Re: [Pixman] [PATCH] Do CPU features detection from 'constructor' function when compiled with gcc

2010-10-04 Thread Siarhei Siamashka
On Monday 04 October 2010 13:19:45 Andrea Canciani wrote: > On Sun, Oct 3, 2010 at 10:52 PM, Siarhei Siamashka > > wrote: > > From: Siarhei Siamashka > > > > There is attribute 'constructor' supported since gcc 2.7 which allows > > to have a constructor function for library initialization. This

Re: [Pixman] [PATCH] Do CPU features detection from 'constructor' function when compiled with gcc

2010-10-04 Thread Andrea Canciani
On Mon, Oct 4, 2010 at 12:58 PM, Luca Barbato wrote: > On 10/04/2010 12:19 PM, Andrea Canciani wrote: >> >> which adds atomic refcounting and (private, automatically called) > > What's missing in libatomic_ops? For atomics, nothing. In fact libatomic_ops is one of the possible ways for simpleops t

[Pixman] [PATCH] Use more unrolling for scaled src_0565_0565 with nearest filter

2010-10-04 Thread Siarhei Siamashka
From: Siarhei Siamashka Benchmark from Intel Core i7 860: == before == op=1, src_fmt=10020565, dst_fmt=10020565, speed=1335.29 MPix/s == after == op=1, src_fmt=10020565, dst_fmt=10020565, speed=1550.96 MPix/s == performance of nonscaled src_0565_0565 operation as a referenc

Re: [Pixman] [PATCH] Do CPU features detection from 'constructor' function when compiled with gcc

2010-10-04 Thread Luca Barbato
On 10/04/2010 12:19 PM, Andrea Canciani wrote: which adds atomic refcounting and (private, automatically called) What's missing in libatomic_ops? lu -- Luca Barbato Gentoo/linux http://dev.gentoo.org/~lu_zero ___ Pixman mailing list Pixman@lists.

Re: [Pixman] FPU-based implementation of the core pixel pipeline

2010-10-04 Thread Dmitri Vorobiev
On Sat, Oct 2, 2010 at 4:45 AM, Soeren Sandmann wrote: > > We need to figure out how a floating point implementation fits in > before it makes sense to review the code in detail, especially > considering that it clearly isn't yet close to being mergable (see > below). Soeren, thanks for your feed

Re: [Pixman] [PATCH] Do CPU features detection from 'constructor' function when compiled with gcc

2010-10-04 Thread Andrea Canciani
On Sun, Oct 3, 2010 at 10:52 PM, Siarhei Siamashka wrote: > From: Siarhei Siamashka > > There is attribute 'constructor' supported since gcc 2.7 which allows > to have a constructor function for library initialization. This eliminates > an extra branch for each composite operation and also helps

[Pixman] [PATCH] ARM: restore fallback to ARMv6 implementation from NEON in the delegate chain

2010-10-04 Thread Siarhei Siamashka
From: Siarhei Siamashka After fast path cache introduction, the overhead of having this fallback is insignificant. On the other hand, some of the ARM assembly optimizations (for example nearest neighbor scaling) do not need NEON. --- pixman/pixman-arm-neon.c |8 ++-- 1 files changed, 6 i

[Pixman] Gradients patches

2010-10-04 Thread Andrea Canciani
This morning I pushed http://cgit.freedesktop.org/~ranma42/pixman/log/?h=radial-for-master It should be ready for master since it is documented, tested (at least on my laptop) and not using any new features (so it should not be broken on other architectures/compilers/etc). I'm reviewing the fix fo