18e90d744e
This is the initial patch for supporting 1/8th pel motion. Currently if we configure with enable-high-precision-mv, all motion vectors would default to 1/8 pel. Encode and decode syncs fine with the current code. In the next phase the code will be refactored so that we can choose the 1/8 pel mode adaptively at a frame/segment/mb level. Derf results: http://www.corp.google.com/~debargha/vp8_results/enhinterp_hpmv.html (about 0.83% better than 8-tap interpoaltion) Patch 3: Rebased. Also adding 1/16th pel interpolation for U and V Patch 4: HD results. http://www.corp.google.com/~debargha/vp8_results/enhinterp_hd_hpmv.html Seems impressive (unless I am doing something wrong). Patch 5: Added mmx/sse for bilateral filtering, as well as enforced use of c-versions of subpel filters with 8-taps and 1/16th pel; Also redesigned the 8-tap filters to reduce the cut-off in order to introduce a denoising effect. There is a new configure option sixteenth-subpel-uv which will use 1/16 th pel interpolation for uv, if the motion vectors have 1/8 pel accuracy. With the fixes the results are promising on the derf set. The enhanced interpolation option with 8-taps alone gives 3% improvement over thei derf set: http://www.corp.google.com/~debargha/vp8_results/enhinterpn.html Results on high precision mv and on the hd set are to follow. Patch 6: Adding a missing condition for CONFIG_SIXTEENTH_SUBPEL_UV in vp8/common/x86/x86_systemdependent.c Patch 7: Cleaning up various debug messages. Patch 8: Merge conflict Change-Id: I5b1d844457aefd7414a9e4e0e06c6ed38fd8cc04
143 lines
5.6 KiB
C
143 lines
5.6 KiB
C
/*
|
|
* 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.
|
|
*/
|
|
|
|
|
|
#include "vpx_config.h"
|
|
#include "vpx_ports/x86.h"
|
|
#include "vp8/common/g_common.h"
|
|
#include "vp8/common/subpixel.h"
|
|
#include "vp8/common/loopfilter.h"
|
|
#include "vp8/common/recon.h"
|
|
#include "vp8/common/idct.h"
|
|
#include "vp8/common/pragmas.h"
|
|
#include "vp8/common/onyxc_int.h"
|
|
|
|
void vp8_arch_x86_common_init(VP8_COMMON *ctx)
|
|
{
|
|
#if CONFIG_RUNTIME_CPU_DETECT
|
|
VP8_COMMON_RTCD *rtcd = &ctx->rtcd;
|
|
int flags = x86_simd_caps();
|
|
|
|
/* Note:
|
|
*
|
|
* This platform can be built without runtime CPU detection as well. If
|
|
* you modify any of the function mappings present in this file, be sure
|
|
* to also update them in static mapings (<arch>/filename_<arch>.h)
|
|
*/
|
|
|
|
/* Override default functions with fastest ones for this CPU. */
|
|
#if HAVE_MMX
|
|
|
|
if (flags & HAS_MMX)
|
|
{
|
|
rtcd->idct.idct1 = vp8_short_idct4x4llm_1_mmx;
|
|
rtcd->idct.idct16 = vp8_short_idct4x4llm_mmx;
|
|
rtcd->idct.idct1_scalar_add = vp8_dc_only_idct_add_mmx;
|
|
rtcd->idct.iwalsh16 = vp8_short_inv_walsh4x4_mmx;
|
|
rtcd->idct.iwalsh1 = vp8_short_inv_walsh4x4_1_mmx;
|
|
|
|
rtcd->recon.recon = vp8_recon_b_mmx;
|
|
rtcd->recon.copy8x8 = vp8_copy_mem8x8_mmx;
|
|
rtcd->recon.copy8x4 = vp8_copy_mem8x4_mmx;
|
|
rtcd->recon.copy16x16 = vp8_copy_mem16x16_mmx;
|
|
|
|
#if CONFIG_ENHANCED_INTERP == 0 && CONFIG_HIGH_PRECISION_MV == 0 && CONFIG_SIXTEENTH_SUBPEL_UV == 0
|
|
rtcd->subpix.sixtap16x16 = vp8_sixtap_predict16x16_mmx;
|
|
rtcd->subpix.sixtap8x8 = vp8_sixtap_predict8x8_mmx;
|
|
rtcd->subpix.sixtap8x4 = vp8_sixtap_predict8x4_mmx;
|
|
rtcd->subpix.sixtap4x4 = vp8_sixtap_predict4x4_mmx;
|
|
#endif
|
|
rtcd->subpix.bilinear16x16 = vp8_bilinear_predict16x16_mmx;
|
|
rtcd->subpix.bilinear8x8 = vp8_bilinear_predict8x8_mmx;
|
|
rtcd->subpix.bilinear8x4 = vp8_bilinear_predict8x4_mmx;
|
|
rtcd->subpix.bilinear4x4 = vp8_bilinear_predict4x4_mmx;
|
|
|
|
rtcd->loopfilter.normal_mb_v = vp8_loop_filter_mbv_mmx;
|
|
rtcd->loopfilter.normal_b_v = vp8_loop_filter_bv_mmx;
|
|
rtcd->loopfilter.normal_mb_h = vp8_loop_filter_mbh_mmx;
|
|
rtcd->loopfilter.normal_b_h = vp8_loop_filter_bh_mmx;
|
|
rtcd->loopfilter.simple_mb_v = vp8_loop_filter_simple_vertical_edge_mmx;
|
|
rtcd->loopfilter.simple_b_v = vp8_loop_filter_bvs_mmx;
|
|
rtcd->loopfilter.simple_mb_h = vp8_loop_filter_simple_horizontal_edge_mmx;
|
|
rtcd->loopfilter.simple_b_h = vp8_loop_filter_bhs_mmx;
|
|
|
|
#if CONFIG_POSTPROC
|
|
rtcd->postproc.down = vp8_mbpost_proc_down_mmx;
|
|
/*rtcd->postproc.across = vp8_mbpost_proc_across_ip_c;*/
|
|
rtcd->postproc.downacross = vp8_post_proc_down_and_across_mmx;
|
|
rtcd->postproc.addnoise = vp8_plane_add_noise_mmx;
|
|
#endif
|
|
}
|
|
|
|
#endif
|
|
#if HAVE_SSE2
|
|
|
|
if (flags & HAS_SSE2)
|
|
{
|
|
rtcd->recon.recon2 = vp8_recon2b_sse2;
|
|
rtcd->recon.recon4 = vp8_recon4b_sse2;
|
|
rtcd->recon.copy16x16 = vp8_copy_mem16x16_sse2;
|
|
rtcd->recon.build_intra_predictors_mbuv =
|
|
vp8_build_intra_predictors_mbuv_sse2;
|
|
rtcd->recon.build_intra_predictors_mbuv_s =
|
|
vp8_build_intra_predictors_mbuv_s_sse2;
|
|
|
|
rtcd->idct.iwalsh16 = vp8_short_inv_walsh4x4_sse2;
|
|
|
|
#if CONFIG_ENHANCED_INTERP == 0 && CONFIG_HIGH_PRECISION_MV == 0 && CONFIG_SIXTEENTH_SUBPEL_UV == 0
|
|
rtcd->subpix.sixtap16x16 = vp8_sixtap_predict16x16_sse2;
|
|
rtcd->subpix.sixtap8x8 = vp8_sixtap_predict8x8_sse2;
|
|
rtcd->subpix.sixtap8x4 = vp8_sixtap_predict8x4_sse2;
|
|
#endif
|
|
rtcd->subpix.bilinear16x16 = vp8_bilinear_predict16x16_sse2;
|
|
rtcd->subpix.bilinear8x8 = vp8_bilinear_predict8x8_sse2;
|
|
|
|
rtcd->loopfilter.normal_mb_v = vp8_loop_filter_mbv_sse2;
|
|
rtcd->loopfilter.normal_b_v = vp8_loop_filter_bv_sse2;
|
|
rtcd->loopfilter.normal_mb_h = vp8_loop_filter_mbh_sse2;
|
|
rtcd->loopfilter.normal_b_h = vp8_loop_filter_bh_sse2;
|
|
rtcd->loopfilter.simple_mb_v = vp8_loop_filter_simple_vertical_edge_sse2;
|
|
rtcd->loopfilter.simple_b_v = vp8_loop_filter_bvs_sse2;
|
|
rtcd->loopfilter.simple_mb_h = vp8_loop_filter_simple_horizontal_edge_sse2;
|
|
rtcd->loopfilter.simple_b_h = vp8_loop_filter_bhs_sse2;
|
|
|
|
#if CONFIG_POSTPROC
|
|
rtcd->postproc.down = vp8_mbpost_proc_down_xmm;
|
|
rtcd->postproc.across = vp8_mbpost_proc_across_ip_xmm;
|
|
rtcd->postproc.downacross = vp8_post_proc_down_and_across_xmm;
|
|
rtcd->postproc.addnoise = vp8_plane_add_noise_wmt;
|
|
#endif
|
|
}
|
|
|
|
#endif
|
|
|
|
#if HAVE_SSSE3
|
|
|
|
if (flags & HAS_SSSE3)
|
|
{
|
|
#if CONFIG_ENHANCED_INTERP == 0 && CONFIG_HIGH_PRECISION_MV == 0
|
|
rtcd->subpix.sixtap16x16 = vp8_sixtap_predict16x16_ssse3;
|
|
rtcd->subpix.sixtap8x8 = vp8_sixtap_predict8x8_ssse3;
|
|
rtcd->subpix.sixtap8x4 = vp8_sixtap_predict8x4_ssse3;
|
|
rtcd->subpix.sixtap4x4 = vp8_sixtap_predict4x4_ssse3;
|
|
rtcd->subpix.bilinear16x16 = vp8_bilinear_predict16x16_ssse3;
|
|
rtcd->subpix.bilinear8x8 = vp8_bilinear_predict8x8_ssse3;
|
|
#endif
|
|
|
|
rtcd->recon.build_intra_predictors_mbuv =
|
|
vp8_build_intra_predictors_mbuv_ssse3;
|
|
rtcd->recon.build_intra_predictors_mbuv_s =
|
|
vp8_build_intra_predictors_mbuv_s_ssse3;
|
|
}
|
|
#endif
|
|
|
|
#endif
|
|
}
|