From:  Andrew McCreight <>
Date:  15 Jul 2017 07:21:43 Hong Kong Time

Re: MemShrink Triage (12 July 2017)


On Wed, Jul 12, 2017 at 4:30 PM, Eric Rahm <> wrote:

MemShrink triage: 2017-07-12

Triage URL:


Note: We're going to just triage via email this week.

Bug List

Vote for:

  • P1 High importance, will follow up periodically
  • P2 Important issue, possibly already being worked on
  • P3 Real issue, someone will get to it if they have time
  • moreinfo We need logs, or clarifying information
  • invalid Misclassified, remove the MemShrink tag

17 bugs to triage

  • 1357598 - Firefox :: General - Poor performance and high memory usage in Firefox after waking up computer from sleep


    erahm, what do you think?

  • 1362767 - Core :: Untriaged - Firefox doesn't release memory after closing tabs


  • 1367999 - Core :: General - Got leak messages when close firefox

P2 or P3? This is a long-standing issue, which isn't great, but I think it doesn't happen if you close the tab first before shutting down, so I don't think it is a leak we need to worry much about.
  • Votes:

  • 1372092 - Firefox :: Untriaged - Long running Firefox eat all memory

Sounds bad, but maybe a bad addon or site?
  • Votes:

  • 1373089 - Core :: JavaScript: Internationalization API - Use Sets rather than objects to store available locales in Intl API


  • 1373340 - Core :: IPC - Intermittent LeakSanitizer | leak at Malloc, nsTArray_base, AppendElements, IPC::ParamTraits


    mccr8, what do you think?

Looks like a fluke. I think it can be P3.

  • 1373418 - Toolkit :: WebExtensions: General - Reduce the amount of compartment overhead in the WebExtension framework

P1 maybe, given that we're going to turn on WebExtensions for all Firefox users? Hopefully JSM merging will fix this.

  • Votes:

  • 1373825 - Core :: JavaScript: GC - Nightly memory use running away over time

Sounds bad, but I don't know if we'll be able to do anything without STR. Maybe one of the other GC issues will fix this.

  • Votes:

    erahm, mccr8, what do you think?

  • 1373895 - Firefox :: Activity Streams: General - Activity Stream leaks a ton of memory

Sounds like this can get marked WFM.

  • Votes:

  • 1374503 - Core :: General - 7+ GB of heap-unclassified on nightly

This sounds vaguely similar to bug 1349991. We really need some better memory reporting for IPC. I filed bug 1381173for that.

  • Votes:

  • 1376013 - Firefox :: Untriaged - Using Firefox for a long time causes it to lag, use over 2.5GB of RAM, and freeze.


  • 1376105 - Toolkit :: Async Tooling - 2.57 - 4.76% Explicit Memory / JS / Resident Memory (linux64, windows10-64-vm, windows7-32-vm) regression on push 6d121d4c2033db893f0c21951f761822d87df960 (Fri Jun 23 2017)

P1 maybe? That is a large regression, and we should at least understand what is happening. It is a very large change so maybe we can't really avoid it.

  • Votes:

    erahm, what do you think?

  • 1376883 - Core :: CSS Parsing and Computation - stylo: Drop thread pool stack size to 100k


  • 1377999 - Core :: DOM - Try out an arena based allocator for DOM nodes again


I don't know if this will really help our memory usage. P2 maybe? I feel like mostly we don't want to regress it.
  • erahm, what do you think?

  • 1378526 - Core :: CSS Parsing and Computation - Measure Stylo memory usage using AWSY tests

dev-memory mailing list