dany/libwebp
1
0
Fork 0
mirror of https://github.com/webmproject/libwebp.git synced 2024-11-20 04:18:26 +01:00
Code Issues Actions Packages Projects Releases Wiki Activity

NEON: fix overflow in SSE NxN calculation

Browse Source 
vmlal_u8() is prone to overflow during the accumulation.
There was a mismatch happening at low q mostly. Because in this
case the distortion is important and the accumulated sum was
later than 16bit-unsigned.

Change-Id: I1a08a2f744bcdf0b26647e61b9ee92a0c2e28fe8
This commit is contained in:
Pascal Massimino 2017-01-17 11:47:36 +01:00
parent db013a8d5c
commit 9b3aca404d
1 changed files with 26 additions and 14 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

40
src/dsp/enc_neon.c
View File

@ -746,9 +746,14 @@ static WEBP_INLINE void AccumulateSSE16(const uint8_t* const a,
const uint8x16_t a0 = vld1q_u8(a);
const uint8x16_t b0 = vld1q_u8(b);
const uint8x16_t abs_diff = vabdq_u8(a0, b0);
uint16x8_t prod = vmull_u8(vget_low_u8(abs_diff), vget_low_u8(abs_diff));
prod = vmlal_u8(prod, vget_high_u8(abs_diff), vget_high_u8(abs_diff));
*sum = vpadalq_u16(*sum, prod); // pair-wise add and accumulate
const uint16x8_t prod1 = vmull_u8(vget_low_u8(abs_diff),
vget_low_u8(abs_diff));
const uint16x8_t prod2 = vmull_u8(vget_high_u8(abs_diff),
vget_high_u8(abs_diff));
/* pair-wise adds and widen */
const uint32x4_t sum1 = vpaddlq_u16(prod1);
const uint32x4_t sum2 = vpaddlq_u16(prod2);
*sum = vaddq_u32(*sum, vaddq_u32(sum1, sum2));
}
// Horizontal sum of all four uint32_t values in 'sum'.
@ -758,7 +763,7 @@ static int SumToInt(uint32x4_t sum) {
return (int)sum3;
}
static int SSE16x16(const uint8_t* a, const uint8_t* b) {
static int SSE16x16_NEON(const uint8_t* a, const uint8_t* b) {
uint32x4_t sum = vdupq_n_u32(0);
int y;
for (y = 0; y < 16; ++y) {
@ -767,7 +772,7 @@ static int SSE16x16(const uint8_t* a, const uint8_t* b) {
return SumToInt(sum);
}
static int SSE16x8(const uint8_t* a, const uint8_t* b) {
static int SSE16x8_NEON(const uint8_t* a, const uint8_t* b) {
uint32x4_t sum = vdupq_n_u32(0);
int y;
for (y = 0; y < 8; ++y) {
@ -776,7 +781,7 @@ static int SSE16x8(const uint8_t* a, const uint8_t* b) {
return SumToInt(sum);
}
static int SSE8x8(const uint8_t* a, const uint8_t* b) {
static int SSE8x8_NEON(const uint8_t* a, const uint8_t* b) {
uint32x4_t sum = vdupq_n_u32(0);
int y;
for (y = 0; y < 8; ++y) {
@ -789,13 +794,18 @@ static int SSE8x8(const uint8_t* a, const uint8_t* b) {
return SumToInt(sum);
}
static int SSE4x4(const uint8_t* a, const uint8_t* b) {
static int SSE4x4_NEON(const uint8_t* a, const uint8_t* b) {
const uint8x16_t a0 = Load4x4(a);
const uint8x16_t b0 = Load4x4(b);
const uint8x16_t abs_diff = vabdq_u8(a0, b0);
uint16x8_t prod = vmull_u8(vget_low_u8(abs_diff), vget_low_u8(abs_diff));
prod = vmlal_u8(prod, vget_high_u8(abs_diff), vget_high_u8(abs_diff));
return SumToInt(vpaddlq_u16(prod));
const uint16x8_t prod1 = vmull_u8(vget_low_u8(abs_diff),
vget_low_u8(abs_diff));
const uint16x8_t prod2 = vmull_u8(vget_high_u8(abs_diff),
vget_high_u8(abs_diff));
/* pair-wise adds and widen */
const uint32x4_t sum1 = vpaddlq_u16(prod1);
const uint32x4_t sum2 = vpaddlq_u16(prod2);
return SumToInt(vaddq_u32(sum1, sum2));
}
//------------------------------------------------------------------------------
@ -903,10 +913,12 @@ WEBP_TSAN_IGNORE_FUNCTION void VP8EncDspInitNEON(void) {
VP8TDisto4x4 = Disto4x4;
VP8TDisto16x16 = Disto16x16;
VP8CollectHistogram = CollectHistogram;
VP8SSE16x16 = SSE16x16;
VP8SSE16x8 = SSE16x8;
VP8SSE8x8 = SSE8x8;
VP8SSE4x4 = SSE4x4;
VP8SSE16x16 = SSE16x16_NEON;
VP8SSE16x8 = SSE16x8_NEON;
VP8SSE8x8 = SSE8x8_NEON;
VP8SSE4x4 = SSE4x4_NEON;
#if !defined(WORK_AROUND_GCC)
VP8EncQuantizeBlock = QuantizeBlock;
VP8EncQuantize2Blocks = Quantize2Blocks;