On Fri 24-08-18 14:27:33, Jan Stancek wrote:
> Since the rework done in a49bd4d71637 ("mm, numa: rework do_pages_move"),
> move_pages() can return also positive value.
Yeah, the original semantic of this syscall was quite fuzzy. The
changelog says
"
Some errors are reported via status parameter but migration failures
are not even though the original `reason' argument suggests there was
an intention to do so. From a quick look into git history this never
worked. I have tried to keep the semantic unchanged.
"
> Signed-off-by: Jan Stancek <jstancek@xxxxxxxxxx>
Thanks for the update
Acked-by: Michal Hocko <mhocko@xxxxxxxx>
> ---
> man2/move_pages.2 | 6 ++++--
> 1 file changed, 4 insertions(+), 2 deletions(-)
>
> https://github.com/linux-test-project/ltp/issues/388
>
> diff --git a/man2/move_pages.2 b/man2/move_pages.2
> index f2c4f7f7a269..35cf6de1ba78 100644
> --- a/man2/move_pages.2
> +++ b/man2/move_pages.2
> @@ -157,9 +157,11 @@ The page is not present.
> .B -ENOMEM
> Unable to allocate memory on target node.
> .SH RETURN VALUE
> -On success
> +On success (since 4.17)
> .BR move_pages ()
> -returns zero.
> +returns the number of pages that could not be moved
> +(i.e., a return of zero means that all pages were successfully moved,
> +older kernels return 0).
> .\" FIXME . Is the following quite true: does the wrapper in numactl
> .\" do the right thing?
> On error, it returns \-1, and sets
> --
> 1.8.3.1
>
--
Michal Hocko
SUSE Labs
