Re: [PATCH 04/15] mm: remove the pgmap field from struct hmm_vma_walk

From: Jason Gunthorpe
Date: Fri Aug 16 2019 - 13:28:49 EST


On Fri, Aug 16, 2019 at 10:21:41AM -0700, Dan Williams wrote:

> > We can do a get_dev_pagemap inside the page_walk and touch the pgmap,
> > or we can do the 'device mutex && retry' pattern and touch the pgmap
> > in the driver, under that lock.
> >
> > However in all cases the current get_dev_pagemap()'s in the page walk
> > are not necessary, and we can delete them.
>
> Yes, as long as 'struct page' instances resulting from that lookup are
> not passed outside of that lock.

Indeed.

Also, I was reflecting over lunch that the hmm_range_fault should only
return DEVICE_PRIVATE pages for the caller's device (see other thread
with HCH), and in this case, the caller should also be responsible to
ensure that the driver is not calling hmm_range_fault at the same time
it is deleting it's own DEVICE_PRIVATE mapping - ie by fencing its
page fault handler.

This does not apply to PCI_P2PDMA, but, lets see how that looks when
we get there.

So the whole thing seems pretty safe.

Jason