kvm_vcpu *vcpu, gpa_t addr, unsigned int len, unsigned long val); +int vgic_uaccess(struct kvm_vcpu *vcpu, struct vgic_io_device *dev, + bool is_write, int offset, u32 *val); + +u64 vgic_read_irq_line_level_info(struct kvm_vcpu *vcpu, u32 intid); + +void vgic_write_irq_line_level_info(struct kvm_vcpu *vcpu, u32 intid, + const u64 val); + unsigned int vgic_v2_init_dist_iodev(struct vgic_io_device *dev); unsigned int vgic_v3_init_dist_iodev(struct vgic_io_device *dev); diff --git a/virt/kvm/arm/vgic/vgic-v2.c b/virt/kvm/arm/vgic/vgic-v2.c index 834137e7b83f..b834ecdf3225 100644 --- a/virt/kvm/arm/vgic/vgic-v2.c +++ b/virt/kvm/arm/vgic/vgic-v2.c @@ -104,7 +104,7 @@ void vgic_v2_fold_lr_state(struct kvm_vcpu *vcpu) /* Edge is the only case where we preserve the pending bit */ if (irq->config == VGIC_CONFIG_EDGE && (val & GICH_LR_PENDING_BIT)) { - irq->pending = true; + irq->pending_latch = true; if (vgic_irq_is_sgi(intid)) { u32 cpuid = val & GICH_LR_PHYSID_CPUID; @@ -120,9 +120,7 @@ void vgic_v2_fold_lr_state(struct kvm_vcpu *vcpu) */ if (irq->config == VGIC_CONFIG_LEVEL) { if (!(val & GICH_LR_PENDING_BIT)) - irq->soft_pending = false; - - irq->pending = irq->line_level || irq->soft_pending; + irq->pending_latch = false; } spin_unlock(&irq->irq_lock); @@ -145,11 +143,11 @@ void vgic_v2_populate_lr(struct kvm_vcpu *vcpu, struct vgic_irq *irq, int lr) { u32 val = irq->intid; - if (irq->pending) { + if (irq_is_pending(irq)) { val |= GICH_LR_PENDING_BIT; if (irq->config == VGIC_CONFIG_EDGE) - irq->pending = false; + irq->pending_latch = false; if (vgic_irq_is_sgi(irq->intid)) { u32 src = ffs(irq->source); @@ -158,7 +156,7 @@ void vgic_v2_populate_lr(struct kvm_vcpu *vcpu, struct vgic_irq *irq, int lr) val |= (src - 1) << GICH_LR_PHYSID_CPUID_SHIFT; irq->source &= ~(1 << (src - 1)); if (irq->source) - irq->pending = true; + irq->pending_latch = true; } } diff --git a/virt/kvm/arm/vgic/vgic-v3.c b/virt/kvm/arm/vgic/vgic-v3.c index e6b03fd8c374..edc6ee2dc852 100644 --- a/virt/kvm/arm/vgic/vgic-v3.c +++ b/virt/kvm/arm/vgic/vgic-v3.c @@ -94,7 +94,7 @@ void vgic_v3_fold_lr_state(struct kvm_vcpu *vcpu) /* Edge is the only case where we preserve the pending bit */ if (irq->config == VGIC_CONFIG_EDGE && (val & ICH_LR_PENDING_BIT)) { - irq->pending = true; + irq->pending_latch = true; if (vgic_irq_is_sgi(intid) && model == KVM_DEV_TYPE_ARM_VGIC_V2) { @@ -111,9 +111,7 @@ void vgic_v3_fold_lr_state(struct kvm_vcpu *vcpu) */ if (irq->config == VGIC_CONFIG_LEVEL) { if (!(val & ICH_LR_PENDING_BIT)) - irq->soft_pending = false; - - irq->pending = irq->line_level || irq->soft_pending; + irq->pending_latch = false; } spin_unlock(&irq->irq_lock); @@ -127,11 +125,11 @@ void vgic_v3_populate_lr(struct kvm_vcpu *vcpu, struct vgic_irq *irq, int lr) u32 model = vcpu->kvm->arch.vgic.vgic_model; u64 val = irq->intid; - if (irq->pending) { + if (irq_is_pending(irq)) { val |= ICH_LR_PENDING_BIT; if (irq->config == VGIC_CONFIG_EDGE) - irq->pending = false; + irq->pending_latch = false; if (vgic_irq_is_sgi(irq->intid) && model == KVM_DEV_TYPE_ARM_VGIC_V2) { @@ -141,7 +139,7 @@ void vgic_v3_populate_lr(struct kvm_vcpu *vcpu, struct vgic_irq *irq, int lr) val |= (src - 1) << GICH_LR_PHYSID_CPUID_SHIFT; irq->source &= ~(1 << (src - 1)); if (irq->source) - irq->pending = true; + irq->pending_latch = true; } } @@ -177,10 +175,18 @@ void vgic_v3_set_vmcr(struct kvm_vcpu *vcpu, struct vgic_vmcr *vmcrp) { u32 vmcr; - vmcr = (vmcrp->ctlr << ICH_VMCR_CTLR_SHIFT) & ICH_VMCR_CTLR_MASK; + /* + * Ignore the FIQen bit, because GIC emulation always implies + * SRE=1 which means the vFIQEn bit is also RES1. + */ + vmcr = ((vmcrp->ctlr >> ICC_CTLR_EL1_EOImode_SHIFT) << + ICH_VMCR_EOIM_SHIFT) & ICH_VMCR_EOIM_MASK; + vmcr |= (vmcrp->ctlr << ICH_VMCR_CBPR_SHIFT) & ICH_VMCR_CBPR_MASK; vmcr |= (vmcrp->abpr << ICH_VMCR_BPR1_SHIFT) & ICH_VMCR_BPR1_MASK; vmcr |= (vmcrp->bpr << ICH_VMCR_BPR0_SHIFT) & ICH_VMCR_BPR0_MASK; vmcr |= (vmcrp->pmr << ICH_VMCR_PMR_SHIFT) & ICH_VMCR_PMR_MASK; + vmcr |= (vmcrp->grpen0 << ICH_VMCR_ENG0_SHIFT) & ICH_VMCR_ENG0_MASK; + vmcr |= (vmcrp->grpen1 << ICH_VMCR_ENG1_SHIFT) & ICH_VMCR_ENG1_MASK; vcpu->arch.vgic_cpu.vgic_v3.vgic_vmcr = vmcr; } @@ -189,10 +195,18 @@ void vgic_v3_get_vmcr(struct kvm_vcpu *vcpu, struct vgic_vmcr *vmcrp) { u32 vmcr = vcpu->arch.vgic_cpu.vgic_v3.vgic_vmcr; - vmcrp->ctlr = (vmcr & ICH_VMCR_CTLR_MASK) >> ICH_VMCR_CTLR_SHIFT; + /* + * Ignore the FIQen bit, because GIC emulation always implies + * SRE=1 which means the vFIQEn bit is also RES1. + */ + vmcrp->ctlr = ((vmcr >> ICH_VMCR_EOIM_SHIFT) << + ICC_CTLR_EL1_EOImode_SHIFT) & ICC_CTLR_EL1_EOImode_MASK; + vmcrp->ctlr |= (vmcr & ICH_VMCR_CBPR_MASK) >> ICH_VMCR_CBPR_SHIFT; vmcrp->abpr = (vmcr & ICH_VMCR_BPR1_MASK) >> ICH_VMCR_BPR1_SHIFT; vmcrp->bpr = (vmcr & ICH_VMCR_BPR0_MASK) >> ICH_VMCR_BPR0_SHIFT; vmcrp->pmr = (vmcr & ICH_VMCR_PMR_MASK) >> ICH_VMCR_PMR_SHIFT; + vmcrp->grpen0 = (vmcr & ICH_VMCR_ENG0_MASK) >> ICH_VMCR_ENG0_SHIFT; + vmcrp->grpen1 = (vmcr & ICH_VMCR_ENG1_MASK) >> ICH_VMCR_ENG1_SHIFT; } #define INITIAL_PENDBASER_VALUE \ @@ -224,6 +238,13 @@ void vgic_v3_enable(struct kvm_vcpu *vcpu) vgic_v3->vgic_sre = 0; } + vcpu->arch.vgic_cpu.num_id_bits = (kvm_vgic_global_state.ich_vtr_el2 & + ICH_VTR_ID_BITS_MASK) >> + ICH_VTR_ID_BITS_SHIFT; + vcpu->arch.vgic_cpu.num_pri_bits = ((kvm_vgic_global_state.ich_vtr_el2 & + ICH_VTR_PRI_BITS_MASK) >> + ICH_VTR_PRI_BITS_SHIFT) + 1; + /* Get the show on the road... */ vgic_v3->vgic_hcr = ICH_HCR_EN; } @@ -322,6 +343,7 @@ int vgic_v3_probe(const struct gic_kvm_info *info) */ kvm_vgic_global_state.nr_lr = (ich_vtr_el2 & 0xf) + 1; kvm_vgic_global_state.can_emulate_gicv2 = false; + kvm_vgic_global_state.ich_vtr_el2 = ich_vtr_el2; if (!info->vcpu.start) { kvm_info("GICv3: no GICV resource entry\n"); diff --git a/virt/kvm/arm/vgic/vgic.c b/virt/kvm/arm/vgic/vgic.c index 6440b56ec90e..654dfd40e449 100644 --- a/virt/kvm/arm/vgic/vgic.c +++ b/virt/kvm/arm/vgic/vgic.c @@ -160,7 +160,7 @@ static struct kvm_vcpu *vgic_target_oracle(struct vgic_irq *irq) * If the distributor is disabled, pending interrupts shouldn't be * forwarded. */ - if (irq->enabled && irq->pending) { + if (irq->enabled && irq_is_pending(irq)) { if (unlikely(irq->target_vcpu && !irq->target_vcpu->kvm->arch.vgic.enabled)) return NULL; @@ -204,8 +204,8 @@ static int vgic_irq_cmp(void *priv, struct list_head *a, struct list_head *b) goto out; } - penda = irqa->enabled && irqa->pending; - pendb = irqb->enabled && irqb->pending; + penda = irqa->enabled && irq_is_pending(irqa); + pendb = irqb->enabled && irq_is_pending(irqb); if (!penda || !pendb) { ret = (int)pendb - (int)penda; @@ -335,9 +335,22 @@ retry: return true; } -static int vgic_update_irq_pending(struct kvm *kvm, int cpuid, - unsigned int intid, bool level, - bool mapped_irq) +/** + * kvm_vgic_inject_irq - Inject an IRQ from a device to the vgic + * @kvm: The VM structure pointer + * @cpuid: The CPU for PPIs + * @intid: The INTID to inject a new state to. + * @level: Edge-triggered: true: to trigger the interrupt + * false: to ignore the call + * Level-sensitive true: raise the input signal + * false: lower the input signal + * + * The VGIC is not concerned with devices being active-LOW or active-HIGH for + * level-sensitive interrupts. You can think of the level parameter as 1 + * being HIGH and 0 being LOW and all devices being active-HIGH. + */ +int kvm_vgic_inject_irq(struct kvm *kvm, int cpuid, unsigned int intid, + bool level) { struct kvm_vcpu *vcpu; struct vgic_irq *irq; @@ -357,11 +370,6 @@ static int vgic_update_irq_pending(struct kvm *kvm, int cpuid, if (!irq) return -EINVAL; - if (irq->hw != mapped_irq) { - vgic_put_irq(kvm, irq); - return -EINVAL; - } - spin_lock(&irq->irq_lock); if (!vgic_validate_injection(irq, level)) { @@ -371,12 +379,10 @@ static int vgic_update_irq_pending(struct kvm *kvm, int cpuid, return 0; } - if (irq->config == VGIC_CONFIG_LEVEL) { + if (irq->config == VGIC_CONFIG_LEVEL) irq->line_level = level; - irq->pending = level || irq->soft_pending; - } else { - irq->pending = true; - } + else + irq->pending_latch = true; vgic_queue_irq_unlock(kvm, irq); vgic_put_irq(kvm, irq); @@ -384,32 +390,6 @@ static int vgic_update_irq_pending(struct kvm *kvm, int cpuid, return 0; } -/** - * kvm_vgic_inject_irq - Inject an IRQ from a device to the vgic - * @kvm: The VM structure pointer - * @cpuid: The CPU for PPIs - * @intid: The INTID to inject a new state to. - * @level: Edge-triggered: true: to trigger the interrupt - * false: to ignore the call - * Level-sensitive true: raise the input signal - * false: lower the input signal - * - * The VGIC is not concerned with devices being active-LOW or active-HIGH for - * level-sensitive interrupts. You can think of the level parameter as 1 - * being HIGH and 0 being LOW and all devices being active-HIGH. - */ -int kvm_vgic_inject_irq(struct kvm *kvm, int cpuid, unsigned int intid, - bool level) -{ - return vgic_update_irq_pending(kvm, cpuid, intid, level, false); -} - -int kvm_vgic_inject_mapped_irq(struct kvm *kvm, int cpuid, unsigned int intid, - bool level) -{ - return vgic_update_irq_pending(kvm, cpuid, intid, level, true); -} - int kvm_vgic_map_phys_irq(struct kvm_vcpu *vcpu, u32 virt_irq, u32 phys_irq) { struct vgic_irq *irq = vgic_get_irq(vcpu->kvm, vcpu, virt_irq); @@ -689,7 +669,7 @@ int kvm_vgic_vcpu_pending_irq(struct kvm_vcpu *vcpu) list_for_each_entry(irq, &vgic_cpu->ap_list_head, ap_list) { spin_lock(&irq->irq_lock); - pending = irq->pending && irq->enabled; + pending = irq_is_pending(irq) && irq->enabled; spin_unlock(&irq->irq_lock); if (pending) diff --git a/virt/kvm/arm/vgic/vgic.h b/virt/kvm/arm/vgic/vgic.h index 859f65c6e056..db28f7cadab2 100644 --- a/virt/kvm/arm/vgic/vgic.h +++ b/virt/kvm/arm/vgic/vgic.h @@ -30,13 +30,79 @@ #define vgic_irq_is_sgi(intid) ((intid) < VGIC_NR_SGIS) +#define VGIC_AFFINITY_0_SHIFT 0 +#define VGIC_AFFINITY_0_MASK (0xffUL << VGIC_AFFINITY_0_SHIFT) +#define VGIC_AFFINITY_1_SHIFT 8 +#define VGIC_AFFINITY_1_MASK (0xffUL << VGIC_AFFINITY_1_SHIFT) +#define VGIC_AFFINITY_2_SHIFT 16 +#define VGIC_AFFINITY_2_MASK (0xffUL << VGIC_AFFINITY_2_SHIFT) +#define VGIC_AFFINITY_3_SHIFT 24 +#define VGIC_AFFINITY_3_MASK (0xffUL << VGIC_AFFINITY_3_SHIFT) + +#define VGIC_AFFINITY_LEVEL(reg, level) \ + ((((reg) & VGIC_AFFINITY_## level ##_MASK) \ + >> VGIC_AFFINITY_## level ##_SHIFT) << MPIDR_LEVEL_SHIFT(level)) + +/* + * The Userspace encodes the affinity differently from the MPIDR, + * Below macro converts vgic userspace format to MPIDR reg format. + */ +#define VGIC_TO_MPIDR(val) (VGIC_AFFINITY_LEVEL(val, 0) | \ + VGIC_AFFINITY_LEVEL(val, 1) | \ + VGIC_AFFINITY_LEVEL(val, 2) | \ + VGIC_AFFINITY_LEVEL(val, 3)) + +/* + * As per Documentation/virtual/kvm/devices/arm-vgic-v3.txt, + * below macros are defined for CPUREG encoding. + */ +#define KVM_REG_ARM_VGIC_SYSREG_OP0_MASK 0x000000000000c000 +#define KVM_REG_ARM_VGIC_SYSREG_OP0_SHIFT 14 +#define KVM_REG_ARM_VGIC_SYSREG_OP1_MASK 0x0000000000003800 +#define KVM_REG_ARM_VGIC_SYSREG_OP1_SHIFT 11 +#define KVM_REG_ARM_VGIC_SYSREG_CRN_MASK 0x0000000000000780 +#define KVM_REG_ARM_VGIC_SYSREG_CRN_SHIFT 7 +#define KVM_REG_ARM_VGIC_SYSREG_CRM_MASK 0x0000000000000078 +#define KVM_REG_ARM_VGIC_SYSREG_CRM_SHIFT 3 +#define KVM_REG_ARM_VGIC_SYSREG_OP2_MASK 0x0000000000000007 +#define KVM_REG_ARM_VGIC_SYSREG_OP2_SHIFT 0 + +#define KVM_DEV_ARM_VGIC_SYSREG_MASK (KVM_REG_ARM_VGIC_SYSREG_OP0_MASK | \ + KVM_REG_ARM_VGIC_SYSREG_OP1_MASK | \ + KVM_REG_ARM_VGIC_SYSREG_CRN_MASK | \ + KVM_REG_ARM_VGIC_SYSREG_CRM_MASK | \ + KVM_REG_ARM_VGIC_SYSREG_OP2_MASK) + +static inline bool irq_is_pending(struct vgic_irq *irq) +{ + if (irq->config == VGIC_CONFIG_EDGE) + return irq->pending_latch; + else + return irq->pending_latch || irq->line_level; +} + struct vgic_vmcr { u32 ctlr; u32 abpr; u32 bpr; u32 pmr; + /* Below member variable are valid only for GICv3 */ + u32 grpen0; + u32 grpen1; +}; + +struct vgic_reg_attr { + struct kvm_vcpu *vcpu; + gpa_t addr; }; +int vgic_v3_parse_attr(struct kvm_device *dev, struct kvm_device_attr *attr, + struct vgic_reg_attr *reg_attr); +int vgic_v2_parse_attr(struct kvm_device *dev, struct kvm_device_attr *attr, + struct vgic_reg_attr *reg_attr); +const struct vgic_register_region * +vgic_get_mmio_region(struct kvm_vcpu *vcpu, struct vgic_io_device *iodev, + gpa_t addr, int len); struct vgic_irq *vgic_get_irq(struct kvm *kvm, struct kvm_vcpu *vcpu, u32 intid); void vgic_put_irq(struct kvm *kvm, struct vgic_irq *irq); @@ -89,9 +155,24 @@ bool vgic_has_its(struct kvm *kvm); int kvm_vgic_register_its_device(void); void vgic_enable_lpis(struct kvm_vcpu *vcpu); int vgic_its_inject_msi(struct kvm *kvm, struct kvm_msi *msi); - +int vgic_v3_has_attr_regs(struct kvm_device *dev, struct kvm_device_attr *attr); +int vgic_v3_dist_uaccess(struct kvm_vcpu *vcpu, bool is_write, + int offset, u32 *val); +int vgic_v3_redist_uaccess(struct kvm_vcpu *vcpu, bool is_write, + int offset, u32 *val); +int vgic_v3_cpu_sysregs_uaccess(struct kvm_vcpu *vcpu, bool is_write, + u64 id, u64 *val); +int vgic_v3_has_cpu_sysregs_attr(struct kvm_vcpu *vcpu, bool is_write, u64 id, + u64 *reg); +int vgic_v3_line_level_info_uaccess(struct kvm_vcpu *vcpu, bool is_write, + u32 intid, u64 *val); int kvm_register_vgic_device(unsigned long type); +void vgic_set_vmcr(struct kvm_vcpu *vcpu, struct vgic_vmcr *vmcr); +void vgic_get_vmcr(struct kvm_vcpu *vcpu, struct vgic_vmcr *vmcr); int vgic_lazy_init(struct kvm *kvm); int vgic_init(struct kvm *kvm); +int vgic_debug_init(struct kvm *kvm); +int vgic_debug_destroy(struct kvm *kvm); + #endif diff --git a/virt/kvm/kvm_main.c b/virt/kvm/kvm_main.c index 482612b4e496..cc4d6e0dd2a2 100644 --- a/virt/kvm/kvm_main.c +++ b/virt/kvm/kvm_main.c @@ -506,11 +506,6 @@ static struct kvm_memslots *kvm_alloc_memslots(void) if (!slots) return NULL; - /* - * Init kvm generation close to the maximum to easily test the - * code of handling generation number wrap-around. - */ - slots->generation = -150; for (i = 0; i < KVM_MEM_SLOTS_NUM; i++) slots->id_to_index[i] = slots->memslots[i].id = i; @@ -641,9 +636,16 @@ static struct kvm *kvm_create_vm(unsigned long type) r = -ENOMEM; for (i = 0; i < KVM_ADDRESS_SPACE_NUM; i++) { - kvm->memslots[i] = kvm_alloc_memslots(); - if (!kvm->memslots[i]) + struct kvm_memslots *slots = kvm_alloc_memslots(); + if (!slots) goto out_err_no_srcu; + /* + * Generations must be different for each address space. + * Init kvm generation close to the maximum to easily test the + * code of handling generation number wrap-around. + */ + slots->generation = i * 2 - 150; + rcu_assign_pointer(kvm->memslots[i], slots); } if (init_srcu_struct(&kvm->srcu)) @@ -870,8 +872,14 @@ static struct kvm_memslots *install_new_memslots(struct kvm *kvm, * Increment the new memslot generation a second time. This prevents * vm exits that race with memslot updates from caching a memslot * generation that will (potentially) be valid forever. + * + * Generations must be unique even across address spaces. We do not need + * a global counter for that, instead the generation space is evenly split + * across address spaces. For example, with two address spaces, address + * space 0 will use generations 0, 4, 8, ... while * address space 1 will + * use generations 2, 6, 10, 14, ... */ - slots->generation++; + slots->generation += KVM_ADDRESS_SPACE_NUM * 2 - 1; kvm_arch_memslots_updated(kvm, slots); @@ -1094,37 +1102,31 @@ int kvm_get_dirty_log(struct kvm *kvm, { struct kvm_memslots *slots; struct kvm_memory_slot *memslot; - int r, i, as_id, id; + int i, as_id, id; unsigned long n; unsigned long any = 0; - r = -EINVAL; as_id = log->slot >> 16; id = (u16)log->slot; if (as_id >= KVM_ADDRESS_SPACE_NUM || id >= KVM_USER_MEM_SLOTS) - goto out; + return -EINVAL; slots = __kvm_memslots(kvm, as_id); memslot = id_to_memslot(slots, id); - r = -ENOENT; if (!memslot->dirty_bitmap) - goto out; + return -ENOENT; n = kvm_dirty_bitmap_bytes(memslot); for (i = 0; !any && i < n/sizeof(long); ++i) any = memslot->dirty_bitmap[i]; - r = -EFAULT; if (copy_to_user(log->dirty_bitmap, memslot->dirty_bitmap, n)) - goto out; + return -EFAULT; if (any) *is_dirty = 1; - - r = 0; -out: - return r; + return 0; } EXPORT_SYMBOL_GPL(kvm_get_dirty_log); @@ -1156,24 +1158,22 @@ int kvm_get_dirty_log_protect(struct kvm *kvm, { struct kvm_memslots *slots; struct kvm_memory_slot *memslot; - int r, i, as_id, id; + int i, as_id, id; unsigned long n; unsigned long *dirty_bitmap; unsigned long *dirty_bitmap_buffer; - r = -EINVAL; as_id = log->slot >> 16; id = (u16)log->slot; if (as_id >= KVM_ADDRESS_SPACE_NUM || id >= KVM_USER_MEM_SLOTS) - goto out; + return -EINVAL; slots = __kvm_memslots(kvm, as_id); memslot = id_to_memslot(slots, id); dirty_bitmap = memslot->dirty_bitmap; - r = -ENOENT; if (!dirty_bitmap) - goto out; + return -ENOENT; n = kvm_dirty_bitmap_bytes(memslot); @@ -1202,14 +1202,9 @@ int kvm_get_dirty_log_protect(struct kvm *kvm, } spin_unlock(&kvm->mmu_lock); - - r = -EFAULT; if (copy_to_user(log->dirty_bitmap, dirty_bitmap_buffer, n)) - goto out; - - r = 0; -out: - return r; + return -EFAULT; + return 0; } EXPORT_SYMBOL_GPL(kvm_get_dirty_log_protect); #endif @@ -1937,10 +1932,10 @@ int kvm_vcpu_write_guest(struct kvm_vcpu *vcpu, gpa_t gpa, const void *data, } EXPORT_SYMBOL_GPL(kvm_vcpu_write_guest); -int kvm_gfn_to_hva_cache_init(struct kvm *kvm, struct gfn_to_hva_cache *ghc, - gpa_t gpa, unsigned long len) +static int __kvm_gfn_to_hva_cache_init(struct kvm_memslots *slots, + struct gfn_to_hva_cache *ghc, + gpa_t gpa, unsigned long len) { - struct kvm_memslots *slots = kvm_memslots(kvm); int offset = offset_in_page(gpa); gfn_t start_gfn = gpa >> PAGE_SHIFT; gfn_t end_gfn = (gpa + len - 1) >> PAGE_SHIFT; @@ -1950,7 +1945,7 @@ int kvm_gfn_to_hva_cache_init(struct kvm *kvm, struct gfn_to_hva_cache *ghc, ghc->gpa = gpa; ghc->generation = slots->generation; ghc->len = len; - ghc->memslot = gfn_to_memslot(kvm, start_gfn); + ghc->memslot = __gfn_to_memslot(slots, start_gfn); ghc->hva = gfn_to_hva_many(ghc->memslot, start_gfn, NULL); if (!kvm_is_error_hva(ghc->hva) && nr_pages_needed <= 1) { ghc->hva += offset; @@ -1960,7 +1955,7 @@ int kvm_gfn_to_hva_cache_init(struct kvm *kvm, struct gfn_to_hva_cache *ghc, * verify that the entire region is valid here. */ while (start_gfn <= end_gfn) { - ghc->memslot = gfn_to_memslot(kvm, start_gfn); + ghc->memslot = __gfn_to_memslot(slots, start_gfn); ghc->hva = gfn_to_hva_many(ghc->memslot, start_gfn, &nr_pages_avail); if (kvm_is_error_hva(ghc->hva)) @@ -1972,22 +1967,29 @@ int kvm_gfn_to_hva_cache_init(struct kvm *kvm, struct gfn_to_hva_cache *ghc, } return 0; } -EXPORT_SYMBOL_GPL(kvm_gfn_to_hva_cache_init); -int kvm_write_guest_offset_cached(struct kvm *kvm, struct gfn_to_hva_cache *ghc, - void *data, int offset, unsigned long len) +int kvm_vcpu_gfn_to_hva_cache_init(struct kvm_vcpu *vcpu, struct gfn_to_hva_cache *ghc, + gpa_t gpa, unsigned long len) { - struct kvm_memslots *slots = kvm_memslots(kvm); + struct kvm_memslots *slots = kvm_vcpu_memslots(vcpu); + return __kvm_gfn_to_hva_cache_init(slots, ghc, gpa, len); +} +EXPORT_SYMBOL_GPL(kvm_vcpu_gfn_to_hva_cache_init); + +int kvm_vcpu_write_guest_offset_cached(struct kvm_vcpu *vcpu, struct gfn_to_hva_cache *ghc, + void *data, int offset, unsigned long len) +{ + struct kvm_memslots *slots = kvm_vcpu_memslots(vcpu); int r; gpa_t gpa = ghc->gpa + offset; BUG_ON(len + offset > ghc->len); if (slots->generation != ghc->generation) - kvm_gfn_to_hva_cache_init(kvm, ghc, ghc->gpa, ghc->len); + __kvm_gfn_to_hva_cache_init(slots, ghc, ghc->gpa, ghc->len); if (unlikely(!ghc->memslot)) - return kvm_write_guest(kvm, gpa, data, len); + return kvm_vcpu_write_guest(vcpu, gpa, data, len); if (kvm_is_error_hva(ghc->hva)) return -EFAULT; @@ -1999,28 +2001,28 @@ int kvm_write_guest_offset_cached(struct kvm *kvm, struct gfn_to_hva_cache *ghc, return 0; } -EXPORT_SYMBOL_GPL(kvm_write_guest_offset_cached); +EXPORT_SYMBOL_GPL(kvm_vcpu_write_guest_offset_cached); -int kvm_write_guest_cached(struct kvm *kvm, struct gfn_to_hva_cache *ghc, - void *data, unsigned long len) +int kvm_vcpu_write_guest_cached(struct kvm_vcpu *vcpu, struct gfn_to_hva_cache *ghc, + void *data, unsigned long len) { - return kvm_write_guest_offset_cached(kvm, ghc, data, 0, len); + return kvm_vcpu_write_guest_offset_cached(vcpu, ghc, data, 0, len); } -EXPORT_SYMBOL_GPL(kvm_write_guest_cached); +EXPORT_SYMBOL_GPL(kvm_vcpu_write_guest_cached); -int kvm_read_guest_cached(struct kvm *kvm, struct gfn_to_hva_cache *ghc, - void *data, unsigned long len) +int kvm_vcpu_read_guest_cached(struct kvm_vcpu *vcpu, struct gfn_to_hva_cache *ghc, + void *data, unsigned long len) { - struct kvm_memslots *slots = kvm_memslots(kvm); + struct kvm_memslots *slots = kvm_vcpu_memslots(vcpu); int r; BUG_ON(len > ghc->len); if (slots->generation != ghc->generation) - kvm_gfn_to_hva_cache_init(kvm, ghc, ghc->gpa, ghc->len); + __kvm_gfn_to_hva_cache_init(slots, ghc, ghc->gpa, ghc->len); if (unlikely(!ghc->memslot)) - return kvm_read_guest(kvm, ghc->gpa, data, len); + return kvm_vcpu_read_guest(vcpu, ghc->gpa, data, len); if (kvm_is_error_hva(ghc->hva)) return -EFAULT; @@ -2031,7 +2033,7 @@ int kvm_read_guest_cached(struct kvm *kvm, struct gfn_to_hva_cache *ghc, return 0; } -EXPORT_SYMBOL_GPL(kvm_read_guest_cached); +EXPORT_SYMBOL_GPL(kvm_vcpu_read_guest_cached); int kvm_clear_guest_page(struct kvm *kvm, gfn_t gfn, int offset, int len) { @@ -3133,10 +3135,9 @@ static long kvm_vm_compat_ioctl(struct file *filp, struct compat_kvm_dirty_log compat_log; struct kvm_dirty_log log; - r = -EFAULT; if (copy_from_user(&compat_log, (void __user *)arg, sizeof(compat_log))) - goto out; + return -EFAULT; log.slot = compat_log.slot; log.padding1 = compat_log.padding1; log.padding2 = compat_log.padding2; @@ -3148,8 +3149,6 @@ static long kvm_vm_compat_ioctl(struct file *filp, default: r = kvm_vm_ioctl(filp, ioctl, arg); } - -out: return r; } #endif |