[PATCH 7/7] OF: set dma_mask for ARM

[PATCH 7/7] OF: set dma_mask for ARM

[Kristoffer Glembo]

Hi,

Have you gotten any feedback on this? I am facing the same decision on SPARC32. I need to add dma_mask to get USB working on non-PCI buses and I'm currently thinking about doing it either the PowerPC way or going with the dma_mask = &coherent_dma_mask route. It would be nice if we could get some community agreement on which is the proper way of doing this.

Best regards,
Kristoffer Glembo


On 11/16/2010 11:24 PM, Grant Likely wrote:
>> On Tue, Nov 16, 2010 at 02:33:55PM -0600, Rob Herring wrote:
>>> From: Rob Herring<rob.herring at calxeda.com>
>>>
>>> Various drivers require dma_mask to be valid, so it needs to be setup
>>> when doing OF probing on ARM.
>>>
>>> Signed-off-by: Rob Herring<rob.herring at calxeda.com>
>>
>> I'll let patches 6&  7 lay fallow for the moment.  I'm not hugely fond
>> of the powerpc approach (which microblaze copies), but I haven't had a
>> chance to look for a better solution.
>>
>
>Looking at this some more, there's a couple of options I see.
>
>Other buses (ISA, PCI, EISA) have a dma_mask in their struct. So perhaps 
>platform_device should just have a dma_mask field.
>
>There are several examples of setting dma_mask = &coherent_dma_mask. I'm 
>not sure if this is considered correct. At least on ARM, I don't think 
>these would need to be different.
>
>Another option would be using bus notifiers to set the dma_mask like 
>platform_data. But that doesn't seem like a good long term solution.
>
>Rob

Re: [PATCH 7/7] OF: set dma_mask for ARM

[Rob Herring]

On 04/20/2011 03:11 AM, Kristoffer Glembo wrote:
> Hi,
>
> Have you gotten any feedback on this? I am facing the same decision
> on SPARC32. I need to add dma_mask to get USB working on non-PCI
> buses and I'm currently thinking about doing it either the PowerPC
> way or going with the dma_mask =&coherent_dma_mask route. It would be
> nice if we could get some community agreement on which is the proper
> way of doing this.
>

No, only Grant's original comment. I did switch to using a bus notifier 
callback to set the dma_mask pointer so I don't need to modify the DT or 
arch code.

Rob

> Best regards, Kristoffer Glembo
>
>
> On 11/16/2010 11:24 PM, Grant Likely wrote:
>>> On Tue, Nov 16, 2010 at 02:33:55PM -0600, Rob Herring wrote:
>>>> From: Rob Herring<rob.herring at calxeda.com>
>>>>
>>>> Various drivers require dma_mask to be valid, so it needs to be
>>>> setup when doing OF probing on ARM.
>>>>
>>>> Signed-off-by: Rob Herring<rob.herring at calxeda.com>
>>>
>>> I'll let patches 6&   7 lay fallow for the moment.  I'm not
>>> hugely fond of the powerpc approach (which microblaze copies),
>>> but I haven't had a chance to look for a better solution.
>>>
>>
>> Looking at this some more, there's a couple of options I see.
>>
>> Other buses (ISA, PCI, EISA) have a dma_mask in their struct. So
>> perhaps platform_device should just have a dma_mask field.
>>
>> There are several examples of setting dma_mask =&coherent_dma_mask.
>> I'm not sure if this is considered correct. At least on ARM, I
>> don't think these would need to be different.
>>
>> Another option would be using bus notifiers to set the dma_mask
>> like platform_data. But that doesn't seem like a good long term
>> solution.
>>
>> Rob

[PATCH 0/7] Various OF fixes for little-endian and ARM

[Rob Herring]

From: Rob Herring <rob.herring-bsGFqQB8/DxBDgjK7y7TUQ@public.gmane.org>

This is a mixture of driver fixes for OF on little-endian and ARM platforms.

The last 2 patches need more thought. Currently, it's just copied from PPC. There 
are some ARM platforms which set 64-bit dma_masks (iop13xx and orion5x)
which is not handled. I would think 64-bit PPC would have a similar issue.

Rob Herring (7):
  OF: make drivers depend on CONFIG_OF instead of CONFIG_PPC_OF
  of: Fixes for OF probing on little endian systems
  ipmi: explicitly include of_address.h and of_irq.h
  mmc: sdhci-of: fix build on non-powerpc platforms
  mmc: sdhci-of: support generic OF controllers
  ARM: add dma_mask to pdev_archdata
  OF: set dma_mask for ARM

 arch/arm/include/asm/device.h    |    1 +
 drivers/char/ipmi/ipmi_si_intf.c |   24 +++++++++++++-----------
 drivers/mmc/host/Kconfig         |    2 +-
 drivers/mmc/host/sdhci-of-core.c |   18 ++++++++++++++++--
 drivers/mtd/Kconfig              |    2 +-
 drivers/mtd/maps/Kconfig         |    2 +-
 drivers/net/can/sja1000/Kconfig  |    4 ++--
 drivers/of/platform.c            |    4 ++--
 drivers/serial/Kconfig           |    2 +-
 drivers/usb/host/isp1760-if.c    |    8 ++++----
 10 files changed, 42 insertions(+), 25 deletions(-)

[PATCH 7/7] OF: set dma_mask for ARM

[Rob Herring]

From: Rob Herring <rob.herring-bsGFqQB8/DxBDgjK7y7TUQ@public.gmane.org>

Various drivers require dma_mask to be valid, so it needs to be setup
when doing OF probing on ARM.

Signed-off-by: Rob Herring <rob.herring-bsGFqQB8/DxBDgjK7y7TUQ@public.gmane.org>
---
 drivers/of/platform.c |    4 ++--
 1 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/of/platform.c b/drivers/of/platform.c
index 5b4a07f..fe2669a 100644
--- a/drivers/of/platform.c
+++ b/drivers/of/platform.c
@@ -614,7 +614,7 @@ struct platform_device *of_device_alloc(struct device_node *np,
 	}
 
 	dev->dev.of_node = of_node_get(np);
-#if defined(CONFIG_PPC) || defined(CONFIG_MICROBLAZE)
+#if defined(CONFIG_PPC) || defined(CONFIG_MICROBLAZE) || defined(CONFIG_ARM)
 	dev->dev.dma_mask = &dev->archdata.dma_mask;
 #endif
 	dev->dev.parent = parent;
@@ -644,7 +644,7 @@ struct platform_device *of_platform_device_create(struct device_node *np,
 	if (!dev)
 		return NULL;
 
-#if defined(CONFIG_PPC) || defined(CONFIG_MICROBLAZE)
+#if defined(CONFIG_PPC) || defined(CONFIG_MICROBLAZE) || defined(CONFIG_ARM)
 	dev->archdata.dma_mask = 0xffffffffUL;
 #endif
 	dev->dev.coherent_dma_mask = DMA_BIT_MASK(32);
-- 
1.7.1

Re: [PATCH 7/7] OF: set dma_mask for ARM

[Grant Likely]

On Tue, Nov 16, 2010 at 02:33:55PM -0600, Rob Herring wrote:
> From: Rob Herring <rob.herring-bsGFqQB8/DxBDgjK7y7TUQ@public.gmane.org>
> 
> Various drivers require dma_mask to be valid, so it needs to be setup
> when doing OF probing on ARM.
> 
> Signed-off-by: Rob Herring <rob.herring-bsGFqQB8/DxBDgjK7y7TUQ@public.gmane.org>

I'll let patches 6 & 7 lay fallow for the moment.  I'm not hugely fond
of the powerpc approach (which microblaze copies), but I haven't had a
chance to look for a better solution.

g.

> ---
>  drivers/of/platform.c |    4 ++--
>  1 files changed, 2 insertions(+), 2 deletions(-)
> 
> diff --git a/drivers/of/platform.c b/drivers/of/platform.c
> index 5b4a07f..fe2669a 100644
> --- a/drivers/of/platform.c
> +++ b/drivers/of/platform.c
> @@ -614,7 +614,7 @@ struct platform_device *of_device_alloc(struct device_node *np,
>  	}
>  
>  	dev->dev.of_node = of_node_get(np);
> -#if defined(CONFIG_PPC) || defined(CONFIG_MICROBLAZE)
> +#if defined(CONFIG_PPC) || defined(CONFIG_MICROBLAZE) || defined(CONFIG_ARM)
>  	dev->dev.dma_mask = &dev->archdata.dma_mask;
>  #endif
>  	dev->dev.parent = parent;
> @@ -644,7 +644,7 @@ struct platform_device *of_platform_device_create(struct device_node *np,
>  	if (!dev)
>  		return NULL;
>  
> -#if defined(CONFIG_PPC) || defined(CONFIG_MICROBLAZE)
> +#if defined(CONFIG_PPC) || defined(CONFIG_MICROBLAZE) || defined(CONFIG_ARM)
>  	dev->archdata.dma_mask = 0xffffffffUL;
>  #endif
>  	dev->dev.coherent_dma_mask = DMA_BIT_MASK(32);
> -- 
> 1.7.1
> 
> _______________________________________________
> devicetree-discuss mailing list
> devicetree-discuss-uLR06cmDAlY/bJ5BZ2RsiQ@public.gmane.org
> https://lists.ozlabs.org/listinfo/devicetree-discuss

Re: [PATCH 7/7] OF: set dma_mask for ARM

[Rob Herring]

Grant,

On 11/16/2010 11:24 PM, Grant Likely wrote:
> On Tue, Nov 16, 2010 at 02:33:55PM -0600, Rob Herring wrote:
>> From: Rob Herring<rob.herring-bsGFqQB8/DxBDgjK7y7TUQ@public.gmane.org>
>>
>> Various drivers require dma_mask to be valid, so it needs to be setup
>> when doing OF probing on ARM.
>>
>> Signed-off-by: Rob Herring<rob.herring-bsGFqQB8/DxBDgjK7y7TUQ@public.gmane.org>
>
> I'll let patches 6&  7 lay fallow for the moment.  I'm not hugely fond
> of the powerpc approach (which microblaze copies), but I haven't had a
> chance to look for a better solution.
>

Looking at this some more, there's a couple of options I see.

Other buses (ISA, PCI, EISA) have a dma_mask in their struct. So perhaps 
platform_device should just have a dma_mask field.

There are several examples of setting dma_mask = &coherent_dma_mask. I'm 
not sure if this is considered correct. At least on ARM, I don't think 
these would need to be different.

Another option would be using bus notifiers to set the dma_mask like 
platform_data. But that doesn't seem like a good long term solution.

Rob