acl: add 512-bit AVX512 classify method
authorKonstantin Ananyev <konstantin.ananyev@intel.com>
Tue, 6 Oct 2020 15:03:12 +0000 (16:03 +0100)
committerDavid Marchand <david.marchand@redhat.com>
Wed, 14 Oct 2020 12:23:01 +0000 (14:23 +0200)
commit45da22e42ec391702d32848949f205f1b363c853
tree8fbc52848cfffbafb6351b7d213813f2164fa39f
parent867d0d3649d8edda60c3fc863f9e491cca2c64d9
acl: add 512-bit AVX512 classify method

Introduce classify implementation that uses AVX512 specific ISA.
rte_acl_classify_avx512x32() is able to process up to 32 flows in parallel.
It uses 512-bit width registers/instructions and provides higher
performance then rte_acl_classify_avx512x16(), but can cause
frequency level change.
Note that for now only 64-bit version is supported.

Signed-off-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
app/test-acl/main.c
app/test/test_acl.c
doc/guides/prog_guide/packet_classif_access_ctrl.rst
doc/guides/rel_notes/release_20_11.rst
lib/librte_acl/acl_run_avx512.c
lib/librte_acl/acl_run_avx512x16.h [new file with mode: 0644]