2011-11-10 15:18:07 +04:00
|
|
|
/*
|
2012-10-09 07:02:04 +04:00
|
|
|
FreeRDP: A Remote Desktop Protocol Implementation
|
2011-11-10 15:18:07 +04:00
|
|
|
RemoteFX Codec Library - NEON Optimizations
|
|
|
|
|
2013-12-04 14:37:57 +04:00
|
|
|
Copyright 2011 Martin Fleisz <martin.fleisz@thincast.com>
|
2011-11-10 15:18:07 +04:00
|
|
|
|
|
|
|
Licensed under the Apache License, Version 2.0 (the "License");
|
|
|
|
you may not use this file except in compliance with the License.
|
|
|
|
You may obtain a copy of the License at
|
|
|
|
|
|
|
|
http://www.apache.org/licenses/LICENSE-2.0
|
|
|
|
|
|
|
|
Unless required by applicable law or agreed to in writing, software
|
|
|
|
distributed under the License is distributed on an "AS IS" BASIS,
|
|
|
|
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
|
|
|
|
See the License for the specific language governing permissions and
|
|
|
|
limitations under the License.
|
|
|
|
*/
|
|
|
|
|
2012-08-15 01:09:01 +04:00
|
|
|
#ifdef HAVE_CONFIG_H
|
|
|
|
#include "config.h"
|
|
|
|
#endif
|
|
|
|
|
2011-11-10 15:18:07 +04:00
|
|
|
#if defined(__ARM_NEON__)
|
|
|
|
|
|
|
|
#include <stdio.h>
|
|
|
|
#include <stdlib.h>
|
|
|
|
#include <string.h>
|
|
|
|
#include <arm_neon.h>
|
2013-02-27 19:04:45 +04:00
|
|
|
#include <winpr/sysinfo.h>
|
2011-11-10 15:18:07 +04:00
|
|
|
|
|
|
|
#include "rfx_types.h"
|
|
|
|
#include "rfx_neon.h"
|
|
|
|
|
2013-01-19 02:32:58 +04:00
|
|
|
/* rfx_decode_YCbCr_to_RGB_NEON code now resides in the primitives library. */
|
2011-11-10 15:18:07 +04:00
|
|
|
|
|
|
|
static __inline void __attribute__((__gnu_inline__, __always_inline__, __artificial__))
|
2012-10-09 11:26:39 +04:00
|
|
|
rfx_quantization_decode_block_NEON(INT16 * buffer, const int buffer_size, const UINT32 factor)
|
2011-11-10 15:18:07 +04:00
|
|
|
{
|
2013-02-21 19:08:46 +04:00
|
|
|
int16x8_t quantFactors = vdupq_n_s16(factor);
|
2011-11-10 15:18:07 +04:00
|
|
|
int16x8_t* buf = (int16x8_t*)buffer;
|
|
|
|
int16x8_t* buf_end = (int16x8_t*)(buffer + buffer_size);
|
|
|
|
|
|
|
|
do
|
|
|
|
{
|
2012-10-09 11:01:37 +04:00
|
|
|
int16x8_t val = vld1q_s16((INT16*)buf);
|
2011-11-10 15:18:07 +04:00
|
|
|
val = vshlq_s16(val, quantFactors);
|
2012-10-09 11:01:37 +04:00
|
|
|
vst1q_s16((INT16*)buf, val);
|
2011-11-10 15:18:07 +04:00
|
|
|
buf++;
|
|
|
|
}
|
|
|
|
while(buf < buf_end);
|
|
|
|
}
|
|
|
|
|
|
|
|
void
|
2012-10-09 11:26:39 +04:00
|
|
|
rfx_quantization_decode_NEON(INT16 * buffer, const UINT32 * quantization_values)
|
2011-11-10 15:18:07 +04:00
|
|
|
{
|
2013-02-21 19:08:46 +04:00
|
|
|
rfx_quantization_decode_block_NEON(buffer, 4096, 5);
|
|
|
|
|
|
|
|
rfx_quantization_decode_block_NEON(buffer, 1024, quantization_values[8] - 6); /* HL1 */
|
|
|
|
rfx_quantization_decode_block_NEON(buffer + 1024, 1024, quantization_values[7] - 6); /* LH1 */
|
|
|
|
rfx_quantization_decode_block_NEON(buffer + 2048, 1024, quantization_values[9] - 6); /* HH1 */
|
|
|
|
rfx_quantization_decode_block_NEON(buffer + 3072, 256, quantization_values[5] - 6); /* HL2 */
|
|
|
|
rfx_quantization_decode_block_NEON(buffer + 3328, 256, quantization_values[4] - 6); /* LH2 */
|
|
|
|
rfx_quantization_decode_block_NEON(buffer + 3584, 256, quantization_values[6] - 6); /* HH2 */
|
|
|
|
rfx_quantization_decode_block_NEON(buffer + 3840, 64, quantization_values[2] - 6); /* HL3 */
|
|
|
|
rfx_quantization_decode_block_NEON(buffer + 3904, 64, quantization_values[1] - 6); /* LH3 */
|
|
|
|
rfx_quantization_decode_block_NEON(buffer + 3968, 64, quantization_values[3] - 6); /* HH3 */
|
|
|
|
rfx_quantization_decode_block_NEON(buffer + 4032, 64, quantization_values[0] - 6); /* LL3 */
|
2011-11-10 15:18:07 +04:00
|
|
|
}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
static __inline void __attribute__((__gnu_inline__, __always_inline__, __artificial__))
|
2012-10-09 11:01:37 +04:00
|
|
|
rfx_dwt_2d_decode_block_horiz_NEON(INT16 * l, INT16 * h, INT16 * dst, int subband_width)
|
2011-11-10 15:18:07 +04:00
|
|
|
{
|
|
|
|
int y, n;
|
2012-10-09 11:01:37 +04:00
|
|
|
INT16 * l_ptr = l;
|
|
|
|
INT16 * h_ptr = h;
|
|
|
|
INT16 * dst_ptr = dst;
|
2011-11-10 15:18:07 +04:00
|
|
|
|
|
|
|
for (y = 0; y < subband_width; y++)
|
|
|
|
{
|
|
|
|
/* Even coefficients */
|
|
|
|
for (n = 0; n < subband_width; n+=8)
|
|
|
|
{
|
|
|
|
// dst[2n] = l[n] - ((h[n-1] + h[n] + 1) >> 1);
|
|
|
|
int16x8_t l_n = vld1q_s16(l_ptr);
|
|
|
|
|
|
|
|
int16x8_t h_n = vld1q_s16(h_ptr);
|
|
|
|
int16x8_t h_n_m = vld1q_s16(h_ptr - 1);
|
|
|
|
|
|
|
|
if (n == 0)
|
|
|
|
{
|
|
|
|
int16_t first = vgetq_lane_s16(h_n_m, 1);
|
|
|
|
h_n_m = vsetq_lane_s16(first, h_n_m, 0);
|
|
|
|
}
|
|
|
|
|
|
|
|
int16x8_t tmp_n = vaddq_s16(h_n, h_n_m);
|
|
|
|
tmp_n = vaddq_s16(tmp_n, vdupq_n_s16(1));
|
|
|
|
tmp_n = vshrq_n_s16(tmp_n, 1);
|
|
|
|
|
|
|
|
int16x8_t dst_n = vsubq_s16(l_n, tmp_n);
|
|
|
|
|
|
|
|
vst1q_s16(l_ptr, dst_n);
|
|
|
|
|
|
|
|
l_ptr+=8;
|
|
|
|
h_ptr+=8;
|
|
|
|
}
|
|
|
|
l_ptr -= subband_width;
|
|
|
|
h_ptr -= subband_width;
|
|
|
|
|
|
|
|
/* Odd coefficients */
|
|
|
|
for (n = 0; n < subband_width; n+=8)
|
|
|
|
{
|
|
|
|
// dst[2n + 1] = (h[n] << 1) + ((dst[2n] + dst[2n + 2]) >> 1);
|
|
|
|
|
|
|
|
int16x8_t h_n = vld1q_s16(h_ptr);
|
|
|
|
|
|
|
|
h_n = vshlq_n_s16(h_n, 1);
|
|
|
|
|
|
|
|
int16x8x2_t dst_n;
|
|
|
|
dst_n.val[0] = vld1q_s16(l_ptr);
|
|
|
|
int16x8_t dst_n_p = vld1q_s16(l_ptr + 1);
|
|
|
|
if (n == subband_width - 8)
|
|
|
|
{
|
|
|
|
int16_t last = vgetq_lane_s16(dst_n_p, 6);
|
|
|
|
dst_n_p = vsetq_lane_s16(last, dst_n_p, 7);
|
|
|
|
}
|
|
|
|
|
|
|
|
dst_n.val[1] = vaddq_s16(dst_n_p, dst_n.val[0]);
|
|
|
|
dst_n.val[1] = vshrq_n_s16(dst_n.val[1], 1);
|
|
|
|
|
|
|
|
dst_n.val[1] = vaddq_s16(dst_n.val[1], h_n);
|
|
|
|
|
|
|
|
vst2q_s16(dst_ptr, dst_n);
|
|
|
|
|
|
|
|
l_ptr+=8;
|
|
|
|
h_ptr+=8;
|
|
|
|
dst_ptr+=16;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
static __inline void __attribute__((__gnu_inline__, __always_inline__, __artificial__))
|
2012-10-09 11:01:37 +04:00
|
|
|
rfx_dwt_2d_decode_block_vert_NEON(INT16 * l, INT16 * h, INT16 * dst, int subband_width)
|
2011-11-10 15:18:07 +04:00
|
|
|
{
|
|
|
|
int x, n;
|
2012-10-09 11:01:37 +04:00
|
|
|
INT16 * l_ptr = l;
|
|
|
|
INT16 * h_ptr = h;
|
|
|
|
INT16 * dst_ptr = dst;
|
2011-11-10 15:18:07 +04:00
|
|
|
|
|
|
|
int total_width = subband_width + subband_width;
|
|
|
|
|
|
|
|
/* Even coefficients */
|
|
|
|
for (n = 0; n < subband_width; n++)
|
|
|
|
{
|
|
|
|
for (x = 0; x < total_width; x+=8)
|
|
|
|
{
|
|
|
|
// dst[2n] = l[n] - ((h[n-1] + h[n] + 1) >> 1);
|
|
|
|
|
|
|
|
int16x8_t l_n = vld1q_s16(l_ptr);
|
|
|
|
int16x8_t h_n = vld1q_s16(h_ptr);
|
|
|
|
|
|
|
|
int16x8_t tmp_n = vaddq_s16(h_n, vdupq_n_s16(1));;
|
|
|
|
if (n == 0)
|
|
|
|
tmp_n = vaddq_s16(tmp_n, h_n);
|
|
|
|
else
|
|
|
|
{
|
|
|
|
int16x8_t h_n_m = vld1q_s16((h_ptr - total_width));
|
|
|
|
tmp_n = vaddq_s16(tmp_n, h_n_m);
|
|
|
|
}
|
|
|
|
tmp_n = vshrq_n_s16(tmp_n, 1);
|
|
|
|
|
|
|
|
int16x8_t dst_n = vsubq_s16(l_n, tmp_n);
|
|
|
|
vst1q_s16(dst_ptr, dst_n);
|
|
|
|
|
|
|
|
l_ptr+=8;
|
|
|
|
h_ptr+=8;
|
|
|
|
dst_ptr+=8;
|
|
|
|
}
|
|
|
|
dst_ptr+=total_width;
|
|
|
|
}
|
|
|
|
|
|
|
|
h_ptr = h;
|
|
|
|
dst_ptr = dst + total_width;
|
|
|
|
|
|
|
|
/* Odd coefficients */
|
|
|
|
for (n = 0; n < subband_width; n++)
|
|
|
|
{
|
|
|
|
for (x = 0; x < total_width; x+=8)
|
|
|
|
{
|
|
|
|
// dst[2n + 1] = (h[n] << 1) + ((dst[2n] + dst[2n + 2]) >> 1);
|
|
|
|
int16x8_t h_n = vld1q_s16(h_ptr);
|
|
|
|
int16x8_t dst_n_m = vld1q_s16(dst_ptr - total_width);
|
|
|
|
|
|
|
|
h_n = vshlq_n_s16(h_n, 1);
|
|
|
|
|
|
|
|
int16x8_t tmp_n = dst_n_m;
|
|
|
|
if (n == subband_width - 1)
|
|
|
|
tmp_n = vaddq_s16(tmp_n, dst_n_m);
|
|
|
|
else
|
|
|
|
{
|
|
|
|
int16x8_t dst_n_p = vld1q_s16((dst_ptr + total_width));
|
|
|
|
tmp_n = vaddq_s16(tmp_n, dst_n_p);
|
|
|
|
}
|
|
|
|
tmp_n = vshrq_n_s16(tmp_n, 1);
|
|
|
|
|
|
|
|
int16x8_t dst_n = vaddq_s16(tmp_n, h_n);
|
|
|
|
vst1q_s16(dst_ptr, dst_n);
|
|
|
|
|
|
|
|
h_ptr+=8;
|
|
|
|
dst_ptr+=8;
|
|
|
|
}
|
|
|
|
dst_ptr+=total_width;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
static __inline void __attribute__((__gnu_inline__, __always_inline__, __artificial__))
|
2012-10-09 11:01:37 +04:00
|
|
|
rfx_dwt_2d_decode_block_NEON(INT16 * buffer, INT16 * idwt, int subband_width)
|
2011-11-10 15:18:07 +04:00
|
|
|
{
|
2012-10-09 11:01:37 +04:00
|
|
|
INT16 * hl, * lh, * hh, * ll;
|
|
|
|
INT16 * l_dst, * h_dst;
|
2011-11-10 15:18:07 +04:00
|
|
|
|
|
|
|
/* Inverse DWT in horizontal direction, results in 2 sub-bands in L, H order in tmp buffer idwt. */
|
|
|
|
/* The 4 sub-bands are stored in HL(0), LH(1), HH(2), LL(3) order. */
|
|
|
|
/* The lower part L uses LL(3) and HL(0). */
|
|
|
|
/* The higher part H uses LH(1) and HH(2). */
|
|
|
|
|
|
|
|
ll = buffer + subband_width * subband_width * 3;
|
|
|
|
hl = buffer;
|
|
|
|
l_dst = idwt;
|
|
|
|
|
|
|
|
rfx_dwt_2d_decode_block_horiz_NEON(ll, hl, l_dst, subband_width);
|
|
|
|
|
|
|
|
lh = buffer + subband_width * subband_width;
|
|
|
|
hh = buffer + subband_width * subband_width * 2;
|
|
|
|
h_dst = idwt + subband_width * subband_width * 2;
|
|
|
|
|
|
|
|
rfx_dwt_2d_decode_block_horiz_NEON(lh, hh, h_dst, subband_width);
|
|
|
|
|
|
|
|
/* Inverse DWT in vertical direction, results are stored in original buffer. */
|
|
|
|
rfx_dwt_2d_decode_block_vert_NEON(l_dst, h_dst, buffer, subband_width);
|
|
|
|
}
|
|
|
|
|
2012-10-23 03:01:19 +04:00
|
|
|
void rfx_dwt_2d_decode_NEON(INT16 * buffer, INT16 * dwt_buffer)
|
2011-11-10 15:18:07 +04:00
|
|
|
{
|
|
|
|
rfx_dwt_2d_decode_block_NEON(buffer + 3840, dwt_buffer, 8);
|
|
|
|
rfx_dwt_2d_decode_block_NEON(buffer + 3072, dwt_buffer, 16);
|
|
|
|
rfx_dwt_2d_decode_block_NEON(buffer, dwt_buffer, 32);
|
|
|
|
}
|
|
|
|
|
|
|
|
void rfx_init_neon(RFX_CONTEXT * context)
|
|
|
|
{
|
2013-02-27 13:59:06 +04:00
|
|
|
if (IsProcessorFeaturePresent(PF_ARM_NEON_INSTRUCTIONS_AVAILABLE))
|
2011-11-10 15:18:07 +04:00
|
|
|
{
|
|
|
|
DEBUG_RFX("Using NEON optimizations");
|
|
|
|
|
2013-01-19 02:32:58 +04:00
|
|
|
IF_PROFILER(context->priv->prof_rfx_ycbcr_to_rgb->name = "rfx_decode_YCbCr_to_RGB_NEON");
|
2011-11-10 15:18:07 +04:00
|
|
|
IF_PROFILER(context->priv->prof_rfx_quantization_decode->name = "rfx_quantization_decode_NEON");
|
|
|
|
IF_PROFILER(context->priv->prof_rfx_dwt_2d_decode->name = "rfx_dwt_2d_decode_NEON");
|
|
|
|
|
|
|
|
context->quantization_decode = rfx_quantization_decode_NEON;
|
|
|
|
context->dwt_2d_decode = rfx_dwt_2d_decode_NEON;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
#endif // __ARM_NEON__
|
|
|
|
|