diff mbox series

rtc: rtc-cmos: Fix wake alarm breakage

Message ID 5887691.lOV4Wx5bFT@kreacher
State New
Headers show
Series rtc: rtc-cmos: Fix wake alarm breakage | expand

Commit Message

Rafael J. Wysocki Oct. 18, 2022, 4:09 p.m. UTC
From: Rafael J. Wysocki <rafael.j.wysocki@intel.com>

Commit 4919d3eb2ec0 ("rtc: cmos: Fix event handler registration
ordering issue") overlooked the fact that cmos_do_probe() depended
on the preparations carried out by cmos_wake_setup() and the wake
alarm stopped working after the ordering of them had been changed.

Address this by partially reverting commit 4919d3eb2ec0 so that
cmos_wake_setup() is called before cmos_do_probe() again and moving
the rtc_wake_setup() invocation from cmos_wake_setup() directly to the
callers of cmos_do_probe() where it will happen after a successful
completion of the latter.

Fixes: 4919d3eb2ec0 ("rtc: cmos: Fix event handler registration ordering issue")
Reported-by: Zhang Rui <rui.zhang@intel.com>
Reported-by: Todd Brandt <todd.e.brandt@linux.intel.com>
Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
---

@Bjorn: This is the minimum fix.  Folding cmos_wake_setup() into cmos_do_probe()
requires changes that are a bit intrusive for post-rc1, but I will do that later.

---
 drivers/rtc/rtc-cmos.c |   11 ++++++++---
 1 file changed, 8 insertions(+), 3 deletions(-)

Comments

Alexandre Belloni Oct. 19, 2022, 6:16 p.m. UTC | #1
On 19/10/2022 18:13:43+0200, Rafael J. Wysocki wrote:
> On Tue, Oct 18, 2022 at 6:39 PM Alexandre Belloni
> <alexandre.belloni@bootlin.com> wrote:
> >
> > On Tue, 18 Oct 2022 18:09:31 +0200, Rafael J. Wysocki wrote:
> > > From: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
> > >
> > > Commit 4919d3eb2ec0 ("rtc: cmos: Fix event handler registration
> > > ordering issue") overlooked the fact that cmos_do_probe() depended
> > > on the preparations carried out by cmos_wake_setup() and the wake
> > > alarm stopped working after the ordering of them had been changed.
> > >
> > > [...]
> >
> > Applied, thanks!
> >
> > [1/1] rtc: rtc-cmos: Fix wake alarm breakage
> >       commit: 0782b66ed2fbb035dda76111df0954515e417b24
> 
> Thank you!
> 
> However, there is a build fix on top of this which has just been posted:
> 
> https://lore.kernel.org/linux-acpi/2677035.mvXUDI8C0e@kreacher/
> 
> Sorry about breaking it again.

I had that in rtc-fixes:

https://git.kernel.org/pub/scm/linux/kernel/git/abelloni/linux.git/commit/?h=rtc-fixes
Rafael J. Wysocki Oct. 19, 2022, 6:20 p.m. UTC | #2
On Wed, Oct 19, 2022 at 8:16 PM Alexandre Belloni
<alexandre.belloni@bootlin.com> wrote:
>
> On 19/10/2022 18:13:43+0200, Rafael J. Wysocki wrote:
> > On Tue, Oct 18, 2022 at 6:39 PM Alexandre Belloni
> > <alexandre.belloni@bootlin.com> wrote:
> > >
> > > On Tue, 18 Oct 2022 18:09:31 +0200, Rafael J. Wysocki wrote:
> > > > From: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
> > > >
> > > > Commit 4919d3eb2ec0 ("rtc: cmos: Fix event handler registration
> > > > ordering issue") overlooked the fact that cmos_do_probe() depended
> > > > on the preparations carried out by cmos_wake_setup() and the wake
> > > > alarm stopped working after the ordering of them had been changed.
> > > >
> > > > [...]
> > >
> > > Applied, thanks!
> > >
> > > [1/1] rtc: rtc-cmos: Fix wake alarm breakage
> > >       commit: 0782b66ed2fbb035dda76111df0954515e417b24
> >
> > Thank you!
> >
> > However, there is a build fix on top of this which has just been posted:
> >
> > https://lore.kernel.org/linux-acpi/2677035.mvXUDI8C0e@kreacher/
> >
> > Sorry about breaking it again.
>
> I had that in rtc-fixes:
>
> https://git.kernel.org/pub/scm/linux/kernel/git/abelloni/linux.git/commit/?h=rtc-fixes

Looks good, thanks!
Mel Gorman Oct. 20, 2022, 8:10 a.m. UTC | #3
On Tue, Oct 18, 2022 at 06:09:31PM +0200, Rafael J. Wysocki wrote:
> From: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
> 
> Commit 4919d3eb2ec0 ("rtc: cmos: Fix event handler registration
> ordering issue") overlooked the fact that cmos_do_probe() depended
> on the preparations carried out by cmos_wake_setup() and the wake
> alarm stopped working after the ordering of them had been changed.
> 
> Address this by partially reverting commit 4919d3eb2ec0 so that
> cmos_wake_setup() is called before cmos_do_probe() again and moving
> the rtc_wake_setup() invocation from cmos_wake_setup() directly to the
> callers of cmos_do_probe() where it will happen after a successful
> completion of the latter.
> 
> Fixes: 4919d3eb2ec0 ("rtc: cmos: Fix event handler registration ordering issue")
> Reported-by: Zhang Rui <rui.zhang@intel.com>
> Reported-by: Todd Brandt <todd.e.brandt@linux.intel.com>
> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>

Boot test that previously hit NULL pointer exceptions also completed successfully.
Todd Brandt Oct. 25, 2022, 3:28 p.m. UTC | #4
On Tue, 2022-10-18 at 18:38 +0200, Alexandre Belloni wrote:
> On Tue, 18 Oct 2022 18:09:31 +0200, Rafael J. Wysocki wrote:
> > From: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
> > 
> > Commit 4919d3eb2ec0 ("rtc: cmos: Fix event handler registration
> > ordering issue") overlooked the fact that cmos_do_probe() depended
> > on the preparations carried out by cmos_wake_setup() and the wake
> > alarm stopped working after the ordering of them had been changed.
> > 
> > [...]
> 
> Applied, thanks!

I did testing yesterday on the 6.1.0-rc2 build and this patch hasn't
made it into rc2. This is an extreme inconvenience to anyone testing
low power modes as the rtc wakealarm doesn't function. I'm a little
surprised more people haven't complained.

Please get this in 6.1.0-rc3.

> [1/1] rtc: rtc-cmos: Fix wake alarm breakage
>       commit: 0782b66ed2fbb035dda76111df0954515e417b24
> 
> Best regards,
>
diff mbox series

Patch

Index: linux-pm/drivers/rtc/rtc-cmos.c
===================================================================
--- linux-pm.orig/drivers/rtc/rtc-cmos.c
+++ linux-pm/drivers/rtc/rtc-cmos.c
@@ -1233,6 +1233,9 @@  static u32 rtc_handler(void *context)
 
 static inline void rtc_wake_setup(struct device *dev)
 {
+	if (acpi_disabled)
+		return;
+
 	acpi_install_fixed_event_handler(ACPI_EVENT_RTC, rtc_handler, dev);
 	/*
 	 * After the RTC handler is installed, the Fixed_RTC event should
@@ -1286,7 +1289,6 @@  static void cmos_wake_setup(struct devic
 
 	use_acpi_alarm_quirks();
 
-	rtc_wake_setup(dev);
 	acpi_rtc_info.wake_on = rtc_wake_on;
 	acpi_rtc_info.wake_off = rtc_wake_off;
 
@@ -1354,6 +1356,8 @@  static int cmos_pnp_probe(struct pnp_dev
 {
 	int irq, ret;
 
+	cmos_wake_setup(&pnp->dev);
+
 	if (pnp_port_start(pnp, 0) == 0x70 && !pnp_irq_valid(pnp, 0)) {
 		irq = 0;
 #ifdef CONFIG_X86
@@ -1372,7 +1376,7 @@  static int cmos_pnp_probe(struct pnp_dev
 	if (ret)
 		return ret;
 
-	cmos_wake_setup(&pnp->dev);
+	rtc_wake_setup(&pnp->dev);
 
 	return 0;
 }
@@ -1461,6 +1465,7 @@  static int __init cmos_platform_probe(st
 	int irq, ret;
 
 	cmos_of_init(pdev);
+	cmos_wake_setup(&pdev->dev);
 
 	if (RTC_IOMAPPED)
 		resource = platform_get_resource(pdev, IORESOURCE_IO, 0);
@@ -1474,7 +1479,7 @@  static int __init cmos_platform_probe(st
 	if (ret)
 		return ret;
 
-	cmos_wake_setup(&pdev->dev);
+	rtc_wake_setup(&pdev->dev);
 
 	return 0;
 }