[PATCH] mmc: sdhci-s3c: pass IRQF ONESHOT to request threaded irq

[PATCH] mmc: sdhci-s3c: pass IRQF ONESHOT to request threaded irq

[Heiko Stübner]

Fix a boot regression in existing kernels causing:

genirq: Threaded irq requested with handler=NULL and !ONESHOT for irq XXX

caused by 1c6c6952 (genirq: Reject bogus threaded irq requests).

Signed-off-by: Heiko Stuebner <heiko@sntech.de>
---
Witnessed on a s3c2416 with 3.5-rc1

 drivers/mmc/host/sdhci-s3c.c |    2 +-
 1 files changed, 1 insertions(+), 1 deletions(-)

diff --git a/drivers/mmc/host/sdhci-s3c.c b/drivers/mmc/host/sdhci-s3c.c
index f473d8b..460ddb5 100644
--- a/drivers/mmc/host/sdhci-s3c.c
+++ b/drivers/mmc/host/sdhci-s3c.c
@@ -408,7 +408,7 @@ static void sdhci_s3c_setup_card_detect_gpio(struct sdhci_s3c *sc)
 		if (sc->ext_cd_irq &&
 		    request_threaded_irq(sc->ext_cd_irq, NULL,
 					 sdhci_s3c_gpio_card_detect_thread,
-					 IRQF_TRIGGER_RISING | IRQF_TRIGGER_FALLING,
+					 IRQF_TRIGGER_RISING | IRQF_TRIGGER_FALLING | IRQF_ONESHOT,
 					 dev_name(dev), sc) == 0) {
 			int status = gpio_get_value(sc->ext_cd_gpio);
 			if (pdata->ext_cd_gpio_invert)
-- 
1.7.2.3

Re: [PATCH] mmc: sdhci-s3c: pass IRQF ONESHOT to request threaded irq

[Mark Brown]

On Sun, Jun 03, 2012 at 03:29:55PM +0200, Heiko St??bner wrote:

>  		    request_threaded_irq(sc->ext_cd_irq, NULL,
>  					 sdhci_s3c_gpio_card_detect_thread,
> -					 IRQF_TRIGGER_RISING | IRQF_TRIGGER_FALLING,
> +					 IRQF_TRIGGER_RISING | IRQF_TRIGGER_FALLING | IRQF_ONESHOT,
>  					 dev_name(dev), sc) == 0) {
>  			int status = gpio_get_value(sc->ext_cd_gpio);

Separate issue but shouldn't this be request_any_context_irq()?  Most
GPIOs work just fine in atomic context, though I don't know if the MMC
core limits things.