diff mbox series

[v2] irqchip/versatile-fpga: Handle chained IRQs properly

Message ID 20200319023448.1479701-1-mans0n@gorani.run (mailing list archive)
State Mainlined
Commit 486562da598c59e9f835b551d7cf19507de2d681
Headers show
Series [v2] irqchip/versatile-fpga: Handle chained IRQs properly | expand

Commit Message

Sungbo Eo March 19, 2020, 2:34 a.m. UTC
Enclose the chained handler with chained_irq_{enter,exit}(), so that the
muxed interrupts get properly acked.

This patch also fixes a reboot bug on OX820 SoC, where the jiffies timer
interrupt is never acked. The kernel waits a clock tick forever in
calibrate_delay_converge(), which leads to a boot hang.

Fixes: c41b16f8c9d9 ("ARM: integrator/versatile: consolidate FPGA IRQ handling code")
Signed-off-by: Sungbo Eo <mans0n@gorani.run>
Cc: Neil Armstrong <narmstrong@baylibre.com>
---
v2: moved readl below chained_irq_enter()
    added Fixes tag

 drivers/irqchip/irq-versatile-fpga.c | 12 ++++++++++--
 1 file changed, 10 insertions(+), 2 deletions(-)

Comments

Marc Zyngier March 19, 2020, 8:44 a.m. UTC | #1
On 2020-03-19 02:34, Sungbo Eo wrote:
> Enclose the chained handler with chained_irq_{enter,exit}(), so that 
> the
> muxed interrupts get properly acked.
> 
> This patch also fixes a reboot bug on OX820 SoC, where the jiffies 
> timer
> interrupt is never acked. The kernel waits a clock tick forever in
> calibrate_delay_converge(), which leads to a boot hang.
> 
> Fixes: c41b16f8c9d9 ("ARM: integrator/versatile: consolidate FPGA IRQ
> handling code")
> Signed-off-by: Sungbo Eo <mans0n@gorani.run>
> Cc: Neil Armstrong <narmstrong@baylibre.com>
> ---
> v2: moved readl below chained_irq_enter()
>     added Fixes tag
> 
>  drivers/irqchip/irq-versatile-fpga.c | 12 ++++++++++--
>  1 file changed, 10 insertions(+), 2 deletions(-)

Queued for 5.7.

Thanks,

         M.
Linus Walleij March 27, 2020, 10:28 a.m. UTC | #2
On Thu, Mar 19, 2020 at 3:36 AM Sungbo Eo <mans0n@gorani.run> wrote:

> Enclose the chained handler with chained_irq_{enter,exit}(), so that the
> muxed interrupts get properly acked.
>
> This patch also fixes a reboot bug on OX820 SoC, where the jiffies timer
> interrupt is never acked. The kernel waits a clock tick forever in
> calibrate_delay_converge(), which leads to a boot hang.
>
> Fixes: c41b16f8c9d9 ("ARM: integrator/versatile: consolidate FPGA IRQ handling code")
> Signed-off-by: Sungbo Eo <mans0n@gorani.run>
> Cc: Neil Armstrong <narmstrong@baylibre.com>
> ---
> v2: moved readl below chained_irq_enter()
>     added Fixes tag

Reviewed-by: Linus Walleij <linus.walleij@linaro.org>

I wonder how Integrator keeps working so well despite
this. I can't test it right now but I'm sure it is fine.

Yours,
Linus Walleij
diff mbox series

Patch

diff --git a/drivers/irqchip/irq-versatile-fpga.c b/drivers/irqchip/irq-versatile-fpga.c
index 928858dada75..70e2cfff8175 100644
--- a/drivers/irqchip/irq-versatile-fpga.c
+++ b/drivers/irqchip/irq-versatile-fpga.c
@@ -6,6 +6,7 @@ 
 #include <linux/irq.h>
 #include <linux/io.h>
 #include <linux/irqchip.h>
+#include <linux/irqchip/chained_irq.h>
 #include <linux/irqchip/versatile-fpga.h>
 #include <linux/irqdomain.h>
 #include <linux/module.h>
@@ -68,12 +69,16 @@  static void fpga_irq_unmask(struct irq_data *d)
 
 static void fpga_irq_handle(struct irq_desc *desc)
 {
+	struct irq_chip *chip = irq_desc_get_chip(desc);
 	struct fpga_irq_data *f = irq_desc_get_handler_data(desc);
-	u32 status = readl(f->base + IRQ_STATUS);
+	u32 status;
+
+	chained_irq_enter(chip, desc);
 
+	status = readl(f->base + IRQ_STATUS);
 	if (status == 0) {
 		do_bad_IRQ(desc);
-		return;
+		goto out;
 	}
 
 	do {
@@ -82,6 +87,9 @@  static void fpga_irq_handle(struct irq_desc *desc)
 		status &= ~(1 << irq);
 		generic_handle_irq(irq_find_mapping(f->domain, irq));
 	} while (status);
+
+out:
+	chained_irq_exit(chip, desc);
 }
 
 /*