On Sep 23, 2014, at 3:24 PM, Søren Sandmann wrote:
> > IIRC, we have already discussed it before. Maybe we should just disable
> > MMX support for x86 and use it only for MIPS Loongson and ARM IWMMXT?
>
> I'd be in favor of that. For a long time the only real use case for MMX/x86
> has been the
On 2013-01-30, at 9:14 PM, Søren Sandmann wrote:
>
>> Obviously this means that in this case, we're not getting the benefits of
>> any platform-specific fast paths. Perhaps what we need is a "pad"
>> equivalent of fast_composite_tiled_repeat()?
>
> This might be a good idea, though it's still w
On 2013-01-28, at 4:28 PM, Siarhei Siamashka wrote:
> On Thu, 24 Jan 2013 15:35:24 -0500
> Jeff Muizelaar wrote:
>
>> Having 4 or fewer bits means we can do two components at
>> a time in a single 32 bit register.
>>
>> Here are the results for firefox-fisht
On 2013-01-27, at 2:43 PM, Siarhei Siamashka wrote:
> I was a bit worried about the projective transforms. The accuracy is
> improved, but the new code uses a rather naive and slow implementation
> for long division. So the performance is going to be worse. But as
> almost nobody seems to be usin
Having 4 or fewer bits means we can do two components at
a time in a single 32 bit register.
Here are the results for firefox-fishtank on a Pandaboard with
4.6.3 and PIXMAN_DISABLE="arm-neon"
Before:
[ # ] backend test min(s) median(s) stddev. count
[ 0]image
And here's the patch.
patch
Description: Binary data
___
Pixman mailing list
Pixman@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/pixman
This patch adds fast paths for bilinear scaling of (SRC, r5g6b5, r5g6b5),
(OVER, a8r8g8b8, r5g6b5), and (OVER, a8r8g8b8, a8r8g8b8). These make a
noticeable
improvement in the performance of Firefox on Android.
-Jeff
___
Pixman mailing list
Pixman@lists
On 2012-06-25, at 7:44 PM, Siarhei Siamashka wrote:
> These are the test patches for switching to 7-bit bilinear
> interpolation precisions. The first patch makes bilinear precision
> configurable. The second patch tweaks SSE2 bilinear scaler for better
> performance using PMADDWD instruction. Bo
Siarhei wrote this patch and we've been using it in the Mozilla tree since May.
Before this patch it was often faster to scale and repeat in two passes because
each pass used a fast path vs. the slow path that the single pass approach
takes. This makes it so that the single pass approach has com
On 2012-05-09, at 12:57 PM, Søren Sandmann wrote:
> Matt Turner writes:
>
>> I started porting my src__0565 MMX function to SSE2, and in the
>> process started thinking about using SSE3+. The useful instructions
>> added post SSE2 that I see are
>> SSE3: lddqu - for unaligned loads a
On 19/04/10 6:34 AM, Soeren Sandmann wrote:
It's probably reasonable to do this given that cairo has the
corresponding feature, but I'd prefer to have to #error in
pixman-compiler.h extended to say something similar to the one in
cairo: That you can defined PIXMAN_NO_TLS, but pixman won't work wi
On 24-Apr-10, at 4:48 PM, Søren Sandmann wrote:
diff --git a/pixman/pixman-private.h b/pixman/pixman-private.h
index d5767af..c4f5064 100644
--- a/pixman/pixman-private.h
+++ b/pixman/pixman-private.h
@@ -686,6 +686,24 @@ pixman_region16_copy_from_region32
(pixman_region16_t *dst,
+/* Byte s
On 22-Mar-10, at 4:27 PM, Siarhei Siamashka wrote:
Comments and suggestions are welcome. Patches review would be
appreciated.
I had a quick look and it seems decent to me.
-Jeff
___
Pixman mailing list
Pixman@lists.freedesktop.org
http://lists.fre
On 22-Feb-10, at 10:35 AM, Rodrigo Kumpera wrote:
On Mon, Feb 22, 2010 at 1:39 PM, Siarhei Siamashka > wrote:
On Friday 19 February 2010, Luca Barbato wrote:
> On 02/19/2010 12:57 PM, Siarhei Siamashka wrote:
> > Adding small increments to the values at the end of loop
iteration could
> >
14 matches
Mail list logo