Chromium Code Reviews
Help | Chromium Project | Sign in
(4)

Issue 11106012: Consistently make the bounds check for AccessArgumentsAt explicit. (Closed)

Can't Edit
Can't Publish+Mail
Start Review
Created:
1 year, 6 months ago by Sven Panne
Modified:
1 year, 6 months ago
Reviewers:
Michael Starzinger
CC:
v8-dev_googlegroups.com
Visibility:
Public.

Description

Consistently make the bounds check for AccessArgumentsAt explicit.

This has the advantage that AccessArgumentsAt itself can't deopt anymore and the
bounds check is visible for the elimination phase. Furthermore, things are
simply more consistent now, a good thing in itself. :-)

Committed: https://code.google.com/p/v8/source/detail?r=12721

Patch Set 1 #

Unified diffs Side-by-side diffs Delta from patch set Stats (+15 lines, -40 lines) Lint Patch
M src/arm/lithium-arm.cc View 1 chunk +2 lines, -4 lines 0 comments 0 errors Download
M src/arm/lithium-codegen-arm.cc View 1 chunk +1 line, -6 lines 0 comments 0 errors Download
M src/hydrogen.cc View 1 chunk +3 lines, -1 line 0 comments 0 errors Download
M src/ia32/lithium-codegen-ia32.cc View 1 chunk +1 line, -4 lines 0 comments 0 errors Download
M src/ia32/lithium-ia32.cc View 1 chunk +2 lines, -4 lines 0 comments 0 errors Download
M src/mips/lithium-codegen-mips.cc View 1 chunk +0 lines, -8 lines 0 comments 0 errors Download
M src/mips/lithium-mips.cc View 1 chunk +2 lines, -4 lines 0 comments 0 errors Download
M src/x64/lithium-codegen-x64.cc View 1 chunk +2 lines, -5 lines 0 comments 0 errors Download
M src/x64/lithium-x64.cc View 1 chunk +2 lines, -4 lines 0 comments 0 errors Download
Trybot results:
Commit:

Messages

Total messages: 2
Michael Starzinger
LGTM. As discussed offline, the benefits most certainly outweigh the additional instructions for the GenerateArguments ...
1 year, 6 months ago #1
Sven Panne
1 year, 6 months ago #2
As discussed offline, the "hot stuff" should be crankshafted, anyway, and this
should use less instructions. In addition, the Lithium instruction can't deopt
anymore, so this should be nicer to the register allocator.

What we would really need here is an optimization across Lithium operation
boundaries, so the cmp/branch/sub would be simplified to a sub/branch. Peephole
optimizations, anyone...?

Landing...
Sign in to reply to this message.

Powered by Google App Engine
RSS Feeds Recent Issues | This issue
This is Rietveld 1280:2d3e6564b7b6