Closed Bug 1604829 Opened 4 years ago Closed 4 years ago

Mark count calculation is wrong in GCMarker::mark

Categories

(Core :: JavaScript: GC, defect, P3)

defect

Tracking

()

RESOLVED FIXED
mozilla74
Tracking Status
firefox-esr68 --- wontfix
firefox72 --- wontfix
firefox73 --- wontfix
firefox74 --- fixed

People

(Reporter: jonco, Assigned: allstars.chh)

References

(Regression)

Details

(Keywords: regression)

Attachments

(1 file)

The calculation of how many cells we mark is wrong because it is incremented whenever we touch a cell, regardless of whether that cell was already marking. I should have spotted this in code review I guess.

  markCount++;
  return cell->markIfUnmarked(color);

We only want to increment the counter if markIfUnmarked() returns true, otherwise we will overcount.

The priority flag is not set for this bug.
:jonco, could you have a look please?

For more information, please visit auto_nag documentation.

Flags: needinfo?(jcoppeard)
Flags: needinfo?(jcoppeard)
Priority: -- → P3

Yoshi, can you take a look at this when you get a chance?

Flags: needinfo?(allstars.chh)
Assignee: nobody → allstars.chh
Flags: needinfo?(allstars.chh)
Pushed by allstars.chh@gmail.com:
https://hg.mozilla.org/integration/autoland/rev/a9b4a0406243
Increment mark count only when the cell is marked. r=jonco
Status: NEW → RESOLVED
Closed: 4 years ago
Resolution: --- → FIXED
Target Milestone: --- → mozilla74

Is there a user impact here which justifies backport consideration or can this fix ride Fx74 to release?

Flags: needinfo?(allstars.chh)

Nope, this only affects telemetry.

Flags: needinfo?(allstars.chh)
Has Regression Range: --- → yes
Flags: needinfo?(jcoppeard)
You need to log in before you can comment on or make changes to this bug.

Attachment

General

Created:
Updated:
Size: