From: Jeff Layton <jlayton@poochiereds.net>
To: bfields@fieldses.org, trond.myklebust@primarydata.com
Cc: linux-nfs@vger.kernel.org, Eric Paris <eparis@parisplace.org>,
Alexander Viro <viro@zeniv.linux.org.uk>,
linux-fsdevel@vger.kernel.org
Subject: [PATCH v1 04/38] fs: rename "delayed_fput" infrastructure to "fput_global"
Date: Tue, 17 Nov 2015 06:52:26 -0500 [thread overview]
Message-ID: <1447761180-4250-5-git-send-email-jeff.layton@primarydata.com> (raw)
In-Reply-To: <1447761180-4250-1-git-send-email-jeff.layton@primarydata.com>
delayed_fput work is only delayed until the workqueue job runs. It's
even possible that it will run immediately after you call it if the
workqueue job happens to be executing at the time. What it does
guarantee is that the final __fput will run in a different task than the
the one that queued it. It's this property that we want to use in the
NFSD code to close files in advance of a setlease attempt by userland.
Change the name of the "delayed_fput" infrastructure to "fput_global"
as that better describes what this code does.
Cc: Al Viro <viro@ZenIV.linux.org.uk>
Signed-off-by: Jeff Layton <jeff.layton@primarydata.com>
---
fs/file_table.c | 38 ++++++++++++++++++++------------------
include/linux/file.h | 2 +-
init/main.c | 2 +-
3 files changed, 22 insertions(+), 20 deletions(-)
diff --git a/fs/file_table.c b/fs/file_table.c
index 8cfeaee6323f..0bf8ddc680ab 100644
--- a/fs/file_table.c
+++ b/fs/file_table.c
@@ -227,10 +227,10 @@ static void __fput(struct file *file)
mntput(mnt);
}
-static LLIST_HEAD(delayed_fput_list);
-static void delayed_fput(struct work_struct *unused)
+static LLIST_HEAD(global_fput_list);
+static void global_fput(struct work_struct *unused)
{
- struct llist_node *node = llist_del_all(&delayed_fput_list);
+ struct llist_node *node = llist_del_all(&global_fput_list);
struct llist_node *next;
for (; node; node = next) {
@@ -244,21 +244,23 @@ static void ____fput(struct callback_head *work)
__fput(container_of(work, struct file, f_u.fu_rcuhead));
}
-static DECLARE_DELAYED_WORK(delayed_fput_work, delayed_fput);
+static DECLARE_DELAYED_WORK(global_fput_work, global_fput);
-/*
- * If kernel thread really needs to have the final fput() it has done
- * to complete, call this. The only user right now is the boot - we
- * *do* need to make sure our writes to binaries on initramfs has
- * not left us with opened struct file waiting for __fput() - execve()
- * won't work without that. Please, don't add more callers without
- * very good reasons; in particular, never call that with locks
- * held and never call that from a thread that might need to do
- * some work on any kind of umount.
+/**
+ * fput_global_flush - ensure that all global_fput work is complete
+ *
+ * If a kernel thread really needs to have the final fput() it has done to
+ * complete, call this. One of the main users is the boot - we *do* need to
+ * make sure our writes to binaries on initramfs has not left us with opened
+ * struct file waiting for __fput() - execve() won't work without that.
+ *
+ * Please, don't add more callers without very good reasons; in particular,
+ * never call that with locks held and never from a thread that might need to
+ * do some work on any kind of umount.
*/
-void flush_delayed_fput(void)
+void fput_global_flush(void)
{
- flush_delayed_work(&delayed_fput_work);
+ flush_delayed_work(&global_fput_work);
}
/**
@@ -296,15 +298,15 @@ void fput(struct file *file)
*/
}
- if (llist_add(&file->f_u.fu_llist, &delayed_fput_list))
- schedule_delayed_work(&delayed_fput_work, 1);
+ if (llist_add(&file->f_u.fu_llist, &global_fput_list))
+ schedule_delayed_work(&global_fput_work, 1);
}
}
EXPORT_SYMBOL(fput);
/*
* synchronous analog of fput(); for kernel threads that might be needed
- * in some umount() (and thus can't use flush_delayed_fput() without
+ * in some umount() (and thus can't use fput_global_flush() without
* risking deadlocks), need to wait for completion of __fput() and know
* for this specific struct file it won't involve anything that would
* need them. Use only if you really need it - at the very least,
diff --git a/include/linux/file.h b/include/linux/file.h
index f87d30882a24..73bb7cee57e9 100644
--- a/include/linux/file.h
+++ b/include/linux/file.h
@@ -70,7 +70,7 @@ extern void put_unused_fd(unsigned int fd);
extern void fd_install(unsigned int fd, struct file *file);
-extern void flush_delayed_fput(void);
+extern void fput_global_flush(void);
extern void __fput_sync(struct file *);
#endif /* __LINUX_FILE_H */
diff --git a/init/main.c b/init/main.c
index 9e64d7097f1a..6cb24a16ddd6 100644
--- a/init/main.c
+++ b/init/main.c
@@ -941,7 +941,7 @@ static int __ref kernel_init(void *unused)
system_state = SYSTEM_RUNNING;
numa_default_policy();
- flush_delayed_fput();
+ fput_global_flush();
if (ramdisk_execute_command) {
ret = run_init_process(ramdisk_execute_command);
--
2.4.3
next prev parent reply other threads:[~2015-11-17 11:53 UTC|newest]
Thread overview: 52+ messages / expand[flat|nested] mbox.gz Atom feed top
2015-11-17 11:52 [PATCH v1 00/38] Allow NFS filesystems to be reexported via knfsd Jeff Layton
2015-11-17 11:52 ` [PATCH v1 01/38] nfsd: add new io class tracepoint Jeff Layton
2015-11-17 11:52 ` [PATCH v1 02/38] fs: have flush_delayed_fput flush the workqueue job Jeff Layton
2015-11-17 11:52 ` [PATCH v1 03/38] fs: add a kerneldoc header to fput Jeff Layton
2015-11-17 11:52 ` Jeff Layton [this message]
2015-11-17 11:52 ` [PATCH v1 05/38] fs: add fput_global Jeff Layton
2015-11-17 11:52 ` [PATCH v1 06/38] fsnotify: fix a sparse warning Jeff Layton
2015-11-17 11:52 ` [PATCH v1 07/38] fsnotify: export several symbols Jeff Layton
2015-11-17 11:52 ` [PATCH v1 08/38] fsnotify: destroy marks with call_srcu instead of dedicated thread Jeff Layton
2015-11-17 11:52 ` [PATCH v1 09/38] fsnotify: add a srcu barrier for fsnotify Jeff Layton
2015-11-17 11:52 ` [PATCH v1 10/38] locks: create a new notifier chain for lease attempts Jeff Layton
2015-11-17 11:52 ` [PATCH v1 11/38] sunrpc: add a new cache_detail operation for when a cache is flushed Jeff Layton
2015-11-17 11:52 ` [PATCH v1 12/38] nfsd: add a new struct file caching facility to nfsd Jeff Layton
2015-11-17 11:52 ` [PATCH v1 13/38] nfsd: keep some rudimentary stats on nfsd_file cache Jeff Layton
2015-11-17 11:52 ` [PATCH v1 14/38] nfsd: allow filecache open to skip fh_verify check Jeff Layton
2015-11-17 11:52 ` [PATCH v1 15/38] nfsd: hook up nfsd_write to the new nfsd_file cache Jeff Layton
2015-11-17 11:52 ` [PATCH v1 16/38] nfsd: hook up nfsd_read to the " Jeff Layton
2015-11-17 11:52 ` [PATCH v1 17/38] nfsd: hook nfsd_commit up " Jeff Layton
2015-11-17 11:52 ` [PATCH v1 18/38] nfsd: convert nfs4_file->fi_fds array to use nfsd_files Jeff Layton
2015-11-17 11:52 ` [PATCH v1 19/38] nfsd: have nfsd_test_lock use the nfsd_file cache Jeff Layton
2015-11-17 11:52 ` [PATCH v1 20/38] nfsd: convert fi_deleg_file and ls_file fields to nfsd_file Jeff Layton
2015-11-17 11:52 ` [PATCH v1 21/38] nfsd: hook up nfs4_preprocess_stateid_op to the nfsd_file cache Jeff Layton
2015-11-17 11:52 ` [PATCH v1 22/38] nfsd: rip out the raparms cache Jeff Layton
2015-11-17 11:52 ` [PATCH v1 23/38] nfsd: add a new EXPORT_OP_NOWCC flag to struct export_operations Jeff Layton
2015-11-17 11:52 ` [PATCH v1 24/38] nfsd: allow lockd to be forcibly disabled Jeff Layton
2015-11-17 11:52 ` [PATCH v1 25/38] nfsd: add errno mapping for EREMOTEIO Jeff Layton
2015-11-17 11:52 ` [PATCH v1 26/38] nfsd: return EREMOTE if we find an S_AUTOMOUNT inode Jeff Layton
2015-11-17 11:52 ` [PATCH v1 27/38] nfsd: allow filesystems to opt out of subtree checking Jeff Layton
2015-11-17 22:53 ` Jeff Layton
2015-11-17 11:52 ` [PATCH v1 28/38] nfsd: close cached files prior to a REMOVE or RENAME that would replace target Jeff Layton
2015-11-17 11:52 ` [PATCH v1 29/38] nfsd: retry once in nfsd_open on an -EOPENSTALE return Jeff Layton
2015-11-17 11:52 ` [PATCH v1 30/38] nfsd: close cached file when underlying file systems says no such file Jeff Layton
2015-11-17 11:52 ` [PATCH v1 31/38] nfs: replace d_add with d_splice_alias in atomic_open Jeff Layton
2015-11-19 20:06 ` J. Bruce Fields
2015-11-19 20:52 ` Trond Myklebust
2015-11-19 20:59 ` Jeff Layton
2015-11-19 22:32 ` J. Bruce Fields
2015-11-17 11:52 ` [PATCH v1 32/38] nfs: add encode_fh export op Jeff Layton
2015-11-17 11:52 ` [PATCH v1 33/38] nfs: add fh_to_dentry " Jeff Layton
2015-11-17 11:52 ` [PATCH v1 34/38] nfs: nfs_fh_to_dentry() make use of inode cache Jeff Layton
2015-11-17 11:52 ` [PATCH v1 35/38] nfs4: add NFSv4 LOOKUPP handlers Jeff Layton
2015-11-17 11:52 ` [PATCH v1 36/38] nfs: add a get_parent export operation for NFS Jeff Layton
2015-11-17 11:52 ` [PATCH v1 37/38] nfs: set export ops Jeff Layton
2015-11-17 11:53 ` [PATCH v1 38/38] nfs: add a Kconfig option for NFS reexporting and documentation Jeff Layton
2015-11-18 20:22 ` J. Bruce Fields
2015-11-18 21:15 ` Jeff Layton
2015-11-18 22:30 ` Frank Filz
2015-11-19 14:01 ` Jeff Layton
2015-11-20 0:04 ` J. Bruce Fields
2015-11-20 0:28 ` Jeff Layton
2016-01-14 22:21 ` J. Bruce Fields
2016-01-15 16:00 ` Jeff Layton
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=1447761180-4250-5-git-send-email-jeff.layton@primarydata.com \
--to=jlayton@poochiereds.net \
--cc=bfields@fieldses.org \
--cc=eparis@parisplace.org \
--cc=linux-fsdevel@vger.kernel.org \
--cc=linux-nfs@vger.kernel.org \
--cc=trond.myklebust@primarydata.com \
--cc=viro@zeniv.linux.org.uk \
/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).