Re: [PATCH v2] PM / sleep: prohibit devices probing during suspend/hibernation

[Rafael J. Wysocki]

On Friday, November 06, 2015 11:58:40 AM Grygorii Strashko wrote:
> On 11/06/2015 02:13 AM, Rafael J. Wysocki wrote:
> > On Thursday, November 05, 2015 11:19:03 PM Pavel Machek wrote:
> >> On Mon 2015-10-19 23:54:24, Grygorii Strashko wrote:
> >>> It is unsafe [1] if probing of devices will happen during suspend or
> >>> hibernation and system behavior will be unpredictable in this case
> >>> (for example: after successful probe the device potentially has a different
> >>>   set of PM callbacks than before [2]).
> >>> So, let's prohibit device's probing in dpm_prepare() and defer their
> >>> probes instead. The normal behavior will be restored in dpm_complete().
> >>>
> >>> This patch introduces new DD core APIs:
> >>>   device_defer_all_probes_enable()
> >>>     It will disable probing of devices and defer their probes.
> >>>   device_defer_all_probes_disable()
> >>>     It will restore normal behavior and trigger re-probing of deferred
> >>>     devices.
> >>>
> >>> [1] https://lkml.org/lkml/2015/9/11/554
> >>> [2] https://lkml.org/lkml/2015/9/15/1039
> >>> Cc: Alan Stern <stern@xxxxxxxxxxxxxxxxxxx>
> >>> Cc: Rafael J. Wysocki <rjw@xxxxxxxxxxxxx>
> >>> Cc: Thierry Reding <thierry.reding@xxxxxxxxx>
> >>> Signed-off-by: Grygorii Strashko <grygorii.strashko@xxxxxx>
> >>> ---
> >>> Changes in v2:
> >>> - DD core API device_defer_all_probes(bool enable) split on two
> >>>    void device_defer_all_probes_enable(void);
> >>>    void device_defer_all_probes_disable(void);
> >>> - more comments added
> >>>
> >>> Link on v1:
> >>> - https://lkml.org/lkml/2015/10/8/681
> >>>
> >>>   drivers/base/base.h       |  2 ++
> >>>   drivers/base/dd.c         | 48 ++++++++++++++++++++++++++++++++++++++++++++++-
> >>>   drivers/base/power/main.c | 17 +++++++++++++++++
> >>>   3 files changed, 66 insertions(+), 1 deletion(-)
> >>>
> >>> diff --git a/drivers/base/base.h b/drivers/base/base.h
> >>> index 1782f3a..c332b68 100644
> >>> --- a/drivers/base/base.h
> >>> +++ b/drivers/base/base.h
> >>> @@ -131,6 +131,8 @@ extern void device_remove_groups(struct device *dev,
> >>>   extern char *make_class_name(const char *name, struct kobject *kobj);
> >>>
> >>>   extern int devres_release_all(struct device *dev);
> >>> +extern void device_defer_all_probes_enable(void);
> >>> +extern void device_defer_all_probes_disable(void);
> >>>
> >>>   /* /sys/devices directory */
> >>>   extern struct kset *devices_kset;
> >>> diff --git a/drivers/base/dd.c b/drivers/base/dd.c
> >>> index be0eb46..b8d9e70 100644
> >>> --- a/drivers/base/dd.c
> >>> +++ b/drivers/base/dd.c
> >>> @@ -55,6 +55,13 @@ static struct workqueue_struct *deferred_wq;
> >>>   static atomic_t deferred_trigger_count = ATOMIC_INIT(0);
> >>>
> >>>   /*
> >>> + * In some cases, like suspend to RAM or hibernation, It might be reasonable
> >>> + * to prohibit probing of devices as it could be unsafe.
> >>> + * Once defer_all_probes is true all drivers probes will be forcibly deferred.
> >>> + */
> >>> +static bool defer_all_probes;
> >>> +
> >>> +/*
> >>>    * deferred_probe_work_func() - Retry probing devices in the active list.
> >>>    */
> >>>   static void deferred_probe_work_func(struct work_struct *work)
> >>> @@ -172,6 +179,30 @@ static void driver_deferred_probe_trigger(void)
> >>>   }
> >>>
> >>>   /**
> >>> + * device_defer_all_probes_enable() - Enable deferring of device's probes
> >>> + *
> >>> + *	It will disable probing of devices and defer their probes.
> >>> + */
> >>> +void device_defer_all_probes_enable(void)
> >>> +{
> >>> +	defer_all_probes = true;
> >>> +	/* sync with probes to avoid races. */
> >>> +	wait_for_device_probe();
> >>> +}
> >>
> >> device_pause_probing()?
> 
> device_stop_probing(). Right?

Or device_block_probing()

> >>
> >>> +/**
> >>> + * device_defer_all_probes_disable() - Disable deferring of device's probes
> >>> + *
> >>> + *	It will restore normal behavior and trigger re-probing of deferred
> >>> + * devices.
> >>> + */
> 
> device_start_probing(). Right?

and device_unblock_probing()

> 
> >>
> >> Hmm. This is not quite a double negative... but it sounds like one.
> >>
> >> device_restart_probing()?
> >
> > I _start/_stop would be fine by me too.
> 
> Ok. So do I need rename/resend it again?

Yes, please.

> Probably, I'll need to wait for -rc1 before resending.

No, you don't.  Please resend when you're ready.

Thanks,
Rafael

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/