Emulation/ppsspp
1
0
Fork 0
mirror of https://github.com/hrydgard/ppsspp.git synced 2025-04-02 11:01:50 -04:00
Code Issues Projects Releases Packages Wiki Activity

Finish up this cleanup (move _M_SSE compat define out of Common.h)

Browse source
This commit is contained in:
Henrik Rydgård 2024-12-19 16:23:13 +01:00
parent d2a96dc7b5
commit d5bfc1b5eb
16 changed files with 43 additions and 41 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

14
Common/Common.h
View file

@ -87,17 +87,3 @@
#define __forceinline inline __attribute__((always_inline))
#endif
#if defined __SSE4_2__
# define _M_SSE 0x402
#elif defined __SSE4_1__
# define _M_SSE 0x401
#elif defined __SSSE3__
# define _M_SSE 0x301
#elif defined __SSE3__
# define _M_SSE 0x300
#elif defined __SSE2__
# define _M_SSE 0x200
#elif !defined(__GNUC__) && (defined(_M_X64) || defined(_M_IX86))
# define _M_SSE 0x402
#endif

14
Common/Math/SIMDHeaders.h
View file

@ -67,6 +67,20 @@ static inline uint32x4_t vcgezq_f32(float32x4_t v) {
#if PPSSPP_ARCH(SSE2)
#if defined __SSE4_2__
# define _M_SSE 0x402
#elif defined __SSE4_1__
# define _M_SSE 0x401
#elif defined __SSSE3__
# define _M_SSE 0x301
#elif defined __SSE3__
# define _M_SSE 0x300
#elif defined __SSE2__
# define _M_SSE 0x200
#elif !defined(__GNUC__) && (defined(_M_X64) || defined(_M_IX86))
# define _M_SSE 0x402
#endif
// These are SSE2 versions of SSE4.1 instructions, for compatibility and ease of
// writing code.
// May later figure out how to use the appropriate ones depending on compile flags.

10
Common/Math/fast/fast_matrix.c
View file

@ -4,9 +4,7 @@
#include "fast_matrix.h"
#if PPSSPP_ARCH(X86) || PPSSPP_ARCH(AMD64)
#include <emmintrin.h>
#if PPSSPP_ARCH(SSE2)
#include "fast_matrix.h"
@ -28,12 +26,6 @@ void fast_matrix_mul_4x4_sse(float *dest, const float *a, const float *b) {
#elif PPSSPP_ARCH(ARM_NEON)
#if defined(_MSC_VER) && PPSSPP_ARCH(ARM64)
#include <arm64_neon.h>
#else
#include <arm_neon.h>
#endif
#if PPSSPP_ARCH(ARM)
static inline float32x4_t vfmaq_laneq_f32(float32x4_t _s, float32x4_t _a, float32x4_t _b, int lane) {
if (lane == 0) return vmlaq_lane_f32(_s, _a, vget_low_f32(_b), 0);

3
Core/MIPS/x86/CompVFPU.cpp
View file

@ -23,11 +23,10 @@
#if PPSSPP_ARCH(X86) || PPSSPP_ARCH(AMD64)
#include <cmath>
#include <limits>
#include <emmintrin.h>
#include "Common/Math/math_util.h"
#include "Common/CPUDetect.h"
#include "Common/Math/SIMDHeaders.h"
#include "Common/Log.h"
#include "Core/Compatibility.h"
#include "Core/Config.h"

3
Core/MIPS/x86/RegCacheFPU.cpp
View file

@ -19,8 +19,7 @@
#if PPSSPP_ARCH(X86) || PPSSPP_ARCH(AMD64)
#include <cstring>
#include <emmintrin.h>
#include "Common/Math/SIMDHeaders.h"
#include "Common/Log.h"
#include "Common/x64Emitter.h"
#include "Core/MIPS/MIPSAnalyst.h"

1
GPU/Common/IndexGenerator.cpp
View file

@ -19,7 +19,6 @@
#include "ppsspp_config.h"
#include "Common/Common.h"
#include "Common/Math/SIMDHeaders.h"
#include "GPU/Common/IndexGenerator.h"

4
GPU/Common/TextureScalerCommon.cpp
View file

@ -25,12 +25,10 @@
#include "Core/Config.h"
#include "Common/Common.h"
#include "Common/Log.h"
#include "Common/Math/SIMDHeaders.h"
#include "Common/Thread/ParallelLoop.h"
#include "ext/xbrz/xbrz.h"
#include "Common/Math/SIMDHeaders.h"
// Report the time and throughput for each larger scaling operation in the log
//#define SCALING_MEASURE_TIME
#include "Common/TimeUtil.h"

4
GPU/Common/VertexDecoderX86.cpp
View file

@ -16,12 +16,12 @@
// https://github.com/hrydgard/ppsspp and http://www.ppsspp.org/.
#include "ppsspp_config.h"
#if PPSSPP_ARCH(X86) || PPSSPP_ARCH(AMD64)
#include <emmintrin.h>
#if PPSSPP_ARCH(X86) || PPSSPP_ARCH(AMD64)
#include "Common/CPUDetect.h"
#include "Common/Data/Convert/ColorConv.h"
#include "Common/Math/SIMDHeaders.h"
#include "Core/Config.h"
#include "GPU/GPUState.h"
#include "GPU/Common/VertexDecoderCommon.h"

2
GPU/GPUState.cpp
View file

@ -17,13 +17,13 @@
#include "ppsspp_config.h"
#include "Common/Common.h"
#include "Common/Math/SIMDHeaders.h"
#include "Common/Serialize/Serializer.h"
#include "Common/Serialize/SerializeFuncs.h"
#include "Core/MemMap.h"
#include "GPU/ge_constants.h"
#include "GPU/GPUCommon.h"
#include "GPU/GPUState.h"
#include "Common/Math/SIMDHeaders.h"
// This must be aligned so that the matrices within are aligned.
alignas(16) GPUgstate gstate;

3
GPU/Math3D.cpp
View file

@ -15,8 +15,9 @@
// Official git repository and contact information can be found at
// https://github.com/hrydgard/ppsspp and http://www.ppsspp.org/.
#include "Common/Common.h"
#include "GPU/Math3D.h"
#include "Common/Common.h"
#include "Common/Math/SIMDHeaders.h"
#if PPSSPP_ARCH(SSE2)
// For the SSE4 stuff.

3
GPU/Software/DrawPixelX86.cpp
View file

@ -16,12 +16,13 @@
// https://github.com/hrydgard/ppsspp and http://www.ppsspp.org/.
#include "ppsspp_config.h"
#if PPSSPP_ARCH(AMD64)
#include <emmintrin.h>
#include "Common/x64Emitter.h"
#include "Common/CPUDetect.h"
#include "Common/LogReporting.h"
#include "Common/Math/SIMDHeaders.h"
#include "GPU/GPUState.h"
#include "GPU/Software/DrawPixel.h"
#include "GPU/Software/SoftGpu.h"

5
GPU/Software/Rasterizer.cpp
View file

@ -38,6 +38,11 @@
#include "Common/Math/SIMDHeaders.h"
// For the SSE4 stuff
#if PPSSPP_ARCH(SSE2)
#include <smmintrin.h>
#endif
namespace Rasterizer {
// Only OK on x64 where our stack is aligned

5
GPU/Software/RasterizerRegCache.h
View file

@ -25,9 +25,8 @@
#include <vector>
#include "Common/Common.h"
#if defined(_M_SSE)
#include <emmintrin.h>
#endif
#include "Common/Math/SIMDHeaders.h"
#if PPSSPP_ARCH(ARM64_NEON)
#if defined(_MSC_VER) && PPSSPP_ARCH(ARM64)
#include <arm64_neon.h>

2
GPU/Software/SamplerX86.cpp
View file

@ -18,7 +18,7 @@
#include "ppsspp_config.h"
#if PPSSPP_ARCH(X86) || PPSSPP_ARCH(AMD64)
#include <emmintrin.h>
#include "Common/Math/SIMDHeaders.h"
#include "Common/x64Emitter.h"
#include "Common/BitScan.h"
#include "Common/CPUDetect.h"

9
GPU/Software/TransformUnit.cpp
View file

@ -15,7 +15,10 @@
// Official git repository and contact information can be found at
// https://github.com/hrydgard/ppsspp and http://www.ppsspp.org/.
#include "ppsspp_config.h"
#include <cmath>
#include "Common/Common.h"
#include "Common/CPUDetect.h"
#include "Common/Math/math_util.h"
@ -25,12 +28,18 @@
#include "GPU/Common/DrawEngineCommon.h"
#include "GPU/Common/VertexDecoderCommon.h"
#include "GPU/Common/SoftwareTransformCommon.h"
#include "Common/Math/SIMDHeaders.h"
#include "GPU/Software/BinManager.h"
#include "GPU/Software/Clipper.h"
#include "GPU/Software/Lighting.h"
#include "GPU/Software/RasterizerRectangle.h"
#include "GPU/Software/TransformUnit.h"
// For the SSE4 stuff
#if PPSSPP_ARCH(SSE2)
#include <smmintrin.h>
#endif
#define TRANSFORM_BUF_SIZE (65536 * 48)
TransformUnit::TransformUnit() {

2
ext/at3_standalone/atrac3plusdsp.cpp
View file

@ -659,7 +659,7 @@ void ff_atrac3p_ipqf(FFTContext *dct_ctx, Atrac3pIPQFChannelCtx *hist,
const float *coeffs2 = ipqf_coeffs2[t];
float *outp = out + s * 16;
#if PPSSPP_ARCH(X86) || PPSSPP_ARCH(AMD64)
#if PPSSPP_ARCH(SSE2)
auto _mm_reverse = [](__m128 x) -> __m128 {
return _mm_shuffle_ps(x, x, _MM_SHUFFLE(0, 1, 2, 3));
};