linux-kernel - Re: [PATCH v2] uart: fix race between uart_put_char() and uart

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:   Fri, 6 Jul 2018 16:39:19 +0200
From:   Greg Kroah-Hartman <gregkh@...uxfoundation.org>
To:     Tycho Andersen <tycho@...ho.ws>
Cc:     Jiri Slaby <jslaby@...e.com>, linux-serial@...r.kernel.org,
        linux-kernel@...r.kernel.org, "Serge E . Hallyn" <serge@...lyn.com>
Subject: Re: [PATCH v2] uart: fix race between uart_put_char() and
 uart_shutdown()

On Fri, Jun 29, 2018 at 10:43:30AM -0600, Tycho Andersen wrote:
> On Fri, Jun 29, 2018 at 04:24:46AM -0600, Tycho Andersen wrote:
> > v2: switch to locking uport->lock on allocation/deallocation instead of
> >     locking the per-port mutex in uart_put_char. Note that since
> >     uport->lock is a spin lock, we have to switch the allocation to
> >     GFP_ATOMIC.
> 
> Serge pointed out off-list that we may want to do the allocation
> before the lock so that it's more likely to be successful. I'm happy
> to send that change to this if it's what we want to do, I don't have a
> strong preference.

That sounds like a much better thing to do.

thanks,

greg k-h