On Tue, Jul 14, 2015 at 12:28:25AM +0000, Barry Song wrote:
> From: Yanchang Li <yl22@xxxxxxx>
patch title should alos say what is changed. "panic under one race" is not
an apt description for this
Can you fix that up, otherwise it looks okay
--
~Vinod
>
> If DMA interrupt comes and is latched by IRQ controller during the
> execution of dma_terminate_all(), dma_irq routine will be executed
> after dma terminated, and it will cause kernel panic.
> We clear DMA interrupts in dma_terminate_all() to avoid this useless
> interrupt.
>
> Signed-off-by: Yanchang Li <yl22@xxxxxxx>
> Signed-off-by: Barry Song <Baohua.Song@xxxxxxx>
> ---
> drivers/dma/sirf-dma.c | 3 +++
> 1 file changed, 3 insertions(+)
>
> diff --git a/drivers/dma/sirf-dma.c b/drivers/dma/sirf-dma.c
> index 8c5186c..7d5598d 100644
> --- a/drivers/dma/sirf-dma.c
> +++ b/drivers/dma/sirf-dma.c
> @@ -455,6 +455,7 @@ static int sirfsoc_dma_terminate_all(struct dma_chan *chan)
> switch (sdma->type) {
> case SIRFSOC_DMA_VER_A7V1:
> writel_relaxed(1 << cid, sdma->base + SIRFSOC_DMA_INT_EN_CLR);
> + writel_relaxed(1 << cid, sdma->base + SIRFSOC_DMA_CH_INT);
> writel_relaxed((1 << cid) | 1 << (cid + 16),
> sdma->base +
> SIRFSOC_DMA_CH_LOOP_CTRL_CLR_ATLAS7);
> @@ -462,6 +463,8 @@ static int sirfsoc_dma_terminate_all(struct dma_chan *chan)
> break;
> case SIRFSOC_DMA_VER_A7V2:
> writel_relaxed(0, sdma->base + SIRFSOC_DMA_INT_EN_ATLAS7);
> + writel_relaxed(SIRFSOC_DMA_INT_ALL_ATLAS7,
> + sdma->base + SIRFSOC_DMA_INT_ATLAS7);
> writel_relaxed(0, sdma->base + SIRFSOC_DMA_LOOP_CTRL_ATLAS7);
> writel_relaxed(0, sdma->base + SIRFSOC_DMA_VALID_ATLAS7);
> break;
> --
> 2.3.5
>
--
--
To unsubscribe from this list: send the line "unsubscribe dmaengine" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
