Re: [PATCH 1/6] MIPS: do not use __GFP_REPEAT for order-0 request

[Vlastimil Babka]

On 06/23/2017 10:53 AM, Michal Hocko wrote:
> From: Michal Hocko <mhocko@xxxxxxxx>
> 
> 3377e227af44 ("MIPS: Add 48-bit VA space (and 4-level page tables) for
> 4K pages.") has added a new __GFP_REPEAT user but using this flag
> doesn't really make any sense for order-0 request which is the case here
> because PUD_ORDER is 0. __GFP_REPEAT has historically effect only on
> allocation requests with order > PAGE_ALLOC_COSTLY_ORDER.
> 
> This doesn't introduce any functional change. This is a preparatory
> patch for later work which renames the flag and redefines its semantic.
> 
> Cc: Alex Belits <alex.belits@xxxxxxxxxx>
> Cc: David Daney <david.daney@xxxxxxxxxx>
> Cc: Ralf Baechle <ralf@xxxxxxxxxxxxxx>
> Signed-off-by: Michal Hocko <mhocko@xxxxxxxx>

Acked-by: Vlastimil Babka <vbabka@xxxxxxx>

> ---
>  arch/mips/include/asm/pgalloc.h | 2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)
> 
> diff --git a/arch/mips/include/asm/pgalloc.h b/arch/mips/include/asm/pgalloc.h
> index a1bdb1ea5234..39b9f311c4ef 100644
> --- a/arch/mips/include/asm/pgalloc.h
> +++ b/arch/mips/include/asm/pgalloc.h
> @@ -116,7 +116,7 @@ static inline pud_t *pud_alloc_one(struct mm_struct *mm, unsigned long address)
>  {
>  	pud_t *pud;
>  
> -	pud = (pud_t *) __get_free_pages(GFP_KERNEL|__GFP_REPEAT, PUD_ORDER);
> +	pud = (pud_t *) __get_free_pages(GFP_KERNEL, PUD_ORDER);
>  	if (pud)
>  		pud_init((unsigned long)pud, (unsigned long)invalid_pmd_table);
>  	return pud;
>