mmc: omap_hsmmc: fix wakeirq handling on removal

commit 3c398f3c3bef21961eaaeb93227fa66d440dc83d upstream.

after unbinding mmc I get things like this:
[  185.294067] mmc1: card 0001 removed
[  185.305206] omap_hsmmc 480b4000.mmc: wake IRQ with no resume: -13

The wakeirq stays in /proc-interrupts

rebinding shows this:
[  289.795959] genirq: Flags mismatch irq 112. 0000200a (480b4000.mmc:wakeup) vs. 0000200a (480b4000.mmc:wakeup)
[  289.808959] omap_hsmmc 480b4000.mmc: Unable to request wake IRQ
[  289.815338] omap_hsmmc 480b4000.mmc: no SDIO IRQ support, falling back to polling

That bug seems to be introduced by switching from devm_request_irq()
to generic wakeirq handling.

So let us cleanup at removal.

Signed-off-by: Andreas Kemnade <andreas@kemnade.info>
Fixes: 5b83b2234be6 ("mmc: omap_hsmmc: Change wake-up interrupt to use generic wakeirq")
Cc: stable@vger.kernel.org # v4.2+
Signed-off-by: Ulf Hansson <ulf.hansson@linaro.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

diff --git a/drivers/mmc/host/omap_hsmmc.c b/drivers/mmc/host/omap_hsmmc.c
index a082aa3..f7d1c8c 100644
--- a/drivers/mmc/host/omap_hsmmc.c
+++ b/drivers/mmc/host/omap_hsmmc.c
@@ -2215,6 +2215,7 @@
 	dma_release_channel(host->tx_chan);
 	dma_release_channel(host->rx_chan);
 
+	dev_pm_clear_wake_irq(host->dev);
 	pm_runtime_dont_use_autosuspend(host->dev);
 	pm_runtime_put_sync(host->dev);
 	pm_runtime_disable(host->dev);