From: Alexandre Belloni <alexandre.belloni@bootlin.com>
To: 최유호 <dbgh9129@gmail.com>
Cc: Daniel Lezcano <daniel.lezcano@kernel.org>,
Thomas Gleixner <tglx@kernel.org>,
Nicolas Ferre <nicolas.ferre@microchip.com>,
Claudiu Beznea <claudiu.beznea@tuxon.dev>,
Boris Brezillon <bbrezillon@kernel.org>,
linux-arm-kernel@lists.infradead.org,
linux-kernel@vger.kernel.org
Subject: Re: [PATCH v1] clocksource/drivers/timer-atmel-pit: Fix init failure cleanup
Date: Tue, 28 Jul 2026 12:01:09 +0200 [thread overview]
Message-ID: <20260728100109ca239fcb@mail.local> (raw)
In-Reply-To: <CACrCO_XQn9WLVVx5NhvzAw=FuW-4R0Du58MGwOQ5q6dgGbtKYg@mail.gmail.com>
Hello,
On 02/07/2026 17:12:48-0400, 최유호 wrote:
> Hi,
>
> Just a gentle ping on this patch.
>
> I would appreciate any feedback when you have a chance to review this.
>
> Thanks
>
> On Mon, 8 Jun 2026 at 14:20, Yuho Choi <dbgh9129@gmail.com> wrote:
> >
> > After clk_prepare_enable(data->mck) succeeds, at91sam926x_pit_dt_init()
> > can still fail while parsing the IRQ, registering the clocksource, or
> > requesting the IRQ. These paths only free the driver data, leaving the
> > master clock enabled.
> >
> > Unwind each initialized state on failure. Stop the PIT after it has been
> > started, dispose the IRQ mapping after it has been created, disable and
> > put the master clock, and unmap the registers before freeing the driver
> > data.
> >
This is a bit moot because if anything in the probe fails, then he
platform will probably not boot at all.
> > Fixes: 699e36e5b8e9 ("clocksource/drivers/timer-atmel-pit: Enable mck clock")
> > Signed-off-by: Yuho Choi <dbgh9129@gmail.com>
> > ---
> > drivers/clocksource/timer-atmel-pit.c | 21 ++++++++++++++++-----
> > 1 file changed, 16 insertions(+), 5 deletions(-)
> >
> > diff --git a/drivers/clocksource/timer-atmel-pit.c b/drivers/clocksource/timer-atmel-pit.c
> > index 888b06731e54..edd427ab93e6 100644
> > --- a/drivers/clocksource/timer-atmel-pit.c
> > +++ b/drivers/clocksource/timer-atmel-pit.c
> > @@ -12,7 +12,9 @@
> > #include <linux/clk.h>
> > #include <linux/clockchips.h>
> > #include <linux/interrupt.h>
> > +#include <linux/io.h>
> > #include <linux/irq.h>
> > +#include <linux/irqdomain.h>
> > #include <linux/kernel.h>
> > #include <linux/of.h>
> > #include <linux/of_address.h>
> > @@ -185,13 +187,13 @@ static int __init at91sam926x_pit_dt_init(struct device_node *node)
> > if (IS_ERR(data->mck)) {
> > pr_err("Unable to get mck clk\n");
> > ret = PTR_ERR(data->mck);
> > - goto exit;
> > + goto exit_iounmap;
> > }
> >
> > ret = clk_prepare_enable(data->mck);
> > if (ret) {
> > pr_err("Unable to enable mck\n");
> > - goto exit;
> > + goto exit_clk_put;
> > }
Switching to devm_clk_get_enabled would avoid most of the error
handling. Similarly, devm_ioremap could be used earlier.
> >
> > /* Get the interrupts property */
> > @@ -199,7 +201,7 @@ static int __init at91sam926x_pit_dt_init(struct device_node *node)
> > if (!data->irq) {
> > pr_err("Unable to get IRQ from DT\n");
> > ret = -EINVAL;
> > - goto exit;
> > + goto exit_clk_disable;
> > }
> >
> > /*
> > @@ -227,7 +229,7 @@ static int __init at91sam926x_pit_dt_init(struct device_node *node)
> > ret = clocksource_register_hz(&data->clksrc, pit_rate);
> > if (ret) {
> > pr_err("Failed to register clocksource\n");
> > - goto exit;
> > + goto exit_pit_disable;
> > }
> >
> > /* Set up irq handler */
> > @@ -237,7 +239,7 @@ static int __init at91sam926x_pit_dt_init(struct device_node *node)
> > if (ret) {
> > pr_err("Unable to setup IRQ\n");
> > clocksource_unregister(&data->clksrc);
> > - goto exit;
> > + goto exit_pit_disable;
> > }
> >
> > /* Set up and register clockevents */
> > @@ -256,6 +258,15 @@ static int __init at91sam926x_pit_dt_init(struct device_node *node)
> >
> > return 0;
> >
> > +exit_pit_disable:
> > + pit_write(data->base, AT91_PIT_MR, 0);
I doubt this write is useful.
> > + irq_dispose_mapping(data->irq);
> > +exit_clk_disable:
> > + clk_disable_unprepare(data->mck);
> > +exit_clk_put:
> > + clk_put(data->mck);
> > +exit_iounmap:
> > + iounmap(data->base);
> > exit:
> > kfree(data);
> > return ret;
> > --
> > 2.43.0
> >
--
Alexandre Belloni, co-owner and COO, Bootlin
Embedded Linux and Kernel engineering
https://bootlin.com
prev parent reply other threads:[~2026-07-28 10:01 UTC|newest]
Thread overview: 3+ messages / expand[flat|nested] mbox.gz Atom feed top
2026-06-08 18:20 [PATCH v1] clocksource/drivers/timer-atmel-pit: Fix init failure cleanup Yuho Choi
2026-07-02 21:12 ` 최유호
2026-07-28 10:01 ` Alexandre Belloni [this message]
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=20260728100109ca239fcb@mail.local \
--to=alexandre.belloni@bootlin.com \
--cc=bbrezillon@kernel.org \
--cc=claudiu.beznea@tuxon.dev \
--cc=daniel.lezcano@kernel.org \
--cc=dbgh9129@gmail.com \
--cc=linux-arm-kernel@lists.infradead.org \
--cc=linux-kernel@vger.kernel.org \
--cc=nicolas.ferre@microchip.com \
--cc=tglx@kernel.org \
/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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox