Hi,
On Sat, Feb 04, 2012 at 06:55:21PM +0100, Sebastian Andrzej Siewior wrote:
> The disconnect code is scheduled in a tasklet. I don't see the point in
> doing this and the comment says that we may want do this in the irq. This
> patch is doing so.
>
> Cc: Thomas Dahlmann <dahlmann.thomas@xxxxxxxx>
> Signed-off-by: Sebastian Andrzej Siewior <sebastian@xxxxxxxxxxxxx>
> ---
> drivers/usb/gadget/amd5536udc.c | 22 +++-------------------
> 1 files changed, 3 insertions(+), 19 deletions(-)
>
> diff --git a/drivers/usb/gadget/amd5536udc.c b/drivers/usb/gadget/amd5536udc.c
> index e9a2c5c..f0a05e3 100644
> --- a/drivers/usb/gadget/amd5536udc.c
> +++ b/drivers/usb/gadget/amd5536udc.c
> @@ -64,8 +64,7 @@
> /* udc specific */
> #include "amd5536udc.h"
>
> -
> -static void udc_tasklet_disconnect(unsigned long);
> +static void udc_tasklet_disconnect(struct udc *dev);
the "tasklet" part of the name doesn't fit anymore.
> @@ -1647,22 +1641,14 @@ static void usb_disconnect(struct udc *dev)
> /* mask interrupts */
> udc_mask_unused_interrupts(dev);
>
> - /* REVISIT there doesn't seem to be a point to having this
> - * talk to a tasklet ... do it directly, we already hold
> - * the spinlock needed to process the disconnect.
> - */
> -
> - tasklet_schedule(&disconnect_tasklet);
> + udc_tasklet_disconnect(dev);
> }
>
> -/* Tasklet for disconnect to be outside of interrupt context */
> -static void udc_tasklet_disconnect(unsigned long par)
> +static void udc_tasklet_disconnect(struct udc *dev)
> {
> - struct udc *dev = (struct udc *)(*((struct udc **) par));
> u32 tmp;
>
> DBG(dev, "Tasklet disconnect\n");
neither here.
--
balbi
Attachment:
signature.asc
Description: Digital signature
