Alexey Serbin has posted comments on this change. ( 
http://gerrit.cloudera.org:8080/17626 )

Change subject: [thirdparty] fix building crcutil with GCC 10
......................................................................


Patch Set 2:

> LGTM. I don't quite understand the rebalancing test failure, but it
 > doesn't look related.

The rebalancing test failed due to SIGSEGV from within libtcmalloc.  Most 
likely reason cited to be double-free or other application programming error 
related to concurrent freeing the same entity: 
https://github.com/gperftools/gperftools/issues/125

However, the stack trace doesn't provides any hints on what exact memory might 
be twice-deallocated, i.e. I don't see an exact place pointed at.  The stack 
hints on something created during static initialization phase, I guess:

PC: @     0x7f59b21cddfb tcmalloc::ThreadCache::ReleaseToCentralCache()
*** SIGSEGV (@0x10000555ae1) received by PID 3837 (TID 0x7f59ab2838c0) from PID 
5593825; stack trace: ***
    @     0x7f59b2d26980 (unknown)
    @     0x7f59b21cddfb tcmalloc::ThreadCache::ReleaseToCentralCache()
    @     0x7f59b21ce185 tcmalloc::ThreadCache::ListTooLong()
    @     0x555ae00fe3ed std::unordered_map<>::~unordered_map()
    @     0x7f59b0aec161 (unknown)
    @     0x7f59b0aec25a exit
    @     0x7f59b0acabfe __libc_start_main
    @     0x555ae0055e4a _start

I'm not going to dig any further, but clearly this failure isn't related to 
this patch.


--
To view, visit http://gerrit.cloudera.org:8080/17626
To unsubscribe, visit http://gerrit.cloudera.org:8080/settings

Gerrit-Project: kudu
Gerrit-Branch: master
Gerrit-MessageType: comment
Gerrit-Change-Id: Ibaba794ce00eb64a29e8038a8a2f4883f35c975a
Gerrit-Change-Number: 17626
Gerrit-PatchSet: 2
Gerrit-Owner: Alexey Serbin <[email protected]>
Gerrit-Reviewer: Alexey Serbin <[email protected]>
Gerrit-Reviewer: Andrew Wong <[email protected]>
Gerrit-Reviewer: Grant Henke <[email protected]>
Gerrit-Reviewer: Kudu Jenkins (120)
Gerrit-Comment-Date: Wed, 07 Jul 2021 23:09:16 +0000
Gerrit-HasComments: No

Reply via email to