There are 2 model barrier options in librte_ring suggested by Jerin:
1. use rte_smp_rmb
2. use load_acquire/store_release
CONFIG_RTE_RING_USE_C11_MEM_MODEL is provided for supporting C11 memory
model barrier in librte_ring. By default it is "y" on arm64, "n" on any
other architectures so far.
Already fuctionally tested on the machines as follows:
- on X86
- on arm64 with CONFIG_RTE_RING_USE_C11_MEM_MODEL=y
- on arm64 with CONFIG_RTE_RING_USE_C11_MEM_MODEL=n
---
Changelog:
V9: remove the SPDX tag and refine commit logs
V8: Change the license to SPDX tag. Change USE_C11_MEM_MODEL to "n" on
any architectures by default
V7: fix check-git-log warnings which is suggested by Jerin
V6: minor change in subject and log
V5: split it into 2 patchset due to the milestone concerns, this is the 2st
one. Also fix checkpatch.pl warnings
V4: split into small patches
V3: arch specific implementation for enqueue/dequeue barrier
V2: let users choose whether using load_acquire/store_release
V1: rte_smp_rmb() between 2 loads
Jia He (3):
eal/arm64: remove the braces {} for dmb() and dsb()
ring: introduce new header file to include common functions
ring: introduce new header file to support C11 memory model
config/common_armv8a_linuxapp | 2 +
config/common_base | 5 +
.../common/include/arch/arm/rte_atomic_64.h | 4 +-
lib/librte_eventdev/rte_event_ring.h | 6 +-
lib/librte_ring/Makefile | 4 +-
lib/librte_ring/rte_ring.h | 173 ++--------------
lib/librte_ring/rte_ring_c11_mem.h | 218 ++++++++++++++++++++
lib/librte_ring/rte_ring_generic.h | 228 +++++++++++++++++++++
8 files changed, 476 insertions(+), 164 deletions(-)
create mode 100644 lib/librte_ring/rte_ring_c11_mem.h
create mode 100644 lib/librte_ring/rte_ring_generic.h
--
2.7.4
