The listing below shows a spu_timing static timing analysis for the optimized SPE thread (process _buffer subroutine only).
.type process_buffer, @function
process_buffer:
0D 0123 shli $2,$3,10
1D 012345 lqa $19,ctx+16
0D 12 ori $6,$3,0
1D 1234 shlqbyi $24,$4,0
0D 23 cgti $3,$4,0
1D 2345 shlqbyi $18,$5,0
0D 34 ila $4,inv_mass
1D 3456 fsmbi $21,0
0 45 ilhu $27,1029
0 56 ilhu $26,2057
0 67 ilhu $25,3085
0 78 ila $28,66051
0 89 a $20,$2,$4
0 90 iohl $27,1543
0D 01 iohl $26,2571
1D 0 lnop
0D 12 iohl $25,3599
1D 1234 brz $3,.L7
0 2345 shli $17,$6,14
0 34 ila $23,pos
0D 45 ila $22,vel
1D 456789 hbra .L10,.L5
1 5 lnop
0 6 nop $127
.L5:
0D 78 ila $43,pos
1D 789012 lqd $41,0($20)
0D 89 ila $42,vel
1D 890123 lqx $40,$17,$23
0 90 a $6,$17,$43
0 01 a $7,$17,$42
0D 12 ai $21,$21,4
1D 123456 lqd $39,16($6)
0D 23 ai $20,$20,16
1D 234567 lqd $38,32($6)
0D 345678 fm $36,$18,$41
1D 345678 lqd $37,48($6)
0D 45 cgt $16,$24,$21
1D 456789 lqx $13,$17,$22
1 567890 lqd $34,16($7)
1 678901 lqd $14,32($7)
1 789012 lqd $15,48($7)
1 -9012 shufb $35,$36,$36,$28
0D 012345 fma $32,$13,$18,$40
1D 0123 shufb $33,$36,$36,$27
0D 123456 fma $10,$34,$18,$39
1D 1234 shufb $31,$36,$36,$26
0D 234567 fma $11,$14,$18,$38
1D 2345 shufb $30,$36,$36,$25
0 345678 fma $8,$15,$18,$37
0 456789 fma $29,$35,$19,$13
0D 567890 fma $5,$33,$19,$34
1D 5 lnop
0D 678901 fma $12,$31,$19,$14
1D 678901 stqx $32,$17,$23
0D 789012 fma $9,$30,$19,$15
1D 789012 stqd $10,16($6)
1 890123 stqd $11,32($6)
1 901234 stqd $8,48($6)
0D 0 nop $127
1D 012345 stqx $29,$17,$22
0D 12 ai $17,$17,64
1D 123456 stqd $5,16($7)
1 234567 stqd $12,32($7)
1 345678 stqd $9,48($7)
0D 4 nop $127
.L10:
1D 4567 brnz $16,.L5
.L7:
0D 5 nop $127
1D 5678 bi $lr