From: "Yan, Zheng" <zheng.z.yan@intel.com>
To: ceph-devel@vger.kernel.org
Cc: sage@inktank.com, "Yan, Zheng" <zheng.z.yan@intel.com>
Subject: [PATCH 2/3] ceph: set caps count after composing cap reconnect message
Date: Sun, 22 Sep 2013 12:46:51 +0800 [thread overview]
Message-ID: <1379825212-10438-2-git-send-email-zheng.z.yan@intel.com> (raw)
In-Reply-To: <1379825212-10438-1-git-send-email-zheng.z.yan@intel.com>
From: "Yan, Zheng" <zheng.z.yan@intel.com>
It's possible that some caps get released while composing the cap
reconnect message.
Signed-off-by: Yan, Zheng <zheng.z.yan@intel.com>
---
fs/ceph/mds_client.c | 25 +++++++++++++++++++------
1 file changed, 19 insertions(+), 6 deletions(-)
diff --git a/fs/ceph/mds_client.c b/fs/ceph/mds_client.c
index 8f8f5c0..6b4ed35 100644
--- a/fs/ceph/mds_client.c
+++ b/fs/ceph/mds_client.c
@@ -43,6 +43,7 @@
*/
struct ceph_reconnect_state {
+ int nr_caps;
struct ceph_pagelist *pagelist;
bool flock;
};
@@ -2549,6 +2550,8 @@ encode_again:
} else {
err = ceph_pagelist_append(pagelist, &rec, reclen);
}
+
+ recon_state->nr_caps++;
out_free:
kfree(path);
out_dput:
@@ -2576,6 +2579,7 @@ static void send_mds_reconnect(struct ceph_mds_client *mdsc,
struct rb_node *p;
int mds = session->s_mds;
int err = -ENOMEM;
+ int s_nr_caps;
struct ceph_pagelist *pagelist;
struct ceph_reconnect_state recon_state;
@@ -2611,10 +2615,12 @@ static void send_mds_reconnect(struct ceph_mds_client *mdsc,
discard_cap_releases(mdsc, session);
/* traverse this session's caps */
- err = ceph_pagelist_encode_32(pagelist, session->s_nr_caps);
+ s_nr_caps = session->s_nr_caps;
+ err = ceph_pagelist_encode_32(pagelist, s_nr_caps);
if (err)
goto fail;
+ recon_state.nr_caps = 0;
recon_state.pagelist = pagelist;
recon_state.flock = session->s_con.peer_features & CEPH_FEATURE_FLOCK;
err = iterate_session_caps(session, encode_caps_cb, &recon_state);
@@ -2643,11 +2649,18 @@ static void send_mds_reconnect(struct ceph_mds_client *mdsc,
if (recon_state.flock)
reply->hdr.version = cpu_to_le16(2);
- if (pagelist->length) {
- /* set up outbound data if we have any */
- reply->hdr.data_len = cpu_to_le32(pagelist->length);
- ceph_msg_data_add_pagelist(reply, pagelist);
- }
+
+ /* raced with cap release? */
+ if (s_nr_caps != recon_state.nr_caps) {
+ struct page *page = list_first_entry(&pagelist->head,
+ struct page, lru);
+ __le32 *addr = kmap_atomic(page);
+ *addr = cpu_to_le32(recon_state.nr_caps);
+ kunmap_atomic(addr);
+ }
+
+ reply->hdr.data_len = cpu_to_le32(pagelist->length);
+ ceph_msg_data_add_pagelist(reply, pagelist);
ceph_con_send(&session->s_con, reply);
mutex_unlock(&session->s_mutex);
--
1.8.1.4
next prev parent reply other threads:[~2013-09-22 4:47 UTC|newest]
Thread overview: 7+ messages / expand[flat|nested] mbox.gz Atom feed top
2013-09-22 4:46 [PATCH 1/3] ceph: queue cap release in __ceph_remove_cap() Yan, Zheng
2013-09-22 4:46 ` Yan, Zheng [this message]
2013-09-22 4:46 ` [PATCH 3/3] ceph: handle race between cap reconnect and cap release Yan, Zheng
2013-09-23 14:31 ` Sage Weil
2013-09-23 16:08 ` Yan, Zheng
2013-09-24 1:05 ` Yan, Zheng
2013-09-23 3:48 ` [PATCH 1/3] ceph: queue cap release in __ceph_remove_cap() Sage Weil
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=1379825212-10438-2-git-send-email-zheng.z.yan@intel.com \
--to=zheng.z.yan@intel.com \
--cc=ceph-devel@vger.kernel.org \
--cc=sage@inktank.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 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.