View Issue Details

IDProjectCategoryView StatusLast Update
0002278Spring engineGeneralpublic2011-01-23 01:27
Reporterdjmad Assigned Tozerver  
PrioritynormalSeveritycrashReproducibilityalways
Status resolvedResolutionfixed 
Product Version0.82.7 
Fixed in Version0.82.7+git 
Summary0002278: Crash if in a AI-Game one of the Bots is destroyed (only in MT)
DescriptionIts not on every of my Computers, only on the one in company

First bot got killed, no problem
Second instacrash

Trace is included

System Info:
8 Threads
QuadroFX 4800
8GB Ram
Additional Information[f=0021971] Removed: LuaUI/Widgets/gui_build_eta.lua
[f=0023295] Removed: LuaUI\Widgets\auto_reclaim_heal_assist.lua
[f=0027170] Team0 (invalid) flatlined
[f=0027170] GML error: Sim thread called gmlDeleteTextures (Unknown)
[f=0027170] GML error: Sim thread called gmlDeleteTextures (Unknown)
[f=0027170] GML error: Sim thread called gmlDeleteTextures (Unknown)
[f=0027171] Skirmish AI "Bot1", which controlled team 0 is now dead
[f=0037569] Slasher is being attacked
[f=0037673] Slasher is being attacked

[f=0043033] Slasher is being attacked
[f=0043136] Slasher is being attacked
[f=0043266] Team1 (invalid) has been played out
[f=0043266] GML error: Sim thread called gmlDeleteTextures (LUA)
[f=0043266] Spring 0.82.7.0 (0.82.7 MT-Sim) has crashed.
[f=0043266] MT with 8 threads.
[f=0043266] Spring 0.82.7.0 (0.82.7 MT-Sim) has crashed.
[f=0043266] MT with 8 threads.
[f=0043266] Exception: Access violation (0xc0000005)
[f=0043266] Exception Address: 0x00b58c98
[f=0043266] DLL information:
[f=0043266] Exception: Unknown exception (0xc0000029)
[f=0043266] Stacktrace:
[f=0043266] Exception Address: 0x778504be
[f=0043266] DLL information:
[f=0043266] Stacktrace:
[f=0043266] (0) Unknown [0x778504BE]
[f=0043266] (1) Unknown [0x770107F0]
[f=0043266] (2) Unknown [0x0028FFC4]
[f=0043266] (3) Unknown [0x770107F0]
[f=0043266] (4) Unknown [0x0D07F058]
TagsNo tags attached.
Attached Files
infolog.txt (Attachment missing)
Checked infolog.txt for Errors

Activities

hoijui

2011-01-05 10:01

reporter   ~0006194

It seems more likely that this is a GFX or Sound problem, then an AI one, as it does not happen on all PCs.
for the record, he is using 7 * "E323AI 3.25.0".

most easy thing to try, would be disabling sound in
"C:\Users\djmad.MARHOLDGMBH\AppData\Local\springsettings.cfg"
with the setting (will be 0 for you):
NoSound=1

(slogic or me should probably try to reproduce this)

djmad

2011-01-07 02:06

reporter   ~0006205

Last edited: 2011-01-07 12:53

as hoijui proposed i tried again with the nosound=1 setting

same error again

[f=0064130] Team0 (invalid) was dominated
[f=0064133] Commanders Remaining: 1
[f=0064189] GML error: Sim thread called gmlDeleteTextures (Unknown)
[f=0064189] LuaRules::RunCallIn: error = 2, DrawWorld, RenderSequence' (a [string "lups/lups.lua"] a nil value
[f=0064189] Spring 0.82.7.0 (0.82.7 MT-Sim) has crashed.
[f=0064189] MT with 4 threads.
[f=0064189] Exception: Unknown exception (0xc0000029)
[f=0064189] Exception Address: 0x77d804be
[f=0064189] DLL information:
[f=0064189] Stacktrace:
[f=0064189] (0) Unknown [0x77D804BE]
[f=0064189] (1) Unknown [0x777107F0]
[f=0064189] (2) Unknown [0x0028FFC4]
[f=0064189] (3) Unknown [0x777107F0]
[f=0064189] (4) Unknown [0x0845EE48]


it seems it only happens in multithreaded Spring, not in single threaded
in this case it was a 4 thread process which was running at a different computer
4 E323ai??s

tested now 4 games, in every game 4 ai??s (2x multithread, 2x singlethread)
always nosound=1

at the destruction of a team, the single thread app is continuing, the multithread is crashing.

Sometimes after first KI destruction spring stays alive (may 50% chance), after the secound destruction it crashes for 100%, may the deletion of a AI-Thread is causing the crash

slogic

2011-01-13 10:37

reporter   ~0006258

djmad, they are not related. Also AIs are run in sequence. They are not multithread-aware. I do not know what exactly functions are run in parallel on multithreaded Spring but I'm sure it is not AIs.

Stacktrace is useless for me.

djmad

2011-01-14 00:14

reporter   ~0006259

Last edited: 2011-01-14 00:17

Slogic, thx for this info
I deleted the note about the 0001927 bug
In this case the dice is rolling back to spring-mt coders

I have Much time for testing so if I can help in any way just tell me what to try

dfreeman

2011-01-19 14:17

reporter   ~0006272

I believe I have experienced this several times myself. I usually use E323ai.

I am running Linux with ATI, and have had many problems. Is it possible that your one machine with the problem is also running ATI?

djmad

2011-01-19 14:39

reporter   ~0006273

yes i have in both of my computers ATI cards,

but the issue only is reproduceable if i run multithreaded Spring, not in a singlethread Spring

dfreeman

2011-01-19 14:51

reporter   ~0006275

I usually run multithreaded. I'm not sure if it has happened to me using singlethreaded.

djmad

2011-01-19 15:05

reporter   ~0006276

may you can test this, so we can exclude it, until now this is a Multithread-issue

thx

zerver

2011-01-23 01:27

reporter   ~0006290

https://github.com/spring/spring/commit/9391d8f99978bcc4a07be9e90e39d89f352212fa

Issue History

Date Modified Username Field Change
2011-01-04 16:58 djmad New Issue
2011-01-04 16:58 djmad File Added: infolog.txt
2011-01-05 10:01 hoijui Note Added: 0006194
2011-01-07 02:06 djmad Note Added: 0006205
2011-01-07 02:41 djmad Note Edited: 0006205
2011-01-07 12:53 djmad Note Edited: 0006205
2011-01-07 17:13 abma Summary Crash if in a AI-Game one of the Bots is destroyed => Crash if in a AI-Game one of the Bots is destroyed (only in MT)
2011-01-13 10:37 slogic Note Added: 0006258
2011-01-14 00:14 djmad Note Added: 0006259
2011-01-14 00:17 djmad Note Edited: 0006259
2011-01-19 14:17 dfreeman Note Added: 0006272
2011-01-19 14:39 djmad Note Added: 0006273
2011-01-19 14:51 dfreeman Note Added: 0006275
2011-01-19 15:05 djmad Note Added: 0006276
2011-01-22 17:25 zerver Status new => assigned
2011-01-22 17:25 zerver Assigned To => zerver
2011-01-23 01:27 zerver Note Added: 0006290
2011-01-23 01:27 zerver Status assigned => resolved
2011-01-23 01:27 zerver Fixed in Version => 0.82.7+git
2011-01-23 01:27 zerver Resolution open => fixed