View Issue Details

IDProjectCategoryView StatusLast Update
0003427Dwarf FortressTechnical -- Generalpublic2014-08-04 09:51
Reportercodeslob Assigned Touser6 
PrioritynormalSeveritycrashReproducibilityhave not tried
Status resolvedResolutionunable to reproduce 
PlatformWindowsOSXPOS VersionSP3
Product Version0.31.16 
Summary0003427: Seg fault attempting to execute code at 0x67656953 (bogus address)
DescriptionStack trace below. The disassembly shows it has just performed a dynamic jump through EAX, which looks more like a string "geiS" than an address.

> 67656953()
     Dwarf Fortress.exe!007dfdd5()
     [Frames below may be incorrect and/or missing, no symbols loaded for Dwarf Fortress.exe]
     msvcr100.dll!78ab0269()
     msvcr100.dll!78ab02d1()
     msvcr100.dll!78ab02d1()
     msvcr100.dll!78ab016a()
     Dwarf Fortress.exe!007e2fcd()
     Dwarf Fortress.exe!007a839a()
     Dwarf Fortress.exe!0053bedf()
     kernel32.dll!7c80262a()
     kernel32.dll!7c802542()
     SDL.dll!6813e35f()
     SDL.dll!6813e3d7()
     Dwarf Fortress.exe!00567962()
     Dwarf Fortress.exe!00566bd5()
     SDL.dll!6810a2ec()
     SDL.dll!6813e4c9()
     msvcrt.dll!77c3a3b0()
     kernel32.dll!7c80b729()

Registers:

EAX = 67656953 EBX = 00000015 ECX = 207D6178 EDX = 00A20D60 ESI = 207D6178 EDI = 00000040 EIP = 67656953 ESP = 01FBF8FC EBP = 0000000D EFL = 00000202

EAX was previously calculated from table at [EDX+0x314] if that helps. Probably not.

Steps To ReproducePlay game.
Watch it crash.
Not doing anything particularly special at the time.
Additional InformationI can provide a mini dump if that helps.
TagsNo tags attached.

Activities

Logical2u

2010-10-15 13:25

manager   ~0013381

Why don't you try the legacy version of DF too? Unfortunately this type of bug report isn't very helpful (at least to me, I haven't heard Toady chime in on stuff like this) unless you can narrow it down to a particular problem...

codeslob

2010-10-15 15:50

reporter   ~0013382

Legacy version?

And narrow down what? In my experience this kind of random crash can sometimes take about five minutes to find with a stack trace and access to source code and map file for the particular version of the executable but is pretty much impossible to track down otherwise..

Issue History

Date Modified Username Field Change
2010-10-15 07:44 codeslob New Issue
2010-10-15 13:25 Logical2u Note Added: 0013381
2010-10-15 15:50 codeslob Note Added: 0013382
2014-08-04 09:51 user6 Status new => resolved
2014-08-04 09:51 user6 Resolution open => unable to reproduce
2014-08-04 09:51 user6 Assigned To => user6