All of lore.kernel.org
 help / color / mirror / Atom feed
From: Oleg Nesterov <oleg@redhat.com>
To: Jiri Olsa <olsajiri@gmail.com>
Cc: Tianyi Liu <i.pear@outlook.com>,
	andrii.nakryiko@gmail.com, mhiramat@kernel.org, ajor@meta.com,
	albancrequy@linux.microsoft.com, bpf@vger.kernel.org,
	flaniel@linux.microsoft.com, linux-trace-kernel@vger.kernel.org,
	linux@jordanrome.com, mathieu.desnoyers@efficios.com
Subject: Re: [PATCH v2] tracing/uprobe: Add missing PID filter for uretprobe
Date: Tue, 27 Aug 2024 18:45:45 +0200	[thread overview]
Message-ID: <20240827164545.GG30765@redhat.com> (raw)
In-Reply-To: <Zs3PdV6nqed1jWC2@krava>

On 08/27, Jiri Olsa wrote:
>
> On Tue, Aug 27, 2024 at 12:29:38AM +0200, Oleg Nesterov wrote:
> >
> > So, can you reproduce the problem reported by Tianyi on your setup?
>
> yes, I can repduce the issue with uretprobe on top of perf event uprobe

...

>    ->     uretprobe-hit
>             handle_swbp
>               uprobe_handle_trampoline
>                 handle_uretprobe_chain
>                 {
>
>                   for_each_uprobe_consumer {
>
>                     // consumer for task 1019
>                     uretprobe_dispatcher
>                       uretprobe_perf_func
>                         -> runs bpf program
>
>                     // consumer for task 1018
>                     uretprobe_dispatcher
>                       uretprobe_perf_func
>                         -> runs bpf program

Confused...

I naively thought that if bpftrace uses bpf_uprobe_multi_link_attach() then
it won't use perf/trace_uprobe, and uretprobe-hit will result in

	// current->pid == 1018

	for_each_uprobe_consumer {
		// consumer for task 1019
		uprobe_multi_link_ret_handler
		    uprobe_prog_run
		       -> current->mm != link->task->mm, return

		// consumer for task 1018
		uprobe_multi_link_ret_handler
		    uprobe_prog_run
		       -> current->mm == link->task->mm, run bpf
	}

> I think the uretprobe_dispatcher could call filter as suggested in the original
> patch..

OK, agreed.

> but I'm not sure we need to remove the uprobe from handle_uretprobe_chain
> like we do in handler_chain..

Me too. In any case this is another issue.

Oleg.


  parent reply	other threads:[~2024-08-27 16:46 UTC|newest]

Thread overview: 57+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2024-08-23 13:53 [PATCH v2] tracing/uprobe: Add missing PID filter for uretprobe Tianyi Liu
2024-08-23 17:44 ` Masami Hiramatsu
2024-08-23 19:07   ` Andrii Nakryiko
2024-08-24  5:49     ` Tianyi Liu
2024-08-24 17:27       ` Masami Hiramatsu
2024-08-25 17:14       ` Oleg Nesterov
2024-08-25 18:43         ` Oleg Nesterov
2024-08-25 22:40         ` Oleg Nesterov
2024-08-26 10:05           ` Jiri Olsa
2024-08-26 11:57             ` Oleg Nesterov
2024-08-26 12:24               ` Oleg Nesterov
2024-08-26 13:48               ` Jiri Olsa
2024-08-26 18:56                 ` Oleg Nesterov
2024-08-26 21:25                 ` Oleg Nesterov
2024-08-26 22:01                   ` Jiri Olsa
2024-08-26 22:08                     ` Andrii Nakryiko
2024-08-26 22:29                     ` Oleg Nesterov
2024-08-27 13:07                       ` Jiri Olsa
2024-08-27 13:45                         ` Jiri Olsa
2024-08-27 16:45                         ` Oleg Nesterov [this message]
2024-08-28 11:40                           ` Jiri Olsa
2024-08-27 20:19                         ` Oleg Nesterov
2024-08-28 11:46                           ` Jiri Olsa
2024-08-29 15:20                             ` Oleg Nesterov
2024-08-29 19:46                               ` Jiri Olsa
2024-08-29 21:12                                 ` Oleg Nesterov
2024-08-29 23:22                                   ` Jiri Olsa
2024-08-27  6:27                   ` Tianyi Liu
2024-08-27 10:08               ` Jiri Olsa
2024-08-27 10:20                 ` Jiri Olsa
2024-08-27 10:54                   ` Oleg Nesterov
2024-08-27 10:40                 ` Oleg Nesterov
2024-08-27 13:32                   ` Jiri Olsa
2024-08-27 14:26                     ` Oleg Nesterov
2024-08-27 14:41                       ` Jiri Olsa
2024-08-26 14:52           ` Tianyi Liu
2024-08-25 17:00     ` Oleg Nesterov
2024-08-30 10:12 ` Oleg Nesterov
2024-08-30 12:23   ` Oleg Nesterov
2024-08-30 13:34   ` Jiri Olsa
2024-08-30 15:51     ` Andrii Nakryiko
2024-09-02  9:11       ` Jiri Olsa
2024-09-03 18:09         ` Andrii Nakryiko
2024-09-03 18:11           ` Andrii Nakryiko
2024-09-03 19:15             ` Jiri Olsa
2024-09-01 19:22   ` Tianyi Liu
2024-09-01 23:26     ` Oleg Nesterov
2024-09-02 17:17       ` Oleg Nesterov
2024-09-03 14:33         ` Jiri Olsa
2024-09-06 10:43     ` Jiri Olsa
2024-09-06 19:18       ` Oleg Nesterov
2024-09-09 10:41         ` Jiri Olsa
2024-09-09 18:34           ` Oleg Nesterov
2024-09-10  8:45             ` Jiri Olsa
2024-09-07 19:19       ` Tianyi Liu
2024-09-08 13:15         ` Oleg Nesterov
2024-09-09  1:16           ` Andrii Nakryiko

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=20240827164545.GG30765@redhat.com \
    --to=oleg@redhat.com \
    --cc=ajor@meta.com \
    --cc=albancrequy@linux.microsoft.com \
    --cc=andrii.nakryiko@gmail.com \
    --cc=bpf@vger.kernel.org \
    --cc=flaniel@linux.microsoft.com \
    --cc=i.pear@outlook.com \
    --cc=linux-trace-kernel@vger.kernel.org \
    --cc=linux@jordanrome.com \
    --cc=mathieu.desnoyers@efficios.com \
    --cc=mhiramat@kernel.org \
    --cc=olsajiri@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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.