[PATCHv2] arm64: Add atomic pool for non-coherent and CMA allocaitons.

On Wed, Jun 04, 2014 at 01:30:18AM +0100, Laura Abbott wrote:

On 6/3/2014 6:28 AM, Will Deacon wrote:

quoted

Hi Laura,

On Mon, Jun 02, 2014 at 09:03:52PM +0100, Laura Abbott wrote:

quoted

Neither CMA nor noncoherent allocations support atomic allocations.
Add a dedicated atomic pool to support this.

Change-Id: I46c8fdffe5e0687403d42b37643137c8cf344259
Signed-off-by: Laura Abbott <redacted>
---

v2: Various bug fixes pointed out by David and Ritesh (CMA dependency, swapping
coherent, noncoherent). I'm still not sure how to address the devicetree
suggestion by Will [1][2]. I added the devicetree mailing list this time around
to get more input on this.

[1] http://lists.infradead.org/pipermail/linux-arm-kernel/2014-April/249180.html
[2] http://lists.infradead.org/pipermail/linux-arm-kernel/2014-April/249528.html

Perhaps that can be done later then, since from what you're saying, we need
the command-line option either way? Have you looked at how this fits in with
the iommu-helper work from Ritesh? We could put the parameter parsing in
there too.

This doesn't seem to overlap with Ritesh's work. The atomic mapping is still
handled in the arm specific code so I assume it would be handled in the arm64
specific code as well. Another question might be is if it would be useful to
make the atomic code common somehow between arm and arm64.

Yeah, that's what I was alluding to. The more of this code that can be
shared between architectures, the better.

Will