> We can't do that right now, anyway.
It seemed to work on XFS when we tested it. Undoutedly after your
patch it won't work anymore.
> If the page is not in use, we don't care about it after this patch
> set is applied - the page is either active in a buffer or it has been
> freed. If it is in use, then we'll shut the filesystem down if we
> detect the memory corruption just like we currently do. Hence I
> don't see any regression here.
I think you're confusing the memory_failure() HWPoison path with some XFS
internal checking. I don't think XFS has any HWPoison checking on its own.
> As it is, there is no way for the filesytem to be notified about
> such failures on active pages in buffers, so in reality we can't
> reliably detect them so there is little point in trying to recover
> from such errors.
Well it works today if the page is in pagecache. memory-failure will
just remove it transparently.
In principle you can check for HWPoison manually yourself, but I'm not sure
that is a good way to do it.
ak@xxxxxxxxxxxxxxx -- Speaking for myself only.