dany/libwebp
1
0
Fork 0
mirror of https://github.com/webmproject/libwebp.git synced 2024-12-28 14:38:21 +01:00
Code Issues Actions Packages Projects Releases Wiki Activity

NEON intrinsics version of CollectHistogram

Browse Source 
apparently faster, but we might save some load/store to/from memory
once we settle for the intrinsics-based FTransform()

(also: fixed some #ifdef USE_INTRINSICS problems)

Change-Id: I426dea299cea0c64eb21c4d81a04a960e0c263c7
This commit is contained in:
skal 2014-04-14 16:41:48 +02:00
parent 8ff96a027a
commit 95203d2d1b
1 changed files with 39 additions and 0 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

39
src/dsp/enc_neon.c
View File

@ -780,6 +780,39 @@ static int Disto16x16(const uint8_t* const a, const uint8_t* const b,
return D; return D;
} }
#if defined(USE_INTRINSICS)
//------------------------------------------------------------------------------
static void CollectHistogram(const uint8_t* ref, const uint8_t* pred,
int start_block, int end_block,
VP8Histogram* const histo) {
const uint16x8_t max_coeff_thresh = vdupq_n_u16(MAX_COEFF_THRESH);
int j;
for (j = start_block; j < end_block; ++j) {
int16_t out[16];
FTransform(ref + VP8DspScan[j], pred + VP8DspScan[j], out);
{
int k;
const int16x8_t a0 = vld1q_s16(out + 0);
const int16x8_t b0 = vld1q_s16(out + 8);
const uint16x8_t a1 = vreinterpretq_u16_s16(vabsq_s16(a0));
const uint16x8_t b1 = vreinterpretq_u16_s16(vabsq_s16(b0));
const uint16x8_t a2 = vshrq_n_u16(a1, 3);
const uint16x8_t b2 = vshrq_n_u16(b1, 3);
const uint16x8_t a3 = vminq_u16(a2, max_coeff_thresh);
const uint16x8_t b3 = vminq_u16(b2, max_coeff_thresh);
vst1q_s16(out + 0, vreinterpretq_s16_u16(a3));
vst1q_s16(out + 8, vreinterpretq_s16_u16(b3));
// Convert coefficients to bin.
for (k = 0; k < 16; ++k) {
histo->distribution[out[k]]++;
}
}
}
}
//------------------------------------------------------------------------------ //------------------------------------------------------------------------------
static WEBP_INLINE void AccumulateSSE16(const uint8_t* const a, static WEBP_INLINE void AccumulateSSE16(const uint8_t* const a,
@ -848,6 +881,8 @@ static int SSE4x4(const uint8_t* a, const uint8_t* b) {
#undef LOAD_LANE_32b #undef LOAD_LANE_32b
#endif // USE_INTRINSICS
//------------------------------------------------------------------------------ //------------------------------------------------------------------------------
// Compilation with gcc4.6.3 is problematic for now. Disable this function then. // Compilation with gcc4.6.3 is problematic for now. Disable this function then.
@ -912,6 +947,7 @@ static int QuantizeBlock(int16_t in[16], int16_t out[16],
if (*(uint64_t*)(out + 12) != 0) return 1; if (*(uint64_t*)(out + 12) != 0) return 1;
return 0; return 0;
} }
#endif // SKIP_QUANTIZE #endif // SKIP_QUANTIZE
#endif // WEBP_USE_NEON #endif // WEBP_USE_NEON
@ -931,10 +967,13 @@ void VP8EncDspInitNEON(void) {
VP8TDisto4x4 = Disto4x4; VP8TDisto4x4 = Disto4x4;
VP8TDisto16x16 = Disto16x16; VP8TDisto16x16 = Disto16x16;
#if defined(USE_INTRINSICS) #if defined(USE_INTRINSICS)
VP8CollectHistogram = CollectHistogram;
VP8SSE16x16 = SSE16x16; VP8SSE16x16 = SSE16x16;
VP8SSE16x8 = SSE16x8; VP8SSE16x8 = SSE16x8;
VP8SSE8x8 = SSE8x8; VP8SSE8x8 = SSE8x8;
VP8SSE4x4 = SSE4x4; VP8SSE4x4 = SSE4x4;
#else
(void)Load4x4; // to avoid a warning
#endif #endif
#if !defined(SKIP_QUANTIZE) #if !defined(SKIP_QUANTIZE)
VP8EncQuantizeBlock = QuantizeBlock; VP8EncQuantizeBlock = QuantizeBlock;