[Ffmpeg-devel] [PATCH] SSE counterpart of ff_imdct_calc_3dn2

Rich Felker dalias
Thu Aug 24 18:12:22 CEST 2006

On Thu, Aug 24, 2006 at 08:50:18AM -0400, Dave Dodge wrote:
> On Thu, Aug 24, 2006 at 09:59:37AM +0200, Guillaume POIRIER wrote:
> > Also, ICC is able to process these intrinsics, whereas it has a hard
> > time with inline asm.
> As an aside: unless they've changed it recently, on IA64 icc provides
> only intrinsics and doesn't allow inline assembly at all.  Granted
> very few people are going to _want_ to write IA64 assembly anyway, so
> it's not that surprising.  A downside is that since they don't provide
> intrinsics for every possible instruction and hint combination, there
> are some obscure but potentially useful instructions that it can't
> generate.

This is idiotic. Inline asm is not just for vector ops. Often you want
it for 'normal' opcodes too. Of course icc sucks so it's no big deal.


More information about the ffmpeg-devel mailing list