Re: [Pixman] [PATCH 1/4] vmx: optimize scaled_nearest_scanline_vmx_8888_8888_OVER

2015-09-17 Thread Siarhei Siamashka
On Sun, 6 Sep 2015 10:08:06 -0700 Matt Turner wrote: > On Sun, Sep 6, 2015 at 8:27 AM, Oded Gabbay wrote: > > reference memcpy speed = 24847.3MB/s (6211.8MP/s for 32bpp fills) > > > > Before After Change > >

Re: [Pixman] [PATCH 1/4] vmx: optimize scaled_nearest_scanline_vmx_8888_8888_OVER

2015-09-17 Thread Siarhei Siamashka
On Mon, 7 Sep 2015 14:07:19 +0300 Oded Gabbay wrote: > On Mon, Sep 7, 2015 at 2:03 PM, Pekka Paalanen wrote: > > On Sun, 6 Sep 2015 18:27:08 +0300 > > Oded Gabbay wrote: > > > >> This patch optimizes

Re: [Pixman] [PATCH 1/4] vmx: optimize scaled_nearest_scanline_vmx_8888_8888_OVER

2015-09-07 Thread Pekka Paalanen
On Sun, 6 Sep 2015 18:27:08 +0300 Oded Gabbay wrote: > This patch optimizes scaled_nearest_scanline_vmx___OVER and all > the functions it calls (combine1, combine4 and > core_combine_over_u_pixel_vmx). > > The optimization is done by removing use of

Re: [Pixman] [PATCH 1/4] vmx: optimize scaled_nearest_scanline_vmx_8888_8888_OVER

2015-09-07 Thread Oded Gabbay
On Mon, Sep 7, 2015 at 2:03 PM, Pekka Paalanen wrote: > On Sun, 6 Sep 2015 18:27:08 +0300 > Oded Gabbay wrote: > >> This patch optimizes scaled_nearest_scanline_vmx___OVER and all >> the functions it calls (combine1, combine4 and >>

[Pixman] [PATCH 1/4] vmx: optimize scaled_nearest_scanline_vmx_8888_8888_OVER

2015-09-06 Thread Oded Gabbay
This patch optimizes scaled_nearest_scanline_vmx___OVER and all the functions it calls (combine1, combine4 and core_combine_over_u_pixel_vmx). The optimization is done by removing use of expand_alpha_1x128 and expand_alpha_2x128 in favor of splat_alpha and MUL/ADD macros from