Sometimes we want to know an accurate number of samples even if it's
lost. Currenlty PERF_RECORD_LOST is generated for a ring-buffer which
might be shared with other events. So it's hard to know per-event
lost count.
Add event->lost_samples field and PERF_FORMAT_LOST to retrieve it from
userspace.
Original-patch-by: Jiri Olsa <jolsa@redhat.com>
Signed-off-by: Namhyung Kim <namhyung@kernel.org>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lkml.kernel.org/r/20220616180623.1358843-1-namhyung@kernel.org
[Problem Statement]
select_idle_cpu() might spend too much time searching for an idle CPU,
when the system is overloaded.
The following histogram is the time spent in select_idle_cpu(),
when running 224 instances of netperf on a system with 112 CPUs
per LLC domain:
@usecs:
[0] 533 | |
[1] 5495 | |
[2, 4) 12008 | |
[4, 8) 239252 | |
[8, 16) 4041924 |@@@@@@@@@@@@@@ |
[16, 32) 12357398 |@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@ |
[32, 64) 14820255 |@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@|
[64, 128) 13047682 |@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@ |
[128, 256) 8235013 |@@@@@@@@@@@@@@@@@@@@@@@@@@@@ |
[256, 512) 4507667 |@@@@@@@@@@@@@@@ |
[512, 1K) 2600472 |@@@@@@@@@ |
[1K, 2K) 927912 |@@@ |
[2K, 4K) 218720 | |
[4K, 8K) 98161 | |
[8K, 16K) 37722 | |
[16K, 32K) 6715 | |
[32K, 64K) 477 | |
[64K, 128K) 7 | |
netperf latency usecs:
=======
case load Lat_99th std%
TCP_RR thread-224 257.39 ( 0.21)
The time spent in select_idle_cpu() is visible to netperf and might have a negative
impact.
[Symptom analysis]
The patch [1] from Mel Gorman has been applied to track the efficiency
of select_idle_sibling. Copy the indicators here:
SIS Search Efficiency(se_eff%):
A ratio expressed as a percentage of runqueues scanned versus
idle CPUs found. A 100% efficiency indicates that the target,
prev or recent CPU of a task was idle at wakeup. The lower the
efficiency, the more runqueues were scanned before an idle CPU
was found.
SIS Domain Search Efficiency(dom_eff%):
Similar, except only for the slower SIS
patch.
SIS Fast Success Rate(fast_rate%):
Percentage of SIS that used target, prev or
recent CPUs.
SIS Success rate(success_rate%):
Percentage of scans that found an idle CPU.
The test is based on Aubrey's schedtests tool, including netperf, hackbench,
schbench and tbench.
Test on vanilla kernel:
schedstat_parse.py -f netperf_vanilla.log
case load se_eff% dom_eff% fast_rate% success_rate%
TCP_RR 28 threads 99.978 18.535 99.995 100.000
TCP_RR 56 threads 99.397 5.671 99.964 100.000
TCP_RR 84 threads 21.721 6.818 73.632 100.000
TCP_RR 112 threads 12.500 5.533 59.000 100.000
TCP_RR 140 threads 8.524 4.535 49.020 100.000
TCP_RR 168 threads 6.438 3.945 40.309 99.999
TCP_RR 196 threads 5.397 3.718 32.320 99.982
TCP_RR 224 threads 4.874 3.661 25.775 99.767
UDP_RR 28 threads 99.988 17.704 99.997 100.000
UDP_RR 56 threads 99.528 5.977 99.970 100.000
UDP_RR 84 threads 24.219 6.992 76.479 100.000
UDP_RR 112 threads 13.907 5.706 62.538 100.000
UDP_RR 140 threads 9.408 4.699 52.519 100.000
UDP_RR 168 threads 7.095 4.077 44.352 100.000
UDP_RR 196 threads 5.757 3.775 35.764 99.991
UDP_RR 224 threads 5.124 3.704 28.748 99.860
schedstat_parse.py -f schbench_vanilla.log
(each group has 28 tasks)
case load se_eff% dom_eff% fast_rate% success_rate%
normal 1 mthread 99.152 6.400 99.941 100.000
normal 2 mthreads 97.844 4.003 99.908 100.000
normal 3 mthreads 96.395 2.118 99.917 99.998
normal 4 mthreads 55.288 1.451 98.615 99.804
normal 5 mthreads 7.004 1.870 45.597 61.036
normal 6 mthreads 3.354 1.346 20.777 34.230
normal 7 mthreads 2.183 1.028 11.257 21.055
normal 8 mthreads 1.653 0.825 7.849 15.549
schedstat_parse.py -f hackbench_vanilla.log
(each group has 28 tasks)
case load se_eff% dom_eff% fast_rate% success_rate%
process-pipe 1 group 99.991 7.692 99.999 100.000
process-pipe 2 groups 99.934 4.615 99.997 100.000
process-pipe 3 groups 99.597 3.198 99.987 100.000
process-pipe 4 groups 98.378 2.464 99.958 100.000
process-pipe 5 groups 27.474 3.653 89.811 99.800
process-pipe 6 groups 20.201 4.098 82.763 99.570
process-pipe 7 groups 16.423 4.156 77.398 99.316
process-pipe 8 groups 13.165 3.920 72.232 98.828
process-sockets 1 group 99.977 5.882 99.999 100.000
process-sockets 2 groups 99.927 5.505 99.996 100.000
process-sockets 3 groups 99.397 3.250 99.980 100.000
process-sockets 4 groups 79.680 4.258 98.864 99.998
process-sockets 5 groups 7.673 2.503 63.659 92.115
process-sockets 6 groups 4.642 1.584 58.946 88.048
process-sockets 7 groups 3.493 1.379 49.816 81.164
process-sockets 8 groups 3.015 1.407 40.845 75.500
threads-pipe 1 group 99.997 0.000 100.000 100.000
threads-pipe 2 groups 99.894 2.932 99.997 100.000
threads-pipe 3 groups 99.611 4.117 99.983 100.000
threads-pipe 4 groups 97.703 2.624 99.937 100.000
threads-pipe 5 groups 22.919 3.623 87.150 99.764
threads-pipe 6 groups 18.016 4.038 80.491 99.557
threads-pipe 7 groups 14.663 3.991 75.239 99.247
threads-pipe 8 groups 12.242 3.808 70.651 98.644
threads-sockets 1 group 99.990 6.667 99.999 100.000
threads-sockets 2 groups 99.940 5.114 99.997 100.000
threads-sockets 3 groups 99.469 4.115 99.977 100.000
threads-sockets 4 groups 87.528 4.038 99.400 100.000
threads-sockets 5 groups 6.942 2.398 59.244 88.337
threads-sockets 6 groups 4.359 1.954 49.448 87.860
threads-sockets 7 groups 2.845 1.345 41.198 77.102
threads-sockets 8 groups 2.871 1.404 38.512 74.312
schedstat_parse.py -f tbench_vanilla.log
case load se_eff% dom_eff% fast_rate% success_rate%
loopback 28 threads 99.976 18.369 99.995 100.000
loopback 56 threads 99.222 7.799 99.934 100.000
loopback 84 threads 19.723 6.819 70.215 100.000
loopback 112 threads 11.283 5.371 55.371 99.999
loopback 140 threads 0.000 0.000 0.000 0.000
loopback 168 threads 0.000 0.000 0.000 0.000
loopback 196 threads 0.000 0.000 0.000 0.000
loopback 224 threads 0.000 0.000 0.000 0.000
According to the test above, if the system becomes busy, the
SIS Search Efficiency(se_eff%) drops significantly. Although some
benchmarks would finally find an idle CPU(success_rate% = 100%), it is
doubtful whether it is worth it to search the whole LLC domain.
[Proposal]
It would be ideal to have a crystal ball to answer this question:
How many CPUs must a wakeup path walk down, before it can find an idle
CPU? Many potential metrics could be used to predict the number.
One candidate is the sum of util_avg in this LLC domain. The benefit
of choosing util_avg is that it is a metric of accumulated historic
activity, which seems to be smoother than instantaneous metrics
(such as rq->nr_running). Besides, choosing the sum of util_avg
would help predict the load of the LLC domain more precisely, because
SIS_PROP uses one CPU's idle time to estimate the total LLC domain idle
time.
In summary, the lower the util_avg is, the more select_idle_cpu()
should scan for idle CPU, and vice versa. When the sum of util_avg
in this LLC domain hits 85% or above, the scan stops. The reason to
choose 85% as the threshold is that this is the imbalance_pct(117)
when a LLC sched group is overloaded.
Introduce the quadratic function:
y = SCHED_CAPACITY_SCALE - p * x^2
and y'= y / SCHED_CAPACITY_SCALE
x is the ratio of sum_util compared to the CPU capacity:
x = sum_util / (llc_weight * SCHED_CAPACITY_SCALE)
y' is the ratio of CPUs to be scanned in the LLC domain,
and the number of CPUs to scan is calculated by:
nr_scan = llc_weight * y'
Choosing quadratic function is because:
[1] Compared to the linear function, it scans more aggressively when the
sum_util is low.
[2] Compared to the exponential function, it is easier to calculate.
[3] It seems that there is no accurate mapping between the sum of util_avg
and the number of CPUs to be scanned. Use heuristic scan for now.
For a platform with 112 CPUs per LLC, the number of CPUs to scan is:
sum_util% 0 5 15 25 35 45 55 65 75 85 86 ...
scan_nr 112 111 108 102 93 81 65 47 25 1 0 ...
For a platform with 16 CPUs per LLC, the number of CPUs to scan is:
sum_util% 0 5 15 25 35 45 55 65 75 85 86 ...
scan_nr 16 15 15 14 13 11 9 6 3 0 0 ...
Furthermore, to minimize the overhead of calculating the metrics in
select_idle_cpu(), borrow the statistics from periodic load balance.
As mentioned by Abel, on a platform with 112 CPUs per LLC, the
sum_util calculated by periodic load balance after 112 ms would
decay to about 0.5 * 0.5 * 0.5 * 0.7 = 8.75%, thus bringing a delay
in reflecting the latest utilization. But it is a trade-off.
Checking the util_avg in newidle load balance would be more frequent,
but it brings overhead - multiple CPUs write/read the per-LLC shared
variable and introduces cache contention. Tim also mentioned that,
it is allowed to be non-optimal in terms of scheduling for the
short-term variations, but if there is a long-term trend in the load
behavior, the scheduler can adjust for that.
When SIS_UTIL is enabled, the select_idle_cpu() uses the nr_scan
calculated by SIS_UTIL instead of the one from SIS_PROP. As Peter and
Mel suggested, SIS_UTIL should be enabled by default.
This patch is based on the util_avg, which is very sensitive to the
CPU frequency invariance. There is an issue that, when the max frequency
has been clamp, the util_avg would decay insanely fast when
the CPU is idle. Commit addca28512 ("cpufreq: intel_pstate: Handle no_turbo
in frequency invariance") could be used to mitigate this symptom, by adjusting
the arch_max_freq_ratio when turbo is disabled. But this issue is still
not thoroughly fixed, because the current code is unaware of the user-specified
max CPU frequency.
[Test result]
netperf and tbench were launched with 25% 50% 75% 100% 125% 150%
175% 200% of CPU number respectively. Hackbench and schbench were launched
by 1, 2 ,4, 8 groups. Each test lasts for 100 seconds and repeats 3 times.
The following is the benchmark result comparison between
baseline:vanilla v5.19-rc1 and compare:patched kernel. Positive compare%
indicates better performance.
Each netperf test is a:
netperf -4 -H 127.0.1 -t TCP/UDP_RR -c -C -l 100
netperf.throughput
=======
case load baseline(std%) compare%( std%)
TCP_RR 28 threads 1.00 ( 0.34) -0.16 ( 0.40)
TCP_RR 56 threads 1.00 ( 0.19) -0.02 ( 0.20)
TCP_RR 84 threads 1.00 ( 0.39) -0.47 ( 0.40)
TCP_RR 112 threads 1.00 ( 0.21) -0.66 ( 0.22)
TCP_RR 140 threads 1.00 ( 0.19) -0.69 ( 0.19)
TCP_RR 168 threads 1.00 ( 0.18) -0.48 ( 0.18)
TCP_RR 196 threads 1.00 ( 0.16) +194.70 ( 16.43)
TCP_RR 224 threads 1.00 ( 0.16) +197.30 ( 7.85)
UDP_RR 28 threads 1.00 ( 0.37) +0.35 ( 0.33)
UDP_RR 56 threads 1.00 ( 11.18) -0.32 ( 0.21)
UDP_RR 84 threads 1.00 ( 1.46) -0.98 ( 0.32)
UDP_RR 112 threads 1.00 ( 28.85) -2.48 ( 19.61)
UDP_RR 140 threads 1.00 ( 0.70) -0.71 ( 14.04)
UDP_RR 168 threads 1.00 ( 14.33) -0.26 ( 11.16)
UDP_RR 196 threads 1.00 ( 12.92) +186.92 ( 20.93)
UDP_RR 224 threads 1.00 ( 11.74) +196.79 ( 18.62)
Take the 224 threads as an example, the SIS search metrics changes are
illustrated below:
vanilla patched
4544492 +237.5% 15338634 sched_debug.cpu.sis_domain_search.avg
38539 +39686.8% 15333634 sched_debug.cpu.sis_failed.avg
128300000 -87.9% 15551326 sched_debug.cpu.sis_scanned.avg
5842896 +162.7% 15347978 sched_debug.cpu.sis_search.avg
There is -87.9% less CPU scans after patched, which indicates lower overhead.
Besides, with this patch applied, there is -13% less rq lock contention
in perf-profile.calltrace.cycles-pp._raw_spin_lock.raw_spin_rq_lock_nested
.try_to_wake_up.default_wake_function.woken_wake_function.
This might help explain the performance improvement - Because this patch allows
the waking task to remain on the previous CPU, rather than grabbing other CPUs'
lock.
Each hackbench test is a:
hackbench -g $job --process/threads --pipe/sockets -l 1000000 -s 100
hackbench.throughput
=========
case load baseline(std%) compare%( std%)
process-pipe 1 group 1.00 ( 1.29) +0.57 ( 0.47)
process-pipe 2 groups 1.00 ( 0.27) +0.77 ( 0.81)
process-pipe 4 groups 1.00 ( 0.26) +1.17 ( 0.02)
process-pipe 8 groups 1.00 ( 0.15) -4.79 ( 0.02)
process-sockets 1 group 1.00 ( 0.63) -0.92 ( 0.13)
process-sockets 2 groups 1.00 ( 0.03) -0.83 ( 0.14)
process-sockets 4 groups 1.00 ( 0.40) +5.20 ( 0.26)
process-sockets 8 groups 1.00 ( 0.04) +3.52 ( 0.03)
threads-pipe 1 group 1.00 ( 1.28) +0.07 ( 0.14)
threads-pipe 2 groups 1.00 ( 0.22) -0.49 ( 0.74)
threads-pipe 4 groups 1.00 ( 0.05) +1.88 ( 0.13)
threads-pipe 8 groups 1.00 ( 0.09) -4.90 ( 0.06)
threads-sockets 1 group 1.00 ( 0.25) -0.70 ( 0.53)
threads-sockets 2 groups 1.00 ( 0.10) -0.63 ( 0.26)
threads-sockets 4 groups 1.00 ( 0.19) +11.92 ( 0.24)
threads-sockets 8 groups 1.00 ( 0.08) +4.31 ( 0.11)
Each tbench test is a:
tbench -t 100 $job 127.0.0.1
tbench.throughput
======
case load baseline(std%) compare%( std%)
loopback 28 threads 1.00 ( 0.06) -0.14 ( 0.09)
loopback 56 threads 1.00 ( 0.03) -0.04 ( 0.17)
loopback 84 threads 1.00 ( 0.05) +0.36 ( 0.13)
loopback 112 threads 1.00 ( 0.03) +0.51 ( 0.03)
loopback 140 threads 1.00 ( 0.02) -1.67 ( 0.19)
loopback 168 threads 1.00 ( 0.38) +1.27 ( 0.27)
loopback 196 threads 1.00 ( 0.11) +1.34 ( 0.17)
loopback 224 threads 1.00 ( 0.11) +1.67 ( 0.22)
Each schbench test is a:
schbench -m $job -t 28 -r 100 -s 30000 -c 30000
schbench.latency_90%_us
========
case load baseline(std%) compare%( std%)
normal 1 mthread 1.00 ( 31.22) -7.36 ( 20.25)*
normal 2 mthreads 1.00 ( 2.45) -0.48 ( 1.79)
normal 4 mthreads 1.00 ( 1.69) +0.45 ( 0.64)
normal 8 mthreads 1.00 ( 5.47) +9.81 ( 14.28)
*Consider the Standard Deviation, this -7.36% regression might not be valid.
Also, a OLTP workload with a commercial RDBMS has been tested, and there
is no significant change.
There were concerns that unbalanced tasks among CPUs would cause problems.
For example, suppose the LLC domain is composed of 8 CPUs, and 7 tasks are
bound to CPU0~CPU6, while CPU7 is idle:
CPU0 CPU1 CPU2 CPU3 CPU4 CPU5 CPU6 CPU7
util_avg 1024 1024 1024 1024 1024 1024 1024 0
Since the util_avg ratio is 87.5%( = 7/8 ), which is higher than 85%,
select_idle_cpu() will not scan, thus CPU7 is undetected during scan.
But according to Mel, it is unlikely the CPU7 will be idle all the time
because CPU7 could pull some tasks via CPU_NEWLY_IDLE.
lkp(kernel test robot) has reported a regression on stress-ng.sock on a
very busy system. According to the sched_debug statistics, it might be caused
by SIS_UTIL terminates the scan and chooses a previous CPU earlier, and this
might introduce more context switch, especially involuntary preemption, which
impacts a busy stress-ng. This regression has shown that, not all benchmarks
in every scenario benefit from idle CPU scan limit, and it needs further
investigation.
Besides, there is slight regression in hackbench's 16 groups case when the
LLC domain has 16 CPUs. Prateek mentioned that we should scan aggressively
in an LLC domain with 16 CPUs. Because the cost to search for an idle one
among 16 CPUs is negligible. The current patch aims to propose a generic
solution and only considers the util_avg. Something like the below could
be applied on top of the current patch to fulfill the requirement:
if (llc_weight <= 16)
nr_scan = nr_scan * 32 / llc_weight;
For LLC domain with 16 CPUs, the nr_scan will be expanded to 2 times large.
The smaller the CPU number this LLC domain has, the larger nr_scan will be
expanded. This needs further investigation.
There is also ongoing work[2] from Abel to filter out the busy CPUs during
wakeup, to further speed up the idle CPU scan. And it could be a following-up
optimization on top of this change.
Suggested-by: Tim Chen <tim.c.chen@intel.com>
Suggested-by: Peter Zijlstra <peterz@infradead.org>
Signed-off-by: Chen Yu <yu.c.chen@intel.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Tested-by: Yicong Yang <yangyicong@hisilicon.com>
Tested-by: Mohini Narkhede <mohini.narkhede@intel.com>
Tested-by: K Prateek Nayak <kprateek.nayak@amd.com>
Link: https://lore.kernel.org/r/20220612163428.849378-1-yu.c.chen@intel.com
sched_setattr(2) issues via kernel/sched/core.c:__sched_setscheduler()
a CAP_SYS_NICE audit event unconditionally, even when the requested
operation does not require that capability / is unprivileged, i.e. for
reducing niceness.
This is relevant in connection with SELinux, where a capability check
results in a policy decision and by default a denial message on
insufficient permission is issued.
It can lead to three undesired cases:
1. A denial message is generated, even in case the operation was an
unprivileged one and thus the syscall succeeded, creating noise.
2. To avoid the noise from 1. the policy writer adds a rule to ignore
those denial messages, hiding future syscalls, where the task
performs an actual privileged operation, leading to hidden limited
functionality of that task.
3. To avoid the noise from 1. the policy writer adds a rule to allow
the task the capability CAP_SYS_NICE, while it does not need it,
violating the principle of least privilege.
Conduct privilged/unprivileged categorization first and perform a
capable test (and at most once) only if needed.
Signed-off-by: Christian Göttsche <cgzones@googlemail.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lkml.kernel.org/r/20220615152505.310488-1-cgzones@googlemail.com
EXPORT_SYMBOL and __init is a bad combination because the .init.text
section is freed up after the initialization. Hence, modules cannot
use symbols annotated __init. The access to a freed symbol may end up
with kernel panic.
modpost used to detect it, but it had been broken for a decade.
Commit 28438794ab ("modpost: fix section mismatch check for exported
init/exit sections") fixed it so modpost started to warn it again, then
this showed up:
MODPOST vmlinux.symvers
WARNING: modpost: vmlinux.o(___ksymtab_gpl+tick_nohz_full_setup+0x0): Section mismatch in reference from the variable __ksymtab_tick_nohz_full_setup to the function .init.text:tick_nohz_full_setup()
The symbol tick_nohz_full_setup is exported and annotated __init
Fix this by removing the __init annotation of tick_nohz_full_setup or drop the export.
Drop the export because tick_nohz_full_setup() is only called from the
built-in code in kernel/sched/isolation.c.
Fixes: ae9e557b5b ("time: Export tick start/stop functions for rcutorture")
Reported-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Masahiro Yamada <masahiroy@kernel.org>
Tested-by: Paul E. McKenney <paulmck@kernel.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Pull hotfixes from Andrew Morton:
"Minor things, mainly - mailmap updates, MAINTAINERS updates, etc.
Fixes for this merge window:
- fix for a damon boot hang, from SeongJae
- fix for a kfence warning splat, from Jason Donenfeld
- fix for zero-pfn pinning, from Alex Williamson
- fix for fallocate hole punch clearing, from Mike Kravetz
Fixes for previous releases:
- fix for a performance regression, from Marcelo
- fix for a hwpoisining BUG from zhenwei pi"
* tag 'mm-hotfixes-stable-2022-06-26' of git://git.kernel.org/pub/scm/linux/kernel/git/akpm/mm:
mailmap: add entry for Christian Marangi
mm/memory-failure: disable unpoison once hw error happens
hugetlbfs: zero partial pages during fallocate hole punch
mm: memcontrol: reference to tools/cgroup/memcg_slabinfo.py
mm: re-allow pinning of zero pfns
mm/kfence: select random number before taking raw lock
MAINTAINERS: add maillist information for LoongArch
MAINTAINERS: update MM tree references
MAINTAINERS: update Abel Vesa's email
MAINTAINERS: add MEMORY HOT(UN)PLUG section and add David as reviewer
MAINTAINERS: add Miaohe Lin as a memory-failure reviewer
mailmap: add alias for jarkko@profian.com
mm/damon/reclaim: schedule 'damon_reclaim_timer' only after 'system_wq' is initialized
kthread: make it clear that kthread_create_on_node() might be terminated by any fatal signal
mm: lru_cache_disable: use synchronize_rcu_expedited
mm/page_isolation.c: fix one kernel-doc comment
Pull dma-mapping fix from Christoph Hellwig:
- pass the correct size to dma_set_encrypted() when freeing memory
(Dexuan Cui)
* tag 'dma-mapping-5.19-2022-06-26' of git://git.infradead.org/users/hch/dma-mapping:
dma-direct: use the correct size for dma_set_encrypted()
Pull block fixes from Jens Axboe:
- Series fixing issues with sysfs locking and name reuse (Christoph)
- NVMe pull request via Christoph:
- Fix the mixed up CRIMS/CRWMS constants (Joel Granados)
- Add another broken identifier quirk (Leo Savernik)
- Fix up a quirk because Samsung reuses PCI IDs over different
products (Christoph Hellwig)
- Remove old WARN_ON() that doesn't apply anymore (Li)
- Fix for using a stale cached request value for rq-qos throttling
mechanisms that may schedule(), like iocost (me)
- Remove unused parameter to blk_independent_access_range() (Damien)
* tag 'block-5.19-2022-06-24' of git://git.kernel.dk/linux-block:
block: remove WARN_ON() from bd_link_disk_holder
nvme: move the Samsung X5 quirk entry to the core quirks
nvme: fix the CRIMS and CRWMS definitions to match the spec
nvme: add a bogus subsystem NQN quirk for Micron MTFDKBA2T0TFH
block: pop cached rq before potentially blocking rq_qos_throttle()
block: remove queue from struct blk_independent_access_range
block: freeze the queue earlier in del_gendisk
block: remove per-disk debugfs files in blk_unregister_queue
block: serialize all debugfs operations using q->debugfs_mutex
block: disable the elevator int del_gendisk
Pull printk kernel thread revert from Petr Mladek:
"Revert printk console kthreads.
The testing of 5.19 release candidates revealed issues that did not
happen when all consoles were serialized using the console semaphore.
More time is needed to check expectations of the existing console
drivers and be confident that they can be safely used in parallel"
* tag 'printk-for-5.19-rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/printk/linux:
Revert "printk: add functions to prefer direct printing"
Revert "printk: add kthread console printers"
Revert "printk: extend console_lock for per-console locking"
Revert "printk: remove @console_locked"
Revert "printk: Block console kthreads when direct printing will be required"
Revert "printk: Wait for the global console lock when the system is going down"
Commit dfd5e3f5fe ("locking/lockdep: Mark local_lock_t") added yet
another lockdep_init_map_*() variant, but forgot to update all the
existing users of the most complicated version.
This could lead to a loss of lock_type and hence an incorrect report.
Given the relative rarity of both local_lock and these annotations,
this is unlikely to happen in practise, still, best fix things.
Fixes: dfd5e3f5fe ("locking/lockdep: Mark local_lock_t")
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lkml.kernel.org/r/YqyEDtoan20K0CVD@worktop.programming.kicks-ass.net
Instead of defaulting to patching NOP opcodes at init time, and leaving
it to the architectures to override this if this is not needed, switch
to a model where doing nothing is the default. This is the common case
by far, as only MIPS requires NOP patching at init time. On all other
architectures, the correct encodings are emitted by the compiler and so
no initial patching is needed.
Signed-off-by: Ard Biesheuvel <ardb@kernel.org>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Acked-by: Mark Rutland <mark.rutland@arm.com>
Acked-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lore.kernel.org/r/20220615154142.1574619-4-ardb@kernel.org
MIPS is the only remaining architecture that needs to patch jump label
NOP encodings to initialize them at load time. So let's move the module
patching part of that from generic code into arch/mips, and drop it from
the others.
Signed-off-by: Ard Biesheuvel <ardb@kernel.org>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lore.kernel.org/r/20220615154142.1574619-3-ardb@kernel.org
Pull power management fix from Rafael Wysocki:
"Fix a recent regression preventing some systems from powering off
after saving a hibernation image (Dmitry Osipenko)"
* tag 'pm-5.19-rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/rafael/linux-pm:
PM: hibernate: Use kernel_can_power_off()
Pull tracing fixes from Steven Rostedt:
- Check for NULL in kretprobe_dispatcher()
NULL can now be passed in, make sure it can handle it
- Clean up unneeded #endif #ifdef of the same preprocessor
check in the middle of the block.
- Comment clean up
- Remove unneeded initialization of the "ret" variable in
__trace_uprobe_create()
* tag 'trace-v5.19-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/rostedt/linux-trace:
tracing/uprobes: Remove unwanted initialization in __trace_uprobe_create()
tracefs: Fix syntax errors in comments
tracing: Simplify conditional compilation code in tracing_set_tracer()
tracing/kprobes: Check whether get_kretprobe() returns NULL in kretprobe_dispatcher()
This reverts commit 2bb2b7b57f.
The testing of 5.19 release candidates revealed missing synchronization
between early and regular console functionality.
It would be possible to start the console kthreads later as a workaround.
But it is clear that console lock serialized console drivers between
each other. It opens a big area of possible problems that were not
considered by people involved in the development and review.
printk() is crucial for debugging kernel issues and console output is
very important part of it. The number of consoles is huge and a proper
review would take some time. As a result it need to be reverted for 5.19.
Link: https://lore.kernel.org/r/YrBdjVwBOVgLfHyb@alley
Signed-off-by: Petr Mladek <pmladek@suse.com>
Link: https://lore.kernel.org/r/20220623145157.21938-7-pmladek@suse.com
This reverts commit 09c5ba0aa2.
This reverts commit b87f02307d.
The testing of 5.19 release candidates revealed missing synchronization
between early and regular console functionality.
It would be possible to start the console kthreads later as a workaround.
But it is clear that console lock serialized console drivers between
each other. It opens a big area of possible problems that were not
considered by people involved in the development and review.
printk() is crucial for debugging kernel issues and console output is
very important part of it. The number of consoles is huge and a proper
review would take some time. As a result it need to be reverted for 5.19.
Link: https://lore.kernel.org/r/YrBdjVwBOVgLfHyb@alley
Signed-off-by: Petr Mladek <pmladek@suse.com>
Link: https://lore.kernel.org/r/20220623145157.21938-6-pmladek@suse.com
This reverts commit 8e27473211.
The testing of 5.19 release candidates revealed missing synchronization
between early and regular console functionality.
It would be possible to start the console kthreads later as a workaround.
But it is clear that console lock serialized console drivers between
each other. It opens a big area of possible problems that were not
considered by people involved in the development and review.
printk() is crucial for debugging kernel issues and console output is
very important part of it. The number of consoles is huge and a proper
review would take some time. As a result it need to be reverted for 5.19.
Link: https://lore.kernel.org/r/YrBdjVwBOVgLfHyb@alley
Signed-off-by: Petr Mladek <pmladek@suse.com>
Link: https://lore.kernel.org/r/20220623145157.21938-5-pmladek@suse.com
This reverts commit ab406816fc.
The testing of 5.19 release candidates revealed missing synchronization
between early and regular console functionality.
It would be possible to start the console kthreads later as a workaround.
But it is clear that console lock serialized console drivers between
each other. It opens a big area of possible problems that were not
considered by people involved in the development and review.
printk() is crucial for debugging kernel issues and console output is
very important part of it. The number of consoles is huge and a proper
review would take some time. As a result it need to be reverted for 5.19.
Link: https://lore.kernel.org/r/YrBdjVwBOVgLfHyb@alley
Signed-off-by: Petr Mladek <pmladek@suse.com>
Link: https://lore.kernel.org/r/20220623145157.21938-4-pmladek@suse.com
This reverts commit c3230283e2.
The testing of 5.19 release candidates revealed missing synchronization
between early and regular console functionality.
It would be possible to start the console kthreads later as a workaround.
But it is clear that console lock serialized console drivers between
each other. It opens a big area of possible problems that were not
considered by people involved in the development and review.
printk() is crucial for debugging kernel issues and console output is
very important part of it. The number of consoles is huge and a proper
review would take some time. As a result it need to be reverted for 5.19.
Link: https://lore.kernel.org/r/YrBdjVwBOVgLfHyb@alley
Signed-off-by: Petr Mladek <pmladek@suse.com>
Link: https://lore.kernel.org/r/20220623145157.21938-3-pmladek@suse.com
This reverts commit b87f02307d.
The testing of 5.19 release candidates revealed missing synchronization
between early and regular console functionality.
It would be possible to start the console kthreads later as a workaround.
But it is clear that console lock serialized console drivers between
each other. It opens a big area of possible problems that were not
considered by people involved in the development and review.
printk() is crucial for debugging kernel issues and console output is
very important part of it. The number of consoles is huge and a proper
review would take some time. As a result it need to be reverted for 5.19.
Link: https://lore.kernel.org/r/YrBdjVwBOVgLfHyb@alley
Signed-off-by: Petr Mladek <pmladek@suse.com>
Link: https://lore.kernel.org/r/20220623145157.21938-2-pmladek@suse.com
Pull networking fixes from Paolo Abeni:
"Including fixes from bpf and netfilter.
Current release - regressions:
- netfilter: cttimeout: fix slab-out-of-bounds read in
cttimeout_net_exit
Current release - new code bugs:
- bpf: ftrace: keep address offset in ftrace_lookup_symbols
- bpf: force cookies array to follow symbols sorting
Previous releases - regressions:
- ipv4: ping: fix bind address validity check
- tipc: fix use-after-free read in tipc_named_reinit
- eth: veth: add updating of trans_start
Previous releases - always broken:
- sock: redo the psock vs ULP protection check
- netfilter: nf_dup_netdev: fix skb_under_panic
- bpf: fix request_sock leak in sk lookup helpers
- eth: igb: fix a use-after-free issue in igb_clean_tx_ring
- eth: ice: prohibit improper channel config for DCB
- eth: at803x: fix null pointer dereference on AR9331 phy
- eth: virtio_net: fix xdp_rxq_info bug after suspend/resume
Misc:
- eth: hinic: replace memcpy() with direct assignment"
* tag 'net-5.19-rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net: (47 commits)
net: openvswitch: fix parsing of nw_proto for IPv6 fragments
sock: redo the psock vs ULP protection check
Revert "net/tls: fix tls_sk_proto_close executed repeatedly"
virtio_net: fix xdp_rxq_info bug after suspend/resume
igb: Make DMA faster when CPU is active on the PCIe link
net: dsa: qca8k: reduce mgmt ethernet timeout
net: dsa: qca8k: reset cpu port on MTU change
MAINTAINERS: Add a maintainer for OCP Time Card
hinic: Replace memcpy() with direct assignment
Revert "drivers/net/ethernet/neterion/vxge: Fix a use-after-free bug in vxge-main.c"
net: phy: smsc: Disable Energy Detect Power-Down in interrupt mode
ice: ethtool: Prohibit improper channel config for DCB
ice: ethtool: advertise 1000M speeds properly
ice: Fix switchdev rules book keeping
ice: ignore protocol field in GTP offload
netfilter: nf_dup_netdev: add and use recursion counter
netfilter: nf_dup_netdev: do not push mac header a second time
selftests: netfilter: correct PKTGEN_SCRIPT_PATHS in nft_concat_range.sh
net/tls: fix tls_sk_proto_close executed repeatedly
erspan: do not assume transport header is always set
...
The third parameter of dma_set_encrypted() is a size in bytes rather than
the number of pages.
Fixes: 4d0564785b ("dma-direct: factor out dma_set_{de,en}crypted helpers")
Signed-off-by: Dexuan Cui <decui@microsoft.com>
Reviewed-by: Robin Murphy <robin.murphy@arm.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Use new kernel_can_power_off() API instead of legacy pm_power_off global
variable to fix regressed hibernation to disk where machine no longer
powers off when it should because ACPI power driver transitioned to the
new sys-off based API and it doesn't use pm_power_off anymore.
Fixes: 98f30d0ecf ("ACPI: power: Switch to sys-off handler API")
Tested-by: Ken Moffat <zarniwhoop@ntlworld.com>
Reported-by: Ken Moffat <zarniwhhop@ntlworld.com>
Signed-off-by: Dmitry Osipenko <dmitry.osipenko@collabora.com>
Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Pull scheduler fix from Thomas Gleixner:
"A single scheduler fix plugging a race between sched_setscheduler()
and balance_push().
sched_setscheduler() spliced the balance callbacks accross a lock
break which makes it possible for an interleaving schedule() to
observe an empty list"
* tag 'sched-urgent-2022-06-19' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
sched: Fix balance_push() vs __sched_setscheduler()
Pull lockdep fix from Thomas Gleixner:
"A RT fix for lockdep.
lockdep invokes prandom_u32() to create cookies. This worked until
prandom_u32() was switched to the real random generator, which takes a
spinlock for extraction, which does not work on RT when invoked from
atomic contexts.
lockdep has no requirement for real random numbers and it turns out
sched_clock() is good enough to create the cookie. That works
everywhere and is faster"
* tag 'locking-urgent-2022-06-19' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
locking/lockdep: Use sched_clock() for random numbers
Pull irq fixes from Thomas Gleixner:
"A set of interrupt subsystem updates:
Core:
- Ensure runtime power management for chained interrupts
Drivers:
- A collection of OF node refcount fixes
- Unbreak MIPS uniprocessor builds
- Fix xilinx interrupt controller Kconfig dependencies
- Add a missing compatible string to the Uniphier driver"
* tag 'irq-urgent-2022-06-19' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
irqchip/loongson-liointc: Use architecture register to get coreid
irqchip/uniphier-aidet: Add compatible string for NX1 SoC
dt-bindings: interrupt-controller/uniphier-aidet: Add bindings for NX1 SoC
irqchip/realtek-rtl: Fix refcount leak in map_interrupts
irqchip/gic-v3: Fix refcount leak in gic_populate_ppi_partitions
irqchip/gic-v3: Fix error handling in gic_populate_ppi_partitions
irqchip/apple-aic: Fix refcount leak in aic_of_ic_init
irqchip/apple-aic: Fix refcount leak in build_fiq_affinity
irqchip/gic/realview: Fix refcount leak in realview_gic_of_init
irqchip/xilinx: Remove microblaze+zynq dependency
genirq: PM: Use runtime PM for chained interrupts
Daniel Borkmann says:
====================
pull-request: bpf 2022-06-17
We've added 12 non-merge commits during the last 4 day(s) which contain
a total of 14 files changed, 305 insertions(+), 107 deletions(-).
The main changes are:
1) Fix x86 JIT tailcall count offset on BPF-2-BPF call, from Jakub Sitnicki.
2) Fix a kprobe_multi link bug which misplaces BPF cookies, from Jiri Olsa.
3) Fix an infinite loop when processing a module's BTF, from Kumar Kartikeya Dwivedi.
4) Fix getting a rethook only in RCU available context, from Masami Hiramatsu.
5) Fix request socket refcount leak in sk lookup helpers, from Jon Maxwell.
6) Fix xsk xmit behavior which wrongly adds skb to already full cq, from Ciara Loftus.
* https://git.kernel.org/pub/scm/linux/kernel/git/bpf/bpf:
rethook: Reject getting a rethook if RCU is not watching
fprobe, samples: Add use_trace option and show hit/missed counter
bpf, docs: Update some of the JIT/maintenance entries
selftest/bpf: Fix kprobe_multi bench test
bpf: Force cookies array to follow symbols sorting
ftrace: Keep address offset in ftrace_lookup_symbols
selftests/bpf: Shuffle cookies symbols in kprobe multi test
selftests/bpf: Test tail call counting with bpf2bpf and data on stack
bpf, x86: Fix tail call count offset calculation on bpf2bpf call
bpf: Limit maximum modifier chain length in btf_check_type_tags
bpf: Fix request_sock leak in sk lookup helpers
xsk: Fix generic transmit when completion queue reservation fails
====================
Link: https://lore.kernel.org/r/20220617202119.2421-1-daniel@iogearbox.net
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
There is a small chance that get_kretprobe(ri) returns NULL in
kretprobe_dispatcher() when another CPU unregisters the kretprobe
right after __kretprobe_trampoline_handler().
To avoid this issue, kretprobe_dispatcher() checks the get_kretprobe()
return value again. And if it is NULL, it returns soon because that
kretprobe is under unregistering process.
This issue has been introduced when the kretprobe is decoupled
from the struct kretprobe_instance by commit d741bf41d7
("kprobes: Remove kretprobe hash"). Before that commit, the
struct kretprob_instance::rp directly points the kretprobe
and it is never be NULL.
Link: https://lkml.kernel.org/r/165366693881.797669.16926184644089588731.stgit@devnote2
Reported-by: Yonghong Song <yhs@fb.com>
Fixes: d741bf41d7 ("kprobes: Remove kretprobe hash")
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Ingo Molnar <mingo@kernel.org>
Cc: bpf <bpf@vger.kernel.org>
Cc: Kernel Team <kernel-team@fb.com>
Cc: stable@vger.kernel.org
Signed-off-by: Masami Hiramatsu (Google) <mhiramat@kernel.org>
Acked-by: Jiri Olsa <jolsa@kernel.org>
Signed-off-by: Steven Rostedt (Google) <rostedt@goodmis.org>
Pull printk fixes from Petr Mladek:
"Make the global console_sem available for CPU that is handling panic()
or shutdown.
This is an old problem when an existing console lock owner might block
console output, but it became more visible with the kthreads"
* tag 'printk-for-5.19-rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/printk/linux:
printk: Wait for the global console lock when the system is going down
printk: Block console kthreads when direct printing will be required
Since the rethook_recycle() will involve the call_rcu() for reclaiming
the rethook_instance, the rethook must be set up at the RCU available
context (non idle). This rethook_recycle() in the rethook trampoline
handler is inevitable, thus the RCU available check must be done before
setting the rethook trampoline.
This adds a rcu_is_watching() check in the rethook_try_get() so that
it will return NULL if it is called when !rcu_is_watching().
Fixes: 54ecbe6f1e ("rethook: Add a generic return hook")
Signed-off-by: Masami Hiramatsu (Google) <mhiramat@kernel.org>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: Steven Rostedt (Google) <rostedt@goodmis.org>
Acked-by: Jiri Olsa <jolsa@kernel.org>
Link: https://lore.kernel.org/bpf/165461827269.280167.7379263615545598958.stgit@devnote2
Various places like I/O schedulers or the QOS infrastructure try to
register debugfs files on demans, which can race with creating and
removing the main queue debugfs directory. Use the existing
debugfs_mutex to serialize all debugfs operations that rely on
q->debugfs_dir or the directories hanging off it.
To make the teardown code a little simpler declare all debugfs dentry
pointers and not just the main one uncoditionally in blkdev.h.
Move debugfs_mutex next to the dentries that it protects and document
what it is used for.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Link: https://lore.kernel.org/r/20220614074827.458955-3-hch@lst.de
Signed-off-by: Jens Axboe <axboe@kernel.dk>
When user specifies symbols and cookies for kprobe_multi link
interface it's very likely the cookies will be misplaced and
returned to wrong functions (via get_attach_cookie helper).
The reason is that to resolve the provided functions we sort
them before passing them to ftrace_lookup_symbols, but we do
not do the same sort on the cookie values.
Fixing this by using sort_r function with custom swap callback
that swaps cookie values as well.
Fixes: 0236fec57a ("bpf: Resolve symbols with ftrace_lookup_symbols for kprobe multi link")
Signed-off-by: Jiri Olsa <jolsa@kernel.org>
Link: https://lore.kernel.org/r/20220615112118.497303-4-jolsa@kernel.org
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
The comments in kernel/kthread.c create a feeling that only SIGKILL is
able to terminate the creation of kernel kthreads by
kthread_create()/_on_node()/_on_cpu() APIs.
In reality, wait_for_completion_killable() might be killed by any fatal
signal that does not have a custom handler:
(!siginmask(signr, SIG_KERNEL_IGNORE_MASK|SIG_KERNEL_STOP_MASK) && \
(t)->sighand->action[(signr)-1].sa.sa_handler == SIG_DFL)
static inline void signal_wake_up(struct task_struct *t, bool resume)
{
signal_wake_up_state(t, resume ? TASK_WAKEKILL : 0);
}
static void complete_signal(int sig, struct task_struct *p, enum pid_type type)
{
[...]
/*
* Found a killable thread. If the signal will be fatal,
* then start taking the whole group down immediately.
*/
if (sig_fatal(p, sig) ...) {
if (!sig_kernel_coredump(sig)) {
[...]
do {
task_clear_jobctl_pending(t, JOBCTL_PENDING_MASK);
sigaddset(&t->pending.signal, SIGKILL);
signal_wake_up(t, 1);
} while_each_thread(p, t);
return;
}
}
}
Update the comments in kernel/kthread.c to make this more obvious.
The motivation for this change was debugging why a module initialization
failed. The module was being loaded from initrd. It "magically" failed
when systemd was switching to the real root. The clean up operations sent
SIGTERM to various pending processed that were started from initrd.
Link: https://lkml.kernel.org/r/20220315102444.2380-1-pmladek@suse.com
Signed-off-by: Petr Mladek <pmladek@suse.com>
Reviewed-by: "Eric W. Biederman" <ebiederm@xmission.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Mathieu Desnoyers <mathieu.desnoyers@polymtl.ca>
Cc: Kees Cook <keescook@chromium.org>
Cc: Marco Elver <elver@google.com>
Cc: Jens Axboe <axboe@kernel.dk>
Cc: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Pull audit fix from Paul Moore:
"A single audit patch to fix a problem where we were not properly
freeing memory allocated when recording information related to a
module load"
* tag 'audit-pr-20220616' of git://git.kernel.org/pub/scm/linux/kernel/git/pcmoore/audit:
audit: free module name
Each cset (css_set) is pinned by its tasks. When we're moving tasks around
across csets for a migration, we need to hold the source and destination
csets to ensure that they don't go away while we're moving tasks about. This
is done by linking cset->mg_preload_node on either the
mgctx->preloaded_src_csets or mgctx->preloaded_dst_csets list. Using the
same cset->mg_preload_node for both the src and dst lists was deemed okay as
a cset can't be both the source and destination at the same time.
Unfortunately, this overloading becomes problematic when multiple tasks are
involved in a migration and some of them are identity noop migrations while
others are actually moving across cgroups. For example, this can happen with
the following sequence on cgroup1:
#1> mkdir -p /sys/fs/cgroup/misc/a/b
#2> echo $$ > /sys/fs/cgroup/misc/a/cgroup.procs
#3> RUN_A_COMMAND_WHICH_CREATES_MULTIPLE_THREADS &
#4> PID=$!
#5> echo $PID > /sys/fs/cgroup/misc/a/b/tasks
#6> echo $PID > /sys/fs/cgroup/misc/a/cgroup.procs
the process including the group leader back into a. In this final migration,
non-leader threads would be doing identity migration while the group leader
is doing an actual one.
After #3, let's say the whole process was in cset A, and that after #4, the
leader moves to cset B. Then, during #6, the following happens:
1. cgroup_migrate_add_src() is called on B for the leader.
2. cgroup_migrate_add_src() is called on A for the other threads.
3. cgroup_migrate_prepare_dst() is called. It scans the src list.
4. It notices that B wants to migrate to A, so it tries to A to the dst
list but realizes that its ->mg_preload_node is already busy.
5. and then it notices A wants to migrate to A as it's an identity
migration, it culls it by list_del_init()'ing its ->mg_preload_node and
putting references accordingly.
6. The rest of migration takes place with B on the src list but nothing on
the dst list.
This means that A isn't held while migration is in progress. If all tasks
leave A before the migration finishes and the incoming task pins it, the
cset will be destroyed leading to use-after-free.
This is caused by overloading cset->mg_preload_node for both src and dst
preload lists. We wanted to exclude the cset from the src list but ended up
inadvertently excluding it from the dst list too.
This patch fixes the issue by separating out cset->mg_preload_node into
->mg_src_preload_node and ->mg_dst_preload_node, so that the src and dst
preloadings don't interfere with each other.
Signed-off-by: Tejun Heo <tj@kernel.org>
Reported-by: Mukesh Ojha <quic_mojha@quicinc.com>
Reported-by: shisiyuan <shisiyuan19870131@gmail.com>
Link: http://lkml.kernel.org/r/1654187688-27411-1-git-send-email-shisiyuan@xiaomi.com
Link: https://www.spinics.net/lists/cgroups/msg33313.html
Fixes: f817de9851 ("cgroup: prepare migration path for unified hierarchy")
Cc: stable@vger.kernel.org # v3.16+
Reset the type of the record last as the helper `audit_free_module()`
depends on it.
unreferenced object 0xffff888153b707f0 (size 16):
comm "modprobe", pid 1319, jiffies 4295110033 (age 1083.016s)
hex dump (first 16 bytes):
62 69 6e 66 6d 74 5f 6d 69 73 63 00 6b 6b 6b a5 binfmt_misc.kkk.
backtrace:
[<ffffffffa07dbf9b>] kstrdup+0x2b/0x50
[<ffffffffa04b0a9d>] __audit_log_kern_module+0x4d/0xf0
[<ffffffffa03b6664>] load_module+0x9d4/0x2e10
[<ffffffffa03b8f44>] __do_sys_finit_module+0x114/0x1b0
[<ffffffffa1f47124>] do_syscall_64+0x34/0x80
[<ffffffffa200007e>] entry_SYSCALL_64_after_hwframe+0x46/0xb0
Cc: stable@vger.kernel.org
Fixes: 12c5e81d3f ("audit: prepare audit_context for use in calling contexts beyond syscalls")
Signed-off-by: Christian Göttsche <cgzones@googlemail.com>
Signed-off-by: Paul Moore <paul@paul-moore.com>
There are reports that the console kthreads block the global console
lock when the system is going down, for example, reboot, panic.
First part of the solution was to block kthreads in these problematic
system states so they stopped handling newly added messages.
Second part of the solution is to wait when for the kthreads when
they are actively printing. It solves the problem when a message
was printed before the system entered the problematic state and
the kthreads managed to step in.
A busy waiting has to be used because panic() can be called in any
context and in an unknown state of the scheduler.
There must be a timeout because the kthread might get stuck or sleeping
and never release the lock. The timeout 10s is an arbitrary value
inspired by the softlockup timeout.
Link: https://lore.kernel.org/r/20220610205038.GA3050413@paulmck-ThinkPad-P17-Gen-1
Link: https://lore.kernel.org/r/CAMdYzYpF4FNTBPZsEFeWRuEwSies36QM_As8osPWZSr2q-viEA@mail.gmail.com
Signed-off-by: Petr Mladek <pmladek@suse.com>
Tested-by: Paul E. McKenney <paulmck@kernel.org>
Link: https://lore.kernel.org/r/20220615162805.27962-3-pmladek@suse.com
There are known situations when the console kthreads are not
reliable or does not work in principle, for example, early boot,
panic, shutdown.
For these situations there is the direct (legacy) mode when printk() tries
to get console_lock() and flush the messages directly. It works very well
during the early boot when the console kthreads are not available at all.
It gets more complicated in the other situations when console kthreads
might be actively printing and block console_trylock() in printk().
The same problem is in the legacy code as well. Any console_lock()
owner could block console_trylock() in printk(). It is solved by
a trick that the current console_lock() owner is responsible for
printing all pending messages. It is actually the reason why there
is the risk of softlockups and why the console kthreads were
introduced.
The console kthreads use the same approach. They are responsible
for printing the messages by definition. So that they handle
the messages anytime when they are awake and see new ones.
The global console_lock is available when there is nothing
to do.
It should work well when the problematic context is correctly
detected and printk() switches to the direct mode. But it seems
that it is not enough in practice. There are reports that
the messages are not printed during panic() or shutdown()
even though printk() tries to use the direct mode here.
The problem seems to be that console kthreads become active in these
situation as well. They steel the job before other CPUs are stopped.
Then they are stopped in the middle of the job and block the global
console_lock.
First part of the solution is to block console kthreads when
the system is in a problematic state and requires the direct
printk() mode.
Link: https://lore.kernel.org/r/20220610205038.GA3050413@paulmck-ThinkPad-P17-Gen-1
Link: https://lore.kernel.org/r/CAMdYzYpF4FNTBPZsEFeWRuEwSies36QM_As8osPWZSr2q-viEA@mail.gmail.com
Suggested-by: John Ogness <john.ogness@linutronix.de>
Tested-by: Paul E. McKenney <paulmck@kernel.org>
Signed-off-by: Petr Mladek <pmladek@suse.com>
Link: https://lore.kernel.org/r/20220615162805.27962-2-pmladek@suse.com
On processing a module BTF of module built for an older kernel, we might
sometimes find that some type points to itself forming a loop. If such a
type is a modifier, btf_check_type_tags's while loop following modifier
chain will be caught in an infinite loop.
Fix this by defining a maximum chain length and bailing out if we spin
any longer than that.
Fixes: eb596b0905 ("bpf: Ensure type tags precede modifiers in BTF")
Reported-by: Daniel Borkmann <daniel@iogearbox.net>
Signed-off-by: Kumar Kartikeya Dwivedi <memxor@gmail.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: Yonghong Song <yhs@fb.com>
Link: https://lore.kernel.org/bpf/20220615042151.2266537-1-memxor@gmail.com
RCU_NONIDLE usage during __cfi_slowpath_diag can result in an invalid
RCU state in the cpuidle code path:
WARNING: CPU: 1 PID: 0 at kernel/rcu/tree.c:613 rcu_eqs_enter+0xe4/0x138
...
Call trace:
rcu_eqs_enter+0xe4/0x138
rcu_idle_enter+0xa8/0x100
cpuidle_enter_state+0x154/0x3a8
cpuidle_enter+0x3c/0x58
do_idle.llvm.6590768638138871020+0x1f4/0x2ec
cpu_startup_entry+0x28/0x2c
secondary_start_kernel+0x1b8/0x220
__secondary_switched+0x94/0x98
Instead, call rcu_irq_enter/exit to wake up RCU only when needed and
disable interrupts for the entire CFI shadow/module check when we do.
Signed-off-by: Sami Tolvanen <samitolvanen@google.com>
Link: https://lore.kernel.org/r/20220531175910.890307-1-samitolvanen@google.com
Fixes: cf68fffb66 ("add support for Clang CFI")
Cc: stable@vger.kernel.org
Signed-off-by: Kees Cook <keescook@chromium.org>
Wakelist can help avoid cache bouncing and offload the overhead of waker
cpu. So far, using wakelist within the same llc only happens on
WF_ON_CPU, and this limitation could be removed to further improve
wakeup performance.
The commit 518cd62341 ("sched: Only queue remote wakeups when
crossing cache boundaries") disabled queuing tasks on wakelist when
the cpus share llc. This is because, at that time, the scheduler must
send IPIs to do ttwu_queue_wakelist. Nowadays, ttwu_queue_wakelist also
supports TIF_POLLING, so this is not a problem now when the wakee cpu is
in idle polling.
Benefits:
Queuing the task on idle cpu can help improving performance on waker cpu
and utilization on wakee cpu, and further improve locality because
the wakee cpu can handle its own rq. This patch helps improving rt on
our real java workloads where wakeup happens frequently.
Consider the normal condition (CPU0 and CPU1 share same llc)
Before this patch:
CPU0 CPU1
select_task_rq() idle
rq_lock(CPU1->rq)
enqueue_task(CPU1->rq)
notify CPU1 (by sending IPI or CPU1 polling)
resched()
After this patch:
CPU0 CPU1
select_task_rq() idle
add to wakelist of CPU1
notify CPU1 (by sending IPI or CPU1 polling)
rq_lock(CPU1->rq)
enqueue_task(CPU1->rq)
resched()
We see CPU0 can finish its work earlier. It only needs to put task to
wakelist and return.
While CPU1 is idle, so let itself handle its own runqueue data.
This patch brings no difference about IPI.
This patch only takes effect when the wakee cpu is:
1) idle polling
2) idle not polling
For 1), there will be no IPI with or without this patch.
For 2), there will always be an IPI before or after this patch.
Before this patch: waker cpu will enqueue task and check preempt. Since
"idle" will be sure to be preempted, waker cpu must send a resched IPI.
After this patch: waker cpu will put the task to the wakelist of wakee
cpu, and send an IPI.
Benchmark:
We've tested schbench, unixbench, and hachbench on both x86 and arm64.
On x86 (Intel Xeon Platinum 8269CY):
schbench -m 2 -t 8
Latency percentiles (usec) before after
50.0000th: 8 6
75.0000th: 10 7
90.0000th: 11 8
95.0000th: 12 8
*99.0000th: 13 10
99.5000th: 15 11
99.9000th: 18 14
Unixbench with full threads (104)
before after
Dhrystone 2 using register variables 3011862938 3009935994 -0.06%
Double-Precision Whetstone 617119.3 617298.5 0.03%
Execl Throughput 27667.3 27627.3 -0.14%
File Copy 1024 bufsize 2000 maxblocks 785871.4 784906.2 -0.12%
File Copy 256 bufsize 500 maxblocks 210113.6 212635.4 1.20%
File Copy 4096 bufsize 8000 maxblocks 2328862.2 2320529.1 -0.36%
Pipe Throughput 145535622.8 145323033.2 -0.15%
Pipe-based Context Switching 3221686.4 3583975.4 11.25%
Process Creation 101347.1 103345.4 1.97%
Shell Scripts (1 concurrent) 120193.5 123977.8 3.15%
Shell Scripts (8 concurrent) 17233.4 17138.4 -0.55%
System Call Overhead 5300604.8 5312213.6 0.22%
hackbench -g 1 -l 100000
before after
Time 3.246 2.251
On arm64 (Ampere Altra):
schbench -m 2 -t 8
Latency percentiles (usec) before after
50.0000th: 14 10
75.0000th: 19 14
90.0000th: 22 16
95.0000th: 23 16
*99.0000th: 24 17
99.5000th: 24 17
99.9000th: 28 25
Unixbench with full threads (80)
before after
Dhrystone 2 using register variables 3536194249 3537019613 0.02%
Double-Precision Whetstone 629383.6 629431.6 0.01%
Execl Throughput 65920.5 65846.2 -0.11%
File Copy 1024 bufsize 2000 maxblocks 1063722.8 1064026.8 0.03%
File Copy 256 bufsize 500 maxblocks 322684.5 318724.5 -1.23%
File Copy 4096 bufsize 8000 maxblocks 2348285.3 2328804.8 -0.83%
Pipe Throughput 133542875.3 131619389.8 -1.44%
Pipe-based Context Switching 3215356.1 3576945.1 11.25%
Process Creation 108520.5 120184.6 10.75%
Shell Scripts (1 concurrent) 122636.3 121888 -0.61%
Shell Scripts (8 concurrent) 17462.1 17381.4 -0.46%
System Call Overhead 4429998.9 4435006.7 0.11%
hackbench -g 1 -l 100000
before after
Time 4.217 2.916
Our patch has improvement on schbench, hackbench
and Pipe-based Context Switching of unixbench
when there exists idle cpus,
and no obvious regression on other tests of unixbench.
This can help improve rt in scenes where wakeup happens frequently.
Signed-off-by: Tianchen Ding <dtcccc@linux.alibaba.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Valentin Schneider <vschneid@redhat.com>
Link: https://lore.kernel.org/r/20220608233412.327341-3-dtcccc@linux.alibaba.com
While doing newidle load balancing, it is possible for new tasks to
arrive, such as with pending wakeups. newidle_balance() already accounts
for this by exiting the sched_domain load_balance() iteration if it
detects these cases. This is very important for minimizing wakeup
latency.
However, if we are already in load_balance(), we may stay there for a
while before returning back to newidle_balance(). This is most
exacerbated if we enter a 'goto redo' loop in the LBF_ALL_PINNED case. A
very straightforward workaround to this is to adjust should_we_balance()
to bail out if we're doing a CPU_NEWLY_IDLE balance and new tasks are
detected.
This was tested with the following reproduction:
- two threads that take turns sleeping and waking each other up are
affined to two cores
- a large number of threads with 100% utilization are pinned to all
other cores
Without this patch, wakeup latency was ~120us for the pair of threads,
almost entirely spent in load_balance(). With this patch, wakeup latency
is ~6us.
Signed-off-by: Josh Don <joshdon@google.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lkml.kernel.org/r/20220609025515.2086253-1-joshdon@google.com
