On 10/19/20 12:06 PM, Thomas Gleixner wrote:
in_interrupt() is a pretty vague context description as it means: hard
interrupt, soft interrupt or bottom half disabled regions.
Replace the vague comment with a proper reasoning why spin_lock_irqsave()
needs to be used.
Signed-off-by: Ahmed S. Darwish <a.darwish@xxxxxxxxxxxxx>
Signed-off-by: Sebastian Andrzej Siewior <bigeasy@xxxxxxxxxxxxx>
Signed-off-by: Thomas Gleixner <tglx@xxxxxxxxxxxxx>
Cc: Duncan Sands <duncan.sands@xxxxxxx>
Cc: Greg Kroah-Hartman <gregkh@xxxxxxxxxxxxxxxxxxx>
Cc: linux-usb@xxxxxxxxxxxxxxx
---
drivers/usb/atm/usbatm.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
--- a/drivers/usb/atm/usbatm.c
+++ b/drivers/usb/atm/usbatm.c
@@ -249,7 +249,7 @@ static void usbatm_complete(struct urb *
/* vdbg("%s: urb 0x%p, status %d, actual_length %d",
__func__, urb, status, urb->actual_length); */
- /* usually in_interrupt(), but not always */
+ /* Can be invoked from task context, protect against interrupts */
spin_lock_irqsave(&channel->lock, flags);
/* must add to the back when receiving; doesn't matter when sending */
Signed-off-by: Duncan Sands <duncan.sands@xxxxxxx>
