pmem / issues

Old issues repo for PMDK.
http://pmem.io
13 stars 7 forks source link

Test: obj_pmalloc_rand_mt/TEST0 fail #1078

Closed Greg091 closed 5 years ago

Greg091 commented 5 years ago

ISSUE:

Environment Information

Please provide a reproduction of the bug:

$ ./RUNTESTS obj_pmalloc_rand_mt -t all
obj_pmalloc_rand_mt/TEST0: SETUP (all/pmem/nondebug/helgrind)
obj_pmalloc_rand_mt/TEST0 failed with Valgrind. See helgrind0.log. Last 20 lines below.
obj_pmalloc_rand_mt/TEST0 helgrind0.log ==18634==    by 0x58E5407: __nptl_deallocate_tsd.part.5 (pthread_create.c:300)
obj_pmalloc_rand_mt/TEST0 helgrind0.log ==18634==    by 0x58E681A: __nptl_deallocate_tsd (futex-internal.h:200)
obj_pmalloc_rand_mt/TEST0 helgrind0.log ==18634==    by 0x58E681A: start_thread (pthread_create.c:473)
obj_pmalloc_rand_mt/TEST0 helgrind0.log ==18634==    by 0x5C1F88E: clone (clone.S:95)
obj_pmalloc_rand_mt/TEST0 helgrind0.log ==18634==  Address 0x6f42ab8 is 40 bytes inside a block of size 3,147,504 alloc'd
obj_pmalloc_rand_mt/TEST0 helgrind0.log ==18634==    at 0x4C30F7B: malloc (vg_replace_malloc.c:299)
obj_pmalloc_rand_mt/TEST0 helgrind0.log ==18634==    by 0x4E5B4FB: heap_boot (in /home/greg/repos/pmdk/src/nondebug/libpmemobj.so.1.0.0)
obj_pmalloc_rand_mt/TEST0 helgrind0.log ==18634==    by 0x4E68E77: pmalloc_boot (in /home/greg/repos/pmdk/src/nondebug/libpmemobj.so.1.0.0)
obj_pmalloc_rand_mt/TEST0 helgrind0.log ==18634==    by 0x4E5CEC7: lane_recover_and_section_boot (in /home/greg/repos/pmdk/src/nondebug/libpmemobj.so.1.0.0)
obj_pmalloc_rand_mt/TEST0 helgrind0.log ==18634==    by 0x4E63826: obj_runtime_init_common (in /home/greg/repos/pmdk/src/nondebug/libpmemobj.so.1.0.0)
obj_pmalloc_rand_mt/TEST0 helgrind0.log ==18634==    by 0x4E63B0B: obj_runtime_init.constprop.10 (in /home/greg/repos/pmdk/src/nondebug/libpmemobj.so.1.0.0)
obj_pmalloc_rand_mt/TEST0 helgrind0.log ==18634==    by 0x4E65434: pmemobj_create (in /home/greg/repos/pmdk/src/nondebug/libpmemobj.so.1.0.0)
obj_pmalloc_rand_mt/TEST0 helgrind0.log ==18634==    by 0x10D456: main (obj_pmalloc_rand_mt.c:112)
obj_pmalloc_rand_mt/TEST0 helgrind0.log ==18634==  Block was alloc'd by thread #1
obj_pmalloc_rand_mt/TEST0 helgrind0.log ==18634== 
obj_pmalloc_rand_mt/TEST0 helgrind0.log ==18634== 
obj_pmalloc_rand_mt/TEST0 helgrind0.log ==18634== For counts of detected and suppressed errors, rerun with: -v
obj_pmalloc_rand_mt/TEST0 helgrind0.log ==18634== Use --history-level=approx or =none to gain increased speed, at
obj_pmalloc_rand_mt/TEST0 helgrind0.log ==18634== the cost of reduced accuracy of conflicting-access information
obj_pmalloc_rand_mt/TEST0 helgrind0.log ==18634== ERROR SUMMARY: 26 errors from 3 contexts (suppressed: 2842 from 58)
RUNTESTS: stopping: obj_pmalloc_rand_mt/TEST0 failed, TEST=all FS=pmem BUILD=nondebug

How often bug is revealed: (always, often, rare): often

Actual behavior:

Test fails.

Expected behavior:

The test should pass.

Details

Additional information about Priority and Help Requested:

Are you willing to submit a pull request with a proposed change? (Yes, No)

Requested priority: (Showstopper, High, Medium, Low)

marcinslusarz commented 5 years ago

full log please

Greg091 commented 5 years ago

helgrind0.log pmem0.log pmemobj0.log

Greg091 commented 5 years ago

The test still fails :( Logs: out0.log trace0.log helgrind0.log

Greg091 commented 5 years ago

Verified on: 1.6-346-g95d96e4c0