resample: vectorize the polyphase tap loop

The tap loop wrapped the ring with a test on every tap:

    if ((idx_out = idx_in + j) >= ringsize) idx_out -= ringsize;

which made the address non-affine and stopped the vectorizer. Split the
walk at the wrap point instead, so both halves are unit-stride, and split
the complex ring into separate I/Q arrays so the taps load contiguously
rather than through a de-interleaving ld2.

The dot product also could not be vectorized as written: reassociating an
fp reduction needs -ffast-math, which this library must not enable (it
relies on IEEE semantics for 0/0 = NaN and x/0 = Inf, see linux_port.h).
Carry four independent accumulator pairs instead, which both breaks the
FMA dependency chain and lets the vectorizer in on any compiler.

Measured on an Apple M1 Pro, 512-sample DSP buffers, best of 5:

  xresample, decimation to 48 kHz     before      after    speedup
    192k -> 48k  (561 taps)          342.0 us    84.2 us     4.06x
    384k -> 48k  (1121 taps)         708.9 us   175.1 us     4.05x
    576k -> 48k  (1681 taps)        1074.4 us   266.9 us     4.03x
    768k -> 48k  (2241 taps)        1438.9 us   360.0 us     4.00x

  full xrxa() chain, 576k input      1148.2 us   337.6 us     3.40x
                                     10.76%       3.16%   of one core

  xresampleF (float, host audio)                          2.6x - 3.4x

Summation order changes, so the double path is not bit-identical: over
400 buffers of the full RX chain the worst deviation is 1.1e-12, an SNR
of 251 dB. The float path is bit-identical, as the cast to float absorbs
the difference. With the resampler bypassed (48k in, 48k out) the chain
is unchanged bit-for-bit.

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>

This commit is contained in:

Uladzimir Karpenka

2026-07-09 22:44:08 +03:00

parent 89c8a0e2b5

commit 9960a1d0e3

3 changed files with 128 additions and 42 deletions

									
										comm.h
									
		+8
		
												View File
												
				@@ -145,6 +145,14 @@ warren@wpratt.com

				#define PI								3.1415926535897932

				#define TWOPI							6.2831853071795864

				// Non-aliasing qualifier for DSP buffers. Spelled __restrict rather than

				// restrict because the JNI translation unit is compiled as -std=gnu89.

				#if defined(__GNUC__) || defined(__clang__) || defined(_MSC_VER)

				#define WDSP_RESTRICT					__restrict

				#else

				#define WDSP_RESTRICT

				#endif

				// miscellaneous

				typedef double complex[2];

				#define PORT							__declspec( dllexport )