| Age | Commit message (Collapse) | Author |
|---|
|
git://git.kernel.org/pub/scm/linux/kernel/git/geert/linux-m68k
Pull m68k updates from Geert Uytterhoeven:
"Summary:
- __put_user_unaligned may/will be used by btrfs
- m68k part of a global cleanup"
* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/geert/linux-m68k:
m68k: Remove deprecated IRQF_DISABLED
m68k/m68knommu: Implement __get_user_unaligned/__put_user_unaligned()
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/deller/parisc-linux
Pull parisc update from Helge Deller:
- a bugfix for sticon (parisc text console driver) to not crash the
64bit kernel on machines with more than 4GB RAM
- added kernel audit support
- made udelay() implementation SMP-safe
- "make install" now does not depend on vmlinux
- added defconfigs for 32- and 64-kernels
* 'parisc-3.13' of git://git.kernel.org/pub/scm/linux/kernel/git/deller/parisc-linux:
parisc: add generic 32- and 64-bit defconfigs
parisc: sticon - unbreak on 64bit kernel
parisc: signal fixup - SIGBUS vs. SIGSEGV
parisc: implement full version of access_ok()
parisc: correctly display number of active CPUs
parisc: do not count IPI calls twice
parisc: make udelay() SMP-safe
parisc: remove duplicate define
parisc: make "make install" not depend on vmlinux
parisc: add kernel audit feature
parisc: provide macro to create exception table entries
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/oleg/misc into perf/core
Pull uprobes fixes from Oleg Nesterov.
Signed-off-by: Ingo Molnar <mingo@kernel.org>
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/arm/arm-soc
Pull ARM SoC DT updates from Olof Johansson:
"Most of this branch consists of updates, additions and general churn
of the device tree source files in the kernel (arch/arm/boot/dts).
Besides that, there are a few things to point out:
- Lots of platform conversion on OMAP2+, with removal of old board
files for various platforms.
- Final conversion of a bunch of ux500 (ST-Ericsson) platforms as
well
- Some updates to pinctrl and other subsystems. Most of these are
for DT-enablement of the various platforms and acks have been
collected"
* tag 'dt-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/arm/arm-soc: (385 commits)
ARM: dts: bcm11351: Use GIC/IRQ defines for sdio interrupts
ARM: dts: bcm: Add missing UARTs for bcm11351 (bcm281xx)
ARM: dts: bcm281xx: Add card detect GPIO
ARM: dts: rename ARCH_BCM to ARCH_BCM_MOBILE (dt)
ARM: bcm281xx: Add device node for the GPIO controller
ARM: mvebu: Add Netgear ReadyNAS 104 board
ARM: tegra: fix Tegra114 IOMMU register address
ARM: kirkwood: add support for OpenBlocks A7 platform
ARM: dts: omap4-panda: add DPI pinmuxing
ARM: dts: AM33xx: Add RNG node
ARM: dts: AM33XX: Add hwspinlock node
ARM: dts: OMAP5: Add hwspinlock node
ARM: dts: OMAP4: Add hwspinlock node
ARM: dts: use 'status' property for PCIe nodes
ARM: dts: sirf: add missed address-cells and size-cells for prima2 I2C
ARM: dts: sirf: add missed cell, cs and dma channel for SPI nodes
ARM: dts: sirf: add missed graphics2d iobg in atlas6 dts
ARM: dts: sirf: add missed chhifbg node in prima2 and atlas6 dts
ARM: dts: sirf: add missed memcontrol-monitor node in prima2 and atlas6 dts
ARM: mvebu: Add the core-divider clock to Armada 370/XP
...
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/arm/arm-soc
Pull ARM driver updates from Olof Johansson:
"Updates of SoC-near drivers and other driver updates that makes more
sense to take through our tree. In this case it's involved:
- Some Davinci driver updates that has required corresponding
platform code changes (gpio mostly)
- CCI bindings and a few driver updates
- Marvell mvebu patches for PCI MSI support (could have gone through
the PCI tree for this release, but they were acked by Bjorn for
3.12 so we kept them through arm-soc).
- Marvell dove switch-over to DT-based PCIe configuration
- Misc updates for Samsung platform dmaengine drivers"
* tag 'drivers-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/arm/arm-soc: (32 commits)
ARM: S3C24XX: add dma pdata for s3c2410, s3c2440 and s3c2442
dmaengine: s3c24xx-dma: add support for the s3c2410 type of controller
ARM: S3C24XX: Fix possible dma selection warning
PCI: mvebu: make local functions static
PCI: mvebu: add I/O access wrappers
PCI: mvebu: Dynamically detect if the PEX link is up to enable hot plug
ARM: mvebu: fix gated clock documentation
ARM: dove: remove legacy pcie and clock init
ARM: dove: switch to DT probed mbus address windows
ARM: SAMSUNG: set s3c24xx_dma_filter for s3c64xx-spi0 device
ARM: S3C24XX: add platform-devices for new dma driver for s3c2412 and s3c2443
dmaengine: add driver for Samsung s3c24xx SoCs
ARM: S3C24XX: number the dma clocks
PCI: mvebu: add support for Marvell Dove SoCs
PCI: mvebu: add support for reset on GPIO
PCI: mvebu: remove subsys_initcall
PCI: mvebu: increment nports only for registered ports
PCI: mvebu: move clock enable before register access
PCI: mvebu: add support for MSI
irqchip: armada-370-xp: implement MSI support
...
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/arm/arm-soc
Pull ARM SoC board updates from Olof Johansson:
"Board-related updates. This branch is getting smaller and smaller,
which is the whole idea so that's reassuring.
Right now by far most of the code is related to shmobile updates, and
they are now switching over to removal of board code and migration to
multiplatform, so we'll see their board code base shrink in the near
future too, I hope.
In addition to that is some defconfig updates, some display updates
for OMAP and a bit of new board support for Rockchip boards"
* tag 'boards-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/arm/arm-soc: (56 commits)
ARM: rockchip: add support for rk3188 and Radxa Rock board
ARM: rockchip: add dts for bqcurie2 tablet
ARM: rockchip: enable arm-global-timer
ARM: rockchip: move shared dt properties to common source file
ARM: OMAP2+: display: Create omap_vout device inside omap_display_init
ARM: OMAP2+: display: Create omapvrfb and omapfb devices inside omap_display_init
ARM: OMAP2+: display: Create omapdrm device inside omap_display_init
ARM: OMAP2+: drm: Don't build device for DMM
ARM: tegra: defconfig updates
RX-51: Add support for OMAP3 ROM Random Number Generator
ARM: OMAP3: RX-51: ARM errata 430973 workaround
ARM: OMAP3: Add secure function omap_smc3() which calling instruction smc #1
ARM: shmobile: marzen: enable INTC IRQ
ARM: shmobile: bockw: add SMSC support on reference
ARM: shmobile: Use SMP on Koelsch
ARM: shmobile: Remove KZM9D reference DTS
ARM: shmobile: Let KZM9D multiplatform boot with KZM9D DTB
ARM: shmobile: Remove non-multiplatform KZM9D reference support
ARM: shmobile: Use KZM9D without reference for multiplatform
ARM: shmobile: Sync KZM9D DTS with KZM9D reference DTS
...
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/arm/arm-soc
Pull ARM SoC platform changes from Olof Johansson:
"New and updated SoC support. Among the things new for this release
are:
- More support for the AM33xx platforms from TI
- Tegra 124 support, and some updates to older tegra families as well
- imx cleanups and updates across the board
- A rename of Broadcom's Mobile platforms which were introduced as
ARCH_BCM, and turned out to be too broad a name. New name is
ARCH_BCM_MOBILE.
- A whole bunch of updates and fixes for integrator, making the
platform code more modern and switches over to DT-only booting.
- Support for two new Renesas shmobile chipsets. Next up for them is
more work on consolidation instead of introduction of new
non-multiplatform SoCs, we're all looking forward to that!
- Misc cleanups for older Samsung platforms, some Allwinner updates,
etc"
* tag 'soc-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/arm/arm-soc: (159 commits)
ARM: bcm281xx: Add ARCH_BCM_MOBILE to bcm config
ARM: bcm_defconfig: Run "make savedefconfig"
ARM: bcm281xx: Add ARCH Timers to config
rename ARCH_BCM to ARCH_BCM_MOBILE (mach-bcm)
ARM: vexpress: Enable platform-specific options in defconfig
ARM: vexpress: Make defconfig work again
ARM: sunxi: remove .init_time hooks
ARM: imx: enable suspend for imx6sl
ARM: imx: ensure dsm_request signal is not asserted when setting LPM
ARM: imx6q: call WB and RBC configuration from imx6q_pm_enter()
ARM: imx6q: move low-power code out of clock driver
ARM: imx: drop extern with function prototypes in common.h
ARM: imx: reset core along with enable/disable operation
ARM: imx: do not return from imx_cpu_die() call
ARM: imx_v6_v7_defconfig: Select CONFIG_PROVE_LOCKING
ARM: imx_v6_v7_defconfig: Enable LEDS_GPIO related options
ARM: mxs_defconfig: Turn off CONFIG_DEBUG_GPIO
ARM: imx: replace imx6q_restart() with mxc_restart()
ARM: mach-imx: mm-imx5: Retrieve iomuxc base address from dt
ARM: mach-imx: mm-imx5: Retrieve tzic base address from dt
...
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/arm/arm-soc
Pull ARM SoC cleanups from Olof Johansson:
"This branch contains code cleanups, moves and removals for 3.13.
Qualcomm msm targets had a bunch of code removal for legacy non-DT
platforms. Nomadik saw more device tree conversions and cleanup of
old code. Tegra has some code refactoring, etc.
One longish patch series from Sebastian Hasselbarth changes the
init_time hooks and tries to use a generic implementation for most
platforms, since they were all doing more or less the same things.
Finally the "shark" platform is removed in this release. It's been
abandoned for a while and nobody seems to care enough to keep it
around. If someone comes along and wants to resurrect it, the removal
can easily be reverted and code brought back.
Beyond this, mostly a bunch of removals of stale content across the
board, etc"
* tag 'cleanup-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/arm/arm-soc: (79 commits)
ARM: gemini: convert to GENERIC_CLOCKEVENTS
ARM: EXYNOS: remove CONFIG_MACH_EXYNOS[4, 5]_DT config options
ARM: OMAP3: control: add API for setting IVA bootmode
ARM: OMAP3: CM/control: move CM scratchpad save to CM driver
ARM: OMAP3: McBSP: do not access CM register directly
ARM: OMAP3: clock: add API to enable/disable autoidle for a single clock
ARM: OMAP2: CM/PM: remove direct register accesses outside CM code
MAINTAINERS: Add patterns for DTS files for AT91
ARM: at91: remove init_machine() as default is suitable
ARM: at91/dt: split sama5d3 peripheral definitions
ARM: at91/dt: split sam9x5 peripheral definitions
ARM: Remove temporary sched_clock.h header
ARM: clps711x: Use linux/sched_clock.h
MAINTAINERS: Add DTS files to patterns for Samsung platform
ARM: EXYNOS: remove unnecessary header inclusions from exynos4/5 dt machine file
ARM: tegra: fix ARCH_TEGRA_114_SOC select sort order
clk: nomadik: fix missing __init on nomadik_src_init
ARM: drop explicit selection of HAVE_CLK and CLKDEV_LOOKUP
ARM: S3C64XX: Kill CONFIG_PLAT_S3C64XX
ASoC: samsung: Use CONFIG_ARCH_S3C64XX to check for S3C64XX support
...
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/arm/arm-soc
Pull ARM SoC low-priority fixes from Olof Johansson:
"A set of fixes for various platforms that weren't considered bad
enough to include in 3.12 (nor -stable). Mostly simple typo fixes,
etc"
* tag 'fixes-nc-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/arm/arm-soc:
ARM: OMAP2+: irq, AM33XX add missing register check
ARM: OMAP2+: wakeupgen: AM43x adaptation
ARM: OMAP1: Fix a bunch of GPIO related section warnings after initdata got corrected
ARM: dts: fix PL330 MDMA1 address in DT for Universal C210 board
ARM: dts: Work around lack of cpufreq regulator lookup for exynos4210-origen and trats boards
ARM: dts: Fix typo earlyprintk in exynos5440-sd5v1 and ssdk5440 boards
ARM: dts: Correct typo in use of samsung,pin-drv for exynos5250
ARM: rockchip: remove obsolete rockchip,config properties
ARM: rockchip: fix wrong use of non-existent CONFIG_LOCAL_TIMERS
ARM: mach-omap1: Fix omap1510_fpga_init_irq() implicit declarations.
ARM: OMAP1: fix incorrect placement of __initdata tag
ARM: OMAP: remove deprecated IRQF_DISABLED
ARM: OMAP2+: throw the die id into the entropy pool
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/cmarinas/linux-aarch64
Pull ARM64 update from Catalin Marinas:
"Main features:
- Ticket-based spinlock implementation and lockless lockref support
- Big endian support
- CPU hotplug support, currently for PSCI (Power State Coordination
Interface) capable firmware
- Virtual address space extended to 42-bit in the 64K page
configuration (maximum VA space with 2 levels of page tables)
- Compat (AArch32) kuser helpers updated to ARMv8 (make use of
load-acquire/store-release instructions)
- Code cleanup, defconfig update and minor fixes"
* tag 'arm64-upstream' of git://git.kernel.org/pub/scm/linux/kernel/git/cmarinas/linux-aarch64: (43 commits)
ARM64: /proc/interrupts: display IPIs of online CPUs only
arm64: locks: Remove CONFIG_GENERIC_LOCKBREAK
arm64: KVM: vgic: byteswap GICv2 access on world switch if BE
arm64: KVM: initialize HYP mode following the kernel endianness
arm64: compat: Clear the IT state independent of the 32-bit ARM or Thumb-2 mode
arm64: Use 42-bit address space with 64K pages
arm64: module: ensure instruction is little-endian before manipulation
arm64: defconfig: Enable CONFIG_PREEMPT by default
arm64: fix access to preempt_count from assembly code
arm64: move enabling of GIC before CPUs are set online
arm64: use generic RW_DATA_SECTION macro in linker script
arm64: Slightly improve the warning on CPU0 enable-method
ARM64: simplify cpu_read_bootcpu_ops using OF/DT helper
ARM64: DT: define ARM64 specific arch_match_cpu_phys_id
arm64: allow ioremap_cache() to use existing RAM mappings
arm64: update 32-bit kuser helpers to ARMv8
arm64: perf: fix event number mask
arm64: kconfig: allow CPU_BIG_ENDIAN to be selected
arm64: Fix the endianness of arch_spinlock_t
arm64: big-endian: write CPU holding pen address as LE
...
|
|
Fixes a suspicious rcu derference warning.
Cc: Florent Fourcot <florent.fourcot@enst-bretagne.fr>
Signed-off-by: Hannes Frederic Sowa <hannes@stressinduktion.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
This is to avoid very silly Kconfig dependencies for modules
using this routine.
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
The max_vfs parameter has a limit of 63 and silently fails (adding 0 vfs) when
it is out of range. This patch adds a warning so that the user knows something
went wrong. Also, this patch moves the warning in ixgbe_enable_sriov() to where
max_vfs is checked, so that even an out of range value will show the deprecated
warning. Previously, an out of range parameter didn't even warn the user to use
the new sysfs interface instead.
Signed-off-by: Jacob Keller <jacob.e.keller@intel.com>
Tested-by: Phil Schmitt <phillip.j.schmitt@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
This patch fixes multiple problems in the link modes display in ethtool.
Newer parts have more complicated methods to determine actual link
capabilities. Older parts cannot communicate with their SFP modules.
Finally, all the available defines are not displayed by ethtool. This
updates the link modes to be as accurate as possible depending on what data
is available to the driver at any given time.
Signed-off-by: Carolyn Wyborny <carolyn.wyborny@intel.com>
Tested-by: Jeff Pieper <jeffrey.e.pieper@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
Pushing original fragments through causes several problems. For example
for matching, frags may not be matched correctly. Take following
example:
<example>
On HOSTA do:
ip6tables -I INPUT -p icmpv6 -j DROP
ip6tables -I INPUT -p icmpv6 -m icmp6 --icmpv6-type 128 -j ACCEPT
and on HOSTB you do:
ping6 HOSTA -s2000 (MTU is 1500)
Incoming echo requests will be filtered out on HOSTA. This issue does
not occur with smaller packets than MTU (where fragmentation does not happen)
</example>
As was discussed previously, the only correct solution seems to be to use
reassembled skb instead of separete frags. Doing this has positive side
effects in reducing sk_buff by one pointer (nfct_reasm) and also the reams
dances in ipvs and conntrack can be removed.
Future plan is to remove net/ipv6/netfilter/nf_conntrack_reasm.c
entirely and use code in net/ipv6/reassembly.c instead.
Signed-off-by: Jiri Pirko <jiri@resnulli.us>
Acked-by: Julian Anastasov <ja@ssi.bg>
Signed-off-by: Marcelo Ricardo Leitner <mleitner@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
If reassembled packet would fit into outdev MTU, it is not fragmented
according the original frag size and it is send as single big packet.
The second case is if skb is gso. In that case fragmentation does not happen
according to the original frag size.
This patch fixes these.
Signed-off-by: Jiri Pirko <jiri@resnulli.us>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
o Changes from v1
Use find_next(_zero)_bit suggested by jg.kim
When f2fs issues discard command, if segment is contiguous,
let's issue more large segment to gather adjacent segments.
** blktrace **
179,1 0 5859 42.619023770 971 C D 131072 + 2097152 [0]
179,1 0 33665 108.840475468 971 C D 2228224 + 2494464 [0]
179,1 0 33671 109.131616427 971 C D 14909440 + 344064 [0]
179,1 0 33677 109.137100677 971 C D 15261696 + 4096 [0]
Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/steve/gfs2-3.0-nmw
Pull gfs2 updates from Steven Whitehouse:
"The main feature of interest this time is quota updates. There are
some clean ups and some patches to use the new generic lru list code.
There is still plenty of scope for some further changes in due course -
faster lookups of quota structures is very much on the todo list.
Also, a start has been made towards the more tricky issue of using the
generic lru code with glocks, but that will have to be completed in a
subsequent merge window.
The other, more minor feature, is that there have been a number of
performance patches which relate to block allocation. In particular
they will improve performance when the disk is nearly full"
* tag 'gfs2-merge-window' of git://git.kernel.org/pub/scm/linux/kernel/git/steve/gfs2-3.0-nmw:
GFS2: Use generic list_lru for quota
GFS2: Rename quota qd_lru_lock qd_lock
GFS2: Use reflink for quota data cache
GFS2: Use lockref for glocks
GFS2: Protect quota sync generation
GFS2: Inline qd_trylock into gfs2_quota_unlock
GFS2: Make two similar quota code fragments into a function
GFS2: Remove obsolete quota tunable
GFS2: Move gfs2_icbit_munge into quota.c
GFS2: Speed up starting point selection for block allocation
GFS2: Add allocation parameters structure
GFS2: Clean up reservation removal
GFS2: fix dentry leaks
GFS2: new function gfs2_rbm_incr
GFS2: Introduce rbm field bii
GFS2: Do not reset flags on active reservations
GFS2: introduce bi_blocks for optimization
GFS2: optimize rbm_from_block wrt bi_start
GFS2: d_splice_alias() can't return error
|
|
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
|
|
Lennart says: "I haven't been able to spend time on mv643xx_eth for a
while now, so if you want to take over maintainership, I'd be fine with
that."
Signed-off-by: Sebastian Hesselbarth <sebastian.hesselbarth@gmail.com>
Acked-by: Lennert Buytenhek <buytenh@wantstofly.org>
Acked-by: Jason Cooper <jason@lakedaemon.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
With psched_ratecfg_precompute(), tbf can deal with 64bit rates.
Add two new attributes so that tc can use them to break the 32bit
limit.
Signed-off-by: Yang Yingliang <yangyingliang@huawei.com>
Suggested-by: Sergei Shtylyov <sergei.shtylyov@cogentembedded.com>
Acked-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
1. copy_insn() doesn't look very nice, all calculations are
confusing and it is not immediately clear why do we read
the 2nd page first.
2. The usage of inode->i_size is wrong on 32-bit machines.
3. "Instruction at end of binary" logic is simply wrong, it
doesn't handle the case when uprobe->offset > inode->i_size.
In this case "bytes" overflows, and __copy_insn() writes to
the memory outside of uprobe->arch.insn.
Yes, uprobe_register() checks i_size_read(), but this file
can be truncated after that. All i_size checks are racy, we
do this only to catch the obvious mistakes.
Change copy_insn() to call __copy_insn() in a loop, simplify
and fix the bytes/nbytes calculations.
Note: we do not care if we read extra bytes after inode->i_size
if we got the valid page. This is fine because the task gets the
same page after page-fault, and arch_uprobe_analyze_insn() can't
know how many bytes were actually read anyway.
Signed-off-by: Oleg Nesterov <oleg@redhat.com>
|
|
Commit aa59c53fd459 "uprobes: Change uprobe_copy_process() to dup
xol_area" has a stupid typo, we need to setup t->utask->vaddr but
the code wrongly uses current->utask.
Even with this bug dup_xol_work() works "in practice", but only
because get_unmapped_area(NULL, TASK_SIZE - PAGE_SIZE) likely
returns the same address every time.
Signed-off-by: Oleg Nesterov <oleg@redhat.com>
|
|
NFSv4 uses leases to guarantee that clients can cache metadata as well
as data.
Cc: Mikulas Patocka <mikulas@artax.karlin.mff.cuni.cz>
Cc: David Howells <dhowells@redhat.com>
Cc: Tyler Hicks <tyhicks@canonical.com>
Cc: Dustin Kirkland <dustin.kirkland@gazzang.com>
Acked-by: Jeff Layton <jlayton@redhat.com>
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
|
|
Cc: Tyler Hicks <tyhicks@canonical.com>
Cc: Dustin Kirkland <dustin.kirkland@gazzang.com>
Acked-by: Jeff Layton <jlayton@redhat.com>
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
|
|
Cc: David Howells <dhowells@redhat.com>
Acked-by: Jeff Layton <jlayton@redhat.com>
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
|
|
We'll need the same logic for rename and link.
Acked-by: Jeff Layton <jlayton@redhat.com>
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
|
|
We need to break delegations on any operation that changes the set of
links pointing to an inode. Start with unlink.
Such operations also hold the i_mutex on a parent directory. Breaking a
delegation may require waiting for a timeout (by default 90 seconds) in
the case of a unresponsive NFS client. To avoid blocking all directory
operations, we therefore drop locks before waiting for the delegation.
The logic then looks like:
acquire locks
...
test for delegation; if found:
take reference on inode
release locks
wait for delegation break
drop reference on inode
retry
It is possible this could never terminate. (Even if we take precautions
to prevent another delegation being acquired on the same inode, we could
get a different inode on each retry.) But this seems very unlikely.
The initial test for a delegation happens after the lock on the target
inode is acquired, but the directory inode may have been acquired
further up the call stack. We therefore add a "struct inode **"
argument to any intervening functions, which we use to pass the inode
back up to the caller in the case it needs a delegation synchronously
broken.
Cc: David Howells <dhowells@redhat.com>
Cc: Tyler Hicks <tyhicks@canonical.com>
Cc: Dustin Kirkland <dustin.kirkland@gazzang.com>
Acked-by: Jeff Layton <jlayton@redhat.com>
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
|
|
We'll be using dentry->d_inode in one more place.
Acked-by: Jeff Layton <jlayton@redhat.com>
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
|
|
Implement NFSv4 delegations at the vfs level using the new FL_DELEG lock
type.
Note nfsd is the only delegation user and is only using read
delegations. Warn on any attempt to set a write delegation for now.
We'll come back to that case later.
Acked-by: Jeff Layton <jlayton@redhat.com>
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
|
|
For now FL_DELEG is just a synonym for FL_LEASE. So this patch doesn't
change behavior.
Next we'll modify break_lease to treat FL_DELEG leases differently, to
account for the fact that NFSv4 delegations should be broken in more
situations than Windows oplocks.
Acked-by: Jeff Layton <jlayton@redhat.com>
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
|
|
A read delegation is used by NFSv4 as a guarantee that a client can
perform local read opens without informing the server.
The open operation takes the last component of the pathname as an
argument, thus is also a lookup operation, and giving the client the
above guarantee means informing the client before we allow anything that
would change the set of names pointing to the inode.
Therefore, we need to break delegations on rename, link, and unlink.
We also need to prevent new delegations from being acquired while one of
these operations is in progress.
We could add some completely new locking for that purpose, but it's
simpler to use the i_mutex, since that's already taken by all the
operations we care about.
The single exception is rename. So, modify rename to take the i_mutex
on the file that is being renamed.
Also fix up lockdep and Documentation/filesystems/directory-locking to
reflect the change.
Acked-by: Jeff Layton <jlayton@redhat.com>
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
|
|
I_MUTEX_QUOTA is now just being used whenever we want to lock two
non-directories. So the name isn't right. I_MUTEX_NONDIR2 isn't
especially elegant but it's the best I could think of.
Also fix some outdated documentation.
Acked-by: Jeff Layton <jlayton@redhat.com>
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
|
|
Reserve I_MUTEX_PARENT and I_MUTEX_CHILD for locking of actual
directories.
(Also I_MUTEX_QUOTA isn't really a meaningful name for this locking
class any more; fixed in a later patch.)
Acked-by: Jeff Layton <jlayton@redhat.com>
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
|
|
We want to do this elsewhere as well.
Also catch any attempts to use it for directories (where this ordering
would conflict with ancestor-first directory ordering in lock_rename).
Cc: Andreas Dilger <adilger.kernel@dilger.ca>
Cc: Dave Chinner <david@fromorbit.com>
Acked-by: Jeff Layton <jlayton@redhat.com>
Acked-by: "Theodore Ts'o" <tytso@mit.edu>
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
|
|
Suppose we're given the filehandle for a directory whose closest
ancestor in the dcache is its Nth ancestor.
The main loop in reconnect_path searches for an IS_ROOT ancestor of
target_dir, reconnects that ancestor to its parent, then recommences the
search for an IS_ROOT ancestor from target_dir.
This behavior is quadratic in N. And there's really no need to restart
the search from target_dir each time: once a directory has been looked
up, it won't become IS_ROOT again. So instead of starting from
target_dir each time, we can continue where we left off.
This simplifies the code and improves performance on very deep directory
heirachies. (I can't think of any reason anyone should need heirarchies
a hundred or more deep, but the performance improvement may be valuable
if only to limit damage in case of abuse.)
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
|
|
Replace another unhelpful acronym.
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
|
|
Also replace 3 easily-confused three-letter acronyms by more helpful
variable names.
Just cleanup, no change in functionality, with one exception: the
dentry_connected() check in the "out_reconnected" case will now only
check the ancestors of the current dentry instead of checking all the
way from target_dir. Since we've already verified connectivity up to
this dentry, that should be sufficient.
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
|
|
Note this counter is now being set to 0 on every pass through the loop,
so it no longer serves any useful purpose.
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
|
|
There are two places here where we could race with a rename or remove:
- We could find the parent, but then be removed or renamed away
from that parent directory before finding our name in that
directory.
- We could find the parent, and find our name in that parent,
but then be renamed or removed before we look ourselves up by
that name in that parent.
In both cases the concurrent rename or remove will take care of
reconnecting the directory that we're currently examining. Our target
directory should then also be connected. Check this and clear
DISCONNECTED in these cases instead of looping around again.
Note: we *do* need to check that this actually happened if we want to be
robust in the face of corrupted filesystems: a corrupted filesystem
could just return a completely wrong parent, and we want to fail with an
error in that case before starting to clear DISCONNECTED on
non-DISCONNECTED filesystems.
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
|
|
Once we've found any connected parent, we know all our parents are
connected--that's true even if there's a concurrent rename. May as well
clear them all at once and be done with it.
Reviewed-by: Cristoph Hellwig <hch@lst.de>
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
|
|
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
|
|
This would indicate a nasty bug in the dcache and has never triggered in
the past 10 years as far as I know.
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
|
|
The DCACHE_NEED_LOOKUP case referred to here was removed with
39e3c9553f34381a1b664c27b0c696a266a5735e "vfs: remove
DCACHE_NEED_LOOKUP".
There are only four real_lookup() callers and all of them pass in an
unhashed dentry just returned from d_alloc.
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
|
|
DCACHE_DISCONNECTED should not be cleared until we're sure the dentry is
connected all the way up to the root of the filesystem. It *shouldn't*
be cleared as soon as the dentry is connected to a parent. That will
cause bugs at least on exportable filesystems.
Acked-by: Christoph Hellwig <hch@infradead.org>
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
|
|
I can't for the life of me see any reason why anyone should care whether
a dentry that is never hooked into the dentry cache would need
DCACHE_DISCONNECTED set.
This originates from 4b936885ab04dc6e0bb0ef35e0e23c1a7364d9e5 "fs:
improve scalability of pseudo filesystems", which probably just made the
false assumption the DCACHE_DISCONNECTED was meant to be set on anything
not connected to a parent somehow.
So this is just confusing. Ideally the only uses of DCACHE_DISCONNECTED
would be in the filehandle-lookup code, which needs it to ensure
dentries are connected into the dentry tree before use.
I left d_alloc_pseudo there even though it's now equivalent to
__d_alloc(), just on the theory the name is better documentation of its
intended use outside dcache.c.
Cc: Nick Piggin <npiggin@kernel.dk>
Acked-by: Christoph Hellwig <hch@infradead.org>
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
|
|
Every hashed dentry is either hashed in the dentry_hashtable, or a
superblock's s_anon list.
__d_drop() assumes it can determine which is the case by checking
DCACHE_DISCONNECTED; this is not true.
It is true that when DCACHE_DISCONNECTED is cleared, the dentry is not
only hashed on dentry_hashtable, but is fully connected to its parents
back to the root.
But the converse is *not* true: fs/exportfs/expfs.c:reconnect_path()
attempts to connect a directory (found by filehandle lookup) back to
root by ascending to parents and performing lookups one at a time. It
does not clear DCACHE_DISCONNECTED until it's done, and that is not at
all an atomic process.
In particular, it is possible for DCACHE_DISCONNECTED to be set on a
dentry which is hashed on the dentry_hashtable.
Instead, use IS_ROOT() to check which hash chain a dentry is on. This
*does* work:
Dentries are hashed only by:
- d_obtain_alias, which adds an IS_ROOT() dentry to sb_anon.
- __d_rehash, called by _d_rehash: hashes to the dentry's
parent, and all callers of _d_rehash appear to have d_parent
set to a "real" parent.
- __d_rehash, called by __d_move: rehashes the moved dentry to
hash chain determined by target, and assigns target's d_parent
to its d_parent, before dropping the dentry's d_lock.
Therefore I believe it's safe for a holder of a dentry's d_lock to
assume that it is hashed on sb_anon if and only if IS_ROOT(dentry) is
true.
I believe the incorrect assumption about DCACHE_DISCONNECTED was
originally introduced by ceb5bdc2d246 "fs: dcache per-bucket dcache hash
locking".
Also add a comment while we're here.
Cc: Nick Piggin <npiggin@kernel.dk>
Acked-by: Christoph Hellwig <hch@infradead.org>
Reviewed-by: NeilBrown <neilb@suse.de>
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
|
|
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
|
|
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
|
|
Symptoms were spurious -ENOENTs on stat of an NFS filesystem from a
32-bit NFS server exporting a very large XFS filesystem, when the
server's cache is cold (so the inodes in question are not in cache).
Reviewed-by: Christoph Hellwig <hch@lst.de>
Reported-by: Trevor Cordes <trevor@tecnopolis.ca>
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
|
