On 4/24/21 1:12 AM, Dan Williams wrote: > On Thu, Mar 25, 2021 at 4:10 PM Joao Martins <joao.m.martins@xxxxxxxxxx> wrote: >> >> memory_failure_dev_pagemap() at the moment assumes base pages (e.g. >> dax_lock_page()). For pagemap with compound pages fetch the >> compound_head in case we are handling a tail page memory failure. >> >> Currently this is a nop, but in the advent of compound pages in >> dev_pagemap it allows memory_failure_dev_pagemap() to keep working. >> >> Reported-by: Jane Chu <jane.chu@xxxxxxxxxx> >> Signed-off-by: Joao Martins <joao.m.martins@xxxxxxxxxx> >> --- >> mm/memory-failure.c | 2 ++ >> 1 file changed, 2 insertions(+) >> >> diff --git a/mm/memory-failure.c b/mm/memory-failure.c >> index 24210c9bd843..94240d772623 100644 >> --- a/mm/memory-failure.c >> +++ b/mm/memory-failure.c >> @@ -1318,6 +1318,8 @@ static int memory_failure_dev_pagemap(unsigned long pfn, int flags, >> goto out; >> } >> >> + page = compound_head(page); > > Unless / until we do compound pages for the filesystem-dax case, I > would add a comment like: > > /* pages instantiated by device-dax (not filesystem-dax) may be > compound pages */ > I've fixed up with the comment. Thanks!
