Commit dbbc1c76 authored Jul 24, 2015 by Deva Ramasubramanian Committed by Gerrit - the friendly Code Review server Jul 30, 2015

msm: vidc: Re-enable IRQ even if we didn't do anything in the IRQ handler

Since we disable the IRQ unconditionally in the IRQ handler, we need to
do likewise in the bottom-half of the interrupt handling routine.
Otherwise, if __power_on() fails, we end up never re-enabling the IRQ,
which eventually results in the next session timing out.

Change-Id: I010c7ebc40d13c310229fa4c06c01259de68f124
Signed-off-by: Deva Ramasubramanian <dramasub@codeaurora.org>

parent 1aa24ca3

drivers/media/platform/msm/vidc/venus_hfi.c

+2 −1

Original line number	Diff line number	Diff line
		@@ -3669,10 +3669,11 @@ static void venus_hfi_core_work_handler(struct work_struct *work)
		__core_clear_interrupt(device);
		num_responses = __response_handler(device);

		err_no_work:
		/* We need re-enable the irq which was disabled in ISR handler */
		if (!(device->intr_status & VIDC_WRAPPER_INTR_STATUS_A2HWD_BMSK))
		enable_irq(device->hal_data->irq);

		err_no_work:
		mutex_unlock(&device->lock);

		/*