netdev - Re: inconsistent lock state with usbnet/asix usb ethernet and xhci

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

Date:   Mon, 05 Mar 2018 11:09:02 -0800
From:   Eric Dumazet <eric.dumazet@...il.com>
To:     Oliver Neukum <oneukum@...e.com>,
        Marek Szyprowski <m.szyprowski@...sung.com>
Cc:     LKML <linux-kernel@...r.kernel.org>,
        'Linux Samsung SOC' <linux-samsung-soc@...r.kernel.org>,
        Linux USB Mailing List <linux-usb@...r.kernel.org>,
        netdev@...r.kernel.org, Dean Jenkins <Dean_Jenkins@...tor.com>
Subject: Re: inconsistent lock state with usbnet/asix usb ethernet and xhci

On Mon, 2018-03-05 at 12:46 +0100, Oliver Neukum wrote:
> On Mon, 2018-03-05 at 08:45 +0100, Marek Szyprowski wrote:
> > Hi Oliver,
> > 
> > On 2018-02-27 17:07, Oliver Neukum wrote:
> > > Am Dienstag, den 27.02.2018, 07:13 -0800 schrieb Eric Dumazet:
> > > > On Tue, 2018-02-27 at 07:09 -0800, Eric Dumazet wrote:
> > > > > 
> > > > > Note that for this one, it seems we also could perform stats
> > > > > updates in
> > > > > BH context, since skb is queued via defer_bh()
> > > > > 
> > > > > But simplicity wins I guess.
> > > > 
> > > > Thinking more about this, I am not sure we have any guarantee
> > > > that TX
> > > > and RX can not run on multiple cpus.
> > > > 
> > > > Using an unique syncp is not going to be safe, even if we make
> > > > lockdep
> > > > happy enough with the local_irq save/restore.
> > > 
> > > Unfortunately you are right. It is not guaranteed for some
> > > hardware.
> > 
> > Does it mean that the fix proposed by Eric is not the proper
> > solution?
> 
> For asix it should work, but asix is unlikely to be the only driver
> with that issue. 32 bit recieves less testing nowadays.

Yes, although the lockdep part could be enforced in 64bit if we really
care.

I will send a patch using two different sync (one for RX, one for TX)