Skip to content

Ignore unknown addresses in sanitizer reports #526

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom
Open
wants to merge 1 commit into
base: master
Choose a base branch
from
Open

Ignore unknown addresses in sanitizer reports #526

wants to merge 1 commit into from

Conversation

kyakdan
Copy link
Contributor

@kyakdan kyakdan commented Apr 19, 2025

Sometimes, the sanitizer logs contain crash addresses marked as unknown. Here is an example ASan report due to an abort signal (ABRT):
==528025==ERROR: AddressSanitizer: ABRT on unknown address 0x... (pc ...

These addresses can differ for the same bug (PC and stack trace addresses are the same), which breaks error deduplication. This PR addresses this issue by ignoring crash addresses from the sanitizer reports if the address is marked as unknown.

The following are two examples resulting from an undefined behavior bug (int n = 32; n <<= 32;). Each time this bug is triggered, we get a different unknown crash address, which seems to be increased by one each time.

=================================================================
==528024==ERROR: AddressSanitizer: ABRT on unknown address 0x03e800080e98 (pc 0x7ffff689eb2c bp 0x7fffffffd620 sp 0x7fffffffd5e0 T0)
    #0 0x7ffff689eb2c in __pthread_kill_implementation nptl/pthread_kill.c:44
    #1 0x7ffff689eb2c in __pthread_kill_internal nptl/pthread_kill.c:78
    #2 0x7ffff689eb2c in __GI___pthread_kill nptl/pthread_kill.c:89
    #3 0x7ffff684527d in __GI_raise ../sysdeps/posix/raise.c:26
    #4 0x7ffff68288fe in __GI_abort stdlib/abort.c:79
    #5 0x7ffff6c20d2f in __sanitizer::Abort() ../../../../src/libsanitizer/sanitizer_common/sanitizer_posix_libcdep.cpp:143
    #6 0x7ffff6c319cc in __sanitizer::Die() ../../../../src/libsanitizer/sanitizer_common/sanitizer_termination.cpp:58
    #7 0x7ffff6c0e6f7 in __ubsan_handle_shift_out_of_bounds_abort ../../../../src/libsanitizer/ubsan/ubsan_handlers.cpp:378
    #8 0x5555555592f6 in LLVMFuzzerTestOneInputNoReturn(unsigned char const*, unsigned long) (/home/khaled/code/CodeIntelligenceTesting/cifuzz/test/testdata/cmake-default/a.out+0x52f6) (BuildId: eba2deebbd9f5cb707cb20371a0bfc127560614e)
    #9 0x555555559055 in LLVMFuzzerTestOneInput (/home/khaled/code/CodeIntelligenceTesting/cifuzz/test/testdata/cmake-default/a.out+0x5055) (BuildId: eba2deebbd9f5cb707cb20371a0bfc127560614e)
    #10 0x55555556415e in HonggfuzzMain (/home/khaled/code/CodeIntelligenceTesting/cifuzz/test/testdata/cmake-default/a.out+0x1015e) (BuildId: eba2deebbd9f5cb707cb20371a0bfc127560614e)
    #11 0x7ffff682a1c9 in __libc_start_call_main ../sysdeps/nptl/libc_start_call_main.h:58
    #12 0x7ffff682a28a in __libc_start_main_impl ../csu/libc-start.c:360
    #13 0x555555558f64 in _start (/home/khaled/code/CodeIntelligenceTesting/cifuzz/test/testdata/cmake-default/a.out+0x4f64) (BuildId: eba2deebbd9f5cb707cb20371a0bfc127560614e)

AddressSanitizer can not provide additional info.
SUMMARY: AddressSanitizer: ABRT nptl/pthread_kill.c:44 in __pthread_kill_implementation
==528024==ABORTING

=================================================================
==528025==ERROR: AddressSanitizer: ABRT on unknown address 0x03e800080e99 (pc 0x7ffff689eb2c bp 0x7fffffffd620 sp 0x7fffffffd5e0 T0)
    #0 0x7ffff689eb2c in __pthread_kill_implementation nptl/pthread_kill.c:44
    #1 0x7ffff689eb2c in __pthread_kill_internal nptl/pthread_kill.c:78
    #2 0x7ffff689eb2c in __GI___pthread_kill nptl/pthread_kill.c:89
    #3 0x7ffff684527d in __GI_raise ../sysdeps/posix/raise.c:26
    #4 0x7ffff68288fe in __GI_abort stdlib/abort.c:79
    #5 0x7ffff6c20d2f in __sanitizer::Abort() ../../../../src/libsanitizer/sanitizer_common/sanitizer_posix_libcdep.cpp:143
    #6 0x7ffff6c319cc in __sanitizer::Die() ../../../../src/libsanitizer/sanitizer_common/sanitizer_termination.cpp:58
    #7 0x7ffff6c0e6f7 in __ubsan_handle_shift_out_of_bounds_abort ../../../../src/libsanitizer/ubsan/ubsan_handlers.cpp:378
    #8 0x5555555592f6 in LLVMFuzzerTestOneInputNoReturn(unsigned char const*, unsigned long) (/home/khaled/code/CodeIntelligenceTesting/cifuzz/test/testdata/cmake-default/a.out+0x52f6) (BuildId: eba2deebbd9f5cb707cb20371a0bfc127560614e)
    #9 0x555555559055 in LLVMFuzzerTestOneInput (/home/khaled/code/CodeIntelligenceTesting/cifuzz/test/testdata/cmake-default/a.out+0x5055) (BuildId: eba2deebbd9f5cb707cb20371a0bfc127560614e)
    #10 0x55555556415e in HonggfuzzMain (/home/khaled/code/CodeIntelligenceTesting/cifuzz/test/testdata/cmake-default/a.out+0x1015e) (BuildId: eba2deebbd9f5cb707cb20371a0bfc127560614e)
    #11 0x7ffff682a1c9 in __libc_start_call_main ../sysdeps/nptl/libc_start_call_main.h:58
    #12 0x7ffff682a28a in __libc_start_main_impl ../csu/libc-start.c:360
    #13 0x555555558f64 in _start (/home/khaled/code/CodeIntelligenceTesting/cifuzz/test/testdata/cmake-default/a.out+0x4f64) (BuildId: eba2deebbd9f5cb707cb20371a0bfc127560614e)

AddressSanitizer can not provide additional info.
SUMMARY: AddressSanitizer: ABRT nptl/pthread_kill.c:44 in __pthread_kill_implementation
==528025==ABORTING

@kyakdan
sanitizers: ignore unknown addresses in reports
28a553a 
In some cases, the sanitizer logs contains addresses marked
as unknown. Here is an exampl ASan report due to an abort signal (ABRT):
 ==528025==ERROR: AddressSanitizer: ABRT on unknown address 0x03e800080e99 (pc ...

These addresses can be different for the same bug (PC and stacktrace are
the same) and as a result breaks error deduplication. Here, we ignore
addresses from the sanitizer reports if the address is markes as
unknown.
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Reviewers
No reviews
Assignees
No one assigned
Labels
None yet
Projects
None yet
Milestone
No milestone
Development

Successfully merging this pull request may close these issues.
1 participant
@kyakdan