preserve-emulation/ppsspp
1
0 Fork 0
mirror of https://github.com/hrydgard/ppsspp.git synced 2026-10-11 05:26:24 +02:00
Code Issues Packages Projects Releases Wiki Activity

Merge separate NEON functions into the normal functions.

Browse Source 
We no longer support non-NEON ARM.

It's nice also to have the NEON and SSE implementations "close" to each
other, easier to port optimizations back and forth etc.
This commit is contained in:
Henrik Rydgård committed 2022-04-12 23:43:21 +02:00
1 parent 6de574104c
commit a68ddd0a8d
34 files changed
+437 -855
Download Patch File Download Diff File Expand all files Collapse all files

No files matched your search

Common/Data/Convert/ColorConv.cpp
+78 -24
View File
@@ -18,8 +18,6 @@
#include "ppsspp_config.h"
#include "Common/Data/Convert/ColorConv.h"
#include "Common/Data/Convert/SmallDataConvert.h"
// NEON is in a separate file so that it can be compiled with a runtime check.
#include "Common/Data/Convert/ColorConvNEON.h"
#include "Common/Common.h"
#include "Common/CPUDetect.h"
@@ -28,6 +26,14 @@
#include <smmintrin.h>
#endif
#if PPSSPP_ARCH(ARM_NEON)
#if defined(_MSC_VER) && PPSSPP_ARCH(ARM64)
#include <arm64_neon.h>
#else
#include <arm_neon.h>
#endif
#endif
inline u16 RGBA8888toRGB565(u32 px) {
return ((px >> 3) & 0x001F) | ((px >> 5) & 0x07E0) | ((px >> 8) & 0xF800);
}
@@ -134,8 +140,6 @@ void convert5551_dx9(u16* data, u32* out, int width, int l, int u) {
}
}
void ConvertBGRA8888ToRGBA8888(u32 *dst, const u32 *src, u32 numPixels) {
#ifdef _M_SSE
const __m128i maskGA = _mm_set1_epi32(0xFF00FF00);
@@ -540,7 +544,7 @@ void ConvertRGB565ToBGRA8888(u32 *dst, const u16 *src, u32 numPixels) {
}
}
void ConvertRGBA4444ToABGR4444Basic(u16 *dst, const u16 *src, u32 numPixels) {
void ConvertRGBA4444ToABGR4444(u16 *dst, const u16 *src, u32 numPixels) {
#ifdef _M_SSE
const __m128i mask0040 = _mm_set1_epi16(0x00F0);
@@ -560,6 +564,28 @@ void ConvertRGBA4444ToABGR4444Basic(u16 *dst, const u16 *src, u32 numPixels) {
}
// The remainder is done in chunks of 2, SSE was chunks of 8.
u32 i = sseChunks * 8 / 2;
#elif PPSSPP_ARCH(ARM_NEON)
const uint16x8_t mask0040 = vdupq_n_u16(0x00F0);
if (((uintptr_t)dst & 15) == 0 && ((uintptr_t)src & 15) == 0) {
u32 simdable = (numPixels / 8) * 8;
for (u32 i = 0; i < simdable; i += 8) {
uint16x8_t c = vld1q_u16(src);
const uint16x8_t a = vshrq_n_u16(c, 12);
const uint16x8_t b = vandq_u16(vshrq_n_u16(c, 4), mask0040);
const uint16x8_t g = vshlq_n_u16(vandq_u16(c, mask0040), 4);
const uint16x8_t r = vshlq_n_u16(c, 12);
uint16x8_t res = vorrq_u16(vorrq_u16(r, g), vorrq_u16(b, a));
vst1q_u16(dst, res);
src += 8;
dst += 8;
}
numPixels -= simdable;
}
u32 i = 0; // already moved the pointers forward
#else
u32 i = 0;
#endif
@@ -584,7 +610,7 @@ void ConvertRGBA4444ToABGR4444Basic(u16 *dst, const u16 *src, u32 numPixels) {
}
}
void ConvertRGBA5551ToABGR1555Basic(u16 *dst, const u16 *src, u32 numPixels) {
void ConvertRGBA5551ToABGR1555(u16 *dst, const u16 *src, u32 numPixels) {
#ifdef _M_SSE
const __m128i maskB = _mm_set1_epi16(0x003E);
const __m128i maskG = _mm_set1_epi16(0x07C0);
@@ -605,6 +631,29 @@ void ConvertRGBA5551ToABGR1555Basic(u16 *dst, const u16 *src, u32 numPixels) {
}
// The remainder is done in chunks of 2, SSE was chunks of 8.
u32 i = sseChunks * 8 / 2;
#elif PPSSPP_ARCH(ARM_NEON)
const uint16x8_t maskB = vdupq_n_u16(0x003E);
const uint16x8_t maskG = vdupq_n_u16(0x07C0);
if (((uintptr_t)dst & 15) == 0 && ((uintptr_t)src & 15) == 0) {
u32 simdable = (numPixels / 8) * 8;
for (u32 i = 0; i < simdable; i += 8) {
uint16x8_t c = vld1q_u16(src);
const uint16x8_t a = vshrq_n_u16(c, 15);
const uint16x8_t b = vandq_u16(vshrq_n_u16(c, 9), maskB);
const uint16x8_t g = vandq_u16(vshlq_n_u16(c, 1), maskG);
const uint16x8_t r = vshlq_n_u16(c, 11);
uint16x8_t res = vorrq_u16(vorrq_u16(r, g), vorrq_u16(b, a));
vst1q_u16(dst, res);
src += 8;
dst += 8;
}
numPixels -= simdable;
}
u32 i = 0;
#else
u32 i = 0;
#endif
@@ -629,7 +678,7 @@ void ConvertRGBA5551ToABGR1555Basic(u16 *dst, const u16 *src, u32 numPixels) {
}
}
void ConvertRGB565ToBGR565Basic(u16 *dst, const u16 *src, u32 numPixels) {
void ConvertRGB565ToBGR565(u16 *dst, const u16 *src, u32 numPixels) {
#ifdef _M_SSE
const __m128i maskG = _mm_set1_epi16(0x07E0);
@@ -648,6 +697,28 @@ void ConvertRGB565ToBGR565Basic(u16 *dst, const u16 *src, u32 numPixels) {
}
// The remainder is done in chunks of 2, SSE was chunks of 8.
u32 i = sseChunks * 8 / 2;
#elif PPSSPP_ARCH(ARM_NEON)
const uint16x8_t maskG = vdupq_n_u16(0x07E0);
if (((uintptr_t)dst & 15) == 0 && ((uintptr_t)src & 15) == 0) {
u32 simdable = (numPixels / 8) * 8;
for (u32 i = 0; i < simdable; i += 8) {
uint16x8_t c = vld1q_u16(src);
const uint16x8_t b = vshrq_n_u16(c, 11);
const uint16x8_t g = vandq_u16(c, maskG);
const uint16x8_t r = vshlq_n_u16(c, 11);
uint16x8_t res = vorrq_u16(vorrq_u16(r, g), b);
vst1q_u16(dst, res);
src += 8;
dst += 8;
}
numPixels -= simdable;
}
u32 i = 0;
#else
u32 i = 0;
#endif
@@ -684,20 +755,3 @@ void ConvertBGRA5551ToABGR1555(u16 *dst, const u16 *src, u32 numPixels) {
dst[i] = (c >> 15) | (c << 1);
}
}
// Reuse the logic from the header - if these aren't defined, we need externs.
#ifndef ConvertRGBA4444ToABGR4444
Convert16bppTo16bppFunc ConvertRGBA4444ToABGR4444 = &ConvertRGBA4444ToABGR4444Basic;
Convert16bppTo16bppFunc ConvertRGBA5551ToABGR1555 = &ConvertRGBA5551ToABGR1555Basic;
Convert16bppTo16bppFunc ConvertRGB565ToBGR565 = &ConvertRGB565ToBGR565Basic;
#endif
void SetupColorConv() {
#if PPSSPP_ARCH(ARM_NEON) && !PPSSPP_ARCH(ARM64)
if (cpu_info.bNEON) {
ConvertRGBA4444ToABGR4444 = &ConvertRGBA4444ToABGR4444NEON;
ConvertRGBA5551ToABGR1555 = &ConvertRGBA5551ToABGR1555NEON;
ConvertRGB565ToBGR565 = &ConvertRGB565ToBGR565NEON;
}
#endif
}