All of lore.kernel.org
 help / color / mirror / Atom feed
From: Alex Elder <elder@inktank.com>
To: ceph-devel@vger.kernel.org
Subject: [PATCH 1/6] libceph: slightly defer registering osd request
Date: Mon, 25 Mar 2013 21:27:11 -0500	[thread overview]
Message-ID: <5151077F.10504@inktank.com> (raw)
In-Reply-To: <5151071C.3000309@inktank.com>

One of the first things ceph_osdc_start_request() does is register
the request.  It then acquires the osd client's map semaphore and
request mutex and proceeds to map and send the request.

There is no reason the request has to be registered before acquiring
the map semaphore.  So hold off doing so until after the map
semaphore is held.

Since register_request() is nothing more than a wrapper around
__register_request(), call the latter function instead, after
acquiring the request mutex.

That leaves register_request() unused, so get rid of it.

This partially resolves:
    http://tracker.ceph.com/issues/4392

Signed-off-by: Alex Elder <elder@inktank.com>
---
 net/ceph/osd_client.c |   11 +----------
 1 file changed, 1 insertion(+), 10 deletions(-)

diff --git a/net/ceph/osd_client.c b/net/ceph/osd_client.c
index 7041906..f9276cb 100644
--- a/net/ceph/osd_client.c
+++ b/net/ceph/osd_client.c
@@ -831,14 +831,6 @@ static void __register_request(struct
ceph_osd_client *osdc,
 	}
 }

-static void register_request(struct ceph_osd_client *osdc,
-			     struct ceph_osd_request *req)
-{
-	mutex_lock(&osdc->request_mutex);
-	__register_request(osdc, req);
-	mutex_unlock(&osdc->request_mutex);
-}
-
 /*
  * called under osdc->request_mutex
  */
@@ -1785,8 +1777,6 @@ int ceph_osdc_start_request(struct ceph_osd_client
*osdc,
 	ceph_osdc_msg_data_set(req->r_reply, &req->r_data_in);
 	ceph_osdc_msg_data_set(req->r_request, &req->r_data_out);

-	register_request(osdc, req);
-
 	down_read(&osdc->map_sem);
 	mutex_lock(&osdc->request_mutex);
 	/*
@@ -1794,6 +1784,7 @@ int ceph_osdc_start_request(struct ceph_osd_client
*osdc,
 	 * while we dropped request_mutex above, so only send now if
 	 * the request still han't been touched yet.
 	 */
+	__register_request(osdc, req);
 	if (req->r_sent == 0) {
 		rc = __map_request(osdc, req, 0);
 		if (rc < 0) {
-- 
1.7.9.5


  reply	other threads:[~2013-03-26  2:27 UTC|newest]

Thread overview: 10+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2013-03-26  2:25 [PATCH 0/6] libceph: send osd requests in tid order Alex Elder
2013-03-26  2:27 ` Alex Elder [this message]
2013-03-26  2:27 ` [PATCH 2/6] libceph: no more kick_requests() race Alex Elder
2013-03-26  2:27 ` [PATCH 3/6] libceph: prepend requests in order when kicking Alex Elder
2013-03-26 14:50   ` [PATCH 3/6, v2] libceph: requeue only sent requests " Alex Elder
2013-03-26  2:27 ` [PATCH 4/6] libceph: keep request lists in tid order Alex Elder
2013-03-26  2:28 ` [PATCH 5/6] libceph: send queued requests when starting new one Alex Elder
2013-03-26  2:28 ` [PATCH 6/6] libceph: verify requests queued in order Alex Elder
2013-03-26 14:50   ` [PATCH 6/6, v2] " Alex Elder
2013-03-26 14:49 ` [PATCH 0/6] libceph: send osd requests in tid order Alex Elder

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=5151077F.10504@inktank.com \
    --to=elder@inktank.com \
    --cc=ceph-devel@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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.