net/mlx5: DR, Sync chunks only during free

When freeing chunks, we want to sync the steering
so that all the "hot" memory will be written to ICM
and all the chunks that are in the hot_list will be
actually destroyed.
When allocating from the pool, we don't have a need
to sync the steering, as we're not freeing anything,
and sync might just hurt the performance in terms of
flow-per-second offloaded.

Signed-off-by: Erez Shitrit <erezsh@nvidia.com>
Signed-off-by: Yevgeny Kliteynik <kliteyn@nvidia.com>
Reviewed-by: Alex Vesker <valex@nvidia.com>
Reviewed-by: Mark Bloch <mbloch@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
This commit is contained in:
Yevgeny Kliteynik 2020-09-14 15:06:49 +03:00 committed by Saeed Mahameed
parent a00cd87880
commit 3eb1006a3b

View File

@ -332,10 +332,6 @@ static int dr_icm_handle_buddies_get_mem(struct mlx5dr_icm_pool *pool,
bool new_mem = false;
int err;
/* Check if we have chunks that are waiting for sync-ste */
if (dr_icm_pool_is_sync_required(pool))
dr_icm_pool_sync_all_buddy_pools(pool);
alloc_buddy_mem:
/* find the next free place from the buddy list */
list_for_each_entry(buddy_mem_pool, &pool->buddy_mem_list, list_node) {
@ -409,12 +405,18 @@ out:
void mlx5dr_icm_free_chunk(struct mlx5dr_icm_chunk *chunk)
{
struct mlx5dr_icm_buddy_mem *buddy = chunk->buddy_mem;
struct mlx5dr_icm_pool *pool = buddy->pool;
/* move the memory to the waiting list AKA "hot" */
mutex_lock(&buddy->pool->mutex);
mutex_lock(&pool->mutex);
list_move_tail(&chunk->chunk_list, &buddy->hot_list);
buddy->hot_memory_size += chunk->byte_size;
mutex_unlock(&buddy->pool->mutex);
/* Check if we have chunks that are waiting for sync-ste */
if (dr_icm_pool_is_sync_required(pool))
dr_icm_pool_sync_all_buddy_pools(pool);
mutex_unlock(&pool->mutex);
}
struct mlx5dr_icm_pool *mlx5dr_icm_pool_create(struct mlx5dr_domain *dmn,