From: Oleg Nesterov <oleg@redhat.com>
To: Denys Vlasenko <dvlasenk@redhat.com>
Cc: Al Viro <viro@zeniv.linux.org.uk>,
Jan Kratochvil <jan.kratochvil@redhat.com>,
linux-kernel@vger.kernel.org
Subject: Re: [PATCH] dcache: error out if the name buffer is too short
Date: Sun, 26 Jan 2014 17:35:36 +0100 [thread overview]
Message-ID: <20140126163536.GA21320@redhat.com> (raw)
In-Reply-To: <20140126155141.GA12249@redhat.com>
On 01/26, Oleg Nesterov wrote:
>
> Al, Denys, unless I am totally confused the "restart" logic is very broken.
> We can't simply restart because the main loop changes dentry?
Plus prepend_name() can't actually return -ENAMETOOLONG,
"int error" inside the loop is wrong.
I believe the minimal fix is something like below. I'll try to test
this change (currently I can't) later and resend. And I still think
it can be cleanuped a bit, but this is minor.
Oleg.
--- a/fs/dcache.c
+++ b/fs/dcache.c
@@ -3103,7 +3103,7 @@ char *simple_dname(struct dentry *dentry, char *buffer, int buflen)
/* these dentries are never renamed, so d_lock is not needed */
if (prepend(&end, &buflen, " (deleted)", 11) ||
prepend(&end, &buflen, dentry->d_name.name, dentry->d_name.len) ||
- prepend(&end, &buflen, "/", 1))
+ prepend(&end, &buflen, "/", 1))
end = ERR_PTR(-ENAMETOOLONG);
return end;
}
@@ -3113,32 +3113,34 @@ char *simple_dname(struct dentry *dentry, char *buffer, int buflen)
*/
static char *__dentry_path(struct dentry *dentry, char *buf, int buflen)
{
+ struct dentry *de;
char *end, *retval;
int len, seq = 0;
int error = 0;
+ if (buflen < 2)
+ goto Elong;
+
rcu_read_lock();
restart:
end = buf + buflen;
len = buflen;
prepend(&end, &len, "\0", 1);
- if (buflen < 1)
- goto Elong;
/* Get '/' right */
retval = end-1;
*retval = '/';
+ de = dentry;
read_seqbegin_or_lock(&rename_lock, &seq);
- while (!IS_ROOT(dentry)) {
- struct dentry *parent = dentry->d_parent;
- int error;
+ while (!IS_ROOT(de)) {
+ struct dentry *parent = de->d_parent;
prefetch(parent);
- error = prepend_name(&end, &len, &dentry->d_name);
+ error = prepend_name(&end, &len, &de->d_name);
if (error)
break;
retval = end;
- dentry = parent;
+ de = parent;
}
if (!(seq & 1))
rcu_read_unlock();
next prev parent reply other threads:[~2014-01-26 16:35 UTC|newest]
Thread overview: 7+ messages / expand[flat|nested] mbox.gz Atom feed top
2014-01-24 14:04 [PATCH] dcache: error out if the name buffer is too short Denys Vlasenko
2014-01-24 16:19 ` Oleg Nesterov
2014-01-26 15:37 ` Oleg Nesterov
2014-01-26 15:51 ` Oleg Nesterov
2014-01-26 16:35 ` Oleg Nesterov [this message]
2014-01-26 17:41 ` Al Viro
2014-01-27 17:40 ` Oleg Nesterov
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=20140126163536.GA21320@redhat.com \
--to=oleg@redhat.com \
--cc=dvlasenk@redhat.com \
--cc=jan.kratochvil@redhat.com \
--cc=linux-kernel@vger.kernel.org \
--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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.