| Message ID | 12541194.33112yYDjW@aspire.rjw.lan (mailing list archive) |
|---|---|
| State | Superseded, archived |
| Delegated to: | Andy Shevchenko |
| Headers | show |
On Wed, 2018-01-10 at 13:26 +0100, Rafael J. Wysocki wrote: > From: Rafael J. Wysocki <rafael.j.wysocki@intel.com> > > Some systems don't support the ACPI_LPS0_ENTRY and ACPI_LPS0_EXIT > functions in their Low Power S0 Idle _DSM, but still expect EC > events to be processed in the suspend-to-idle state for power button > wakeup (among other things) to work. Surface Pro3 turns out to be > one of them. > > Fortunately, it still provides Low Power S0 Idle _DSM with the screen > on/off functions supported, so modify the ACPI suspend-to-idle to use > the Low Power S0 Idle code path for all systems supporting the > ACPI_LPS0_ENTRY and ACPI_LPS0_EXIT or the ACPI_LPS0_SCREEN_OFF and > ACPI_LPS0_SCREEN_ON functions in their Low Power S0 Idle _DSM. > > Potentially, that will cause more systems to use suspend-to-idle by > default, so some future corrections may be necessary if it leads > to issues, but let it remain more straightforward for now. > -#define ACPI_S2IDLE_FUNC_MASK ((1 << ACPI_LPS0_ENTRY) | (1 << > ACPI_LPS0_EXIT)) > +#define ACPI_LPS0_SCREEN_MASK ((1 << ACPI_LPS0_SCREEN_OFF) | > (1 << ACPI_LPS0_SCREEN_ON)) > +#define ACPI_LPS0_S2I_MASK ((1 << ACPI_LPS0_ENTRY) | (1 << > ACPI_LPS0_EXIT)) Just a nitpick: Can we leave S2IDLE instead of S2I? Would it make sense for potential code readers?
> -----Original Message----- > From: platform-driver-x86-owner@vger.kernel.org [mailto:platform-driver-x86- > owner@vger.kernel.org] On Behalf Of Rafael J. Wysocki > Sent: Wednesday, January 10, 2018 6:26 AM > To: Linux ACPI <linux-acpi@vger.kernel.org> > Cc: Andy Shevchenko <andriy.shevchenko@linux.intel.com>; Darren Hart > <dvhart@infradead.org>; LKML <linux-kernel@vger.kernel.org>; Linux PM <linux- > pm@vger.kernel.org>; Platform Driver <platform-driver-x86@vger.kernel.org>; > Valentin Manea <valy@mrs.ro> > Subject: [PATCH 1/2] ACPI / PM: Use Low Power S0 Idle on more systems > > From: Rafael J. Wysocki <rafael.j.wysocki@intel.com> > > Some systems don't support the ACPI_LPS0_ENTRY and ACPI_LPS0_EXIT > functions in their Low Power S0 Idle _DSM, but still expect EC > events to be processed in the suspend-to-idle state for power button > wakeup (among other things) to work. Surface Pro3 turns out to be > one of them. > > Fortunately, it still provides Low Power S0 Idle _DSM with the screen > on/off functions supported, so modify the ACPI suspend-to-idle to use > the Low Power S0 Idle code path for all systems supporting the > ACPI_LPS0_ENTRY and ACPI_LPS0_EXIT or the ACPI_LPS0_SCREEN_OFF and > ACPI_LPS0_SCREEN_ON functions in their Low Power S0 Idle _DSM. > > Potentially, that will cause more systems to use suspend-to-idle by > default, so some future corrections may be necessary if it leads > to issues, but let it remain more straightforward for now. > > Link: https://bugzilla.kernel.org/show_bug.cgi?id=198389 > Reported-by: Valentin Manea <valy@mrs.ro> > Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com> > --- > drivers/acpi/sleep.c | 6 ++++-- > 1 file changed, 4 insertions(+), 2 deletions(-) > > Index: linux-pm/drivers/acpi/sleep.c > =================================================================== > --- linux-pm.orig/drivers/acpi/sleep.c > +++ linux-pm/drivers/acpi/sleep.c > @@ -707,7 +707,8 @@ static const struct acpi_device_id lps0_ > #define ACPI_LPS0_ENTRY 5 > #define ACPI_LPS0_EXIT 6 > > -#define ACPI_S2IDLE_FUNC_MASK ((1 << ACPI_LPS0_ENTRY) | (1 << > ACPI_LPS0_EXIT)) > +#define ACPI_LPS0_SCREEN_MASK ((1 << ACPI_LPS0_SCREEN_OFF) | (1 << > ACPI_LPS0_SCREEN_ON)) > +#define ACPI_LPS0_S2I_MASK ((1 << ACPI_LPS0_ENTRY) | (1 << ACPI_LPS0_EXIT)) > > static acpi_handle lps0_device_handle; > static guid_t lps0_dsm_guid; > @@ -910,7 +911,8 @@ static int lps0_device_attach(struct acp > if (out_obj && out_obj->type == ACPI_TYPE_BUFFER) { > char bitmask = *(char *)out_obj->buffer.pointer; > > - if ((bitmask & ACPI_S2IDLE_FUNC_MASK) == > ACPI_S2IDLE_FUNC_MASK) { > + if ((bitmask & ACPI_LPS0_S2I_MASK) == ACPI_LPS0_S2I_MASK || > + (bitmask & ACPI_LPS0_SCREEN_MASK) == > ACPI_LPS0_SCREEN_MASK) { > lps0_dsm_func_mask = bitmask; > lps0_device_handle = adev->handle; > /* In making this change I believe you'll need to cache the values that you found from the function mask to test them later too. Here: https://github.com/torvalds/linux/blob/master/drivers/acpi/sleep.c#L943 This is because later on both ACPI_LPS0_SCREEN_OFF and ACPI_LPS0_ENTRY are called whether or not they both exist. Fortunately looking at the DSDT in the attached bug nothing happens if calling the undefined Arg2 == 0x05/Arg2 == 0x06 but that might not always be the case.
On Wed, Jan 10, 2018 at 6:38 PM, <Mario.Limonciello@dell.com> wrote: > > >> -----Original Message----- >> From: platform-driver-x86-owner@vger.kernel.org [mailto:platform-driver-x86- >> owner@vger.kernel.org] On Behalf Of Rafael J. Wysocki >> Sent: Wednesday, January 10, 2018 6:26 AM >> To: Linux ACPI <linux-acpi@vger.kernel.org> >> Cc: Andy Shevchenko <andriy.shevchenko@linux.intel.com>; Darren Hart >> <dvhart@infradead.org>; LKML <linux-kernel@vger.kernel.org>; Linux PM <linux- >> pm@vger.kernel.org>; Platform Driver <platform-driver-x86@vger.kernel.org>; >> Valentin Manea <valy@mrs.ro> >> Subject: [PATCH 1/2] ACPI / PM: Use Low Power S0 Idle on more systems >> >> From: Rafael J. Wysocki <rafael.j.wysocki@intel.com> >> >> Some systems don't support the ACPI_LPS0_ENTRY and ACPI_LPS0_EXIT >> functions in their Low Power S0 Idle _DSM, but still expect EC >> events to be processed in the suspend-to-idle state for power button >> wakeup (among other things) to work. Surface Pro3 turns out to be >> one of them. >> >> Fortunately, it still provides Low Power S0 Idle _DSM with the screen >> on/off functions supported, so modify the ACPI suspend-to-idle to use >> the Low Power S0 Idle code path for all systems supporting the >> ACPI_LPS0_ENTRY and ACPI_LPS0_EXIT or the ACPI_LPS0_SCREEN_OFF and >> ACPI_LPS0_SCREEN_ON functions in their Low Power S0 Idle _DSM. >> >> Potentially, that will cause more systems to use suspend-to-idle by >> default, so some future corrections may be necessary if it leads >> to issues, but let it remain more straightforward for now. >> >> Link: https://bugzilla.kernel.org/show_bug.cgi?id=198389 >> Reported-by: Valentin Manea <valy@mrs.ro> >> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com> >> --- >> drivers/acpi/sleep.c | 6 ++++-- >> 1 file changed, 4 insertions(+), 2 deletions(-) >> >> Index: linux-pm/drivers/acpi/sleep.c >> =================================================================== >> --- linux-pm.orig/drivers/acpi/sleep.c >> +++ linux-pm/drivers/acpi/sleep.c >> @@ -707,7 +707,8 @@ static const struct acpi_device_id lps0_ >> #define ACPI_LPS0_ENTRY 5 >> #define ACPI_LPS0_EXIT 6 >> >> -#define ACPI_S2IDLE_FUNC_MASK ((1 << ACPI_LPS0_ENTRY) | (1 << >> ACPI_LPS0_EXIT)) >> +#define ACPI_LPS0_SCREEN_MASK ((1 << ACPI_LPS0_SCREEN_OFF) | (1 << >> ACPI_LPS0_SCREEN_ON)) >> +#define ACPI_LPS0_S2I_MASK ((1 << ACPI_LPS0_ENTRY) | (1 << ACPI_LPS0_EXIT)) >> >> static acpi_handle lps0_device_handle; >> static guid_t lps0_dsm_guid; >> @@ -910,7 +911,8 @@ static int lps0_device_attach(struct acp >> if (out_obj && out_obj->type == ACPI_TYPE_BUFFER) { >> char bitmask = *(char *)out_obj->buffer.pointer; >> >> - if ((bitmask & ACPI_S2IDLE_FUNC_MASK) == >> ACPI_S2IDLE_FUNC_MASK) { >> + if ((bitmask & ACPI_LPS0_S2I_MASK) == ACPI_LPS0_S2I_MASK || >> + (bitmask & ACPI_LPS0_SCREEN_MASK) == >> ACPI_LPS0_SCREEN_MASK) { >> lps0_dsm_func_mask = bitmask; >> lps0_device_handle = adev->handle; >> /* > > In making this change I believe you'll need to cache the values that you found from the > function mask to test them later too. But that's what lps0_dsm_func_mask is for if I understand you correctly. > Here: > https://github.com/torvalds/linux/blob/master/drivers/acpi/sleep.c#L943 > > This is because later on both ACPI_LPS0_SCREEN_OFF and ACPI_LPS0_ENTRY are called > whether or not they both exist. No, that's not the case. acpi_sleep_run_lps0_dsm() checks if the given function is there in the mask returned by function 0 and it doesn't evaluate the _DSM otherwise. Thanks, Rafael
On Wed, Jan 10, 2018 at 2:24 PM, Andy Shevchenko <andriy.shevchenko@linux.intel.com> wrote: > On Wed, 2018-01-10 at 13:26 +0100, Rafael J. Wysocki wrote: >> From: Rafael J. Wysocki <rafael.j.wysocki@intel.com> >> >> Some systems don't support the ACPI_LPS0_ENTRY and ACPI_LPS0_EXIT >> functions in their Low Power S0 Idle _DSM, but still expect EC >> events to be processed in the suspend-to-idle state for power button >> wakeup (among other things) to work. Surface Pro3 turns out to be >> one of them. >> >> Fortunately, it still provides Low Power S0 Idle _DSM with the screen >> on/off functions supported, so modify the ACPI suspend-to-idle to use >> the Low Power S0 Idle code path for all systems supporting the >> ACPI_LPS0_ENTRY and ACPI_LPS0_EXIT or the ACPI_LPS0_SCREEN_OFF and >> ACPI_LPS0_SCREEN_ON functions in their Low Power S0 Idle _DSM. >> >> Potentially, that will cause more systems to use suspend-to-idle by >> default, so some future corrections may be necessary if it leads >> to issues, but let it remain more straightforward for now. > >> -#define ACPI_S2IDLE_FUNC_MASK ((1 << ACPI_LPS0_ENTRY) | (1 << >> ACPI_LPS0_EXIT)) >> +#define ACPI_LPS0_SCREEN_MASK ((1 << ACPI_LPS0_SCREEN_OFF) | >> (1 << ACPI_LPS0_SCREEN_ON)) > >> +#define ACPI_LPS0_S2I_MASK ((1 << ACPI_LPS0_ENTRY) | (1 << >> ACPI_LPS0_EXIT)) > > Just a nitpick: Can we leave S2IDLE instead of S2I? > Would it make sense for potential code readers? I wanted it to be shorter, but if that is a problem, I'd rather call it PLATFORM than S2IDLE (as technically they are related to the low-power mode of the platform). I'll send an update shortly.
On Wed, 2018-01-10 at 23:25 +0100, Rafael J. Wysocki wrote: > On Wed, Jan 10, 2018 at 2:24 PM, Andy Shevchenko > <andriy.shevchenko@linux.intel.com> wrote: > > On Wed, 2018-01-10 at 13:26 +0100, Rafael J. Wysocki wrote: > > Just a nitpick: Can we leave S2IDLE instead of S2I? > > Would it make sense for potential code readers? > > I wanted it to be shorter, but if that is a problem, I'd rather call > it PLATFORM than S2IDLE (as technically they are related to the > low-power mode of the platform). It's not problem per se, though without a context it would take time to get into S2I acronym from the code for not familiar reader. > I'll send an update shortly. Thanks!
> -----Original Message----- > From: rjwysocki@gmail.com [mailto:rjwysocki@gmail.com] On Behalf Of Rafael J. > Wysocki > Sent: Wednesday, January 10, 2018 4:23 PM > To: Limonciello, Mario <Mario_Limonciello@Dell.com> > Cc: Rafael J. Wysocki <rjw@rjwysocki.net>; ACPI Devel Maling List <linux- > acpi@vger.kernel.org>; Andy Shevchenko <andriy.shevchenko@linux.intel.com>; > Darren Hart <dvhart@infradead.org>; Linux Kernel Mailing List <linux- > kernel@vger.kernel.org>; Linux PM <linux-pm@vger.kernel.org>; Platform Driver > <platform-driver-x86@vger.kernel.org>; valy@mrs.ro > Subject: Re: [PATCH 1/2] ACPI / PM: Use Low Power S0 Idle on more systems > > On Wed, Jan 10, 2018 at 6:38 PM, <Mario.Limonciello@dell.com> wrote: > > > > > >> -----Original Message----- > >> From: platform-driver-x86-owner@vger.kernel.org [mailto:platform-driver-x86- > >> owner@vger.kernel.org] On Behalf Of Rafael J. Wysocki > >> Sent: Wednesday, January 10, 2018 6:26 AM > >> To: Linux ACPI <linux-acpi@vger.kernel.org> > >> Cc: Andy Shevchenko <andriy.shevchenko@linux.intel.com>; Darren Hart > >> <dvhart@infradead.org>; LKML <linux-kernel@vger.kernel.org>; Linux PM > <linux- > >> pm@vger.kernel.org>; Platform Driver <platform-driver-x86@vger.kernel.org>; > >> Valentin Manea <valy@mrs.ro> > >> Subject: [PATCH 1/2] ACPI / PM: Use Low Power S0 Idle on more systems > >> > >> From: Rafael J. Wysocki <rafael.j.wysocki@intel.com> > >> > >> Some systems don't support the ACPI_LPS0_ENTRY and ACPI_LPS0_EXIT > >> functions in their Low Power S0 Idle _DSM, but still expect EC > >> events to be processed in the suspend-to-idle state for power button > >> wakeup (among other things) to work. Surface Pro3 turns out to be > >> one of them. > >> > >> Fortunately, it still provides Low Power S0 Idle _DSM with the screen > >> on/off functions supported, so modify the ACPI suspend-to-idle to use > >> the Low Power S0 Idle code path for all systems supporting the > >> ACPI_LPS0_ENTRY and ACPI_LPS0_EXIT or the ACPI_LPS0_SCREEN_OFF and > >> ACPI_LPS0_SCREEN_ON functions in their Low Power S0 Idle _DSM. > >> > >> Potentially, that will cause more systems to use suspend-to-idle by > >> default, so some future corrections may be necessary if it leads > >> to issues, but let it remain more straightforward for now. > >> > >> Link: https://bugzilla.kernel.org/show_bug.cgi?id=198389 > >> Reported-by: Valentin Manea <valy@mrs.ro> > >> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com> > >> --- > >> drivers/acpi/sleep.c | 6 ++++-- > >> 1 file changed, 4 insertions(+), 2 deletions(-) > >> > >> Index: linux-pm/drivers/acpi/sleep.c > >> =================================================================== > >> --- linux-pm.orig/drivers/acpi/sleep.c > >> +++ linux-pm/drivers/acpi/sleep.c > >> @@ -707,7 +707,8 @@ static const struct acpi_device_id lps0_ > >> #define ACPI_LPS0_ENTRY 5 > >> #define ACPI_LPS0_EXIT 6 > >> > >> -#define ACPI_S2IDLE_FUNC_MASK ((1 << ACPI_LPS0_ENTRY) | (1 << > >> ACPI_LPS0_EXIT)) > >> +#define ACPI_LPS0_SCREEN_MASK ((1 << ACPI_LPS0_SCREEN_OFF) | (1 << > >> ACPI_LPS0_SCREEN_ON)) > >> +#define ACPI_LPS0_S2I_MASK ((1 << ACPI_LPS0_ENTRY) | (1 << > ACPI_LPS0_EXIT)) > >> > >> static acpi_handle lps0_device_handle; > >> static guid_t lps0_dsm_guid; > >> @@ -910,7 +911,8 @@ static int lps0_device_attach(struct acp > >> if (out_obj && out_obj->type == ACPI_TYPE_BUFFER) { > >> char bitmask = *(char *)out_obj->buffer.pointer; > >> > >> - if ((bitmask & ACPI_S2IDLE_FUNC_MASK) == > >> ACPI_S2IDLE_FUNC_MASK) { > >> + if ((bitmask & ACPI_LPS0_S2I_MASK) == ACPI_LPS0_S2I_MASK || > >> + (bitmask & ACPI_LPS0_SCREEN_MASK) == > >> ACPI_LPS0_SCREEN_MASK) { > >> lps0_dsm_func_mask = bitmask; > >> lps0_device_handle = adev->handle; > >> /* > > > > In making this change I believe you'll need to cache the values that you found > from the > > function mask to test them later too. > > But that's what lps0_dsm_func_mask is for if I understand you correctly. > > > Here: > > https://github.com/torvalds/linux/blob/master/drivers/acpi/sleep.c#L943 > > > > This is because later on both ACPI_LPS0_SCREEN_OFF and ACPI_LPS0_ENTRY are > called > > whether or not they both exist. > > No, that's not the case. > > acpi_sleep_run_lps0_dsm() checks if the given function is there in the > mask returned by function 0 and it doesn't evaluate the _DSM > otherwise. > > Thanks, > Rafael Thanks yes, I see this more closely now you're right.
Index: linux-pm/drivers/acpi/sleep.c =================================================================== --- linux-pm.orig/drivers/acpi/sleep.c +++ linux-pm/drivers/acpi/sleep.c @@ -707,7 +707,8 @@ static const struct acpi_device_id lps0_ #define ACPI_LPS0_ENTRY 5 #define ACPI_LPS0_EXIT 6 -#define ACPI_S2IDLE_FUNC_MASK ((1 << ACPI_LPS0_ENTRY) | (1 << ACPI_LPS0_EXIT)) +#define ACPI_LPS0_SCREEN_MASK ((1 << ACPI_LPS0_SCREEN_OFF) | (1 << ACPI_LPS0_SCREEN_ON)) +#define ACPI_LPS0_S2I_MASK ((1 << ACPI_LPS0_ENTRY) | (1 << ACPI_LPS0_EXIT)) static acpi_handle lps0_device_handle; static guid_t lps0_dsm_guid; @@ -910,7 +911,8 @@ static int lps0_device_attach(struct acp if (out_obj && out_obj->type == ACPI_TYPE_BUFFER) { char bitmask = *(char *)out_obj->buffer.pointer; - if ((bitmask & ACPI_S2IDLE_FUNC_MASK) == ACPI_S2IDLE_FUNC_MASK) { + if ((bitmask & ACPI_LPS0_S2I_MASK) == ACPI_LPS0_S2I_MASK || + (bitmask & ACPI_LPS0_SCREEN_MASK) == ACPI_LPS0_SCREEN_MASK) { lps0_dsm_func_mask = bitmask; lps0_device_handle = adev->handle; /*