Re: [libav-devel] [PATCH] dca: change the core to work with integer coefficients.

2015-12-23 Thread Janne Grunau
On 2015-12-17 15:52:47 +0100, Alexandra Hájková wrote: > The DCA core decoder converts integer coefficients read from the > bitstream to floats just after reading them (along with dequantization). > All the other steps of the audio reconstruction are done with floats > which makes the output for

Re: [libav-devel] [libav-commits] checkasm: add fmtconvert tests

2015-12-23 Thread Henrik Gramner
On Tue, Dec 22, 2015 at 10:44 PM, Janne Grunau wrote: >> Intel's current documentation is very clear on cvtpi2ps: "This >> instruction causes a transition from x87 FPU to MMX technology >> operation". > > every tested silicon (nothing ancient or SSE only though) and the

Re: [libav-devel] [PATCH] dca: remove unused decode_hf function and quant_d tables

2015-12-23 Thread James Almer
On 12/18/2015 10:54 AM, Alexandra Hájková wrote: > diff --git a/libavcodec/dcadsp.h b/libavcodec/dcadsp.h > index eac9de9..b249b4a 100644 > --- a/libavcodec/dcadsp.h > +++ b/libavcodec/dcadsp.h > @@ -32,11 +32,6 @@ typedef struct DCADSPContext { > int

Re: [libav-devel] [PATCH] dca: remove unused decode_hf function and quant_d tables

2015-12-23 Thread Janne Grunau
On 2015-12-18 14:54:02 +0100, Alexandra Hájková wrote: > They were superseded with their integer equivalents. > --- > The patch with amended commit message. > > libavcodec/aarch64/dcadsp_init.c | 7 > libavcodec/aarch64/dcadsp_neon.S | 60 >