diff mbox series

[v2] pinctrl: renesas: rzg2l: Use spin_{lock,unlock}_irq{save,restore}

Message ID 20240522055421.2842689-1-claudiu.beznea.uj@bp.renesas.com
State New
Headers show
Series [v2] pinctrl: renesas: rzg2l: Use spin_{lock,unlock}_irq{save,restore} | expand

Commit Message

Claudiu Beznea May 22, 2024, 5:54 a.m. UTC 
From: Claudiu Beznea <claudiu.beznea.uj@bp.renesas.com>

On PREEMPT_RT kernels the spinlock_t maps to an rtmutex. Using
raw_spin_lock_irqsave()/raw_spin_unlock_irqrestore() on
&pctrl->lock.rlock breaks the PREEMPT_RT builds. To fix this use
spin_lock_irqsave()/spin_unlock_irqrestore() on &pctrl->lock.

Fixes: 02cd2d3be1c3 ("pinctrl: renesas: rzg2l: Configure the interrupt type on resume")
Reported-by: Diederik de Haas <didi.debian@cknow.org>
Closes: https://lore.kernel.org/all/131999629.KQPSlr0Zke@bagend
Signed-off-by: Claudiu Beznea <claudiu.beznea.uj@bp.renesas.com>
---

Changes in v2:
- use proper commit id in "Fixes" section
- s/use/Use in patch title

 drivers/pinctrl/renesas/pinctrl-rzg2l.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

Comments

Sebastian Andrzej Siewior June 5, 2024, 11 a.m. UTC | #1 
On 2024-05-22 11:27:55 [+0200], Geert Uytterhoeven wrote:
> 
> Reviewed-by: Geert Uytterhoeven <geert+renesas@glider.be>
> i.e. will queue in renesas-pinctrl-fixes for v6.10.

I picked this up for RT and see that it is not part of v6.10-rc2. I
assume that this will be applied within the v6.10 cycle once you have
more in your tree, right?

> Gr{oetje,eeting}s,
> 
>                         Geert

Sebastian
Geert Uytterhoeven June 5, 2024, 11:22 a.m. UTC | #2 
Hi Sebastian,

On Wed, Jun 5, 2024 at 1:00 PM Sebastian Andrzej Siewior
<bigeasy@linutronix.de> wrote:
> On 2024-05-22 11:27:55 [+0200], Geert Uytterhoeven wrote:
> > Reviewed-by: Geert Uytterhoeven <geert+renesas@glider.be>
> > i.e. will queue in renesas-pinctrl-fixes for v6.10.
>
> I picked this up for RT and see that it is not part of v6.10-rc2. I
> assume that this will be applied within the v6.10 cycle once you have
> more in your tree, right?

I have already sent a pull request for it to LinusW, and it is now part
of pinctrl/for-next.  I assume LinusW will send a PR to Linus(T ;-)
in time later.

Gr{oetje,eeting}s,

                        Geert
Sebastian Andrzej Siewior June 5, 2024, 1:26 p.m. UTC | #3 
On 2024-06-05 13:22:50 [+0200], Geert Uytterhoeven wrote:
> Hi Sebastian,
Hi Geert,

> I have already sent a pull request for it to LinusW, and it is now part
> of pinctrl/for-next.  I assume LinusW will send a PR to Linus(T ;-)
> in time later.

Thanks for the update.

> Gr{oetje,eeting}s,
> 
>                         Geert

Sebastian
diff mbox series

Patch

diff --git a/drivers/pinctrl/renesas/pinctrl-rzg2l.c b/drivers/pinctrl/renesas/pinctrl-rzg2l.c
index 20425afc6b33..78f947a8e26e 100644
--- a/drivers/pinctrl/renesas/pinctrl-rzg2l.c
+++ b/drivers/pinctrl/renesas/pinctrl-rzg2l.c
@@ -2069,11 +2069,11 @@  static void rzg2l_gpio_irq_restore(struct rzg2l_pinctrl *pctrl)
 		 * This has to be atomically executed to protect against a concurrent
 		 * interrupt.
 		 */
-		raw_spin_lock_irqsave(&pctrl->lock.rlock, flags);
+		spin_lock_irqsave(&pctrl->lock, flags);
 		ret = rzg2l_gpio_irq_set_type(data, irqd_get_trigger_type(data));
 		if (!ret && !irqd_irq_disabled(data))
 			rzg2l_gpio_irq_enable(data);
-		raw_spin_unlock_irqrestore(&pctrl->lock.rlock, flags);
+		spin_unlock_irqrestore(&pctrl->lock, flags);
 
 		if (ret)
 			dev_crit(pctrl->dev, "Failed to set IRQ type for virq=%u\n", virq);