ERROR: AddressSanitizer: SEGV on unknown address 0x000000809ca0

[Svelar]

Hi experts,

As you know, ceph crimson is using seastar framework, recently, I observed that sometimes ceph ARM ci reported,

Rate-limit: suppressed 137 backtraces on shard 0
...
...
Reactor stalled for 420 ms on shard 0. Backtrace: 0x6f87523 0xcd8d3cb 0xcd8c8eb 0xcb84273 0xcb7e823 0xcb7e2ef 0xcb7ec83 0xcb856f3 0x7db 0x74c8243 0x74c8003 0x74c7c73 0x74c62cf 0xca7fde7 0xca8019b 0xca7fb97 0xca7f4f7 0xca7e5eb 0xca7a7df 0xca43283 0xca42e47 0xca43927 0xca3f08f 0xca3ee23 0xca3f207 0xca6829f 0xca4fb9f 0xca3636f 0xca362cb 0x86c471b 0x86808eb 0x84ce38b 0x83a252b 0x83a2597 0x7f9efdf 0x7f9ebc3 0x7bddacb 0x773d35f 0x779145b 0x7791293 0x779106b 0x7790e1f 0x7790873 0x77906b7 0x7c15e9b 0x7c15bbf 0x7c1560b 0x7c153eb 0x7c15087 0xcba500f 0xcbad427 0xcbb0d3b 0xcbaef83 0xc943acf 0xc94157f 0xc94428b 0x760019b 0x75fff9f 0x75ffe3b 0x75ffac7 0x75ff8a7 0x75ff60f 0x75ff1b7 0xdbc6b 0x7d5c7 0xe5edb
AddressSanitizer:DEADLYSIGNAL
=================================================================
==2965808==ERROR: AddressSanitizer: SEGV on unknown address 0x000000809ca0 (pc 0xffff7fd0f8b0 bp 0xffff77aee550 sp 0xffff77aee550 T1)
==2965808==The signal is caused by a READ memory access.
    <empty stack>

AddressSanitizer can not provide additional info.
SUMMARY: AddressSanitizer: SEGV
Thread T1 created by T0 here:
    #0 0xaaaabf4216a8 in __interceptor_pthread_create (/home/jenkins-build/build/workspace/ceph-pull-requests-arm64/build/bin/unittest-seastore+0x6fb16a8) (BuildId: 6331e1e6842a4d603461a18cc08c47a8a1d72bfb)
    #1 0xffff7feabfbc in std::thread::_M_start_thread(std::unique_ptr<std::thread::_State, std::default_delete<std::thread::_State> >, void (*)()) (/lib/aarch64-linux-gnu/libstdc++.so.6+0xdbfbc) (BuildId: c1d2833c18219b32e8bcc5c5f00173dfa24daea2)
    #2 0xaaaabfa6e754 in std::thread::thread<SeastarRunner::init(int, char**)::'lambda'(), void>(SeastarRunner::init(int, char**)::'lambda'()&&) /usr/bin/../lib/gcc/aarch64-linux-gnu/11/../../../../include/c++/11/bits/std_thread.h:143:2
    #3 0xaaaabf92111c in SeastarRunner::init(int, char**) /home/jenkins-build/build/workspace/ceph-pull-requests-arm64/src/test/crimson/seastar_runner.h:41:14
    #4 0xaaaabf912b04 in main /home/jenkins-build/build/workspace/ceph-pull-requests-arm64/src/test/crimson/gtest_seastar.cc:41:47
    #5 0xffff7fb773f8 in __libc_start_call_main csu/../sysdeps/nptl/libc_start_call_main.h:58:16
    #6 0xffff7fb774c8 in __libc_start_main csu/../csu/libc-start.c:392:3
    #7 0xaaaabf3c142c in _start (/home/jenkins-build/build/workspace/ceph-pull-requests-arm64/build/bin/unittest-seastore+0x6f5142c) (BuildId: 6331e1e6842a4d603461a18cc08c47a8a1d72bfb)

==2965808==ABORTING

Any idea to get more useful info or any hypothesis here? Thanks~

Ref:
https://jenkins.ceph.com/job/ceph-pull-requests-arm64/58814/consoleFull#11047942376733401c-e9d0-4737-9832-6594c5da0afa