[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <56997ACE.80604@redhat.com>
Date: Fri, 15 Jan 2016 15:03:42 -0800
From: Laura Abbott <labbott@...hat.com>
To: Chen Feng <puck.chen@...ilicon.com>, gregkh@...uxfoundation.org,
arve@...roid.com, riandrews@...roid.com,
paul.gortmaker@...driver.com, gioh.kim@....com,
tranmanphong@...il.com, mitchelh@...eaurora.org,
devel@...verdev.osuosl.org, linux-kernel@...r.kernel.org,
yudongbin@...ilicon.com, saberlily.xia@...ilicon.com,
suzhuangluan@...ilicon.com, kong.kongxinwei@...ilicon.com,
xuyiping@...ilicon.com, z.liuxinliang@...ilicon.com,
weidong2@...ilicon.com, w.f@...wei.com, puck.chen@...mail.com,
shimingxing@...ilicon.com, oliver.fu@...ilicon.com,
albert.lubing@...ilicon.com, chenxiang9@...wei.com,
liuzixing@...wei.com, haojian.zhuang@...aro.org,
Catalin Marinas <catalin.marinas@....com>,
Will Deacon <will.deacon@....com>
Cc: qijiwen@...ilicon.com, peter.panshilin@...ilicon.com,
dan.zhao@...ilicon.com, linuxarm@...wei.com,
dev@...ts.96boards.org,
linux-arm-kernel <linux-arm-kernel@...ts.infradead.org>
Subject: Re: [PATCH] staging: ion: make the pte default none PTE_RDONLY
(adding linux-arm and a few people)
On 01/14/2016 06:42 PM, Chen Feng wrote:
> The page is already alloc at ion_alloc function,
> ion_mmap map the alloced pages to user-space.
>
> The default prot can be PTE_RDONLY. Take a look at
> here:
> set_pte_at()
> arch/arm64/include/asm:
> if (pte_dirty(pte) && pte_write(pte))
> pte_val(pte) &= ~PTE_RDONLY;
> else
> pte_val(pte) |= PTE_RDONLY;
>
> So with the dirty bit,it can improve the efficiency
> and donnot need to handle memory fault when use access.
>
> Signed-off-by: Chen Feng <puck.chen@...ilicon.com>
> Signed-off-by: Wei Dong <weidong2@...ilicon.com>
> Reviewed-by: Zhuangluan Su <suzhuangluan@...ilicon.com>
> ---
> drivers/staging/android/ion/ion.c | 3 +++
> 1 file changed, 3 insertions(+)
>
> diff --git a/drivers/staging/android/ion/ion.c b/drivers/staging/android/ion/ion.c
> index e237e9f..dba5942 100644
> --- a/drivers/staging/android/ion/ion.c
> +++ b/drivers/staging/android/ion/ion.c
> @@ -1026,6 +1026,9 @@ static int ion_mmap(struct dma_buf *dmabuf, struct vm_area_struct *vma)
> if (!(buffer->flags & ION_FLAG_CACHED))
> vma->vm_page_prot = pgprot_writecombine(vma->vm_page_prot);
>
> + /*Default writeable*/
> + vma->vm_page_prot = pte_mkdirty(vma->vm_page_prot);
> +
> mutex_lock(&buffer->lock);
> /* now map it to userspace */
> ret = buffer->heap->ops->map_user(buffer->heap, buffer, vma);
>
The extra fault is unfortunate but I'm skeptical about just setting
pte_mkdirty.
Catalin/Will, do you have any thoughts? Right now it seems like any
range mapped with remap_pfn_range will have this extra fault
behavior. Is marking the range dirty the best solution?
Thanks,
Laura
Powered by blists - more mailing lists