Hi,
On 10/14/2010 9:13 PM, Hari Kanigeri wrote:
@@ -252,41 +253,39 @@ static int omap_mbox_startup(struct omap_mbox *mbox)
...
+ if (!mbox->use_count++) {
+ ret = request_irq(mbox->irq, mbox_interrupt, IRQF_SHARED,
+ mbox->name, mbox);
...
@@ -296,29 +295,36 @@ fail_alloc_txq:
...
static void omap_mbox_fini(struct omap_mbox *mbox)
{
+ if (!--mbox->use_count) {
+ tasklet_kill(&mbox->txq->tasklet);
+ flush_work(&mbox->rxq->work);
+ mbox_queue_free(mbox->txq);
+ mbox_queue_free(mbox->rxq);
+ }
+
+ if (likely(mbox->ops->shutdown)) {
+ if (!--mbox_configured) {
+ free_irq(mbox->irq, mbox);
Above hunks will create an imbalance of free_irq, as request_irq can be
called per registered mailbox and free_irq is only done for the last
caller releasing the mailbox handle.
e.g.: mbox-1, mbox-N will request a shared irq on the same interrupt
line, but only the last caller of omap_mbox_put will free its irq,
leaving the other one there.
This can be fixed if the free is moved to be executed within the
following block:
if (!--mbox->use_count) {
...
free_irq(mbox->irq, mbox);
}
Regards,
Omar
--
To unsubscribe from this list: send the line "unsubscribe linux-omap" in
the body of a message to majord...@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
