Chromium Code Reviews
chromiumcodereview-hr@appspot.gserviceaccount.com (chromiumcodereview-hr) | Please choose your nickname with Settings | Help | Chromium Project | Gerrit Changes | Sign out
(266)

Issues Search
    My Issues | Starred     Open | Closed | All

Unified Diff: source/libvpx/vp9/encoder/x86/vp9_highbd_variance_sse2.c

Issue 1162573005: libvpx: Pull from upstream (Closed) Base URL: https://chromium.googlesource.com/chromium/deps/libvpx.git@master
Patch Set: Created 5 years, 7 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 | « source/libvpx/vp9/encoder/x86/vp9_highbd_variance_impl_sse2.asm ('k') | source/libvpx/vp9/encoder/x86/vp9_quantize_sse2.c » ('j') | no next file with comments »
Expand Comments ('e') | Collapse Comments ('c') | Show Comments Hide Comments ('s')
Index: source/libvpx/vp9/encoder/x86/vp9_highbd_variance_sse2.c
diff --git a/source/libvpx/vp9/encoder/x86/vp9_highbd_variance_sse2.c b/source/libvpx/vp9/encoder/x86/vp9_highbd_variance_sse2.c
index 4bc3e7e2d15a184700b9175c1ded9737bbd0f7ae..29b7b278217e80ff608fe3080b79aaebc395593c 100644
--- a/source/libvpx/vp9/encoder/x86/vp9_highbd_variance_sse2.c
+++ b/source/libvpx/vp9/encoder/x86/vp9_highbd_variance_sse2.c
@@ -13,237 +13,6 @@
#include "vp9/encoder/vp9_variance.h"
#include "vpx_ports/mem.h"
-typedef uint32_t (*high_variance_fn_t) (const uint16_t *src, int src_stride,
- const uint16_t *ref, int ref_stride,
- uint32_t *sse, int *sum);
-
-uint32_t vp9_highbd_calc8x8var_sse2(const uint16_t *src, int src_stride,
- const uint16_t *ref, int ref_stride,
- uint32_t *sse, int *sum);
-
-uint32_t vp9_highbd_calc16x16var_sse2(const uint16_t *src, int src_stride,
- const uint16_t *ref, int ref_stride,
- uint32_t *sse, int *sum);
-
-static void highbd_variance_sse2(const uint16_t *src, int src_stride,
- const uint16_t *ref, int ref_stride,
- int w, int h, uint32_t *sse, int *sum,
- high_variance_fn_t var_fn, int block_size) {
- int i, j;
-
- *sse = 0;
- *sum = 0;
-
- for (i = 0; i < h; i += block_size) {
- for (j = 0; j < w; j += block_size) {
- unsigned int sse0;
- int sum0;
- var_fn(src + src_stride * i + j, src_stride,
- ref + ref_stride * i + j, ref_stride, &sse0, &sum0);
- *sse += sse0;
- *sum += sum0;
- }
- }
-}
-
-static void highbd_10_variance_sse2(const uint16_t *src, int src_stride,
- const uint16_t *ref, int ref_stride,
- int w, int h, uint32_t *sse, int *sum,
- high_variance_fn_t var_fn, int block_size) {
- int i, j;
- uint64_t sse_long = 0;
- int64_t sum_long = 0;
-
- for (i = 0; i < h; i += block_size) {
- for (j = 0; j < w; j += block_size) {
- unsigned int sse0;
- int sum0;
- var_fn(src + src_stride * i + j, src_stride,
- ref + ref_stride * i + j, ref_stride, &sse0, &sum0);
- sse_long += sse0;
- sum_long += sum0;
- }
- }
- *sum = ROUND_POWER_OF_TWO(sum_long, 2);
- *sse = ROUND_POWER_OF_TWO(sse_long, 4);
-}
-
-static void highbd_12_variance_sse2(const uint16_t *src, int src_stride,
- const uint16_t *ref, int ref_stride,
- int w, int h, uint32_t *sse, int *sum,
- high_variance_fn_t var_fn, int block_size) {
- int i, j;
- uint64_t sse_long = 0;
- int64_t sum_long = 0;
-
- for (i = 0; i < h; i += block_size) {
- for (j = 0; j < w; j += block_size) {
- unsigned int sse0;
- int sum0;
- var_fn(src + src_stride * i + j, src_stride,
- ref + ref_stride * i + j, ref_stride, &sse0, &sum0);
- sse_long += sse0;
- sum_long += sum0;
- }
- }
- *sum = ROUND_POWER_OF_TWO(sum_long, 4);
- *sse = ROUND_POWER_OF_TWO(sse_long, 8);
-}
-
-
-#define HIGH_GET_VAR(S) \
-void vp9_highbd_get##S##x##S##var_sse2(const uint8_t *src8, int src_stride, \
- const uint8_t *ref8, int ref_stride, \
- uint32_t *sse, int *sum) { \
- uint16_t *src = CONVERT_TO_SHORTPTR(src8); \
- uint16_t *ref = CONVERT_TO_SHORTPTR(ref8); \
- vp9_highbd_calc##S##x##S##var_sse2(src, src_stride, ref, ref_stride, \
- sse, sum); \
-} \
-\
-void vp9_highbd_10_get##S##x##S##var_sse2(const uint8_t *src8, int src_stride, \
- const uint8_t *ref8, int ref_stride, \
- uint32_t *sse, int *sum) { \
- uint16_t *src = CONVERT_TO_SHORTPTR(src8); \
- uint16_t *ref = CONVERT_TO_SHORTPTR(ref8); \
- vp9_highbd_calc##S##x##S##var_sse2(src, src_stride, ref, ref_stride, \
- sse, sum); \
- *sum = ROUND_POWER_OF_TWO(*sum, 2); \
- *sse = ROUND_POWER_OF_TWO(*sse, 4); \
-} \
-\
-void vp9_highbd_12_get##S##x##S##var_sse2(const uint8_t *src8, int src_stride, \
- const uint8_t *ref8, int ref_stride, \
- uint32_t *sse, int *sum) { \
- uint16_t *src = CONVERT_TO_SHORTPTR(src8); \
- uint16_t *ref = CONVERT_TO_SHORTPTR(ref8); \
- vp9_highbd_calc##S##x##S##var_sse2(src, src_stride, ref, ref_stride, \
- sse, sum); \
- *sum = ROUND_POWER_OF_TWO(*sum, 4); \
- *sse = ROUND_POWER_OF_TWO(*sse, 8); \
-}
-
-HIGH_GET_VAR(16);
-HIGH_GET_VAR(8);
-
-#undef HIGH_GET_VAR
-
-#define VAR_FN(w, h, block_size, shift) \
-uint32_t vp9_highbd_variance##w##x##h##_sse2( \
- const uint8_t *src8, int src_stride, \
- const uint8_t *ref8, int ref_stride, uint32_t *sse) { \
- int sum; \
- uint16_t *src = CONVERT_TO_SHORTPTR(src8); \
- uint16_t *ref = CONVERT_TO_SHORTPTR(ref8); \
- highbd_variance_sse2(src, src_stride, ref, ref_stride, w, h, sse, &sum, \
- vp9_highbd_calc##block_size##x##block_size##var_sse2, \
- block_size); \
- return *sse - (((int64_t)sum * sum) >> shift); \
-} \
-\
-uint32_t vp9_highbd_10_variance##w##x##h##_sse2( \
- const uint8_t *src8, int src_stride, \
- const uint8_t *ref8, int ref_stride, uint32_t *sse) { \
- int sum; \
- uint16_t *src = CONVERT_TO_SHORTPTR(src8); \
- uint16_t *ref = CONVERT_TO_SHORTPTR(ref8); \
- highbd_10_variance_sse2( \
- src, src_stride, ref, ref_stride, w, h, sse, &sum, \
- vp9_highbd_calc##block_size##x##block_size##var_sse2, block_size); \
- return *sse - (((int64_t)sum * sum) >> shift); \
-} \
-\
-uint32_t vp9_highbd_12_variance##w##x##h##_sse2( \
- const uint8_t *src8, int src_stride, \
- const uint8_t *ref8, int ref_stride, uint32_t *sse) { \
- int sum; \
- uint16_t *src = CONVERT_TO_SHORTPTR(src8); \
- uint16_t *ref = CONVERT_TO_SHORTPTR(ref8); \
- highbd_12_variance_sse2( \
- src, src_stride, ref, ref_stride, w, h, sse, &sum, \
- vp9_highbd_calc##block_size##x##block_size##var_sse2, block_size); \
- return *sse - (((int64_t)sum * sum) >> shift); \
-}
-
-VAR_FN(64, 64, 16, 12);
-VAR_FN(64, 32, 16, 11);
-VAR_FN(32, 64, 16, 11);
-VAR_FN(32, 32, 16, 10);
-VAR_FN(32, 16, 16, 9);
-VAR_FN(16, 32, 16, 9);
-VAR_FN(16, 16, 16, 8);
-VAR_FN(16, 8, 8, 7);
-VAR_FN(8, 16, 8, 7);
-VAR_FN(8, 8, 8, 6);
-
-#undef VAR_FN
-
-unsigned int vp9_highbd_mse16x16_sse2(const uint8_t *src8, int src_stride,
- const uint8_t *ref8, int ref_stride,
- unsigned int *sse) {
- int sum;
- uint16_t *src = CONVERT_TO_SHORTPTR(src8);
- uint16_t *ref = CONVERT_TO_SHORTPTR(ref8);
- highbd_variance_sse2(src, src_stride, ref, ref_stride, 16, 16,
- sse, &sum, vp9_highbd_calc16x16var_sse2, 16);
- return *sse;
-}
-
-unsigned int vp9_highbd_10_mse16x16_sse2(const uint8_t *src8, int src_stride,
- const uint8_t *ref8, int ref_stride,
- unsigned int *sse) {
- int sum;
- uint16_t *src = CONVERT_TO_SHORTPTR(src8);
- uint16_t *ref = CONVERT_TO_SHORTPTR(ref8);
- highbd_10_variance_sse2(src, src_stride, ref, ref_stride, 16, 16,
- sse, &sum, vp9_highbd_calc16x16var_sse2, 16);
- return *sse;
-}
-
-unsigned int vp9_highbd_12_mse16x16_sse2(const uint8_t *src8, int src_stride,
- const uint8_t *ref8, int ref_stride,
- unsigned int *sse) {
- int sum;
- uint16_t *src = CONVERT_TO_SHORTPTR(src8);
- uint16_t *ref = CONVERT_TO_SHORTPTR(ref8);
- highbd_12_variance_sse2(src, src_stride, ref, ref_stride, 16, 16,
- sse, &sum, vp9_highbd_calc16x16var_sse2, 16);
- return *sse;
-}
-
-unsigned int vp9_highbd_mse8x8_sse2(const uint8_t *src8, int src_stride,
- const uint8_t *ref8, int ref_stride,
- unsigned int *sse) {
- int sum;
- uint16_t *src = CONVERT_TO_SHORTPTR(src8);
- uint16_t *ref = CONVERT_TO_SHORTPTR(ref8);
- highbd_variance_sse2(src, src_stride, ref, ref_stride, 8, 8,
- sse, &sum, vp9_highbd_calc8x8var_sse2, 8);
- return *sse;
-}
-
-unsigned int vp9_highbd_10_mse8x8_sse2(const uint8_t *src8, int src_stride,
- const uint8_t *ref8, int ref_stride,
- unsigned int *sse) {
- int sum;
- uint16_t *src = CONVERT_TO_SHORTPTR(src8);
- uint16_t *ref = CONVERT_TO_SHORTPTR(ref8);
- highbd_10_variance_sse2(src, src_stride, ref, ref_stride, 8, 8,
- sse, &sum, vp9_highbd_calc8x8var_sse2, 8);
- return *sse;
-}
-
-unsigned int vp9_highbd_12_mse8x8_sse2(const uint8_t *src8, int src_stride,
- const uint8_t *ref8, int ref_stride,
- unsigned int *sse) {
- int sum;
- uint16_t *src = CONVERT_TO_SHORTPTR(src8);
- uint16_t *ref = CONVERT_TO_SHORTPTR(ref8);
- highbd_12_variance_sse2(src, src_stride, ref, ref_stride, 8, 8,
- sse, &sum, vp9_highbd_calc8x8var_sse2, 8);
- return *sse;
-}
-
#define DECL(w, opt) \
int vp9_highbd_sub_pixel_variance##w##xh_##opt(const uint16_t *src, \
ptrdiff_t src_stride, \
« no previous file with comments | « source/libvpx/vp9/encoder/x86/vp9_highbd_variance_impl_sse2.asm ('k') | source/libvpx/vp9/encoder/x86/vp9_quantize_sse2.c » ('j') | no next file with comments »

Powered by Google App Engine
This is Rietveld 408576698