Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]


Groups > linux.kernel > #1550203 > unrolled thread

Re: [PATCH] drivers/virt: use get_user_pages_unlocked()

Started byLorenzo Stoakes <lstoakes@gmail.com>
First post2017-01-03 22:20 +0100
Last post2017-01-03 22:40 +0100
Articles 2 — 2 participants

Back to article view | Back to linux.kernel

This discussion starts older than the indexed window; earlier articles aren't shown. The article labeled Started by below is the oldest one visible, not the original post.


Contents

  Re: [PATCH] drivers/virt: use get_user_pages_unlocked() Lorenzo Stoakes <lstoakes@gmail.com> - 2017-01-03 22:20 +0100
    Re: [PATCH] drivers/virt: use get_user_pages_unlocked() Michal Hocko <mhocko@kernel.org> - 2017-01-03 22:40 +0100

#1550203 — Re: [PATCH] drivers/virt: use get_user_pages_unlocked()

FromLorenzo Stoakes <lstoakes@gmail.com>
Date2017-01-03 22:20 +0100
SubjectRe: [PATCH] drivers/virt: use get_user_pages_unlocked()
Message-ID<sVEEW-3Cx-9@gated-at.bofh.it>
Just a gentle ping on this :) I think this might be a slightly
abandoned corner of the kernel so not sure who else to ping to get
this moving.

On 1 November 2016 at 19:43, Lorenzo Stoakes <lstoakes@gmail.com> wrote:
> Moving from get_user_pages() to get_user_pages_unlocked() simplifies the code
> and takes advantage of VM_FAULT_RETRY functionality when faulting in pages.
>
> Signed-off-by: Lorenzo Stoakes <lstoakes@gmail.com>
> ---
>  drivers/virt/fsl_hypervisor.c | 7 ++-----
>  1 file changed, 2 insertions(+), 5 deletions(-)
>
> diff --git a/drivers/virt/fsl_hypervisor.c b/drivers/virt/fsl_hypervisor.c
> index 150ce2a..d3eca87 100644
> --- a/drivers/virt/fsl_hypervisor.c
> +++ b/drivers/virt/fsl_hypervisor.c
> @@ -243,11 +243,8 @@ static long ioctl_memcpy(struct fsl_hv_ioctl_memcpy __user *p)
>         sg_list = PTR_ALIGN(sg_list_unaligned, sizeof(struct fh_sg_list));
>
>         /* Get the physical addresses of the source buffer */
> -       down_read(&current->mm->mmap_sem);
> -       num_pinned = get_user_pages(param.local_vaddr - lb_offset,
> -               num_pages, (param.source == -1) ? 0 : FOLL_WRITE,
> -               pages, NULL);
> -       up_read(&current->mm->mmap_sem);
> +       num_pinned = get_user_pages_unlocked(param.local_vaddr - lb_offset,
> +               num_pages, pages, (param.source == -1) ? 0 : FOLL_WRITE);
>
>         if (num_pinned != num_pages) {
>                 /* get_user_pages() failed */
> --
> 2.10.2
>



-- 
Lorenzo Stoakes
https://ljs.io

[toc] | [next] | [standalone]


#1550226

FromMichal Hocko <mhocko@kernel.org>
Date2017-01-03 22:40 +0100
Message-ID<sVEYi-3J6-51@gated-at.bofh.it>
In reply to#1550203
On Tue 03-01-17 21:14:20, Lorenzo Stoakes wrote:
> Just a gentle ping on this :) I think this might be a slightly
> abandoned corner of the kernel so not sure who else to ping to get
> this moving.

Maybe Andrew can pick it up?
http://lkml.kernel.org/r/20161101194332.23961-1-lstoakes@gmail.com

> On 1 November 2016 at 19:43, Lorenzo Stoakes <lstoakes@gmail.com> wrote:
> > Moving from get_user_pages() to get_user_pages_unlocked() simplifies the code
> > and takes advantage of VM_FAULT_RETRY functionality when faulting in pages.
> >
> > Signed-off-by: Lorenzo Stoakes <lstoakes@gmail.com>
> > ---
> >  drivers/virt/fsl_hypervisor.c | 7 ++-----
> >  1 file changed, 2 insertions(+), 5 deletions(-)
> >
> > diff --git a/drivers/virt/fsl_hypervisor.c b/drivers/virt/fsl_hypervisor.c
> > index 150ce2a..d3eca87 100644
> > --- a/drivers/virt/fsl_hypervisor.c
> > +++ b/drivers/virt/fsl_hypervisor.c
> > @@ -243,11 +243,8 @@ static long ioctl_memcpy(struct fsl_hv_ioctl_memcpy __user *p)
> >         sg_list = PTR_ALIGN(sg_list_unaligned, sizeof(struct fh_sg_list));
> >
> >         /* Get the physical addresses of the source buffer */
> > -       down_read(&current->mm->mmap_sem);
> > -       num_pinned = get_user_pages(param.local_vaddr - lb_offset,
> > -               num_pages, (param.source == -1) ? 0 : FOLL_WRITE,
> > -               pages, NULL);
> > -       up_read(&current->mm->mmap_sem);
> > +       num_pinned = get_user_pages_unlocked(param.local_vaddr - lb_offset,
> > +               num_pages, pages, (param.source == -1) ? 0 : FOLL_WRITE);
> >
> >         if (num_pinned != num_pages) {
> >                 /* get_user_pages() failed */
> > --
> > 2.10.2
> >
> 
> 
> 
> -- 
> Lorenzo Stoakes
> https://ljs.io

-- 
Michal Hocko
SUSE Labs

[toc] | [prev] | [standalone]


Back to top | Article view | linux.kernel


csiph-web