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

RDMA/qedr: Use for_each_sg_dma_page iterator on umem SGL

Browse Source 
Use the for_each_sg_dma_page iterator variant to walk the umem DMA-mapped
SGL and get the page DMA address. This avoids the extra loop to iterate
pages in the SGE when for_each_sg iterator is used.

Additionally, purge umem->page_shift usage in the driver as its only
relevant for ODP MRs. Use system page size and shift instead.

Signed-off-by: Shiraz, Saleem <shiraz.saleem@intel.com>
Acked-by: Michal Kalderon <michal.kalderon@marvell.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
This commit is contained in:
Shiraz, Saleem 2019-02-11 09:25:04 -06:00 committed by Jason Gunthorpe
parent f3e6d31179
commit 95ad233ffb
1 changed files with 27 additions and 34 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

61
drivers/infiniband/hw/qedr/verbs.c
View File

@ -620,13 +620,12 @@ static void qedr_populate_pbls(struct qedr_dev *dev, struct ib_umem *umem,
struct qedr_pbl *pbl, struct qedr_pbl *pbl,
struct qedr_pbl_info *pbl_info, u32 pg_shift) struct qedr_pbl_info *pbl_info, u32 pg_shift)
{ {
int shift, pg_cnt, pages, pbe_cnt, total_num_pbes = 0; int pbe_cnt, total_num_pbes = 0;
u32 fw_pg_cnt, fw_pg_per_umem_pg; u32 fw_pg_cnt, fw_pg_per_umem_pg;
struct qedr_pbl *pbl_tbl; struct qedr_pbl *pbl_tbl;
struct scatterlist *sg; struct sg_dma_page_iter sg_iter;
struct regpair *pbe; struct regpair *pbe;
u64 pg_addr; u64 pg_addr;
int entry;
if (!pbl_info->num_pbes) if (!pbl_info->num_pbes)
return; return;
@ -647,38 +646,32 @@ static void qedr_populate_pbls(struct qedr_dev *dev, struct ib_umem *umem,
pbe_cnt = 0; pbe_cnt = 0;
shift = umem->page_shift; fw_pg_per_umem_pg = BIT(PAGE_SHIFT - pg_shift);
fw_pg_per_umem_pg = BIT(umem->page_shift - pg_shift); for_each_sg_dma_page (umem->sg_head.sgl, &sg_iter, umem->nmap, 0) {
pg_addr = sg_page_iter_dma_address(&sg_iter);
for (fw_pg_cnt = 0; fw_pg_cnt < fw_pg_per_umem_pg;) {
pbe->lo = cpu_to_le32(pg_addr);
pbe->hi = cpu_to_le32(upper_32_bits(pg_addr));
for_each_sg(umem->sg_head.sgl, sg, umem->nmap, entry) { pg_addr += BIT(pg_shift);
pages = sg_dma_len(sg) >> shift; pbe_cnt++;
pg_addr = sg_dma_address(sg); total_num_pbes++;
for (pg_cnt = 0; pg_cnt < pages; pg_cnt++) { pbe++;
for (fw_pg_cnt = 0; fw_pg_cnt < fw_pg_per_umem_pg;) {
pbe->lo = cpu_to_le32(pg_addr);
pbe->hi = cpu_to_le32(upper_32_bits(pg_addr));
pg_addr += BIT(pg_shift); if (total_num_pbes == pbl_info->num_pbes)
pbe_cnt++; return;
total_num_pbes++;
pbe++;
if (total_num_pbes == pbl_info->num_pbes) /* If the given pbl is full storing the pbes,
return; * move to next pbl.
*/
/* If the given pbl is full storing the pbes, if (pbe_cnt == (pbl_info->pbl_size / sizeof(u64))) {
* move to next pbl. pbl_tbl++;
*/ pbe = (struct regpair *)pbl_tbl->va;
if (pbe_cnt == pbe_cnt = 0;
(pbl_info->pbl_size / sizeof(u64))) {
pbl_tbl++;
pbe = (struct regpair *)pbl_tbl->va;
pbe_cnt = 0;
}
fw_pg_cnt++;
} }
fw_pg_cnt++;
} }
} }
} }
@ -739,7 +732,7 @@ static inline int qedr_init_user_queue(struct ib_udata *udata,
} }
fw_pages = ib_umem_page_count(q->umem) << fw_pages = ib_umem_page_count(q->umem) <<
(q->umem->page_shift - FW_PAGE_SHIFT); (PAGE_SHIFT - FW_PAGE_SHIFT);
rc = qedr_prepare_pbl_tbl(dev, &q->pbl_info, fw_pages, 0); rc = qedr_prepare_pbl_tbl(dev, &q->pbl_info, fw_pages, 0);
if (rc) if (rc)
@ -1455,7 +1448,7 @@ struct ib_srq *qedr_create_srq(struct ib_pd *ibpd,
page_cnt = srq->usrq.pbl_info.num_pbes; page_cnt = srq->usrq.pbl_info.num_pbes;
pbl_base_addr = srq->usrq.pbl_tbl->pa; pbl_base_addr = srq->usrq.pbl_tbl->pa;
phy_prod_pair_addr = hw_srq->phy_prod_pair_addr; phy_prod_pair_addr = hw_srq->phy_prod_pair_addr;
page_size = BIT(srq->usrq.umem->page_shift); page_size = PAGE_SIZE;
} else { } else {
struct qed_chain *pbl; struct qed_chain *pbl;
@ -2707,7 +2700,7 @@ struct ib_mr *qedr_reg_user_mr(struct ib_pd *ibpd, u64 start, u64 len,
goto err1; goto err1;
qedr_populate_pbls(dev, mr->umem, mr->info.pbl_table, qedr_populate_pbls(dev, mr->umem, mr->info.pbl_table,
&mr->info.pbl_info, mr->umem->page_shift); &mr->info.pbl_info, PAGE_SHIFT);
rc = dev->ops->rdma_alloc_tid(dev->rdma_ctx, &mr->hw_mr.itid); rc = dev->ops->rdma_alloc_tid(dev->rdma_ctx, &mr->hw_mr.itid);
if (rc) { if (rc) {
@ -2728,7 +2721,7 @@ struct ib_mr *qedr_reg_user_mr(struct ib_pd *ibpd, u64 start, u64 len,
mr->hw_mr.pbl_ptr = mr->info.pbl_table[0].pa; mr->hw_mr.pbl_ptr = mr->info.pbl_table[0].pa;
mr->hw_mr.pbl_two_level = mr->info.pbl_info.two_layered; mr->hw_mr.pbl_two_level = mr->info.pbl_info.two_layered;
mr->hw_mr.pbl_page_size_log = ilog2(mr->info.pbl_info.pbl_size); mr->hw_mr.pbl_page_size_log = ilog2(mr->info.pbl_info.pbl_size);
mr->hw_mr.page_size_log = mr->umem->page_shift; mr->hw_mr.page_size_log = PAGE_SHIFT;
mr->hw_mr.fbo = ib_umem_offset(mr->umem); mr->hw_mr.fbo = ib_umem_offset(mr->umem);
mr->hw_mr.length = len; mr->hw_mr.length = len;
mr->hw_mr.vaddr = usr_addr; mr->hw_mr.vaddr = usr_addr;