mm,hwpoison: try to narrow window race for free pages

Aristeu Rozanski reported that a customer test case started to report
-EBUSY after the hwpoison rework patchset.

There is a race window between spotting a free page and taking it off its
buddy freelist, so it might be that by the time we try to take it off, the
page has been already allocated.

This patch tries to handle such race window by trying to handle the new
type of page again if the page was allocated under us.

Reported-by: Aristeu Rozanski <[email protected]>
Signed-off-by: Oscar Salvador <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Tested-by: Aristeu Rozanski <[email protected]>
Acked-by: Naoya Horiguchi <[email protected]>
Cc: "Aneesh Kumar K.V" <[email protected]>
Cc: Aneesh Kumar K.V <[email protected]>
Cc: Dave Hansen <[email protected]>
Cc: David Hildenbrand <[email protected]>
Cc: Dmitry Yakunin <[email protected]>
Cc: Michal Hocko <[email protected]>
Cc: Mike Kravetz <[email protected]>
Cc: Oscar Salvador <[email protected]>
Cc: Qian Cai <[email protected]>
Cc: Tony Luck <[email protected]>
Link: https://lkml.kernel.org/r/[email protected]
Signed-off-by: Linus Torvalds <[email protected]>

1 files changed, 6 insertions, 1 deletions

diff --git a/mm/memory-failure.c b/mm/memory-failure.c
index 7c63ba9ff6e7..a2184b721fbf 100644
--- a/mm/memory-failure.c
+++ b/mm/memory-failure.c
@@ -1903,6 +1903,7 @@ int soft_offline_page(unsigned long pfn, int flags)
 {
 	int ret;
 	struct page *page;
+	bool try_again = true;
 
 	if (!pfn_valid(pfn))
 		return -ENXIO;
@@ -1918,6 +1919,7 @@ int soft_offline_page(unsigned long pfn, int flags)
 		return 0;
 	}
 
+retry:
 	get_online_mems();
 	ret = get_any_page(page, pfn, flags);
 	put_online_mems();
@@ -1925,7 +1927,10 @@ int soft_offline_page(unsigned long pfn, int flags)
 	if (ret > 0)
 		ret = soft_offline_in_use_page(page);
 	else if (ret == 0)
-		ret = soft_offline_free_page(page);
+		if (soft_offline_free_page(page) && try_again) {
+			try_again = false;
+			goto retry;
+		}
 
 	return ret;
 }