LKML Archive on lore.kernel.org
help / color / mirror / Atom feed
From: Alex Williamson <alex.williamson@redhat.com>
To: "Jason Cai (Xiang Feng)" <jason.cai@linux.alibaba.com>
Cc: pbonzini@redhat.com, kvm@vger.kernel.org,
	linux-kernel@vger.kernel.org, linux-mm@kvack.org,
	gnehzuil@linux.alibaba.com
Subject: Re: [PATCH] vfio iommu type1: improve memory pinning process for raw PFN mapping
Date: Tue, 20 Mar 2018 15:33:23 -0600	[thread overview]
Message-ID: <20180320153323.41c58c19@t450s.home> (raw)
In-Reply-To: <7F93BB33-4ABF-468F-8814-78DE9D23FA08@linux.alibaba.com>

On Mon, 19 Mar 2018 10:30:24 +0800
"Jason Cai (Xiang Feng)" <jason.cai@linux.alibaba.com> wrote:

> When using vfio to pass through a PCIe device (e.g. a GPU card) that
> has a huge BAR (e.g. 16GB), a lot of cycles are wasted on memory
> pinning because PFNs of PCI BAR are not backed by struct page, and
> the corresponding VMA has flag VM_PFNMAP.
> 
> With this change, when pinning a region which is a raw PFN mapping,
> it can skip unnecessary user memory pinning process, and thus, can
> significantly improve VM's boot up time when passing through devices
> via VFIO. In my test on a Xeon E5 2.6GHz, the time mapping a 16GB
> BAR was reduced from about 0.4s to 1.5us.
> 
> Signed-off-by: Jason Cai (Xiang Feng) <jason.cai@linux.alibaba.com>
> ---
>  drivers/vfio/vfio_iommu_type1.c | 24 ++++++++++++++----------
>  1 file changed, 14 insertions(+), 10 deletions(-)
> 
> diff --git a/drivers/vfio/vfio_iommu_type1.c b/drivers/vfio/vfio_iommu_type1.c
> index 45657e2b1ff7..0658f35318b8 100644
> --- a/drivers/vfio/vfio_iommu_type1.c
> +++ b/drivers/vfio/vfio_iommu_type1.c
> @@ -397,7 +397,6 @@ static long vfio_pin_pages_remote(struct vfio_dma *dma, unsigned long vaddr,
>  {
>         unsigned long pfn = 0;
>         long ret, pinned = 0, lock_acct = 0;
> -       bool rsvd;
>         dma_addr_t iova = vaddr - dma->vaddr + dma->iova;
> 
>         /* This code path is only user initiated */
> @@ -408,14 +407,22 @@ static long vfio_pin_pages_remote(struct vfio_dma *dma, unsigned long vaddr,
>         if (ret)
>                 return ret;
> 
> +       if (is_invalid_reserved_pfn(*pfn_base)) {
> +               struct vm_area_struct *vma;
> +               down_read(&current->mm->mmap_sem);
> +               vma = find_vma_intersection(current->mm, vaddr, vaddr + 1);
> +               pinned = min(npage, (long)vma_pages(vma));
> +               up_read(&current->mm->mmap_sem);
> +               return pinned;
> +       }
> +
>         pinned++;
> -       rsvd = is_invalid_reserved_pfn(*pfn_base);
> 
>         /*
>          * Reserved pages aren't counted against the user, externally pinned
>          * pages are already counted against the user.
>          */
> -       if (!rsvd && !vfio_find_vpfn(dma, iova)) {
> +       if (!vfio_find_vpfn(dma, iova)) {
>                 if (!lock_cap && current->mm->locked_vm + 1 > limit) {
>                         put_pfn(*pfn_base, dma->prot);
>                         pr_warn("%s: RLIMIT_MEMLOCK (%ld) exceeded\n", __func__,
> @@ -435,13 +442,12 @@ static long vfio_pin_pages_remote(struct vfio_dma *dma, unsigned long vaddr,
>                 if (ret)
>                         break;
> 
> -               if (pfn != *pfn_base + pinned ||
> -                   rsvd != is_invalid_reserved_pfn(pfn)) {
> +               if (pfn != *pfn_base + pinned) {
>                         put_pfn(pfn, dma->prot);
>                         break;
>                 }
> 
> -               if (!rsvd && !vfio_find_vpfn(dma, iova)) {
> +               if (!vfio_find_vpfn(dma, iova)) {
>                         if (!lock_cap &&
>                             current->mm->locked_vm + lock_acct + 1 > limit) {
>                                 put_pfn(pfn, dma->prot);
> @@ -459,10 +465,8 @@ static long vfio_pin_pages_remote(struct vfio_dma *dma, unsigned long vaddr,
> 
>  unpin_out:
>         if (ret) {
> -               if (!rsvd) {
> -                       for (pfn = *pfn_base ; pinned ; pfn++, pinned--)
> -                               put_pfn(pfn, dma->prot);
> -               }
> +               for (pfn = *pfn_base ; pinned ; pfn++, pinned--)
> +                       put_pfn(pfn, dma->prot);
> 
>                 return ret;
>         }

Hi Jason,

Something is wrong with your mail setup, the patch looks normal above,
but when I view the source or save it to try to apply it, the diff is
corrupt, as below.  It looks like maybe you're pasting the patch into
your mailer and it's wrapping lines (ending with '=') and actual '='
are replaced with '=3D' and tabs are converted to spaces.  Please fix
your mailer and resend.  Thanks,

Alex

diff --git a/drivers/vfio/vfio_iommu_type1.c =
b/drivers/vfio/vfio_iommu_type1.c
index 45657e2b1ff7..0658f35318b8 100644
--- a/drivers/vfio/vfio_iommu_type1.c
+++ b/drivers/vfio/vfio_iommu_type1.c
@@ -397,7 +397,6 @@ static long vfio_pin_pages_remote(struct vfio_dma =
*dma, unsigned long vaddr,
 {
        unsigned long pfn =3D 0;
        long ret, pinned =3D 0, lock_acct =3D 0;
-       bool rsvd;
        dma_addr_t iova =3D vaddr - dma->vaddr + dma->iova;

        /* This code path is only user initiated */
@@ -408,14 +407,22 @@ static long vfio_pin_pages_remote(struct vfio_dma =
*dma, unsigned long vaddr,
        if (ret)
                return ret;

+       if (is_invalid_reserved_pfn(*pfn_base)) {
+               struct vm_area_struct *vma;
+               down_read(&current->mm->mmap_sem);
+               vma =3D find_vma_intersection(current->mm, vaddr, vaddr =
+ 1);
+               pinned =3D min(npage, (long)vma_pages(vma));
+               up_read(&current->mm->mmap_sem);
+               return pinned;
+       }
+
        pinned++;
-       rsvd =3D is_invalid_reserved_pfn(*pfn_base);

        /*
         * Reserved pages aren't counted against the user, externally =
pinned
         * pages are already counted against the user.
         */
-       if (!rsvd && !vfio_find_vpfn(dma, iova)) {
+       if (!vfio_find_vpfn(dma, iova)) {
                if (!lock_cap && current->mm->locked_vm + 1 > limit) {
                        put_pfn(*pfn_base, dma->prot);
                        pr_warn("%s: RLIMIT_MEMLOCK (%ld) exceeded\n", =
__func__,
@@ -435,13 +442,12 @@ static long vfio_pin_pages_remote(struct vfio_dma =
*dma, unsigned long vaddr,
                if (ret)
                        break;

-               if (pfn !=3D *pfn_base + pinned ||
-                   rsvd !=3D is_invalid_reserved_pfn(pfn)) {
+               if (pfn !=3D *pfn_base + pinned) {
                        put_pfn(pfn, dma->prot);
                        break;
                }

-               if (!rsvd && !vfio_find_vpfn(dma, iova)) {
+               if (!vfio_find_vpfn(dma, iova)) {
                        if (!lock_cap &&
                            current->mm->locked_vm + lock_acct + 1 > =
limit) {
                                put_pfn(pfn, dma->prot);
@@ -459,10 +465,8 @@ static long vfio_pin_pages_remote(struct vfio_dma =
*dma, unsigned long vaddr,

 unpin_out:
        if (ret) {
-               if (!rsvd) {
-                       for (pfn =3D *pfn_base ; pinned ; pfn++, =
pinned--)
-                               put_pfn(pfn, dma->prot);
-               }
+               for (pfn =3D *pfn_base ; pinned ; pfn++, pinned--)
+                       put_pfn(pfn, dma->prot);

                return ret;
        }
--
2.13.6

  reply	other threads:[~2018-03-20 21:33 UTC|newest]

Thread overview: 5+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2018-03-19  2:30 Jason Cai (Xiang Feng)
2018-03-20 21:33 ` Alex Williamson [this message]
2018-03-20 22:04   ` Alex Williamson
2018-03-22  4:52 Jason Cai (Xiang Feng)
2018-03-22 21:48 ` Alex Williamson

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20180320153323.41c58c19@t450s.home \
    --to=alex.williamson@redhat.com \
    --cc=gnehzuil@linux.alibaba.com \
    --cc=jason.cai@linux.alibaba.com \
    --cc=kvm@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-mm@kvack.org \
    --cc=pbonzini@redhat.com \
    --subject='Re: [PATCH] vfio iommu type1: improve memory pinning process for raw PFN mapping' \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).