[FFmpeg-cvslog] r17543 - trunk/libavcodec/x86/cpuid.c

Michael Niedermayer michaelni
Tue Feb 24 17:17:36 CET 2009


On Tue, Feb 24, 2009 at 11:18:48AM +0800, Zuxy Meng wrote:
> Hi,
> 
> "ReimarD?ffinger" <Reimar.Doeffinger at stud.uni-karlsruhe.de> ???? 
> news:20090223165111.GA13666 at 1und1.de...
> > On Mon, Feb 23, 2009 at 04:53:40PM +0100, zuxy wrote:
> >> Author: zuxy
> >> Date: Mon Feb 23 16:53:39 2009
> >> New Revision: 17543
> >>
> >> Log:
> >> Remove CPUID availability check on AMD64 as it's architectural.
> >
> > Do you actually have a reference for that?
> > My architecture reference manuals state no such thing, which is why it
> > was done like this (not that I expect any CPU to not support it).
> > The statements on SSE etc. support are a bit ambiguous, too, but the
> > CPUID description is the one that sound like you absolutely have to
> > check for it always...
> 
> Long mode support itself is reported thru CPUID, and an AMD64 kernel won't 
> ever boot if it can't detect the existence of CPUID.
> 

> Besides, GCC thinks it's architectural.

GCC didnt design x86_64/AMD64 only intel and amd can decide what is
part of their architecture.

[...]

-- 
Michael     GnuPG fingerprint: 9FF2128B147EF6730BADF133611EC787040B0FAB

Those who are too smart to engage in politics are punished by being
governed by those who are dumber. -- Plato 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 189 bytes
Desc: Digital signature
URL: <http://lists.mplayerhq.hu/pipermail/ffmpeg-cvslog/attachments/20090224/6721d66f/attachment.pgp>



More information about the ffmpeg-cvslog mailing list