Re: [PATCH] mm/gup: continue VM_FAULT_RETRY processing event for pre-faults

[Mike Rapoport <rppt@linux.ibm.com>]

Hi,

Any comments on this?

On Tue, May 14, 2019 at 05:29:55PM +0300, Mike Rapoport wrote:
> When get_user_pages*() is called with pages = NULL, the processing of
> VM_FAULT_RETRY terminates early without actually retrying to fault-in all
> the pages.
> 
> If the pages in the requested range belong to a VMA that has userfaultfd
> registered, handle_userfault() returns VM_FAULT_RETRY *after* user space
> has populated the page, but for the gup pre-fault case there's no actual
> retry and the caller will get no pages although they are present.
> 
> This issue was uncovered when running post-copy memory restore in CRIU
> after commit d9c9ce34ed5c ("x86/fpu: Fault-in user stack if
> copy_fpstate_to_sigframe() fails").
> 
> After this change, the copying of FPU state to the sigframe switched from
> copy_to_user() variants which caused a real page fault to get_user_pages()
> with pages parameter set to NULL.
> 
> In post-copy mode of CRIU, the destination memory is managed with
> userfaultfd and lack of the retry for pre-fault case in get_user_pages()
> causes a crash of the restored process.
> 
> Making the pre-fault behavior of get_user_pages() the same as the "normal"
> one fixes the issue.
> 
> Fixes: d9c9ce34ed5c ("x86/fpu: Fault-in user stack if copy_fpstate_to_sigframe() fails")
> Signed-off-by: Mike Rapoport <rppt@linux.ibm.com>
> ---
>  mm/gup.c | 15 ++++++++-------
>  1 file changed, 8 insertions(+), 7 deletions(-)
> 
> diff --git a/mm/gup.c b/mm/gup.c
> index 91819b8..c32ae5a 100644
> --- a/mm/gup.c
> +++ b/mm/gup.c
> @@ -936,10 +936,6 @@ static __always_inline long __get_user_pages_locked(struct task_struct *tsk,
>  			BUG_ON(ret >= nr_pages);
>  		}
>  
> -		if (!pages)
> -			/* If it's a prefault don't insist harder */
> -			return ret;
> -
>  		if (ret > 0) {
>  			nr_pages -= ret;
>  			pages_done += ret;
> @@ -955,8 +951,12 @@ static __always_inline long __get_user_pages_locked(struct task_struct *tsk,
>  				pages_done = ret;
>  			break;
>  		}
> -		/* VM_FAULT_RETRY triggered, so seek to the faulting offset */
> -		pages += ret;
> +		/*
> +		 * VM_FAULT_RETRY triggered, so seek to the faulting offset.
> +		 * For the prefault case (!pages) we only update counts.
> +		 */
> +		if (likely(pages))
> +			pages += ret;
>  		start += ret << PAGE_SHIFT;
>  
>  		/*
> @@ -979,7 +979,8 @@ static __always_inline long __get_user_pages_locked(struct task_struct *tsk,
>  		pages_done++;
>  		if (!nr_pages)
>  			break;
> -		pages++;
> +		if (likely(pages))
> +			pages++;
>  		start += PAGE_SIZE;
>  	}
>  	if (lock_dropped && *locked) {
> -- 
> 2.7.4
> 

-- 
Sincerely yours,
Mike.