On Thu, Jan 27, 2022 at 08:40:51PM +0800, Shiyang Ruan wrote:
> +static int mf_generic_kill_procs(unsigned long long pfn, int flags,
> + struct dev_pagemap *pgmap)
> +{
> + struct page *page = pfn_to_page(pfn);
> + LIST_HEAD(to_kill);
> + dax_entry_t cookie;
> +
> + /*
> + * Prevent the inode from being freed while we are interrogating
> + * the address_space, typically this would be handled by
> + * lock_page(), but dax pages do not use the page lock. This
> + * also prevents changes to the mapping of this pfn until
> + * poison signaling is complete.
> + */
> + cookie = dax_lock_page(page);
> + if (!cookie)
> + return -EBUSY;
> +
> + if (hwpoison_filter(page))
> + return 0;
> +
> + if (pgmap->type == MEMORY_DEVICE_PRIVATE) {
> + /*
> + * TODO: Handle HMM pages which may need coordination
> + * with device-side memory.
> + */
> + return -EBUSY;
> + }
> +
> + /*
> + * Use this flag as an indication that the dax page has been
> + * remapped UC to prevent speculative consumption of poison.
> + */
> + SetPageHWPoison(page);
> +
> + /*
> + * Unlike System-RAM there is no possibility to swap in a
> + * different physical page at a given virtual address, so all
> + * userspace consumption of ZONE_DEVICE memory necessitates
> + * SIGBUS (i.e. MF_MUST_KILL)
> + */
> + flags |= MF_ACTION_REQUIRED | MF_MUST_KILL;
> + collect_procs(page, &to_kill, true);
> +
> + unmap_and_kill(&to_kill, pfn, page->mapping, page->index, flags);
> + dax_unlock_page(page, cookie);
> + return 0;
> +}
There's an assumption here that fsdax only has order-0 pages.
pfn_to_page() is going to return the precise page for that pfn, but then
page->mapping and page->index are not valid for tail pages.
I'm currently trying to folio-ise memory-failure.c, and it is not
going well! There are several places where it's hard to tell
what should happen.
