[PATCH 03/10] x86: add initialization code for DMA-API debugging

FUJITA Tomonori fujita.tomonori at lab.ntt.co.jp
Sat Nov 22 04:28:00 PST 2008


On Sat, 22 Nov 2008 19:16:05 +0900
FUJITA Tomonori <fujita.tomonori at lab.ntt.co.jp> wrote:

> On Sat, 22 Nov 2008 10:40:32 +0100
> Joerg Roedel <joro at 8bytes.org> wrote:
> 
> > > > +static int hash_fn(struct dma_debug_entry *entry)
> > > > +{
> > > > +	/*
> > > > +	 * Hash function is based on the dma address.
> > > > +	 * We use bits 20-27 here as the index into the hash
> > > > +	 */
> > > > +	BUG_ON(entry->dev_addr == bad_dma_address);
> > > 
> > > 'bad_dma_address' is x86 specific. You already found it though.
> > 
> > Interesting. Is there another value for dma_addr_t which drivers can
> > check for to find out if a dma-api operation failed?
> 
> They are architecture dependent. But only X86 uses a variable because
> of GART and swiotlb, I think.

BTW, this code doesn't work even on x86 (swiotlb). dma_map_error
should be used, which is an architecture-independent way to test a dma
address.


More information about the iommu mailing list