Re: fftw3: Patch for powerpc build failure

[Colin Watson <cjwatson@debian.org>]

On Thu, Jan 01, 2004 at 01:19:14PM +0000, Colin Watson wrote:
> On Mon, Dec 29, 2003 at 12:32:34PM -0500, Lukas Geyer wrote:
> > the problem with fftw3 on powerpc is that the current debian/rules
> > assumes that all powerpc CPUs know Altivec instructions. That is
> > plainly untrue (Altivec is only available in the most recent CPUs) and
> > leads to an illegal instruction error here. The buildd fails with sig
> > 11, but I suspect the reason is the same.
[...]
> Although you can't see it here, "still work" has an emphasis tag around
> it. It seems clear to me that FFTW is intended to work if you
> --enable-altivec but aren't using an AltiVec-supporting CPU, so I'm
> suspicious of fixing this by removing --enable-altivec.

Another thing to note is that voltaire, the buildd in question, claims
AltiVec support in /proc/cpuinfo:

  processor       : 0
  cpu             : 7400, altivec supported
  temperature     : 26-28 C (uncalibrated)
  clock           : 500MHz
  revision        : 2.9 (pvr 000c 0209)
  bogomips        : 992.87
  
  processor       : 1
  cpu             : 7400, altivec supported
  temperature     : 48-51 C (uncalibrated)
  clock           : 500MHz
  revision        : 2.9 (pvr 000c 0209)
  bogomips        : 992.87
  
  total bogomips  : 1985.74
  machine         : PowerMac3,3
  motherboard     : PowerMac3,3 MacRISC Power Macintosh
  detected as     : 65 (PowerMac G4 AGP Graphics)
  pmac flags      : 00000004
  L2 cache        : 1024K unified
  memory          : 320MB
  pmac-generation : NewWorld

Unfortunately the user unstable chroot on voltaire isn't working right
now (I've mailed -admin), and fftw3 builds without any problems on my
laptop, so I'm stalled for the moment. Once I can get to voltaire's
chroot I'll try digging into a build there.

I'm cc'ing -powerpc just in case anyone has any other ideas.
http://bugs.debian.org/223601 has the history. I suppose I *could* just
upload the successful build from my laptop but I'm not very happy about
doing that if it's just going to fail again on the buildd next time
around, or if it indicates some bug which will bite programs linked
against fftw3 ...

Cheers,

-- 
Colin Watson                                  [cjwatson@flatline.org.uk]