Skip to contentSkip to conversation

gpio: pl061: Fix the issue failed to register the ACPI interruption

PATCH Superseded Wei Xu 3 participants 5 messages Aug 12, 2019 – Aug 16, 2019 GPIO

v1 of 3 Superseded Latest is v3 →

lore
Share & tools
lore

Wei Xu

author
#0 lore raw

PATCH 1/1 gpio: pl061: Fix the issue failed to register the ACPI interruption

Invoke acpi_gpiochip_request_interrupts after the acpi data has been attached to the pl061 acpi node to register interruption.

Otherwise it will be failed to register interruption for the ACPI case. Because in the gpiochip_add_data_with_key, acpi_gpiochip_add is invoked after gpiochip_add_irqchip but at that time the acpi data has not been attached yet.

Tested with below steps on QEMU v4.1.0-rc3 and Linux kernel v5.3-rc4, and found pl061 interruption is missed in the /proc/interrupts:
1.
qemu-system-aarch64 \
-machine virt,gic-version=3 -cpu cortex-a57 \
-m 1G,maxmem=4G,slots=4 \
-kernel Image -initrd rootfs.cpio.gz \
-net none -nographic \
-bios QEMU_EFI.fd \
-append "console=ttyAMA0 acpi=force earlycon=pl011,0x9000000"

2. cat /proc/interrupts in the guest console:

estuary:/$ cat /proc/interrupts
CPU0
2:       3228     GICv3  27 Level     arch_timer
4:         15     GICv3  33 Level     uart-pl011
42:          0     GICv3  23 Level     arm-pmu
IPI0:         0       Rescheduling interrupts
IPI1:         0       Function call interrupts
IPI2:         0       CPU stop interrupts
IPI3:         0       CPU stop (for crash dump) interrupts
IPI4:         0       Timer broadcast interrupts
IPI5:         0       IRQ work interrupts
IPI6:         0       CPU wake-up interrupts
Err:          0
Fixes 04ce935c6b2a ("gpio: pl061: Pass irqchip when adding gpiochip")
---
 drivers/gpio/gpio-pl061.c | 7 +++++++
 1 file changed, 7 insertions(+)
drivers/gpio/gpio-pl061.c +7 -0
index 722ce5c..e1a434e 100644
@@ -8,6 +8,7 @@
  *
  * Data sheet: ARM DDI 0190B, September 2000
  */
+#include <linux/acpi.h>
 #include <linux/spinlock.h>
 #include <linux/errno.h>
 #include <linux/init.h>
@@ -24,6 +25,9 @@
 #include <linux/pinctrl/consumer.h>
 #include <linux/pm.h>
 
+#include "gpiolib.h"
+#include "gpiolib-acpi.h"
+
 #define GPIODIR 0x400
 #define GPIOIS  0x404
 #define GPIOIBE 0x408
@@ -345,6 +349,9 @@ static int pl061_probe(struct amba_device *adev, const struct amba_id *id)
     if (ret)
         return ret;
 
+    if (has_acpi_companion(dev))
+        acpi_gpiochip_request_interrupts(&pl061->gc);
+
     amba_set_drvdata(adev, pl061);
     dev_info(dev, "PL061 GPIO chip registered\n");
 
Signature · 4 lines
2.8.1


.
#1 lore raw

Hi Wei,

thanks for your patch!

This doesn't apply for my "devel" branch, can you rebase on this:
https://git.kernel.org/pub/scm/linux/kernel/git/linusw/linux-gpio.git/log/?h=devel

We have moved some ACPI headers around recently.

Quoted 2 lines · Wei Xu

Invoke acpi_gpiochip_request_interrupts after the acpi data has been attached to the pl061 acpi node to register interruption.

Makes sense.

Quoted 1 line
Fixes 04ce935c6b2a ("gpio: pl061: Pass irqchip when adding gpiochip")

I doubt this is a regression since I haven't seen anyone use this gpiochip with ACPI before.

Please rename the patch "gpio: pl061: Add ACPI support" unless you can convince me it worked without changes before.

Please include some ACPI people on review of this. From MAINTAINERS:
ACPI

M:      "Rafael J. Wysocki" <[email protected]>
M:      Len Brown <[email protected]>
L:      [email protected]

I would also include Andy Shevchenko and Mika Westerberg for the GPIO aspects.

Thanks!
Linus Walleij

author
#2 lore raw

Hi Linus,

Quoted 9 lines · Linus Walleij

Hi Wei,

thanks for your patch!

This doesn't apply for my "devel" branch, can you rebase on this:
https://git.kernel.org/pub/scm/linux/kernel/git/linusw/linux-gpio.git/log/?h=devel

We have moved some ACPI headers around recently.

Thanks to review!
I just sent out the v2 based on that.

Quoted 12 lines · Wei Xu
Quoted 3 lines

Invoke acpi_gpiochip_request_interrupts after the acpi data has been attached to the pl061 acpi node to register interruption.

Makes sense.

Quoted 1 line
Fixes 04ce935c6b2a ("gpio: pl061: Pass irqchip when adding gpiochip")

I doubt this is a regression since I haven't seen anyone use this gpiochip with ACPI before.

Please rename the patch "gpio: pl061: Add ACPI support" unless you can convince me it worked without changes before.

In the v2, I attached the log on QEMU v3.0.0 and Linux kernel v5.2.0-rc7 and the pl061 driver can register ACPI interruption.
Based on that, I did not rename the patch but I am OK to rename it if you have any doubt.

Quoted 9 lines

Please include some ACPI people on review of this. From MAINTAINERS:
ACPI

M:      "Rafael J. Wysocki" <[email protected]>
M:      Len Brown <[email protected]>
L:      [email protected]

I would also include Andy Shevchenko and Mika Westerberg for the GPIO aspects.

Copied to all of them in the v2.
Thanks!

Best Regards,
Wei

Quoted 5 lines

Thanks!
Linus Walleij

.

#3 lore raw
In reply to Wei Xu
Quoted 35 lines · Wei Xu

Invoke acpi_gpiochip_request_interrupts after the acpi data has been attached to the pl061 acpi node to register interruption.

Otherwise it will be failed to register interruption for the ACPI case. Because in the gpiochip_add_data_with_key, acpi_gpiochip_add is invoked after gpiochip_add_irqchip but at that time the acpi data has not been attached yet.

Tested with below steps on QEMU v4.1.0-rc3 and Linux kernel v5.3-rc4, and found pl061 interruption is missed in the /proc/interrupts:
1.
qemu-system-aarch64 \

(quoted diff) +0 -6
-machine virt,gic-version=3 -cpu cortex-a57 \
-m 1G,maxmem=4G,slots=4 \
-kernel Image -initrd rootfs.cpio.gz \
-net none -nographic  \
-bios QEMU_EFI.fd  \
-append "console=ttyAMA0 acpi=force earlycon=pl011,0x9000000"

2. cat /proc/interrupts in the guest console:

estuary:/$ cat /proc/interrupts
CPU0
2:       3228     GICv3  27 Level     arch_timer
4:         15     GICv3  33 Level     uart-pl011
42:          0     GICv3  23 Level     arm-pmu
IPI0:         0       Rescheduling interrupts
IPI1:         0       Function call interrupts
IPI2:         0       CPU stop interrupts
IPI3:         0       CPU stop (for crash dump) interrupts
IPI4:         0       Timer broadcast interrupts
IPI5:         0       IRQ work interrupts
IPI6:         0       CPU wake-up interrupts
Err:          0
Quoted 1 line
Fixes 04ce935c6b2a ("gpio: pl061: Pass irqchip when adding gpiochip")

Linus, I'm wondering if we can do this for all inside the GPIO library. Thoughts?

--
With Best Regards,
Andy Shevchenko

#4 lore raw
Quoted 4 lines · Andy Shevchenko

On Mon, Aug 12, 2019 at 2:30 PM Wei Xu <[email protected]> wrote:

Linus, I'm wondering if we can do this for all inside the GPIO library. Thoughts?

If it's supposed to happen exactly the same way on all ACPI-enabled gpiochips, I think it is more or less mandatory :D

We need to be sure we're not gonna have to quirks for misc variants of GPIO controllers down in the gpiolib-acpi.c as a result because some ACPI-thing is "standard not quite standard" though.

Yours,
Linus Walleij