From: Jiri Olsa <olsajiri@gmail.com>
To: Dmitry Dolgov <9erthalion6@gmail.com>
Cc: Jiri Olsa <olsajiri@gmail.com>,
bpf@vger.kernel.org, ast@kernel.org, daniel@iogearbox.net,
andrii@kernel.org, martin.lau@linux.dev, song@kernel.org,
yonghong.song@linux.dev, dan.carpenter@linaro.org
Subject: Re: [PATCH bpf-next v4 1/3] bpf: Relax tracing prog recursive attach rules
Date: Thu, 30 Nov 2023 23:34:05 +0100 [thread overview]
Message-ID: <ZWkN3QMnAGwFo7JF@krava> (raw)
In-Reply-To: <20231130185752.2fepd5wlospfzc53@erthalion.local>
On Thu, Nov 30, 2023 at 07:57:52PM +0100, Dmitry Dolgov wrote:
> > On Thu, Nov 30, 2023 at 03:30:38PM +0100, Jiri Olsa wrote:
> > > diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c
> > > index 8e7b6072e3f4..31ffcffb7198 100644
> > > --- a/kernel/bpf/verifier.c
> > > +++ b/kernel/bpf/verifier.c
> > > @@ -20109,6 +20109,12 @@ int bpf_check_attach_target(struct bpf_verifier_log *log,
> > > if (tgt_prog) {
> > > struct bpf_prog_aux *aux = tgt_prog->aux;
> > >
> > > + if (aux->attach_depth >= 32) {
> > > + bpf_log(log, "Target program attach depth is %d. Too large\n",
> > > + aux->attach_depth);
> > > + return -EINVAL;
> > > + }
> >
> > IIUC the use case you have is to be able to attach fentry program on
> > another fentry program.. do you have use case that needs more than
> > that?
> >
> > could we allow just single nesting? that might perhaps endup in easier
> > code while still allowing your use case?
>
> Right, there is no use case beyond attaching an fentry to another one,
> so having just a single nesting should be fine.
>
> > > + /*
> > > + * To avoid potential call chain cycles, prevent attaching programs
> > > + * of the same type. The only exception is standalone fentry/fexit
> > > + * programs that themselves are not attachment targets.
> > > + * That means:
> > > + * - Cannot attach followed fentry/fexit to another
> > > + * fentry/fexit program.
> > > + * - Cannot attach program extension to another extension.
> > > * It's ok to attach fentry/fexit to extension program.
> >
> > next condition below denies extension on fentry/fexit and the reason
> > is the possibility:
> > "... to create long call chain * fentry->extension->fentry->extension
> > beyond reasonable stack size ..."
> >
> > that might be problem also here with 32 allowed nesting
>
> Reducing nesting to only one level probably will lift this question, but
> for posterity, what kind of problem similar to "fentry->extension->fentry->..."
> do you have in mind?
I meant that allowing that amount of nesting will make it easier to get
to a point where we use all the available stack size
jirka
next prev parent reply other threads:[~2023-11-30 22:34 UTC|newest]
Thread overview: 18+ messages / expand[flat|nested] mbox.gz Atom feed top
2023-11-29 19:52 [PATCH bpf-next v4 0/3] Relax tracing prog recursive attach rules Dmitrii Dolgov
2023-11-29 19:52 ` [PATCH bpf-next v4 1/3] bpf: " Dmitrii Dolgov
2023-11-29 23:58 ` Song Liu
2023-11-30 10:08 ` Dmitry Dolgov
2023-11-30 20:19 ` Song Liu
2023-11-30 20:41 ` Dmitry Dolgov
2023-12-01 9:55 ` Artem Savkov
2023-12-01 14:29 ` Dmitry Dolgov
2023-11-30 14:30 ` Jiri Olsa
2023-11-30 18:57 ` Dmitry Dolgov
2023-11-30 22:34 ` Jiri Olsa [this message]
2023-11-29 19:52 ` [PATCH bpf-next v4 2/3] selftests/bpf: Add test for recursive attachment of tracing progs Dmitrii Dolgov
2023-11-30 14:47 ` Jiri Olsa
2023-11-29 19:52 ` [PATCH bpf-next v4 3/3] bpf, selftest/bpf: Fix re-attachment branch in bpf_tracing_prog_attach Dmitrii Dolgov
2023-11-30 15:14 ` Jiri Olsa
2023-11-30 22:30 ` Jiri Olsa
2023-12-01 14:21 ` Dmitry Dolgov
2023-12-01 14:52 ` Jiri Olsa
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=ZWkN3QMnAGwFo7JF@krava \
--to=olsajiri@gmail.com \
--cc=9erthalion6@gmail.com \
--cc=andrii@kernel.org \
--cc=ast@kernel.org \
--cc=bpf@vger.kernel.org \
--cc=dan.carpenter@linaro.org \
--cc=daniel@iogearbox.net \
--cc=martin.lau@linux.dev \
--cc=song@kernel.org \
--cc=yonghong.song@linux.dev \
/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