Ryan Houdek 791c7d5a84 [AArch64] Clean up bogus vector FCVT{N,L} instruction usage.
Replace the instruction with the scalar variant FCVT instruction.
FCVT{N,L} 8 cycles latency on the Cortex A57
FCVT has five cycle latency and slightly higher throughput

On the A72 all three of these instructions will have three cycle latency,
While FCVT{N,L} will have half the throughput.
2015-09-04 19:41:54 -05:00
..
2015-09-03 06:10:01 -07:00
2015-09-04 02:43:38 -04:00