Closed
Bug 862501
Opened 8 years ago
Closed 5 years ago
Baseline (BC) should provide more accurate SP for better interleaving
Categories
(Core :: Gecko Profiler, defect)
Tracking
()
RESOLVED
INVALID
People
(Reporter: BenWa, Assigned: djvj)
Details
Attachments
(1 file)
|
2.51 KB,
patch
|
jandem
:
review+
|
Details | Diff | Splinter Review |
Right now the baseline pseudo entries are attached before RunScript. If we give better SP estimate they will get attached to their native JIT frames. FYI: You might want to update the author field before landing
|
Reporter | |
Updated•8 years ago
|
Attachment #738129 -
Attachment is patch: true
|
Assignee | |
Updated•8 years ago
|
Attachment #738129 -
Flags: review?(jdemooij)
|
||
Updated•8 years ago
|
Attachment #738129 -
Flags: review?(jdemooij) → review+
|
|
Assignee | |
Comment 1•8 years ago
|
||
Pushed: https://hg.mozilla.org/integration/mozilla-inbound/rev/36ffb85842ac
|
||
Comment 2•8 years ago
|
||
Backed out for mochitest crashes. https://hg.mozilla.org/integration/mozilla-inbound/rev/fce173d8800d https://tbpl.mozilla.org/php/getParsedLog.php?id=21968482&tree=Mozilla-Inbound
|
|
Assignee | |
Comment 3•8 years ago
|
||
Ah, it seems that the presence of a NULL sp is used as an indicator of whether the frame is JS or not. Talked to Benoit about it, he's busy with other stuff, and so am I so pushing this down the priority stack.
|
|
Assignee | |
Comment 4•5 years ago
|
||
This bug is irrelevant after the changes to the profiling infrastructure to let the profiler walk jit stacks directly instead of needing a pseudostack.
Status: NEW → RESOLVED
Closed: 5 years ago
Resolution: --- → INVALID
You need to log in
before you can comment on or make changes to this bug.
Description
•