Bug 110536

Summary: DFG ArgumentsSimplificationPhase should not have to rethread the graph
Product: WebKit Reporter: Filip Pizlo <fpizlo>
Component: JavaScriptCoreAssignee: Filip Pizlo <fpizlo>
Status: RESOLVED INVALID    
Severity: Normal CC: barraclough, ggaren, mark.lam, mhahnenberg, msaboff, oliver, sam
Priority: P2    
Version: 528+ (Nightly build)   
Hardware: All   
OS: All   
Bug Depends on: 110535    
Bug Blocks:    
Attachments:
Description Flags
the patch ggaren: review+

Filip Pizlo
Reported 2013-02-21 17:36:40 PST
This is an investigation of sorts. I may be wrong. But if true, then this should be a speed win.
Attachments
the patch (7.71 KB, patch)
2013-02-21 17:37 PST, Filip Pizlo 		ggaren: review+
DetailsFormatted DiffDiff
View All Add attachment proposed patch, testcase, etc.
Filip Pizlo
Comment 1 2013-02-21 17:37:35 PST
Created attachment 189647 [details] the patch
Filip Pizlo
Comment 2 2013-02-21 17:38:11 PST
Comment on attachment 189647 [details] the patch View in context: https://bugs.webkit.org/attachment.cgi?id=189647&action=review > Source/JavaScriptCore/dfg/DFGVariableAccessData.h:280 > + > + Oops. I will back this out.
Geoffrey Garen
Comment 3 2013-02-21 17:44:46 PST
Comment on attachment 189647 [details] the patch r=me
Filip Pizlo
Comment 4 2013-02-22 00:28:46 PST
I've come to the conclusion that despite its weirdness, the current state of the arguments simplification phase is actually right. In fact: - PhantomArguments is always live, because it is always used to store into the arguments register. - The arguments alias register is always dead. So, everything is consistent as it is. I'm glad that I tried to "fix" the code; it gave me an appreciation for how awfully right the code was to begin with.
Filip Pizlo
Comment 5 2013-02-22 00:29:16 PST
*** Bug 110533 has been marked as a duplicate of this bug. ***
Note You need to log in before you can comment on or make changes to this bug.