BUG: "error: out of memory" Win32

Index > Compiler Internals > BUG: "error: out of memory" Win32

Author

Thread

FlashBurn

Joined: 06 Jan 2005
Posts: 87

FlashBurn 21 Jun 2005, 16:52

Today I wanted to code some string functions for my os and changes also a little bit my code, but as I wanted to compile my code I got this error! I can still compile small files. The problem is, why are 800MB free RAM not enough?

21 Jun 2005, 16:52

coconut

Joined: 02 Apr 2004
Posts: 326
Location: US

coconut 21 Jun 2005, 21:08

did you try to change the memory size in Options > Compiler setup menu?

21 Jun 2005, 21:08

FlashBurn

Joined: 06 Jan 2005
Posts: 87

FlashBurn 22 Jun 2005, 04:19

Yes, and this also didn´t work and normally I´m using the commandline version of fasm!

22 Jun 2005, 04:19

Tomasz Grysztar

Joined: 16 Jun 2003
Posts: 8356
Location: Kraków, Poland

Tomasz Grysztar 22 Jun 2005, 06:18

You have to show the source that causes the problem.

22 Jun 2005, 06:18

FlashBurn

Joined: 06 Jan 2005
Posts: 87

FlashBurn 22 Jun 2005, 20:44

I will see when I can load it up, because it is a little bit big (109KiB)!

22 Jun 2005, 20:44

f0dder

Joined: 19 Feb 2004
Posts: 3175
Location: Denmark

f0dder 22 Jun 2005, 21:15

flashburn, compress it? Smile

22 Jun 2005, 21:15

coconut

Joined: 02 Apr 2004
Posts: 326
Location: US

coconut 22 Jun 2005, 23:28

commandline also has -m switch to specify memory settings look in the manual

22 Jun 2005, 23:28

FlashBurn

Joined: 06 Jan 2005
Posts: 87

FlashBurn 23 Jun 2005, 10:59

I compressed it, but I have to share my internet connection with my brother and upload is like hell Sad

The -m switch didn´t helped as I tested it with fasmw!

Description:	This is all of the source you need to compile my kernel	Download
Filename:	OS.zip
Filesize:	31.05 KB
Downloaded:	732 Time(s)

23 Jun 2005, 10:59

Tomasz Grysztar

Joined: 16 Jun 2003
Posts: 8356
Location: Kraków, Poland

Tomasz Grysztar 23 Jun 2005, 11:38

The problem is caused by this line in your task_t structure:

Code:

        .name                   rb (256 - .name)

which, when the current offset is above 256, tries to reserve the negative amount of bytes. That was also indeed a flaw in fasm, since it was signaling the wrong error message - I've updated the 1.62 with the fix. Thanks for the report.

23 Jun 2005, 11:38

FlashBurn

Joined: 06 Jan 2005
Posts: 87

FlashBurn 23 Jun 2005, 13:46

OK, but what have I to write that I get the value "256 - offset of .name in the structure"?

23 Jun 2005, 13:46

Tomasz Grysztar

Joined: 16 Jun 2003
Posts: 8356
Location: Kraków, Poland

Tomasz Grysztar 23 Jun 2005, 13:50

In general, the offset of .name is the address of .name minus the address of the structure. You can use some additional label:

Code:

struc foo
 {
   .start:
   .name  rb 256 - (.name-.start)
 }

But since you define also "task_t task_t" at address 0 to get the structure offset, the simplest should be just:

Code:

.name                   rb (256 - task_t.name)

23 Jun 2005, 13:50

vid
Verbosity in development

Joined: 05 Sep 2003
Posts: 7105
Location: Slovakia

vid 23 Jun 2005, 14:21

to the first one

Code:

struc foo{
local ..start
..start:
.name rb 256 - ..start
}

just like first privalov's, but this way you won't get defined additional label task_t.start and won't change namespace for some very special causes.

23 Jun 2005, 14:21

Tomasz Grysztar

Joined: 16 Jun 2003
Posts: 8356
Location: Kraków, Poland

Tomasz Grysztar 23 Jun 2005, 14:56

Even better:

Code:

struc foo
 {
   .:
   .name  rb 256 - (.name-.)
 }

this one doesn't use any additional label at all.

23 Jun 2005, 14:56

< Last Thread | Next Thread >

Forum Rules:

You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot vote in polls in this forum
You cannot attach files in this forum
You can download files in this forum