vmalloc_to_page is contiguous ?

vmalloc_to_page is contiguous ?

[Ranran]

Hello,

Does vmalloc_to_page() returns contiguous physical memory ?
I ask becuase I see that its return value is used for DMA and DMA uses
contiguous memory.
see here:
https://elixir.bootlin.com/linux/latest/source/drivers/mtd/nand/onenand/samsung.c#L657
Yet, in vmalloc_to_page implementation:
https://elixir.bootlin.com/linux/v4.5/source/mm/vmalloc.c#L234
it does not seem that it creates a contiguous allocation.

Thank you,
ranran

_______________________________________________
Kernelnewbies mailing list
Kernelnewbies@kernelnewbies.org
https://lists.kernelnewbies.org/mailman/listinfo/kernelnewbies

Re: vmalloc_to_page is contiguous ?

[Rik van Riel]

[-- Attachment #1.1: Type: text/plain, Size: 800 bytes --]

On Mon, 2019-03-18 at 12:55 +0200, Ranran wrote:
> Hello,
> 
> Does vmalloc_to_page() returns contiguous physical memory ?
> I ask becuase I see that its return value is used for DMA and DMA
> uses
> contiguous memory.
> see here:
> 
https://elixir.bootlin.com/linux/latest/source/drivers/mtd/nand/onenand/samsung.c#L657
> Yet, in vmalloc_to_page implementation:
> https://elixir.bootlin.com/linux/v4.5/source/mm/vmalloc.c#L234
> it does not seem that it creates a contiguous allocation.

That is correct, vmalloc_to_page does not create anything.

All it does is find (and return) the page address of
a single vmalloc address.

If you have a vmalloc buffer larger than PAGE_SIZE, you
will have to call vmalloc_to_page to find each individual
page.

-- 
All Rights Reversed.

[-- Attachment #1.2: This is a digitally signed message part --]
[-- Type: application/pgp-signature, Size: 488 bytes --]

[-- Attachment #2: Type: text/plain, Size: 170 bytes --]

_______________________________________________
Kernelnewbies mailing list
Kernelnewbies@kernelnewbies.org
https://lists.kernelnewbies.org/mailman/listinfo/kernelnewbies