summaryrefslogtreecommitdiff
path: root/vp9/common/mips/dspr2
diff options
context:
space:
mode:
authorJingning Han <jingning@google.com>2015-07-17 12:31:53 -0700
committerJingning Han <jingning@google.com>2015-07-17 15:55:02 -0700
commit2992739b5d5ee0b87e8654b3caaeee3a7b468465 (patch)
tree8fed5b7c55159c5dded56e8870ff83eb64c70479 /vp9/common/mips/dspr2
parent4735edd00f6a7b4131b9b9c33b220bfaf0d5a68a (diff)
downloadlibvpx-2992739b5d5ee0b87e8654b3caaeee3a7b468465.tar
libvpx-2992739b5d5ee0b87e8654b3caaeee3a7b468465.tar.gz
libvpx-2992739b5d5ee0b87e8654b3caaeee3a7b468465.tar.bz2
libvpx-2992739b5d5ee0b87e8654b3caaeee3a7b468465.zip
Rename loop filter function from vp9_ to vpx_
Change-Id: I6f424bb8daec26bf8482b5d75dd9b0e45c11a665
Diffstat (limited to 'vp9/common/mips/dspr2')
-rw-r--r--vp9/common/mips/dspr2/vp9_convolve2_avg_dspr2.c10
-rw-r--r--vp9/common/mips/dspr2/vp9_convolve2_avg_horiz_dspr2.c38
-rw-r--r--vp9/common/mips/dspr2/vp9_convolve2_dspr2.c24
-rw-r--r--vp9/common/mips/dspr2/vp9_convolve2_horiz_dspr2.c40
-rw-r--r--vp9/common/mips/dspr2/vp9_convolve2_vert_dspr2.c8
-rw-r--r--vp9/common/mips/dspr2/vp9_convolve8_avg_dspr2.c54
-rw-r--r--vp9/common/mips/dspr2/vp9_convolve8_avg_horiz_dspr2.c38
-rw-r--r--vp9/common/mips/dspr2/vp9_convolve8_dspr2.c68
-rw-r--r--vp9/common/mips/dspr2/vp9_convolve8_horiz_dspr2.c40
-rw-r--r--vp9/common/mips/dspr2/vp9_convolve8_vert_dspr2.c10
-rw-r--r--vp9/common/mips/dspr2/vp9_itrans16_dspr2.c22
-rw-r--r--vp9/common/mips/dspr2/vp9_itrans32_cols_dspr2.c16
-rw-r--r--vp9/common/mips/dspr2/vp9_itrans32_dspr2.c4
-rw-r--r--vp9/common/mips/dspr2/vp9_itrans4_dspr2.c16
-rw-r--r--vp9/common/mips/dspr2/vp9_itrans8_dspr2.c16
15 files changed, 202 insertions, 202 deletions
diff --git a/vp9/common/mips/dspr2/vp9_convolve2_avg_dspr2.c b/vp9/common/mips/dspr2/vp9_convolve2_avg_dspr2.c
index 91d62bc49..aad7c4514 100644
--- a/vp9/common/mips/dspr2/vp9_convolve2_avg_dspr2.c
+++ b/vp9/common/mips/dspr2/vp9_convolve2_avg_dspr2.c
@@ -44,7 +44,7 @@ static void convolve_bi_avg_vert_4_dspr2(const uint8_t *src,
for (y = h; y--;) {
/* prefetch data to cache memory */
- vp9_prefetch_store(dst + dst_stride);
+ prefetch_store(dst + dst_stride);
for (x = 0; x < w; x += 4) {
src_ptr = src + x;
@@ -148,8 +148,8 @@ static void convolve_bi_avg_vert_64_dspr2(const uint8_t *src,
for (y = h; y--;) {
/* prefetch data to cache memory */
- vp9_prefetch_store(dst + dst_stride);
- vp9_prefetch_store(dst + dst_stride + 32);
+ prefetch_store(dst + dst_stride);
+ prefetch_store(dst + dst_stride + 32);
for (x = 0; x < 64; x += 4) {
src_ptr = src + x;
@@ -245,7 +245,7 @@ void vp9_convolve2_avg_vert_dspr2(const uint8_t *src, ptrdiff_t src_stride,
: [pos] "r" (pos)
);
- vp9_prefetch_store(dst);
+ prefetch_store(dst);
switch (w) {
case 4:
@@ -257,7 +257,7 @@ void vp9_convolve2_avg_vert_dspr2(const uint8_t *src, ptrdiff_t src_stride,
filter_y, w, h);
break;
case 64:
- vp9_prefetch_store(dst + 32);
+ prefetch_store(dst + 32);
convolve_bi_avg_vert_64_dspr2(src, src_stride,
dst, dst_stride,
filter_y, h);
diff --git a/vp9/common/mips/dspr2/vp9_convolve2_avg_horiz_dspr2.c b/vp9/common/mips/dspr2/vp9_convolve2_avg_horiz_dspr2.c
index 148b20fba..bc60e9332 100644
--- a/vp9/common/mips/dspr2/vp9_convolve2_avg_horiz_dspr2.c
+++ b/vp9/common/mips/dspr2/vp9_convolve2_avg_horiz_dspr2.c
@@ -40,9 +40,9 @@ static void convolve_bi_avg_horiz_4_dspr2(const uint8_t *src,
for (y = h; y--;) {
/* prefetch data to cache memory */
- vp9_prefetch_load(src + src_stride);
- vp9_prefetch_load(src + src_stride + 32);
- vp9_prefetch_store(dst + dst_stride);
+ prefetch_load(src + src_stride);
+ prefetch_load(src + src_stride + 32);
+ prefetch_store(dst + dst_stride);
__asm__ __volatile__ (
"ulw %[tp1], 0(%[src]) \n\t"
@@ -135,9 +135,9 @@ static void convolve_bi_avg_horiz_8_dspr2(const uint8_t *src,
for (y = h; y--;) {
/* prefetch data to cache memory */
- vp9_prefetch_load(src + src_stride);
- vp9_prefetch_load(src + src_stride + 32);
- vp9_prefetch_store(dst + dst_stride);
+ prefetch_load(src + src_stride);
+ prefetch_load(src + src_stride + 32);
+ prefetch_store(dst + dst_stride);
__asm__ __volatile__ (
"ulw %[tp1], 0(%[src]) \n\t"
@@ -290,9 +290,9 @@ static void convolve_bi_avg_horiz_16_dspr2(const uint8_t *src_ptr,
dst = dst_ptr;
/* prefetch data to cache memory */
- vp9_prefetch_load(src_ptr + src_stride);
- vp9_prefetch_load(src_ptr + src_stride + 32);
- vp9_prefetch_store(dst_ptr + dst_stride);
+ prefetch_load(src_ptr + src_stride);
+ prefetch_load(src_ptr + src_stride + 32);
+ prefetch_store(dst_ptr + dst_stride);
for (c = 0; c < count; c++) {
__asm__ __volatile__ (
@@ -539,11 +539,11 @@ static void convolve_bi_avg_horiz_64_dspr2(const uint8_t *src_ptr,
dst = dst_ptr;
/* prefetch data to cache memory */
- vp9_prefetch_load(src_ptr + src_stride);
- vp9_prefetch_load(src_ptr + src_stride + 32);
- vp9_prefetch_load(src_ptr + src_stride + 64);
- vp9_prefetch_store(dst_ptr + dst_stride);
- vp9_prefetch_store(dst_ptr + dst_stride + 32);
+ prefetch_load(src_ptr + src_stride);
+ prefetch_load(src_ptr + src_stride + 32);
+ prefetch_load(src_ptr + src_stride + 64);
+ prefetch_store(dst_ptr + dst_stride);
+ prefetch_store(dst_ptr + dst_stride + 32);
for (c = 0; c < 4; c++) {
__asm__ __volatile__ (
@@ -781,9 +781,9 @@ void vp9_convolve2_avg_horiz_dspr2(const uint8_t *src, ptrdiff_t src_stride,
);
/* prefetch data to cache memory */
- vp9_prefetch_load(src);
- vp9_prefetch_load(src + 32);
- vp9_prefetch_store(dst);
+ prefetch_load(src);
+ prefetch_load(src + 32);
+ prefetch_store(dst);
switch (w) {
case 4:
@@ -807,8 +807,8 @@ void vp9_convolve2_avg_horiz_dspr2(const uint8_t *src, ptrdiff_t src_stride,
filter_x, h, 2);
break;
case 64:
- vp9_prefetch_load(src + 64);
- vp9_prefetch_store(dst + 32);
+ prefetch_load(src + 64);
+ prefetch_store(dst + 32);
convolve_bi_avg_horiz_64_dspr2(src, src_stride,
dst, dst_stride,
diff --git a/vp9/common/mips/dspr2/vp9_convolve2_dspr2.c b/vp9/common/mips/dspr2/vp9_convolve2_dspr2.c
index 92644f2c8..b714f9a4d 100644
--- a/vp9/common/mips/dspr2/vp9_convolve2_dspr2.c
+++ b/vp9/common/mips/dspr2/vp9_convolve2_dspr2.c
@@ -41,8 +41,8 @@ static void convolve_bi_horiz_4_transposed_dspr2(const uint8_t *src,
for (y = h; y--;) {
dst_ptr = dst;
/* prefetch data to cache memory */
- vp9_prefetch_load(src + src_stride);
- vp9_prefetch_load(src + src_stride + 32);
+ prefetch_load(src + src_stride);
+ prefetch_load(src + src_stride + 32);
__asm__ __volatile__ (
"ulw %[tp1], 0(%[src]) \n\t"
@@ -132,8 +132,8 @@ static void convolve_bi_horiz_8_transposed_dspr2(const uint8_t *src,
for (y = h; y--;) {
/* prefetch data to cache memory */
- vp9_prefetch_load(src + src_stride);
- vp9_prefetch_load(src + src_stride + 32);
+ prefetch_load(src + src_stride);
+ prefetch_load(src + src_stride + 32);
dst_ptr = dst;
odd_dst = (dst_ptr + dst_stride);
@@ -272,8 +272,8 @@ static void convolve_bi_horiz_16_transposed_dspr2(const uint8_t *src_ptr,
for (y = h; y--;) {
/* prefetch data to cache memory */
- vp9_prefetch_load(src_ptr + src_stride);
- vp9_prefetch_load(src_ptr + src_stride + 32);
+ prefetch_load(src_ptr + src_stride);
+ prefetch_load(src_ptr + src_stride + 32);
src = src_ptr;
dst = dst_ptr;
@@ -504,9 +504,9 @@ static void convolve_bi_horiz_64_transposed_dspr2(const uint8_t *src_ptr,
for (y = h; y--;) {
/* prefetch data to cache memory */
- vp9_prefetch_load(src_ptr + src_stride);
- vp9_prefetch_load(src_ptr + src_stride + 32);
- vp9_prefetch_load(src_ptr + src_stride + 64);
+ prefetch_load(src_ptr + src_stride);
+ prefetch_load(src_ptr + src_stride + 32);
+ prefetch_load(src_ptr + src_stride + 64);
src = src_ptr;
dst = dst_ptr;
@@ -747,8 +747,8 @@ void vp9_convolve2_dspr2(const uint8_t *src, ptrdiff_t src_stride,
);
/* prefetch data to cache memory */
- vp9_prefetch_load(src);
- vp9_prefetch_load(src + 32);
+ prefetch_load(src);
+ prefetch_load(src + 32);
switch (w) {
case 4:
@@ -769,7 +769,7 @@ void vp9_convolve2_dspr2(const uint8_t *src, ptrdiff_t src_stride,
(w/16));
break;
case 64:
- vp9_prefetch_load(src + 32);
+ prefetch_load(src + 32);
convolve_bi_horiz_64_transposed_dspr2(src, src_stride,
dst, dst_stride,
filter, h);
diff --git a/vp9/common/mips/dspr2/vp9_convolve2_horiz_dspr2.c b/vp9/common/mips/dspr2/vp9_convolve2_horiz_dspr2.c
index 1debdb4c0..27ea100cd 100644
--- a/vp9/common/mips/dspr2/vp9_convolve2_horiz_dspr2.c
+++ b/vp9/common/mips/dspr2/vp9_convolve2_horiz_dspr2.c
@@ -39,9 +39,9 @@ static void convolve_bi_horiz_4_dspr2(const uint8_t *src,
for (y = h; y--;) {
/* prefetch data to cache memory */
- vp9_prefetch_load(src + src_stride);
- vp9_prefetch_load(src + src_stride + 32);
- vp9_prefetch_store(dst + dst_stride);
+ prefetch_load(src + src_stride);
+ prefetch_load(src + src_stride + 32);
+ prefetch_store(dst + dst_stride);
__asm__ __volatile__ (
"ulw %[tp1], 0(%[src]) \n\t"
@@ -122,9 +122,9 @@ static void convolve_bi_horiz_8_dspr2(const uint8_t *src,
for (y = h; y--;) {
/* prefetch data to cache memory */
- vp9_prefetch_load(src + src_stride);
- vp9_prefetch_load(src + src_stride + 32);
- vp9_prefetch_store(dst + dst_stride);
+ prefetch_load(src + src_stride);
+ prefetch_load(src + src_stride + 32);
+ prefetch_store(dst + dst_stride);
__asm__ __volatile__ (
"ulw %[tp1], 0(%[src]) \n\t"
@@ -252,9 +252,9 @@ static void convolve_bi_horiz_16_dspr2(const uint8_t *src_ptr,
dst = dst_ptr;
/* prefetch data to cache memory */
- vp9_prefetch_load(src_ptr + src_stride);
- vp9_prefetch_load(src_ptr + src_stride + 32);
- vp9_prefetch_store(dst_ptr + dst_stride);
+ prefetch_load(src_ptr + src_stride);
+ prefetch_load(src_ptr + src_stride + 32);
+ prefetch_store(dst_ptr + dst_stride);
for (c = 0; c < count; c++) {
__asm__ __volatile__ (
@@ -459,11 +459,11 @@ static void convolve_bi_horiz_64_dspr2(const uint8_t *src_ptr,
dst = dst_ptr;
/* prefetch data to cache memory */
- vp9_prefetch_load(src_ptr + src_stride);
- vp9_prefetch_load(src_ptr + src_stride + 32);
- vp9_prefetch_load(src_ptr + src_stride + 64);
- vp9_prefetch_store(dst_ptr + dst_stride);
- vp9_prefetch_store(dst_ptr + dst_stride + 32);
+ prefetch_load(src_ptr + src_stride);
+ prefetch_load(src_ptr + src_stride + 32);
+ prefetch_load(src_ptr + src_stride + 64);
+ prefetch_store(dst_ptr + dst_stride);
+ prefetch_store(dst_ptr + dst_stride + 32);
for (c = 0; c < 4; c++) {
__asm__ __volatile__ (
@@ -651,7 +651,7 @@ void vp9_convolve2_horiz_dspr2(const uint8_t *src, ptrdiff_t src_stride,
if (16 == x_step_q4) {
uint32_t pos = 38;
- vp9_prefetch_load((const uint8_t *)filter_x);
+ prefetch_load((const uint8_t *)filter_x);
/* bit positon for extract from acc */
__asm__ __volatile__ (
@@ -661,9 +661,9 @@ void vp9_convolve2_horiz_dspr2(const uint8_t *src, ptrdiff_t src_stride,
);
/* prefetch data to cache memory */
- vp9_prefetch_load(src);
- vp9_prefetch_load(src + 32);
- vp9_prefetch_store(dst);
+ prefetch_load(src);
+ prefetch_load(src + 32);
+ prefetch_store(dst);
switch (w) {
case 4:
@@ -687,8 +687,8 @@ void vp9_convolve2_horiz_dspr2(const uint8_t *src, ptrdiff_t src_stride,
filter_x, (int32_t)h, 2);
break;
case 64:
- vp9_prefetch_load(src + 64);
- vp9_prefetch_store(dst + 32);
+ prefetch_load(src + 64);
+ prefetch_store(dst + 32);
convolve_bi_horiz_64_dspr2(src, (int32_t)src_stride,
dst, (int32_t)dst_stride,
diff --git a/vp9/common/mips/dspr2/vp9_convolve2_vert_dspr2.c b/vp9/common/mips/dspr2/vp9_convolve2_vert_dspr2.c
index bf01f1154..32f5fb663 100644
--- a/vp9/common/mips/dspr2/vp9_convolve2_vert_dspr2.c
+++ b/vp9/common/mips/dspr2/vp9_convolve2_vert_dspr2.c
@@ -44,7 +44,7 @@ static void convolve_bi_vert_4_dspr2(const uint8_t *src,
for (y = h; y--;) {
/* prefetch data to cache memory */
- vp9_prefetch_store(dst + dst_stride);
+ prefetch_store(dst + dst_stride);
for (x = 0; x < w; x += 4) {
src_ptr = src + x;
@@ -141,7 +141,7 @@ static void convolve_bi_vert_64_dspr2(const uint8_t *src,
for (y = h; y--;) {
/* prefetch data to cache memory */
- vp9_prefetch_store(dst + dst_stride);
+ prefetch_store(dst + dst_stride);
for (x = 0; x < 64; x += 4) {
src_ptr = src + x;
@@ -230,7 +230,7 @@ void vp9_convolve2_vert_dspr2(const uint8_t *src, ptrdiff_t src_stride,
: [pos] "r" (pos)
);
- vp9_prefetch_store(dst);
+ prefetch_store(dst);
switch (w) {
case 4 :
@@ -242,7 +242,7 @@ void vp9_convolve2_vert_dspr2(const uint8_t *src, ptrdiff_t src_stride,
filter_y, w, h);
break;
case 64 :
- vp9_prefetch_store(dst + 32);
+ prefetch_store(dst + 32);
convolve_bi_vert_64_dspr2(src, src_stride,
dst, dst_stride,
filter_y, h);
diff --git a/vp9/common/mips/dspr2/vp9_convolve8_avg_dspr2.c b/vp9/common/mips/dspr2/vp9_convolve8_avg_dspr2.c
index 17422798c..d9cbfe68f 100644
--- a/vp9/common/mips/dspr2/vp9_convolve8_avg_dspr2.c
+++ b/vp9/common/mips/dspr2/vp9_convolve8_avg_dspr2.c
@@ -49,7 +49,7 @@ static void convolve_avg_vert_4_dspr2(const uint8_t *src,
for (y = h; y--;) {
/* prefetch data to cache memory */
- vp9_prefetch_store(dst + dst_stride);
+ prefetch_store(dst + dst_stride);
for (x = 0; x < w; x += 4) {
src_ptr = src + x;
@@ -210,8 +210,8 @@ static void convolve_avg_vert_64_dspr2(const uint8_t *src,
for (y = h; y--;) {
/* prefetch data to cache memory */
- vp9_prefetch_store(dst + dst_stride);
- vp9_prefetch_store(dst + dst_stride + 32);
+ prefetch_store(dst + dst_stride);
+ prefetch_store(dst + dst_stride + 32);
for (x = 0; x < 64; x += 4) {
src_ptr = src + x;
@@ -372,7 +372,7 @@ void vp9_convolve8_avg_vert_dspr2(const uint8_t *src, ptrdiff_t src_stride,
: [pos] "r" (pos)
);
- vp9_prefetch_store(dst);
+ prefetch_store(dst);
switch (w) {
case 4:
@@ -384,7 +384,7 @@ void vp9_convolve8_avg_vert_dspr2(const uint8_t *src, ptrdiff_t src_stride,
filter_y, w, h);
break;
case 64:
- vp9_prefetch_store(dst + 32);
+ prefetch_store(dst + 32);
convolve_avg_vert_64_dspr2(src, src_stride,
dst, dst_stride,
filter_y, h);
@@ -452,17 +452,17 @@ void vp9_convolve_avg_dspr2(const uint8_t *src, ptrdiff_t src_stride,
uint32_t tp3, tp4, tn2;
/* prefetch data to cache memory */
- vp9_prefetch_load(src);
- vp9_prefetch_load(src + 32);
- vp9_prefetch_store(dst);
+ prefetch_load(src);
+ prefetch_load(src + 32);
+ prefetch_store(dst);
switch (w) {
case 4:
/* 1 word storage */
for (y = h; y--; ) {
- vp9_prefetch_load(src + src_stride);
- vp9_prefetch_load(src + src_stride + 32);
- vp9_prefetch_store(dst + dst_stride);
+ prefetch_load(src + src_stride);
+ prefetch_load(src + src_stride + 32);
+ prefetch_store(dst + dst_stride);
__asm__ __volatile__ (
"ulw %[tp1], 0(%[src]) \n\t"
@@ -482,9 +482,9 @@ void vp9_convolve_avg_dspr2(const uint8_t *src, ptrdiff_t src_stride,
case 8:
/* 2 word storage */
for (y = h; y--; ) {
- vp9_prefetch_load(src + src_stride);
- vp9_prefetch_load(src + src_stride + 32);
- vp9_prefetch_store(dst + dst_stride);
+ prefetch_load(src + src_stride);
+ prefetch_load(src + src_stride + 32);
+ prefetch_store(dst + dst_stride);
__asm__ __volatile__ (
"ulw %[tp1], 0(%[src]) \n\t"
@@ -509,9 +509,9 @@ void vp9_convolve_avg_dspr2(const uint8_t *src, ptrdiff_t src_stride,
case 16:
/* 4 word storage */
for (y = h; y--; ) {
- vp9_prefetch_load(src + src_stride);
- vp9_prefetch_load(src + src_stride + 32);
- vp9_prefetch_store(dst + dst_stride);
+ prefetch_load(src + src_stride);
+ prefetch_load(src + src_stride + 32);
+ prefetch_store(dst + dst_stride);
__asm__ __volatile__ (
"ulw %[tp1], 0(%[src]) \n\t"
@@ -544,9 +544,9 @@ void vp9_convolve_avg_dspr2(const uint8_t *src, ptrdiff_t src_stride,
case 32:
/* 8 word storage */
for (y = h; y--; ) {
- vp9_prefetch_load(src + src_stride);
- vp9_prefetch_load(src + src_stride + 32);
- vp9_prefetch_store(dst + dst_stride);
+ prefetch_load(src + src_stride);
+ prefetch_load(src + src_stride + 32);
+ prefetch_store(dst + dst_stride);
__asm__ __volatile__ (
"ulw %[tp1], 0(%[src]) \n\t"
@@ -593,16 +593,16 @@ void vp9_convolve_avg_dspr2(const uint8_t *src, ptrdiff_t src_stride,
}
break;
case 64:
- vp9_prefetch_load(src + 64);
- vp9_prefetch_store(dst + 32);
+ prefetch_load(src + 64);
+ prefetch_store(dst + 32);
/* 16 word storage */
for (y = h; y--; ) {
- vp9_prefetch_load(src + src_stride);
- vp9_prefetch_load(src + src_stride + 32);
- vp9_prefetch_load(src + src_stride + 64);
- vp9_prefetch_store(dst + dst_stride);
- vp9_prefetch_store(dst + dst_stride + 32);
+ prefetch_load(src + src_stride);
+ prefetch_load(src + src_stride + 32);
+ prefetch_load(src + src_stride + 64);
+ prefetch_store(dst + dst_stride);
+ prefetch_store(dst + dst_stride + 32);
__asm__ __volatile__ (
"ulw %[tp1], 0(%[src]) \n\t"
diff --git a/vp9/common/mips/dspr2/vp9_convolve8_avg_horiz_dspr2.c b/vp9/common/mips/dspr2/vp9_convolve8_avg_horiz_dspr2.c
index 69da1cfd6..cdb831237 100644
--- a/vp9/common/mips/dspr2/vp9_convolve8_avg_horiz_dspr2.c
+++ b/vp9/common/mips/dspr2/vp9_convolve8_avg_horiz_dspr2.c
@@ -43,9 +43,9 @@ static void convolve_avg_horiz_4_dspr2(const uint8_t *src,
for (y = h; y--;) {
/* prefetch data to cache memory */
- vp9_prefetch_load(src + src_stride);
- vp9_prefetch_load(src + src_stride + 32);
- vp9_prefetch_store(dst + dst_stride);
+ prefetch_load(src + src_stride);
+ prefetch_load(src + src_stride + 32);
+ prefetch_store(dst + dst_stride);
__asm__ __volatile__ (
"ulw %[tp1], 0(%[src]) \n\t"
@@ -165,9 +165,9 @@ static void convolve_avg_horiz_8_dspr2(const uint8_t *src,
for (y = h; y--;) {
/* prefetch data to cache memory */
- vp9_prefetch_load(src + src_stride);
- vp9_prefetch_load(src + src_stride + 32);
- vp9_prefetch_store(dst + dst_stride);
+ prefetch_load(src + src_stride);
+ prefetch_load(src + src_stride + 32);
+ prefetch_store(dst + dst_stride);
__asm__ __volatile__ (
"ulw %[tp1], 0(%[src]) \n\t"
@@ -357,9 +357,9 @@ static void convolve_avg_horiz_16_dspr2(const uint8_t *src_ptr,
dst = dst_ptr;
/* prefetch data to cache memory */
- vp9_prefetch_load(src_ptr + src_stride);
- vp9_prefetch_load(src_ptr + src_stride + 32);
- vp9_prefetch_store(dst_ptr + dst_stride);
+ prefetch_load(src_ptr + src_stride);
+ prefetch_load(src_ptr + src_stride + 32);
+ prefetch_store(dst_ptr + dst_stride);
for (c = 0; c < count; c++) {
__asm__ __volatile__ (
@@ -668,11 +668,11 @@ static void convolve_avg_horiz_64_dspr2(const uint8_t *src_ptr,
dst = dst_ptr;
/* prefetch data to cache memory */
- vp9_prefetch_load(src_ptr + src_stride);
- vp9_prefetch_load(src_ptr + src_stride + 32);
- vp9_prefetch_load(src_ptr + src_stride + 64);
- vp9_prefetch_store(dst_ptr + dst_stride);
- vp9_prefetch_store(dst_ptr + dst_stride + 32);
+ prefetch_load(src_ptr + src_stride);
+ prefetch_load(src_ptr + src_stride + 32);
+ prefetch_load(src_ptr + src_stride + 64);
+ prefetch_store(dst_ptr + dst_stride);
+ prefetch_store(dst_ptr + dst_stride + 32);
for (c = 0; c < 4; c++) {
__asm__ __volatile__ (
@@ -985,9 +985,9 @@ void vp9_convolve8_avg_horiz_dspr2(const uint8_t *src, ptrdiff_t src_stride,
);
/* prefetch data to cache memory */
- vp9_prefetch_load(src);
- vp9_prefetch_load(src + 32);
- vp9_prefetch_store(dst);
+ prefetch_load(src);
+ prefetch_load(src + 32);
+ prefetch_store(dst);
switch (w) {
case 4:
@@ -1011,8 +1011,8 @@ void vp9_convolve8_avg_horiz_dspr2(const uint8_t *src, ptrdiff_t src_stride,
filter_x, h, 2);
break;
case 64:
- vp9_prefetch_load(src + 64);
- vp9_prefetch_store(dst + 32);
+ prefetch_load(src + 64);
+ prefetch_store(dst + 32);
convolve_avg_horiz_64_dspr2(src, src_stride,
dst, dst_stride,
diff --git a/vp9/common/mips/dspr2/vp9_convolve8_dspr2.c b/vp9/common/mips/dspr2/vp9_convolve8_dspr2.c
index 58b50d2df..a1309d1ee 100644
--- a/vp9/common/mips/dspr2/vp9_convolve8_dspr2.c
+++ b/vp9/common/mips/dspr2/vp9_convolve8_dspr2.c
@@ -60,8 +60,8 @@ static void convolve_horiz_4_transposed_dspr2(const uint8_t *src,
for (y = h; y--;) {
dst_ptr = dst;
/* prefetch data to cache memory */
- vp9_prefetch_load(src + src_stride);
- vp9_prefetch_load(src + src_stride + 32);
+ prefetch_load(src + src_stride);
+ prefetch_load(src + src_stride + 32);
__asm__ __volatile__ (
"ulw %[tp1], 0(%[src]) \n\t"
@@ -176,8 +176,8 @@ static void convolve_horiz_8_transposed_dspr2(const uint8_t *src,
for (y = h; y--;) {
/* prefetch data to cache memory */
- vp9_prefetch_load(src + src_stride);
- vp9_prefetch_load(src + src_stride + 32);
+ prefetch_load(src + src_stride);
+ prefetch_load(src + src_stride + 32);
dst_ptr = dst;
odd_dst = (dst_ptr + dst_stride);
@@ -355,8 +355,8 @@ static void convolve_horiz_16_transposed_dspr2(const uint8_t *src_ptr,
for (y = h; y--;) {
/* prefetch data to cache memory */
- vp9_prefetch_load(src_ptr + src_stride);
- vp9_prefetch_load(src_ptr + src_stride + 32);
+ prefetch_load(src_ptr + src_stride);
+ prefetch_load(src_ptr + src_stride + 32);
src = src_ptr;
dst = dst_ptr;
@@ -645,9 +645,9 @@ static void convolve_horiz_64_transposed_dspr2(const uint8_t *src_ptr,
for (y = h; y--;) {
/* prefetch data to cache memory */
- vp9_prefetch_load(src_ptr + src_stride);
- vp9_prefetch_load(src_ptr + src_stride + 32);
- vp9_prefetch_load(src_ptr + src_stride + 64);
+ prefetch_load(src_ptr + src_stride);
+ prefetch_load(src_ptr + src_stride + 32);
+ prefetch_load(src_ptr + src_stride + 64);
src = src_ptr;
dst = dst_ptr;
@@ -993,8 +993,8 @@ void vp9_convolve8_dspr2(const uint8_t *src, ptrdiff_t src_stride,
src -= (src_stride * 3 + 3);
/* prefetch data to cache memory */
- vp9_prefetch_load(src);
- vp9_prefetch_load(src + 32);
+ prefetch_load(src);
+ prefetch_load(src + 32);
switch (w) {
case 4:
@@ -1015,7 +1015,7 @@ void vp9_convolve8_dspr2(const uint8_t *src, ptrdiff_t src_stride,
(w/16));
break;
case 64:
- vp9_prefetch_load(src + 32);
+ prefetch_load(src + 32);
convolve_horiz_64_transposed_dspr2(src, src_stride,
temp, intermediate_height,
filter_x, intermediate_height);
@@ -1078,9 +1078,9 @@ void vp9_convolve_copy_dspr2(const uint8_t *src, ptrdiff_t src_stride,
int x, y;
/* prefetch data to cache memory */
- vp9_prefetch_load(src);
- vp9_prefetch_load(src + 32);
- vp9_prefetch_store(dst);
+ prefetch_load(src);
+ prefetch_load(src + 32);
+ prefetch_store(dst);
switch (w) {
case 4:
@@ -1089,9 +1089,9 @@ void vp9_convolve_copy_dspr2(const uint8_t *src, ptrdiff_t src_stride,
/* 1 word storage */
for (y = h; y--; ) {
- vp9_prefetch_load(src + src_stride);
- vp9_prefetch_load(src + src_stride + 32);
- vp9_prefetch_store(dst + dst_stride);
+ prefetch_load(src + src_stride);
+ prefetch_load(src + src_stride + 32);
+ prefetch_store(dst + dst_stride);
__asm__ __volatile__ (
"ulw %[tp1], (%[src]) \n\t"
@@ -1112,9 +1112,9 @@ void vp9_convolve_copy_dspr2(const uint8_t *src, ptrdiff_t src_stride,
/* 2 word storage */
for (y = h; y--; ) {
- vp9_prefetch_load(src + src_stride);
- vp9_prefetch_load(src + src_stride + 32);
- vp9_prefetch_store(dst + dst_stride);
+ prefetch_load(src + src_stride);
+ prefetch_load(src + src_stride + 32);
+ prefetch_store(dst + dst_stride);
__asm__ __volatile__ (
"ulw %[tp1], 0(%[src]) \n\t"
@@ -1137,9 +1137,9 @@ void vp9_convolve_copy_dspr2(const uint8_t *src, ptrdiff_t src_stride,
/* 4 word storage */
for (y = h; y--; ) {
- vp9_prefetch_load(src + src_stride);
- vp9_prefetch_load(src + src_stride + 32);
- vp9_prefetch_store(dst + dst_stride);
+ prefetch_load(src + src_stride);
+ prefetch_load(src + src_stride + 32);
+ prefetch_store(dst + dst_stride);
__asm__ __volatile__ (
"ulw %[tp1], 0(%[src]) \n\t"
@@ -1169,9 +1169,9 @@ void vp9_convolve_copy_dspr2(const uint8_t *src, ptrdiff_t src_stride,
/* 8 word storage */
for (y = h; y--; ) {
- vp9_prefetch_load(src + src_stride);
- vp9_prefetch_load(src + src_stride + 32);
- vp9_prefetch_store(dst + dst_stride);
+ prefetch_load(src + src_stride);
+ prefetch_load(src + src_stride + 32);
+ prefetch_store(dst + dst_stride);
__asm__ __volatile__ (
"ulw %[tp1], 0(%[src]) \n\t"
@@ -1209,16 +1209,16 @@ void vp9_convolve_copy_dspr2(const uint8_t *src, ptrdiff_t src_stride,
uint32_t tp1, tp2, tp3, tp4;
uint32_t tp5, tp6, tp7, tp8;
- vp9_prefetch_load(src + 64);
- vp9_prefetch_store(dst + 32);
+ prefetch_load(src + 64);
+ prefetch_store(dst + 32);
/* 16 word storage */
for (y = h; y--; ) {
- vp9_prefetch_load(src + src_stride);
- vp9_prefetch_load(src + src_stride + 32);
- vp9_prefetch_load(src + src_stride + 64);
- vp9_prefetch_store(dst + dst_stride);
- vp9_prefetch_store(dst + dst_stride + 32);
+ prefetch_load(src + src_stride);
+ prefetch_load(src + src_stride + 32);
+ prefetch_load(src + src_stride + 64);
+ prefetch_store(dst + dst_stride);
+ prefetch_store(dst + dst_stride + 32);
__asm__ __volatile__ (
"ulw %[tp1], 0(%[src]) \n\t"
diff --git a/vp9/common/mips/dspr2/vp9_convolve8_horiz_dspr2.c b/vp9/common/mips/dspr2/vp9_convolve8_horiz_dspr2.c
index 030389615..d0e3095b3 100644
--- a/vp9/common/mips/dspr2/vp9_convolve8_horiz_dspr2.c
+++ b/vp9/common/mips/dspr2/vp9_convolve8_horiz_dspr2.c
@@ -43,9 +43,9 @@ static void convolve_horiz_4_dspr2(const uint8_t *src,
for (y = h; y--;) {
/* prefetch data to cache memory */
- vp9_prefetch_load(src + src_stride);
- vp9_prefetch_load(src + src_stride + 32);
- vp9_prefetch_store(dst + dst_stride);
+ prefetch_load(src + src_stride);
+ prefetch_load(src + src_stride + 32);
+ prefetch_store(dst + dst_stride);
__asm__ __volatile__ (
"ulw %[tp1], 0(%[src]) \n\t"
@@ -154,9 +154,9 @@ static void convolve_horiz_8_dspr2(const uint8_t *src,
for (y = h; y--;) {
/* prefetch data to cache memory */
- vp9_prefetch_load(src + src_stride);
- vp9_prefetch_load(src + src_stride + 32);
- vp9_prefetch_store(dst + dst_stride);
+ prefetch_load(src + src_stride);
+ prefetch_load(src + src_stride + 32);
+ prefetch_store(dst + dst_stride);
__asm__ __volatile__ (
"ulw %[tp1], 0(%[src]) \n\t"
@@ -323,9 +323,9 @@ static void convolve_horiz_16_dspr2(const uint8_t *src_ptr,
dst = dst_ptr;
/* prefetch data to cache memory */
- vp9_prefetch_load(src_ptr + src_stride);
- vp9_prefetch_load(src_ptr + src_stride + 32);
- vp9_prefetch_store(dst_ptr + dst_stride);
+ prefetch_load(src_ptr + src_stride);
+ prefetch_load(src_ptr + src_stride + 32);
+ prefetch_store(dst_ptr + dst_stride);
for (c = 0; c < count; c++) {
__asm__ __volatile__ (
@@ -593,11 +593,11 @@ static void convolve_horiz_64_dspr2(const uint8_t *src_ptr,
dst = dst_ptr;
/* prefetch data to cache memory */
- vp9_prefetch_load(src_ptr + src_stride);
- vp9_prefetch_load(src_ptr + src_stride + 32);
- vp9_prefetch_load(src_ptr + src_stride + 64);
- vp9_prefetch_store(dst_ptr + dst_stride);
- vp9_prefetch_store(dst_ptr + dst_stride + 32);
+ prefetch_load(src_ptr + src_stride);
+ prefetch_load(src_ptr + src_stride + 32);
+ prefetch_load(src_ptr + src_stride + 64);
+ prefetch_store(dst_ptr + dst_stride);
+ prefetch_store(dst_ptr + dst_stride + 32);
for (c = 0; c < 4; c++) {
__asm__ __volatile__ (
@@ -859,7 +859,7 @@ void vp9_convolve8_horiz_dspr2(const uint8_t *src, ptrdiff_t src_stride,
if (16 == x_step_q4) {
uint32_t pos = 38;
- vp9_prefetch_load((const uint8_t *)filter_x);
+ prefetch_load((const uint8_t *)filter_x);
src -= 3;
/* bit positon for extract from acc */
@@ -870,9 +870,9 @@ void vp9_convolve8_horiz_dspr2(const uint8_t *src, ptrdiff_t src_stride,
);
/* prefetch data to cache memory */
- vp9_prefetch_load(src);
- vp9_prefetch_load(src + 32);
- vp9_prefetch_store(dst);
+ prefetch_load(src);
+ prefetch_load(src + 32);
+ prefetch_store(dst);
switch (w) {
case 4:
@@ -896,8 +896,8 @@ void vp9_convolve8_horiz_dspr2(const uint8_t *src, ptrdiff_t src_stride,
filter_x, (int32_t)h, 2);
break;
case 64:
- vp9_prefetch_load(src + 64);
- vp9_prefetch_store(dst + 32);
+ prefetch_load(src + 64);
+ prefetch_store(dst + 32);
convolve_horiz_64_dspr2(src, (int32_t)src_stride,
dst, (int32_t)dst_stride,
diff --git a/vp9/common/mips/dspr2/vp9_convolve8_vert_dspr2.c b/vp9/common/mips/dspr2/vp9_convolve8_vert_dspr2.c
index 0930bb3d8..98acb81ba 100644
--- a/vp9/common/mips/dspr2/vp9_convolve8_vert_dspr2.c
+++ b/vp9/common/mips/dspr2/vp9_convolve8_vert_dspr2.c
@@ -49,7 +49,7 @@ static void convolve_vert_4_dspr2(const uint8_t *src,
for (y = h; y--;) {
/* prefetch data to cache memory */
- vp9_prefetch_store(dst + dst_stride);
+ prefetch_store(dst + dst_stride);
for (x = 0; x < w; x += 4) {
src_ptr = src + x;
@@ -203,8 +203,8 @@ static void convolve_vert_64_dspr2(const uint8_t *src,
for (y = h; y--;) {
/* prefetch data to cache memory */
- vp9_prefetch_store(dst + dst_stride);
- vp9_prefetch_store(dst + dst_stride + 32);
+ prefetch_store(dst + dst_stride);
+ prefetch_store(dst + dst_stride + 32);
for (x = 0; x < 64; x += 4) {
src_ptr = src + x;
@@ -358,7 +358,7 @@ void vp9_convolve8_vert_dspr2(const uint8_t *src, ptrdiff_t src_stride,
: [pos] "r" (pos)
);
- vp9_prefetch_store(dst);
+ prefetch_store(dst);
switch (w) {
case 4 :
@@ -370,7 +370,7 @@ void vp9_convolve8_vert_dspr2(const uint8_t *src, ptrdiff_t src_stride,
filter_y, w, h);
break;
case 64 :
- vp9_prefetch_store(dst + 32);
+ prefetch_store(dst + 32);
convolve_vert_64_dspr2(src, src_stride,
dst, dst_stride,
filter_y, h);
diff --git a/vp9/common/mips/dspr2/vp9_itrans16_dspr2.c b/vp9/common/mips/dspr2/vp9_itrans16_dspr2.c
index 202d91381..10a24f33d 100644
--- a/vp9/common/mips/dspr2/vp9_itrans16_dspr2.c
+++ b/vp9/common/mips/dspr2/vp9_itrans16_dspr2.c
@@ -34,7 +34,7 @@ static void idct16_rows_dspr2(const int16_t *input, int16_t *output,
for (i = no_rows; i--; ) {
/* prefetch row */
- vp9_prefetch_load((const uint8_t *)(input + 16));
+ prefetch_load((const uint8_t *)(input + 16));
__asm__ __volatile__ (
"lh %[load1], 0(%[input]) \n\t"
@@ -421,14 +421,14 @@ static void idct16_cols_add_blk_dspr2(int16_t *input, uint8_t *dest,
uint8_t *cm = vp9_ff_cropTbl;
/* prefetch vp9_ff_cropTbl */
- vp9_prefetch_load(vp9_ff_cropTbl);
- vp9_prefetch_load(vp9_ff_cropTbl + 32);
- vp9_prefetch_load(vp9_ff_cropTbl + 64);
- vp9_prefetch_load(vp9_ff_cropTbl + 96);
- vp9_prefetch_load(vp9_ff_cropTbl + 128);
- vp9_prefetch_load(vp9_ff_cropTbl + 160);
- vp9_prefetch_load(vp9_ff_cropTbl + 192);
- vp9_prefetch_load(vp9_ff_cropTbl + 224);
+ prefetch_load(vp9_ff_cropTbl);
+ prefetch_load(vp9_ff_cropTbl + 32);
+ prefetch_load(vp9_ff_cropTbl + 64);
+ prefetch_load(vp9_ff_cropTbl + 96);
+ prefetch_load(vp9_ff_cropTbl + 128);
+ prefetch_load(vp9_ff_cropTbl + 160);
+ prefetch_load(vp9_ff_cropTbl + 192);
+ prefetch_load(vp9_ff_cropTbl + 224);
for (i = 0; i < 16; ++i) {
dest_pix = (dest + i);
@@ -1124,7 +1124,7 @@ void vp9_iht16x16_256_add_dspr2(const int16_t *input, uint8_t *dest,
for (i = 0; i < 16; ++i) {
/* prefetch row */
- vp9_prefetch_load((const uint8_t *)(input + 16));
+ prefetch_load((const uint8_t *)(input + 16));
iadst16(input, outptr);
input += 16;
@@ -1144,7 +1144,7 @@ void vp9_iht16x16_256_add_dspr2(const int16_t *input, uint8_t *dest,
for (i = 0; i < 16; ++i) {
/* prefetch row */
- vp9_prefetch_load((const uint8_t *)(input + 16));
+ prefetch_load((const uint8_t *)(input + 16));
iadst16(input, outptr);
input += 16;
diff --git a/vp9/common/mips/dspr2/vp9_itrans32_cols_dspr2.c b/vp9/common/mips/dspr2/vp9_itrans32_cols_dspr2.c
index 7ceebb6d8..a25614581 100644
--- a/vp9/common/mips/dspr2/vp9_itrans32_cols_dspr2.c
+++ b/vp9/common/mips/dspr2/vp9_itrans32_cols_dspr2.c
@@ -44,14 +44,14 @@ void vp9_idct32_cols_add_blk_dspr2(int16_t *input, uint8_t *dest,
uint8_t *cm = vp9_ff_cropTbl;
/* prefetch vp9_ff_cropTbl */
- vp9_prefetch_load(vp9_ff_cropTbl);
- vp9_prefetch_load(vp9_ff_cropTbl + 32);
- vp9_prefetch_load(vp9_ff_cropTbl + 64);
- vp9_prefetch_load(vp9_ff_cropTbl + 96);
- vp9_prefetch_load(vp9_ff_cropTbl + 128);
- vp9_prefetch_load(vp9_ff_cropTbl + 160);
- vp9_prefetch_load(vp9_ff_cropTbl + 192);
- vp9_prefetch_load(vp9_ff_cropTbl + 224);
+ prefetch_load(vp9_ff_cropTbl);
+ prefetch_load(vp9_ff_cropTbl + 32);
+ prefetch_load(vp9_ff_cropTbl + 64);
+ prefetch_load(vp9_ff_cropTbl + 96);
+ prefetch_load(vp9_ff_cropTbl + 128);
+ prefetch_load(vp9_ff_cropTbl + 160);
+ prefetch_load(vp9_ff_cropTbl + 192);
+ prefetch_load(vp9_ff_cropTbl + 224);
for (i = 0; i < 32; ++i) {
dest_pix = dest + i;
diff --git a/vp9/common/mips/dspr2/vp9_itrans32_dspr2.c b/vp9/common/mips/dspr2/vp9_itrans32_dspr2.c
index 74a90b02c..dd18831fc 100644
--- a/vp9/common/mips/dspr2/vp9_itrans32_dspr2.c
+++ b/vp9/common/mips/dspr2/vp9_itrans32_dspr2.c
@@ -96,8 +96,8 @@ static void idct32_rows_dspr2(const int16_t *input, int16_t *output,
}
/* prefetch row */
- vp9_prefetch_load((const uint8_t *)(input + 32));
- vp9_prefetch_load((const uint8_t *)(input + 48));
+ prefetch_load((const uint8_t *)(input + 32));
+ prefetch_load((const uint8_t *)(input + 48));
__asm__ __volatile__ (
"lh %[load1], 2(%[input]) \n\t"
diff --git a/vp9/common/mips/dspr2/vp9_itrans4_dspr2.c b/vp9/common/mips/dspr2/vp9_itrans4_dspr2.c
index 280190a39..4e31f9fee 100644
--- a/vp9/common/mips/dspr2/vp9_itrans4_dspr2.c
+++ b/vp9/common/mips/dspr2/vp9_itrans4_dspr2.c
@@ -115,14 +115,14 @@ static void vp9_idct4_columns_add_blk_dspr2(int16_t *input, uint8_t *dest,
uint8_t *cm = vp9_ff_cropTbl;
/* prefetch vp9_ff_cropTbl */
- vp9_prefetch_load(vp9_ff_cropTbl);
- vp9_prefetch_load(vp9_ff_cropTbl + 32);
- vp9_prefetch_load(vp9_ff_cropTbl + 64);
- vp9_prefetch_load(vp9_ff_cropTbl + 96);
- vp9_prefetch_load(vp9_ff_cropTbl + 128);
- vp9_prefetch_load(vp9_ff_cropTbl + 160);
- vp9_prefetch_load(vp9_ff_cropTbl + 192);
- vp9_prefetch_load(vp9_ff_cropTbl + 224);
+ prefetch_load(vp9_ff_cropTbl);
+ prefetch_load(vp9_ff_cropTbl + 32);
+ prefetch_load(vp9_ff_cropTbl + 64);
+ prefetch_load(vp9_ff_cropTbl + 96);
+ prefetch_load(vp9_ff_cropTbl + 128);
+ prefetch_load(vp9_ff_cropTbl + 160);
+ prefetch_load(vp9_ff_cropTbl + 192);
+ prefetch_load(vp9_ff_cropTbl + 224);
for (i = 0; i < 4; ++i) {
dest_pix = (dest + i);
diff --git a/vp9/common/mips/dspr2/vp9_itrans8_dspr2.c b/vp9/common/mips/dspr2/vp9_itrans8_dspr2.c
index 04d226663..6898d569c 100644
--- a/vp9/common/mips/dspr2/vp9_itrans8_dspr2.c
+++ b/vp9/common/mips/dspr2/vp9_itrans8_dspr2.c
@@ -211,14 +211,14 @@ static void idct8_columns_add_blk_dspr2(int16_t *input, uint8_t *dest,
uint8_t *cm = vp9_ff_cropTbl;
/* prefetch vp9_ff_cropTbl */
- vp9_prefetch_load(vp9_ff_cropTbl);
- vp9_prefetch_load(vp9_ff_cropTbl + 32);
- vp9_prefetch_load(vp9_ff_cropTbl + 64);
- vp9_prefetch_load(vp9_ff_cropTbl + 96);
- vp9_prefetch_load(vp9_ff_cropTbl + 128);
- vp9_prefetch_load(vp9_ff_cropTbl + 160);
- vp9_prefetch_load(vp9_ff_cropTbl + 192);
- vp9_prefetch_load(vp9_ff_cropTbl + 224);
+ prefetch_load(vp9_ff_cropTbl);
+ prefetch_load(vp9_ff_cropTbl + 32);
+ prefetch_load(vp9_ff_cropTbl + 64);
+ prefetch_load(vp9_ff_cropTbl + 96);
+ prefetch_load(vp9_ff_cropTbl + 128);
+ prefetch_load(vp9_ff_cropTbl + 160);
+ prefetch_load(vp9_ff_cropTbl + 192);
+ prefetch_load(vp9_ff_cropTbl + 224);
for (i = 0; i < 8; ++i) {
dest_pix = (dest + i);