Comment #27 on issue 17569 by [email protected]: Crash in  
SkARGB32_Black_Blitter::blitMask
http://code.google.com/p/chromium/issues/detail?id=17569

After adding the sanity check in MaskSuperBlitter::blitH(), the crash has  
no longer
occurred in any subsequent reliability runs.

http://code.google.com/p/skia/source/detail?r=330

The new asserts traced the root cause to path.getBounds().roundOut(&ir),  
which means
somewhere in floorf() or the float->int cast, something is going wrong  
(61.5 -> 62).
I pored over the disassembly a bit, but I couldn't see anything obvious,  
although
Googling for _ftol2_sse (the cast code in MSVCRT) yields this compiler bug:

connect.microsoft.com/VisualStudio/feedback/ViewFeedback.aspx?FeedbackID=262389

Note also that the cast code paths are different for SSE, non-SSE and  
Pentium4.  It's
possible that one of the machines in the reliability pool is taking a  
different
code path than my machine, which would explain the difficulty repro'ing on  
my
machine.

At any rate, since the crash is fixed, I'm going to remove the trace from
known_crashes.txt, and close this bug.

--
You received this message because you are listed in the owner
or CC fields of this issue, or because you starred this issue.
You may adjust your issue notification preferences at:
http://code.google.com/hosting/settings

--~--~---------~--~----~------------~-------~--~----~
Automated mail from issue updates at http://crbug.com/
Subscription options: http://groups.google.com/group/chromium-bugs
-~----------~----~----~----~------~----~------~--~---

Reply via email to