Re: [PATCH 22 of 31] split_huge_page paging

[Mel Gorman <mel@csn.ul.ie>]

On Tue, Jan 26, 2010 at 02:52:17PM +0100, Andrea Arcangeli wrote:
> From: Andrea Arcangeli <aarcange@redhat.com>
> 
> Paging logic that splits the page before it is unmapped and added to swap to
> ensure backwards compatibility with the legacy swap code. Eventually swap
> should natively pageout the hugepages to increase performance and decrease
> seeking and fragmentation of swap space.

To be honest, I'm not sure how much of a win that would even be.
SWAP_CLUSTER_MAX number of pages is 128K on x86-64 which is far short of the
2MB. There is no guarantee that the cost of 2M-128K of potentialy unnecessary
IO to satisfy a watermark would be offset by seekier swap or fragmentation of
swap space. Even if it was, it would make sense to try and fix swap-layout
so that virtually-contiguous pages are swap-contiguous. Just splitting the
page and swapping out what's necessary seems reasonable on its own.

> swapoff can just skip over huge pmd as
> they cannot be part of swap yet. In add_to_swap be careful to split the page
> only if we got a valid swap entry so we don't split hugepages with a full swap.
> 
> Signed-off-by: Andrea Arcangeli <aarcange@redhat.com>

Acked-by: Mel Gorman <mel@csn.ul.ie>

> ---
> 
> diff --git a/mm/memory-failure.c b/mm/memory-failure.c
> --- a/mm/memory-failure.c
> +++ b/mm/memory-failure.c
> @@ -378,6 +378,8 @@ static void collect_procs_anon(struct pa
>  	struct task_struct *tsk;
>  	struct anon_vma *av;
>  
> +	if (unlikely(split_huge_page(page)))
> +		return;
>  	read_lock(&tasklist_lock);
>  	av = page_lock_anon_vma(page);
>  	if (av == NULL)	/* Not actually mapped anymore */
> diff --git a/mm/rmap.c b/mm/rmap.c
> --- a/mm/rmap.c
> +++ b/mm/rmap.c
> @@ -1174,6 +1174,7 @@ int try_to_unmap(struct page *page, enum
>  	int ret;
>  
>  	BUG_ON(!PageLocked(page));
> +	BUG_ON(PageTransHuge(page));
>  
>  	if (unlikely(PageKsm(page)))
>  		ret = try_to_unmap_ksm(page, flags);
> diff --git a/mm/swap_state.c b/mm/swap_state.c
> --- a/mm/swap_state.c
> +++ b/mm/swap_state.c
> @@ -156,6 +156,12 @@ int add_to_swap(struct page *page)
>  	if (!entry.val)
>  		return 0;
>  
> +	if (unlikely(PageTransHuge(page)))
> +		if (unlikely(split_huge_page(page))) {
> +			swapcache_free(entry, NULL);
> +			return 0;
> +		}
> +
>  	/*
>  	 * Radix-tree node allocations from PF_MEMALLOC contexts could
>  	 * completely exhaust the page allocator. __GFP_NOMEMALLOC
> diff --git a/mm/swapfile.c b/mm/swapfile.c
> --- a/mm/swapfile.c
> +++ b/mm/swapfile.c
> @@ -905,6 +905,8 @@ static inline int unuse_pmd_range(struct
>  	pmd = pmd_offset(pud, addr);
>  	do {
>  		next = pmd_addr_end(addr, end);
> +		if (unlikely(pmd_trans_huge(*pmd)))
> +			continue;
>  		if (pmd_none_or_clear_bad(pmd))
>  			continue;
>  		ret = unuse_pte_range(vma, pmd, addr, next, entry, page);
> 

-- 
Mel Gorman
Part-time Phd Student                          Linux Technology Center
University of Limerick                         IBM Dublin Software Lab

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org.  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>