return type of wait_for_completion_timeout is unsigned long not int. The
rc variable is in use for other calls so an additional timeout variable
of type unsigned long is added.
Signed-off-by: Nicholas Mc Guire <[email protected]>
---
Note that the timeout case is being ignored - not sure if this should not
do something like:
if (!timeout)
return -ETIMEOUT;
in case resume timed-out ?
Patch was only compile tested with x86_64_defconfig + CONFIG_MMC=m,
CONFIG_MMC_TIFM_SD=m, (the later implies CONFIG_TIFM_7XX1=m)
Patch is against 3.19.0-rc7 (localversion-next is -next-20150209)
drivers/misc/tifm_7xx1.c | 5 +++--
1 file changed, 3 insertions(+), 2 deletions(-)
diff --git a/drivers/misc/tifm_7xx1.c b/drivers/misc/tifm_7xx1.c
index a606c89..252530a 100644
--- a/drivers/misc/tifm_7xx1.c
+++ b/drivers/misc/tifm_7xx1.c
@@ -236,6 +236,7 @@ static int tifm_7xx1_resume(struct pci_dev *dev)
{
struct tifm_adapter *fm = pci_get_drvdata(dev);
int rc;
+ unsigned long timeout;
unsigned int good_sockets = 0, bad_sockets = 0;
unsigned long flags;
unsigned char new_ids[fm->num_sockets];
@@ -272,8 +273,8 @@ static int tifm_7xx1_resume(struct pci_dev *dev)
if (good_sockets) {
fm->finish_me = &finish_resume;
spin_unlock_irqrestore(&fm->lock, flags);
- rc = wait_for_completion_timeout(&finish_resume, HZ);
- dev_dbg(&dev->dev, "wait returned %d\n", rc);
+ timeout = wait_for_completion_timeout(&finish_resume, HZ);
+ dev_dbg(&dev->dev, "wait returned %lu\n", timeout);
writel(TIFM_IRQ_FIFOMASK(good_sockets)
| TIFM_IRQ_CARDMASK(good_sockets),
fm->addr + FM_CLEAR_INTERRUPT_ENABLE);
--
1.7.10.4