[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: Bug#830180: PowerPC without altivec causes crash

On Sat, Nov 11, 2017 at 10:07:50AM +0100, Riccardo Mottola wrote:
> Hi,
> On 09/11/2017 17:41, John Paul Adrian Glaubitz wrote:
> > On 11/09/2017 05:30 PM, Ondřej Surý wrote:
> > > this has been already fixed by upstream:
> > > 
> > > https://github.com/libjpeg-turbo/libjpeg-turbo/commit/02fa8f244e549edd3f3acf174b97157590d1b71e
> > > 
> > 
> > Ah, nice. I didn't see that. Thanks for reporting the issue and
> > getting it fixed. Much appreciated!
> Do we have this biting us again?
> https://github.com/libjpeg-turbo/libjpeg-turbo/issues/88
> A year ago I had libjpeg-turbo issues in debian on G3, libjpeg used
> runtime-detection and failed.
> How is it now?
> Maybe your CPU has a different /proc/cpuinfo and runtime detection fails.

I hope that you don't parse /proc/cpuinfo to know the capabilities of
your CPU. There is a better way to do this, namely use getauxval(3).

Speaking of Altivec, there is another library that claims to be compiled
for Altivec but is actually compiled for VSX: libx265, which means that
I get crashes with "illegal instruction" on my G5.

I had a look at the source code, and it's really confusing, many
functions have Altivec in the name, but it's actually using vsx
instructions. A disassembly of the code reveals it: just grep 
for "vs[0-9]" patterns.


Reply to: