Linear Assembly Resource Allocation - Texas Instruments TMS320C6000 Programmer's Manual

Loop Carry Paths
6.7.4.2 New 'C6x Instructions (Inner Loop)
Example 6–43. Linear Assembly for IIR Inner Loop With Reduced Loop Carry Path
LDH *xptr++,xi
MPY c1,xi,p0
LDH *xptr,xi+1
MPY c2,xi+1,p1
ADD p0,p1,s0
MPY c3,y,p2
ADD s0,p2,s1
SHR s1,15,y
STH y,*yptr++
[cntr]SUB cntr,1,cntr
[cntr]B LOOP
6.7.5

Linear Assembly Resource Allocation

Example 6–44. Linear Assembly for IIR Inner Loop (With Allocated Resources)
LDH .D1
MPY .M1
LDH .D1
MPY .M1X
ADD .L1
MPY .M2X
ADD .L2X
SHR .S2
STH .D2
[A1] SUB .L1
[A1] B .S1
6-82
Example 6–43 shows the new linear assembly from the graph in Figure 6–16,
where LDH yi was removed. The one variable y that is read and written is yi
for the MPY p2 instruction and yi+1 for the SHR and STH instructions.
; xi+1
; c1 * xi
; xi+1
; c2 * xi+1
; c1 * xi + c2 * xi+1
; c3 * yi
; c1 * xi + c2 * xi+1 + c3 * yi
; yi+1
; store yi+1
; decrement loop counter
; branch to loop
Example 6–44 shows the same linear assembly instructions as those in
Example 6–43 with the functional units and registers assigned.
*A4++,A2 ; xi+1
A6,A2,A5 ; c1 * xi
*A4,A3 ; xi+1
B6,A3,A7 ; c2 * xi+1
A5,A7,A9 ; c1 * xi + c2 * xi+1
A8,B2,B3 ; c3 * yi
B3,A9,B5 ; c1 * xi + c2 * xi+1 + c3 * yi
B5,15,B2 ; yi+1
B2,*B4++ ; store yi+1
A1,1,A1 ; decrement loop counter
LOOP ; branch to loop