performance regression on sphereflake

Gerolf Hoflehner ghoflehner at apple.com
Wed Feb 5 16:33:38 PST 2014


Hi Rafael

it seems that  one of your commits modifies stack alignment (and consequently instruction selection) of not just APCS and results in a performance regression on sphereflake. Could you please take a look? 

Thanks
Gerolf

— svn rev 197526  
Author: Rafael Espindola <rafael.espindola at gmail.com>
Date:   Tue Dec 17 21:36:54 2013 +0000

    On APCS, only try to align aggregates to 32 bits instead of 64.
    
    This matches clang's behavior and since it is only a preference, it is not
    an ABI issue.
—

Sphereflake sample snippet (main()) with 32b (after) and 64b (before your change) stack alignment:

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20140205/e5da296b/attachment.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: PastedGraphic-2.png
Type: image/png
Size: 67847 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20140205/e5da296b/attachment.png>


More information about the llvm-commits mailing list