On Tue, Feb 09, 2021 at 04:17:38PM -0500, Jerome Glisse wrote: > > > Yes, I would like to avoid the long term pin constraints as well if possible I > > > just haven't found a solution yet. Are you suggesting it might be possible to > > > add a callback in the page migration logic to specially deal with moving these > > > pages? > > > > How would migration even find the page? > > Migration can scan memory from physical address (isolate_migratepages_range()) > So the CPU mapping is not the only path to get to a page. I mean find out that the page is now owned by the GPU driver to tell it that it needs to migrate that reference. Normally that would go through the VMA to the mmu notifier, but with the page removed from the normal VMA it can't get to a mmu notifier or the GPU driver. Jason
