LGTM

Yes, it looks good. I downloaded the patch and ran the test - had some doubts about 'otool -L %t |grep -q "dynamic.dylib"' - but verified with other ASAN tests binaries and convinced myself that it's the right thing to do.

Thanks Arthur, was going to do that but you beat me to it

s/lean/lsan/ above. My spell checker is driving me nuts.

This test was failing before w. Tsan and Ubsan, but these failures disappeared, only failures were lean. I'd guess that just disabling lean will leave things clean, but it apparently didn't.

LGTM

Vitaly - what is the change which caused his test to become broken? Our bots also show failures in symbolize_pc.cpp tests like this:

Yes, it makes sense sense to me. Thanks for doing that.

@kubamracek got it. I missed the place where the COMMON_INTERCEPTOR_LIBRARY_LOADED macro is actually used (I believe that's the only place). It all looks good then.

What ScopedIgnoreInterceptors does, is:
..

cur_thread()->ignore_interceptors++;

..

LGTM, perhaps with addressing @RKSimon remark about release notes

LGTM. I verified using debugger and the longjmp3 Tsan test case that he setjmp Tsan interceptor is being correctly called with tis change.

Thanks for reverting. Julian is right, the failure was caused by the virtual keyword being added to the destructor.

Addressed the latest feedback from yln

Apply code review feedback

Before you land this - I just noticed that your #elifs spacing is slightly off.

LGTM

LGTM

I am actually not very familiar with the memset_pattern API - is it available on every OS? What about various OS versions, when was is introduced? I see above that the diff build failed on Debian (not sure if that's reliable, but it's a hint).

I wonder why the implementation is different from memset interceptor, looking at sanitizer_common_interceptors(868):

There are no changes in strncat-verlap.cpp test, correcting.

After discussion I decided to come up with a different, separate fix.

Reapproving commit message change

I mean, I'm good accepting it too

LGTM

• Simplify expression deciding that high memory limit is low
• Fine tune the IS_HIGH_MEM_LOW comment

+1 to delcypher's suggestion

LVGTM - very DRY :)

LGTM

Will do (add a comment why revert) , sorry.

Okay, will revert and investigate - thanks for notifying.

Added SANITIZER_WATCHOS and SANITIZER_TVOS for the non-Apple case

Re-adding missing #else line

Removed unneccessary conditional