After commit e1866b3 "PM / Runtime: Rework runtime PM handling
during driver removal" was introduced, the driver core stopped
incrementing the runtime PM usage counter of the device during
the invocation of the ->remove() callback.
This indirectly broke SDIO's runtime PM path during driver removal,
because no one calls _put_sync() anymore after ->remove() completes.
This means that the power of runtime-PM-managed SDIO cards is kept
high after their driver is removed (even if it was powered down
beforehand).
Fix that by directly calling _put_sync() when the last usage
counter is downref'ed by the SDIO bus.
Reported-and-tested-by: Daniel Drake <[email protected]>
Signed-off-by: Ohad Ben-Cohen <[email protected]>
---
drivers/mmc/core/sdio_bus.c | 2 +-
1 files changed, 1 insertions(+), 1 deletions(-)
diff --git a/drivers/mmc/core/sdio_bus.c b/drivers/mmc/core/sdio_bus.c
index d29b9c3..d2565df 100644
--- a/drivers/mmc/core/sdio_bus.c
+++ b/drivers/mmc/core/sdio_bus.c
@@ -189,7 +189,7 @@ static int sdio_bus_remove(struct device *dev)
/* Then undo the runtime PM settings in sdio_bus_probe() */
if (func->card->host->caps & MMC_CAP_POWER_OFF_CARD)
- pm_runtime_put_noidle(dev);
+ pm_runtime_put_sync(dev);
out:
return ret;
--
1.7.1
Hi Ohad,
On Sun, Jun 26 2011, Ohad Ben-Cohen wrote:
> On Fri, Jun 10, 2011 at 2:40 AM, Ohad Ben-Cohen <[email protected]> wrote:
>> After commit e1866b3 "PM / Runtime: Rework runtime PM handling
>> during driver removal" was introduced, the driver core stopped
>> incrementing the runtime PM usage counter of the device during
>> the invocation of the ->remove() callback.
>>
>> This indirectly broke SDIO's runtime PM path during driver removal,
>> because no one calls _put_sync() anymore after ->remove() completes.
>>
>> This means that the power of runtime-PM-managed SDIO cards is kept
>> high after their driver is removed (even if it was powered down
>> beforehand).
>>
>> Fix that by directly calling _put_sync() when the last usage
>> counter is downref'ed by the SDIO bus.
>
> Can you please take this 1-liner into 3.0-rc ?
>
> It fixes SDIO runtime PM after a breakage was introduced in 3.0.
Merged, thanks. I'll be sending a pull request on my for-linus branch
shortly:
http://git.kernel.org/?p=linux/kernel/git/cjb/mmc.git;a=shortlog;h=refs/heads/for-linus
- Chris.
--
Chris Ball <[email protected]> <http://printf.net/>
One Laptop Per Child
Hi Chris,
On Fri, Jun 10, 2011 at 2:40 AM, Ohad Ben-Cohen <[email protected]> wrote:
> After commit e1866b3 "PM / Runtime: Rework runtime PM handling
> during driver removal" was introduced, the driver core stopped
> incrementing the runtime PM usage counter of the device during
> the invocation of the ->remove() callback.
>
> This indirectly broke SDIO's runtime PM path during driver removal,
> because no one calls _put_sync() anymore after ->remove() completes.
>
> This means that the power of runtime-PM-managed SDIO cards is kept
> high after their driver is removed (even if it was powered down
> beforehand).
>
> Fix that by directly calling _put_sync() when the last usage
> counter is downref'ed by the SDIO bus.
Can you please take this 1-liner into 3.0-rc ?
It fixes SDIO runtime PM after a breakage was introduced in 3.0.
Thanks,
Ohad.