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;
MIME-Version: 1.0
References: <20210908061611.69823-1-mie@igel.co.jp> <20210908061611.69823-2-mie@igel.co.jp>
 <YThXe4WxHErNiwgE@infradead.org> <CANXvt5ojNPpyPVnE0D5o9873hGz6ijF7QfTd9z08Ds-ex3Ye-Q@mail.gmail.com>
 <YThj70ByPvZNQjgU@infradead.org> <CANXvt5rCCBku7LpAG5TV7LxkQ1bZnB6ACybKxJnTrRA1LE8e6Q@mail.gmail.com>
 <20210908111804.GX1200268@ziepe.ca> <1c0356f5-19cf-e883-3d96-82a87d0cffcb@amd.com>
In-Reply-To: <1c0356f5-19cf-e883-3d96-82a87d0cffcb@amd.com>
From:   Daniel Vetter <daniel.vetter@ffwll.ch>
Date:   Wed, 8 Sep 2021 21:22:37 +0200
Message-ID: <CAKMK7uE=mQwgcSaTcT8U3GgCeeKOmPqS=YOqkn+SEnbbUNM1=A@mail.gmail.com>
Subject: Re: [RFC PATCH 1/3] RDMA/umem: Change for rdma devices has not dma device
To:     =?UTF-8?Q?Christian_K=C3=B6nig?= <christian.koenig@amd.com>
Cc:     Jason Gunthorpe <jgg@ziepe.ca>, Shunsuke Mie <mie@igel.co.jp>,
        Christoph Hellwig <hch@infradead.org>,
        Zhu Yanjun <zyjzyj2000@gmail.com>,
        Alex Deucher <alexander.deucher@amd.com>,
        Doug Ledford <dledford@redhat.com>,
        Jianxin Xiong <jianxin.xiong@intel.com>,
        Leon Romanovsky <leon@kernel.org>,
        Linux Kernel Mailing List <linux-kernel@vger.kernel.org>,
        linux-rdma <linux-rdma@vger.kernel.org>,
        Damian Hobson-Garcia <dhobsong@igel.co.jp>,
        Takanari Hayama <taki@igel.co.jp>,
        Tomohito Esaki <etom@igel.co.jp>
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable
Precedence: bulk

On Wed, Sep 8, 2021 at 3:33 PM Christian K=C3=B6nig <christian.koenig@amd.c=
om> wrote:
> Am 08.09.21 um 13:18 schrieb Jason Gunthorpe:
> > On Wed, Sep 08, 2021 at 05:41:39PM +0900, Shunsuke Mie wrote:
> >> 2021=E5=B9=B49=E6=9C=888=E6=97=A5(=E6=B0=B4) 16:20 Christoph Hellwig <=
hch@infradead.org>:
> >>> On Wed, Sep 08, 2021 at 04:01:14PM +0900, Shunsuke Mie wrote:
> >>>> Thank you for your comment.
> >>>>> On Wed, Sep 08, 2021 at 03:16:09PM +0900, Shunsuke Mie wrote:
> >>>>>> To share memory space using dma-buf, a API of the dma-buf requires=
 dma
> >>>>>> device, but devices such as rxe do not have a dma device. For thos=
e case,
> >>>>>> change to specify a device of struct ib instead of the dma device.
> >>>>> So if dma-buf doesn't actually need a device to dma map why do we e=
ver
> >>>>> pass the dma_device here?  Something does not add up.
> >>>> As described in the dma-buf api guide [1], the dma_device is used by=
 dma-buf
> >>>> exporter to know the device buffer constraints of importer.
> >>>> [1] https://nam11.safelinks.protection.outlook.com/?url=3Dhttps%3A%2=
F%2Flwn.net%2FArticles%2F489703%2F&amp;data=3D04%7C01%7Cchristian.koenig%40=
amd.com%7C4d18470a94df4ed24c8108d972ba5591%7C3dd8961fe4884e608e11a82d994e18=
3d%7C0%7C0%7C637666967356417448%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDA=
iLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C2000&amp;sdata=3DARwQyo%2=
BCjMohaNbyREofToHIj2bndL5L0HaU9cOrYq4%3D&amp;reserved=3D0
> >>> Which means for rxe you'd also have to pass the one for the underlyin=
g
> >>> net device.
> >> I thought of that way too. In that case, the memory region is constrai=
ned by the
> >> net device, but rxe driver copies data using CPU. To avoid the constra=
ints, I
> >> decided to use the ib device.
> > Well, that is the whole problem.
> >
> > We can't mix the dmabuf stuff people are doing that doesn't fill in
> > the CPU pages in the SGL with RXE - it is simply impossible as things
> > currently are for RXE to acess this non-struct page memory.
>
> Yeah, agree that doesn't make much sense.
>
> When you want to access the data with the CPU then why do you want to
> use DMA-buf in the first place?
>
> Please keep in mind that there is work ongoing to replace the sg table
> with an DMA address array and so make the underlying struct page
> inaccessible for importers.

Also if you do have a dma-buf, you can just dma_buf_vmap() the buffer
for cpu access. Which intentionally does not require any device. No
idea why there's a dma_buf_attach involved. Now not all exporters
support this, but that's fixable, and you must call
dma_buf_begin/end_cpu_access for cache management if the allocation
isn't cpu coherent. But it's all there, no need to apply hacks of
allowing a wrong device or other fun things.
-Daniel
--=20
Daniel Vetter
Software Engineer, Intel Corporation
http://blog.ffwll.ch