Duke4.net Forums: This usermap crashes on load - Duke4.net Forums

Jump to content

Page 1 of 1
  • You cannot start a new topic
  • You cannot reply to this topic

This usermap crashes on load  "no log"

User is offline   necroslut 

#1

EDuke32 crashed quietly upon loading this map ("E1L2" from Alien Invasion). The loading screen gets displayed, but it doesn't look to me like the progress bar starts moving. There's nothing out of the ordinary in the logs.
I've tried the map in DOS Duke (1.3D and 1.5), as well as an older EDuke (2018), and it works fine there. It's a TC map, but skipping the arts/cons make no difference to whether it crashes or not in any of the versions. I've also checked it in Mapster and there doesn't seem to be any corruptions.
Is this some known issue? I'm attaching the map file below.

Edit: running it in software mode it actually starts, running for a second or two before crashing. Still nothing that looks interesting in the logs (I've tried the debug build too). There's some GL warning earlier, no idea if it's related.
(" debug: DwmGetCompositionTimingInfo() FAILED! HRESULT: 0x80263001 (Unknown error) ")

Attached File(s)

  • Attached File  E2L2.zip (50.68K)
    Number of downloads: 227


This post has been edited by necroslut: 19 April 2020 - 06:15 AM

0

User is offline   LeoD 

  • Duke4.net topic/3513

#2

View Postnecroslut, on 19 April 2020 - 06:06 AM, said:

[...] nothing that looks interesting in the logs (I've tried the debug build too).
Without the [crash] logs your issue report doesn't look interesting either...
0

User is offline   necroslut 

#3

View PostLeoD, on 19 April 2020 - 06:23 AM, said:

Without the [crash] logs your issue report doesn't look interesting either...

There's no "crash log" since nothing gets printed to it after the map loads. Here it is, anyway...

Attached File(s)


0

User is offline   LeoD 

  • Duke4.net topic/3513

#4

View Postnecroslut, on 19 April 2020 - 07:17 AM, said:

There's no "crash log" since nothing gets printed to it after the map loads. Here it is, anyway...
Did you forget the debug DLL (ebacktrace1-64.dll)?
Just a few lines above this thread there's something you might want to read.
However, issue confirmed. My [crash]logs look like this:
Attached File  eduke32-E2L2-synthesis.log (2.29K)
Number of downloads: 164
Attached File  eduke32-E2L2-synthesis.crash.log (1.38K)
Number of downloads: 181
Attached File  eduke32-E2L2-leod.log (1.33K)
Number of downloads: 183
Attached File  eduke32-E2L2-leod.crash.log (2.4K)
Number of downloads: 168
1

User is offline   necroslut 

#5

View PostLeoD, on 19 April 2020 - 07:46 AM, said:

Did you forget the debug DLL (ebacktrace1-64.dll)?
Just a few lines above this thread there's something you might want to read.

Yes, I did... There's been a few years since I reported a bug. So, confirmed? Do you need any more information?
0

User is offline   LeoD 

  • Duke4.net topic/3513

#6

View Postnecroslut, on 19 April 2020 - 07:57 AM, said:

Yes, I did... There's been a few years since I reported a bug.
I guess the debug DLL needs an extra mention in the Crash section of that wiki page. Done.

View Postnecroslut, on 19 April 2020 - 07:57 AM, said:

So, confirmed?
Definitely.

View Postnecroslut, on 19 April 2020 - 07:57 AM, said:

Do you need any more information?
Nope. It's not for me, btw., but for the actual developers. The better you deliver, the greater the chance for a fix.

Bisect-compiled to r7813-a78e49c23, aka r7820 in the old Subversion repository. Thanks to oasiz for pointing me into the right direction. ;)
Attached File  eduke32-r7813-a78e49c23.log (1.42K)
Number of downloads: 157
Attached File  eduke32-r7813-a78e49c23.crash.log (2.74K)
Number of downloads: 184

This post has been edited by LeoD: 19 April 2020 - 12:28 PM

0

User is offline   necroslut 

#7

View PostLeoD, on 19 April 2020 - 12:08 PM, said:

I guess the debug DLL needs an extra mention in the Crash section of that wiki page. Done.
Nope. It's not for me, btw., but for the actual developers. The better you deliver, the greater the chance for a fix.

Bisect-compiled to r7813-a78e49c23, aka r7820 in the old Subversion repository. Thanks to oasiz for pointing me into the right direction.

Ah, I got the impression you were a developer.
I was prepared to bisect and shit, but I wanted to know if it was a known issue already before I wasted the time.
For myself, I already "solved" it by using an older version.

This post has been edited by necroslut: 19 April 2020 - 08:57 PM

0

User is offline   LeoD 

  • Duke4.net topic/3513

#8

View Postnecroslut, on 19 April 2020 - 08:56 PM, said:

Ah, I got the impression you were a developer.
Not for software. But I know my way around hunting issues via brute force methods like bisecting, or the printf() debugger.
0

User is offline   Ninety-Six 

#9

View PostLeoD, on 20 April 2020 - 01:55 AM, said:

Not for software. But I know my way around hunting issues via brute force methods like bisecting, or the printf() debugger.


Out of curiosity, what is bisecting? (and from my thread, trisecting?)
0

User is offline   LeoD 

  • Duke4.net topic/3513

#10

View PostNinety-Six, on 20 April 2020 - 03:17 AM, said:

Out of curiosity, what is bisecting? (and from my thread, trisecting?)


View PostLeoD, on 19 April 2020 - 07:46 AM, said:

Just a few lines above this thread there's something you might want to read.
;)
0

User is offline   LeoD 

  • Duke4.net topic/3513

#11

It seems that r7813-a78e49c2 / svn7820 was a bad commit. It started causing a Mapster32 crash as well, which I've reported last month.
0

User is offline   necroslut 

#12

Ok, this seems to be fixed now.
0

Share this topic:


Page 1 of 1
  • You cannot start a new topic
  • You cannot reply to this topic


All copyrights and trademarks not owned by Voidpoint, LLC are the sole property of their respective owners. Play Ion Fury! ;) © Voidpoint, LLC

Enter your sign in name and password


Sign in options