ImageMagick bit depth and compiling

Problems with packages? Post here, using [tags] of the package name.

ImageMagick bit depth and compiling

Postby qashar » Mon Aug 13, 2012 7:40 pm

Hello there,

I was just wondering why ImageMagick for ALARM is compiled with 32 bit depth, while it looks to me that the default is 16. 32bit version requires twice the memory to uncompress an image. This creates some problems given the small RAM for some devices like the Dockstar. Any particular reason for that?

Anyway, I got the PKGBUILD from the repository and try to compile ImageMagick with 16bit. The compilation failed (both makepkg and manual), but mostly failed also for 8bit and 32bit versions with the same exact error (some weird symbols due to "screen" log):
$this->bbcode_second_pass_code('', '
CC wand/wand_libMagickWand_la-convert.lo
gcc: internal compiler error: Killed (program cc1)
Please submit a full bug report,
with preprocessed source if appropriate.
See <https://bugs.archlinux.org/> for instructions.
make[1]: *** [wand/wand_libMagickWand_la-convert.lo] Error 1
make[1]: Leaving directory `/home/cenci/packages/imagemagick-q16_aur/src/ImageMagick-6.7.8-6'
make: *** [all] Error 2
==> ERROR: A failure occurred in build().
 Aborting...
')

I attached below the full output.

Do we have any log for the compiling of the version that is available through pacman?

Thank you very much for any help or advices...
qashar
 
Posts: 11
Joined: Thu Apr 21, 2011 3:39 pm

Re: ImageMagick bit depth and compiling

Postby WarheadsSE » Mon Aug 13, 2012 7:42 pm

My guess: OOM chewed it, and spit it out.
Core Developer
Remember: Arch Linux ARM is entirely community donation supported!
WarheadsSE
Developer
 
Posts: 6807
Joined: Mon Oct 18, 2010 2:12 pm

Re: ImageMagick bit depth and compiling

Postby qashar » Mon Aug 13, 2012 8:45 pm

Thanks for reading. I was compiling on a PogoE02, so maybe 256mb are not enough.
Which machine is used to compile the official packages? does it run behind an arm emulator?

Any chance that the default package can be changed from 32bit to 16 bit?
Even the default ArchLinux i686 is 16bit...
http://www.archlinux.org/packages/extra/i686/imagemagick/files/
qashar
 
Posts: 11
Joined: Thu Apr 21, 2011 3:39 pm

Re: ImageMagick bit depth and compiling

Postby pepedog » Mon Aug 13, 2012 8:55 pm

I thought this error was just on v7 cubox compiling kernel (700 mb ram + 1 gb swap), then I get on trimslice compiling gcc, now v5 see it
pepedog
Developer
 
Posts: 2431
Joined: Mon Jun 07, 2010 3:30 pm
Location: London UK

Re: ImageMagick bit depth and compiling

Postby WarheadsSE » Mon Aug 13, 2012 9:42 pm

If someone can get me a *real error, I can do something with it. There are a few packages that are huge enough to require swap. This may/may not be one of them.
Core Developer
Remember: Arch Linux ARM is entirely community donation supported!
WarheadsSE
Developer
 
Posts: 6807
Joined: Mon Oct 18, 2010 2:12 pm

Re: ImageMagick bit depth and compiling

Postby qashar » Fri Aug 17, 2012 4:05 pm

Thanks again for replying. I'll try to produce a real error later, because I found a temporary fix and I need a small application running in the next few weeks (an automatic timelapse).

I also spent few days in the corner reading everything about the cross compiling (I was completely unaware of that, my bad).
I'll try to setup a linux box for that later and compile a 16bit version of ImageMagick.
qashar
 
Posts: 11
Joined: Thu Apr 21, 2011 3:39 pm

Re: ImageMagick bit depth and compiling

Postby WarheadsSE » Fri Aug 17, 2012 4:09 pm

Psst, VM's work fine.
Core Developer
Remember: Arch Linux ARM is entirely community donation supported!
WarheadsSE
Developer
 
Posts: 6807
Joined: Mon Oct 18, 2010 2:12 pm


Return to Packages

Who is online

Users browsing this forum: No registered users and 10 guests