linux/tools/perf, branch v5.17

perf parse-events: Ignore case in topdown.slots check

2022-03-18T21:39:09Z

An issue with icelakex metrics: https://git.kernel.org/pub/scm/linux/kernel/git/acme/linux.git/tree/tools/perf/pmu-events/arch/x86/icelakex/icx-metrics.json?h=perf/core&id=65eab2bc7dab326ee892ec5a4c749470b368b51a#n48 That causes the slots not to be first. Fixes: 94dbfd6781a0e87b ("perf parse-events: Architecture specific leader override") Reported-by: Caleb Biggers Signed-off-by: Ian Rogers Cc: Alexander Shishkin Cc: Alexandre Torgue Cc: Andi Kleen Cc: Ingo Molnar Cc: James Clark Cc: Jiri Olsa Cc: John Garry Cc: Kan Liang Cc: Mark Rutland Cc: Maxime Coquelin Cc: Namhyung Kim Cc: Peter Zijlstra Cc: Stephane Eranian Cc: Zhengjun Xing Link: https://lore.kernel.org/r/20220317224309.543736-1-irogers@google.com Signed-off-by: Arnaldo Carvalho de Melo

perf evlist: Avoid iteration for empty evlist.

2022-03-18T21:39:09Z

As seen with 'perf stat --null ..' and reported in: https://lore.kernel.org/lkml/YjCLcpcX2peeQVCH@kernel.org/ v2. Avoids setting evsel in the empty list case as suggested by Jiri Olsa. Committer testing: Before: $ perf stat --null sleep 1 Segmentation fault (core dumped) $ After: $ perf stat --null sleep 1 Performance counter stats for 'sleep 1': 1.010340646 seconds time elapsed 0.001420000 seconds user 0.000000000 seconds sys $ Fixes: 472832d2c000b961 ("perf evlist: Refactor evlist__for_each_cpu()") Reported-by: Arnaldo Carvalho de Melo Signed-off-by: Ian Rogers Tested-by: Arnaldo Carvalho de Melo Acked-by: Jiri Olsa Cc: Adrian Hunter Cc: Ian Rogers Cc: Namhyung Kim Link: https://lore.kernel.org/r/20220317231643.550902-1-irogers@google.com Signed-off-by: Arnaldo Carvalho de Melo

perf symbols: Fix symbol size calculation condition

2022-03-18T21:39:09Z

Before this patch, the symbol end address fixup to be called, needed two conditions being met: if (prev->end == prev->start && prev->end != curr->start) Where "prev->end == prev->start" means that prev is zero-long (and thus needs a fixup) and "prev->end != curr->start" means that fixup hasn't been applied yet However, this logic is incorrect in the following situation: *curr = {rb_node = {__rb_parent_color = 278218928, rb_right = 0x0, rb_left = 0x0}, start = 0xc000000000062354, end = 0xc000000000062354, namelen = 40, type = 2 '\002', binding = 0 '\000', idle = 0 '\000', ignore = 0 '\000', inlined = 0 '\000', arch_sym = 0 '\000', annotate2 = false, name = 0x1159739e "kprobe_optinsn_page\t[__builtin__kprobes]"} *prev = {rb_node = {__rb_parent_color = 278219041, rb_right = 0x109548b0, rb_left = 0x109547c0}, start = 0xc000000000062354, end = 0xc000000000062354, namelen = 12, type = 2 '\002', binding = 1 '\001', idle = 0 '\000', ignore = 0 '\000', inlined = 0 '\000', arch_sym = 0 '\000', annotate2 = false, name = 0x1095486e "optinsn_slot"} In this case, prev->start == prev->end == curr->start == curr->end, thus the condition above thinks that "we need a fixup due to zero length of prev symbol, but it has been probably done, since the prev->end == curr->start", which is wrong. After the patch, the execution path proceeds to arch__symbols__fixup_end function which fixes up the size of prev symbol by adding page_size to its end offset. Fixes: 3b01a413c196c910 ("perf symbols: Improve kallsyms symbol end addr calculation") Signed-off-by: Michael Petlan Cc: Athira Jajeev Cc: Jiri Olsa Cc: Kajol Jain Cc: Madhavan Srinivasan Link: http://lore.kernel.org/lkml/20220317135536.805-1-mpetlan@redhat.com Signed-off-by: Arnaldo Carvalho de Melo

perf parse: Fix event parser error for hybrid systems

2022-03-12T14:04:55Z

This bug happened on hybrid systems when both cpu_core and cpu_atom have the same event name such as "UOPS_RETIRED.MS" while their event terms are different, then during perf stat, the event for cpu_atom will parse fail and then no output for cpu_atom. UOPS_RETIRED.MS -> cpu_core/period=0x1e8483,umask=0x4,event=0xc2,frontend=0x8/ UOPS_RETIRED.MS -> cpu_atom/period=0x1e8483,umask=0x1,event=0xc2/ It is because event terms in the "head" of parse_events_multi_pmu_add will be changed to event terms for cpu_core after parsing UOPS_RETIRED.MS for cpu_core, then when parsing the same event for cpu_atom, it still uses the event terms for cpu_core, but event terms for cpu_atom are different with cpu_core, the event parses for cpu_atom will fail. This patch fixes it, the event terms should be parsed from the original event. This patch can work for the hybrid systems that have the same event in more than 2 PMUs. It also can work in non-hybrid systems. Before: # perf stat -v -e UOPS_RETIRED.MS -a sleep 1 Using CPUID GenuineIntel-6-97-1 UOPS_RETIRED.MS -> cpu_core/period=0x1e8483,umask=0x4,event=0xc2,frontend=0x8/ Control descriptor is not initialized UOPS_RETIRED.MS: 2737845 16068518485 16068518485 Performance counter stats for 'system wide': 2,737,845 cpu_core/UOPS_RETIRED.MS/ 1.002553850 seconds time elapsed After: # perf stat -v -e UOPS_RETIRED.MS -a sleep 1 Using CPUID GenuineIntel-6-97-1 UOPS_RETIRED.MS -> cpu_core/period=0x1e8483,umask=0x4,event=0xc2,frontend=0x8/ UOPS_RETIRED.MS -> cpu_atom/period=0x1e8483,umask=0x1,event=0xc2/ Control descriptor is not initialized UOPS_RETIRED.MS: 1977555 16076950711 16076950711 UOPS_RETIRED.MS: 568684 8038694234 8038694234 Performance counter stats for 'system wide': 1,977,555 cpu_core/UOPS_RETIRED.MS/ 568,684 cpu_atom/UOPS_RETIRED.MS/ 1.004758259 seconds time elapsed Fixes: fb0811535e92c6c1 ("perf parse-events: Allow config on kernel PMU events") Reviewed-by: Kan Liang Signed-off-by: Zhengjun Xing Cc: Adrian Hunter Cc: Alexander Shishkin Cc: Andi Kleen Cc: Ian Rogers Cc: Ingo Molnar Cc: Jiri Olsa Cc: Peter Zijlstra Link: https://lore.kernel.org/r/20220307151627.30049-1-zhengjun.xing@linux.intel.com Signed-off-by: Arnaldo Carvalho de Melo

perf bench: Fix NULL check against wrong variable

2022-03-12T13:49:13Z

We did a NULL check after "epollfdp = calloc(...)", but we checked "epollfd" instead of "epollfdp". Signed-off-by: Weiguo Li Acked-by: Davidlohr Bueso Cc: Alexander Shishkin Cc: Davidlohr Bueso Cc: Ingo Molnar Cc: Jiri Olsa Cc: Mark Rutland Cc: Namhyung Kim Cc: Peter Zijlstra Link: https://lore.kernel.org/r/tencent_B5D64530EB9C7DBB8D2C88A0C790F1489D0A@qq.com Signed-off-by: Arnaldo Carvalho de Melo

perf parse-events: Fix NULL check against wrong variable

2022-03-12T13:47:52Z

We did a null check after "tmp->symbol = strdup(...)", but we checked "list->symbol" other than "tmp->symbol". Reviewed-by: John Garry Signed-off-by: Weiguo Li Cc: Alexander Shishkin Cc: Ingo Molnar Cc: Jiri Olsa Cc: Mark Rutland Cc: Namhyung Kim Cc: Peter Zijlstra Link: https://lore.kernel.org/r/tencent_DF39269807EC9425E24787E6DB632441A405@qq.com Signed-off-by: Arnaldo Carvalho de Melo

perf script: Fix error when printing 'weight' field

2022-02-23T00:17:55Z

In SPE traces the 'weight' field can't be printed in 'perf script' because the 'dummy:u' event doesn't have the WEIGHT attribute set. Use evsel__do_check_stype(..) to check this field, as it's done with other fields such as "phys_addr". Before: $ perf record -e arm_spe_0// -- sleep 1 $ perf script -F event,ip,weight Samples for 'dummy:u' event do not have WEIGHT attribute set. Cannot print 'weight' field. After: $ perf script -F event,ip,weight l1d-access: 12 ffffaf629d4cb320 tlb-access: 12 ffffaf629d4cb320 memory: 12 ffffaf629d4cb320 Fixes: b0fde9c6e291e528 ("perf arm-spe: Add SPE total latency as PERF_SAMPLE_WEIGHT") Signed-off-by: German Gomez Acked-by: Jiri Olsa Cc: Alexander Shishkin Cc: Mark Rutland Cc: Namhyung Kim Link: https://lore.kernel.org/r/20220221171707.62960-1-german.gomez@arm.com Signed-off-by: Arnaldo Carvalho de Melo

perf data: Fix double free in perf_session__delete()

2022-02-22T20:34:16Z

When perf_data__create_dir() fails, it calls close_dir(), but perf_session__delete() also calls close_dir() and since dir.version and dir.nr were initialized by perf_data__create_dir(), a double free occurs. This patch moves the initialization of dir.version and dir.nr after successful initialization of dir.files, that prevents double freeing. This behavior is already implemented in perf_data__open_dir(). Fixes: 145520631130bd64 ("perf data: Add perf_data__(create_dir|close_dir) functions") Signed-off-by: Alexey Bayduraev Acked-by: Jiri Olsa Cc: Adrian Hunter Cc: Alexander Antonov Cc: Alexander Shishkin Cc: Alexei Budankov Cc: Andi Kleen Cc: Ingo Molnar Cc: Namhyung Kim Cc: Peter Zijlstra Link: https://lore.kernel.org/r/20220218152341.5197-2-alexey.v.bayduraev@linux.intel.com Signed-off-by: Arnaldo Carvalho de Melo

perf evlist: Fix failed to use cpu list for uncore events

2022-02-18T12:59:26Z

The 'perf record' and 'perf stat' commands have supported the option '-C/--cpus' to count or collect only on the list of CPUs provided. Commit 1d3351e631fc34d7 ("perf tools: Enable on a list of CPUs for hybrid") add it to be supported for hybrid. For hybrid support, it checks the cpu list are available on hybrid PMU. But when we test only uncore events(or events not in cpu_core and cpu_atom), there is a bug: Before: # perf stat -C0 -e uncore_clock/clockticks/ sleep 1 failed to use cpu list 0 In this case, for uncore event, its pmu_name is not cpu_core or cpu_atom, so in evlist__fix_hybrid_cpus, perf_pmu__find_hybrid_pmu should return NULL,both events_nr and unmatched_count should be 0 ,then the cpu list check function evlist__fix_hybrid_cpus return -1 and the error "failed to use cpu list 0" will happen. Bypass "events_nr=0" case then the issue is fixed. After: # perf stat -C0 -e uncore_clock/clockticks/ sleep 1 Performance counter stats for 'CPU(s) 0': 195,476,873 uncore_clock/clockticks/ 1.004518677 seconds time elapsed When testing with at least one core event and uncore events, it has no issue. # perf stat -C0 -e cpu_core/cpu-cycles/,uncore_clock/clockticks/ sleep 1 Performance counter stats for 'CPU(s) 0': 5,993,774 cpu_core/cpu-cycles/ 301,025,912 uncore_clock/clockticks/ 1.003964934 seconds time elapsed Fixes: 1d3351e631fc34d7 ("perf tools: Enable on a list of CPUs for hybrid") Reviewed-by: Kan Liang Signed-off-by: Zhengjun Xing Cc: Adrian Hunter Cc: alexander.shishkin@intel.com Cc: Andi Kleen Cc: Ian Rogers Cc: Jin Yao Cc: Jiri Olsa Cc: Peter Zijlstra Link: http://lore.kernel.org/lkml/20220218093127.1844241-1-zhengjun.xing@linux.intel.com Signed-off-by: Arnaldo Carvalho de Melo

perf test: Skip failing sigtrap test for arm+aarch64

2022-02-18T12:54:50Z

Skip the Sigtrap test for arm + arm64, same as was done for s390 in commit a840974e96fd ("perf test: Test 73 Sig_trap fails on s390"). For this, reuse BP_SIGNAL_IS_SUPPORTED - meaning that the arch can use BP to generate signals - instead of BP_ACCOUNT_IS_SUPPORTED, which is appropriate. As described by Will at [0], in the test we get stuck in a loop of handling the HW breakpoint exception and never making progress. GDB handles this by stepping over the faulting instruction, but with perf the kernel is expected to handle the step (which it doesn't for arm). Dmitry made an attempt to get this work, also mentioned in the same thread as [0], which was appreciated. But the best thing to do is skip the test for now. [0] https://lore.kernel.org/linux-perf-users/20220118124343.GC98966@leoy-ThinkPad-X240s/T/#m13b06c39d2a5100d340f009435df6f4d8ee57b5a Fixes: 5504f67944484495 ("perf test sigtrap: Add basic stress test for sigtrap handling") Signed-off-by: John Garry Tested-by: Leo Yan Acked-by: Marco Elver Cc: Dmitriy Vyukov Cc: Ian Rogers Cc: Ingo Molnar Cc: Jiri Olsa Cc: Marco Elver Cc: Mark Rutland Cc: Namhyung Kim Cc: Peter Zijlstra Cc: Thomas Richter Cc: Will Deacon Cc: linux-arm-kernel@lists.infradead.org Cc: linux@armlinux.org.uk Link: https://lore.kernel.org/r/1645176813-202756-1-git-send-email-john.garry@huawei.com Signed-off-by: Arnaldo Carvalho de Melo