2021-12-16 19:30:23

by Rafael J. Wysocki

Subject: [PATCH] PM: sleep: Fix error handling in dpm_prepare()

From: Rafael J. Wysocki <[email protected]>

Commit 2aa36604e824 ("PM: sleep: Avoid calling put_device() under
dpm_list_mtx") forgot to update the while () loop termination
condition to also break the loop if error is nonzero, which
causes the loop to become infinite if device_prepare() returns
an error for one device.

Add the missing !error check.

Fixes: 2aa36604e824 ("PM: sleep: Avoid calling put_device() under dpm_list_mtx")
Signed-off-by: Rafael J. Wysocki <[email protected]>
Thomas Hellström <[email protected]>
---
drivers/base/power/main.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)

Index: linux-pm/drivers/base/power/main.c
===================================================================
--- linux-pm.orig/drivers/base/power/main.c
+++ linux-pm/drivers/base/power/main.c
@@ -1902,7 +1902,7 @@ int dpm_prepare(pm_message_t state)
device_block_probing();

mutex_lock(&dpm_list_mtx);
- while (!list_empty(&dpm_list)) {
+ while (!list_empty(&dpm_list) && !error) {
struct device *dev = to_device(dpm_list.next);

get_device(dev);

2021-12-16 19:44:19

by Thomas Hellström

Subject: Re: [PATCH] PM: sleep: Fix error handling in dpm_prepare()

On Thu, 2021-12-16 at 20:30 +0100, Rafael J. Wysocki wrote:
> From: Rafael J. Wysocki <[email protected]>
>
> Commit 2aa36604e824 ("PM: sleep: Avoid calling put_device() under
> dpm_list_mtx") forgot to update the while () loop termination
> condition to also break the loop if error is nonzero, which
> causes the loop to become infinite if device_prepare() returns
> an error for one device.
>
> Add the missing !error check.
>
> Fixes: 2aa36604e824 ("PM: sleep: Avoid calling put_device() under
> dpm_list_mtx")
> Signed-off-by: Rafael J. Wysocki <[email protected]>
> Thomas Hellström <[email protected]>

Was this meant to be a Reported-by:?
Also Cc stable? IIRC 2aa36604e824 was.

In any case,
Reviewed-by: Thomas Hellström <[email protected]>

Thanks,
Thomas


> ---
>  drivers/base/power/main.c |    2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)
>
> Index: linux-pm/drivers/base/power/main.c
> ===================================================================
> --- linux-pm.orig/drivers/base/power/main.c
> +++ linux-pm/drivers/base/power/main.c
> @@ -1902,7 +1902,7 @@ int dpm_prepare(pm_message_t state)
>         device_block_probing();
>  
>         mutex_lock(&dpm_list_mtx);
> -       while (!list_empty(&dpm_list)) {
> +       while (!list_empty(&dpm_list) && !error) {
>                 struct device *dev = to_device(dpm_list.next);
>  
>                 get_device(dev);
>
>
>

2021-12-16 19:49:20

by Rafael J. Wysocki

Subject: Re: [PATCH] PM: sleep: Fix error handling in dpm_prepare()

On Thu, Dec 16, 2021 at 8:44 PM Thomas Hellström
<[email protected]> wrote:
>
> On Thu, 2021-12-16 at 20:30 +0100, Rafael J. Wysocki wrote:
> > From: Rafael J. Wysocki <[email protected]>
> >
> > Commit 2aa36604e824 ("PM: sleep: Avoid calling put_device() under
> > dpm_list_mtx") forgot to update the while () loop termination
> > condition to also break the loop if error is nonzero, which
> > causes the loop to become infinite if device_prepare() returns
> > an error for one device.
> >
> > Add the missing !error check.
> >
> > Fixes: 2aa36604e824 ("PM: sleep: Avoid calling put_device() under
> > dpm_list_mtx")
> > Signed-off-by: Rafael J. Wysocki <[email protected]>
> > Thomas Hellström <[email protected]>
>
> Was this meant to be a Reported-by:?

Yes, it was, sorry.

> Also Cc stable? IIRC 2aa36604e824 was.

That gets added later and -stable can figure it out.

> In any case,
> Reviewed-by: Thomas Hellström <[email protected]>

Thanks!

> > ---
> > drivers/base/power/main.c | 2 +-
> > 1 file changed, 1 insertion(+), 1 deletion(-)
> >
> > Index: linux-pm/drivers/base/power/main.c
> > ===================================================================
> > --- linux-pm.orig/drivers/base/power/main.c
> > +++ linux-pm/drivers/base/power/main.c
> > @@ -1902,7 +1902,7 @@ int dpm_prepare(pm_message_t state)
> > device_block_probing();
> >
> > mutex_lock(&dpm_list_mtx);
> > - while (!list_empty(&dpm_list)) {
> > + while (!list_empty(&dpm_list) && !error) {
> > struct device *dev = to_device(dpm_list.next);
> >
> > get_device(dev);
> >
> >
> >
>
>

2021-12-17 17:21:55

by Ulf Hansson

Subject: Re: [PATCH] PM: sleep: Fix error handling in dpm_prepare()

On Thu, 16 Dec 2021 at 20:30, Rafael J. Wysocki <[email protected]> wrote:
>
> From: Rafael J. Wysocki <[email protected]>
>
> Commit 2aa36604e824 ("PM: sleep: Avoid calling put_device() under
> dpm_list_mtx") forgot to update the while () loop termination
> condition to also break the loop if error is nonzero, which
> causes the loop to become infinite if device_prepare() returns
> an error for one device.
>
> Add the missing !error check.
>
> Fixes: 2aa36604e824 ("PM: sleep: Avoid calling put_device() under dpm_list_mtx")
> Signed-off-by: Rafael J. Wysocki <[email protected]>
> Thomas Hellström <[email protected]>

Reviewed-by: Ulf Hansson <[email protected]>

Kind regards
Uffe

> ---
> drivers/base/power/main.c | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
> Index: linux-pm/drivers/base/power/main.c
> ===================================================================
> --- linux-pm.orig/drivers/base/power/main.c
> +++ linux-pm/drivers/base/power/main.c
> @@ -1902,7 +1902,7 @@ int dpm_prepare(pm_message_t state)
> device_block_probing();
>
> mutex_lock(&dpm_list_mtx);
> - while (!list_empty(&dpm_list)) {
> + while (!list_empty(&dpm_list) && !error) {
> struct device *dev = to_device(dpm_list.next);
>
> get_device(dev);
>
>
>