- 04 Apr, 2023 40 commits
-
-
Ian Rogers authored
Don't wrap stdin and stdout of subprocess with streams, use the api/io library for buffering. Signed-off-by:
Ian Rogers <irogers@google.com> Acked-by:
Namhyung Kim <namhyung@kernel.org> Cc: Adrian Hunter <adrian.hunter@intel.com> Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com> Cc: Ingo Molnar <mingo@redhat.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Mark Rutland <mark.rutland@arm.com> Cc: Nathan Chancellor <nathan@kernel.org> Cc: Nick Desaulniers <ndesaulniers@google.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Tom Rix <trix@redhat.com> Cc: llvm@lists.linux.dev Link: https://lore.kernel.org/r/20230403184033.1836023-2-irogers@google.comSigned-off-by:
Arnaldo Carvalho de Melo <acme@redhat.com>
-
Ian Rogers authored
Reads a line to allocated memory up to a newline following the getline API. Committer notes: It also adds this new function to the 'api io' 'perf test' entry: $ perf test "api io" 64: Test api io : Ok $ Signed-off-by:
-
Namhyung Kim authored
Intel-PT calls perf_pmu__scan_file() a lot, let's use relative address when it accesses multiple files at one place. Signed-off-by:
Adrian Hunter <adrian.hunter@intel.com> Acked-by:
-
Namhyung Kim authored
These two helpers will also use openat() to reduce the overhead with relative pathnames. Convert other functions in pmu_lookup() to use the new helpers. Committer testing: Before: ⬢[acme@toolbox perf-tools-next]$ perf bench internals pmu-scan # Running 'internals/pmu-scan' benchmark: Computing performance of sysfs PMU event scan for 100 times Average PMU scanning took: 2729.040 usec (+- 7.117 usec) ⬢[acme@toolbox perf-tools-next]$ After: ⬢[acme@toolbox perf-tools-next]$ perf bench internals pmu-scan # Running 'internals/pmu-scan' benchmark: Computing performance of sysfs PMU event scan for 100 times Average PMU scanning took: 2419.870 usec (+- 9.057 usec) ⬢[acme@toolbox perf-tools-next]$ Signed-off-by: -
Namhyung Kim authored
Likewise, x86 needs to traverse the PMU list to build alias. Let's use the new helpers to use relative paths. Signed-off-by:
-
Namhyung Kim authored
Likewise, it needs to traverse the pmu/caps directory, let's use openat() with the dirfd instead of open() using the absolute path. Signed-off-by:
-
Namhyung Kim authored
The PMU information is in the kernel sysfs so it needs to scan the directory to get the whole information like event aliases, formats and so on. During the traversal, it opens a lot of files and directories like below: dir = opendir("/sys/bus/event_source/devices"); while (dentry = readdir(dir)) { char buf[PATH_MAX]; snprintf(buf, sizeof(buf), "%s/%s", "/sys/bus/event_source/devices", dentry->d_name); fd = open(buf, O_RDONLY); ... } But this is not good since it needs to copy the string to build the absolute pathname, and it makes redundant pathname walk (from the /sys) unnecessarily. We can use openat(2) to open the file in the given directory. While it's not a problem ususally, it can be a problem when the kernel has contentions on the sysfs. Add a couple of new helper to return the file descriptor of PMU directory so that it can use it with relative paths. * perf_pmu__event_source_devices_fd() - returns a fd for the PMU root ("/sys/bus/event_source/devices") * perf_pmu__pathname_fd() - returns a fd for "<pmu>/<file>" under the PMU root Now the above code can be converted something like below: dirfd = perf_pmu__event_source_devices_fd(); dir = fdopendir(dirfd); while (dentry = readdir(dir)) { fd = openat(dirfd, dentry->d_name, O_RDONLY); ... } Signed-off-by: -
Namhyung Kim authored
The pmu-scan benchmark will repeatedly scan the sysfs to get the available PMU information. $ ./perf bench internals pmu-scan # Running 'internals/pmu-scan' benchmark: Computing performance of sysfs PMU event scan for 100 times Average PMU scanning took: 6850.990 usec (+- 48.445 usec) Signed-off-by: -
Namhyung Kim authored
It seems there's no function to delete the perf pmu struct. Add the perf_pmu__destroy() to do the job. While at it, add some more helper functions to delete pmu aliases and caps. Signed-off-by:
-
Namhyung Kim authored
In the parse_events_multi_pmu_add() it passes the 'config' variable twice to parse_events_term__num() - one for config and another for loc_term. I'm not sure about the second one as it's converted to YYLTYPE variable. Asan reports it like below: In function ‘parse_events_term__num’, inlined from ‘parse_events_multi_pmu_add’ at util/parse-events.c:1602:6: util/parse-events.c:2653:64: error: array subscript ‘YYLTYPE[0]’ is partly outside array bounds of ‘char[8]’ [-Werror=array-bounds] 2653 | .err_term = loc_term ? loc_term->first_column : 0, | ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~^~~ util/parse-events.c: In function ‘parse_events_multi_pmu_add’: util/parse-events.c:1587:15: note: object ‘config’ of size 8 1587 | char *config; | ^~~~~~ cc1: all warnings being treated as errors Signed-off-by: -
Namhyung Kim authored
Committer notes: Added missing #include <unistd.h> for the close() prototype to fix this on Alma Linux 8: 1 21.54 almalinux:8 : FAIL gcc version 8.5.0 20210514 (Red Hat 8.5.0-16) (GCC) util/print-events.c: In function 'print_tracepoint_events': util/print-events.c:103:4: error: implicit declaration of function 'close'; did you mean 'clone'? [-Werror=implicit-function-declaration] close(evt_fd); ^~~~~ clone Also use the newly added scandirat feature test to check if that function is available, providing a HAVE_SCANDIRAT_SUPPORT conditional warning to the user if it isn't available. Signed-off-by: -
Arnaldo Carvalho de Melo authored
We use it just when listing tracepoint events, and for root, so just emit a warning about it to get users to ask the library maintainers to implement it, as suggested in this systemd ticket: https://github.com/systemd/casync/issues/129 Cc: Adrian Hunter <adrian.hunter@intel.com> Cc: Ian Rogers <irogers@google.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Namhyung Kim <namhyung@kernel.org> Link: https://lore.kernel.org/lkml/ZCwv4z5Dh%2FdHUMG6@kernel.org/Signed-off-by:
-
Adrian Hunter authored
After a standalone CBR (not associated with TSC), update the cycles reference timestamp and reset the cycle count, so that CYC timestamps are calculated relative to that point with the new frequency. Fixes: cc336186 ("perf tools: Add Intel PT support for decoding CYC packets") Signed-off-by:
-
Adrian Hunter authored
kallsyms is not completely in address order. In find_entire_kern_cb(), calculate the kernel end from the maximum address not the last symbol. Example: Before: $ sudo cat /proc/kallsyms | grep ' [twTw] ' | tail -1 ffffffffc00b8bd0 t bpf_prog_6deef7357e7b4530 [bpf] $ sudo cat /proc/kallsyms | grep ' [twTw] ' | sort | tail -1 ffffffffc15e0cc0 t iwl_mvm_exit [iwlmvm] $ perf.d093603a05aa record -v --kcore -e intel_pt// --filter 'filter *' -- uname |& grep filter Address filter: filter 0xffffffff93200000/0x2ceba000 After: $ perf.8fb0f7a01f8e record -v --kcore -e intel_pt// --filter 'filter *' -- uname |& grep filter Address filter: filter 0xffffffff93200000/0x2e3e2000 Fixes: 1b36c03e ("perf record: Add support for using symbols in address filters") Signed-off-by: -
Rob Herring authored
Arm SPEv1.3 adds new load/store operation subclasses for Memory Tagging Extension (MTE) and memory operations (MOPS). The memory operations are memcpy and memset. Add support for decoding these new subclasses in the raw decoding. Reviewed-by: Leo Yan <leo.yan@linaro.org Signed-off-by:
Rob Herring <robh@kernel.org> Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com> Cc: Ian Rogers <irogers@google.com> Cc: Ingo Molnar <mingo@redhat.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Leo Yan <leo.yan@linaro.org> Cc: Mark Rutland <mark.rutland@arm.com> Cc: Namhyung Kim <namhyung@kernel.org> Cc: Peter Zijlstra <peterz@infradead.org> Link: https://lore.kernel.org/r/20230327162057.4057188-1-robh@kernel.orgSigned-off-by:
-
Mike Leach authored
When using dynamically assigned CoreSight trace IDs the drivers can output the ID / CPU association as a PERF_RECORD_AUX_OUTPUT_HW_ID packet. Update cs-etm decoder to handle this packet by setting the CPU/Trace ID mapping. Reviewed-by:
James Clark <james.clark@arm.com> Signed-off-by:
Mike Leach <mike.leach@linaro.org> Acked-by:
Suzuki Poulouse <suzuki.poulose@arm.com> Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com> Cc: Darren Hart <darren@os.amperecomputing.com> Cc: Ganapatrao Kulkarni <gankulkarni@os.amperecomputing.com> Cc: Ingo Molnar <mingo@redhat.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Leo Yan <leo.yan@linaro.org> Cc: Mark Rutland <mark.rutland@arm.com> Cc: Namhyung Kim <namhyung@kernel.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Will Deacon <will@kernel.org> Cc: coresight@lists.linaro.org Cc: linux-arm-kernel@lists.infradead.org Link: https://lore.kernel.org/r/20230331055645.26918-2-mike.leach@linaro.orgSigned-off-by:
-
Mike Leach authored
Trace IDs are now dynamically allocated. Previously used the static association algorithm that is no longer used. The 'cpu * 2 + seed' was outdated and broken for systems with high core counts (>46). as it did not scale and was broken for larger core counts. Trace ID will now be sent in PERF_RECORD_AUX_OUTPUT_HW_ID record. Legacy ID algorithm renamed and retained for limited backward compatibility use. Reviewed-by:
-
Mike Leach authored
The information to associate Trace ID and CPU will be changing. Drivers will start outputting this as a hardware ID packet in the data file which if present will be used in preference to the AUXINFO values. To prepare for this we provide a helper functions to do the individual ID mapping, and one to extract the IDs from the completed metadata blocks. Reviewed-by:
-
Namhyung Kim authored
It can fail to collect lock stat from BPF for various reasons. For example, I've got a report that sometimes time calculation seems wrong in case of contended spinlocks. I suspect the time delta went negative for some reason. Count them separately and show in the output like below: $ sudo perf lock contention -abE5 sleep 10 contended total wait max wait avg wait type caller 13 785.61 us 79.36 us 60.43 us spinlock remove_wait_queue+0x14 10 469.02 us 87.51 us 46.90 us spinlock prepare_to_wait+0x27 9 289.09 us 69.08 us 32.12 us spinlock finish_wait+0x36 114 251.05 us 8.56 us 2.20 us spinlock try_to_wake_up+0x1f5 132 188.63 us 5.01 us 1.43 us spinlock __wake_up_common_lock+0x62 === output for debug === bad: 1, total: 279 bad rate: 0.36 % histogram of failure reasons task: 1 stack: 0 time: 0 Signed-off-by: -
Namhyung Kim authored
It should not divide if the total number is 0. Otherwise it'd show NaN in the bad rate output. Also add a whitespace in the "output for debug" message. $ sudo perf lock contention -abv true Looking at the vmlinux_path (8 entries long) symsrc__init: cannot get elf header. Using /proc/kcore for kernel data Using /proc/kallsyms for symbols contended total wait max wait avg wait type caller === output for debug=== bad: 0, total: 0 bad rate: -nan % <------------------------- (here) histogram of events caused bad sequence acquire: 0 acquired: 0 contended: 0 release: 0 Signed-off-by: -
Ian Rogers authored
Update to versions 24 and 23 respectively. Adds the event BR_MISP_EXEC.INDIRECT. Signed-off-by:
-
Andreas Herrmann authored
'j' is of type int and start/end are of type 'long'. Thus 'j' might become negative and cause segfault in access_data(). Fix it by using 'long' for 'j' as well. Reviewed-by:
Andreas Herrmann <aherrmann@suse.de> Link: https://lore.kernel.org/r/20230330074202.14052-1-aherrmann@suse.deSigned-off-by:
-
Adrian Hunter authored
branch_callstack was added by commit 8b7bad58 ("perf callchain: Support handling complete branch stacks as histograms") but never used. Signed-off-by:
-
Adrian Hunter authored
Add --branch-history option, to act the same as that option does for perf report. Example: $ cat tcallf.c volatile a = 10000, b = 100000, c; __attribute__((noinline)) f2() { c = a / b; } __attribute__((noinline)) f1() { f2(); f2(); } main() { while (1) f1(); } $ gcc -w -g -o tcallf tcallf.c $ ./tcallf & [1] 29409 $ perf top -e cycles:u -t $(pidof tcallf) --stdio --no-children --branch-history PerfTop: 3819 irqs/sec kernel: 0.0% exact: 0.0% lost: 0/0 drop: 0/0 [4000Hz cycles:u], (target_tid: 29409) -------------------------------------------------------------------------------------------------------------------- 49.01% tcallf.c:5 [.] f2 tcallf | |--24.91%--f2 tcallf.c:4 | | | |--17.14%--f1 tcallf.c:11 (cycles:1) | | f1 tcallf.c:11 | | f2 tcallf.c:6 (cycles:3) | | f2 tcallf.c:4 | | f1 tcallf.c:10 (cycles:2) | | f1 tcallf.c:9 | | main tcallf.c:16 (cycles:1) | | main tcallf.c:16 | | main tcallf.c:16 (cycles:1) | | main tcallf.c:16 | | f1 tcallf.c:12 (cycles:1) | | f1 tcallf.c:12 | | f2 tcallf.c:6 (cycles:3) | | f2 tcallf.c:4 | | f1 tcallf.c:11 (cycles:1 iter:1 avg_cycles:12) | | f1 tcallf.c:11 | | f2 tcallf.c:6 (cycles:3 iter:1 avg_cycles:12) | | f2 tcallf.c:4 | | f1 tcallf.c:10 (cycles:2 iter:1 avg_cycles:12) | | | --7.78%--f1 tcallf.c:10 (cycles:2) | f1 tcallf.c:9 | main tcallf.c:16 (cycles:1) | main tcallf.c:16 | main tcallf.c:16 (cycles:1) | main tcallf.c:16 | f1 tcallf.c:12 (cycles:1) | f1 tcallf.c:12 | f2 tcallf.c:6 (cycles:3) | f2 tcallf.c:4 | f1 tcallf.c:11 (cycles:1) | f1 tcallf.c:11 | f2 tcallf.c:6 (cycles:3) | f2 tcallf.c:4 | f1 tcallf.c:10 (cycles:2 iter:1 avg_cycles:12) | f1 tcallf.c:9 | main tcallf.c:16 (cycles:1 iter:1 avg_cycles:12) | main tcallf.c:16 | main tcallf.c:16 (cycles:1 iter:1 avg_cycles:12) ... $ pkill tcallf [1]+ Terminated ./tcallf Signed-off-by: -
Ian Rogers authored
When a build is done without DEBUG=1 then define NDEBUG. This will compile out asserts and other debug code. Signed-off-by:
-
Ian Rogers authored
Make good on a comment and avoid a unused-but-set-variable warning. Signed-off-by:
-
Ian Rogers authored
With NDEBUG set the asserts are compiled out. This yields "unused-but-set-variable" variables. Move these variables behind NDEBUG to avoid the warning. Signed-off-by:
-
Ian Rogers authored
https://github.com/intel/perfmon/pull/65 Generated by: https://github.com/intel/perfmon/blob/main/scripts/create_perf_json.py The PR notes state: - E-Core TMA version 2.3. - FP_UOPS changed to FPDIV_Uops - Added BR_MISP breakdown stats - Frontend_Bandwidth/Latency changed to Fetch_Bandwidth/Latency - Load_Store_Bound changed to Memory_Bound - Icache changed to ICache_Misses - ITLB changed to ITLB_Misses - Store_Fwd changed to Store_Fwd_Blk Reviewed-by:
Kan Liang <kan.liang@linux.intel.com> Signed-off-by:
-
Ian Rogers authored
Allow addr2line to be set either on the command line or via the perfconfig file. This doesn't currently work with llvm-addr2line as the addr2line code emits two things: 1) the address to decode, 2) a bogus ',' value. The expectation is the bogus value will generate: ?? ??:0 that terminates the addr2line reading. However, the output from llvm-addr2line is a single line with just the input ',' locking up the addr2line reading that is expecting a second line. Signed-off-by:
-
Ian Rogers authored
Allow the setting of the objdump command in the perfconfig. Update man page for this new option. Signed-off-by:
-
Ian Rogers authored
Make struct annotation_options own the strings objdump_path and disassembler_style, freeing them on exit. Add missing strdup for disassembler_style when read from a config file. Committer notes: Converted free(obj->member) to zfree(&obj->member) in annotation_options__exit() Signed-off-by:
-
Ian Rogers authored
The annotation__default_options global variable was used to initialize annotation_options. Switch to the init/exit pattern as later changes will give ownership over strings and this will be necessary to avoid memory leaks. Committer note: Fix the GTK2=1 build, hist_entry__gtk_annotate() needs to receive a 'struct annotation_options' pointer. Signed-off-by:
-
Ian Rogers authored
If the default_sort_order isn't correctly strdup-ed warn and return an error. Debug warn if no option is matched. Signed-off-by:
-
Ian Rogers authored
Use the debug build indicator as the guide to free the session. This implements a behavior described in a comment, which is consequentially removed. Signed-off-by:
-
Adrian Hunter authored
do_realloc_array_as_needed() used memcpy() of zero size with a NULL pointer. Check the size first to avoid sanitize warning. Discovered using EXTRA_CFLAGS="-fsanitize=undefined -fsanitize=address". Reported-by:
kernel test robot <yujie.liu@intel.com> Signed-off-by:
-
Adrian Hunter authored
Use memcpy() to avoid unaligned access. Discovered using EXTRA_CFLAGS="-fsanitize=undefined -fsanitize=address". Fixes: ce4c8e79 ("perf symbols: Get symbols for .plt.got for x86-64") Reported-by:
-
Adrian Hunter authored
Fix use-after-free in get_plt_got_name(). Discovered using EXTRA_CFLAGS="-fsanitize=undefined -fsanitize=address". Fixes: ce4c8e79 ("perf symbols: Get symbols for .plt.got for x86-64") Reported-by:
-
Kajol Jain authored
Commit 3c22ba52 ("perf vendor events powerpc: Update POWER9 events") added and updated power9 PMU JSON events. However some of the JSON events which are part of other.json and pipeline.json files, contains UTF-8 characters in their brief description. Having UTF-8 character could breaks the perf build on some distros. Fix this issue by removing the UTF-8 characters from other.json and pipeline.json files. Result without the fix: [command]# file -i pmu-events/arch/powerpc/power9/* pmu-events/arch/powerpc/power9/cache.json: application/json; charset=us-ascii pmu-events/arch/powerpc/power9/floating-point.json: application/json; charset=us-ascii pmu-events/arch/powerpc/power9/frontend.json: application/json; charset=us-ascii pmu-events/arch/powerpc/power9/marked.json: application/json; charset=us-ascii pmu-events/arch/powerpc/power9/memory.json: application/json; charset=us-ascii pmu-events/arch/powerpc/power9/metrics.json: application/json; charset=us-ascii pmu-events/arch/powerpc/power9/nest_metrics.json: application/json; charset=us-ascii pmu-events/arch/powerpc/power9/other.json: application/json; charset=utf-8 pmu-events/arch/powerpc/power9/pipeline.json: application/json; charset=utf-8 pmu-events/arch/powerpc/power9/pmc.json: application/json; charset=us-ascii pmu-events/arch/powerpc/power9/translation.json: application/json; charset=us-ascii [command]# Result with the fix: [command]# file -i pmu-events/arch/powerpc/power9/* pmu-events/arch/powerpc/power9/cache.json: application/json; charset=us-ascii pmu-events/arch/powerpc/power9/floating-point.json: application/json; charset=us-ascii pmu-events/arch/powerpc/power9/frontend.json: application/json; charset=us-ascii pmu-events/arch/powerpc/power9/marked.json: application/json; charset=us-ascii pmu-events/arch/powerpc/power9/memory.json: application/json; charset=us-ascii pmu-events/arch/powerpc/power9/metrics.json: application/json; charset=us-ascii pmu-events/arch/powerpc/power9/nest_metrics.json: application/json; charset=us-ascii pmu-events/arch/powerpc/power9/other.json: application/json; charset=us-ascii pmu-events/arch/powerpc/power9/pipeline.json: application/json; charset=us-ascii pmu-events/arch/powerpc/power9/pmc.json: application/json; charset=us-ascii pmu-events/arch/powerpc/power9/translation.json: application/json; charset=us-ascii [command]# Fixes: 3c22ba52 ("perf vendor events powerpc: Update POWER9 events") Reported-by:
Arnaldo Carvalho de Melo <acme@kernel.com> Signed-off-by:
Kajol Jain <kjain@linux.ibm.com> Acked-by:
-
Yang Jihong authored
If no target is specified for 'latency' subcommand, the execution fails because - 1 (invalid value) is written to set_ftrace_pid tracefs file. Make system wide the default target, which is the same as the default behavior of 'trace' subcommand. Before the fix: # perf ftrace latency -T schedule failed to set ftrace pid After the fix: # perf ftrace latency -T schedule ^C# DURATION | COUNT | GRAPH | 0 - 1 us | 0 | | 1 - 2 us | 0 | | 2 - 4 us | 0 | | 4 - 8 us | 2828 | #### | 8 - 16 us | 23953 | ######################################## | 16 - 32 us | 408 | | 32 - 64 us | 318 | | 64 - 128 us | 4 | | 128 - 256 us | 3 | | 256 - 512 us | 0 | | 512 - 1024 us | 1 | | 1 - 2 ms | 4 | | 2 - 4 ms | 0 | | 4 - 8 ms | 0 | | 8 - 16 ms | 0 | | 16 - 32 ms | 0 | | 32 - 64 ms | 0 | | 64 - 128 ms | 0 | | 128 - 256 ms | 4 | | 256 - 512 ms | 2 | | 512 - 1024 ms | 0 | | 1 - ... s | 0 | | Fixes: 53be5028 ("perf ftrace: Add 'latency' subcommand") Signed-off-by:Yang Jihong <yangjihong1@huawei.com> Acked-by:
-
Tiezhu Yang authored
This is a follow up patch for the execve bench which is actually fork + execve, it makes sense to add the fork syscall benchmark to compare the execve part precisely. Some archs have no __NR_fork definition which is used only as a check condition to call test_fork(), let us just define it as -1 to avoid build error. Suggested-by:
Tiezhu Yang <yangtiezhu@loongson.cn> Acked-by:
-
