From e9cc5d48d4f463fbea19dd93253e98af3b612b7c Mon Sep 17 00:00:00 2001 From: Arnaldo Carvalho de Melo Date: Wed, 12 Feb 2020 11:04:23 -0300 Subject: tools include UAPI: Sync sound/asound.h copy with the kernel sources Picking the changes from: 55b71f6c29f2a78a ("ALSA: uapi: use C90 comment style instead of C99 style") fb6723daf89083a0 ("ALSA: pcm: comment about relation between msbits hw parameter and [S|U]32 formats") b456abe63f60ad93 ("ALSA: pcm: introduce INFO_NO_REWINDS flag") 5aec579e08e4f2be ("ALSA: uapi: Fix a C++ style comment in asound.h") Which entails no changes in the tooling side as it doesn't introduce new SNDRV_PCM_IOCTL_ ioctls. To silence this perf tools build warning: Warning: Kernel ABI header at 'tools/include/uapi/sound/asound.h' differs from latest version at 'include/uapi/sound/asound.h' diff -u tools/include/uapi/sound/asound.h include/uapi/sound/asound.h Cc: Mark Brown Cc: Pierre-Louis Bossart Cc: Takashi Iwai Cc: Takashi Sakamoto Link: https://lore.kernel.org/all/YflN0j09T+6ODHIh@kernel.org Signed-off-by: Arnaldo Carvalho de Melo --- tools/include/uapi/sound/asound.h | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-) (limited to 'tools') diff --git a/tools/include/uapi/sound/asound.h b/tools/include/uapi/sound/asound.h index 5fbb79e30819..ef0cafe295b2 100644 --- a/tools/include/uapi/sound/asound.h +++ b/tools/include/uapi/sound/asound.h @@ -202,6 +202,11 @@ typedef int __bitwise snd_pcm_format_t; #define SNDRV_PCM_FORMAT_S24_BE ((__force snd_pcm_format_t) 7) /* low three bytes */ #define SNDRV_PCM_FORMAT_U24_LE ((__force snd_pcm_format_t) 8) /* low three bytes */ #define SNDRV_PCM_FORMAT_U24_BE ((__force snd_pcm_format_t) 9) /* low three bytes */ +/* + * For S32/U32 formats, 'msbits' hardware parameter is often used to deliver information about the + * available bit count in most significant bit. It's for the case of so-called 'left-justified' or + * `right-padding` sample which has less width than 32 bit. + */ #define SNDRV_PCM_FORMAT_S32_LE ((__force snd_pcm_format_t) 10) #define SNDRV_PCM_FORMAT_S32_BE ((__force snd_pcm_format_t) 11) #define SNDRV_PCM_FORMAT_U32_LE ((__force snd_pcm_format_t) 12) @@ -300,7 +305,7 @@ typedef int __bitwise snd_pcm_subformat_t; #define SNDRV_PCM_INFO_HAS_LINK_ESTIMATED_ATIME 0x04000000 /* report estimated link audio time */ #define SNDRV_PCM_INFO_HAS_LINK_SYNCHRONIZED_ATIME 0x08000000 /* report synchronized audio/system time */ #define SNDRV_PCM_INFO_EXPLICIT_SYNC 0x10000000 /* needs explicit sync of pointers and data */ - +#define SNDRV_PCM_INFO_NO_REWINDS 0x20000000 /* hardware can only support monotonic changes of appl_ptr */ #define SNDRV_PCM_INFO_DRAIN_TRIGGER 0x40000000 /* internal kernel flag - trigger in drain */ #define SNDRV_PCM_INFO_FIFO_IN_FRAMES 0x80000000 /* internal kernel flag - FIFO size is in frames */ -- cgit v1.2.3-58-ga151 From 88443d3f79b8d2b5679f2a1df1482fa024be2353 Mon Sep 17 00:00:00 2001 From: Arnaldo Carvalho de Melo Date: Fri, 21 May 2021 16:00:31 -0300 Subject: tools headers UAPI: Sync linux/perf_event.h with the kernel sources To pick the trivial change in: cb1c4aba055f928f ("perf: Add new macros for mem_hops field") Just comment source code alignment. This silences this perf build warning: Warning: Kernel ABI header at 'tools/include/uapi/linux/perf_event.h' differs from latest version at 'include/uapi/linux/perf_event.h' diff -u tools/include/uapi/linux/perf_event.h include/uapi/linux/perf_event.h Cc: Kajol Jain Cc: Michael Ellerman Link: https://lore.kernel.org/lkml/YflPKLhu2AtHmPov@kernel.org/ Signed-off-by: Arnaldo Carvalho de Melo --- tools/include/uapi/linux/perf_event.h | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) (limited to 'tools') diff --git a/tools/include/uapi/linux/perf_event.h b/tools/include/uapi/linux/perf_event.h index 4cd39aaccbe7..1b65042ab1db 100644 --- a/tools/include/uapi/linux/perf_event.h +++ b/tools/include/uapi/linux/perf_event.h @@ -1332,9 +1332,9 @@ union perf_mem_data_src { /* hop level */ #define PERF_MEM_HOPS_0 0x01 /* remote core, same node */ -#define PERF_MEM_HOPS_1 0x02 /* remote node, same socket */ -#define PERF_MEM_HOPS_2 0x03 /* remote socket, same board */ -#define PERF_MEM_HOPS_3 0x04 /* remote board */ +#define PERF_MEM_HOPS_1 0x02 /* remote node, same socket */ +#define PERF_MEM_HOPS_2 0x03 /* remote socket, same board */ +#define PERF_MEM_HOPS_3 0x04 /* remote board */ /* 5-7 available */ #define PERF_MEM_HOPS_SHIFT 43 -- cgit v1.2.3-58-ga151 From d5381cc9f123d64bee1d1a124cd98faa5fa36ca6 Mon Sep 17 00:00:00 2001 From: Arnaldo Carvalho de Melo Date: Thu, 1 Jul 2021 13:39:15 -0300 Subject: tools headers cpufeatures: Sync with the kernel sources To pick the changes from: 690a757d610e50c2 ("kvm: x86: Add CPUID support for Intel AMX") This only causes these perf files to be rebuilt: CC /tmp/build/perf/bench/mem-memcpy-x86-64-asm.o CC /tmp/build/perf/bench/mem-memset-x86-64-asm.o And addresses this perf build warning: Warning: Kernel ABI header at 'tools/arch/x86/include/asm/cpufeatures.h' differs from latest version at 'arch/x86/include/asm/cpufeatures.h' diff -u tools/arch/x86/include/asm/cpufeatures.h arch/x86/include/asm/cpufeatures.h Cc: Jing Liu Cc: Paolo Bonzini Link: https://lore.kernel.org/lkml/YflQCEO9FRLeTmlB@kernel.org/ Signed-off-by: Arnaldo Carvalho de Melo --- tools/arch/x86/include/asm/cpufeatures.h | 2 ++ 1 file changed, 2 insertions(+) (limited to 'tools') diff --git a/tools/arch/x86/include/asm/cpufeatures.h b/tools/arch/x86/include/asm/cpufeatures.h index 18de5f76f198..6db4e2932b3d 100644 --- a/tools/arch/x86/include/asm/cpufeatures.h +++ b/tools/arch/x86/include/asm/cpufeatures.h @@ -299,7 +299,9 @@ /* Intel-defined CPU features, CPUID level 0x00000007:1 (EAX), word 12 */ #define X86_FEATURE_AVX_VNNI (12*32+ 4) /* AVX VNNI instructions */ #define X86_FEATURE_AVX512_BF16 (12*32+ 5) /* AVX512 BFLOAT16 instructions */ +#define X86_FEATURE_AMX_BF16 (18*32+22) /* AMX bf16 Support */ #define X86_FEATURE_AMX_TILE (18*32+24) /* AMX tile Support */ +#define X86_FEATURE_AMX_INT8 (18*32+25) /* AMX int8 Support */ /* AMD-defined CPU features, CPUID level 0x80000008 (EBX), word 13 */ #define X86_FEATURE_CLZERO (13*32+ 0) /* CLZERO instruction */ -- cgit v1.2.3-58-ga151 From 100198322b2eb7fb38750cb0fcae5cd533907410 Mon Sep 17 00:00:00 2001 From: Arnaldo Carvalho de Melo Date: Tue, 1 Feb 2022 12:53:16 -0300 Subject: perf beauty: Make the prctl arg regexp more strict to cope with PR_SET_VMA This new PR_SET_VMA value isn't in sequence with all the other prctl arguments and instead uses a big, 0x prefixed hex number: 0x53564d41 (S V M A). This makes it harder to generate a string table as it would be rather sparse, so make the regexp more stricter to avoid catching those. A followup patch for 'perf trace' to cope with such oddities will be needed, but then its a matter for the next merge window. The next patch will update the prctl.h file to cope with this perf build warning: Warning: Kernel ABI header at 'tools/include/uapi/linux/prctl.h' differs from latest version at 'include/uapi/linux/prctl.h' diff -u tools/include/uapi/linux/prctl.h include/uapi/linux/prctl.h Here is the output of this script: $ tools/perf/trace/beauty/prctl_option.sh static const char *prctl_options[] = { [1] = "SET_PDEATHSIG", [2] = "GET_PDEATHSIG", [3] = "GET_DUMPABLE", [4] = "SET_DUMPABLE", [5] = "GET_UNALIGN", [6] = "SET_UNALIGN", [7] = "GET_KEEPCAPS", [8] = "SET_KEEPCAPS", [9] = "GET_FPEMU", [10] = "SET_FPEMU", [11] = "GET_FPEXC", [12] = "SET_FPEXC", [13] = "GET_TIMING", [14] = "SET_TIMING", [15] = "SET_NAME", [16] = "GET_NAME", [19] = "GET_ENDIAN", [20] = "SET_ENDIAN", [21] = "GET_SECCOMP", [22] = "SET_SECCOMP", [23] = "CAPBSET_READ", [24] = "CAPBSET_DROP", [25] = "GET_TSC", [26] = "SET_TSC", [27] = "GET_SECUREBITS", [28] = "SET_SECUREBITS", [29] = "SET_TIMERSLACK", [30] = "GET_TIMERSLACK", [31] = "TASK_PERF_EVENTS_DISABLE", [32] = "TASK_PERF_EVENTS_ENABLE", [33] = "MCE_KILL", [34] = "MCE_KILL_GET", [35] = "SET_MM", [36] = "SET_CHILD_SUBREAPER", [37] = "GET_CHILD_SUBREAPER", [38] = "SET_NO_NEW_PRIVS", [39] = "GET_NO_NEW_PRIVS", [40] = "GET_TID_ADDRESS", [41] = "SET_THP_DISABLE", [42] = "GET_THP_DISABLE", [43] = "MPX_ENABLE_MANAGEMENT", [44] = "MPX_DISABLE_MANAGEMENT", [45] = "SET_FP_MODE", [46] = "GET_FP_MODE", [47] = "CAP_AMBIENT", [50] = "SVE_SET_VL", [51] = "SVE_GET_VL", [52] = "GET_SPECULATION_CTRL", [53] = "SET_SPECULATION_CTRL", [54] = "PAC_RESET_KEYS", [55] = "SET_TAGGED_ADDR_CTRL", [56] = "GET_TAGGED_ADDR_CTRL", [57] = "SET_IO_FLUSHER", [58] = "GET_IO_FLUSHER", [59] = "SET_SYSCALL_USER_DISPATCH", [60] = "PAC_SET_ENABLED_KEYS", [61] = "PAC_GET_ENABLED_KEYS", [62] = "SCHED_CORE", }; static const char *prctl_set_mm_options[] = { [1] = "START_CODE", [2] = "END_CODE", [3] = "START_DATA", [4] = "END_DATA", [5] = "START_STACK", [6] = "START_BRK", [7] = "BRK", [8] = "ARG_START", [9] = "ARG_END", [10] = "ENV_START", [11] = "ENV_END", [12] = "AUXV", [13] = "EXE_FILE", [14] = "MAP", [15] = "MAP_SIZE", }; $ Cc: Adrian Hunter Cc: Colin Cross Cc: Ian Rogers Cc: Jiri Olsa Cc: Kees Cook Cc: Namhyung Kim Cc: Suren Baghdasaryan Link: https://lore.kernel.org/lkml/YflZqY0rYQ3d1bKt@kernel.org Signed-off-by: Arnaldo Carvalho de Melo --- tools/perf/trace/beauty/prctl_option.sh | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'tools') diff --git a/tools/perf/trace/beauty/prctl_option.sh b/tools/perf/trace/beauty/prctl_option.sh index 3109d7b05e11..3d278785fe57 100755 --- a/tools/perf/trace/beauty/prctl_option.sh +++ b/tools/perf/trace/beauty/prctl_option.sh @@ -4,7 +4,7 @@ [ $# -eq 1 ] && header_dir=$1 || header_dir=tools/include/uapi/linux/ printf "static const char *prctl_options[] = {\n" -regex='^#define[[:space:]]+PR_(\w+)[[:space:]]*([[:xdigit:]]+).*' +regex='^#define[[:space:]]{1}PR_(\w+)[[:space:]]*([[:xdigit:]]+)([[:space:]]*\/.*)?$' egrep $regex ${header_dir}/prctl.h | grep -v PR_SET_PTRACER | \ sed -r "s/$regex/\2 \1/g" | \ sort -n | xargs printf "\t[%s] = \"%s\",\n" -- cgit v1.2.3-58-ga151 From fc45e6588d57b65378612fce07089276141509dc Mon Sep 17 00:00:00 2001 From: Arnaldo Carvalho de Melo Date: Thu, 11 Feb 2021 12:50:52 -0300 Subject: tools headers UAPI: Sync linux/prctl.h with the kernel sources To pick the changes in: 9a10064f5625d557 ("mm: add a field to store names for private anonymous memory") That don't result in any changes in tooling: $ tools/perf/trace/beauty/prctl_option.sh > before $ cp include/uapi/linux/prctl.h tools/include/uapi/linux/prctl.h $ tools/perf/trace/beauty/prctl_option.sh > after $ diff -u before after $ This actually adds a new prctl arg, but it has to be dealt with differently, as it is not in sequence with the other arguments. Just silences this perf tools build warning: Warning: Kernel ABI header at 'tools/include/uapi/linux/prctl.h' differs from latest version at 'include/uapi/linux/prctl.h' diff -u tools/include/uapi/linux/prctl.h include/uapi/linux/prctl.h Cc: Adrian Hunter Cc: Colin Cross Cc: Ian Rogers Cc: Jiri Olsa Cc: Kees Cook Cc: Namhyung Kim Cc: Suren Baghdasaryan Signed-off-by: Arnaldo Carvalho de Melo --- tools/include/uapi/linux/prctl.h | 3 +++ 1 file changed, 3 insertions(+) (limited to 'tools') diff --git a/tools/include/uapi/linux/prctl.h b/tools/include/uapi/linux/prctl.h index bb73e9a0b24f..e998764f0262 100644 --- a/tools/include/uapi/linux/prctl.h +++ b/tools/include/uapi/linux/prctl.h @@ -272,4 +272,7 @@ struct prctl_mm_map { # define PR_SCHED_CORE_SCOPE_THREAD_GROUP 1 # define PR_SCHED_CORE_SCOPE_PROCESS_GROUP 2 +#define PR_SET_VMA 0x53564d41 +# define PR_SET_VMA_ANON_NAME 0 + #endif /* _LINUX_PRCTL_H */ -- cgit v1.2.3-58-ga151 From b7b9825fbee708e17ab7ea8b583561587c8ff7df Mon Sep 17 00:00:00 2001 From: Arnaldo Carvalho de Melo Date: Sun, 9 May 2021 09:39:02 -0300 Subject: tools headers UAPI: Sync linux/kvm.h with the kernel sources To pick the changes in: f6c6804c43fa18d3 ("kvm: Move KVM_GET_XSAVE2 IOCTL definition at the end of kvm.h") That just rebuilds perf, as these patches don't add any new KVM ioctl to be harvested for the the 'perf trace' ioctl syscall argument beautifiers. This is also by now used by tools/testing/selftests/kvm/, a simple test build succeeded. This silences this perf build warning: Warning: Kernel ABI header at 'tools/include/uapi/linux/kvm.h' differs from latest version at 'include/uapi/linux/kvm.h' diff -u tools/include/uapi/linux/kvm.h include/uapi/linux/kvm.h Cc: Janosch Frank Cc: Paolo Bonzini Link: http://lore.kernel.org/lkml/Yf+4k5Fs5Q3HdSG9@kernel.org Signed-off-by: Arnaldo Carvalho de Melo --- tools/include/uapi/linux/kvm.h | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) (limited to 'tools') diff --git a/tools/include/uapi/linux/kvm.h b/tools/include/uapi/linux/kvm.h index b46bcdb0cab1..5191b57e1562 100644 --- a/tools/include/uapi/linux/kvm.h +++ b/tools/include/uapi/linux/kvm.h @@ -1624,9 +1624,6 @@ struct kvm_enc_region { #define KVM_S390_NORMAL_RESET _IO(KVMIO, 0xc3) #define KVM_S390_CLEAR_RESET _IO(KVMIO, 0xc4) -/* Available with KVM_CAP_XSAVE2 */ -#define KVM_GET_XSAVE2 _IOR(KVMIO, 0xcf, struct kvm_xsave) - struct kvm_s390_pv_sec_parm { __u64 origin; __u64 length; @@ -2048,4 +2045,7 @@ struct kvm_stats_desc { #define KVM_GET_STATS_FD _IO(KVMIO, 0xce) +/* Available with KVM_CAP_XSAVE2 */ +#define KVM_GET_XSAVE2 _IOR(KVMIO, 0xcf, struct kvm_xsave) + #endif /* __LINUX_KVM_H */ -- cgit v1.2.3-58-ga151 From ae65b443f03fca3620cb37c5e019ddca3f89a1ce Mon Sep 17 00:00:00 2001 From: Anshuman Khandual Date: Wed, 2 Feb 2022 16:27:23 +0530 Subject: perf tools: Add missing branch_sample_type to perf_event_attr__fprintf() This updates branch sample type with missing PERF_SAMPLE_BRANCH_TYPE_SAVE. Suggested-by: James Clark Signed-off-by: Anshuman Khandual Acked-by: Jiri Olsa Cc: James Clark Cc: Mark Rutland Cc: Peter Zijlstra Cc: linux-arm-kernel@lists.infradead.org Link: http://lore.kernel.org/lkml/1643799443-15109-1-git-send-email-anshuman.khandual@arm.com Signed-off-by: Arnaldo Carvalho de Melo --- tools/perf/util/perf_event_attr_fprintf.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'tools') diff --git a/tools/perf/util/perf_event_attr_fprintf.c b/tools/perf/util/perf_event_attr_fprintf.c index 47b7531f51da..98af3fa4ea35 100644 --- a/tools/perf/util/perf_event_attr_fprintf.c +++ b/tools/perf/util/perf_event_attr_fprintf.c @@ -52,7 +52,7 @@ static void __p_branch_sample_type(char *buf, size_t size, u64 value) bit_name(ABORT_TX), bit_name(IN_TX), bit_name(NO_TX), bit_name(COND), bit_name(CALL_STACK), bit_name(IND_JUMP), bit_name(CALL), bit_name(NO_FLAGS), bit_name(NO_CYCLES), - bit_name(HW_INDEX), + bit_name(TYPE_SAVE), bit_name(HW_INDEX), { .name = NULL, } }; #undef bit_name -- cgit v1.2.3-58-ga151 From a663520fcc4bce2814032e3de6c4e2665b9555e5 Mon Sep 17 00:00:00 2001 From: Namhyung Kim Date: Tue, 1 Feb 2022 23:08:25 -0800 Subject: perf annotate: Set error stream of objdump process for TUI The stderr should be set to a pipe when using TUI. Otherwise it'd print to stdout and break TUI windows with an error message. Signed-off-by: Namhyung Kim Cc: Andi Kleen Cc: Ian Rogers Cc: Jiri Olsa Cc: Peter Zijlstra Link: http://lore.kernel.org/lkml/20220202070828.143303-2-namhyung@kernel.org Signed-off-by: Arnaldo Carvalho de Melo --- tools/perf/util/annotate.c | 1 + 1 file changed, 1 insertion(+) (limited to 'tools') diff --git a/tools/perf/util/annotate.c b/tools/perf/util/annotate.c index 01900689dc00..8190a124b99d 100644 --- a/tools/perf/util/annotate.c +++ b/tools/perf/util/annotate.c @@ -2036,6 +2036,7 @@ static int symbol__disassemble(struct symbol *sym, struct annotate_args *args) memset(&objdump_process, 0, sizeof(objdump_process)); objdump_process.argv = objdump_argv; objdump_process.out = -1; + objdump_process.err = -1; if (start_command(&objdump_process)) { pr_err("Failure starting to run %s\n", command); err = -1; -- cgit v1.2.3-58-ga151 From d792a7a94c2c3ba045247266bee2e2bced7b495a Mon Sep 17 00:00:00 2001 From: Ameer Hamza Date: Tue, 25 Jan 2022 17:11:41 +0500 Subject: perf session: Check for NULL pointer before dereference Move NULL pointer check before dereferencing the variable. Addresses-Coverity: 1497622 ("Derereference before null check") Reviewed-by: James Clark Signed-off-by: Ameer Hamza Cc: Adrian Hunter Cc: Alexander Shishkin Cc: Alexey Bayduraev Cc: German Gomez Cc: Ingo Molnar Cc: Jiri Olsa Cc: Leo Yan Cc: Mark Rutland Cc: Namhyung Kim Cc: Peter Zijlstra Cc: Riccardo Mancini Link: https://lore.kernel.org/r/20220125121141.18347-1-amhamza.mgc@gmail.com Signed-off-by: Arnaldo Carvalho de Melo --- tools/perf/util/session.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'tools') diff --git a/tools/perf/util/session.c b/tools/perf/util/session.c index 2c0d30f08e78..498b05708db5 100644 --- a/tools/perf/util/session.c +++ b/tools/perf/util/session.c @@ -1503,11 +1503,12 @@ static int machines__deliver_event(struct machines *machines, ++evlist->stats.nr_unknown_id; return 0; } - dump_sample(evsel, event, sample, perf_env__arch(machine->env)); if (machine == NULL) { ++evlist->stats.nr_unprocessable_samples; + dump_sample(evsel, event, sample, perf_env__arch(NULL)); return 0; } + dump_sample(evsel, event, sample, perf_env__arch(machine->env)); return evlist__deliver_sample(evlist, tool, event, sample, evsel, machine); case PERF_RECORD_MMAP: return tool->mmap(tool, event, sample, machine); -- cgit v1.2.3-58-ga151 From bc9c806e524429a4b98de257179af5e3fc2cb57d Mon Sep 17 00:00:00 2001 From: Leo Yan Date: Fri, 24 Dec 2021 20:40:13 +0800 Subject: perf synthetic-events: Return error if procfs isn't mounted for PID namespaces For perf recording, it retrieves process info by iterating nodes in proc fs. If we run perf in a non-root PID namespace with command: # unshare --fork --pid perf record -e cycles -a -- test_program ... in this case, unshare command creates a child PID namespace and launches perf tool in it, but the issue is the proc fs is not mounted for the non-root PID namespace, this leads to the perf tool gathering process info from its parent PID namespace. We can use below command to observe the process nodes under proc fs: # unshare --pid --fork ls /proc 1 137 1968 2128 3 342 48 62 78 crypto kcore net uptime 10 138 2 2142 30 35 49 63 8 devices keys pagetypeinfo version 11 139 20 2143 304 36 50 64 82 device-tree key-users partitions vmallocinfo 12 14 2011 22 305 37 51 65 83 diskstats kmsg self vmstat 128 140 2038 23 307 39 52 656 84 driver kpagecgroup slabinfo zoneinfo 129 15 2074 24 309 4 53 67 9 execdomains kpagecount softirqs 13 16 2094 241 31 40 54 68 asound fb kpageflags stat 130 164 2096 242 310 41 55 69 buddyinfo filesystems loadavg swaps 131 17 2098 25 317 42 56 70 bus fs locks sys 132 175 21 26 32 43 57 71 cgroups interrupts meminfo sysrq-trigger 133 179 2102 263 329 44 58 75 cmdline iomem misc sysvipc 134 1875 2103 27 330 45 59 76 config.gz ioports modules thread-self 135 19 2117 29 333 46 6 77 consoles irq mounts timer_list 136 1941 2121 298 34 47 60 773 cpuinfo kallsyms mtd tty So it shows many existed tasks, since unshared command has not mounted the proc fs for the new created PID namespace, it still accesses the proc fs of the root PID namespace. This leads to two prominent issues: - Firstly, PID values are mismatched between thread info and samples. The gathered thread info are coming from the proc fs of the root PID namespace, but samples record its PID from the child PID namespace. - The second issue is profiled program 'test_program' returns its forked PID number from the child PID namespace, perf tool wrongly uses this PID number to retrieve the process info via the proc fs of the root PID namespace. To avoid issues, we need to mount proc fs for the child PID namespace with the option '--mount-proc' when use unshare command: # unshare --fork --pid --mount-proc perf record -e cycles -a -- test_program Conversely, when the proc fs of the root PID namespace is used by child namespace, perf tool can detect the multiple PID levels and nsinfo__is_in_root_namespace() returns false, this patch reports error for this case: # unshare --fork --pid perf record -e cycles -a -- test_program Couldn't synthesize bpf events. Perf runs in non-root PID namespace but it tries to gather process info from its parent PID namespace. Please mount the proc file system properly, e.g. add the option '--mount-proc' for unshare command. Reviewed-by: James Clark Signed-off-by: Leo Yan Cc: Alexander Shishkin Cc: Alexei Starovoitov Cc: Andrii Nakryiko Cc: Daniel Borkmann Cc: Ian Rogers Cc: Ingo Molnar Cc: Jiri Olsa Cc: John Fastabend Cc: KP Singh Cc: Mark Rutland Cc: Martin KaFai Lau Cc: Namhyung Kim Cc: Peter Zijlstra Cc: Song Liu Cc: Yonghong Song Link: https://lore.kernel.org/r/20211224124014.2492751-1-leo.yan@linaro.org Signed-off-by: Arnaldo Carvalho de Melo --- tools/perf/util/synthetic-events.c | 19 +++++++++++++++++++ 1 file changed, 19 insertions(+) (limited to 'tools') diff --git a/tools/perf/util/synthetic-events.c b/tools/perf/util/synthetic-events.c index 70f095624a0b..b654de0841f8 100644 --- a/tools/perf/util/synthetic-events.c +++ b/tools/perf/util/synthetic-events.c @@ -1784,6 +1784,25 @@ int __machine__synthesize_threads(struct machine *machine, struct perf_tool *too perf_event__handler_t process, bool needs_mmap, bool data_mmap, unsigned int nr_threads_synthesize) { + /* + * When perf runs in non-root PID namespace, and the namespace's proc FS + * is not mounted, nsinfo__is_in_root_namespace() returns false. + * In this case, the proc FS is coming for the parent namespace, thus + * perf tool will wrongly gather process info from its parent PID + * namespace. + * + * To avoid the confusion that the perf tool runs in a child PID + * namespace but it synthesizes thread info from its parent PID + * namespace, returns failure with warning. + */ + if (!nsinfo__is_in_root_namespace()) { + pr_err("Perf runs in non-root PID namespace but it tries to "); + pr_err("gather process info from its parent PID namespace.\n"); + pr_err("Please mount the proc file system properly, e.g. "); + pr_err("add the option '--mount-proc' for unshare command.\n"); + return -EPERM; + } + if (target__has_task(target)) return perf_event__synthesize_thread_map(tool, threads, process, machine, needs_mmap, data_mmap); -- cgit v1.2.3-58-ga151 From a2887b9b8d1db7be971e5951e08ffe8563ea412f Mon Sep 17 00:00:00 2001 From: Masanari Iida Date: Sat, 25 Dec 2021 09:55:58 +0900 Subject: perf bpf: Fix a typo in bpf_counter_cgroup.c This patch fixes a spelling typo in error message. Signed-off-by: Masanari Iida Acked-by: Namhyung Kim Cc: Alexander Shishkin Cc: Ingo Molnar Cc: Jiri Olsa Cc: Mark Rutland Cc: Peter Zijlstra Link: https://lore.kernel.org/r/20211225005558.503935-1-standby24x7@gmail.com Signed-off-by: Arnaldo Carvalho de Melo --- tools/perf/util/bpf_counter_cgroup.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'tools') diff --git a/tools/perf/util/bpf_counter_cgroup.c b/tools/perf/util/bpf_counter_cgroup.c index 631e34a0b66f..ac60c08e8e2a 100644 --- a/tools/perf/util/bpf_counter_cgroup.c +++ b/tools/perf/util/bpf_counter_cgroup.c @@ -266,7 +266,7 @@ static int bperf_cgrp__read(struct evsel *evsel) idx = evsel->core.idx; err = bpf_map_lookup_elem(reading_map_fd, &idx, values); if (err) { - pr_err("bpf map lookup falied: idx=%u, event=%s, cgrp=%s\n", + pr_err("bpf map lookup failed: idx=%u, event=%s, cgrp=%s\n", idx, evsel__name(evsel), evsel->cgrp->name); goto out; } -- cgit v1.2.3-58-ga151 From 05b5a9d6285412d97fc61b8ec113d1d4f6b950c2 Mon Sep 17 00:00:00 2001 From: German Gomez Date: Wed, 26 Jan 2022 10:59:26 +0000 Subject: perf tools: Apply correct label to user/kernel symbols in branch mode In branch mode, the branch symbols were being displayed with incorrect cpumode labels. So fix this. For example, before: # perf record -b -a -- sleep 1 # perf report -b Overhead Command Source Shared Object Source Symbol Target Symbol 0.08% swapper [kernel.kallsyms] [k] rcu_idle_enter [k] cpuidle_enter_state ==> 0.08% cmd0 [kernel.kallsyms] [.] psi_group_change [.] psi_group_change 0.08% cmd1 [kernel.kallsyms] [k] psi_group_change [k] psi_group_change After: # perf report -b Overhead Command Source Shared Object Source Symbol Target Symbol 0.08% swapper [kernel.kallsyms] [k] rcu_idle_enter [k] cpuidle_enter_state 0.08% cmd0 [kernel.kallsyms] [k] psi_group_change [k] pei_group_change 0.08% cmd1 [kernel.kallsyms] [k] psi_group_change [k] psi_group_change Reviewed-by: James Clark Signed-off-by: German Gomez Cc: Alexander Shishkin Cc: Jiri Olsa Cc: Mark Rutland Cc: Namhyung Kim Link: https://lore.kernel.org/r/20220126105927.3411216-1-german.gomez@arm.com Signed-off-by: Arnaldo Carvalho de Melo --- tools/perf/util/machine.c | 2 ++ tools/perf/util/map_symbol.h | 1 + tools/perf/util/sort.c | 4 ++-- 3 files changed, 5 insertions(+), 2 deletions(-) (limited to 'tools') diff --git a/tools/perf/util/machine.c b/tools/perf/util/machine.c index f70ba56912d4..394550003693 100644 --- a/tools/perf/util/machine.c +++ b/tools/perf/util/machine.c @@ -2073,6 +2073,7 @@ static void ip__resolve_ams(struct thread *thread, ams->addr = ip; ams->al_addr = al.addr; + ams->al_level = al.level; ams->ms.maps = al.maps; ams->ms.sym = al.sym; ams->ms.map = al.map; @@ -2092,6 +2093,7 @@ static void ip__resolve_data(struct thread *thread, ams->addr = addr; ams->al_addr = al.addr; + ams->al_level = al.level; ams->ms.maps = al.maps; ams->ms.sym = al.sym; ams->ms.map = al.map; diff --git a/tools/perf/util/map_symbol.h b/tools/perf/util/map_symbol.h index 7d22ade082c8..e08817b0c30f 100644 --- a/tools/perf/util/map_symbol.h +++ b/tools/perf/util/map_symbol.h @@ -18,6 +18,7 @@ struct addr_map_symbol { struct map_symbol ms; u64 addr; u64 al_addr; + char al_level; u64 phys_addr; u64 data_page_size; }; diff --git a/tools/perf/util/sort.c b/tools/perf/util/sort.c index cfba8c337783..2da081ef532b 100644 --- a/tools/perf/util/sort.c +++ b/tools/perf/util/sort.c @@ -915,7 +915,7 @@ static int hist_entry__sym_from_snprintf(struct hist_entry *he, char *bf, struct addr_map_symbol *from = &he->branch_info->from; return _hist_entry__sym_snprintf(&from->ms, from->al_addr, - he->level, bf, size, width); + from->al_level, bf, size, width); } return repsep_snprintf(bf, size, "%-*.*s", width, width, "N/A"); @@ -928,7 +928,7 @@ static int hist_entry__sym_to_snprintf(struct hist_entry *he, char *bf, struct addr_map_symbol *to = &he->branch_info->to; return _hist_entry__sym_snprintf(&to->ms, to->al_addr, - he->level, bf, size, width); + to->al_level, bf, size, width); } return repsep_snprintf(bf, size, "%-*.*s", width, width, "N/A"); -- cgit v1.2.3-58-ga151 From b2b1aa73ade982c175ac926a1fd34e76ad628b94 Mon Sep 17 00:00:00 2001 From: Ian Rogers Date: Fri, 4 Feb 2022 17:09:41 -0800 Subject: perf stat: Fix display of grouped aliased events An event may have a number of uncore aliases that when added to the evlist are consecutive. If there are multiple uncore events in a group then parse_events__set_leader_for_uncore_aliase will reorder the evlist so that events on the same PMU are adjacent. The collect_all_aliases function assumes that aliases are in blocks so that only the first counter is printed and all others are marked merged. The reordering for groups breaks the assumption and so all counts are printed. This change removes the assumption from collect_all_aliases that the events are in blocks and instead processes the entire evlist. Before: ``` $ perf stat -e '{UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD_REMOTE,UNC_CHA_TOR_INSERTS.IA_MISS_DRD_REMOTE},duration_time' -a -A -- sleep 1 Performance counter stats for 'system wide': CPU0 256,866 UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD_REMOTE CPU36 494,413 UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD_REMOTE CPU0 967 UNC_CHA_TOR_INSERTS.IA_MISS_DRD_REMOTE CPU36 1,738 UNC_CHA_TOR_INSERTS.IA_MISS_DRD_REMOTE CPU0 285,161 UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD_REMOTE CPU36 429,920 UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD_REMOTE CPU0 955 UNC_CHA_TOR_INSERTS.IA_MISS_DRD_REMOTE CPU36 1,443 UNC_CHA_TOR_INSERTS.IA_MISS_DRD_REMOTE CPU0 310,753 UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD_REMOTE CPU36 416,657 UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD_REMOTE CPU0 1,231 UNC_CHA_TOR_INSERTS.IA_MISS_DRD_REMOTE CPU36 1,573 UNC_CHA_TOR_INSERTS.IA_MISS_DRD_REMOTE CPU0 416,067 UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD_REMOTE CPU36 405,966 UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD_REMOTE CPU0 1,481 UNC_CHA_TOR_INSERTS.IA_MISS_DRD_REMOTE CPU36 1,447 UNC_CHA_TOR_INSERTS.IA_MISS_DRD_REMOTE CPU0 312,911 UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD_REMOTE CPU36 408,154 UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD_REMOTE CPU0 1,086 UNC_CHA_TOR_INSERTS.IA_MISS_DRD_REMOTE CPU36 1,380 UNC_CHA_TOR_INSERTS.IA_MISS_DRD_REMOTE CPU0 333,994 UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD_REMOTE CPU36 370,349 UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD_REMOTE CPU0 1,287 UNC_CHA_TOR_INSERTS.IA_MISS_DRD_REMOTE CPU36 1,335 UNC_CHA_TOR_INSERTS.IA_MISS_DRD_REMOTE CPU0 188,107 UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD_REMOTE CPU36 302,423 UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD_REMOTE CPU0 701 UNC_CHA_TOR_INSERTS.IA_MISS_DRD_REMOTE CPU36 1,070 UNC_CHA_TOR_INSERTS.IA_MISS_DRD_REMOTE CPU0 307,221 UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD_REMOTE CPU36 383,642 UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD_REMOTE CPU0 1,036 UNC_CHA_TOR_INSERTS.IA_MISS_DRD_REMOTE CPU36 1,158 UNC_CHA_TOR_INSERTS.IA_MISS_DRD_REMOTE CPU0 318,479 UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD_REMOTE CPU36 821,545 UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD_REMOTE CPU0 1,028 UNC_CHA_TOR_INSERTS.IA_MISS_DRD_REMOTE CPU36 2,550 UNC_CHA_TOR_INSERTS.IA_MISS_DRD_REMOTE CPU0 227,618 UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD_REMOTE CPU36 372,272 UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD_REMOTE CPU0 903 UNC_CHA_TOR_INSERTS.IA_MISS_DRD_REMOTE CPU36 1,456 UNC_CHA_TOR_INSERTS.IA_MISS_DRD_REMOTE CPU0 376,783 UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD_REMOTE CPU36 419,827 UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD_REMOTE CPU0 1,406 UNC_CHA_TOR_INSERTS.IA_MISS_DRD_REMOTE CPU36 1,453 UNC_CHA_TOR_INSERTS.IA_MISS_DRD_REMOTE CPU0 286,583 UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD_REMOTE CPU36 429,956 UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD_REMOTE CPU0 999 UNC_CHA_TOR_INSERTS.IA_MISS_DRD_REMOTE CPU36 1,436 UNC_CHA_TOR_INSERTS.IA_MISS_DRD_REMOTE CPU0 313,867 UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD_REMOTE CPU36 370,159 UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD_REMOTE CPU0 1,114 UNC_CHA_TOR_INSERTS.IA_MISS_DRD_REMOTE CPU36 1,291 UNC_CHA_TOR_INSERTS.IA_MISS_DRD_REMOTE CPU0 342,083 UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD_REMOTE CPU36 409,111 UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD_REMOTE CPU0 1,399 UNC_CHA_TOR_INSERTS.IA_MISS_DRD_REMOTE CPU36 1,684 UNC_CHA_TOR_INSERTS.IA_MISS_DRD_REMOTE CPU0 365,828 UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD_REMOTE CPU36 376,037 UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD_REMOTE CPU0 1,378 UNC_CHA_TOR_INSERTS.IA_MISS_DRD_REMOTE CPU36 1,411 UNC_CHA_TOR_INSERTS.IA_MISS_DRD_REMOTE CPU0 382,456 UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD_REMOTE CPU36 621,743 UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD_REMOTE CPU0 1,232 UNC_CHA_TOR_INSERTS.IA_MISS_DRD_REMOTE CPU36 1,955 UNC_CHA_TOR_INSERTS.IA_MISS_DRD_REMOTE CPU0 342,316 UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD_REMOTE CPU36 385,067 UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD_REMOTE CPU0 1,176 UNC_CHA_TOR_INSERTS.IA_MISS_DRD_REMOTE CPU36 1,268 UNC_CHA_TOR_INSERTS.IA_MISS_DRD_REMOTE CPU0 373,588 UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD_REMOTE CPU36 386,163 UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD_REMOTE CPU0 1,394 UNC_CHA_TOR_INSERTS.IA_MISS_DRD_REMOTE CPU36 1,464 UNC_CHA_TOR_INSERTS.IA_MISS_DRD_REMOTE CPU0 381,206 UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD_REMOTE CPU36 546,891 UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD_REMOTE CPU0 1,266 UNC_CHA_TOR_INSERTS.IA_MISS_DRD_REMOTE CPU36 1,712 UNC_CHA_TOR_INSERTS.IA_MISS_DRD_REMOTE CPU0 221,176 UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD_REMOTE CPU36 392,069 UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD_REMOTE CPU0 831 UNC_CHA_TOR_INSERTS.IA_MISS_DRD_REMOTE CPU36 1,456 UNC_CHA_TOR_INSERTS.IA_MISS_DRD_REMOTE CPU0 355,401 UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD_REMOTE CPU36 705,595 UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD_REMOTE CPU0 1,235 UNC_CHA_TOR_INSERTS.IA_MISS_DRD_REMOTE CPU36 2,216 UNC_CHA_TOR_INSERTS.IA_MISS_DRD_REMOTE CPU0 371,436 UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD_REMOTE CPU36 428,103 UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD_REMOTE CPU0 1,306 UNC_CHA_TOR_INSERTS.IA_MISS_DRD_REMOTE CPU36 1,442 UNC_CHA_TOR_INSERTS.IA_MISS_DRD_REMOTE CPU0 384,352 UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD_REMOTE CPU36 504,200 UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD_REMOTE CPU0 1,468 UNC_CHA_TOR_INSERTS.IA_MISS_DRD_REMOTE CPU36 1,860 UNC_CHA_TOR_INSERTS.IA_MISS_DRD_REMOTE CPU0 228,856 UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD_REMOTE CPU36 287,976 UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD_REMOTE CPU0 832 UNC_CHA_TOR_INSERTS.IA_MISS_DRD_REMOTE CPU36 1,060 UNC_CHA_TOR_INSERTS.IA_MISS_DRD_REMOTE CPU0 215,121 UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD_REMOTE CPU36 334,162 UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD_REMOTE CPU0 681 UNC_CHA_TOR_INSERTS.IA_MISS_DRD_REMOTE CPU36 1,026 UNC_CHA_TOR_INSERTS.IA_MISS_DRD_REMOTE CPU0 296,179 UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD_REMOTE CPU36 436,083 UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD_REMOTE CPU0 1,084 UNC_CHA_TOR_INSERTS.IA_MISS_DRD_REMOTE CPU36 1,525 UNC_CHA_TOR_INSERTS.IA_MISS_DRD_REMOTE CPU0 262,296 UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD_REMOTE CPU36 416,573 UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD_REMOTE CPU0 986 UNC_CHA_TOR_INSERTS.IA_MISS_DRD_REMOTE CPU36 1,533 UNC_CHA_TOR_INSERTS.IA_MISS_DRD_REMOTE CPU0 285,852 UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD_REMOTE CPU36 359,842 UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD_REMOTE CPU0 1,073 UNC_CHA_TOR_INSERTS.IA_MISS_DRD_REMOTE CPU36 1,326 UNC_CHA_TOR_INSERTS.IA_MISS_DRD_REMOTE CPU0 303,379 UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD_REMOTE CPU36 367,222 UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD_REMOTE CPU0 1,008 UNC_CHA_TOR_INSERTS.IA_MISS_DRD_REMOTE CPU36 1,156 UNC_CHA_TOR_INSERTS.IA_MISS_DRD_REMOTE CPU0 273,487 UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD_REMOTE CPU36 425,449 UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD_REMOTE CPU0 932 UNC_CHA_TOR_INSERTS.IA_MISS_DRD_REMOTE CPU36 1,367 UNC_CHA_TOR_INSERTS.IA_MISS_DRD_REMOTE CPU0 297,596 UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD_REMOTE CPU36 414,793 UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD_REMOTE CPU0 1,140 UNC_CHA_TOR_INSERTS.IA_MISS_DRD_REMOTE CPU36 1,601 UNC_CHA_TOR_INSERTS.IA_MISS_DRD_REMOTE CPU0 342,365 UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD_REMOTE CPU36 360,422 UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD_REMOTE CPU0 1,291 UNC_CHA_TOR_INSERTS.IA_MISS_DRD_REMOTE CPU36 1,342 UNC_CHA_TOR_INSERTS.IA_MISS_DRD_REMOTE CPU0 327,196 UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD_REMOTE CPU36 580,858 UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD_REMOTE CPU0 1,122 UNC_CHA_TOR_INSERTS.IA_MISS_DRD_REMOTE CPU36 2,014 UNC_CHA_TOR_INSERTS.IA_MISS_DRD_REMOTE CPU0 296,564 UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD_REMOTE CPU36 452,817 UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD_REMOTE CPU0 1,087 UNC_CHA_TOR_INSERTS.IA_MISS_DRD_REMOTE CPU36 1,694 UNC_CHA_TOR_INSERTS.IA_MISS_DRD_REMOTE CPU0 375,002 UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD_REMOTE CPU36 389,393 UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD_REMOTE CPU0 1,478 UNC_CHA_TOR_INSERTS.IA_MISS_DRD_REMOTE CPU36 1,540 UNC_CHA_TOR_INSERTS.IA_MISS_DRD_REMOTE CPU0 365,213 UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD_REMOTE CPU36 594,685 UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD_REMOTE CPU0 1,401 UNC_CHA_TOR_INSERTS.IA_MISS_DRD_REMOTE CPU36 2,222 UNC_CHA_TOR_INSERTS.IA_MISS_DRD_REMOTE CPU0 1,000,749,060 ns duration_time 1.000749060 seconds time elapsed ``` After: ``` Performance counter stats for 'system wide': CPU0 20,547,434 UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD_REMOTE CPU36 45,202,862 UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD_REMOTE CPU0 82,001 UNC_CHA_TOR_INSERTS.IA_MISS_DRD_REMOTE CPU36 159,688 UNC_CHA_TOR_INSERTS.IA_MISS_DRD_REMOTE CPU0 1,000,464,828 ns duration_time 1.000464828 seconds time elapsed ``` Fixes: 3cdc5c2cb924acb4 ("perf parse-events: Handle uncore event aliases in small groups properly") Reviewed-by: Andi Kleen Signed-off-by: Ian Rogers Cc: Alexander Shishkin Cc: Alexandre Torgue Cc: Asaf Yaffe Cc: Caleb Biggers Cc: Ingo Molnar Cc: James Clark Cc: Jiri Olsa Cc: John Garry Cc: Kan Liang Cc: Kshipra Bopardikar Cc: Mark Rutland Cc: Maxime Coquelin Cc: Namhyung Kim Cc: Perry Taylor Cc: Peter Zijlstra Cc: Stephane Eranian Cc: Vineet Singh Cc: Zhengjun Xing Link: https://lore.kernel.org/r/20220205010941.1065469-1-irogers@google.com Signed-off-by: Arnaldo Carvalho de Melo --- tools/perf/util/stat-display.c | 19 ++++++++++--------- 1 file changed, 10 insertions(+), 9 deletions(-) (limited to 'tools') diff --git a/tools/perf/util/stat-display.c b/tools/perf/util/stat-display.c index 5db83e51ceef..9cbe351b141f 100644 --- a/tools/perf/util/stat-display.c +++ b/tools/perf/util/stat-display.c @@ -585,15 +585,16 @@ static void collect_all_aliases(struct perf_stat_config *config, struct evsel *c alias = list_prepare_entry(counter, &(evlist->core.entries), core.node); list_for_each_entry_continue (alias, &evlist->core.entries, core.node) { - if (strcmp(evsel__name(alias), evsel__name(counter)) || - alias->scale != counter->scale || - alias->cgrp != counter->cgrp || - strcmp(alias->unit, counter->unit) || - evsel__is_clock(alias) != evsel__is_clock(counter) || - !strcmp(alias->pmu_name, counter->pmu_name)) - break; - alias->merged_stat = true; - cb(config, alias, data, false); + /* Merge events with the same name, etc. but on different PMUs. */ + if (!strcmp(evsel__name(alias), evsel__name(counter)) && + alias->scale == counter->scale && + alias->cgrp == counter->cgrp && + !strcmp(alias->unit, counter->unit) && + evsel__is_clock(alias) == evsel__is_clock(counter) && + strcmp(alias->pmu_name, counter->pmu_name)) { + alias->merged_stat = true; + cb(config, alias, data, false); + } } } -- cgit v1.2.3-58-ga151 From 4f2492731ada9d702ffdfaa6ec1ff64820a1664c Mon Sep 17 00:00:00 2001 From: Arnaldo Carvalho de Melo Date: Wed, 12 Feb 2020 11:04:23 -0300 Subject: tools include UAPI: Sync sound/asound.h copy with the kernel sources Picking the changes from: 06feec6005c9d950 ("ASoC: hdmi-codec: Fix OOB memory accesses") Which entails no changes in the tooling side as it doesn't introduce new SNDRV_PCM_IOCTL_ ioctls. To silence this perf tools build warning: Warning: Kernel ABI header at 'tools/include/uapi/sound/asound.h' differs from latest version at 'include/uapi/sound/asound.h' diff -u tools/include/uapi/sound/asound.h include/uapi/sound/asound.h Cc: Dmitry Osipenko Cc: Mark Brown Cc: Takashi Iwai Link: https://lore.kernel.org/lkml/Yf+6OT+2eMrYDEeX@kernel.org Signed-off-by: Arnaldo Carvalho de Melo --- tools/include/uapi/sound/asound.h | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) (limited to 'tools') diff --git a/tools/include/uapi/sound/asound.h b/tools/include/uapi/sound/asound.h index ef0cafe295b2..2d3e5df39a59 100644 --- a/tools/include/uapi/sound/asound.h +++ b/tools/include/uapi/sound/asound.h @@ -56,8 +56,10 @@ * * ****************************************************************************/ +#define AES_IEC958_STATUS_SIZE 24 + struct snd_aes_iec958 { - unsigned char status[24]; /* AES/IEC958 channel status bits */ + unsigned char status[AES_IEC958_STATUS_SIZE]; /* AES/IEC958 channel status bits */ unsigned char subcode[147]; /* AES/IEC958 subcode bits */ unsigned char pad; /* nothing */ unsigned char dig_subframe[4]; /* AES/IEC958 subframe bits */ -- cgit v1.2.3-58-ga151 From 407eb43ae87c969d98746c3274ae5d0f977b102e Mon Sep 17 00:00:00 2001 From: Rob Herring Date: Tue, 1 Feb 2022 15:40:56 -0600 Subject: libperf: Add arm64 support to perf_mmap__read_self() Add the arm64 variants for read_perf_counter() and read_timestamp(). Unfortunately the counter number is encoded into the instruction, so the code is a bit verbose to enumerate all possible counters. Tested-by: Masayoshi Mizuma Signed-off-by: Rob Herring Acked-by: Jiri Olsa Tested-by: John Garry Signed-off-by: Arnaldo Carvalho de Melo Link: https://lore.kernel.org/r/20220201214056.702854-1-robh@kernel.org Cc: Mark Rutland Cc: Peter Zijlstra Cc: Arnaldo Carvalho de Melo Cc: Namhyung Kim Cc: Will Deacon Cc: Alexander Shishkin Cc: Ingo Molnar Cc: linux-kernel@vger.kernel.org Cc: linux-perf-users@vger.kernel.org --- tools/lib/perf/mmap.c | 98 +++++++++++++++++++++++++++++++++++++++ tools/lib/perf/tests/test-evsel.c | 5 +- 2 files changed, 102 insertions(+), 1 deletion(-) (limited to 'tools') diff --git a/tools/lib/perf/mmap.c b/tools/lib/perf/mmap.c index f7ee07cb5818..0d1634cedf44 100644 --- a/tools/lib/perf/mmap.c +++ b/tools/lib/perf/mmap.c @@ -13,6 +13,7 @@ #include #include #include +#include #include "internal.h" void perf_mmap__init(struct perf_mmap *map, struct perf_mmap *prev, @@ -294,6 +295,103 @@ static u64 read_timestamp(void) return low | ((u64)high) << 32; } +#elif defined(__aarch64__) +#define read_sysreg(r) ({ \ + u64 __val; \ + asm volatile("mrs %0, " __stringify(r) : "=r" (__val)); \ + __val; \ +}) + +static u64 read_pmccntr(void) +{ + return read_sysreg(pmccntr_el0); +} + +#define PMEVCNTR_READ(idx) \ + static u64 read_pmevcntr_##idx(void) { \ + return read_sysreg(pmevcntr##idx##_el0); \ + } + +PMEVCNTR_READ(0); +PMEVCNTR_READ(1); +PMEVCNTR_READ(2); +PMEVCNTR_READ(3); +PMEVCNTR_READ(4); +PMEVCNTR_READ(5); +PMEVCNTR_READ(6); +PMEVCNTR_READ(7); +PMEVCNTR_READ(8); +PMEVCNTR_READ(9); +PMEVCNTR_READ(10); +PMEVCNTR_READ(11); +PMEVCNTR_READ(12); +PMEVCNTR_READ(13); +PMEVCNTR_READ(14); +PMEVCNTR_READ(15); +PMEVCNTR_READ(16); +PMEVCNTR_READ(17); +PMEVCNTR_READ(18); +PMEVCNTR_READ(19); +PMEVCNTR_READ(20); +PMEVCNTR_READ(21); +PMEVCNTR_READ(22); +PMEVCNTR_READ(23); +PMEVCNTR_READ(24); +PMEVCNTR_READ(25); +PMEVCNTR_READ(26); +PMEVCNTR_READ(27); +PMEVCNTR_READ(28); +PMEVCNTR_READ(29); +PMEVCNTR_READ(30); + +/* + * Read a value direct from PMEVCNTR + */ +static u64 read_perf_counter(unsigned int counter) +{ + static u64 (* const read_f[])(void) = { + read_pmevcntr_0, + read_pmevcntr_1, + read_pmevcntr_2, + read_pmevcntr_3, + read_pmevcntr_4, + read_pmevcntr_5, + read_pmevcntr_6, + read_pmevcntr_7, + read_pmevcntr_8, + read_pmevcntr_9, + read_pmevcntr_10, + read_pmevcntr_11, + read_pmevcntr_13, + read_pmevcntr_12, + read_pmevcntr_14, + read_pmevcntr_15, + read_pmevcntr_16, + read_pmevcntr_17, + read_pmevcntr_18, + read_pmevcntr_19, + read_pmevcntr_20, + read_pmevcntr_21, + read_pmevcntr_22, + read_pmevcntr_23, + read_pmevcntr_24, + read_pmevcntr_25, + read_pmevcntr_26, + read_pmevcntr_27, + read_pmevcntr_28, + read_pmevcntr_29, + read_pmevcntr_30, + read_pmccntr + }; + + if (counter < ARRAY_SIZE(read_f)) + return (read_f[counter])(); + + return 0; +} + +static u64 read_timestamp(void) { return read_sysreg(cntvct_el0); } + #else static u64 read_perf_counter(unsigned int counter __maybe_unused) { return 0; } static u64 read_timestamp(void) { return 0; } diff --git a/tools/lib/perf/tests/test-evsel.c b/tools/lib/perf/tests/test-evsel.c index 33ae9334861a..89be89afb24d 100644 --- a/tools/lib/perf/tests/test-evsel.c +++ b/tools/lib/perf/tests/test-evsel.c @@ -130,6 +130,9 @@ static int test_stat_user_read(int event) struct perf_event_attr attr = { .type = PERF_TYPE_HARDWARE, .config = event, +#ifdef __aarch64__ + .config1 = 0x2, /* Request user access */ +#endif }; int err, i; @@ -150,7 +153,7 @@ static int test_stat_user_read(int event) pc = perf_evsel__mmap_base(evsel, 0, 0); __T("failed to get mmapped address", pc); -#if defined(__i386__) || defined(__x86_64__) +#if defined(__i386__) || defined(__x86_64__) || defined(__aarch64__) __T("userspace counter access not supported", pc->cap_user_rdpmc); __T("userspace counter access not enabled", pc->index); __T("userspace counter width not set", pc->pmc_width >= 32); -- cgit v1.2.3-58-ga151 From fceb62124d8fe1f6fb4b64e8f11c095dca8e7ea7 Mon Sep 17 00:00:00 2001 From: Changbin Du Date: Thu, 27 Jan 2022 21:20:10 +0800 Subject: perf ftrace: system_wide collection is not effective by default The ftrace.target.system_wide must be set before invoking evlist__create_maps(), otherwise it has no effect. Fixes: 53be50282269b46c ("perf ftrace: Add 'latency' subcommand") Signed-off-by: Changbin Du Acked-by: Namhyung Kim Cc: Ingo Molnar Cc: Jiri Olsa Cc: Namhyung Kim Cc: Peter Zijlstra Link: https://lore.kernel.org/r/20220127132010.4836-1-changbin.du@gmail.com Signed-off-by: Arnaldo Carvalho de Melo --- tools/perf/builtin-ftrace.c | 45 ++++++++++++++++++++++++--------------------- 1 file changed, 24 insertions(+), 21 deletions(-) (limited to 'tools') diff --git a/tools/perf/builtin-ftrace.c b/tools/perf/builtin-ftrace.c index dec24dc0e767..a8785dec5ca6 100644 --- a/tools/perf/builtin-ftrace.c +++ b/tools/perf/builtin-ftrace.c @@ -1115,6 +1115,7 @@ enum perf_ftrace_subcommand { int cmd_ftrace(int argc, const char **argv) { int ret; + int (*cmd_func)(struct perf_ftrace *) = NULL; struct perf_ftrace ftrace = { .tracer = DEFAULT_TRACER, .target = { .uid = UINT_MAX, }, @@ -1221,6 +1222,28 @@ int cmd_ftrace(int argc, const char **argv) goto out_delete_filters; } + switch (subcmd) { + case PERF_FTRACE_TRACE: + if (!argc && target__none(&ftrace.target)) + ftrace.target.system_wide = true; + cmd_func = __cmd_ftrace; + break; + case PERF_FTRACE_LATENCY: + if (list_empty(&ftrace.filters)) { + pr_err("Should provide a function to measure\n"); + parse_options_usage(ftrace_usage, options, "T", 1); + ret = -EINVAL; + goto out_delete_filters; + } + cmd_func = __cmd_latency; + break; + case PERF_FTRACE_NONE: + default: + pr_err("Invalid subcommand\n"); + ret = -EINVAL; + goto out_delete_filters; + } + ret = target__validate(&ftrace.target); if (ret) { char errbuf[512]; @@ -1248,27 +1271,7 @@ int cmd_ftrace(int argc, const char **argv) goto out_delete_evlist; } - switch (subcmd) { - case PERF_FTRACE_TRACE: - if (!argc && target__none(&ftrace.target)) - ftrace.target.system_wide = true; - ret = __cmd_ftrace(&ftrace); - break; - case PERF_FTRACE_LATENCY: - if (list_empty(&ftrace.filters)) { - pr_err("Should provide a function to measure\n"); - parse_options_usage(ftrace_usage, options, "T", 1); - ret = -EINVAL; - goto out_delete_evlist; - } - ret = __cmd_latency(&ftrace); - break; - case PERF_FTRACE_NONE: - default: - pr_err("Invalid subcommand\n"); - ret = -EINVAL; - break; - } + ret = cmd_func(&ftrace); out_delete_evlist: evlist__delete(ftrace.evlist); -- cgit v1.2.3-58-ga151