Re: [linus:master] [perf test] c9cd0c7e52: perf-sanity-tests.python_profiling_with_jitdump.fail
From: Ian Rogers
Date: Mon Mar 02 2026 - 17:08:46 EST
On Fri, Feb 27, 2026 at 1:39 PM Namhyung Kim <namhyung@xxxxxxxxxx> wrote:
>
> On Fri, Feb 27, 2026 at 04:42:59PM +0800, kernel test robot wrote:
> >
> >
> > Hello,
> >
> >
> > we failed to run this new added test. not sure if we miss any dependencies?
> >
> >
> > below is full report FYI.
> >
> >
> > kernel test robot noticed "perf-sanity-tests.python_profiling_with_jitdump.fail" on:
> >
> > commit: c9cd0c7e529e80e9be79867d2ebb874f67dbc35e ("perf test: Add python JIT dump test")
> > https://git.kernel.org/cgit/linux/kernel/git/torvalds/linux.git master
> >
> > in testcase: perf-sanity-tests
> > version:
> > with following parameters:
> >
> > perf_compiler: clang
> > group: group-02
> >
> >
> > config: x86_64-rhel-9.4-bpf
> > compiler: gcc-14
> > test machine: 16 threads Intel(R) Core(TM) i7-13620H (Raptor Lake) with 32G memory
> >
> > (please refer to attached dmesg/kmsg for entire log/backtrace)
> >
> >
> > If you fix the issue in a separate patch/commit (i.e. not just a new version of
> > the same patch/commit), kindly add following tags
> > | Reported-by: kernel test robot <oliver.sang@xxxxxxxxx>
> > | Closes: https://lore.kernel.org/oe-lkp/202602270932.50ed0a9f-lkp@xxxxxxxxx
>
> Looks like a reference leak to map during libdw unwinding.
>
> Ian, can you please take a look?
>
> Thanks,
> Namhyung
>
> >
> >
> > 2026-01-26 15:25:04 sudo ASAN_OPTIONS=fast_unwind_on_malloc=0 /usr/src/linux-perf-x86_64-rhel-9.4-bpf-c9cd0c7e529e80e9be79867d2ebb874f67dbc35e/tools/perf/perf test 84 -v
> > 84: python profiling with jitdump : Running (1 active)
> > --- start ---
> > test child forked, pid 12390
> > Run python with -Xperf_jit
> > [ perf record: Woken up 5 times to write data ]
> > =================================================================
> > ==12400==ERROR: AddressSanitizer: heap-use-after-free on address 0x50200002ba30 at pc 0x55c306993987 bp 0x7ffd98998720 sp 0x7ffd98998718
> > READ of size 8 at 0x50200002ba30 thread T0
> > #0 0x55c306993986 in map__start util/map.h:54
> > #1 0x55c306993cd0 in map__dso_map_ip util/map.h:110
> > #2 0x55c3069b1ecb in unwind_entry util/machine.c:2950
> > #3 0x55c306e4197b in unwind__get_entries util/unwind-libdw.c:321
> > #4 0x55c3069b2398 in thread__resolve_callchain_unwind util/machine.c:2976
> > #5 0x55c3069b24ec in __thread__resolve_callchain util/machine.c:3003
> > #6 0x55c306987e78 in sample__for_each_callchain_node util/callchain.c:1808
> > #7 0x55c3067d4ece in build_id__mark_dso_hit util/build-id.c:77
> > #8 0x55c3062f7004 in process_sample_event /usr/src/perf_selftests-x86_64-rhel-9.4-bpf-c9cd0c7e529e80e9be79867d2ebb874f67dbc35e/tools/perf/builtin-record.c:1504
> > #9 0x55c3069dbec3 in evlist__deliver_sample util/session.c:1259
> > #10 0x55c3069dc425 in machines__deliver_event util/session.c:1299
> > #11 0x55c3069de009 in perf_session__deliver_event util/session.c:1386
> > #12 0x55c3069cb296 in ordered_events__deliver_event util/session.c:134
> > #13 0x55c306a06aac in do_flush util/ordered-events.c:245
> > #14 0x55c306a07951 in __ordered_events__flush util/ordered-events.c:324
> > #15 0x55c306a07e62 in ordered_events__flush util/ordered-events.c:342
> > #16 0x55c3069d6863 in perf_event__process_finished_round util/session.c:786
> > #17 0x55c3069deb7e in perf_session__process_user_event util/session.c:1446
> > #18 0x55c3069e190e in perf_session__process_event util/session.c:1668
> > #19 0x55c3069e9198 in process_simple util/session.c:2259
> > #20 0x55c3069e8158 in reader__read_event util/session.c:2188
> > #21 0x55c3069e8afe in reader__process_events util/session.c:2237
> > #22 0x55c3069e9987 in __perf_session__process_events util/session.c:2282
> > #23 0x55c3069eca43 in perf_session__process_events util/session.c:2446
> > #24 0x55c3062f7328 in process_buildids /usr/src/perf_selftests-x86_64-rhel-9.4-bpf-c9cd0c7e529e80e9be79867d2ebb874f67dbc35e/tools/perf/builtin-record.c:1534
> > #25 0x55c3062fbdf0 in record__finish_output /usr/src/perf_selftests-x86_64-rhel-9.4-bpf-c9cd0c7e529e80e9be79867d2ebb874f67dbc35e/tools/perf/builtin-record.c:1858
> > #26 0x55c306309ebf in __cmd_record /usr/src/perf_selftests-x86_64-rhel-9.4-bpf-c9cd0c7e529e80e9be79867d2ebb874f67dbc35e/tools/perf/builtin-record.c:2930
> > #27 0x55c306317a8f in cmd_record /usr/src/perf_selftests-x86_64-rhel-9.4-bpf-c9cd0c7e529e80e9be79867d2ebb874f67dbc35e/tools/perf/builtin-record.c:4417
> > #28 0x55c30652bf7b in run_builtin /usr/src/perf_selftests-x86_64-rhel-9.4-bpf-c9cd0c7e529e80e9be79867d2ebb874f67dbc35e/tools/perf/perf.c:349
> > #29 0x55c30652c86c in handle_internal_command /usr/src/perf_selftests-x86_64-rhel-9.4-bpf-c9cd0c7e529e80e9be79867d2ebb874f67dbc35e/tools/perf/perf.c:401
> > #30 0x55c30652cdc2 in run_argv /usr/src/perf_selftests-x86_64-rhel-9.4-bpf-c9cd0c7e529e80e9be79867d2ebb874f67dbc35e/tools/perf/perf.c:445
> > #31 0x55c30652d569 in main /usr/src/perf_selftests-x86_64-rhel-9.4-bpf-c9cd0c7e529e80e9be79867d2ebb874f67dbc35e/tools/perf/perf.c:553
> > #32 0x7f026b873ca7 (/lib/x86_64-linux-gnu/libc.so.6+0x29ca7) (BuildId: def5460e3cee00bfee25b429c97bcc4853e5b3a8)
> > #33 0x7f026b873d64 in __libc_start_main (/lib/x86_64-linux-gnu/libc.so.6+0x29d64) (BuildId: def5460e3cee00bfee25b429c97bcc4853e5b3a8)
> > #34 0x55c306293210 in _start (/usr/src/perf_selftests-x86_64-rhel-9.4-bpf-c9cd0c7e529e80e9be79867d2ebb874f67dbc35e/tools/perf/perf+0x1023210) (BuildId: 47c1eede4a6cf8e0c379d2104e8b035638daa1b4)
> >
> > 0x50200002ba30 is located 0 bytes inside of 8-byte region [0x50200002ba30,0x50200002ba38)
> > freed by thread T0 here:
> > #0 0x7f0275b3d8f8 in free ../../../../src/libsanitizer/asan/asan_malloc_linux.cpp:52
> > #1 0x55c3069ba258 in map__put util/map.c:312
> > #2 0x55c30679b7f8 in __map__zput util/map.h:198
> > #3 0x55c30679c0df in addr_location__exit util/addr_location.c:31
> > #4 0x55c306e3f7fe in entry util/unwind-libdw.c:147
This free is happening here (note I've set the commit to be that in
the report here):
https://web.git.kernel.org/pub/scm/linux/kernel/git/perf/perf-tools-next.git/tree/tools/perf/util/unwind-libdw.c?h=perf-tools-next&id=c9cd0c7e529e80e9be79867d2ebb874f67dbc35e#n147
```
static int entry(u64 ip, struct unwind_info *ui)
{
struct unwind_entry *e = &ui->entries[ui->idx++];
struct addr_location al;
addr_location__init(&al);
if (__report_module(&al, ip, ui)) {
addr_location__exit(&al);
return -1;
}
e->ip = ip;
e->ms.maps = al.maps;
e->ms.map = al.map;
e->ms.sym = al.sym;
pr_debug("unwind: %s:ip = 0x%" PRIx64 " (0x%" PRIx64 ")\n",
al.sym ? al.sym->name : "''",
ip,
al.map ? map__map_ip(al.map, ip) : (u64) 0);
addr_location__exit(&al); // <-- free happens here
return 0;
}
```
This was fixed by the first of my unwind-libdw changes:
https://web.git.kernel.org/pub/scm/linux/kernel/git/perf/perf-tools-next.git/commit/tools/perf/util/unwind-libdw.c?h=perf-tools-next&id=f815fc0c66e777c727689666cfb46b8d461c2f99
That is the new test exposed a latent bug, which we've already fixed,
with a "fixes" tag for backports.
Thanks,
Ian
> > #5 0x55c306e40b1b in frame_callback util/unwind-libdw.c:259
> > #6 0x7f02758daae2 in dwfl_thread_getframes (/lib/x86_64-linux-gnu/libdw.so.1+0x78ae2) (BuildId: bafe9cb5bbbe6cee778d0d137730ea5c223d553e)
> > #7 0x7f02758dae53 in dwfl_getthread_frames (/lib/x86_64-linux-gnu/libdw.so.1+0x78e53) (BuildId: bafe9cb5bbbe6cee778d0d137730ea5c223d553e)
> > #8 0x55c306e414af in unwind__get_entries util/unwind-libdw.c:307
> > #9 0x55c3069b2398 in thread__resolve_callchain_unwind util/machine.c:2976
> > #10 0x55c3069b24ec in __thread__resolve_callchain util/machine.c:3003
> > #11 0x55c306987e78 in sample__for_each_callchain_node util/callchain.c:1808
> > #12 0x55c3067d4ece in build_id__mark_dso_hit util/build-id.c:77
> > #13 0x55c3062f7004 in process_sample_event /usr/src/perf_selftests-x86_64-rhel-9.4-bpf-c9cd0c7e529e80e9be79867d2ebb874f67dbc35e/tools/perf/builtin-record.c:1504
> > #14 0x55c3069dbec3 in evlist__deliver_sample util/session.c:1259
> > #15 0x55c3069dc425 in machines__deliver_event util/session.c:1299
> > #16 0x55c3069de009 in perf_session__deliver_event util/session.c:1386
> > #17 0x55c3069cb296 in ordered_events__deliver_event util/session.c:134
> > #18 0x55c306a06aac in do_flush util/ordered-events.c:245
> > #19 0x55c306a07951 in __ordered_events__flush util/ordered-events.c:324
> > #20 0x55c306a07e62 in ordered_events__flush util/ordered-events.c:342
> > #21 0x55c3069d6863 in perf_event__process_finished_round util/session.c:786
> > #22 0x55c3069deb7e in perf_session__process_user_event util/session.c:1446
> > #23 0x55c3069e190e in perf_session__process_event util/session.c:1668
> > #24 0x55c3069e9198 in process_simple util/session.c:2259
> > #25 0x55c3069e8158 in reader__read_event util/session.c:2188
> > #26 0x55c3069e8afe in reader__process_events util/session.c:2237
> > #27 0x55c3069e9987 in __perf_session__process_events util/session.c:2282
> >
> > previously allocated by thread T0 here:
> > #0 0x7f0275b3ec57 in malloc ../../../../src/libsanitizer/asan/asan_malloc_linux.cpp:69
> > #1 0x55c3069bd747 in map__get util/map.h:188
> > #2 0x55c3069c72c4 in maps__find util/maps.c:1112
> > #3 0x55c30683b979 in thread__find_map util/event.c:687
> > #4 0x55c30683be37 in thread__find_symbol util/event.c:723
> > #5 0x55c306e3e7c3 in __report_module util/unwind-libdw.c:54
> > #6 0x55c306e3f428 in entry util/unwind-libdw.c:133
> > #7 0x55c306e40b1b in frame_callback util/unwind-libdw.c:259
> > #8 0x7f02758daae2 in dwfl_thread_getframes (/lib/x86_64-linux-gnu/libdw.so.1+0x78ae2) (BuildId: bafe9cb5bbbe6cee778d0d137730ea5c223d553e)
> > #9 0x7f02758dae53 in dwfl_getthread_frames (/lib/x86_64-linux-gnu/libdw.so.1+0x78e53) (BuildId: bafe9cb5bbbe6cee778d0d137730ea5c223d553e)
> > #10 0x55c306e414af in unwind__get_entries util/unwind-libdw.c:307
> > #11 0x55c3069b2398 in thread__resolve_callchain_unwind util/machine.c:2976
> > #12 0x55c3069b24ec in __thread__resolve_callchain util/machine.c:3003
> > #13 0x55c306987e78 in sample__for_each_callchain_node util/callchain.c:1808
> > #14 0x55c3067d4ece in build_id__mark_dso_hit util/build-id.c:77
> > #15 0x55c3062f7004 in process_sample_event /usr/src/perf_selftests-x86_64-rhel-9.4-bpf-c9cd0c7e529e80e9be79867d2ebb874f67dbc35e/tools/perf/builtin-record.c:1504
> > #16 0x55c3069dbec3 in evlist__deliver_sample util/session.c:1259
> > #17 0x55c3069dc425 in machines__deliver_event util/session.c:1299
> > #18 0x55c3069de009 in perf_session__deliver_event util/session.c:1386
> > #19 0x55c3069cb296 in ordered_events__deliver_event util/session.c:134
> > #20 0x55c306a06aac in do_flush util/ordered-events.c:245
> > #21 0x55c306a07951 in __ordered_events__flush util/ordered-events.c:324
> > #22 0x55c306a07e62 in ordered_events__flush util/ordered-events.c:342
> > #23 0x55c3069d6863 in perf_event__process_finished_round util/session.c:786
> > #24 0x55c3069deb7e in perf_session__process_user_event util/session.c:1446
> > #25 0x55c3069e190e in perf_session__process_event util/session.c:1668
> > #26 0x55c3069e9198 in process_simple util/session.c:2259
> > #27 0x55c3069e8158 in reader__read_event util/session.c:2188
> >
> > SUMMARY: AddressSanitizer: heap-use-after-free util/map.h:54 in map__start
> > Shadow bytes around the buggy address:
> > 0x50200002b780: fa fa fd fa fa fa fd fd fa fa 00 06 fa fa 00 fa
> > 0x50200002b800: fa fa 00 fa fa fa fd fa fa fa 00 fa fa fa 00 fa
> > 0x50200002b880: fa fa 00 fa fa fa fd fa fa fa fd fa fa fa fd fa
> > 0x50200002b900: fa fa fd fa fa fa fd fa fa fa fd fa fa fa fd fa
> > 0x50200002b980: fa fa fd fa fa fa fd fa fa fa fd fa fa fa fd fa
> > =>0x50200002ba00: fa fa fd fa fa fa[fd]fa fa fa fd fa fa fa fd fa
> > 0x50200002ba80: fa fa fd fa fa fa fd fa fa fa fd fa fa fa fd fa
> > 0x50200002bb00: fa fa fd fa fa fa fd fa fa fa fd fa fa fa fd fa
> > 0x50200002bb80: fa fa fd fa fa fa fd fa fa fa fd fa fa fa fd fa
> > 0x50200002bc00: fa fa fd fa fa fa fd fa fa fa fd fa fa fa fd fa
> > 0x50200002bc80: fa fa fa fa fa fa fa fa fa fa fa fa fa fa fa fa
> > Shadow byte legend (one shadow byte represents 8 application bytes):
> > Addressable: 00
> > Partially addressable: 01 02 03 04 05 06 07
> > Heap left redzone: fa
> > Freed heap region: fd
> > Stack left redzone: f1
> > Stack mid redzone: f2
> > Stack right redzone: f3
> > Stack after return: f5
> > Stack use after scope: f8
> > Global redzone: f9
> > Global init order: f6
> > Poisoned by user: f7
> > Container overflow: fc
> > Array cookie: ac
> > Intra object redzone: bb
> > ASan internal: fe
> > Left alloca redzone: ca
> > Right alloca redzone: cb
> > ==12400==ABORTING
> > Error:
> > failed to process sample
> > Generate JIT-ed DSOs using perf inject
> > WARNING: The /tmp/__perf_test.perf.data.W1jHYt file's data size field is 0 which is unexpected.
> > Was the 'perf record' command properly terminated?
> > Add JIT-ed DSOs to the build-ID cache
> > Couldn't add /tmp/jitted--*.so: No such file or directory
> > Check the symbol containing the function/module name
> > zero-sized data (/tmp/__perf_test.perf.data.W1jHYt.jit), nothing to do!
> > Found 0 matching lines
> > Remove JIT-ed DSOs from the build-ID cache
> > Cleaning up files...
> > ---- end(-1) ----
> > 84: python profiling with jitdump : FAILED!
> >
> >
> >
> > The kernel config and materials to reproduce are available at:
> > https://download.01.org/0day-ci/archive/20260227/202602270932.50ed0a9f-lkp@xxxxxxxxx
> >
> >
> >
> > --
> > 0-DAY CI Kernel Test Service
> > https://github.com/intel/lkp-tests/wiki
> >