[BUG] pgprot_noncached() is -NOT- safe for mapping vmalloc buffers into userspace

[Benjamin Herrenschmidt]

On Fri, 2011-03-25 at 09:01 +0100, Takashi Iwai wrote:
> > In fact, cache coherent architectures often don't support mapping memory
> > uncached -at-all- so something like snd_pcm_lib_mmap_noncached()
> > shouldn't exist, or at least be under arch control. There's no case
> > where it's "always safe". There will almost always be a cache alias in
> > the linear mapping unless special arch specific sauce has been applied.
> 
> I see.  I'll take your removal patch.
> It should be applied to stable kernel, too, right?

I suppose so :-)

> > Now, there's another problem on top of that, which is that
> > snd_pcm_default_mmap() will not work properly the "other way around" on
> > powerpc, where the mapping -needs- to be uncached bcs you are running on
> > a non cache coherent embedded CPU and trying to mmap DMA memory, but
> > that's something that needs fixing inside powerpc by properly defining
> > dma_mmap_coherent() & ARCH_HAS_DMA_MMAP_COHERENT (I thought we had added
> > it a while back but it's not upstream, patch must have got lost).
> 
> Mea culpa.  I moved to a different team and have had little time for
> the upstream development since then...

That's ok, I did a new one, which you have noticed already anyways :-)

> > We
> > must also make sure we don't go down that path for vmalloc memory
> > though.
>
> Yes.

I haven't actually checked, but I assume that the test

substream->dma_buffer.dev.type == SNDRV_DMA_TYPE_DEV

In snd_pcm_default_mmap() takes care of that, please correct me if
I'm wrong in which case we'll need something else there.

> Your patch looks good.  Thanks for taking care of this! 

Are you taking care of sending it upstream ?

Cheers,
Ben.