iv/linux
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

net/mlx5e: RX, Remove internal page_cache

Browse Source 
This patch removes the internal rx page_cache and uses the generic
page_pool api only. It used to be that the page_pool couldn't handle all
the mlx5 driver usecases, but with the introduction of skb recycling and
page fragmentaton in the page_pool full switch can now be made. Some
benfits of this transition:
* Better page recycling in the cases when the page_cache was suffering
  from head of queue blocking. The page_pool doesn't have this issue.
* DMA mapping/unmapping can be managed by the page_pool.
* mlx5e_rq size reduced by more than 50% due to the page_cache array
  being deleted.

This patch only removes the page_cache. Downstream patches will enable
the required page_pool features and will add further fine-tuning.

Signed-off-by: Dragos Tatulea <dtatulea@nvidia.com>
Reviewed-by: Tariq Toukan <tariqt@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
This commit is contained in:
Dragos Tatulea 2022-12-13 14:37:07 +02:00 committed by Saeed Mahameed
parent ca6ef9f031
commit 08c9b61b07
3 changed files with 0 additions and 72 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

6
drivers/net/ethernet/mellanox/mlx5/core/en.h
View File

@ -628,11 +628,6 @@ struct mlx5e_mpw_info {
#define MLX5E_CACHE_UNIT (MLX5_MPWRQ_MAX_PAGES_PER_WQE > NAPI_POLL_WEIGHT ? \
MLX5_MPWRQ_MAX_PAGES_PER_WQE : NAPI_POLL_WEIGHT)
#define MLX5E_CACHE_SIZE (4 * roundup_pow_of_two(MLX5E_CACHE_UNIT))
struct mlx5e_page_cache {
u32 head;
u32 tail;
struct page *page_cache[MLX5E_CACHE_SIZE];
};
struct mlx5e_rq;
typedef void (*mlx5e_fp_handle_rx_cqe)(struct mlx5e_rq*, struct mlx5_cqe64*);
@ -745,7 +740,6 @@ struct mlx5e_rq {
struct mlx5e_rq_stats *stats;
struct mlx5e_cq cq;
struct mlx5e_cq_decomp cqd;
struct mlx5e_page_cache page_cache;
struct hwtstamp_config *tstamp;
struct mlx5_clock *clock;
struct mlx5e_icosq *icosq;

13
drivers/net/ethernet/mellanox/mlx5/core/en_main.c
View File

@ -900,9 +900,6 @@ static int mlx5e_alloc_rq(struct mlx5e_params *params,
rq->dim.mode = DIM_CQ_PERIOD_MODE_START_FROM_EQE;
}
rq->page_cache.head = 0;
rq->page_cache.tail = 0;
return 0;
err_destroy_page_pool:
@ -933,7 +930,6 @@ err_rq_xdp_prog:
static void mlx5e_free_rq(struct mlx5e_rq *rq)
{
struct bpf_prog *old_prog;
int i;
if (xdp_rxq_info_is_reg(&rq->xdp_rxq)) {
old_prog = rcu_dereference_protected(rq->xdp_prog,
@ -953,15 +949,6 @@ static void mlx5e_free_rq(struct mlx5e_rq *rq)
mlx5e_free_wqe_alloc_info(rq);
}
for (i = rq->page_cache.head; i != rq->page_cache.tail;
i = (i + 1) & (MLX5E_CACHE_SIZE - 1)) {
/* With AF_XDP, page_cache is not used, so this loop is not
* entered, and it's safe to call mlx5e_page_release_dynamic
* directly.
*/
mlx5e_page_release_dynamic(rq, rq->page_cache.page_cache[i], false);
}
xdp_rxq_info_unreg(&rq->xdp_rxq);
page_pool_destroy(rq->page_pool);
mlx5_wq_destroy(&rq->wq_ctrl);

53
drivers/net/ethernet/mellanox/mlx5/core/en_rx.c
View File

@ -271,60 +271,10 @@ static inline u32 mlx5e_decompress_cqes_start(struct mlx5e_rq *rq,
return mlx5e_decompress_cqes_cont(rq, wq, 1, budget_rem);
}
static inline bool mlx5e_rx_cache_put(struct mlx5e_rq *rq, struct page *page)
{
struct mlx5e_page_cache *cache = &rq->page_cache;
u32 tail_next = (cache->tail + 1) & (MLX5E_CACHE_SIZE - 1);
struct mlx5e_rq_stats *stats = rq->stats;
if (tail_next == cache->head) {
stats->cache_full++;
return false;
}
if (!dev_page_is_reusable(page)) {
stats->cache_waive++;
return false;
}
cache->page_cache[cache->tail] = page;
cache->tail = tail_next;
return true;
}
static inline bool mlx5e_rx_cache_get(struct mlx5e_rq *rq, struct page **pagep)
{
struct mlx5e_page_cache *cache = &rq->page_cache;
struct mlx5e_rq_stats *stats = rq->stats;
dma_addr_t addr;
if (unlikely(cache->head == cache->tail)) {
stats->cache_empty++;
return false;
}
if (page_ref_count(cache->page_cache[cache->head]) != 1) {
stats->cache_busy++;
return false;
}
*pagep = cache->page_cache[cache->head];
cache->head = (cache->head + 1) & (MLX5E_CACHE_SIZE - 1);
stats->cache_reuse++;
addr = page_pool_get_dma_addr(*pagep);
/* Non-XSK always uses PAGE_SIZE. */
dma_sync_single_for_device(rq->pdev, addr, PAGE_SIZE, rq->buff.map_dir);
return true;
}
static inline int mlx5e_page_alloc_pool(struct mlx5e_rq *rq, struct page **pagep)
{
dma_addr_t addr;
if (mlx5e_rx_cache_get(rq, pagep))
return 0;
*pagep = page_pool_dev_alloc_pages(rq->page_pool);
if (unlikely(!*pagep))
return -ENOMEM;
@ -353,9 +303,6 @@ void mlx5e_page_dma_unmap(struct mlx5e_rq *rq, struct page *page)
void mlx5e_page_release_dynamic(struct mlx5e_rq *rq, struct page *page, bool recycle)
{
if (likely(recycle)) {
if (mlx5e_rx_cache_put(rq, page))
return;
mlx5e_page_dma_unmap(rq, page);
page_pool_recycle_direct(rq->page_pool, page);
} else {