Dream Maker crashes at end of compilation (with crash data) in Bug Reports

ID:85711
 
Nov 15 2009, 8:25 am (Edited on Nov 15 2009, 1:22 pm)
BYOND Version:458
Operating System:Windows XP Pro
Web Browser:Firefox 3.5.5
Status: Unverified
Thus far we've been unable to verify or reproduce this bug. Or, it has been observed but it cannot be triggered with a reliable test case. You can help us out by editing your report or adding a comment with more information.
Descriptive Problem Summary:

Dream Maker crashes at the last phase of compilation, although it seems to have compiled fully since any changes made to code or resource is applied when running the project after compiling.
(the project to compile is very large)

Crash information:
http://nielsdg.webs.com/compile%20crash.jpg
http://nielsdg.webs.com/ed8a_appcompat.txt

Does the problem occur:
Every time? Or how often? It started happening at a certain point of time in the past.
For other projects? Not that I know of.
In other user accounts? Sort of irrelevant.
On other computers? Well, one of the other developers of the project claims it doesn't occur on his computer.

When does the problem NOT occur?

When compiling a different project

Did the problem NOT occur in any earlier versions? If so, what was the last version that worked? (Visit http://www.byond.com/download/build to download old versions for testing.)

It has happened in every version since the problem first occurred.

Workarounds:

Ignoring it the crash, although it's never appealing to see any kind of crash, and one big issue this crash brings, is that I can't jump to any code errors/warnings after compiling. I can't even see all of the detected errors when there are more than the error field fits.

#1 Nov 16 2009, 2:04 pm
It has happened in every version since the problem first occurred.

Can you clarify that please? Which version was the earliest to show the problem?

I traced the address you provided in the crash data (thank you for including that info!) and found it happens in a mostly innocuous place in the core, right after a memory allocation. This makes me wonder if the problem is that you're running out of memory. (How much RAM does your system have?)

I'll need a copy of your project to see if I can produce the problem on my own system. If it's a memory issue I likely won't have any problem with it, but it's important to find out for sure. You can send me the source in a .zip file at LummoxJR@aol.com. Until I can test and verify this issue myself though (the latter of which might be unlikely), I'll mark this issue unverified.
#2 Nov 16 2009, 3:26 pm
My system has 1Gb of RAM, which should be enough, I think...

I can't recall in which version of BYOND the crash did not occur; it doesn't really seem to be a version-related issue, it started occurring after a session of editing (code and map) of the source, not after a BYOND update.

I'll send a copy of the source soon.
#3 Nov 16 2009, 3:49 pm
Mail sent.

It would be surprising to me if the crash is caused by my system running out of memory, because I've done a quick check and during compiling, the memory usage only increased from 580 Mb to 660 Mb or so. (out of 2461 Mb max, including the 1.4 Gb of pagefile)

Could there be a memory leak or something alike?
#4 Nov 16 2009, 4:03 pm
Oh and, sometimes - but very rarely - it compiles without crashing, but then Dream Maker crashes as soon as I close it, as in, the same error message pops up ( http://nielsdg.webs.com/compile%20crash.jpg )
#5 Nov 17 2009, 9:12 am
I was able to compile your source without problems. The memory usage of the compiler never seemed to go above 100M during the first compile, and hovered around 83-84 after compiling. Compiling a second time in the same session did not increase the memory usage after it was finished, so I think a memory leak is unlikely.
#6 Nov 17 2009, 11:00 am
Lummox JR wrote:
I was able to compile your source without problems. The memory usage of the compiler never seemed to go above 100M during the first compile, and hovered around 83-84 after compiling. Compiling a second time in the same session did not increase the memory usage after it was finished, so I think a memory leak is unlikely.

I see...
Do you have any insight in how memory works? If so, think you can theorize what's wrong in my system that could be causing this crash on my end? Based on what I wrote in post #3 and #4? (and of course, the crash data)