Nam Se Hyun authored 7 months ago and Tristan Matthews committed 7 months ago

Signed-off-by: Tristan Matthews <tmatth@videolan.org>

It's just an internal gcc/clang type

MSVC doesn't have a real __m128i_u, so it would generate an aligned
store, resulting in a segfault. Adding explicit loadu/stureu
intrinsics to make sure the compiler generates unaligned load/store

Matches the C code and avoids undefined behaviour

Timothy B. Terriberry authored 1 year ago and Jean-Marc Valin committed 1 year ago

The existing code in vec_avx.h produced
  warning: dereferencing type-punned pointer will break
   strict-aliasing rules
 with gcc 6.4.0.
We already had a macro to work around this within the rules of the
 C standard, but trying to use that here does not get optimized
 into a single MOVD like we were hoping.
Replacing it with memcpy() instead does get optimized correctly,
 but requires switching from a macro to an inline function in order
 to be able to declare a local variable and return a value.
We already have such an inline function in NSQ_del_dec_avx2.c, so
 hoist that out and use it everywhere, and then convert vec_avx.h
 to use it also.

Fixes regression in 83368e6.
vcgez_s16() is A64-only, but vcge_s16(..., vdup_n_s16(0)) works
everywhere.

Timothy B. Terriberry authored 1 year ago and Jean-Marc Valin committed 1 year ago

Since any value of dQ > 0 will cause the initial quantizer to
 degrade to the format-implied maximum (15) with a sufficient
 number of DRED frames, allow signaling a maximum smaller than 15.
This allows encoders to improve the minimum quality of long DRED
 sequences (at the expense of bitrate) without requiring a constant
 quantizer for all frames (dQ == 0).

Silences NONTHREADSAFE_PSEUDOSTACK warnings

We don't need redundancy for the first active frame
since we already have the main Opus payload.

Allows us to exclude the most recent silence from DRED

Use the neon version of silk_noise_shape_quantizer_short_prediction()

Matches the C version (see 4a7027b2)

Adjust q0, qD and duration based on bitrate and loss.

Still don't quite know what I'm doing

1) Enable asm/intrinsics even for floating-point
2) Make sure ARMv8 asimd enables EDSP/MEDIA/Neon
3) Add dotp architecture to rtcd table since AArch *can* have dotp

Not yet with rtcd

Fixes warnings, undefined behaviour, and check-asm failure

Victor Ding authored 1 year ago and Jean-Marc Valin committed 1 year ago

The optimization is bit-exact with C function.

This optimization speeds up SILK encoder (floating point) as following:

AMD Zen:
Complexity 0-5 :      0%
Complexity 6-7 : 3 -  7%
Complexity 8-10: 8 - 15%

Intel Skylake:
Complexity 0-5 :       0%
Complexity 6-7 : 14 - 18%
Complexity 8-10: 17 - 22%

Adapted by Jean-Marc Valin

Not hooked up

Reducing the dependency chain between tmp1 and tmp2 at the
cost of an extra multiply.

Starting with compute_linear()