linux-nfs.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Chuck Lever <chuck.lever@oracle.com>
To: linux-rdma@vger.kernel.org, linux-nfs@vger.kernel.org
Subject: [PATCH v2 11/18] xprtrdma: Save I/O direction in struct rpcrdma_frwr
Date: Mon, 25 Apr 2016 15:22:18 -0400	[thread overview]
Message-ID: <20160425192218.3566.13233.stgit@manet.1015granger.net> (raw)
In-Reply-To: <20160425185956.3566.64142.stgit@manet.1015granger.net>

Move the the I/O direction field from rpcrdma_mr_seg into the
rpcrdma_frmr. This makes it possible to perform the DMA unmapping
long after the rpcrdma_mr_seg has been released and re-used.

Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
---
 net/sunrpc/xprtrdma/frwr_ops.c  |    6 +++---
 net/sunrpc/xprtrdma/xprt_rdma.h |    1 +
 2 files changed, 4 insertions(+), 3 deletions(-)

diff --git a/net/sunrpc/xprtrdma/frwr_ops.c b/net/sunrpc/xprtrdma/frwr_ops.c
index 7c22e9e..e1e6ac1 100644
--- a/net/sunrpc/xprtrdma/frwr_ops.c
+++ b/net/sunrpc/xprtrdma/frwr_ops.c
@@ -416,6 +416,7 @@ frwr_op_map(struct rpcrdma_xprt *r_xprt, struct rpcrdma_mr_seg *seg,
 			break;
 	}
 	frmr->fr_nents = i;
+	frmr->fr_dir = direction;
 
 	dma_nents = ib_dma_map_sg(device, frmr->fr_sg, frmr->fr_nents, direction);
 	if (!dma_nents) {
@@ -455,7 +456,6 @@ frwr_op_map(struct rpcrdma_xprt *r_xprt, struct rpcrdma_mr_seg *seg,
 	if (rc)
 		goto out_senderr;
 
-	seg1->mr_dir = direction;
 	seg1->rl_mw = mw;
 	seg1->mr_rkey = mr->rkey;
 	seg1->mr_base = mr->iova;
@@ -500,7 +500,7 @@ __frwr_dma_unmap(struct rpcrdma_xprt *r_xprt, struct rpcrdma_mr_seg *seg,
 
 	seg->rl_mw = NULL;
 
-	ib_dma_unmap_sg(device, f->fr_sg, f->fr_nents, seg->mr_dir);
+	ib_dma_unmap_sg(device, f->fr_sg, f->fr_nents, f->fr_dir);
 
 	if (!rc)
 		rpcrdma_put_mw(r_xprt, mw);
@@ -611,7 +611,7 @@ frwr_op_unmap(struct rpcrdma_xprt *r_xprt, struct rpcrdma_mr_seg *seg)
 	invalidate_wr->ex.invalidate_rkey = frmr->fr_mr->rkey;
 	DECR_CQCOUNT(&r_xprt->rx_ep);
 
-	ib_dma_unmap_sg(ia->ri_device, frmr->fr_sg, frmr->fr_nents, seg1->mr_dir);
+	ib_dma_unmap_sg(ia->ri_device, frmr->fr_sg, frmr->fr_nents, frmr->fr_dir);
 	read_lock(&ia->ri_qplock);
 	rc = ib_post_send(ia->ri_id->qp, invalidate_wr, &bad_wr);
 	read_unlock(&ia->ri_qplock);
diff --git a/net/sunrpc/xprtrdma/xprt_rdma.h b/net/sunrpc/xprtrdma/xprt_rdma.h
index abfce80..0b3aa54 100644
--- a/net/sunrpc/xprtrdma/xprt_rdma.h
+++ b/net/sunrpc/xprtrdma/xprt_rdma.h
@@ -222,6 +222,7 @@ enum rpcrdma_frmr_state {
 struct rpcrdma_frmr {
 	struct scatterlist		*fr_sg;
 	int				fr_nents;
+	enum dma_data_direction		fr_dir;
 	struct ib_mr			*fr_mr;
 	struct ib_cqe			fr_cqe;
 	enum rpcrdma_frmr_state		fr_state;


  parent reply	other threads:[~2016-04-25 19:22 UTC|newest]

Thread overview: 50+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2016-04-25 19:20 [PATCH v2 00/18] NFS/RDMA client patches for v4.7 Chuck Lever
2016-04-25 19:20 ` [PATCH v2 01/18] sunrpc: Advertise maximum backchannel payload size Chuck Lever
2016-04-25 19:21 ` [PATCH v2 02/18] xprtrdma: Bound the inline threshold values Chuck Lever
2016-04-25 19:21 ` [PATCH v2 03/18] xprtrdma: Limit number of RDMA segments in RPC-over-RDMA headers Chuck Lever
2016-04-26 19:43   ` Sagi Grimberg
2016-04-25 19:21 ` [PATCH v2 04/18] xprtrdma: Prevent inline overflow Chuck Lever
2016-04-26 19:55   ` Sagi Grimberg
2016-04-26 20:04     ` Chuck Lever
2016-04-26 20:42       ` Sagi Grimberg
2016-04-26 20:56         ` Chuck Lever
2016-04-25 19:21 ` [PATCH v2 05/18] xprtrdma: Avoid using Write list for small NFS READ requests Chuck Lever
2016-04-26 19:56   ` Sagi Grimberg
2016-04-25 19:21 ` [PATCH v2 06/18] xprtrdma: Update comments in rpcrdma_marshal_req() Chuck Lever
2016-04-26 19:57   ` Sagi Grimberg
2016-04-25 19:21 ` [PATCH v2 07/18] xprtrdma: Allow Read list and Reply chunk simultaneously Chuck Lever
2016-04-26 20:04   ` Sagi Grimberg
2016-04-25 19:21 ` [PATCH v2 08/18] xprtrdma: Remove rpcrdma_create_chunks() Chuck Lever
2016-04-26 20:04   ` Sagi Grimberg
2016-04-25 19:22 ` [PATCH v2 09/18] xprtrdma: Use core ib_drain_qp() API Chuck Lever
2016-04-26 20:07   ` Sagi Grimberg
2016-04-25 19:22 ` [PATCH v2 10/18] xprtrdma: Rename rpcrdma_frwr::sg and sg_nents Chuck Lever
2016-04-26 20:08   ` Sagi Grimberg
2016-04-25 19:22 ` Chuck Lever [this message]
2016-04-26 20:12   ` [PATCH v2 11/18] xprtrdma: Save I/O direction in struct rpcrdma_frwr Sagi Grimberg
2016-04-26 20:14     ` Chuck Lever
2016-04-25 19:22 ` [PATCH v2 12/18] xprtrdma: Reset MRs in frwr_op_unmap_sync() Chuck Lever
2016-04-26 20:13   ` Sagi Grimberg
2016-04-25 19:22 ` [PATCH v2 13/18] xprtrdma: Refactor the FRWR recovery worker Chuck Lever
2016-04-26 20:16   ` Sagi Grimberg
2016-04-26 20:30     ` Chuck Lever
2016-04-26 20:33       ` Sagi Grimberg
2016-04-25 19:22 ` [PATCH v2 14/18] xprtrdma: Move fr_xprt and fr_worker to struct rpcrdma_mw Chuck Lever
2016-04-26 20:18   ` Sagi Grimberg
2016-04-25 19:22 ` [PATCH v2 15/18] xprtrdma: Refactor __fmr_dma_unmap() Chuck Lever
2016-04-26 20:21   ` Sagi Grimberg
2016-04-25 19:22 ` [PATCH v2 16/18] xprtrdma: Add ro_unmap_safe memreg method Chuck Lever
2016-04-26 20:26   ` Sagi Grimberg
2016-04-26 20:44     ` Chuck Lever
2016-04-27 15:59       ` Removing NFS/RDMA client support for PHYSICAL memory registration Chuck Lever
2016-04-28 10:59         ` Sagi Grimberg
2016-04-25 19:23 ` [PATCH v2 17/18] xprtrdma: Remove ro_unmap() from all registration modes Chuck Lever
2016-04-26 20:29   ` Sagi Grimberg
2016-04-26 20:46     ` Chuck Lever
2016-04-26 20:50       ` Sagi Grimberg
2016-04-25 19:23 ` [PATCH v2 18/18] xprtrdma: Faster server reboot recovery Chuck Lever
2016-04-26 20:31   ` Sagi Grimberg
2016-04-26 14:13 ` [PATCH v2 00/18] NFS/RDMA client patches for v4.7 Steve Wise
2016-04-26 14:57   ` Chuck Lever
2016-04-26 16:45     ` Steve Wise
2016-04-26 17:15       ` Chuck Lever

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=20160425192218.3566.13233.stgit@manet.1015granger.net \
    --to=chuck.lever@oracle.com \
    --cc=linux-nfs@vger.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).