On Thu, Jan 22, 2026 at 3:06 AM Florian Fainelli
[off-list ref] wrote:

From: Doug Berger <opendmb@gmail.com>

The irq_mask_ack operation is slightly more efficient than doing
irq_mask and irq_ack separately.

I would refer to the callbacks as

.irq_mask()
.irq_ack()

et cetera.

More importantly for this driver it bypasses the check of
irqd_irq_masked ensuring a previously masked but still active
interrupt gets remasked if unmasked at the hardware level. This
allows the driver to more efficiently unmask the wake capable
interrupts when quiescing without needing to enable the irqs
individually to clear the irqd_irq_masked state.

...

-// Copyright (C) 2015-2017 Broadcom
+// Copyright (C) 2015-2026 Broadcom

Shouldn't it be rather 2015-2017,2026 ? (In one case when I updated a
driver for Intel, I went via Git history to gather the info.)

...

 static void brcmstb_gpio_set_imask(struct brcmstb_gpio_bank *bank,
-               unsigned int hwirq, bool enable)
+               unsigned int hwirq, bool enable, bool ack)

This type of interface is usually discouraged as it makes code harder
to read and follow. Since there are a lot of duplication, I recommend
to move the ack op to a separate helper.

...

-       gpio_generic_write_reg(&bank->chip,
-                              priv->reg_base + GIO_MASK(bank->id), imask);
+       if (ack)
+               gpio_generic_write_reg(&bank->chip,
+                                      priv->reg_base + GIO_MASK(bank->id),
+                                      imask);

Id est this piece...

+static void brcmstb_gpio_irq_mask_ack(struct irq_data *d)
+{
+       struct gpio_chip *gc = irq_data_get_irq_chip_data(d);
+       struct brcmstb_gpio_bank *bank = gpiochip_get_data(gc);
+
+       brcmstb_gpio_set_imask(bank, d->hwirq, false, true);

...and call it here explicitly (seems the only place for it, so it can
even be just moved here without an intermediate helper).

 }

-- 
With Best Regards,
Andy Shevchenko