Texas Instruments TMS320C6000 Programmer's Manual page 191

Software Pipelining
Example 6–31. Assembly Code for Floating-Point Dot Product (Software Pipelined With
Removal of Prolog and Epilog) (Continued)
[A1] B
||[A1] SUB
[A1] B
||[A1] SUB
LOOP:
LDDW
|| LDDW
|| MPYSP
|| MPYSP
|| ADDSP
|| ADDSP
||[A1] B
||[A1] SUB
; Branch occurs here
ADDSP
ADDSP
ADDSP
ADDSP
NOP
ADDSP
NOP
ADDSP
NOP
ADDSP
NOP
6-54
.S2 LOOP
.S1 A1,1,A1
.S2 LOOP
.S1 A1,1,A1
.D1 A4++,A7:A6
.D2 B4++,B7:B6
.M1X A6,B6,A5
.M2X A7,B7,B5
.L1 A5,A8,A8
.L2 B5,B8,B8
.S2 LOOP
.S1 A1,1,A1
.L1X A8,B8,A0
.L2X A8,B8,B0
.L1X A8,B8,A0
.L2X A8,B8,B0
.L1X A0,B0,A5
.L2X A0,B0,B5
3
.L1X A5,B5,A4
3
;*** branch to loop
;**** decrement loop counter
;**** branch to loop
;***** decrement loop counter
;********* load ai & ai + 1 from memory
;********* load bi & bi + 1 from memory
;**** pi = a0 b0
;**** pi1 = a1 b1
; sum0 += (ai bi)
; sum1 += (ai+1 bi+1)
;***** branch to loop
;****** decrement loop counter
; sum(0) = sum0(0) + sum1(0)
; sum(1) = sum0(1) + sum1(1)
; sum(2) = sum0(2) + sum1(2)
; sum(3) = sum0(3) + sum1(3)
; wait for B0
; sum(01) = sum(0) + sum(1)
; wait for next B0
; sum(23) = sum(2) + sum(3)
; sum = sum(01) + sum(23)
;