[RFC v1 1/3] mm/mmu_notifier: Add a new notifier for mapping updates (new pages)
Jason Gunthorpe
jgg at nvidia.com
Wed Aug 30 16:02:17 UTC 2023
On Mon, Aug 28, 2023 at 04:38:01AM +0000, Kasireddy, Vivek wrote:
> Turns out, calling hmm_range_fault() from the invalidate callback was indeed
> a problem and the reason why new pages were not faulted-in. In other words,
> it looks like the invalidate callback is not the right place to invoke hmm_range_fault()
> as the PTEs may not have been cleared.
Yes, that is correct, you can't even do the locking properly if you
invoke things like this. mmu_interval_read_retry() should continually
fail while you are within the invalidate callback.
Jason
More information about the dri-devel
mailing list