[dpdk-dev] [PATCH v2] ring: guarantee ordering of cons/prod loading when doing

[Jia He]

Hi Jerin


On 11/3/2017 8:56 PM, Jerin Jacob Wrote:
> -----Original Message-----
>>
[...]
>> g like that.
>> Ok, but how to distinguish following 2 options?
> No clearly understood this question. For arm64 case, you can add
> CONFIG_RTE_RING_USE_C11_MEM_MODEL=y in config/defconfig_arm64-armv8a-*
Sorry for my unclear expressions.
I mean there should be one additional config macro besides 
CONFIG_RTE_RING_USE_C11_MEM_MODEL
for users to choose?

i.e.
  - On X86:CONFIG_RTE_RING_USE_C11_MEM_MODEL=n
include rte_ring_generic.h, no changes
- On arm64，CONFIG_RTE_RING_USE_C11_MEM_MODEL=y
include rte_ring_c11_mem.h by default.
In rte_ring_c11_mem.h, implement new version of 
__rte_ring_move_prod_head/__rte_ring_move_cons_head/update_tail

Then, how to distinguish the option of using rte_smp_rmb() or 
__atomic_load/store_n()?

Thanks for the clarification.

-- 
Cheers,
Jia