diff options
Diffstat (limited to 'vp8')
| -rw-r--r-- | vp8/common/postproc.c | 12 | ||||
| -rw-r--r-- | vp8/common/rtcd_defs.sh | 199 | ||||
| -rw-r--r-- | vp8/encoder/arm/arm_csystemdependent.c | 52 | ||||
| -rw-r--r-- | vp8/encoder/arm/variance_arm.c | 1 | ||||
| -rw-r--r-- | vp8/encoder/arm/variance_arm.h | 155 | ||||
| -rw-r--r-- | vp8/encoder/encodeframe.c | 2 | ||||
| -rw-r--r-- | vp8/encoder/encodeintra.c | 2 | ||||
| -rw-r--r-- | vp8/encoder/firstpass.c | 4 | ||||
| -rw-r--r-- | vp8/encoder/generic/csystemdependent.c | 51 | ||||
| -rw-r--r-- | vp8/encoder/onyx_if.c | 136 | ||||
| -rw-r--r-- | vp8/encoder/onyx_int.h | 1 | ||||
| -rw-r--r-- | vp8/encoder/pickinter.c | 18 | ||||
| -rw-r--r-- | vp8/encoder/picklpf.c | 25 | ||||
| -rw-r--r-- | vp8/encoder/rdopt.c | 14 | ||||
| -rw-r--r-- | vp8/encoder/ssim.c | 38 | ||||
| -rw-r--r-- | vp8/encoder/variance.h | 461 | ||||
| -rw-r--r-- | vp8/encoder/x86/variance_x86.h | 332 | ||||
| -rw-r--r-- | vp8/encoder/x86/x86_csystemdependent.c | 84 | ||||
| -rw-r--r-- | vp8/vp8cx.mk | 1 | ||||
| -rw-r--r-- | vp8/vp8cx_arm.mk | 1 |
20 files changed, 385 insertions, 1204 deletions
diff --git a/vp8/common/postproc.c b/vp8/common/postproc.c index 8e2f683ae..0dc14b8bd 100644 --- a/vp8/common/postproc.c +++ b/vp8/common/postproc.c @@ -728,18 +728,18 @@ static void multiframe_quality_enhance_block unsigned int act, sse, sad, thr; if (blksize == 16) { - act = (vp8_variance_var16x16(yd, yd_stride, VP8_ZEROS, 0, &sse)+128)>>8; - sad = (vp8_variance_sad16x16(y, y_stride, yd, yd_stride, 0)+128)>>8; + act = (vp8_variance16x16(yd, yd_stride, VP8_ZEROS, 0, &sse)+128)>>8; + sad = (vp8_sad16x16(y, y_stride, yd, yd_stride, 0)+128)>>8; } else if (blksize == 8) { - act = (vp8_variance_var8x8(yd, yd_stride, VP8_ZEROS, 0, &sse)+32)>>6; - sad = (vp8_variance_sad8x8(y, y_stride, yd, yd_stride, 0)+32)>>6; + act = (vp8_variance8x8(yd, yd_stride, VP8_ZEROS, 0, &sse)+32)>>6; + sad = (vp8_sad8x8(y, y_stride, yd, yd_stride, 0)+32)>>6; } else { - act = (vp8_variance_var4x4(yd, yd_stride, VP8_ZEROS, 0, &sse)+8)>>4; - sad = (vp8_variance_sad4x4(y, y_stride, yd, yd_stride, 0)+8)>>4; + act = (vp8_variance4x4(yd, yd_stride, VP8_ZEROS, 0, &sse)+8)>>4; + sad = (vp8_sad4x4(y, y_stride, yd, yd_stride, 0)+8)>>4; } /* thr = qdiff/8 + log2(act) + log4(qprev) */ thr = (qdiff>>3); diff --git a/vp8/common/rtcd_defs.sh b/vp8/common/rtcd_defs.sh index 62030f7c3..60af88801 100644 --- a/vp8/common/rtcd_defs.sh +++ b/vp8/common/rtcd_defs.sh @@ -195,3 +195,202 @@ vp8_bilinear_predict8x4_media=vp8_bilinear_predict8x4_armv6 prototype void vp8_bilinear_predict4x4 "unsigned char *src, int src_pitch, int xofst, int yofst, unsigned char *dst, int dst_pitch" specialize vp8_bilinear_predict4x4 mmx media neon vp8_bilinear_predict4x4_media=vp8_bilinear_predict4x4_armv6 + +# +# Encoder functions below this point. +# +if [ "$CONFIG_VP8_ENCODER" = "yes" ]; then + +# +# Whole-pixel Variance +# +prototype unsigned int vp8_variance4x4 "const unsigned char *src_ptr, int source_stride, const unsigned char *ref_ptr, int ref_stride, unsigned int *sse" +specialize vp8_variance4x4 mmx sse2 +vp8_variance4x4_sse2=vp8_variance4x4_wmt + +prototype unsigned int vp8_variance8x8 "const unsigned char *src_ptr, int source_stride, const unsigned char *ref_ptr, int ref_stride, unsigned int *sse" +specialize vp8_variance8x8 mmx sse2 media neon +vp8_variance8x8_sse2=vp8_variance8x8_wmt +vp8_variance8x8_media=vp8_variance8x8_armv6 + +prototype unsigned int vp8_variance8x16 "const unsigned char *src_ptr, int source_stride, const unsigned char *ref_ptr, int ref_stride, unsigned int *sse" +specialize vp8_variance8x16 mmx sse2 neon +vp8_variance8x16_sse2=vp8_variance8x16_wmt + +prototype unsigned int vp8_variance16x8 "const unsigned char *src_ptr, int source_stride, const unsigned char *ref_ptr, int ref_stride, unsigned int *sse" +specialize vp8_variance16x8 mmx sse2 neon +vp8_variance16x8_sse2=vp8_variance16x8_wmt + +prototype unsigned int vp8_variance16x16 "const unsigned char *src_ptr, int source_stride, const unsigned char *ref_ptr, int ref_stride, unsigned int *sse" +specialize vp8_variance16x16 mmx sse2 media neon +vp8_variance16x16_sse2=vp8_variance16x16_wmt +vp8_variance16x16_media=vp8_variance16x16_armv6 + +# +# Sub-pixel Variance +# +prototype unsigned int vp8_sub_pixel_variance4x4 "const unsigned char *src_ptr, int source_stride, int xoffset, int yoffset, const unsigned char *ref_ptr, int Refstride, unsigned int *sse" +specialize vp8_sub_pixel_variance4x4 mmx sse2 +vp8_sub_pixel_variance4x4_sse2=vp8_sub_pixel_variance4x4_wmt + +prototype unsigned int vp8_sub_pixel_variance8x8 "const unsigned char *src_ptr, int source_stride, int xoffset, int yoffset, const unsigned char *ref_ptr, int Refstride, unsigned int *sse" +specialize vp8_sub_pixel_variance8x8 mmx sse2 media neon +vp8_sub_pixel_variance8x8_sse2=vp8_sub_pixel_variance8x8_wmt +vp8_sub_pixel_variance8x8_media=vp8_sub_pixel_variance8x8_armv6 + +prototype unsigned int vp8_sub_pixel_variance8x16 "const unsigned char *src_ptr, int source_stride, int xoffset, int yoffset, const unsigned char *ref_ptr, int Refstride, unsigned int *sse" +specialize vp8_sub_pixel_variance8x16 mmx sse2 +vp8_sub_pixel_variance8x16_sse2=vp8_sub_pixel_variance8x16_wmt + +prototype unsigned int vp8_sub_pixel_variance16x8 "const unsigned char *src_ptr, int source_stride, int xoffset, int yoffset, const unsigned char *ref_ptr, int Refstride, unsigned int *sse" +specialize vp8_sub_pixel_variance16x8 mmx sse2 ssse3 +vp8_sub_pixel_variance16x8_sse2=vp8_sub_pixel_variance16x8_wmt + +prototype unsigned int vp8_sub_pixel_variance16x16 "const unsigned char *src_ptr, int source_stride, int xoffset, int yoffset, const unsigned char *ref_ptr, int Refstride, unsigned int *sse" +specialize vp8_sub_pixel_variance16x16 mmx sse2 ssse3 media neon +vp8_sub_pixel_variance16x16_sse2=vp8_sub_pixel_variance16x16_wmt +vp8_sub_pixel_variance16x16_media=vp8_sub_pixel_variance16x16_armv6 + +prototype unsigned int vp8_variance_halfpixvar16x16_h "const unsigned char *src_ptr, int source_stride, const unsigned char *ref_ptr, int ref_stride, unsigned int *sse" +specialize vp8_variance_halfpixvar16x16_h mmx sse2 media neon +vp8_variance_halfpixvar16x16_h_sse2=vp8_variance_halfpixvar16x16_h_wmt +vp8_variance_halfpixvar16x16_h_media=vp8_variance_halfpixvar16x16_h_armv6 + +prototype unsigned int vp8_variance_halfpixvar16x16_v "const unsigned char *src_ptr, int source_stride, const unsigned char *ref_ptr, int ref_stride, unsigned int *sse" +specialize vp8_variance_halfpixvar16x16_v mmx sse2 media neon +vp8_variance_halfpixvar16x16_v_sse2=vp8_variance_halfpixvar16x16_v_wmt +vp8_variance_halfpixvar16x16_v_media=vp8_variance_halfpixvar16x16_v_armv6 + +prototype unsigned int vp8_variance_halfpixvar16x16_hv "const unsigned char *src_ptr, int source_stride, const unsigned char *ref_ptr, int ref_stride, unsigned int *sse" +specialize vp8_variance_halfpixvar16x16_hv mmx sse2 media neon +vp8_variance_halfpixvar16x16_hv_sse2=vp8_variance_halfpixvar16x16_hv_wmt +vp8_variance_halfpixvar16x16_hv_media=vp8_variance_halfpixvar16x16_hv_armv6 + +# +# Sum of squares (vector) +# +prototype unsigned int vp8_get_mb_ss "const short *" +specialize vp8_get_mb_ss mmx sse2 + +# +# SSE (Sum Squared Error) +# +prototype unsigned int vp8_sub_pixel_mse16x16 "const unsigned char *src_ptr, int source_stride, int xoffset, int yoffset, const unsigned char *ref_ptr, int Refstride, unsigned int *sse" +specialize vp8_sub_pixel_mse16x16 mmx sse2 +vp8_sub_pixel_mse16x16_sse2=vp8_sub_pixel_mse16x16_wmt + +prototype unsigned int vp8_mse16x16 "const unsigned char *src_ptr, int source_stride, const unsigned char *ref_ptr, int ref_stride, unsigned int *sse" +specialize vp8_mse16x16 mmx sse2 media neon +vp8_mse16x16_sse2=vp8_mse16x16_wmt +vp8_mse16x16_media=vp8_mse16x16_armv6 + +prototype unsigned int vp8_get4x4sse_cs "const unsigned char *src_ptr, int source_stride, const unsigned char *ref_ptr, int ref_stride" +specialize vp8_get4x4sse_cs mmx neon + +# +# Single block SAD +# +prototype unsigned int vp8_sad4x4 "const unsigned char *src_ptr, int source_stride, const unsigned char *ref_ptr, int ref_stride, int max_sad" +specialize vp8_sad4x4 mmx sse2 neon +vp8_sad4x4_sse2=vp8_sad4x4_wmt + +prototype unsigned int vp8_sad8x8 "const unsigned char *src_ptr, int source_stride, const unsigned char *ref_ptr, int ref_stride, int max_sad" +specialize vp8_sad8x8 mmx sse2 neon +vp8_sad8x8_sse2=vp8_sad8x8_wmt + +prototype unsigned int vp8_sad8x16 "const unsigned char *src_ptr, int source_stride, const unsigned char *ref_ptr, int ref_stride, int max_sad" +specialize vp8_sad8x16 mmx sse2 neon +vp8_sad8x16_sse2=vp8_sad8x16_wmt + +prototype unsigned int vp8_sad16x8 "const unsigned char *src_ptr, int source_stride, const unsigned char *ref_ptr, int ref_stride, int max_sad" +specialize vp8_sad16x8 mmx sse2 neon +vp8_sad16x8_sse2=vp8_sad16x8_wmt + +prototype unsigned int vp8_sad16x16 "const unsigned char *src_ptr, int source_stride, const unsigned char *ref_ptr, int ref_stride, int max_sad" +specialize vp8_sad16x16 mmx sse2 sse3 media neon +vp8_sad16x16_sse2=vp8_sad16x16_wmt +vp8_sad16x16_media=vp8_sad16x16_armv6 + +# +# Multi-block SAD, comparing a reference to N blocks 1 pixel apart horizontally +# +prototype void vp8_sad4x4x3 "const unsigned char *src_ptr, int source_stride, const unsigned char *ref_ptr, int ref_stride, unsigned int *sad_array" +specialize vp8_sad4x4x3 sse3 + +prototype void vp8_sad8x8x3 "const unsigned char *src_ptr, int source_stride, const unsigned char *ref_ptr, int ref_stride, unsigned int *sad_array" +specialize vp8_sad8x8x3 sse3 + +prototype void vp8_sad8x16x3 "const unsigned char *src_ptr, int source_stride, const unsigned char *ref_ptr, int ref_stride, unsigned int *sad_array" +specialize vp8_sad8x16x3 sse3 + +prototype void vp8_sad16x8x3 "const unsigned char *src_ptr, int source_stride, const unsigned char *ref_ptr, int ref_stride, unsigned int *sad_array" +specialize vp8_sad16x8x3 sse3 ssse3 + +prototype void vp8_sad16x16x3 "const unsigned char *src_ptr, int source_stride, const unsigned char *ref_ptr, int ref_stride, unsigned int *sad_array" +specialize vp8_sad16x16x3 sse3 ssse3 + +# Note the only difference in the following prototypes is that they return into +# an array of short +prototype void vp8_sad4x4x8 "const unsigned char *src_ptr, int source_stride, const unsigned char *ref_ptr, int ref_stride, unsigned short *sad_array" +specialize vp8_sad4x4x8 sse4_1 +vp8_sad4x4x8_sse4_1=vp8_sad4x4x8_sse4 + +prototype void vp8_sad8x8x8 "const unsigned char *src_ptr, int source_stride, const unsigned char *ref_ptr, int ref_stride, unsigned short *sad_array" +specialize vp8_sad8x8x8 sse4_1 +vp8_sad8x8x8_sse4_1=vp8_sad8x8x8_sse4 + +prototype void vp8_sad8x16x8 "const unsigned char *src_ptr, int source_stride, const unsigned char *ref_ptr, int ref_stride, unsigned short *sad_array" +specialize vp8_sad8x16x8 sse4_1 +vp8_sad8x16x8_sse4_1=vp8_sad8x16x8_sse4 + +prototype void vp8_sad16x8x8 "const unsigned char *src_ptr, int source_stride, const unsigned char *ref_ptr, int ref_stride, unsigned short *sad_array" +specialize vp8_sad16x8x8 sse4_1 +vp8_sad16x8x8_sse4_1=vp8_sad16x8x8_sse4 + +prototype void vp8_sad16x16x8 "const unsigned char *src_ptr, int source_stride, const unsigned char *ref_ptr, int ref_stride, unsigned short *sad_array" +specialize vp8_sad16x16x8 sse4_1 +vp8_sad16x16x8_sse4_1=vp8_sad16x16x8_sse4 + +# +# Multi-block SAD, comparing a reference to N independent blocks +# +prototype void vp8_sad4x4x4d "const unsigned char *src_ptr, int source_stride, unsigned char *ref_ptr[4], int ref_stride, unsigned int *sad_array" +specialize vp8_sad4x4x4d sse3 + +prototype void vp8_sad8x8x4d "const unsigned char *src_ptr, int source_stride, unsigned char *ref_ptr[4], int ref_stride, unsigned int *sad_array" +specialize vp8_sad8x8x4d sse3 + +prototype void vp8_sad8x16x4d "const unsigned char *src_ptr, int source_stride, unsigned char *ref_ptr[4], int ref_stride, unsigned int *sad_array" +specialize vp8_sad8x16x4d sse3 + +prototype void vp8_sad16x8x4d "const unsigned char *src_ptr, int source_stride, unsigned char *ref_ptr[4], int ref_stride, unsigned int *sad_array" +specialize vp8_sad16x8x4d sse3 + +prototype void vp8_sad16x16x4d "const unsigned char *src_ptr, int source_stride, unsigned char *ref_ptr[4], int ref_stride, unsigned int *sad_array" +specialize vp8_sad16x16x4d sse3 + +# +# Block copy +# +case $arch in + x86*) + prototype void vp8_copy32xn "const unsigned char *src_ptr, int source_stride, const unsigned char *ref_ptr, int ref_stride, int n" + specialize vp8_copy32xn sse2 sse3 + ;; +esac + +# +# Structured Similarity (SSIM) +# +if [ "$CONFIG_INTERNAL_STATS" = "yes" ]; then + [ $arch = "x86_64" ] && sse2_on_x86_64=sse2 + + prototype void vp8_ssim_parms_8x8 "unsigned char *s, int sp, unsigned char *r, int rp, unsigned long *sum_s, unsigned long *sum_r, unsigned long *sum_sq_s, unsigned long *sum_sq_r, unsigned long *sum_sxr" + specialize vp8_ssim_parms_8x8 $sse2_on_x86_64 + + prototype void vp8_ssim_parms_16x16 "unsigned char *s, int sp, unsigned char *r, int rp, unsigned long *sum_s, unsigned long *sum_r, unsigned long *sum_sq_s, unsigned long *sum_sq_r, unsigned long *sum_sxr" + specialize vp8_ssim_parms_16x16 $sse2_on_x86_64 +fi + +# End of encoder only functions +fi diff --git a/vp8/encoder/arm/arm_csystemdependent.c b/vp8/encoder/arm/arm_csystemdependent.c index c6f46f424..60c3da948 100644 --- a/vp8/encoder/arm/arm_csystemdependent.c +++ b/vp8/encoder/arm/arm_csystemdependent.c @@ -32,32 +32,6 @@ void vp8_arch_arm_encoder_init(VP8_COMP *cpi) #if HAVE_MEDIA if (flags & HAS_MEDIA) { - cpi->rtcd.variance.sad16x16 = vp8_sad16x16_armv6; - /*cpi->rtcd.variance.sad16x8 = vp8_sad16x8_c; - cpi->rtcd.variance.sad8x16 = vp8_sad8x16_c; - cpi->rtcd.variance.sad8x8 = vp8_sad8x8_c; - cpi->rtcd.variance.sad4x4 = vp8_sad4x4_c;*/ - - /*cpi->rtcd.variance.var4x4 = vp8_variance4x4_c;*/ - cpi->rtcd.variance.var8x8 = vp8_variance8x8_armv6; - /*cpi->rtcd.variance.var8x16 = vp8_variance8x16_c; - cpi->rtcd.variance.var16x8 = vp8_variance16x8_c;*/ - cpi->rtcd.variance.var16x16 = vp8_variance16x16_armv6; - - /*cpi->rtcd.variance.subpixvar4x4 = vp8_sub_pixel_variance4x4_c;*/ - cpi->rtcd.variance.subpixvar8x8 = vp8_sub_pixel_variance8x8_armv6; - /*cpi->rtcd.variance.subpixvar8x16 = vp8_sub_pixel_variance8x16_c; - cpi->rtcd.variance.subpixvar16x8 = vp8_sub_pixel_variance16x8_c;*/ - cpi->rtcd.variance.subpixvar16x16 = vp8_sub_pixel_variance16x16_armv6; - cpi->rtcd.variance.halfpixvar16x16_h = vp8_variance_halfpixvar16x16_h_armv6; - cpi->rtcd.variance.halfpixvar16x16_v = vp8_variance_halfpixvar16x16_v_armv6; - cpi->rtcd.variance.halfpixvar16x16_hv = vp8_variance_halfpixvar16x16_hv_armv6; - - cpi->rtcd.variance.mse16x16 = vp8_mse16x16_armv6; - /*cpi->rtcd.variance.getmbss = vp8_get_mb_ss_c;*/ - - /*cpi->rtcd.variance.get4x4sse_cs = vp8_get4x4sse_cs_c;*/ - cpi->rtcd.fdct.short4x4 = vp8_short_fdct4x4_armv6; cpi->rtcd.fdct.short8x4 = vp8_short_fdct8x4_armv6; cpi->rtcd.fdct.fast4x4 = vp8_short_fdct4x4_armv6; @@ -79,32 +53,6 @@ void vp8_arch_arm_encoder_init(VP8_COMP *cpi) #if HAVE_NEON if (flags & HAS_NEON) { - cpi->rtcd.variance.sad16x16 = vp8_sad16x16_neon; - cpi->rtcd.variance.sad16x8 = vp8_sad16x8_neon; - cpi->rtcd.variance.sad8x16 = vp8_sad8x16_neon; - cpi->rtcd.variance.sad8x8 = vp8_sad8x8_neon; - cpi->rtcd.variance.sad4x4 = vp8_sad4x4_neon; - - /*cpi->rtcd.variance.var4x4 = vp8_variance4x4_c;*/ - cpi->rtcd.variance.var8x8 = vp8_variance8x8_neon; - cpi->rtcd.variance.var8x16 = vp8_variance8x16_neon; - cpi->rtcd.variance.var16x8 = vp8_variance16x8_neon; - cpi->rtcd.variance.var16x16 = vp8_variance16x16_neon; - - /*cpi->rtcd.variance.subpixvar4x4 = vp8_sub_pixel_variance4x4_c;*/ - cpi->rtcd.variance.subpixvar8x8 = vp8_sub_pixel_variance8x8_neon; - /*cpi->rtcd.variance.subpixvar8x16 = vp8_sub_pixel_variance8x16_c; - cpi->rtcd.variance.subpixvar16x8 = vp8_sub_pixel_variance16x8_c;*/ - cpi->rtcd.variance.subpixvar16x16 = vp8_sub_pixel_variance16x16_neon; - cpi->rtcd.variance.halfpixvar16x16_h = vp8_variance_halfpixvar16x16_h_neon; - cpi->rtcd.variance.halfpixvar16x16_v = vp8_variance_halfpixvar16x16_v_neon; - cpi->rtcd.variance.halfpixvar16x16_hv = vp8_variance_halfpixvar16x16_hv_neon; - - cpi->rtcd.variance.mse16x16 = vp8_mse16x16_neon; - /*cpi->rtcd.variance.getmbss = vp8_get_mb_ss_c;*/ - - cpi->rtcd.variance.get4x4sse_cs = vp8_get4x4sse_cs_neon; - cpi->rtcd.fdct.short4x4 = vp8_short_fdct4x4_neon; cpi->rtcd.fdct.short8x4 = vp8_short_fdct8x4_neon; cpi->rtcd.fdct.fast4x4 = vp8_short_fdct4x4_neon; diff --git a/vp8/encoder/arm/variance_arm.c b/vp8/encoder/arm/variance_arm.c index 5faa1048e..052a2578a 100644 --- a/vp8/encoder/arm/variance_arm.c +++ b/vp8/encoder/arm/variance_arm.c @@ -9,6 +9,7 @@ */ #include "vpx_config.h" +#include "vpx_rtcd.h" #include "vp8/encoder/variance.h" #include "vp8/common/filter.h" diff --git a/vp8/encoder/arm/variance_arm.h b/vp8/encoder/arm/variance_arm.h deleted file mode 100644 index 99eb26b28..000000000 --- a/vp8/encoder/arm/variance_arm.h +++ /dev/null @@ -1,155 +0,0 @@ -/* - * Copyright (c) 2010 The WebM project authors. All Rights Reserved. - * - * Use of this source code is governed by a BSD-style license - * that can be found in the LICENSE file in the root of the source - * tree. An additional intellectual property rights grant can be found - * in the file PATENTS. All contributing project authors may - * be found in the AUTHORS file in the root of the source tree. - */ - - -#ifndef VARIANCE_ARM_H -#define VARIANCE_ARM_H - -#if HAVE_MEDIA - -extern prototype_sad(vp8_sad16x16_armv6); -extern prototype_variance(vp8_variance16x16_armv6); -extern prototype_variance(vp8_variance8x8_armv6); -extern prototype_subpixvariance(vp8_sub_pixel_variance16x16_armv6); -extern prototype_subpixvariance(vp8_sub_pixel_variance8x8_armv6); -extern prototype_variance(vp8_variance_halfpixvar16x16_h_armv6); -extern prototype_variance(vp8_variance_halfpixvar16x16_v_armv6); -extern prototype_variance(vp8_variance_halfpixvar16x16_hv_armv6); -extern prototype_variance(vp8_mse16x16_armv6); - -#if !CONFIG_RUNTIME_CPU_DETECT - -#undef vp8_variance_sad16x16 -#define vp8_variance_sad16x16 vp8_sad16x16_armv6 - -#undef vp8_variance_subpixvar16x16 -#define vp8_variance_subpixvar16x16 vp8_sub_pixel_variance16x16_armv6 - -#undef vp8_variance_subpixvar8x8 -#define vp8_variance_subpixvar8x8 vp8_sub_pixel_variance8x8_armv6 - -#undef vp8_variance_var16x16 -#define vp8_variance_var16x16 vp8_variance16x16_armv6 - -#undef vp8_variance_mse16x16 -#define vp8_variance_mse16x16 vp8_mse16x16_armv6 - -#undef vp8_variance_var8x8 -#define vp8_variance_var8x8 vp8_variance8x8_armv6 - -#undef vp8_variance_halfpixvar16x16_h -#define vp8_variance_halfpixvar16x16_h vp8_variance_halfpixvar16x16_h_armv6 - -#undef vp8_variance_halfpixvar16x16_v -#define vp8_variance_halfpixvar16x16_v vp8_variance_halfpixvar16x16_v_armv6 - -#undef vp8_variance_halfpixvar16x16_hv -#define vp8_variance_halfpixvar16x16_hv vp8_variance_halfpixvar16x16_hv_armv6 - -#endif /* !CONFIG_RUNTIME_CPU_DETECT */ - -#endif /* HAVE_MEDIA */ - - -#if HAVE_NEON -extern prototype_sad(vp8_sad4x4_neon); -extern prototype_sad(vp8_sad8x8_neon); -extern prototype_sad(vp8_sad8x16_neon); -extern prototype_sad(vp8_sad16x8_neon); -extern prototype_sad(vp8_sad16x16_neon); - -//extern prototype_variance(vp8_variance4x4_c); -extern prototype_variance(vp8_variance8x8_neon); -extern prototype_variance(vp8_variance8x16_neon); -extern prototype_variance(vp8_variance16x8_neon); -extern prototype_variance(vp8_variance16x16_neon); - -//extern prototype_subpixvariance(vp8_sub_pixel_variance4x4_c); -extern prototype_subpixvariance(vp8_sub_pixel_variance8x8_neon); -//extern prototype_subpixvariance(vp8_sub_pixel_variance8x16_c); -//extern prototype_subpixvariance(vp8_sub_pixel_variance16x8_c); -extern prototype_subpixvariance(vp8_sub_pixel_variance16x16_neon); -extern prototype_subpixvariance(vp8_sub_pixel_variance16x16_neon_func); -extern prototype_variance(vp8_variance_halfpixvar16x16_h_neon); -extern prototype_variance(vp8_variance_halfpixvar16x16_v_neon); -extern prototype_variance(vp8_variance_halfpixvar16x16_hv_neon); - -//extern prototype_getmbss(vp8_get_mb_ss_c); -extern prototype_variance(vp8_mse16x16_neon); -extern prototype_get16x16prederror(vp8_get4x4sse_cs_neon); - -#if !CONFIG_RUNTIME_CPU_DETECT -#undef vp8_variance_sad4x4 -#define vp8_variance_sad4x4 vp8_sad4x4_neon - -#undef vp8_variance_sad8x8 -#define vp8_variance_sad8x8 vp8_sad8x8_neon - -#undef vp8_variance_sad8x16 -#define vp8_variance_sad8x16 vp8_sad8x16_neon - -#undef vp8_variance_sad16x8 -#define vp8_variance_sad16x8 vp8_sad16x8_neon - -#undef vp8_variance_sad16x16 -#define vp8_variance_sad16x16 vp8_sad16x16_neon - -//#undef vp8_variance_var4x4 -//#define vp8_variance_var4x4 vp8_variance4x4_c - -#undef vp8_variance_var8x8 -#define vp8_variance_var8x8 vp8_variance8x8_neon - -#undef vp8_variance_var8x16 -#define vp8_variance_var8x16 vp8_variance8x16_neon - -#undef vp8_variance_var16x8 -#define vp8_variance_var16x8 vp8_variance16x8_neon - -#undef vp8_variance_var16x16 -#define vp8_variance_var16x16 vp8_variance16x16_neon - -//#undef vp8_variance_subpixvar4x4 -//#define vp8_variance_subpixvar4x4 vp8_sub_pixel_variance4x4_c - -#undef vp8_variance_subpixvar8x8 -#define vp8_variance_subpixvar8x8 vp8_sub_pixel_variance8x8_neon - -//#undef vp8_variance_subpixvar8x16 -//#define vp8_variance_subpixvar8x16 vp8_sub_pixel_variance8x16_c - -//#undef vp8_variance_subpixvar16x8 -//#define vp8_variance_subpixvar16x8 vp8_sub_pixel_variance16x8_c - -#undef vp8_variance_subpixvar16x16 -#define vp8_variance_subpixvar16x16 vp8_sub_pixel_variance16x16_neon - -#undef vp8_variance_halfpixvar16x16_h -#define vp8_variance_halfpixvar16x16_h vp8_variance_halfpixvar16x16_h_neon - -#undef vp8_variance_halfpixvar16x16_v -#define vp8_variance_halfpixvar16x16_v vp8_variance_halfpixvar16x16_v_neon - -#undef vp8_variance_halfpixvar16x16_hv -#define vp8_variance_halfpixvar16x16_hv vp8_variance_halfpixvar16x16_hv_neon - -//#undef vp8_variance_getmbss -//#define vp8_variance_getmbss vp8_get_mb_ss_c - -#undef vp8_variance_mse16x16 -#define vp8_variance_mse16x16 vp8_mse16x16_neon - -#undef vp8_variance_get4x4sse_cs -#define vp8_variance_get4x4sse_cs vp8_get4x4sse_cs_neon -#endif /* !CONFIG_RUNTIME_CPU_DETECT */ - -#endif /* HAVE_NEON */ - -#endif diff --git a/vp8/encoder/encodeframe.c b/vp8/encoder/encodeframe.c index 9eed6ccaf..41da9abf5 100644 --- a/vp8/encoder/encodeframe.c +++ b/vp8/encoder/encodeframe.c @@ -96,7 +96,7 @@ static unsigned int tt_activity_measure( VP8_COMP *cpi, MACROBLOCK *x ) * lambda using a non-linear combination (e.g., the smallest, or second * smallest, etc.). */ - act = VARIANCE_INVOKE(&cpi->rtcd.variance, var16x16)(x->src.y_buffer, + act = vp8_variance16x16(x->src.y_buffer, x->src.y_stride, VP8_VAR_OFFS, 0, &sse); act = act<<4; diff --git a/vp8/encoder/encodeintra.c b/vp8/encoder/encodeintra.c index 14c93eb46..4ae4e46df 100644 --- a/vp8/encoder/encodeintra.c +++ b/vp8/encoder/encodeintra.c @@ -53,7 +53,7 @@ int vp8_encode_intra(VP8_COMP *cpi, MACROBLOCK *x, int use_dc_pred) } } - intra_pred_var = VARIANCE_INVOKE(&cpi->rtcd.variance, getmbss)(x->src_diff); + intra_pred_var = vp8_get_mb_ss(x->src_diff); return intra_pred_var; } diff --git a/vp8/encoder/firstpass.c b/vp8/encoder/firstpass.c index 03d639765..09f184c6b 100644 --- a/vp8/encoder/firstpass.c +++ b/vp8/encoder/firstpass.c @@ -409,7 +409,7 @@ static void zz_motion_search( VP8_COMP *cpi, MACROBLOCK * x, YV12_BUFFER_CONFIG ref_ptr = (unsigned char *)(*(d->base_pre) + d->pre ); - VARIANCE_INVOKE(IF_RTCD(&cpi->rtcd.variance), mse16x16) ( src_ptr, src_stride, ref_ptr, ref_stride, (unsigned int *)(best_motion_err)); + vp8_mse16x16 ( src_ptr, src_stride, ref_ptr, ref_stride, (unsigned int *)(best_motion_err)); } static void first_pass_motion_search(VP8_COMP *cpi, MACROBLOCK *x, @@ -433,7 +433,7 @@ static void first_pass_motion_search(VP8_COMP *cpi, MACROBLOCK *x, int new_mv_mode_penalty = 256; // override the default variance function to use MSE - v_fn_ptr.vf = VARIANCE_INVOKE(IF_RTCD(&cpi->rtcd.variance), mse16x16); + v_fn_ptr.vf = vp8_mse16x16; // Set up pointers for this macro block recon buffer xd->pre.y_buffer = recon_buffer->y_buffer + recon_yoffset; diff --git a/vp8/encoder/generic/csystemdependent.c b/vp8/encoder/generic/csystemdependent.c index 89e8dc2bc..59324d0df 100644 --- a/vp8/encoder/generic/csystemdependent.c +++ b/vp8/encoder/generic/csystemdependent.c @@ -25,53 +25,6 @@ extern void vp8_yv12_copy_partial_frame(YV12_BUFFER_CONFIG *src_ybc, void vp8_cmachine_specific_config(VP8_COMP *cpi) { #if CONFIG_RUNTIME_CPU_DETECT - cpi->rtcd.variance.sad16x16 = vp8_sad16x16_c; - cpi->rtcd.variance.sad16x8 = vp8_sad16x8_c; - cpi->rtcd.variance.sad8x16 = vp8_sad8x16_c; - cpi->rtcd.variance.sad8x8 = vp8_sad8x8_c; - cpi->rtcd.variance.sad4x4 = vp8_sad4x4_c; - - cpi->rtcd.variance.sad16x16x3 = vp8_sad16x16x3_c; - cpi->rtcd.variance.sad16x8x3 = vp8_sad16x8x3_c; - cpi->rtcd.variance.sad8x16x3 = vp8_sad8x16x3_c; - cpi->rtcd.variance.sad8x8x3 = vp8_sad8x8x3_c; - cpi->rtcd.variance.sad4x4x3 = vp8_sad4x4x3_c; - - cpi->rtcd.variance.sad16x16x8 = vp8_sad16x16x8_c; - cpi->rtcd.variance.sad16x8x8 = vp8_sad16x8x8_c; - cpi->rtcd.variance.sad8x16x8 = vp8_sad8x16x8_c; - cpi->rtcd.variance.sad8x8x8 = vp8_sad8x8x8_c; - cpi->rtcd.variance.sad4x4x8 = vp8_sad4x4x8_c; - - cpi->rtcd.variance.sad16x16x4d = vp8_sad16x16x4d_c; - cpi->rtcd.variance.sad16x8x4d = vp8_sad16x8x4d_c; - cpi->rtcd.variance.sad8x16x4d = vp8_sad8x16x4d_c; - cpi->rtcd.variance.sad8x8x4d = vp8_sad8x8x4d_c; - cpi->rtcd.variance.sad4x4x4d = vp8_sad4x4x4d_c; -#if ARCH_X86 || ARCH_X86_64 - cpi->rtcd.variance.copy32xn = vp8_copy32xn_c; -#endif - cpi->rtcd.variance.var4x4 = vp8_variance4x4_c; - cpi->rtcd.variance.var8x8 = vp8_variance8x8_c; - cpi->rtcd.variance.var8x16 = vp8_variance8x16_c; - cpi->rtcd.variance.var16x8 = vp8_variance16x8_c; - cpi->rtcd.variance.var16x16 = vp8_variance16x16_c; - - cpi->rtcd.variance.subpixvar4x4 = vp8_sub_pixel_variance4x4_c; - cpi->rtcd.variance.subpixvar8x8 = vp8_sub_pixel_variance8x8_c; - cpi->rtcd.variance.subpixvar8x16 = vp8_sub_pixel_variance8x16_c; - cpi->rtcd.variance.subpixvar16x8 = vp8_sub_pixel_variance16x8_c; - cpi->rtcd.variance.subpixvar16x16 = vp8_sub_pixel_variance16x16_c; - cpi->rtcd.variance.halfpixvar16x16_h = vp8_variance_halfpixvar16x16_h_c; - cpi->rtcd.variance.halfpixvar16x16_v = vp8_variance_halfpixvar16x16_v_c; - cpi->rtcd.variance.halfpixvar16x16_hv = vp8_variance_halfpixvar16x16_hv_c; - cpi->rtcd.variance.subpixmse16x16 = vp8_sub_pixel_mse16x16_c; - - cpi->rtcd.variance.mse16x16 = vp8_mse16x16_c; - cpi->rtcd.variance.getmbss = vp8_get_mb_ss_c; - - cpi->rtcd.variance.get4x4sse_cs = vp8_get4x4sse_cs_c; - cpi->rtcd.fdct.short4x4 = vp8_short_fdct4x4_c; cpi->rtcd.fdct.short8x4 = vp8_short_fdct8x4_c; cpi->rtcd.fdct.fast4x4 = vp8_short_fdct4x4_c; @@ -95,10 +48,6 @@ void vp8_cmachine_specific_config(VP8_COMP *cpi) #if !(CONFIG_REALTIME_ONLY) cpi->rtcd.temporal.apply = vp8_temporal_filter_apply_c; #endif -#if CONFIG_INTERNAL_STATS - cpi->rtcd.variance.ssimpf_8x8 = vp8_ssim_parms_8x8_c; - cpi->rtcd.variance.ssimpf_16x16 = vp8_ssim_parms_16x16_c; -#endif #endif // Pure C: diff --git a/vp8/encoder/onyx_if.c b/vp8/encoder/onyx_if.c index 65ae45375..416d4daa1 100644 --- a/vp8/encoder/onyx_if.c +++ b/vp8/encoder/onyx_if.c @@ -69,7 +69,7 @@ extern void vp8_yv12_copy_src_frame_func_neon(YV12_BUFFER_CONFIG *src_ybc, YV12_ int vp8_estimate_entropy_savings(VP8_COMP *cpi); -int vp8_calc_ss_err(YV12_BUFFER_CONFIG *source, YV12_BUFFER_CONFIG *dest, const vp8_variance_rtcd_vtable_t *rtcd); +int vp8_calc_ss_err(YV12_BUFFER_CONFIG *source, YV12_BUFFER_CONFIG *dest); extern void vp8_temporal_filter_prepare_c(VP8_COMP *cpi, int distance); @@ -85,8 +85,7 @@ extern double vp8_calc_ssim YV12_BUFFER_CONFIG *source, YV12_BUFFER_CONFIG *dest, int lumamask, - double *weight, - const vp8_variance_rtcd_vtable_t *rtcd + double *weight ); @@ -96,8 +95,7 @@ extern double vp8_calc_ssimg YV12_BUFFER_CONFIG *dest, double *ssim_y, double *ssim_u, - double *ssim_v, - const vp8_variance_rtcd_vtable_t *rtcd + double *ssim_v ); @@ -1947,62 +1945,62 @@ struct VP8_COMP* vp8_create_compressor(VP8_CONFIG *oxcf) vp8cx_create_encoder_threads(cpi); #endif - cpi->fn_ptr[BLOCK_16X16].sdf = VARIANCE_INVOKE(&cpi->rtcd.variance, sad16x16); - cpi->fn_ptr[BLOCK_16X16].vf = VARIANCE_INVOKE(&cpi->rtcd.variance, var16x16); - cpi->fn_ptr[BLOCK_16X16].svf = VARIANCE_INVOKE(&cpi->rtcd.variance, subpixvar16x16); - cpi->fn_ptr[BLOCK_16X16].svf_halfpix_h = VARIANCE_INVOKE(&cpi->rtcd.variance, halfpixvar16x16_h); - cpi->fn_ptr[BLOCK_16X16].svf_halfpix_v = VARIANCE_INVOKE(&cpi->rtcd.variance, halfpixvar16x16_v); - cpi->fn_ptr[BLOCK_16X16].svf_halfpix_hv = VARIANCE_INVOKE(&cpi->rtcd.variance, halfpixvar16x16_hv); - cpi->fn_ptr[BLOCK_16X16].sdx3f = VARIANCE_INVOKE(&cpi->rtcd.variance, sad16x16x3); - cpi->fn_ptr[BLOCK_16X16].sdx8f = VARIANCE_INVOKE(&cpi->rtcd.variance, sad16x16x8); - cpi->fn_ptr[BLOCK_16X16].sdx4df = VARIANCE_INVOKE(&cpi->rtcd.variance, sad16x16x4d); - - cpi->fn_ptr[BLOCK_16X8].sdf = VARIANCE_INVOKE(&cpi->rtcd.variance, sad16x8); - cpi->fn_ptr[BLOCK_16X8].vf = VARIANCE_INVOKE(&cpi->rtcd.variance, var16x8); - cpi->fn_ptr[BLOCK_16X8].svf = VARIANCE_INVOKE(&cpi->rtcd.variance, subpixvar16x8); + cpi->fn_ptr[BLOCK_16X16].sdf = vp8_sad16x16; + cpi->fn_ptr[BLOCK_16X16].vf = vp8_variance16x16; + cpi->fn_ptr[BLOCK_16X16].svf = vp8_sub_pixel_variance16x16; + cpi->fn_ptr[BLOCK_16X16].svf_halfpix_h = vp8_variance_halfpixvar16x16_h; + cpi->fn_ptr[BLOCK_16X16].svf_halfpix_v = vp8_variance_halfpixvar16x16_v; + cpi->fn_ptr[BLOCK_16X16].svf_halfpix_hv = vp8_variance_halfpixvar16x16_hv; + cpi->fn_ptr[BLOCK_16X16].sdx3f = vp8_sad16x16x3; + cpi->fn_ptr[BLOCK_16X16].sdx8f = vp8_sad16x16x8; + cpi->fn_ptr[BLOCK_16X16].sdx4df = vp8_sad16x16x4d; + + cpi->fn_ptr[BLOCK_16X8].sdf = vp8_sad16x8; + cpi->fn_ptr[BLOCK_16X8].vf = vp8_variance16x8; + cpi->fn_ptr[BLOCK_16X8].svf = vp8_sub_pixel_variance16x8; cpi->fn_ptr[BLOCK_16X8].svf_halfpix_h = NULL; cpi->fn_ptr[BLOCK_16X8].svf_halfpix_v = NULL; cpi->fn_ptr[BLOCK_16X8].svf_halfpix_hv = NULL; - cpi->fn_ptr[BLOCK_16X8].sdx3f = VARIANCE_INVOKE(&cpi->rtcd.variance, sad16x8x3); - cpi->fn_ptr[BLOCK_16X8].sdx8f = VARIANCE_INVOKE(&cpi->rtcd.variance, sad16x8x8); - cpi->fn_ptr[BLOCK_16X8].sdx4df = VARIANCE_INVOKE(&cpi->rtcd.variance, sad16x8x4d); + cpi->fn_ptr[BLOCK_16X8].sdx3f = vp8_sad16x8x3; + cpi->fn_ptr[BLOCK_16X8].sdx8f = vp8_sad16x8x8; + cpi->fn_ptr[BLOCK_16X8].sdx4df = vp8_sad16x8x4d; - cpi->fn_ptr[BLOCK_8X16].sdf = VARIANCE_INVOKE(&cpi->rtcd.variance, sad8x16); - cpi->fn_ptr[BLOCK_8X16].vf = VARIANCE_INVOKE(&cpi->rtcd.variance, var8x16); - cpi->fn_ptr[BLOCK_8X16].svf = VARIANCE_INVOKE(&cpi->rtcd.variance, subpixvar8x16); + cpi->fn_ptr[BLOCK_8X16].sdf = vp8_sad8x16; + cpi->fn_ptr[BLOCK_8X16].vf = vp8_variance8x16; + cpi->fn_ptr[BLOCK_8X16].svf = vp8_sub_pixel_variance8x16; cpi->fn_ptr[BLOCK_8X16].svf_halfpix_h = NULL; cpi->fn_ptr[BLOCK_8X16].svf_halfpix_v = NULL; cpi->fn_ptr[BLOCK_8X16].svf_halfpix_hv = NULL; - cpi->fn_ptr[BLOCK_8X16].sdx3f = VARIANCE_INVOKE(&cpi->rtcd.variance, sad8x16x3); - cpi->fn_ptr[BLOCK_8X16].sdx8f = VARIANCE_INVOKE(&cpi->rtcd.variance, sad8x16x8); - cpi->fn_ptr[BLOCK_8X16].sdx4df = VARIANCE_INVOKE(&cpi->rtcd.variance, sad8x16x4d); + cpi->fn_ptr[BLOCK_8X16].sdx3f = vp8_sad8x16x3; + cpi->fn_ptr[BLOCK_8X16].sdx8f = vp8_sad8x16x8; + cpi->fn_ptr[BLOCK_8X16].sdx4df = vp8_sad8x16x4d; - cpi->fn_ptr[BLOCK_8X8].sdf = VARIANCE_INVOKE(&cpi->rtcd.variance, sad8x8); - cpi->fn_ptr[BLOCK_8X8].vf = VARIANCE_INVOKE(&cpi->rtcd.variance, var8x8); - cpi->fn_ptr[BLOCK_8X8].svf = VARIANCE_INVOKE(&cpi->rtcd.variance, subpixvar8x8); + cpi->fn_ptr[BLOCK_8X8].sdf = vp8_sad8x8; + cpi->fn_ptr[BLOCK_8X8].vf = vp8_variance8x8; + cpi->fn_ptr[BLOCK_8X8].svf = vp8_sub_pixel_variance8x8; cpi->fn_ptr[BLOCK_8X8].svf_halfpix_h = NULL; cpi->fn_ptr[BLOCK_8X8].svf_halfpix_v = NULL; cpi->fn_ptr[BLOCK_8X8].svf_halfpix_hv = NULL; - cpi->fn_ptr[BLOCK_8X8].sdx3f = VARIANCE_INVOKE(&cpi->rtcd.variance, sad8x8x3); - cpi->fn_ptr[BLOCK_8X8].sdx8f = VARIANCE_INVOKE(&cpi->rtcd.variance, sad8x8x8); - cpi->fn_ptr[BLOCK_8X8].sdx4df = VARIANCE_INVOKE(&cpi->rtcd.variance, sad8x8x4d); + cpi->fn_ptr[BLOCK_8X8].sdx3f = vp8_sad8x8x3; + cpi->fn_ptr[BLOCK_8X8].sdx8f = vp8_sad8x8x8; + cpi->fn_ptr[BLOCK_8X8].sdx4df = vp8_sad8x8x4d; - cpi->fn_ptr[BLOCK_4X4].sdf = VARIANCE_INVOKE(&cpi->rtcd.variance, sad4x4); - cpi->fn_ptr[BLOCK_4X4].vf = VARIANCE_INVOKE(&cpi->rtcd.variance, var4x4); - cpi->fn_ptr[BLOCK_4X4].svf = VARIANCE_INVOKE(&cpi->rtcd.variance, subpixvar4x4); + cpi->fn_ptr[BLOCK_4X4].sdf = vp8_sad4x4; + cpi->fn_ptr[BLOCK_4X4].vf = vp8_variance4x4; + cpi->fn_ptr[BLOCK_4X4].svf = vp8_sub_pixel_variance4x4; cpi->fn_ptr[BLOCK_4X4].svf_halfpix_h = NULL; cpi->fn_ptr[BLOCK_4X4].svf_halfpix_v = NULL; cpi->fn_ptr[BLOCK_4X4].svf_halfpix_hv = NULL; - cpi->fn_ptr[BLOCK_4X4].sdx3f = VARIANCE_INVOKE(&cpi->rtcd.variance, sad4x4x3); - cpi->fn_ptr[BLOCK_4X4].sdx8f = VARIANCE_INVOKE(&cpi->rtcd.variance, sad4x4x8); - cpi->fn_ptr[BLOCK_4X4].sdx4df = VARIANCE_INVOKE(&cpi->rtcd.variance, sad4x4x4d); + cpi->fn_ptr[BLOCK_4X4].sdx3f = vp8_sad4x4x3; + cpi->fn_ptr[BLOCK_4X4].sdx8f = vp8_sad4x4x8; + cpi->fn_ptr[BLOCK_4X4].sdx4df = vp8_sad4x4x4d; #if ARCH_X86 || ARCH_X86_64 - cpi->fn_ptr[BLOCK_16X16].copymem = VARIANCE_INVOKE(&cpi->rtcd.variance, copy32xn); - cpi->fn_ptr[BLOCK_16X8].copymem = VARIANCE_INVOKE(&cpi->rtcd.variance, copy32xn); - cpi->fn_ptr[BLOCK_8X16].copymem = VARIANCE_INVOKE(&cpi->rtcd.variance, copy32xn); - cpi->fn_ptr[BLOCK_8X8].copymem = VARIANCE_INVOKE(&cpi->rtcd.variance, copy32xn); - cpi->fn_ptr[BLOCK_4X4].copymem = VARIANCE_INVOKE(&cpi->rtcd.variance, copy32xn); + cpi->fn_ptr[BLOCK_16X16].copymem = vp8_copy32xn; + cpi->fn_ptr[BLOCK_16X8].copymem = vp8_copy32xn; + cpi->fn_ptr[BLOCK_8X16].copymem = vp8_copy32xn; + cpi->fn_ptr[BLOCK_8X8].copymem = vp8_copy32xn; + cpi->fn_ptr[BLOCK_4X4].copymem = vp8_copy32xn; #endif cpi->full_search_sad = SEARCH_INVOKE(&cpi->rtcd.search, full_search); @@ -2334,8 +2332,7 @@ void vp8_remove_compressor(VP8_COMP **ptr) static uint64_t calc_plane_error(unsigned char *orig, int orig_stride, unsigned char *recon, int recon_stride, - unsigned int cols, unsigned int rows, - vp8_variance_rtcd_vtable_t *rtcd) + unsigned int cols, unsigned int rows) { unsigned int row, col; uint64_t total_sse = 0; @@ -2347,7 +2344,7 @@ static uint64_t calc_plane_error(unsigned char *orig, int orig_stride, { unsigned int sse; - VARIANCE_INVOKE(rtcd, mse16x16)(orig + col, orig_stride, + vp8_mse16x16(orig + col, orig_stride, recon + col, recon_stride, &sse); total_sse += sse; @@ -2408,8 +2405,7 @@ static void generate_psnr_packet(VP8_COMP *cpi) pkt.kind = VPX_CODEC_PSNR_PKT; sse = calc_plane_error(orig->y_buffer, orig->y_stride, recon->y_buffer, recon->y_stride, - width, height, - IF_RTCD(&cpi->rtcd.variance)); + width, height); pkt.data.psnr.sse[0] = sse; pkt.data.psnr.sse[1] = sse; pkt.data.psnr.samples[0] = width * height; @@ -2420,8 +2416,7 @@ static void generate_psnr_packet(VP8_COMP *cpi) sse = calc_plane_error(orig->u_buffer, orig->uv_stride, recon->u_buffer, recon->uv_stride, - width, height, - IF_RTCD(&cpi->rtcd.variance)); + width, height); pkt.data.psnr.sse[0] += sse; pkt.data.psnr.sse[2] = sse; pkt.data.psnr.samples[0] += width * height; @@ -2429,8 +2424,7 @@ static void generate_psnr_packet(VP8_COMP *cpi) sse = calc_plane_error(orig->v_buffer, orig->uv_stride, recon->v_buffer, recon->uv_stride, - width, height, - IF_RTCD(&cpi->rtcd.variance)); + width, height); pkt.data.psnr.sse[0] += sse; pkt.data.psnr.sse[3] = sse; pkt.data.psnr.samples[0] += width * height; @@ -3819,8 +3813,7 @@ static void encode_frame_to_data_rate { int last_q = Q; int kf_err = vp8_calc_ss_err(cpi->Source, - &cm->yv12_fb[cm->new_fb_idx], - IF_RTCD(&cpi->rtcd.variance)); + &cm->yv12_fb[cm->new_fb_idx]); // The key frame is not good enough if ( kf_err > ((cpi->ambient_err * 7) >> 3) ) @@ -4016,8 +4009,7 @@ static void encode_frame_to_data_rate if ( cpi->next_key_frame_forced && (cpi->twopass.frames_to_key == 0) ) { cpi->ambient_err = vp8_calc_ss_err(cpi->Source, - &cm->yv12_fb[cm->new_fb_idx], - IF_RTCD(&cpi->rtcd.variance)); + &cm->yv12_fb[cm->new_fb_idx]); } /* This frame's MVs are saved and will be used in next frame's MV predictor. @@ -4961,16 +4953,13 @@ int vp8_get_compressed_data(VP8_COMP *cpi, unsigned int *frame_flags, unsigned l int64_t sq_error, sq_error2; ye = calc_plane_error(orig->y_buffer, orig->y_stride, - recon->y_buffer, recon->y_stride, orig->y_width, orig->y_height, - IF_RTCD(&cpi->rtcd.variance)); + recon->y_buffer, recon->y_stride, orig->y_width, orig->y_height); ue = calc_plane_error(orig->u_buffer, orig->uv_stride, - recon->u_buffer, recon->uv_stride, orig->uv_width, orig->uv_height, - IF_RTCD(&cpi->rtcd.variance)); + recon->u_buffer, recon->uv_stride, orig->uv_width, orig->uv_height); ve = calc_plane_error(orig->v_buffer, orig->uv_stride, - recon->v_buffer, recon->uv_stride, orig->uv_width, orig->uv_height, - IF_RTCD(&cpi->rtcd.variance)); + recon->v_buffer, recon->uv_stride, orig->uv_width, orig->uv_height); sq_error = ye + ue + ve; @@ -4985,20 +4974,17 @@ int vp8_get_compressed_data(VP8_COMP *cpi, unsigned int *frame_flags, unsigned l double frame_psnr2, frame_ssim2 = 0; double weight = 0; - vp8_deblock(cm->frame_to_show, &cm->post_proc_buffer, cm->filter_level * 10 / 6, 1, 0, IF_RTCD(&cm->rtcd.postproc)); + vp8_deblock(cm->frame_to_show, &cm->post_proc_buffer, cm->filter_level * 10 / 6, 1, 0); vp8_clear_system_state(); ye = calc_plane_error(orig->y_buffer, orig->y_stride, - pp->y_buffer, pp->y_stride, orig->y_width, orig->y_height, - IF_RTCD(&cpi->rtcd.variance)); + pp->y_buffer, pp->y_stride, orig->y_width, orig->y_height); ue = calc_plane_error(orig->u_buffer, orig->uv_stride, - pp->u_buffer, pp->uv_stride, orig->uv_width, orig->uv_height, - IF_RTCD(&cpi->rtcd.variance)); + pp->u_buffer, pp->uv_stride, orig->uv_width, orig->uv_height); ve = calc_plane_error(orig->v_buffer, orig->uv_stride, - pp->v_buffer, pp->uv_stride, orig->uv_width, orig->uv_height, - IF_RTCD(&cpi->rtcd.variance)); + pp->v_buffer, pp->uv_stride, orig->uv_width, orig->uv_height); sq_error2 = ye + ue + ve; @@ -5011,8 +4997,7 @@ int vp8_get_compressed_data(VP8_COMP *cpi, unsigned int *frame_flags, unsigned l cpi->totalp += frame_psnr2; frame_ssim2 = vp8_calc_ssim(cpi->Source, - &cm->post_proc_buffer, 1, &weight, - IF_RTCD(&cpi->rtcd.variance)); + &cm->post_proc_buffer, 1, &weight); cpi->summed_quality += frame_ssim2 * weight; cpi->summed_weights += weight; @@ -5042,7 +5027,7 @@ int vp8_get_compressed_data(VP8_COMP *cpi, unsigned int *frame_flags, unsigned l { double y, u, v, frame_all; frame_all = vp8_calc_ssimg(cpi->Source, cm->frame_to_show, - &y, &u, &v, IF_RTCD(&cpi->rtcd.variance)); + &y, &u, &v); if (cpi->oxcf.number_of_layers > 1) { @@ -5222,14 +5207,13 @@ int vp8_set_internal_size(VP8_COMP *cpi, VPX_SCALING horiz_mode, VPX_SCALING ver -int vp8_calc_ss_err(YV12_BUFFER_CONFIG *source, YV12_BUFFER_CONFIG *dest, const vp8_variance_rtcd_vtable_t *rtcd) +int vp8_calc_ss_err(YV12_BUFFER_CONFIG *source, YV12_BUFFER_CONFIG *dest) { int i, j; int Total = 0; unsigned char *src = source->y_buffer; unsigned char *dst = dest->y_buffer; - (void)rtcd; // Loop through the Y plane raw and reconstruction data summing (square differences) for (i = 0; i < source->y_height; i += 16) @@ -5237,7 +5221,7 @@ int vp8_calc_ss_err(YV12_BUFFER_CONFIG *source, YV12_BUFFER_CONFIG *dest, const for (j = 0; j < source->y_width; j += 16) { unsigned int sse; - Total += VARIANCE_INVOKE(rtcd, mse16x16)(src + j, source->y_stride, dst + j, dest->y_stride, &sse); + Total += vp8_mse16x16(src + j, source->y_stride, dst + j, dest->y_stride, &sse); } src += 16 * source->y_stride; diff --git a/vp8/encoder/onyx_int.h b/vp8/encoder/onyx_int.h index eeadf62a6..8a0c3df06 100644 --- a/vp8/encoder/onyx_int.h +++ b/vp8/encoder/onyx_int.h @@ -224,7 +224,6 @@ typedef struct typedef struct VP8_ENCODER_RTCD { - vp8_variance_rtcd_vtable_t variance; vp8_fdct_rtcd_vtable_t fdct; vp8_encodemb_rtcd_vtable_t encodemb; vp8_quantize_rtcd_vtable_t quantize; diff --git a/vp8/encoder/pickinter.c b/vp8/encoder/pickinter.c index e8a5d9522..4f69a5a89 100644 --- a/vp8/encoder/pickinter.c +++ b/vp8/encoder/pickinter.c @@ -31,7 +31,7 @@ #define IF_RTCD(x) NULL #endif -extern int VP8_UVSSE(MACROBLOCK *x, const vp8_variance_rtcd_vtable_t *rtcd); +extern int VP8_UVSSE(MACROBLOCK *x); #ifdef SPEEDSTATS extern unsigned int cnt_pm; @@ -40,7 +40,6 @@ extern unsigned int cnt_pm; extern const int vp8_ref_frame_order[MAX_MODES]; extern const MB_PREDICTION_MODE vp8_mode_order[MAX_MODES]; -extern unsigned int (*vp8_get4x4sse_cs)(unsigned char *src_ptr, int source_stride, unsigned char *ref_ptr, int recon_stride); extern int vp8_cost_mv_ref(MB_PREDICTION_MODE m, const int near_mv_ref_ct[4]); @@ -120,14 +119,14 @@ unsigned int vp8_get4x4sse_cs_c return distortion; } -static int get_prediction_error(BLOCK *be, BLOCKD *b, const vp8_variance_rtcd_vtable_t *rtcd) +static int get_prediction_error(BLOCK *be, BLOCKD *b) { unsigned char *sptr; unsigned char *dptr; sptr = (*(be->base_src) + be->src); dptr = b->predictor; - return VARIANCE_INVOKE(rtcd, get4x4sse_cs)(sptr, be->src_stride, dptr, 16); + return vp8_get4x4sse_cs(sptr, be->src_stride, dptr, 16); } @@ -157,7 +156,7 @@ static int pick_intra4x4block( vp8_intra4x4_predict (*(b->base_dst) + b->dst, b->dst_stride, mode, b->predictor, 16); - distortion = get_prediction_error(be, b, &rtcd->variance); + distortion = get_prediction_error(be, b); this_rd = RDCOST(x->rdmult, x->rddiv, rate, distortion); if (this_rd < best_rd) @@ -674,8 +673,7 @@ void vp8_pick_inter_mode(VP8_COMP *cpi, MACROBLOCK *x, int recon_yoffset, else { rate2 += rate; - distortion2 = VARIANCE_INVOKE - (&cpi->rtcd.variance, var16x16)( + distortion2 = vp8_variance16x16( *(b->base_src), b->src_stride, x->e_mbd.predictor, 16, &sse); this_rd = RDCOST(x->rdmult, x->rddiv, rate2, distortion2); @@ -700,7 +698,7 @@ void vp8_pick_inter_mode(VP8_COMP *cpi, MACROBLOCK *x, int recon_yoffset, case TM_PRED: vp8_build_intra_predictors_mby (&x->e_mbd); - distortion2 = VARIANCE_INVOKE(&cpi->rtcd.variance, var16x16) + distortion2 = vp8_variance16x16 (*(b->base_src), b->src_stride, x->e_mbd.predictor, 16, &sse); rate2 += x->mbmode_cost[x->e_mbd.frame_type][x->e_mbd.mode_info_context->mbmi.mode]; @@ -937,7 +935,7 @@ void vp8_pick_inter_mode(VP8_COMP *cpi, MACROBLOCK *x, int recon_yoffset, // Check u and v to make sure skip is ok int sse2 = 0; - sse2 = VP8_UVSSE(x, IF_RTCD(&cpi->rtcd.variance)); + sse2 = VP8_UVSSE(x); if (sse2 * 2 < x->encode_breakout) x->skip = 1; @@ -1071,7 +1069,7 @@ void vp8_pick_intra_mode(VP8_COMP *cpi, MACROBLOCK *x, int *rate_) x->e_mbd.mode_info_context->mbmi.mode = mode; vp8_build_intra_predictors_mby (&x->e_mbd); - distortion = VARIANCE_INVOKE(&cpi->rtcd.variance, var16x16) + distortion = vp8_variance16x16 (*(b->base_src), b->src_stride, x->e_mbd.predictor, 16, &sse); rate = x->mbmode_cost[x->e_mbd.frame_type][mode]; this_rd = RDCOST(x->rdmult, x->rddiv, rate, distortion); diff --git a/vp8/encoder/picklpf.c b/vp8/encoder/picklpf.c index 2449ae540..09e45bea6 100644 --- a/vp8/encoder/picklpf.c +++ b/vp8/encoder/picklpf.c @@ -21,7 +21,7 @@ #include "vpx_ports/arm.h" #endif -extern int vp8_calc_ss_err(YV12_BUFFER_CONFIG *source, YV12_BUFFER_CONFIG *dest, const vp8_variance_rtcd_vtable_t *rtcd); +extern int vp8_calc_ss_err(YV12_BUFFER_CONFIG *source, YV12_BUFFER_CONFIG *dest); #if CONFIG_RUNTIME_CPU_DETECT #define IF_RTCD(x) (x) @@ -64,8 +64,7 @@ void vp8_yv12_copy_partial_frame(YV12_BUFFER_CONFIG *src_ybc, } static int calc_partial_ssl_err(YV12_BUFFER_CONFIG *source, - YV12_BUFFER_CONFIG *dest, - const vp8_variance_rtcd_vtable_t *rtcd) + YV12_BUFFER_CONFIG *dest) { int i, j; int Total = 0; @@ -93,7 +92,7 @@ static int calc_partial_ssl_err(YV12_BUFFER_CONFIG *source, for (j = 0; j < source->y_width; j += 16) { unsigned int sse; - Total += VARIANCE_INVOKE(rtcd, mse16x16)(src + j, source->y_stride, + Total += vp8_mse16x16(src + j, source->y_stride, dst + j, dest->y_stride, &sse); } @@ -183,8 +182,7 @@ void vp8cx_pick_filter_level_fast(YV12_BUFFER_CONFIG *sd, VP8_COMP *cpi) vp8_yv12_copy_partial_frame_ptr(saved_frame, cm->frame_to_show); vp8_loop_filter_partial_frame(cm, &cpi->mb.e_mbd, filt_val); - best_err = calc_partial_ssl_err(sd, cm->frame_to_show, - IF_RTCD(&cpi->rtcd.variance)); + best_err = calc_partial_ssl_err(sd, cm->frame_to_show); filt_val -= 1 + (filt_val > 10); @@ -196,8 +194,7 @@ void vp8cx_pick_filter_level_fast(YV12_BUFFER_CONFIG *sd, VP8_COMP *cpi) vp8_loop_filter_partial_frame(cm, &cpi->mb.e_mbd, filt_val); // Get the err for filtered frame - filt_err = calc_partial_ssl_err(sd, cm->frame_to_show, - IF_RTCD(&cpi->rtcd.variance)); + filt_err = calc_partial_ssl_err(sd, cm->frame_to_show); // Update the best case record or exit loop. if (filt_err < best_err) @@ -228,8 +225,7 @@ void vp8cx_pick_filter_level_fast(YV12_BUFFER_CONFIG *sd, VP8_COMP *cpi) vp8_loop_filter_partial_frame(cm, &cpi->mb.e_mbd, filt_val); // Get the err for filtered frame - filt_err = calc_partial_ssl_err(sd, cm->frame_to_show, - IF_RTCD(&cpi->rtcd.variance)); + filt_err = calc_partial_ssl_err(sd, cm->frame_to_show); // Update the best case record or exit loop. if (filt_err < best_err) @@ -322,8 +318,7 @@ void vp8cx_pick_filter_level(YV12_BUFFER_CONFIG *sd, VP8_COMP *cpi) vp8cx_set_alt_lf_level(cpi, filt_mid); vp8_loop_filter_frame_yonly(cm, &cpi->mb.e_mbd, filt_mid); - best_err = vp8_calc_ss_err(sd, cm->frame_to_show, - IF_RTCD(&cpi->rtcd.variance)); + best_err = vp8_calc_ss_err(sd, cm->frame_to_show); ss_err[filt_mid] = best_err; @@ -349,8 +344,7 @@ void vp8cx_pick_filter_level(YV12_BUFFER_CONFIG *sd, VP8_COMP *cpi) vp8cx_set_alt_lf_level(cpi, filt_low); vp8_loop_filter_frame_yonly(cm, &cpi->mb.e_mbd, filt_low); - filt_err = vp8_calc_ss_err(sd, cm->frame_to_show, - IF_RTCD(&cpi->rtcd.variance)); + filt_err = vp8_calc_ss_err(sd, cm->frame_to_show); ss_err[filt_low] = filt_err; } else @@ -376,8 +370,7 @@ void vp8cx_pick_filter_level(YV12_BUFFER_CONFIG *sd, VP8_COMP *cpi) vp8cx_set_alt_lf_level(cpi, filt_high); vp8_loop_filter_frame_yonly(cm, &cpi->mb.e_mbd, filt_high); - filt_err = vp8_calc_ss_err(sd, cm->frame_to_show, - IF_RTCD(&cpi->rtcd.variance)); + filt_err = vp8_calc_ss_err(sd, cm->frame_to_show); ss_err[filt_high] = filt_err; } else diff --git a/vp8/encoder/rdopt.c b/vp8/encoder/rdopt.c index 278445535..015409d48 100644 --- a/vp8/encoder/rdopt.c +++ b/vp8/encoder/rdopt.c @@ -453,7 +453,7 @@ int vp8_mbuverror_c(MACROBLOCK *mb) return error; } -int VP8_UVSSE(MACROBLOCK *x, const vp8_variance_rtcd_vtable_t *rtcd) +int VP8_UVSSE(MACROBLOCK *x) { unsigned char *uptr, *vptr; unsigned char *upred_ptr = (*(x->block[16].base_src) + x->block[16].src); @@ -486,17 +486,17 @@ int VP8_UVSSE(MACROBLOCK *x, const vp8_variance_rtcd_vtable_t *rtcd) if ((mv_row | mv_col) & 7) { - VARIANCE_INVOKE(rtcd, subpixvar8x8)(uptr, pre_stride, + vp8_sub_pixel_variance8x8(uptr, pre_stride, mv_col & 7, mv_row & 7, upred_ptr, uv_stride, &sse2); - VARIANCE_INVOKE(rtcd, subpixvar8x8)(vptr, pre_stride, + vp8_sub_pixel_variance8x8(vptr, pre_stride, mv_col & 7, mv_row & 7, vpred_ptr, uv_stride, &sse1); sse2 += sse1; } else { - VARIANCE_INVOKE(rtcd, var8x8)(uptr, pre_stride, + vp8_variance8x8(uptr, pre_stride, upred_ptr, uv_stride, &sse2); - VARIANCE_INVOKE(rtcd, var8x8)(vptr, pre_stride, + vp8_variance8x8(vptr, pre_stride, vpred_ptr, uv_stride, &sse1); sse2 += sse1; } @@ -2137,7 +2137,7 @@ void vp8_rd_pick_inter_mode(VP8_COMP *cpi, MACROBLOCK *x, int recon_yoffset, if(threshold < x->encode_breakout) threshold = x->encode_breakout; - var = VARIANCE_INVOKE(&cpi->rtcd.variance, var16x16) + var = vp8_variance16x16 (*(b->base_src), b->src_stride, x->e_mbd.predictor, 16, &sse); @@ -2150,7 +2150,7 @@ void vp8_rd_pick_inter_mode(VP8_COMP *cpi, MACROBLOCK *x, int recon_yoffset, (sse /2 > var && sse-var < 64)) { // Check u and v to make sure skip is ok - int sse2= VP8_UVSSE(x, IF_RTCD(&cpi->rtcd.variance)); + int sse2= VP8_UVSSE(x); if (sse2 * 2 < threshold) { x->skip = 1; diff --git a/vp8/encoder/ssim.c b/vp8/encoder/ssim.c index d0f8e490a..e75160836 100644 --- a/vp8/encoder/ssim.c +++ b/vp8/encoder/ssim.c @@ -94,25 +94,22 @@ static double similarity return ssim_n * 1.0 / ssim_d; } -static double ssim_16x16(unsigned char *s,int sp, unsigned char *r,int rp, - const vp8_variance_rtcd_vtable_t *rtcd) +static double ssim_16x16(unsigned char *s,int sp, unsigned char *r,int rp) { unsigned long sum_s=0,sum_r=0,sum_sq_s=0,sum_sq_r=0,sum_sxr=0; - SSIMPF_INVOKE(rtcd,16x16)(s, sp, r, rp, &sum_s, &sum_r, &sum_sq_s, &sum_sq_r, &sum_sxr); + vp8_ssim_parms_16x16(s, sp, r, rp, &sum_s, &sum_r, &sum_sq_s, &sum_sq_r, &sum_sxr); return similarity(sum_s, sum_r, sum_sq_s, sum_sq_r, sum_sxr, 256); } -static double ssim_8x8(unsigned char *s,int sp, unsigned char *r,int rp, - const vp8_variance_rtcd_vtable_t *rtcd) +static double ssim_8x8(unsigned char *s,int sp, unsigned char *r,int rp) { unsigned long sum_s=0,sum_r=0,sum_sq_s=0,sum_sq_r=0,sum_sxr=0; - SSIMPF_INVOKE(rtcd,8x8)(s, sp, r, rp, &sum_s, &sum_r, &sum_sq_s, &sum_sq_r, &sum_sxr); + vp8_ssim_parms_8x8(s, sp, r, rp, &sum_s, &sum_r, &sum_sq_s, &sum_sq_r, &sum_sxr); return similarity(sum_s, sum_r, sum_sq_s, sum_sq_r, sum_sxr, 64); } // TODO: (jbb) tried to scale this function such that we may be able to use it // for distortion metric in mode selection code ( provided we do a reconstruction) -long dssim(unsigned char *s,int sp, unsigned char *r,int rp, - const vp8_variance_rtcd_vtable_t *rtcd) +long dssim(unsigned char *s,int sp, unsigned char *r,int rp) { unsigned long sum_s=0,sum_r=0,sum_sq_s=0,sum_sq_r=0,sum_sxr=0; int64_t ssim3; @@ -125,7 +122,7 @@ long dssim(unsigned char *s,int sp, unsigned char *r,int rp, c1 = cc1*16; c2 = cc2*16; - SSIMPF_INVOKE(rtcd,16x16)(s, sp, r, rp, &sum_s, &sum_r, &sum_sq_s, &sum_sq_r, &sum_sxr); + vp8_ssim_parms_16x16(s, sp, r, rp, &sum_s, &sum_r, &sum_sq_s, &sum_sq_r, &sum_sxr); ssim_n1 = (2*sum_s*sum_r+ c1); ssim_n2 =((int64_t) 2*256*sum_sxr-(int64_t) 2*sum_s*sum_r+c2); @@ -154,8 +151,7 @@ double vp8_ssim2 int stride_img1, int stride_img2, int width, - int height, - const vp8_variance_rtcd_vtable_t *rtcd + int height ) { int i,j; @@ -167,7 +163,7 @@ double vp8_ssim2 { for(j=0; j < width-8; j+=4 ) { - double v = ssim_8x8(img1+j, stride_img1, img2+j, stride_img2, rtcd); + double v = ssim_8x8(img1+j, stride_img1, img2+j, stride_img2); ssim_total += v; samples++; } @@ -180,8 +176,7 @@ double vp8_calc_ssim YV12_BUFFER_CONFIG *source, YV12_BUFFER_CONFIG *dest, int lumamask, - double *weight, - const vp8_variance_rtcd_vtable_t *rtcd + double *weight ) { double a, b, c; @@ -189,15 +184,15 @@ double vp8_calc_ssim a = vp8_ssim2(source->y_buffer, dest->y_buffer, source->y_stride, dest->y_stride, source->y_width, - source->y_height, rtcd); + source->y_height); b = vp8_ssim2(source->u_buffer, dest->u_buffer, source->uv_stride, dest->uv_stride, source->uv_width, - source->uv_height, rtcd); + source->uv_height); c = vp8_ssim2(source->v_buffer, dest->v_buffer, source->uv_stride, dest->uv_stride, source->uv_width, - source->uv_height, rtcd); + source->uv_height); ssimv = a * .8 + .1 * (b + c); @@ -212,8 +207,7 @@ double vp8_calc_ssimg YV12_BUFFER_CONFIG *dest, double *ssim_y, double *ssim_u, - double *ssim_v, - const vp8_variance_rtcd_vtable_t *rtcd + double *ssim_v ) { double ssim_all = 0; @@ -221,15 +215,15 @@ double vp8_calc_ssimg a = vp8_ssim2(source->y_buffer, dest->y_buffer, source->y_stride, dest->y_stride, source->y_width, - source->y_height, rtcd); + source->y_height); b = vp8_ssim2(source->u_buffer, dest->u_buffer, source->uv_stride, dest->uv_stride, source->uv_width, - source->uv_height, rtcd); + source->uv_height); c = vp8_ssim2(source->v_buffer, dest->v_buffer, source->uv_stride, dest->uv_stride, source->uv_width, - source->uv_height, rtcd); + source->uv_height); *ssim_y = a; *ssim_u = b; *ssim_v = c; diff --git a/vp8/encoder/variance.h b/vp8/encoder/variance.h index d9bf66975..e216a2222 100644 --- a/vp8/encoder/variance.h +++ b/vp8/encoder/variance.h @@ -12,392 +12,89 @@ #ifndef VARIANCE_H #define VARIANCE_H -#define prototype_sad(sym)\ - unsigned int (sym)\ - (\ - const unsigned char *src_ptr, \ - int source_stride, \ - const unsigned char *ref_ptr, \ - int ref_stride, \ - int max_sad\ - ) - -#define prototype_sad_multi_same_address(sym)\ - void (sym)\ - (\ - const unsigned char *src_ptr, \ - int source_stride, \ - const unsigned char *ref_ptr, \ - int ref_stride, \ - unsigned int *sad_array\ - ) - -#define prototype_sad_multi_same_address_1(sym)\ - void (sym)\ - (\ - const unsigned char *src_ptr, \ - int source_stride, \ - const unsigned char *ref_ptr, \ - int ref_stride, \ - unsigned short *sad_array\ - ) +typedef unsigned int(*vp8_sad_fn_t) + ( + const unsigned char *src_ptr, + int source_stride, + const unsigned char *ref_ptr, + int ref_stride, + int max_sad + ); -#define prototype_sad_multi_dif_address(sym)\ - void (sym)\ - (\ - const unsigned char *src_ptr, \ - int source_stride, \ - unsigned char *ref_ptr[4], \ - int ref_stride, \ - unsigned int *sad_array\ - ) +typedef void (*vp8_copy32xn_fn_t)( + const unsigned char *src_ptr, + int source_stride, + const unsigned char *ref_ptr, + int ref_stride, + int n); + +typedef void (*vp8_sad_multi_fn_t)( + const unsigned char *src_ptr, + int source_stride, + const unsigned char *ref_ptr, + int ref_stride, + unsigned int *sad_array); + +typedef void (*vp8_sad_multi1_fn_t) + ( + const unsigned char *src_ptr, + int source_stride, + const unsigned char *ref_ptr, + int ref_stride, + unsigned short *sad_array + ); -#define prototype_variance(sym) \ - unsigned int (sym) \ - (\ - const unsigned char *src_ptr, \ - int source_stride, \ - const unsigned char *ref_ptr, \ - int ref_stride, \ - unsigned int *sse\ - ) +typedef void (*vp8_sad_multi_d_fn_t) + ( + const unsigned char *src_ptr, + int source_stride, + unsigned char *ref_ptr[4], + int ref_stride, + unsigned int *sad_array + ); -#define prototype_variance2(sym) \ - unsigned int (sym) \ - (\ - const unsigned char *src_ptr, \ - int source_stride, \ - const unsigned char *ref_ptr, \ - int ref_stride, \ - unsigned int *sse,\ - int *sum\ - ) +typedef unsigned int (*vp8_variance_fn_t) + ( + const unsigned char *src_ptr, + int source_stride, + const unsigned char *ref_ptr, + int ref_stride, + unsigned int *sse + ); -#define prototype_subpixvariance(sym) \ - unsigned int (sym) \ - ( \ - const unsigned char *src_ptr, \ - int source_stride, \ - int xoffset, \ - int yoffset, \ - const unsigned char *ref_ptr, \ - int Refstride, \ - unsigned int *sse \ +typedef unsigned int (*vp8_subpixvariance_fn_t) + ( + const unsigned char *src_ptr, + int source_stride, + int xoffset, + int yoffset, + const unsigned char *ref_ptr, + int Refstride, + unsigned int *sse ); -#define prototype_ssimpf(sym) \ - void (sym) \ - ( \ - unsigned char *s, \ - int sp, \ - unsigned char *r, \ - int rp, \ - unsigned long *sum_s, \ - unsigned long *sum_r, \ - unsigned long *sum_sq_s, \ - unsigned long *sum_sq_r, \ - unsigned long *sum_sxr \ +typedef void (*vp8_ssimpf_fn_t) + ( + unsigned char *s, + int sp, + unsigned char *r, + int rp, + unsigned long *sum_s, + unsigned long *sum_r, + unsigned long *sum_sq_s, + unsigned long *sum_sq_r, + unsigned long *sum_sxr ); -#define prototype_getmbss(sym) unsigned int (sym)(const short *) - -#define prototype_get16x16prederror(sym)\ - unsigned int (sym)\ - (\ - const unsigned char *src_ptr, \ - int source_stride, \ - const unsigned char *ref_ptr, \ - int ref_stride \ - ) - -#if ARCH_X86 || ARCH_X86_64 -#include "x86/variance_x86.h" -#endif - -#if ARCH_ARM -#include "arm/variance_arm.h" -#endif - -#ifndef vp8_variance_sad4x4 -#define vp8_variance_sad4x4 vp8_sad4x4_c -#endif -extern prototype_sad(vp8_variance_sad4x4); - -#ifndef vp8_variance_sad8x8 -#define vp8_variance_sad8x8 vp8_sad8x8_c -#endif -extern prototype_sad(vp8_variance_sad8x8); - -#ifndef vp8_variance_sad8x16 -#define vp8_variance_sad8x16 vp8_sad8x16_c -#endif -extern prototype_sad(vp8_variance_sad8x16); - -#ifndef vp8_variance_sad16x8 -#define vp8_variance_sad16x8 vp8_sad16x8_c -#endif -extern prototype_sad(vp8_variance_sad16x8); - -#ifndef vp8_variance_sad16x16 -#define vp8_variance_sad16x16 vp8_sad16x16_c -#endif -extern prototype_sad(vp8_variance_sad16x16); - -//-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=- - -#ifndef vp8_variance_sad16x16x3 -#define vp8_variance_sad16x16x3 vp8_sad16x16x3_c -#endif -extern prototype_sad_multi_same_address(vp8_variance_sad16x16x3); - -#ifndef vp8_variance_sad16x8x3 -#define vp8_variance_sad16x8x3 vp8_sad16x8x3_c -#endif -extern prototype_sad_multi_same_address(vp8_variance_sad16x8x3); - -#ifndef vp8_variance_sad8x8x3 -#define vp8_variance_sad8x8x3 vp8_sad8x8x3_c -#endif -extern prototype_sad_multi_same_address(vp8_variance_sad8x8x3); - -#ifndef vp8_variance_sad8x16x3 -#define vp8_variance_sad8x16x3 vp8_sad8x16x3_c -#endif -extern prototype_sad_multi_same_address(vp8_variance_sad8x16x3); - -#ifndef vp8_variance_sad4x4x3 -#define vp8_variance_sad4x4x3 vp8_sad4x4x3_c -#endif -extern prototype_sad_multi_same_address(vp8_variance_sad4x4x3); - -#ifndef vp8_variance_sad16x16x8 -#define vp8_variance_sad16x16x8 vp8_sad16x16x8_c -#endif -extern prototype_sad_multi_same_address_1(vp8_variance_sad16x16x8); - -#ifndef vp8_variance_sad16x8x8 -#define vp8_variance_sad16x8x8 vp8_sad16x8x8_c -#endif -extern prototype_sad_multi_same_address_1(vp8_variance_sad16x8x8); - -#ifndef vp8_variance_sad8x8x8 -#define vp8_variance_sad8x8x8 vp8_sad8x8x8_c -#endif -extern prototype_sad_multi_same_address_1(vp8_variance_sad8x8x8); - -#ifndef vp8_variance_sad8x16x8 -#define vp8_variance_sad8x16x8 vp8_sad8x16x8_c -#endif -extern prototype_sad_multi_same_address_1(vp8_variance_sad8x16x8); - -#ifndef vp8_variance_sad4x4x8 -#define vp8_variance_sad4x4x8 vp8_sad4x4x8_c -#endif -extern prototype_sad_multi_same_address_1(vp8_variance_sad4x4x8); - -//-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=- - -#ifndef vp8_variance_sad16x16x4d -#define vp8_variance_sad16x16x4d vp8_sad16x16x4d_c -#endif -extern prototype_sad_multi_dif_address(vp8_variance_sad16x16x4d); - -#ifndef vp8_variance_sad16x8x4d -#define vp8_variance_sad16x8x4d vp8_sad16x8x4d_c -#endif -extern prototype_sad_multi_dif_address(vp8_variance_sad16x8x4d); - -#ifndef vp8_variance_sad8x8x4d -#define vp8_variance_sad8x8x4d vp8_sad8x8x4d_c -#endif -extern prototype_sad_multi_dif_address(vp8_variance_sad8x8x4d); - -#ifndef vp8_variance_sad8x16x4d -#define vp8_variance_sad8x16x4d vp8_sad8x16x4d_c -#endif -extern prototype_sad_multi_dif_address(vp8_variance_sad8x16x4d); - -#ifndef vp8_variance_sad4x4x4d -#define vp8_variance_sad4x4x4d vp8_sad4x4x4d_c -#endif -extern prototype_sad_multi_dif_address(vp8_variance_sad4x4x4d); - -#if ARCH_X86 || ARCH_X86_64 -#ifndef vp8_variance_copy32xn -#define vp8_variance_copy32xn vp8_copy32xn_c -#endif -extern prototype_sad(vp8_variance_copy32xn); -#endif - -//-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=- +typedef unsigned int (*vp8_getmbss_fn_t)(const short *); -#ifndef vp8_variance_var4x4 -#define vp8_variance_var4x4 vp8_variance4x4_c -#endif -extern prototype_variance(vp8_variance_var4x4); - -#ifndef vp8_variance_var8x8 -#define vp8_variance_var8x8 vp8_variance8x8_c -#endif -extern prototype_variance(vp8_variance_var8x8); - -#ifndef vp8_variance_var8x16 -#define vp8_variance_var8x16 vp8_variance8x16_c -#endif -extern prototype_variance(vp8_variance_var8x16); - -#ifndef vp8_variance_var16x8 -#define vp8_variance_var16x8 vp8_variance16x8_c -#endif -extern prototype_variance(vp8_variance_var16x8); - -#ifndef vp8_variance_var16x16 -#define vp8_variance_var16x16 vp8_variance16x16_c -#endif -extern prototype_variance(vp8_variance_var16x16); - -//-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=- - -#ifndef vp8_variance_subpixvar4x4 -#define vp8_variance_subpixvar4x4 vp8_sub_pixel_variance4x4_c -#endif -extern prototype_subpixvariance(vp8_variance_subpixvar4x4); - -#ifndef vp8_variance_subpixvar8x8 -#define vp8_variance_subpixvar8x8 vp8_sub_pixel_variance8x8_c -#endif -extern prototype_subpixvariance(vp8_variance_subpixvar8x8); - -#ifndef vp8_variance_subpixvar8x16 -#define vp8_variance_subpixvar8x16 vp8_sub_pixel_variance8x16_c -#endif -extern prototype_subpixvariance(vp8_variance_subpixvar8x16); - -#ifndef vp8_variance_subpixvar16x8 -#define vp8_variance_subpixvar16x8 vp8_sub_pixel_variance16x8_c -#endif -extern prototype_subpixvariance(vp8_variance_subpixvar16x8); - -#ifndef vp8_variance_subpixvar16x16 -#define vp8_variance_subpixvar16x16 vp8_sub_pixel_variance16x16_c -#endif -extern prototype_subpixvariance(vp8_variance_subpixvar16x16); - -#ifndef vp8_variance_halfpixvar16x16_h -#define vp8_variance_halfpixvar16x16_h vp8_variance_halfpixvar16x16_h_c -#endif -extern prototype_variance(vp8_variance_halfpixvar16x16_h); - -#ifndef vp8_variance_halfpixvar16x16_v -#define vp8_variance_halfpixvar16x16_v vp8_variance_halfpixvar16x16_v_c -#endif -extern prototype_variance(vp8_variance_halfpixvar16x16_v); - -#ifndef vp8_variance_halfpixvar16x16_hv -#define vp8_variance_halfpixvar16x16_hv vp8_variance_halfpixvar16x16_hv_c -#endif -extern prototype_variance(vp8_variance_halfpixvar16x16_hv); - -#ifndef vp8_variance_subpixmse16x16 -#define vp8_variance_subpixmse16x16 vp8_sub_pixel_mse16x16_c -#endif -extern prototype_subpixvariance(vp8_variance_subpixmse16x16); - -//-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=- - -#ifndef vp8_variance_getmbss -#define vp8_variance_getmbss vp8_get_mb_ss_c -#endif -extern prototype_getmbss(vp8_variance_getmbss); - -#ifndef vp8_variance_mse16x16 -#define vp8_variance_mse16x16 vp8_mse16x16_c -#endif -extern prototype_variance(vp8_variance_mse16x16); - -#ifndef vp8_variance_get4x4sse_cs -#define vp8_variance_get4x4sse_cs vp8_get4x4sse_cs_c -#endif -extern prototype_get16x16prederror(vp8_variance_get4x4sse_cs); - -#ifndef vp8_ssimpf_8x8 -#define vp8_ssimpf_8x8 vp8_ssim_parms_8x8_c -#endif -extern prototype_ssimpf(vp8_ssimpf_8x8) - -#ifndef vp8_ssimpf_16x16 -#define vp8_ssimpf_16x16 vp8_ssim_parms_16x16_c -#endif -extern prototype_ssimpf(vp8_ssimpf_16x16) - -typedef prototype_sad(*vp8_sad_fn_t); -typedef prototype_sad_multi_same_address(*vp8_sad_multi_fn_t); -typedef prototype_sad_multi_same_address_1(*vp8_sad_multi1_fn_t); -typedef prototype_sad_multi_dif_address(*vp8_sad_multi_d_fn_t); -typedef prototype_variance(*vp8_variance_fn_t); -typedef prototype_variance2(*vp8_variance2_fn_t); -typedef prototype_subpixvariance(*vp8_subpixvariance_fn_t); -typedef prototype_getmbss(*vp8_getmbss_fn_t); -typedef prototype_ssimpf(*vp8_ssimpf_fn_t); -typedef prototype_get16x16prederror(*vp8_get16x16prederror_fn_t); - -typedef struct -{ - vp8_sad_fn_t sad4x4; - vp8_sad_fn_t sad8x8; - vp8_sad_fn_t sad8x16; - vp8_sad_fn_t sad16x8; - vp8_sad_fn_t sad16x16; - - vp8_variance_fn_t var4x4; - vp8_variance_fn_t var8x8; - vp8_variance_fn_t var8x16; - vp8_variance_fn_t var16x8; - vp8_variance_fn_t var16x16; - - vp8_subpixvariance_fn_t subpixvar4x4; - vp8_subpixvariance_fn_t subpixvar8x8; - vp8_subpixvariance_fn_t subpixvar8x16; - vp8_subpixvariance_fn_t subpixvar16x8; - vp8_subpixvariance_fn_t subpixvar16x16; - vp8_variance_fn_t halfpixvar16x16_h; - vp8_variance_fn_t halfpixvar16x16_v; - vp8_variance_fn_t halfpixvar16x16_hv; - vp8_subpixvariance_fn_t subpixmse16x16; - - vp8_getmbss_fn_t getmbss; - vp8_variance_fn_t mse16x16; - - vp8_get16x16prederror_fn_t get4x4sse_cs; - - vp8_sad_multi_fn_t sad16x16x3; - vp8_sad_multi_fn_t sad16x8x3; - vp8_sad_multi_fn_t sad8x16x3; - vp8_sad_multi_fn_t sad8x8x3; - vp8_sad_multi_fn_t sad4x4x3; - - vp8_sad_multi1_fn_t sad16x16x8; - vp8_sad_multi1_fn_t sad16x8x8; - vp8_sad_multi1_fn_t sad8x16x8; - vp8_sad_multi1_fn_t sad8x8x8; - vp8_sad_multi1_fn_t sad4x4x8; - - vp8_sad_multi_d_fn_t sad16x16x4d; - vp8_sad_multi_d_fn_t sad16x8x4d; - vp8_sad_multi_d_fn_t sad8x16x4d; - vp8_sad_multi_d_fn_t sad8x8x4d; - vp8_sad_multi_d_fn_t sad4x4x4d; - -#if ARCH_X86 || ARCH_X86_64 - vp8_sad_fn_t copy32xn; -#endif - -#if CONFIG_INTERNAL_STATS - vp8_ssimpf_fn_t ssimpf_8x8; - vp8_ssimpf_fn_t ssimpf_16x16; -#endif - -} vp8_variance_rtcd_vtable_t; +typedef unsigned int (*vp8_get16x16prederror_fn_t) + ( + const unsigned char *src_ptr, + int source_stride, + const unsigned char *ref_ptr, + int ref_stride + ); typedef struct { @@ -411,16 +108,8 @@ typedef struct vp8_sad_multi1_fn_t sdx8f; vp8_sad_multi_d_fn_t sdx4df; #if ARCH_X86 || ARCH_X86_64 - vp8_sad_fn_t copymem; + vp8_copy32xn_fn_t copymem; #endif } vp8_variance_fn_ptr_t; -#if CONFIG_RUNTIME_CPU_DETECT -#define VARIANCE_INVOKE(ctx,fn) (ctx)->fn -#define SSIMPF_INVOKE(ctx,fn) (ctx)->ssimpf_##fn -#else -#define VARIANCE_INVOKE(ctx,fn) vp8_variance_##fn -#define SSIMPF_INVOKE(ctx,fn) vp8_ssimpf_##fn -#endif - #endif diff --git a/vp8/encoder/x86/variance_x86.h b/vp8/encoder/x86/variance_x86.h deleted file mode 100644 index 4b41b5436..000000000 --- a/vp8/encoder/x86/variance_x86.h +++ /dev/null @@ -1,332 +0,0 @@ -/* - * Copyright (c) 2010 The WebM project authors. All Rights Reserved. - * - * Use of this source code is governed by a BSD-style license - * that can be found in the LICENSE file in the root of the source - * tree. An additional intellectual property rights grant can be found - * in the file PATENTS. All contributing project authors may - * be found in the AUTHORS file in the root of the source tree. - */ - - -#ifndef VARIANCE_X86_H -#define VARIANCE_X86_H - - -/* Note: - * - * This platform is commonly built for runtime CPU detection. If you modify - * any of the function mappings present in this file, be sure to also update - * them in the function pointer initialization code - */ -#if HAVE_MMX -extern prototype_sad(vp8_sad4x4_mmx); -extern prototype_sad(vp8_sad8x8_mmx); -extern prototype_sad(vp8_sad8x16_mmx); -extern prototype_sad(vp8_sad16x8_mmx); -extern prototype_sad(vp8_sad16x16_mmx); -extern prototype_variance(vp8_variance4x4_mmx); -extern prototype_variance(vp8_variance8x8_mmx); -extern prototype_variance(vp8_variance8x16_mmx); -extern prototype_variance(vp8_variance16x8_mmx); -extern prototype_variance(vp8_variance16x16_mmx); -extern prototype_subpixvariance(vp8_sub_pixel_variance4x4_mmx); -extern prototype_subpixvariance(vp8_sub_pixel_variance8x8_mmx); -extern prototype_subpixvariance(vp8_sub_pixel_variance8x16_mmx); -extern prototype_subpixvariance(vp8_sub_pixel_variance16x8_mmx); -extern prototype_subpixvariance(vp8_sub_pixel_variance16x16_mmx); -extern prototype_variance(vp8_variance_halfpixvar16x16_h_mmx); -extern prototype_variance(vp8_variance_halfpixvar16x16_v_mmx); -extern prototype_variance(vp8_variance_halfpixvar16x16_hv_mmx); -extern prototype_subpixvariance(vp8_sub_pixel_mse16x16_mmx); -extern prototype_getmbss(vp8_get_mb_ss_mmx); -extern prototype_variance(vp8_mse16x16_mmx); -extern prototype_variance2(vp8_get8x8var_mmx); -extern prototype_get16x16prederror(vp8_get4x4sse_cs_mmx); - -#if !CONFIG_RUNTIME_CPU_DETECT -#undef vp8_variance_sad4x4 -#define vp8_variance_sad4x4 vp8_sad4x4_mmx - -#undef vp8_variance_sad8x8 -#define vp8_variance_sad8x8 vp8_sad8x8_mmx - -#undef vp8_variance_sad8x16 -#define vp8_variance_sad8x16 vp8_sad8x16_mmx - -#undef vp8_variance_sad16x8 -#define vp8_variance_sad16x8 vp8_sad16x8_mmx - -#undef vp8_variance_sad16x16 -#define vp8_variance_sad16x16 vp8_sad16x16_mmx - -#undef vp8_variance_var4x4 -#define vp8_variance_var4x4 vp8_variance4x4_mmx - -#undef vp8_variance_var8x8 -#define vp8_variance_var8x8 vp8_variance8x8_mmx - -#undef vp8_variance_var8x16 -#define vp8_variance_var8x16 vp8_variance8x16_mmx - -#undef vp8_variance_var16x8 -#define vp8_variance_var16x8 vp8_variance16x8_mmx - -#undef vp8_variance_var16x16 -#define vp8_variance_var16x16 vp8_variance16x16_mmx - -#undef vp8_variance_subpixvar4x4 -#define vp8_variance_subpixvar4x4 vp8_sub_pixel_variance4x4_mmx - -#undef vp8_variance_subpixvar8x8 -#define vp8_variance_subpixvar8x8 vp8_sub_pixel_variance8x8_mmx - -#undef vp8_variance_subpixvar8x16 -#define vp8_variance_subpixvar8x16 vp8_sub_pixel_variance8x16_mmx - -#undef vp8_variance_subpixvar16x8 -#define vp8_variance_subpixvar16x8 vp8_sub_pixel_variance16x8_mmx - -#undef vp8_variance_subpixvar16x16 -#define vp8_variance_subpixvar16x16 vp8_sub_pixel_variance16x16_mmx - -#undef vp8_variance_halfpixvar16x16_h -#define vp8_variance_halfpixvar16x16_h vp8_variance_halfpixvar16x16_h_mmx - -#undef vp8_variance_halfpixvar16x16_v -#define vp8_variance_halfpixvar16x16_v vp8_variance_halfpixvar16x16_v_mmx - -#undef vp8_variance_halfpixvar16x16_hv -#define vp8_variance_halfpixvar16x16_hv vp8_variance_halfpixvar16x16_hv_mmx - -#undef vp8_variance_subpixmse16x16 -#define vp8_variance_subpixmse16x16 vp8_sub_pixel_mse16x16_mmx - -#undef vp8_variance_getmbss -#define vp8_variance_getmbss vp8_get_mb_ss_mmx - -#undef vp8_variance_mse16x16 -#define vp8_variance_mse16x16 vp8_mse16x16_mmx - -#undef vp8_variance_get4x4sse_cs -#define vp8_variance_get4x4sse_cs vp8_get4x4sse_cs_mmx - -#endif -#endif - - -#if HAVE_SSE2 -extern prototype_sad(vp8_sad4x4_wmt); -extern prototype_sad(vp8_sad8x8_wmt); -extern prototype_sad(vp8_sad8x16_wmt); -extern prototype_sad(vp8_sad16x8_wmt); -extern prototype_sad(vp8_sad16x16_wmt); -extern prototype_sad(vp8_copy32xn_sse2); -extern prototype_variance(vp8_variance4x4_wmt); -extern prototype_variance(vp8_variance8x8_wmt); -extern prototype_variance(vp8_variance8x16_wmt); -extern prototype_variance(vp8_variance16x8_wmt); -extern prototype_variance(vp8_variance16x16_wmt); -extern prototype_subpixvariance(vp8_sub_pixel_variance4x4_wmt); -extern prototype_subpixvariance(vp8_sub_pixel_variance8x8_wmt); -extern prototype_subpixvariance(vp8_sub_pixel_variance8x16_wmt); -extern prototype_subpixvariance(vp8_sub_pixel_variance16x8_wmt); -extern prototype_subpixvariance(vp8_sub_pixel_variance16x16_wmt); -extern prototype_variance(vp8_variance_halfpixvar16x16_h_wmt); -extern prototype_variance(vp8_variance_halfpixvar16x16_v_wmt); -extern prototype_variance(vp8_variance_halfpixvar16x16_hv_wmt); -extern prototype_subpixvariance(vp8_sub_pixel_mse16x16_wmt); -extern prototype_getmbss(vp8_get_mb_ss_sse2); -extern prototype_variance(vp8_mse16x16_wmt); -extern prototype_variance2(vp8_get8x8var_sse2); -extern prototype_variance2(vp8_get16x16var_sse2); -extern prototype_ssimpf(vp8_ssim_parms_8x8_sse2) -extern prototype_ssimpf(vp8_ssim_parms_16x16_sse2) - -#if !CONFIG_RUNTIME_CPU_DETECT -#undef vp8_variance_sad4x4 -#define vp8_variance_sad4x4 vp8_sad4x4_wmt - -#undef vp8_variance_sad8x8 -#define vp8_variance_sad8x8 vp8_sad8x8_wmt - -#undef vp8_variance_sad8x16 -#define vp8_variance_sad8x16 vp8_sad8x16_wmt - -#undef vp8_variance_sad16x8 -#define vp8_variance_sad16x8 vp8_sad16x8_wmt - -#undef vp8_variance_sad16x16 -#define vp8_variance_sad16x16 vp8_sad16x16_wmt - -#undef vp8_variance_copy32xn -#define vp8_variance_copy32xn vp8_copy32xn_sse2 - -#undef vp8_variance_var4x4 -#define vp8_variance_var4x4 vp8_variance4x4_wmt - -#undef vp8_variance_var8x8 -#define vp8_variance_var8x8 vp8_variance8x8_wmt - -#undef vp8_variance_var8x16 -#define vp8_variance_var8x16 vp8_variance8x16_wmt - -#undef vp8_variance_var16x8 -#define vp8_variance_var16x8 vp8_variance16x8_wmt - -#undef vp8_variance_var16x16 -#define vp8_variance_var16x16 vp8_variance16x16_wmt - -#undef vp8_variance_subpixvar4x4 -#define vp8_variance_subpixvar4x4 vp8_sub_pixel_variance4x4_wmt - -#undef vp8_variance_subpixvar8x8 -#define vp8_variance_subpixvar8x8 vp8_sub_pixel_variance8x8_wmt - -#undef vp8_variance_subpixvar8x16 -#define vp8_variance_subpixvar8x16 vp8_sub_pixel_variance8x16_wmt - -#undef vp8_variance_subpixvar16x8 -#define vp8_variance_subpixvar16x8 vp8_sub_pixel_variance16x8_wmt - -#undef vp8_variance_subpixvar16x16 -#define vp8_variance_subpixvar16x16 vp8_sub_pixel_variance16x16_wmt - -#undef vp8_variance_halfpixvar16x16_h -#define vp8_variance_halfpixvar16x16_h vp8_variance_halfpixvar16x16_h_wmt - -#undef vp8_variance_halfpixvar16x16_v -#define vp8_variance_halfpixvar16x16_v vp8_variance_halfpixvar16x16_v_wmt - -#undef vp8_variance_halfpixvar16x16_hv -#define vp8_variance_halfpixvar16x16_hv vp8_variance_halfpixvar16x16_hv_wmt - -#undef vp8_variance_subpixmse16x16 -#define vp8_variance_subpixmse16x16 vp8_sub_pixel_mse16x16_wmt - -#undef vp8_variance_getmbss -#define vp8_variance_getmbss vp8_get_mb_ss_sse2 - -#undef vp8_variance_mse16x16 -#define vp8_variance_mse16x16 vp8_mse16x16_wmt - -#if ARCH_X86_64 -#undef vp8_ssimpf_8x8 -#define vp8_ssimpf_8x8 vp8_ssim_parms_8x8_sse2 - -#undef vp8_ssimpf_16x16 -#define vp8_ssimpf_16x16 vp8_ssim_parms_16x16_sse2 -#endif - -#endif -#endif - - -#if HAVE_SSE3 -extern prototype_sad(vp8_sad16x16_sse3); -extern prototype_sad(vp8_sad16x8_sse3); -extern prototype_sad_multi_same_address(vp8_sad16x16x3_sse3); -extern prototype_sad_multi_same_address(vp8_sad16x8x3_sse3); -extern prototype_sad_multi_same_address(vp8_sad8x16x3_sse3); -extern prototype_sad_multi_same_address(vp8_sad8x8x3_sse3); -extern prototype_sad_multi_same_address(vp8_sad4x4x3_sse3); - -extern prototype_sad_multi_dif_address(vp8_sad16x16x4d_sse3); -extern prototype_sad_multi_dif_address(vp8_sad16x8x4d_sse3); -extern prototype_sad_multi_dif_address(vp8_sad8x16x4d_sse3); -extern prototype_sad_multi_dif_address(vp8_sad8x8x4d_sse3); -extern prototype_sad_multi_dif_address(vp8_sad4x4x4d_sse3); -extern prototype_sad(vp8_copy32xn_sse3); - -#if !CONFIG_RUNTIME_CPU_DETECT - -#undef vp8_variance_sad16x16 -#define vp8_variance_sad16x16 vp8_sad16x16_sse3 - -#undef vp8_variance_sad16x16x3 -#define vp8_variance_sad16x16x3 vp8_sad16x16x3_sse3 - -#undef vp8_variance_sad16x8x3 -#define vp8_variance_sad16x8x3 vp8_sad16x8x3_sse3 - -#undef vp8_variance_sad8x16x3 -#define vp8_variance_sad8x16x3 vp8_sad8x16x3_sse3 - -#undef vp8_variance_sad8x8x3 -#define vp8_variance_sad8x8x3 vp8_sad8x8x3_sse3 - -#undef vp8_variance_sad4x4x3 -#define vp8_variance_sad4x4x3 vp8_sad4x4x3_sse3 - -#undef vp8_variance_sad16x16x4d -#define vp8_variance_sad16x16x4d vp8_sad16x16x4d_sse3 - -#undef vp8_variance_sad16x8x4d -#define vp8_variance_sad16x8x4d vp8_sad16x8x4d_sse3 - -#undef vp8_variance_sad8x16x4d -#define vp8_variance_sad8x16x4d vp8_sad8x16x4d_sse3 - -#undef vp8_variance_sad8x8x4d -#define vp8_variance_sad8x8x4d vp8_sad8x8x4d_sse3 - -#undef vp8_variance_sad4x4x4d -#define vp8_variance_sad4x4x4d vp8_sad4x4x4d_sse3 - -#undef vp8_variance_copy32xn -#define vp8_variance_copy32xn vp8_copy32xn_sse3 - -#endif -#endif - - -#if HAVE_SSSE3 -extern prototype_sad_multi_same_address(vp8_sad16x16x3_ssse3); -extern prototype_sad_multi_same_address(vp8_sad16x8x3_ssse3); -extern prototype_subpixvariance(vp8_sub_pixel_variance16x8_ssse3); -extern prototype_subpixvariance(vp8_sub_pixel_variance16x16_ssse3); - -#if !CONFIG_RUNTIME_CPU_DETECT -#undef vp8_variance_sad16x16x3 -#define vp8_variance_sad16x16x3 vp8_sad16x16x3_ssse3 - -#undef vp8_variance_sad16x8x3 -#define vp8_variance_sad16x8x3 vp8_sad16x8x3_ssse3 - -#undef vp8_variance_subpixvar16x8 -#define vp8_variance_subpixvar16x8 vp8_sub_pixel_variance16x8_ssse3 - -#undef vp8_variance_subpixvar16x16 -#define vp8_variance_subpixvar16x16 vp8_sub_pixel_variance16x16_ssse3 - -#endif -#endif - - -#if HAVE_SSE4_1 -extern prototype_sad_multi_same_address_1(vp8_sad16x16x8_sse4); -extern prototype_sad_multi_same_address_1(vp8_sad16x8x8_sse4); -extern prototype_sad_multi_same_address_1(vp8_sad8x16x8_sse4); -extern prototype_sad_multi_same_address_1(vp8_sad8x8x8_sse4); -extern prototype_sad_multi_same_address_1(vp8_sad4x4x8_sse4); - -#if !CONFIG_RUNTIME_CPU_DETECT -#undef vp8_variance_sad16x16x8 -#define vp8_variance_sad16x16x8 vp8_sad16x16x8_sse4 - -#undef vp8_variance_sad16x8x8 -#define vp8_variance_sad16x8x8 vp8_sad16x8x8_sse4 - -#undef vp8_variance_sad8x16x8 -#define vp8_variance_sad8x16x8 vp8_sad8x16x8_sse4 - -#undef vp8_variance_sad8x8x8 -#define vp8_variance_sad8x8x8 vp8_sad8x8x8_sse4 - -#undef vp8_variance_sad4x4x8 -#define vp8_variance_sad4x4x8 vp8_sad4x4x8_sse4 - -#endif -#endif - -#endif diff --git a/vp8/encoder/x86/x86_csystemdependent.c b/vp8/encoder/x86/x86_csystemdependent.c index 7f5208461..8fb794696 100644 --- a/vp8/encoder/x86/x86_csystemdependent.c +++ b/vp8/encoder/x86/x86_csystemdependent.c @@ -127,33 +127,6 @@ void vp8_arch_x86_encoder_init(VP8_COMP *cpi) #if HAVE_MMX if (flags & HAS_MMX) { - cpi->rtcd.variance.sad16x16 = vp8_sad16x16_mmx; - cpi->rtcd.variance.sad16x8 = vp8_sad16x8_mmx; - cpi->rtcd.variance.sad8x16 = vp8_sad8x16_mmx; - cpi->rtcd.variance.sad8x8 = vp8_sad8x8_mmx; - cpi->rtcd.variance.sad4x4 = vp8_sad4x4_mmx; - - cpi->rtcd.variance.var4x4 = vp8_variance4x4_mmx; - cpi->rtcd.variance.var8x8 = vp8_variance8x8_mmx; - cpi->rtcd.variance.var8x16 = vp8_variance8x16_mmx; - cpi->rtcd.variance.var16x8 = vp8_variance16x8_mmx; - cpi->rtcd.variance.var16x16 = vp8_variance16x16_mmx; - - cpi->rtcd.variance.subpixvar4x4 = vp8_sub_pixel_variance4x4_mmx; - cpi->rtcd.variance.subpixvar8x8 = vp8_sub_pixel_variance8x8_mmx; - cpi->rtcd.variance.subpixvar8x16 = vp8_sub_pixel_variance8x16_mmx; - cpi->rtcd.variance.subpixvar16x8 = vp8_sub_pixel_variance16x8_mmx; - cpi->rtcd.variance.subpixvar16x16 = vp8_sub_pixel_variance16x16_mmx; - cpi->rtcd.variance.halfpixvar16x16_h = vp8_variance_halfpixvar16x16_h_mmx; - cpi->rtcd.variance.halfpixvar16x16_v = vp8_variance_halfpixvar16x16_v_mmx; - cpi->rtcd.variance.halfpixvar16x16_hv = vp8_variance_halfpixvar16x16_hv_mmx; - cpi->rtcd.variance.subpixmse16x16 = vp8_sub_pixel_mse16x16_mmx; - - cpi->rtcd.variance.mse16x16 = vp8_mse16x16_mmx; - cpi->rtcd.variance.getmbss = vp8_get_mb_ss_mmx; - - cpi->rtcd.variance.get4x4sse_cs = vp8_get4x4sse_cs_mmx; - cpi->rtcd.fdct.short4x4 = vp8_short_fdct4x4_mmx; cpi->rtcd.fdct.short8x4 = vp8_short_fdct8x4_mmx; cpi->rtcd.fdct.fast4x4 = vp8_short_fdct4x4_mmx; @@ -175,34 +148,6 @@ void vp8_arch_x86_encoder_init(VP8_COMP *cpi) #if HAVE_SSE2 if (flags & HAS_SSE2) { - cpi->rtcd.variance.sad16x16 = vp8_sad16x16_wmt; - cpi->rtcd.variance.sad16x8 = vp8_sad16x8_wmt; - cpi->rtcd.variance.sad8x16 = vp8_sad8x16_wmt; - cpi->rtcd.variance.sad8x8 = vp8_sad8x8_wmt; - cpi->rtcd.variance.sad4x4 = vp8_sad4x4_wmt; - cpi->rtcd.variance.copy32xn = vp8_copy32xn_sse2; - - cpi->rtcd.variance.var4x4 = vp8_variance4x4_wmt; - cpi->rtcd.variance.var8x8 = vp8_variance8x8_wmt; - cpi->rtcd.variance.var8x16 = vp8_variance8x16_wmt; - cpi->rtcd.variance.var16x8 = vp8_variance16x8_wmt; - cpi->rtcd.variance.var16x16 = vp8_variance16x16_wmt; - - cpi->rtcd.variance.subpixvar4x4 = vp8_sub_pixel_variance4x4_wmt; - cpi->rtcd.variance.subpixvar8x8 = vp8_sub_pixel_variance8x8_wmt; - cpi->rtcd.variance.subpixvar8x16 = vp8_sub_pixel_variance8x16_wmt; - cpi->rtcd.variance.subpixvar16x8 = vp8_sub_pixel_variance16x8_wmt; - cpi->rtcd.variance.subpixvar16x16 = vp8_sub_pixel_variance16x16_wmt; - cpi->rtcd.variance.halfpixvar16x16_h = vp8_variance_halfpixvar16x16_h_wmt; - cpi->rtcd.variance.halfpixvar16x16_v = vp8_variance_halfpixvar16x16_v_wmt; - cpi->rtcd.variance.halfpixvar16x16_hv = vp8_variance_halfpixvar16x16_hv_wmt; - cpi->rtcd.variance.subpixmse16x16 = vp8_sub_pixel_mse16x16_wmt; - - cpi->rtcd.variance.mse16x16 = vp8_mse16x16_wmt; - cpi->rtcd.variance.getmbss = vp8_get_mb_ss_sse2; - - /* cpi->rtcd.variance.get4x4sse_cs not implemented for wmt */; - cpi->rtcd.fdct.short4x4 = vp8_short_fdct4x4_sse2; cpi->rtcd.fdct.short8x4 = vp8_short_fdct8x4_sse2; cpi->rtcd.fdct.fast4x4 = vp8_short_fdct4x4_sse2; @@ -224,31 +169,13 @@ void vp8_arch_x86_encoder_init(VP8_COMP *cpi) cpi->rtcd.temporal.apply = vp8_temporal_filter_apply_sse2; #endif -#if CONFIG_INTERNAL_STATS -#if ARCH_X86_64 - cpi->rtcd.variance.ssimpf_8x8 = vp8_ssim_parms_8x8_sse2; - cpi->rtcd.variance.ssimpf_16x16 = vp8_ssim_parms_16x16_sse2; -#endif -#endif } #endif #if HAVE_SSE3 if (flags & HAS_SSE3) { - cpi->rtcd.variance.sad16x16 = vp8_sad16x16_sse3; - cpi->rtcd.variance.sad16x16x3 = vp8_sad16x16x3_sse3; - cpi->rtcd.variance.sad16x8x3 = vp8_sad16x8x3_sse3; - cpi->rtcd.variance.sad8x16x3 = vp8_sad8x16x3_sse3; - cpi->rtcd.variance.sad8x8x3 = vp8_sad8x8x3_sse3; - cpi->rtcd.variance.sad4x4x3 = vp8_sad4x4x3_sse3; cpi->rtcd.search.full_search = vp8_full_search_sadx3; - cpi->rtcd.variance.sad16x16x4d = vp8_sad16x16x4d_sse3; - cpi->rtcd.variance.sad16x8x4d = vp8_sad16x8x4d_sse3; - cpi->rtcd.variance.sad8x16x4d = vp8_sad8x16x4d_sse3; - cpi->rtcd.variance.sad8x8x4d = vp8_sad8x8x4d_sse3; - cpi->rtcd.variance.sad4x4x4d = vp8_sad4x4x4d_sse3; - cpi->rtcd.variance.copy32xn = vp8_copy32xn_sse3; cpi->rtcd.search.diamond_search = vp8_diamond_search_sadx4; cpi->rtcd.search.refining_search = vp8_refining_search_sadx4; } @@ -257,12 +184,6 @@ void vp8_arch_x86_encoder_init(VP8_COMP *cpi) #if HAVE_SSSE3 if (flags & HAS_SSSE3) { - cpi->rtcd.variance.sad16x16x3 = vp8_sad16x16x3_ssse3; - cpi->rtcd.variance.sad16x8x3 = vp8_sad16x8x3_ssse3; - - cpi->rtcd.variance.subpixvar16x8 = vp8_sub_pixel_variance16x8_ssse3; - cpi->rtcd.variance.subpixvar16x16 = vp8_sub_pixel_variance16x16_ssse3; - cpi->rtcd.quantize.fastquantb = vp8_fast_quantize_b_ssse3; } #endif @@ -272,11 +193,6 @@ void vp8_arch_x86_encoder_init(VP8_COMP *cpi) #if HAVE_SSE4_1 if (flags & HAS_SSE4_1) { - cpi->rtcd.variance.sad16x16x8 = vp8_sad16x16x8_sse4; - cpi->rtcd.variance.sad16x8x8 = vp8_sad16x8x8_sse4; - cpi->rtcd.variance.sad8x16x8 = vp8_sad8x16x8_sse4; - cpi->rtcd.variance.sad8x8x8 = vp8_sad8x8x8_sse4; - cpi->rtcd.variance.sad4x4x8 = vp8_sad4x4x8_sse4; cpi->rtcd.search.full_search = vp8_full_search_sadx8; cpi->rtcd.quantize.quantb = vp8_regular_quantize_b_sse4; diff --git a/vp8/vp8cx.mk b/vp8/vp8cx.mk index 2d99981f5..f56328960 100644 --- a/vp8/vp8cx.mk +++ b/vp8/vp8cx.mk @@ -97,7 +97,6 @@ endif VP8_CX_SRCS-$(ARCH_X86)$(ARCH_X86_64) += encoder/x86/encodemb_x86.h VP8_CX_SRCS-$(ARCH_X86)$(ARCH_X86_64) += encoder/x86/dct_x86.h VP8_CX_SRCS-$(ARCH_X86)$(ARCH_X86_64) += encoder/x86/mcomp_x86.h -VP8_CX_SRCS-$(ARCH_X86)$(ARCH_X86_64) += encoder/x86/variance_x86.h VP8_CX_SRCS-$(ARCH_X86)$(ARCH_X86_64) += encoder/x86/quantize_x86.h VP8_CX_SRCS-$(ARCH_X86)$(ARCH_X86_64) += encoder/x86/temporal_filter_x86.h VP8_CX_SRCS-$(ARCH_X86)$(ARCH_X86_64) += encoder/x86/x86_csystemdependent.c diff --git a/vp8/vp8cx_arm.mk b/vp8/vp8cx_arm.mk index 7429e16d7..680842239 100644 --- a/vp8/vp8cx_arm.mk +++ b/vp8/vp8cx_arm.mk @@ -21,7 +21,6 @@ VP8_CX_SRCS-$(ARCH_ARM) += encoder/arm/encodemb_arm.h VP8_CX_SRCS-$(ARCH_ARM) += encoder/arm/quantize_arm.c VP8_CX_SRCS-$(ARCH_ARM) += encoder/arm/quantize_arm.h VP8_CX_SRCS-$(ARCH_ARM) += encoder/arm/variance_arm.c -VP8_CX_SRCS-$(ARCH_ARM) += encoder/arm/variance_arm.h #File list for edsp # encoder |