Linux Container Development
 help / color / mirror / Atom feed
From: ebiederm-aS9lmoZGLiVWk0Htik3J/w@public.gmane.org (Eric W. Biederman)
To: Al Viro <viro-3bDd1+5oDREiFSDQTTA3OLVCufUGDwFn@public.gmane.org>
Cc: Aditya Kali <adityakali-hpIqsD4AKlfQT0dZR+AlfA@public.gmane.org>,
	Neil Brown <neilb-l3A5Bk7waGM@public.gmane.org>,
	Containers
	<containers-cunTk1MwBs9QetFLy7KEm3xJsTq8ys+cHZ5vskTnxNA@public.gmane.org>,
	Oleg Nesterov <oleg-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org>,
	Andy Lutomirski <luto-kltTT9wpgjJwATOyAt5JVQ@public.gmane.org>,
	Eric Dumazet <edumazet-hpIqsD4AKlfQT0dZR+AlfA@public.gmane.org>,
	linux-fsdevel-u79uwXL29TY76Z2rM5mHXA@public.gmane.org,
	Linus Torvalds
	<torvalds-de/tnXTf+JLsfHDXvbKv3WD2FQJk+8+b@public.gmane.org>
Subject: [PATCH 3/4] vfs: Move the call of d_op->d_dname from d_path to prepend_path
Date: Thu, 06 Feb 2014 18:24:27 -0800	[thread overview]
Message-ID: <8738jvhd7o.fsf_-_@xmission.com> (raw)
In-Reply-To: <87iosrhdc0.fsf_-_-aS9lmoZGLiVWk0Htik3J/w@public.gmane.org> (Eric W. Biederman's message of "Thu, 06 Feb 2014 18:21:51 -0800")


For clarity move the call of d_dname from d_path into prepend_path.
There are only 4 callers of prepend_path and this change does not
affect the results returned by any of them.  d_absolute_path
and __d_path return an error and ignore the output of prepend_path
error == 2. getcwd only operates on directories and all of the
implementations of d_dname are on files.

For d_path the argument of no changes in output is a little trickier.
path_with_deleted does not trigger because pseudo dentries are match
IS_ROOT so is_unlinked is false.  All of current implementations of
d_dname are files so they are not currently anyone's d_parent, which
means today that either the first iteration of prepend_path will call
d_dname or none of the iterations of prepend_path will call d_dname.

d_dname is called from the while loop in prepend_path so that if it
happens in the future that any of the implementors of d_dname are a
directory instead of a file then prepend_path will work properly on them.

Signed-off-by: "Eric W. Biederman" <ebiederm-aS9lmoZGLiVWk0Htik3J/w@public.gmane.org>
---
 fs/dcache.c |   32 +++++++++++++++++++++-----------
 1 files changed, 21 insertions(+), 11 deletions(-)

diff --git a/fs/dcache.c b/fs/dcache.c
index c250d97befe4..c5c7847ff84b 100644
--- a/fs/dcache.c
+++ b/fs/dcache.c
@@ -2892,6 +2892,27 @@ restart:
 	while (dentry != root->dentry || vfsmnt != root->mnt) {
 		struct dentry * parent;
 
+		/*
+		 * We have various synthetic files allocated with
+		 * d_alloc_pseudo that are not available through
+		 * ordinary path lookup and don't need a name until
+		 * a user wants to identify the object in
+		 * /proc/pid/fd/ or similiar.
+		 */
+		if (IS_ROOT(dentry) && dentry != vfsmnt->mnt_root &&
+		    dentry->d_op && dentry->d_op->d_dname) {
+			char *buf  = bptr - blen;
+			char *name = dentry->d_op->d_dname(dentry, buf, blen);
+			if (IS_ERR(name)) {
+				error = PTR_ERR(name);
+				break;
+			}
+			blen = name - buf;
+			bptr = name;
+			error = 2; /* unmounted by definition */
+			break;
+		}
+
 		if (dentry == vfsmnt->mnt_root || IS_ROOT(dentry)) {
 			struct mount *parent = ACCESS_ONCE(mnt->mnt_parent);
 			/* Global root? */
@@ -3055,17 +3076,6 @@ char *d_path(const struct path *path, char *buf, int buflen)
 	struct path root;
 	int error;
 
-	/*
-	 * We have various synthetic files allocated with
-	 * d_alloc_pseudo that are not available through
-	 * ordinary path lookup and don't need a name until
-	 * a user wants to identify the object in
-	 * /proc/pid/fd/ or similiar.
-	 */
-	if (path->dentry->d_op && path->dentry->d_op->d_dname &&
-	    IS_ROOT(path->dentry) && path->dentry != path->mnt->mnt_root)
-		return path->dentry->d_op->d_dname(path->dentry, buf, buflen);
-
 	rcu_read_lock();
 	get_fs_root_rcu(current->fs, &root);
 	error = path_with_deleted(path, &root, &res, &buflen);
-- 
1.7.5.4

  parent reply	other threads:[~2014-02-07  2:24 UTC|newest]

Thread overview: 51+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2013-11-15 16:41 Regression wrt mounting /proc in user namespace in 3.13 Daniel P. Berrange
     [not found] ` <20131115164123.GN28794-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org>
2013-11-16 16:48   ` Serge E. Hallyn
     [not found]     ` <20131116164840.GA4441-7LNsyQBKDXoIagZqoN9o3w@public.gmane.org>
2013-11-17  3:06       ` Serge E. Hallyn
     [not found]         ` <20131117030653.GA7670-7LNsyQBKDXoIagZqoN9o3w@public.gmane.org>
2013-11-18  3:19           ` Serge E. Hallyn
     [not found]             ` <20131118031932.GA17621-7LNsyQBKDXoIagZqoN9o3w@public.gmane.org>
2013-11-18  4:52               ` Gao feng
     [not found]                 ` <52899D09.5080202-BthXqXjhjHXQFUHtdCDX3A@public.gmane.org>
2013-11-18 14:08                   ` Serge E. Hallyn
     [not found]                     ` <20131118140830.GA22075-7LNsyQBKDXoIagZqoN9o3w@public.gmane.org>
2013-11-18 18:01                       ` Serge E. Hallyn
     [not found]                         ` <20131118180134.GA24156-7LNsyQBKDXoIagZqoN9o3w@public.gmane.org>
2013-11-19  1:51                           ` Eric W. Biederman
     [not found]                             ` <87k3g5gnuv.fsf-aS9lmoZGLiVWk0Htik3J/w@public.gmane.org>
2013-11-19  3:47                               ` Serge E. Hallyn
2013-11-26 18:10                               ` Serge E. Hallyn
     [not found]                                 ` <20131126181043.GA25492-7LNsyQBKDXoIagZqoN9o3w@public.gmane.org>
2013-11-27  0:14                                   ` [REVIEW][PATCH 0/3] userns fixes for v3.13-rc1 Eric W. Biederman
     [not found]                                     ` <87siui1z1g.fsf_-_-aS9lmoZGLiVWk0Htik3J/w@public.gmane.org>
2013-11-27  0:16                                       ` [REVIEW][PATCH 1/3] vfs: In d_path don't call d_dname on a mount point Eric W. Biederman
     [not found]                                         ` <8738mi1yya.fsf_-_-aS9lmoZGLiVWk0Htik3J/w@public.gmane.org>
2013-11-27  1:58                                           ` Serge E. Hallyn
2013-11-30  6:15                                           ` Al Viro
     [not found]                                             ` <20131130061525.GY10323-3bDd1+5oDREiFSDQTTA3OLVCufUGDwFn@public.gmane.org>
2013-11-30 17:02                                               ` Al Viro
     [not found]                                                 ` <20131130170226.GZ10323-3bDd1+5oDREiFSDQTTA3OLVCufUGDwFn@public.gmane.org>
2013-11-30 21:51                                                   ` Eric W. Biederman
     [not found]                                                     ` <87a9glh838.fsf-aS9lmoZGLiVWk0Htik3J/w@public.gmane.org>
2013-11-30 22:43                                                       ` Al Viro
     [not found]                                                         ` <20131130224340.GA10323-3bDd1+5oDREiFSDQTTA3OLVCufUGDwFn@public.gmane.org>
2013-12-02  7:29                                                           ` Al Viro
2014-01-17  3:29                                                         ` Eric W. Biederman
     [not found]                                                         ` <874n53gub7.fsf@xmission.com>
     [not found]                                                           ` <874n53gub7.fsf-aS9lmoZGLiVWk0Htik3J/w@public.gmane.org>
2014-01-17  8:39                                                             ` Al Viro
     [not found]                                                               ` <20140117083901.GA10323-3bDd1+5oDREiFSDQTTA3OLVCufUGDwFn@public.gmane.org>
2014-02-07  2:21                                                                 ` [PATCH 0/4] d_dname cleanups Eric W. Biederman
     [not found]                                                                   ` <87iosrhdc0.fsf_-_-aS9lmoZGLiVWk0Htik3J/w@public.gmane.org>
2014-02-07  2:23                                                                     ` [PATCH 1/4] perfmon: Use d_alloc_pseudo like all of the d_dname callers Eric W. Biederman
2014-02-07  2:23                                                                     ` [PATCH 2/4] vfs: Simply when d_alloc_dname is called Eric W. Biederman
2014-02-07  2:24                                                                     ` Eric W. Biederman [this message]
2014-02-07  2:24                                                                     ` [PATCH 4/4] vfs: Call d_dname from dentry_path Eric W. Biederman
2013-12-01  5:09                                                   ` [REVIEW][PATCH 1/3] vfs: In d_path don't call d_dname on a mount point Al Viro
     [not found]                                                     ` <20131201050930.GB10323-3bDd1+5oDREiFSDQTTA3OLVCufUGDwFn@public.gmane.org>
2013-12-01  6:15                                                       ` [REVIEW][PATCH 1/3] vfs: In d_path don't call d_dname on a mountpoint Tetsuo Handa
2013-12-02  5:43                                                   ` [REVIEW][PATCH 1/3] vfs: In d_path don't call d_dname on a mount point NeilBrown
     [not found]                                                     ` <20131202164359.4f4f2c94-wvvUuzkyo1EYVZTmpyfIwg@public.gmane.org>
2013-12-02 16:23                                                       ` J.Bruce Fields
2013-11-27  0:16                                       ` [REVIEW][PATCH 2/3] fork: Allow CLONE_PARENT after setns(CLONE_NEWPID) Eric W. Biederman
     [not found]                                         ` <87vbzezojq.fsf_-_-aS9lmoZGLiVWk0Htik3J/w@public.gmane.org>
2013-11-27  1:58                                           ` Serge E. Hallyn
2013-11-27  0:17                                       ` [REVIEW][PATCH 3/3] vfs: Fix a regression in mounting proc Eric W. Biederman
     [not found]                                         ` <87pppmzoin.fsf_-_-aS9lmoZGLiVWk0Htik3J/w@public.gmane.org>
2013-11-27  0:21                                           ` Andy Lutomirski
     [not found]                                             ` <CALCETrVp78EfzY3Oa-LV1Hm8A4Y35apehcxrxdyrzvTb5sp=pA-JsoAwUIsXosN+BqQ9rBEUg@public.gmane.org>
2013-11-27  0:36                                               ` Eric W. Biederman
2013-11-27  2:00                                           ` Serge E. Hallyn
2013-11-27  3:19                                           ` Gao feng
     [not found]                                             ` <529564AA.8050100-BthXqXjhjHXQFUHtdCDX3A@public.gmane.org>
2013-11-27  5:00                                               ` Eric W. Biederman
2013-11-27 16:13                                           ` Oleg Nesterov
     [not found]                                         ` <20131127161300.GA24773@redhat.com>
     [not found]                                           ` <20131127161300.GA24773-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org>
2013-11-27 16:29                                             ` Serge E. Hallyn
     [not found]                                               ` <20131127162928.GB7358-7LNsyQBKDXoIagZqoN9o3w@public.gmane.org>
2013-11-27 18:09                                                 ` Oleg Nesterov
2013-11-27 16:41                                             ` Andy Lutomirski
     [not found]                                               ` <CALCETrXFnw63=JoEaQxM+Opj+kCXSL=9XppymzGKhLzOnp3WaQ-JsoAwUIsXosN+BqQ9rBEUg@public.gmane.org>
2013-11-27 18:10                                                 ` Oleg Nesterov
2013-11-27 18:51                                             ` Eric W. Biederman
     [not found]                                               ` <871u21oeyr.fsf-aS9lmoZGLiVWk0Htik3J/w@public.gmane.org>
2013-11-27 19:47                                                 ` Oleg Nesterov
     [not found]                                               ` <20131127194722.GA32673@redhat.com>
     [not found]                                                 ` <20131127194722.GA32673-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org>
2013-11-27 19:52                                                   ` Eric W. Biederman
     [not found]                                                 ` <87iovdmxl7.fsf@xmission.com>
     [not found]                                                   ` <87iovdmxl7.fsf-aS9lmoZGLiVWk0Htik3J/w@public.gmane.org>
2013-11-27 20:01                                                     ` Oleg Nesterov
2013-11-27 20:07                                                     ` Eric W. Biederman
     [not found]                                                       ` <CALCETrUwjK7iLMMJaCvKUbBwEqV58oXY4dWzTGJohYgg4DwjWQ@mail.gmail.com>
     [not found]                                                         ` <CALCETrUwjK7iLMMJaCvKUbBwEqV58oXY4dWzTGJohYgg4DwjWQ-JsoAwUIsXosN+BqQ9rBEUg@public.gmane.org>
2013-11-29 14:56                                                           ` Serge E. Hallyn
     [not found]                                                       ` <87wqjtlic3.fsf-aS9lmoZGLiVWk0Htik3J/w@public.gmane.org>
2013-11-27 20:41                                                         ` Andy Lutomirski
2013-11-29 19:53                                                         ` Oleg Nesterov
     [not found]                                                           ` <20131129195327.GA12974-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org>
2013-12-13 22:07                                                             ` Richard Weinberger

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=8738jvhd7o.fsf_-_@xmission.com \
    --to=ebiederm-as9lmozglivwk0htik3j/w@public.gmane.org \
    --cc=adityakali-hpIqsD4AKlfQT0dZR+AlfA@public.gmane.org \
    --cc=containers-cunTk1MwBs9QetFLy7KEm3xJsTq8ys+cHZ5vskTnxNA@public.gmane.org \
    --cc=edumazet-hpIqsD4AKlfQT0dZR+AlfA@public.gmane.org \
    --cc=linux-fsdevel-u79uwXL29TY76Z2rM5mHXA@public.gmane.org \
    --cc=luto-kltTT9wpgjJwATOyAt5JVQ@public.gmane.org \
    --cc=neilb-l3A5Bk7waGM@public.gmane.org \
    --cc=oleg-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org \
    --cc=torvalds-de/tnXTf+JLsfHDXvbKv3WD2FQJk+8+b@public.gmane.org \
    --cc=viro-3bDd1+5oDREiFSDQTTA3OLVCufUGDwFn@public.gmane.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