- 17 Jul, 2024 6 commits
-
-
James Clark authored
Now that symsrc_filename is always accessed through an accessor, we also need a free() function for it to avoid the following compilation error: util/unwind-libunwind-local.c:416:12: error: lvalue required as unary ‘&’ operand 416 | zfree(&dso__symsrc_filename(dso)); Fixes: 1553419c ("perf dso: Fix address sanitizer build") Signed-off-by:James Clark <james.clark@linaro.org> Reviewed-by:
Ian Rogers <irogers@google.com> Tested-by:
Leo Yan <leo.yan@arm.com> Tested-by:
Florian Fainelli <florian.fainelli@broadcom.com> Cc: Yunseong Kim <yskelg@gmail.com> Cc: Athira Rajeev <atrajeev@linux.vnet.ibm.com> Link: https://lore.kernel.org/r/20240715094715.3914813-1-james.clark@linaro.orgSigned-off-by:
Namhyung Kim <namhyung@kernel.org>
-
Guilherme Amadio authored
This allows to build against libtraceevent and libtracefs installed in non-standard locations. Signed-off-by:
Guilherme Amadio <amadio@gentoo.org> Tested-by:
Thorsten Leemhuis <linux@leemhuis.info> Tested-by:
Steven Rostedt (Google) <rostedt@goodmis.org> Cc: linux-trace-devel@vger.kernel.org Link: https://lore.kernel.org/r/20240717174739.186988-6-amadio@gentoo.orgSigned-off-by:
-
Guilherme Amadio authored
This allows to build against libtraceevent and libtracefs installed in non-standard locations. Signed-off-by:
-
Guilherme Amadio authored
This allows to build against libtraceevent and libtracefs installed in non-standard locations. Signed-off-by:
-
Guilherme Amadio authored
Other tools, in tools/verification and tools/tracing, make use of libtraceevent and libtracefs as dependencies. This allows setting up the feature check flags for them as well. Signed-off-by:
-
Guilherme Amadio authored
Signed-off-by:
-
- 12 Jul, 2024 9 commits
-
-
Howard Chu authored
This is a bug found when implementing pretty-printing for the landlock_add_rule system call, I decided to send this patch separately because this is a serious bug that should be fixed fast. I wrote a test program to do landlock_add_rule syscall in a loop, yet perf trace -e landlock_add_rule freezes, giving no output. This bug is introduced by the false understanding of the variable "key" below: ``` for (key = 0; key < trace->sctbl->syscalls.nr_entries; ++key) { struct syscall *sc = trace__syscall_info(trace, NULL, key); ... } ``` The code above seems right at the beginning, but when looking at syscalltbl.c, I found these lines: ``` for (i = 0; i <= syscalltbl_native_max_id; ++i) if (syscalltbl_native[i]) ++nr_entries; entries = tbl->syscalls.entries = malloc(sizeof(struct syscall) * nr_entries); ... for (i = 0, j = 0; i <= syscalltbl_native_max_id; ++i) { if (syscalltbl_native[i]) { entries[j].name = syscalltbl_native[i]; entries[j].id = i; ++j; } } ``` meaning the key is merely an index to traverse the syscall table, instead of the actual syscall id for this particular syscall. So if one uses key to do trace__syscall_info(trace, NULL, key), because key only goes up to trace->sctbl->syscalls.nr_entries, for example, on my X86_64 machine, this number is 373, it will end up neglecting all the rest of the syscall, in my case, everything after `rseq`, because the traversal will stop at 373, and `rseq` is the last syscall whose id is lower than 373 in tools/perf/arch/x86/include/generated/asm/syscalls_64.c: ``` ... [334] = "rseq", [424] = "pidfd_send_signal", ... ``` The reason why the key is scrambled but perf trace works well is that key is used in trace__syscall_info(trace, NULL, key) to do trace->syscalls.table[id], this makes sure that the struct syscall returned actually has an id the same value as key, making the later bpf_prog matching all correct. After fixing this bug, I can do perf trace on 38 more syscalls, and because more syscalls are visible, we get 8 more syscalls that can be augmented. before: perf $ perf trace -vv --max-events=1 |& grep Reusing Reusing "open" BPF sys_enter augmenter for "stat" Reusing "open" BPF sys_enter augmenter for "lstat" Reusing "open" BPF sys_enter augmenter for "access" Reusing "connect" BPF sys_enter augmenter for "accept" Reusing "sendto" BPF sys_enter augmenter for "recvfrom" Reusing "connect" BPF sys_enter augmenter for "bind" Reusing "connect" BPF sys_enter augmenter for "getsockname" Reusing "connect" BPF sys_enter augmenter for "getpeername" Reusing "open" BPF sys_enter augmenter for "execve" Reusing "open" BPF sys_enter augmenter for "truncate" Reusing "open" BPF sys_enter augmenter for "chdir" Reusing "open" BPF sys_enter augmenter for "mkdir" Reusing "open" BPF sys_enter augmenter for "rmdir" Reusing "open" BPF sys_enter augmenter for "creat" Reusing "open" BPF sys_enter augmenter for "link" Reusing "open" BPF sys_enter augmenter for "unlink" Reusing "open" BPF sys_enter augmenter for "symlink" Reusing "open" BPF sys_enter augmenter for "readlink" Reusing "open" BPF sys_enter augmenter for "chmod" Reusing "open" BPF sys_enter augmenter for "chown" Reusing "open" BPF sys_enter augmenter for "lchown" Reusing "open" BPF sys_enter augmenter for "mknod" Reusing "open" BPF sys_enter augmenter for "statfs" Reusing "open" BPF sys_enter augmenter for "pivot_root" Reusing "open" BPF sys_enter augmenter for "chroot" Reusing "open" BPF sys_enter augmenter for "acct" Reusing "open" BPF sys_enter augmenter for "swapon" Reusing "open" BPF sys_enter augmenter for "swapoff" Reusing "open" BPF sys_enter augmenter for "delete_module" Reusing "open" BPF sys_enter augmenter for "setxattr" Reusing "open" BPF sys_enter augmenter for "lsetxattr" Reusing "openat" BPF sys_enter augmenter for "fsetxattr" Reusing "open" BPF sys_enter augmenter for "getxattr" Reusing "open" BPF sys_enter augmenter for "lgetxattr" Reusing "openat" BPF sys_enter augmenter for "fgetxattr" Reusing "open" BPF sys_enter augmenter for "listxattr" Reusing "open" BPF sys_enter augmenter for "llistxattr" Reusing "open" BPF sys_enter augmenter for "removexattr" Reusing "open" BPF sys_enter augmenter for "lremovexattr" Reusing "fsetxattr" BPF sys_enter augmenter for "fremovexattr" Reusing "open" BPF sys_enter augmenter for "mq_open" Reusing "open" BPF sys_enter augmenter for "mq_unlink" Reusing "fsetxattr" BPF sys_enter augmenter for "add_key" Reusing "fremovexattr" BPF sys_enter augmenter for "request_key" Reusing "fremovexattr" BPF sys_enter augmenter for "inotify_add_watch" Reusing "fremovexattr" BPF sys_enter augmenter for "mkdirat" Reusing "fremovexattr" BPF sys_enter augmenter for "mknodat" Reusing "fremovexattr" BPF sys_enter augmenter for "fchownat" Reusing "fremovexattr" BPF sys_enter augmenter for "futimesat" Reusing "fremovexattr" BPF sys_enter augmenter for "newfstatat" Reusing "fremovexattr" BPF sys_enter augmenter for "unlinkat" Reusing "fremovexattr" BPF sys_enter augmenter for "linkat" Reusing "open" BPF sys_enter augmenter for "symlinkat" Reusing "fremovexattr" BPF sys_enter augmenter for "readlinkat" Reusing "fremovexattr" BPF sys_enter augmenter for "fchmodat" Reusing "fremovexattr" BPF sys_enter augmenter for "faccessat" Reusing "fremovexattr" BPF sys_enter augmenter for "utimensat" Reusing "connect" BPF sys_enter augmenter for "accept4" Reusing "fremovexattr" BPF sys_enter augmenter for "name_to_handle_at" Reusing "fremovexattr" BPF sys_enter augmenter for "renameat2" Reusing "open" BPF sys_enter augmenter for "memfd_create" Reusing "fremovexattr" BPF sys_enter augmenter for "execveat" Reusing "fremovexattr" BPF sys_enter augmenter for "statx" after perf $ perf trace -vv --max-events=1 |& grep Reusing Reusing "open" BPF sys_enter augmenter for "stat" Reusing "open" BPF sys_enter augmenter for "lstat" Reusing "open" BPF sys_enter augmenter for "access" Reusing "connect" BPF sys_enter augmenter for "accept" Reusing "sendto" BPF sys_enter augmenter for "recvfrom" Reusing "connect" BPF sys_enter augmenter for "bind" Reusing "connect" BPF sys_enter augmenter for "getsockname" Reusing "connect" BPF sys_enter augmenter for "getpeername" Reusing "open" BPF sys_enter augmenter for "execve" Reusing "open" BPF sys_enter augmenter for "truncate" Reusing "open" BPF sys_enter augmenter for "chdir" Reusing "open" BPF sys_enter augmenter for "mkdir" Reusing "open" BPF sys_enter augmenter for "rmdir" Reusing "open" BPF sys_enter augmenter for "creat" Reusing "open" BPF sys_enter augmenter for "link" Reusing "open" BPF sys_enter augmenter for "unlink" Reusing "open" BPF sys_enter augmenter for "symlink" Reusing "open" BPF sys_enter augmenter for "readlink" Reusing "open" BPF sys_enter augmenter for "chmod" Reusing "open" BPF sys_enter augmenter for "chown" Reusing "open" BPF sys_enter augmenter for "lchown" Reusing "open" BPF sys_enter augmenter for "mknod" Reusing "open" BPF sys_enter augmenter for "statfs" Reusing "open" BPF sys_enter augmenter for "pivot_root" Reusing "open" BPF sys_enter augmenter for "chroot" Reusing "open" BPF sys_enter augmenter for "acct" Reusing "open" BPF sys_enter augmenter for "swapon" Reusing "open" BPF sys_enter augmenter for "swapoff" Reusing "open" BPF sys_enter augmenter for "delete_module" Reusing "open" BPF sys_enter augmenter for "setxattr" Reusing "open" BPF sys_enter augmenter for "lsetxattr" Reusing "openat" BPF sys_enter augmenter for "fsetxattr" Reusing "open" BPF sys_enter augmenter for "getxattr" Reusing "open" BPF sys_enter augmenter for "lgetxattr" Reusing "openat" BPF sys_enter augmenter for "fgetxattr" Reusing "open" BPF sys_enter augmenter for "listxattr" Reusing "open" BPF sys_enter augmenter for "llistxattr" Reusing "open" BPF sys_enter augmenter for "removexattr" Reusing "open" BPF sys_enter augmenter for "lremovexattr" Reusing "fsetxattr" BPF sys_enter augmenter for "fremovexattr" Reusing "open" BPF sys_enter augmenter for "mq_open" Reusing "open" BPF sys_enter augmenter for "mq_unlink" Reusing "fsetxattr" BPF sys_enter augmenter for "add_key" Reusing "fremovexattr" BPF sys_enter augmenter for "request_key" Reusing "fremovexattr" BPF sys_enter augmenter for "inotify_add_watch" Reusing "fremovexattr" BPF sys_enter augmenter for "mkdirat" Reusing "fremovexattr" BPF sys_enter augmenter for "mknodat" Reusing "fremovexattr" BPF sys_enter augmenter for "fchownat" Reusing "fremovexattr" BPF sys_enter augmenter for "futimesat" Reusing "fremovexattr" BPF sys_enter augmenter for "newfstatat" Reusing "fremovexattr" BPF sys_enter augmenter for "unlinkat" Reusing "fremovexattr" BPF sys_enter augmenter for "linkat" Reusing "open" BPF sys_enter augmenter for "symlinkat" Reusing "fremovexattr" BPF sys_enter augmenter for "readlinkat" Reusing "fremovexattr" BPF sys_enter augmenter for "fchmodat" Reusing "fremovexattr" BPF sys_enter augmenter for "faccessat" Reusing "fremovexattr" BPF sys_enter augmenter for "utimensat" Reusing "connect" BPF sys_enter augmenter for "accept4" Reusing "fremovexattr" BPF sys_enter augmenter for "name_to_handle_at" Reusing "fremovexattr" BPF sys_enter augmenter for "renameat2" Reusing "open" BPF sys_enter augmenter for "memfd_create" Reusing "fremovexattr" BPF sys_enter augmenter for "execveat" Reusing "fremovexattr" BPF sys_enter augmenter for "statx" TL;DR: These are the new syscalls that can be augmented Reusing "openat" BPF sys_enter augmenter for "open_tree" Reusing "openat" BPF sys_enter augmenter for "openat2" Reusing "openat" BPF sys_enter augmenter for "mount_setattr" Reusing "openat" BPF sys_enter augmenter for "move_mount" Reusing "open" BPF sys_enter augmenter for "fsopen" Reusing "openat" BPF sys_enter augmenter for "fspick" Reusing "openat" BPF sys_enter augmenter for "faccessat2" Reusing "openat" BPF sys_enter augmenter for "fchmodat2" as for the perf trace output: before perf $ perf trace -e faccessat2 --max-events=1 [no output] after perf $ ./perf trace -e faccessat2 --max-events=1 0.000 ( 0.037 ms): waybar/958 faccessat2(dfd: 40, filename: "uevent") = 0 P.S. The reason why this bug was not found in the past five years is probably because it only happens to the newer syscalls whose id is greater, for instance, faccessat2 of id 439, which not a lot of people care about when using perf trace. [Arnaldo]: notes That and the fact that the BPF code was hidden before having to use -e, that got changed kinda recently when we switched to using BPF skels for augmenting syscalls in 'perf trace': ⬢[acme@toolbox perf-tools-next]$ git log --oneline tools/perf/util/bpf_skel/augmented_raw_syscalls.bpf.c a9f4c6c9 perf trace: Collect sys_nanosleep first argument 29d16de2 perf augmented_raw_syscalls.bpf: Move 'struct timespec64' to vmlinux.h 5069211e perf trace: Use the right bpf_probe_read(_str) variant for reading user data 33b725ce perf trace: Avoid compile error wrt redefining bool 7d964231 perf bpf augmented_raw_syscalls: Add an assert to make sure sizeof(augmented_arg->value) is a power of two. 262b54b6 perf bpf augmented_raw_syscalls: Add an assert to make sure sizeof(saddr) is a power of two. 18364804 perf bpf_skel augmented_raw_syscalls: Cap the socklen parameter using &= sizeof(saddr) cd2cece6 perf trace: Tidy comments related to BPF + syscall augmentation 5e6da6be perf trace: Migrate BPF augmentation to use a skeleton ⬢[acme@toolbox perf-tools-next]$ ⬢[acme@toolbox perf-tools-next]$ git show --oneline --pretty=reference 5e6da6be | head -1 5e6da6be (perf trace: Migrate BPF augmentation to use a skeleton, 2023-08-10) ⬢[acme@toolbox perf-tools-next]$ I.e. from August, 2023. One had as well to ask for BUILD_BPF_SKEL=1, which now is default if all it needs is available on the system. I simplified the code to not expose the 'struct syscall' outside of tools/perf/util/syscalltbl.c, instead providing a function to go from the index to the syscall id: int syscalltbl__id_at_idx(struct syscalltbl *tbl, int idx); Signed-off-by:Howard Chu <howardchu95@gmail.com> Tested-by:
Arnaldo Carvalho de Melo <acme@redhat.com> Link: https://lore.kernel.org/lkml/ZmhlAxbVcAKoPTg8@x1 Link: https://lore.kernel.org/r/20240705132059.853205-2-howardchu95@gmail.comSigned-off-by:
-
Ian Rogers authored
Various files had been missed from having accessor functions added for the sake of dso reference count checking. Add the function calls and missing dso accessor functions. Fixes: ee756ef7 ("perf dso: Add reference count checking and accessor functions") Signed-off-by:
-
Leo Yan authored
It is possible that memory events are not supported on all CPUs. Prints a warning by dumping the enabled CPU maps in this case. Signed-off-by:
-
Leo Yan authored
A platform can have more than one Arm SPE PMU. For example, a system with multiple clusters may have each cluster enabled with its own Arm SPE instance. In such case, the PMU devices will be named 'arm_spe_0', 'arm_spe_1', and so on. Currently, the tool only supports 'arm_spe_0'. This commit extends support to multiple Arm SPE PMUs by detecting the substring 'arm_spe_'. Signed-off-by:
-
Haoze Xie authored
Change the unused var in 'arch/x86/entry/syscalls/syscalltbl.sh' to '_' when reading from '$sorted_table'. This change allows the script to pass tests of ShellCheck before and after version 0.7.2 at the same time. When building in arch x86, syscalltbl.sh got a ShellCheck warning, which makes compilation error: In arch/x86/entry/syscalls/syscalltbl.sh line 27: while read nr _abi name entry _compat; do ^-^ SC2034: abi appears unused. Verify use (or export if used externally). ^----^ SC2034: compat appears unused. Verify use (or export if used externally). The script reads unused param abi and compat. It uses format '_xxx' to indicate dummy vars, which won't work properly when ShellCheck <= 0.7.2. According to SC2034, the more general way of writing is to use directly '_' to indicate discarding vars. 'entry' is also replaced by '_' because it just happens to be defined in emit function, otherwise it will lead to some misunderstandings. Link: https://www.shellcheck.net/wiki/SC2034Signed-off-by:Haoze Xie <royenheart@gmail.com> Signed-off-by:
Yuan Tan <tanyuan@tinylab.org> Link: https://lore.kernel.org/r/2143cab4cd8468c88860f4e5e382d0e6b4d89ac9.1720372178.git.royenheart@gmail.comSigned-off-by:
-
Haoze Xie authored
Modified the object of 'memset' from '&lost.lost' to '&lost' in record__read_lost_samples. This allows 'memset' to access memory properly without causing out-of-bounds problems. The problems got from builtin-record.c are: In file included from /usr/include/string.h:495, from util/parse-events.h:13, from builtin-record.c:14: In function 'memset', inlined from 'record__read_lost_samples' at builtin-record.c:1958:6, inlined from '__cmd_record.constprop' at builtin-record.c:2817:2: /usr/include/x86_64-linux-gnu/bits/string_fortified.h:71:10: error: '__builtin_memset' offset [17, 64] from the object at 'lost' is out of the bounds of referenced subobject 'lost' with type 'struct perf_record_lost_samples' at offset 0 [-Werror=array-bounds] 71|return __builtin___memset_chk (__dest,__ch,__len,__bos0 (__dest)); | ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ The error arised when performing a memset operation on the 'lost' variable, the bytes of 'sizeof(lost)' exceeds that of '&lost.lost', which are 64 and 16. Fixes: 6c1785cd ("perf record: Ensure space for lost samples") Signed-off-by: -
Madadi Vineeth Reddy authored
The --fuzzy-name option can be used if fuzzy name matching is required. For example, "taskname" can be matched to any string that contains "taskname" as its substring. Sample output for --task-name wdav --fuzzy-name ============= . *A0 . . . . - . 131040.641346 secs A0 => wdavdaemon:62509 . A0 *B0 . . . - . 131040.641378 secs B0 => wdavdaemon:62274 . *- B0 . . . - . 131040.641379 secs *C0 . B0 . . . . . 131040.641572 secs C0 => wdavdaemon:62283 C0 . B0 . *D0 . . . 131040.641572 secs D0 => wdavdaemon:62277 C0 . B0 . D0 . *E0 . 131040.641578 secs E0 => wdavdaemon:62270 *- . B0 . D0 . E0 . 131040.641581 secs Suggested-by:
Chen Yu <yu.c.chen@intel.com> Reviewed-and-tested-by:
Athira Rajeev <atrajeev@linux.vnet.ibm.com> Signed-off-by:
Madadi Vineeth Reddy <vineethr@linux.ibm.com> Link: https://lore.kernel.org/r/20240707182716.22054-4-vineethr@linux.ibm.comSigned-off-by:
-
Madadi Vineeth Reddy authored
To track the scheduling patterns of multiple tasks simultaneously, multiple task names can be specified using a comma separator without any whitespace. Sample output for --task-name perf,wdavdaemon ============= . *A0 . . . . - . 131040.641346 secs A0 => wdavdaemon:62509 . A0 *B0 . . . - . 131040.641378 secs B0 => wdavdaemon:62274 . *- B0 . . . - . 131040.641379 secs *C0 . B0 . . . . . 131040.641572 secs C0 => wdavdaemon:62283 ... . *- . . . . . . 131041.395649 secs . . . . . . . *X2 131041.403969 secs X2 => perf:70211 . . . . . . . *- 131041.404006 secs Suggested-by:
-
Madadi Vineeth Reddy authored
By default, perf sched map prints sched-in events for all the tasks which may not be required all the time as it prints lot of symbols and rows to the terminal. With --task-name option, one could specify the specific task name for which the map has to be shown. This would help in analyzing the CPU usage patterns easier for that specific task. Since multiple PID's might have the same task name, using task-name filter would be more useful for debugging. For other tasks, instead of printing the symbol, '-' is printed and the same '.' is used to represent idle. '-' is used instead of symbol for other tasks because it helps in clear visualization of task of interest and secondly the symbol itself doesn't mean anything because the sched-in of that symbol will not be printed(first sched-in contains pid and the corresponding symbol). When using the --task-name option, the sched-out time is represented by a '*-'. Since not all task sched-in events are printed, the sched-out time of the relevant task might be lost. This representation ensures that the sched-out time of the interested task is not overlooked. 6.10.0-rc1 ========== *A0 131040.639793 secs A0 => migration/0:19 *. 131040.639801 secs . => swapper:0 . *B0 131040.639830 secs B0 => migration/1:24 . *. 131040.639836 secs . . *C0 131040.640108 secs C0 => migration/2:30 . . *. 131040.640163 secs . . . *D0 131040.640386 secs D0 => migration/3:36 . . . *. 131040.640395 secs 6.10.0-rc1 + patch (--task-name wdavdaemon) ============= . *A0 . . . . - . 131040.641346 secs A0 => wdavdaemon:62509 . A0 *B0 . . . - . 131040.641378 secs B0 => wdavdaemon:62274 - *- B0 . . . - . 131040.641379 secs *C0 . B0 . . . . . 131040.641572 secs C0 => wdavdaemon:62283 C0 . B0 . *D0 . . . 131040.641572 secs D0 => wdavdaemon:62277 C0 . B0 . D0 . *E0 . 131040.641578 secs E0 => wdavdaemon:62270 *- . B0 . D0 . E0 . 131040.641581 secs . . B0 . D0 . *- . 131040.641583 secs Reviewed-and-tested-by:
-
- 11 Jul, 2024 1 commit
-
-
Guilherme Amadio authored
This avoids reported warnings when the packages are not installed. [namhyung]: Removed the dummy assignment and unnecessary ifeq checks. Fixes: 0f0e1f44 ("perf build: Use pkg-config for feature check for libtrace{event,fs}") Signed-off-by:
-
- 04 Jul, 2024 1 commit
-
-
Nicolas Schier authored
Explicitly reset 'subdir' variable when descending to tools/perf/Documentation. Similar to commit f89fb557 ("perf build: Don't propagate subdir to submakes for install_headers", 2023-01-02), calling the 'tools/perf_install' target via top-levels Makefile results in repeated subdir components when attempting to call the perf documentation installation rules: $ make tools/perf_install NO_LIBTRACEEVENT=1 JOBS=1 [...] /bin/sh: 1: cd: can't cd to /data/linux/kbuild/tools/perf/tools/perf/ ../../scripts/Makefile.include:17: *** output directory "/data/linux/kbuild/tools/perf/tools/perf/" does not exist. Stop. make[5]: *** [Makefile.perf:1096: try-install-man] Error 2 make[4]: *** [Makefile.perf:264: sub-make] Error 2 make[3]: *** [Makefile:113: install] Error 2 make[2]: *** [Makefile:131: perf_install] Error 2 Resetting 'subdir' fixes the call from top-level Makefile. Reported-by:
Uwe Kleine-König <u.kleine-koenig@pengutronix.de> Signed-off-by:
Nicolas Schier <n.schier@avm.de> Acked-by:
Uwe Kleine-König <u.kleine-koenig@baylibre.com> Tested-by:
-
- 03 Jul, 2024 4 commits
-
-
Xu Yang authored
Add JSON metrics for i.MX95 DDR Performance Monitor. Reviewed-by:
John Garry <john.g.garry@oracle.com> Reviewed-by:
Frank Li <Frank.Li@nxp.com> Signed-off-by:
Xu Yang <xu.yang_2@nxp.com> Cc: festevam@gmail.com Cc: conor+dt@kernel.org Cc: robh+dt@kernel.org Cc: shawnguo@kernel.org Cc: will@kernel.org Cc: krzysztof.kozlowski+dt@linaro.org Cc: mike.leach@linaro.org Cc: linux-arm-kernel@lists.infradead.org Cc: imx@lists.linux.dev Cc: kernel@pengutronix.de Cc: s.hauer@pengutronix.de Cc: devicetree@vger.kernel.org Link: https://lore.kernel.org/r/20240529080358.703784-8-xu.yang_2@nxp.comSigned-off-by:
-
Xu Yang authored
Add JSON metrics for i.MX93 DDR Performance Monitor. Reviewed-by:
-
Ian Rogers authored
dsos__add would add at the end of the dso array possibly requiring a later find to re-sort the array. Patterns of find then add were becoming O(n*log n) due to the sorts. Change the add routine to be O(n) rather than O(1) but to maintain the sorted-ness of the dsos array so that later finds don't need the O(n*log n) sort. Fixes: 3f4ac23a ("perf dsos: Switch backing storage to array from rbtree/list") Reported-by:
-
Ian Rogers authored
The array is sorted, so just move the elements and insert in order. Fixes: 13ca6287 ("perf comm: Add reference count checking to 'struct comm_str'") Reported-by:
Matt Fleming <matt@readmodwrite.com> Signed-off-by:
-
- 02 Jul, 2024 4 commits
-
-
Abhishek Dubey authored
For printing dump_trace, just use existing stats_print() function. Signed-off-by:
Abhishek Dubey <adubey@linux.ibm.com> Link: https://lore.kernel.org/r/20240628183224.452055-1-adubey@linux.ibm.comSigned-off-by:
-
Adrian Hunter authored
In the past, the exclude_guest setting has had no effect on Intel PT tracing, but that may not be the case in the future. Set the flag correctly based upon whether KVM is using Intel PT "Host/Guest" mode, which is determined by the kvm_intel module parameter pt_mode: pt_mode=0 System-wide mode : host and guest output to host buffer pt_mode=1 Host/Guest mode : host/guest output to host/guest buffers respectively Fixes: 6e86bfdc ("perf intel-pt: Support decoding of guest kernel") Signed-off-by:Adrian Hunter <adrian.hunter@intel.com> Link: https://lore.kernel.org/r/20240625104532.11990-3-adrian.hunter@intel.comSigned-off-by:
-
Adrian Hunter authored
aux_watermark is a u32. For a 64-bit size, cap the aux_watermark calculation at UINT_MAX instead of truncating it to 32-bits. Fixes: 874fc35c ("perf intel-pt: Use aux_watermark") Signed-off-by:
-
Namhyung Kim authored
Merge fixes and updates in v6.10 into perf-tools-next to resolve changes in synthesizing the LOST_SAMPLES records and build fixes. Signed-off-by:
-
- 28 Jun, 2024 7 commits
-
-
Madadi Vineeth Reddy authored
Currently, the -r/--repeat option accepts values from 0 and complains for -1. The help section specifies: -r, --repeat <n> repeat the workload replay N times (-1: infinite) The -r -1 option raises an error because replay_repeat is defined as an unsigned int. In the current implementation, the workload is repeated n times when -r <n> is used, except when n is 0. When -r is set to 0, the workload is also repeated once. This happens because when -r=0, the run_one_test function is not called. (Note that mutex unlocking, which is essential for child threads spawned to emulate the workload, happens in run_one_test.) However, mutex unlocking is still performed in the destroy_tasks function. Thus, -r=0 results in the workload running once coincidentally. To clarify and maintain the existing logic for -r >= 1 (which runs the workload the specified number of times) and to fix the issue with infinite runs, make -r=0 perform an infinite run. Reviewed-by:
James Clark <james.clark@arm.com> Signed-off-by:
-
Yang Li authored
./tools/perf/util/pmu.c:1776:49-50: Unneeded semicolon Reported-by:
Abaci Robot <abaci@linux.alibaba.com> Closes: https://bugzilla.openanolis.cn/show_bug.cgi?id=9443Signed-off-by:
Yang Li <yang.lee@linux.alibaba.com> Reviewed-by:
-
Namhyung Kim authored
It didn't use the passed field separator (using -x option) when it prints the metric headers and always put "," between the fields. Before: $ sudo ./perf stat -a -x : --per-core -M tma_core_bound --metric-only true core,cpus,% tma_core_bound: <<<--- here: "core,cpus," but ":" expected S0-D0-C0:2:10.5: S0-D0-C1:2:14.8: S0-D0-C2:2:9.9: S0-D0-C3:2:13.2: After: $ sudo ./perf stat -a -x : --per-core -M tma_core_bound --metric-only true core:cpus:% tma_core_bound: S0-D0-C0:2:10.5: S0-D0-C1:2:15.0: S0-D0-C2:2:16.5: S0-D0-C3:2:12.5: Tested-by:
-
Namhyung Kim authored
The new --per-cluster option was added recently but it forgot to update the aggr_header fields which are used for --metric-only option. And it resulted in a segfault due to NULL string in fputs(). Fixes: cbc917a1 ("perf stat: Support per-cluster aggregation") Reviewed-by:
Yicong Yang <yangyicong@hisilicon.com> Tested-by:
-
James Clark authored
Arm PMUs have a suffix, either a single decimal (armv8_pmuv3_0) or 3 hex digits which (armv8_cortex_a53) which Perf assumes are both strippable suffixes for the purposes of deduplication. S390 "cpum_cf" is a similarly suffixed core PMU but is only two characters so is not treated as strippable because the rules are a minimum of 3 hex characters or 1 decimal character. There are two paths involved in listing PMU events: * HW/cache event printing assumes core PMUs don't have suffixes so doesn't try to strip. * Sysfs PMU events share the printing function with uncore PMUs which strips. This results in slightly inconsistent Perf list behavior if a core PMU has a suffix: # perf list ... armv8_pmuv3_0/branch-load-misses/ armv8_pmuv3/l3d_cache_wb/ [Kernel PMU event] ... Fix it by partially reverting back to the old list behavior where stripping was only done for uncore PMUs. For example commit 8d9f5146 ("perf pmus: Sort pmus by name then suffix") mentions that only PMUs starting 'uncore_' are considered to have a potential suffix. This change doesn't go back that far, but does only strip PMUs that are !is_core. This keeps the desirable behavior where the many possibly duplicated uncore PMUs aren't repeated, but it doesn't break listing for core PMUs. Searching for a PMU continues to use the new stripped comparison functions, meaning that it's still possible to request an event by specifying the common part of a PMU name, or even open events on multiple similarly named PMUs. For example: # perf stat -e armv8_cortex/inst_retired/ 5777173628 armv8_cortex_a53/inst_retired/ (99.93%) 7469626951 armv8_cortex_a57/inst_retired/ (49.88%) Fixes: 3241d46f ("perf pmus: Sort/merge/aggregate PMUs like mrvl_ddr_pmu") Suggested-by:
-
James Clark authored
Commit b2b9d3a3 ("perf pmu: Support wildcards on pmu name in dynamic pmu events") gives the following example for wildcarding a subset of PMUs: E.g., in a system with the following dynamic pmus: mypmu_0 mypmu_1 mypmu_2 mypmu_4 perf stat -e mypmu_[01]/<config>/ Since commit f91fa2ae ("perf pmu: Refactor perf_pmu__match()"), only "*" has been supported, removing the ability to subset PMUs, even though parse-events.l still supports ? and [] characters. Fix it by using fnmatch() when any glob character is detected and add a test which covers that and other scenarios of perf_pmu__match_ignoring_suffix(). Fixes: f91fa2ae ("perf pmu: Refactor perf_pmu__match()") Signed-off-by:
-
Namhyung Kim authored
It's possible to save pipe output of perf record into a file. $ perf record -o- ... > pipe.data And you can use the data same as the normal perf data. $ perf report -i pipe.data In that case, perf tools will treat the input as a pipe, but it can get the total size of the input. This means it can show the progress bar unlike the normal pipe input (which doesn't know the total size in advance). While at it, fix the string in __perf_session__process_dir_events(). Reviewed-by:
-
- 26 Jun, 2024 8 commits
-
-
Veronika Molnarova authored
The test has been failing for some time when two separate runs of perf benchmarks are recorded for cycles events and their counts are compared, while once the recording was done with option --bpf-counters and once without it. It is expected that the count of the samples should be within a certain range, firstly the difference was set to be within 10%, which was then later raised to 20%. However, the test case keeps failing on certain architectures as recording the provided benchmark can produce completely different counts based on the current load of the system. Sampling two separate runs on intel-eaglestream-spr-13 of "perf stat --no-big-num -e cycles -- perf bench sched messaging -g 1 -l 100 -t": Performance counter stats for 'perf bench sched messaging -g 1 -l 100 -t': 396782898 cycles 0.010051983 seconds time elapsed 0.008664000 seconds user 0.097058000 seconds sys Performance counter stats for 'perf bench sched messaging -g 1 -l 100 -t': 1431133032 cycles 0.021803714 seconds time elapsed 0.023377000 seconds user 0.349918000 seconds sys , which is ranging from 400mil to 1400mil samples. Instead of recording the cycles use instructions event, which provides more stable values. At the same time change the tested workload to one of the provided testing workloads by perf that is not based on a scheduler, which can provide another dependency on the current load. Sampling instructions event with the new workload provide much more stable results on intel-eaglestream-spr-13 of "perf stat --no-big-num -e instructions -- perf test -w brstack": Performance counter stats for 'perf test -w brstack': 64584494 instructions 0.009173945 seconds time elapsed 0.007262000 seconds user 0.002071000 seconds sys Performance counter stats for 'perf test -w brstack': 64672669 instructions 0.008888135 seconds time elapsed 0.005018000 seconds user 0.004018000 seconds sys Signed-off-by:Veronika Molnarova <vmolnaro@redhat.com> Acked-by:
-
Ian Rogers authored
The python build now depends on libraries and doesn't use python-ext-sources except for the util/python.c dependency. Switch to just directly depending on that file and util/setup.py. This allows the removal of python-ext-sources. Signed-off-by:
-
Ian Rogers authored
setup.py was building most perf sources causing setup.py to mimic the Makefile logic as well as flex/bison code to be stubbed out, due to complexity building. By using libraries fewer functions are stubbed out, the build is faster and the Makefile logic is reused which should simplify updating. The libraries are passed through LDFLAGS to avoid complexity in python. Force the -fPIC flag for libbpf.a to ensure it is suitable for linking into the perf python module. Signed-off-by:
-
Ian Rogers authored
Make the util directory into its own library. This is done to avoid compiling code twice, once for the perf tool and once for the perf python module. For convenience: arch/common.c scripts/perl/Perf-Trace-Util/Context.c scripts/python/Perf-Trace-Util/Context.c are made part of this library. Signed-off-by:
-
Ian Rogers authored
Make the benchmark code into a library so it may be linked against things like the python module to avoid compiling code twice. Signed-off-by:
-
Ian Rogers authored
Make the tests code its own library. This is done to avoid compiling code twice, once for the perf tool and once for the perf python module. Signed-off-by:
-
Ian Rogers authored
Make pmu-events into a library so it may be linked against things like the python module and not built from source. Signed-off-by:
-
Ian Rogers authored
Make the ui code its own library. This is done to avoid compiling code twice, once for the perf tool and once for the perf python module. Signed-off-by:
-
