Re: [PATCH v1] mm: hwpoison: disable memory error handling on 1GB hugepage

[Naoya Horiguchi]

My apology, I forgot to CC to the mailing lists.

On Mon, Jan 29, 2018 at 03:28:03PM +0900, Naoya Horiguchi wrote:
> Recently the following BUG was reported:
> 
>     Injecting memory failure for pfn 0x3c0000 at process virtual address 0x7fe300000000
>     Memory failure: 0x3c0000: recovery action for huge page: Recovered
>     BUG: unable to handle kernel paging request at ffff8dfcc0003000
>     IP: gup_pgd_range+0x1f0/0xc20
>     PGD 17ae72067 P4D 17ae72067 PUD 0
>     Oops: 0000 [#1] SMP PTI
>     ...
>     CPU: 3 PID: 5467 Comm: hugetlb_1gb Not tainted 4.15.0-rc8-mm1-abc+ #3
>     Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.9.3-1.fc25 04/01/2014
> 
> You can easily reproduce this by calling madvise(MADV_HWPOISON) twice on
> a 1GB hugepage. This happens because get_user_pages_fast() is not aware
> of a migration entry on pud that was created in the 1st madvise() event.
> 
> I think that conversion to pud-aligned migration entry is working,
> but other MM code walking over page table isn't prepared for it.
> We need some time and effort to make all this work properly, so
> this patch avoids the reported bug by just disabling error handling
> for 1GB hugepage.
> 
> Signed-off-by: Naoya Horiguchi <n-horiguchi@xxxxxxxxxxxxx>
> ---
>  include/linux/mm.h  | 1 +
>  mm/memory-failure.c | 7 +++++++
>  2 files changed, 8 insertions(+)
> 
> diff --git v4.15-rc8-mmotm-2018-01-18-16-31/include/linux/mm.h v4.15-rc8-mmotm-2018-01-18-16-31_patched/include/linux/mm.h
> index 63f7ba1..166864e 100644
> --- v4.15-rc8-mmotm-2018-01-18-16-31/include/linux/mm.h
> +++ v4.15-rc8-mmotm-2018-01-18-16-31_patched/include/linux/mm.h
> @@ -2607,6 +2607,7 @@ enum mf_action_page_type {
>  	MF_MSG_POISONED_HUGE,
>  	MF_MSG_HUGE,
>  	MF_MSG_FREE_HUGE,
> +	MF_MSG_GIGANTIC,
>  	MF_MSG_UNMAP_FAILED,
>  	MF_MSG_DIRTY_SWAPCACHE,
>  	MF_MSG_CLEAN_SWAPCACHE,
> diff --git v4.15-rc8-mmotm-2018-01-18-16-31/mm/memory-failure.c v4.15-rc8-mmotm-2018-01-18-16-31_patched/mm/memory-failure.c
> index d530ac1..c497588 100644
> --- v4.15-rc8-mmotm-2018-01-18-16-31/mm/memory-failure.c
> +++ v4.15-rc8-mmotm-2018-01-18-16-31_patched/mm/memory-failure.c
> @@ -508,6 +508,7 @@ static const char * const action_page_types[] = {
>  	[MF_MSG_POISONED_HUGE]		= "huge page already hardware poisoned",
>  	[MF_MSG_HUGE]			= "huge page",
>  	[MF_MSG_FREE_HUGE]		= "free huge page",
> +	[MF_MSG_GIGANTIC]		= "gigantic page",
>  	[MF_MSG_UNMAP_FAILED]		= "unmapping failed page",
>  	[MF_MSG_DIRTY_SWAPCACHE]	= "dirty swapcache page",
>  	[MF_MSG_CLEAN_SWAPCACHE]	= "clean swapcache page",
> @@ -1090,6 +1091,12 @@ static int memory_failure_hugetlb(unsigned long pfn, int trapno, int flags)
>  		return 0;
>  	}
>  
> +	if (hstate_is_gigantic(page_hstate(head))) {
> +		action_result(pfn, MF_MSG_GIGANTIC, MF_IGNORED);
> +		res = -EBUSY;
> +		goto out;
> +	}
> +
>  	if (!hwpoison_user_mappings(p, pfn, trapno, flags, &head)) {
>  		action_result(pfn, MF_MSG_UNMAP_FAILED, MF_IGNORED);
>  		res = -EBUSY;
> -- 
> 2.7.0
> 
>