On 7/30/12 6:25 PM, Ronald S. Bultje wrote:
Hi,
On Sun, Jul 29, 2012 at 3:00 PM, Ronald S. Bultje wrote:
On Sat, Jul 28, 2012 at 10:11 AM, Ronald S. Bultje wrote:
From: "Ronald S. Bultje"
This completes the conversion of h264dsp to yasm; note that h264 also
uses some dsputil functions, mos
Hi,
On Sun, Jul 29, 2012 at 3:00 PM, Ronald S. Bultje wrote:
> On Sat, Jul 28, 2012 at 10:11 AM, Ronald S. Bultje wrote:
>> From: "Ronald S. Bultje"
>>
>> This completes the conversion of h264dsp to yasm; note that h264 also
>> uses some dsputil functions, most notably qpel. Performance-wise, t
Hi,
On Sat, Jul 28, 2012 at 10:11 AM, Ronald S. Bultje wrote:
> From: "Ronald S. Bultje"
>
> This completes the conversion of h264dsp to yasm; note that h264 also
> uses some dsputil functions, most notably qpel. Performance-wise, the
> yasm-version is ~10 cycles faster (182->172) on x86-64, and
From: "Ronald S. Bultje"
This completes the conversion of h264dsp to yasm; note that h264 also
uses some dsputil functions, most notably qpel. Performance-wise, the
yasm-version is ~10 cycles faster (182->172) on x86-64, and ~8 cycles
faster (201->193) on x86-32.
---
libavcodec/x86/h264_deblock.
On Sat, 28 Jul 2012, Ronald S. Bultje wrote:
> On Fri, Jul 27, 2012 at 1:01 PM, Loren Merritt
> wrote:
>
>> %%.b_idx_loop:
>> Automatically generates a different label for each instantiation of the
>> macro.
>
> My disassembly now looks like this:
>
> 0x0001004c43b4 :jne
>0x1004c44d1
Hi,
On Fri, Jul 27, 2012 at 1:01 PM, Loren Merritt wrote:
> On Fri, 27 Jul 2012, Ronald S. Bultje wrote:
>
>> From: "Ronald S. Bultje"
>>
>> This completes the conversion of h264dsp to yasm; note that h264 also
>> uses some dsputil functions, most notably qpel. Performance-wise, the
>> yasm-vers
On Fri, 27 Jul 2012, Ronald S. Bultje wrote:
> From: "Ronald S. Bultje"
>
> This completes the conversion of h264dsp to yasm; note that h264 also
> uses some dsputil functions, most notably qpel. Performance-wise, the
> yasm-version is ~10 cycles faster (182->172) on x86-64, and ~8 cycles
> faste
From: "Ronald S. Bultje"
This completes the conversion of h264dsp to yasm; note that h264 also
uses some dsputil functions, most notably qpel. Performance-wise, the
yasm-version is ~10 cycles faster (182->172) on x86-64, and ~8 cycles
faster (201->193) on x86-32.
---
libavcodec/x86/h264_deblock.
On Sun, Jul 22, 2012 at 1:16 PM, Ronald S. Bultje wrote:
> From: "Ronald S. Bultje"
>
> This completes the conversion of h264dsp to yasm; note that h264 also
> uses some dsputil functions, most notably qpel. Performance-wise, the
> yasm-version is ~10 cycles faster (182->172) on x86-64, and ~8 cy
On 23/07/2012 10:12 AM, Ronald S. Bultje wrote:
> Ping.
>From what I can tell, it looks OK, so long as you've
tested each possibility + FATE.
Someone better acquainted with (Y)ASM should look, though.
- Derek
___
libav-devel mailing list
libav-devel@l
Hi,
On Sun, Jul 22, 2012 at 1:16 PM, Ronald S. Bultje wrote:
> From: "Ronald S. Bultje"
>
> This completes the conversion of h264dsp to yasm; note that h264 also
> uses some dsputil functions, most notably qpel. Performance-wise, the
> yasm-version is ~10 cycles faster (182->172) on x86-64, and
From: "Ronald S. Bultje"
This completes the conversion of h264dsp to yasm; note that h264 also
uses some dsputil functions, most notably qpel. Performance-wise, the
yasm-version is ~10 cycles faster (182->172) on x86-64, and ~8 cycles
faster (201->193) on x86-32.
---
libavcodec/x86/h264_deblock.
From: "Ronald S. Bultje"
This completes the conversion of h264dsp to yasm; note that h264 also
uses some dsputil functions, most notably qpel. Performance-wise, the
yasm-version is ~10 cycles faster (182->172) on x86-64, and ~8 cycles
faster (201->193) on x86-32.
---
libavcodec/x86/h264_deblock.
13 matches
Mail list logo