Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 23.128.96.18 as permitted sender) client-ip=23.128.96.18;
MIME-Version: 1.0
References: <20200320093125.23092-1-brgl@bgdev.pl> <CACRpkdZgWUwmmuXn12DS3TsQS0yQxcweqK6HGxBm=V_2LBLBMw@mail.gmail.com>
 <CAMpxmJUb09KGreHw6Bdz79rbnQE7oZnWg_5qN_FhzoS2-XccFA@mail.gmail.com> <CAHp75Vdpb=hNiR3c7G_yTeSt70Vcy3DWHin0B5+WYV1hbRMBJQ@mail.gmail.com>
In-Reply-To: <CAHp75Vdpb=hNiR3c7G_yTeSt70Vcy3DWHin0B5+WYV1hbRMBJQ@mail.gmail.com>
From:   Bartosz Golaszewski <bgolaszewski@baylibre.com>
Date:   Wed, 29 Apr 2020 08:40:03 +0200
Message-ID: <CAMpxmJVT4wm26hQB-_BV73tC5_nqH5JG9KmDuNQ2OJe+tE+gLg@mail.gmail.com>
Subject: Re: [PATCH] gpiolib: don't call sleeping functions with a spinlock taken
To:     Andy Shevchenko <andy.shevchenko@gmail.com>
Cc:     Linus Walleij <linus.walleij@linaro.org>,
        Bartosz Golaszewski <brgl@bgdev.pl>,
        Geert Uytterhoeven <geert@linux-m68k.org>,
        "open list:GPIO SUBSYSTEM" <linux-gpio@vger.kernel.org>,
        "linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk

wt., 28 kwi 2020 o 17:53 Andy Shevchenko <andy.shevchenko@gmail.com> napisa=
=C5=82(a):
>
> On Tue, Apr 14, 2020 at 6:35 PM Bartosz Golaszewski
> <bgolaszewski@baylibre.com> wrote:
> >
> > wt., 14 kwi 2020 o 14:00 Linus Walleij <linus.walleij@linaro.org> napis=
a=C5=82(a):
> > >
> > > On Fri, Mar 20, 2020 at 10:31 AM Bartosz Golaszewski <brgl@bgdev.pl> =
wrote:
> > >
> > > > From: Bartosz Golaszewski <bgolaszewski@baylibre.com>
> > > >
> > > > We must not call pinctrl_gpio_can_use_line() with the gpio_lock tak=
en
> > > > as it takes a mutex internally. Let's move the call before taking t=
he
> > > > spinlock and store the return value.
> > > >
> > > > This isn't perfect - there's a moment between calling
> > > > pinctrl_gpio_can_use_line() and taking the spinlock where the situa=
tion
> > > > can change but it isn't a regression either: previously this part w=
asn't
> > > > protected at all and it only affects the information user-space is
> > > > seeing.
>
> It seems I have no original at hand, so, commenting here.
>
> It looks like we need a mutex less function which can be used here and
> in the call you are considering racy.

The thing is this mutex is in pinctrl - we'd need to export it too so
that gpio can use it.

Bart