Commit 4f56868b authored by Dragos Tatulea's avatar Dragos Tatulea Committed by Jakub Kicinski
Browse files

net/mlx5e: SHAMPO, Change frag page setup order during allocation 



Now that the UMR allocation has been simplified, it is no longer
possible to have a leftover page from a previous call to
mlx5e_build_shampo_hd_umr().

This patch simplifies the code by switching the order of operations:
first take the frag page and then increment the index. This is more
straightforward and it also paves the way for dropping the info
array.

Signed-off-by: default avatarDragos Tatulea <dtatulea@nvidia.com>
Signed-off-by: default avatarTariq Toukan <tariqt@nvidia.com>
Link: https://patch.msgid.link/20241107194357.683732-11-tariqt@nvidia.com


Signed-off-by: default avatarJakub Kicinski <kuba@kernel.org>
parent 1a4b5885

drivers/net/ethernet/mellanox/mlx5/core/en_rx.c

+2 −4
Original line number Diff line number Diff line
@@ -651,7 +651,7 @@ static int mlx5e_build_shampo_hd_umr(struct mlx5e_rq *rq, 
	u16 pi, header_offset, err, wqe_bbs;

	u32 lkey = rq->mdev->mlx5e_res.hw_objs.mkey;

	u16 page_index = shampo->curr_page_index;

	struct mlx5e_frag_page *frag_page;

	struct mlx5e_frag_page *frag_page = NULL;

	struct mlx5e_dma_info *dma_info;

	struct mlx5e_umr_wqe *umr_wqe;

	int headroom, i;
@@ -663,16 +663,14 @@ static int mlx5e_build_shampo_hd_umr(struct mlx5e_rq *rq, 
	umr_wqe = mlx5_wq_cyc_get_wqe(&sq->wq, pi);

	build_ksm_umr(sq, umr_wqe, shampo->key, index, ksm_entries);



	frag_page = &shampo->pages[page_index];



	WARN_ON_ONCE(ksm_entries & (MLX5E_SHAMPO_WQ_HEADER_PER_PAGE - 1));

	for (i = 0; i < ksm_entries; i++, index++) {

		dma_info = &shampo->info[index];

		header_offset = (index & (MLX5E_SHAMPO_WQ_HEADER_PER_PAGE - 1)) <<

			MLX5E_SHAMPO_LOG_MAX_HEADER_ENTRY_SIZE;

		if (!(header_offset & (PAGE_SIZE - 1))) {

			page_index = (page_index + 1) & (shampo->pages_per_wq - 1);

			frag_page = &shampo->pages[page_index];

			page_index = (page_index + 1) & (shampo->pages_per_wq - 1);



			err = mlx5e_page_alloc_fragmented(rq, frag_page);

			if (unlikely(err))