Bug 117049 - [sh4] Optimize NaN checks in LLINT for floating point comparisons
Summary: [sh4] Optimize NaN checks in LLINT for floating point comparisons
Status: RESOLVED FIXED
Alias: None
Product: WebKit
Classification: Unclassified
Component: JavaScriptCore (show other bugs)
Version: 528+ (Nightly build)
Hardware: Unspecified Unspecified
: P2 Normal
Assignee: Nobody
URL:
Keywords:
Depends on:
Blocks:
 
Reported: 2013-05-30 14:56 PDT by Julien Brianceau
Modified: 2013-05-30 16:21 PDT (History)
2 users (show)

See Also:

Attachments
Optimize NaN checks in sh4 LLINT for floating point comparisons (4.62 KB, patch)
2013-05-30 15:16 PDT, Julien Brianceau 		no flags Details | Formatted Diff | Diff
View All Add an attachment (proposed patch, testcase, etc.)

Note You need to log in before you can comment on or make changes to this bug.
Description Julien Brianceau 2013-05-30 14:56:55 PDT 
Use the fcmp/eq opcode in sh4 LLINT to test if a double is NaN.

This is more efficient, doesn't require two tmp registers and requires less code than current implementation (which converts double to float, then checks 'E = Emax + 1' and 'f != 0').
Comment 1 Julien Brianceau 2013-05-30 15:16:27 PDT 
Created attachment 203384 [details]
Optimize NaN checks in sh4 LLINT for floating point comparisons
Comment 2 WebKit Commit Bot 2013-05-30 16:21:08 PDT 
Comment on attachment 203384 [details]
Optimize NaN checks in sh4 LLINT for floating point comparisons

Clearing flags on attachment: 203384

Committed r150992: <http://trac.webkit.org/changeset/150992>
Comment 3 WebKit Commit Bot 2013-05-30 16:21:10 PDT 
All reviewed patches have been landed.  Closing bug.