Re: [PATCH v3] arm64: v8.4: Support for new floating point multiplication instructions

2018-01-05 Thread gengdongjiu
> > > New features should not be going into 4.15-rc, that should be a > > > 4.16-rc1 thing, right? > > > > It is also great if it can be applied to 4.16-rc1. Thanks a lot! > > I will queue it for 4.16-rc1. Thanks very much to Catalin. > > -- > Catalin

Re: [PATCH v3] arm64: v8.4: Support for new floating point multiplication instructions

2018-01-05 Thread Catalin Marinas
On Fri, Jan 05, 2018 at 04:22:24PM +0800, gengdongjiu wrote: > On 2018/1/5 15:57, Greg KH wrote: > > On Fri, Jan 05, 2018 at 09:22:54AM +0800, gengdongjiu wrote: > >> Hi will/catalin > >> > >> On 2017/12/13 18:09, Suzuki K Poulose wrote: > >>> On 13/12/17 10:13, Dongjiu Geng wrote: > ARM v8.4

Re: [PATCH v3] arm64: v8.4: Support for new floating point multiplication instructions

2018-01-05 Thread gengdongjiu
On 2018/1/5 15:57, Greg KH wrote: > On Fri, Jan 05, 2018 at 09:22:54AM +0800, gengdongjiu wrote: >> Hi will/catalin >> >> On 2017/12/13 18:09, Suzuki K Poulose wrote: >>> On 13/12/17 10:13, Dongjiu Geng wrote: ARM v8.4 extensions add new neon instructions for performing a multiplication o

Re: [PATCH v3] arm64: v8.4: Support for new floating point multiplication instructions

2018-01-04 Thread Greg KH
On Fri, Jan 05, 2018 at 09:22:54AM +0800, gengdongjiu wrote: > Hi will/catalin > > On 2017/12/13 18:09, Suzuki K Poulose wrote: > > On 13/12/17 10:13, Dongjiu Geng wrote: > >> ARM v8.4 extensions add new neon instructions for performing a > >> multiplication of each FP16 element of one vector with

Re: [PATCH v3] arm64: v8.4: Support for new floating point multiplication instructions

2018-01-04 Thread gengdongjiu
Hi will/catalin On 2017/12/13 18:09, Suzuki K Poulose wrote: > On 13/12/17 10:13, Dongjiu Geng wrote: >> ARM v8.4 extensions add new neon instructions for performing a >> multiplication of each FP16 element of one vector with the corresponding >> FP16 element of a second vector, and to add or subt

Re: [PATCH v3] arm64: v8.4: Support for new floating point multiplication instructions

2017-12-13 Thread gengdongjiu
On 2017/12/13 18:09, Suzuki K Poulose wrote: >> Reviewed-by: Dave Martin > > Looks good to me. > > Reviewed-by: Suzuki K Poulose Thanks a lot to Suzuki's review.

Re: [PATCH v3] arm64: v8.4: Support for new floating point multiplication instructions

2017-12-13 Thread Suzuki K Poulose
On 13/12/17 10:13, Dongjiu Geng wrote: ARM v8.4 extensions add new neon instructions for performing a multiplication of each FP16 element of one vector with the corresponding FP16 element of a second vector, and to add or subtract this without an intermediate rounding to the corresponding FP32 el

[PATCH v3] arm64: v8.4: Support for new floating point multiplication instructions

2017-12-12 Thread Dongjiu Geng
ARM v8.4 extensions add new neon instructions for performing a multiplication of each FP16 element of one vector with the corresponding FP16 element of a second vector, and to add or subtract this without an intermediate rounding to the corresponding FP32 element in a third vector. This patch dete