Aristeu Rozanski reported that a customer test case started to report -EBUSY after the hwpoison report 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. After this patch, Aristeu said the test cases work properly. Signed-off-by: Oscar Salvador <[email protected]> Reported-by: Aristeu Rozanski <[email protected]> --- mm/memory-failure.c | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-) diff --git a/mm/memory-failure.c b/mm/memory-failure.c index db61bdee9734..a2ccd3ba4015 100644 --- a/mm/memory-failure.c +++ b/mm/memory-failure.c @@ -1917,6 +1917,7 @@ int soft_offline_page(unsigned long pfn, int flags) { int ret; struct page *page; + bool try_again = true; if (!pfn_valid(pfn)) return -ENXIO; @@ -1932,6 +1933,7 @@ 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(); @@ -1939,7 +1941,10 @@ 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; } -- 2.26.2
