From: Andrii Nakryiko <andrii.nakryiko@gmail.com>
To: Roman Gushchin <guro@fb.com>
Cc: bpf <bpf@vger.kernel.org>, Networking <netdev@vger.kernel.org>,
Alexei Starovoitov <ast@kernel.org>,
Daniel Borkmann <daniel@iogearbox.net>,
Kernel Team <kernel-team@fb.com>,
open list <linux-kernel@vger.kernel.org>
Subject: Re: [PATCH bpf-next v2 29/35] bpf: libbpf: cleanup RLIMIT_MEMLOCK usage
Date: Thu, 30 Jul 2020 12:39:40 -0700 [thread overview]
Message-ID: <CAEf4BzaZhyus7Kd-08vrVW9sr6gHGj1mCBgUY-NCWUOfdEJgHw@mail.gmail.com> (raw)
In-Reply-To: <20200730013836.GA637520@carbon.dhcp.thefacebook.com>
On Wed, Jul 29, 2020 at 6:38 PM Roman Gushchin <guro@fb.com> wrote:
>
> On Mon, Jul 27, 2020 at 10:59:33PM -0700, Andrii Nakryiko wrote:
> > On Mon, Jul 27, 2020 at 4:15 PM Roman Gushchin <guro@fb.com> wrote:
> > >
> > > On Mon, Jul 27, 2020 at 03:05:11PM -0700, Andrii Nakryiko wrote:
> > > > On Mon, Jul 27, 2020 at 12:21 PM Roman Gushchin <guro@fb.com> wrote:
> > > > >
> > > > > As bpf is not using memlock rlimit for memory accounting anymore,
> > > > > let's remove the related code from libbpf.
> > > > >
> > > > > Bpf operations can't fail because of exceeding the limit anymore.
> > > > >
> > > >
> > > > They can't in the newest kernel, but libbpf will keep working and
> > > > supporting old kernels for a very long time now. So please don't
> > > > remove any of this.
> > >
> > > Yeah, good point, agree.
> > > So we just can drop this patch from the series, no other changes
> > > are needed.
> > >
> > > >
> > > > But it would be nice to add a detection of whether kernel needs a
> > > > RLIMIT_MEMLOCK bump or not. Is there some simple and reliable way to
> > > > detect this from user-space?
>
> Btw, do you mean we should add a new function to the libbpf API?
> Or just extend pr_perm_msg() to skip guessing on new kernels?
>
I think we have to do both. There is libbpf_util.h in libbpf, we could
add two functions there:
- libbpf_needs_memlock() that would return true/false if kernel is old
and needs RLIMIT_MEMLOCK
- as a convenience, we can also add libbpf_inc_memlock_by() and
libbpf_set_memlock_to(), which will optionally (if kernel needs it)
adjust RLIMIT_MEMLOCK?
I think for your patch set, given it's pretty big already, let's not
touch runqslower, libbpf, and perf code (I think samples/bpf are fine
to just remove memlock adjustment), and we'll deal with detection and
optional bumping of RLIMIT_MEMLOCK as a separate patch once your
change land.
> The problem with the latter one is that it's called on a failed attempt
> to create a map, so unlikely we'll be able to create a new one just to test
> for the "memlock" value. But it also raises a question what should we do
> if the creation of this temporarily map fails? Assume the old kernel and
> bump the limit?
Yeah, I think we'll have to make assumptions like that. Ideally, of
course, detection of this would be just a simple sysfs value or
something, don't know. Maybe there is already a way for kernel to
communicate something like that?
> Idk, maybe it's better to just leave the userspace code as it is for some time.
>
> Thanks!
next prev parent reply other threads:[~2020-07-30 19:39 UTC|newest]
Thread overview: 88+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-07-27 18:44 [PATCH bpf-next v2 00/35] bpf: switch to memcg-based memory accounting Roman Gushchin
2020-07-27 18:44 ` [PATCH bpf-next v2 01/35] bpf: memcg-based memory accounting for bpf progs Roman Gushchin
2020-07-27 22:11 ` Song Liu
2020-07-28 0:08 ` Roman Gushchin
2020-07-28 4:42 ` Song Liu
2020-07-27 18:44 ` [PATCH bpf-next v2 02/35] bpf: memcg-based memory accounting for bpf maps Roman Gushchin
2020-07-27 22:12 ` Song Liu
2020-07-27 18:44 ` [PATCH bpf-next v2 03/35] bpf: refine memcg-based memory accounting for arraymap maps Roman Gushchin
2020-07-27 22:30 ` Song Liu
2020-07-27 18:44 ` [PATCH bpf-next v2 04/35] bpf: refine memcg-based memory accounting for cpumap maps Roman Gushchin
2020-07-27 22:48 ` Song Liu
2020-07-27 18:44 ` [PATCH bpf-next v2 05/35] bpf: memcg-based memory accounting for cgroup storage maps Roman Gushchin
2020-07-27 23:05 ` Song Liu
2020-07-27 18:44 ` [PATCH bpf-next v2 06/35] bpf: refine memcg-based memory accounting for devmap maps Roman Gushchin
2020-07-27 23:35 ` Song Liu
2020-07-27 18:44 ` [PATCH bpf-next v2 07/35] bpf: refine memcg-based memory accounting for hashtab maps Roman Gushchin
2020-07-27 23:36 ` Song Liu
2020-07-27 18:44 ` [PATCH bpf-next v2 08/35] bpf: memcg-based memory accounting for lpm_trie maps Roman Gushchin
2020-07-27 23:55 ` Song Liu
2020-07-27 18:44 ` [PATCH bpf-next v2 09/35] bpf: memcg-based memory accounting for bpf ringbuffer Roman Gushchin
2020-07-27 23:56 ` Song Liu
2020-07-27 18:44 ` [PATCH bpf-next v2 10/35] bpf: memcg-based memory accounting for socket storage maps Roman Gushchin
2020-07-27 23:57 ` Song Liu
2020-07-27 18:44 ` [PATCH bpf-next v2 11/35] bpf: refine memcg-based memory accounting for sockmap and sockhash maps Roman Gushchin
2020-07-27 23:58 ` Song Liu
2020-07-27 18:44 ` [PATCH bpf-next v2 12/35] bpf: refine memcg-based memory accounting for xskmap maps Roman Gushchin
2020-07-28 0:01 ` Song Liu
2020-07-27 18:44 ` [PATCH bpf-next v2 13/35] bpf: eliminate rlimit-based memory accounting for arraymap maps Roman Gushchin
2020-07-28 0:04 ` Song Liu
2020-07-27 18:44 ` [PATCH bpf-next v2 14/35] bpf: eliminate rlimit-based memory accounting for bpf_struct_ops maps Roman Gushchin
2020-07-28 5:29 ` Song Liu
2020-07-27 18:44 ` [PATCH bpf-next v2 15/35] bpf: eliminate rlimit-based memory accounting for cpumap maps Roman Gushchin
2020-07-28 5:30 ` Song Liu
2020-07-27 18:44 ` [PATCH bpf-next v2 16/35] bpf: eliminate rlimit-based memory accounting for cgroup storage maps Roman Gushchin
2020-07-28 5:31 ` Song Liu
2020-07-27 18:44 ` [PATCH bpf-next v2 17/35] bpf: eliminate rlimit-based memory accounting for devmap maps Roman Gushchin
2020-07-28 5:31 ` Song Liu
2020-07-27 18:44 ` [PATCH bpf-next v2 18/35] bpf: eliminate rlimit-based memory accounting for hashtab maps Roman Gushchin
2020-07-28 5:32 ` Song Liu
2020-07-27 18:44 ` [PATCH bpf-next v2 19/35] bpf: eliminate rlimit-based memory accounting for lpm_trie maps Roman Gushchin
2020-07-28 5:32 ` Song Liu
2020-07-27 18:44 ` [PATCH bpf-next v2 20/35] bpf: eliminate rlimit-based memory accounting for queue_stack_maps maps Roman Gushchin
2020-07-28 5:35 ` Song Liu
2020-07-27 18:44 ` [PATCH bpf-next v2 21/35] bpf: eliminate rlimit-based memory accounting for reuseport_array maps Roman Gushchin
2020-07-28 5:36 ` Song Liu
2020-07-27 18:44 ` [PATCH bpf-next v2 22/35] bpf: eliminate rlimit-based memory accounting for bpf ringbuffer Roman Gushchin
2020-07-28 5:37 ` Song Liu
2020-07-28 5:56 ` Andrii Nakryiko
2020-07-27 18:44 ` [PATCH bpf-next v2 23/35] bpf: eliminate rlimit-based memory accounting for sockmap and sockhash maps Roman Gushchin
2020-07-28 5:37 ` Song Liu
2020-07-27 18:44 ` [PATCH bpf-next v2 24/35] bpf: eliminate rlimit-based memory accounting for stackmap maps Roman Gushchin
2020-07-28 5:38 ` Song Liu
2020-07-27 18:44 ` [PATCH bpf-next v2 25/35] bpf: eliminate rlimit-based memory accounting for socket storage maps Roman Gushchin
2020-07-28 5:41 ` Song Liu
2020-07-27 18:44 ` [PATCH bpf-next v2 26/35] bpf: eliminate rlimit-based memory accounting for xskmap maps Roman Gushchin
2020-07-28 5:42 ` Song Liu
2020-07-27 18:44 ` [PATCH bpf-next v2 27/35] bpf: eliminate rlimit-based memory accounting infra for bpf maps Roman Gushchin
2020-07-28 5:47 ` Song Liu
2020-07-28 5:58 ` Andrii Nakryiko
2020-07-28 6:06 ` Song Liu
2020-07-28 19:08 ` Roman Gushchin
2020-07-28 19:16 ` Andrii Nakryiko
2020-07-27 18:44 ` [PATCH bpf-next v2 28/35] bpf: eliminate rlimit-based memory accounting for bpf progs Roman Gushchin
2020-07-28 5:55 ` Song Liu
2020-07-27 18:45 ` [PATCH bpf-next v2 29/35] bpf: libbpf: cleanup RLIMIT_MEMLOCK usage Roman Gushchin
2020-07-27 22:05 ` Andrii Nakryiko
2020-07-27 22:44 ` Song Liu
2020-07-27 23:15 ` Roman Gushchin
2020-07-28 5:59 ` Andrii Nakryiko
2020-07-30 1:38 ` Roman Gushchin
2020-07-30 19:39 ` Andrii Nakryiko [this message]
2020-07-30 20:46 ` Roman Gushchin
2020-07-27 18:45 ` [PATCH bpf-next v2 30/35] bpf: bpftool: do not touch RLIMIT_MEMLOCK Roman Gushchin
2020-07-28 6:00 ` Song Liu
2020-07-28 6:00 ` Andrii Nakryiko
2020-07-27 18:45 ` [PATCH bpf-next v2 31/35] bpf: runqslower: don't " Roman Gushchin
2020-07-28 6:03 ` Andrii Nakryiko
2020-07-27 18:45 ` [PATCH bpf-next v2 32/35] bpf: selftests: delete bpf_rlimit.h Roman Gushchin
2020-07-28 6:06 ` Andrii Nakryiko
2020-07-28 6:11 ` Song Liu
2020-07-28 18:30 ` Andrii Nakryiko
2020-07-27 18:45 ` [PATCH bpf-next v2 33/35] bpf: selftests: don't touch RLIMIT_MEMLOCK Roman Gushchin
2020-07-28 6:08 ` Andrii Nakryiko
2020-07-27 18:45 ` [PATCH bpf-next v2 34/35] bpf: samples: do not " Roman Gushchin
2020-07-28 6:14 ` Song Liu
2020-07-27 18:45 ` [PATCH bpf-next v2 35/35] perf: don't " Roman Gushchin
2020-07-28 6:09 ` Andrii Nakryiko
2020-07-28 12:13 ` Arnaldo Carvalho de Melo
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=CAEf4BzaZhyus7Kd-08vrVW9sr6gHGj1mCBgUY-NCWUOfdEJgHw@mail.gmail.com \
--to=andrii.nakryiko@gmail.com \
--cc=ast@kernel.org \
--cc=bpf@vger.kernel.org \
--cc=daniel@iogearbox.net \
--cc=guro@fb.com \
--cc=kernel-team@fb.com \
--cc=linux-kernel@vger.kernel.org \
--cc=netdev@vger.kernel.org \
/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).