Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 23.128.96.18 as permitted sender) client-ip=23.128.96.18;
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: quoted-printable
From:   Andy Lutomirski <luto@amacapital.net>
Mime-Version: 1.0 (1.0)
Subject: Re: [RFC PATCH v3 1/2] mempinfd: Add new syscall to provide memory pin
Date:   Sun, 7 Feb 2021 14:02:36 -0800
Message-Id: <ED58431F-5972-47D1-BF50-93A20AD86C46@amacapital.net>
References: <1612685884-19514-2-git-send-email-wangzhou1@hisilicon.com>
Cc:     linux-kernel@vger.kernel.org, iommu@lists.linux-foundation.org,
        linux-mm@kvack.org, linux-arm-kernel@lists.infradead.org,
        linux-api@vger.kernel.org,
        Andrew Morton <akpm@linux-foundation.org>,
        Alexander Viro <viro@zeniv.linux.org.uk>,
        gregkh@linuxfoundation.org, song.bao.hua@hisilicon.com,
        jgg@ziepe.ca, kevin.tian@intel.com, jean-philippe@linaro.org,
        eric.auger@redhat.com, liguozhu@hisilicon.com,
        zhangfei.gao@linaro.org, Sihang Chen <chensihang1@hisilicon.com>
In-Reply-To: <1612685884-19514-2-git-send-email-wangzhou1@hisilicon.com>
To:     Zhou Wang <wangzhou1@hisilicon.com>
Precedence: bulk


> On Feb 7, 2021, at 12:31 AM, Zhou Wang <wangzhou1@hisilicon.com> wrote:
>=20
> =EF=BB=BFSVA(share virtual address) offers a way for device to share proce=
ss virtual
> address space safely, which makes more convenient for user space device
> driver coding. However, IO page faults may happen when doing DMA
> operations. As the latency of IO page fault is relatively big, DMA
> performance will be affected severely when there are IO page faults.
> =46rom a long term view, DMA performance will be not stable.
>=20
> In high-performance I/O cases, accelerators might want to perform
> I/O on a memory without IO page faults which can result in dramatically
> increased latency. Current memory related APIs could not achieve this
> requirement, e.g. mlock can only avoid memory to swap to backup device,
> page migration can still trigger IO page fault.
>=20
> Various drivers working under traditional non-SVA mode are using
> their own specific ioctl to do pin. Such ioctl can be seen in v4l2,
> gpu, infiniband, media, vfio, etc. Drivers are usually doing dma
> mapping while doing pin.
>=20
> But, in SVA mode, pin could be a common need which isn't necessarily
> bound with any drivers, and neither is dma mapping needed by drivers
> since devices are using the virtual address of CPU. Thus, It is better
> to introduce a new common syscall for it.
>=20
> This patch leverages the design of userfaultfd and adds mempinfd for pin
> to avoid messing up mm_struct. A fd will be got by mempinfd, then user
> space can do pin/unpin pages by ioctls of this fd, all pinned pages under
> one file will be unpinned in file release process. Like pin page cases in
> other places, can_do_mlock is used to check permission and input
> parameters.


Can you document what the syscall does?

Userfaultfd is an fd because one program controls another.  Is mempinfd like=
 this?=