Issue 10270032: Sort functions when doing megamorphic dispatch/inlining so their

Keyboard Shortcuts

	File
u :	up to issue
j / k :	jump to file after / before current file
J / K :	jump to next file with a comment after / before current file
	Side-by-side diff
i :	toggle intra-line diffs
e :	expand all comments
c :	collapse all comments
s :	toggle showing all comments
n / p :	next / previous diff chunk or comment
N / P :	next / previous comment
<Up> / <Down> :	next / previous line

	Issue
u :	up to list of issues
j / k :	jump to patch after / before current patch
o / <Enter> :	open current patch in side-by-side view
i :	open current patch in unified diff view

	Issue List
j / k :	jump to issue after / before current issue
o / <Enter> :	open current issue

Issue 10270032: Sort functions when doing megamorphic dispatch/inlining so their (Closed)

Created:
8 years, 7 months ago by Erik Corry

Modified:
8 years, 7 months ago

Reviewers:
Michael Starzinger

CC:
v8-dev

Base URL:
http://v8.googlecode.com/svn/branches/bleeding_edge/

Visibility:
Public.

More Reviews

Description

Sort functions when doing megamorphic dispatch/inlining so their order does not depend on the hash seed. Committed: https://code.google.com/p/v8/source/detail?r=11482

Patch Set 1 #

Total comments: 2

Created: 8 years, 7 months ago

Download [raw] [tar.bz2]

		Unified diffs	Side-by-side diffs	Delta from patch set	Stats (+116 lines, -45 lines)			Patch
	M	src/hydrogen.h	View		1 chunk	+1 line, -0 lines	0 comments	Download
	M	src/hydrogen.cc	View		4 chunks	+115 lines, -45 lines	2 comments	Download

Messages

Total messages: 3 (0 generated)

Expand Messages | Collapse Messages

Michael Starzinger

https://chromiumcodereview.appspot.com/10270032/diff/1/src/hydrogen.cc File src/hydrogen.cc (right): https://chromiumcodereview.appspot.com/10270032/diff/1/src/hydrogen.cc#newcode5717 src/hydrogen.cc:5717: InliningAstSize(expr->target()), Could we just use "expr->target()->shared()->ast_node_count()" and disregard whether ...

8 years, 7 months ago (2012-05-02 08:10:35 UTC) #2

Erik Corry

8 years, 7 months ago (2012-05-08 12:13:38 UTC) #3

https://chromiumcodereview.appspot.com/10270032/diff/1/src/hydrogen.cc
File src/hydrogen.cc (right):

https://chromiumcodereview.appspot.com/10270032/diff/1/src/hydrogen.cc#newcod...
src/hydrogen.cc:5717: InliningAstSize(expr->target()),
On 2012/05/02 08:10:35, Michael Starzinger wrote:
> Could we just use "expr->target()->shared()->ast_node_count()" and disregard
> whether we potentially inline the target or not? That would simplify the code.
> Also inlining decision is affected by other factors as well.

The reason I don't do this is that we already have code in place to try to avoid
parsing up the function in the case where it is very large.  I don't want to
duplicate that logic to avoid parsing up the function just in order to sort it.

Expand Messages | Collapse Messages