For a typical Linux server, probably there are several hardware modules.
For example, numa node0 has a compressor, numa node2 has a same module.
Some drivers are automatically using the module near the CPU calling
acomp_alloc.
But it isn't necessarily correct. Just like memory allocation API like
kmalloc and kmalloc_node. Similar optimization may be done for crypto.
-v2:
* fix kern-doc and some codingstyle issues according to Jonathan's comment
* patch 3/3 is rebased againest "[PATCH] mm/zswap: careful error path
implementation in comp_prepare"[1]
[1] https://lkml.org/lkml/2020/6/22/347
Barry Song (3):
crypto: permit users to specify numa node of acomp hardware
crypto: hisilicon/zip - permit users to specify NUMA node
mm/zswap: allocate acomp on the numa node committing acomp_req
crypto/acompress.c | 8 ++++++++
crypto/api.c | 22 ++++++++++++++--------
crypto/internal.h | 23 +++++++++++++++++++----
drivers/crypto/hisilicon/zip/zip.h | 2 +-
drivers/crypto/hisilicon/zip/zip_crypto.c | 6 +++---
drivers/crypto/hisilicon/zip/zip_main.c | 5 +++--
include/crypto/acompress.h | 18 ++++++++++++++++++
include/linux/crypto.h | 2 ++
mm/zswap.c | 2 +-
9 files changed, 69 insertions(+), 19 deletions(-)
--
2.27.0
