linux-fsdevel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Al Viro <viro@ZenIV.linux.org.uk>
To: linux-fsdevel@vger.kernel.org
Cc: Alexey Dobriyan <adobriyan@gmail.com>
Subject: [PATCH 2/5] proc_lookupfd_common(): don't bother with instantiate unless the file is open
Date: Sat, 26 May 2018 01:04:07 +0100	[thread overview]
Message-ID: <20180526000410.10804-2-viro@ZenIV.linux.org.uk> (raw)
In-Reply-To: <20180526000410.10804-1-viro@ZenIV.linux.org.uk>

From: Al Viro <viro@zeniv.linux.org.uk>

... and take the "check if file is open, pick ->f_mode" into a helper;
tid_fd_revalidate() can use it.

The next patch will get rid of tid_fd_revalidate() calls in instantiate
callbacks.

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
---
 fs/proc/fd.c | 63 ++++++++++++++++++++++++++++++++----------------------------
 1 file changed, 34 insertions(+), 29 deletions(-)

diff --git a/fs/proc/fd.c b/fs/proc/fd.c
index 6b80cd1e419a..d38845ecc408 100644
--- a/fs/proc/fd.c
+++ b/fs/proc/fd.c
@@ -81,12 +81,29 @@ static const struct file_operations proc_fdinfo_file_operations = {
 	.release	= single_release,
 };
 
+static bool tid_fd_mode(struct task_struct *task, unsigned fd, fmode_t *mode)
+{
+	struct files_struct *files = get_files_struct(task);
+	struct file *file;
+
+	if (!files)
+		return false;
+
+	rcu_read_lock();
+	file = fcheck_files(files, fd);
+	if (file)
+		*mode = file->f_mode;
+	rcu_read_unlock();
+	put_files_struct(files);
+	return !!file;
+}
+
 static int tid_fd_revalidate(struct dentry *dentry, unsigned int flags)
 {
-	struct files_struct *files;
 	struct task_struct *task;
 	struct inode *inode;
 	unsigned int fd;
+	fmode_t f_mode;
 
 	if (flags & LOOKUP_RCU)
 		return -ECHILD;
@@ -96,35 +113,20 @@ static int tid_fd_revalidate(struct dentry *dentry, unsigned int flags)
 	fd = proc_fd(inode);
 
 	if (task) {
-		files = get_files_struct(task);
-		if (files) {
-			struct file *file;
-
-			rcu_read_lock();
-			file = fcheck_files(files, fd);
-			if (file) {
-				unsigned f_mode = file->f_mode;
-
-				rcu_read_unlock();
-				put_files_struct(files);
-
-				task_dump_owner(task, 0, &inode->i_uid, &inode->i_gid);
-
-				if (S_ISLNK(inode->i_mode)) {
-					unsigned i_mode = S_IFLNK;
-					if (f_mode & FMODE_READ)
-						i_mode |= S_IRUSR | S_IXUSR;
-					if (f_mode & FMODE_WRITE)
-						i_mode |= S_IWUSR | S_IXUSR;
-					inode->i_mode = i_mode;
-				}
-
-				security_task_to_inode(task, inode);
-				put_task_struct(task);
-				return 1;
+		if (tid_fd_mode(task, fd, &f_mode)) {
+			task_dump_owner(task, 0, &inode->i_uid, &inode->i_gid);
+
+			if (S_ISLNK(inode->i_mode)) {
+				unsigned i_mode = S_IFLNK;
+				if (f_mode & FMODE_READ)
+					i_mode |= S_IRUSR | S_IXUSR;
+				if (f_mode & FMODE_WRITE)
+					i_mode |= S_IWUSR | S_IXUSR;
+				inode->i_mode = i_mode;
 			}
-			rcu_read_unlock();
-			put_files_struct(files);
+			security_task_to_inode(task, inode);
+			put_task_struct(task);
+			return 1;
 		}
 		put_task_struct(task);
 	}
@@ -203,11 +205,14 @@ static struct dentry *proc_lookupfd_common(struct inode *dir,
 	struct task_struct *task = get_proc_task(dir);
 	int result = -ENOENT;
 	unsigned fd = name_to_int(&dentry->d_name);
+	fmode_t f_mode;
 
 	if (!task)
 		goto out_no_task;
 	if (fd == ~0U)
 		goto out;
+	if (!tid_fd_mode(task, fd, &f_mode))
+		goto out;
 
 	result = instantiate(dir, dentry, task, (void *)(unsigned long)fd);
 out:
-- 
2.11.0

  reply	other threads:[~2018-05-26  0:04 UTC|newest]

Thread overview: 46+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2018-05-13 21:26 [RFC][PATCHES] reducing d_add() use, part 1 Al Viro
2018-05-13 21:30 ` [PATCH 01/15] bfs_lookup(): use d_splice_alias() Al Viro
2018-05-13 21:30   ` [PATCH 02/15] bfs_find_entry: pass name/len as qstr pointer Al Viro
2018-05-13 21:30   ` [PATCH 03/15] bfs_add_entry: " Al Viro
2018-05-13 21:30   ` [PATCH 04/15] cramfs_lookup(): use d_splice_alias() Al Viro
2018-05-13 21:30   ` [PATCH 05/15] freevxfs_lookup(): " Al Viro
2018-05-14  8:41     ` Christoph Hellwig
2018-05-14 15:26       ` Al Viro
2018-05-13 21:30   ` [PATCH 06/15] minix_lookup: " Al Viro
2018-05-13 21:30   ` [PATCH 07/15] qnx4_lookup: " Al Viro
2018-05-14 10:48     ` Anders Larsen
2018-05-13 21:30   ` [PATCH 08/15] sysv_lookup: " Al Viro
2018-05-14  8:41     ` Christoph Hellwig
2018-05-13 21:30   ` [PATCH 09/15] ubifs_lookup: " Al Viro
2018-05-14  6:48     ` Richard Weinberger
2018-05-13 21:30   ` [PATCH 10/15] qnx6_lookup: switch to d_splice_alias() Al Viro
2018-05-13 21:30   ` [PATCH 11/15] romfs_lookup: hash negative lookups, use d_splice_alias() Al Viro
2018-05-13 21:30   ` [PATCH 12/15] adfs_lookup_byname: .. *is* taken care of in fs/namei.c Al Viro
2018-05-13 21:30   ` [PATCH 13/15] adfs_lookup: do not fail with ENOENT on negatives, use d_splice_alias() Al Viro
2018-05-13 21:30   ` [PATCH 14/15] xfs_vn_lookup: simplify a bit Al Viro
2018-05-14  8:41     ` Christoph Hellwig
2018-05-13 21:30   ` [PATCH 15/15] openpromfs: switch to d_splice_alias() Al Viro
2018-05-16  9:45   ` [PATCH 01/15] bfs_lookup(): use d_splice_alias() Tigran Aivazian
2018-05-25 23:53 ` [RFC][PATCHES] reducing d_add() use, part 2 Al Viro
2018-05-25 23:54   ` [PATCH 01/10] openpromfs: switch to d_splice_alias() Al Viro
2018-05-25 23:54     ` [PATCH 02/10] orangefs_lookup: simplify Al Viro
2018-05-31 18:23       ` Mike Marshall
2018-05-25 23:54     ` [PATCH 03/10] omfs_lookup(): report IO errors, use d_splice_alias() Al Viro
2018-05-25 23:54     ` [PATCH 04/10] hfs: " Al Viro
2018-05-25 23:54     ` [PATCH 05/10] hfs: don't allow mounting over .../rsrc Al Viro
2018-05-25 23:54     ` [PATCH 06/10] hfsplus: switch to d_splice_alias() Al Viro
2018-05-25 23:54     ` [PATCH 07/10] ncp_lookup(): use d_splice_alias() Al Viro
2018-05-25 23:54     ` [PATCH 08/10] 9p: unify paths in v9fs_vfs_lookup() Al Viro
2018-05-25 23:54     ` [PATCH 09/10] cifs_lookup(): cifs_get_inode_...() never returns 0 with *inode left NULL Al Viro
2018-05-25 23:54     ` [PATCH 10/10] cifs_lookup(): switch to d_splice_alias() Al Viro
2018-05-26  0:03   ` [RFC][PATCHES] reducing d_add() use, part 3 (procfs) Al Viro
2018-05-26  0:04     ` [PATCH 1/5] procfs: get rid of ancient BS in pid_revalidate() uses Al Viro
2018-05-26  0:04       ` Al Viro [this message]
2018-05-26  0:04       ` [PATCH 3/5] don't bother with tid_fd_revalidate() in lookups Al Viro
2018-05-26  0:04       ` [PATCH 4/5] procfs: switch instantiate_t to d_splice_alias() Al Viro
2018-05-26  0:04       ` [PATCH 5/5] switch the rest of procfs lookups " Al Viro
2018-05-31  8:28       ` [PATCH 1/5] procfs: get rid of ancient BS in pid_revalidate() uses Alexey Dobriyan
2018-05-26 13:07     ` [RFC][PATCHES] reducing d_add() use, part 3 (procfs) Alexey Dobriyan
2018-05-26 13:56       ` Alexey Dobriyan
2018-05-26 18:20         ` Al Viro
2018-05-26 18:38           ` Matthew Wilcox

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=20180526000410.10804-2-viro@ZenIV.linux.org.uk \
    --to=viro@zeniv.linux.org.uk \
    --cc=adobriyan@gmail.com \
    --cc=linux-fsdevel@vger.kernel.org \
    /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).