diff options
| author | Maamoun TK <maamoun.tk@googlemail.com> | 2020-07-14 14:43:06 +0300 |
|---|---|---|
| committer | Niels Möller <nisse@lysator.liu.se> | 2020-08-26 18:13:51 +0200 |
| commit | 9df0079edc6c2552ee2c162d9b569f7cccb5e215 (patch) | |
| tree | f0c8e2d7caa51cb0d81a30ed5e31168e065b0dcd /Makefile.in | |
| parent | 95b29fe510cc35cf617e730ed7fa951e982e1532 (diff) | |
| download | nettle-9df0079edc6c2552ee2c162d9b569f7cccb5e215.tar.gz | |
"PowerPC64" Add optimized AES [Enc|Dec]
I measured the latency and throughput of vcipher/vncipher/vxor instructions
for POWER8
vcipher/vncipher
throughput 6 instructions per cycle
latency 0.91 clock cycles
vxor
throughput 6 instructions per cycle
latency 0.32 clock cycles
So the ideal option for POWER8 is processing 8 blocks, it has +12%
performance over processing 4 blocks.
Diffstat (limited to 'Makefile.in')
0 files changed, 0 insertions, 0 deletions