swscale/x86/output.asm: add x86-optimized planer gbr yuv2anyX functions

changes since v2:
 * fixed label
changes since v1:
 * remove vex intruction on sse4 path
 * some load/pack marcos use less intructions
 * fixed some typos

yuv2gbrp_full_X_4_512_c: 12757.6
yuv2gbrp_full_X_4_512_sse2: 8946.6
yuv2gbrp_full_X_4_512_sse4: 5138.6
yuv2gbrp_full_X_4_512_avx2: 3889.6
yuv2gbrap_full_X_4_512_c: 15368.6
yuv2gbrap_full_X_4_512_sse2: 11916.1
yuv2gbrap_full_X_4_512_sse4: 6294.6
yuv2gbrap_full_X_4_512_avx2: 3477.1
yuv2gbrp9be_full_X_4_512_c: 14381.6
yuv2gbrp9be_full_X_4_512_sse2: 9139.1
yuv2gbrp9be_full_X_4_512_sse4: 5150.1
yuv2gbrp9be_full_X_4_512_avx2: 2834.6
yuv2gbrp9le_full_X_4_512_c: 12990.1
yuv2gbrp9le_full_X_4_512_sse2: 9118.1
yuv2gbrp9le_full_X_4_512_sse4: 5132.1
yuv2gbrp9le_full_X_4_512_avx2: 2833.1
yuv2gbrp10be_full_X_4_512_c: 14401.6
yuv2gbrp10be_full_X_4_512_sse2: 9133.1
yuv2gbrp10be_full_X_4_512_sse4: 5126.1
yuv2gbrp10be_full_X_4_512_avx2: 2837.6
yuv2gbrp10le_full_X_4_512_c: 12718.1
yuv2gbrp10le_full_X_4_512_sse2: 9106.1
yuv2gbrp10le_full_X_4_512_sse4: 5120.1
yuv2gbrp10le_full_X_4_512_avx2: 2826.1
yuv2gbrap10be_full_X_4_512_c: 18535.6
yuv2gbrap10be_full_X_4_512_sse2: 33617.6
yuv2gbrap10be_full_X_4_512_sse4: 6264.1
yuv2gbrap10be_full_X_4_512_avx2: 3422.1
yuv2gbrap10le_full_X_4_512_c: 16724.1
yuv2gbrap10le_full_X_4_512_sse2: 11787.1
yuv2gbrap10le_full_X_4_512_sse4: 6282.1
yuv2gbrap10le_full_X_4_512_avx2: 3441.6
yuv2gbrp12be_full_X_4_512_c: 13723.6
yuv2gbrp12be_full_X_4_512_sse2: 9128.1
yuv2gbrp12be_full_X_4_512_sse4: 7997.6
yuv2gbrp12be_full_X_4_512_avx2: 2844.1
yuv2gbrp12le_full_X_4_512_c: 12257.1
yuv2gbrp12le_full_X_4_512_sse2: 9107.6
yuv2gbrp12le_full_X_4_512_sse4: 5142.6
yuv2gbrp12le_full_X_4_512_avx2: 2837.6
yuv2gbrap12be_full_X_4_512_c: 18511.1
yuv2gbrap12be_full_X_4_512_sse2: 12156.6
yuv2gbrap12be_full_X_4_512_sse4: 6251.1
yuv2gbrap12be_full_X_4_512_avx2: 3444.6
yuv2gbrap12le_full_X_4_512_c: 16687.1
yuv2gbrap12le_full_X_4_512_sse2: 11785.1
yuv2gbrap12le_full_X_4_512_sse4: 6243.6
yuv2gbrap12le_full_X_4_512_avx2: 3446.1
yuv2gbrp14be_full_X_4_512_c: 13690.6
yuv2gbrp14be_full_X_4_512_sse2: 9120.6
yuv2gbrp14be_full_X_4_512_sse4: 5138.1
yuv2gbrp14be_full_X_4_512_avx2: 2843.1
yuv2gbrp14le_full_X_4_512_c: 14995.6
yuv2gbrp14le_full_X_4_512_sse2: 9119.1
yuv2gbrp14le_full_X_4_512_sse4: 5126.1
yuv2gbrp14le_full_X_4_512_avx2: 2843.1
yuv2gbrp16be_full_X_4_512_c: 12367.1
yuv2gbrp16be_full_X_4_512_sse2: 8233.6
yuv2gbrp16be_full_X_4_512_sse4: 4820.1
yuv2gbrp16be_full_X_4_512_avx2: 2666.6
yuv2gbrp16le_full_X_4_512_c: 10904.1
yuv2gbrp16le_full_X_4_512_sse2: 8214.1
yuv2gbrp16le_full_X_4_512_sse4: 4824.1
yuv2gbrp16le_full_X_4_512_avx2: 2629.1
yuv2gbrap16be_full_X_4_512_c: 26569.6
yuv2gbrap16be_full_X_4_512_sse2: 10884.1
yuv2gbrap16be_full_X_4_512_sse4: 5488.1
yuv2gbrap16be_full_X_4_512_avx2: 3272.1
yuv2gbrap16le_full_X_4_512_c: 14010.1
yuv2gbrap16le_full_X_4_512_sse2: 10562.1
yuv2gbrap16le_full_X_4_512_sse4: 5463.6
yuv2gbrap16le_full_X_4_512_avx2: 3255.1
yuv2gbrpf32be_full_X_4_512_c: 14524.1
yuv2gbrpf32be_full_X_4_512_sse2: 8552.6
yuv2gbrpf32be_full_X_4_512_sse4: 4636.1
yuv2gbrpf32be_full_X_4_512_avx2: 2474.6
yuv2gbrpf32le_full_X_4_512_c: 13060.6
yuv2gbrpf32le_full_X_4_512_sse2: 9682.6
yuv2gbrpf32le_full_X_4_512_sse4: 4298.1
yuv2gbrpf32le_full_X_4_512_avx2: 2453.1
yuv2gbrapf32be_full_X_4_512_c: 18629.6
yuv2gbrapf32be_full_X_4_512_sse2: 11363.1
yuv2gbrapf32be_full_X_4_512_sse4: 15201.6
yuv2gbrapf32be_full_X_4_512_avx2: 3727.1
yuv2gbrapf32le_full_X_4_512_c: 16677.6
yuv2gbrapf32le_full_X_4_512_sse2: 10221.6
yuv2gbrapf32le_full_X_4_512_sse4: 5693.6
yuv2gbrapf32le_full_X_4_512_avx2: 3656.6

Reviewed-by: Paul B Mahol <onemda@gmail.com>
Signed-off-by: James Almer <jamrial@gmail.com>

5 files changed, 202 insertions, 1 deletions

diff --git a/tests/checkasm/Makefile b/tests/checkasm/Makefile
index 4ef5fa87da..f768b1144e 100644
--- a/tests/checkasm/Makefile
+++ b/tests/checkasm/Makefile
@@ -45,7 +45,7 @@ AVFILTEROBJS-$(CONFIG_NLMEANS_FILTER)    += vf_nlmeans.o
 CHECKASMOBJS-$(CONFIG_AVFILTER) += $(AVFILTEROBJS-yes)
 
 # swscale tests
-SWSCALEOBJS                             += sw_rgb.o sw_scale.o
+SWSCALEOBJS                             += sw_gbrp.o sw_rgb.o sw_scale.o
 
 CHECKASMOBJS-$(CONFIG_SWSCALE)  += $(SWSCALEOBJS)
 
diff --git a/tests/checkasm/checkasm.c b/tests/checkasm/checkasm.c
index 90d080de02..f74125e810 100644
--- a/tests/checkasm/checkasm.c
+++ b/tests/checkasm/checkasm.c
@@ -183,6 +183,7 @@ static const struct {
     #endif
 #endif
 #if CONFIG_SWSCALE
+    { "sw_gbrp", checkasm_check_sw_gbrp },
     { "sw_rgb", checkasm_check_sw_rgb },
     { "sw_scale", checkasm_check_sw_scale },
 #endif
diff --git a/tests/checkasm/checkasm.h b/tests/checkasm/checkasm.h
index 68b0697d3e..c3192d8c23 100644
--- a/tests/checkasm/checkasm.h
+++ b/tests/checkasm/checkasm.h
@@ -72,6 +72,7 @@ void checkasm_check_opusdsp(void);
 void checkasm_check_pixblockdsp(void);
 void checkasm_check_sbrdsp(void);
 void checkasm_check_synth_filter(void);
+void checkasm_check_sw_gbrp(void);
 void checkasm_check_sw_rgb(void);
 void checkasm_check_sw_scale(void);
 void checkasm_check_utvideodsp(void);
diff --git a/tests/checkasm/sw_gbrp.c b/tests/checkasm/sw_gbrp.c
new file mode 100644
index 0000000000..61b2b9d6a1
--- /dev/null
+++ b/tests/checkasm/sw_gbrp.c
@@ -0,0 +1,198 @@
+/*
+ *
+ * This file is part of FFmpeg.
+ *
+ * FFmpeg is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 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 General Public License for more details.
+ *
+ * You should have received a copy of the GNU 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 <string.h>
+
+#include "libavutil/common.h"
+#include "libavutil/intreadwrite.h"
+#include "libavutil/mem_internal.h"
+#include "libavutil/pixdesc.h"
+
+#include "libswscale/swscale.h"
+#include "libswscale/swscale_internal.h"
+
+#include "checkasm.h"
+
+#define randomize_buffers(buf, size)      \
+    do {                                  \
+        int j;                            \
+        for (j = 0; j < size; j+=4)       \
+            AV_WN32(buf + j, rnd());      \
+    } while (0)
+
+static const int planar_fmts[] = {
+    AV_PIX_FMT_GBRP,
+    AV_PIX_FMT_GBRP9BE,
+    AV_PIX_FMT_GBRP9LE,
+    AV_PIX_FMT_GBRP10BE,
+    AV_PIX_FMT_GBRP10LE,
+    AV_PIX_FMT_GBRP12BE,
+    AV_PIX_FMT_GBRP12LE,
+    AV_PIX_FMT_GBRP14BE,
+    AV_PIX_FMT_GBRP14LE,
+    AV_PIX_FMT_GBRAP,
+    AV_PIX_FMT_GBRAP10BE,
+    AV_PIX_FMT_GBRAP10LE,
+    AV_PIX_FMT_GBRAP12BE,
+    AV_PIX_FMT_GBRAP12LE,
+    AV_PIX_FMT_GBRP16BE,
+    AV_PIX_FMT_GBRP16LE,
+    AV_PIX_FMT_GBRAP16BE,
+    AV_PIX_FMT_GBRAP16LE,
+    AV_PIX_FMT_GBRPF32BE,
+    AV_PIX_FMT_GBRPF32LE,
+    AV_PIX_FMT_GBRAPF32BE,
+    AV_PIX_FMT_GBRAPF32LE
+};
+
+static void check_output_yuv2gbrp(void)
+{
+    struct SwsContext *ctx;
+    const AVPixFmtDescriptor *desc;
+    int fmi, fsi, isi, i;
+    int dstW, byte_size, luma_filter_size, chr_filter_size;
+#define LARGEST_FILTER 16
+#define FILTER_SIZES 4
+    static const int filter_sizes[] = {1, 4, 8, 16};
+#define LARGEST_INPUT_SIZE 512
+#define INPUT_SIZES 6
+    static const int input_sizes[] = {8, 24, 128, 144, 256, 512};
+    uint8_t *dst0[4];
+    uint8_t *dst1[4];
+
+    declare_func(void, void *c, const int16_t *lumFilter,
+                       const int16_t **lumSrcx, int lumFilterSize,
+                       const int16_t *chrFilter, const int16_t **chrUSrcx,
+                       const int16_t **chrVSrcx, int chrFilterSize,
+                       const int16_t **alpSrcx, uint8_t **dest,
+                       int dstW, int y);
+
+    const int16_t *luma[LARGEST_FILTER];
+    const int16_t *chru[LARGEST_FILTER];
+    const int16_t *chrv[LARGEST_FILTER];
+    const int16_t *alpha[LARGEST_FILTER];
+
+    LOCAL_ALIGNED_8(int16_t, luma_filter, [LARGEST_FILTER]);
+    LOCAL_ALIGNED_8(int16_t, chr_filter, [LARGEST_FILTER]);
+
+    LOCAL_ALIGNED_8(int32_t, src_y, [LARGEST_FILTER * LARGEST_INPUT_SIZE]);
+    LOCAL_ALIGNED_8(int32_t, src_u, [LARGEST_FILTER * LARGEST_INPUT_SIZE]);
+    LOCAL_ALIGNED_8(int32_t, src_v, [LARGEST_FILTER * LARGEST_INPUT_SIZE]);
+    LOCAL_ALIGNED_8(int32_t, src_a, [LARGEST_FILTER * LARGEST_INPUT_SIZE]);
+
+    LOCAL_ALIGNED_8(uint8_t, dst0_r, [LARGEST_INPUT_SIZE * sizeof(int32_t)]);
+    LOCAL_ALIGNED_8(uint8_t, dst0_g, [LARGEST_INPUT_SIZE * sizeof(int32_t)]);
+    LOCAL_ALIGNED_8(uint8_t, dst0_b, [LARGEST_INPUT_SIZE * sizeof(int32_t)]);
+    LOCAL_ALIGNED_8(uint8_t, dst0_a, [LARGEST_INPUT_SIZE * sizeof(int32_t)]);
+
+    LOCAL_ALIGNED_8(uint8_t, dst1_r, [LARGEST_INPUT_SIZE * sizeof(int32_t)]);
+    LOCAL_ALIGNED_8(uint8_t, dst1_g, [LARGEST_INPUT_SIZE * sizeof(int32_t)]);
+    LOCAL_ALIGNED_8(uint8_t, dst1_b, [LARGEST_INPUT_SIZE * sizeof(int32_t)]);
+    LOCAL_ALIGNED_8(uint8_t, dst1_a, [LARGEST_INPUT_SIZE * sizeof(int32_t)]);
+
+    randomize_buffers((uint8_t*)src_y, LARGEST_FILTER * LARGEST_INPUT_SIZE * sizeof(int32_t));
+    randomize_buffers((uint8_t*)src_u, LARGEST_FILTER * LARGEST_INPUT_SIZE * sizeof(int32_t));
+    randomize_buffers((uint8_t*)src_v, LARGEST_FILTER * LARGEST_INPUT_SIZE * sizeof(int32_t));
+    randomize_buffers((uint8_t*)src_a, LARGEST_FILTER * LARGEST_INPUT_SIZE * sizeof(int32_t));
+    randomize_buffers((uint8_t*)luma_filter, LARGEST_FILTER * sizeof(int16_t));
+    randomize_buffers((uint8_t*)chr_filter, LARGEST_FILTER * sizeof(int16_t));
+
+    dst0[0] = (uint8_t*)dst0_g;
+    dst0[1] = (uint8_t*)dst0_b;
+    dst0[2] = (uint8_t*)dst0_r;
+    dst0[3] = (uint8_t*)dst0_a;
+
+    dst1[0] = (uint8_t*)dst1_g;
+    dst1[1] = (uint8_t*)dst1_b;
+    dst1[2] = (uint8_t*)dst1_r;
+    dst1[3] = (uint8_t*)dst1_a;
+
+    for (i = 0; i < LARGEST_FILTER; i++) {
+        luma[i] =  (int16_t *)(src_y + i*LARGEST_INPUT_SIZE);
+        chru[i] =  (int16_t *)(src_u + i*LARGEST_INPUT_SIZE);
+        chrv[i] =  (int16_t *)(src_v + i*LARGEST_INPUT_SIZE);
+        alpha[i] = (int16_t *)(src_a + i*LARGEST_INPUT_SIZE);
+    }
+
+    ctx = sws_alloc_context();
+    if (sws_init_context(ctx, NULL, NULL) < 0)
+        fail();
+
+    ctx->flags |= SWS_FULL_CHR_H_INT;
+    ctx->yuv2rgb_y_offset  = rnd();
+    ctx->yuv2rgb_y_coeff   = rnd();
+    ctx->yuv2rgb_v2r_coeff = rnd();
+    ctx->yuv2rgb_v2g_coeff = rnd();
+    ctx->yuv2rgb_u2g_coeff = rnd();
+    ctx->yuv2rgb_u2b_coeff = rnd();
+
+    for (fmi = 0; fmi < FF_ARRAY_ELEMS(planar_fmts); fmi++) {
+        for (fsi = 0; fsi < FILTER_SIZES; fsi++) {
+            for (isi = 0; isi < INPUT_SIZES; isi++ ) {
+                desc = av_pix_fmt_desc_get(planar_fmts[fmi]);
+                ctx->dstFormat = planar_fmts[fmi];
+
+                dstW = input_sizes[isi];
+                luma_filter_size = filter_sizes[fsi];
+                chr_filter_size = filter_sizes[fsi];
+
+                if (desc->comp[0].depth > 16) {
+                    byte_size = 4;
+                } else if (desc->comp[0].depth > 8) {
+                    byte_size = 2;
+                } else {
+                    byte_size = 1;
+                }
+
+                ff_sws_init_scale(ctx);
+                if (check_func(ctx->yuv2anyX, "yuv2%s_full_X_%d_%d", desc->name, luma_filter_size, dstW)) {
+                    for (i = 0; i < 4; i ++) {
+                        memset(dst0[i], 0xFF, LARGEST_INPUT_SIZE * sizeof(int32_t));
+                        memset(dst1[i], 0xFF, LARGEST_INPUT_SIZE * sizeof(int32_t));
+                    }
+
+                    call_ref(ctx, luma_filter, luma, luma_filter_size,
+                             chr_filter, chru, chrv, chr_filter_size,
+                             alpha, dst0, dstW, 0);
+                    call_new(ctx, luma_filter, luma, luma_filter_size,
+                             chr_filter, chru, chrv, chr_filter_size,
+                             alpha, dst1, dstW, 0);
+
+                    if (memcmp(dst0[0], dst1[0], dstW * byte_size) ||
+                        memcmp(dst0[1], dst1[1], dstW * byte_size) ||
+                        memcmp(dst0[2], dst1[2], dstW * byte_size) ||
+                        memcmp(dst0[3], dst1[3], dstW * byte_size) )
+                        fail();
+
+                    bench_new(ctx, luma_filter, luma, luma_filter_size,
+                              chr_filter, chru, chrv, chr_filter_size,
+                              alpha, dst1, dstW, 0);
+                }
+            }
+        }
+    }
+    sws_freeContext(ctx);
+}
+
+
+void checkasm_check_sw_gbrp(void)
+{
+    check_output_yuv2gbrp();
+    report("output_yuv2gbrp");
+}
diff --git a/tests/fate/checkasm.mak b/tests/fate/checkasm.mak
index 6e7edbe655..6db8f09d12 100644
--- a/tests/fate/checkasm.mak
+++ b/tests/fate/checkasm.mak
@@ -26,6 +26,7 @@ FATE_CHECKASM = fate-checkasm-aacpsdsp                                  \
                 fate-checkasm-pixblockdsp                               \
                 fate-checkasm-sbrdsp                                    \
                 fate-checkasm-synth_filter                              \
+                fate-checkasm-sw_gbrp                                   \
                 fate-checkasm-sw_rgb                                    \
                 fate-checkasm-sw_scale                                  \
                 fate-checkasm-utvideodsp                                \