Re: [FFmpeg-devel] [PATCH 3/4] x86util: import MOVHL macro

2017-02-16 Thread Paul B Mahol
On 2/16/17, James Darnley wrote: > Originally committed to x264 in 1637239a by Henrik Gramner who has > agreed to re-license it as LGPL. Original commit message follows. > > x86: Avoid some bypass delays and false dependencies > > A bypass delay of 1-3 clock cycles may occur on some CPUs

[FFmpeg-devel] [PATCH 3/4] x86util: import MOVHL macro

2017-02-16 Thread James Darnley
Originally committed to x264 in 1637239a by Henrik Gramner who has agreed to re-license it as LGPL. Original commit message follows. x86: Avoid some bypass delays and false dependencies A bypass delay of 1-3 clock cycles may occur on some CPUs when transitioning between int and float

Re: [FFmpeg-devel] [PATCH 3/4] x86util: import MOVHL macro

2017-02-15 Thread James Darnley
On 2017-02-14 17:21, Henrik Gramner wrote: > On Mon, Feb 13, 2017 at 1:44 PM, James Darnley wrote: >> Originally committed to x264 in 1637239a by Henrik Gramner who has >> agreed to re-license it as LGPL. Original commit message follows. >> >> x86: Avoid some bypass delays and false dependenc

Re: [FFmpeg-devel] [PATCH 3/4] x86util: import MOVHL macro

2017-02-14 Thread Henrik Gramner
On Mon, Feb 13, 2017 at 1:44 PM, James Darnley wrote: > Originally committed to x264 in 1637239a by Henrik Gramner who has > agreed to re-license it as LGPL. Original commit message follows. > > x86: Avoid some bypass delays and false dependencies > > A bypass delay of 1-3 clock cycles ma

[FFmpeg-devel] [PATCH 3/4] x86util: import MOVHL macro

2017-02-13 Thread James Darnley
Originally committed to x264 in 1637239a by Henrik Gramner who has agreed to re-license it as LGPL. Original commit message follows. x86: Avoid some bypass delays and false dependencies A bypass delay of 1-3 clock cycles may occur on some CPUs when transitioning between int and float