Skip to content
GitLab
Commit 1ade44b3 authored by John Koleszar's avatar John Koleszar Committed by Code Review
Browse files

Merge "fix --disable-runtime-cpu-detect on x86"

parents e7e5a58d 92b0e544
v1.14.0-linphone 1.4.X cayuga eider experimental feature/update_to_v1.9.0-linphone feature/uwp_nuget forest frame_parallel highbitdepth indianrunnerduck javanwhistlingduck khakicampbell linphone linphone-android linphone-old longtailedduck m29-baseline m31-baseline m49-2623 m52-2743 m54-2840 m56-2924 m66-3359 m68-3440 mandarinduck mcw mcw2 nextgen nextgenv2 pcs-2013 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/jkoleszar/cached-multibit sandbox/jkoleszar/new-rate-control sandbox/jkoleszar/new-rtcd sandbox/jkoleszar/reuse-modemv sandbox/jzern@google.com/test sandbox/wangch@google.com/vp9 sandbox/yaowu@google.com/mergeaom stable-vp9-decoder v1.12.0-linphone v1.6.1_linphone v1.7.0-linphone v1.9.0-linphone vp9-preview 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 v1.2.0 v1.1.0 v1.0.0 v0.9.7 v0.9.7-p1
No related merge requests found
Hide whitespace changes
Inline Side-by-side
Showing
with 26 additions and 20 deletions
vp8/decoder/x86/x86_dsystemdependent.c
+ 3
− 3
View file @ 1ade44b3
......@@ -17,7 +17,7 @@
#if HAVE_MMX
void vp8_dequantize_b_impl_mmx(short *sq, short *dq, short *q);
static void dequantize_b_mmx(BLOCKD *d)
void vp8_dequantize_b_mmx(BLOCKD *d)
{
short *sq = (short *) d->qcoeff;
short *dq = (short *) d->dqcoeff;
......@@ -28,6 +28,7 @@ static void dequantize_b_mmx(BLOCKD *d)
void vp8_arch_x86_decode_init(VP8D_COMP *pbi)
{
#if CONFIG_RUNTIME_CPU_DETECT
int flags = x86_simd_caps();
/* Note:
......@@ -36,12 +37,11 @@ void vp8_arch_x86_decode_init(VP8D_COMP *pbi)
* you modify any of the function mappings present in this file, be sure
* to also update them in static mapings (<arch>/filename_<arch>.h)
*/
#if CONFIG_RUNTIME_CPU_DETECT
/* Override default functions with fastest ones for this CPU. */
#if HAVE_MMX
if (flags & HAS_MMX)
{
pbi->dequant.block = dequantize_b_mmx;
pbi->dequant.block = vp8_dequantize_b_mmx;
pbi->dequant.idct_add = vp8_dequant_idct_add_mmx;
pbi->dequant.dc_idct_add = vp8_dequant_dc_idct_add_mmx;
pbi->dequant.dc_idct_add_y_block = vp8_dequant_dc_idct_add_y_block_mmx;
......
vp8/encoder/x86/dct_x86.h
+ 6
− 0
View file @ 1ade44b3
......@@ -31,6 +31,12 @@ extern prototype_fdct(vp8_short_fdct8x4_mmx);
#undef vp8_fdct_short8x4
#define vp8_fdct_short8x4 vp8_short_fdct8x4_mmx
#undef vp8_fdct_fast4x4
#define vp8_fdct_fast4x4 vp8_short_fdct4x4_mmx
#undef vp8_fdct_fast8x4
#define vp8_fdct_fast8x4 vp8_short_fdct8x4_mmx
#endif
#endif
......
vp8/encoder/x86/x86_csystemdependent.c
+ 17
− 17
View file @ 1ade44b3
......@@ -16,7 +16,7 @@
#if HAVE_MMX
static void short_fdct8x4_mmx(short *input, short *output, int pitch)
void vp8_short_fdct8x4_mmx(short *input, short *output, int pitch)
{
vp8_short_fdct4x4_mmx(input, output, pitch);
vp8_short_fdct4x4_mmx(input + 4, output + 16, pitch);
......@@ -26,7 +26,7 @@ int vp8_fast_quantize_b_impl_mmx(short *coeff_ptr, short *zbin_ptr,
short *qcoeff_ptr, short *dequant_ptr,
short *scan_mask, short *round_ptr,
short *quant_ptr, short *dqcoeff_ptr);
static void fast_quantize_b_mmx(BLOCK *b, BLOCKD *d)
void vp8_fast_quantize_b_mmx(BLOCK *b, BLOCKD *d)
{
short *scan_mask = vp8_default_zig_zag_mask;//d->scan_order_mask_ptr;
short *coeff_ptr = b->coeff;
......@@ -51,7 +51,7 @@ static void fast_quantize_b_mmx(BLOCK *b, BLOCKD *d)
}
int vp8_mbblock_error_mmx_impl(short *coeff_ptr, short *dcoef_ptr, int dc);
static int mbblock_error_mmx(MACROBLOCK *mb, int dc)
int vp8_mbblock_error_mmx(MACROBLOCK *mb, int dc)
{
short *coeff_ptr = mb->block[0].coeff;
short *dcoef_ptr = mb->e_mbd.block[0].dqcoeff;
......@@ -59,7 +59,7 @@ static int mbblock_error_mmx(MACROBLOCK *mb, int dc)
}
int vp8_mbuverror_mmx_impl(short *s_ptr, short *d_ptr);
static int mbuverror_mmx(MACROBLOCK *mb)
int vp8_mbuverror_mmx(MACROBLOCK *mb)
{
short *s_ptr = &mb->coeff[256];
short *d_ptr = &mb->e_mbd.dqcoeff[256];
......@@ -69,7 +69,7 @@ static int mbuverror_mmx(MACROBLOCK *mb)
void vp8_subtract_b_mmx_impl(unsigned char *z, int src_stride,
short *diff, unsigned char *predictor,
int pitch);
static void subtract_b_mmx(BLOCK *be, BLOCKD *bd, int pitch)
void vp8_subtract_b_mmx(BLOCK *be, BLOCKD *bd, int pitch)
{
unsigned char *z = *(be->base_src) + be->src;
unsigned int src_stride = be->src_stride;
......@@ -82,7 +82,7 @@ static void subtract_b_mmx(BLOCK *be, BLOCKD *bd, int pitch)
#if HAVE_SSE2
int vp8_mbblock_error_xmm_impl(short *coeff_ptr, short *dcoef_ptr, int dc);
static int mbblock_error_xmm(MACROBLOCK *mb, int dc)
int vp8_mbblock_error_xmm(MACROBLOCK *mb, int dc)
{
short *coeff_ptr = mb->block[0].coeff;
short *dcoef_ptr = mb->e_mbd.block[0].dqcoeff;
......@@ -90,7 +90,7 @@ static int mbblock_error_xmm(MACROBLOCK *mb, int dc)
}
int vp8_mbuverror_xmm_impl(short *s_ptr, short *d_ptr);
static int mbuverror_xmm(MACROBLOCK *mb)
int vp8_mbuverror_xmm(MACROBLOCK *mb)
{
short *s_ptr = &mb->coeff[256];
short *d_ptr = &mb->e_mbd.dqcoeff[256];
......@@ -100,7 +100,7 @@ static int mbuverror_xmm(MACROBLOCK *mb)
void vp8_subtract_b_sse2_impl(unsigned char *z, int src_stride,
short *diff, unsigned char *predictor,
int pitch);
static void subtract_b_sse2(BLOCK *be, BLOCKD *bd, int pitch)
void vp8_subtract_b_sse2(BLOCK *be, BLOCKD *bd, int pitch)
{
unsigned char *z = *(be->base_src) + be->src;
unsigned int src_stride = be->src_stride;
......@@ -178,20 +178,20 @@ void vp8_arch_x86_encoder_init(VP8_COMP *cpi)
cpi->rtcd.variance.get4x4sse_cs = vp8_get4x4sse_cs_mmx;
cpi->rtcd.fdct.short4x4 = vp8_short_fdct4x4_mmx;
cpi->rtcd.fdct.short8x4 = short_fdct8x4_mmx;
cpi->rtcd.fdct.short8x4 = vp8_short_fdct8x4_mmx;
cpi->rtcd.fdct.fast4x4 = vp8_short_fdct4x4_mmx;
cpi->rtcd.fdct.fast8x4 = short_fdct8x4_mmx;
cpi->rtcd.fdct.fast8x4 = vp8_short_fdct8x4_mmx;
cpi->rtcd.fdct.walsh_short4x4 = vp8_short_walsh4x4_c;
cpi->rtcd.encodemb.berr = vp8_block_error_mmx;
cpi->rtcd.encodemb.mberr = mbblock_error_mmx;
cpi->rtcd.encodemb.mbuverr = mbuverror_mmx;
cpi->rtcd.encodemb.subb = subtract_b_mmx;
cpi->rtcd.encodemb.mberr = vp8_mbblock_error_mmx;
cpi->rtcd.encodemb.mbuverr = vp8_mbuverror_mmx;
cpi->rtcd.encodemb.subb = vp8_subtract_b_mmx;
cpi->rtcd.encodemb.submby = vp8_subtract_mby_mmx;
cpi->rtcd.encodemb.submbuv = vp8_subtract_mbuv_mmx;
/*cpi->rtcd.quantize.fastquantb = fast_quantize_b_mmx;*/
/*cpi->rtcd.quantize.fastquantb = vp8_fast_quantize_b_mmx;*/
}
#endif
......@@ -233,9 +233,9 @@ void vp8_arch_x86_encoder_init(VP8_COMP *cpi)
cpi->rtcd.fdct.walsh_short4x4 = vp8_short_walsh4x4_sse2 ;
cpi->rtcd.encodemb.berr = vp8_block_error_xmm;
cpi->rtcd.encodemb.mberr = mbblock_error_xmm;
cpi->rtcd.encodemb.mbuverr = mbuverror_xmm;
cpi->rtcd.encodemb.subb = subtract_b_sse2;
cpi->rtcd.encodemb.mberr = vp8_mbblock_error_xmm;
cpi->rtcd.encodemb.mbuverr = vp8_mbuverror_xmm;
cpi->rtcd.encodemb.subb = vp8_subtract_b_sse2;
cpi->rtcd.encodemb.submby = vp8_subtract_mby_sse2;
cpi->rtcd.encodemb.submbuv = vp8_subtract_mbuv_sse2;
......
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