#6697 closed defect (invalid)
[macOS] 0ad drives up memory and CPU usage making the computer unusable
| Reported by: | Langbart | Owned by: | |
|---|---|---|---|
| Priority: | Should Have | Milestone: | |
| Component: | Core engine | Keywords: | |
| Cc: | Patch: |
Description (last modified by )
- I have built
spidermonkey 91on the 11/jan/23 (GitHub) on commit [27409]- I did not notice any performance issues, while debugging and playing for over a week
- on the
18/Jan/23I did a rebuild of spidermonkey, ever since then I noticed my RAM is being filled up by some processes when the game crashes or throws errors or I just play the game, they are calledReportCrash,spindumpandsandboxed- killing those processes manually solves the issue and my RAM reaches normal levels again
- had no problems with such processes in the past
- in the last 24hrs I needed to reboot my PC 4x, that is the reason for the ticket.
Note: The steps for reproduction could not be clearly determined. This is not a good bug report. Think of it more as documentation of a problem that is not yet fully understood.
questions
- is it possible that 0ad is the cause of those processes mentioned above to eat my memory and CPU?
- is it possible the issue was introduced between
10-18/Jan/23?- list of commits that touched the library folder
❯ git log HEAD...HEAD~10 --oneline --format="%cd %h %an %s" libraries/ 17/Jan/23 f76a92f4d0 Itms Fix and cleanup the SM91 build, fixes #6676. 14/Jan/23 cdfe6cc87e Stan Fix typo noticed by @Itms Refs #6676 Phab:rP27409 Differential Revision: https://code.wildfiregames.com/D4888 14/Jan/23 56690ca834 Stan Fix a bug with Python >= 3.11 where -rU was removed as an open flag. Patch by: @sera Refs #6676 Phab:rP27409 Differential Revision: https://code.wildfiregames.com/D4888 14/Jan/23 255d879edb Stan Fix a case sensitive issue on macOS. macOS usually do not use a case sensitive filesystem, hence why it got unnoticed. Reported by: @Freagarach Refs: Phab:rP27409 12/Jan/23 a9e965f2cb Stan Commit the Spidermonkey tarball, as it puts extra strain on CI, and package manager builds don't have internet. Reported by: @Norse_Harold 10/Jan/23 9513e7eb79 Stan [SM91] Update to Spidermonkey 91.1.3 APIs Fixes: #5986 Patch by: @wraitii Comments by: @nwtour, @Stan
image of ReportCrash process
video of spindump process
- https://ttm.sh/SqL.8x_1500p (link will be dead within a year)
related articles
Attachments (5)
Change History (11)
by , 15 months ago
by , 15 months ago
| Attachment: | game_19_Jan_23.png added |
|---|
by , 15 months ago
| Attachment: | ram_19_Jan_23.png added |
|---|
by , 15 months ago
| Attachment: | replay_19_Jan_23.txt added |
|---|
comment:2 by , 15 months ago
| Description: | modified (diff) |
|---|
Update description, add the case when simply playing the game
comment:3 by , 15 months ago
Reproducible steps: This is reproducible over and over again, "spindump" accumulates huge amounts of memory.
Date: 20/Jan/23 ~3:00
Action: Watching a replay, with a debug message being placed in chatHelper.js, like this:
-
binaries/data/mods/public/simulation/ai/petra/chatHelper.js
a b PETRA.chatNewDiplomacy = function(gameState, player, newDiplomaticStance) 220 220 221 221 PETRA.chatAnswerRequestDiplomacy = function(gameState, player, requestType, response, requiredTribute) 222 222 { 223 warn(uneval(['debug', gameState, player, requestType, response, requiredTribute])) 223 224 Engine.PostCommand(PlayerID, { 224 225 "type": "aichat", 225 226 "message": "/msg " + gameState.sharedScript.playersData[player].name + " " +
- start the replay, wait for it to freeze
- check activity monitor for
spindump
Replay
Screenshots
by , 15 months ago
| Attachment: | mem_20_Jan_23.png added |
|---|
comment:4 by , 15 months ago
| Priority: | Should Have → Release Blocker |
|---|
follow-up: 6 comment:5 by , 15 months ago
Does the in-game profiler (f11) show anything? Does profiler2 ? Maybe it's related to https://trac.wildfiregames.com/ticket/6193 ? Maybe we have a memory leak?
comment:6 by , 15 months ago
| Milestone: | Alpha 27 |
|---|---|
| Priority: | Release Blocker → Should Have |
| Resolution: | → invalid |
| Status: | new → closed |
Note:
See TracTickets
for help on using tickets.
I will just document when it happens and what I did.
Date: 19/Jan/23 ~13:00
Action: Playing a game with 4 AI's on a tiny random map, at about 25min the game starts having some slow downs, about 30 seconds later it's unbearable, but I managed to take a screenshot of the RAM before forcing a restart.
System info
Replay
Screenshots