Re: USB transfer_buffer allocations on 64bit systems

[Chris Wright <chrisw@sous-sol.org>]

* Alan Stern (stern@rowland.harvard.edu) wrote:
> Since using mem=4096M or GFP_DMA stopped the symptoms, it seems very 
> likely that a buffer is getting allocated above the 4 GB line and not 
> bounced or IOMMU-mapped correctly.
> 
> David, do you have anything to suggest?  Any ways to check for IOMMU or 
> related errors?

Well if the IOMMU is enabled, dmesg will show you if you're getting DMA
faults due to IOMMU.  Doesn't sound like that's the case.

> The problem, in short, is that USB audio doesn't work properly when
> Pedro boots a 64-bit kernel on his 4-GB machine.  With a 32-bit kernel
> it works okay, and it also works if we use dma_alloc_coherent().  The
> host controller is limited to 32-bit DMA, and the DMA addresses
> generated by dma_map_single() appear to be normal.

So dma_map_single is the case that's failing, but you think the mask is
correct?  What about the direction?

> At the moment we don't even know if this is caused by a bug in the 
> kernel or a bug in Pedro's hardware.  But he has observed the same 
> problem on two different machines, both using the ICH9 chipset.

Is the IOMMU enabled?

$ dmesg | grep -e DMAR -e IOMMU

If it's on, you can boot w/out (intel_iommu=off) or in passthrough mode
(intel_iommu=on iommu=pt) and see if that makes a difference.

If it's not on (but there) you can enable it (intel_iommu=on) and look
for DMA faults (pointing to driver bug).

thanks,
-chris