compiler, libraries, packaging and other related tools version(s): TBD
Please provide a reproduction of the bug:
$ ./RUNTESTS.py pmem2_integration -u 39 -g byte --force-enable pmemcheck -b debug
pmem2_integration/TEST39: SETUP (medium/debug/pmemcheck/byte)
Last 16 lines of /home/jmichal/work/pmdk/src/test/pmem2_integration/pmemcheck39.log below (whole file has 16 lines):
==1998050== pmemcheck-1.0, a simple persistent store checker
==1998050== Copyright (c) 2014-2020, Intel Corporation
==1998050== Using Valgrind-3.20.0 and LibVEX; rerun with -h for copyright info
==1998050== Command: /home/jmichal/work/pmdk/src/test/pmem2_integration/pmem2_integration test_map_from_existing /tmp/pmem2_integration_39/testfile
==1998050== Parent PID: 1998038
==1998050==
==1998050==
==1998050== Number of stores not made persistent: 1
==1998050== Stores not made persistent properly:
==1998050== [0] at 0x401481A: _dl_fixup (dl-machine.h:239)
==1998050== by 0x4001C3D: _dl_runtime_resolve_xsave (dl-trampoline.h:126)
==1998050== by 0x402883: TEST_CASE_PROCESS (unittest.h:700)
==1998050== by 0x405A7A: main (pmem2_integration.c:1030)
==1998050== Address: 0x60f0c8 size: 8 state: DIRTY
==1998050== Total memory not made persistent: 8
==1998050== ERROR SUMMARY: 1 errors
How often bug is revealed: (always, often, rare): always
ISSUE: Sporadic:
pmem2_integration -u 39 --force-enable pmemcheck
Environment Information
Please provide a reproduction of the bug:
How often bug is revealed: (always, often, rare): always
But it seems that not all users experience this issue: https://github.com/pmem/pmdk/pull/5605#pullrequestreview-1415567258
Actual behavior:
Valgrind/pmemcheck finds DIRTY bytes.
Expected behavior:
Pass.
Additional information about Priority and Help Requested:
Requested priority: (Showstopper, High, Medium, Low) Low