Re: [PATCH] [media] dma-mapping: enable no mmu support in dma_common_mmap

[Marek Szyprowski <m.szyprowski@samsung.com>]

Hello,

On 3/6/2013 12:40 AM, Scott Jiang wrote:
> No MMU systems also make use of this function to do mmap.
>
> Signed-off-by: Scott Jiang <scott.jiang.linux@gmail.com>
> ---
>   drivers/base/dma-mapping.c |    2 --
>   1 files changed, 0 insertions(+), 2 deletions(-)
>
> diff --git a/drivers/base/dma-mapping.c b/drivers/base/dma-mapping.c
> index 0ce39a3..ae655b2 100644
> --- a/drivers/base/dma-mapping.c
> +++ b/drivers/base/dma-mapping.c
> @@ -245,7 +245,6 @@ int dma_common_mmap(struct device *dev, struct vm_area_struct *vma,
>   		    void *cpu_addr, dma_addr_t dma_addr, size_t size)
>   {
>   	int ret = -ENXIO;
> -#ifdef CONFIG_MMU
>   	unsigned long user_count = (vma->vm_end - vma->vm_start) >> PAGE_SHIFT;
>   	unsigned long count = PAGE_ALIGN(size) >> PAGE_SHIFT;
>   	unsigned long pfn = page_to_pfn(virt_to_page(cpu_addr));
> @@ -262,7 +261,6 @@ int dma_common_mmap(struct device *dev, struct vm_area_struct *vma,
>   				      user_count << PAGE_SHIFT,
>   				      vma->vm_page_prot);
>   	}
> -#endif	/* CONFIG_MMU */
>   
>   	return ret;
>   }

I really have no experience with NO-MMU kernels, could you explain a bit 
more how this
is useful for handling mmap on such systems? How remap_pfn_range() is 
handled on no-mmu
systems?

I've thought that mmap on no-mmu systems is silently replaced by a call to
get_unmapped_area(), but it looks that there is still a call to mmap 
function.

Best regards
-- 
Marek Szyprowski
Samsung Poland R&D Center