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;
Subject: Re: [RFC PATCH 1/7] drivers: base: Add resource managed version of
 delayed work init
From:   Hans de Goede <hdegoede@redhat.com>
To:     Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Cc:     Matti Vaittinen <matti.vaittinen@fi.rohmeurope.com>,
        mazziesaccount@gmail.com, "Rafael J. Wysocki" <rafael@kernel.org>,
        MyungJoo Ham <myungjoo.ham@samsung.com>,
        Chanwoo Choi <cw00.choi@samsung.com>,
        Andy Gross <agross@kernel.org>,
        Bjorn Andersson <bjorn.andersson@linaro.org>,
        Jean Delvare <jdelvare@suse.com>,
        Guenter Roeck <linux@roeck-us.net>,
        Mark Gross <mgross@linux.intel.com>,
        Sebastian Reichel <sre@kernel.org>,
        Chen-Yu Tsai <wens@csie.org>,
        Liam Girdwood <lgirdwood@gmail.com>,
        Mark Brown <broonie@kernel.org>,
        Wim Van Sebroeck <wim@linux-watchdog.org>,
        Saravana Kannan <saravanak@google.com>,
        Heikki Krogerus <heikki.krogerus@linux.intel.com>,
        Andy Shevchenko <andriy.shevchenko@linux.intel.com>,
        Joerg Roedel <jroedel@suse.de>,
        Dan Williams <dan.j.williams@intel.com>,
        Bartosz Golaszewski <bgolaszewski@baylibre.com>,
        linux-kernel@vger.kernel.org, linux-arm-msm@vger.kernel.org,
        linux-hwmon@vger.kernel.org, platform-driver-x86@vger.kernel.org,
        linux-pm@vger.kernel.org, linux-watchdog@vger.kernel.org
References: <cover.1613216412.git.matti.vaittinen@fi.rohmeurope.com>
 <1230b0d2ba99ad546d72ab079e76cb1b3df32afb.1613216412.git.matti.vaittinen@fi.rohmeurope.com>
 <YCfDAly9b0zHMpJT@kroah.com>
 <284d4a13-5cc8-e23c-7e99-c03db5415bf1@redhat.com>
 <YCfVKyXbeJXNbMsd@kroah.com>
 <eecf8ad6-a99d-8ad0-fc7e-f87d015add3f@redhat.com>
Message-ID: <529a5122-ff3a-2af0-52ed-b13216fa49d1@redhat.com>
Date:   Sat, 13 Feb 2021 15:52:06 +0100
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:78.0) Gecko/20100101
 Thunderbird/78.7.0
MIME-Version: 1.0
In-Reply-To: <eecf8ad6-a99d-8ad0-fc7e-f87d015add3f@redhat.com>
Content-Type: text/plain; charset=utf-8
Content-Language: en-US
Content-Transfer-Encoding: 7bit
Precedence: bulk

Hi,

On 2/13/21 3:38 PM, Hans de Goede wrote:
> Hi,
> 
> On 2/13/21 2:33 PM, Greg Kroah-Hartman wrote:

<snip>

> Having this new devm_delayed_work_autocancel() helper will allow a
> bunch of drivers to move away from mixing the 2, which is a good thing
> in my book.
> 
> As I said above I believe that having devm_delayed_work_autocancel() (1)
> in our toolbox will be a good thing to have. Driver authors can then choose
> to use it; or they can choose to not use it if they don't like it.
> 
> I know that the reason why I did not use it in the
> drivers/extcon/extcon-intel-int3496.c driver is because it was not available
> if it had been available then I would definitely have used it, as it avoids the
> mixing of resource-management styles which that driver is currently doing.
> 
> And I think that that is what this is ultimately about, there are 2 styles
> of resource-management:
> 
> 1. manual
> 2. devm based
> 
> And they both have their pros and cons, problems mostly arise when mixing them
> and adding new devm helpers for commonly used cleanup patterns is a good thing
> as it helps to get rid of mixing these 2 styles in a single driver.

I just noticed that I forgot to fill in the (1) footnote above:

1) And we probably will want one for non delayed work items to: devm_work_autocancel(),
but lets cross that bridge when we get there.

Also when reviewing: "[RFC PATCH 2/7] extconn: Clean-up few drivers by using managed work init"
I noticed that the extcon-qcom-spmi-misc.c and extcon-palmas.c follow the same standard
pattern of having an IRQ which queues a delayed work and they both miss the devm_free_irq
call before the cancel_delayed_work_sync() on driver release. So just patch 2/7 here
fixes 3 driver-release race conditions (fixing 3/4 drivers which it touches) as a
bonus to the code-cleanup which it does.

So as this clearly seems to be fixing a bunch of bugs, by simply completely removing the
buggy code driver remove callbacks, this really seems like a good idea to me.

Regards,

Hans