fix(windows): vendor audiopus_sys + patch libopus for clang-cl SIMD
cargo-xwin drives the Windows MSVC cross-compile via clang-cl, under which CMake sets MSVC=1 — causing libopus 1.3.1's `if(NOT MSVC)` guards to skip the per-file `-msse4.1` / `-mssse3` COMPILE_FLAGS that its x86 SIMD source files need. Clang-cl (unlike real cl.exe) still honors Clang's target-feature system, so those files then fail to compile with "always_inline function '_mm_cvtepi16_epi32' requires target feature 'sse4.1'" errors across silk/NSQ_sse4_1.c, NSQ_del_dec_sse4_1.c, and VQ_WMat_EC_sse4_1.c. Earlier attempts to fix this downstream (cargo-xwin toolchain file, override.cmake CMAKE_C_COMPILE_OBJECT <FLAGS> replace, CFLAGS env vars) all failed because cargo-xwin rewrites override.cmake from scratch on every `cargo xwin build` invocation and cmake-rs's -DCMAKE_C_FLAGS= assembly happens before toolchain FORCE sets propagate. Fixing it upstream at the source: vendor audiopus_sys 0.2.2 into vendor/audiopus_sys, patch its bundled opus/CMakeLists.txt to introduce an MSVC_CL var (true only when CMAKE_C_COMPILER_ID == "MSVC", i.e. real cl.exe), and flip the eight `if(NOT MSVC)` SIMD guards to `if(NOT MSVC_CL)`. Clang-cl then gets the GCC-style per-file flags and the SSE4.1 sources build cleanly. Also flip the `if(MSVC)` global /arch block at line 445 to `if(MSVC_CL)` so only cl.exe applies /arch:AVX and clang-cl relies purely on per-file flags (no global/per-file mixing). Wire via [patch.crates-io] in the workspace root Cargo.toml; the patch is resolved relative to the workspace root as `vendor/audiopus_sys`. Upstream context: xiph/opus#256, xiph/opus PR #257 (both stale). Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
This commit is contained in:
Siavash Sameni
188
vendor/audiopus_sys/opus/celt/fixed_generic.h
vendored
Normal file
188
vendor/audiopus_sys/opus/celt/fixed_generic.h
vendored
Normal file
@@ -0,0 +1,188 @@
|
||||
/* Copyright (C) 2007-2009 Xiph.Org Foundation
|
||||
Copyright (C) 2003-2008 Jean-Marc Valin
|
||||
Copyright (C) 2007-2008 CSIRO */
|
||||
/**
|
||||
@file fixed_generic.h
|
||||
@brief Generic fixed-point operations
|
||||
*/
|
||||
/*
|
||||
Redistribution and use in source and binary forms, with or without
|
||||
modification, are permitted provided that the following conditions
|
||||
are met:
|
||||
|
||||
- Redistributions of source code must retain the above copyright
|
||||
notice, this list of conditions and the following disclaimer.
|
||||
|
||||
- Redistributions in binary form must reproduce the above copyright
|
||||
notice, this list of conditions and the following disclaimer in the
|
||||
documentation and/or other materials provided with the distribution.
|
||||
|
||||
THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
|
||||
``AS IS'' AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
|
||||
LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
|
||||
A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER
|
||||
OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL,
|
||||
EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO,
|
||||
PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR
|
||||
PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF
|
||||
LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING
|
||||
NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS
|
||||
SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
|
||||
*/
|
||||
|
||||
#ifndef FIXED_GENERIC_H
|
||||
#define FIXED_GENERIC_H
|
||||
|
||||
/** Multiply a 16-bit signed value by a 16-bit unsigned value. The result is a 32-bit signed value */
|
||||
#define MULT16_16SU(a,b) ((opus_val32)(opus_val16)(a)*(opus_val32)(opus_uint16)(b))
|
||||
|
||||
/** 16x32 multiplication, followed by a 16-bit shift right. Results fits in 32 bits */
|
||||
#if OPUS_FAST_INT64
|
||||
#define MULT16_32_Q16(a,b) ((opus_val32)SHR((opus_int64)((opus_val16)(a))*(b),16))
|
||||
#else
|
||||
#define MULT16_32_Q16(a,b) ADD32(MULT16_16((a),SHR((b),16)), SHR(MULT16_16SU((a),((b)&0x0000ffff)),16))
|
||||
#endif
|
||||
|
||||
/** 16x32 multiplication, followed by a 16-bit shift right (round-to-nearest). Results fits in 32 bits */
|
||||
#if OPUS_FAST_INT64
|
||||
#define MULT16_32_P16(a,b) ((opus_val32)PSHR((opus_int64)((opus_val16)(a))*(b),16))
|
||||
#else
|
||||
#define MULT16_32_P16(a,b) ADD32(MULT16_16((a),SHR((b),16)), PSHR(MULT16_16SU((a),((b)&0x0000ffff)),16))
|
||||
#endif
|
||||
|
||||
/** 16x32 multiplication, followed by a 15-bit shift right. Results fits in 32 bits */
|
||||
#if OPUS_FAST_INT64
|
||||
#define MULT16_32_Q15(a,b) ((opus_val32)SHR((opus_int64)((opus_val16)(a))*(b),15))
|
||||
#else
|
||||
#define MULT16_32_Q15(a,b) ADD32(SHL(MULT16_16((a),SHR((b),16)),1), SHR(MULT16_16SU((a),((b)&0x0000ffff)),15))
|
||||
#endif
|
||||
|
||||
/** 32x32 multiplication, followed by a 16-bit shift right. Results fits in 32 bits */
|
||||
#if OPUS_FAST_INT64
|
||||
#define MULT32_32_Q16(a,b) ((opus_val32)SHR((opus_int64)(a)*(opus_int64)(b),16))
|
||||
#else
|
||||
#define MULT32_32_Q16(a,b) (ADD32(ADD32(ADD32((opus_val32)(SHR32(((opus_uint32)((a)&0x0000ffff)*(opus_uint32)((b)&0x0000ffff)),16)), MULT16_16SU(SHR32(a,16),((b)&0x0000ffff))), MULT16_16SU(SHR32(b,16),((a)&0x0000ffff))), SHL32(MULT16_16(SHR32(a,16),SHR32(b,16)),16)))
|
||||
#endif
|
||||
|
||||
/** 32x32 multiplication, followed by a 31-bit shift right. Results fits in 32 bits */
|
||||
#if OPUS_FAST_INT64
|
||||
#define MULT32_32_Q31(a,b) ((opus_val32)SHR((opus_int64)(a)*(opus_int64)(b),31))
|
||||
#else
|
||||
#define MULT32_32_Q31(a,b) ADD32(ADD32(SHL(MULT16_16(SHR((a),16),SHR((b),16)),1), SHR(MULT16_16SU(SHR((a),16),((b)&0x0000ffff)),15)), SHR(MULT16_16SU(SHR((b),16),((a)&0x0000ffff)),15))
|
||||
#endif
|
||||
|
||||
/** Compile-time conversion of float constant to 16-bit value */
|
||||
#define QCONST16(x,bits) ((opus_val16)(.5+(x)*(((opus_val32)1)<<(bits))))
|
||||
|
||||
/** Compile-time conversion of float constant to 32-bit value */
|
||||
#define QCONST32(x,bits) ((opus_val32)(.5+(x)*(((opus_val32)1)<<(bits))))
|
||||
|
||||
/** Negate a 16-bit value */
|
||||
#define NEG16(x) (-(x))
|
||||
/** Negate a 32-bit value */
|
||||
#define NEG32(x) (-(x))
|
||||
|
||||
/** Change a 32-bit value into a 16-bit value. The value is assumed to fit in 16-bit, otherwise the result is undefined */
|
||||
#define EXTRACT16(x) ((opus_val16)(x))
|
||||
/** Change a 16-bit value into a 32-bit value */
|
||||
#define EXTEND32(x) ((opus_val32)(x))
|
||||
|
||||
/** Arithmetic shift-right of a 16-bit value */
|
||||
#define SHR16(a,shift) ((a) >> (shift))
|
||||
/** Arithmetic shift-left of a 16-bit value */
|
||||
#define SHL16(a,shift) ((opus_int16)((opus_uint16)(a)<<(shift)))
|
||||
/** Arithmetic shift-right of a 32-bit value */
|
||||
#define SHR32(a,shift) ((a) >> (shift))
|
||||
/** Arithmetic shift-left of a 32-bit value */
|
||||
#define SHL32(a,shift) ((opus_int32)((opus_uint32)(a)<<(shift)))
|
||||
|
||||
/** 32-bit arithmetic shift right with rounding-to-nearest instead of rounding down */
|
||||
#define PSHR32(a,shift) (SHR32((a)+((EXTEND32(1)<<((shift))>>1)),shift))
|
||||
/** 32-bit arithmetic shift right where the argument can be negative */
|
||||
#define VSHR32(a, shift) (((shift)>0) ? SHR32(a, shift) : SHL32(a, -(shift)))
|
||||
|
||||
/** "RAW" macros, should not be used outside of this header file */
|
||||
#define SHR(a,shift) ((a) >> (shift))
|
||||
#define SHL(a,shift) SHL32(a,shift)
|
||||
#define PSHR(a,shift) (SHR((a)+((EXTEND32(1)<<((shift))>>1)),shift))
|
||||
#define SATURATE(x,a) (((x)>(a) ? (a) : (x)<-(a) ? -(a) : (x)))
|
||||
|
||||
#define SATURATE16(x) (EXTRACT16((x)>32767 ? 32767 : (x)<-32768 ? -32768 : (x)))
|
||||
|
||||
/** Shift by a and round-to-nearest 32-bit value. Result is a 16-bit value */
|
||||
#define ROUND16(x,a) (EXTRACT16(PSHR32((x),(a))))
|
||||
/** Shift by a and round-to-nearest 32-bit value. Result is a saturated 16-bit value */
|
||||
#define SROUND16(x,a) EXTRACT16(SATURATE(PSHR32(x,a), 32767));
|
||||
|
||||
/** Divide by two */
|
||||
#define HALF16(x) (SHR16(x,1))
|
||||
#define HALF32(x) (SHR32(x,1))
|
||||
|
||||
/** Add two 16-bit values */
|
||||
#define ADD16(a,b) ((opus_val16)((opus_val16)(a)+(opus_val16)(b)))
|
||||
/** Subtract two 16-bit values */
|
||||
#define SUB16(a,b) ((opus_val16)(a)-(opus_val16)(b))
|
||||
/** Add two 32-bit values */
|
||||
#define ADD32(a,b) ((opus_val32)(a)+(opus_val32)(b))
|
||||
/** Subtract two 32-bit values */
|
||||
#define SUB32(a,b) ((opus_val32)(a)-(opus_val32)(b))
|
||||
|
||||
/** Add two 32-bit values, ignore any overflows */
|
||||
#define ADD32_ovflw(a,b) ((opus_val32)((opus_uint32)(a)+(opus_uint32)(b)))
|
||||
/** Subtract two 32-bit values, ignore any overflows */
|
||||
#define SUB32_ovflw(a,b) ((opus_val32)((opus_uint32)(a)-(opus_uint32)(b)))
|
||||
/* Avoid MSVC warning C4146: unary minus operator applied to unsigned type */
|
||||
/** Negate 32-bit value, ignore any overflows */
|
||||
#define NEG32_ovflw(a) ((opus_val32)(0-(opus_uint32)(a)))
|
||||
|
||||
/** 16x16 multiplication where the result fits in 16 bits */
|
||||
#define MULT16_16_16(a,b) ((((opus_val16)(a))*((opus_val16)(b))))
|
||||
|
||||
/** 32x32 multiplication where the result fits in 32 bits */
|
||||
#define MULT32_32_32(a,b) ((((opus_val32)(a))*((opus_val32)(b))))
|
||||
|
||||
/* (opus_val32)(opus_val16) gives TI compiler a hint that it's 16x16->32 multiply */
|
||||
/** 16x16 multiplication where the result fits in 32 bits */
|
||||
#define MULT16_16(a,b) (((opus_val32)(opus_val16)(a))*((opus_val32)(opus_val16)(b)))
|
||||
|
||||
/** 16x16 multiply-add where the result fits in 32 bits */
|
||||
#define MAC16_16(c,a,b) (ADD32((c),MULT16_16((a),(b))))
|
||||
/** 16x32 multiply, followed by a 15-bit shift right and 32-bit add.
|
||||
b must fit in 31 bits.
|
||||
Result fits in 32 bits. */
|
||||
#define MAC16_32_Q15(c,a,b) ADD32((c),ADD32(MULT16_16((a),SHR((b),15)), SHR(MULT16_16((a),((b)&0x00007fff)),15)))
|
||||
|
||||
/** 16x32 multiplication, followed by a 16-bit shift right and 32-bit add.
|
||||
Results fits in 32 bits */
|
||||
#define MAC16_32_Q16(c,a,b) ADD32((c),ADD32(MULT16_16((a),SHR((b),16)), SHR(MULT16_16SU((a),((b)&0x0000ffff)),16)))
|
||||
|
||||
#define MULT16_16_Q11_32(a,b) (SHR(MULT16_16((a),(b)),11))
|
||||
#define MULT16_16_Q11(a,b) (SHR(MULT16_16((a),(b)),11))
|
||||
#define MULT16_16_Q13(a,b) (SHR(MULT16_16((a),(b)),13))
|
||||
#define MULT16_16_Q14(a,b) (SHR(MULT16_16((a),(b)),14))
|
||||
#define MULT16_16_Q15(a,b) (SHR(MULT16_16((a),(b)),15))
|
||||
|
||||
#define MULT16_16_P13(a,b) (SHR(ADD32(4096,MULT16_16((a),(b))),13))
|
||||
#define MULT16_16_P14(a,b) (SHR(ADD32(8192,MULT16_16((a),(b))),14))
|
||||
#define MULT16_16_P15(a,b) (SHR(ADD32(16384,MULT16_16((a),(b))),15))
|
||||
|
||||
/** Divide a 32-bit value by a 16-bit value. Result fits in 16 bits */
|
||||
#define DIV32_16(a,b) ((opus_val16)(((opus_val32)(a))/((opus_val16)(b))))
|
||||
|
||||
/** Divide a 32-bit value by a 32-bit value. Result fits in 32 bits */
|
||||
#define DIV32(a,b) (((opus_val32)(a))/((opus_val32)(b)))
|
||||
|
||||
#if defined(MIPSr1_ASM)
|
||||
#include "mips/fixed_generic_mipsr1.h"
|
||||
#endif
|
||||
|
||||
static OPUS_INLINE opus_val16 SIG2WORD16_generic(celt_sig x)
|
||||
{
|
||||
x = PSHR32(x, SIG_SHIFT);
|
||||
x = MAX32(x, -32768);
|
||||
x = MIN32(x, 32767);
|
||||
return EXTRACT16(x);
|
||||
}
|
||||
#define SIG2WORD16(x) (SIG2WORD16_generic(x))
|
||||
|
||||
#endif
|
||||
Reference in New Issue
Block a user