Re: [PATCH 6/7] mm, fs: introduce file_operations->post_mmap()

[Dan Williams]

On Wed, Sep 27, 2017 at 8:39 AM, Ross Zwisler
<ross.zwisler@xxxxxxxxxxxxxxx> wrote:
> On Wed, Sep 27, 2017 at 01:35:27PM +0200, Jan Kara wrote:
[..]
>> Hum, this is an interesting option. So do you suggest that filesystems
>> supporting DAX would always setup mappings with VM_MIXEDMAP and without
>> VM_HUGEPAGE and thus we'd get rid of dependency on S_DAX flag in ->mmap?
>> That could actually work. The only possible issue I can see is that
>> VM_MIXEDMAP is still slightly different from normal page mappings and it
>> could have some performance implications - e.g. copy_page_range() does more
>> work on VM_MIXEDMAP mappings but not on normal page mappings.
>
> It looks like having VM_MIXEDMAP always set for filesystems that support DAX
> might affect their memory's NUMA migration in the non-DAX case?
>
> 8e76d4e sched, numa: do not hint for NUMA balancing on VM_MIXEDMAP mappings

Addressed separately here:

c1ef8e2c0235 mm: disable numa migration faults for dax vmas