kernel/git/stable/linux.git/tools/perf, branch linux-6.0.y

perf stat: Fix handling of --for-each-cgroup with --bpf-counters to match non BPF mode

2023-01-12T11:00:41+00:00

[ Upstream commit 54b353a20c7e8be98414754f5aff98c8a68fcc1f ] The --for-each-cgroup can have the same cgroup multiple times, but this confuses BPF counters (since they have the same cgroup id), making only the last cgroup events to be counted. Let's check the cgroup name before adding a new entry to the cgroups list. Before: $ sudo ./perf stat -a --bpf-counters --for-each-cgroup /,/ sleep 1 Performance counter stats for 'system wide': msec cpu-clock / context-switches / cpu-migrations / page-faults / cycles / instructions / branches / branch-misses / 8,016.04 msec cpu-clock / # 7.998 CPUs utilized 6,152 context-switches / # 767.461 /sec 250 cpu-migrations / # 31.187 /sec 442 page-faults / # 55.139 /sec 613,111,487 cycles / # 0.076 GHz 280,599,604 instructions / # 0.46 insn per cycle 57,692,724 branches / # 7.197 M/sec 3,385,168 branch-misses / # 5.87% of all branches 1.002220125 seconds time elapsed After it becomes similar to the non-BPF mode: $ sudo ./perf stat -a --bpf-counters --for-each-cgroup /,/ sleep 1 Performance counter stats for 'system wide': 8,013.38 msec cpu-clock / # 7.998 CPUs utilized 6,859 context-switches / # 855.944 /sec 334 cpu-migrations / # 41.680 /sec 345 page-faults / # 43.053 /sec 782,326,119 cycles / # 0.098 GHz 471,645,724 instructions / # 0.60 insn per cycle 94,963,430 branches / # 11.851 M/sec 3,685,511 branch-misses / # 3.88% of all branches 1.001864539 seconds time elapsed Committer notes: As a reminder, to test with BPF counters one has to use BUILD_BPF_SKEL=1 in the make command line and have clang/llvm installed when building perf, otherwise the --bpf-counters option will not be available: # perf stat -a --bpf-counters --for-each-cgroup /,/ sleep 1 Error: unknown option `bpf-counters' Usage: perf stat [] [] -a, --all-cpus system-wide collection from all CPUs # Fixes: bb1c15b60b981d10 ("perf stat: Support regex pattern in --for-each-cgroup") Signed-off-by: Namhyung Kim Tested-by: Arnaldo Carvalho de Melo Cc: Adrian Hunter Cc: bpf@vger.kernel.org Cc: Ian Rogers Cc: Ingo Molnar Cc: Jiri Olsa Cc: Namhyung Kim Cc: Peter Zijlstra Cc: Song Liu Link: https://lore.kernel.org/r/20230104064402.1551516-5-namhyung@kernel.org Signed-off-by: Arnaldo Carvalho de Melo Signed-off-by: Sasha Levin

perf stat: Fix handling of unsupported cgroup events when using BPF counters

2023-01-12T11:00:41+00:00

[ Upstream commit 2d656b0f81b22101db0447f890e39fdd736b745e ] When --for-each-cgroup option is used, it fails when any of events is not supported and exits immediately. This is not how 'perf stat' handles unsupported events. Let's ignore the failure and proceed with others so that the output is similar to when BPF counters are not used: Before: $ sudo ./perf stat -a --bpf-counters -e L1-icache-loads,L1-dcache-loads --for-each-cgroup system.slice,user.slice sleep 1 Failed to open first cgroup events $ After it shows output similat to when --bpf-counters isn't specified: $ sudo ./perf stat -a --bpf-counters -e L1-icache-loads,L1-dcache-loads --for-each-cgroup system.slice,user.slice sleep 1 Performance counter stats for 'system wide': L1-icache-loads system.slice 29,892,418 L1-dcache-loads system.slice L1-icache-loads user.slice 52,497,220 L1-dcache-loads user.slice $ Fixes: 944138f048f7d759 ("perf stat: Enable BPF counter with --for-each-cgroup") Signed-off-by: Namhyung Kim Tested-by: Arnaldo Carvalho de Melo Cc: Adrian Hunter Cc: Ian Rogers Cc: Ingo Molnar Cc: Jiri Olsa Cc: Peter Zijlstra Cc: Song Liu Link: https://lore.kernel.org/r/20230104064402.1551516-4-namhyung@kernel.org Signed-off-by: Arnaldo Carvalho de Melo Signed-off-by: Sasha Levin

perf lock contention: Fix core dump related to not finding the "__sched_text_end" symbol on s/390

2023-01-12T11:00:40+00:00

[ Upstream commit d8d85ce86dc82de4f88b821a78f533b9d5b22a45 ] The test case perf lock contention dumps core on s390. Run the following commands: # ./perf lock record -- ./perf bench sched messaging # Running 'sched/messaging' benchmark: # 20 sender and receiver processes per group # 10 groups == 400 processes run Total time: 2.799 [sec] [ perf record: Woken up 1 times to write data ] [ perf record: Captured and wrote 0.073 MB perf.data (100 samples) ] # # ./perf lock contention Segmentation fault (core dumped) # The function call stack is lengthy, here are the top 5 functions: # gdb ./perf core.24048 GNU gdb (GDB) Fedora Linux 12.1-6.fc37 Core was generated by `./perf lock contention'. Program terminated with signal SIGSEGV, Segmentation fault. #0 0x00000000011dd25c in machine__is_lock_function (machine=0x3029e28, addr=1789230) at util/machine.c:3356 3356 machine->sched.text_end = kmap->unmap_ip(kmap, sym->start); (gdb) where #0 0x00000000011dd25c in machine__is_lock_function (machine=0x3029e28, addr=1789230) at util/machine.c:3356 #1 0x000000000109f244 in callchain_id (evsel=0x30313e0, sample=0x3ffea4f77d0) at builtin-lock.c:957 #2 0x000000000109e094 in get_key_by_aggr_mode (key=0x3ffea4f7290, addr=27758136, evsel=0x30313e0, sample=0x3ffea4f77d0) at builtin-lock.c:586 #3 0x000000000109f4d0 in report_lock_contention_begin_event (evsel=0x30313e0, sample=0x3ffea4f77d0) at builtin-lock.c:1004 #4 0x00000000010a00ae in evsel__process_contention_begin (evsel=0x30313e0, sample=0x3ffea4f77d0) at builtin-lock.c:1254 #5 0x00000000010a0e14 in process_sample_event (tool=0x3ffea4f8480, event=0x3ff85601ef8, sample=0x3ffea4f77d0, evsel=0x30313e0, machine=0x3029e28) at builtin-lock.c:1464 ..... The issue is in function machine__is_lock_function() in file ./util/machine.c lines 3355: /* should not fail from here */ sym = machine__find_kernel_symbol_by_name(machine, "__sched_text_end", &kmap); machine->sched.text_end = kmap->unmap_ip(kmap, sym->start) On s390 the symbol __sched_text_end is *NOT* in the symbol list and the resulting pointer sym is set to NULL. The sym->start is then a NULL pointer access and generates the core dump. The reason why __sched_text_end is not in the symbol list on s390 is simple: When the symbol list is created at perf start up with function calls dso__load +--> dso__load_vmlinux_path +--> dso__load_vmlinux +--> dso__load_sym +--> dso__load_sym_internal (reads kernel symbols) +--> symbols__fixup_end +--> symbols__fixup_duplicate The issue is in function symbols__fixup_duplicate(). It deletes all symbols with have the same address. On s390: # nm -g ~/linux/vmlinux| fgrep c68390 0000000000c68390 T __cpuidle_text_start 0000000000c68390 T __sched_text_end # two symbols have identical addresses and __sched_text_end is considered duplicate (in ascending sort order) and removed from the symbol list. Therefore it is missing and an invalid pointer reference occurs. The code checks for symbol __sched_text_start and when it exists assumes symbol __sched_text_end is also in the symbol table. However this is not the case on s390. Same situation exists for symbol __lock_text_start: 0000000000c68770 T __cpuidle_text_end 0000000000c68770 T __lock_text_start This symbol is also removed from the symbol table but used in function machine__is_lock_function(). To fix this and keep duplicate symbols in the symbol table, set symbol_conf.allow_aliases to true. This prevents the removal of duplicate symbols in function symbols__fixup_duplicate(). Output After: # ./perf lock contention contended total wait max wait avg wait type caller 48 124.39 ms 123.99 ms 2.59 ms rwsem:W unlink_anon_vmas+0x24a 47 83.68 ms 83.26 ms 1.78 ms rwsem:W free_pgtables+0x132 5 41.22 us 10.55 us 8.24 us rwsem:W free_pgtables+0x140 4 40.12 us 20.55 us 10.03 us rwsem:W copy_process+0x1ac8 # Fixes: 0d2997f750d1de39 ("perf lock: Look up callchain for the contended locks") Signed-off-by: Thomas Richter Acked-by: Namhyung Kim Cc: Heiko Carstens Cc: Sumanth Korikkar Cc: Sven Schnelle Cc: Vasily Gorbik Link: https://lore.kernel.org/r/20221230102627.2410847-1-tmricht@linux.ibm.com Signed-off-by: Arnaldo Carvalho de Melo Signed-off-by: Sasha Levin

perf tools: Fix resources leak in perf_data__open_dir()

2023-01-12T11:00:40+00:00

[ Upstream commit 0a6564ebd953c4590663c9a3c99a3ea9920ade6f ] In perf_data__open_dir(), opendir() opens the directory stream. Add missing closedir() to release it after use. Fixes: eb6176709b235b96 ("perf data: Add perf_data__open_dir_data function") Reviewed-by: Adrian Hunter Signed-off-by: Miaoqian Lin Cc: Alexander Shishkin Cc: Alexey Bayduraev Cc: Ingo Molnar Cc: Jiri Olsa Cc: Mark Rutland Cc: Namhyung Kim Cc: Peter Zijlstra Link: https://lore.kernel.org/r/20221229090903.1402395-1-linmq006@gmail.com Signed-off-by: Arnaldo Carvalho de Melo Signed-off-by: Sasha Levin

perf probe: Fix to get the DW_AT_decl_file and DW_AT_call_file as unsinged data

2023-01-12T11:00:28+00:00

[ Upstream commit a9dfc46c67b52ad43b8e335e28f4cf8002c67793 ] DWARF version 5 standard Sec 2.14 says that Any debugging information entry representing the declaration of an object, module, subprogram or type may have DW_AT_decl_file, DW_AT_decl_line and DW_AT_decl_column attributes, each of whose value is an unsigned integer constant. So it should be an unsigned integer data. Also, even though the standard doesn't clearly say the DW_AT_call_file is signed or unsigned, the elfutils (eu-readelf) interprets it as unsigned integer data and it is natural to handle it as unsigned integer data as same as DW_AT_decl_file. This changes the DW_AT_call_file as unsigned integer data too. Fixes: 3f4460a28fb2f73d ("perf probe: Filter out redundant inline-instances") Signed-off-by: Masami Hiramatsu Acked-by: Namhyung Kim Cc: Alexander Shishkin Cc: Ingo Molnar Cc: Jiri Olsa Cc: Mark Rutland Cc: Masami Hiramatsu Cc: Peter Zijlstra Cc: stable@vger.kernel.org Cc: Steven Rostedt (VMware) Link: https://lore.kernel.org/r/166761727445.480106.3738447577082071942.stgit@devnote3 Signed-off-by: Arnaldo Carvalho de Melo Signed-off-by: Sasha Levin

perf probe: Use dwarf_attr_integrate as generic DWARF attr accessor

2023-01-12T11:00:28+00:00

[ Upstream commit f828929ab7f0dc3353e4a617f94f297fa8f3dec3 ] Use dwarf_attr_integrate() instead of dwarf_attr() for generic attribute acccessor functions, so that it can find the specified attribute from abstact origin DIE etc. Signed-off-by: Masami Hiramatsu Acked-by: Namhyung Kim Cc: Alexander Shishkin Cc: Ingo Molnar Cc: Jiri Olsa Cc: Mark Rutland Cc: Peter Zijlstra Cc: Steven Rostedt (VMware) Link: https://lore.kernel.org/r/166731051988.2100653.13595339994343449770.stgit@devnote3 Signed-off-by: Arnaldo Carvalho de Melo Stable-dep-of: a9dfc46c67b5 ("perf probe: Fix to get the DW_AT_decl_file and DW_AT_call_file as unsinged data") Signed-off-by: Sasha Levin

perf test: Fix "all PMU test" to skip parametrized events

2022-12-31T12:26:50+00:00

[ Upstream commit b50d691e50e600fab82b423be871860537d75dc9 ] Parametrized events are not only a powerpc domain. They occur on other platforms too (e.g. aarch64). They should be ignored in this testcase, since proper setup of the parameters is out of scope of this script. Let's not filter them out by PMU name, but rather based on the fact that they expect a parameter. Fixes: 451ed8058c69a3fe ("perf test: Fix "all PMU test" to skip hv_24x7/hv_gpci tests on powerpc") Signed-off-by: Michael Petlan Cc: Athira Rajeev Cc: Disha Goel Cc: Ian Rogers Cc: Michael Ellerman Cc: Nageswara R Sastry Link: https://lore.kernel.org/r/20221219163008.9691-1-mpetlan@redhat.com Signed-off-by: Arnaldo Carvalho de Melo Signed-off-by: Sasha Levin

perf debug: Set debug_peo_args and redirect_to_stderr variable to correct values in perf_quiet_option()

2022-12-31T12:26:50+00:00

[ Upstream commit 188ac720d364035008a54d249cf47b4cc100f819 ] When perf uses quiet mode, perf_quiet_option() sets the 'debug_peo_args' variable to -1, and display_attr() incorrectly determines the value of 'debug_peo_args'. As a result, unexpected information is displayed. Before: # perf record --quiet -- ls > /dev/null ------------------------------------------------------------ perf_event_attr: size 128 { sample_period, sample_freq } 4000 sample_type IP|TID|TIME|PERIOD read_format ID|LOST disabled 1 inherit 1 mmap 1 comm 1 freq 1 enable_on_exec 1 task 1 precise_ip 3 sample_id_all 1 exclude_guest 1 mmap2 1 comm_exec 1 ksymbol 1 bpf_event 1 ------------------------------------------------------------ ... After: # perf record --quiet -- ls > /dev/null # redirect_to_stderr is a similar problem. Fixes: f78eaef0e0493f60 ("perf tools: Allow to force redirect pr_debug to stderr.") Fixes: ccd26741f5e6bdf2 ("perf tool: Provide an option to print perf_event_open args and return value") Suggested-by: Adrian Hunter Reviewed-by: Adrian Hunter Signed-off-by: Yang Jihong Cc: Alexander Shishkin Cc: Andi Kleen Cc: Carsten Haitzler Cc: Ian Rogers Cc: Ingo Molnar Cc: Jiri Olsa Cc: Leo Yan Cc: Mark Rutland Cc: martin.lau@kernel.org Cc: Masami Hiramatsu Cc: Namhyung Kim Cc: Peter Zijlstra Cc: Ravi Bangoria Cc: Ravi Bangoria Link: https://lore.kernel.org/r/20221220035702.188413-2-yangjihong1@huawei.com Signed-off-by: Arnaldo Carvalho de Melo Signed-off-by: Sasha Levin

perf stat: Do not delay the workload with --delay

2022-12-31T12:26:31+00:00

[ Upstream commit c587e77e100fa40eb6af10e00497c67acf493f33 ] The -D/--delay option is to delay the measure after the program starts. But the current code goes to sleep before starting the program so the program is delayed too. This is not the intention, let's fix it. Before: $ time sudo ./perf stat -a -e cycles -D 3000 sleep 4 Events disabled Events enabled Performance counter stats for 'system wide': 4,326,949,337 cycles 4.007494118 seconds time elapsed real 0m7.474s user 0m0.356s sys 0m0.120s It ran the workload for 4 seconds and gave the 3 second delay. So it should skip the first 3 second and measure the last 1 second only. But as you can see, it delays 3 seconds and ran the workload after that for 4 seconds. So the total time (real) was 7 seconds. After: $ time sudo ./perf stat -a -e cycles -D 3000 sleep 4 Events disabled Events enabled Performance counter stats for 'system wide': 1,063,551,013 cycles 1.002769510 seconds time elapsed real 0m4.484s user 0m0.385s sys 0m0.086s The bug was introduced when it changed enablement of system-wide events with a command line workload. But it should've considered the initial delay case. The code was reworked since then (in bb8bc52e7578) so I'm afraid it won't be applied cleanly. Fixes: d0a0a511493d2695 ("perf stat: Fix forked applications enablement of counters") Reported-by: Kevin Nomura Signed-off-by: Namhyung Kim Tested-by: Thomas Richter Cc: Adrian Hunter Cc: Ian Rogers Cc: Ingo Molnar Cc: Jiri Olsa Cc: Peter Zijlstra Cc: Sumanth Korikkar Link: https://lore.kernel.org/r/20221212230820.901382-1-namhyung@kernel.org Signed-off-by: Arnaldo Carvalho de Melo Signed-off-by: Sasha Levin

perf off_cpu: Fix a typo in BTF tracepoint name, it should be 'btf_trace_sched_switch'

2022-12-31T12:26:31+00:00

[ Upstream commit 167b266bf66c5b93171011ef9d1f09b070c2c537 ] In BTF, tracepoint definitions have the "btf_trace_" prefix. The off-cpu profiler needs to check the signature of the sched_switch event using that definition. But there's a typo (s/bpf/btf/) so it failed always. Fixes: b36888f71c8542cd ("perf record: Handle argument change in sched_switch") Signed-off-by: Namhyung Kim Cc: Adrian Hunter Cc: bpf@vger.kernel.org Cc: Ian Rogers Cc: Ingo Molnar Cc: Jiri Olsa Cc: Namhyung Kim Cc: Peter Zijlstra Cc: Song Liu Link: https://lore.kernel.org/r/20221208182636.524139-1-namhyung@kernel.org Signed-off-by: Arnaldo Carvalho de Melo Signed-off-by: Sasha Levin