Skip to content
GitLab
Commit e404db44 authored by Johann's avatar Johann Committed by Gerrit Code Review
Browse files

Merge "Remove Windows-style newlines using dos2unix"

parents 9b05f23e 67cf81b1
v1.14.0-linphone 1.4.X feature/update_to_v1.9.0-linphone feature/uwp_nuget forest frame_parallel highbitdepth indianrunnerduck javanwhistlingduck khakicampbell linphone linphone-android linphone-old longtailedduck m49-2623 m52-2743 m54-2840 m56-2924 m66-3359 m68-3440 mandarinduck mcw mcw2 nextgen nextgenv2 playground sandbox/Jingning/experimental sandbox/Jingning/transcode sandbox/Jingning/vpx sandbox/aconverse@google.com/ansbench sandbox/debargha/playground sandbox/hkuang/frame_parallel sandbox/hkuang@google.com/decode sandbox/jimbankoski@google.com/proposed-aom sandbox/jingning@google.com/decoder_test_suite sandbox/jingning@google.com/experimental sandbox/jzern@google.com/test sandbox/wangch@google.com/vp9 sandbox/yaowu@google.com/mergeaom v1.12.0-linphone v1.6.1_linphone v1.7.0-linphone v1.9.0-linphone v1.9.0 v1.9.0-rc1 v1.8.2 v1.8.1 v1.8.0 v1.7.0 v1.6.1 v1.6.0 v1.5.0 v1.4.0 v1.3.0
No related merge requests found
Expand all Hide whitespace changes
Inline Side-by-side
Showing
with 874 additions and 874 deletions
vp9/common/mips/dspr2/vp9_convolve2_dspr2.c
+ 688
− 688
View file @ e404db44
This diff is collapsed.
vp9/common/mips/dspr2/vp9_convolve2_vert_dspr2.c
+ 186
− 186
View file @ e404db44
......@@ -20,103 +20,103 @@
#include "vp9/common/mips/dspr2/vp9_common_dspr2.h"
#if HAVE_DSPR2
static void convolve_bi_vert_4_dspr2(const uint8_t *src,
int32_t src_stride,
uint8_t *dst,
int32_t dst_stride,
const int16_t *filter_y,
int32_t w,
int32_t h) {
int32_t x, y;
const uint8_t *src_ptr;
uint8_t *dst_ptr;
uint8_t *cm = vp9_ff_cropTbl;
uint32_t vector4a = 64;
uint32_t load1, load2;
uint32_t p1, p2;
uint32_t scratch1;
uint32_t store1, store2;
int32_t Temp1, Temp2;
const int16_t *filter = &filter_y[3];
uint32_t filter45;
filter45 = ((const int32_t *)filter)[0];
for (y = h; y--;) {
/* prefetch data to cache memory */
vp9_prefetch_store(dst + dst_stride);
for (x = 0; x < w; x += 4) {
src_ptr = src + x;
dst_ptr = dst + x;
__asm__ __volatile__ (
"ulw %[load1], 0(%[src_ptr]) \n\t"
"add %[src_ptr], %[src_ptr], %[src_stride] \n\t"
"ulw %[load2], 0(%[src_ptr]) \n\t"
"mtlo %[vector4a], $ac0 \n\t"
"mtlo %[vector4a], $ac1 \n\t"
"mtlo %[vector4a], $ac2 \n\t"
"mtlo %[vector4a], $ac3 \n\t"
"mthi $zero, $ac0 \n\t"
"mthi $zero, $ac1 \n\t"
"mthi $zero, $ac2 \n\t"
"mthi $zero, $ac3 \n\t"
"preceu.ph.qbr %[scratch1], %[load1] \n\t"
"preceu.ph.qbr %[p1], %[load2] \n\t"
"precrq.ph.w %[p2], %[p1], %[scratch1] \n\t" /* pixel 2 */
"append %[p1], %[scratch1], 16 \n\t" /* pixel 1 */
"dpa.w.ph $ac0, %[p1], %[filter45] \n\t"
"dpa.w.ph $ac1, %[p2], %[filter45] \n\t"
"preceu.ph.qbl %[scratch1], %[load1] \n\t"
"preceu.ph.qbl %[p1], %[load2] \n\t"
"precrq.ph.w %[p2], %[p1], %[scratch1] \n\t" /* pixel 2 */
"append %[p1], %[scratch1], 16 \n\t" /* pixel 1 */
"dpa.w.ph $ac2, %[p1], %[filter45] \n\t"
"dpa.w.ph $ac3, %[p2], %[filter45] \n\t"
"extp %[Temp1], $ac0, 31 \n\t"
"extp %[Temp2], $ac1, 31 \n\t"
"lbux %[store1], %[Temp1](%[cm]) \n\t"
"extp %[Temp1], $ac2, 31 \n\t"
"lbux %[store2], %[Temp2](%[cm]) \n\t"
"extp %[Temp2], $ac3, 31 \n\t"
"sb %[store1], 0(%[dst_ptr]) \n\t"
"sb %[store2], 1(%[dst_ptr]) \n\t"
"lbux %[store1], %[Temp1](%[cm]) \n\t"
"lbux %[store2], %[Temp2](%[cm]) \n\t"
"sb %[store1], 2(%[dst_ptr]) \n\t"
"sb %[store2], 3(%[dst_ptr]) \n\t"
: [load1] "=&r" (load1), [load2] "=&r" (load2),
[p1] "=&r" (p1), [p2] "=&r" (p2),
[scratch1] "=&r" (scratch1),
[Temp1] "=&r" (Temp1), [Temp2] "=&r" (Temp2),
[store1] "=&r" (store1), [store2] "=&r" (store2),
[src_ptr] "+r" (src_ptr)
: [filter45] "r" (filter45),[vector4a] "r" (vector4a),
[src_stride] "r" (src_stride),
[cm] "r" (cm), [dst_ptr] "r" (dst_ptr)
);
}
/* Next row... */
src += src_stride;
dst += dst_stride;
}
}
static void convolve_bi_vert_4_dspr2(const uint8_t *src,
int32_t src_stride,
uint8_t *dst,
int32_t dst_stride,
const int16_t *filter_y,
int32_t w,
int32_t h) {
int32_t x, y;
const uint8_t *src_ptr;
uint8_t *dst_ptr;
uint8_t *cm = vp9_ff_cropTbl;
uint32_t vector4a = 64;
uint32_t load1, load2;
uint32_t p1, p2;
uint32_t scratch1;
uint32_t store1, store2;
int32_t Temp1, Temp2;
const int16_t *filter = &filter_y[3];
uint32_t filter45;
filter45 = ((const int32_t *)filter)[0];
for (y = h; y--;) {
/* prefetch data to cache memory */
vp9_prefetch_store(dst + dst_stride);
for (x = 0; x < w; x += 4) {
src_ptr = src + x;
dst_ptr = dst + x;
__asm__ __volatile__ (
"ulw %[load1], 0(%[src_ptr]) \n\t"
"add %[src_ptr], %[src_ptr], %[src_stride] \n\t"
"ulw %[load2], 0(%[src_ptr]) \n\t"
"mtlo %[vector4a], $ac0 \n\t"
"mtlo %[vector4a], $ac1 \n\t"
"mtlo %[vector4a], $ac2 \n\t"
"mtlo %[vector4a], $ac3 \n\t"
"mthi $zero, $ac0 \n\t"
"mthi $zero, $ac1 \n\t"
"mthi $zero, $ac2 \n\t"
"mthi $zero, $ac3 \n\t"
"preceu.ph.qbr %[scratch1], %[load1] \n\t"
"preceu.ph.qbr %[p1], %[load2] \n\t"
"precrq.ph.w %[p2], %[p1], %[scratch1] \n\t" /* pixel 2 */
"append %[p1], %[scratch1], 16 \n\t" /* pixel 1 */
"dpa.w.ph $ac0, %[p1], %[filter45] \n\t"
"dpa.w.ph $ac1, %[p2], %[filter45] \n\t"
"preceu.ph.qbl %[scratch1], %[load1] \n\t"
"preceu.ph.qbl %[p1], %[load2] \n\t"
"precrq.ph.w %[p2], %[p1], %[scratch1] \n\t" /* pixel 2 */
"append %[p1], %[scratch1], 16 \n\t" /* pixel 1 */
"dpa.w.ph $ac2, %[p1], %[filter45] \n\t"
"dpa.w.ph $ac3, %[p2], %[filter45] \n\t"
"extp %[Temp1], $ac0, 31 \n\t"
"extp %[Temp2], $ac1, 31 \n\t"
"lbux %[store1], %[Temp1](%[cm]) \n\t"
"extp %[Temp1], $ac2, 31 \n\t"
"lbux %[store2], %[Temp2](%[cm]) \n\t"
"extp %[Temp2], $ac3, 31 \n\t"
"sb %[store1], 0(%[dst_ptr]) \n\t"
"sb %[store2], 1(%[dst_ptr]) \n\t"
"lbux %[store1], %[Temp1](%[cm]) \n\t"
"lbux %[store2], %[Temp2](%[cm]) \n\t"
"sb %[store1], 2(%[dst_ptr]) \n\t"
"sb %[store2], 3(%[dst_ptr]) \n\t"
: [load1] "=&r" (load1), [load2] "=&r" (load2),
[p1] "=&r" (p1), [p2] "=&r" (p2),
[scratch1] "=&r" (scratch1),
[Temp1] "=&r" (Temp1), [Temp2] "=&r" (Temp2),
[store1] "=&r" (store1), [store2] "=&r" (store2),
[src_ptr] "+r" (src_ptr)
: [filter45] "r" (filter45),[vector4a] "r" (vector4a),
[src_stride] "r" (src_stride),
[cm] "r" (cm), [dst_ptr] "r" (dst_ptr)
);
}
/* Next row... */
src += src_stride;
dst += dst_stride;
}
}
static void convolve_bi_vert_64_dspr2(const uint8_t *src,
int32_t src_stride,
......@@ -124,95 +124,95 @@ static void convolve_bi_vert_64_dspr2(const uint8_t *src,
int32_t dst_stride,
const int16_t *filter_y,
int32_t h) {
int32_t x, y;
const uint8_t *src_ptr;
uint8_t *dst_ptr;
uint8_t *cm = vp9_ff_cropTbl;
uint32_t vector4a = 64;
uint32_t load1, load2;
uint32_t p1, p2;
uint32_t scratch1;
uint32_t store1, store2;
int32_t Temp1, Temp2;
const int16_t *filter = &filter_y[3];
uint32_t filter45;
filter45 = ((const int32_t *)filter)[0];
for (y = h; y--;) {
/* prefetch data to cache memory */
vp9_prefetch_store(dst + dst_stride);
for (x = 0; x < 64; x += 4) {
src_ptr = src + x;
dst_ptr = dst + x;
__asm__ __volatile__ (
"ulw %[load1], 0(%[src_ptr]) \n\t"
"add %[src_ptr], %[src_ptr], %[src_stride] \n\t"
"ulw %[load2], 0(%[src_ptr]) \n\t"
"mtlo %[vector4a], $ac0 \n\t"
"mtlo %[vector4a], $ac1 \n\t"
"mtlo %[vector4a], $ac2 \n\t"
"mtlo %[vector4a], $ac3 \n\t"
"mthi $zero, $ac0 \n\t"
"mthi $zero, $ac1 \n\t"
"mthi $zero, $ac2 \n\t"
"mthi $zero, $ac3 \n\t"
"preceu.ph.qbr %[scratch1], %[load1] \n\t"
"preceu.ph.qbr %[p1], %[load2] \n\t"
"precrq.ph.w %[p2], %[p1], %[scratch1] \n\t" /* pixel 2 */
"append %[p1], %[scratch1], 16 \n\t" /* pixel 1 */
"dpa.w.ph $ac0, %[p1], %[filter45] \n\t"
"dpa.w.ph $ac1, %[p2], %[filter45] \n\t"
"preceu.ph.qbl %[scratch1], %[load1] \n\t"
"preceu.ph.qbl %[p1], %[load2] \n\t"
"precrq.ph.w %[p2], %[p1], %[scratch1] \n\t" /* pixel 2 */
"append %[p1], %[scratch1], 16 \n\t" /* pixel 1 */
"dpa.w.ph $ac2, %[p1], %[filter45] \n\t"
"dpa.w.ph $ac3, %[p2], %[filter45] \n\t"
"extp %[Temp1], $ac0, 31 \n\t"
"extp %[Temp2], $ac1, 31 \n\t"
"lbux %[store1], %[Temp1](%[cm]) \n\t"
"extp %[Temp1], $ac2, 31 \n\t"
"lbux %[store2], %[Temp2](%[cm]) \n\t"
"extp %[Temp2], $ac3, 31 \n\t"
"sb %[store1], 0(%[dst_ptr]) \n\t"
"sb %[store2], 1(%[dst_ptr]) \n\t"
"lbux %[store1], %[Temp1](%[cm]) \n\t"
"lbux %[store2], %[Temp2](%[cm]) \n\t"
"sb %[store1], 2(%[dst_ptr]) \n\t"
"sb %[store2], 3(%[dst_ptr]) \n\t"
: [load1] "=&r" (load1), [load2] "=&r" (load2),
[p1] "=&r" (p1), [p2] "=&r" (p2),
[scratch1] "=&r" (scratch1),
[Temp1] "=&r" (Temp1), [Temp2] "=&r" (Temp2),
[store1] "=&r" (store1), [store2] "=&r" (store2),
[src_ptr] "+r" (src_ptr)
: [filter45] "r" (filter45),[vector4a] "r" (vector4a),
[src_stride] "r" (src_stride),
[cm] "r" (cm), [dst_ptr] "r" (dst_ptr)
);
}
/* Next row... */
src += src_stride;
dst += dst_stride;
}
int32_t x, y;
const uint8_t *src_ptr;
uint8_t *dst_ptr;
uint8_t *cm = vp9_ff_cropTbl;
uint32_t vector4a = 64;
uint32_t load1, load2;
uint32_t p1, p2;
uint32_t scratch1;
uint32_t store1, store2;
int32_t Temp1, Temp2;
const int16_t *filter = &filter_y[3];
uint32_t filter45;
filter45 = ((const int32_t *)filter)[0];
for (y = h; y--;) {
/* prefetch data to cache memory */
vp9_prefetch_store(dst + dst_stride);
for (x = 0; x < 64; x += 4) {
src_ptr = src + x;
dst_ptr = dst + x;
__asm__ __volatile__ (
"ulw %[load1], 0(%[src_ptr]) \n\t"
"add %[src_ptr], %[src_ptr], %[src_stride] \n\t"
"ulw %[load2], 0(%[src_ptr]) \n\t"
"mtlo %[vector4a], $ac0 \n\t"
"mtlo %[vector4a], $ac1 \n\t"
"mtlo %[vector4a], $ac2 \n\t"
"mtlo %[vector4a], $ac3 \n\t"
"mthi $zero, $ac0 \n\t"
"mthi $zero, $ac1 \n\t"
"mthi $zero, $ac2 \n\t"
"mthi $zero, $ac3 \n\t"
"preceu.ph.qbr %[scratch1], %[load1] \n\t"
"preceu.ph.qbr %[p1], %[load2] \n\t"
"precrq.ph.w %[p2], %[p1], %[scratch1] \n\t" /* pixel 2 */
"append %[p1], %[scratch1], 16 \n\t" /* pixel 1 */
"dpa.w.ph $ac0, %[p1], %[filter45] \n\t"
"dpa.w.ph $ac1, %[p2], %[filter45] \n\t"
"preceu.ph.qbl %[scratch1], %[load1] \n\t"
"preceu.ph.qbl %[p1], %[load2] \n\t"
"precrq.ph.w %[p2], %[p1], %[scratch1] \n\t" /* pixel 2 */
"append %[p1], %[scratch1], 16 \n\t" /* pixel 1 */
"dpa.w.ph $ac2, %[p1], %[filter45] \n\t"
"dpa.w.ph $ac3, %[p2], %[filter45] \n\t"
"extp %[Temp1], $ac0, 31 \n\t"
"extp %[Temp2], $ac1, 31 \n\t"
"lbux %[store1], %[Temp1](%[cm]) \n\t"
"extp %[Temp1], $ac2, 31 \n\t"
"lbux %[store2], %[Temp2](%[cm]) \n\t"
"extp %[Temp2], $ac3, 31 \n\t"
"sb %[store1], 0(%[dst_ptr]) \n\t"
"sb %[store2], 1(%[dst_ptr]) \n\t"
"lbux %[store1], %[Temp1](%[cm]) \n\t"
"lbux %[store2], %[Temp2](%[cm]) \n\t"
"sb %[store1], 2(%[dst_ptr]) \n\t"
"sb %[store2], 3(%[dst_ptr]) \n\t"
: [load1] "=&r" (load1), [load2] "=&r" (load2),
[p1] "=&r" (p1), [p2] "=&r" (p2),
[scratch1] "=&r" (scratch1),
[Temp1] "=&r" (Temp1), [Temp2] "=&r" (Temp2),
[store1] "=&r" (store1), [store2] "=&r" (store2),
[src_ptr] "+r" (src_ptr)
: [filter45] "r" (filter45),[vector4a] "r" (vector4a),
[src_stride] "r" (src_stride),
[cm] "r" (cm), [dst_ptr] "r" (dst_ptr)
);
}
/* Next row... */
src += src_stride;
dst += dst_stride;
}
}
void vp9_convolve2_vert_dspr2(const uint8_t *src, ptrdiff_t src_stride,
......
Supports Markdown
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment