Konstantin Ananyev
b64c2295f7
acl: add 256-bit AVX512 classify method ...
Introduce classify implementation that uses AVX512 specific ISA.
rte_acl_classify_avx512x16() is able to process up to 16 flows in parallel.
It uses 256-bit width registers/instructions only
(to avoid frequency level change).
Note that for now only 64-bit version is supported.
Signed-off-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
2020-10-14 14:23:00 +02:00
2020-10-14 14:23:00 +02:00
2020-09-25 15:37:55 +02:00
2020-09-08 00:09:50 +02:00
2020-09-16 18:52:55 +02:00
2020-09-08 00:09:50 +02:00
2020-09-08 00:09:50 +02:00
2020-09-08 00:09:50 +02:00
2020-09-08 00:09:50 +02:00
2020-09-08 00:09:50 +02:00
2020-10-14 14:23:00 +02:00
2020-09-25 11:13:57 +02:00
2020-10-09 13:17:43 +02:00
2020-09-08 00:09:50 +02:00
2020-09-08 00:09:50 +02:00
2020-09-08 00:09:50 +02:00
2020-09-08 00:09:50 +02:00
2020-10-06 21:51:03 +02:00
2020-09-18 18:55:12 +02:00
2020-09-25 11:13:57 +02:00
2020-09-08 00:09:50 +02:00
2020-09-08 00:09:50 +02:00
2020-09-08 00:09:50 +02:00
2020-09-19 00:25:35 +02:00
2020-09-08 00:09:50 +02:00
2020-09-08 00:09:50 +02:00
2020-10-13 16:35:48 +02:00
2020-09-19 00:25:37 +02:00
2020-09-25 11:13:57 +02:00
2020-10-06 23:44:15 +02:00
2020-10-05 11:11:59 +02:00
2020-09-08 00:09:50 +02:00
2020-10-13 19:26:15 +02:00
2020-09-25 11:13:57 +02:00
2020-09-30 19:19:11 +02:00
2020-09-08 00:09:50 +02:00
2020-10-08 15:09:28 +02:00
2020-10-06 23:39:34 +02:00
2020-10-07 14:51:52 +02:00
2020-10-06 22:54:58 +02:00
2020-10-06 10:31:13 +02:00
2020-09-08 00:09:50 +02:00
2020-09-08 00:09:50 +02:00
2020-09-08 00:09:50 +02:00
2020-09-08 00:11:30 +02:00
2020-09-25 11:13:57 +02:00
2020-09-08 00:09:50 +02:00
2020-10-05 11:56:17 +02:00
2020-10-01 18:43:10 +02:00
2020-10-06 22:55:00 +02:00
2020-10-08 09:43:57 +02:00
2020-09-30 23:16:56 +02:00
2020-10-05 09:49:55 +02:00
b64c2295f7