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