Bug 66730 - [LLVM radeonsi] Infinite loop in LLVM backend possible when using "double" arguments
Summary: [LLVM radeonsi] Infinite loop in LLVM backend possible when using "double" ar...
Status: RESOLVED FIXED
Alias: None
Product: Mesa
Classification: Unclassified
Component: Drivers/Gallium/radeonsi (show other bugs)
Version: git
Hardware: Other All
: medium normal
Assignee: Default DRI bug account
QA Contact:
URL:
Whiteboard:
Keywords:
Depends on:
Blocks:
 
Reported: 2013-07-09 10:46 UTC by Niels Ole Salscheider
Modified: 2013-11-16 02:34 UTC (History)
0 users

See Also:
i915 platform:
i915 features:

Attachments
testcase (189 bytes, text/plain)
2013-07-09 10:46 UTC, Niels Ole Salscheider 		Details
View All

Description Niels Ole Salscheider 2013-07-09 10:46:22 UTC 
Created attachment 82223 [details]
testcase

Some expressions containing multiple double precision kernel arguments cause llvm to run until all memory is used. It works fine as long as I pass pointers to double values.

I have attached an LLVM IR testcase that triggers the described behaviour.
Comment 1 Tom Stellard 2013-11-16 02:34:08 UTC 
This is working for me now with the latest LLVM code.  Please re-open if it still doesn't work for you.

Use of freedesktop.org services, including Bugzilla, is subject to our Code of Conduct. How we collect and use information is described in our Privacy Policy.