Status: Available
Owner: ----
CC: [email protected]
Labels: Type-Bug Priority-Medium
New issue 4236 by [email protected]: mjsunit/regress-1132 is unstable
on ASAN
https://code.google.com/p/v8/issues/detail?id=4236
Recent changes including Object.assign and Harmony Array/TypedArray methods
increased the JS code size, and this somehow made regress-1132 segfault,
but only on ASAN builds. The test seems to be trying to get at exception
propagation from lazy compilation, and the error looks like something that
ASAN doesn't quite understand, rather than something wrong. Increasing the
stack size seems to suppress the error, but it's not clear how much to
increase the stack size by while continuing to test what the test is
testing. Failures look like this:
Run #1
Exit code: 1
Result: FAIL
Expected outcomes: PASS
Duration: 00:00:598
Stderr:
Exception thrown during bootstrapping
Extension or internal compilation error.
ASAN:SIGSEGV
=================================================================
==35238==ERROR: AddressSanitizer: SEGV on unknown address 0x000000000000
(pc 0x000100526db9 bp 0x7fff5fbfe8d0 sp 0x7fff5fbfe850 T0)
#0 0x100526db8 in v8::Context::Enter()
(/Volumes/data/b/build/slave/mac64-asan/build/v8/out/Release/d8+0x100526db8)
#1 0x10000950f in v8::Shell::CreateEvaluationContext(v8::Isolate*)
(/Volumes/data/b/build/slave/mac64-asan/build/v8/out/Release/d8+0x10000950f)
#2 0x10000cea9 in v8::Shell::RunMain(v8::Isolate*, int, char**)
(/Volumes/data/b/build/slave/mac64-asan/build/v8/out/Release/d8+0x10000cea9)
#3 0x10001384b in v8::Shell::Main(int, char**)
(/Volumes/data/b/build/slave/mac64-asan/build/v8/out/Release/d8+0x10001384b)
#4 0x100001133 in start
(/Volumes/data/b/build/slave/mac64-asan/build/v8/out/Release/d8+0x100001133)
#5 0xc
(/Volumes/data/b/build/slave/mac64-asan/build/v8/out/Release/d8+0xc)
AddressSanitizer can not provide additional info.
SUMMARY: AddressSanitizer: SEGV
(/Volumes/data/b/build/slave/mac64-asan/build/v8/out/Release/d8+0x100526db8)
in v8::Context::Enter()
==35238==ABORTING
I'm going to mail a patch to disable regress-1132 on ASAN runs until
someone can dig in a fix the underlying issue. I don't think it's the fault
of the TypedArray methods that this is failing, and I don't want to just
increase the stack size significantly to work around it.
--
You received this message because this project is configured to send all
issue notifications to this address.
You may adjust your notification preferences at:
https://code.google.com/hosting/settings
--
--
v8-dev mailing list
[email protected]
http://groups.google.com/group/v8-dev
---
You received this message because you are subscribed to the Google Groups "v8-dev" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to [email protected].
For more options, visit https://groups.google.com/d/optout.
