From: Mina Almasry <almasrymina@google.com>
To: Yunsheng Lin <linyunsheng@huawei.com>, YiFei Zhu <zhuyifei@google.com>
Cc: netdev@vger.kernel.org, linux-kernel@vger.kernel.org,
linux-doc@vger.kernel.org, linux-alpha@vger.kernel.org,
linux-mips@vger.kernel.org, linux-parisc@vger.kernel.org,
sparclinux@vger.kernel.org, linux-trace-kernel@vger.kernel.org,
linux-arch@vger.kernel.org, bpf@vger.kernel.org,
linux-kselftest@vger.kernel.org, linux-media@vger.kernel.org,
dri-devel@lists.freedesktop.org,
"David S. Miller" <davem@davemloft.net>,
"Eric Dumazet" <edumazet@google.com>,
"Jakub Kicinski" <kuba@kernel.org>,
"Paolo Abeni" <pabeni@redhat.com>,
"Jonathan Corbet" <corbet@lwn.net>,
"Richard Henderson" <richard.henderson@linaro.org>,
"Ivan Kokshaysky" <ink@jurassic.park.msu.ru>,
"Matt Turner" <mattst88@gmail.com>,
"Thomas Bogendoerfer" <tsbogend@alpha.franken.de>,
"James E.J. Bottomley" <James.Bottomley@hansenpartnership.com>,
"Helge Deller" <deller@gmx.de>,
"Andreas Larsson" <andreas@gaisler.com>,
"Jesper Dangaard Brouer" <hawk@kernel.org>,
"Ilias Apalodimas" <ilias.apalodimas@linaro.org>,
"Steven Rostedt" <rostedt@goodmis.org>,
"Masami Hiramatsu" <mhiramat@kernel.org>,
"Mathieu Desnoyers" <mathieu.desnoyers@efficios.com>,
"Arnd Bergmann" <arnd@arndb.de>,
"Alexei Starovoitov" <ast@kernel.org>,
"Daniel Borkmann" <daniel@iogearbox.net>,
"Andrii Nakryiko" <andrii@kernel.org>,
"Martin KaFai Lau" <martin.lau@linux.dev>,
"Eduard Zingerman" <eddyz87@gmail.com>,
"Song Liu" <song@kernel.org>,
"Yonghong Song" <yonghong.song@linux.dev>,
"John Fastabend" <john.fastabend@gmail.com>,
"KP Singh" <kpsingh@kernel.org>,
"Stanislav Fomichev" <sdf@google.com>,
"Hao Luo" <haoluo@google.com>, "Jiri Olsa" <jolsa@kernel.org>,
"David Ahern" <dsahern@kernel.org>,
"Willem de Bruijn" <willemdebruijn.kernel@gmail.com>,
"Shuah Khan" <shuah@kernel.org>,
"Sumit Semwal" <sumit.semwal@linaro.org>,
"Christian König" <christian.koenig@amd.com>,
"Pavel Begunkov" <asml.silence@gmail.com>,
"David Wei" <dw@davidwei.uk>, "Jason Gunthorpe" <jgg@ziepe.ca>,
"Shailend Chand" <shailend@google.com>,
"Harshitha Ramamurthy" <hramamurthy@google.com>,
"Shakeel Butt" <shakeelb@google.com>,
"Jeroen de Borst" <jeroendb@google.com>,
"Praveen Kaligineedi" <pkaligineedi@google.com>
Subject: Re: [RFC PATCH net-next v6 00/15] Device Memory TCP
Date: Mon, 25 Mar 2024 17:28:12 -0700 [thread overview]
Message-ID: <CAHS8izPR+SioMKNv3=2ajK=GGOE26BTaxOMykHJfjttqYjx1wQ@mail.gmail.com> (raw)
In-Reply-To: <CAHS8izMwTRyqUS0iRtErfAqDVsXRia5Ajx9PRK3vcfo8utJoUA@mail.gmail.com>
On Tue, Mar 5, 2024 at 11:38 AM Mina Almasry <almasrymina@google.com> wrote:
>
> On Tue, Mar 5, 2024 at 4:54 AM Yunsheng Lin <linyunsheng@huawei.com> wrote:
> >
> > On 2024/3/5 10:01, Mina Almasry wrote:
> >
> > ...
> >
> > >
> > > Perf - page-pool benchmark:
> > > ---------------------------
> > >
> > > bench_page_pool_simple.ko tests with and without these changes:
> > > https://pastebin.com/raw/ncHDwAbn
> > >
> > > AFAIK the number that really matters in the perf tests is the
> > > 'tasklet_page_pool01_fast_path Per elem'. This one measures at about 8
> > > cycles without the changes but there is some 1 cycle noise in some
> > > results.
> > >
> > > With the patches this regresses to 9 cycles with the changes but there
> > > is 1 cycle noise occasionally running this test repeatedly.
> > >
> > > Lastly I tried disable the static_branch_unlikely() in
> > > netmem_is_net_iov() check. To my surprise disabling the
> > > static_branch_unlikely() check reduces the fast path back to 8 cycles,
> > > but the 1 cycle noise remains.
> > >
> >
> > The last sentence seems to be suggesting the above 1 ns regresses is caused
> > by the static_branch_unlikely() checking?
>
> Note it's not a 1ns regression, it's looks like maybe a 1 cycle
> regression (slightly less than 1ns if I'm reading the output of the
> test correctly):
>
> # clean net-next
> time_bench: Type:tasklet_page_pool01_fast_path Per elem: 8 cycles(tsc)
> 2.993 ns (step:0)
>
> # with patches
> time_bench: Type:tasklet_page_pool01_fast_path Per elem: 9 cycles(tsc)
> 3.679 ns (step:0)
>
> # with patches and with diff that disables static branching:
> time_bench: Type:tasklet_page_pool01_fast_path Per elem: 8 cycles(tsc)
> 3.248 ns (step:0)
>
> I do see noise in the test results between run and run, and any
> regression (if any) is slightly obfuscated by the noise, so it's a bit
> hard to make confident statements. So far it looks like a ~0.25ns
> regression without static branch and about ~0.65ns with static branch.
>
> Honestly when I saw all 3 results were within some noise I did not
> investigate more, but if this looks concerning to you I can dig
> further. I likely need to gather a few test runs to filter out the
> noise and maybe investigate the assembly my compiler is generating to
> maybe narrow down what changes there.
>
I did some more investigation here to gather more data to filter out
the noise, and recorded the summary here:
https://pastebin.com/raw/v5dYRg8L
Long story short, the page_pool benchmark results are consistent with
some outlier noise results that I'm discounting here. Currently
page_pool fast path is at 8 cycles
[ 2115.724510] time_bench: Type:tasklet_page_pool01_fast_path Per
elem: 8 cycles(tsc) 3.187 ns (step:0) - (measurement period
time:0.031870585 sec time_interval:31870585) - (invoke count:10000000
tsc_interval:86043192)
and with this patch series it degrades to 10 cycles, or about a 0.7ns
degradation or so:
[ 498.226127] time_bench: Type:tasklet_page_pool01_fast_path Per
elem: 10 cycles(tsc) 3.944 ns (step:0) - (measurement period
time:0.039442539 sec time_interval:39442539) - (invoke count:10000000
tsc_interval:106485268)
I took the time to dig into where the degradation comes from, and to
my surprise we can shave off 1 cycle in perf by removing the
static_branch_unlikely check in netmem_is_net_iov() like so:
diff --git a/include/net/netmem.h b/include/net/netmem.h
index fe354d11a421..2b4310ac1115 100644
--- a/include/net/netmem.h
+++ b/include/net/netmem.h
@@ -122,8 +122,7 @@ typedef unsigned long __bitwise netmem_ref;
static inline bool netmem_is_net_iov(const netmem_ref netmem)
{
#ifdef CONFIG_PAGE_POOL
- return static_branch_unlikely(&page_pool_mem_providers) &&
- (__force unsigned long)netmem & NET_IOV;
+ return (__force unsigned long)netmem & NET_IOV;
#else
return false;
#endif
With this change, the fast path is 9 cycles, only a 1 cycle (~0.35ns)
regression:
[ 199.184429] time_bench: Type:tasklet_page_pool01_fast_path Per
elem: 9 cycles(tsc) 3.552 ns (step:0) - (measurement period
time:0.035524013 sec time_interval:35524013) - (invoke count:10000000
tsc_interval:95907775)
I did some digging with YiFei on why the static_branch_unlikely
appears to be causing a 1 cycle regression, but could not get an
answer that makes sense. The # of instructions in
page_pool_return_page() with the static_branch_unlikely and without is
about the same in the compiled .o file, and my understanding is that
static_branch will cause code re-writing anyway so looking at the
compiled code may not be representative.
Worthy of note is that I get ~95% line rate of devmem TCP regardless
of the static_branch_unlikely() or not, so impact of the static_branch
is not large enough to be measurable end-to-end. I'm thinking I want
to drop the static_branch_unlikely() in the next RFC since it doesn't
improve the end-to-end throughput number and is resulting in a
measurable improvement in the page pool benchmark.
--
Thanks,
Mina
next prev parent reply other threads:[~2024-03-26 0:28 UTC|newest]
Thread overview: 73+ messages / expand[flat|nested] mbox.gz Atom feed top
2024-03-05 2:01 [RFC PATCH net-next v6 00/15] Device Memory TCP Mina Almasry
2024-03-05 2:01 ` [RFC PATCH net-next v6 01/15] queue_api: define queue api Mina Almasry
2024-03-08 1:30 ` Jakub Kicinski
2024-03-08 2:08 ` Mina Almasry
2024-03-08 3:36 ` Jakub Kicinski
2024-03-08 23:47 ` David Wei
2024-03-09 0:27 ` Mina Almasry
2024-03-11 1:12 ` David Ahern
2024-03-05 2:01 ` [RFC PATCH net-next v6 02/15] net: page_pool: create hooks for custom page providers Mina Almasry
2024-03-05 21:54 ` David Wei
2024-03-05 22:36 ` Mina Almasry
2024-03-06 14:29 ` Pavel Begunkov
2024-03-06 17:04 ` Mina Almasry
2024-03-06 19:12 ` Pavel Begunkov
2024-03-06 21:59 ` Mina Almasry
2024-03-07 14:25 ` Pavel Begunkov
2024-03-08 4:57 ` David Wei
2024-03-08 19:53 ` Mina Almasry
2024-03-18 2:02 ` Christoph Hellwig
2024-03-18 2:49 ` David Wei
2024-03-18 23:22 ` Christoph Hellwig
2024-03-22 17:40 ` Mina Almasry
2024-03-22 23:19 ` Jakub Kicinski
2024-03-24 23:35 ` Christoph Hellwig
2024-03-24 23:35 ` Christoph Hellwig
2024-03-22 17:54 ` Mina Almasry
2024-03-24 23:37 ` Christoph Hellwig
2024-03-26 20:19 ` Mina Almasry
2024-03-28 7:31 ` Christoph Hellwig
2024-04-01 19:22 ` Mina Almasry
2024-04-08 15:34 ` Cong Wang
2024-03-05 2:01 ` [RFC PATCH net-next v6 03/15] net: page_pool: factor out page_pool recycle check Mina Almasry
2024-03-05 12:55 ` Yunsheng Lin
2024-03-05 2:01 ` [RFC PATCH net-next v6 04/15] net: netdev netlink api to bind dma-buf to a net device Mina Almasry
2024-03-05 2:01 ` [RFC PATCH net-next v6 05/15] netdev: support binding dma-buf to netdevice Mina Almasry
2024-03-05 9:04 ` Arnd Bergmann
2024-03-05 20:00 ` Mina Almasry
2024-03-05 21:42 ` Arnd Bergmann
2024-03-05 12:55 ` Yunsheng Lin
2024-03-05 21:17 ` Mina Almasry
2024-03-06 12:38 ` Yunsheng Lin
2024-03-06 22:10 ` Mina Almasry
2024-03-07 12:15 ` Yunsheng Lin
2024-03-08 3:58 ` Jakub Kicinski
2024-03-05 2:01 ` [RFC PATCH net-next v6 06/15] netdev: netdevice devmem allocator Mina Almasry
2024-03-05 2:01 ` [RFC PATCH net-next v6 07/15] page_pool: convert to use netmem Mina Almasry
2024-03-05 21:30 ` Mina Almasry
2024-03-05 2:01 ` [RFC PATCH net-next v6 08/15] page_pool: devmem support Mina Almasry
2024-03-05 21:42 ` Mina Almasry
2024-03-05 2:01 ` [RFC PATCH net-next v6 09/15] memory-provider: dmabuf devmem memory provider Mina Almasry
2024-03-06 2:28 ` David Wei
2024-03-06 2:42 ` Mina Almasry
2024-03-06 2:46 ` David Wei
2024-03-06 2:54 ` Mina Almasry
2024-03-06 14:58 ` Pavel Begunkov
2024-03-06 16:51 ` Mina Almasry
2024-03-05 2:01 ` [RFC PATCH net-next v6 10/15] net: support non paged skb frags Mina Almasry
2024-03-05 2:01 ` [RFC PATCH net-next v6 11/15] net: add support for skbs with unreadable frags Mina Almasry
2024-03-05 2:01 ` [RFC PATCH net-next v6 12/15] tcp: RX path for devmem TCP Mina Almasry
2024-03-05 8:41 ` Arnd Bergmann
2024-03-05 19:22 ` Mina Almasry
2024-03-05 19:39 ` Arnd Bergmann
2024-03-05 2:01 ` [RFC PATCH net-next v6 13/15] net: add SO_DEVMEM_DONTNEED setsockopt to release RX frags Mina Almasry
2024-03-05 2:01 ` [RFC PATCH net-next v6 14/15] net: add devmem TCP documentation Mina Almasry
2024-03-08 1:52 ` Jakub Kicinski
2024-03-05 2:01 ` [RFC PATCH net-next v6 15/15] selftests: add ncdevmem, netcat for devmem TCP Mina Almasry
2024-03-05 7:16 ` [RFC PATCH net-next v6 07/15] page_pool: convert to use netmem David Howells
2024-03-05 12:54 ` [RFC PATCH net-next v6 00/15] Device Memory TCP Yunsheng Lin
2024-03-05 19:38 ` Mina Almasry
2024-03-06 12:37 ` Yunsheng Lin
2024-03-26 0:28 ` Mina Almasry [this message]
2024-03-26 12:47 ` Yunsheng Lin
2024-03-26 20:14 ` Mina Almasry
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='CAHS8izPR+SioMKNv3=2ajK=GGOE26BTaxOMykHJfjttqYjx1wQ@mail.gmail.com' \
--to=almasrymina@google.com \
--cc=James.Bottomley@hansenpartnership.com \
--cc=andreas@gaisler.com \
--cc=andrii@kernel.org \
--cc=arnd@arndb.de \
--cc=asml.silence@gmail.com \
--cc=ast@kernel.org \
--cc=bpf@vger.kernel.org \
--cc=christian.koenig@amd.com \
--cc=corbet@lwn.net \
--cc=daniel@iogearbox.net \
--cc=davem@davemloft.net \
--cc=deller@gmx.de \
--cc=dri-devel@lists.freedesktop.org \
--cc=dsahern@kernel.org \
--cc=dw@davidwei.uk \
--cc=eddyz87@gmail.com \
--cc=edumazet@google.com \
--cc=haoluo@google.com \
--cc=hawk@kernel.org \
--cc=hramamurthy@google.com \
--cc=ilias.apalodimas@linaro.org \
--cc=ink@jurassic.park.msu.ru \
--cc=jeroendb@google.com \
--cc=jgg@ziepe.ca \
--cc=john.fastabend@gmail.com \
--cc=jolsa@kernel.org \
--cc=kpsingh@kernel.org \
--cc=kuba@kernel.org \
--cc=linux-alpha@vger.kernel.org \
--cc=linux-arch@vger.kernel.org \
--cc=linux-doc@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-kselftest@vger.kernel.org \
--cc=linux-media@vger.kernel.org \
--cc=linux-mips@vger.kernel.org \
--cc=linux-parisc@vger.kernel.org \
--cc=linux-trace-kernel@vger.kernel.org \
--cc=linyunsheng@huawei.com \
--cc=martin.lau@linux.dev \
--cc=mathieu.desnoyers@efficios.com \
--cc=mattst88@gmail.com \
--cc=mhiramat@kernel.org \
--cc=netdev@vger.kernel.org \
--cc=pabeni@redhat.com \
--cc=pkaligineedi@google.com \
--cc=richard.henderson@linaro.org \
--cc=rostedt@goodmis.org \
--cc=sdf@google.com \
--cc=shailend@google.com \
--cc=shakeelb@google.com \
--cc=shuah@kernel.org \
--cc=song@kernel.org \
--cc=sparclinux@vger.kernel.org \
--cc=sumit.semwal@linaro.org \
--cc=tsbogend@alpha.franken.de \
--cc=willemdebruijn.kernel@gmail.com \
--cc=yonghong.song@linux.dev \
--cc=zhuyifei@google.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).