This bug makes 'lock free reader/writer concurrency hash' unusable.
There are two reasons:
1] memory leak because we cannot free keys which indexes greater than the
number of total entries
2] the ring of free_slots may have unexpected key conflict with in use one
The patch fixes these 2 issues, both of which are in the same API:
int __rte_experimental
rte_hash_free_key_with_position(const struct rte_hash *h,
const int32_t position)
I don't think it is necessarily to split in 2 patches.
Thanks,
Linfan
At 2019-05-09 16:29:56, "Thomas Monjalon" <tho...@monjalon.net> wrote:
>09/05/2019 00:59, Dharmik Thakkar:
>> This patch fixes 2 bugs-
>> 1] Incorrect position returned to the free slots.
>> 2] Incorrect computation of total_entries
>
>Is it possible to split in 2 patches?
>
>How critical is this bug? It looks old.
>I'm afraid it will miss 19.05.
>
>
