On 07/16/2016 10:13 AM, John Stultz wrote:
From: Andy Green <[email protected]>
As it was before, as soon as the DMAC IP felt there was an error
he would return IRQ_NONE since no actual transfer had completed.
After spinning on that for 100K interrupts, Linux yanks the IRQ with
a "nobody cared" error.
This patch lets it handle the interrupt and keep the IRQ alive.
Cc: Zhangfei Gao <[email protected]>
Cc: Jingoo Han <[email protected]>
Cc: Krzysztof Kozlowski <[email protected]>
Cc: Maxime Ripard <[email protected]>
Cc: Vinod Koul <[email protected]>
Cc: Dan Williams <[email protected]>
Cc: Liam Girdwood <[email protected]>
Cc: Mark Brown <[email protected]>
Cc: Jaroslav Kysela <[email protected]>
Cc: Takashi Iwai <[email protected]>
Cc: Wei Xu <[email protected]>
Cc: Rob Herring <[email protected]>
Cc: Andy Green <[email protected]>
Cc: Dave Long <[email protected]>
Cc: Guodong Xu <[email protected]>
Signed-off-by: Andy Green <[email protected]>
[jstultz: Forward ported to mainline]
Signed-off-by: John Stultz <[email protected]>
Acked-by: Zhangfei Gao <[email protected]>
---
drivers/dma/k3dma.c | 8 +++++---
1 file changed, 5 insertions(+), 3 deletions(-)
diff --git a/drivers/dma/k3dma.c b/drivers/dma/k3dma.c
index 8dd050c..c2906a82 100644
--- a/drivers/dma/k3dma.c
+++ b/drivers/dma/k3dma.c
@@ -220,11 +220,13 @@ static irqreturn_t k3_dma_int_handler(int irq, void
*dev_id)
writel_relaxed(err1, d->base + INT_ERR1_RAW);
writel_relaxed(err2, d->base + INT_ERR2_RAW);
- if (irq_chan) {
+ if (irq_chan)
tasklet_schedule(&d->task);
+
+ if (irq_chan || err1 || err2)
return IRQ_HANDLED;
- } else
- return IRQ_NONE;
+
+ return IRQ_NONE;
}
static int k3_dma_start_txd(struct k3_dma_chan *c)
