Re: [RFC v3 02/45] dma-mapping: Use unsigned long for dma_attrs

Re: [RFC v3 02/45] dma-mapping: Use unsigned long for dma_attrs

[Geert Uytterhoeven]

Hi Krzysztof,

On Thu, Jun 2, 2016 at 5:39 PM, Krzysztof Kozlowski
<k.kozlowski-Sze3O3UU22JBDgjK7y7TUQ@public.gmane.org> wrote:
> --- a/include/linux/dma-mapping.h
> +++ b/include/linux/dma-mapping.h
> @@ -5,13 +5,25 @@

> +/**
> + * List of possible attributes associated with a DMA mapping. The semantics
> + * of each attribute should be defined in Documentation/DMA-attributes.txt.
> + */
> +#define DMA_ATTR_WRITE_BARRIER         (1UL << 1)

Any particular reason they start at 2, not 1?

> +#define DMA_ATTR_WEAK_ORDERING         (1UL << 2)
> +#define DMA_ATTR_WRITE_COMBINE         (1UL << 3)
> +#define DMA_ATTR_NON_CONSISTENT                (1UL << 4)
> +#define DMA_ATTR_NO_KERNEL_MAPPING     (1UL << 5)
> +#define DMA_ATTR_SKIP_CPU_SYNC         (1UL << 6)
> +#define DMA_ATTR_FORCE_CONTIGUOUS      (1UL << 7)
> +#define DMA_ATTR_ALLOC_SINGLE_PAGES    (1UL << 8)

Gr{oetje,eeting}s,

                        Geert

--
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- geert-Td1EMuHUCqxL1ZNQvxDV9g@public.gmane.org

In personal conversations with technical people, I call myself a hacker. But
when I'm talking to journalists I just say "programmer" or something like that.
                                -- Linus Torvalds

Re: [RFC v3 29/45] m68k: dma-mapping: Use unsigned long for dma_attrs

[Geert Uytterhoeven]

On Thu, Jun 2, 2016 at 5:39 PM, Krzysztof Kozlowski
<k.kozlowski-Sze3O3UU22JBDgjK7y7TUQ@public.gmane.org> wrote:
> Split out subsystem specific changes for easier reviews. This will be
> squashed with main commit.
>
> Signed-off-by: Krzysztof Kozlowski <k.kozlowski-Sze3O3UU22JBDgjK7y7TUQ@public.gmane.org>

Looks good.

Acked-by: Geert Uytterhoeven <geert-Td1EMuHUCqxL1ZNQvxDV9g@public.gmane.org>

Gr{oetje,eeting}s,

                        Geert

--
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- geert-Td1EMuHUCqxL1ZNQvxDV9g@public.gmane.org

In personal conversations with technical people, I call myself a hacker. But
when I'm talking to journalists I just say "programmer" or something like that.
                                -- Linus Torvalds

Re: [RFC v3 02/45] dma-mapping: Use unsigned long for dma_attrs

[Krzysztof Kozlowski]

On 06/03/2016 09:17 AM, Geert Uytterhoeven wrote:
> Hi Krzysztof,
> 
> On Thu, Jun 2, 2016 at 5:39 PM, Krzysztof Kozlowski
> <k.kozlowski-Sze3O3UU22JBDgjK7y7TUQ@public.gmane.org> wrote:
>> --- a/include/linux/dma-mapping.h
>> +++ b/include/linux/dma-mapping.h
>> @@ -5,13 +5,25 @@
> 
>> +/**
>> + * List of possible attributes associated with a DMA mapping. The semantics
>> + * of each attribute should be defined in Documentation/DMA-attributes.txt.
>> + */
>> +#define DMA_ATTR_WRITE_BARRIER         (1UL << 1)
> 
> Any particular reason they start at 2, not 1?

No reason. I'll fix this in next version (and trim Cc-list). Anyway the
values of constants won't match old ones but that should not be problem
(unless they are hard-coded somewhere).

Best regards,
Krzysztof

Re: [RFC v3 10/45] cris: dma-mapping: Use unsigned long for dma_attrs

[Jesper Nilsson]

On Thu, Jun 02, 2016 at 05:39:12PM +0200, Krzysztof Kozlowski wrote:
> Split out subsystem specific changes for easier reviews. This will be
> squashed with main commit.

Acked-by: Jesper Nilsson <jesper.nilsson@axis.com>

> Signed-off-by: Krzysztof Kozlowski <k.kozlowski@samsung.com>

/^JN - Jesper Nilsson
-- 
               Jesper Nilsson -- jesper.nilsson@axis.com

Re: [RFC v3 07/45] avr32: dma-mapping: Use unsigned long for dma_attrs

[Hans-Christian Noren Egtvedt]

Around Thu 02 Jun 2016 17:39:09 +0200 or thereabout, Krzysztof Kozlowski wrote:
> Split out subsystem specific changes for easier reviews. This will be
> squashed with main commit.
> 
> Signed-off-by: Krzysztof Kozlowski <k.kozlowski@samsung.com>

Acked-by: Hans-Christian Noren Egtvedt <egtvedt@samfundet.no>

> ---
>  arch/avr32/mm/dma-coherent.c | 8 ++++----
>  1 file changed, 4 insertions(+), 4 deletions(-)
> 
> diff --git a/arch/avr32/mm/dma-coherent.c b/arch/avr32/mm/dma-coherent.c
> index 92cf1fb2b3e6..fc51f4421933 100644
> --- a/arch/avr32/mm/dma-coherent.c
> +++ b/arch/avr32/mm/dma-coherent.c
> @@ -99,7 +99,7 @@ static void __dma_free(struct device *dev, size_t size,
>  }
>  
>  static void *avr32_dma_alloc(struct device *dev, size_t size,
> -		dma_addr_t *handle, gfp_t gfp, struct dma_attrs *attrs)
> +		dma_addr_t *handle, gfp_t gfp, unsigned long attrs)
>  {
>  	struct page *page;
>  	dma_addr_t phys;
> @@ -119,7 +119,7 @@ static void *avr32_dma_alloc(struct device *dev, size_t size,
>  }
>  
>  static void avr32_dma_free(struct device *dev, size_t size,
> -		void *cpu_addr, dma_addr_t handle, struct dma_attrs *attrs)
> +		void *cpu_addr, dma_addr_t handle, unsigned long attrs)
>  {
>  	struct page *page;
>  
> @@ -142,7 +142,7 @@ static void avr32_dma_free(struct device *dev, size_t size,
>  
>  static dma_addr_t avr32_dma_map_page(struct device *dev, struct page *page,
>  		unsigned long offset, size_t size,
> -		enum dma_data_direction direction, struct dma_attrs *attrs)
> +		enum dma_data_direction direction, unsigned long attrs)
>  {
>  	void *cpu_addr = page_address(page) + offset;
>  
> @@ -152,7 +152,7 @@ static dma_addr_t avr32_dma_map_page(struct device *dev, struct page *page,
>  
>  static int avr32_dma_map_sg(struct device *dev, struct scatterlist *sglist,
>  		int nents, enum dma_data_direction direction,
> -		struct dma_attrs *attrs)
> +		unsigned long attrs)
>  {
>  	int i;
>  	struct scatterlist *sg;
-- 
mvh
Hans-Christian Noren Egtvedt

Re: [RFC v3 44/45] dma-mapping: Remove dma_get_attr

[Hans-Christian Noren Egtvedt]

Around Thu 02 Jun 2016 17:39:46 +0200 or thereabout, Krzysztof Kozlowski wrote:
> After switching DMA attributes to unsigned long it is easier to just
> compare the bits.
> 
> Signed-off-by: Krzysztof Kozlowski <k.kozlowski@samsung.com>
> ---
>  Documentation/DMA-API.txt                      |  4 +--
>  arch/arc/mm/dma.c                              |  4 +--
>  arch/arm/mm/dma-mapping.c                      | 36 ++++++++++++--------------
>  arch/arm/xen/mm.c                              |  4 +--
>  arch/arm64/mm/dma-mapping.c                    | 10 +++----
>  arch/avr32/mm/dma-coherent.c                   |  4 +--

For the AVR32 related change

Acked-by: Hans-Christian Noren Egtvedt <egtvedt@samfundet.no>

>  arch/ia64/sn/pci/pci_dma.c                     | 10 ++-----
>  arch/metag/kernel/dma.c                        |  2 +-
>  arch/mips/mm/dma-default.c                     |  6 ++---
>  arch/openrisc/kernel/dma.c                     |  4 +--
>  arch/parisc/kernel/pci-dma.c                   |  2 +-
>  arch/powerpc/platforms/cell/iommu.c            | 10 +++----
>  drivers/gpu/drm/rockchip/rockchip_drm_gem.c    |  2 +-
>  drivers/iommu/dma-iommu.c                      |  2 +-
>  drivers/media/v4l2-core/videobuf2-dma-contig.c |  2 +-
>  include/linux/dma-mapping.h                    | 13 ----------
>  16 files changed, 46 insertions(+), 69 deletions(-)

<snipp non-AVR32>

> diff --git a/arch/avr32/mm/dma-coherent.c b/arch/avr32/mm/dma-coherent.c
> index fc51f4421933..58610d0df7ed 100644
> --- a/arch/avr32/mm/dma-coherent.c
> +++ b/arch/avr32/mm/dma-coherent.c
> @@ -109,7 +109,7 @@ static void *avr32_dma_alloc(struct device *dev, size_t size,
>  		return NULL;
>  	phys = page_to_phys(page);
>  
> -	if (dma_get_attr(DMA_ATTR_WRITE_COMBINE, attrs)) {
> +	if (attrs & DMA_ATTR_WRITE_COMBINE) {
>  		/* Now, map the page into P3 with write-combining turned on */
>  		*handle = phys;
>  		return __ioremap(phys, size, _PAGE_BUFFER);
> @@ -123,7 +123,7 @@ static void avr32_dma_free(struct device *dev, size_t size,
>  {
>  	struct page *page;
>  
> -	if (dma_get_attr(DMA_ATTR_WRITE_COMBINE, attrs)) {
> +	if (attrs & DMA_ATTR_WRITE_COMBINE) {
>  		iounmap(cpu_addr);
>  
>  		page = phys_to_page(handle);

<snipp non-AVR32>

-- 
mvh
Hans-Christian Noren Egtvedt