From: Long Li <longli@microsoft.com>
To: Tom Talpey <tom@talpey.com>, Steve French <sfrench@samba.org>,
"linux-cifs@vger.kernel.org" <linux-cifs@vger.kernel.org>,
"samba-technical@lists.samba.org"
<samba-technical@lists.samba.org>,
"linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>,
"linux-rdma@vger.kernel.org" <linux-rdma@vger.kernel.org>
Subject: RE: [Patch v2 09/15] CIFS: SMBD: Support page offset in RDMA recv
Date: Mon, 25 Jun 2018 21:29:17 +0000 [thread overview]
Message-ID: <CY4PR2101MB07221AFF7B75D21C00C60035CE4A0@CY4PR2101MB0722.namprd21.prod.outlook.com> (raw)
In-Reply-To: <1e6b7cb0-98e8-d1fd-cb59-d3344fc70b19@talpey.com>
> Subject: Re: [Patch v2 09/15] CIFS: SMBD: Support page offset in RDMA recv
>
> On 5/30/2018 3:48 PM, Long Li wrote:
> > From: Long Li <longli@microsoft.com>
> >
> > RDMA recv function needs to place data to the correct place starting
> > at page offset.
> >
> > Signed-off-by: Long Li <longli@microsoft.com>
> > ---
> > fs/cifs/smbdirect.c | 18 +++++++++++-------
> > 1 file changed, 11 insertions(+), 7 deletions(-)
> >
> > diff --git a/fs/cifs/smbdirect.c b/fs/cifs/smbdirect.c index
> > 6141e3c..ba53c52 100644
> > --- a/fs/cifs/smbdirect.c
> > +++ b/fs/cifs/smbdirect.c
> > @@ -2004,10 +2004,12 @@ static int smbd_recv_buf(struct
> smbd_connection *info, char *buf,
> > * return value: actual data read
> > */
> > static int smbd_recv_page(struct smbd_connection *info,
> > - struct page *page, unsigned int to_read)
> > + struct page *page, unsigned int page_offset,
> > + unsigned int to_read)
> > {
> > int ret;
> > char *to_address;
> > + void *page_address;
> >
> > /* make sure we have the page ready for read */
> > ret = wait_event_interruptible(
> > @@ -2015,16 +2017,17 @@ static int smbd_recv_page(struct
> smbd_connection *info,
> > info->reassembly_data_length >= to_read ||
> > info->transport_status != SMBD_CONNECTED);
> > if (ret)
> > - return 0;
> > + return ret;
> >
> > /* now we can read from reassembly queue and not sleep */
> > - to_address = kmap_atomic(page);
> > + page_address = kmap_atomic(page);
> > + to_address = (char *) page_address + page_offset;
> >
> > log_read(INFO, "reading from page=%p address=%p to_read=%d\n",
> > page, to_address, to_read);
> >
> > ret = smbd_recv_buf(info, to_address, to_read);
> > - kunmap_atomic(to_address);
> > + kunmap_atomic(page_address);
>
> Is "page" truly not mapped? This kmap/kunmap for each received 4KB is very
> expensive. Is there not a way to keep a kva for the reassembly queue
> segments?
I will find a way to not map those upper layer pages when doing RDMA receive. The reason for doing this is to use a common layer (that will greatly simplifies the code) for transport. It's probably a waste of time when only pages are involved, and not need to be mapped.
I will address those in a separate patch.
>
> >
> > return ret;
> > }
> > @@ -2038,7 +2041,7 @@ int smbd_recv(struct smbd_connection *info,
> struct msghdr *msg)
> > {
> > char *buf;
> > struct page *page;
> > - unsigned int to_read;
> > + unsigned int to_read, page_offset;
> > int rc;
> >
> > info->smbd_recv_pending++;
> > @@ -2052,15 +2055,16 @@ int smbd_recv(struct smbd_connection *info,
> > struct msghdr *msg)
> >
> > case READ | ITER_BVEC:
> > page = msg->msg_iter.bvec->bv_page;
> > + page_offset = msg->msg_iter.bvec->bv_offset;
> > to_read = msg->msg_iter.bvec->bv_len;
> > - rc = smbd_recv_page(info, page, to_read);
> > + rc = smbd_recv_page(info, page, page_offset, to_read);
> > break;
> >
> > default:
> > /* It's a bug in upper layer to get there */
> > cifs_dbg(VFS, "CIFS: invalid msg type %d\n",
> > msg->msg_iter.type);
> > - rc = -EIO;
> > + rc = -EINVAL;
> > }
> >
> > info->smbd_recv_pending--;
> >
next prev parent reply other threads:[~2018-06-25 21:29 UTC|newest]
Thread overview: 50+ messages / expand[flat|nested] mbox.gz Atom feed top
2018-05-30 19:47 [Patch v2 00/15] CIFS: Add direct I/O support Long Li
2018-05-30 19:47 ` [Patch v2 01/15] CIFS: Introduce offset for the 1st page in data transfer structures Long Li
2018-05-30 19:47 ` [Patch v2 02/15] CIFS: Add support for direct pages in rdata Long Li
2018-05-30 20:27 ` Ruhl, Michael J
2018-05-30 20:57 ` Long Li
2018-06-24 1:50 ` Tom Talpey
2018-06-25 20:25 ` Long Li
2018-06-25 21:01 ` Jason Gunthorpe
2018-06-26 15:13 ` Tom Talpey
2018-06-27 3:21 ` Long Li
2018-05-30 19:47 ` [Patch v2 03/15] CIFS: Use offset when reading pages Long Li
2018-06-24 1:58 ` Tom Talpey
2018-06-25 20:27 ` Long Li
2018-05-30 19:47 ` [Patch v2 04/15] CIFS: Add support for direct pages in wdata Long Li
2018-06-24 2:01 ` Tom Talpey
2018-06-25 20:34 ` Long Li
2018-05-30 19:47 ` [Patch v2 05/15] CIFS: Calculate the correct request length based on page offset and tail size Long Li
2018-06-24 2:07 ` Tom Talpey
2018-06-25 21:07 ` Long Li
2018-05-30 19:47 ` [Patch v2 06/15] CIFS: Introduce helper function to get page offset and length in smb_rqst Long Li
2018-06-24 2:09 ` Tom Talpey
2018-06-25 21:14 ` Long Li
2018-06-26 13:16 ` Tom Talpey
2018-06-27 3:24 ` Long Li
2018-05-30 19:47 ` [Patch v2 07/15] CIFS: When sending data on socket, pass the correct page offset Long Li
2018-05-30 19:48 ` [Patch v2 08/15] CIFS: SMBD: Support page offset in RDMA send Long Li
2018-06-24 2:11 ` Tom Talpey
2018-06-25 21:23 ` Long Li
2018-05-30 19:48 ` [Patch v2 09/15] CIFS: SMBD: Support page offset in RDMA recv Long Li
2018-06-24 2:16 ` Tom Talpey
2018-06-25 21:29 ` Long Li [this message]
2018-05-30 19:48 ` [Patch v2 10/15] CIFS: SMBD: Support page offset in memory registration Long Li
2018-06-24 2:24 ` Tom Talpey
2018-05-30 19:48 ` [Patch v2 11/15] CIFS: Pass page offset for calculating signature Long Li
2018-06-24 2:27 ` Tom Talpey
2018-06-26 4:15 ` Long Li
2018-05-30 19:48 ` [Patch v2 12/15] CIFS: Pass page offset for encrypting Long Li
2018-06-24 2:28 ` Tom Talpey
2018-05-30 19:48 ` [Patch v2 13/15] CIFS: Add support for direct I/O read Long Li
2018-06-02 5:51 ` kbuild test robot
2018-06-02 7:15 ` kbuild test robot
2018-06-24 2:39 ` Tom Talpey
2018-06-26 4:34 ` Long Li
2018-05-30 19:48 ` [Patch v2 14/15] CIFS: Add support for direct I/O write Long Li
2018-06-24 2:48 ` Tom Talpey
2018-06-26 4:39 ` Long Li
2018-06-26 13:29 ` Tom Talpey
2018-06-27 3:44 ` Long Li
2018-05-30 19:48 ` [Patch v2 15/15] CIFS: Add direct I/O functions to file_operations Long Li
2018-06-07 11:17 ` Pavel Shilovsky
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=CY4PR2101MB07221AFF7B75D21C00C60035CE4A0@CY4PR2101MB0722.namprd21.prod.outlook.com \
--to=longli@microsoft.com \
--cc=linux-cifs@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-rdma@vger.kernel.org \
--cc=samba-technical@lists.samba.org \
--cc=sfrench@samba.org \
--cc=tom@talpey.com \
/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).