On Tue, 30 Apr 2013, Linus Walleij wrote:
> On Wed, Apr 24, 2013 at 12:58 PM, Lee Jones wrote:
>
> > Currently, if DMA information isn't passed from platform data, then DMA
> > will not be used. This patch allows DMA information obtained though Device
> > Tree to be used as well.
> >
> > Cc: Rus
On Wed, Apr 24, 2013 at 12:58 PM, Lee Jones wrote:
> Currently, if DMA information isn't passed from platform data, then DMA
> will not be used. This patch allows DMA information obtained though Device
> Tree to be used as well.
>
> Cc: Russell King
> Cc: Chris Ball
> Cc: linux-...@vger.kernel.
Currently, if DMA information isn't passed from platform data, then DMA
will not be used. This patch allows DMA information obtained though Device
Tree to be used as well.
Cc: Russell King
Cc: Chris Ball
Cc: linux-...@vger.kernel.org
Signed-off-by: Lee Jones
---
drivers/mmc/host/mmci.c | 43
On Thu, Apr 18, 2013 at 11:25:44AM +0200, Arnd Bergmann wrote:
> +static void mmci_dma_setup(struct device *dev, struct mmci_host *host)
> +{
> + const char *rxname, *txname;
> +
> + host->dma_rx_channel = dma_request_slave_channel(dev, "rx");
> + host->dma_tx_channel = dma_request_slav
On Thursday 18 April 2013, Russell King - ARM Linux wrote:
> On Thu, Apr 18, 2013 at 09:02:38AM +0100, Lee Jones wrote:
> > @@ -321,19 +323,21 @@ static void mmci_dma_setup(struct mmci_host *host)
> >* attempt to use it bidirectionally, however if it is
> >* is specified but cannot
On 18 April 2013 10:02, Lee Jones wrote:
>> I think you can further simplify this, given that in the DT case we always
>> allocate a zeroed mmci_platform_data for host->plat, so !plat cannot happen
>> when we get here.
>
> Okay, third time lucky. :)
>
> diff --git a/drivers/mmc/host/mmci.c b/drive
On Thu, Apr 18, 2013 at 09:02:38AM +0100, Lee Jones wrote:
> @@ -321,19 +323,21 @@ static void mmci_dma_setup(struct mmci_host *host)
>* attempt to use it bidirectionally, however if it is
>* is specified but cannot be located, DMA will be disabled.
>*/
> - if (plat->dma
> I think you can further simplify this, given that in the DT case we always
> allocate a zeroed mmci_platform_data for host->plat, so !plat cannot happen
> when we get here.
Okay, third time lucky. :)
diff --git a/drivers/mmc/host/mmci.c b/drivers/mmc/host/mmci.c
index 372e921..3260ab4 100644
--
On Wednesday 17 April 2013, Lee Jones wrote:
> > This looks unnecessarily complex.
>
> That thought did cross my mind.
>
> > Why not just do dma_request_slave_channel_compat() unconditionally here?
>
> So how about something like this instead, as it keeps the current
> semantics, and only differ
> This looks unnecessarily complex.
That thought did cross my mind.
> Why not just do dma_request_slave_channel_compat() unconditionally here?
So how about something like this instead, as it keeps the current
semantics, and only differs in the case of DT.
diff --git a/drivers/mmc/host/mmci.c b/
On Wednesday 17 April 2013, Lee Jones wrote:
> #ifdef CONFIG_DMA_ENGINE
> -static void mmci_dma_setup(struct mmci_host *host)
> +static void mmci_dma_setup(struct amba_device *dev,
> +struct mmci_host *host)
> {
> + struct device_node *np = dev->dev.of_node;
> s
Currently, if DMA information isn't passed from platform data, then DMA
will not be used. This patch allows DMA information obtained though Device
Tree to be used as well.
Cc: Russell King
Cc: Chris Ball
Cc: linux-...@vger.kernel.org
Signed-off-by: Lee Jones
---
drivers/mmc/host/mmci.c | 43
12 matches
Mail list logo