From: Oleg Nesterov <oleg@redhat.com>
To: Yafang Shao <laoar.shao@gmail.com>
Cc: Jiri Olsa <jolsa@kernel.org>, Alexei Starovoitov <ast@kernel.org>,
Daniel Borkmann <daniel@iogearbox.net>,
Andrii Nakryiko <andrii@kernel.org>,
bpf@vger.kernel.org, Martin KaFai Lau <kafai@fb.com>,
Song Liu <songliubraving@fb.com>, Yonghong Song <yhs@fb.com>,
John Fastabend <john.fastabend@gmail.com>,
KP Singh <kpsingh@chromium.org>,
Stanislav Fomichev <sdf@google.com>, Hao Luo <haoluo@google.com>
Subject: Re: [PATCHv4 bpf-next 05/28] bpf: Add pid filter support for uprobe_multi link
Date: Fri, 21 Jul 2023 10:31:41 +0200 [thread overview]
Message-ID: <20230721083140.GA10521@redhat.com> (raw)
In-Reply-To: <CALOAHbB3_qTzi+2_0=pFjyDXFUh_MGMJt6gz7eh0Z=He4guPow@mail.gmail.com>
On 07/21, Yafang Shao wrote:
>
> On Thu, Jul 20, 2023 at 7:36 PM Jiri Olsa <jolsa@kernel.org> wrote:
> >
> > Adding support to specify pid for uprobe_multi link and the uprobes
> > are created only for task with given pid value.
>
> Is it possible to use tgid as the filter?
> It would be helpful when we uprobe a library file but want to filter
> out a multi-threaded task only.
I leave this to you and Jiri, just some notes...
I think it is possible but needs some complications.
uprobe_prog_run() needs a trivial change,
- if (link->task && current != link->task)
+ if (link->task && same_thread_group(current, link->task)
->signal is tied to task_struct so this should work even if link->task has
already exited.
bpf_uprobe_multi_link_attach() should probably use
get_pid_task(find_vpid(pid), PIDTYPE_TGID);
to ensure that 'pid' is actually tgid.
But. uprobe_multi_link_filter() can return F if the group leader has already
exited, in this case uprobe->link->task->mm == NULL.
Perhaps we can simply ignore this case, a zombie group-leader is not that common.
Or we can add uprobe->link->mm = get_task_mm(uprobe->link->task->mm) for
uprobe_multi_link_filter().
Or. We can simply kill uprobe_multi_link_filter(). In this case uprobe_register()
can touch CLONE_VM'ed mm's we are not interested in, but everything should work
correctly.
Oleg.
next prev parent reply other threads:[~2023-07-21 8:32 UTC|newest]
Thread overview: 36+ messages / expand[flat|nested] mbox.gz Atom feed top
2023-07-20 11:35 [PATCHv4 bpf-next 00/28] bpf: Add multi uprobe link Jiri Olsa
2023-07-20 11:35 ` [PATCHv4 bpf-next 01/28] bpf: Switch BPF_F_KPROBE_MULTI_RETURN macro to enum Jiri Olsa
2023-07-20 11:35 ` [PATCHv4 bpf-next 02/28] bpf: Add attach_type checks under bpf_prog_attach_check_attach_type Jiri Olsa
2023-07-20 11:35 ` [PATCHv4 bpf-next 03/28] bpf: Add multi uprobe link Jiri Olsa
2023-07-20 11:35 ` [PATCHv4 bpf-next 04/28] bpf: Add cookies support for uprobe_multi link Jiri Olsa
2023-07-21 2:18 ` Yafang Shao
2023-07-21 7:39 ` Jiri Olsa
2023-07-20 11:35 ` [PATCHv4 bpf-next 05/28] bpf: Add pid filter " Jiri Olsa
2023-07-21 2:55 ` Yafang Shao
2023-07-21 8:31 ` Oleg Nesterov [this message]
2023-07-21 8:54 ` Oleg Nesterov
2023-07-21 9:05 ` Yafang Shao
2023-07-21 10:41 ` Jiri Olsa
2023-07-20 11:35 ` [PATCHv4 bpf-next 06/28] bpf: Add bpf_get_func_ip helper support for uprobe link Jiri Olsa
2023-07-20 11:35 ` [PATCHv4 bpf-next 07/28] libbpf: Add uprobe_multi attach type and link names Jiri Olsa
2023-07-20 11:35 ` [PATCHv4 bpf-next 08/28] libbpf: Move elf_find_func_offset* functions to elf object Jiri Olsa
2023-07-20 11:35 ` [PATCHv4 bpf-next 09/28] libbpf: Add elf_open/elf_close functions Jiri Olsa
2023-07-20 11:35 ` [PATCHv4 bpf-next 10/28] libbpf: Add elf symbol iterator Jiri Olsa
2023-07-20 11:35 ` [PATCHv4 bpf-next 11/28] libbpf: Add elf_resolve_syms_offsets function Jiri Olsa
2023-07-20 11:35 ` [PATCHv4 bpf-next 12/28] libbpf: Add elf_resolve_pattern_offsets function Jiri Olsa
2023-07-20 11:35 ` [PATCHv4 bpf-next 13/28] libbpf: Add bpf_link_create support for multi uprobes Jiri Olsa
2023-07-20 11:35 ` [PATCHv4 bpf-next 14/28] libbpf: Add bpf_program__attach_uprobe_multi function Jiri Olsa
2023-07-20 11:35 ` [PATCHv4 bpf-next 15/28] libbpf: Add support for u[ret]probe.multi[.s] program sections Jiri Olsa
2023-07-20 11:35 ` [PATCHv4 bpf-next 16/28] libbpf: Add uprobe multi link detection Jiri Olsa
2023-07-20 11:35 ` [PATCHv4 bpf-next 17/28] libbpf: Add uprobe multi link support to bpf_program__attach_usdt Jiri Olsa
2023-07-20 11:35 ` [PATCHv4 bpf-next 18/28] selftests/bpf: Move get_time_ns to testing_helpers.h Jiri Olsa
2023-07-20 11:35 ` [PATCHv4 bpf-next 19/28] selftests/bpf: Add uprobe_multi skel test Jiri Olsa
2023-07-20 11:35 ` [PATCHv4 bpf-next 20/28] selftests/bpf: Add uprobe_multi api test Jiri Olsa
2023-07-20 11:35 ` [PATCHv4 bpf-next 21/28] selftests/bpf: Add uprobe_multi link test Jiri Olsa
2023-07-20 11:35 ` [PATCHv4 bpf-next 22/28] selftests/bpf: Add uprobe_multi test program Jiri Olsa
2023-07-20 11:35 ` [PATCHv4 bpf-next 23/28] selftests/bpf: Add uprobe_multi bench test Jiri Olsa
2023-07-20 11:35 ` [PATCHv4 bpf-next 24/28] selftests/bpf: Add uprobe_multi usdt test code Jiri Olsa
2023-07-20 11:35 ` [PATCHv4 bpf-next 25/28] selftests/bpf: Add uprobe_multi usdt bench test Jiri Olsa
2023-07-20 11:35 ` [PATCHv4 bpf-next 26/28] selftests/bpf: Add uprobe_multi cookie test Jiri Olsa
2023-07-20 11:35 ` [PATCHv4 bpf-next 27/28] selftests/bpf: Add uprobe_multi pid filter tests Jiri Olsa
2023-07-20 11:35 ` [PATCHv4 bpf-next 28/28] selftests/bpf: Add extra link to uprobe_multi tests 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=20230721083140.GA10521@redhat.com \
--to=oleg@redhat.com \
--cc=andrii@kernel.org \
--cc=ast@kernel.org \
--cc=bpf@vger.kernel.org \
--cc=daniel@iogearbox.net \
--cc=haoluo@google.com \
--cc=john.fastabend@gmail.com \
--cc=jolsa@kernel.org \
--cc=kafai@fb.com \
--cc=kpsingh@chromium.org \
--cc=laoar.shao@gmail.com \
--cc=sdf@google.com \
--cc=songliubraving@fb.com \
--cc=yhs@fb.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).