All of lore.kernel.org
 help / color / mirror / Atom feed
From: Gregory Fong <gregory.0xf0@gmail.com>
To: Doug Berger <opendmb@gmail.com>
Cc: Linus Walleij <linus.walleij@linaro.org>,
	Brian Norris <computersforpeace@gmail.com>,
	Florian Fainelli <f.fainelli@gmail.com>,
	bcm-kernel-feedback-list <bcm-kernel-feedback-list@broadcom.com>,
	linux-gpio@vger.kernel.org,
	"linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>,
	"linux-arm-kernel@lists.infradead.org"
	<linux-arm-kernel@lists.infradead.org>
Subject: Re: [PATCH 2/7] gpio: brcmstb: release the bgpio lock during irq handlers
Date: Tue, 3 Oct 2017 18:55:51 -0700	[thread overview]
Message-ID: <CADtm3G58SBF0FskUPJCs_8km8oXrKYs-+hrJoV4osyw9mvQbiQ@mail.gmail.com> (raw)
In-Reply-To: <20170930034057.15166-3-opendmb@gmail.com>

Hi Doug,

On Fri, Sep 29, 2017 at 8:40 PM, Doug Berger <opendmb@gmail.com> wrote:
> The basic memory-mapped GPIO controller lock must be released
> before calling the registered GPIO interrupt handlers to allow
> the interrupt handlers to access the hardware.  Otherwise, the
> hardware accesses will deadlock when they attempt to grab the
> lock.

I was having some trouble understanding exactly what the problem was
here, but I think I see it now.  Since this locks the entire bank,
where some GPIOs might be set as inputs and some as inputs (and
interrupt sources), then an interrupt on a GPIO that is supposed to
set another GPIO in the bank would result in deadlock.  Is that
correct?  If so, please update the commit message to make that clear,
and nice fix.  If not that, it would be nice to know what scenario can
cause a problem.

Thanks,
Gregory

WARNING: multiple messages have this Message-ID (diff)
From: gregory.0xf0@gmail.com (Gregory Fong)
To: linux-arm-kernel@lists.infradead.org
Subject: [PATCH 2/7] gpio: brcmstb: release the bgpio lock during irq handlers
Date: Tue, 3 Oct 2017 18:55:51 -0700	[thread overview]
Message-ID: <CADtm3G58SBF0FskUPJCs_8km8oXrKYs-+hrJoV4osyw9mvQbiQ@mail.gmail.com> (raw)
In-Reply-To: <20170930034057.15166-3-opendmb@gmail.com>

Hi Doug,

On Fri, Sep 29, 2017 at 8:40 PM, Doug Berger <opendmb@gmail.com> wrote:
> The basic memory-mapped GPIO controller lock must be released
> before calling the registered GPIO interrupt handlers to allow
> the interrupt handlers to access the hardware.  Otherwise, the
> hardware accesses will deadlock when they attempt to grab the
> lock.

I was having some trouble understanding exactly what the problem was
here, but I think I see it now.  Since this locks the entire bank,
where some GPIOs might be set as inputs and some as inputs (and
interrupt sources), then an interrupt on a GPIO that is supposed to
set another GPIO in the bank would result in deadlock.  Is that
correct?  If so, please update the commit message to make that clear,
and nice fix.  If not that, it would be nice to know what scenario can
cause a problem.

Thanks,
Gregory

  reply	other threads:[~2017-10-04  1:56 UTC|newest]

Thread overview: 76+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2017-09-30  3:40 [PATCH 0/7] gpio: brcmstb: improved interrupt and wake support Doug Berger
2017-09-30  3:40 ` Doug Berger
2017-09-30  3:40 ` [PATCH 1/7] gpio: brcmstb: allow all instances to be wakeup sources Doug Berger
2017-09-30  3:40   ` Doug Berger
2017-10-04  1:40   ` Gregory Fong
2017-10-04  1:40     ` Gregory Fong
2017-10-04  1:40     ` Gregory Fong
2017-10-07 10:52   ` Linus Walleij
2017-10-07 10:52     ` Linus Walleij
2017-10-07 10:52     ` Linus Walleij
2017-09-30  3:40 ` [PATCH 2/7] gpio: brcmstb: release the bgpio lock during irq handlers Doug Berger
2017-09-30  3:40   ` Doug Berger
2017-10-04  1:55   ` Gregory Fong [this message]
2017-10-04  1:55     ` Gregory Fong
2017-10-04  1:55     ` Gregory Fong
2017-10-04  2:09     ` Doug Berger
2017-10-04  2:09       ` Doug Berger
2017-10-04  2:09       ` Doug Berger
2017-10-04  3:07       ` Gregory Fong
2017-10-04  3:07         ` Gregory Fong
2017-10-04  3:07         ` Gregory Fong
2017-09-30  3:40 ` [PATCH 3/7] gpio: brcmstb: switch to handle_level_irq flow Doug Berger
2017-09-30  3:40   ` Doug Berger
2017-10-04  1:59   ` Gregory Fong
2017-10-04  1:59     ` Gregory Fong
2017-10-04  1:59     ` Gregory Fong
2017-09-30  3:40 ` [PATCH 4/7] gpio: brcmstb: correct the configuration of level interrupts Doug Berger
2017-09-30  3:40   ` Doug Berger
2017-10-04  2:03   ` Gregory Fong
2017-10-04  2:03     ` Gregory Fong
2017-10-04  2:03     ` Gregory Fong
2017-09-30  3:40 ` [PATCH 5/7] gpio: brcmstb: enable masking of interrupts when changing type Doug Berger
2017-09-30  3:40   ` Doug Berger
2017-10-04  2:10   ` Gregory Fong
2017-10-04  2:10     ` Gregory Fong
2017-10-04  2:10     ` Gregory Fong
2017-10-04  2:22     ` Doug Berger
2017-10-04  2:22       ` Doug Berger
2017-10-04  2:22       ` Doug Berger
2017-10-04  3:15       ` Gregory Fong
2017-10-04  3:15         ` Gregory Fong
2017-10-04  3:15         ` Gregory Fong
2017-09-30  3:40 ` [PATCH 6/7] gpio: brcmstb: consolidate interrupt domains Doug Berger
2017-09-30  3:40   ` Doug Berger
2017-10-04  3:03   ` Gregory Fong
2017-10-04  3:03     ` Gregory Fong
2017-10-04  3:03     ` Gregory Fong
2017-10-04 21:24     ` Doug Berger
2017-10-04 21:24       ` Doug Berger
2017-10-04 21:24       ` Doug Berger
2017-10-16 23:04       ` Doug Berger
2017-10-16 23:04         ` Doug Berger
2017-10-16 23:04         ` Doug Berger
2017-10-19  7:57       ` Gregory Fong
2017-10-19  7:57         ` Gregory Fong
2017-10-19  7:57         ` Gregory Fong
2017-10-19 18:25         ` Doug Berger
2017-10-19 18:25           ` Doug Berger
2017-10-19 18:25           ` Doug Berger
2017-09-30  3:40 ` [PATCH 7/7] gpio: brcmstb: implement suspend/resume/shutdown Doug Berger
2017-09-30  3:40   ` Doug Berger
2017-10-19  9:03   ` Gregory Fong
2017-10-19  9:03     ` Gregory Fong
2017-10-19 18:39     ` Doug Berger
2017-10-19 18:39       ` Doug Berger
2017-10-21  0:54       ` Gregory Fong
2017-10-21  0:54         ` Gregory Fong
2017-10-21  0:54         ` Gregory Fong
2017-10-23 23:06         ` Doug Berger
2017-10-23 23:06           ` Doug Berger
2017-10-23 23:06           ` Doug Berger
2017-09-30  5:34 ` [PATCH 0/7] gpio: brcmstb: improved interrupt and wake support Florian Fainelli
2017-09-30  5:34   ` Florian Fainelli
2017-10-07 10:54 ` Linus Walleij
2017-10-07 10:54   ` Linus Walleij
2017-10-07 10:54   ` Linus Walleij

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=CADtm3G58SBF0FskUPJCs_8km8oXrKYs-+hrJoV4osyw9mvQbiQ@mail.gmail.com \
    --to=gregory.0xf0@gmail.com \
    --cc=bcm-kernel-feedback-list@broadcom.com \
    --cc=computersforpeace@gmail.com \
    --cc=f.fainelli@gmail.com \
    --cc=linus.walleij@linaro.org \
    --cc=linux-arm-kernel@lists.infradead.org \
    --cc=linux-gpio@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=opendmb@gmail.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body. 
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.