Issue 465473002: Use actual incremental marking throughput in IdleNotification to estimate marking step size.

Issue 465473002: Use actual incremental marking throughput in IdleNotification to estimate marking step size. (Closed)

Created:
6 years, 4 months ago by Hannes Payer (out of office)

Modified:
6 years, 4 months ago

Reviewers:
jochen (gone - plz use gerrit), ulan, ernstm

CC:
v8-dev

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

Project:
v8

Visibility:
Public.

More Reviews

Description

Use actual incremental marking throughput in IdleNotification to estimate marking step size. BUG= R=jochen@chromium.org, ulan@chromium.org Committed: https://code.google.com/p/v8/source/detail?r=23224

Patch Set 8 : #

Created: 6 years, 4 months ago

Download [raw] [tar.bz2]

	Unified diffs	Side-by-side diffs	Delta from patch set	Stats (+125 lines, -26 lines)			Patch
M	BUILD.gn	View	1 2 3 4	1 chunk	+2 lines, -0 lines	0 comments	Download
A	src/heap/gc-idle-time-handler.h	View	1 2 3 4 5 6 7	1 chunk	+35 lines, -0 lines	0 comments	Download
A	src/heap/gc-idle-time-handler.cc	View	1 2 3 4 5 6	1 chunk	+37 lines, -0 lines	0 comments	Download
M	src/heap/heap.h	View	1 2 3 4 5 6 7	2 chunks	+2 lines, -2 lines	0 comments	Download
M	src/heap/heap.cc	View	1 2 3 4	5 chunks	+15 lines, -21 lines	0 comments	Download
M	test/heap-unittests/heap-unittest.cc	View	1 2 3 4 5 6 7	1 chunk	+32 lines, -3 lines	0 comments	Download
M	tools/gyp/v8.gyp	View	1 2 3 4	1 chunk	+2 lines, -0 lines	0 comments	Download

Messages

Total messages: 9 (0 generated)

Expand Messages | Collapse Messages

jochen (gone - plz use gerrit)

https://codereview.chromium.org/465473002/diff/1/src/heap/heap.cc File src/heap/heap.cc (right): https://codereview.chromium.org/465473002/diff/1/src/heap/heap.cc#newcode4295 src/heap/heap.cc:4295: ((tracer_.IncrementalMarkingSpeedInBytesPerMillisecond() / 100) * why not just * 0.9 ...

6 years, 4 months ago (2014-08-12 08:35:24 UTC) #2

Hannes Payer (out of office)

Moreover, note that I moved the calculation into AdvanceIdleIncrementalMarking. https://codereview.chromium.org/465473002/diff/1/src/heap/heap.cc File src/heap/heap.cc (right): https://codereview.chromium.org/465473002/diff/1/src/heap/heap.cc#newcode4295 src/heap/heap.cc:4295: ...

6 years, 4 months ago (2014-08-12 09:48:56 UTC) #3

ulan

https://codereview.chromium.org/465473002/diff/60001/src/heap/heap.cc File src/heap/heap.cc (right): https://codereview.chromium.org/465473002/diff/60001/src/heap/heap.cc#newcode4274 src/heap/heap.cc:4274: return static_cast<intptr_t>(marking_speed_in_bytes_per_millisecond * Potential integer overflow here, maybe use ...

6 years, 4 months ago (2014-08-18 12:32:13 UTC) #5

Hannes Payer (out of office)

Because of the potential bug you found, I thought it would be nice to have ...

6 years, 4 months ago (2014-08-19 15:24:04 UTC) #6

ulan

lgtm https://codereview.chromium.org/465473002/diff/80001/src/heap/gc-idle-time-handler.cc File src/heap/gc-idle-time-handler.cc (right): https://codereview.chromium.org/465473002/diff/80001/src/heap/gc-idle-time-handler.cc#newcode19 src/heap/gc-idle-time-handler.cc:19: int idle_time_in_ms, intptr_t marking_speed_in_bytes_per_millisecond) { Nit: marking_speed_in_bytes_per_ms would ...

6 years, 4 months ago (2014-08-20 08:21:16 UTC) #7

Hannes Payer (out of office)

https://codereview.chromium.org/465473002/diff/80001/src/heap/gc-idle-time-handler.cc File src/heap/gc-idle-time-handler.cc (right): https://codereview.chromium.org/465473002/diff/80001/src/heap/gc-idle-time-handler.cc#newcode19 src/heap/gc-idle-time-handler.cc:19: int idle_time_in_ms, intptr_t marking_speed_in_bytes_per_millisecond) { On 2014/08/20 08:21:16, ulan ...

6 years, 4 months ago (2014-08-20 10:20:15 UTC) #8

Hannes Payer (out of office)

6 years, 4 months ago (2014-08-20 10:33:13 UTC) #9

Message was sent while issue was closed.

Committed patchset #8 manually as 23224 (presubmit successful).

Expand Messages | Collapse Messages