| Age | Commit message (Collapse) | Author |
|---|
|
tclock_gettime() is a wrapper around clock_gettime().
The first parameter of clock_gettime() is of type "clockid_t",
not "clock_t".
Use the correct type instead.
Link: https://lore.kernel.org/r/20250502-selftests-timens-fixes-v1-3-fb517c76f04d@linutronix.de
Signed-off-by: Thomas Weißschuh <thomas.weissschuh@linutronix.de>
Reviewed-by: Muhammad Usama Anjum <usama.anjum@collabora.com>
Signed-off-by: Shuah Khan <skhan@linuxfoundation.org>
|
|
These functions are never used outside their defining compilation unit and
can be made static.
Link: https://lore.kernel.org/r/20250502-selftests-timens-fixes-v1-2-fb517c76f04d@linutronix.de
Signed-off-by: Thomas Weißschuh <thomas.weissschuh@linutronix.de>
Reviewed-by: Muhammad Usama Anjum <usama.anjum@collabora.com>
Signed-off-by: Shuah Khan <skhan@linuxfoundation.org>
|
|
The TAP specification requires that the output begins with a header line.
These headers lines are missing in the timens tests.
Print such a line.
Link: https://lore.kernel.org/r/20250502-selftests-timens-fixes-v1-1-fb517c76f04d@linutronix.de
Signed-off-by: Thomas Weißschuh <thomas.weissschuh@linutronix.de>
Reviewed-by: Muhammad Usama Anjum <usama.anjum@collabora.com>
Signed-off-by: Shuah Khan <skhan@linuxfoundation.org>
|
|
Fix compile on openSUSE Tumbleweed (gcc-14.2.1, glibc-2.40):
- add missing sys/mount.h include
Fixes:
pid_max.c: In function ‘pid_max_cb’:
pid_max.c:42:15: error: implicit declaration of function ‘mount’ [-Wimplicit-function-declaration]
42 | ret = mount("", "/", NULL, MS_PRIVATE | MS_REC, 0);
| ^~~~~
Link: https://lore.kernel.org/r/20250115105211.390370-3-ps.report@gmx.net
Signed-off-by: Peter Seiderer <ps.report@gmx.net>
Reviewed-by: T.J. Mercier <tjmercier@google.com>
Signed-off-by: Shuah Khan <skhan@linuxfoundation.org>
|
|
https://gitlab.freedesktop.org/drm/i915/kernel into drm-fixes
drm/i915 fixes for v6.15-rc6:
- Fix oops on resume after disconnecting DP MST sinks during suspend
- Fix SPLC num_waiters refcounting
Signed-off-by: Dave Airlie <airlied@redhat.com>
From: Jani Nikula <jani.nikula@intel.com>
Link: https://lore.kernel.org/r/87tt5umeaw.fsf@intel.com
|
|
https://gitlab.freedesktop.org/drm/xe/kernel into drm-fixes
Driver Changes:
- Prevent PF queue overflow
- Hold all forcewake during mocs test
- Remove GSC flush on reset path
- Fix forcewake put on error path
- Fix runtime warning when building without svm
Signed-off-by: Dave Airlie <airlied@redhat.com>
From: Lucas De Marchi <lucas.demarchi@intel.com>
Link: https://lore.kernel.org/r/jffqa56f2zp4i5ztz677cdspgxhnw7qfop3dd3l2epykfpfvza@q2nw6wapsphz
|
|
Coverity scan reported the usage of "mode->clock * 1000" may lead to
integer overflow. Use "1000ULL" instead of "1000"
when utilizing it to avoid potential integer overflow issue.
Link: https://scan5.scan.coverity.com/#/project-view/10074/10063?selectedIssue=1646759
Signed-off-by: I Hsin Cheng <richard120310@gmail.com>
Reviewed-by: Martin Blumenstingl <martin.blumenstingl@googlemail.com>
Fixes: 1017560164b6 ("drm/meson: use unsigned long long / Hz for frequency types")
Link: https://lore.kernel.org/r/20250505184338.678540-1-richard120310@gmail.com
Signed-off-by: Neil Armstrong <neil.armstrong@linaro.org>
|
|
Commit 0b631ed3ce92 ("kselftest: cpufreq: Add RTC wakeup alarm") added
support for automatic wakeup in the suspend routine of the cpufreq
kselftest by using rtcwake, however it left the manual power state
change in the common path. The end result is that when running the
cpufreq kselftest with '-t suspend_rtc' or '-t hibernate_rtc', the
system will go to sleep and be woken up by the RTC, but then immediately
go to sleep again with no wakeup programmed, so it will sleep forever in
an automated testing setup.
Fix this by moving the manual power state change so that it only happens
when not using rtcwake.
Link: https://lore.kernel.org/r/20250430-ksft-cpufreq-suspend-rtc-double-fix-v1-1-dc17a729c5a7@collabora.com
Fixes: 0b631ed3ce92 ("kselftest: cpufreq: Add RTC wakeup alarm")
Signed-off-by: Nícolas F. R. A. Prado <nfraprado@collabora.com>
Signed-off-by: Shuah Khan <skhan@linuxfoundation.org>
|
|
In cpufreq basic selftests, one of the testcases is to read all cpufreq
sysfs files and print the values. This testcase assumes all the cpufreq
sysfs files have read permissions. However certain cpufreq sysfs files
(eg. stats/reset) are write only files and this testcase errors out
when it is not able to read the file.
Similarily, there is one more testcase which reads the cpufreq sysfs
file data and write it back to same file. This testcase also errors out
for sysfs files without read permission.
Fix these testcases by adding proper read permission checks.
Link: https://lore.kernel.org/r/20250430171433.10866-1-swapnil.sapkal@amd.com
Reported-by: Narasimhan V <narasimhan.v@amd.com>
Signed-off-by: Swapnil Sapkal <swapnil.sapkal@amd.com>
Signed-off-by: Shuah Khan <skhan@linuxfoundation.org>
|
|
Poll program is a helper to ftracetest, thus make it a
generic file and remove it from being run as a test.
Currently when executing tests using
$ make run_tests
CC poll
TAP version 13
1..2
# timeout set to 0
# selftests: ftrace: poll
# Error: Polling file is not specified
not ok 1 selftests: ftrace: poll # exit=255
Fix this by using TEST_GEN_FILES to build the 'poll' binary as a helper
rather than as a test.
Fixes: 80c3e28528ff ("selftests/tracing: Add hist poll() support test")
Link: https://lore.kernel.org/r/20250409044632.363285-1-Ayush.jain3@amd.com
Signed-off-by: Ayush Jain <Ayush.jain3@amd.com>
Acked-by: Steven Rostedt (Google) <rostedt@goodmis.org>
Signed-off-by: Shuah Khan <skhan@linuxfoundation.org>
|
|
Marek reported that the rework of handle_nested_irq() introduced a inverted
condition, which prevents handling of interrupts. Fix it up.
Fixes: 2ef2e13094c7 ("genirq/chip: Rework handle_nested_irq()")
Reported-by: Marek Szyprowski <m.szyprowski@samsung.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Closes: https://lore.kernel/org/all/46ed4040-ca11-4157-8bd7-13c04c113734@samsung.com
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/arm64/linux
Pull arm64 fix from Catalin Marinas:
"Move the arm64_use_ng_mappings variable from the .bss to the .data
section as it is accessed very early during boot with the MMU off and
before the .bss has been initialised.
This could lead to incorrect idmap page table"
* tag 'arm64-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/arm64/linux:
arm64: cpufeature: Move arm64_use_ng_mappings to the .data section to prevent wrong idmap generation
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/riscv/linux
Pull RISC-V fixes from Palmer Dabbelt:
- The compressed half-word misaligned access instructions (c.lhu, c.lh,
and c.sh) from the Zcb extension are now properly emulated
- A series of fixes to properly emulate permissions while handling
userspace misaligned accesses
- A pair of fixes for PR_GET_TAGGED_ADDR_CTRL to avoid accessing the
envcfg CSR on systems that don't support that CSR, and to report
those failures up to userspace
- The .rela.dyn section is no longer stripped from vmlinux, as it is
necessary to relocate the kernel under some conditions (including
kexec)
* tag 'riscv-for-linus-6.15-rc6' of git://git.kernel.org/pub/scm/linux/kernel/git/riscv/linux:
riscv: Disallow PR_GET_TAGGED_ADDR_CTRL without Supm
scripts: Do not strip .rela.dyn section
riscv: Fix kernel crash due to PR_SET_TAGGED_ADDR_CTRL
riscv: misaligned: use get_user() instead of __get_user()
riscv: misaligned: enable IRQs while handling misaligned accesses
riscv: misaligned: factorize trap handling
riscv: misaligned: Add handling for ZCB instructions
|
|
Commit ec5fbdfb99d1 ("cgroup/cpuset: Enable update_tasks_cpumask()
on top_cpuset") enabled us to pull CPUs dedicated to child partitions
from tasks in top_cpuset by ignoring per cpu kthreads. However, there
can be other kthreads that are not per cpu but have PF_NO_SETAFFINITY
flag set to indicate that we shouldn't mess with their CPU affinity.
For other kthreads, their affinity will be changed to skip CPUs dedicated
to child partitions whether it is an isolating or a scheduling one.
As all the per cpu kthreads have PF_NO_SETAFFINITY set, the
PF_NO_SETAFFINITY tasks are essentially a superset of per cpu kthreads.
Fix this issue by dropping the kthread_is_per_cpu() check and checking
the PF_NO_SETAFFINITY flag instead.
Fixes: ec5fbdfb99d1 ("cgroup/cpuset: Enable update_tasks_cpumask() on top_cpuset")
Signed-off-by: Waiman Long <longman@redhat.com>
Acked-by: Frederic Weisbecker <frederic@kernel.org>
Signed-off-by: Tejun Heo <tj@kernel.org>
|
|
Pull block fixes from Jens Axboe:
- Fix for a regression in this series for loop and read/write iterator
handling
- zone append block update tweak
- remove a broken IO priority test
- NVMe pull request via Christoph:
- unblock ctrl state transition for firmware update (Daniel
Wagner)
* tag 'block-6.15-20250509' of git://git.kernel.dk/linux:
block: remove test of incorrect io priority level
nvme: unblock ctrl state transition for firmware update
block: only update request sector if needed
loop: Add sanity check for read/write_iter
|
|
Pull io_uring fixes from Jens Axboe:
- Fix for linked timeouts arming and firing wrt prep and issue of the
request being managed by the linked timeout
- Fix for a CQE ordering issue between requests with multishot and
using the same buffer group. This is a dumbed down version for this
release and for stable, it'll get improved for v6.16
- Tweak the SQPOLL submit batch size. A previous commit made SQPOLL
manage its own task_work and chose a tiny batch size, bump it from 8
to 32 to fix a performance regression due to that
* tag 'io_uring-6.15-20250509' of git://git.kernel.dk/linux:
io_uring/sqpoll: Increase task_work submission batch size
io_uring: ensure deferred completions are flushed for multishot
io_uring: always arm linked timeouts prior to issue
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/modules/linux
Pull modules fix from Petr Pavlu:
"A single fix to prevent use of an uninitialized completion pointer
when releasing a module_kobject in specific situations.
This addresses a latent bug exposed by commit f95bbfe18512 ("drivers:
base: handle module_kobject creation")"
* tag 'modules-6.15-rc6' of git://git.kernel.org/pub/scm/linux/kernel/git/modules/linux:
module: ensure that kobject_put() is safe for module type kobjects
|
|
arm/fixes
Apple SoC fixes for 6.15
This tag contains two small commits since rc1:
- Add a .mailmap entry requested by Asahi Lina to better filter her
emails
- Mark the power domains for the touchbar support introduced with 6.15
as always on since the driver cannot initialize the touchbar from
scratch after the domains are powered off (e.g. during suspend).
* tag 'asahi-soc-fixes-6.15' of https://github.com/AsahiLinux/linux:
arm64: dts: apple: touchbar: Mark ps_dispdfr_be as always-on
mailmap: Update email for Asahi Lina
Link: https://lore.kernel.org/r/20250423145047.3098-1-sven@svenpeter.dev
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
|
|
https://github.com/sophgo/linux into arm/fixes
RISC-V Sophgo Devicetree fixes for v6.15-rc1
Just one minor fix to correct DMA data-width
configuration for CV18xx.
Signed-off-by: Chen Wang <unicorn_wang@outlook.com>
* tag 'riscv-sophgo-dt-fixes-for-v6.15-rc1' of https://github.com/sophgo/linux:
riscv: dts: sophgo: fix DMA data-width configuration for CV18xx
Link: https://lore.kernel.org/r/MA0P287MB2262454C19B8899BC1694D04FE832@MA0P287MB2262.INDP287.PROD.OUTLOOK.COM
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
|
|
https://git.kernel.org/pub/scm/linux/kernel/git/amlogic/linux into arm/fixes
Amlogic Fixes for v6.15:
- fix reference to unknown/untested PWM clock on ARM/ARM64 boards
- fix missing clkc_audio node on dreambox ARM64 DT
* tag 'amlogic-fixes-for-v6.15' of https://git.kernel.org/pub/scm/linux/kernel/git/amlogic/linux:
arm64: dts: amlogic: dreambox: fix missing clkc_audio node
arm64: dts: amlogic: g12: fix reference to unknown/untested PWM clock
arm64: dts: amlogic: gx: fix reference to unknown/untested PWM clock
ARM: dts: amlogic: meson8b: fix reference to unknown/untested PWM clock
ARM: dts: amlogic: meson8: fix reference to unknown/untested PWM clock
Link: https://lore.kernel.org/r/e9c520a1-b986-49e1-b9b1-67511c187716@linaro.org
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
|
|
https://git.kernel.org/pub/scm/linux/kernel/git/mmind/linux-rockchip into arm/fixes
Removal of operating-points above what the rk3588j soc is rated for, and
a number of smaller fixes: Turing RK1 fan can spin down again, fixed pins,
pinmuxing and clocks and some devicetree-correctnes improvements.
* tag 'v6.15-rockchip-dtsfixes1' of https://git.kernel.org/pub/scm/linux/kernel/git/mmind/linux-rockchip:
arm64: dts: rockchip: fix Sige5 RTC interrupt pin
arm64: dts: rockchip: Assign RT5616 MCLK rate on rk3588-friendlyelec-cm3588
arm64: dts: rockchip: Align wifi node name with bindings in CB2
arm64: dts: rockchip: Fix mmc-pwrseq clock name on rock-pi-4
arm64: dts: rockchip: Use "regulator-fixed" for btreg on px30-engicam for vcc3v3-btreg
arm64: dts: rockchip: Add pinmuxing for eMMC on QNAP TS433
arm64: dts: rockchip: Remove overdrive-mode OPPs from RK3588J SoC dtsi
arm64: dts: rockchip: Allow Turing RK1 cooling fan to spin down
Link: https://lore.kernel.org/r/2923598.88bMQJbFj6@diego
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
|
|
tl;dr: There is a window in the mm switching code where the new CR3 is
set and the CPU should be getting TLB flushes for the new mm. But
should_flush_tlb() has a bug and suppresses the flush. Fix it by
widening the window where should_flush_tlb() sends an IPI.
Long Version:
=== History ===
There were a few things leading up to this.
First, updating mm_cpumask() was observed to be too expensive, so it was
made lazier. But being lazy caused too many unnecessary IPIs to CPUs
due to the now-lazy mm_cpumask(). So code was added to cull
mm_cpumask() periodically[2]. But that culling was a bit too aggressive
and skipped sending TLB flushes to CPUs that need them. So here we are
again.
=== Problem ===
The too-aggressive code in should_flush_tlb() strikes in this window:
// Turn on IPIs for this CPU/mm combination, but only
// if should_flush_tlb() agrees:
cpumask_set_cpu(cpu, mm_cpumask(next));
next_tlb_gen = atomic64_read(&next->context.tlb_gen);
choose_new_asid(next, next_tlb_gen, &new_asid, &need_flush);
load_new_mm_cr3(need_flush);
// ^ After 'need_flush' is set to false, IPIs *MUST*
// be sent to this CPU and not be ignored.
this_cpu_write(cpu_tlbstate.loaded_mm, next);
// ^ Not until this point does should_flush_tlb()
// become true!
should_flush_tlb() will suppress TLB flushes between load_new_mm_cr3()
and writing to 'loaded_mm', which is a window where they should not be
suppressed. Whoops.
=== Solution ===
Thankfully, the fuzzy "just about to write CR3" window is already marked
with loaded_mm==LOADED_MM_SWITCHING. Simply checking for that state in
should_flush_tlb() is sufficient to ensure that the CPU is targeted with
an IPI.
This will cause more TLB flush IPIs. But the window is relatively small
and I do not expect this to cause any kind of measurable performance
impact.
Update the comment where LOADED_MM_SWITCHING is written since it grew
yet another user.
Peter Z also raised a concern that should_flush_tlb() might not observe
'loaded_mm' and 'is_lazy' in the same order that switch_mm_irqs_off()
writes them. Add a barrier to ensure that they are observed in the
order they are written.
Signed-off-by: Dave Hansen <dave.hansen@linux.intel.com>
Acked-by: Rik van Riel <riel@surriel.com>
Link: https://lore.kernel.org/oe-lkp/202411282207.6bd28eae-lkp@intel.com/ [1]
Fixes: 6db2526c1d69 ("x86/mm/tlb: Only trim the mm_cpumask once a second") [2]
Reported-by: Stephen Dolan <sdolan@janestreet.com>
Cc: stable@vger.kernel.org
Acked-by: Ingo Molnar <mingo@kernel.org>
Acked-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
|
|
Fix SD card timeout issue caused by LDO5 regulator getting disabled
after boot.
The kernel log shows LDO5 being disabled, which leads to a timeout
on USDHC2:
[ 33.760561] LDO5: disabling
[ 81.119861] mmc1: Timeout waiting for hardware interrupt.
To prevent this, set regulator-boot-on and regulator-always-on for
LDO5. Also add the vqmmc regulator to properly support 1.8V/3.3V
signaling for USDHC2 using a GPIO-controlled regulator.
Fixes: 6c2a1f4f71258 ("arm64: dts: imx8mp-var-som-symphony: Add Variscite Symphony board and VAR-SOM-MX8MP SoM")
Signed-off-by: Himanshu Bhavani <himanshu.bhavani@siliconsignals.io>
Acked-by: Tarang Raval <tarang.raval@siliconsignals.io>
Signed-off-by: Shawn Guo <shawnguo@kernel.org>
|
|
Keep track of the number requests a ring currently has allocated (and
not freed), it'll be needed in the next patch.
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Link: https://lore.kernel.org/r/c8f8308294dc2a1cb8925d984d937d4fc14ab5d4.1746788718.git.asml.silence@gmail.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
|
|
io_account_cq_overflow() doesn't help explaining what's going on in
there, and it'll become even smaller with following patches, so open
code it.
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Link: https://lore.kernel.org/r/e4333fa0d371f519e52a71148ebdffed4b8d3aa9.1746788718.git.asml.silence@gmail.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
|
|
We check sequences when queuing drained requests as well when flushing
them. Instead, always queue and immediately try to flush, so that all
seq handling can be kept contained in the flushing code.
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Link: https://lore.kernel.org/r/d4651f742e671af5b3216581e539ea5d31bc7125.1746788718.git.asml.silence@gmail.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
|
|
Currently io_drain_req() has two steps. The first is fast path checking
sequence numbers. The second is allocations, rechecking and actual
queuing. Further simplify it by removing the first step.
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Link: https://lore.kernel.org/r/4d06e89ed07611993d7bf89182de2300858379bd.1746788718.git.asml.silence@gmail.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
|
|
"ret" in io_drain_req() is only used in one place, remove it and pass
-ENOMEM directly.
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Link: https://lore.kernel.org/r/ece724b77e66e6caabcc215e0032ee7ff140f289.1746788718.git.asml.silence@gmail.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
|
|
io_queue_deferred() is not tolerant to spurious calls not completing
some requests. You can have an inflight drain-marked request and another
request that came after and got queued into the drain list. Now, if
io_queue_deferred() is called before the first request completes, it'll
check the 2nd req with req_need_defer(), find that there is no drain
flag set, and queue it for execution.
To make io_queue_deferred() work, it should at least check sequences for
the first request, and then we need also need to check if there is
another drain request creating another bubble.
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Link: https://lore.kernel.org/r/972bde11b7d4ef25b3f5e3fd34f80e4d2aa345b8.1746788718.git.asml.silence@gmail.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
|
|
Account drain allocations against memcg. It's not a big problem as each
such allocation is paired with a request, which is accounted, but it's
nicer to follow the limits more closely.
Cc: stable@vger.kernel.org # 6.1
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Link: https://lore.kernel.org/r/f8dfdbd755c41fd9c75d12b858af07dfba5bbb68.1746788718.git.asml.silence@gmail.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
|
|
io_add_aux_cqe() can only be called for rings with uring_lock protected
completion queues, add a couple of assertions in regards to that.
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Link: https://lore.kernel.org/r/c010eab7b94a187c00a9d46d8b67bf7fcad18af4.1746788592.git.asml.silence@gmail.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
|
|
Instruct Makefile to never try to compile net.c without CONFIG_NET and
kill ifdefs in the file.
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Link: https://lore.kernel.org/r/f466400e20c3f536191bfd559b1f3cd2a2ab5a1e.1746788579.git.asml.silence@gmail.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
|
|
Rename first parameter in io_pin_pages from ubuf to uaddr for consistency
between declaration and implementation.
Signed-off-by: Long Li <leo.lilong@huawei.com>
Link: https://lore.kernel.org/r/20250509063015.3799255-1-leo.lilong@huaweicloud.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
|
|
Our QA team reported a 10%-23%, throughput reduction on an io_uring
sqpoll testcase doing IO to a null_blk, that I traced back to a
reduction of the device submission queue depth utilization. It turns out
that, after commit af5d68f8892f ("io_uring/sqpoll: manage task_work
privately"), we capped the number of task_work entries that can be
completed from a single spin of sqpoll to only 8 entries, before the
sqpoll goes around to (potentially) sleep. While this cap doesn't drive
the submission side directly, it impacts the completion behavior, which
affects the number of IO queued by fio per sqpoll cycle on the
submission side, and io_uring ends up seeing less ios per sqpoll cycle.
As a result, block layer plugging is less effective, and we see more
time spent inside the block layer in profilings charts, and increased
submission latency measured by fio.
There are other places that have increased overhead once sqpoll sleeps
more often, such as the sqpoll utilization calculation. But, in this
microbenchmark, those were not representative enough in perf charts, and
their removal didn't yield measurable changes in throughput. The major
overhead comes from the fact we plug less, and less often, when submitting
to the block layer.
My benchmark is:
fio --ioengine=io_uring --direct=1 --iodepth=128 --runtime=300 --bs=4k \
--invalidate=1 --time_based --ramp_time=10 --group_reporting=1 \
--filename=/dev/nullb0 --name=RandomReads-direct-nullb-sqpoll-4k-1 \
--rw=randread --numjobs=1 --sqthread_poll
In one machine, tested on top of Linux 6.15-rc1, we have the following
baseline:
READ: bw=4994MiB/s (5236MB/s), 4994MiB/s-4994MiB/s (5236MB/s-5236MB/s), io=439GiB (471GB), run=90001-90001msec
With this patch:
READ: bw=5762MiB/s (6042MB/s), 5762MiB/s-5762MiB/s (6042MB/s-6042MB/s), io=506GiB (544GB), run=90001-90001msec
which is a 15% improvement in measured bandwidth. The average
submission latency is noticeably lowered too. As measured by
fio:
Baseline:
lat (usec): min=20, max=241, avg=99.81, stdev=3.38
Patched:
lat (usec): min=26, max=226, avg=86.48, stdev=4.82
If we look at blktrace, we can also see the plugging behavior is
improved. In the baseline, we end up limited to plugging 8 requests in
the block layer regardless of the device queue depth size, while after
patching we can drive more io, and we manage to utilize the full device
queue.
In the baseline, after a stabilization phase, an ordinary submission
looks like:
254,0 1 49942 0.016028795 5977 U N [iou-sqp-5976] 7
After patching, I see consistently more requests per unplug.
254,0 1 4996 0.001432872 3145 U N [iou-sqp-3144] 32
Ideally, the cap size would at least be the deep enough to fill the
device queue, but we can't predict that behavior, or assume all IO goes
to a single device, and thus can't guess the ideal batch size. We also
don't want to let the tw run unbounded, though I'm not sure it would
really be a problem. Instead, let's just give it a more sensible value
that will allow for more efficient batching. I've tested with different
cap values, and initially proposed to increase the cap to 1024. Jens
argued it is too big of a bump and I observed that, with 32, I'm no
longer able to observe this bottleneck in any of my machines.
Fixes: af5d68f8892f ("io_uring/sqpoll: manage task_work privately")
Signed-off-by: Gabriel Krisman Bertazi <krisman@suse.de>
Link: https://lore.kernel.org/r/20250508181203.3785544-1-krisman@suse.de
Signed-off-by: Jens Axboe <axboe@kernel.dk>
|
|
Cong Wang says:
====================
net_sched: Fix gso_skb flushing during qdisc change
This patchset contains a bug fix and its test cases, please check each
patch description for more details. To keep the bug fix minimum, I
intentionally limit the code changes to the cases reported here.
---
v2: added a missing qlen--
fixed the new boolean parameter for two qdiscs
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
Added new test cases for FQ, FQ_CODEL, FQ_PIE, and HHF qdiscs to verify queue
trimming behavior when the qdisc limit is dynamically reduced.
Each test injects packets, reduces the qdisc limit, and checks that the new
limit is enforced. This is still best effort since timing qdisc backlog
is not easy.
Signed-off-by: Cong Wang <xiyou.wangcong@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
Previously, when reducing a qdisc's limit via the ->change() operation, only
the main skb queue was trimmed, potentially leaving packets in the gso_skb
list. This could result in NULL pointer dereference when we only check
sch->limit against sch->q.qlen.
This patch introduces a new helper, qdisc_dequeue_internal(), which ensures
both the gso_skb list and the main queue are properly flushed when trimming
excess packets. All relevant qdiscs (codel, fq, fq_codel, fq_pie, hhf, pie)
are updated to use this helper in their ->change() routines.
Fixes: 76e3cc126bb2 ("codel: Controlled Delay AQM")
Fixes: 4b549a2ef4be ("fq_codel: Fair Queue Codel AQM")
Fixes: afe4fd062416 ("pkt_sched: fq: Fair Queue packet scheduler")
Fixes: ec97ecf1ebe4 ("net: sched: add Flow Queue PIE packet scheduler")
Fixes: 10239edf86f1 ("net-qdisc-hhf: Heavy-Hitter Filter (HHF) qdisc")
Fixes: d4b36210c2e6 ("net: pkt_sched: PIE AQM scheme")
Reported-by: Will <willsroot@protonmail.com>
Reported-by: Savy <savy@syst3mfailure.io>
Signed-off-by: Cong Wang <xiyou.wangcong@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
Joel Savitz <jsavitz@redhat.com> says:
The two patches are independent of each other. The first patch removes
unnecssary NULL guards from free_nsproxy() and create_new_namespaces()
in line with other usage of the put_*_ns() call sites. The second patch
slightly reduces the size of the kernel when CONFIG_CGROUPS is not
selected.
* patches from https://lore.kernel.org/20250508184930.183040-1-jsavitz@redhat.com:
include/cgroup: separate {get,put}_cgroup_ns no-op case
kernel/nsproxy: remove unnecessary guards
Link: https://lore.kernel.org/20250508184930.183040-1-jsavitz@redhat.com
Signed-off-by: Christian Brauner <brauner@kernel.org>
|
|
When CONFIG_CGROUPS is not selected, {get,put}_cgroup_ns become no-ops
and therefore it is not necessary to compile in the code for changing
the reference count.
When CONFIG_CGROUP is selected, there is no valid case where
either of {get,put}_cgroup_ns() will be called with a NULL argument.
Signed-off-by: Joel Savitz <jsavitz@redhat.com>
Link: https://lore.kernel.org/20250508184930.183040-3-jsavitz@redhat.com
Signed-off-by: Christian Brauner <brauner@kernel.org>
|
|
In free_nsproxy() and the error path of create_new_namesapces() the
put_*_ns() calls are guarded by unnecessary NULL checks.
put_pid_ns(), put_ipc_ns(), put_uts_ns(), and put_time_ns() will never
receive a NULL argument unless their namespace type is disabled, and in
this case all four become no-ops at compile time anyway. put_mnt_ns()
will never receive a null argument at any time.
This unguarded usage is in line with other call sites of put_*_ns().
Signed-off-by: Joel Savitz <jsavitz@redhat.com>
Link: https://lore.kernel.org/20250508184930.183040-2-jsavitz@redhat.com
Signed-off-by: Christian Brauner <brauner@kernel.org>
|
|
Using FREEZE_HOLDER_USERSPACE has two consequences:
(1) If userspace freezes the filesystem after mnt_drop_write_file() but
before freeze_super() was called filesystem resizing will fail
because the freeze isn't marked as nestable.
(2) If the kernel has successfully frozen the filesystem via
FREEZE_HOLDER_USERSPACE userspace can simply undo it by using the
FITHAW ioctl.
Fix both issues by using FREEZE_HOLDER_KERNEL. It will nest with
FREEZE_HOLDER_USERSPACE and cannot be undone by userspace.
And it is the correct thing to do because the kernel temporarily freezes
the filesystem.
Signed-off-by: Christian Brauner <brauner@kernel.org>
|
|
Christian Brauner <brauner@kernel.org> says:
Now all the pieces are in place to actually allow the power subsystem to
freeze/thaw filesystems during suspend/resume. Filesystems are only
frozen and thawed if the power subsystem does actually own the freeze.
Othwerwise it risks thawing filesystems it didn't own. This could be
done differently be e.g., keeping the filesystems that were actually
frozen on a list and then unfreezing them from that list. This is
disgustingly unclean though and reeks of an ugly hack.
If the filesystem is already frozen by the time we've frozen all
userspace processes we don't care to freeze it again. That's userspace's
job once the process resumes. We only actually freeze filesystems if we
absolutely have to and we ignore other failures to freeze.
We could bubble up errors and fail suspend/resume if the error isn't
EBUSY (aka it's already frozen) but I don't think that this is worth it.
Filesystem freezing during suspend/resume is best-effort. If the user
has 500 ext4 filesystems mounted and 4 fail to freeze for whatever
reason then we simply skip them.
What we have now is already a big improvement and let's see how we fare
with it before making our lives even harder (and uglier) than we have
to.
* patches from https://lore.kernel.org/r/20250402-work-freeze-v2-0-6719a97b52ac@kernel.org:
kernfs: add warning about implementing freeze/thaw
power: freeze filesystems during suspend/resume
Link: https://lore.kernel.org/r/20250402-work-freeze-v2-0-6719a97b52ac@kernel.org
Signed-off-by: Christian Brauner <brauner@kernel.org>
|
|
Christian Brauner <brauner@kernel.org> says:
Allow efivarfs to partake to resync variable state during system
hibernation and suspend. Add freeze/thaw support.
This is a pretty straightforward implementation. We simply add regular
freeze/thaw support for both userspace and the kernel. This works
without any big issues and congrats afaict efivars is the first
pseudofilesystem that adds support for filesystem freezing and thawing.
The simplicity comes from the fact that we simply always resync variable
state after efivarfs has been frozen. It doesn't matter whether that's
because of suspend, userspace initiated freeze or hibernation. Efivars
is simple enough that it doesn't matter that we walk all dentries. There
are no directories and there aren't insane amounts of entries and both
freeze/thaw are already heavy-handed operations. If userspace initiated
a freeze/thaw cycle they would need CAP_SYS_ADMIN in the initial user
namespace (as that's where efivarfs is mounted) so it can't be triggered
by random userspace. IOW, we really really don't care.
* patches from https://lore.kernel.org/r/20250331-work-freeze-v1-0-6dfbe8253b9f@kernel.org:
efivarfs: support freeze/thaw
libfs: export find_next_child()
Link: https://lore.kernel.org/r/20250331-work-freeze-v1-0-6dfbe8253b9f@kernel.org
Signed-off-by: Christian Brauner <brauner@kernel.org>
|
|
Sysfs is built on top of kernfs and sysfs provides the power management
infrastructure to support suspend/hibernate by writing to various files
in /sys/power/. As filesystems may be automatically frozen during
suspend/hibernate implementing freeze/thaw support for kernfs
generically will cause deadlocks as the suspending/hibernation
initiating task will hold a VFS lock that it will then wait upon to be
released. If freeze/thaw for kernfs is needed talk to the VFS.
Link: https://lore.kernel.org/r/20250402-work-freeze-v2-4-6719a97b52ac@kernel.org
Signed-off-by: Christian Brauner <brauner@kernel.org>
|
|
Allow efivarfs to partake to resync variable state during system
hibernation and suspend. Add freeze/thaw support.
This is a pretty straightforward implementation. We simply add regular
freeze/thaw support for both userspace and the kernel. This works
without any big issues and congrats afaict efivars is the first
pseudofilesystem that adds support for filesystem freezing and thawing.
The simplicity comes from the fact that we simply always resync variable
state after efivarfs has been frozen. It doesn't matter whether that's
because of suspend, userspace initiated freeze or hibernation. Efivars
is simple enough that it doesn't matter that we walk all dentries. There
are no directories and there aren't insane amounts of entries and both
freeze/thaw are already heavy-handed operations. We really really don't
need to care.
Link: https://lore.kernel.org/r/20250331-work-freeze-v1-2-6dfbe8253b9f@kernel.org
Signed-off-by: Christian Brauner <brauner@kernel.org>
|
|
Now all the pieces are in place to actually allow the power subsystem
to freeze/thaw filesystems during suspend/resume. Filesystems are only
frozen and thawed if the power subsystem does actually own the freeze.
We could bubble up errors and fail suspend/resume if the error isn't
EBUSY (aka it's already frozen) but I don't think that this is worth it.
Filesystem freezing during suspend/resume is best-effort. If the user
has 500 ext4 filesystems mounted and 4 fail to freeze for whatever
reason then we simply skip them.
What we have now is already a big improvement and let's see how we fare
with it before making our lives even harder (and uglier) than we have
to.
We add a new sysctl know /sys/power/freeze_filesystems that will allow
userspace to freeze filesystems during suspend/hibernate. For now it
defaults to off. The thaw logic doesn't require checking whether
freezing is enabled because the power subsystem exclusively owns frozen
filesystems for the duration of suspend/hibernate and is able to skip
filesystems it doesn't need to freeze.
Also it is technically possible that filesystem
filesystem_freeze_enabled is true and power freezes the filesystems but
before freezing all processes another process disables
filesystem_freeze_enabled. If power were to place the filesystems_thaw()
call under filesystems_freeze_enabled it would fail to thaw the
fileystems it frozw. The exclusive holder mechanism makes it possible to
iterate through the list without any concern making sure that no
filesystems are left frozen.
Link: https://lore.kernel.org/r/20250402-work-freeze-v2-3-6719a97b52ac@kernel.org
Signed-off-by: Christian Brauner <brauner@kernel.org>
|
|
Export find_next_child() so it can be used by efivarfs.
Keep it internal for now. There's no reason to advertise this
kernel-wide.
Link: https://lore.kernel.org/r/20250331-work-freeze-v1-1-6dfbe8253b9f@kernel.org
Signed-off-by: Christian Brauner <brauner@kernel.org>
|
|
Christian Brauner <brauner@kernel.org> says:
Add the necessary infrastructure changes to support freezing for suspend
and hibernate. This should all that's needed to wire up power.
* patches from https://lore.kernel.org/r/20250329-work-freeze-v2-0-a47af37ecc3d@kernel.org:
super: add filesystem freezing helpers for suspend and hibernate
gfs2: pass through holder from the VFS for freeze/thaw
super: use common iterator (Part 2)
super: use a common iterator (Part 1)
super: skip dying superblocks early
super: simplify user_get_super()
super: remove pointless s_root checks
Link: https://lore.kernel.org/r/20250329-work-freeze-v2-0-a47af37ecc3d@kernel.org
Signed-off-by: Christian Brauner <brauner@kernel.org>
|
|
Allow the power subsystem to support filesystem freeze for
suspend and hibernate.
For some kernel subsystems it is paramount that they are guaranteed that
they are the owner of the freeze to avoid any risk of deadlocks. This is
the case for the power subsystem. Enable it to recognize whether it did
actually freeze the filesystem.
If userspace has 10 filesystems and suspend/hibernate manges to freeze 5
and then fails on the 6th for whatever odd reason (current or future)
then power needs to undo the freeze of the first 5 filesystems. It can't
just walk the list again because while it's unlikely that a new
filesystem got added in the meantime it still cannot tell which
filesystems the power subsystem actually managed to get a freeze
reference count on that needs to be dropped during thaw.
There's various ways out of this ugliness. For example, record the
filesystems the power subsystem managed to freeze on a temporary list in
the callbacks and then walk that list backwards during thaw to undo the
freezing or make sure that the power subsystem just actually exclusively
freezes things it can freeze and marking such filesystems as being owned
by power for the duration of the suspend or resume cycle. I opted for
the latter as that seemed the clean thing to do even if it means more
code changes.
If hibernation races with filesystem freezing (e.g. DM reconfiguration),
then hibernation need not freeze a filesystem because it's already
frozen but userspace may thaw the filesystem before hibernation actually
happens.
If the race happens the other way around, DM reconfiguration may
unexpectedly fail with EBUSY.
So allow FREEZE_EXCL to nest with other holders. An exclusive freezer
cannot be undone by any of the other concurrent freezers.
Link: https://lore.kernel.org/r/20250329-work-freeze-v2-6-a47af37ecc3d@kernel.org
Reviewed-by: Jan Kara <jack@suse.cz>
Signed-off-by: Christian Brauner <brauner@kernel.org>
|
|
Stop using write_cache_pages and use writeback_iter directly. This
removes an indirect call per written folio and makes the code easier
to follow.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Link: https://lore.kernel.org/20250507062124.3933305-1-hch@lst.de
Reviewed-by: Jan Kara <jack@suse.cz>
Reviewed-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Signed-off-by: Christian Brauner <brauner@kernel.org>
|
