[PATCH 0/2] dmaengine: intel_mid_dma: minor fixes

[PATCH 0/2] dmaengine: intel_mid_dma: minor fixes

[Adrian Hunter]

Hi

Here are 2 minor fixes for intel_mid_dma.


Adrian Hunter (2):
      dmaengine: intel_mid_dma: fix error status mask
      dmaengine: intel_mid_dma: move pm_runtime_put

 drivers/dma/intel_mid_dma.c |    5 +++--
 1 files changed, 3 insertions(+), 2 deletions(-)


Regards
Adrian Hunter

[PATCH 1/2] dmaengine: intel_mid_dma: fix error status mask

[Adrian Hunter]

The error status mask (MASK_ERR) has the same format as the other
masks (MASK_TFR, MASK_BLOCK etc) and must be cleared the same way.

Signed-off-by: Adrian Hunter <adrian.hunter@intel.com>
---
 drivers/dma/intel_mid_dma.c |    3 ++-
 1 files changed, 2 insertions(+), 1 deletions(-)

diff --git a/drivers/dma/intel_mid_dma.c b/drivers/dma/intel_mid_dma.c
index 74f70aa..512e2ab 100644
--- a/drivers/dma/intel_mid_dma.c
+++ b/drivers/dma/intel_mid_dma.c
@@ -1056,7 +1056,8 @@ static irqreturn_t intel_mid_dma_interrupt(int irq, void *data)
 	}
 	err_status &= mid->intr_mask;
 	if (err_status) {
-		iowrite32(MASK_INTR_REG(err_status), mid->dma_base + MASK_ERR);
+		iowrite32((err_status << INT_MASK_WE),
+			  mid->dma_base + MASK_ERR);
 		call_tasklet = 1;
 	}
 	if (call_tasklet)
-- 
1.7.6.4

[PATCH 2/2] dmaengine: intel_mid_dma: move pm_runtime_put

[Adrian Hunter]

Move pm_runtime_put() to the end of intel_mid_dma_free_chan_resources()
because there is no sense in allowing runtime suspend while the driver
is still accessing the device.

Signed-off-by: Adrian Hunter <adrian.hunter@intel.com>
---
 drivers/dma/intel_mid_dma.c |    2 +-
 1 files changed, 1 insertions(+), 1 deletions(-)

diff --git a/drivers/dma/intel_mid_dma.c b/drivers/dma/intel_mid_dma.c
index 512e2ab..923476d 100644
--- a/drivers/dma/intel_mid_dma.c
+++ b/drivers/dma/intel_mid_dma.c
@@ -832,7 +832,6 @@ static void intel_mid_dma_free_chan_resources(struct dma_chan *chan)
 		/*trying to free ch in use!!!!!*/
 		pr_err("ERR_MDMA: trying to free ch in use\n");
 	}
-	pm_runtime_put(&mid->pdev->dev);
 	spin_lock_bh(&midc->lock);
 	midc->descs_allocated = 0;
 	list_for_each_entry_safe(desc, _desc, &midc->active_list, desc_node) {
@@ -853,6 +852,7 @@ static void intel_mid_dma_free_chan_resources(struct dma_chan *chan)
 	/* Disable CH interrupts */
 	iowrite32(MASK_INTR_REG(midc->ch_id), mid->dma_base + MASK_BLOCK);
 	iowrite32(MASK_INTR_REG(midc->ch_id), mid->dma_base + MASK_ERR);
+	pm_runtime_put(&mid->pdev->dev);
 }
 
 /**
-- 
1.7.6.4

Re: [PATCH 1/2] dmaengine: intel_mid_dma: fix error status mask

[Vinod Koul]

On Tue, 2012-01-31 at 12:48 +0200, Adrian Hunter wrote:
> The error status mask (MASK_ERR) has the same format as the other
> masks (MASK_TFR, MASK_BLOCK etc) and must be cleared the same way.
> 
> Signed-off-by: Adrian Hunter <adrian.hunter@intel.com>
> ---
>  drivers/dma/intel_mid_dma.c |    3 ++-
>  1 files changed, 2 insertions(+), 1 deletions(-)
> 
> diff --git a/drivers/dma/intel_mid_dma.c b/drivers/dma/intel_mid_dma.c
> index 74f70aa..512e2ab 100644
> --- a/drivers/dma/intel_mid_dma.c
> +++ b/drivers/dma/intel_mid_dma.c
> @@ -1056,7 +1056,8 @@ static irqreturn_t intel_mid_dma_interrupt(int irq, void *data)
>  	}
>  	err_status &= mid->intr_mask;
>  	if (err_status) {
> -		iowrite32(MASK_INTR_REG(err_status), mid->dma_base + MASK_ERR);
> +		iowrite32((err_status << INT_MASK_WE),
> +			  mid->dma_base + MASK_ERR);
why not use the macro?
>  		call_tasklet = 1;
>  	}
>  	if (call_tasklet)


-- 
~Vinod

Re: [PATCH 1/2] dmaengine: intel_mid_dma: fix error status mask

[Adrian Hunter]

On 06/02/12 13:39, Vinod Koul wrote:
> On Tue, 2012-01-31 at 12:48 +0200, Adrian Hunter wrote:
>> The error status mask (MASK_ERR) has the same format as the other
>> masks (MASK_TFR, MASK_BLOCK etc) and must be cleared the same way.
>>
>> Signed-off-by: Adrian Hunter <adrian.hunter@intel.com>
>> ---
>>  drivers/dma/intel_mid_dma.c |    3 ++-
>>  1 files changed, 2 insertions(+), 1 deletions(-)
>>
>> diff --git a/drivers/dma/intel_mid_dma.c b/drivers/dma/intel_mid_dma.c
>> index 74f70aa..512e2ab 100644
>> --- a/drivers/dma/intel_mid_dma.c
>> +++ b/drivers/dma/intel_mid_dma.c
>> @@ -1056,7 +1056,8 @@ static irqreturn_t intel_mid_dma_interrupt(int irq, void *data)
>>  	}
>>  	err_status &= mid->intr_mask;
>>  	if (err_status) {
>> -		iowrite32(MASK_INTR_REG(err_status), mid->dma_base + MASK_ERR);
>> +		iowrite32((err_status << INT_MASK_WE),
>> +			  mid->dma_base + MASK_ERR);
> why not use the macro?

The parameter to the macro MASK_INTR_REG is the channel number.
err_status is not a channel number.

This is the same as the way it is done for MASK_TFR a few lines up.

>>  		call_tasklet = 1;
>>  	}
>>  	if (call_tasklet)
> 
> 

Re: [PATCH 0/2] dmaengine: intel_mid_dma: minor fixes

[Vinod Koul]

On Tue, 2012-01-31 at 12:48 +0200, Adrian Hunter wrote:
> Hi
> 
> Here are 2 minor fixes for intel_mid_dma.
> 
> 
> Adrian Hunter (2):
>       dmaengine: intel_mid_dma: fix error status mask
>       dmaengine: intel_mid_dma: move pm_runtime_put
> 
>  drivers/dma/intel_mid_dma.c |    5 +++--
>  1 files changed, 3 insertions(+), 2 deletions(-)
> 
Applied, Thanks


-- 
~Vinod