From: Erez Zadok <ezk@cs.sunysb.edu>
To: akpm@linux-foundation.org
Cc: linux-kernel@vger.kernel.org, linux-fsdevel@vger.kernel.org,
viro@ftp.linux.org.uk, hch@infradead.org,
Erez Zadok <ezk@cs.sunysb.edu>
Subject: [PATCH 08/12] Unionfs: copyup only if file is being written to
Date: Fri, 25 Apr 2008 18:19:04 -0400 [thread overview]
Message-ID: <12091619521881-git-send-email-ezk@cs.sunysb.edu> (raw)
In-Reply-To: <12091619483888-git-send-email-ezk@cs.sunysb.edu>
Before, we used to copyup on an open(2) call which used flags implying
writing (O_RDWR, O_WRONLY, O_APPEND). This meant that a file being opened
for writing, then immediately closed (without actually writing to the file),
will still have been copied up. Now, we don't copyup such files in ->open,
but defer the copyup till later. [Bug #591].
Signed-off-by: Erez Zadok <ezk@cs.sunysb.edu>
---
fs/unionfs/commonfops.c | 4 ++--
fs/unionfs/file.c | 1 +
fs/unionfs/union.h | 1 +
3 files changed, 4 insertions(+), 2 deletions(-)
diff --git a/fs/unionfs/commonfops.c b/fs/unionfs/commonfops.c
index 83034c2..b7d3e38 100644
--- a/fs/unionfs/commonfops.c
+++ b/fs/unionfs/commonfops.c
@@ -635,7 +635,7 @@ int unionfs_file_release(struct inode *inode, struct file *file)
* This is important for open-but-unlinked files, as well as mmap
* support.
*/
- err = unionfs_file_revalidate(file, true);
+ err = unionfs_file_revalidate(file, UNIONFS_F(file)->wrote_to_file);
if (unlikely(err))
goto out;
unionfs_check_file(file);
@@ -819,7 +819,7 @@ int unionfs_flush(struct file *file, fl_owner_t id)
unionfs_read_lock(dentry->d_sb, UNIONFS_SMUTEX_PARENT);
unionfs_lock_dentry(dentry, UNIONFS_DMUTEX_CHILD);
- err = unionfs_file_revalidate(file, true);
+ err = unionfs_file_revalidate(file, UNIONFS_F(file)->wrote_to_file);
if (unlikely(err))
goto out;
unionfs_check_file(file);
diff --git a/fs/unionfs/file.c b/fs/unionfs/file.c
index f14b38b..09f594d 100644
--- a/fs/unionfs/file.c
+++ b/fs/unionfs/file.c
@@ -67,6 +67,7 @@ static ssize_t unionfs_write(struct file *file, const char __user *buf,
lower_file->f_path.dentry->d_inode);
fsstack_copy_attr_times(dentry->d_inode,
lower_file->f_path.dentry->d_inode);
+ UNIONFS_F(file)->wrote_to_file = true; /* for delayed copyup */
unionfs_check_file(file);
}
diff --git a/fs/unionfs/union.h b/fs/unionfs/union.h
index 4ca4e76..ab877ee 100644
--- a/fs/unionfs/union.h
+++ b/fs/unionfs/union.h
@@ -91,6 +91,7 @@ struct unionfs_file_info {
struct file **lower_files;
int *saved_branch_ids; /* IDs of branches when file was opened */
struct vm_operations_struct *lower_vm_ops;
+ bool wrote_to_file; /* for delayed copyup */
};
/* unionfs inode data in memory */
--
1.5.2.2
next prev parent reply other threads:[~2008-04-25 22:20 UTC|newest]
Thread overview: 13+ messages / expand[flat|nested] mbox.gz Atom feed top
2008-04-25 22:18 [GIT PULL -mm] 00/12 Unionfs updates/fixes/cleanups Erez Zadok
2008-04-25 22:18 ` [PATCH 01/12] Unionfs: minor code cleanups Erez Zadok
2008-04-25 22:18 ` [PATCH 02/12] Unionfs: prevent races in unionfs_fault Erez Zadok
2008-04-25 22:18 ` [PATCH 03/12] Unionfs: copy lower times in fsync/fasync only when needed Erez Zadok
2008-04-25 22:19 ` [PATCH 04/12] Unionfs: lock inode around calls to notify_change() Erez Zadok
2008-04-25 22:19 ` [PATCH 05/12] Unionfs: stop as soon as first writeable branch is found Erez Zadok
2008-04-25 22:19 ` [PATCH 06/12] Unionfs: don't dereference dentry without lower branches in d_release Erez Zadok
2008-04-25 22:19 ` [PATCH 07/12] Unionfs: set append offset correctly for copied-up files Erez Zadok
2008-04-25 22:19 ` Erez Zadok [this message]
2008-04-25 22:19 ` [PATCH 09/12] Unionfs: reorganize file_revalidate for un/locking callers Erez Zadok
2008-04-25 22:19 ` [PATCH 10/12] Unionfs: maintain one-open-file invariant for non-directories Erez Zadok
2008-04-25 22:19 ` [PATCH 11/12] Unionfs: set lower file to NULL in file_release Erez Zadok
2008-04-25 22:19 ` [PATCH 12/12] Unionfs: lock parent dentry branch config in write Erez Zadok
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=12091619521881-git-send-email-ezk@cs.sunysb.edu \
--to=ezk@cs.sunysb.edu \
--cc=akpm@linux-foundation.org \
--cc=hch@infradead.org \
--cc=linux-fsdevel@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=viro@ftp.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).