Re: [PATCH v2 7/7] pmem, dax: have direct_access use __pmem annotation

On Sat, Aug 15, 2015 at 2:11 AM, Christoph Hellwig [off-list ref] wrote:

On Fri, Aug 14, 2015 at 09:58:16AM -0700, Dan Williams wrote:

quoted

quoted

I'll merge with your code for v3.

Sounds, let me go rebase the __pfn_t patches on -mm so we'all lined up
and collision free.

I'm doubt that we'll have PFN mapping ready for 4.3.  I'd rather see
Ross series goes first, and move the patch to remove the size argument
from ->direct access [1] over to this series as well.

[1] https://git.kernel.org/cgit/linux/kernel/git/djbw/nvdimm.git/commit/?h=pfn&id=8e15e69fb9e61ac563c5a7ffd9dd9a7b545cced3

Yes, let's do it.  The need for DAX persistence guarantees is a higher
priority than solving the DAX vs PMEM unbind bug.  Especially if we
can defer __pfn_t and get some basic struct page mapping into 4.3 as
well.