[LeoD]

Helixhorned, on 16 February 2012 - 09:13 AM, said:

Saving with one CON code and loading into another is highly illegal, however.

I thought the current CONs were saved together with the map?

[Helixhorned]

Sure, they are, but there may be dependencies between data that is not stored in the savegame. Besides, what are you trying to achieve? Loading a vanilla save in DukePlus with the goal of continuing it in DukePlus or vanilla? Both options seem weird to me...

[LeoD]

Helixhorned, on 16 February 2012 - 01:38 PM, said:

Sure, they are, but there may be dependencies between data that is not stored in the savegame. Besides, what are you trying to achieve? Loading a vanilla save in DukePlus with the goal of continuing it in DukePlus or vanilla? Both options seem weird to me...

It happened unintentional while testing some Polymost HRP stuff ingame. But when I noticed I could reproduce it - why not post a crash report then? Instead of calling Dr. Watson EDuke32 should tell me when I'm doing silly things.

[Micky C]

Suddenly when I look in certain directions on the map I'm working on, eduke32 crashes to desktop. Using 2743. Is this an eduke32 bug or perhaps something to do with cons?

0x515c46 : C:\Users\Michael\Documents\DUKE STUFF\AMC TC dev2\eduke32.debug.exe : Z:\var\www\synthesis\eduke32\eduke32\build/src/engine.
c (6216) : in function (drawsprite) 
0x51f188 : C:\Users\Michael\Documents\DUKE STUFF\AMC TC dev2\eduke32.debug.exe : Z:\var\www\synthesis\eduke32\eduke32\build/src/engine.
c (8729) : in function (drawmasks) 
0x411acf : C:\Users\Michael\Documents\DUKE STUFF\AMC TC dev2\eduke32.debug.exe : Z:\var\www\synthesis\eduke32\eduke32/source/game.c (3713) : in function (G_DrawRooms) 
0x4261b2 : C:\Users\Michael\Documents\DUKE STUFF\AMC TC dev2\eduke32.debug.exe : Z:\var\www\synthesis\eduke32\eduke32/source/game.c (10718) : in function (app_main) 
0x5695eb : C:\Users\Michael\Documents\DUKE STUFF\AMC TC dev2\eduke32.debug.exe : Z:\var\www\synthesis\eduke32\eduke32\build/src/winlayer.
c (540) : in function (WinMain@16) 
0x5ed4f2 : C:\Users\Michael\Documents\DUKE STUFF\AMC TC dev2\eduke32.debug.exe : C:\MinGW\msys\1.0\src\mingwrt/../mingw/main.c (73) : in function (main) 
0x4010b9 : C:\Users\Michael\Documents\DUKE STUFF\AMC TC dev2\eduke32.debug.exe : C:\MinGW\msys\1.0\src\mingwrt/../mingw/crt1.c (244) : in function (__mingw_CRTStartup) 
0x40129c : C:\Users\Michael\Documents\DUKE STUFF\AMC TC dev2\eduke32.debug.exe : C:\MinGW\msys\1.0\src\mingwrt/../mingw/crt1.c (287) : in function (atexit) 
Failed to init bfd from (C:\Windows\syswow64\kernel32.dll): 1 1 0
0x75fb339a : C:\Windows\syswow64\kernel32.dll : BaseThreadInitThunk 
Failed to init bfd from (C:\Windows\SysWOW64\ntdll.dll): 1 1 0
0x77469ef2 : C:\Windows\SysWOW64\ntdll.dll : RtlInitializeExceptionChain 
Failed to init bfd from (C:\Windows\SysWOW64\ntdll.dll): 1 1 0
0x77469ec5 : C:\Windows\SysWOW64\ntdll.dll : RtlInitializeExceptionChain 
0x401284 : C:\Users\Michael\Documents\DUKE STUFF\AMC TC dev2\eduke32.debug.exe : C:\MinGW\msys\1.0\src\mingwrt/../mingw/crt1.c (274) : in function (WinMainCRTStartup) 
Thu Jun 07 23:03:05 2012
---------------
0x515c46 : C:\Users\Michael\Documents\DUKE STUFF\AMC TC dev2\eduke32.debug.exe : Z:\var\www\synthesis\eduke32\eduke32\build/src/engine.
c (6216) : in function (drawsprite) 
0x51f188 : C:\Users\Michael\Documents\DUKE STUFF\AMC TC dev2\eduke32.debug.exe : Z:\var\www\synthesis\eduke32\eduke32\build/src/engine.
c (8729) : in function (drawmasks) 
0x4ff75d : C:\Users\Michael\Documents\DUKE STUFF\AMC TC dev2\eduke32.debug.exe : Z:\var\www\synthesis\eduke32\eduke32\build/src/engine.
c (1056) : in function (yax_drawrooms) 
0x4117bd : C:\Users\Michael\Documents\DUKE STUFF\AMC TC dev2\eduke32.debug.exe : Z:\var\www\synthesis\eduke32\eduke32/source/game.c (3680) : in function (G_DrawRooms) 
0x4261b2 : C:\Users\Michael\Documents\DUKE STUFF\AMC TC dev2\eduke32.debug.exe : Z:\var\www\synthesis\eduke32\eduke32/source/game.c (10718) : in function (app_main) 
0x5695eb : C:\Users\Michael\Documents\DUKE STUFF\AMC TC dev2\eduke32.debug.exe : Z:\var\www\synthesis\eduke32\eduke32\build/src/winlayer.
c (540) : in function (WinMain@16) 
0x5ed4f2 : C:\Users\Michael\Documents\DUKE STUFF\AMC TC dev2\eduke32.debug.exe : C:\MinGW\msys\1.0\src\mingwrt/../mingw/main.c (73) : in function (main) 
0x4010b9 : C:\Users\Michael\Documents\DUKE STUFF\AMC TC dev2\eduke32.debug.exe : C:\MinGW\msys\1.0\src\mingwrt/../mingw/crt1.c (244) : in function (__mingw_CRTStartup) 
0x40129c : C:\Users\Michael\Documents\DUKE STUFF\AMC TC dev2\eduke32.debug.exe : C:\MinGW\msys\1.0\src\mingwrt/../mingw/crt1.c (287) : in function (atexit) 
Failed to init bfd from (C:\Windows\syswow64\kernel32.dll): 1 1 0
0x75fb339a : C:\Windows\syswow64\kernel32.dll : BaseThreadInitThunk 
Failed to init bfd from (C:\Windows\SysWOW64\ntdll.dll): 1 1 0
0x77469ef2 : C:\Windows\SysWOW64\ntdll.dll : RtlInitializeExceptionChain 
Failed to init bfd from (C:\Windows\SysWOW64\ntdll.dll): 1 1 0
0x77469ec5 : C:\Windows\SysWOW64\ntdll.dll : RtlInitializeExceptionChain 
0x401284 : C:\Users\Michael\Documents\DUKE STUFF\AMC TC dev2\eduke32.debug.exe : C:\MinGW\msys\1.0\src\mingwrt/../mingw/crt1.c (274) : in function (WinMainCRTStartup) 
Fri Jun 08 00:35:04 2012
---------------

[Helixhorned]

Yikes! It happens when drawing a voxel. This is unexpected because I constructed a map containing all of the AMC TC ones just yesterday and encountered nothing exceptional whatsoever. Have more voxels been added since James gave me a dev snapshot?

[Jblade]

Helixhorned, on 07 June 2012 - 08:58 AM, said:

Yikes! It happens when drawing a voxel. This is unexpected because I constructed a map containing all of the AMC TC ones just yesterday and encountered nothing exceptional whatsoever. Have more voxels been added since James gave me a dev snapshot?

No - it must be a wrong reading or something, I just removed all the vox files from the def files and the crash still occured. I'll send you a package containing the map and the relevant files (I have asked Plagman to set up a AMC SVN, but heard nothing back yet)

EDIT: Ok scratch that just got a PM as I finished this message haha.

[Helixhorned]

Alright, I know what's most likely causing it. Internally, the engine considers sprites with bits 32 and 16 both set as voxels and tried to draw one accordingly when they were set accidentally from CON. Now they'll draw as huge black face sprites, so it should be relatively easy to track them down.

[Jblade]

Excellent, nice work! Glad we can be of assistance by messing things up The game's showing me the offending sprite as you described it (large and with pal 4) so I'm gonna take a look and see what's what with it.

[LeoD]

Weird crashlog:
 eduke32_or_mapster32.crash.log (4.32K)
Number of downloads: 664
 eduke32.log (21.11K)
Number of downloads: 702

Instead of opening a new thread, here comes two compiler warnings I got when fiddling with some GCC options:

source/game.c: In function 'G_GameQuit':
source/game.c:624:8: warning: assuming signed overflow does not occur when assuming that (X + c) < X is always false [-Wstrict-overflow]
source/game.c: In function 'app_crashhandler':
source/game.c:624:8: warning: assuming signed overflow does not occur when assuming that (X + c) < X is always false [-Wstrict-overflow]

src/osd.c: In function 'OSD_Dispatch':...
src/osd.c:1658:8: warning: 'wtp' may be used uninitialized in this function [-Wmaybe-uninitialized]
src/osd.c:1742:26: note: 'wtp' was declared here
src/osd.c:1661:48: warning: 'state' may be used uninitialized in this function [-Wmaybe-uninitialized]
src/osd.c:1742:32: note: 'state' was declared here

[Helixhorned]

LeoD, on 26 September 2012 - 10:50 AM, said:

Weird crashlog:

Thanks, that clears up the seemingly illogical "Abyss crash prevention code" comment in engine.c! Crash should be gone as of r3028, though the real problem is (in this case, floor-aligned) sprites with huge x/y coordinates.

[The Commander]

Might be something fucked up with ATI drivers on my end, but I cant launch eduke32 in 32bit mode. (Polymost or Polymer) 8 bit works fine.
Must also be noted that I havent tried to launch eduke32 in about 2 month but it worked fine last time I tried.

Failed to init bfd from (B:\Windows\system32\atioglxx.dll): 1 1 0
0x67f1302d : B:\Windows\system32\atioglxx.dll : DrvPresentBuffers 
Failed to init bfd from (B:\Windows\system32\atioglxx.dll): 1 1 0
0x67f12313 : B:\Windows\system32\atioglxx.dll : DrvPresentBuffers 
Failed to init bfd from (B:\Windows\system32\atioglxx.dll): 1 1 0
0x675a6dbd : B:\Windows\system32\atioglxx.dll : eglCopyBuffers 
Failed to init bfd from (B:\Windows\system32\atioglxx.dll): 1 1 0
0x675a9bbd : B:\Windows\system32\atioglxx.dll : DrvDescribePixelFormat 
Failed to init bfd from (B:\Windows\system32\opengl32.dll): 1 1 0
0x73cac654 : B:\Windows\system32\opengl32.dll : wglGetProcAddress 
Failed to init bfd from (B:\Windows\system32\opengl32.dll): 1 1 0
0x73cac6b7 : B:\Windows\system32\opengl32.dll : wglGetProcAddress 
Failed to init bfd from (B:\Windows\system32\opengl32.dll): 1 1 0
0x73cb5658 : B:\Windows\system32\opengl32.dll : wglGetPixelFormat 
Failed to init bfd from (B:\Windows\system32\opengl32.dll): 1 1 0
0x73cb5dd4 : B:\Windows\system32\opengl32.dll : wglDescribePixelFormat 
Failed to init bfd from (B:\Windows\system32\opengl32.dll): 1 1 0
0x73cb6559 : B:\Windows\system32\opengl32.dll : wglChoosePixelFormat 
Failed to init bfd from (B:\Windows\syswow64\GDI32.dll): 1 1 0
0x75a658b7 : B:\Windows\syswow64\GDI32.dll : ChoosePixelFormat 
0x5658b6 : B:\Users\Commander\Desktop\Test\eduke32.debug.exe : Z:\var\www\synthesis\eduke32\eduke32\build/src/winlayer.
c (2885) : in function (SetupOpenGL) 
0x56652e : B:\Users\Commander\Desktop\Test\eduke32.debug.exe : Z:\var\www\synthesis\eduke32\eduke32\build/src/winlayer.
c (3253) : in function (CreateAppWindow) 
0x563714 : B:\Users\Commander\Desktop\Test\eduke32.debug.exe : Z:\var\www\synthesis\eduke32\eduke32\build/src/winlayer.
c (1841) : in function (setvideomode) 
0x51db9b : B:\Users\Commander\Desktop\Test\eduke32.debug.exe : Z:\var\www\synthesis\eduke32\eduke32\build/src/engine.
c (10494) : in function (setgamemode) 
0x4243e2 : B:\Users\Commander\Desktop\Test\eduke32.debug.exe : Z:\var\www\synthesis\eduke32\eduke32/source/game.c (10431) : in function (app_main) 
0x560e37 : B:\Users\Commander\Desktop\Test\eduke32.debug.exe : Z:\var\www\synthesis\eduke32\eduke32\build/src/winlayer.
c (540) : in function (WinMain@16) 
0x608b02 : B:\Users\Commander\Desktop\Test\eduke32.debug.exe : C:\MinGW\msys\1.0\src\mingwrt/../mingw/main.c (73) : in function (main) 
0x4010b9 : B:\Users\Commander\Desktop\Test\eduke32.debug.exe : C:\MinGW\msys\1.0\src\mingwrt/../mingw/crt1.c (244) : in function (__mingw_CRTStartup) 
0x40129c : B:\Users\Commander\Desktop\Test\eduke32.debug.exe : C:\MinGW\msys\1.0\src\mingwrt/../mingw/crt1.c (287) : in function (atexit) 
Failed to init bfd from (B:\Windows\syswow64\kernel32.dll): 1 1 0
0x7592339a : B:\Windows\syswow64\kernel32.dll : BaseThreadInitThunk 
Failed to init bfd from (B:\Windows\SysWOW64\ntdll.dll): 1 1 0
0x779a9ef2 : B:\Windows\SysWOW64\ntdll.dll : RtlInitializeExceptionChain 
Failed to init bfd from (B:\Windows\SysWOW64\ntdll.dll): 1 1 0
0x779a9ec5 : B:\Windows\SysWOW64\ntdll.dll : RtlInitializeExceptionChain 
0x401284 : B:\Users\Commander\Desktop\Test\eduke32.debug.exe : C:\MinGW\msys\1.0\src\mingwrt/../mingw/crt1.c (274) : in function (WinMainCRTStartup) 
Thu Sep 27 17:04:01 2012
---------------

[The Commander]

Ah fuck it never mind, I went out and bought a new PSU so I could run my other card correctly

OpenGL Information:
Version: 4.2.0
Vendor: NVIDIA Corporation
Renderer: GeForce GTX 570/PCIe/SSE2

[LeoD]

Mapster32 crash when using the zoom function (seems reproducible).
Maybe it's a known issue but a quick forum search gave no results.

Attached File(s)

•  mapster32.log (1.1K)
  Number of downloads: 680
•  eduke32_or_mapster32.crash.log (1.8K)
  Number of downloads: 656

[Helixhorned]

I couldn't reproduce this yet. Does it happen at a specific place in CBP8.map? Also, what happens without the packs like HRP etc.?

[Micky C]

That used to happen to me all the time, then it stopped, and how it still happens here and there but not consistently. When it does happen it's annoying, of course. Haven't had time to try to narrow down the cause.

[LeoD]

Helixhorned, on 14 November 2012 - 03:57 AM, said:

I couldn't reproduce this yet. Does it happen at a specific place in CBP8.map?

Yes, it seems to be a specific place. It's quite crowded due to some writing on a wall.

Reproduce:
-start mapster32.debug.exe
-open CBP8.map
-zoom out for full map overview
-move to ca. -13900 / 97500 ("Merlijns Spacebar", lower left)
-zoom in very close -> crash

Helixhorned, on 14 November 2012 - 03:57 AM, said:

Also, what happens without the packs like HRP etc.?

Happens as well without autoload contents.

Attached File(s)

•  mapster32.log (746bytes)
  Number of downloads: 633
•  eduke32_or_mapster32.crash.log (1.8K)
  Number of downloads: 634
•  mapster32.cfg.zip (2.62K)
  Number of downloads: 526

[Helixhorned]

This one's really creepy: I managed to reproduce it exactly once in GDB, understood the cause (an overflow) and hopefully fixed it r3188, but couldn't get that code path to be hit again to be sure.
Excellent bug report, as usual.

[LeoD]

Helixhorned, on 15 November 2012 - 01:16 PM, said:

This one's really creepy: I managed to reproduce it exactly once in GDB, understood the cause (an overflow) and hopefully fixed it r3188, but couldn't get that code path to be hit again to be sure.

Fixed as far as I'm concerned.

[Kyanos]

That's awesome!!! I was just coming here to report the same crash. It happens to me a lot mapping with WGR, I thought it was just the mod and dealt with it, but finally had enough. Glad to see it's fixed.

Sorry, it's still doing it for me.

http://pastebin.com/JgZMumTX

[aethius]

Hi everybody! I tried searching for this bug but I can't find any solution on the forums.
My problem is. When I try to load game I get this crash.

http://img713.images...949/errorqw.png

I have legit Duke Nukem 3D: Atomic Edition from GOG.com. And I tried re-instaling the game and copying all the eduke32 files in the folder.
Anybody knows what the problem is? MAybe I need to update the game with some patch? Thank you in advance.

EDIT: found the topic now.

[LeoD]

Random Mapster32 crashlog r3255. Can't remember the crash, just found the file.
Plus an assertion. Are those of any use in terms of bugfixing?

Attached thumbnail(s)

• 

Attached File(s)

•  eduke32_or_mapster32.crash.log (1.76K)
  Number of downloads: 605

[Helixhorned]

LeoD, on 15 December 2012 - 10:47 AM, said:

Random Mapster32 crashlog r3255. Can't remember the crash, just found the file.
Plus an assertion. Are those of any use in terms of bugfixing?

The expression

(unsigned)sprite[i].sectnum < (unsigned)numsectors && (unsigned)sprite[i].statnum < MAXSTATUS

bound-checks the newly inserted sprites' sectnum and statnum, so the map that was saved with the undo/redo system must have been corrupt.

(

(unsigned)x < BOUND

is just a cute way of writing

x >= 0 && x < BOUND

)

[LeoD]

Random crashlog (hope it's useful):
 eduke32_or_mapster32.crash.log (2K)
Number of downloads: 554  eduke32.log (10.38K)
Number of downloads: 594
The frame name errors don't matter, the models are redefined later.

[LeoD]

The debug executables appear to be less stable than the regular ones ... regarding Nightshade Army at least (reproducible).
 eduke32_or_mapster32.crash-A.log (1.62K)
Number of downloads: 550
 eduke32-A.log (1.73K)
Number of downloads: 566
 eduke32_or_mapster32.crash-B.log (1.93K)
Number of downloads: 527
 eduke32-B.log (1.67K)
Number of downloads: 552

Playing some other maps will result in assertions (debug builds only, reproducible) like this:

[Hendricks266]

IIRC, assertions are only enabled for debug builds. Either way, they warn us of a larger problem.

[Helixhorned]

LeoD, on 20 March 2013 - 02:08 PM, said:

The debug executables appear to be less stable than the regular ones ... regarding Nightshade Army at least (reproducible).

Thanks for the bug report! Fixed in SVN HEAD. That debug executables crash on more occasions doesn't mean that the release ones are more stable: they have the same bugs as the debug ones, but they may manifest themselves in different ways (or not at all, but it doesn't mean it's not there). C is an unsafe language: if you do bad stuff like accessing arrays out of their bounds, any behavior is permitted, like invoking nasal demons. Crashes are really a courtesy by the OS, telling the programmer that there's something wrong.

Quote

Playing some other maps will result in assertions (debug builds only, reproducible) like this:
assertion-1A.TRAIN.MAP.jpg

Interesting, that one is happening in the ray-intersection routine. It's a bit delicate that all the geometric calculations are written using fixed point (integer) types, so there's probably some overflow happening somewhere. Which maps are those and at which coordinates does the assertion fail?

[LeoD]

Helixhorned, on 22 March 2013 - 02:34 AM, said:

Which maps are those and at which coordinates does the assertion fail?

Cathedral: Fly up to the ledge at 39000/17000, walk up to the shotgun ammo and turn right around.
Infestation in Time Revisited - Train: Right at the beginning turn left, get the health pack and look around.
Both give the same assertion message as posted above. Use the respective command line parameters from the attached log files.

Attached File(s)

•  eduke32-CATH.log (2.81K)
  Number of downloads: 566
•  eduke32-IITTC-TRAIN.log (4.53K)
  Number of downloads: 559
•  CFG Files.zip (3.44K)
  Number of downloads: 465
•  eduke32-IITTC-TRAIN.crash.log (3.28K)
  Number of downloads: 535

[LeoD]

Battlefield 3 - Tamachazam
eduke32.debug.exe -s3 -c3 -server -map tamachaz.map -m : Crashes as soon as you start moving.
May related to this compiler warning:

src/osd.c: In function 'OSD_Dispatch':ex.o...
src/osd.c:1671:8: warning: 'wtp' may be used uninitialized in this function [-Wmaybe-uninitialized]
src/osd.c:1755:26: note: 'wtp' was declared here
src/osd.c:1674:48: warning: 'state' may be used uninitialized in this function [-Wmaybe-uninitialized]
src/osd.c:1755:32: note: 'state' was declared here

Another warning:

source/game.c: In function 'G_GameQuit':angry:...
source/game.c:614:8: warning: assuming signed overflow does not occur when assuming that (X + c) < X is always false [-Wstrict-overflow]
source/game.c: In function 'app_crashhandler':
source/game.c:614:8: warning: assuming signed overflow does not occur when assuming that (X + c) < X is always false [-Wstrict-overflow]

Using MinGW gcc 4.7.2

Attached File(s)

•  eduke32_or_mapster32.crash.log (2.25K)
  Number of downloads: 561
•  eduke32.log (2.05K)
  Number of downloads: 528

[Helixhorned]

LeoD, on 22 March 2013 - 01:09 PM, said:

Infestation in Time Revisited - Train: Right at the beginning turn left, get the health pack and look around.
Both give the same assertion message as posted above. Use the respective command line parameters from the attached log files.

EDuke32 just got a little more correct. Nice!

LeoD, on 24 March 2013 - 08:19 AM, said:

Battlefield 3 - Tamachazam
eduke32.debug.exe -s3 -c3 -server -map tamachaz.map -m : Crashes as soon as you start moving.

Caused by net.c:1098, or rather, the wrong order of the blocks at lines 1096 and 1103.

[Forge]

extreme circumstances, but it's reproducible.
walking through vent systems is where this is normally encountered

walk out on top of a floor aligned vent grate sprite/fan sprite, crouch and look all the way down, shoot or kick the sprite to break it, game crashes when duke starts to fall through the sprite

(must be standing directly on the sprite to increase frequency of crashes - otherwise looking all the way down and crouch-walking forward to fall produces a more random result)


edit: kinda funny that .crashlog and .map extensions aren't authorized attachments
x2 files removed