Issue 11275145: Decouple allocation and creation of deopt tables

Issue 11275145: Decouple allocation and creation of deopt tables (Closed)

Created:
8 years, 1 month ago by danno

Modified:
8 years, 1 month ago

Reviewers:
Jakob Kummerow

CC:
v8-dev

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

Visibility:
Public.

More Reviews

Description

Decouple allocation and creation of deopt tables This makes it possible to calculate the future address of a deopt entry before it is possible to generate the deopt table. Committed: https://code.google.com/p/v8/source/detail?r=12877

Patch Set 1 #

Patch Set 2 : Grow deopt tables on demand #

Patch Set 3 : Revert MIPS support #

Patch Set 4 : nits #

Patch Set 5 : Ready for review #

Total comments: 5

Patch Set 6 : Review feedback #

Created: 8 years, 1 month ago

Download [raw] [tar.bz2]

	Unified diffs	Side-by-side diffs	Delta from patch set	Stats (+87 lines, -45 lines)			Patch
M	src/api.cc	View	1 2 3 4	1 chunk	+1 line, -0 lines	0 comments	Download
M	src/deoptimizer.h	View	1 2 3 4	4 chunks	+21 lines, -5 lines	0 comments	Download
M	src/deoptimizer.cc	View	1 2 3 4 5	5 chunks	+65 lines, -40 lines	0 comments	Download

Messages

Total messages: 3 (0 generated)

Expand Messages | Collapse Messages

Jakob Kummerow

LGTM with nits. http://codereview.chromium.org/11275145/diff/8001/src/deoptimizer.cc File src/deoptimizer.cc (right): http://codereview.chromium.org/11275145/diff/8001/src/deoptimizer.cc#newcode58 src/deoptimizer.cc:58: if (eager_deoptimization_entry_code_ != NULL) { nit: ...

8 years, 1 month ago (2012-11-06 11:40:49 UTC) #2

danno

8 years, 1 month ago (2012-11-06 11:59:03 UTC) #3

Feedback addressed, I will land this after verifying there's not performance
regression.

http://codereview.chromium.org/11275145/diff/8001/src/deoptimizer.cc
File src/deoptimizer.cc (right):

http://codereview.chromium.org/11275145/diff/8001/src/deoptimizer.cc#newcode58
src/deoptimizer.cc:58: if (eager_deoptimization_entry_code_ != NULL) {
On 2012/11/06 11:40:50, Jakob wrote:
> nit: you can simplify this a bit by removing the condition. |delete| is safe
to
> call on NULL pointers.
> 
> Again below.

Done.

http://codereview.chromium.org/11275145/diff/8001/src/deoptimizer.cc#newcode478
src/deoptimizer.cc:478: Address Deoptimizer::GetDeoptimizationEntry(int id,
BailoutType type,
On 2012/11/06 11:40:50, Jakob wrote:
> nit: one line per argument

Done.

Expand Messages | Collapse Messages