linux/tools/build, branch v6.17

tools: Remove libcrypto dependency

2025-06-26T17:51:41Z

Remove all occurrence of libcrypto in the build system. Signed-off-by: Yuzhuo Jing Signed-off-by: Eric Biggers Reviewed-by: Ian Rogers Link: https://lore.kernel.org/r/20250625202311.23244-5-ebiggers@kernel.org Signed-off-by: Namhyung Kim

perf build: The bfd features are opt-in, stop testing for them by default

2025-06-24T17:27:50Z

These are leftovers noticed while updating a build container. We don't need those so that test-all.c can build and thus speed up the feature detection. Test for those features only if the user asks for BUILD_NONDISTRO=1 to build with libbfd. Signed-off-by: Arnaldo Carvalho de Melo Reviewed-by: James Clark Reviewed-by: Ian Rogers Link: https://lore.kernel.org/r/20250620212435.93846-4-acme@kernel.org Signed-off-by: Namhyung Kim

perf build: detect support for libbpf's emit_strings option

2025-06-20T21:48:14Z

This creates a config option that detects libbpf's ability to display character arrays as strings, which was just added to the BPF tree (https://git.kernel.org/bpf/bpf-next/c/87c9c79a02b4). To test this change, I built perf (from later in this patch set) with: - static libbpf (default, using source from kernel tree) - dynamic libbpf (LIBBPF_DYNAMIC=1 LIBBPF_INCLUDE=/usr/local/include) For both the static and dynamic versions, I used headers with and without the ".emit_strings" option. I verified that of the four resulting binaries, the two with ".emit_strings" would successfully record BPF_METADATA events, and the two without wouldn't. All four binaries would successfully display BPF_METADATA events, because the relevant bit of libbpf code is only used during "perf record". Signed-off-by: Blake Jones Link: https://lore.kernel.org/r/20250612194939.162730-2-blakejones@google.com Signed-off-by: Namhyung Kim

tools/build: Remove some unused libbpf pre-1.0 feature test logic

2025-06-09T18:18:19Z

Commit 76a97cf2e169 ("perf build: Remove libbpf pre-1.0 feature tests") removed the libbpf feature test logic used by perf in favor of using LIBBPF_MAJOR_VERSION. Remove some build targets that should have been removed as part of that clean up. Fixes: 76a97cf2e169 ("perf build: Remove libbpf pre-1.0 feature tests") Signed-off-by: Ian Rogers Link: https://lore.kernel.org/r/20250603221358.2562167-1-irogers@google.com Signed-off-by: Namhyung Kim

tools build: Don't show libbfd build status as it is opt-in

2025-04-10T13:45:30Z

Since dd317df072071903 ("perf build: Make binutil libraries opt in") doesn't try to build with binutils libraries, so showing that it is OFF when building causes just distraction, remove it from FEATURES_DISPLAY. For people that for some reason notice that there is always 'perf -vv', a short hand for 'perf version --build-options' and 'perf check feature libbfd' that now explains why it is not built: $ perf check feature libbfd libbfd: [ OFF ] # HAVE_LIBBFD_SUPPORT ( tip: Deprecated, license incompatibility, use BUILD_NONDISTRO=1 and install binutils-dev[el] ) $ Fixes: dd317df072071903 ("perf build: Make binutil libraries opt in") Tested-by: Ingo Molnar Cc: Adrian Hunter Cc: Dmitriy Vyukov Cc: Howard Chu Cc: Ian Rogers Cc: Jiri Olsa Cc: Kan Liang Cc: Namhyung Kim Cc: Peter Zijlstra Link: https://lore.kernel.org/r/Z--pWmTHGb62_83e@gmail.com Signed-off-by: Arnaldo Carvalho de Melo

tools build: Don't show libunwind build status as it is opt-in

2025-04-10T13:44:55Z

Since 13e17c9ff49119aa ("perf build: Make libunwind opt-in rather than opt-out") doesn't try to build with libunwind, so showing that it is OFF when building causes just distraction, remove it from FEATURES_DISPLAY. For people that for some reason notice that there is always 'perf -vv', a short hand for 'perf version --build-options' and 'perf check feature libunwind' that now explains why it is not built: $ perf check feature libunwind libunwind: [ OFF ] # HAVE_LIBUNWIND_SUPPORT ( tip: Deprecated, use LIBUNWIND=1 and install libunwind-dev[el] to build with it ) $ Fixes: 13e17c9ff49119aa ("perf build: Make libunwind opt-in rather than opt-out") Reported-by: Ingo Molnar Tested-by: Ingo Molnar Cc: Adrian Hunter Cc: Dmitriy Vyukov Cc: Howard Chu Cc: Ian Rogers Cc: Jiri Olsa Cc: Kan Liang Cc: Namhyung Kim Cc: Peter Zijlstra Link: https://lore.kernel.org/r/Z--pWmTHGb62_83e@gmail.com Signed-off-by: Arnaldo Carvalho de Melo

tools build: Don't set libunwind as available if test-all.c build succeeds

2025-04-10T13:43:43Z

The tools/build/feature/test-all.c file tries to detect the expected, most common set of libraries/features we expect to have available to build perf with. At some point libunwind was deemed not to be part of that set of libries, but the patches making it to be opt-in ended up forgetting some details, fix one more. Testing it: $ rm -rf /tmp/build/$(basename $PWD)/ ; mkdir -p /tmp/build/$(basename $PWD)/ $ rpm -q libunwind-devel libunwind-devel-1.8.0-3.fc40.x86_64 $ make -k LIBUNWIND=1 CORESIGHT=1 O=/tmp/build/$(basename $PWD)/ -C tools/perf install-bin |& grep unwind && ldd ~/bin/perf | grep unwind ... libunwind: [ on ] CC /tmp/build/perf-tools-next/arch/x86/tests/dwarf-unwind.o CC /tmp/build/perf-tools-next/arch/x86/util/unwind-libunwind.o CC /tmp/build/perf-tools-next/util/arm64-frame-pointer-unwind-support.o CC /tmp/build/perf-tools-next/tests/dwarf-unwind.o CC /tmp/build/perf-tools-next/util/unwind-libunwind-local.o CC /tmp/build/perf-tools-next/util/unwind-libunwind.o libunwind-x86_64.so.8 => /lib64/libunwind-x86_64.so.8 (0x00007f615a549000) libunwind.so.8 => /lib64/libunwind.so.8 (0x00007f615a52f000) $ sudo rpm -e libunwind-devel $ rm -rf /tmp/build/$(basename $PWD)/ ; mkdir -p /tmp/build/$(basename $PWD)/ $ make -k LIBUNWIND=1 CORESIGHT=1 O=/tmp/build/$(basename $PWD)/ -C tools/perf install-bin |& grep unwind && ldd ~/bin/perf | grep unwind Makefile.config:653: No libunwind found. Please install libunwind-dev[el] >= 1.1 and/or set LIBUNWIND_DIR ... libunwind: [ OFF ] CC /tmp/build/perf-tools-next/arch/x86/tests/dwarf-unwind.o CC /tmp/build/perf-tools-next/arch/x86/util/unwind-libdw.o CC /tmp/build/perf-tools-next/util/arm64-frame-pointer-unwind-support.o CC /tmp/build/perf-tools-next/tests/dwarf-unwind.o CC /tmp/build/perf-tools-next/util/unwind-libdw.o $ Should be in a separate patch, but tired now, so also adding a message about the need to use LIBUNWIND=1 in the output when its not available, so done here as well. So, now when the devel files are not available we get: $ make -k LIBUNWIND=1 CORESIGHT=1 O=/tmp/build/$(basename $PWD)/ -C tools/perf install-bin |& grep unwind && ldd ~/bin/perf | grep unwind Makefile.config:653: No libunwind found. Please install libunwind-dev[el] >= 1.1 and/or set LIBUNWIND_DIR and set LIBUNWIND=1 in the make command line as it is opt-in now ... libunwind: [ OFF ] $ Fixes: 13e17c9ff49119aa ("perf build: Make libunwind opt-in rather than opt-out") Reported-by: Ingo Molnar Tested-by: Ingo Molnar Cc: Adrian Hunter Cc: Dmitriy Vyukov Cc: Ian Rogers Cc: James Clark Cc: Jiri Olsa Cc: Kan Liang Cc: Namhyung Kim Cc: Peter Zijlstra Link: https://lore.kernel.org/r/Z_AnsW9oJzFbhIFC@x1 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 ...

Merge tag 'trace-tools-v6.15' of git://git.kernel.org/pub/scm/linux/kernel/git/trace/linux-trace

2025-03-28T00:03:01Z

Pull tracing tooling updates from Steven Rostedt: - Allow RTLA to collect data via BPF The current implementation of rtla uses libtracefs and libtraceevent to pull sample events generated by the timerlat tracer from the trace buffer. rtla then processes the sample by updating the histogram and summary (current, maximum, minimum, and sum values) as well as checks if tracing has been stopped due to threshold overflow. In use cases where a large number of samples is being generated, that is, with measurements running on many CPUs and with a low interval, this sample processing design causes a significant CPU load on the rtla side. Furthermore, with >100 CPUs and 100us interval, rtla was reported as not being able to keep up with the samples and dropping most of them, leading to it being unusable. Change the way the timerlat trace processes samples by attaching a BPF program to the trace event using the BPF skeleton feature of bpftool. Unlike the current implementation, the BPF implementation does not check whether tracing is stopped (in BPF mode, tracing is always off to improve performance), but waits for a write to a BPF ringbuffer instead. This allows rtla to exit immediately when a threshold is violated, without waiting for the next iteration of the while loop. If the requirements for the BPF implementation are not met, either at build time or at run time, the current implementation is used as fallback. Which implementation is being used can be seen when running rtla timerlat with "-D" option. rtla can be forced to run in non-BPF mode by setting the RTLA_NO_BPF option to 1, for debugging purposes. - Fix LD_FLAGS from being dropped in build - Refactor code to remove duplication of save_trace_to_file - Always set options and do not rely on default settings Do not rely on the default kernel settings of the tracers when starting. They could have been changed by the user which gives inconsistent results. Always set the options that rtla expects. - Add creation of ctags and TAGS for traversing code * tag 'trace-tools-v6.15' of git://git.kernel.org/pub/scm/linux/kernel/git/trace/linux-trace: rtla: Add the ability to create ctags and etags rtla/tests: Test setting default options rtla/tests: Reset osnoise options before check rtla: Always set all tracer options rtla/osnoise: Set OSNOISE_WORKLOAD to true rtla: Unify apply_config between top and hist rtla/osnoise: Unify params struct rtla: Fix segfault in save_trace_to_file call tools/build: Use SYSTEM_BPFTOOL for system bpftool rtla: Refactor save_trace_to_file tools/rv: Keep user LDFLAGS in build rtla/timerlat: Test BPF mode rtla/timerlat_top: Use BPF to collect samples rtla/timerlat_top: Move divisor to update rtla/timerlat_hist: Use BPF to collect samples rtla/timerlat: Add BPF skeleton to collect samples rtla: Add optional dependency on BPF tooling tools/build: Add bpftool-skeletons feature test rtla/timerlat: Unify params struct

tools/build: Use SYSTEM_BPFTOOL for system bpftool

2025-03-26T14:31:31Z

The feature test for system bpftool uses BPFTOOL as the variable to set its path, defaulting to just "bpftool" if not set by the user. This conflicts with selftests and a few other utilities, which expect BPFTOOL to be set to the in-tree bpftool path by default. For example, bpftool selftests fail to build: $ make -C tools/testing/selftests/bpf/ make: Entering directory '/home/tglozar/dev/linux/tools/testing/selftests/bpf' make: *** No rule to make target 'bpftool', needed by '/home/tglozar/dev/linux/tools/testing/selftests/bpf/tools/include/vmlinux.h'. Stop. make: Leaving directory '/home/tglozar/dev/linux/tools/testing/selftests/bpf' Fix the problem by renaming the variable used for system bpftool from BPFTOOL to SYSTEM_BPFTOOL, so that the new usage does not conflict with the existing one of BPFTOOL. Cc: John Kacur Cc: Luis Goncalves Link: https://lore.kernel.org/20250326004018.248357-1-tglozar@redhat.com Fixes: 8a635c3856dd ("tools/build: Add bpftool-skeletons feature test") Closes: https://lore.kernel.org/linux-kernel/5df6968a-2e5f-468e-b457-fc201535dd4c@linux.ibm.com/ Reported-by: Venkat Rao Bagalkote Tested-by: Venkat Rao Bagalkote Suggested-by: Quentin Monnet Acked-by: Quentin Monnet Signed-off-by: Tomas Glozar Signed-off-by: Steven Rostedt (Google)