From: Chuck Lever <chuck.lever@oracle.com>
To: bfields@fieldses.org
Cc: linux-nfs@vger.kernel.org, linux-rdma@vger.kernel.org
Subject: [PATCH v2 08/10] svcrdma: Support RDMA_NOMSG requests
Date: Tue, 13 Jan 2015 11:03:37 -0500 [thread overview]
Message-ID: <20150113160336.8118.32990.stgit@klimt.1015granger.net> (raw)
In-Reply-To: <20150113155904.8118.57718.stgit@klimt.1015granger.net>
Currently the Linux server can not decode RDMA_NOMSG type requests.
Operations whose length exceeds the fixed size of RDMA SEND buffers,
like large NFSv4 CREATE(NF4LNK) operations, must be conveyed via
RDMA_NOMSG.
For an RDMA_MSG type request, the client sends the RPC/RDMA, RPC
headers, and some or all of the NFS arguments via RDMA SEND.
For an RDMA_NOMSG type request, the client sends just the RPC/RDMA
header via RDMA SEND. The request's read list contains elements for
the entire RPC message, including the RPC header.
NFSD expects the RPC/RMDA header and RPC header to be contiguous in
page zero of the XDR buffer. Add logic in the RDMA READ path to make
the read list contents land where the server prefers, when the
incoming message is a type RDMA_NOMSG message.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
---
include/linux/sunrpc/svc_rdma.h | 1 +
net/sunrpc/xprtrdma/svc_rdma_recvfrom.c | 39 +++++++++++++++++++++++++++++--
2 files changed, 37 insertions(+), 3 deletions(-)
diff --git a/include/linux/sunrpc/svc_rdma.h b/include/linux/sunrpc/svc_rdma.h
index f161e30..c343a94 100644
--- a/include/linux/sunrpc/svc_rdma.h
+++ b/include/linux/sunrpc/svc_rdma.h
@@ -79,6 +79,7 @@ struct svc_rdma_op_ctxt {
enum ib_wr_opcode wr_op;
enum ib_wc_status wc_status;
u32 byte_len;
+ u32 position;
struct svcxprt_rdma *xprt;
unsigned long flags;
enum dma_data_direction direction;
diff --git a/net/sunrpc/xprtrdma/svc_rdma_recvfrom.c b/net/sunrpc/xprtrdma/svc_rdma_recvfrom.c
index a67dd1a..36cf51a 100644
--- a/net/sunrpc/xprtrdma/svc_rdma_recvfrom.c
+++ b/net/sunrpc/xprtrdma/svc_rdma_recvfrom.c
@@ -60,6 +60,7 @@ static void rdma_build_arg_xdr(struct svc_rqst *rqstp,
struct svc_rdma_op_ctxt *ctxt,
u32 byte_count)
{
+ struct rpcrdma_msg *rmsgp;
struct page *page;
u32 bc;
int sge_no;
@@ -82,7 +83,14 @@ static void rdma_build_arg_xdr(struct svc_rqst *rqstp,
/* If data remains, store it in the pagelist */
rqstp->rq_arg.page_len = bc;
rqstp->rq_arg.page_base = 0;
- rqstp->rq_arg.pages = &rqstp->rq_pages[1];
+
+ /* RDMA_NOMSG: RDMA READ data should land just after RDMA RECV data */
+ rmsgp = (struct rpcrdma_msg *)rqstp->rq_arg.head[0].iov_base;
+ if (be32_to_cpu(rmsgp->rm_type) == RDMA_NOMSG)
+ rqstp->rq_arg.pages = &rqstp->rq_pages[0];
+ else
+ rqstp->rq_arg.pages = &rqstp->rq_pages[1];
+
sge_no = 1;
while (bc && sge_no < ctxt->count) {
page = ctxt->pages[sge_no];
@@ -383,7 +391,6 @@ static int rdma_read_chunks(struct svcxprt_rdma *xprt,
*/
head->arg.head[0] = rqstp->rq_arg.head[0];
head->arg.tail[0] = rqstp->rq_arg.tail[0];
- head->arg.pages = &head->pages[head->count];
head->hdr_count = head->count;
head->arg.page_base = 0;
head->arg.page_len = 0;
@@ -393,9 +400,17 @@ static int rdma_read_chunks(struct svcxprt_rdma *xprt,
ch = (struct rpcrdma_read_chunk *)&rmsgp->rm_body.rm_chunks[0];
position = be32_to_cpu(ch->rc_position);
+ /* RDMA_NOMSG: RDMA READ data should land just after RDMA RECV data */
+ if (position == 0) {
+ head->arg.pages = &head->pages[0];
+ page_offset = head->byte_len;
+ } else {
+ head->arg.pages = &head->pages[head->count];
+ page_offset = 0;
+ }
+
ret = 0;
page_no = 0;
- page_offset = 0;
for (; ch->rc_discrim != xdr_zero; ch++) {
if (be32_to_cpu(ch->rc_position) != position)
goto err;
@@ -418,7 +433,10 @@ static int rdma_read_chunks(struct svcxprt_rdma *xprt,
head->arg.buflen += ret;
}
}
+
ret = 1;
+ head->position = position;
+
err:
/* Detach arg pages. svc_recv will replenish them */
for (page_no = 0;
@@ -465,6 +483,21 @@ static int rdma_read_complete(struct svc_rqst *rqstp,
put_page(rqstp->rq_pages[page_no]);
rqstp->rq_pages[page_no] = head->pages[page_no];
}
+
+ /* Adjustments made for RDMA_NOMSG type requests */
+ if (head->position == 0) {
+ if (head->arg.len <= head->sge[0].length) {
+ head->arg.head[0].iov_len = head->arg.len -
+ head->byte_len;
+ head->arg.page_len = 0;
+ } else {
+ head->arg.head[0].iov_len = head->sge[0].length -
+ head->byte_len;
+ head->arg.page_len = head->arg.len -
+ head->sge[0].length;
+ }
+ }
+
/* Point rq_arg.pages past header */
rdma_fix_xdr_pad(&head->arg);
rqstp->rq_arg.pages = &rqstp->rq_pages[head->hdr_count];
next prev parent reply other threads:[~2015-01-13 16:03 UTC|newest]
Thread overview: 13+ messages / expand[flat|nested] mbox.gz Atom feed top
2015-01-13 16:02 [PATCH v2 00/10] NFS/RDMA server for 3.20 Chuck Lever
2015-01-13 16:02 ` [PATCH v2 01/10] svcrdma: Clean up dprintk Chuck Lever
2015-01-13 16:02 ` [PATCH v2 02/10] svcrdma: Remove unused variable Chuck Lever
2015-01-13 16:02 ` [PATCH v2 03/10] svcrdma: Clean up read chunk counting Chuck Lever
2015-01-13 16:03 ` [PATCH v2 04/10] svcrdma: Scrub BUG_ON() and WARN_ON() call sites Chuck Lever
2015-01-13 16:03 ` [PATCH v2 05/10] svcrdma: Find rmsgp more reliably Chuck Lever
2015-01-13 16:03 ` [PATCH v2 06/10] svcrdma: Plant reader function in struct svcxprt_rdma Chuck Lever
2015-01-13 16:03 ` [PATCH v2 07/10] svcrdma: rc_position sanity checking Chuck Lever
2015-01-13 16:03 ` Chuck Lever [this message]
2015-01-13 16:03 ` [PATCH v2 09/10] svcrdma: Move read list XDR round-up logic Chuck Lever
2015-01-13 16:03 ` [PATCH v2 10/10] svcrdma: Handle additional inline content Chuck Lever
2015-01-13 17:31 ` [PATCH v2 00/10] NFS/RDMA server for 3.20 Steve Wise
2015-01-15 19:01 ` J. Bruce Fields
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=20150113160336.8118.32990.stgit@klimt.1015granger.net \
--to=chuck.lever@oracle.com \
--cc=bfields@fieldses.org \
--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