[chromium-bugs] Re: Issue 28749 in chromium: gcc 4.4-specific renderer crash in malloc, memory corruption while loading plugins (?)

Comment #46 on issue 28749 by senorbla...@chromium.org: gcc 4.4-specific renderer crash in malloc, memory corruption while loading plugins (?) http://code.google.com/p/chromium/issues/detail?id=28749 -fno-strict-aliasing will work, but I think if you want to get around it without turning

[chromium-bugs] Re: Issue 28749 in chromium: gcc 4.4-specific renderer crash in malloc, memory corruption while loading plugins (?)

Comment #47 on issue 28749 by craig.schlen...@chromium.org: gcc 4.4-specific renderer crash in malloc, memory corruption while loading plugins (?) http://code.google.com/p/chromium/issues/detail?id=28749 O ... that's tempting ... I was only thinking about doing union for Type, not

[chromium-bugs] Re: Issue 28749 in chromium: gcc 4.4-specific renderer crash in malloc, memory corruption while loading plugins (?)

Comment #48 on issue 28749 by craig.schlen...@chromium.org: gcc 4.4-specific renderer crash in malloc, memory corruption while loading plugins (?) http://code.google.com/p/chromium/issues/detail?id=28749 Bah ... it's not happy. We can't fool it. cc1plus: warnings being treated as errors

[chromium-bugs] Re: Issue 28749 in chromium: gcc 4.4-specific renderer crash in malloc, memory corruption while loading plugins (?)

Comment #49 on issue 28749 by e...@chromium.org: gcc 4.4-specific renderer crash in malloc, memory corruption while loading plugins (?) http://code.google.com/p/chromium/issues/detail?id=28749 Yes, the strict aliasing rule is about using the same memory as two different types, not some

[chromium-bugs] Re: Issue 28749 in chromium: gcc 4.4-specific renderer crash in malloc, memory corruption while loading plugins (?)

Updates: Status: Started Owner: craig.sc...@chromium.org Comment #52 on issue 28749 by craig.schlen...@chromium.org: gcc 4.4-specific renderer crash in malloc, memory corruption while loading plugins (?) http://code.google.com/p/chromium/issues/detail?id=28749 Joel I'm

[chromium-bugs] Re: Issue 28749 in chromium: gcc 4.4-specific renderer crash in malloc, memory corruption while loading plugins (?)

Comment #43 on issue 28749 by yu...@chromium.org: gcc 4.4-specific renderer crash in malloc, memory corruption while loading plugins (?) http://code.google.com/p/chromium/issues/detail?id=28749 Yes, the current default compiler for Fedora is gcc-4.4. I also like the idea of fixing the

[chromium-bugs] Re: Issue 28749 in chromium: gcc 4.4-specific renderer crash in malloc, memory corruption while loading plugins (?)

Comment #44 on issue 28749 by craig.schlen...@chromium.org: gcc 4.4-specific renderer crash in malloc, memory corruption while loading plugins (?) http://code.google.com/p/chromium/issues/detail?id=28749 What do you think is incorrect about the patch btw. (other than the fact that it

[chromium-bugs] Re: Issue 28749 in chromium: gcc 4.4-specific renderer crash in malloc, memory corruption while loading plugins (?)

Comment #45 on issue 28749 by yu...@chromium.org: gcc 4.4-specific renderer crash in malloc, memory corruption while loading plugins (?) http://code.google.com/p/chromium/issues/detail?id=28749 What I meant was I couldn't judge the correctness due to lack of my knowledge about

[chromium-bugs] Re: Issue 28749 in chromium: gcc 4.4-specific renderer crash in malloc, memory corruption while loading plugins (?)

Comment #40 on issue 28749 by yu...@chromium.org: gcc 4.4-specific renderer crash in malloc, memory corruption while loading plugins (?) http://code.google.com/p/chromium/issues/detail?id=28749 This issue affects the official Google Chrome binary for Fedora (and probably Ubuntu Karmic too)

[chromium-bugs] Re: Issue 28749 in chromium: gcc 4.4-specific renderer crash in malloc, memory corruption while loading plugins (?)

Updates: Status: Available Cc: jap...@chromium.org Comment #41 on issue 28749 by e...@chromium.org: gcc 4.4-specific renderer crash in malloc, memory corruption while loading plugins (?) http://code.google.com/p/chromium/issues/detail?id=28749 CC japhet FYI since he

[chromium-bugs] Re: Issue 28749 in chromium: gcc 4.4-specific renderer crash in malloc, memory corruption while loading plugins (?)

Comment #42 on issue 28749 by craig.schlen...@chromium.org: gcc 4.4-specific renderer crash in malloc, memory corruption while loading plugins (?) http://code.google.com/p/chromium/issues/detail?id=28749 yutak: I was under the impression that only gcc 4.4 compiled code is a problem - are

[chromium-bugs] Re: Issue 28749 in chromium: gcc 4.4-specific renderer crash in malloc, memory corruption while loading plugins (?)

Comment #38 on issue 28749 by craig.schlen...@chromium.org: gcc 4.4-specific renderer crash in malloc, memory corruption while loading plugins (?) http://code.google.com/p/chromium/issues/detail?id=28749 How about something that tries to use the returned Type* from placement new rather

[chromium-bugs] Re: Issue 28749 in chromium: gcc 4.4-specific renderer crash in malloc, memory corruption while loading plugins (?)

Comment #32 on issue 28749 by de...@chromium.org: gcc 4.4-specific renderer crash in malloc, memory corruption while loading plugins (?) http://code.google.com/p/chromium/issues/detail?id=28749 Using a union is actually against the C spec, and is only a GCCism. The actual spec specifies

[chromium-bugs] Re: Issue 28749 in chromium: gcc 4.4-specific renderer crash in malloc, memory corruption while loading plugins (?)

Comment #33 on issue 28749 by de...@chromium.org: gcc 4.4-specific renderer crash in malloc, memory corruption while loading plugins (?) http://code.google.com/p/chromium/issues/detail?id=28749 Actually, looking at this code again, it could be probably just rewritten as: Type* Pointer()

[chromium-bugs] Re: Issue 28749 in chromium: gcc 4.4-specific renderer crash in malloc, memory corruption while loading plugins (?)

Comment #34 on issue 28749 by craig.schlen...@chromium.org: gcc 4.4-specific renderer crash in malloc, memory corruption while loading plugins (?) http://code.google.com/p/chromium/issues/detail?id=28749 Thanks for the comments Dean. I'll try out a couple of variations shortly and report

[chromium-bugs] Re: Issue 28749 in chromium: gcc 4.4-specific renderer crash in malloc, memory corruption while loading plugins (?)

Comment #35 on issue 28749 by craig.schlen...@chromium.org: gcc 4.4-specific renderer crash in malloc, memory corruption while loading plugins (?) http://code.google.com/p/chromium/issues/detail?id=28749 Still broken unfortunately :( Using char instead of int8 doesn't help and using the

[chromium-bugs] Re: Issue 28749 in chromium: gcc 4.4-specific renderer crash in malloc, memory corruption while loading plugins (?)

Comment #36 on issue 28749 by craig.schlen...@chromium.org: gcc 4.4-specific renderer crash in malloc, memory corruption while loading plugins (?) http://code.google.com/p/chromium/issues/detail?id=28749 If all else fails, shuffle some code around and hope the compiler optimizes things

[chromium-bugs] Re: Issue 28749 in chromium: gcc 4.4-specific renderer crash in malloc, memory corruption while loading plugins (?)

Comment #37 on issue 28749 by de...@chromium.org: gcc 4.4-specific renderer crash in malloc, memory corruption while loading plugins (?) http://code.google.com/p/chromium/issues/detail?id=28749 You can't do work in the constructor, otherwise it no longer becomes LinkerInitialized and defeats

[chromium-bugs] Re: Issue 28749 in chromium: gcc 4.4-specific renderer crash in malloc, memory corruption while loading plugins (?)

Comment #30 on issue 28749 by craig.schlen...@chromium.org: gcc 4.4-specific renderer crash in malloc, memory corruption while loading plugins (?) http://code.google.com/p/chromium/issues/detail?id=28749 I had considered using a union to avoid the aliasing issue but apparently when using

[chromium-bugs] Re: Issue 28749 in chromium: gcc 4.4-specific renderer crash in malloc, memory corruption while loading plugins (?)

Comment #31 on issue 28749 by craig.schlen...@chromium.org: gcc 4.4-specific renderer crash in malloc, memory corruption while loading plugins (?) http://code.google.com/p/chromium/issues/detail?id=28749 http://codereview.chromium.org/519034/show ( use no_strict_aliasing=1 with gcc 4.4 )

[chromium-bugs] Re: Issue 28749 in chromium: gcc 4.4-specific renderer crash in malloc, memory corruption while loading plugins (?)

Comment #28 on issue 28749 by craig.schlen...@chromium.org: gcc 4.4-specific renderer crash in malloc, memory corruption while loading plugins (?) http://code.google.com/p/chromium/issues/detail?id=28749 Issue 31044 has been merged into this issue. -- You received this message because you

[chromium-bugs] Re: Issue 28749 in chromium: gcc 4.4-specific renderer crash in malloc, memory corruption while loading plugins (?)

Updates: Cc: de...@chromium.org Comment #29 on issue 28749 by e...@chromium.org: gcc 4.4-specific renderer crash in malloc, memory corruption while loading plugins (?) http://code.google.com/p/chromium/issues/detail?id=28749 Why not set no_strict_aliasing=1? It seems that not setting

[chromium-bugs] Re: Issue 28749 in chromium: gcc 4.4-specific renderer crash in malloc, memory corruption while loading plugins (?)

Updates: Cc: j...@chromium.org Comment #27 on issue 28749 by e...@chromium.org: gcc 4.4-specific renderer crash in malloc, memory corruption while loading plugins (?) http://code.google.com/p/chromium/issues/detail?id=28749 Issue 28543 has been merged into this issue. -- You received

[chromium-bugs] Re: Issue 28749 in chromium: gcc 4.4-specific renderer crash in malloc, memory corruption while loading plugins (?)

Comment #26 on issue 28749 by craig.schlen...@chromium.org: gcc 4.4-specific renderer crash in malloc, memory corruption while loading plugins (?) http://code.google.com/p/chromium/issues/detail?id=28749 Issue 31011 has been merged into this issue. -- You received this message because you

[chromium-bugs] Re: Issue 28749 in chromium: gcc 4.4-specific renderer crash in malloc, memory corruption while loading plugins (?)

Comment #25 on issue 28749 by f...@sofaraway.org: gcc 4.4-specific renderer crash in malloc, memory corruption while loading plugins (?) http://code.google.com/p/chromium/issues/detail?id=28749 i reverted my change shortly afterwards (meaning i'm back with no_strict_aliasing=1 gcc_version=44

[chromium-bugs] Re: Issue 28749 in chromium: gcc 4.4-specific renderer crash in malloc, memory corruption while loading plugins (?)

Updates: Cc: f...@sofaraway.org Comment #23 on issue 28749 by craig.schlen...@chromium.org: gcc 4.4-specific renderer crash in malloc, memory corruption while loading plugins (?) http://code.google.com/p/chromium/issues/detail?id=28749 Sifting through IRC logs, I see fta seems to be

[chromium-bugs] Re: Issue 28749 in chromium: gcc 4.4-specific renderer crash in malloc, memory corruption while loading plugins (?)

Comment #24 on issue 28749 by craig.schlen...@chromium.org: gcc 4.4-specific renderer crash in malloc, memory corruption while loading plugins (?) http://code.google.com/p/chromium/issues/detail?id=28749 So adding -fno-strict-aliasing to my test program in comment #22, solves the problem

[chromium-bugs] Re: Issue 28749 in chromium: gcc 4.4-specific renderer crash in malloc, memory corruption while loading plugins (?)

Updates: Cc: yusu...@chromium.org Comment #19 on issue 28749 by craig.schlen...@chromium.org: gcc 4.4-specific renderer crash in malloc, memory corruption while loading plugins (?) http://code.google.com/p/chromium/issues/detail?id=28749 Issue 30539 has been merged into this issue.

[chromium-bugs] Re: Issue 28749 in chromium: gcc 4.4-specific renderer crash in malloc, memory corruption while loading plugins (?)

Comment #20 on issue 28749 by craig.schlen...@chromium.org: gcc 4.4-specific renderer crash in malloc, memory corruption while loading plugins (?) http://code.google.com/p/chromium/issues/detail?id=28749 So the following change fixes things for me: http://codereview.chromium.org/506072 I'm

[chromium-bugs] Re: Issue 28749 in chromium: gcc 4.4-specific renderer crash in malloc, memory corruption while loading plugins (?)

Comment #21 on issue 28749 by craig.schlen...@chromium.org: gcc 4.4-specific renderer crash in malloc, memory corruption while loading plugins (?) http://code.google.com/p/chromium/issues/detail?id=28749 Here's the first error from running valgrind on the renderer btw.: ==24618== ==24618==

[chromium-bugs] Re: Issue 28749 in chromium: gcc 4.4-specific renderer crash in malloc, memory corruption while loading plugins (?)

Comment #22 on issue 28749 by craig.schlen...@chromium.org: gcc 4.4-specific renderer crash in malloc, memory corruption while loading plugins (?) http://code.google.com/p/chromium/issues/detail?id=28749 So the attached file is a super scaled down version of what sort of happens in

[chromium-bugs] Re: Issue 28749 in chromium: gcc 4.4-specific renderer crash in malloc, memory corruption while loading plugins (?)

Updates: Summary: gcc 4.4-specific renderer crash in malloc, memory corruption while loading plugins (?) Labels: -Area-Misc Area-Plugins Comment #17 on issue 28749 by e...@chromium.org: gcc 4.4-specific renderer crash in malloc, memory corruption while loading plugins (?)