[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20220713015352-mutt-send-email-mst@kernel.org>
Date: Wed, 13 Jul 2022 01:57:23 -0400
From: "Michael S. Tsirkin" <mst@...hat.com>
To: Xie Yongji <xieyongji@...edance.com>
Cc: jasowang@...hat.com, xiaodong.liu@...el.com,
maxime.coquelin@...hat.com, stefanha@...hat.com,
songmuchun@...edance.com,
virtualization@...ts.linux-foundation.org,
linux-kernel@...r.kernel.org
Subject: Re: [PATCH v2 1/5] vduse: Remove unnecessary spin lock protection
On Wed, Jul 06, 2022 at 01:04:59PM +0800, Xie Yongji wrote:
> Taking iotlb lock to access bounce page in page fault
> handler is meaningless since vduse_domain_free_bounce_pages()
> would only be called during file release.
>
> Signed-off-by: Xie Yongji <xieyongji@...edance.com>
vduse_domain_free_bounce_pages is not the
only one taking this lock. This commit log needs more
analysis documenting all points of access to bounce_maps
and why vduse_domain_get_bounce_page and file
release are the only two.
> ---
> drivers/vdpa/vdpa_user/iova_domain.c | 7 ++-----
> 1 file changed, 2 insertions(+), 5 deletions(-)
>
> diff --git a/drivers/vdpa/vdpa_user/iova_domain.c b/drivers/vdpa/vdpa_user/iova_domain.c
> index 6daa3978d290..bca1f0b8850c 100644
> --- a/drivers/vdpa/vdpa_user/iova_domain.c
> +++ b/drivers/vdpa/vdpa_user/iova_domain.c
> @@ -211,17 +211,14 @@ static struct page *
> vduse_domain_get_bounce_page(struct vduse_iova_domain *domain, u64 iova)
> {
> struct vduse_bounce_map *map;
> - struct page *page = NULL;
> + struct page *page;
>
> - spin_lock(&domain->iotlb_lock);
> map = &domain->bounce_maps[iova >> PAGE_SHIFT];
> if (!map->bounce_page)
> - goto out;
> + return NULL;
>
> page = map->bounce_page;
> get_page(page);
> -out:
> - spin_unlock(&domain->iotlb_lock);
>
> return page;
> }
> --
> 2.20.1
Powered by blists - more mailing lists