- 23 May, 2022 33 commits
-
-
Nick Forrington authored
Add PMU events for Arm Cortex-A65 Update mapfile.csv Event data based on: https://github.com/ARM-software/data/tree/master/pmu/cortex-a65.json which is based on PMU event descriptions from the Arm Cortex-A65 Technical Reference Manual. Mapping data (for mapfile.csv) based on: https://github.com/ARM-software/data/blob/master/cpus.json which is based on Main ID Register (MIDR) information found in the Arm Technical Reference Manuals for individual CPUs. Reviewed-by:
John Garry <john.garry@huawei.com> Signed-off-by:
Nick Forrington <nick.forrington@arm.com> Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com> Cc: Andi Kleen <ak@linux.intel.com> Cc: Andrew Kilroy <andrew.kilroy@arm.com> Cc: Ingo Molnar <mingo@redhat.com> Cc: James Clark <james.clark@arm.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Kajol Jain <kjain@linux.ibm.com> Cc: Leo Yan <leo.yan@linaro.org> Cc: Mark Rutland <mark.rutland@arm.com> Cc: Mike Leach <mike.leach@linaro.org> Cc: Namhyung Kim <namhyung@kernel.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Will Deacon <will@kernel.org> Cc: linux-arm-kernel@lists.infradead.org Link: https://lore.kernel.org/r/20220520181455.340344-6-nick.forrington@arm.comSigned-off-by:
Arnaldo Carvalho de Melo <acme@redhat.com>
-
Nick Forrington authored
Add PMU events for Arm Cortex-A510 Add corresponding common events Update mapfile.csv Event data based on: https://github.com/ARM-software/data/tree/master/pmu/cortex-a510.json which is based on PMU event descriptions from the Arm Cortex-A510 Technical Reference Manual. Common event data based on: https://github.com/ARM-software/data/blob/master/pmu/common_armv9.json which is based on PMU event descriptions found in the Arm Architecture Reference Manual: https://developer.arm.com/documentation/ddi0487/ Mapping data (for mapfile.csv) based on: https://github.com/ARM-software/data/blob/master/cpus.json which is based on Main ID Register (MIDR) information found in the Arm Technical Reference Manuals for individual CPUs. Reviewed-by:
-
Nick Forrington authored
Add PMU events for Arm Cortex-A55 Add corresponding common events Update mapfile.csv Event data based on: https://github.com/ARM-software/data/tree/master/pmu/cortex-a55.json which is based on PMU event descriptions from the Arm Cortex-A55 Technical Reference Manual. Common event data based on: https://github.com/ARM-software/data/blob/master/pmu/common_armv9.json which is based on PMU event descriptions found in the Arm Architecture Reference Manual: https://developer.arm.com/documentation/ddi0487/ Mapping data (for mapfile.csv) based on: https://github.com/ARM-software/data/blob/master/cpus.json which is based on Main ID Register (MIDR) information found in the Arm Technical Reference Manuals for individual CPUs. Reviewed-by:
-
Nick Forrington authored
Add PMU events for Arm Cortex-A35 Update mapfile.csv Event data based on: https://github.com/ARM-software/data/tree/master/pmu/cortex-a35.json which is based on PMU event descriptions from the Arm Cortex-A35 Technical Reference Manual. Mapping data (for mapfile.csv) based on: https://github.com/ARM-software/data/blob/master/cpus.json which is based on Main ID Register (MIDR) information found in the Arm Technical Reference Manuals for individual CPUs. Reviewed-by:
-
Nick Forrington authored
Add PMU events for Arm Cortex-A34 Add corresponding common events Update mapfile.csv Event data based on: https://github.com/ARM-software/data/tree/master/pmu/cortex-a34.json which is based on PMU event descriptions from the Arm Cortex-A34 Technical Reference Manual. Common event data based on: https://github.com/ARM-software/data/blob/master/pmu/common_armv9.json which is based on PMU event descriptions found in the Arm Architecture Reference Manual: https://developer.arm.com/documentation/ddi0487/ Mapping data (for mapfile.csv) based on: https://github.com/ARM-software/data/blob/master/cpus.json which is based on Main ID Register (MIDR) information found in the Arm Technical Reference Manuals for individual CPUs. Reviewed-by:
-
Adrian Hunter authored
If the input perf.data has a kcore_dir, copy it into the output, since at least the kallsyms in the kcore_dir will be useful to the output. Example: Before: $ ls -lR perf.data-from-desktop perf.data-from-desktop: total 916 -rw------- 1 user user 931756 May 19 09:55 data drwx------ 2 user user 4096 May 19 09:55 kcore_dir perf.data-from-desktop/kcore_dir: total 42952 -r-------- 1 user user 7582467 May 19 09:55 kallsyms -r-------- 1 user user 36388864 May 19 09:55 kcore -r-------- 1 user user 4828 May 19 09:55 modules $ perf inject -i perf.data-from-desktop -o injected-perf.data $ ls -lR injected-perf.data -rw------- 1 user user 931320 May 20 15:08 injected-perf.data After: $ perf inject -i perf.data-from-desktop -o injected-perf.data $ ls -lR injected-perf.data injected-perf.data: total 916 -rw------- 1 user user 931320 May 20 15:21 data drwx------ 2 user user 4096 May 20 15:21 kcore_dir injected-perf.data/kcore_dir: total 42952 -r-------- 1 user user 7582467 May 20 15:21 kallsyms -r-------- 1 user user 36388864 May 20 15:21 kcore -r-------- 1 user user 4828 May 20 15:21 modules Signed-off-by:
Adrian Hunter <adrian.hunter@intel.com> Acked-by:
Jiri Olsa <jolsa@kernel.org> Cc: Namhyung Kim <namhyung@kernel.org> Link: https://lore.kernel.org/r/20220520132404.25853-6-adrian.hunter@intel.comSigned-off-by:
-
Adrian Hunter authored
Add a helper function has_kcore_dir(), so that perf inject can determine if it needs to keep the kcore_dir. Signed-off-by:
-
Adrian Hunter authored
perf inject overwrites feature sections with information from the current machine. It makes more sense to keep original information that describes the machine or software when perf record was run. Example: perf.data from "Desktop" injected on "nuc11" Before: $ perf script --header-only -i perf.data-from-desktop | head -15 # ======== # captured on : Thu May 19 09:55:50 2022 # header version : 1 # data offset : 1208 # data size : 837480 # feat offset : 838688 # hostname : Desktop # os release : 5.13.0-41-generic # perf version : 5.18.rc5.gac837f7ca7ed # arch : x86_64 # nrcpus online : 28 # nrcpus avail : 28 # cpudesc : Intel(R) Core(TM) i9-9940X CPU @ 3.30GHz # cpuid : GenuineIntel,6,85,4 # total memory : 65548656 kB $ perf inject -i perf.data-from-desktop -o injected-perf.data $ perf script --header-only -i injected-perf.data | head -15 # ======== # captured on : Fri May 20 15:06:55 2022 # header version : 1 # data offset : 1208 # data size : 837480 # feat offset : 838688 # hostname : nuc11 # os release : 5.17.5-local # perf version : 5.18.rc5.g0f828fdeb9af # arch : x86_64 # nrcpus online : 8 # nrcpus avail : 8 # cpudesc : 11th Gen Intel(R) Core(TM) i7-1165G7 @ 2.80GHz # cpuid : GenuineIntel,6,140,1 # total memory : 16012124 kB After: $ perf inject -i perf.data-from-desktop -o injected-perf.data $ perf script --header-only -i injected-perf.data | head -15 # ======== # captured on : Fri May 20 15:08:54 2022 # header version : 1 # data offset : 1208 # data size : 837480 # feat offset : 838688 # hostname : Desktop # os release : 5.13.0-41-generic # perf version : 5.18.rc5.gac837f7ca7ed # arch : x86_64 # nrcpus online : 28 # nrcpus avail : 28 # cpudesc : Intel(R) Core(TM) i9-9940X CPU @ 3.30GHz # cpuid : GenuineIntel,6,85,4 # total memory : 65548656 kB Signed-off-by:
-
Adrian Hunter authored
Add preadn() to provide pread() and readn() semantics. Signed-off-by:
-
Adrian Hunter authored
Many feature sections should not be re-written during perf inject. In preparation to support that, add callbacks that a tool can use to copy a feature section from elsewhere. perf inject will use this facility to copy features sections from the input file. Signed-off-by:
-
Ian Rogers authored
The match_field function looks for json values to append to the event string. As the C code processes these in order the output order matches that in the json dictionary. Python json readers read the entire dictionary and lose the ordering. To make the python and C output comparable make the C code first read the extra fields then append them to the event in an order not determined by their order in the file. Modify the pmu-events test so that test expectations match the new order. Reviewed-by:
Ian Rogers <irogers@google.com> Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com> Cc: Ananth Narayan <ananth.narayan@amd.com> Cc: Andi Kleen <ak@linux.intel.com> Cc: Andrew Kilroy <andrew.kilroy@arm.com> Cc: Caleb Biggers <caleb.biggers@intel.com> Cc: Felix Fietkau <nbd@nbd.name> Cc: Ingo Molnar <mingo@redhat.com> Cc: James Clark <james.clark@arm.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Kajol Jain <kjain@linux.ibm.com> Cc: Kan Liang <kan.liang@linux.intel.com> Cc: Kshipra Bopardikar <kshipra.bopardikar@intel.com> Cc: Like Xu <likexu@tencent.com> Cc: Mark Rutland <mark.rutland@arm.com> Cc: Mathieu Poirier <mathieu.poirier@linaro.org> Cc: Namhyung Kim <namhyung@kernel.org> Cc: Nick Forrington <nick.forrington@arm.com> Cc: Paul Clarke <pc@us.ibm.com> Cc: Perry Taylor <perry.taylor@intel.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Qi Liu <liuqi115@huawei.com> Cc: Ravi Bangoria <ravi.bangoria@amd.com> Cc: Sandipan Das <sandipan.das@amd.com> Cc: Santosh Shukla <santosh.shukla@amd.com> Cc: Stephane Eranian <eranian@google.com> Cc: Will Deacon <will@kernel.org> Cc: Xing Zhengjun <zhengjun.xing@linux.intel.com> Link: https://lore.kernel.org/r/20220511211526.1021908-5-irogers@google.comSigned-off-by:
-
Ian Rogers authored
The event had two umasks with the umask of 3 being correct. Note: this change wasn't automatically generated as there is no CSV for Ivytown uncore events at: https://github.com/intel/event-converter-for-linux-perfReviewed-by:
-
Ian Rogers authored
Remove unnecessary empty groups. Reviewed-by:
-
Ian Rogers authored
Append the PMU information from "Unit" to the description later. This avoids a problem when "Unit" appears early in a json event and the information prepends the description rather than being the expected suffix. Update the pmu-events test so that expectations now match the improved output. Reviewed-by:
-
Ian Rogers authored
Check if the error code is EACCES and make the test a skip with a "permissions" skip reason if so. Committer testing: Before: $ perf test PERF_RECORD 8: PERF_RECORD_* events & perf_sample fields : FAILED! $ After: $ perf test PERF_RECORD 8: PERF_RECORD_* events & perf_sample fields : Skip (permissions) $ Signed-off-by: -
Ian Rogers authored
Break multiple tests in the main test into individual test cases. Make better use of skip and add reasons. Skip also for parse event permission issues (detected by searching the error string). Rather than break out of tests on the first failure, keep going and logging to pr_debug. Signed-off-by:
-
Ian Rogers authored
Remove two unused variables. Make structs const. Also fix the array index (aka id) for the event software/r0x1a/. Signed-off-by:
-
Ian Rogers authored
Remove an unused variables. Make structs const. Fix checkpatch issue wrt unsigned not being with an int. Signed-off-by:
-
Ian Rogers authored
If opening the event fails for basic mmap with EACCES it is more likely permission related that a true error. Mark the test as skip in this case and add a skip reason. Committer testing: Before: $ perf test "mmap interface" 4: Read samples using the mmap interface : FAILED! $ After: $ perf test "mmap interface" 4: Read samples using the mmap interface : Skip (permissions) $ Signed-off-by: -
Ian Rogers authored
Failures to open the tracepoint cause this test to fail, however, typically such failures are permission related. Lower the failure to just skipping the test in those cases and add a skip reason. Committer testing: Before: $ perf test "openat syscall" 2: Detect openat syscall event : FAILED! 3: Detect openat syscall event on all cpus : FAILED! $ After: $ perf test "openat syscall" 2: Detect openat syscall event : Skip (permissions) 3: Detect openat syscall event on all cpus : Skip (permissions) $ Signed-off-by: -
Ian Rogers authored
Currently failures in reading vmlinux or kallsyms result in a test failure. However, the failure is typically permission related. Prefer to flag these failures as skip. Committer testing: Before: $ perf test vmlinux 1: vmlinux symtab matches kallsyms : FAILED! $ After: $ perf test vmlinux 1: vmlinux symtab matches kallsyms : Skip $ Signed-off-by: -
Ian Rogers authored
When a suite has just 1 subtest, the subtest number is given as -1 to avoid indented printing. When this subtest number is seen for the skip reason, use the reason of the first test. Reviewed-by:
Namhyung Kim <namhyung@kernel.org> Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com> Cc: Carsten Haitzler <carsten.haitzler@arm.com> Cc: Ingo Molnar <mingo@redhat.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Marco Elver <elver@google.com> Cc: Mark Rutland <mark.rutland@arm.com> Cc: Michael Petlan <mpetlan@redhat.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Riccardo Mancini <rickyman7@gmail.com> Cc: Sohaib Mohamed <sohaib.amhmd@gmail.com> Cc: Stephane Eranian <eranian@google.com> Link: https://lore.kernel.org/r/20220518042027.836799-2-irogers@google.comSigned-off-by:
-
Ian Rogers authored
Try to disambiguate further when perf_counts is being accessed it is with a cpu map index rather than a CPU. Signed-off-by:
-
Ian Rogers authored
BPF counters are typically running across all CPUs and so the CPU map index and CPU number are the same. There may be cases with offline CPUs where this isn't the case and so ensure the cpu map index for perf_counts is going to be a valid index by explicitly iterating over the CPU map. This also makes it clearer that users of perf_counts are using an index. Collapse some multiple uses of perf_counts into single uses. Signed-off-by:
-
Ian Rogers authored
A variant of perf_cpu_map__for_each_cpu() that just iterates index values without the corresponding load of the CPU. Signed-off-by:
-
Ian Rogers authored
This would have caught: "Subject: Re: perf stat report segfaults" https://lore.kernel.org/linux-perf-users/CAP-5=fWQR=sCuiSMktvUtcbOLidEpUJLCybVF6=BRvORcDOq+g@mail.gmail.com/Signed-off-by:
-
Namhyung Kim authored
The -t option is to show per-thread lock stat like below: $ perf lock report -t -F acquired,contended,avg_wait Name acquired contended avg wait (ns) perf 240569 9 5784 swapper 106610 19 543 :15789 17370 2 14538 ContainerMgr 8981 6 874 sleep 5275 1 11281 ContainerThread 4416 4 944 RootPressureThr 3215 5 1215 rcu_preempt 2954 0 0 ContainerMgr 2560 0 0 unnamed 1873 0 0 EventManager_De 1845 1 636 futex-default-S 1609 0 0 ... Committer notes: Add that option to the 'perf lock report' man page. Signed-off-by: -
Namhyung Kim authored
Currently it discards a lock_stat for a lock instance when there's a broken lock_seq_stat in a single task for the lock. But it also means that the existing (and later) valid lock stat info for that lock will be discarded as well. This is not ideal since we can lose many valuable info because of a single failure. Actually those failures are indepent to the existing stat. So we can only discard the broken lock_seq_stat but keep the valid lock_stat. The discarded lock_seq_stat will be reallocated in a subsequent event with SEQ_STATE_UNINITIALIZED which will be ignored until it see the start of the next sequence. So it should be ok just free it. Before: $ perf lock report -F acquired,contended,avg_wait Warning: Processed 1401603 events and lost 18 chunks! Check IO/CPU overload! Name acquired contended avg wait (ns) rcu_read_lock 251225 0 0 &(ei->i_block_re... 8731 0 0 &sb->s_type->i_l... 8731 0 0 hrtimer_bases.lock 5261 0 0 hrtimer_bases.lock 2626 0 0 hrtimer_bases.lock 1953 0 0 hrtimer_bases.lock 1382 0 0 cpu_hotplug_lock 1350 0 0 hrtimer_bases.lock 1273 0 0 hrtimer_bases.lock 1269 0 0 hrtimer_bases.lock 1198 0 0 ... New: Name acquired contended avg wait (ns) rcu_read_lock 251225 0 0 tk_core.seq.seqc... 54074 0 0 &xa->xa_lock 17470 0 0 &ei->i_es_lock 17464 0 0 &ei->i_raw_lock 9391 0 0 &mapping->privat... 8734 0 0 &ei->i_data_sem 8731 0 0 &(ei->i_block_re... 8731 0 0 &sb->s_type->i_l... 8731 0 0 jiffies_seq.seqc... 6953 0 0 &mm->mmap_lock 6889 0 0 balancing 5768 0 0 hrtimer_bases.lock 5261 0 0 ... Signed-off-by: -
Leo Yan authored
The 'N/A' metric is added for store operations, update documentation to reflect changes in the report table. Signed-off-by:
Leo Yan <leo.yan@linaro.org> Acked-by:
-
Leo Yan authored
Since now we have the statistics 'st_na' for store operations, add dimensions for the 'N/A' (no available memory level) metrics and the associated percentage calculation for the single cache line view. Signed-off-by:
-
Leo Yan authored
Sometimes we don't know memory store operations happen on exactly which memory (or cache) level, the memory level flag is set to PERF_MEM_LVL_NA in this case; a practical example is Arm SPE AUX trace sets this flag for all store operations due to absent info for cache level. This patch is to add a new item "st_na" in structure c2c_stats to add statistics for store operations with no available cache level. Signed-off-by:
-
Ian Rogers authored
LIBBPF requires LIBELF so doing "make BUILD_BPF_SKEL=1 NO_LIBELF=1" fails with compiler errors about missing declarations. Similar could happen if libbpf feature detection fails. Prefer to error when BUILD_BPF_SKEL is enabled but LIBBPF isn't. Signed-off-by:
-
Arnaldo Carvalho de Melo authored
To get the rest of 5.18. Signed-off-by:
-
- 22 May, 2022 4 commits
-
-
Linus Torvalds authored
-
David Howells authored
If a callback break occurs (change notification), afs_getattr() needs to issue an FS.FetchStatus RPC operation to update the status of the file being examined by the stat-family of system calls. Fix afs_getattr() to do this if AFS_VNODE_CB_PROMISED has been cleared on a vnode by a callback break. Skip this if AT_STATX_DONT_SYNC is set. This can be tested by appending to a file on one AFS client and then using "stat -L" to examine its length on a machine running kafs. This can also be watched through tracing on the kafs machine. The callback break is seen: kworker/1:1-46 [001] ..... 978.910812: afs_cb_call: c=0000005f YFSCB.CallBack kworker/1:1-46 [001] ...1. 978.910829: afs_cb_break: 100058:23b4c:242d2c2 b=2 s=1 break-cb kworker/1:1-46 [001] ..... 978.911062: afs_call_done: c=0000005f ret=0 ab=0 [0000000082994ead] And then the stat command generated no traffic if unpatched, but with this change a call to fetch the status can be observed: stat-4471 [000] ..... 986.744122: afs_make_fs_call: c=000000ab 100058:023b4c:242d2c2 YFS.FetchStatus stat-4471 [000] ..... 986.745578: afs_call_done: c=000000ab ret=0 ab=0 [0000000087fc8c84] Fixes: 08e0e7c8 ("[AF_RXRPC]: Make the in-kernel AFS filesystem use AF_RXRPC.") Reported-by:Markus Suvanto <markus.suvanto@gmail.com> Signed-off-by:
David Howells <dhowells@redhat.com> cc: Marc Dionne <marc.dionne@auristor.com> cc: linux-afs@lists.infradead.org Tested-by:
Linus Torvalds <torvalds@linux-foundation.org>
-
git://git.kernel.org/pub/scm/linux/kernel/git/wsa/linuxLinus Torvalds authored
Pull i2c fixes from Wolfram Sang: "Some I2C driver bugfixes for 5.18. Nothing spectacular but worth fixing" * 'i2c/for-current' of git://git.kernel.org/pub/scm/linux/kernel/git/wsa/linux: drivers: i2c: thunderx: Allow driver to work with ACPI defined TWSI controllers i2c: ismt: Provide a DMA buffer for Interrupt Cause Logging i2c: mt7621: fix missing clk_disable_unprepare() on error in mtk_i2c_probe()
-
Linus Torvalds authored
Merge tag 'perf-tools-fixes-for-v5.18-2022-05-21' of git://git.kernel.org/pub/scm/linux/kernel/git/acme/linux Pull perf tools fixes from Arnaldo Carvalho de Melo: - Fix and validate CPU map inputs in synthetic PERF_RECORD_STAT events in 'perf stat'. - Fix x86's arch__intr_reg_mask() for the hybrid platform. - Address 'perf bench numa' compiler error on s390. - Fix check for btf__load_from_kernel_by_id() in libbpf. - Fix "all PMU test" 'perf test' to skip hv_24x7/hv_gpci tests on powerpc. - Fix session topology test to skip the test in guest environment. - Skip BPF 'perf test' if clang is not present. - Avoid shell test description infinite loop in 'perf test'. - Fix Intel LBR callstack entries and nr print message. * tag 'perf-tools-fixes-for-v5.18-2022-05-21' of git://git.kernel.org/pub/scm/linux/kernel/git/acme/linux: perf session: Fix Intel LBR callstack entries and nr print message perf test bpf: Skip test if clang is not present perf test session topology: Fix test to skip the test in guest environment perf bench numa: Address compiler error on s390 perf test: Avoid shell test description infinite loop perf regs x86: Fix arch__intr_reg_mask() for the hybrid platform perf test: Fix "all PMU test" to skip hv_24x7/hv_gpci tests on powerpc perf stat: Fix and validate CPU map inputs in synthetic PERF_RECORD_STAT events perf build: Fix check for btf__load_from_kernel_by_id() in libbpf
-
- 21 May, 2022 3 commits
-
-
git://git.kernel.org/pub/scm/linux/kernel/git/dtor/inputLinus Torvalds authored
Pull input fixes from Dmitry Torokhov: "A small fixup to ili210x touchscreen driver, and updated maintainer entry for the device tree binding of Mediatek 6779 keypad: - fix reset timing of Ilitek touchscreens - update maintainer entry of DT binding of Mediatek 6779 keypad" * tag 'input-for-v5.18-rc7' of git://git.kernel.org/pub/scm/linux/kernel/git/dtor/input: Input: ili210x - use one common reset implementation Input: ili210x - fix reset timing dt-bindings: input: mediatek,mt6779-keypad: update maintainer
-
git://git.kernel.org/pub/scm/linux/kernel/git/jejb/scsiLinus Torvalds authored
Pull SCSI fixes from James Bottomley: "Two patches, both in drivers. The iscsi one is fixing the cpumask issue you commented on and the ufs one is a late arriving fix for conditions that can occur in Host Performance Booster reads" * tag 'scsi-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/jejb/scsi: scsi: ufs: core: Fix referencing invalid rsp field scsi: target: Fix incorrect use of cpumask_t
-
Chengdong Li authored
When generating callstack information from branch_stack(Intel LBR), the actual number of callstack entry should be bigger than the number of branch_stack, for example: branch_stack records: B() -> C() A() -> B() converted callstack records should be: C() B() A() though, the number of callstack equals to the number of branch stack plus 1. This patch fixes above issue in branch_stack__printf(). For example, # echo 'scale=2000; 4*a(1)' > cmd # perf record --call-graph lbr bc -l < cmd Before applying this patch, `perf script -D` output: 1220022677386876 0x2a40 [0xd8]: PERF_RECORD_SAMPLE(IP, 0x4002): 17990/17990: 0x40a6d6 period: 894172 addr: 0 ... LBR call chain: nr:8 ..... 0: fffffffffffffe00 ..... 1: 000000000040a410 ..... 2: 000000000040573c ..... 3: 0000000000408650 ..... 4: 00000000004022f2 ..... 5: 00000000004015f5 ..... 6: 00007f5ed6dcb553 ..... 7: 0000000000401698 ... FP chain: nr:2 ..... 0: fffffffffffffe00 ..... 1: 000000000040a6d8 ... branch callstack: nr:6 # which is not consistent with LBR records. ..... 0: 000000000040a410 ..... 1: 0000000000408650 # ditto ..... 2: 00000000004022f2 ..... 3: 00000000004015f5 ..... 4: 00007f5ed6dcb553 ..... 5: 0000000000401698 ... thread: bc:17990 ...... dso: /usr/bin/bc bc 17990 1220022.677386: 894172 cycles: 40a410 [unknown] (/usr/bin/bc) 40573c [unknown] (/usr/bin/bc) 408650 [unknown] (/usr/bin/bc) 4022f2 [unknown] (/usr/bin/bc) 4015f5 [unknown] (/usr/bin/bc) 7f5ed6dcb553 __libc_start_main+0xf3 (/usr/lib64/libc-2.17.so) 401698 [unknown] (/usr/bin/bc) After applied: 1220022677386876 0x2a40 [0xd8]: PERF_RECORD_SAMPLE(IP, 0x4002): 17990/17990: 0x40a6d6 period: 894172 addr: 0 ... LBR call chain: nr:8 ..... 0: fffffffffffffe00 ..... 1: 000000000040a410 ..... 2: 000000000040573c ..... 3: 0000000000408650 ..... 4: 00000000004022f2 ..... 5: 00000000004015f5 ..... 6: 00007f5ed6dcb553 ..... 7: 0000000000401698 ... FP chain: nr:2 ..... 0: fffffffffffffe00 ..... 1: 000000000040a6d8 ... branch callstack: nr:7 ..... 0: 000000000040a410 ..... 1: 000000000040573c ..... 2: 0000000000408650 ..... 3: 00000000004022f2 ..... 4: 00000000004015f5 ..... 5: 00007f5ed6dcb553 ..... 6: 0000000000401698 ... thread: bc:17990 ...... dso: /usr/bin/bc bc 17990 1220022.677386: 894172 cycles: 40a410 [unknown] (/usr/bin/bc) 40573c [unknown] (/usr/bin/bc) 408650 [unknown] (/usr/bin/bc) 4022f2 [unknown] (/usr/bin/bc) 4015f5 [unknown] (/usr/bin/bc) 7f5ed6dcb553 __libc_start_main+0xf3 (/usr/lib64/libc-2.17.so) 401698 [unknown] (/usr/bin/bc) Change from v1: - refined code style according to Jiri's review comments. Signed-off-by:
Chengdong Li <chengdongli@tencent.com> Acked-by:
-
