netdev.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Marcel Holtmann <marcel@holtmann.org>
To: Stephen Rothwell <sfr@canb.auug.org.au>
Cc: Stephen Hemminger <shemminger@vyatta.com>,
	vibi_sreenivasan@cms.com, netdev <netdev@vger.kernel.org>,
	Johannes Berg <johannes@sipsolutions.net>,
	Thomas Graf <tgraf@suug.ch>,
	linux-next <linux-next@vger.kernel.org>
Subject: Re: [PATCH] Fix Warnings from net/netlink/genetlink.c
Date: Tue, 11 Aug 2009 21:03:55 -0700	[thread overview]
Message-ID: <1250049835.30166.47.camel@localhost.localdomain> (raw)
In-Reply-To: <20090812135031.a38be3f6.sfr@canb.auug.org.au>

Hi Stephen,

> > > > @@ -136,7 +136,7 @@ int genl_register_mc_group(struct genl_family *family,
> > > >  {
> > > >         int id;
> > > >         unsigned long *new_groups;
> > > > -       int err;
> > > > +       int err = 0;
> > > >  
> > > >         BUG_ON(grp->name[0] == '\0');
> > > 
> > > this looks fishy. How does gcc thinks this variable is uninitialized. If
> > > I look at the code in Linus' tree, I don't see it.
> > 
> > Agreed, and the line numbers are off.
> 
> In the -next tree, it looks like this:

it would have been nice if the patch actually indicates that it is for
-next since otherwise just shutting up a compiler warning is a bad idea
in this case.

> int genl_register_mc_group(struct genl_family *family,
> 			   struct genl_multicast_group *grp)
> {
> 	int id;
> 	unsigned long *new_groups;
> 	int err;
> 
> 	BUG_ON(grp->name[0] == '\0');
> 
> 	genl_lock();
> 
> 	/* special-case our own group */
> 	if (grp == &notify_grp)
> 		id = GENL_ID_CTRL;
> 	else
> 		id = find_first_zero_bit(mc_groups,
> 					 mc_groups_longs * BITS_PER_LONG);
> 
> 
> 	if (id >= mc_groups_longs * BITS_PER_LONG) {
> 		size_t nlen = (mc_groups_longs + 1) * sizeof(unsigned long);
> 
> 		if (mc_groups == &mc_group_start) {
> 			new_groups = kzalloc(nlen, GFP_KERNEL);
> 			if (!new_groups) {
> 				err = -ENOMEM;
> 				goto out;
> 			}
> 			mc_groups = new_groups;
> 			*mc_groups = mc_group_start;
> 		} else {
> 			new_groups = krealloc(mc_groups, nlen, GFP_KERNEL);
> 			if (!new_groups) {
> 				err = -ENOMEM;
> 				goto out;
> 			}
> 			mc_groups = new_groups;
> 			mc_groups[mc_groups_longs] = 0;
> 		}
> 		mc_groups_longs++;
> 	}
> 
> 	if (family->netnsok) {
> 		struct net *net;
> 
> 		rcu_read_lock();
> 		for_each_net_rcu(net) {
> 			err = netlink_change_ngroups(net->genl_sock,
> 					mc_groups_longs * BITS_PER_LONG);
> 			if (err) {
> 				rcu_read_unlock();
> 				goto out;
> 			}
> 		}
> 		rcu_read_unlock();
> 	} else {
> 		err = netlink_change_ngroups(init_net.genl_sock,
> 					     mc_groups_longs * BITS_PER_LONG);
> 		if (err)
> 			goto out;
> 	}
> 
> 	grp->id = id;
> 	set_bit(id, mc_groups);
> 	list_add_tail(&grp->list, &family->mcast_groups);
> 	grp->family = family;
> 
> 	genl_ctrl_event(CTRL_CMD_NEWMCAST_GRP, grp);
>  out:
> 	genl_unlock();
> 	return err;
> }
> 
> so if family->netnsok is true and the for_each_net_rcu loop executes 0
> times, err will not be set ...  Now, that may not be logically possible,
> but I can't tell that from this code.

I prefer we add a err = 0 in the if (family->netnsok) { block instead of
just globally setting it to a value.

Regards

Marcel



  reply	other threads:[~2009-08-12  4:04 UTC|newest]

Thread overview: 8+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2009-08-11 10:07 [PATCH] Fix Warnings from net/netlink/genetlink.c vibi sreenivasan
2009-08-11 23:57 ` Marcel Holtmann
2009-08-12  3:24   ` Stephen Hemminger
2009-08-12  3:50     ` Stephen Rothwell
2009-08-12  4:03       ` Marcel Holtmann [this message]
     [not found] <1249975630.2400.7.camel@HunTEr>
2009-08-11  7:36 ` Johannes Berg
2009-08-11  9:15   ` vibi sreenivasan
2009-08-11  9:11     ` Johannes Berg

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=1250049835.30166.47.camel@localhost.localdomain \
    --to=marcel@holtmann.org \
    --cc=johannes@sipsolutions.net \
    --cc=linux-next@vger.kernel.org \
    --cc=netdev@vger.kernel.org \
    --cc=sfr@canb.auug.org.au \
    --cc=shemminger@vyatta.com \
    --cc=tgraf@suug.ch \
    --cc=vibi_sreenivasan@cms.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;
as well as URLs for NNTP newsgroup(s).