WebKit Bugzilla
New
Browse
Log In
×
Sign in with GitHub
or
Remember my login
Create Account
·
Forgot Password
Forgotten password account recovery
RESOLVED FIXED
216717
[JSC] PreciseAllocation's isNewlyAllocated flag should be propagated from isMarked at GC begin phase to make isLive correct
https://bugs.webkit.org/show_bug.cgi?id=216717
Summary
[JSC] PreciseAllocation's isNewlyAllocated flag should be propagated from isM...
Yusuke Suzuki
Reported
2020-09-18 15:04:32 PDT
[JSC] PreciseAllocation's isNewlyAllocated flag should be propagated from isMarked at GC begin phase to make isLive correct
Attachments
Patch
(3.61 KB, patch)
2020-09-18 15:13 PDT
,
Yusuke Suzuki
no flags
Details
Formatted Diff
Diff
Patch
(4.05 KB, patch)
2020-09-18 15:25 PDT
,
Yusuke Suzuki
no flags
Details
Formatted Diff
Diff
Patch
(4.05 KB, patch)
2020-09-18 16:32 PDT
,
Yusuke Suzuki
mark.lam
: review+
Details
Formatted Diff
Diff
Show Obsolete
(2)
View All
Add attachment
proposed patch, testcase, etc.
Yusuke Suzuki
Comment 1
2020-09-18 15:13:43 PDT
Created
attachment 409173
[details]
Patch
Yusuke Suzuki
Comment 2
2020-09-18 15:25:08 PDT
Created
attachment 409175
[details]
Patch
Radar WebKit Bug Importer
Comment 3
2020-09-18 15:50:02 PDT
<
rdar://problem/69179885
>
Mark Lam
Comment 4
2020-09-18 16:26:38 PDT
Comment on
attachment 409175
[details]
Patch View in context:
https://bugs.webkit.org/attachment.cgi?id=409175&action=review
Nice. Looks good so far but I still need to check this against the MarkedBlock code. Here are some typos while I continue reviewing.
> Source/JavaScriptCore/ChangeLog:9 > + However, this means that HeapCell::isLive will see this object dead until it is marked.
/object dead/object as dead/
> Source/JavaScriptCore/heap/PreciseAllocation.cpp:218 > + // We do not need to care about concurrency here since marking thread is stopped right now. This is followin to the logic
/followin/equivalent/
Yusuke Suzuki
Comment 5
2020-09-18 16:31:18 PDT
Comment on
attachment 409175
[details]
Patch View in context:
https://bugs.webkit.org/attachment.cgi?id=409175&action=review
>> Source/JavaScriptCore/ChangeLog:9 >> + However, this means that HeapCell::isLive will see this object dead until it is marked. > > /object dead/object as dead/
Fixed.
>> Source/JavaScriptCore/heap/PreciseAllocation.cpp:218 >> + // We do not need to care about concurrency here since marking thread is stopped right now. This is followin to the logic > > /followin/equivalent/
Fixed.
Yusuke Suzuki
Comment 6
2020-09-18 16:32:30 PDT
Created
attachment 409178
[details]
Patch
Mark Lam
Comment 7
2020-09-18 17:36:35 PDT
Comment on
attachment 409178
[details]
Patch r=me. Nice fix, and nice comments documenting the reasoning behind all this.
Yusuke Suzuki
Comment 8
2020-09-18 18:22:49 PDT
Committed
r267304
: <
https://trac.webkit.org/changeset/267304
>
Note
You need to
log in
before you can comment on or make changes to this bug.
Top of Page
Format For Printing
XML
Clone This Bug