Bug 127538 - DFG should allow inlining of op_call_varargs calls
Summary: DFG should allow inlining of op_call_varargs calls
Status: RESOLVED FIXED
Alias: None
Product: WebKit
Classification: Unclassified
Component: JavaScriptCore (show other bugs)
Version: 528+ (Nightly build)
Hardware: All All
: P2 Normal
Assignee: Filip Pizlo
URL:
Keywords:
Depends on:
Blocks: 127335
  Show dependency tree / graph
 
Reported: 2014-01-23 20:07 PST by Filip Pizlo
Modified: 2014-01-24 16:56 PST (History)
10 users (show)

See Also:

Attachments
getting there (10.01 KB, patch)
2014-01-24 11:43 PST, Filip Pizlo 		no flags Details | Formatted Diff | Diff
the patch (15.43 KB, patch)
2014-01-24 13:03 PST, Filip Pizlo 		msaboff: review+
Details | Formatted Diff | Diff
Show Obsolete (1) 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 Filip Pizlo 2014-01-23 20:07:49 PST 
Patch forthcoming.
Comment 1 Filip Pizlo 2014-01-23 20:14:06 PST 
This is freakishly weird because of how op_call_varargs gets its arguments.  I think we can make it work, though.  The important thing is that the ByteCodeParser needs to use the firstFreeRegister operand to decode how to lay out the call as if it was a normal call, and then to emit a sled of SetLocal's representing the arguments - again, as if this had been a normal call.  Those SetLocals would be ImmediateSet.

Anyway, it'l be super amazing.
Comment 2 Filip Pizlo 2014-01-24 11:43:39 PST 
Created attachment 222130 [details]
getting there

It still has OSR exit issues but it's starting to look like something.
Comment 3 Filip Pizlo 2014-01-24 13:03:53 PST 
Created attachment 222144 [details]
the patch
Comment 4 Filip Pizlo 2014-01-24 16:56:20 PST 
Landed in http://trac.webkit.org/changeset/162739