linux-kernel - Re: [PATCH 4.9 41/51] fib_rules: return 0 directly if an exactly same rule exists when NLM_F

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [day] [month] [year] [list]

Message-ID: <20190520091131.GA1593@kroah.com>
Date:   Mon, 20 May 2019 11:11:31 +0200
From:   Greg Kroah-Hartman <gregkh@...uxfoundation.org>
To:     David Ahern <dsahern@...il.com>
Cc:     Nathan Chancellor <natechancellor@...il.com>,
        linux-kernel@...r.kernel.org, stable@...r.kernel.org,
        Thomas Haller <thaller@...hat.com>,
        Hangbin Liu <liuhangbin@...il.com>,
        "David S. Miller" <davem@...emloft.net>, netdev@...r.kernel.org
Subject: Re: [PATCH 4.9 41/51] fib_rules: return 0 directly if an exactly
 same rule exists when NLM_F_EXCL not supplied

On Mon, May 20, 2019 at 11:04:29AM +0200, Greg Kroah-Hartman wrote:
> On Sun, May 19, 2019 at 06:29:19PM -0600, David Ahern wrote:
> > On 5/19/19 9:43 AM, Nathan Chancellor wrote:
> > > Hi all,
> > > 
> > > This commit is causing issues on Android devices when Wi-Fi and mobile
> > > data are both enabled. The device will do a soft reboot consistently.
> > > So far, I've had reports on the Pixel 3 XL, OnePlus 6, Pocophone, and
> > > Note 9 and I can reproduce on my OnePlus 6.
> > > 
> > > Sorry for taking so long to report this, I just figured out how to
> > > reproduce it today and I didn't want to report it without that.
> > > 
> > > Attached is a full dmesg and the relevant snippet from Android's logcat.
> > > 
> > > Let me know what I can do to help debug,
> > > Nathan
> > > 
> > 
> > It's a backport problem. err needs to be reset to 0 before the goto.
> 
> Ah, I see it, let me go queue up a fix for this.

Here's the fix I'm queueing up now:


>From b42f0ebbe4431ff7ce99c916555418f4a4c2be67 Mon Sep 17 00:00:00 2001
From: Greg Kroah-Hartman <gregkh@...uxfoundation.org>
Date: Mon, 20 May 2019 11:07:29 +0200
Subject: [PATCH] fib_rules: fix error in backport of e9919a24d302 ("fib_rules:
 return 0...")

When commit e9919a24d302 ("fib_rules: return 0 directly if an exactly
same rule exists when NLM_F_EXCL not supplied") was backported to 4.9.y,
it changed the logic a bit as err should have been reset before exiting
the test, like it happens in the original logic.

If this is not set, errors happen :(

Reported-by: Nathan Chancellor <natechancellor@...il.com>
Reported-by: David Ahern <dsahern@...il.com>
Reported-by: Florian Westphal <fw@...len.de>
Cc: Hangbin Liu <liuhangbin@...il.com>
Cc: David S. Miller <davem@...emloft.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@...uxfoundation.org>
---
 net/core/fib_rules.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/net/core/fib_rules.c b/net/core/fib_rules.c
index bb26457e8c21..c03dd2104d33 100644
--- a/net/core/fib_rules.c
+++ b/net/core/fib_rules.c
@@ -430,6 +430,7 @@ int fib_nl_newrule(struct sk_buff *skb, struct nlmsghdr *nlh)
 		goto errout_free;
 
 	if (rule_exists(ops, frh, tb, rule)) {
+		err = 0;
 		if (nlh->nlmsg_flags & NLM_F_EXCL)
 			err = -EEXIST;
 		goto errout_free;
-- 
2.21.0