From: Jianxin Xiong <jianxin.xiong@intel.com>
To: linux-rdma@vger.kernel.org, dri-devel@lists.freedesktop.org
Cc: Leon Romanovsky <leon@kernel.org>, Jason Gunthorpe <jgg@ziepe.ca>,
Doug Ledford <dledford@redhat.com>,
Daniel Vetter <daniel.vetter@intel.com>,
Christian Koenig <christian.koenig@amd.com>,
Jianxin Xiong <jianxin.xiong@intel.com>
Subject: [RFC PATCH v3 3/4] RDMA/mlx5: Support dma-buf based userspace memory region
Date: Sun, 4 Oct 2020 12:12:30 -0700 [thread overview]
Message-ID: <1601838751-148544-4-git-send-email-jianxin.xiong@intel.com> (raw)
In-Reply-To: <1601838751-148544-1-git-send-email-jianxin.xiong@intel.com>
Recognize the new access flag and call the core function to import
dma-buf based memory region.
Since the invalidation callback is called from the dma-buf driver
instead of mmu_interval_notifier, the part inside the ODP pagefault
handler that checks for on-going invalidation is modified to handle
the difference.
Signed-off-by: Jianxin Xiong <jianxin.xiong@intel.com>
Reviewed-by: Sean Hefty <sean.hefty@intel.com>
Acked-by: Michael J. Ruhl <michael.j.ruhl@intel.com>
---
drivers/infiniband/hw/mlx5/mr.c | 50 +++++++++++++++++++++++++++++++++++++---
drivers/infiniband/hw/mlx5/odp.c | 22 ++++++++++++++++--
2 files changed, 67 insertions(+), 5 deletions(-)
diff --git a/drivers/infiniband/hw/mlx5/mr.c b/drivers/infiniband/hw/mlx5/mr.c
index 3c91e32..d58be20 100644
--- a/drivers/infiniband/hw/mlx5/mr.c
+++ b/drivers/infiniband/hw/mlx5/mr.c
@@ -912,6 +912,44 @@ static int mr_umem_get(struct mlx5_ib_dev *dev, u64 start, u64 length,
return 0;
}
+static int mr_umem_dmabuf_get(struct mlx5_ib_dev *dev, u64 start, u64 length,
+ int dmabuf_fd, int access_flags,
+ struct ib_umem **umem, int *npages,
+ int *page_shift, int *ncont, int *order)
+{
+ struct ib_umem *u;
+ struct ib_umem_odp *odp;
+
+ *umem = NULL;
+
+ u = ib_umem_dmabuf_get(&dev->ib_dev, start, length, dmabuf_fd,
+ access_flags, &mlx5_mn_ops);
+ if (IS_ERR(u)) {
+ mlx5_ib_dbg(dev, "umem get failed (%ld)\n", PTR_ERR(u));
+ return PTR_ERR(u);
+ }
+
+ odp = to_ib_umem_odp(u);
+ *page_shift = odp->page_shift;
+ *ncont = ib_umem_odp_num_pages(odp);
+ *npages = *ncont << (*page_shift - PAGE_SHIFT);
+ if (order)
+ *order = ilog2(roundup_pow_of_two(*ncont));
+
+ if (!*npages) {
+ mlx5_ib_warn(dev, "avoid zero region\n");
+ ib_umem_release(u);
+ return -EINVAL;
+ }
+
+ *umem = u;
+
+ mlx5_ib_dbg(dev, "npages %d, ncont %d, order %d, page_shift %d\n",
+ *npages, *ncont, *order, *page_shift);
+
+ return 0;
+}
+
static void mlx5_ib_umr_done(struct ib_cq *cq, struct ib_wc *wc)
{
struct mlx5_ib_umr_context *context =
@@ -1382,9 +1420,15 @@ struct ib_mr *mlx5_ib_reg_user_mr(struct ib_pd *pd,
return &mr->ibmr;
}
- err = mr_umem_get(dev, attr->start, attr->length,
- attr->access_flags, &umem,
- &npages, &page_shift, &ncont, &order);
+ if (attr->access_flags & IB_ACCESS_DMABUF)
+ err = mr_umem_dmabuf_get(dev, attr->start, attr->length,
+ attr->fd, attr->access_flags,
+ &umem, &npages, &page_shift, &ncont,
+ &order);
+ else
+ err = mr_umem_get(dev, attr->start, attr->length,
+ attr->access_flags, &umem,
+ &npages, &page_shift, &ncont, &order);
if (err < 0)
return ERR_PTR(err);
diff --git a/drivers/infiniband/hw/mlx5/odp.c b/drivers/infiniband/hw/mlx5/odp.c
index cfd7efa..f2ca3f8 100644
--- a/drivers/infiniband/hw/mlx5/odp.c
+++ b/drivers/infiniband/hw/mlx5/odp.c
@@ -665,6 +665,24 @@ void mlx5_ib_fence_odp_mr(struct mlx5_ib_mr *mr)
dma_fence_odp_mr(mr);
}
+static inline unsigned long notifier_read_begin(struct ib_umem_odp *odp)
+{
+ if (odp->umem.is_dmabuf)
+ return ib_umem_dmabuf_notifier_read_begin(odp);
+ else
+ return mmu_interval_read_begin(&odp->notifier);
+}
+
+static inline int notifier_read_retry(struct ib_umem_odp *odp,
+ unsigned long current_seq)
+{
+ if (odp->umem.is_dmabuf) {
+ return ib_umem_dmabuf_notifier_read_retry(odp, current_seq);
+ } else {
+ return mmu_interval_read_retry(&odp->notifier, current_seq);
+ }
+}
+
#define MLX5_PF_FLAGS_DOWNGRADE BIT(1)
static int pagefault_real_mr(struct mlx5_ib_mr *mr, struct ib_umem_odp *odp,
u64 user_va, size_t bcnt, u32 *bytes_mapped,
@@ -683,7 +701,7 @@ static int pagefault_real_mr(struct mlx5_ib_mr *mr, struct ib_umem_odp *odp,
if (odp->umem.writable && !downgrade)
access_mask |= ODP_WRITE_ALLOWED_BIT;
- current_seq = mmu_interval_read_begin(&odp->notifier);
+ current_seq = notifier_read_begin(odp);
np = ib_umem_odp_map_dma_pages(odp, user_va, bcnt, access_mask,
current_seq);
@@ -691,7 +709,7 @@ static int pagefault_real_mr(struct mlx5_ib_mr *mr, struct ib_umem_odp *odp,
return np;
mutex_lock(&odp->umem_mutex);
- if (!mmu_interval_read_retry(&odp->notifier, current_seq)) {
+ if (!notifier_read_retry(odp, current_seq)) {
/*
* No need to check whether the MTTs really belong to
* this MR, since ib_umem_odp_map_dma_pages already
--
1.8.3.1
_______________________________________________
dri-devel mailing list
dri-devel@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/dri-devel
next prev parent reply other threads:[~2020-10-04 19:00 UTC|newest]
Thread overview: 22+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-10-04 19:12 [RFC PATCH v3 0/4] RDMA: Add dma-buf support Jianxin Xiong
2020-10-04 19:12 ` [RFC PATCH v3 1/4] RDMA/umem: Support importing dma-buf as user memory region Jianxin Xiong
2020-10-05 10:54 ` Christian König
2020-10-05 16:19 ` Xiong, Jianxin
2020-10-05 13:13 ` Jason Gunthorpe
2020-10-05 16:18 ` Xiong, Jianxin
2020-10-05 16:33 ` Jason Gunthorpe
2020-10-05 19:41 ` Xiong, Jianxin
2020-10-06 9:22 ` Daniel Vetter
2020-10-06 15:26 ` Xiong, Jianxin
2020-10-06 15:49 ` Jason Gunthorpe
2020-10-06 16:34 ` Daniel Vetter
2020-10-06 17:24 ` Daniel Vetter
2020-10-06 18:02 ` Jason Gunthorpe
2020-10-06 18:17 ` Daniel Vetter
2020-10-06 18:38 ` Jason Gunthorpe
2020-10-06 19:12 ` Daniel Vetter
2020-10-07 7:13 ` Christian König
2020-10-06 16:40 ` Daniel Vetter
2020-10-04 19:12 ` [RFC PATCH v3 2/4] RDMA: Expand driver memory registration methods to support dma-buf Jianxin Xiong
2020-10-04 19:12 ` Jianxin Xiong [this message]
2020-10-04 19:12 ` [RFC PATCH v3 4/4] RDMA/uverbs: Add uverbs command for dma-buf based MR registration Jianxin Xiong
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=1601838751-148544-4-git-send-email-jianxin.xiong@intel.com \
--to=jianxin.xiong@intel.com \
--cc=christian.koenig@amd.com \
--cc=daniel.vetter@intel.com \
--cc=dledford@redhat.com \
--cc=dri-devel@lists.freedesktop.org \
--cc=jgg@ziepe.ca \
--cc=leon@kernel.org \
--cc=linux-rdma@vger.kernel.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).