Re: Fwd: vmalloc error: btrfs-delalloc btrfs_work_helper [btrfs] in kernel 6.3.x

[Forza <forza@tnonline.net>]

---- From: Uladzislau Rezki <urezki@gmail.com> -- Sent: 2023-05-23 - 12:28 ----

> On Tue, May 23, 2023 at 08:52:21AM +0700, Bagas Sanjaya wrote:
>> On Mon, May 22, 2023 at 09:04:05PM +0200, Forza wrote:
>> > I have a similar experience with kernel 6.3 where vmalloc fails in a similar way. I was able to reproduce it in a QEMU VM as well as on my system. 
>> > 
>> > https://lore.kernel.org/all/d11418b6-38e5-eb78-1537-c39245dc0b78@tnonline.net/T/
>> 
>> Thanks for your similar report. Telling regzbot about it:
>> 
>> #regzbot link: https://lore.kernel.org/all/d11418b6-38e5-eb78-1537-c39245dc0b78@tnonline.net/
>> 
> It is good that you can reproduce it. Could you please test below patch?

Yes, applied it to my test VM and will let it run over night to see how it turns out. I'll post again tomorrow. 

Thanks. 
> 
> <snip>
> diff --git a/mm/vmalloc.c b/mm/vmalloc.c
> index 31ff782d368b..7a06452f7807 100644
> --- a/mm/vmalloc.c
> +++ b/mm/vmalloc.c
> @@ -2957,14 +2957,18 @@ vm_area_alloc_pages(gfp_t gfp, int nid,
>                         page = alloc_pages(alloc_gfp, order);
>                 else
>                         page = alloc_pages_node(nid, alloc_gfp, order);
> +
>                 if (unlikely(!page)) {
> -                       if (!nofail)
> -                               break;
> +                       if (nofail)
> +                               alloc_gfp |= __GFP_NOFAIL;
> 
> -                       /* fall back to the zero order allocations */
> -                       alloc_gfp |= __GFP_NOFAIL;
> -                       order = 0;
> -                       continue;
> +                       /* Fall back to the zero order allocations. */
> +                       if (order || nofail) {
> +                               order = 0;
> +                               continue;
> +                       }
> +
> +                       break;
>                 }
> 
>                 /*
> <snip>
> 
> Thanks!
> 
> --
> Uladzislau Rezki