| Message ID | 20200403154834.303105-2-hdegoede@redhat.com (mailing list archive) |
|---|---|
| State | Mainlined, archived |
| Headers | show |
| Series | [v3,1/2] ACPI: PM: Add acpi_[un]register_wakeup_handler() | expand |
On Friday, April 3, 2020 5:48:34 PM CEST Hans de Goede wrote: > The Power Management Events (PMEs) the INT0002 driver listens for get > signalled by the Power Management Controller (PMC) using the same IRQ > as used for the ACPI SCI. > > Since commit fdde0ff8590b ("ACPI: PM: s2idle: Prevent spurious SCIs from > waking up the system") the SCI triggering, without there being a wakeup > cause recognized by the ACPI sleep code, will no longer wakeup the system. > > This breaks PMEs / wakeups signalled to the INT0002 driver, the system > never leaves the s2idle_loop() now. > > Use acpi_register_wakeup_handler() to register a function which checks > the GPE0a_STS register for a PME and trigger a wakeup when a PME has > been signalled. > > Fixes: fdde0ff8590b ("ACPI: PM: s2idle: Prevent spurious SCIs from waking up the system") > Cc: 5.4+ <stable@vger.kernel.org> # 5.4+ > Signed-off-by: Hans de Goede <hdegoede@redhat.com> Andy, any objections? > --- > Changes in v3: > - Keep the pm_wakeup_hard_event() call > > Changes in v2: > - Adjust for the wakeup-handler registration function being renamed to > acpi_register_wakeup_handler() > --- > drivers/platform/x86/intel_int0002_vgpio.c | 10 ++++++++++ > 1 file changed, 10 insertions(+) > > diff --git a/drivers/platform/x86/intel_int0002_vgpio.c b/drivers/platform/x86/intel_int0002_vgpio.c > index f14e2c5f9da5..55f088f535e2 100644 > --- a/drivers/platform/x86/intel_int0002_vgpio.c > +++ b/drivers/platform/x86/intel_int0002_vgpio.c > @@ -127,6 +127,14 @@ static irqreturn_t int0002_irq(int irq, void *data) > return IRQ_HANDLED; > } > > +static bool int0002_check_wake(void *data) > +{ > + u32 gpe_sts_reg; > + > + gpe_sts_reg = inl(GPE0A_STS_PORT); > + return (gpe_sts_reg & GPE0A_PME_B0_STS_BIT); > +} > + > static struct irq_chip int0002_byt_irqchip = { > .name = DRV_NAME, > .irq_ack = int0002_irq_ack, > @@ -220,6 +228,7 @@ static int int0002_probe(struct platform_device *pdev) > return ret; > } > > + acpi_register_wakeup_handler(irq, int0002_check_wake, NULL); > device_init_wakeup(dev, true); > return 0; > } > @@ -227,6 +236,7 @@ static int int0002_probe(struct platform_device *pdev) > static int int0002_remove(struct platform_device *pdev) > { > device_init_wakeup(&pdev->dev, false); > + acpi_unregister_wakeup_handler(int0002_check_wake, NULL); > return 0; > } > >
On Fri, Apr 3, 2020 at 7:08 PM Rafael J. Wysocki <rjw@rjwysocki.net> wrote: > > On Friday, April 3, 2020 5:48:34 PM CEST Hans de Goede wrote: > > The Power Management Events (PMEs) the INT0002 driver listens for get > > signalled by the Power Management Controller (PMC) using the same IRQ > > as used for the ACPI SCI. > > > > Since commit fdde0ff8590b ("ACPI: PM: s2idle: Prevent spurious SCIs from > > waking up the system") the SCI triggering, without there being a wakeup > > cause recognized by the ACPI sleep code, will no longer wakeup the system. > > > > This breaks PMEs / wakeups signalled to the INT0002 driver, the system > > never leaves the s2idle_loop() now. > > > > Use acpi_register_wakeup_handler() to register a function which checks > > the GPE0a_STS register for a PME and trigger a wakeup when a PME has > > been signalled. > > > > Fixes: fdde0ff8590b ("ACPI: PM: s2idle: Prevent spurious SCIs from waking up the system") > > Cc: 5.4+ <stable@vger.kernel.org> # 5.4+ > > Signed-off-by: Hans de Goede <hdegoede@redhat.com> > > Andy, any objections? No, Acked-by: Andy Shevchenko <andy.shevchenko@gmail.com> Hans, just today when testing some other stuff noticed this [ 49.279001] irq 9: nobody cared (try booting with the "irqpoll" option) [ 49.289176] CPU: 0 PID: 168 Comm: irq/123-ATML100 Not tainted 5.6.0-next-20200403+ #212 [ 49.300915] Hardware name: Intel Corporation CHERRYVIEW D0 PLATFORM/Braswell CRB, BIOS BRAS.X64.B082.R00.150727 0557 07/27/2015 [ 49.316520] Call Trace: [ 49.322093] <IRQ> [ 49.327193] dump_stack+0x50/0x70 [ 49.333744] __report_bad_irq+0x30/0xa2 [ 49.340858] note_interrupt.cold+0xb/0x62 ... [ 49.685087] handlers: [ 49.690307] [<000000000ab3cf88>] acpi_irq [ 49.697463] [<00000000e5d78029>] int0002_irq [intel_int0002_vgpio] [ 49.707063] Disabling IRQ #9 Is this what your series fixes? > > > --- > > Changes in v3: > > - Keep the pm_wakeup_hard_event() call > > > > Changes in v2: > > - Adjust for the wakeup-handler registration function being renamed to > > acpi_register_wakeup_handler() > > --- > > drivers/platform/x86/intel_int0002_vgpio.c | 10 ++++++++++ > > 1 file changed, 10 insertions(+) > > > > diff --git a/drivers/platform/x86/intel_int0002_vgpio.c b/drivers/platform/x86/intel_int0002_vgpio.c > > index f14e2c5f9da5..55f088f535e2 100644 > > --- a/drivers/platform/x86/intel_int0002_vgpio.c > > +++ b/drivers/platform/x86/intel_int0002_vgpio.c > > @@ -127,6 +127,14 @@ static irqreturn_t int0002_irq(int irq, void *data) > > return IRQ_HANDLED; > > } > > > > +static bool int0002_check_wake(void *data) > > +{ > > + u32 gpe_sts_reg; > > + > > + gpe_sts_reg = inl(GPE0A_STS_PORT); > > + return (gpe_sts_reg & GPE0A_PME_B0_STS_BIT); > > +} > > + > > static struct irq_chip int0002_byt_irqchip = { > > .name = DRV_NAME, > > .irq_ack = int0002_irq_ack, > > @@ -220,6 +228,7 @@ static int int0002_probe(struct platform_device *pdev) > > return ret; > > } > > > > + acpi_register_wakeup_handler(irq, int0002_check_wake, NULL); > > device_init_wakeup(dev, true); > > return 0; > > } > > @@ -227,6 +236,7 @@ static int int0002_probe(struct platform_device *pdev) > > static int int0002_remove(struct platform_device *pdev) > > { > > device_init_wakeup(&pdev->dev, false); > > + acpi_unregister_wakeup_handler(int0002_check_wake, NULL); > > return 0; > > } > > > > > > > >
Hi, On 4/3/20 6:23 PM, Andy Shevchenko wrote: > On Fri, Apr 3, 2020 at 7:08 PM Rafael J. Wysocki <rjw@rjwysocki.net> wrote: >> >> On Friday, April 3, 2020 5:48:34 PM CEST Hans de Goede wrote: >>> The Power Management Events (PMEs) the INT0002 driver listens for get >>> signalled by the Power Management Controller (PMC) using the same IRQ >>> as used for the ACPI SCI. >>> >>> Since commit fdde0ff8590b ("ACPI: PM: s2idle: Prevent spurious SCIs from >>> waking up the system") the SCI triggering, without there being a wakeup >>> cause recognized by the ACPI sleep code, will no longer wakeup the system. >>> >>> This breaks PMEs / wakeups signalled to the INT0002 driver, the system >>> never leaves the s2idle_loop() now. >>> >>> Use acpi_register_wakeup_handler() to register a function which checks >>> the GPE0a_STS register for a PME and trigger a wakeup when a PME has >>> been signalled. >>> >>> Fixes: fdde0ff8590b ("ACPI: PM: s2idle: Prevent spurious SCIs from waking up the system") >>> Cc: 5.4+ <stable@vger.kernel.org> # 5.4+ >>> Signed-off-by: Hans de Goede <hdegoede@redhat.com> >> >> Andy, any objections? > > No, > Acked-by: Andy Shevchenko <andy.shevchenko@gmail.com> > > Hans, just today when testing some other stuff noticed this > > [ 49.279001] irq 9: nobody cared (try booting with the "irqpoll" option) > [ 49.289176] CPU: 0 PID: 168 Comm: irq/123-ATML100 Not tainted > 5.6.0-next-20200403+ #212 > [ 49.300915] Hardware name: Intel Corporation CHERRYVIEW D0 > PLATFORM/Braswell CRB, BIOS BRAS.X64.B082.R00.150727 0557 07/27/2015 > [ 49.316520] Call Trace: > [ 49.322093] <IRQ> > [ 49.327193] dump_stack+0x50/0x70 > [ 49.333744] __report_bad_irq+0x30/0xa2 > [ 49.340858] note_interrupt.cold+0xb/0x62 > ... > [ 49.685087] handlers: > [ 49.690307] [<000000000ab3cf88>] acpi_irq > [ 49.697463] [<00000000e5d78029>] int0002_irq [intel_int0002_vgpio] > [ 49.707063] Disabling IRQ #9 > > Is this what your series fixes? I don't think that my series fixes this. My series fixes CHT devices no longer waking up from s2idle when woken by pressing a key on the USB attached keyboard dock of various 2-in-1s. Before we were relying on the ACPI code also waking up on IRQs which from the ACPI SCI handler pov where spurious. Rafael fixed the ACPI SCI handling during s2idle also waking the system on spurious events. My patch hooks the INT0002 driver into the ACPI SCI handling / wakeup checks during s2idle. I guess Rafael may have made some related changes elsewhere though where the SCI no longer returns IRQ_HANDLED in some cases? Rafael? Regards, Hans > >> >>> --- >>> Changes in v3: >>> - Keep the pm_wakeup_hard_event() call >>> >>> Changes in v2: >>> - Adjust for the wakeup-handler registration function being renamed to >>> acpi_register_wakeup_handler() >>> --- >>> drivers/platform/x86/intel_int0002_vgpio.c | 10 ++++++++++ >>> 1 file changed, 10 insertions(+) >>> >>> diff --git a/drivers/platform/x86/intel_int0002_vgpio.c b/drivers/platform/x86/intel_int0002_vgpio.c >>> index f14e2c5f9da5..55f088f535e2 100644 >>> --- a/drivers/platform/x86/intel_int0002_vgpio.c >>> +++ b/drivers/platform/x86/intel_int0002_vgpio.c >>> @@ -127,6 +127,14 @@ static irqreturn_t int0002_irq(int irq, void *data) >>> return IRQ_HANDLED; >>> } >>> >>> +static bool int0002_check_wake(void *data) >>> +{ >>> + u32 gpe_sts_reg; >>> + >>> + gpe_sts_reg = inl(GPE0A_STS_PORT); >>> + return (gpe_sts_reg & GPE0A_PME_B0_STS_BIT); >>> +} >>> + >>> static struct irq_chip int0002_byt_irqchip = { >>> .name = DRV_NAME, >>> .irq_ack = int0002_irq_ack, >>> @@ -220,6 +228,7 @@ static int int0002_probe(struct platform_device *pdev) >>> return ret; >>> } >>> >>> + acpi_register_wakeup_handler(irq, int0002_check_wake, NULL); >>> device_init_wakeup(dev, true); >>> return 0; >>> } >>> @@ -227,6 +236,7 @@ static int int0002_probe(struct platform_device *pdev) >>> static int int0002_remove(struct platform_device *pdev) >>> { >>> device_init_wakeup(&pdev->dev, false); >>> + acpi_unregister_wakeup_handler(int0002_check_wake, NULL); >>> return 0; >>> } >>> >>> >> >> >> >> > >
diff --git a/drivers/platform/x86/intel_int0002_vgpio.c b/drivers/platform/x86/intel_int0002_vgpio.c index f14e2c5f9da5..55f088f535e2 100644 --- a/drivers/platform/x86/intel_int0002_vgpio.c +++ b/drivers/platform/x86/intel_int0002_vgpio.c @@ -127,6 +127,14 @@ static irqreturn_t int0002_irq(int irq, void *data) return IRQ_HANDLED; } +static bool int0002_check_wake(void *data) +{ + u32 gpe_sts_reg; + + gpe_sts_reg = inl(GPE0A_STS_PORT); + return (gpe_sts_reg & GPE0A_PME_B0_STS_BIT); +} + static struct irq_chip int0002_byt_irqchip = { .name = DRV_NAME, .irq_ack = int0002_irq_ack, @@ -220,6 +228,7 @@ static int int0002_probe(struct platform_device *pdev) return ret; } + acpi_register_wakeup_handler(irq, int0002_check_wake, NULL); device_init_wakeup(dev, true); return 0; } @@ -227,6 +236,7 @@ static int int0002_probe(struct platform_device *pdev) static int int0002_remove(struct platform_device *pdev) { device_init_wakeup(&pdev->dev, false); + acpi_unregister_wakeup_handler(int0002_check_wake, NULL); return 0; }
The Power Management Events (PMEs) the INT0002 driver listens for get signalled by the Power Management Controller (PMC) using the same IRQ as used for the ACPI SCI. Since commit fdde0ff8590b ("ACPI: PM: s2idle: Prevent spurious SCIs from waking up the system") the SCI triggering, without there being a wakeup cause recognized by the ACPI sleep code, will no longer wakeup the system. This breaks PMEs / wakeups signalled to the INT0002 driver, the system never leaves the s2idle_loop() now. Use acpi_register_wakeup_handler() to register a function which checks the GPE0a_STS register for a PME and trigger a wakeup when a PME has been signalled. Fixes: fdde0ff8590b ("ACPI: PM: s2idle: Prevent spurious SCIs from waking up the system") Cc: 5.4+ <stable@vger.kernel.org> # 5.4+ Signed-off-by: Hans de Goede <hdegoede@redhat.com> --- Changes in v3: - Keep the pm_wakeup_hard_event() call Changes in v2: - Adjust for the wakeup-handler registration function being renamed to acpi_register_wakeup_handler() --- drivers/platform/x86/intel_int0002_vgpio.c | 10 ++++++++++ 1 file changed, 10 insertions(+)