Message-ID: <1478811873.31531.4.camel@corsac.net>
Subject: Re: [PATCH] firmware: fix async/manual firmware loading
From: Yves-Alexis Perez <corsac@corsac.net>
To: "Luis R. Rodriguez" <mcgrof@kernel.org>,
        Bjorn Andersson <bjorn.andersson@linaro.org>
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
        "linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>,
        Ming Lei <ming.lei@canonical.com>,
        Johannes Berg <johannes@sipsolutions.net>, Jouni Malinen <j@w1.fi>,
        Kees Cook <keescook@chromium.org>, Jiri Kosina <jkosina@suse.cz>,
        Jiri Slaby <jslaby@suse.com>, Tom Gundersen <teg@jklm.no>,
        Kay Sievers <kay@vrfy.org>, Josh Boyer <jwboyer@fedoraproject.org>,
        Dmitry Torokhov <dmitry.torokhov@gmail.com>,
        Andy Lutomirski <luto@amacapital.net>,
        Harald Hoyer <harald@redhat.com>,
        Seth Forshee <seth.forshee@canonical.com>,
        Daniel Wagner <wagi@monom.org>, "4.2+" <stable@vger.kernel.org>
Date: Thu, 10 Nov 2016 22:04:33 +0100
In-Reply-To: <CAB=NE6UNZLDhEFURvY_zP61WYNJU6CtpB+6FGXCvMGLvSfTd2w@mail.gmail.com>
References: <20161030145048.6291-1-corsac@corsac.net>
         <20161030145048.6291-2-corsac@corsac.net>
         <20161109203921.GH13978@wotan.suse.de> <20161110155511.GA10806@kroah.com>
         <CAB=NE6XBdwe7o5MKHEiNHwNVfRZ39FREwyz1gNoFXoqf=J8qYw@mail.gmail.com>
         <20161110185212.GE11179@tuxbot>
         <CAB=NE6UNZLDhEFURvY_zP61WYNJU6CtpB+6FGXCvMGLvSfTd2w@mail.gmail.com>
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: 8BIT
Mime-Version: 1.0
Sender: linux-kernel-owner@vger.kernel.org
Content-Length: 1390
Lines: 39

On Thu, 2016-11-10 at 11:48 -0800, Luis R. Rodriguez wrote:
> > I haven't verified that this particular use case actually worked before,
> > but this code works with lower timeout values (e.g. 60 in the fallback
> > case), so this looks isolated.
> 
> This is true, but as I noted the broken aspect was when the timeout
> was set to the max value.
> 
> > The bug was clearly introduced in v4.0 by:
> > 
> > 68ff2a00dbf5 "firmware_loader: handle timeout via
> > wait_for_completion_interruptible_timeout()"
> > 
> > So please add a Fixes: and
> > 
> > Reviewed-by: Bjorn Andersson <bjorn.andersson@linaro.org>
> 
> This I agree with, thanks for that, and because of this then:
> 
> Acked-by: Luis R. Rodriguez <mcgrof@kernel.org>
> 
> And because of this do recommend it for stable. I would still prefer
> at least a new re-submit with the respected tags and a changed commit
> log describing the reason for the fix, how the cast is an issue
> exactly, and how this is a regression.

Hi all,

I'm still slightly confused, but I can certainly re-submit it. I've added the
CC: stable because I first experienced it on a stable box, but indeed it was
during coding a kernel driver so it might not be what's expected in
“regression” here.

I'll try to collect the tag and rewrite the commit log, then re-submit,
hopefully not messing with git send-email this time.

Regards,
-- 
Yves-Alexis