From: David Laight <David.Laight@ACULAB.COM>
To: 'Alexei Starovoitov' <alexei.starovoitov@gmail.com>,
Cong Wang <xiyou.wangcong@gmail.com>
Cc: Daniel Borkmann <daniel@iogearbox.net>,
Andrii Nakryiko <andrii.nakryiko@gmail.com>,
Networking <netdev@vger.kernel.org>, bpf <bpf@vger.kernel.org>,
Cong Wang <cong.wang@bytedance.com>,
"Alexei Starovoitov" <ast@kernel.org>,
Dongdong Wang <wangdongdong.6@bytedance.com>
Subject: RE: [Patch bpf-next v2 2/5] bpf: introduce timeout map
Date: Wed, 16 Dec 2020 10:38:56 +0000 [thread overview]
Message-ID: <a73e4e66b8a541d7bf74fe8a66ebbc6a@AcuMS.aculab.com> (raw)
In-Reply-To: <CAADnVQL70bVdms6_D_ep1L2v-OcgXu-9KTtLULQdfCMftLhENQ@mail.gmail.com>
From: Alexei Starovoitov
> Sent: 16 December 2020 02:36
...
> > The problem is never about granularity, it is about how efficient we can
> > GC. User-space has to scan the whole table one by one, while the kernel
> > can just do this behind the scene with a much lower overhead.
> >
> > Let's say we arm a timer for each entry in user-space, it requires a syscall
> > and locking buckets each time for each entry. Kernel could do it without
> > any additional syscall and batching. Like I said above, we could have
> > millions of entries, so the overhead would be big in this scenario.
>
> and the user space can pick any other implementation instead
> of trivial entry by entry gc with timer.
The kernel can also gc entries when scanning hash lists during insert
(or even during lookup if not using rw locks).
Apart from the memory use there isn't really a problem having timed-out
entries in the hash table if nothing is looking at them.
David
-
Registered Address Lakeside, Bramley Road, Mount Farm, Milton Keynes, MK1 1PT, UK
Registration No: 1397386 (Wales)
next prev parent reply other threads:[~2020-12-16 10:40 UTC|newest]
Thread overview: 27+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-12-14 20:11 [Patch bpf-next v2 0/5] bpf: introduce timeout map Cong Wang
2020-12-14 20:11 ` [Patch bpf-next v2 1/5] bpf: use index instead of hash for map_locked[] Cong Wang
2020-12-14 20:11 ` [Patch bpf-next v2 2/5] bpf: introduce timeout map Cong Wang
2020-12-15 19:27 ` Andrii Nakryiko
2020-12-15 20:06 ` Cong Wang
2020-12-15 22:03 ` Andrii Nakryiko
2020-12-15 23:23 ` Daniel Borkmann
2020-12-16 0:22 ` Cong Wang
2020-12-16 1:14 ` Alexei Starovoitov
2020-12-16 2:10 ` Cong Wang
2020-12-16 2:35 ` Alexei Starovoitov
2020-12-16 10:38 ` David Laight [this message]
2020-12-17 5:06 ` Cong Wang
2020-12-17 22:39 ` Daniel Borkmann
2020-12-16 0:15 ` Cong Wang
2020-12-16 18:35 ` Andrii Nakryiko
2020-12-17 6:29 ` Cong Wang
2020-12-17 21:14 ` Cong Wang
2020-12-18 19:14 ` Andrii Nakryiko
2020-12-18 19:13 ` Andrii Nakryiko
2020-12-14 20:11 ` [Patch bpf-next v2 3/5] selftests/bpf: update elem_size check in map ptr test Cong Wang
2020-12-14 20:40 ` Andrey Ignatov
2020-12-14 20:11 ` [Patch bpf-next v2 4/5] selftests/bpf: add a test case for bpf timeout map Cong Wang
2020-12-14 20:11 ` [Patch bpf-next v2 5/5] selftests/bpf: add timeout map check in map_ptr tests Cong Wang
2020-12-14 20:41 ` Andrey Ignatov
2020-12-15 19:28 ` [Patch bpf-next v2 0/5] bpf: introduce timeout map Andrii Nakryiko
2020-12-15 20:12 ` Cong Wang
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=a73e4e66b8a541d7bf74fe8a66ebbc6a@AcuMS.aculab.com \
--to=david.laight@aculab.com \
--cc=alexei.starovoitov@gmail.com \
--cc=andrii.nakryiko@gmail.com \
--cc=ast@kernel.org \
--cc=bpf@vger.kernel.org \
--cc=cong.wang@bytedance.com \
--cc=daniel@iogearbox.net \
--cc=netdev@vger.kernel.org \
--cc=wangdongdong.6@bytedance.com \
--cc=xiyou.wangcong@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).