* Dan Williams <[email protected]> wrote:
> Kirill points out that the calls to {get,put}_dev_pagemap() can be
> removed from the mm fast path if we take a single get_dev_pagemap()
> reference to signify that the page is alive and use the final put of the
> page to drop that reference.
>
> This does require some care to make sure that any waits for the
> percpu_ref to drop to zero occur *after* devm_memremap_page_release(),
> since it now maintains its own elevated reference.
>
> Cc: Ingo Molnar <[email protected]>
> Cc: Jérôme Glisse <[email protected]>
> Cc: Andrew Morton <[email protected]>
> Reviewed-by: Logan Gunthorpe <[email protected]>
> Suggested-by: Kirill Shutemov <[email protected]>
> Tested-by: Kirill Shutemov <[email protected]>
> Signed-off-by: Dan Williams <[email protected]>
This changelog is lacking an explanation about how this solves the crashes you
were seeing.
Thanks,
Ingo
