On Mon, 17 Nov 2008 10:02:59 +0100

Ingo Molnar  wrote:
>

quoted text
> * FUJITA Tomonori  wrote:

>

> > On Mon, 17 Nov 2008 09:15:26 +0100

> > Ingo Molnar  wrote:

> >

> > >

> > > * FUJITA Tomonori  wrote:

> > >

> > > > This patch fixes swiotlb to use dev->coherent_dma_mask in

> > > > alloc_coherent. Currently, swiotlb uses dev->dma_mask in

> > > > alloc_coherent but alloc_coherent is supposed to use

> > > > coherent_dma_mask. It could break drivers that uses smaller

> > > > coherent_dma_mask than dma_mask (though the current code works for

> > > > the majority that use the same mask for coherent_dma_mask and

> > > > dma_mask).

> > > >

> > > > Signed-off-by: FUJITA Tomonori

> > > > ---

> > > >  lib/swiotlb.c |   10 +++++++---

> > > >  1 files changed, 7 insertions(+), 3 deletions(-)

> > >

> > > Applied it with the changelog below to tip/core/urgent, thanks!

> > >

> > > I also flagged it for v2.6.28 inclusion. This bug was caused by the

> > > removal of the GFP_DMA hack in swiotlb_alloc_coherent() in this cycle.

> > > I havent seen it actually reported anywhere - have you perhaps?Or have

> > > you found this via code review?

> >

> > This wasn't introduced by the removal of the GFP_DMA hack. It has

> > been for ages, I think.

>

> Yeah, what i mean is that our GFP_DMA hack (which we indeed had for

> years) definitely _hid_ the problem: on x86 for example it limits

> coherent DMA buffers into the DMA zone: the first 16 MB of RAM.

Ah, I see. I misunderstood what you meant.
> ( Other platforms are pretty narrow about GFP_DMA too - it implies at

quoted text
>   least DMA32 which is in practice often the real limit for

>   cache-coherent DMA addresses. )

>

> So the removal of GFP_DMA flag from coherent allocations exposed us to

> this long-standing (but hidden) problem.

Yes, you are right. But it's pretty hard to hit this bug since x86_64

rarely uses swiotlb_alloc_coherent. pci-swiotlb_64.c tries

dma_generic_alloc_coherent first and then, only if it fails, it uses

swiotlb_alloc_coherent. dma_generic_alloc_coherent usually succeeds.
Before removing the GFP_DMA hack (the dma_alloc_coherent rewrite),

x86_64 rarely used swiotlb_alloc_coherent too because

swiotlb_alloc_coherent was used only if dma_alloc_coherent fails.

This issue was hidden by dma_alloc_coherent and GFP_DMA hack.

--

unsubscribe notice
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in

the body of a message to majordomo@vger.kernel.org

More majordomo info at  http://vger.kernel.org/majordomo-info.html

Please read the FAQ at  http://www.tux.org/lkml/