public inbox for linux-kernel@vger.kernel.org
 help / color / mirror / Atom feed
From: Jeffle Xu <jefflexu@linux.alibaba.com>
To: dhowells@redhat.com, linux-cachefs@redhat.com, xiang@kernel.org,
	chao@kernel.org, linux-erofs@lists.ozlabs.org
Cc: torvalds@linux-foundation.org, gregkh@linuxfoundation.org,
	willy@infradead.org, linux-fsdevel@vger.kernel.org,
	joseph.qi@linux.alibaba.com, bo.liu@linux.alibaba.com,
	tao.peng@linux.alibaba.com, gerry@linux.alibaba.com,
	eguan@linux.alibaba.com, linux-kernel@vger.kernel.org,
	luodaowen.backend@bytedance.com, tianzichen@kuaishou.com,
	yinxin.x@bytedance.com, zhangjiachen.jaycee@bytedance.com,
	zhujia.zj@bytedance.com
Subject: [PATCH v11 03/22] cachefiles: unbind cachefiles gracefully in on-demand mode
Date: Mon,  9 May 2022 15:40:09 +0800	[thread overview]
Message-ID: <20220509074028.74954-4-jefflexu@linux.alibaba.com> (raw)
In-Reply-To: <20220509074028.74954-1-jefflexu@linux.alibaba.com>

Add a refcount to avoid the deadlock in on-demand read mode. The
on-demand read mode will pin the corresponding cachefiles object for
each anonymous fd. The cachefiles object is unpinned when the anonymous
fd gets closed. When the user daemon exits and the fd of
"/dev/cachefiles" device node gets closed, it will wait for all
cahcefiles objects getting withdrawn. Then if there's any anonymous fd
getting closed after the fd of the device node, the user daemon will
hang forever, waiting for all objects getting withdrawn.

To fix this, add a refcount indicating if there's any object pinned by
anonymous fds. The cachefiles cache gets unbound and withdrawn when the
refcount is decreased to 0. It won't change the behaviour of the
original mode, in which case the cachefiles cache gets unbound and
withdrawn as long as the fd of the device node gets closed.

Signed-off-by: Jeffle Xu <jefflexu@linux.alibaba.com>
---
 fs/cachefiles/daemon.c   | 19 ++++++++++++++++---
 fs/cachefiles/internal.h |  3 +++
 fs/cachefiles/ondemand.c |  3 +++
 3 files changed, 22 insertions(+), 3 deletions(-)

diff --git a/fs/cachefiles/daemon.c b/fs/cachefiles/daemon.c
index d5417da7f792..5b1d0642c749 100644
--- a/fs/cachefiles/daemon.c
+++ b/fs/cachefiles/daemon.c
@@ -111,6 +111,7 @@ static int cachefiles_daemon_open(struct inode *inode, struct file *file)
 	INIT_LIST_HEAD(&cache->volumes);
 	INIT_LIST_HEAD(&cache->object_list);
 	spin_lock_init(&cache->object_list_lock);
+	refcount_set(&cache->unbind_pincount, 1);
 	xa_init_flags(&cache->reqs, XA_FLAGS_ALLOC);
 	xa_init_flags(&cache->ondemand_ids, XA_FLAGS_ALLOC1);
 
@@ -164,6 +165,20 @@ static void cachefiles_flush_reqs(struct cachefiles_cache *cache)
 	xa_destroy(&cache->ondemand_ids);
 }
 
+void cachefiles_put_unbind_pincount(struct cachefiles_cache *cache)
+{
+	if (refcount_dec_and_test(&cache->unbind_pincount)) {
+		cachefiles_daemon_unbind(cache);
+		cachefiles_open = 0;
+		kfree(cache);
+	}
+}
+
+void cachefiles_get_unbind_pincount(struct cachefiles_cache *cache)
+{
+	refcount_inc(&cache->unbind_pincount);
+}
+
 /*
  * Release a cache.
  */
@@ -179,14 +194,12 @@ static int cachefiles_daemon_release(struct inode *inode, struct file *file)
 
 	if (cachefiles_in_ondemand_mode(cache))
 		cachefiles_flush_reqs(cache);
-	cachefiles_daemon_unbind(cache);
 
 	/* clean up the control file interface */
 	cache->cachefilesd = NULL;
 	file->private_data = NULL;
-	cachefiles_open = 0;
 
-	kfree(cache);
+	cachefiles_put_unbind_pincount(cache);
 
 	_leave("");
 	return 0;
diff --git a/fs/cachefiles/internal.h b/fs/cachefiles/internal.h
index 4f5150a96849..e5c612888f84 100644
--- a/fs/cachefiles/internal.h
+++ b/fs/cachefiles/internal.h
@@ -109,6 +109,7 @@ struct cachefiles_cache {
 	char				*rootdirname;	/* name of cache root directory */
 	char				*secctx;	/* LSM security context */
 	char				*tag;		/* cache binding tag */
+	refcount_t			unbind_pincount;/* refcount to do daemon unbind */
 	struct xarray			reqs;		/* xarray of pending on-demand requests */
 	struct xarray			ondemand_ids;	/* xarray for ondemand_id allocation */
 	u32				ondemand_id_next;
@@ -171,6 +172,8 @@ extern int cachefiles_has_space(struct cachefiles_cache *cache,
  * daemon.c
  */
 extern const struct file_operations cachefiles_daemon_fops;
+extern void cachefiles_get_unbind_pincount(struct cachefiles_cache *cache);
+extern void cachefiles_put_unbind_pincount(struct cachefiles_cache *cache);
 
 /*
  * error_inject.c
diff --git a/fs/cachefiles/ondemand.c b/fs/cachefiles/ondemand.c
index 64fc312b16d3..7946ee6c40be 100644
--- a/fs/cachefiles/ondemand.c
+++ b/fs/cachefiles/ondemand.c
@@ -14,6 +14,7 @@ static int cachefiles_ondemand_fd_release(struct inode *inode,
 	object->ondemand_id = CACHEFILES_ONDEMAND_ID_CLOSED;
 	xa_erase(&cache->ondemand_ids, object_id);
 	cachefiles_put_object(object, cachefiles_obj_put_ondemand_fd);
+	cachefiles_put_unbind_pincount(cache);
 	return 0;
 }
 
@@ -169,6 +170,8 @@ static int cachefiles_ondemand_get_fd(struct cachefiles_req *req)
 	load->fd = fd;
 	req->msg.object_id = object_id;
 	object->ondemand_id = object_id;
+
+	cachefiles_get_unbind_pincount(cache);
 	return 0;
 
 err_put_fd:
-- 
2.27.0


  parent reply	other threads:[~2022-05-09  7:51 UTC|newest]

Thread overview: 30+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2022-05-09  7:40 [PATCH v11 00/22] fscache,erofs: fscache-based on-demand read semantics Jeffle Xu
2022-05-09  7:40 ` [PATCH v11 01/22] cachefiles: extract write routine Jeffle Xu
2022-05-09  7:40 ` [PATCH v11 02/22] cachefiles: notify the user daemon when looking up cookie Jeffle Xu
2022-05-10 12:50   ` David Howells
2022-05-09  7:40 ` Jeffle Xu [this message]
2022-05-10 12:53   ` [PATCH v11 03/22] cachefiles: unbind cachefiles gracefully in on-demand mode David Howells
2022-05-09  7:40 ` [PATCH v11 04/22] cachefiles: notify the user daemon when withdrawing cookie Jeffle Xu
2022-05-09  7:40 ` [PATCH v11 05/22] cachefiles: implement on-demand read Jeffle Xu
2022-05-09  7:40 ` [PATCH v11 06/22] cachefiles: enable on-demand read mode Jeffle Xu
2022-05-10 12:56   ` David Howells
2022-05-10 13:29     ` Gao Xiang
2022-05-09  7:40 ` [PATCH v11 07/22] cachefiles: add tracepoints for " Jeffle Xu
2022-05-09  7:40 ` [PATCH v11 08/22] cachefiles: document " Jeffle Xu
2022-05-10 13:01   ` David Howells
2022-05-09  7:40 ` [PATCH v11 09/22] erofs: make erofs_map_blocks() generally available Jeffle Xu
2022-05-09  7:40 ` [PATCH v11 10/22] erofs: add fscache mode check helper Jeffle Xu
2022-05-09  7:40 ` [PATCH v11 11/22] erofs: register fscache volume Jeffle Xu
2022-05-09  7:40 ` [PATCH v11 12/22] erofs: add fscache context helper functions Jeffle Xu
2022-05-09  7:40 ` [PATCH v11 13/22] erofs: add anonymous inode caching metadata for data blobs Jeffle Xu
2022-05-09  7:40 ` [PATCH v11 14/22] erofs: add erofs_fscache_read_folios() helper Jeffle Xu
2022-05-09  7:40 ` [PATCH v11 15/22] erofs: register fscache context for primary data blob Jeffle Xu
2022-05-09  7:40 ` [PATCH v11 16/22] erofs: register fscache context for extra data blobs Jeffle Xu
2022-05-09  7:40 ` [PATCH v11 17/22] erofs: implement fscache-based metadata read Jeffle Xu
2022-05-09  7:40 ` [PATCH v11 18/22] erofs: implement fscache-based data read for non-inline layout Jeffle Xu
2022-05-09  7:40 ` [PATCH v11 19/22] erofs: implement fscache-based data read for inline layout Jeffle Xu
2022-05-09  7:40 ` [PATCH v11 20/22] erofs: implement fscache-based data readahead Jeffle Xu
2022-05-09  7:40 ` [PATCH v11 21/22] erofs: add 'fsid' mount option Jeffle Xu
2022-05-09  7:40 ` [PATCH v11 22/22] erofs: change to use asynchronous io for fscache readpage/readahead Jeffle Xu
2022-05-10  6:48 ` [PATCH v11 00/22] fscache, erofs: fscache-based on-demand read semantics 严松
2022-05-10 14:14 ` [PATCH v11 00/22] fscache,erofs: " Chao Yu

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=20220509074028.74954-4-jefflexu@linux.alibaba.com \
    --to=jefflexu@linux.alibaba.com \
    --cc=bo.liu@linux.alibaba.com \
    --cc=chao@kernel.org \
    --cc=dhowells@redhat.com \
    --cc=eguan@linux.alibaba.com \
    --cc=gerry@linux.alibaba.com \
    --cc=gregkh@linuxfoundation.org \
    --cc=joseph.qi@linux.alibaba.com \
    --cc=linux-cachefs@redhat.com \
    --cc=linux-erofs@lists.ozlabs.org \
    --cc=linux-fsdevel@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=luodaowen.backend@bytedance.com \
    --cc=tao.peng@linux.alibaba.com \
    --cc=tianzichen@kuaishou.com \
    --cc=torvalds@linux-foundation.org \
    --cc=willy@infradead.org \
    --cc=xiang@kernel.org \
    --cc=yinxin.x@bytedance.com \
    --cc=zhangjiachen.jaycee@bytedance.com \
    --cc=zhujia.zj@bytedance.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