Re: [PATCH 13/19] s390: get rid of superfluous __GFP_REPEAT
From: Cornelia Huck
Date: Mon Apr 11 2016 - 07:29:07 EST
On Mon, 11 Apr 2016 13:08:06 +0200
Michal Hocko <mhocko@xxxxxxxxxx> wrote:
> From: Michal Hocko <mhocko@xxxxxxxx>
>
> __GFP_REPEAT has a rather weak semantic but since it has been introduced
> around 2.6.12 it has been ignored for low order allocations.
>
> arch_dup_task_struct uses __GFP_REPEAT for fpu_regs_size which is either
> sizeof(__vector128) * __NUM_VXRS = 4069B resp.
> sizeof(freg_t) * __NUM_FPRS = 1024B AFAICS. page_table_alloc then uses
> the flag for a single page allocation. This means that this flag has
> never been actually useful here because it has always been used only for
> PAGE_ALLOC_COSTLY requests.
>
> Cc: Christian Borntraeger <borntraeger@xxxxxxxxxx>
> Cc: Cornelia Huck <cornelia.huck@xxxxxxxxxx>
Let's cc: Martin/Heiko instead :)
> Cc: linux-arch@xxxxxxxxxxxxxxx
> Signed-off-by: Michal Hocko <mhocko@xxxxxxxx>
> ---
> arch/s390/kernel/process.c | 2 +-
> arch/s390/mm/pgalloc.c | 2 +-
> 2 files changed, 2 insertions(+), 2 deletions(-)
>
> diff --git a/arch/s390/kernel/process.c b/arch/s390/kernel/process.c
> index f8e79824e284..1837a1901d4b 100644
> --- a/arch/s390/kernel/process.c
> +++ b/arch/s390/kernel/process.c
> @@ -102,7 +102,7 @@ int arch_dup_task_struct(struct task_struct *dst, struct task_struct *src)
> */
> fpu_regs_size = MACHINE_HAS_VX ? sizeof(__vector128) * __NUM_VXRS
> : sizeof(freg_t) * __NUM_FPRS;
> - dst->thread.fpu.regs = kzalloc(fpu_regs_size, GFP_KERNEL|__GFP_REPEAT);
> + dst->thread.fpu.regs = kzalloc(fpu_regs_size, GFP_KERNEL);
> if (!dst->thread.fpu.regs)
> return -ENOMEM;
>
> diff --git a/arch/s390/mm/pgalloc.c b/arch/s390/mm/pgalloc.c
> index f6c3de26cda8..3f716741797a 100644
> --- a/arch/s390/mm/pgalloc.c
> +++ b/arch/s390/mm/pgalloc.c
> @@ -198,7 +198,7 @@ unsigned long *page_table_alloc(struct mm_struct *mm)
> return table;
> }
> /* Allocate a fresh page */
> - page = alloc_page(GFP_KERNEL|__GFP_REPEAT);
> + page = alloc_page(GFP_KERNEL);
> if (!page)
> return NULL;
> if (!pgtable_page_ctor(page)) {