📄 fir_r8.asm
字号:
* ========================================================================= *
* TEXAS INSTRUMENTS, INC. *
* *
* NAME *
* fir_r8 -- Block FIR *
* *
* *
* REVISION DATE *
* 09-May-2005 *
* *
* USAGE *
* This routine has following C prototype: *
* void fir_r8_asm *
* ( *
* const short *restrict x, /* Input array [nr+nh-1 elements] */ *
* const short *restrict h, /* Coeff array [nh elements] */ *
* /* Reversed order */ *
* short *restrict r, /* Output array [nr elements] */ *
* int nh, /* Number of coefficients. */ *
* int nr /* Number of output samples. */ *
* ) *
* *
* *
* *
* DESCRIPTION *
* Computes a real FIR filter (direct-form) using coefficients *
* stored in vector h. The real data input is stored in vector x. *
* The filter output result is stored in vector r. Input data and *
* filter taps are 16-bit, with intermediate values kept at 32-bit *
* precision. Filter taps are expected in Q15 format. *
* *
* *
* TECHNIQUES *
* 1. The inner loop is unrolled eight times and the outer loop is *
* unrolled 8 times, computing 8 outputs. *
* *
* 2. The inner and outer loops are completely overlaped with *
* SPLOOP reload; thus no overhead exists for outer loop *
* execution. *
* *
* 3. For a case where this function is used with circular *
* addressing, A4 and B7 are allocated to A_X_addr and B_X_addr, *
* respectively. *
* *
* ASSUMPTIONS *
* Number of taps: 'nh' >= 8, multiple of 8. *
* Number of samples: 'nr' >= 8, multiple of 8. *
* Array 'r' is double-word aligned. *
* *
* *
* C CODE *
* void fir_r8 *
* ( *
* const short *restrict x, *
* const short *restrict h, *
* short *restrict r, *
* int nh, *
* int nr *
* ) *
* { *
* int i, j, sum; *
* *
* for (j = 0; j < nr; j++) *
* { *
* sum = 0; *
* for (i = 0; i < nh; i++) *
* sum += x[i + j] * h[i]; *
* r[j] = sum >> 15; *
* } *
* } *
* *
* *
* NOTES *
* This function is fully interruptible. *
* *
* *
* CYCLES *
* When nh>=32, nh * nr / 8 + 22 *
* Otherwise, 32 * nr / 8 + 22 *
* *
* For nh = 32 and nr = 256, cycles = 1046 *
* *
* CODESIZE *
* 640 bytes *
* ------------------------------------------------------------------------- *
* Copyright (c) 2005 Texas Instruments, Incorporated. *
* All Rights Reserved. *
* ========================================================================= *
.text .global _fir_r8_fir_r8:
.asg A4, A_X_addr
.asg B4, B_H_addr
.asg A6, A_R_addr
.asg B6, B_T
.asg A8, A_N
* ===================== SYMBOLIC REGISTER ASSIGNMENTS ===================== *
.asg A4, A_X_addr
.asg B7, B_X_addr
.asg A8, A_COEFaddr
.asg B29, B_sum3
.asg B28, B_sum2
.asg B31, B_sum1
.asg B30, B_sum0
.asg A27, A_sum5
.asg A26, A_sum4
.asg A29, A_sum7
.asg A28, A_sum6
.asg A15, A_d3d2
.asg A14, A_d1d0
.asg B27, B_d7d6
.asg B26, B_d5d4
.asg A23, A_dbda
.asg A22, A_d9d8
.asg B11, B_dfde
.asg B10, B_dddc
.asg A25, A_c3c2
.asg A24, A_c1c0
.asg B9, B_COEFaddr
.asg B21, B_c7c6
.asg B20, B_c5c4
.asg B23, B_c3c2
.asg B22, B_c1c0
.asg A21, A_c7c6
.asg A20, A_c5c4
.asg B25, B_d5d4_
.asg B24, B_d3d2_
.asg A17, A_d9d8_
.asg A16, A_d7d6_
.asg B13, B_dddc_
.asg B12, B_dbda_
.asg B17, B_s7d
.asg B16, B_s6d
.asg B11, B_s7c
.asg B10, B_s6c
.asg A19, A_s7b
.asg A18, A_s6b
.asg A21, A_s7a
.asg A20, A_s6a
.asg A5, A_s6ab
.asg A15, A_s7ab
.asg B2, B_s6cd
.asg B8, B_s7cd
.asg A5, A_s6abcd
.asg A7, A_s7abcd
.asg B19, B_s5d
.asg B18, B_s4d
.asg A15, A_s5c
.asg A14, A_s4c
.asg A13, A_s5b
.asg A12, A_s4b
.asg B13, B_s5a
.asg B12, B_s4a
.asg B10, B_s4ad
.asg B11, B_s5ad
.asg A24, A_s4bc
.asg A13, A_s5bc
.asg A7, A_s4abcd
.asg A25, A_s5abcd
.asg A13, A_s3d
.asg A12, A_s2d
.asg A17, A_s3c
.asg A16, A_s2c
.asg B17, B_s3b
.asg B16, B_s2b
.asg B19, B_s3a
.asg B18, B_s2a
.asg B20, B_s2ab
.asg B8, B_s3ab
.asg A9, A_s2cd
.asg A11, A_s3cd
.asg B21, B_s2abcd
.asg B25, B_s3abcd
.asg A19, A_s1d
.asg A18, A_s0d
.asg B19, B_s1c
.asg B18, B_s0c
.asg B13, B_s1b
.asg B12, B_s0b
.asg A31, A_s1a
.asg A30, A_s0a
.asg A15, A_s0ad
.asg A0, A_s1ad
.asg B0, B_s0bc
.asg B14, B_s1bc
.asg B0, B_s0abcd
.asg B23, B_s1abcd
.asg B3, B_ret
.asg B21, B_R_addr
.asg B1, B_i
.asg B5, B_sum32
.asg B4, B_sum10
.asg A3, A_sum76
.asg A2, A_sum54
.asg B6, B_TI
.asg B3, B_TC
.asg A1, A_TI
.asg A10, A_TC
.asg B15, B_SP
* ========================================================================= *
SHRU .S2 B_T, 3, B_T
|| SHRU .S1 B_T, 2, A_TC
|| MV .L2X A_N, B_i
|| STDW .D2 A11:A10, *--B_SP[10]
MVC .S2 B_T, RILC
|| SUB .L2 B_T, 1, B_T
|| STW .D2 B3, *+B_SP[6*2]
[B_i] SPLOOPD 8
|| SUB .D1 A_X_addr, 16, A_X_addr
|| MVC .S2 B_T, ILC
|| SUB .L2 A_TC, 2, B_TC
|| SUB .S1 B_H_addr, 16, A_COEFaddr
|| STDW .D2T1 A15:A14, *+B_SP[2]
*- Stage 0 -----------------------------------------------------------------*
SPMASK
|| LDDW .D1T1 *++A_X_addr[2], A_d3d2:A_d1d0 ;[ 1,1]
||^ SUB .S2 B_H_addr, 8, B_COEFaddr ;
||^ SUB .L1 A_TC, 2, A_TC
||^ STW .D2T2 B14, *+B_SP[3*2]
SPMASK
|| LDDW .D2T2 *++B_COEFaddr[2], B_c7c6:B_c5c4 ;[ 2,1]
|| LDDW .D1T1 *++A_COEFaddr[2], A_c3c2:A_c1c0 ;[ 2,1]
||^ SUB .S2 A_X_addr, 8, B_X_addr ;
SPMASK
|| LDDW .D2T2 *++B_X_addr[2], B_d7d6:B_d5d4 ;[ 3,1]
||^ SUB .S2 B_TC, 2, B_TI
||^ SUB .S1 A_TC, 2, A_TI
LDDW .D2T2 *B_X_addr[2], B_dfde:B_dddc ;[ 4,1]
|| LDDW .D1T1 *A_X_addr[2], A_dbda:A_d9d8 ;[ 4,1]
LDDW .D2T1 *B_COEFaddr, A_c7c6:A_c5c4 ;[ 5,1]
|| LDDW .D1T2 *A_COEFaddr, B_c3c2:B_c1c0 ;[ 5,1]
⌨️ 快捷键说明
复制代码
Ctrl + C
搜索代码
Ctrl + F
全屏模式
F11
切换主题
Ctrl + Shift + D
显示快捷键
?
增大字号
Ctrl + =
减小字号
Ctrl + -