- 08 Aug, 2023 1 commit
-
-
Ivan Babrou authored
Commit 3fd7a168 ("perf script: Add 'cgroup' field for output") added support for printing cgroup path in perf script output. It was okay if you didn't want any stacks: $ sudo perf script --comms jpegtran:23f4bf -F comm,tid,cpu,time,cgroup jpegtran:23f4bf 3321915 [013] 404718.587488: /idle.slice/polish.service jpegtran:23f4bf 3321915 [031] 404718.592073: /idle.slice/polish.service With stacks it gets messier as cgroup is printed after the stack: $ perf script --comms jpegtran:23f4bf -F comm,tid,cpu,time,cgroup,ip,sym jpegtran:23f4bf 3321915 [013] 404718.587488: 5c554 compress_output 570d9 jpeg_finish_compress 3476e jpegtran_main 330ee jpegtran::main 326e2 core::ops::function::FnOnce::call_once (inlined) 326e2 std::sys_common::backtrace::__rust_begin_short_backtrace /idle.slice/polish.service jpegtran:23f4bf 3321915 [031] 404718.592073: 8474d jsimd_encode_mcu_AC_first_prepare_sse2.PADDING 55af68e62fff [unknown] /idle.slice/polish.service Let's instead print cgroup on the same line as comm: $ perf script --comms jpegtran:23f4bf -F comm,tid,cpu,time,cgroup,ip,sym jpegtran:23f4bf 3321915 [013] 404718.587488: /idle.slice/polish.service 5c554 compress_output 570d9 jpeg_finish_compress 3476e jpegtran_main 330ee jpegtran::main 326e2 core::ops::function::FnOnce::call_once (inlined) 326e2 std::sys_common::backtrace::__rust_begin_short_backtrace jpegtran:23f4bf 3321915 [031] 404718.592073: /idle.slice/polish.service 8474d jsimd_encode_mcu_AC_first_prepare_sse2.PADDING 55af68e62fff [unknown] Fixes: 3fd7a168 ("perf script: Add 'cgroup' field for output") Signed-off-by:
Ivan Babrou <ivan@cloudflare.com> Acked-by:
Ian Rogers <irogers@google.com> Acked-by:
Namhyung Kim <namhyung@kernel.org> Cc: Adrian Hunter <adrian.hunter@intel.com> Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Mark Rutland <mark.rutland@arm.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: kernel-team@cloudflare.com Link: https://lore.kernel.org/r/20230718000737.49077-1-ivan@cloudflare.comSigned-off-by:
Arnaldo Carvalho de Melo <acme@redhat.com>
-
- 07 Aug, 2023 3 commits
-
-
Arnaldo Carvalho de Melo authored
Not used in any other place, so just make it static. Cc: Adrian Hunter <adrian.hunter@intel.com> Cc: Ian Rogers <irogers@google.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Masami Hiramatsu <mhiramat@kernel.org> Cc: Namhyung Kim <namhyung@kernel.org> Link: https://lore.kernel.org/lkml/ZM0pjfOe6R4X%2Fcql@kernel.org/Signed-off-by:
-
Arnaldo Carvalho de Melo authored
perf probe: Free string returned by synthesize_perf_probe_point() on failure in synthesize_perf_probe_command() Building perf with EXTRA_CFLAGS="-fsanitize=address" a leak was detected elsewhere and lead to an audit, where we found that synthesize_perf_probe_command() may leak synthesize_perf_probe_point() return on failure, fix it. Cc: Adrian Hunter <adrian.hunter@intel.com> Cc: Ian Rogers <irogers@google.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Masami Hiramatsu <mhiramat@kernel.org> Cc: Namhyung Kim <namhyung@kernel.org> Link: https://lore.kernel.org/lkml/ZM0mzpQktHnhXJXr@kernel.orgSigned-off-by:
-
Arnaldo Carvalho de Melo authored
Building perf with EXTRA_CFLAGS="-fsanitize=address" a leak is detect when trying to add a probe to a non-existent function: # perf probe -x ~/bin/perf dso__neW Probe point 'dso__neW' not found. Error: Failed to add events. ================================================================= ==296634==ERROR: LeakSanitizer: detected memory leaks Direct leak of 128 byte(s) in 1 object(s) allocated from: #0 0x7f67642ba097 in calloc (/lib64/libasan.so.8+0xba097) #1 0x7f67641a76f1 in allocate_cfi (/lib64/libdw.so.1+0x3f6f1) Direct leak of 65 byte(s) in 1 object(s) allocated from: #0 0x7f67642b95b5 in __interceptor_realloc.part.0 (/lib64/libasan.so.8+0xb95b5) #1 0x6cac75 in strbuf_grow util/strbuf.c:64 #2 0x6ca934 in strbuf_init util/strbuf.c:25 #3 0x9337d2 in synthesize_perf_probe_point util/probe-event.c:2018 #4 0x92be51 in try_to_find_probe_trace_events util/probe-event.c:964 #5 0x93d5c6 in convert_to_probe_trace_events util/probe-event.c:3512 #6 0x93d6d5 in convert_perf_probe_events util/probe-event.c:3529 #7 0x56f37f in perf_add_probe_events /var/home/acme/git/perf-tools-next/tools/perf/builtin-probe.c:354 #8 0x572fbc in __cmd_probe /var/home/acme/git/perf-tools-next/tools/perf/builtin-probe.c:738 #9 0x5730f2 in cmd_probe /var/home/acme/git/perf-tools-next/tools/perf/builtin-probe.c:766 #10 0x635d81 in run_builtin /var/home/acme/git/perf-tools-next/tools/perf/perf.c:323 #11 0x6362c1 in handle_internal_command /var/home/acme/git/perf-tools-next/tools/perf/perf.c:377 #12 0x63667a in run_argv /var/home/acme/git/perf-tools-next/tools/perf/perf.c:421 #13 0x636b8d in main /var/home/acme/git/perf-tools-next/tools/perf/perf.c:537 #14 0x7f676302950f in __libc_start_call_main (/lib64/libc.so.6+0x2950f) SUMMARY: AddressSanitizer: 193 byte(s) leaked in 2 allocation(s). # synthesize_perf_probe_point() returns a "detachec" strbuf, i.e. a malloc'ed string that needs to be free'd. An audit will be performed to find other such cases. Acked-by:Masami Hiramatsu <mhiramat@kernel.org> Cc: Adrian Hunter <adrian.hunter@intel.com> Cc: Ian Rogers <irogers@google.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Namhyung Kim <namhyung@kernel.org> Link: https://lore.kernel.org/lkml/ZM0l1Oxamr4SVjfY@kernel.orgSigned-off-by:
-
- 04 Aug, 2023 1 commit
-
-
Arnaldo Carvalho de Melo authored
To pick up the fixes that were just merged from perf-tools/perf-tools for v6.5. Signed-off-by:
-
- 03 Aug, 2023 35 commits
-
-
Linus Torvalds authored
Merge tag 'perf-tools-fixes-for-v6.5-2-2023-08-03' of git://git.kernel.org/pub/scm/linux/kernel/git/perf/perf-tools Pull perf tools fixes from Arnaldo Carvalho de Melo: - Fix segfault in the powerpc specific arch_skip_callchain_idx function. The patch doing the reference count init/exit that went into 6.5 missed this function. - Fix regression reading the arm64 PMU cpu slots in sysfs, a patch removing some code duplication ended up duplicating the /sysfs prefix for these files. - Fix grouping of events related to topdown, addressing a regression on the CSV output produced by 'perf stat' noticed on the downstream tool toplev. - Fix the uprobe_from_different_cu 'perf test' entry, it is failing when gcc isn't available, so we need to check that and skip the test if it is not installed. * tag 'perf-tools-fixes-for-v6.5-2-2023-08-03' of git://git.kernel.org/pub/scm/linux/kernel/git/perf/perf-tools: perf test parse-events: Test complex name has required event format perf pmus: Create placholder regardless of scanning core_only perf test uprobe_from_different_cu: Skip if there is no gcc perf parse-events: Only move force grouped evsels when sorting perf parse-events: When fixing group leaders always set the leader perf parse-events: Extra care around force grouped events perf callchain powerpc: Fix addr location init during arch_skip_callchain_idx function perf pmu arm64: Fix reading the PMU cpu slots in sysfs
-
git://git.kernel.org/pub/scm/linux/kernel/git/cxl/cxlLinus Torvalds authored
Pull cxl fixes from Vishal Verma: - Fixup the Sanitixe device ABI that was merged for v6.5 to hide some sysfs files when the necessary support is missing. Update the ABI documentation around this as well. * tag 'cxl-fixes-6.5-rc5' of git://git.kernel.org/pub/scm/linux/kernel/git/cxl/cxl: cxl/memdev: Only show sanitize sysfs files when supported cxl/memdev: Document security state in kern-doc cxl/memdev: Improve sanitize ABI descriptions
-
git://git.kernel.org/pub/scm/linux/kernel/git/netdev/netLinus Torvalds authored
Pull networking fixes from Jakub Kicinski: "Including fixes from bpf and wireless. Nothing scary here. Feels like the first wave of regressions from v6.5 is addressed - one outstanding fix still to come in TLS for the sendpage rework. Current release - regressions: - udp: fix __ip_append_data()'s handling of MSG_SPLICE_PAGES - dsa: fix older DSA drivers using phylink Previous releases - regressions: - gro: fix misuse of CB in udp socket lookup - mlx5: unregister devlink params in case interface is down - Revert "wifi: ath11k: Enable threaded NAPI" Previous releases - always broken: - sched: cls_u32: fix match key mis-addressing - sched: bind logic fixes for cls_fw, cls_u32 and cls_route - add bound checks to a number of places which hand-parse netlink - bpf: disable preemption in perf_event_output helpers code - qed: fix scheduling in a tasklet while getting stats - avoid using APIs which are not hardirq-safe in couple of drivers, when we may be in a hard IRQ (netconsole) - wifi: cfg80211: fix return value in scan logic, avoid page allocator warning - wifi: mt76: mt7615: do not advertise 5 GHz on first PHY of MT7615D (DBDC) Misc: - drop handful of inactive maintainers, put some new in place" * tag 'net-6.5-rc5' of git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net: (98 commits) MAINTAINERS: update TUN/TAP maintainers test/vsock: remove vsock_perf executable on `make clean` tcp_metrics: fix data-race in tcpm_suck_dst() vs fastopen tcp_metrics: annotate data-races around tm->tcpm_net tcp_metrics: annotate data-races around tm->tcpm_vals[] tcp_metrics: annotate data-races around tm->tcpm_lock tcp_metrics: annotate data-races around tm->tcpm_stamp tcp_metrics: fix addr_same() helper prestera: fix fallback to previous version on same major version udp: Fix __ip_append_data()'s handling of MSG_SPLICE_PAGES net/mlx5e: Set proper IPsec source port in L4 selector net/mlx5: fs_core: Skip the FTs in the same FS_TYPE_PRIO_CHAINS fs_prio net/mlx5: fs_core: Make find_closest_ft more generic wifi: brcmfmac: Fix field-spanning write in brcmf_scan_params_v2_to_v1() vxlan: Fix nexthop hash size ip6mr: Fix skb_under_panic in ip6mr_cache_report() s390/qeth: Don't call dev_close/dev_open (DOWN/UP) net: tap_open(): set sk_uid from current_fsuid() net: tun_chr_open(): set sk_uid from current_fsuid() net: dcb: choose correct policy to parse DCB_ATTR_BCN ... -
Namhyung Kim authored
The 'e' key is to toggle expand/collapse the selected entry only. But the current code has a bug that it only increases the number of entries by 1 in the hierarchy mode so users cannot move under the current entry after the key stroke. This is due to a wrong assumption in the hist_entry__set_folding(). The commit b33f9226 ("perf hists browser: Put hist_entry folding logic into single function") factored out the code, but actually it should be handled separately. The hist_browser__set_folding() is to update fold state for each entry so it needs to traverse all (child) entries regardless of the current fold state. So it increases the number of entries by 1. But the hist_entry__set_folding() only cares the currently selected entry and its all children. So it should count all unfolded child entries. This code is implemented in hist_browser__toggle_fold() already so we can just call it. Fixes: b33f9226 ("perf hists browser: Put hist_entry folding logic into single function") Signed-off-by:
-
Namhyung Kim authored
The commit ef9ff601 ("perf ui browser: Move the extra title lines from the hists browser") introduced ui_browser__gotorc_title() to help moving non-title lines easily. But it missed to update the title for the hierarchy mode so it won't print the header line on TUI at all. $ perf report --hierarchy Fixes: ef9ff601 ("perf ui browser: Move the extra title lines from the hists browser") Signed-off-by:
-
Arnaldo Carvalho de Melo authored
In 616b14b4 ("perf build: Conditionally define NDEBUG") we started using NDEBUG=1 when DEBUG=1 isn't present, so code that is enclosed with assert() is not called. In dd317df0 ("perf build: Make binutil libraries opt in") we stopped linking against binutils-devel, for licensing reasons. Recently people asked me why annotation of BPF programs wasn't working, i.e. this: $ perf annotate bpf_prog_5280546344e3f45c_kfree_skb was returning: case SYMBOL_ANNOTATE_ERRNO__NO_LIBOPCODES_FOR_BPF: scnprintf(buf, buflen, "Please link with binutils's libopcode to enable BPF annotation"); This was on a fedora rpm, so its new enough that I had to try to test by rebuilding using BUILD_NONDISTRO=1, only to get it segfaulting on me. This combination made this libopcode function not to be called: assert(bfd_check_format(bfdf, bfd_object)); Changing it to: if (!bfd_check_format(bfdf, bfd_object)) abort(); Made it work, looking at this "check" function made me realize it changes the 'bfdf' internal state, i.e. we better call it. So stop using assert() on it, just call it and abort if it fails. Probably it is better to propagate the error, etc, but it seems it is unlikely to fail from the usage done so far and we really need to stop using libopcodes, so do the quick fix above and move on. With it we have BPF annotation back working when built with BUILD_NONDISTRO=1: ⬢[acme@toolbox perf-tools-next]$ perf annotate --stdio2 bpf_prog_5280546344e3f45c_kfree_skb | head No kallsyms or vmlinux with build-id 939bc71a1a51cdc434e60af93c7e734f7d5c0e7e was found Samples: 12 of event 'cpu-clock:ppp', 4000 Hz, Event count (approx.): 3000000, [percent: local period] bpf_prog_5280546344e3f45c_kfree_skb() bpf_prog_5280546344e3f45c_kfree_skb Percent int kfree_skb(struct trace_event_raw_kfree_skb *args) { nop 33.33 xchg %ax,%ax push %rbp mov %rsp,%rbp sub $0x180,%rsp push %rbx push %r13 ⬢[acme@toolbox perf-tools-next]$ Fixes: 6987561c ("perf annotate: Enable annotation of BPF programs") Cc: Adrian Hunter <adrian.hunter@intel.com> Cc: Ian Rogers <irogers@google.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Mohamed Mahmoud <mmahmoud@redhat.com> Cc: Namhyung Kim <namhyung@kernel.org> Cc: Dave Tucker <datucker@redhat.com> Cc: Derek Barbosa <debarbos@redhat.com> Cc: Song Liu <songliubraving@fb.com> Link: https://lore.kernel.org/lkml/ZMrMzoQBe0yqMek1@kernel.orgSigned-off-by:
-
Thomas Richter authored
Perf build suports llvm and clang support compiled in. Test case 56 builtin clang support provides a test case which is always skipped. Link perf with the latest llvm and clang libraries and enable this test case. Use 'make LIBCLANGLLVM=1' to include this support. V2: Add Library patch before -lclang-cpp Output before: # ./perf test 56 56: builtin clang support : 56.1: builtin clang compile C source to IR : Skip (not compiled in) 56.2: builtin clang compile C source to ELF object: Skip (not compiled in) Output after: # ./perf test 56 56: builtin clang support : 56.1: builtin clang compile C source to IR : Ok 56.2: builtin clang compile C source to ELF object : Ok # From Ian Rogers: Build tested with LLVM 14 and 15 using: BUILD_BPF_SKEL=1 LIBCLANGLLVM=1 LLVM_CONFIG=llvm-config-14 BUILD_BPF_SKEL=1 LIBCLANGLLVM=1 LLVM_CONFIG=llvm-config-15 Signed-off-by:
Thomas Richter <tmricht@linux.ibm.com> Tested-by:
-
Thomas Richter authored
Perf build auto-detects features and packages already installed for its build. This is done in directory tools/build/feature. This directory contains small sample programs. When they successfully compile the necessary prereqs in form of libraries and header files are present. Such a check is also done for llvm and clang. And the checks fail. Fix this and update to the latest C++ standard and use the new library provided by clang (which contains new packaging) s/ee this link for reference: https://fedoraproject.org/wiki/Changes/Stop-Shipping-Individual-Component-Libraries-In-clang-lib-Package Output before: # rm -f ./test-clang.bin; make test-clang.bin; ./test-clang.bin; \ ll test-clang.make.output g++ -MD -Wall -Werror -o test-clang.bin test-clang.cpp \ > test-clang.make.output 2>&1 -std=gnu++14 \ -I/usr/include \ -L/usr/lib64 \ -Wl,--start-group -lclangBasic -lclangDriver \ -lclangFrontend -lclangEdit -lclangLex \ -lclangAST -Wl,--end-group \ -lLLVM-16 \ \ > test-clang.make.output 2>&1 make: *** [Makefile:356: test-clang.bin] Error 1 -bash: ./test-clang.bin: No such file or directory -rw-r--r--. 1 root root 252041 Jul 12 09:56 test-clang.make.output # File test-clang.make.output contains many lines of unreferenced symbols. Output after: # rm -f ./test-clang.bin; make test-clang.bin; ./test-clang.bin; \ cat test-clang.make.output g++ -MD -Wall -Werror -o test-clang.bin test-clang.cpp \ > test-clang.make.output 2>&1 -std=gnu++17 \ -I/usr/include \ -L/usr/lib64 \ -Wl,--start-group -lclang-cpp -Wl,--end-group \ -lLLVM-16 \ \ > test-clang.make.output 2>&1 # Committer notes: Test it in the tools/build/feature directory, and have clang-devel and llvm-devel installed. Signed-off-by:
-
Arnaldo Carvalho de Melo authored
With -Werror the build was failing on fedora rawhide: [perfbuilder@27cfe44d67ed perf-6.5.0-rc2]$ gcc -v Using built-in specs. COLLECT_GCC=gcc COLLECT_LTO_WRAPPER=/usr/libexec/gcc/x86_64-redhat-linux/13/lto-wrapper OFFLOAD_TARGET_NAMES=nvptx-none OFFLOAD_TARGET_DEFAULT=1 Target: x86_64-redhat-linux Configured with: ../configure --enable-bootstrap --enable-languages=c,c++,fortran,objc,obj-c++,ada,go,d,m2,lto --prefix=/usr --mandir=/usr/share/man --infodir=/usr/share/info --with-bugurl=http://bugzilla.redhat.com/bugzilla --enable-shared --enable-threads=posix --enable-checking=release --enable-multilib --with-system-zlib --enable-__cxa_atexit --disable-libunwind-exceptions --enable-gnu-unique-object --enable-linker-build-id --with-gcc-major-version-only --enable-libstdcxx-backtrace --with-libstdcxx-zoneinfo=/usr/share/zoneinfo --with-linker-hash-style=gnu --enable-plugin --enable-initfini-array --with-isl=/builddir/build/BUILD/gcc-13.2.1-20230728/obj-x86_64-redhat-linux/isl-install --enable-offload-targets=nvptx-none --without-cuda-driver --enable-offload-defaulted --enable-gnu-indirect-function --enable-cet --with-tune=generic --with-arch_32=i686 --build=x86_64-redhat-linux --with-build-config=bootstrap-lto --enable-link-serialization=1 Thread model: posix Supported LTO compression algorithms: zlib zstd gcc version 13.2.1 20230728 (Red Hat 13.2.1-1) (GCC) [perfbuilder@27cfe44d67ed perf-6.5.0-rc2]$ In file included from /usr/include/python3.12/Python.h:44, from scripts/python/Perf-Trace-Util/Context.c:14: /usr/include/python3.12/object.h: In function 'Py_SIZE': /usr/include/python3.12/object.h:217:5: error: ISO C90 forbids mixed declarations and code [-Werror=declaration-after-statement] 217 | PyVarObject *var_ob = _PyVarObject_CAST(ob); | ^~~~~~~~~~~ In file included from /usr/include/python3.12/Python.h:53: /usr/include/python3.12/cpython/longintrepr.h: In function '_PyLong_CompactValue': /usr/include/python3.12/cpython/longintrepr.h:121:5: error: ISO C90 forbids mixed declarations and code [-Werror=declaration-after-statement] 121 | Py_ssize_t sign = 1 - (op->long_value.lv_tag & _PyLong_SIGN_MASK); | ^~~~~~~~~~ <SNIP> In file included from /usr/include/python3.12/Python.h:44, from util/scripting-engines/trace-event-python.c:22: /usr/include/python3.12/object.h: In function 'Py_SIZE': /usr/include/python3.12/object.h:217:5: error: ISO C90 forbids mixed declarations and code [-Werror=declaration-after-statement] 217 | PyVarObject *var_ob = _PyVarObject_CAST(ob); | ^~~~~~~~~~~ CC /tmp/build/perf/util/units.o CC /tmp/build/perf/util/time-utils.o In file included from /usr/include/python3.12/Python.h:53: /usr/include/python3.12/cpython/longintrepr.h: In function '_PyLong_CompactValue': /usr/include/python3.12/cpython/longintrepr.h:121:5: error: ISO C90 forbids mixed declarations and code [-Werror=declaration-after-statement] 121 | Py_ssize_t sign = 1 - (op->long_value.lv_tag & _PyLong_SIGN_MASK); | ^~~~~~~~~~ So add -Wno-declaration-after-statement to the python scripting CFLAGS. Reviewed-by:
-
Arnaldo Carvalho de Melo authored
With -Werror the build was failing on fedora rawhide: [perfbuilder@27cfe44d67ed perf-6.5.0-rc2]$ gcc -v Using built-in specs. COLLECT_GCC=gcc COLLECT_LTO_WRAPPER=/usr/libexec/gcc/x86_64-redhat-linux/13/lto-wrapper OFFLOAD_TARGET_NAMES=nvptx-none OFFLOAD_TARGET_DEFAULT=1 Target: x86_64-redhat-linux Configured with: ../configure --enable-bootstrap --enable-languages=c,c++,fortran,objc,obj-c++,ada,go,d,m2,lto --prefix=/usr --mandir=/usr/share/man --infodir=/usr/share/info --with-bugurl=http://bugzilla.redhat.com/bugzilla --enable-shared --enable-threads=posix --enable-checking=release --enable-multilib --with-system-zlib --enable-__cxa_atexit --disable-libunwind-exceptions --enable-gnu-unique-object --enable-linker-build-id --with-gcc-major-version-only --enable-libstdcxx-backtrace --with-libstdcxx-zoneinfo=/usr/share/zoneinfo --with-linker-hash-style=gnu --enable-plugin --enable-initfini-array --with-isl=/builddir/build/BUILD/gcc-13.2.1-20230728/obj-x86_64-redhat-linux/isl-install --enable-offload-targets=nvptx-none --without-cuda-driver --enable-offload-defaulted --enable-gnu-indirect-function --enable-cet --with-tune=generic --with-arch_32=i686 --build=x86_64-redhat-linux --with-build-config=bootstrap-lto --enable-link-serialization=1 Thread model: posix Supported LTO compression algorithms: zlib zstd gcc version 13.2.1 20230728 (Red Hat 13.2.1-1) (GCC) [perfbuilder@27cfe44d67ed perf-6.5.0-rc2]$ In file included from /usr/include/python3.12/Python.h:44, from /git/perf-6.5.0-rc2/tools/perf/util/python.c:2: /usr/include/python3.12/object.h: In function ‘Py_SIZE’: /usr/include/python3.12/object.h:217:5: error: ISO C90 forbids mixed declarations and code [-Werror=declaration-after-statement] 217 | PyVarObject *var_ob = _PyVarObject_CAST(ob); | ^~~~~~~~~~~ LD /tmp/build/perf/arch/perf-in.o In file included from /usr/include/python3.12/Python.h:53: /usr/include/python3.12/cpython/longintrepr.h: In function ‘_PyLong_CompactValue’: /usr/include/python3.12/cpython/longintrepr.h:121:5: error: ISO C90 forbids mixed declarations and code [-Werror=declaration-after-statement] 121 | Py_ssize_t sign = 1 - (op->long_value.lv_tag & _PyLong_SIGN_MASK); | ^~~~~~~~~~ So add -Wno-declaration-after-statement to the python binding CFLAGS. Reviewed-by:
-
Ian Rogers authored
Avoid grouping events especially in cases where the kernel's PMU driver fails to not open the events, causing the events to report back as "<not counted>". This update comes from: https://github.com/intel/perfmon/pull/94 Fixes issues reported with patch: https://lore.kernel.org/lkml/20230719001836.198363-3-irogers@google.com/Reviewed-by:
Kan Liang <kan.liang@linux.intel.com> Signed-off-by:
-
Ian Rogers authored
1.15 events were released in: https://github.com/intel/perfmon/commit/76dfb81a1148ec049fd9caae9c62529404da63df Adds the events OCR.DEMAND_DATA_RD.LOCAL_SOCKET_PMM and OCR.DEMAND_DATA_RD.PMM. Reviewed-by:
-
Ian Rogers authored
1.04 events were released in: https://github.com/intel/perfmon/commit/44fe3681501f43fc515577aced8e944b187c8e51 Addition of 51 core events. Reviewed-by:
-
Ian Rogers authored
As topdown.slots may appear as slots it may get confused with uops_retired.slots which is an invalid perf metric event group leader. Special case uops_retired.slots to avoid this confusion. Reviewed-by:
-
Masami Hiramatsu authored
Since @symbol variable access is not supported by uprobe event, it must be correctly warn user instead of kernel version update. Committer testing: With/without the patch: [root@quaco ~]# perf probe -x ~/bin/perf -L sigtrap_handler <sigtrap_handler@/home/acme/git/perf-tools-next/tools/perf/tests/sigtrap.c:0> 0 sigtrap_handler(int signum __maybe_unused, siginfo_t *info, void *ucontext __maybe_unused) 1 { 2 if (!__atomic_fetch_add(&ctx.signal_count, 1, __ATOMIC_RELAXED)) 3 ctx.first_siginfo = *info; 4 __atomic_fetch_sub(&ctx.tids_want_signal, syscall(SYS_gettid), __ATOMIC_RELAXED); 5 } static void *test_thread(void *arg) { [root@quaco ~]# perf probe -x ~/bin/perf sigtrap_handler:4 "ctx.signal_count" Without the patch: [root@quaco ~]# perf probe -x ~/bin/perf sigtrap_handler:4 "ctx.signal_count" Failed to write event: Invalid argument Please upgrade your kernel to at least 3.14 to have access to feature @ctx Error: Failed to add events. [root@quaco ~]# With the patch: [root@quaco ~]# Failed to write event: Invalid argument @ctx accesses a variable by symbol name, but that is not supported for user application probe. Error: Failed to add events. [root@quaco ~]# Reported-by:Arnaldo Carvalho de Melo <acme@kernel.org> Signed-off-by:
-
Arnaldo Carvalho de Melo authored
First noticed on Fedora Rawhide: tests/bpf.c: In function ‘epoll_pwait_loop’: tests/bpf.c:36:17: error: argument 2 null where non-null expected [-Werror=nonnull] 36 | epoll_pwait(-(i + 1), NULL, 0, 0, NULL); | ^~~~~~~~~~~ In file included from tests/bpf.c:5: /usr/include/sys/epoll.h:134:12: note: in a call to function ‘epoll_pwait’ declared ‘nonnull’ 134 | extern int epoll_pwait (int __epfd, struct epoll_event *__events, | ^~~~~~~~~~~ [perfbuilder@27cfe44d67ed perf-6.5.0-rc2]$ gcc -v Using built-in specs. COLLECT_GCC=gcc COLLECT_LTO_WRAPPER=/usr/libexec/gcc/x86_64-redhat-linux/13/lto-wrapper OFFLOAD_TARGET_NAMES=nvptx-none OFFLOAD_TARGET_DEFAULT=1 Target: x86_64-redhat-linux Configured with: ../configure --enable-bootstrap --enable-languages=c,c++,fortran,objc,obj-c++,ada,go,d,m2,lto --prefix=/usr --mandir=/usr/share/man --infodir=/usr/share/info --with-bugurl=http://bugzilla.redhat.com/bugzilla --enable-shared --enable-threads=posix --enable-checking=release --enable-multilib --with-system-zlib --enable-__cxa_atexit --disable-libunwind-exceptions --enable-gnu-unique-object --enable-linker-build-id --with-gcc-major-version-only --enable-libstdcxx-backtrace --with-libstdcxx-zoneinfo=/usr/share/zoneinfo --with-linker-hash-style=gnu --enable-plugin --enable-initfini-array --with-isl=/builddir/build/BUILD/gcc-13.2.1-20230728/obj-x86_64-redhat-linux/isl-install --enable-offload-targets=nvptx-none --without-cuda-driver --enable-offload-defaulted --enable-gnu-indirect-function --enable-cet --with-tune=generic --with-arch_32=i686 --build=x86_64-redhat-linux --with-build-config=bootstrap-lto --enable-link-serialization=1 Thread model: posix Supported LTO compression algorithms: zlib zstd gcc version 13.2.1 20230728 (Red Hat 13.2.1-1) (GCC) [perfbuilder@27cfe44d67ed perf-6.5.0-rc2]$ Just add that argument to address this compiler warning. Cc: Adrian Hunter <adrian.hunter@intel.com> Cc: Ian Rogers <irogers@google.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Namhyung Kim <namhyung@kernel.org> Link: https://lore.kernel.org/lkml/ZMj8+bvN86D0ZKiB@kernel.orgSigned-off-by: -
Athira Rajeev authored
perf tests stat+std_output: Fix shellcheck warnings about word splitting/quoting and local variables Running shellcheck on stat_std_output testcase throws below warning: In tests/shell/stat+std_output.sh line 9: . $(dirname $0)/lib/stat_output.sh ^-----------^ SC2046 (warning): Quote this to prevent word splitting. In tests/shell/stat+std_output.sh line 32: local -i cnt=0 ^-^ SC2034 (warning): cnt appears unused. Verify use (or export if used externally). Fixed the warning by adding quotes to avoid word splitting and removed unused variable "cnt" at line 32. Signed-off-by:Athira Rajeev <atrajeev@linux.vnet.ibm.com> Acked-by:
-
Athira Rajeev authored
Running shellcheck on stat+csv_output.sh throws below warning: In tests/shell/stat+csv_output.sh line 9: . $(dirname $0)/lib/stat_output.sh ^-----------^ SC2046 (warning): Quote this to prevent word splitting. Fixed the warning by adding quotes to avoid word splitting. Signed-off-by: -
Kajol Jain authored
Running shellcheck on stat_output.sh throws below warning: In tests/shell/lib/stat_output.sh line 1: ^-- SC2148 (error): Tips depend on target shell and yours is unknown. Add a shebang or a 'shell' directive. Fixed the warning by adding shell directive. Signed-off-by:
Kajol Jain <kjain@linux.ibm.com> Acked-by:
-
Kajol Jain authored
Running shellcheck on thread_loop_check_tid_2.sh throws below warning: In tests/shell/coresight/thread_loop_check_tid_2.sh line 8: . $(dirname $0)/../lib/coresight.sh ^-----------^ SC2046 (warning): Quote this to prevent word splitting. Fixed the warning by adding quotes to avoid word splitting. Signed-off-by:
-
Athira Rajeev authored
Running shellcheck on record+zstd_comp_decomp.sh testcases throws below warning: In tests/shell/record+zstd_comp_decomp.sh line 16: $perf_tool record -o $trace_file $gflag -z -F 5000 -- \ ^---------^ SC2086 (info): Double quote to prevent globbing and word splitting. Did you mean: $perf_tool record -o "$trace_file" $gflag -z -F 5000 -- \ In tests/shell/record+zstd_comp_decomp.sh line 22: $perf_tool report -i $trace_file --header --stats | \ ^---------^ SC2086 (info): Double quote to prevent globbing and word splitting. Added double quote around file names to fix these shellcheck reported issues. Signed-off-by: -
Athira Rajeev authored
Running shellcheck on syscalltbl.sh generates below warning: In ./tools/perf/arch/x86/entry/syscalls/syscalltbl.sh line 27: while read nr abi name entry compat; do ^-^ SC2034 (warning): abi appears unused. Verify use (or export if used externally). ^----^ SC2034 (warning): compat appears unused. Verify use (or export if used externally). These variables are intentionally unused since they are needed to parse through the output. Use "_" as a prefix for these throw away variables. Signed-off-by: -
Athira Rajeev authored
Running shellcheck on x86_arch_prctl.sh generates below warning: In ./tools/perf/trace/beauty/x86_arch_prctl.sh line 10: local idx=$1 ^-------^ SC3043 (warning): In POSIX sh, 'local' is undefined. In ./tools/perf/trace/beauty/x86_arch_prctl.sh line 11: local prefix=$2 ^----------^ SC3043 (warning): In POSIX sh, 'local' is undefined. In ./tools/perf/trace/beauty/x86_arch_prctl.sh line 12: local first_entry=$3 ^---------------^ SC3043 (warning): In POSIX sh, 'local' is undefined. Fix this by removing local since these are variables used only in specific function Signed-off-by: -
Athira Rajeev authored
Running shellcheck in "lib/waiting.sh" generates below warning: In ./tools/perf/tests/shell/lib/waiting.sh line 1: # SPDX-License-Identifier: GPL-2.0 ^-- SC2148 (error): Tips depend on target shell and yours is unknown. Add a shebang or a 'shell' directive. Fix this by adding shebang in the beginning of the script. Signed-off-by:
-
Athira Rajeev authored
Running shellcheck on probe_vfs_getname fails with below warning: In ./tools/perf/tests/shell/lib/probe_vfs_getname.sh line 1: # Arnaldo Carvalho de Melo <acme@kernel.org>, 2017 ^-- SC2148 (error): Tips depend on target shell and yours is unknown. Add a shebang or a 'shell' directive. In ./tools/perf/tests/shell/lib/probe_vfs_getname.sh line 14: local verbose=$1 ^-----------^ SC3043 (warning): In POSIX sh, 'local' is undefined. Fix this: - by adding shebang in the beginning of the file and - rename variable verbose to "add_probe_verbose" after removing local Signed-off-by: -
Athira Rajeev authored
Fix the shellcheck warnings for unroll_loop_thread_10.sh Add quotes to prevent word splitting which are caused by unquoted command expansions. Signed-off-by:
-
Athira Rajeev authored
Fix the shellcheck warnings for thread_loop_check_tid_10.sh In ./tools/perf/tests/shell/coresight/thread_loop_check_tid_10.sh line 8: . $(dirname $0)/../lib/coresight.sh ^-----------^ SC2046 (warning): Quote this to prevent word splitting. Add quotes to prevent word splitting which are caused by unquoted command expansions. Signed-off-by: -
Kajol Jain authored
Running shellcheck on check-headers.sh generates below warning: In check-headers.sh line 126: check_2 "tools/$file" "$file" $* ^-- SC2048 (warning): Use "$@" (with quotes) to prevent whitespace problems. In check-headers.sh line 134: check_2 "tools/perf/trace/beauty/$file" "$file" $* ^-- SC2048 (warning): Use "$@" (with quotes) to prevent whitespace problems. In check-headers.sh line 186: cd tools/perf ^-----------^ SC2164 (warning): Use 'cd ... || exit' or 'cd ... || return' in case cd fails. Fixed the warnings by: - Using "$@" instead of $* - Adding exit condition with cd command Signed-off-by: -
Kajol Jain authored
Running shellcheck on arch_errno_names.sh generates below warning: In arch_errno_names.sh line 20: local arch="$1" ^--------^ SC3043 (warning): In POSIX sh, 'local' is undefined. ...... In arch_errno_names.sh line 61: local arch ^--------^ SC3043 (warning): In POSIX sh, 'local' is undefined. In arch_errno_names.sh line 67: printf '\t\treturn errno_to_name__%s(err);\n' $(arch_string "$arch") ^--------------------^ SC2046 (warning): Quote this to prevent word splitting. In arch_errno_names.sh line 69: printf '\treturn errno_to_name__%s(err);\n' $(arch_string "$default") ^-----------------------^ SC2046 (warning): Quote this to prevent word splitting. Fixed the warnings by: - Fixing shellcheck warnings for local usage, by removing local from the variable names - Adding quotes to avoid word splitting Signed-off-by: -
Kajol Jain authored
Running shellcheck on probe.sh throws below warning: In lib/probe.sh line 1: ^-- SC2148 (error): Tips depend on target shell and yours is unknown. Add a shebang or a 'shell' directive. Fixed the warnings by adding shell directive. Signed-off-by:
-
Kajol Jain authored
Running shellcheck on memcpy_thread_16k_10.sh throws below warning: In memcpy_thread_16k_10.sh line 8: . $(dirname $0)/../lib/coresight.sh ^-----------^ SC2046 (warning): Quote this to prevent word splitting. Fixed the warning by adding quotes to avoid word splitting. ShellCheck result with patch: # shellcheck -S warning coresight/memcpy_thread_16k_10.sh # Signed-off-by: -
Kajol Jain authored
Running shellcheck on asm_pure_loop.sh throws below warning: In coresight/asm_pure_loop.sh line 8: . $(dirname $0)/../lib/coresight.sh ^-----------^ SC2046 (warning): Quote this to prevent word splitting. Fixed the warning by adding quotes to avoid word splitting. ShellCheck result with patch: # shellcheck -S warning coresight/asm_pure_loop.sh # Signed-off-by: -
Athira Rajeev authored
Running shellcheck on stat+shadow_stat.sh generates below warning: In tests/shell/stat+shadow_stat.sh line 48: while read cpu num evt hash ipc rest ^--^ SC2034 (warning): hash appears unused. Verify use (or export if used externally). This variable is intentionally unused since it is needed to parse through the output. Use "_" as a prefix for this throw away variable. Signed-off-by: -
Athira Rajeev authored
Running shellcheck on stat_bpf_counter.sh generates below warning: In tests/shell/stat_bpf_counters.sh line 34: if [ "$base_cycles" == "<not" ]; then ^-- SC3014 (warning): In POSIX sh, == in place of = is undefined. In tests/shell/stat_bpf_counters.sh line 39: if [ "$bpf_cycles" == "<not" ]; then ^-- SC3014 (warning): In POSIX sh, == in place of = is undefined. Fix this by using "=" instead of "==" to work for all shells. Signed-off-by: -
Athira Rajeev authored
Running shellcheck on test_perf_data_converter_json.sh throws below warning: In tests/shell/test_perf_data_converter_json.sh line 42: if [ $(cat "${result}" | wc -l) -gt "0" ] ; then ^------------------------^ SC2046 (warning): Quote this to prevent word splitting. Fixed the warning by adding quotes to avoid word splitting. ShellCheck result with patch: # shellcheck -S warning test_perf_data_converter_json.sh # Signed-off-by:
-
