[Wormux-gna] [bug #17738] Fails to build on s390

2011-02-26 Thread Kurosu

Update of bug #17738 (project warmux):

  Status:None => Fixed  
 Open/Closed:Open => Closed 

___

Follow-up Comment #5:

Thanks. Due to the number and different nature of build systems/architectures
it was tested on, I'm closing it as fixed.

___

Reply to this item at:

  

___
  Message posté via/par Gna!
  http://gna.org/


___
Wormux-gna mailing list
wormux-...@gna.org
https://mail.gna.org/listinfo/warmux-gna


[Wormux-gna] [bug #17738] Fails to build on s390

2011-02-25 Thread Felix Geyer

Follow-up Comment #3, bug #17738 (project warmux):

As I said the patch would only break if size_t is neither unsigned int nor
unsigned long.

On 64bit architectures it's usually unsigned long.

___

Reply to this item at:

  

___
  Nachricht geschickt von/durch Gna!
  http://gna.org/


___
Wormux-gna mailing list
wormux-...@gna.org
https://mail.gna.org/listinfo/warmux-gna


[Wormux-gna] [bug #17738] Fails to build on s390

2011-02-25 Thread Felix Geyer

Follow-up Comment #2, bug #17738 (project warmux):

It doesn't break 64bit systems.
The warmux package with that patch builds fine on all platforms Ubuntu
supports: amd64, armel, i386, powerpc.

___

Reply to this item at:

  

___
  Nachricht geschickt von/durch Gna!
  http://gna.org/


___
Wormux-gna mailing list
wormux-...@gna.org
https://mail.gna.org/listinfo/warmux-gna


[Wormux-gna] [bug #17738] Fails to build on s390

2011-02-25 Thread Kurosu

Follow-up Comment #1, bug #17738 (project warmux):

This will break 64bits system on the other hand. Could you check that using
"#if SIZE_MAX != UINT_MAX" works?

I'm unsure that SIZE_MAX is available everywhere (C99 afaik and C++ with
proper macro).

___

Reply to this item at:

  

___
  Message posté via/par Gna!
  http://gna.org/


___
Wormux-gna mailing list
wormux-...@gna.org
https://mail.gna.org/listinfo/warmux-gna


[Wormux-gna] [bug #17738] Fails to build on s390

2011-02-14 Thread Felix Geyer

URL:
  

 Summary: Fails to build on s390
 Project: Warmux
Submitted by: debfx
Submitted on: Di 15 Feb 2011 00:32:12 GMT
Category: Compilation
Severity: 3 - Normal
Priority: 5 - Normal
  Status: None
 Assigned to: None
Originator Email: 
 Open/Closed: Open
 Release: 11.01
 Discussion Lock: Any

___

Details:

Warmux 11.01 fails to build on the Debian architecture s390:

graphic/spritecache.cpp: In member function Surface
SpriteSubframeCache::GetSurfaceForAngle(Double):
graphic/spritecache.cpp:48: error: call of overloaded fixed_point(size_t) is
ambiguous
../lib/fixedpoint/fixed_class.h:79: note: candidates are:
fixedpoint::fixed_point::fixed_point(unsigned int) [with int p = 16]
../lib/fixedpoint/fixed_class.h:78: note:
fixedpoint::fixed_point::fixed_point(double) [with int p = 16]
../lib/fixedpoint/fixed_class.h:77: note:
fixedpoint::fixed_point::fixed_point(float) [with int p = 16]
../lib/fixedpoint/fixed_class.h:75: note:
fixedpoint::fixed_point::fixed_point(int32_t) [with int p = 16]
../lib/fixedpoint/fixed_class.h:73: note:
fixedpoint::fixed_point::fixed_point(fixedpoint::fixint_t) [with int p =
16]
../lib/fixedpoint/fixed_class.h:55: note:
fixedpoint::fixed_point<16>::fixed_point(const fixedpoint::fixed_point<16>&)

On s390 the following is true:
size_t = unsigned int
sizeof(int) = 4
sizeof(long) = 8

Attached is a patch that fixes this issue.
It assumes that size_t is either unsigned int or long.
I've tested the patch on s390, i386 and amd64.



___

File Attachments:


---
Date: Di 15 Feb 2011 00:32:12 GMT  Name: warmux_ftbfs_s390.patch  Size: 744B 
 By: debfx



___

Reply to this item at:

  

___
  Nachricht geschickt von/durch Gna!
  http://gna.org/


___
Wormux-gna mailing list
wormux-...@gna.org
https://mail.gna.org/listinfo/warmux-gna