| Age | Commit message (Collapse) | Author |
|---|
|
Enable the list IOCTL to provide lists longer than one page (85 entries).
The list IOCTL now accepts any argument length in page granularity.
It fills the argument up to this length with entries until the list
ends. User space unaware of this enhancement will still receive one page
of data and an uv_rc 0x0100.
Signed-off-by: Steffen Eiden <seiden@linux.ibm.com>
Reviewed-by: Janosch Frank <frankja@linux.ibm.com>
Link: https://lore.kernel.org/r/20241104153609.1361388-1-seiden@linux.ibm.com
Signed-off-by: Janosch Frank <frankja@linux.ibm.com>
Message-ID: <20241104153609.1361388-1-seiden@linux.ibm.com>
Signed-off-by: Heiko Carstens <hca@linux.ibm.com>
|
|
Add a trival phys_to_target_node() implementation which always returns 0 if
CONFIG_NUMA is enabled, since the s390 NUMA implementation only supports
node 0.
This is similar to memory_add_physaddr_to_nid() in order to avoid runtime
warnings.
Signed-off-by: Heiko Carstens <hca@linux.ibm.com>
|
|
Signed-off-by: Heiko Carstens <hca@linux.ibm.com>
|
|
David Hildenbrand says:
====================
virtio-mem: s390 support
Let's finally add s390 support for virtio-mem; my last RFC was sent
4 years ago, and a lot changed in the meantime.
The latest QEMU series is available at [1], which contains some more
details and a usage example on s390 (last patch).
There is not too much in here: The biggest part is querying a new diag(500)
STORAGE_LIMIT hypercall to obtain the proper "max_physmem_end".
The last three patches are not strictly required but certainly nice-to-have.
Note that -- in contrast to standby memory -- virtio-mem memory must be
configured to be automatically onlined as soon as hotplugged. The easiest
approach is using the "memhp_default_state=" kernel parameter or by using
proper udev rules. More details can be found at [2].
I have reviving+upstreaming a systemd service to handle configuring
that on my todo list, but for some reason I keep getting distracted ...
I tested various things, including:
* Various memory hotplug/hotunplug combinations
* Device hotplug/hotunplug
* /proc/iomem output
* reboot
* kexec
* kdump: make sure we properly enter the "kdump mode" in the virtio-mem
driver
kdump support for virtio-mem memory on s390 will be sent out separately.
v2 -> v3
* "s390/kdump: make is_kdump_kernel() consistently return "true" in kdump
environments only"
-> Sent out separately [3]
* "s390/physmem_info: query diag500(STORAGE LIMIT) to support QEMU/KVM memory
devices"
-> No query function for diag500 for now.
-> Update comment above setup_ident_map_size().
-> Optimize/rewrite diag500_storage_limit() [Heiko]
-> Change handling in detect_physmem_online_ranges [Alexander]
-> Improve documentation.
* "s390/sparsemem: provide memory_add_physaddr_to_nid() with CONFIG_NUMA"
-> Added after testing on systems with CONFIG_NUMA=y
v1 -> v2:
* Document the new diag500 subfunction
* Use "s390" instead of "s390x" consistently
[1] https://lkml.kernel.org/r/20241008105455.2302628-1-david@redhat.com
[2] https://virtio-mem.gitlab.io/user-guide/user-guide-linux.html
[3] https://lkml.kernel.org/r/20241023090651.1115507-1-david@redhat.com
====================
Link: https://lore.kernel.org/r/20241025141453.1210600-1-david@redhat.com/
Signed-off-by: Heiko Carstens <hca@linux.ibm.com>
|
|
virtio-mem uses memory_add_physaddr_to_nid() to determine the NID to use
for memory it adds.
We currently fallback to the dummy implementation in mm/numa.c with
CONFIG_NUMA, which will end up triggering an undesired pr_info_once():
Unknown online node for memory at 0x100000000, assuming node 0
On s390, we map all cpus and memory to node 0, so let's add a simple
memory_add_physaddr_to_nid() implementation that does exactly that,
but without complaining.
Signed-off-by: David Hildenbrand <david@redhat.com>
Reviewed-by: Heiko Carstens <hca@linux.ibm.com>
Tested-by: Sumanth Korikkar <sumanthk@linux.ibm.com>
Acked-by: Christian Borntraeger <borntraeger@linux.ibm.com>
Link: https://lore.kernel.org/r/20241025141453.1210600-8-david@redhat.com
Signed-off-by: Heiko Carstens <hca@linux.ibm.com>
|
|
Ever since commit 421c175c4d609 ("[S390] Add support for memory hot-add.")
we've been using a section size of 256 MiB on s390 and 32 MiB on s390.
Before that, we were using a section size of 32 MiB on both
architectures.
Now that we have a new mechanism to expose additional memory to a VM --
virtio-mem -- reduce the section size to 128 MiB to allow for more
flexibility and reduce the metadata overhead when dealing with hot(un)plug
granularity smaller than 256 MiB.
128 MiB has been used by x86-64 since the very beginning. arm64 with 4k
base pages switched to 128 MiB as well: it's just big enough on these
architectures to allows for using a huge page (2 MiB) in the vmemmap in
sane setups with sizeof(struct page) == 64 bytes and a huge page mapping
in the direct mapping, while still allowing for small hot(un)plug
granularity.
For s390, we could even switch to a 64 MiB section size, as our huge page
size is 1 MiB: but the smaller the section size, the more sections we'll
have to manage especially on bigger machines. Making it consistent with
x86-64 and arm64 feels like the right thing for now.
Note that the smallest memory hot(un)plug granularity is also limited by
the memory block size, determined by extracting the memory increment
size from SCLP. Under QEMU/KVM, implementing virtio-mem, we expose 0;
therefore, we'll end up with a memory block size of 128 MiB with a
128 MiB section size.
Tested-by: Mario Casquero <mcasquer@redhat.com>
Acked-by: Heiko Carstens <hca@linux.ibm.com>
Signed-off-by: David Hildenbrand <david@redhat.com>
Tested-by: Sumanth Korikkar <sumanthk@linux.ibm.com>
Acked-by: Christian Borntraeger <borntraeger@linux.ibm.com>
Link: https://lore.kernel.org/r/20241025141453.1210600-7-david@redhat.com
Signed-off-by: Heiko Carstens <hca@linux.ibm.com>
|
|
virtio-mem currently depends on !DEVMEM | STRICT_DEVMEM. Let's default
STRICT_DEVMEM to "y" just like we do for arm64 and x86.
There could be ways in the future to filter access to virtio-mem device
memory even without STRICT_DEVMEM, but for now let's just keep it
simple.
Tested-by: Mario Casquero <mcasquer@redhat.com>
Acked-by: Heiko Carstens <hca@linux.ibm.com>
Signed-off-by: David Hildenbrand <david@redhat.com>
Tested-by: Sumanth Korikkar <sumanthk@linux.ibm.com>
Acked-by: Christian Borntraeger <borntraeger@linux.ibm.com>
Link: https://lore.kernel.org/r/20241025141453.1210600-6-david@redhat.com
Signed-off-by: Heiko Carstens <hca@linux.ibm.com>
|
|
Now that s390 code is prepared for memory devices that reside above the
maximum storage increment exposed through SCLP, everything is in place
to unlock virtio-mem support.
As virtio-mem in Linux currently supports logically onlining/offlining
memory in pageblock granularity, we have an effective hot(un)plug
granularity of 1 MiB on s390.
As virito-mem adds/removes individual Linux memory blocks (256MB), we
will currently never use gigantic pages in the identity mapping.
It is worth noting that neither storage keys nor storage attributes (e.g.,
data / nodat) are touched when onlining memory blocks, which is good
because we are not supposed to touch these parts for unplugged device
blocks that are logically offline in Linux.
We will currently never initialize storage keys for virtio-mem
memory -- IOW, storage_key_init_range() is never called. It could be added
in the future when plugging device blocks. But as that function
essentially does nothing without modifying the code (changing
PAGE_DEFAULT_ACC), that's just fine for now.
kexec should work as intended and just like on other architectures that
support virtio-mem: we will never place kexec binaries on virtio-mem
memory, and never indicate virtio-mem memory to the 2nd kernel. The
device driver in the 2nd kernel can simply reset the device --
turning all memory unplugged, to then start plugging memory and adding
them to Linux, without causing trouble because the memory is already
used elsewhere.
The special s390 kdump mode, whereby the 2nd kernel creates the ELF
core header, won't currently dump virtio-mem memory. The virtio-mem
driver has a special kdump mode, from where we can detect memory ranges
to dump. Based on this, support for dumping virtio-mem memory can be
added in the future fairly easily.
Acked-by: Michael S. Tsirkin <mst@redhat.com>
Tested-by: Mario Casquero <mcasquer@redhat.com>
Signed-off-by: David Hildenbrand <david@redhat.com>
Acked-by: Heiko Carstens <hca@linux.ibm.com>
Tested-by: Sumanth Korikkar <sumanthk@linux.ibm.com>
Acked-by: Christian Borntraeger <borntraeger@linux.ibm.com>
Link: https://lore.kernel.org/r/20241025141453.1210600-5-david@redhat.com
Signed-off-by: Heiko Carstens <hca@linux.ibm.com>
|
|
devices
To support memory devices under QEMU/KVM, such as virtio-mem,
we have to prepare our kernel virtual address space accordingly and
have to know the highest possible physical memory address we might see
later: the storage limit. The good old SCLP interface is not suitable for
this use case.
In particular, memory owned by memory devices has no relationship to
storage increments, it is always detected using the device driver, and
unaware OSes (no driver) must never try making use of that memory.
Consequently this memory is located outside of the "maximum storage
increment"-indicated memory range.
Let's use our new diag500 STORAGE_LIMIT subcode to query this storage
limit that can exceed the "maximum storage increment", and use the
existing interfaces (i.e., SCLP) to obtain information about the initial
memory that is not owned+managed by memory devices.
If a hypervisor does not support such memory devices, the address exposed
through diag500 STORAGE_LIMIT will correspond to the maximum storage
increment exposed through SCLP.
To teach kdump on s390 to include memory owned by memory devices, there
will be ways to query the relevant memory ranges from the device via a
driver running in special kdump mode (like virtio-mem already implements
to filter /proc/vmcore access so we don't end up reading from unplugged
device blocks).
Update setup_ident_map_size(), to clarify that there can be more than
just online and standby memory.
Tested-by: Mario Casquero <mcasquer@redhat.com>
Signed-off-by: David Hildenbrand <david@redhat.com>
Acked-by: Heiko Carstens <hca@linux.ibm.com>
Reviewed-by: Alexander Gordeev <agordeev@linux.ibm.com>
Tested-by: Sumanth Korikkar <sumanthk@linux.ibm.com>
Acked-by: Christian Borntraeger <borntraeger@linux.ibm.com>
Link: https://lore.kernel.org/r/20241025141453.1210600-4-david@redhat.com
Signed-off-by: Heiko Carstens <hca@linux.ibm.com>
|
|
Let's document our new diag500 subfunction that can be implemented by
userspace.
Signed-off-by: David Hildenbrand <david@redhat.com>
Reviewed-by: Heiko Carstens <hca@linux.ibm.com>
Reviewed-by: Eric Farman <farman@linux.ibm.com>
Tested-by: Sumanth Korikkar <sumanthk@linux.ibm.com>
Acked-by: Christian Borntraeger <borntraeger@linux.ibm.com>
Link: https://lore.kernel.org/r/20241025141453.1210600-3-david@redhat.com
Signed-off-by: Heiko Carstens <hca@linux.ibm.com>
|
|
Let's make it a generic KVM hypercall, allowing other subfunctions to
be more independent of virtio.
While at it, document that unsupported/unimplemented subfunctions result
in a SPECIFICATION exception.
This is a preparation for documenting a new subfunction.
Signed-off-by: David Hildenbrand <david@redhat.com>
Reviewed-by: Heiko Carstens <hca@linux.ibm.com>
Reviewed-by: Eric Farman <farman@linux.ibm.com>
Tested-by: Sumanth Korikkar <sumanthk@linux.ibm.com>
Acked-by: Christian Borntraeger <borntraeger@linux.ibm.com>
Link: https://lore.kernel.org/r/20241025141453.1210600-2-david@redhat.com
Signed-off-by: Heiko Carstens <hca@linux.ibm.com>
|
|
The program interrupt code has some extra bits that are sometimes set
by hardware for various reasons; those bits should be ignored when the
program interrupt number is needed for interrupt handling.
Fixes: 05066cafa925 ("s390/mm/fault: Handle guest-related program interrupts in KVM")
Reported-by: Christian Borntraeger <borntraeger@linux.ibm.com>
Signed-off-by: Claudio Imbrenda <imbrenda@linux.ibm.com>
Tested-by: Christian Borntraeger <borntraeger@linux.ibm.com>
Link: https://lore.kernel.org/r/20241031120316.25462-1-imbrenda@linux.ibm.com
Signed-off-by: Heiko Carstens <hca@linux.ibm.com>
|
|
Commit a23aa0404218 ("net: stmmac: ethtool: Fixed calltrace caused by
unbalanced disable_irq_wake calls") introduced checks to prevent
unbalanced enable and disable IRQ wake calls. However it only
initialized the auxiliary variable on one of the paths,
stmmac_request_irq_multi_msi(), missing the other,
stmmac_request_irq_single().
Add the same initialization on stmmac_request_irq_single() to prevent
"Unbalanced IRQ <x> wake disable" warnings from being printed the first
time disable_irq_wake() is called on platforms that run on that code
path.
Fixes: a23aa0404218 ("net: stmmac: ethtool: Fixed calltrace caused by unbalanced disable_irq_wake calls")
Signed-off-by: Nícolas F. R. A. Prado <nfraprado@collabora.com>
Reviewed-by: Simon Horman <horms@kernel.org>
Link: https://patch.msgid.link/20241101-stmmac-unbalanced-wake-single-fix-v1-1-5952524c97f0@collabora.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
|
|
The scope of the TX skb is wider than just mse102x_tx_frame_spi(),
so in case the TX skb room needs to be expanded, we should free the
the temporary skb instead of the original skb. Otherwise the original
TX skb pointer would be freed again in mse102x_tx_work(), which leads
to crashes:
Internal error: Oops: 0000000096000004 [#2] PREEMPT SMP
CPU: 0 PID: 712 Comm: kworker/0:1 Tainted: G D 6.6.23
Hardware name: chargebyte Charge SOM DC-ONE (DT)
Workqueue: events mse102x_tx_work [mse102x]
pstate: 20400009 (nzCv daif +PAN -UAO -TCO -DIT -SSBS BTYPE=--)
pc : skb_release_data+0xb8/0x1d8
lr : skb_release_data+0x1ac/0x1d8
sp : ffff8000819a3cc0
x29: ffff8000819a3cc0 x28: ffff0000046daa60 x27: ffff0000057f2dc0
x26: ffff000005386c00 x25: 0000000000000002 x24: 00000000ffffffff
x23: 0000000000000000 x22: 0000000000000001 x21: ffff0000057f2e50
x20: 0000000000000006 x19: 0000000000000000 x18: ffff00003fdacfcc
x17: e69ad452d0c49def x16: 84a005feff870102 x15: 0000000000000000
x14: 000000000000024a x13: 0000000000000002 x12: 0000000000000000
x11: 0000000000000400 x10: 0000000000000930 x9 : ffff00003fd913e8
x8 : fffffc00001bc008
x7 : 0000000000000000 x6 : 0000000000000008
x5 : ffff00003fd91340 x4 : 0000000000000000 x3 : 0000000000000009
x2 : 00000000fffffffe x1 : 0000000000000000 x0 : 0000000000000000
Call trace:
skb_release_data+0xb8/0x1d8
kfree_skb_reason+0x48/0xb0
mse102x_tx_work+0x164/0x35c [mse102x]
process_one_work+0x138/0x260
worker_thread+0x32c/0x438
kthread+0x118/0x11c
ret_from_fork+0x10/0x20
Code: aa1303e0 97fffab6 72001c1f 54000141 (f9400660)
Cc: stable@vger.kernel.org
Fixes: 2f207cbf0dd4 ("net: vertexcom: Add MSE102x SPI support")
Signed-off-by: Stefan Wahren <wahrenst@gmx.net>
Link: https://patch.msgid.link/20241105163101.33216-1-wahrenst@gmx.net
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
hrtimer_init_on_stack() is now unused. Delete it.
Signed-off-by: Nam Cao <namcao@linutronix.de>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Link: https://lore.kernel.org/all/510ce0d2944c4a382ea51e51d03dcfb73ba0f4f7.1730386209.git.namcao@linutronix.de
|
|
hrtimer_setup() and hrtimer_setup_on_stack() take the callback function
pointer as argument and initialize the timer completely.
Replace the hrtimer_init*() variants and the open coded initialization of
hrtimer::function with the new setup mechanism.
Switch to use the new functions.
Signed-off-by: Nam Cao <namcao@linutronix.de>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Link: https://lore.kernel.org/all/2bae912336103405adcdab96b88d3ea0353b4228.1730386209.git.namcao@linutronix.de
|
|
hrtimer_setup_on_stack() takes the callback function pointer as argument
and initializes the timer completely.
Replace hrtimer_init_on_stack() and the open coded initialization of
hrtimer::function with the new setup mechanism.
Signed-off-by: Nam Cao <namcao@linutronix.de>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Link: https://lore.kernel.org/all/f0d4ac32ec4050710a656cee8385fa4427be33aa.1730386209.git.namcao@linutronix.de
|
|
hrtimer_setup_on_stack() takes the callback function pointer as argument
and initializes the timer completely.
Replace hrtimer_init_on_stack() and the open coded initialization of
hrtimer::function with the new setup mechanism.
The conversion was done with Coccinelle.
Signed-off-by: Nam Cao <namcao@linutronix.de>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Link: https://lore.kernel.org/all/17f9421fed6061df4ad26a4cc91873d2c078cb0f.1730386209.git.namcao@linutronix.de
|
|
hrtimer_init_sleeper_on_stack() is now unused. Delete it.
Signed-off-by: Nam Cao <namcao@linutronix.de>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Link: https://lore.kernel.org/all/52549846635c0b3a2abf82101f539efdabcd9778.1730386209.git.namcao@linutronix.de
|
|
hrtimer_setup_sleeper_on_stack() replaces hrtimer_init_sleeper_on_stack()
to keep the naming convention consistent.
Convert the usage site over to it.
Signed-off-by: Nam Cao <namcao@linutronix.de>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Link: https://lore.kernel.org/all/fc91182375df81120a88dbe0263267e24d1bf19e.1730386209.git.namcao@linutronix.de
|
|
hrtimer_setup_sleeper_on_stack() replaces hrtimer_init_sleeper_on_stack()
to keep the naming convention consistent.
Convert the usage sites over to it. The conversion was done with
Coccinelle.
Signed-off-by: Nam Cao <namcao@linutronix.de>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Link: https://lore.kernel.org/all/299c07f0f96af8ab3a7631b47b6ca22b06b20577.1730386209.git.namcao@linutronix.de
|
|
hrtimer_setup_sleeper_on_stack() replaces hrtimer_init_sleeper_on_stack()
to keep the naming convention consistent.
Convert the usage site over to it. The conversion was done with Coccinelle.
Signed-off-by: Nam Cao <namcao@linutronix.de>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Link: https://lore.kernel.org/all/c4b40b8fef250b6a325e1b8bd6057005fb3cb660.1730386209.git.namcao@linutronix.de
|
|
hrtimer_setup_sleeper_on_stack() replaces hrtimer_init_sleeper_on_stack()
to keep the naming convention consistent.
Convert the usage site over to it. The conversion was done with Coccinelle.
Signed-off-by: Nam Cao <namcao@linutronix.de>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Link: https://lore.kernel.org/all/d92116a17313dee283ebc959869bea80fbf94cdb.1730386209.git.namcao@linutronix.de
|
|
hrtimer_setup_sleeper_on_stack() replaces hrtimer_init_sleeper_on_stack()
to keep the naming convention consistent.
Convert the usage site over to it. The conversion was done with Coccinelle.
Signed-off-by: Nam Cao <namcao@linutronix.de>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Link: https://lore.kernel.org/all/5f10c259fa43ba2fe774de5b2cedc22f5e9cfd2d.1730386209.git.namcao@linutronix.de
|
|
Some users of hrtimer need to change the callback function after the
initial setup. They write to hrtimer::function directly.
That's not safe under all circumstances as the write is lockless and a
concurrent timer expiry might end up using the wrong function pointer.
Introduce hrtimer_update_function(), which also performs runtime checks
whether it is safe to modify the callback.
This allows to make hrtimer::function private once all users are converted.
Signed-off-by: Nam Cao <namcao@linutronix.de>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Link: https://lore.kernel.org/all/20a937b0ae09ad54b5b6d86eabead7c570f1b72e.1730386209.git.namcao@linutronix.de
|
|
The hrtimer_init*() API is replaced by hrtimer_setup*() variants to
initialize the timer including the callback function at once.
hrtimer_init_sleeper_on_stack() does not need user to setup the callback
function separately, so a new variant would not be strictly necessary.
Nonetheless, to keep the naming convention consistent, introduce
hrtimer_setup_sleeper_on_stack(). hrtimer_init_on_stack() will be removed
once all users are converted.
Signed-off-by: Nam Cao <namcao@linutronix.de>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Link: https://lore.kernel.org/all/7b5e18e6dd0ace9eaa211201528cb9dc23752454.1730386209.git.namcao@linutronix.de
|
|
To initialize hrtimer on stack, hrtimer_init_on_stack() needs to be called
and also hrtimer::function must be set. This is error-prone and awkward to
use.
Introduce hrtimer_setup_on_stack() which does both of these things, so that
users of hrtimer can be simplified.
The new setup function also has a sanity check for the provided function
pointer. If NULL, a warning is emitted and a dummy callback installed.
hrtimer_init_on_stack() will be removed as soon as all of its users have
been converted to the new function.
Signed-off-by: Nam Cao <namcao@linutronix.de>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Link: https://lore.kernel.org/all/4b05e2ab3a82c517adf67fabc0f0cd8fe118b97c.1730386209.git.namcao@linutronix.de
|
|
To initialize hrtimer, hrtimer_init() needs to be called and also
hrtimer::function must be set. This is error-prone and awkward to use.
Introduce hrtimer_setup() which does both of these things, so that users of
hrtimer can be simplified.
The new setup function also has a sanity check for the provided function
pointer. If NULL, a warning is emitted and a dummy callback installed.
hrtimer_init() will be removed as soon as all of its users have been
converted to the new function.
Signed-off-by: Nam Cao <namcao@linutronix.de>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Link: https://lore.kernel.org/all/5057c1ddbfd4b92033cd93d37fe38e6b069d5ba6.1730386209.git.namcao@linutronix.de
|
|
The IORING_OP_TIMEOUT command uses hrtimer underneath. The timer's callback
function is setup in io_timeout(), and then the callback function is setup
again when the timer is rearmed.
Since the callback function is the same for both cases, the latter setup is
redundant, therefore remove it.
Signed-off-by: Nam Cao <namcao@linutronix.de>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Jens Axboe <axboe@kernel.dk:
Link: https://lore.kernel.org/all/07b28dfd5691478a2d250f379c8b90dd37f9bb9a.1730386209.git.namcao@linutronix.de
|
|
rt2x00usb_probe() executes a hrtimer_init() for txstatus_timer. Afterwards,
rt2x00lib_probe_dev() is called which also initializes this txstatus_timer
with the same settings.
Remove the redundant hrtimer_init() call in rt2x00usb_probe().
Signed-off-by: Nam Cao <namcao@linutronix.de>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Acked-by: Kalle Valo <kvalo@kernel.org>
Link: https://lore.kernel.org/all/66116057f788e18a6603d50a554417eee459e02c.1730386209.git.namcao@linutronix.de
|
|
The hrtimer is initialized in the KVM_XEN_VCPU_SET_ATTR ioctl. That caused
problem in the past, because the hrtimer can be initialized multiple times,
which was fixed by commit af735db31285 ("KVM: x86/xen: Initialize Xen timer
only once"). This commit avoids initializing the timer multiple times by
checking the field 'function' of struct hrtimer to determine if it has
already been initialized.
This is not required and in the way to make the function field private.
Move the hrtimer initialization into kvm_xen_init_vcpu() so that it will
only be initialized once.
Signed-off-by: Nam Cao <namcao@linutronix.de>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Acked-by: Sean Christopherson <seanjc@google.com>
Link: https://lore.kernel.org/all/9c33c7224d97d08f4fa30d3cc8687981c1d3e953.1730386209.git.namcao@linutronix.de
|
|
When a request is created, the hrtimer is not initialized and only its
'function' field is set to NULL. The hrtimer is only initialized when the
request is enqueued. The point of setting 'function' to NULL is that, it
can be used to check whether hrtimer_try_to_cancel() should be called while
retiring the request.
This "trick" is unnecessary, because hrtimer_try_to_cancel() already does
its own check whether the timer is armed. If the timer is not armed,
hrtimer_try_to_cancel() returns 0.
Fully initialize the timer when the request is created, which allows to
make the hrtimer::function field private once all users of hrtimer_init()
are converted to hrtimer_setup(), which requires a valid callback function
to be set.
Because hrtimer_try_to_cancel() returns 0 if the timer is not armed, the
logic to check whether to call i915_request_put() remains equivalent.
Signed-off-by: Nam Cao <namcao@linutronix.de>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Link: https://lore.kernel.org/all/50f865045aa672a9730343ad131543da332b1d8d.1730386209.git.namcao@linutronix.de
|
|
hrtimer_init*_on_stack() is not covered by tracing when
CONFIG_DEBUG_OBJECTS_TIMERS=y.
Rework the functions similar to hrtimer_init() and hrtimer_init_sleeper()
so that the hrtimer_init() tracepoint is unconditionally available.
The rework makes hrtimer_init_sleeper() unused. Delete it.
Signed-off-by: Nam Cao <namcao@linutronix.de>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Link: https://lore.kernel.org/all/74528e8abf2bb96e8bee85ffacbf14e15cf89f0d.1730386209.git.namcao@linutronix.de
|
|
The timer and hrtimer soft interrupts are raised in hard interrupt
context. With threaded interrupts force enabled or on PREEMPT_RT this leads
to waking the ksoftirqd for the processing of the soft interrupt.
ksoftirqd runs as SCHED_OTHER task which means it will compete with other
tasks for CPU resources. This can introduce long delays for timer
processing on heavy loaded systems and is not desired.
Split the TIMER_SOFTIRQ and HRTIMER_SOFTIRQ processing into a dedicated
timers thread and let it run at the lowest SCHED_FIFO priority.
Wake-ups for RT tasks happen from hardirq context so only timer_list timers
and hrtimers for "regular" tasks are processed here. The higher priority
ensures that wakeups are performed before scheduling SCHED_OTHER tasks.
Using a dedicated variable to store the pending softirq bits values ensure
that the timer are not accidentally picked up by ksoftirqd and other
threaded interrupts.
It shouldn't be picked up by ksoftirqd since it runs at lower priority.
However if ksoftirqd is already running while a timer fires, then ksoftird
will be PI-boosted due to the BH-lock to ktimer's priority.
The timer thread can pick up pending softirqs from ksoftirqd but only
if the softirq load is high. It is not be desired that the picked up
softirqs are processed at SCHED_FIFO priority under high softirq load
but this can already happen by a PI-boost by a force-threaded interrupt.
[ frederic@kernel.org: rcutorture.c fixes, storm fix by introduction of
local_timers_pending() for tick_nohz_next_event() ]
[ junxiao.chang@intel.com: Ensure ktimersd gets woken up even if a
softirq is currently served. ]
Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Paul E. McKenney <paulmck@kernel.org> [rcutorture]
Reviewed-by: Frederic Weisbecker <frederic@kernel.org>
Link: https://lore.kernel.org/all/20241106150419.2593080-4-bigeasy@linutronix.de
|
|
Raising the timer soft interrupt is always done from hard interrupt
context, so it can be reduced to just setting the TIMER soft interrupt
flag. The soft interrupt will be invoked on return from interrupt.
Use therefore __raise_softirq_irqoff() to raise the TIMER soft interrupt,
which is a trivial optimization.
Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Frederic Weisbecker <frederic@kernel.org>
Link: https://lore.kernel.org/all/20241106150419.2593080-3-bigeasy@linutronix.de
|
|
Raising the hrtimer soft interrupt is always done from hard interrupt
context, so it can be reduced to just setting the HRTIMER soft interrupt
flag. The soft interrupt will be invoked on return from interrupt.
Use therefore __raise_softirq_irqoff() to raise the HRTIMER soft interrupt,
which is a trivial optimization.
Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Frederic Weisbecker <frederic@kernel.org>
Link: https://lore.kernel.org/all/20241106150419.2593080-2-bigeasy@linutronix.de
|
|
Now that the SIG_IGN problem is solved in the core code, the alarmtimer
callbacks do not require a return value anymore.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Acked-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Acked-by: Frederic Weisbecker <frederic@kernel.org>
Link: https://lore.kernel.org/all/20241105064214.318837272@linutronix.de
|
|
Now that ignored posix timer signals are requeued and the timers are
rearmed on signal delivery the workaround to keep such timers alive and
self rearm them is not longer required.
Remove the unused alarm timer parts.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Frederic Weisbecker <frederic@kernel.org>
Acked-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lore.kernel.org/all/20241105064214.252443020@linutronix.de
|
|
Now that ignored posix timer signals are requeued and the timers are
rearmed on signal delivery the workaround to keep such timers alive and
self rearm them is not longer required.
Remove the relevant hacks and the not longer required return values from
the related functions. The alarm timer workarounds will be cleaned up in a
separate step.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Frederic Weisbecker <frederic@kernel.org>
Acked-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lore.kernel.org/all/20241105064214.187239060@linutronix.de
|
|
Queue posixtimers which have their signal ignored on the ignored list:
1) When the timer fires and the signal has SIG_IGN set
2) When SIG_IGN is installed via sigaction() and a timer signal
is already queued
This only happens when the signal is for a valid timer, which delivered the
signal in periodic mode. One-shot timer signals are correctly dropped.
Due to the lock order constraints (sighand::siglock nests inside
timer::lock) the signal code cannot access any of the timer fields which
are relevant to make this decision, e.g. timer::it_status.
This is addressed by establishing a protection scheme which requires to
lock both locks on the timer side for modifying decision fields in the
timer struct and therefore makes it possible for the signal delivery to
evaluate with only sighand:siglock being held:
1) Move the NULLification of timer->it_signal into the sighand::siglock
protected section of timer_delete() and check timer::it_signal in the
code path which determines whether the signal is dropped or queued on
the ignore list.
This ensures that a deleted timer cannot be moved onto the ignore
list, which would prevent it from being freed on exit() as it is not
longer in the process' posix timer list.
If the timer got moved to the ignored list before deletion then it is
removed from the ignored list under sighand lock in timer_delete().
2) Provide a new timer::it_sig_periodic flag, which gets set in the
signal queue path with both timer and sighand locks held if the timer
is actually in periodic mode at expiry time.
The ignore list code checks this flag under sighand::siglock and drops
the signal when it is not set.
If it is set, then the signal is moved to the ignored list independent
of the actual state of the timer.
When the signal is un-ignored later then the signal is moved back to
the signal queue. On signal delivery the posix timer side decides
about dropping the signal if the timer was re-armed, dis-armed or
deleted based on the signal sequence counter check.
If the thread/process exits then not yet delivered signals are
discarded which means the reference of the timer containing the
sigqueue is dropped and frees the timer.
This is way cheaper than requiring all code paths to lock
sighand::siglock of the target thread/process on any modification of
timer::it_status or going all the way and removing pending signals
from the signal queues on every rearm, disarm or delete operation.
So the protection scheme here is that on the timer side both timer::lock
and sighand::siglock have to be held for modifying
timer::it_signal
timer::it_sig_periodic
which means that on the signal side holding sighand::siglock is enough to
evaluate these fields.
In posixtimer_deliver_signal() holding timer::lock is sufficient to do the
sequence validation against timer::it_signal_seq because a concurrent
expiry is waiting on timer::lock to be released.
This completes the SIG_IGN handling and such timers are not longer self
rearmed which avoids pointless wakeups.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Frederic Weisbecker <frederic@kernel.org>
Acked-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lore.kernel.org/all/20241105064214.120756416@linutronix.de
|
|
When a real handler (including SIG_DFL) is installed for a signal, which
had previously SIG_IGN set, then the list of ignored posix timers has to be
checked for timers which are affected by this change.
Add a list walk function which checks for the matching signal number and if
found requeues the timers signal, so the timer is rearmed on signal
delivery.
Rearming the timer right away is not possible because that requires to drop
sighand lock.
No functional change as the counter part which queues the timers on the
ignored list is still missing.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Frederic Weisbecker <frederic@kernel.org>
Acked-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lore.kernel.org/all/20241105064214.054091076@linutronix.de
|
|
To handle posix timer signals on sigaction(SIG_IGN) properly, the timers
will be queued on a separate ignored list.
Add the necessary cleanup code for timer_delete() and exit_itimers().
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Frederic Weisbecker <frederic@kernel.org>
Acked-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lore.kernel.org/all/20241105064213.987530588@linutronix.de
|
|
To prepare for handling posix timer signals on sigaction(SIG_IGN) properly,
add a list to task::signal.
This list will be used to queue posix timers so their signal can be
requeued when SIG_IGN is lifted later.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Frederic Weisbecker <frederic@kernel.org>
Acked-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lore.kernel.org/all/20241105064213.920101900@linutronix.de
|
|
The posix timer signal handling uses siginfo::si_sys_private for handling
the sequence counter check. That indirection is not longer required and the
sequence count value at signal queueing time can be stored in struct
k_itimer itself.
This removes the requirement of treating siginfo::si_sys_private special as
it's now always zero as the kernel does not touch it anymore.
Suggested-by: Eric W. Biederman <ebiederm@xmission.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Frederic Weisbecker <frederic@kernel.org>
Acked-by: "Eric W. Biederman" <ebiederm@xmission.com>
Link: https://lore.kernel.org/all/20241105064213.852619866@linutronix.de
|
|
Remove the leftovers of sigqueue preallocation as it's not longer used.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Frederic Weisbecker <frederic@kernel.org>
Acked-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lore.kernel.org/all/20241105064213.786506636@linutronix.de
|
|
To cure the SIG_IGN handling for posix interval timers, the preallocated
sigqueue needs to be embedded into struct k_itimer to prevent life time
races of all sorts.
Now that the prerequisites are in place, embed the sigqueue into struct
k_itimer and fixup the relevant usage sites.
Aside of preparing for proper SIG_IGN handling, this spares an extra
allocation.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Frederic Weisbecker <frederic@kernel.org>
Acked-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lore.kernel.org/all/20241105064213.719695194@linutronix.de
|
|
In preparation for handling ignored posix timer signals correctly and
embedding the sigqueue struct into struct k_itimer, hand down a pointer to
the sigqueue struct into posix_timer_deliver_signal() instead of just
having a boolean flag.
No functional change.
Suggested-by: Eric W. Biederman <ebiederm@xmission.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Frederic Weisbecker <frederic@kernel.org>
Acked-by: "Eric W. Biederman" <ebiederm@xmission.com>
Link: https://lore.kernel.org/all/20241105064213.652658158@linutronix.de
|
|
To handle posix timers which have their signal ignored via SIG_IGN properly
it is required to requeue a ignored signal for delivery when SIG_IGN is
lifted so the timer gets rearmed.
Split the required code out of send_sigqueue() so it can be reused in
context of sigaction().
While at it rename send_sigqueue() to posixtimer_send_sigqueue() so its
clear what this is about.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Frederic Weisbecker <frederic@kernel.org>
Acked-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lore.kernel.org/all/20241105064213.586453412@linutronix.de
|
|
instead of re-evaluating the signal delivery mode everywhere.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Frederic Weisbecker <frederic@kernel.org>
Acked-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lore.kernel.org/all/20241105064213.519086500@linutronix.de
|
|
To cure the SIG_IGN handling for posix interval timers, the preallocated
sigqueue needs to be embedded into struct k_itimer to prevent life time
races of all sorts.
Provide a new function to initialize the embedded sigqueue to prepare for
that.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Frederic Weisbecker <frederic@kernel.org>
Acked-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lore.kernel.org/all/20241105064213.450427515@linutronix.de
|
