RepoMirrors
/
ffmpeg
mirror of https://git.ffmpeg.org/ffmpeg.git
1
0
Fork 0
Code Issues Projects Releases Wiki Activity

Merge commit '7014b65995b1fe6188fb53447bf61b08e3963355' 

* commit '7014b65995b1fe6188fb53447bf61b08e3963355':
  ppc: pixblockdsp: Use the abriged vector types

Merged-by: Michael Niedermayer <michaelni@gmx.at>
This commit is contained in:
Michael Niedermayer 2015-05-31 21:48:06 +02:00
parent e8676c758a 7014b65995
commit 08f229d81e
1 changed files with 20 additions and 23 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

43
libavcodec/ppc/pixblockdsp.c
View File

@ -67,24 +67,22 @@ static void get_pixels_altivec(int16_t *restrict block, const uint8_t *pixels,
ptrdiff_t line_size)
{
int i;
vector unsigned char perm = vec_lvsl(0, pixels);
const vector unsigned char zero =
(const vector unsigned char) vec_splat_u8(0);
vec_u8 perm = vec_lvsl(0, pixels);
const vec_u8 zero = (const vec_u8)vec_splat_u8(0);
for (i = 0; i < 8; i++) {
/* Read potentially unaligned pixels.
* We're reading 16 pixels, and actually only want 8,
* but we simply ignore the extras. */
vector unsigned char pixl = vec_ld(0, pixels);
vector unsigned char pixr = vec_ld(7, pixels);
vector unsigned char bytes = vec_perm(pixl, pixr, perm);
vec_u8 pixl = vec_ld(0, pixels);
vec_u8 pixr = vec_ld(7, pixels);
vec_u8 bytes = vec_perm(pixl, pixr, perm);
// Convert the bytes into shorts.
vector signed short shorts = (vector signed short) vec_mergeh(zero,
bytes);
vec_s16 shorts = (vec_s16)vec_mergeh(zero, bytes);
// Save the data to the block, we assume the block is 16-byte aligned.
vec_st(shorts, i * 16, (vector signed short *) block);
vec_st(shorts, i * 16, (vec_s16 *)block);
pixels += line_size;
}
@ -159,22 +157,21 @@ static void diff_pixels_altivec(int16_t *restrict block, const uint8_t *s1,
const uint8_t *s2, int stride)
{
int i;
vector unsigned char perm1 = vec_lvsl(0, s1);
vector unsigned char perm2 = vec_lvsl(0, s2);
const vector unsigned char zero =
(const vector unsigned char) vec_splat_u8(0);
vector signed short shorts1, shorts2;
vec_u8 perm1 = vec_lvsl(0, s1);
vec_u8 perm2 = vec_lvsl(0, s2);
const vec_u8 zero = (const vec_u8)vec_splat_u8(0);
vec_s16 shorts1, shorts2;
for (i = 0; i < 4; i++) {
/* Read potentially unaligned pixels.
* We're reading 16 pixels, and actually only want 8,
* but we simply ignore the extras. */
vector unsigned char pixl = vec_ld(0, s1);
vector unsigned char pixr = vec_ld(15, s1);
vector unsigned char bytes = vec_perm(pixl, pixr, perm1);
vec_u8 pixl = vec_ld(0, s1);
vec_u8 pixr = vec_ld(15, s1);
vec_u8 bytes = vec_perm(pixl, pixr, perm1);
// Convert the bytes into shorts.
shorts1 = (vector signed short) vec_mergeh(zero, bytes);
shorts1 = (vec_s16)vec_mergeh(zero, bytes);
// Do the same for the second block of pixels.
pixl = vec_ld(0, s2);
@ -182,13 +179,13 @@ static void diff_pixels_altivec(int16_t *restrict block, const uint8_t *s1,
bytes = vec_perm(pixl, pixr, perm2);
// Convert the bytes into shorts.
shorts2 = (vector signed short) vec_mergeh(zero, bytes);
shorts2 = (vec_s16)vec_mergeh(zero, bytes);
// Do the subtraction.
shorts1 = vec_sub(shorts1, shorts2);
// Save the data to the block, we assume the block is 16-byte aligned.
vec_st(shorts1, 0, (vector signed short *) block);
vec_st(shorts1, 0, (vec_s16 *)block);
s1 += stride;
s2 += stride;
@ -205,7 +202,7 @@ static void diff_pixels_altivec(int16_t *restrict block, const uint8_t *s1,
bytes = vec_perm(pixl, pixr, perm1);
// Convert the bytes into shorts.
shorts1 = (vector signed short) vec_mergeh(zero, bytes);
shorts1 = (vec_s16)vec_mergeh(zero, bytes);
// Do the same for the second block of pixels.
pixl = vec_ld(0, s2);
@ -213,13 +210,13 @@ static void diff_pixels_altivec(int16_t *restrict block, const uint8_t *s1,
bytes = vec_perm(pixl, pixr, perm2);
// Convert the bytes into shorts.
shorts2 = (vector signed short) vec_mergeh(zero, bytes);
shorts2 = (vec_s16)vec_mergeh(zero, bytes);
// Do the subtraction.
shorts1 = vec_sub(shorts1, shorts2);
// Save the data to the block, we assume the block is 16-byte aligned.
vec_st(shorts1, 0, (vector signed short *) block);
vec_st(shorts1, 0, (vec_s16 *)block);
s1 += stride;
s2 += stride;