[PATCH] mmc: fix rmmod race for card-detection polling hosts

[PATCH] mmc: fix rmmod race for card-detection polling hosts

[Guennadi Liakhovetski]

MMC hosts, using polling for card detection by defining the MMC_CAP_NEEDS_POLL
flag, have a race on rmmod, where the delayed work is cancelled without
waiting for completed polling. To prevent this a _sync version of the work
cancellation has to be used.

Signed-off-by: Guennadi Liakhovetski <g.liakhovetski@gmx.de>
---

Chris, if this bug is old enough, this should also go to stable 
eventually.

 drivers/mmc/core/core.c |    2 +-
 1 files changed, 1 insertions(+), 1 deletions(-)

diff --git a/drivers/mmc/core/core.c b/drivers/mmc/core/core.c
index 8f86d70..31ae07a 100644
--- a/drivers/mmc/core/core.c
+++ b/drivers/mmc/core/core.c
@@ -1559,7 +1559,7 @@ void mmc_stop_host(struct mmc_host *host)
 
 	if (host->caps & MMC_CAP_DISABLE)
 		cancel_delayed_work(&host->disable);
-	cancel_delayed_work(&host->detect);
+	cancel_delayed_work_sync(&host->detect);
 	mmc_flush_scheduled_work();
 
 	/* clear pm flags now and let card drivers set them as needed */
-- 
1.7.2.3

Re: [PATCH] mmc: fix rmmod race for card-detection polling hosts

[Chris Ball]

Hi Guennadi,

On Thu, Nov 11, 2010 at 05:32:25PM +0100, Guennadi Liakhovetski wrote:
> MMC hosts, using polling for card detection by defining the MMC_CAP_NEEDS_POLL
> flag, have a race on rmmod, where the delayed work is cancelled without
> waiting for completed polling. To prevent this a _sync version of the work
> cancellation has to be used.
> 
> Signed-off-by: Guennadi Liakhovetski <g.liakhovetski@gmx.de>
> ---
> 
> Chris, if this bug is old enough, this should also go to stable 
> eventually.
> 
>  drivers/mmc/core/core.c |    2 +-
>  1 files changed, 1 insertions(+), 1 deletions(-)
> 
> diff --git a/drivers/mmc/core/core.c b/drivers/mmc/core/core.c
> index 8f86d70..31ae07a 100644
> --- a/drivers/mmc/core/core.c
> +++ b/drivers/mmc/core/core.c
> @@ -1559,7 +1559,7 @@ void mmc_stop_host(struct mmc_host *host)
>  
>  	if (host->caps & MMC_CAP_DISABLE)
>  		cancel_delayed_work(&host->disable);
> -	cancel_delayed_work(&host->detect);
> +	cancel_delayed_work_sync(&host->detect);
>  	mmc_flush_scheduled_work();
>  
>  	/* clear pm flags now and let card drivers set them as needed */

Thanks very much, pushed to for-linus/mmc-next with a Cc: to stable@,
and queued for .37.  The bug is old -- the cancel_delayed_work() call
was introduced in .31, and before that we weren't cancelling it at all.

- Chris.
-- 
Chris Ball   <cjb@laptop.org>   <http://printf.net/>
One Laptop Per Child