search for: silk_rshift_round

...silk_assert( ( stride == 1 ) || ( stride == 2 ) ); if( stride == 1) { opus_int32 out32_Q14; for( k = 0; k < len; k++ ) { /* S[ 0 ], S[ 1 ]: Q12 */ out32_Q14 = silk_LSHIFT( silk_SMLAWB( S[ 0 ], B_Q28[ 0 ], in[ k ] ), 2 ); S[ 0 ] = S[ 1 ] + silk_RSHIFT_ROUND( (opus_int64)out32_Q14 * (-A_Q28[ 0 ]), 30 ); S[ 0 ] = silk_SMLAWB( S[ 0 ], B_Q28[ 1 ], in[ k ] ); S[ 1 ] = silk_RSHIFT_ROUND( (opus_int64)out32_Q14 * (-A_Q28[ 1 ]) , 30 ); S[ 1 ] = silk_SMLAWB( S[ 1 ], B_Q28[ 2 ], in[ k ] ); /* Scale back to Q0 and...

...stride == 1) { > opus_int32 out32_Q14; > for( k = 0; k < len; k++ ) { > /* S[ 0 ], S[ 1 ]: Q12 */ > out32_Q14 = silk_LSHIFT( silk_SMLAWB( S[ 0 ], B_Q28[ 0 > ], in[ k ] ), 2 ); > > S[ 0 ] = S[ 1 ] + silk_RSHIFT_ROUND( > (opus_int64)out32_Q14 * (-A_Q28[ 0 ]), 30 ); > S[ 0 ] = silk_SMLAWB( S[ 0 ], B_Q28[ 1 ], in[ k ] ); > > S[ 1 ] = silk_RSHIFT_ROUND( (opus_int64)out32_Q14 * > (-A_Q28[ 1 ]) , 30 ); > S[ 1 ] = silk_SMLAWB( S[ 1 ], B_Q28[ 2...

...2 ) ); > > if( stride == 1) { > opus_int32 out32_Q14; > for( k = 0; k < len; k++ ) { > /* S[ 0 ], S[ 1 ]: Q12 */ > out32_Q14 = silk_LSHIFT( silk_SMLAWB( S[ 0 ], B_Q28[ 0 ], in[ > k ] ), 2 ); > > S[ 0 ] = S[ 1 ] + silk_RSHIFT_ROUND( (opus_int64)out32_Q14 * > (-A_Q28[ 0 ]), 30 ); > S[ 0 ] = silk_SMLAWB( S[ 0 ], B_Q28[ 1 ], in[ k ] ); > > S[ 1 ] = silk_RSHIFT_ROUND( (opus_int64)out32_Q14 * (-A_Q28[ > 1 ]) , 30 ); > S[ 1 ] = silk_SMLAWB( S[ 1 ], B_Q28[ 2 ], in[ k ] ); > &g...

...opus_int32 out32_Q14; > > for( k = 0; k < len; k++ ) { > > /* S[ 0 ], S[ 1 ]: Q12 */ > > out32_Q14 = silk_LSHIFT( silk_SMLAWB( S[ 0 ], B_Q28[ 0 > > ], in[ k ] ), 2 ); > > > > S[ 0 ] = S[ 1 ] + silk_RSHIFT_ROUND( > > (opus_int64)out32_Q14 * (-A_Q28[ 0 ]), 30 ); > > S[ 0 ] = silk_SMLAWB( S[ 0 ], B_Q28[ 1 ], in[ k ] ); > > > > S[ 1 ] = silk_RSHIFT_ROUND( (opus_int64)out32_Q14 * > > (-A_Q28[ 1 ]) , 30 ); > > S[ 1 ] = s...

...A_Q28[ 0 ] ) & 0x00003FFF; /* lower part */ A0_U_Q28 = silk_RSHIFT( -A_Q28[ 0 ], 14 ); /* upper part */ A1_L_Q28 = ( -A_Q28[ 1 ] ) & 0x00003FFF; /* lower part */ A1_U_Q28 = silk_RSHIFT( -A_Q28[ 1 ], 14 ); /* upper part */ ... S[ 0 ] = S[1] + silk_RSHIFT_ROUND( silk_SMULWB( out32_Q14, A0_L_Q28 ), 14 ); S[ 0 ] = silk_SMLAWB( S[ 0 ], out32_Q14, A0_U_Q28 ); S[ 0 ] = silk_SMLAWB( S[ 0 ], B_Q28[ 1 ], inval); S[ 1 ] = silk_RSHIFT_ROUND( silk_SMULWB( out32_Q14, A1_L_Q28 ), 14 ); S[ 1 ] = silk_SMLAWB( S[ 1 ], out32_Q14, A1_U_Q28...

...)))) /* (a32 * b32) >> 16 */ -#if defined(__x86_64__) || defined(__LP64__) || defined(_WIN64) +#if OPUS_FAST_INT64 #define silk_SMULWW(a32, b32) (((opus_int64)(a32) * (b32)) >> 16) #else #define silk_SMULWW(a32, b32) silk_MLA(silk_SMULWB((a32), (b32)), (a32), silk_RSHIFT_ROUND((b32), 16)) #endif /* a32 + ((b32 * c32) >> 16) */ -#if defined(__x86_64__) || defined(__LP64__) || defined(_WIN64) +#if OPUS_FAST_INT64 #define silk_SMLAWW(a32, b32, c32) ((a32) + (((opus_int64)(b32) * (c32)) >> 16)) #else #define silk_SMLAWW(a32, b32, c32) silk_MLA(...

...6 */ #if OPUS_FAST_INT64 -#define silk_SMULWW(a32, b32) (((opus_int64)(a32) * (b32)) >> 16) +#define silk_SMULWW(a32, b32) ((opus_int32)(((opus_int64)(a32) * (b32)) >> 16)) #else #define silk_SMULWW(a32, b32) silk_MLA(silk_SMULWB((a32), (b32)), (a32), silk_RSHIFT_ROUND((b32), 16)) #endif /* a32 + ((b32 * c32) >> 16) */ #if OPUS_FAST_INT64 -#define silk_SMLAWW(a32, b32, c32) ((a32) + (((opus_int64)(b32) * (c32)) >> 16)) +#define silk_SMLAWW(a32, b32, c32) ((opus_int32)((a32) + (((opus_int64)(b32) * (c32)) >> 16))) #else #define...

--- celt/arch.h | 5 +++++ silk/macros.h | 4 +--- 2 files changed, 6 insertions(+), 3 deletions(-) diff --git a/celt/arch.h b/celt/arch.h index 9f74ddd..670527b 100644 --- a/celt/arch.h +++ b/celt/arch.h @@ -78,6 +78,11 @@ static OPUS_INLINE void _celt_fatal(const char *str, const char *file, int line) #define UADD32(a,b) ((a)+(b)) #define USUB32(a,b) ((a)-(b)) +/* Set this if opus_int64

Hi Jean-Marc, Tested on my chromebook, when stride (channel) == 1, the optimization has no gain compared with C function. When stride (channel) == 2, the optimization is 1.2%-1.8% faster (1.6% at Complexity 8) compared with C function. Please let me know and I can remove the optimization of stride 1 case. If it's allowed to skip the split of A_Q28 and replace by 32-bit multiplication

...psDelDec, i - decisionDelay, last_smple_idx, Winner_ind, gain_lo_s32x2, gain_hi_s32x2, shift_s32x4, t0_s32x4, t1_s32x4, pulses, pxq, NSQ ); + } + for( ; ( i < decisionDelay ) && ( last_smple_idx >= 0 ); i++, last_smple_idx-- ) { + pulses[ i - decisionDelay ] = (opus_int8)silk_RSHIFT_ROUND( psDelDec->Q_Q10[ last_smple_idx ][ Winner_ind ], 10 ); + pxq[ i - decisionDelay ] = (opus_int16)silk_SAT16( silk_RSHIFT_ROUND( silk_SMULWW( psDelDec->Xq_Q14[ last_smple_idx ][ Winner_ind ], gain ), shift ) ); + NSQ->sLTP_shp_Q14[ NSQ->sLTP_shp_buf_idx - decisionDelay + i...

NSQ_LPC_BUF_LENGTH is independent of DECISION_DELAY. --- silk/define.h | 4 ---- 1 file changed, 4 deletions(-) diff --git a/silk/define.h b/silk/define.h index 781cfdc..1286048 100644 --- a/silk/define.h +++ b/silk/define.h @@ -173,11 +173,7 @@ extern "C" #define MAX_MATRIX_SIZE MAX_LPC_ORDER /* Max of LPC Order and LTP order */ -#if( MAX_LPC_ORDER >

...t;(c), "r"(a) + ); + return res; +} + +/* a64 + (b32 * c32) */ +#define silk_SMLAL(a64, b32, c32) (silk_ADD64((a64), ((opus_int64)(b32) * (opus_int64)(c32)))) + +/* (a32 * b32) >> 16 */ +#define silk_SMULWW(a32, b32) silk_MLA(silk_SMULWB((a32), (b32)), (a32), silk_RSHIFT_ROUND((b32), 16)) + +/* a32 + ((b32 * c32) >> 16) */ +#define silk_SMLAWW(a32, b32, c32) silk_MLA(silk_SMLAWB((a32), (b32), (c32)), (b32), silk_RSHIFT_ROUND((c32), 16)) + +/* add/subtract with output saturated */ +static inline opus_int32 silk_ADD_SAT32(opus_int32 a, opus_int32 b) +{ + int r...

...all input voice silent. I have applied the recent bugfix 2349213003, but that didn't help. I tried windows and android build, both ends keep quiet even both builds happily seems to generate silence there. There is a line: frame[ i ] = (opus_int16)silk_ADD_SAT16( frame[ i ], silk_SAT16( silk_RSHIFT_ROUND( silk_SMULWW( CNG_sig_Q14[ MAX_LPC_ORDER + i ], gain_Q10 ), 8 ) ) ); When I mute both ends, the frame[i] is 0 before and after that line. Opus 1.1.3 used with PJSip 2.5.5. Windows 7 desktop, Android 5 phone Any help what is the trick to get Comfort Noise to be generated? Thx, Markus

--- configure.ac | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/configure.ac b/configure.ac index f52d2c2..e1a6e9b 100644 --- a/configure.ac +++ b/configure.ac @@ -190,7 +190,7 @@ AC_ARG_ENABLE([rtcd], [enable_rtcd=yes]) AC_ARG_ENABLE([intrinsics], - [AS_HELP_STRING([--disable-intrinsics], [Disable intrinsics optimizations for ARM(float) X86(fixed)])],, +