kernel/tools/perf/Makefile.perf, branch linux-6.16.y

perf tests: Fix 'perf report' tests installation

2025-05-09T17:10:08Z

There was a copy-paste mistake in the installation commands. Also, we need to install stderr-whitelist.txt file, which contains allowed messages that are printed on stderr and should not cause test fail. Fixes: 097fe67df1aa9cc7 ("perf testsuite: Install perf-report tests in the 'make install-tests -C tools/perf' target") Signed-off-by: Michael Petlan Cc: Ian Rogers Cc: Namhyung Kim Link: https://lore.kernel.org/r/20250113182605.130719-6-vmolnaro@redhat.com Signed-off-by: Veronika Molnarova Signed-off-by: Arnaldo Carvalho de Melo

perf trace: Implement syscall summary in BPF

2025-04-28T19:53:11Z

When -s/--summary option is used, it doesn't need (augmented) arguments of syscalls. Let's skip the augmentation and load another small BPF program to collect the statistics in the kernel instead of copying the data to the ring-buffer to calculate the stats in userspace. This will be much more light-weight than the existing approach and remove any lost events. Let's add a new option --bpf-summary to control this behavior. I cannot make it default because there's no way to get e_machine in the BPF which is needed for detecting different ABIs like 32-bit compat mode. No functional changes intended except for no more LOST events. :) $ sudo ./perf trace -as --summary-mode=total --bpf-summary sleep 1 Summary of events: total, 6194 events syscall calls errors total min avg max stddev (msec) (msec) (msec) (msec) (%) --------------- -------- ------ -------- --------- --------- --------- ------ epoll_wait 561 0 4530.843 0.000 8.076 520.941 18.75% futex 693 45 4317.231 0.000 6.230 500.077 21.98% poll 300 0 1040.109 0.000 3.467 120.928 17.02% clock_nanosleep 1 0 1000.172 1000.172 1000.172 1000.172 0.00% ppoll 360 0 872.386 0.001 2.423 253.275 41.91% epoll_pwait 14 0 384.349 0.001 27.453 380.002 98.79% pselect6 14 0 108.130 7.198 7.724 8.206 0.85% nanosleep 39 0 43.378 0.069 1.112 10.084 44.23% ... Reviewed-by: Howard Chu Signed-off-by: Namhyung Kim Tested-by: Arnaldo Carvalho de Melo Cc: Adrian Hunter Cc: Howard Chu Cc: Ian Rogers Cc: Ingo Molnar Cc: Jiri Olsa Cc: Kan Liang Cc: Peter Zijlstra Cc: Song Liu Link: https://lore.kernel.org/r/20250326044001.3503432-1-namhyung@kernel.org [ Added fixup sent from Namhyung in response to my report to make it also dependent on CONFIG_TRACE ] Signed-off-by: Arnaldo Carvalho de Melo

Merge tag 'perf-tools-for-v6.15-2025-03-27' of git://git.kernel.org/pub/scm/linux/kernel/git/perf/perf-tools

2025-03-31T15:52:33Z

Pull perf tools updates from Namhyung Kim: "perf record: - Introduce latency profiling using scheduler information. The latency profiling is to show impacts on wall-time rather than cpu-time. By tracking context switches, it can weight samples and find which part of the code contributed more to the execution latency. The value (period) of the sample is weighted by dividing it by the number of parallel execution at the moment. The parallelism is tracked in perf report with sched-switch records. This will reduce the portion that are run in parallel and in turn increase the portion of serial executions. For now, it's limited to profile processes, IOW system-wide profiling is not supported. You can add --latency option to enable this. $ perf record --latency -- make -C tools/perf I've run the above command for perf build which adds -j option to make with the number of CPUs in the system internally. Normally it'd show something like below: $ perf report -F overhead,comm ... # # Overhead Command # ........ ............... # 78.97% cc1 6.54% python3 4.21% shellcheck 3.28% ld 1.80% as 1.37% cc1plus 0.80% sh 0.62% clang 0.56% gcc 0.44% perl 0.39% make ... The cc1 takes around 80% of the overhead as it's the actual compiler. However it runs in parallel so its contribution to latency may be less than that. Now, perf report will show both overhead and latency (if --latency was given at record time) like below: $ perf report -s comm ... # # Overhead Latency Command # ........ ........ ............... # 78.97% 48.66% cc1 6.54% 25.68% python3 4.21% 0.39% shellcheck 3.28% 13.70% ld 1.80% 2.56% as 1.37% 3.08% cc1plus 0.80% 0.98% sh 0.62% 0.61% clang 0.56% 0.33% gcc 0.44% 1.71% perl 0.39% 0.83% make ... You can see latency of cc1 goes down to around 50% and python3 and ld contribute a lot more than their overhead. You can use --latency option in perf report to get the same result but ordered by latency. $ perf report --latency -s comm perf report: - As a side effect of the latency profiling work, it adds a new output field 'latency' and a sort key 'parallelism'. The below is a result from my system with 64 CPUs. The build was well-parallelized but contained some serial portions. $ perf report -s parallelism ... # # Overhead Latency Parallelism # ........ ........ ........... # 16.95% 1.54% 62 13.38% 1.24% 61 12.50% 70.47% 1 11.81% 1.06% 63 7.59% 0.71% 60 4.33% 12.20% 2 3.41% 0.33% 59 2.05% 0.18% 64 1.75% 1.09% 9 1.64% 1.85% 5 ... - Support Feodra mini-debuginfo which is a LZMA compressed symbol table inside ".gnu_debugdata" ELF section. perf annotate: - Add --code-with-type option to enable data-type profiling with the usual annotate output. Instead of focusing on data structure, it shows code annotation together with data type it accesses in case the instruction refers to a memory location (and it was able to resolve the target data type). Currently it only works with --stdio. $ perf annotate --stdio --code-with-type ... Percent | Source code & Disassembly of vmlinux for cpu/mem-loads,ldlat=30/pp (18 samples, percent: local period) ---------------------------------------------------------------------------------------------------------------------- : 0 0xffffffff81050610 <__fdget>: 0.00 : ffffffff81050610: callq 0xffffffff81c01b80 <__fentry__> # data-type: (stack operation) 0.00 : ffffffff81050615: pushq %rbp # data-type: (stack operation) 0.00 : ffffffff81050616: movq %rsp, %rbp 0.00 : ffffffff81050619: pushq %r15 # data-type: (stack operation) 0.00 : ffffffff8105061b: pushq %r14 # data-type: (stack operation) 0.00 : ffffffff8105061d: pushq %rbx # data-type: (stack operation) 0.00 : ffffffff8105061e: subq $0x10, %rsp 0.00 : ffffffff81050622: movl %edi, %ebx 0.00 : ffffffff81050624: movq %gs:0x7efc4814(%rip), %rax # 0x14e40 # data-type: struct task_struct* +0 0.00 : ffffffff8105062c: movq 0x8d0(%rax), %r14 # data-type: struct task_struct +0x8d0 (files) 0.00 : ffffffff81050633: movl (%r14), %eax # data-type: struct files_struct +0 (count.counter) 0.00 : ffffffff81050636: cmpl $0x1, %eax 0.00 : ffffffff81050639: je 0xffffffff810506a9 <__fdget+0x99> 0.00 : ffffffff8105063b: movq 0x20(%r14), %rcx # data-type: struct files_struct +0x20 (fdt) 0.00 : ffffffff8105063f: movl (%rcx), %eax # data-type: struct fdtable +0 (max_fds) 0.00 : ffffffff81050641: cmpl %ebx, %eax 0.00 : ffffffff81050643: jbe 0xffffffff810506ef <__fdget+0xdf> 0.00 : ffffffff81050649: movl %ebx, %r15d 5.56 : ffffffff8105064c: movq 0x8(%rcx), %rdx # data-type: struct fdtable +0x8 (fd) ... The "# data-type:" part was added with this change. The first few entries are not very interesting. But later you can it accesses a couple of fields in the task_struct, files_struct and fdtable. perf trace: - Support syscall tracing for different ABI. For example it can trace system calls for 32-bit applications on 64-bit kernel transparently. - Add --summary-mode=total option to show global syscall summary. The default is 'thread' to show per-thread syscall summary. Python support: - Add more interfaces to 'perf' module to parse events, and config, enable or disable the event list properly so that it can implement basic functionalities purely in Python. There is an example code for these new interfaces in python/tracepoint.py. - Add mypy and pylint support to enable build time checking. Fix some code based on the findings from these tools. Internals: - Introduce io_dir__readdir() API to make directory traveral (usually for proc or sysfs) efficient with less memory footprint. JSON vendor events: - Add events and metrics for ARM Neoverse N3 and V3 - Update events and metrics on various Intel CPUs - Add/update events for a number of SiFive processors" * tag 'perf-tools-for-v6.15-2025-03-27' of git://git.kernel.org/pub/scm/linux/kernel/git/perf/perf-tools: (229 commits) perf bpf-filter: Fix a parsing error with comma perf report: Fix a memory leak for perf_env on AMD perf trace: Fix wrong size to bpf_map__update_elem call perf tools: annotate asm_pure_loop.S perf python: Fix setup.py mypy errors perf test: Address attr.py mypy error perf build: Add pylint build tests perf build: Add mypy build tests perf build: Rename TEST_LOGS to SHELL_TEST_LOGS tools/build: Don't pass test log files to linker perf bench sched pipe: fix enforced blocking reads in worker_thread perf tools: Fix is_compat_mode build break in ppc64 perf build: filter all combinations of -flto for libperl perf vendor events arm64 AmpereOneX: Fix frontend_bound calculation perf vendor events arm64: AmpereOne/AmpereOneX: Mark LD_RETIRED impacted by errata perf trace: Fix evlist memory leak perf trace: Fix BTF memory leak perf trace: Make syscall table stable perf syscalltbl: Mask off ABI type for MIPS system calls perf build: Remove Makefile.syscalls ...

perf build: Add pylint build tests

2025-03-24T16:38:20Z

If PYLINT=1 is passed to the build then run pylint over python code in perf. Unlike shellcheck this isn't default on as there are currently too many errors. An example of an error: ``` ************* Module setup util/setup.py:19:0: C0301: Line too long (127/100) (line-too-long) util/setup.py:20:0: C0301: Line too long (138/100) (line-too-long) util/setup.py:63:0: C0301: Line too long (106/100) (line-too-long) util/setup.py:1:0: C0114: Missing module docstring (missing-module-docstring) util/setup.py:24:4: W0622: Redefining built-in 'vars' (redefined-builtin) util/setup.py:11:4: C0103: Constant name "cc_options" doesn't conform to UPPER_CASE naming style (invalid-name) util/setup.py:13:4: C0103: Constant name "cc_options" doesn't conform to UPPER_CASE naming style (invalid-name) util/setup.py:15:34: R1732: Consider using 'with' for resource-allocating operations (consider-using-with) util/setup.py:18:0: C0116: Missing function or method docstring (missing-function-docstring) util/setup.py:19:16: R1732: Consider using 'with' for resource-allocating operations (consider-using-with) util/setup.py:44:0: C0413: Import "from setuptools import setup, Extension" should be placed at the top of the module (wrong-import-position) util/setup.py:46:0: C0413: Import "from setuptools.command.build_ext import build_ext as _build_ext" should be placed at the top of the module (wrong-import-position) util/setup.py:47:0: C0413: Import "from setuptools.command.install_lib import install_lib as _install_lib" should be placed at the top of the module (wrong-import-position) util/setup.py:49:0: C0115: Missing class docstring (missing-class-docstring) util/setup.py:49:0: C0103: Class name "build_ext" doesn't conform to PascalCase naming style (invalid-name) util/setup.py:52:8: W0201: Attribute 'build_lib' defined outside __init__ (attribute-defined-outside-init) util/setup.py:53:8: W0201: Attribute 'build_temp' defined outside __init__ (attribute-defined-outside-init) util/setup.py:55:0: C0115: Missing class docstring (missing-class-docstring) util/setup.py:55:0: C0103: Class name "install_lib" doesn't conform to PascalCase naming style (invalid-name) util/setup.py:58:8: W0201: Attribute 'build_dir' defined outside __init__ (attribute-defined-outside-init) *----------------------------------------------------------------- Your code has been rated at 6.67/10 (previous run: 6.51/10, +0.16) make[4]: *** [util/Build:442: util/setup.py.pylint_log] Error 1 ``` Reviewed-by: James Clark Signed-off-by: Ian Rogers Link: https://lore.kernel.org/r/20250311213628.569562-5-irogers@google.com Signed-off-by: Namhyung Kim

perf build: Add mypy build tests

2025-03-24T16:38:20Z

If MYPY=1 is passed to the build then run mypy over python code in perf. Unlike shellcheck this isn't default on as there are currently too many errors. An example of an error: ``` util/setup.py:8: error: Item "None" of "str | None" has no attribute "split" [union-attr] util/setup.py:15: error: Item "None" of "IO[bytes] | None" has no attribute "readline" [union-attr] util/setup.py:15: error: List item 0 has incompatible type "str | None"; expected "str | bytes | PathLike[str] | PathLike[bytes]" [list-item] util/setup.py:16: error: Unsupported left operand type for + ("None") [operator] util/setup.py:16: note: Left operand is of type "str | None" util/setup.py:74: error: Unsupported left operand type for + ("None") [operator] util/setup.py:74: note: Left operand is of type "str | None" Found 5 errors in 1 file (checked 1 source file) make[4]: *** [util/Build:430: util/setup.py.mypy_log] Error 1 ``` Reviewed-by: James Clark Signed-off-by: Ian Rogers Link: https://lore.kernel.org/r/20250311213628.569562-4-irogers@google.com Signed-off-by: Namhyung Kim

perf build: Rename TEST_LOGS to SHELL_TEST_LOGS

2025-03-24T16:38:20Z

Rename TEST_LOGS to SHELL_TEST_LOGS as later changes will add more kinds of test logs. Minor comment tweak in Makefile.perf as more than just test shell tests are checked. Reviewed-by: James Clark Signed-off-by: Ian Rogers Link: https://lore.kernel.org/r/20250311213628.569562-3-irogers@google.com Signed-off-by: Namhyung Kim

perf build: Remove Makefile.syscalls

2025-03-21T05:58:20Z

Now a single beauty file is generated and used by all architectures, remove the per-architecture Makefiles, Kbuild files and previous generator script. Note: there was conversation with Charlie Jenkins and they'd written an alternate approach to support multiple architectures: https://lore.kernel.org/all/20250114-perf_syscall_arch_runtime-v1-1-5b304e408e11@rivosinc.com/ It would have been better to have helped Charlie fix their series (my apologies) but they agreed that the approach taken here was likely best for longer term maintainability: https://lore.kernel.org/lkml/Z6Jk_UN9i69QGqUj@ghost/ Signed-off-by: Ian Rogers Reviewed-by: Howard Chu Reviewed-by: Charlie Jenkins Reviewed-by: Namhyung Kim Acked-by: Arnaldo Carvalho de Melo Link: https://lore.kernel.org/r/20250319050741.269828-11-irogers@google.com Signed-off-by: Namhyung Kim

perf trace beauty: Add syscalltbl.sh generating all system call tables

2025-03-21T05:58:08Z

Rather than generating individual syscall header files generate a single trace/beauty/generated/syscalltbl.c. In a syscalltbls array have references to each architectures tables along with the corresponding e_machine. When the 32-bit or 64-bit table is ambiguous, match the perf binary's type. For ARM32 don't use the arm64 32-bit table which is smaller. EM_NONE is present for is no machine matches. Conditionally compile the tables, only having the appropriate 32 and 64-bit table. If ALL_SYSCALLTBL is defined all tables can be compiled. Add comment for noreturn column suggested by Arnd Bergmann: https://lore.kernel.org/lkml/d47c35dd-9c52-48e7-a00d-135572f11fbb@app.fastmail.com/ and added in commit 9142be9e6443 ("x86/syscall: Mark exit[_group] syscall handlers __noreturn"). Signed-off-by: Ian Rogers Reviewed-by: Howard Chu Reviewed-by: Charlie Jenkins Reviewed-by: Namhyung Kim Acked-by: Arnaldo Carvalho de Melo Link: https://lore.kernel.org/r/20250319050741.269828-9-irogers@google.com Signed-off-by: Namhyung Kim

perf build: Fix in-tree build due to symbolic link

2025-03-01T00:17:41Z

Building perf in-tree is broken after commit 890a1961c812 ("perf tools: Create source symlink in perf object dir") which added a 'source' symlink in the output dir pointing to the source dir. With in-tree builds, the added 'SOURCE = ...' line is executed multiple times (I observed 2 during the build plus 2 during installation). This is a minor inefficiency, in theory not harmful because symlink creation is assumed to be idempotent. But it is not. Considering with in-tree builds: srctree=/absolute/path/to/linux OUTPUT=/absolute/path/to/linux/tools/perf here's what happens: 1. ln -sf $(srctree)/tools/perf $(OUTPUT)/source -> creates /absolute/path/to/linux/tools/perf/source link to /absolute/path/to/linux/tools/perf => OK, that's what was intended 2. ln -sf $(srctree)/tools/perf $(OUTPUT)/source # same command as 1 -> creates /absolute/path/to/linux/tools/perf/perf link to /absolute/path/to/linux/tools/perf => Not what was intended, not idempotent 3. Now the build _should_ create the 'perf' executable, but it fails The reason is the tricky 'ln' command line. At the first invocation 'ln' uses the 1st form: ln [OPTION]... [-T] TARGET LINK_NAME and creates a link to TARGET *called LINK_NAME*. At the second invocation $(OUTPUT)/source exists, so 'ln' uses the 3rd form: ln [OPTION]... TARGET... DIRECTORY and creates a link to TARGET *called TARGET* inside DIRECTORY. Fix by adding -n/--no-dereference to "treat LINK_NAME as a normal file if it is a symbolic link to a directory", as the manpage says. Closes: https://lore.kernel.org/all/20241125182506.38af9907@booty/ Fixes: 890a1961c812 ("perf tools: Create source symlink in perf object dir") Signed-off-by: Luca Ceresoli Reviewed-by: Charlie Jenkins Tested-by: Charlie Jenkins Link: https://lore.kernel.org/r/20250124-perf-fix-intree-build-v1-1-485dd7a855e4@bootlin.com Signed-off-by: Namhyung Kim

tools: Unify top-level quiet infrastructure

2025-02-18T18:31:45Z

Commit f2868b1a66d4f40f ("perf tools: Expose quiet/verbose variables in Makefile.perf") moved the quiet infrastructure out of tools/build/Makefile.build and into the top-level Makefile.perf file so that the quiet infrastructure could be used throughout perf and not just in Makefile.build. Extract out the quiet infrastructure into Makefile.include so that it can be leveraged outside of perf. Fixes: f2868b1a66d4f40f ("perf tools: Expose quiet/verbose variables in Makefile.perf") Reviewed-by: Jiri Olsa Signed-off-by: Charlie Jenkins Acked-by: Andrii Nakryiko Cc: Adrian Hunter Cc: Alexander Shishkin Cc: Alexei Starovoitov Cc: Benjamin Tissoires Cc: Daniel Borkmann Cc: Daniel Lezcano Cc: Eduard Zingerman Cc: Hao Luo Cc: Ian Rogers Cc: Ingo Molnar Cc: Jiri Kosina Cc: John Fastabend Cc: Josh Poimboeuf Cc: KP Singh Cc: Lukasz Luba Cc: Mark Rutland Cc: Martin KaFai Lau Cc: Mykola Lysenko Cc: Namhyung Kim Cc: Peter Zijlstra Cc: Quentin Monnet Cc: Rafael J. Wysocki Cc: Shuah Khan Cc: Song Liu Cc: Stanislav Fomichev Cc: Steven Rostedt (VMware) Cc: Yonghong Song Cc: Zhang Rui Link: https://lore.kernel.org/r/20250213-quiet_tools-v3-1-07de4482a581@rivosinc.com Signed-off-by: Arnaldo Carvalho de Melo