From: Shaun McDowell <shaunjmcdowell@gmail.com>
To: linux-block@vger.kernel.org
Cc: jbacik@fb.com, nbd-general@lists.sourceforge.net, bsdkurt@gmail.com
Subject: [PATCH] nbd: add FUA op support
Date: Thu, 25 May 2017 23:55:54 -0400 [thread overview]
Message-ID: <20170526035554.GA31566@auir> (raw)
NBD userland client and server have FUA (forced unit access) support
and flags defined. Make NBD kernel module recognize NBD_FLAG_SEND_FUA,
enable FUA on the queue, and forward FUA requests to the server.
Signed-off-by: Shaun McDowell <shaunjmcdowell@gmail.com>
---
drivers/block/nbd.c | 16 +++++++++++++---
include/uapi/linux/nbd.h | 4 ++++
2 files changed, 17 insertions(+), 3 deletions(-)
diff --git a/drivers/block/nbd.c b/drivers/block/nbd.c
index 9a7bb2c..c5e52f6 100644
--- a/drivers/block/nbd.c
+++ b/drivers/block/nbd.c
@@ -400,6 +400,7 @@ static int nbd_send_cmd(struct nbd_device *nbd, struct nbd_cmd *cmd, int index)
unsigned long size = blk_rq_bytes(req);
struct bio *bio;
u32 type;
+ u32 nbd_cmd_flags = 0;
u32 tag = blk_mq_unique_tag(req);
int sent = nsock->sent, skip = 0;
@@ -429,6 +430,9 @@ static int nbd_send_cmd(struct nbd_device *nbd, struct nbd_cmd *cmd, int index)
return -EIO;
}
+ if (req->cmd_flags & REQ_FUA)
+ nbd_cmd_flags |= NBD_CMD_FLAG_FUA;
+
/* We did a partial send previously, and we at least sent the whole
* request struct, so just go and send the rest of the pages in the
* request.
@@ -442,7 +446,7 @@ static int nbd_send_cmd(struct nbd_device *nbd, struct nbd_cmd *cmd, int index)
}
cmd->index = index;
cmd->cookie = nsock->cookie;
- request.type = htonl(type);
+ request.type = htonl(type | nbd_cmd_flags);
if (type != NBD_CMD_FLUSH) {
request.from = cpu_to_be64((u64)blk_rq_pos(req) << 9);
request.len = htonl(size);
@@ -965,8 +969,12 @@ static void nbd_parse_flags(struct nbd_device *nbd)
set_disk_ro(nbd->disk, false);
if (config->flags & NBD_FLAG_SEND_TRIM)
queue_flag_set_unlocked(QUEUE_FLAG_DISCARD, nbd->disk->queue);
- if (config->flags & NBD_FLAG_SEND_FLUSH)
- blk_queue_write_cache(nbd->disk->queue, true, false);
+ if (config->flags & NBD_FLAG_SEND_FLUSH) {
+ if (config->flags & NBD_FLAG_SEND_FUA)
+ blk_queue_write_cache(nbd->disk->queue, true, true);
+ else
+ blk_queue_write_cache(nbd->disk->queue, true, false);
+ }
else
blk_queue_write_cache(nbd->disk->queue, false, false);
}
@@ -1309,6 +1317,8 @@ static int nbd_dbg_flags_show(struct seq_file *s, void *unused)
seq_puts(s, "NBD_FLAG_READ_ONLY\n");
if (flags & NBD_FLAG_SEND_FLUSH)
seq_puts(s, "NBD_FLAG_SEND_FLUSH\n");
+ if (flags & NBD_FLAG_SEND_FUA)
+ seq_puts(s, "NBD_FLAG_SEND_FUA\n");
if (flags & NBD_FLAG_SEND_TRIM)
seq_puts(s, "NBD_FLAG_SEND_TRIM\n");
diff --git a/include/uapi/linux/nbd.h b/include/uapi/linux/nbd.h
index 155e33f..a50527e 100644
--- a/include/uapi/linux/nbd.h
+++ b/include/uapi/linux/nbd.h
@@ -41,10 +41,14 @@ enum {
#define NBD_FLAG_HAS_FLAGS (1 << 0) /* nbd-server supports flags */
#define NBD_FLAG_READ_ONLY (1 << 1) /* device is read-only */
#define NBD_FLAG_SEND_FLUSH (1 << 2) /* can flush writeback cache */
+#define NBD_FLAG_SEND_FUA (1 << 3) /* send FUA (forced unit access) */
/* there is a gap here to match userspace */
#define NBD_FLAG_SEND_TRIM (1 << 5) /* send trim/discard */
#define NBD_FLAG_CAN_MULTI_CONN (1 << 8) /* Server supports multiple connections per export. */
+/* values for cmd flags in the upper 16 bits of request type */
+#define NBD_CMD_FLAG_FUA (1 << 16) /* FUA (forced unit access) op */
+
/* These are client behavior specific flags. */
#define NBD_CFLAG_DESTROY_ON_DISCONNECT (1 << 0) /* delete the nbd device on
disconnect. */
--
2.7.4
next reply other threads:[~2017-05-26 3:56 UTC|newest]
Thread overview: 3+ messages / expand[flat|nested] mbox.gz Atom feed top
2017-05-26 3:55 Shaun McDowell [this message]
2017-05-30 13:03 ` [PATCH] nbd: add FUA op support Josef Bacik
2017-05-30 14:20 ` Jens Axboe
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=20170526035554.GA31566@auir \
--to=shaunjmcdowell@gmail.com \
--cc=bsdkurt@gmail.com \
--cc=jbacik@fb.com \
--cc=linux-block@vger.kernel.org \
--cc=nbd-general@lists.sourceforge.net \
/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