linux-fsdevel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Miklos Szeredi <mszeredi@redhat.com>
To: linux-unionfs@vger.kernel.org
Cc: linux-fsdevel@vger.kernel.org, linux-kernel@vger.kernel.org
Subject: [PATCH 05/39] vfs: optionally don't account file in nr_files
Date: Tue, 29 May 2018 16:43:05 +0200	[thread overview]
Message-ID: <20180529144339.16538-6-mszeredi@redhat.com> (raw)
In-Reply-To: <20180529144339.16538-1-mszeredi@redhat.com>

Stacking file operations in overlay will store an extra open file for each
overlay file opened.

The overhead is just that of "struct file" which is about 256bytes, because
overlay already pins an extra dentry and inode when the file is open, which
add up to a much larger overhead.

For fear of breaking working setups, don't start accounting the extra file.

The implementation adds a bool argument to path_open() to control whether
the returned file is to be accounted or not.  If the file is not accounted,
f_mode will contain FMODE_NOACCOUNT, so that when freeing the file the
count is not decremented.

Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
---
 fs/file_table.c    | 13 +++++++++----
 fs/internal.h      |  7 ++++++-
 fs/open.c          | 10 +++++-----
 include/linux/fs.h |  5 ++++-
 4 files changed, 24 insertions(+), 11 deletions(-)

diff --git a/fs/file_table.c b/fs/file_table.c
index 7ec0b3e5f05d..60376bfa04cf 100644
--- a/fs/file_table.c
+++ b/fs/file_table.c
@@ -51,7 +51,8 @@ static void file_free_rcu(struct rcu_head *head)
 
 static inline void file_free(struct file *f)
 {
-	percpu_counter_dec(&nr_files);
+	if (!(f->f_mode & FMODE_NOACCOUNT))
+		percpu_counter_dec(&nr_files);
 	call_rcu(&f->f_u.fu_rcuhead, file_free_rcu);
 }
 
@@ -100,7 +101,7 @@ int proc_nr_files(struct ctl_table *table, int write,
  * done, you will imbalance int the mount's writer count
  * and a warning at __fput() time.
  */
-struct file *get_empty_filp(void)
+struct file *__get_empty_filp(bool account)
 {
 	const struct cred *cred = current_cred();
 	static long old_max;
@@ -110,7 +111,8 @@ struct file *get_empty_filp(void)
 	/*
 	 * Privileged users can go above max_files
 	 */
-	if (get_nr_files() >= files_stat.max_files && !capable(CAP_SYS_ADMIN)) {
+	if (account &&
+	    get_nr_files() >= files_stat.max_files && !capable(CAP_SYS_ADMIN)) {
 		/*
 		 * percpu_counters are inaccurate.  Do an expensive check before
 		 * we go and fail.
@@ -123,7 +125,10 @@ struct file *get_empty_filp(void)
 	if (unlikely(!f))
 		return ERR_PTR(-ENOMEM);
 
-	percpu_counter_inc(&nr_files);
+	if (account)
+		percpu_counter_inc(&nr_files);
+	else
+		f->f_mode = FMODE_NOACCOUNT;
 	f->f_cred = get_cred(cred);
 	error = security_file_alloc(f);
 	if (unlikely(error)) {
diff --git a/fs/internal.h b/fs/internal.h
index e08972db0303..b82725ba3054 100644
--- a/fs/internal.h
+++ b/fs/internal.h
@@ -93,7 +93,12 @@ extern void chroot_fs_refs(const struct path *, const struct path *);
 /*
  * file_table.c
  */
-extern struct file *get_empty_filp(void);
+extern struct file *__get_empty_filp(bool account);
+
+static inline struct file *get_empty_filp(void)
+{
+	return __get_empty_filp(true);
+}
 
 /*
  * super.c
diff --git a/fs/open.c b/fs/open.c
index d0bf7f061a1a..6e52fd6fea7c 100644
--- a/fs/open.c
+++ b/fs/open.c
@@ -732,8 +732,8 @@ static int do_dentry_open(struct file *f,
 	static const struct file_operations empty_fops = {};
 	int error;
 
-	f->f_mode = OPEN_FMODE(f->f_flags) | FMODE_LSEEK |
-				FMODE_PREAD | FMODE_PWRITE;
+	f->f_mode = (f->f_mode & FMODE_NOACCOUNT) | OPEN_FMODE(f->f_flags) |
+		FMODE_LSEEK | FMODE_PREAD | FMODE_PWRITE;
 
 	path_get(&f->f_path);
 	f->f_inode = inode;
@@ -743,7 +743,7 @@ static int do_dentry_open(struct file *f,
 	f->f_wb_err = filemap_sample_wb_err(f->f_mapping);
 
 	if (unlikely(f->f_flags & O_PATH)) {
-		f->f_mode = FMODE_PATH;
+		f->f_mode = (f->f_mode & FMODE_NOACCOUNT) | FMODE_PATH;
 		f->f_op = &empty_fops;
 		goto done;
 	}
@@ -917,12 +917,12 @@ int vfs_open(const struct path *path, struct file *file,
  * Return: A pointer to a struct file or an IS_ERR pointer.  Cannot return NULL.
  */
 struct file *path_open(const struct path *path, int flags, struct inode *inode,
-		       const struct cred *cred)
+		       const struct cred *cred, bool account)
 {
 	struct file *file;
 	int retval;
 
-	file = get_empty_filp();
+	file = __get_empty_filp(account);
 	if (IS_ERR(file))
 		return file;
 
diff --git a/include/linux/fs.h b/include/linux/fs.h
index 9473e68280d0..ecc854c75611 100644
--- a/include/linux/fs.h
+++ b/include/linux/fs.h
@@ -153,6 +153,9 @@ typedef int (dio_iodone_t)(struct kiocb *iocb, loff_t offset,
 /* File is capable of returning -EAGAIN if I/O will block */
 #define FMODE_NOWAIT	((__force fmode_t)0x8000000)
 
+/* File does not contribute to nr_files count */
+#define FMODE_NOACCOUNT	((__force fmode_t)0x10000000)
+
 /*
  * Flag for rw_copy_check_uvector and compat_rw_copy_check_uvector
  * that indicates that they should check the contents of the iovec are
@@ -2402,7 +2405,7 @@ extern struct file *file_open_root(struct dentry *, struct vfsmount *,
 				   const char *, int, umode_t);
 extern struct file * dentry_open(const struct path *, int, const struct cred *);
 extern struct file *path_open(const struct path *, int, struct inode *,
-			      const struct cred *);
+			      const struct cred *, bool);
 extern int filp_close(struct file *, fl_owner_t id);
 
 extern struct filename *getname_flags(const char __user *, int, int *);
-- 
2.14.3

  parent reply	other threads:[~2018-05-29 14:43 UTC|newest]

Thread overview: 83+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2018-05-29 14:43 [PATCH 00/39] overlayfs: stack file operations Miklos Szeredi
2018-05-29 14:43 ` [PATCH 01/39] vfs: dedpue: return loff_t Miklos Szeredi
2018-06-04  8:43   ` Christoph Hellwig
2018-06-05  8:33     ` Miklos Szeredi
2018-06-06 15:09       ` Darrick J. Wong
2018-06-18 20:08         ` Miklos Szeredi
2018-05-29 14:43 ` [PATCH 02/39] vfs: dedupe: rationalize args Miklos Szeredi
2018-06-06 15:02   ` Darrick J. Wong
2018-05-29 14:43 ` [PATCH 03/39] vfs: dedupe: extract helper for a single dedup Miklos Szeredi
2018-05-29 15:41   ` Amir Goldstein
2018-05-29 16:04     ` Amir Goldstein
2018-06-04  8:44   ` Christoph Hellwig
2018-05-29 14:43 ` [PATCH 04/39] vfs: add path_open() Miklos Szeredi
2018-06-04  8:46   ` Christoph Hellwig
2018-06-10  4:36     ` Al Viro
2018-05-29 14:43 ` Miklos Szeredi [this message]
2018-06-04  8:47   ` [PATCH 05/39] vfs: optionally don't account file in nr_files Christoph Hellwig
2018-06-04  8:57     ` Miklos Szeredi
2018-06-10  4:41   ` Al Viro
2018-05-29 14:43 ` [PATCH 06/39] vfs: add f_op->pre_mmap() Miklos Szeredi
2018-06-04  8:48   ` Christoph Hellwig
2018-06-05 11:36     ` Miklos Szeredi
2018-05-29 14:43 ` [PATCH 07/39] vfs: export vfs_ioctl() to modules Miklos Szeredi
2018-06-04  8:49   ` Christoph Hellwig
2018-06-10  4:57     ` Al Viro
2018-06-11  7:19       ` Miklos Szeredi
2018-06-11 16:24         ` Christoph Hellwig
2018-06-19 14:04           ` Miklos Szeredi
2018-06-19 14:24             ` Christoph Hellwig
2018-06-19 14:34               ` Miklos Szeredi
2018-06-19 14:54                 ` Al Viro
2018-05-29 14:43 ` [PATCH 08/39] vfs: export vfs_dedupe_file_range_one() " Miklos Szeredi
2018-05-29 14:43 ` [PATCH 09/39] ovl: copy up times Miklos Szeredi
2018-05-29 14:43 ` [PATCH 10/39] ovl: copy up inode flags Miklos Szeredi
2018-05-29 14:43 ` [PATCH 11/39] Revert "Revert "ovl: get_write_access() in truncate"" Miklos Szeredi
2018-05-29 14:43 ` [PATCH 12/39] ovl: copy up file size as well Miklos Szeredi
2018-05-29 14:43 ` [PATCH 13/39] ovl: deal with overlay files in ovl_d_real() Miklos Szeredi
2018-05-29 14:43 ` [PATCH 14/39] ovl: stack file ops Miklos Szeredi
2018-06-10  4:13   ` Al Viro
2018-06-11  7:09     ` Miklos Szeredi
2018-06-12  2:29       ` Al Viro
2018-06-12  2:40         ` Al Viro
2018-06-12  9:24           ` Miklos Szeredi
2018-06-12 18:24             ` Al Viro
2018-06-12 18:31               ` Al Viro
2018-06-13  9:21                 ` Miklos Szeredi
2018-06-15  5:47                   ` Al Viro
2018-06-18 11:50                     ` Miklos Szeredi
2018-06-13 11:56               ` J. R. Okajima
2018-05-29 14:43 ` [PATCH 15/39] ovl: add helper to return real file Miklos Szeredi
2018-06-10  5:42   ` Al Viro
2018-06-11  8:11     ` Miklos Szeredi
2018-05-29 14:43 ` [PATCH 16/39] ovl: add ovl_read_iter() Miklos Szeredi
2018-05-29 14:43 ` [PATCH 17/39] ovl: add ovl_write_iter() Miklos Szeredi
2018-05-29 14:43 ` [PATCH 18/39] ovl: add ovl_fsync() Miklos Szeredi
2018-05-29 14:43 ` [PATCH 19/39] ovl: add ovl_mmap() Miklos Szeredi
2018-06-10  5:24   ` Al Viro
2018-06-11  7:58     ` Miklos Szeredi
2018-05-29 14:43 ` [PATCH 20/39] ovl: add ovl_fallocate() Miklos Szeredi
2018-05-29 14:43 ` [PATCH 21/39] ovl: add lsattr/chattr support Miklos Szeredi
2018-05-29 14:43 ` [PATCH 22/39] ovl: add ovl_fiemap() Miklos Szeredi
2018-05-29 14:43 ` [PATCH 23/39] ovl: add O_DIRECT support Miklos Szeredi
2018-06-10  5:31   ` Al Viro
2018-06-11  8:08     ` Miklos Szeredi
2018-05-29 14:43 ` [PATCH 24/39] ovl: add reflink/copyfile/dedup support Miklos Szeredi
2018-05-29 14:43 ` [PATCH 25/39] vfs: don't open real Miklos Szeredi
2018-05-29 14:43 ` [PATCH 26/39] ovl: copy-up on MAP_SHARED Miklos Szeredi
2018-05-29 14:43 ` [PATCH 27/39] ovl: obsolete "check_copy_up" module option Miklos Szeredi
2018-05-29 15:13   ` Amir Goldstein
2018-05-30  8:26     ` Miklos Szeredi
2018-05-29 14:43 ` [PATCH 28/39] ovl: fix documentation of non-standard behavior Miklos Szeredi
2018-05-29 14:43 ` [PATCH 29/39] vfs: simplify dentry_open() Miklos Szeredi
2018-05-29 14:43 ` [PATCH 30/39] Revert "ovl: fix may_write_real() for overlayfs directories" Miklos Szeredi
2018-05-29 14:43 ` [PATCH 31/39] Revert "ovl: don't allow writing ioctl on lower layer" Miklos Szeredi
2018-05-29 14:43 ` [PATCH 32/39] vfs: fix freeze protection in mnt_want_write_file() for overlayfs Miklos Szeredi
2018-06-04  8:50   ` Christoph Hellwig
2018-05-29 14:43 ` [PATCH 33/39] Revert "ovl: fix relatime for directories" Miklos Szeredi
2018-05-29 14:43 ` [PATCH 34/39] Revert "vfs: update ovl inode before relatime check" Miklos Szeredi
2018-05-29 14:43 ` [PATCH 35/39] Revert "vfs: add flags to d_real()" Miklos Szeredi
2018-05-29 14:43 ` [PATCH 36/39] Revert "vfs: do get_write_access() on upper layer of overlayfs" Miklos Szeredi
2018-05-29 14:43 ` [PATCH 37/39] Partially revert "locks: fix file locking on overlayfs" Miklos Szeredi
2018-05-29 14:43 ` [PATCH 38/39] Revert "fsnotify: support overlayfs" Miklos Szeredi
2018-05-29 14:43 ` [PATCH 39/39] vfs: remove open_flags from d_real() Miklos Szeredi

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=20180529144339.16538-6-mszeredi@redhat.com \
    --to=mszeredi@redhat.com \
    --cc=linux-fsdevel@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-unionfs@vger.kernel.org \
    --subject='Re: [PATCH 05/39] vfs: optionally don'\''t account file in nr_files' \
    /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

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).