On Tue, Aug 26, 2008 at 05:25:24PM +0200, Takashi Iwai wrote:
> Now updated my git tree:
> I'll post each patch again if preferred.
#define SNDRV_DMA_TYPE_DEV_SG 3 /* generic device SG-buffer */
+#define SNDRV_DMA_TYPE_DEV_SG SNDRV_DMA_TYPE_DEV /* no SG-buf support */
I had to look at a previous patch to figure out CONFIG_SND_COHERENT_DMA
is an arch dependent flag:
+ def_bool y
+ depends on !PPC32 || !NOT_COHERENT_CACHE
+ depends on !ARM
+ depends on !MIPS
+ depends on !PARISC
In general, I don't expect this to be a compile time option.
I'm wondering if extending the DMA API to provide an
interface for user space to also be DMA coherent.
Maybe something to talk about at Linux Plumbers Conf
or kernel summit...
> Do you guys see any pending issues? I'd love to merge these patches
> into the upstream for 2.6.28.
SPARC/SPARC64 usually falls into the same category as parisc/mips.
> To get things clear -- I don't intend to fix the problem of mmap on
> every non-coherent platform perfectly (yet). Instead, this patch
> series is intended to fix the current behavior, at least, for the
> sound drivers not to crash unconditionally. It provides a (minimal)
> way to mmap the pages taken via dma_alloc_coherent().
*nod* that's reasonable.
> To unsubscribe from this list: send the line "unsubscribe linux-parisc" in
> the body of a message to firstname.lastname@example.org
> More majordomo info at http://vger.kernel.org/majordomo-info.html