Re: [PATCH] swiotlb: enlarge iotlb buffer on demand

[Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>]

On Saturday 31 July 2010 23:03:11 FUJITA Tomonori wrote:
> On Fri, 30 Jul 2010 21:07:06 -0400
>
> Konrad Rzeszutek Wilk <konrad@kernel.org> wrote:
> > I took your patch and was trying to fit it over the
> > stable/swiotlb-0.8.4 branch and when I did so a found couple of things..
> >
> > > > @@ -215,14 +222,14 @@ swiotlb_late_init_with_default_size(size_t
> > > > default_size) bytes = io_tlb_nslabs << IO_TLB_SHIFT;
> >
> > You should also initialize the __io_tlb_start array first:
>
> Yeah, I know. As I wrote, this patchset breaks IA64.
>
> I really merge to swiotlb's two memory allocator mechanisms
> (swiotlb_init_with_default_size and
> swiotlb_late_init_with_default_size). I need to look at the x86 memory
> boot code after memblock surgery finishes.

<nods>
>
> And as you know, I've not fixed the error path and swiotlb_free. I'll
> do later if people are not against swiotlb dynamic allocation.

It looks to me like it would be a good patch.

I am curious about the handling of the -ENOMEM stage. Naturally we would 
return an error the device - are the most common ones (ahci, r8169, 
ata_piix - those that are DMA_32) equipped to deal with unavailable memory?