From: Stanislav Fomichev <sdf@google.com>
To: Willem de Bruijn <willemdebruijn.kernel@gmail.com>
Cc: "Mina Almasry" <almasrymina@google.com>,
netdev@vger.kernel.org, linux-kernel@vger.kernel.org,
linux-arch@vger.kernel.org, linux-kselftest@vger.kernel.org,
linux-media@vger.kernel.org, dri-devel@lists.freedesktop.org,
linaro-mm-sig@lists.linaro.org,
"David S. Miller" <davem@davemloft.net>,
"Eric Dumazet" <edumazet@google.com>,
"Jakub Kicinski" <kuba@kernel.org>,
"Paolo Abeni" <pabeni@redhat.com>,
"Jesper Dangaard Brouer" <hawk@kernel.org>,
"Ilias Apalodimas" <ilias.apalodimas@linaro.org>,
"Arnd Bergmann" <arnd@arndb.de>,
"David Ahern" <dsahern@kernel.org>,
"Shuah Khan" <shuah@kernel.org>,
"Sumit Semwal" <sumit.semwal@linaro.org>,
"Christian König" <christian.koenig@amd.com>,
"Shakeel Butt" <shakeelb@google.com>,
"Jeroen de Borst" <jeroendb@google.com>,
"Praveen Kaligineedi" <pkaligineedi@google.com>,
"Willem de Bruijn" <willemb@google.com>,
"Kaiyuan Zhang" <kaiyuanz@google.com>
Subject: Re: [RFC PATCH v3 10/12] tcp: RX path for devmem TCP
Date: Mon, 6 Nov 2023 15:32:22 -0800 [thread overview]
Message-ID: <CAKH8qBueYgpxQTvTwngOs6RNjy9yvLF92s1p5nFrobw_UprNMQ@mail.gmail.com> (raw)
In-Reply-To: <CAF=yD-JZ88j+44MYgX-=oYJngz4Z0zw6Y0V3nHXisZJtNu7q6A@mail.gmail.com>
On Mon, Nov 6, 2023 at 2:56 PM Willem de Bruijn
<willemdebruijn.kernel@gmail.com> wrote:
>
> On Mon, Nov 6, 2023 at 2:34 PM Stanislav Fomichev <sdf@google.com> wrote:
> >
> > On 11/06, Willem de Bruijn wrote:
> > > > > IMHO, we need a better UAPI to receive the tokens and give them back to
> > > > > the kernel. CMSG + setsockopt(SO_DEVMEM_DONTNEED) get the job done,
> > > > > but look dated and hacky :-(
> > > > >
> > > > > We should either do some kind of user/kernel shared memory queue to
> > > > > receive/return the tokens (similar to what Jonathan was doing in his
> > > > > proposal?)
> > > >
> > > > I'll take a look at Jonathan's proposal, sorry, I'm not immediately
> > > > familiar but I wanted to respond :-) But is the suggestion here to
> > > > build a new kernel-user communication channel primitive for the
> > > > purpose of passing the information in the devmem cmsg? IMHO that seems
> > > > like an overkill. Why add 100-200 lines of code to the kernel to add
> > > > something that can already be done with existing primitives? I don't
> > > > see anything concretely wrong with cmsg & setsockopt approach, and if
> > > > we switch to something I'd prefer to switch to an existing primitive
> > > > for simplicity?
> > > >
> > > > The only other existing primitive to pass data outside of the linear
> > > > buffer is the MSG_ERRQUEUE that is used for zerocopy. Is that
> > > > preferred? Any other suggestions or existing primitives I'm not aware
> > > > of?
> > > >
> > > > > or bite the bullet and switch to io_uring.
> > > > >
> > > >
> > > > IMO io_uring & socket support are orthogonal, and one doesn't preclude
> > > > the other. As you know we like to use sockets and I believe there are
> > > > issues with io_uring adoption at Google that I'm not familiar with
> > > > (and could be wrong). I'm interested in exploring io_uring support as
> > > > a follow up but I think David Wei will be interested in io_uring
> > > > support as well anyway.
> > >
> > > I also disagree that we need to replace a standard socket interface
> > > with something "faster", in quotes.
> > >
> > > This interface is not the bottleneck to the target workload.
> > >
> > > Replacing the synchronous sockets interface with something more
> > > performant for workloads where it is, is an orthogonal challenge.
> > > However we do that, I think that traditional sockets should continue
> > > to be supported.
> > >
> > > The feature may already even work with io_uring, as both recvmsg with
> > > cmsg and setsockopt have io_uring support now.
> >
> > I'm not really concerned with faster. I would prefer something cleaner :-)
> >
> > Or maybe we should just have it documented. With some kind of path
> > towards beautiful world where we can create dynamic queues..
>
> I suppose we just disagree on the elegance of the API.
Yeah, I might be overly sensitive to the apis that use get/setsockopt
for something more involved than setting a flag.
Probably because I know that bpf will (unnecessarily) trigger on these :-D
I had to implement that bpf "bypass" (or fastpath) for
TCP_ZEROCOPY_RECEIVE and it looks like this token recycle might also
benefit from something similar.
> The concise notification API returns tokens as a range for
> compression, encoding as two 32-bit unsigned integers start + length.
> It allows for even further batching by returning multiple such ranges
> in a single call.
Tangential: should tokens be u64? Otherwise we can't have more than
4gb unacknowledged. Or that's a reasonable constraint?
> This is analogous to the MSG_ZEROCOPY notification mechanism from
> kernel to user.
>
> The synchronous socket syscall interface can be replaced by something
> asynchronous like io_uring. This already works today? Whatever
> asynchronous ring-based API would be selected, io_uring or otherwise,
> I think the concise notification encoding would remain as is.
>
> Since this is an operation on a socket, I find a setsockopt the
> fitting interface.
next prev parent reply other threads:[~2023-11-06 23:32 UTC|newest]
Thread overview: 126+ messages / expand[flat|nested] mbox.gz Atom feed top
2023-11-06 2:43 [RFC PATCH v3 00/12] Device Memory TCP Mina Almasry
2023-11-06 2:44 ` [RFC PATCH v3 01/12] net: page_pool: factor out releasing DMA from releasing the page Mina Almasry
2023-11-06 2:44 ` [RFC PATCH v3 02/12] net: page_pool: create hooks for custom page providers Mina Almasry
2023-11-07 7:44 ` Yunsheng Lin
2023-11-09 11:09 ` Paolo Abeni
2023-11-10 23:19 ` Jakub Kicinski
2023-11-13 3:28 ` Mina Almasry
2023-11-13 22:10 ` Jakub Kicinski
2023-11-06 2:44 ` [RFC PATCH v3 03/12] net: netdev netlink api to bind dma-buf to a net device Mina Almasry
2023-11-10 23:16 ` Jakub Kicinski
2023-11-06 2:44 ` [RFC PATCH v3 04/12] netdev: support binding dma-buf to netdevice Mina Almasry
2023-11-07 7:46 ` Yunsheng Lin
2023-11-07 21:59 ` Mina Almasry
2023-11-08 3:40 ` Yunsheng Lin
2023-11-09 2:22 ` Mina Almasry
2023-11-09 9:29 ` Yunsheng Lin
2023-11-08 23:47 ` David Wei
2023-11-09 2:25 ` Mina Almasry
2023-11-09 8:29 ` Paolo Abeni
2023-11-10 2:59 ` Mina Almasry
2023-11-10 7:38 ` Yunsheng Lin
2023-11-10 9:45 ` Mina Almasry
2023-11-10 23:19 ` Jakub Kicinski
2023-11-11 2:19 ` Mina Almasry
2023-11-06 2:44 ` [RFC PATCH v3 05/12] netdev: netdevice devmem allocator Mina Almasry
2023-11-06 23:44 ` David Ahern
2023-11-07 22:10 ` Mina Almasry
2023-11-07 22:55 ` David Ahern
2023-11-07 23:03 ` Mina Almasry
2023-11-09 1:15 ` David Wei
2023-11-10 14:26 ` Pavel Begunkov
2023-11-11 17:19 ` David Ahern
2023-11-14 16:09 ` Pavel Begunkov
2023-11-09 1:00 ` David Wei
2023-11-08 3:48 ` Yunsheng Lin
2023-11-09 1:41 ` Mina Almasry
2023-11-07 7:45 ` Yunsheng Lin
2023-11-09 8:44 ` Paolo Abeni
2023-11-06 2:44 ` [RFC PATCH v3 06/12] memory-provider: dmabuf devmem memory provider Mina Almasry
2023-11-06 21:02 ` Stanislav Fomichev
2023-11-06 23:49 ` David Ahern
2023-11-08 0:02 ` Mina Almasry
2023-11-08 0:10 ` David Ahern
2023-11-10 23:16 ` Jakub Kicinski
2023-11-13 4:54 ` Mina Almasry
2023-11-06 2:44 ` [RFC PATCH v3 07/12] page-pool: device memory support Mina Almasry
2023-11-07 8:00 ` Yunsheng Lin
2023-11-07 21:56 ` Mina Almasry
2023-11-08 10:56 ` Yunsheng Lin
2023-11-09 3:20 ` Mina Almasry
2023-11-09 9:30 ` Yunsheng Lin
2023-11-09 12:20 ` Mina Almasry
2023-11-09 13:23 ` Yunsheng Lin
2023-11-09 9:01 ` Paolo Abeni
2023-11-06 2:44 ` [RFC PATCH v3 08/12] net: support non paged skb frags Mina Almasry
2023-11-07 9:00 ` Yunsheng Lin
2023-11-07 21:19 ` Mina Almasry
2023-11-08 11:25 ` Yunsheng Lin
2023-11-09 9:14 ` Paolo Abeni
2023-11-10 4:06 ` Mina Almasry
2023-11-10 23:19 ` Jakub Kicinski
2023-11-13 6:05 ` Mina Almasry
2023-11-13 22:17 ` Jakub Kicinski
2023-11-06 2:44 ` [RFC PATCH v3 09/12] net: add support for skbs with unreadable frags Mina Almasry
2023-11-06 18:47 ` Stanislav Fomichev
2023-11-06 19:34 ` David Ahern
2023-11-06 20:31 ` Mina Almasry
2023-11-06 21:59 ` Stanislav Fomichev
2023-11-06 22:18 ` Mina Almasry
2023-11-06 22:59 ` Stanislav Fomichev
2023-11-06 23:27 ` Mina Almasry
2023-11-06 23:55 ` Stanislav Fomichev
2023-11-07 0:07 ` Willem de Bruijn
2023-11-07 0:14 ` Stanislav Fomichev
2023-11-07 0:59 ` Stanislav Fomichev
2023-11-07 2:23 ` Willem de Bruijn
2023-11-07 17:44 ` Stanislav Fomichev
2023-11-07 17:57 ` Willem de Bruijn
2023-11-07 18:14 ` Stanislav Fomichev
2023-11-07 0:20 ` Mina Almasry
2023-11-07 1:06 ` Stanislav Fomichev
2023-11-07 19:53 ` Mina Almasry
2023-11-07 21:05 ` Stanislav Fomichev
2023-11-07 21:17 ` Eric Dumazet
2023-11-07 22:23 ` Stanislav Fomichev
2023-11-10 23:17 ` Jakub Kicinski
2023-11-10 23:19 ` Jakub Kicinski
2023-11-07 1:09 ` David Ahern
2023-11-06 23:37 ` David Ahern
2023-11-07 0:03 ` Mina Almasry
2023-11-06 20:56 ` Stanislav Fomichev
2023-11-07 0:16 ` David Ahern
2023-11-07 0:23 ` Mina Almasry
2023-11-08 14:43 ` David Laight
2023-11-06 2:44 ` [RFC PATCH v3 10/12] tcp: RX path for devmem TCP Mina Almasry
2023-11-06 18:44 ` Stanislav Fomichev
2023-11-06 19:29 ` Mina Almasry
2023-11-06 21:14 ` Willem de Bruijn
2023-11-06 22:34 ` Stanislav Fomichev
2023-11-06 22:55 ` Willem de Bruijn
2023-11-06 23:32 ` Stanislav Fomichev [this message]
2023-11-06 23:55 ` David Ahern
2023-11-07 0:02 ` Willem de Bruijn
2023-11-07 23:55 ` Mina Almasry
2023-11-08 0:01 ` David Ahern
2023-11-09 2:39 ` Mina Almasry
2023-11-09 16:07 ` Edward Cree
2023-12-08 20:12 ` Pavel Begunkov
2023-11-09 11:05 ` Paolo Abeni
2023-11-10 23:16 ` Jakub Kicinski
2023-12-08 20:28 ` Pavel Begunkov
2023-12-08 20:09 ` Pavel Begunkov
2023-11-06 21:17 ` Stanislav Fomichev
2023-11-08 15:36 ` Edward Cree
2023-11-09 10:52 ` Paolo Abeni
2023-11-10 23:19 ` Jakub Kicinski
2023-11-06 2:44 ` [RFC PATCH v3 11/12] net: add SO_DEVMEM_DONTNEED setsockopt to release RX pages Mina Almasry
2023-11-06 2:44 ` [RFC PATCH v3 12/12] selftests: add ncdevmem, netcat for devmem TCP Mina Almasry
2023-11-09 11:03 ` Paolo Abeni
2023-11-10 23:13 ` Jakub Kicinski
2023-11-11 2:27 ` Mina Almasry
2023-11-11 2:35 ` Jakub Kicinski
2023-11-13 4:08 ` Mina Almasry
2023-11-13 22:20 ` Jakub Kicinski
2023-11-10 23:17 ` Jakub Kicinski
2023-11-07 15:18 ` [RFC PATCH v3 00/12] Device Memory TCP David Ahern
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=CAKH8qBueYgpxQTvTwngOs6RNjy9yvLF92s1p5nFrobw_UprNMQ@mail.gmail.com \
--to=sdf@google.com \
--cc=almasrymina@google.com \
--cc=arnd@arndb.de \
--cc=christian.koenig@amd.com \
--cc=davem@davemloft.net \
--cc=dri-devel@lists.freedesktop.org \
--cc=dsahern@kernel.org \
--cc=edumazet@google.com \
--cc=hawk@kernel.org \
--cc=ilias.apalodimas@linaro.org \
--cc=jeroendb@google.com \
--cc=kaiyuanz@google.com \
--cc=kuba@kernel.org \
--cc=linaro-mm-sig@lists.linaro.org \
--cc=linux-arch@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-kselftest@vger.kernel.org \
--cc=linux-media@vger.kernel.org \
--cc=netdev@vger.kernel.org \
--cc=pabeni@redhat.com \
--cc=pkaligineedi@google.com \
--cc=shakeelb@google.com \
--cc=shuah@kernel.org \
--cc=sumit.semwal@linaro.org \
--cc=willemb@google.com \
--cc=willemdebruijn.kernel@gmail.com \
/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
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
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).