Linux NFS development
 help / color / mirror / Atom feed
From: Chuck Lever III <chuck.lever@oracle.com>
To: Jeff Layton <jlayton@kernel.org>
Cc: Linux NFS Mailing List <linux-nfs@vger.kernel.org>,
	Benjamin Coddington <bcodding@redhat.com>,
	Zhi Li <yieli@redhat.com>
Subject: Re: [PATCH] sunrpc: only free unix grouplist after RCU settles
Date: Thu, 30 Mar 2023 19:33:27 +0000	[thread overview]
Message-ID: <D3F3D553-C252-47FB-9D41-9C9A254557DB@oracle.com> (raw)
In-Reply-To: <12c783873abcd0f9c631e2f367cc1dd97949e0fb.camel@kernel.org>



> On Mar 30, 2023, at 2:49 PM, Jeff Layton <jlayton@kernel.org> wrote:
> 
> On Thu, 2023-03-30 at 18:31 +0000, Chuck Lever III wrote:
>> Hi Jeff-
>> 
>>> On Mar 30, 2023, at 2:24 PM, Jeff Layton <jlayton@kernel.org> wrote:
>>> 
>>> While the unix_gid object is rcu-freed, the group_info list that it
>>> contains is not. Ensure that we only put the group list reference once
>>> we are really freeing the unix_gid object.
>>> 
>>> Reported-by: Zhi Li <yieli@redhat.com>
>> 
>> Should we also add
>> 
>> Fixes: fd5d2f78261b ("SUNRPC: Make server side AUTH_UNIX use lockless lookups") ?
>> 
>> 
> 
> Sure. That does look like when that particular bug crept in.
> 
>>> Link: https://bugzilla.redhat.com/show_bug.cgi?id=2183056
>> 
>> This bug isn't publicly accessible, fwiw.
>> 
> 
> Thanks. It should be now.

OK, applied to nfsd-fixes!


>>> Signed-off-by: Jeff Layton <jlayton@kernel.org>
>>> ---
>>> net/sunrpc/svcauth_unix.c | 17 +++++++++++++----
>>> 1 file changed, 13 insertions(+), 4 deletions(-)
>>> 
>>> diff --git a/net/sunrpc/svcauth_unix.c b/net/sunrpc/svcauth_unix.c
>>> index 50e2eb579194..4485088ce27b 100644
>>> --- a/net/sunrpc/svcauth_unix.c
>>> +++ b/net/sunrpc/svcauth_unix.c
>>> @@ -416,14 +416,23 @@ static int unix_gid_hash(kuid_t uid)
>>> 	return hash_long(from_kuid(&init_user_ns, uid), GID_HASHBITS);
>>> }
>>> 
>>> -static void unix_gid_put(struct kref *kref)
>>> +static void unix_gid_free(struct rcu_head *rcu)
>>> {
>>> -	struct cache_head *item = container_of(kref, struct cache_head, ref);
>>> -	struct unix_gid *ug = container_of(item, struct unix_gid, h);
>>> +	struct unix_gid *ug = container_of(rcu, struct unix_gid, rcu);
>>> +	struct cache_head *item = &ug->h;
>>> +
>>> 	if (test_bit(CACHE_VALID, &item->flags) &&
>>> 	    !test_bit(CACHE_NEGATIVE, &item->flags))
>>> 		put_group_info(ug->gi);
>>> -	kfree_rcu(ug, rcu);
>>> +	kfree(ug);
>>> +}
>>> +
>>> +static void unix_gid_put(struct kref *kref)
>>> +{
>>> +	struct cache_head *item = container_of(kref, struct cache_head, ref);
>>> +	struct unix_gid *ug = container_of(item, struct unix_gid, h);
>>> +
>>> +	call_rcu(&ug->rcu, unix_gid_free);
>>> }
>>> 
>>> static int unix_gid_match(struct cache_head *corig, struct cache_head *cnew)
>>> -- 
>>> 2.39.2
>>> 
>> 
>> --
>> Chuck Lever
>> 
>> 
> 
> -- 
> Jeff Layton <jlayton@kernel.org>

--
Chuck Lever



      reply	other threads:[~2023-03-30 19:33 UTC|newest]

Thread overview: 4+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2023-03-30 18:24 [PATCH] sunrpc: only free unix grouplist after RCU settles Jeff Layton
2023-03-30 18:31 ` Chuck Lever III
2023-03-30 18:49   ` Jeff Layton
2023-03-30 19:33     ` Chuck Lever III [this message]

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=D3F3D553-C252-47FB-9D41-9C9A254557DB@oracle.com \
    --to=chuck.lever@oracle.com \
    --cc=bcodding@redhat.com \
    --cc=jlayton@kernel.org \
    --cc=linux-nfs@vger.kernel.org \
    --cc=yieli@redhat.com \
    /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