On Mon, 24 Jan 2022, Shakeel Butt wrote:
> On an overcommitted system which is running multiple workloads of
> varying priorities, it is preferred to trigger an oom-killer to kill a
> low priority workload than to let the high priority workload receiving
> ENOMEMs. On our memory overcommitted systems, we are seeing a lot of
> ENOMEMs instead of oom-kills because io_uring_setup callchain is using
> __GFP_NORETRY gfp flag which avoids the oom-killer. Let's remove it and
> allow the oom-killer to kill a lower priority job.
>
What is the size of the allocations that io_mem_alloc() is doing?
If get_order(size) > PAGE_ALLOC_COSTLY_ORDER, then this will fail even
without the __GFP_NORETRY. To make the guarantee that workloads are not
receiving ENOMEM, it seems like we'd need to guarantee that allocations
going through io_mem_alloc() are sufficiently small.
(And if we're really serious about it, then even something like a
BUILD_BUG_ON().)
> Signed-off-by: Shakeel Butt <shakeelb@xxxxxxxxxx>
> ---
> fs/io_uring.c | 5 ++---
> 1 file changed, 2 insertions(+), 3 deletions(-)
>
> diff --git a/fs/io_uring.c b/fs/io_uring.c
> index e54c4127422e..d9eeb202363c 100644
> --- a/fs/io_uring.c
> +++ b/fs/io_uring.c
> @@ -8928,10 +8928,9 @@ static void io_mem_free(void *ptr)
>
> static void *io_mem_alloc(size_t size)
> {
> - gfp_t gfp_flags = GFP_KERNEL | __GFP_ZERO | __GFP_NOWARN | __GFP_COMP |
> - __GFP_NORETRY | __GFP_ACCOUNT;
> + gfp_t gfp = GFP_KERNEL_ACCOUNT | __GFP_ZERO | __GFP_NOWARN | __GFP_COMP;
>
> - return (void *) __get_free_pages(gfp_flags, get_order(size));
> + return (void *) __get_free_pages(gfp, get_order(size));
> }
>
> static unsigned long rings_size(unsigned sq_entries, unsigned cq_entries,
> --
> 2.35.0.rc0.227.g00780c9af4-goog
>
>
>
