@@ -8,4 +8,5 @@ LSX-OBJS-$(CONFIG_SWSCALE) += loongarch/swscale.o \
loongarch/swscale_lsx.o \
loongarch/input.o \
loongarch/output.o \
- loongarch/output_lsx.o
+ loongarch/output_lsx.o \
+ loongarch/yuv2rgb_lsx.o
@@ -90,8 +90,8 @@ av_cold void rgb2rgb_init_loongarch(void)
av_cold SwsFunc ff_yuv2rgb_init_loongarch(SwsContext *c)
{
-#if HAVE_LASX
int cpu_flags = av_get_cpu_flags();
+#if HAVE_LASX
if (have_lasx(cpu_flags)) {
switch (c->dstFormat) {
case AV_PIX_FMT_RGB24:
@@ -121,5 +121,33 @@ av_cold SwsFunc ff_yuv2rgb_init_loongarch(SwsContext *c)
}
}
#endif // #if HAVE_LASX
+ if (have_lsx(cpu_flags)) {
+ switch (c->dstFormat) {
+ case AV_PIX_FMT_RGB24:
+ return yuv420_rgb24_lsx;
+ case AV_PIX_FMT_BGR24:
+ return yuv420_bgr24_lsx;
+ case AV_PIX_FMT_RGBA:
+ if (CONFIG_SWSCALE_ALPHA && isALPHA(c->srcFormat)) {
+ break;
+ } else
+ return yuv420_rgba32_lsx;
+ case AV_PIX_FMT_ARGB:
+ if (CONFIG_SWSCALE_ALPHA && isALPHA(c->srcFormat)) {
+ break;
+ } else
+ return yuv420_argb32_lsx;
+ case AV_PIX_FMT_BGRA:
+ if (CONFIG_SWSCALE_ALPHA && isALPHA(c->srcFormat)) {
+ break;
+ } else
+ return yuv420_bgra32_lsx;
+ case AV_PIX_FMT_ABGR:
+ if (CONFIG_SWSCALE_ALPHA && isALPHA(c->srcFormat)) {
+ break;
+ } else
+ return yuv420_abgr32_lsx;
+ }
+ }
return NULL;
}
@@ -62,6 +62,24 @@ void ff_yuv2planeX_8_lsx(const int16_t *filter, int filterSize,
av_cold void ff_sws_init_output_lsx(SwsContext *c);
+int yuv420_rgb24_lsx(SwsContext *c, const uint8_t *src[], int srcStride[],
+ int srcSliceY, int srcSliceH, uint8_t *dst[], int dstStride[]);
+
+int yuv420_bgr24_lsx(SwsContext *c, const uint8_t *src[], int srcStride[],
+ int srcSliceY, int srcSliceH, uint8_t *dst[], int dstStride[]);
+
+int yuv420_rgba32_lsx(SwsContext *c, const uint8_t *src[], int srcStride[],
+ int srcSliceY, int srcSliceH, uint8_t *dst[], int dstStride[]);
+
+int yuv420_bgra32_lsx(SwsContext *c, const uint8_t *src[], int srcStride[],
+ int srcSliceY, int srcSliceH, uint8_t *dst[], int dstStride[]);
+
+int yuv420_argb32_lsx(SwsContext *c, const uint8_t *src[], int srcStride[],
+ int srcSliceY, int srcSliceH, uint8_t *dst[], int dstStride[]);
+
+int yuv420_abgr32_lsx(SwsContext *c, const uint8_t *src[], int srcStride[],
+ int srcSliceY, int srcSliceH, uint8_t *dst[], int dstStride[]);
+
#if HAVE_LASX
void ff_hscale_8_to_15_lasx(SwsContext *c, int16_t *dst, int dstW,
const uint8_t *src, const int16_t *filter,
new file mode 100644
@@ -0,0 +1,361 @@
+/*
+ * Copyright (C) 2023 Loongson Technology Co. Ltd.
+ * Contributed by Bo Jin(jinbo@loongson.cn)
+ * All rights reserved.
+ *
+ * This file is part of FFmpeg.
+ *
+ * FFmpeg is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU Lesser General Public
+ * License as published by the Free Software Foundation; either
+ * version 2.1 of the License, or (at your option) any later version.
+ *
+ * FFmpeg is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
+ * Lesser General Public License for more details.
+ *
+ * You should have received a copy of the GNU Lesser General Public
+ * License along with FFmpeg; if not, write to the Free Software
+ * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
+ */
+
+#include "swscale_loongarch.h"
+#include "libavutil/loongarch/loongson_intrinsics.h"
+
+#define YUV2RGB_LOAD_COE \
+ /* Load x_offset */ \
+ __m128i y_offset = __lsx_vreplgr2vr_d(c->yOffset); \
+ __m128i u_offset = __lsx_vreplgr2vr_d(c->uOffset); \
+ __m128i v_offset = __lsx_vreplgr2vr_d(c->vOffset); \
+ /* Load x_coeff */ \
+ __m128i ug_coeff = __lsx_vreplgr2vr_d(c->ugCoeff); \
+ __m128i vg_coeff = __lsx_vreplgr2vr_d(c->vgCoeff); \
+ __m128i y_coeff = __lsx_vreplgr2vr_d(c->yCoeff); \
+ __m128i ub_coeff = __lsx_vreplgr2vr_d(c->ubCoeff); \
+ __m128i vr_coeff = __lsx_vreplgr2vr_d(c->vrCoeff); \
+
+#define LOAD_YUV_16 \
+ m_y1 = __lsx_vld(py_1, 0); \
+ m_y2 = __lsx_vld(py_2, 0); \
+ m_u = __lsx_vldrepl_d(pu, 0); \
+ m_v = __lsx_vldrepl_d(pv, 0); \
+ DUP2_ARG2(__lsx_vilvl_b, m_u, m_u, m_v, m_v, m_u, m_v); \
+ DUP2_ARG2(__lsx_vilvh_b, zero, m_u, zero, m_v, m_u_h, m_v_h); \
+ DUP2_ARG2(__lsx_vilvl_b, zero, m_u, zero, m_v, m_u, m_v); \
+ DUP2_ARG2(__lsx_vilvh_b, zero, m_y1, zero, m_y2, m_y1_h, m_y2_h); \
+ DUP2_ARG2(__lsx_vilvl_b, zero, m_y1, zero, m_y2, m_y1, m_y2); \
+
+/* YUV2RGB method
+ * The conversion method is as follows:
+ * R = Y' * y_coeff + V' * vr_coeff
+ * G = Y' * y_coeff + V' * vg_coeff + U' * ug_coeff
+ * B = Y' * y_coeff + U' * ub_coeff
+ *
+ * where X' = X * 8 - x_offset
+ *
+ */
+
+#define YUV2RGB(y1, y2, u, v, r1, g1, b1, r2, g2, b2) \
+{ \
+ y1 = __lsx_vslli_h(y1, 3); \
+ y2 = __lsx_vslli_h(y2, 3); \
+ u = __lsx_vslli_h(u, 3); \
+ v = __lsx_vslli_h(v, 3); \
+ y1 = __lsx_vsub_h(y1, y_offset); \
+ y2 = __lsx_vsub_h(y2, y_offset); \
+ u = __lsx_vsub_h(u, u_offset); \
+ v = __lsx_vsub_h(v, v_offset); \
+ y_1 = __lsx_vmuh_h(y1, y_coeff); \
+ y_2 = __lsx_vmuh_h(y2, y_coeff); \
+ u2g = __lsx_vmuh_h(u, ug_coeff); \
+ u2b = __lsx_vmuh_h(u, ub_coeff); \
+ v2r = __lsx_vmuh_h(v, vr_coeff); \
+ v2g = __lsx_vmuh_h(v, vg_coeff); \
+ r1 = __lsx_vsadd_h(y_1, v2r); \
+ v2g = __lsx_vsadd_h(v2g, u2g); \
+ g1 = __lsx_vsadd_h(y_1, v2g); \
+ b1 = __lsx_vsadd_h(y_1, u2b); \
+ r2 = __lsx_vsadd_h(y_2, v2r); \
+ g2 = __lsx_vsadd_h(y_2, v2g); \
+ b2 = __lsx_vsadd_h(y_2, u2b); \
+ DUP4_ARG1(__lsx_vclip255_h, r1, g1, b1, r2, r1, g1, b1, r2); \
+ DUP2_ARG1(__lsx_vclip255_h, g2, b2, g2, b2); \
+}
+
+#define RGB_PACK(r, g, b, rgb_l, rgb_h) \
+{ \
+ __m128i rg; \
+ rg = __lsx_vpackev_b(g, r); \
+ DUP2_ARG3(__lsx_vshuf_b, b, rg, shuf2, b, rg, shuf3, rgb_l, rgb_h); \
+}
+
+#define RGB32_PACK(a, r, g, b, rgb_l, rgb_h) \
+{ \
+ __m128i ra, bg; \
+ ra = __lsx_vpackev_b(r, a); \
+ bg = __lsx_vpackev_b(b, g); \
+ rgb_l = __lsx_vilvl_h(bg, ra); \
+ rgb_h = __lsx_vilvh_h(bg, ra); \
+}
+
+#define RGB_STORE(rgb_l, rgb_h, image) \
+{ \
+ __lsx_vstelm_d(rgb_l, image, 0, 0); \
+ __lsx_vstelm_d(rgb_l, image, 8, 1); \
+ __lsx_vstelm_d(rgb_h, image, 16, 0); \
+}
+
+#define RGB32_STORE(rgb_l, rgb_h, image) \
+{ \
+ __lsx_vst(rgb_l, image, 0); \
+ __lsx_vst(rgb_h, image, 16); \
+}
+
+#define YUV2RGBFUNC(func_name, dst_type, alpha) \
+ int func_name(SwsContext *c, const uint8_t *src[], \
+ int srcStride[], int srcSliceY, int srcSliceH, \
+ uint8_t *dst[], int dstStride[]) \
+{ \
+ int x, y, h_size, vshift, res; \
+ __m128i m_y1, m_y2, m_u, m_v; \
+ __m128i m_y1_h, m_y2_h, m_u_h, m_v_h; \
+ __m128i y_1, y_2, u2g, v2g, u2b, v2r, rgb1_l, rgb1_h; \
+ __m128i rgb2_l, rgb2_h, r1, g1, b1, r2, g2, b2; \
+ __m128i shuf2 = {0x0504120302100100, 0x0A18090816070614}; \
+ __m128i shuf3 = {0x1E0F0E1C0D0C1A0B, 0x0101010101010101}; \
+ __m128i zero = __lsx_vldi(0); \
+ \
+ YUV2RGB_LOAD_COE \
+ \
+ h_size = c->dstW >> 4; \
+ res = (c->dstW & 15) >> 1; \
+ vshift = c->srcFormat != AV_PIX_FMT_YUV422P; \
+ for (y = 0; y < srcSliceH; y += 2) { \
+ dst_type av_unused *r, *g, *b; \
+ dst_type *image1 = (dst_type *)(dst[0] + (y + srcSliceY) * dstStride[0]);\
+ dst_type *image2 = (dst_type *)(image1 + dstStride[0]);\
+ const uint8_t *py_1 = src[0] + y * srcStride[0]; \
+ const uint8_t *py_2 = py_1 + srcStride[0]; \
+ const uint8_t *pu = src[1] + (y >> vshift) * srcStride[1]; \
+ const uint8_t *pv = src[2] + (y >> vshift) * srcStride[2]; \
+ for(x = 0; x < h_size; x++) { \
+
+#define YUV2RGBFUNC32(func_name, dst_type, alpha) \
+ int func_name(SwsContext *c, const uint8_t *src[], \
+ int srcStride[], int srcSliceY, int srcSliceH, \
+ uint8_t *dst[], int dstStride[]) \
+{ \
+ int x, y, h_size, vshift, res; \
+ __m128i m_y1, m_y2, m_u, m_v; \
+ __m128i m_y1_h, m_y2_h, m_u_h, m_v_h; \
+ __m128i y_1, y_2, u2g, v2g, u2b, v2r, rgb1_l, rgb1_h; \
+ __m128i rgb2_l, rgb2_h, r1, g1, b1, r2, g2, b2; \
+ __m128i a = __lsx_vldi(0xFF); \
+ __m128i zero = __lsx_vldi(0); \
+ \
+ YUV2RGB_LOAD_COE \
+ \
+ h_size = c->dstW >> 4; \
+ res = (c->dstW & 15) >> 1; \
+ vshift = c->srcFormat != AV_PIX_FMT_YUV422P; \
+ for (y = 0; y < srcSliceH; y += 2) { \
+ int yd = y + srcSliceY; \
+ dst_type av_unused *r, *g, *b; \
+ dst_type *image1 = (dst_type *)(dst[0] + (yd) * dstStride[0]); \
+ dst_type *image2 = (dst_type *)(dst[0] + (yd + 1) * dstStride[0]); \
+ const uint8_t *py_1 = src[0] + y * srcStride[0]; \
+ const uint8_t *py_2 = py_1 + srcStride[0]; \
+ const uint8_t *pu = src[1] + (y >> vshift) * srcStride[1]; \
+ const uint8_t *pv = src[2] + (y >> vshift) * srcStride[2]; \
+ for(x = 0; x < h_size; x++) { \
+
+#define DEALYUV2RGBREMAIN \
+ py_1 += 16; \
+ py_2 += 16; \
+ pu += 8; \
+ pv += 8; \
+ image1 += 48; \
+ image2 += 48; \
+ } \
+ for (x = 0; x < res; x++) { \
+ int av_unused U, V, Y; \
+ U = pu[0]; \
+ V = pv[0]; \
+ r = (void *)c->table_rV[V+YUVRGB_TABLE_HEADROOM]; \
+ g = (void *)(c->table_gU[U+YUVRGB_TABLE_HEADROOM] \
+ + c->table_gV[V+YUVRGB_TABLE_HEADROOM]); \
+ b = (void *)c->table_bU[U+YUVRGB_TABLE_HEADROOM];
+
+#define DEALYUV2RGBREMAIN32 \
+ py_1 += 16; \
+ py_2 += 16; \
+ pu += 8; \
+ pv += 8; \
+ image1 += 16; \
+ image2 += 16; \
+ } \
+ for (x = 0; x < res; x++) { \
+ int av_unused U, V, Y; \
+ U = pu[0]; \
+ V = pv[0]; \
+ r = (void *)c->table_rV[V+YUVRGB_TABLE_HEADROOM]; \
+ g = (void *)(c->table_gU[U+YUVRGB_TABLE_HEADROOM] \
+ + c->table_gV[V+YUVRGB_TABLE_HEADROOM]); \
+ b = (void *)c->table_bU[U+YUVRGB_TABLE_HEADROOM]; \
+
+#define PUTRGB24(dst, src) \
+ Y = src[0]; \
+ dst[0] = r[Y]; \
+ dst[1] = g[Y]; \
+ dst[2] = b[Y]; \
+ Y = src[1]; \
+ dst[3] = r[Y]; \
+ dst[4] = g[Y]; \
+ dst[5] = b[Y];
+
+#define PUTBGR24(dst, src) \
+ Y = src[0]; \
+ dst[0] = b[Y]; \
+ dst[1] = g[Y]; \
+ dst[2] = r[Y]; \
+ Y = src[1]; \
+ dst[3] = b[Y]; \
+ dst[4] = g[Y]; \
+ dst[5] = r[Y];
+
+#define PUTRGB(dst, src) \
+ Y = src[0]; \
+ dst[0] = r[Y] + g[Y] + b[Y]; \
+ Y = src[1]; \
+ dst[1] = r[Y] + g[Y] + b[Y]; \
+
+#define ENDRES \
+ pu += 1; \
+ pv += 1; \
+ py_1 += 2; \
+ py_2 += 2; \
+ image1 += 6; \
+ image2 += 6; \
+
+#define ENDRES32 \
+ pu += 1; \
+ pv += 1; \
+ py_1 += 2; \
+ py_2 += 2; \
+ image1 += 2; \
+ image2 += 2; \
+
+#define END_FUNC() \
+ } \
+ } \
+ return srcSliceH; \
+}
+
+YUV2RGBFUNC(yuv420_rgb24_lsx, uint8_t, 0)
+ LOAD_YUV_16
+ YUV2RGB(m_y1, m_y2, m_u, m_v, r1, g1, b1, r2, g2, b2);
+ RGB_PACK(r1, g1, b1, rgb1_l, rgb1_h);
+ RGB_PACK(r2, g2, b2, rgb2_l, rgb2_h);
+ RGB_STORE(rgb1_l, rgb1_h, image1);
+ RGB_STORE(rgb2_l, rgb2_h, image2);
+ YUV2RGB(m_y1_h, m_y2_h, m_u_h, m_v_h, r1, g1, b1, r2, g2, b2);
+ RGB_PACK(r1, g1, b1, rgb1_l, rgb1_h);
+ RGB_PACK(r2, g2, b2, rgb2_l, rgb2_h);
+ RGB_STORE(rgb1_l, rgb1_h, image1 + 24);
+ RGB_STORE(rgb2_l, rgb2_h, image2 + 24);
+ DEALYUV2RGBREMAIN
+ PUTRGB24(image1, py_1);
+ PUTRGB24(image2, py_2);
+ ENDRES
+ END_FUNC()
+
+YUV2RGBFUNC(yuv420_bgr24_lsx, uint8_t, 0)
+ LOAD_YUV_16
+ YUV2RGB(m_y1, m_y2, m_u, m_v, r1, g1, b1, r2, g2, b2);
+ RGB_PACK(b1, g1, r1, rgb1_l, rgb1_h);
+ RGB_PACK(b2, g2, r2, rgb2_l, rgb2_h);
+ RGB_STORE(rgb1_l, rgb1_h, image1);
+ RGB_STORE(rgb2_l, rgb2_h, image2);
+ YUV2RGB(m_y1_h, m_y2_h, m_u_h, m_v_h, r1, g1, b1, r2, g2, b2);
+ RGB_PACK(b1, g1, r1, rgb1_l, rgb1_h);
+ RGB_PACK(b2, g2, r2, rgb2_l, rgb2_h);
+ RGB_STORE(rgb1_l, rgb1_h, image1 + 24);
+ RGB_STORE(rgb2_l, rgb2_h, image2 + 24);
+ DEALYUV2RGBREMAIN
+ PUTBGR24(image1, py_1);
+ PUTBGR24(image2, py_2);
+ ENDRES
+ END_FUNC()
+
+YUV2RGBFUNC32(yuv420_rgba32_lsx, uint32_t, 0)
+ LOAD_YUV_16
+ YUV2RGB(m_y1, m_y2, m_u, m_v, r1, g1, b1, r2, g2, b2);
+ RGB32_PACK(r1, g1, b1, a, rgb1_l, rgb1_h);
+ RGB32_PACK(r2, g2, b2, a, rgb2_l, rgb2_h);
+ RGB32_STORE(rgb1_l, rgb1_h, image1);
+ RGB32_STORE(rgb2_l, rgb2_h, image2);
+ YUV2RGB(m_y1_h, m_y2_h, m_u_h, m_v_h, r1, g1, b1, r2, g2, b2);
+ RGB32_PACK(r1, g1, b1, a, rgb1_l, rgb1_h);
+ RGB32_PACK(r2, g2, b2, a, rgb2_l, rgb2_h);
+ RGB32_STORE(rgb1_l, rgb1_h, image1 + 8);
+ RGB32_STORE(rgb2_l, rgb2_h, image2 + 8);
+ DEALYUV2RGBREMAIN32
+ PUTRGB(image1, py_1);
+ PUTRGB(image2, py_2);
+ ENDRES32
+ END_FUNC()
+
+YUV2RGBFUNC32(yuv420_bgra32_lsx, uint32_t, 0)
+ LOAD_YUV_16
+ YUV2RGB(m_y1, m_y2, m_u, m_v, r1, g1, b1, r2, g2, b2);
+ RGB32_PACK(b1, g1, r1, a, rgb1_l, rgb1_h);
+ RGB32_PACK(b2, g2, r2, a, rgb2_l, rgb2_h);
+ RGB32_STORE(rgb1_l, rgb1_h, image1);
+ RGB32_STORE(rgb2_l, rgb2_h, image2);
+ YUV2RGB(m_y1_h, m_y2_h, m_u_h, m_v_h, r1, g1, b1, r2, g2, b2);
+ RGB32_PACK(b1, g1, r1, a, rgb1_l, rgb1_h);
+ RGB32_PACK(b2, g2, r2, a, rgb2_l, rgb2_h);
+ RGB32_STORE(rgb1_l, rgb1_h, image1 + 8);
+ RGB32_STORE(rgb2_l, rgb2_h, image2 + 8);
+ DEALYUV2RGBREMAIN32
+ PUTRGB(image1, py_1);
+ PUTRGB(image2, py_2);
+ ENDRES32
+ END_FUNC()
+
+YUV2RGBFUNC32(yuv420_argb32_lsx, uint32_t, 0)
+ LOAD_YUV_16
+ YUV2RGB(m_y1, m_y2, m_u, m_v, r1, g1, b1, r2, g2, b2);
+ RGB32_PACK(a, r1, g1, b1, rgb1_l, rgb1_h);
+ RGB32_PACK(a, r2, g2, b2, rgb2_l, rgb2_h);
+ RGB32_STORE(rgb1_l, rgb1_h, image1);
+ RGB32_STORE(rgb2_l, rgb2_h, image2);
+ YUV2RGB(m_y1_h, m_y2_h, m_u_h, m_v_h, r1, g1, b1, r2, g2, b2);
+ RGB32_PACK(a, r1, g1, b1, rgb1_l, rgb1_h);
+ RGB32_PACK(a, r2, g2, b2, rgb2_l, rgb2_h);
+ RGB32_STORE(rgb1_l, rgb1_h, image1 + 8);
+ RGB32_STORE(rgb2_l, rgb2_h, image2 + 8);
+ DEALYUV2RGBREMAIN32
+ PUTRGB(image1, py_1);
+ PUTRGB(image2, py_2);
+ ENDRES32
+ END_FUNC()
+
+YUV2RGBFUNC32(yuv420_abgr32_lsx, uint32_t, 0)
+ LOAD_YUV_16
+ YUV2RGB(m_y1, m_y2, m_u, m_v, r1, g1, b1, r2, g2, b2);
+ RGB32_PACK(a, b1, g1, r1, rgb1_l, rgb1_h);
+ RGB32_PACK(a, b2, g2, r2, rgb2_l, rgb2_h);
+ RGB32_STORE(rgb1_l, rgb1_h, image1);
+ RGB32_STORE(rgb2_l, rgb2_h, image2);
+ YUV2RGB(m_y1_h, m_y2_h, m_u_h, m_v_h, r1, g1, b1, r2, g2, b2);
+ RGB32_PACK(a, b1, g1, r1, rgb1_l, rgb1_h);
+ RGB32_PACK(a, b2, g2, r2, rgb2_l, rgb2_h);
+ RGB32_STORE(rgb1_l, rgb1_h, image1 + 8);
+ RGB32_STORE(rgb2_l, rgb2_h, image2 + 8);
+ DEALYUV2RGBREMAIN32
+ PUTRGB(image1, py_1);
+ PUTRGB(image2, py_2);
+ ENDRES32
+ END_FUNC()
From: Jin Bo <jinbo@loongson.cn> yuv420_rgb24_lsx yuv420_bgr24_lsx yuv420_rgba32_lsx yuv420_argb32_lsx yuv420_bgra32_lsx yuv420_abgr32_lsx ./configure --disable-lasx ffmpeg -i ~/media/1_h264_1080p_30fps_3Mbps.mp4 -f rawvideo -pix_fmt rgb24 -y /dev/null -an before: 184fps after: 207fps --- libswscale/loongarch/Makefile | 3 +- libswscale/loongarch/swscale_init_loongarch.c | 30 +- libswscale/loongarch/swscale_loongarch.h | 18 + libswscale/loongarch/yuv2rgb_lsx.c | 361 ++++++++++++++++++ 4 files changed, 410 insertions(+), 2 deletions(-) create mode 100644 libswscale/loongarch/yuv2rgb_lsx.c