Techwizz/kernel-fxtec-pro1x
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity

[AVR32] Implement dma_{alloc,free}_writecombine()

Browse source 
Implement dma_alloc_writecombine() and its dma_free_writecombine()
counterpart. These will do basically the same thing as
dma_alloc_coherent() except that the virtual mapping will allow
write buffering, causing better performance for certain use cases
like frame buffers.

The same API is already available on ARM.

Signed-off-by: Haavard Skinnemoen <hskinnemoen@atmel.com>
This commit is contained in:
Haavard Skinnemoen 2007-03-20 14:41:13 +01:00
parent e89b064a4f
commit a492dbb9e3
1 changed files with 8 additions and 4 deletions

12
arch/avr32/mm/dma-coherent.c
View file

@ -112,16 +112,21 @@ void dma_free_coherent(struct device *dev, size_t size,
}
EXPORT_SYMBOL(dma_free_coherent);
#if 0
void *dma_alloc_writecombine(struct device *dev, size_t size,
dma_addr_t *handle, gfp_t gfp)
{
struct page *page;
dma_addr_t phys;
page = __dma_alloc(dev, size, handle, gfp);
if (!page)
return NULL;
phys = page_to_phys(page);
*handle = phys;
/* Now, map the page into P3 with write-combining turned on */
return __ioremap(page_to_phys(page), size, _PAGE_BUFFER);
return __ioremap(phys, size, _PAGE_BUFFER);
}
EXPORT_SYMBOL(dma_alloc_writecombine);
@ -132,8 +137,7 @@ void dma_free_writecombine(struct device *dev, size_t size,
iounmap(cpu_addr);
page = bus_to_page(handle);
page = phys_to_page(handle);
__dma_free(dev, size, page, handle);
}
EXPORT_SYMBOL(dma_free_writecombine);
#endif