PowerPC e500 Core Family Reference Manual, Rev. 1
4-40 Freescale Semiconductor
Execution Timing
evandc SU1 1
evcmpeq SU1 1
evcmpgts SU1 1
evcmpgtu SU1 1
evcmplts SU1 1
evcmpltu SU1 1
evcntlsw SU1 1
evcntlzw SU1 1
evdivws
evdivwu
MU Both the lower and upper words match the criteria described for the divw
x
4-cycle case.
1
Assuming the 4-cycle evdivw
x
case does not apply, the lower and upper words match the
criteria described for the divw
x
4- or 11-cycle case.
1
Assuming neither the 4- or 11-cycle evdivw
x
cases apply, the lower and upper words match
the criteria described for the divw
x
4-, 11-, or 19-cycle case.
1
All other cases
1
1eveqv SU1
evextsb SU1 1
evextsh SU1 1
evfsabs SU1 1
evfsadd MU 4:1
evfscfsf MU 4:1
evfscfsi MU 4:1
evfscfuf MU 4:1
evfscfui MU 4:1
evfscmpeq MU 4:1
evfscmpgt MU 4:1
evfscmplt MU 4:1
evfsctsf MU 4:1
evfsctsi MU 4:1
evfsctsiz MU 4:1
evfsctuf MU 4:1
evfsctui MU 4:1
evfsctuiz MU 4:1
evfsdiv MU 4 (if either rA or rB is 0.0)
29 (all other cases)
evfsmul MU 4:1
evfsnabs SU1 1
evfsneg SU1 1
evfssub MU 4:1
Table 4-8. SPE and Embedded Floating-Point APU Instruction Latencies (continued)
Mnemonic Unit Cycles (Latency:Throughput)