The 'perf record' and 'perf stat' commands have supported the option
'-C/--cpus' to count or collect only on the list of CPUs provided. This
option needs to be supported for hybrid as well.
For hybrid support, it needs to check that the cpu list are available
on hybrid PMU. One example for AlderLake, cpu0-7 is 'cpu_core', cpu8-11
is 'cpu_atom'.
Before:
# perf stat -e cpu_core/cycles/ -C11 -- sleep 1
Performance counter stats for 'CPU(s) 11':
<not supported> cpu_core/cycles/
1.006179431 seconds time elapsed
The 'perf stat' command silently returned "<not supported>" without any
helpful information. It should error out pointing out that that cpu11
was not 'cpu_core'.
After:
# perf stat -e cpu_core/cycles/ -C11 -- sleep 1
WARNING: 11 isn't a 'cpu_core', please use a CPU list in the 'cpu_core' range (0-7)
failed to use cpu list 11
We also need to support the events without pmu prefix specified.
# perf stat -e cycles -C11 -- sleep 1
WARNING: 11 isn't a 'cpu_core', please use a CPU list in the 'cpu_core' range (0-7)
Performance counter stats for 'CPU(s) 11':
1,067,373 cpu_atom/cycles/
1.005544738 seconds time elapsed
The perf tool creates two cycles events automatically, cpu_core/cycles/ and
cpu_atom/cycles/. It checks that cpu11 is not 'cpu_core', then shows a warning
for cpu_core/cycles/ and only count the cpu_atom/cycles/.
If part of cpus are 'cpu_core' and part of cpus are 'cpu_atom', for example,
# perf stat -e cycles -C0,11 -- sleep 1
WARNING: use 0 in 'cpu_core' for 'cycles', skip other cpus in list.
WARNING: use 11 in 'cpu_atom' for 'cycles', skip other cpus in list.
Performance counter stats for 'CPU(s) 0,11':
1,914,704 cpu_core/cycles/
2,036,983 cpu_atom/cycles/
1.005815641 seconds time elapsed
It now automatically selects cpu0 for cpu_core/cycles/, selects cpu11 for
cpu_atom/cycles/, and output with some warnings.
Some more complex examples,
# perf stat -e cycles,instructions -C0,11 -- sleep 1
WARNING: use 0 in 'cpu_core' for 'cycles', skip other cpus in list.
WARNING: use 11 in 'cpu_atom' for 'cycles', skip other cpus in list.
WARNING: use 0 in 'cpu_core' for 'instructions', skip other cpus in list.
WARNING: use 11 in 'cpu_atom' for 'instructions', skip other cpus in list.
Performance counter stats for 'CPU(s) 0,11':
2,780,387 cpu_core/cycles/
1,583,432 cpu_atom/cycles/
3,957,277 cpu_core/instructions/
1,167,089 cpu_atom/instructions/
1.006005124 seconds time elapsed
# perf stat -e cycles,cpu_atom/instructions/ -C0,11 -- sleep 1
WARNING: use 0 in 'cpu_core' for 'cycles', skip other cpus in list.
WARNING: use 11 in 'cpu_atom' for 'cycles', skip other cpus in list.
WARNING: use 11 in 'cpu_atom' for 'cpu_atom/instructions/', skip other cpus in list.
Performance counter stats for 'CPU(s) 0,11':
3,290,301 cpu_core/cycles/
1,953,073 cpu_atom/cycles/
1,407,869 cpu_atom/instructions/
1.006260912 seconds time elapsed
Signed-off-by: Jin Yao <yao.jin@linux.intel.com>
Acked-by: Jiri Olsa <jolsa@redhat.com>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Jin Yao <yao.jin@intel.com>
Cc: Kan Liang <kan.liang@intel.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: https //lore.kernel.org/r/20210723063433.7318-4-yao.jin@linux.intel.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
The user may count or collect only on a cpu list via '-C/--cpus' option.
Previously cpus for an evsel were retrieved from PMU's sysfs. But if the
target cpu list is defined, the retrieved cpus are not kept and the
target cpu list is used instead.
But for hybrid system, we can't directly use target cpu list. The cpu
list may not be available on hybrid pmu (e.g. cpu_core or cpu_atom). So
we should not set the 'has_user_cpus' flag for hybrid system.
The difficulity is that we can't call perf_pmu__has_hybrid() in evlist.c
to check hybrid system otherwise 'perf test python' would be failed
(undefined symbol for perf_pmu__has_hybrid). If we add pmu.c to
python-ext-sources, too many symbol dependencies are hard to resolve.
We use an alternative method by using a new 'hybrid' flag in target
for hybrid system checking.
Signed-off-by: Jin Yao <yao.jin@linux.intel.com>
Acked-by: Jiri Olsa <jolsa@redhat.com>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Jin Yao <yao.jin@intel.com>
Cc: Kan Liang <kan.liang@intel.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: https //lore.kernel.org/r/20210723063433.7318-3-yao.jin@linux.intel.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
perf-test has the option --skip to provide a list of tests to skip.
However, this option does not work with shell scripts.
This patch passes the skiplist to run_shell_tests, so that also shell
scripts could be skipped using --skip.
Committer tests:
Tests 79 onwards are shell tests:
Before:
# perf test --skip 1,2,81,82,84,88,90
1: vmlinux symtab matches kallsyms : Skip (user override)
2: Detect openat syscall event : Skip (user override)
3: Detect openat syscall event on all cpus : Ok
4: Read samples using the mmap interface : Ok
5: Test data source output : Ok
<SNIP>
78: x86 Sample parsing : Ok
79: build id cache operations : Ok
80: daemon operations : Ok
81: perf pipe recording and injection test : Ok
82: Add vfs_getname probe to get syscall args filenames : FAILED!
83: probe libc's inet_pton & backtrace it with ping : Ok
84: Use vfs_getname probe to get syscall args filenames : FAILED!
85: Zstd perf.data compression/decompression : Ok
86: perf stat csv summary test : Ok
87: perf stat metrics (shadow stat) test : Ok
88: perf stat --bpf-counters test : Ok
89: Check Arm CoreSight trace data recording and synthesized samples: Skip
90: Check open filename arg using perf trace + vfs_getname : FAILED!
#
After:
# perf test --skip 1,2,81,82,84,88,90
1: vmlinux symtab matches kallsyms : Skip (user override)
2: Detect openat syscall event : Skip (user override)
3: Detect openat syscall event on all cpus : Ok
4: Read samples using the mmap interface : Ok
5: Test data source output : Ok
<SNIP>
78: x86 Sample parsing : Ok
79: build id cache operations : Ok
80: daemon operations : Ok
81: perf pipe recording and injection test : Skip (user override)
82: Add vfs_getname probe to get syscall args filenames : Skip (user override)
83: probe libc's inet_pton & backtrace it with ping : Ok
84: Use vfs_getname probe to get syscall args filenames : Skip (user override)
85: Zstd perf.data compression/decompression : Ok
86: perf stat csv summary test : Ok
87: perf stat metrics (shadow stat) test : Ok
88: perf stat --bpf-counters test : Skip (user override)
89: Check Arm CoreSight trace data recording and synthesized samples: Skip
90: Check open filename arg using perf trace + vfs_getname : Skip (user override)
#
Signed-off-by: Riccardo Mancini <rickyman7@gmail.com>
Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Ian Rogers <irogers@google.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Mark Rutland <mark.rutland@arm.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: http://lore.kernel.org/lkml/20210811180625.160944-1-rickyman7@gmail.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Add a perf test to test the dlfilter C API.
A perf.data file is synthesized and then processed by perf script with a
dlfilter named dlfilter-test-api-v0.so. Also a C file is compiled to
provide a dso to match the synthesized perf.data file.
Committer testing:
[root@five ~]# perf test dlfilter
72: dlfilter C API : Ok
[root@five ~]# perf test -v dlfilter
72: dlfilter C API :
--- start ---
test child forked, pid 3387712
Checking for gcc
Command: gcc --version
gcc (GCC) 11.1.1 20210531 (Red Hat 11.1.1-3)
Copyright (C) 2021 Free Software Foundation, Inc.
This is free software; see the source for copying conditions. There is NO
warranty; not even for MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.
dlfilters path: /var/home/acme/libexec/perf-core/dlfilters
Command: gcc -g -o /tmp/dlfilter-test-3387712-prog /tmp/dlfilter-test-3387712-prog.c
Creating new host machine structure
Command: /var/home/acme/bin/perf script -i /tmp/dlfilter-test-3387712-perf-data --dlfilter /var/home/acme/libexec/perf-core/dlfilters/dlfilter-test-api-v0.so --dlarg first --dlarg 1 --dlarg 4198669 --dlarg 4198662 --dlarg 0 --dlarg last
start API
filter_event_early API
filter_event API
stop API
Command: /var/home/acme/bin/perf script -i /tmp/dlfilter-test-3387712-perf-data --dlfilter /var/home/acme/libexec/perf-core/dlfilters/dlfilter-test-api-v0.so --dlarg first --dlarg 1 --dlarg 4198669 --dlarg 4198662 --dlarg 1 --dlarg last
start API
filter_event_early API
filter_event API
stop API
Command: /var/home/acme/bin/perf script -i /tmp/dlfilter-test-3387712-perf-data --dlfilter /var/home/acme/libexec/perf-core/dlfilters/dlfilter-test-api-v0.so --dlarg first --dlarg 1 --dlarg 4198669 --dlarg 4198662 --dlarg 2 --dlarg last
start API
filter_event_early API
stop API
test child finished with 0
---- end ----
dlfilter C API: Ok
[root@five ~]#
Signed-off-by: Adrian Hunter <adrian.hunter@intel.com>
Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Link: https //lore.kernel.org/r/20210811101036.17986-7-adrian.hunter@intel.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Like all locally-built programs, dlfilters may need to be re-built if
shared libraries they use change. Also there may be unexpected results
if the dfilter uses different versions of the shared libraries that perf
uses.
Note those things in the documentation.
Signed-off-by: Adrian Hunter <adrian.hunter@intel.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Link: https //lore.kernel.org/r/20210811101036.17986-5-adrian.hunter@intel.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
perf_dlfilter_fns must not be const, because it is not.
Declaring it const can result in it being mapped read-only, causing a
segfaullt when it is written. Update documentation accordingly.
Signed-off-by: Adrian Hunter <adrian.hunter@intel.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Fixes: 8defa7147d5572 ("perf script Add API for filtering via dynamically loaded shared object")
Link: https //lore.kernel.org/r/20210811101036.17986-2-adrian.hunter@intel.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Pull ARC fixes from Vineet Gupta:
- Fix FPU_STATUS update
- Update my email address
- Other spellos and fixes
* tag 'arc-5.14-rc6' of git://git.kernel.org/pub/scm/linux/kernel/git/vgupta/arc:
MAINTAINERS: update Vineet's email address
ARC: fp: set FPU_STATUS.FWE to enable FPU_STATUS update on context switch
ARC: Fix CONFIG_STACKDEPOT
arc: Fix spelling mistake and grammar in Kconfig
arc: Prefer unsigned int to bare use of unsigned
Pull x86 platform driver fixes from Hans de Goede:
"Small set of pdx86 fixes for 5.14"
* tag 'platform-drivers-x86-v5.14-3' of git://git.kernel.org/pub/scm/linux/kernel/git/pdx86/platform-drivers-x86:
platform/x86: pcengines-apuv2: Add missing terminating entries to gpio-lookup tables
platform/x86: Make dual_accel_detect() KIOX010A + KIOX020A detect more robust
platform/x86: Add and use a dual_accel_detect() helper
Pull overlayfs fixes from Miklos Szeredi:
"Fix several bugs in overlayfs"
* tag 'ovl-fixes-5.14-rc6-v2' of git://git.kernel.org/pub/scm/linux/kernel/git/mszeredi/vfs:
ovl: prevent private clone if bind mount is not allowed
ovl: fix uninitialized pointer read in ovl_lookup_real_one()
ovl: fix deadlock in splice write
ovl: skip stale entries in merge dir cache iteration
This new benchmark finds the total time that is taken to open, mmap,
enable, disable, munmap, close an evlist (time taken for new,
create_maps, config, delete is not counted in).
The evlist can be configured as in perf-record using the
-a,-C,-e,-u,--per-thread,-t,-p options.
The events can be duplicated in the evlist to quickly test performance
with many events using the -n options.
Furthermore, also the number of iterations used to calculate the
statistics is customizable.
Examples:
- Open one dummy event system-wide:
$ sudo ./perf bench internals evlist-open-close
Number of cpus: 4
Number of threads: 1
Number of events: 1 (4 fds)
Number of iterations: 100
Average open-close took: 613.870 usec (+- 32.852 usec)
- Open the group '{cs,cycles}' on CPU 0
$ sudo ./perf bench internals evlist-open-close -e '{cs,cycles}' -C 0
Number of cpus: 1
Number of threads: 1
Number of events: 2 (2 fds)
Number of iterations: 100
Average open-close took: 8503.220 usec (+- 252.652 usec)
- Open 10 'cycles' events for user 0, calculate average over 100 runs
$ sudo ./perf bench internals evlist-open-close -e cycles -n 10 -u 0 -i 100
Number of cpus: 4
Number of threads: 328
Number of events: 10 (13120 fds)
Number of iterations: 100
Average open-close took: 180043.140 usec (+- 2295.889 usec)
Committer notes:
Replaced a deprecated bzero() call with designated initialized zeroing.
Added some missing evlist allocation checks, one noted by Riccardo on
the mailing list.
Minor cosmetic changes (sent in private).
Signed-off-by: Riccardo Mancini <rickyman7@gmail.com>
Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Ian Rogers <irogers@google.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Mark Rutland <mark.rutland@arm.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: http://lore.kernel.org/lkml/20210809201101.277594-1-rickyman7@gmail.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Add the following checks from __do_loopback() to clone_private_mount() as
well:
- verify that the mount is in the current namespace
- verify that there are no locked children
Reported-by: Alois Wohlschlager <alois1@gmx-topmail.de>
Fixes: c771d683a6 ("vfs: introduce clone_private_mount()")
Cc: <stable@vger.kernel.org> # v3.18
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
One error path can result in release_dentry_name_snapshot() being called
before "name" was initialized by take_dentry_name_snapshot().
Fix by moving the release_dentry_name_snapshot() to immediately after the
only use.
Reported-by: Colin Ian King <colin.king@canonical.com>
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
There's possibility of an ABBA deadlock in case of a splice write to an
overlayfs file and a concurrent splice write to a corresponding real file.
The call chain for splice to an overlay file:
-> do_splice [takes sb_writers on overlay file]
-> do_splice_from
-> iter_file_splice_write [takes pipe->mutex]
-> vfs_iter_write
...
-> ovl_write_iter [takes sb_writers on real file]
And the call chain for splice to a real file:
-> do_splice [takes sb_writers on real file]
-> do_splice_from
-> iter_file_splice_write [takes pipe->mutex]
Syzbot successfully bisected this to commit 82a763e61e ("ovl: simplify
file splice").
Fix by reverting the write part of the above commit and by adding missing
bits from ovl_write_iter() into ovl_splice_write().
Fixes: 82a763e61e ("ovl: simplify file splice")
Reported-and-tested-by: syzbot+579885d1a9a833336209@syzkaller.appspotmail.com
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
On the first getdents call, ovl_iterate() populates the readdir cache
with a list of entries, but for upper entries with origin lower inode,
p->ino remains zero.
Following getdents calls traverse the readdir cache list and call
ovl_cache_update_ino() for entries with zero p->ino to lookup the entry
in the overlay and return d_ino that is consistent with st_ino.
If the upper file was unlinked between the first getdents call and the
getdents call that lists the file entry, ovl_cache_update_ino() will not
find the entry and fall back to setting d_ino to the upper real st_ino,
which is inconsistent with how this object was presented to users.
Instead of listing a stale entry with inconsistent d_ino, simply skip
the stale entry, which is better for users.
xfstest overlay/077 is failing without this patch.
Signed-off-by: Amir Goldstein <amir73il@gmail.com>
Link: https://lore.kernel.org/fstests/CAOQ4uxgR_cLnC_vdU5=seP3fwqVkuZM_-WfD6maFTMbMYq=a9w@mail.gmail.com/
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
Pull cgroup fix from Tejun Heo:
"One commit to fix a possible A-A deadlock around u64_stats_sync on
32bit machines caused by updating it without disabling IRQ when it may
be read from IRQ context"
* 'for-5.14-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/tj/cgroup:
cgroup: rstat: fix A-A deadlock on 32bit around u64_stats_sync
