Chromium Code Reviews
chromiumcodereview-hr@appspot.gserviceaccount.com (chromiumcodereview-hr) | Please choose your nickname with Settings | Help | Chromium Project | Gerrit Changes | Sign out
(764)

Issues Search
    My Issues | Starred     Open | Closed | All

Unified Diff: libvpx/source/libvpx/vp8/encoder/x86/x86_csystemdependent.c

Issue 7624054: Revert r97185 "Update libvpx snapshot to v0.9.7-p1 (Cayuga)." (Closed) Base URL: svn://chrome-svn/chrome/trunk/deps/third_party
Patch Set: Created 9 years, 4 months ago
Use n/p to move between diff chunks; N/P to move between comments. Draft comments are only viewable by you.
Jump to:
View side-by-side diff with in-line comments
Download patch
« no previous file with comments | « libvpx/source/libvpx/vp8/encoder/x86/variance_x86.h ('k') | libvpx/source/libvpx/vp8/vp8_common.mk » ('j') | no next file with comments »
Expand Comments ('e') | Collapse Comments ('c') | Show Comments Hide Comments ('s')
Index: libvpx/source/libvpx/vp8/encoder/x86/x86_csystemdependent.c
diff --git a/libvpx/source/libvpx/vp8/encoder/x86/x86_csystemdependent.c b/libvpx/source/libvpx/vp8/encoder/x86/x86_csystemdependent.c
index badb9f04451b040f08713ca76d0263af16e82bda..2b6bd98eb436110a1bc306040b97b183d92a5057 100644
--- a/libvpx/source/libvpx/vp8/encoder/x86/x86_csystemdependent.c
+++ b/libvpx/source/libvpx/vp8/encoder/x86/x86_csystemdependent.c
@@ -16,7 +16,7 @@
#if HAVE_MMX
-void vp8_short_fdct8x4_mmx(short *input, short *output, int pitch)
+static void short_fdct8x4_mmx(short *input, short *output, int pitch)
{
vp8_short_fdct4x4_mmx(input, output, pitch);
vp8_short_fdct4x4_mmx(input + 4, output + 16, pitch);
@@ -26,7 +26,7 @@ int vp8_fast_quantize_b_impl_mmx(short *coeff_ptr, short *zbin_ptr,
short *qcoeff_ptr, short *dequant_ptr,
short *scan_mask, short *round_ptr,
short *quant_ptr, short *dqcoeff_ptr);
-void vp8_fast_quantize_b_mmx(BLOCK *b, BLOCKD *d)
+static void fast_quantize_b_mmx(BLOCK *b, BLOCKD *d)
{
short *scan_mask = vp8_default_zig_zag_mask;//d->scan_order_mask_ptr;
short *coeff_ptr = b->coeff;
@@ -51,7 +51,7 @@ void vp8_fast_quantize_b_mmx(BLOCK *b, BLOCKD *d)
}
int vp8_mbblock_error_mmx_impl(short *coeff_ptr, short *dcoef_ptr, int dc);
-int vp8_mbblock_error_mmx(MACROBLOCK *mb, int dc)
+static int mbblock_error_mmx(MACROBLOCK *mb, int dc)
{
short *coeff_ptr = mb->block[0].coeff;
short *dcoef_ptr = mb->e_mbd.block[0].dqcoeff;
@@ -59,7 +59,7 @@ int vp8_mbblock_error_mmx(MACROBLOCK *mb, int dc)
}
int vp8_mbuverror_mmx_impl(short *s_ptr, short *d_ptr);
-int vp8_mbuverror_mmx(MACROBLOCK *mb)
+static int mbuverror_mmx(MACROBLOCK *mb)
{
short *s_ptr = &mb->coeff[256];
short *d_ptr = &mb->e_mbd.dqcoeff[256];
@@ -69,7 +69,7 @@ int vp8_mbuverror_mmx(MACROBLOCK *mb)
void vp8_subtract_b_mmx_impl(unsigned char *z, int src_stride,
short *diff, unsigned char *predictor,
int pitch);
-void vp8_subtract_b_mmx(BLOCK *be, BLOCKD *bd, int pitch)
+static void subtract_b_mmx(BLOCK *be, BLOCKD *bd, int pitch)
{
unsigned char *z = *(be->base_src) + be->src;
unsigned int src_stride = be->src_stride;
@@ -81,8 +81,33 @@ void vp8_subtract_b_mmx(BLOCK *be, BLOCKD *bd, int pitch)
#endif
#if HAVE_SSE2
+int vp8_fast_quantize_b_impl_sse2(short *coeff_ptr,
+ short *qcoeff_ptr, short *dequant_ptr,
+ const short *inv_scan_order, short *round_ptr,
+ short *quant_ptr, short *dqcoeff_ptr);
+static void fast_quantize_b_sse2(BLOCK *b, BLOCKD *d)
+{
+ short *scan_mask = vp8_default_zig_zag_mask;//d->scan_order_mask_ptr;
+ short *coeff_ptr = b->coeff;
+ short *round_ptr = b->round;
+ short *quant_ptr = b->quant_fast;
+ short *qcoeff_ptr = d->qcoeff;
+ short *dqcoeff_ptr = d->dqcoeff;
+ short *dequant_ptr = d->dequant;
+
+ d->eob = vp8_fast_quantize_b_impl_sse2(
+ coeff_ptr,
+ qcoeff_ptr,
+ dequant_ptr,
+ vp8_default_inv_zig_zag,
+ round_ptr,
+ quant_ptr,
+ dqcoeff_ptr
+ );
+}
+
int vp8_mbblock_error_xmm_impl(short *coeff_ptr, short *dcoef_ptr, int dc);
-int vp8_mbblock_error_xmm(MACROBLOCK *mb, int dc)
+static int mbblock_error_xmm(MACROBLOCK *mb, int dc)
{
short *coeff_ptr = mb->block[0].coeff;
short *dcoef_ptr = mb->e_mbd.block[0].dqcoeff;
@@ -90,7 +115,7 @@ int vp8_mbblock_error_xmm(MACROBLOCK *mb, int dc)
}
int vp8_mbuverror_xmm_impl(short *s_ptr, short *d_ptr);
-int vp8_mbuverror_xmm(MACROBLOCK *mb)
+static int mbuverror_xmm(MACROBLOCK *mb)
{
short *s_ptr = &mb->coeff[256];
short *d_ptr = &mb->e_mbd.dqcoeff[256];
@@ -100,7 +125,7 @@ int vp8_mbuverror_xmm(MACROBLOCK *mb)
void vp8_subtract_b_sse2_impl(unsigned char *z, int src_stride,
short *diff, unsigned char *predictor,
int pitch);
-void vp8_subtract_b_sse2(BLOCK *be, BLOCKD *bd, int pitch)
+static void subtract_b_sse2(BLOCK *be, BLOCKD *bd, int pitch)
{
unsigned char *z = *(be->base_src) + be->src;
unsigned int src_stride = be->src_stride;
@@ -112,7 +137,22 @@ void vp8_subtract_b_sse2(BLOCK *be, BLOCKD *bd, int pitch)
#endif
#if HAVE_SSSE3
-#if CONFIG_INTERNAL_STATS
+int vp8_fast_quantize_b_impl_ssse3(short *coeff_ptr,
+ short *qcoeff_ptr, short *dequant_ptr,
+ short *round_ptr,
+ short *quant_ptr, short *dqcoeff_ptr);
+static void fast_quantize_b_ssse3(BLOCK *b, BLOCKD *d)
+{
+ d->eob = vp8_fast_quantize_b_impl_ssse3(
+ b->coeff,
+ d->qcoeff,
+ d->dequant,
+ b->round,
+ b->quant_fast,
+ d->dqcoeff
+ );
+}
+#if CONFIG_PSNR
#if ARCH_X86_64
typedef void ssimpf
(
@@ -138,6 +178,12 @@ void vp8_arch_x86_encoder_init(VP8_COMP *cpi)
{
#if CONFIG_RUNTIME_CPU_DETECT
int flags = x86_simd_caps();
+ int mmx_enabled = flags & HAS_MMX;
+ int xmm_enabled = flags & HAS_SSE;
+ int wmt_enabled = flags & HAS_SSE2;
+ int SSE3Enabled = flags & HAS_SSE3;
+ int SSSE3Enabled = flags & HAS_SSSE3;
+ int SSE4_1Enabled = flags & HAS_SSE4_1;
/* Note:
*
@@ -148,7 +194,7 @@ void vp8_arch_x86_encoder_init(VP8_COMP *cpi)
/* Override default functions with fastest ones for this CPU. */
#if HAVE_MMX
- if (flags & HAS_MMX)
+ if (mmx_enabled)
{
cpi->rtcd.variance.sad16x16 = vp8_sad16x16_mmx;
cpi->rtcd.variance.sad16x8 = vp8_sad16x8_mmx;
@@ -175,35 +221,37 @@ void vp8_arch_x86_encoder_init(VP8_COMP *cpi)
cpi->rtcd.variance.mse16x16 = vp8_mse16x16_mmx;
cpi->rtcd.variance.getmbss = vp8_get_mb_ss_mmx;
+ cpi->rtcd.variance.get16x16prederror = vp8_get16x16pred_error_mmx;
+ cpi->rtcd.variance.get8x8var = vp8_get8x8var_mmx;
+ cpi->rtcd.variance.get16x16var = vp8_get16x16var_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.short8x4 = short_fdct8x4_mmx;
cpi->rtcd.fdct.fast4x4 = vp8_short_fdct4x4_mmx;
- cpi->rtcd.fdct.fast8x4 = vp8_short_fdct8x4_mmx;
+ cpi->rtcd.fdct.fast8x4 = short_fdct8x4_mmx;
cpi->rtcd.fdct.walsh_short4x4 = vp8_short_walsh4x4_c;
cpi->rtcd.encodemb.berr = vp8_block_error_mmx;
- cpi->rtcd.encodemb.mberr = vp8_mbblock_error_mmx;
- cpi->rtcd.encodemb.mbuverr = vp8_mbuverror_mmx;
- cpi->rtcd.encodemb.subb = vp8_subtract_b_mmx;
+ cpi->rtcd.encodemb.mberr = mbblock_error_mmx;
+ cpi->rtcd.encodemb.mbuverr = mbuverror_mmx;
+ cpi->rtcd.encodemb.subb = subtract_b_mmx;
cpi->rtcd.encodemb.submby = vp8_subtract_mby_mmx;
cpi->rtcd.encodemb.submbuv = vp8_subtract_mbuv_mmx;
- /*cpi->rtcd.quantize.fastquantb = vp8_fast_quantize_b_mmx;*/
+ /*cpi->rtcd.quantize.fastquantb = fast_quantize_b_mmx;*/
}
#endif
#if HAVE_SSE2
- if (flags & HAS_SSE2)
+ if (wmt_enabled)
{
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;
@@ -224,6 +272,11 @@ void vp8_arch_x86_encoder_init(VP8_COMP *cpi)
cpi->rtcd.variance.mse16x16 = vp8_mse16x16_wmt;
cpi->rtcd.variance.getmbss = vp8_get_mb_ss_sse2;
+ cpi->rtcd.variance.get16x16prederror = vp8_get16x16pred_error_sse2;
+ cpi->rtcd.variance.get8x8var = vp8_get8x8var_sse2;
+ cpi->rtcd.variance.get16x16var = vp8_get16x16var_sse2;
+
+
/* cpi->rtcd.variance.get4x4sse_cs not implemented for wmt */;
cpi->rtcd.fdct.short4x4 = vp8_short_fdct4x4_sse2;
@@ -234,14 +287,14 @@ void vp8_arch_x86_encoder_init(VP8_COMP *cpi)
cpi->rtcd.fdct.walsh_short4x4 = vp8_short_walsh4x4_sse2 ;
cpi->rtcd.encodemb.berr = vp8_block_error_xmm;
- cpi->rtcd.encodemb.mberr = vp8_mbblock_error_xmm;
- cpi->rtcd.encodemb.mbuverr = vp8_mbuverror_xmm;
- cpi->rtcd.encodemb.subb = vp8_subtract_b_sse2;
+ cpi->rtcd.encodemb.mberr = mbblock_error_xmm;
+ cpi->rtcd.encodemb.mbuverr = mbuverror_xmm;
+ cpi->rtcd.encodemb.subb = subtract_b_sse2;
cpi->rtcd.encodemb.submby = vp8_subtract_mby_sse2;
cpi->rtcd.encodemb.submbuv = vp8_subtract_mbuv_sse2;
cpi->rtcd.quantize.quantb = vp8_regular_quantize_b_sse2;
- cpi->rtcd.quantize.fastquantb = vp8_fast_quantize_b_sse2;
+ cpi->rtcd.quantize.fastquantb = fast_quantize_b_sse2;
#if !(CONFIG_REALTIME_ONLY)
cpi->rtcd.temporal.apply = vp8_temporal_filter_apply_sse2;
@@ -250,7 +303,7 @@ void vp8_arch_x86_encoder_init(VP8_COMP *cpi)
#endif
#if HAVE_SSE3
- if (flags & HAS_SSE3)
+ if (SSE3Enabled)
{
cpi->rtcd.variance.sad16x16 = vp8_sad16x16_sse3;
cpi->rtcd.variance.sad16x16x3 = vp8_sad16x16x3_sse3;
@@ -258,20 +311,20 @@ void vp8_arch_x86_encoder_init(VP8_COMP *cpi)
cpi->rtcd.variance.sad8x16x3 = vp8_sad8x16x3_sse3;
cpi->rtcd.variance.sad8x8x3 = vp8_sad8x8x3_sse3;
cpi->rtcd.variance.sad4x4x3 = vp8_sad4x4x3_sse3;
+#if !(CONFIG_REALTIME_ONLY)
cpi->rtcd.search.full_search = vp8_full_search_sadx3;
+#endif
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;
}
#endif
#if HAVE_SSSE3
- if (flags & HAS_SSSE3)
+ if (SSSE3Enabled)
{
cpi->rtcd.variance.sad16x16x3 = vp8_sad16x16x3_ssse3;
cpi->rtcd.variance.sad16x8x3 = vp8_sad16x8x3_ssse3;
@@ -279,9 +332,9 @@ void vp8_arch_x86_encoder_init(VP8_COMP *cpi)
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;
+ cpi->rtcd.quantize.fastquantb = fast_quantize_b_ssse3;
-#if CONFIG_INTERNAL_STATS
+#if CONFIG_PSNR
#if ARCH_X86_64
cpi->rtcd.variance.ssimpf_8x8 = vp8_ssim_parms_8x8_sse3;
cpi->rtcd.variance.ssimpf = vp8_ssim_parms_16x16_sse3;
@@ -294,16 +347,16 @@ void vp8_arch_x86_encoder_init(VP8_COMP *cpi)
#if HAVE_SSE4_1
- if (flags & HAS_SSE4_1)
+ if (SSE4_1Enabled)
{
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;
+#if !(CONFIG_REALTIME_ONLY)
cpi->rtcd.search.full_search = vp8_full_search_sadx8;
-
- cpi->rtcd.quantize.quantb = vp8_regular_quantize_b_sse4;
+#endif
}
#endif
« no previous file with comments | « libvpx/source/libvpx/vp8/encoder/x86/variance_x86.h ('k') | libvpx/source/libvpx/vp8/vp8_common.mk » ('j') | no next file with comments »

Powered by Google App Engine
This is Rietveld 408576698