View Issue Details
| ID | Project | Category | View Status | Date Submitted | Last Update |
|---|---|---|---|---|---|
| 0001937 | Spring engine | General | public | 2010-06-04 20:52 | 2010-06-28 19:48 |
| Reporter | hoijui | Assigned To | Kloot | ||
| Priority | normal | Severity | crash | Reproducibility | have not tried |
| Status | resolved | Resolution | fixed | ||
| Product Version | 0.81.2+git | ||||
| Target Version | 0.82.0 | Fixed in Version | 0.81.0.0+git | ||
| Summary | 0001937: crash in CMiniMap::GetUnitIcon | ||||
| Description | [ 62511] Exception: Access violation (0xc0000005) [ 62511] Exception Address: 0x004c75bf translated stack-trace: http://buildbot.eat-peet.net/spring/stacktrace/%5Bno_new_cpp_ais_4%5Dpureint_spring_stktrc_0.81.2.1-1272-g30a72ba_1275593413.html looks like an other CUnit == NULL issue, though it is not a SIGSEGV. how would we find the source of all these? we had quite a few of these lately, and putting != NULL tests everywhere does not seem like the optimal way to fix this. Side-note: There was a sync error in the game already (most likely still caused by pathing by AIs :/ ). will investigate more when i have time, and open an other report. | ||||
| Tags | No tags attached. | ||||
| Attached Files | |||||
| Checked infolog.txt for Errors | |||||
|
|
an otherone: http://buildbot.eat-peet.net/spring/stacktrace/[no_new_cpp_ais_4]pureint_spring_stktrc_0.81.2.1-1322-gce3b201_1275939765.html same class, different method, and again CUnit involved. |
|
|
It is not a NULL unit, since unit is being dereferenced multiple times before that, and there is also checks e.g. in CMiniMap::DrawUnit. Could be some memory corruption though. |
|
|
yeah. must have been blind the first time ... so what do we have: * Two crashes in different points in "void CMiniMap::DrawUnit(const CUnit* unit)" * it looks like "unit" is corrupted (memory) * it is NOT a GML version of spring * the call seems to come from within Lua, but i guess that does not matter i guess... we just have to try to reproduce, running in gdb, and when it happens, look at addresses involved? |
|
|
got yet an other crash that might be related, this time in master (rest was in pureint). see file infolog_UnitDrawer_memCor.txt |
|
|
So... do these crashes occur without AI/bots? |
|
|
i did not test that, but i am pretty sure that yes: they also seem to happen in master, and there were really few/small changes only in the AI Interface there. also, i tested once with only E323AI, and the other time with RAI, KAIK and AAI. So it is unlikely that one of the AIs is to blame. but we will only really know when we do human only test games of course. |
| Date Modified | Username | Field | Change |
|---|---|---|---|
| 2010-06-04 20:52 | hoijui | New Issue | |
| 2010-06-07 21:44 | hoijui | Note Added: 0004955 | |
| 2010-06-07 21:44 | hoijui | Target Version | => 0.82 |
| 2010-06-10 15:13 | zerver | Note Added: 0004966 | |
| 2010-06-11 17:17 | hoijui | Note Added: 0004973 | |
| 2010-06-18 23:09 | hoijui | Relationship added | related to 0001952 |
| 2010-06-21 14:37 | hoijui | File Added: infolog_UnitDrawer_memCor.txt | |
| 2010-06-21 14:38 | hoijui | Note Added: 0004998 | |
| 2010-06-21 18:18 | zerver | Note Added: 0005001 | |
| 2010-06-21 18:41 | hoijui | Note Added: 0005002 | |
| 2010-06-28 19:48 | Kloot | Relationship added | related to 0001966 |
| 2010-06-28 19:48 | Kloot | Status | new => resolved |
| 2010-06-28 19:48 | Kloot | Fixed in Version | => 0.81.0.0+git |
| 2010-06-28 19:48 | Kloot | Resolution | open => fixed |
| 2010-06-28 19:48 | Kloot | Assigned To | => Kloot |
| 2010-09-03 00:55 | Kloot | Relationship added | related to 0002074 |
