From: Peter Zijlstra <peterz@infradead.org>
To: Andrii Nakryiko <andrii@kernel.org>
Cc: bpf@vger.kernel.org, ast@kernel.org, daniel@iogearbox.net,
kernel-team@fb.com
Subject: Re: [PATCH v2 bpf-next 05/14] bpf: allow to specify user-provided context value for BPF perf links
Date: Tue, 27 Jul 2021 11:11:48 +0200 [thread overview]
Message-ID: <YP/N1HR6GAanBd9m@hirez.programming.kicks-ass.net> (raw)
In-Reply-To: <20210726161211.925206-6-andrii@kernel.org>
On Mon, Jul 26, 2021 at 09:12:02AM -0700, Andrii Nakryiko wrote:
> Add ability for users to specify custom u64 value when creating BPF link for
> perf_event-backed BPF programs (kprobe/uprobe, perf_event, tracepoints).
If I read this right, the value is dependent on the link, not the
program. In which case:
> diff --git a/include/linux/perf_event.h b/include/linux/perf_event.h
> index 2d510ad750ed..97ab46802800 100644
> --- a/include/linux/perf_event.h
> +++ b/include/linux/perf_event.h
> @@ -762,6 +762,7 @@ struct perf_event {
> #ifdef CONFIG_BPF_SYSCALL
> perf_overflow_handler_t orig_overflow_handler;
> struct bpf_prog *prog;
> + u64 user_ctx;
> #endif
>
> #ifdef CONFIG_EVENT_TRACING
> diff --git a/include/linux/trace_events.h b/include/linux/trace_events.h
> index 8ac92560d3a3..4543852f1480 100644
> --- a/include/linux/trace_events.h
> +++ b/include/linux/trace_events.h
> @@ -675,7 +675,7 @@ trace_trigger_soft_disabled(struct trace_event_file *file)
>
> #ifdef CONFIG_BPF_EVENTS
> unsigned int trace_call_bpf(struct trace_event_call *call, void *ctx);
> -int perf_event_attach_bpf_prog(struct perf_event *event, struct bpf_prog *prog);
> +int perf_event_attach_bpf_prog(struct perf_event *event, struct bpf_prog *prog, u64 user_ctx);
This API would be misleading, because it is about setting the program.
> void perf_event_detach_bpf_prog(struct perf_event *event);
> int perf_event_query_prog_array(struct perf_event *event, void __user *info);
> int bpf_probe_register(struct bpf_raw_event_map *btp, struct bpf_prog *prog);
> @@ -9966,6 +9968,7 @@ static int perf_event_set_bpf_handler(struct perf_event *event, struct bpf_prog
> }
>
> event->prog = prog;
> + event->user_ctx = user_ctx;
> event->orig_overflow_handler = READ_ONCE(event->overflow_handler);
> WRITE_ONCE(event->overflow_handler, bpf_overflow_handler);
> return 0;
Also, the name @user_ctx is a bit confusing. Would something like
@bpf_cookie or somesuch not be a better name?
Combined would it not make more sense to add something like:
extern int perf_event_set_bpf_cookie(struct perf_event *event, u64 cookie);
next prev parent reply other threads:[~2021-07-27 9:14 UTC|newest]
Thread overview: 43+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-07-26 16:11 [PATCH v2 bpf-next 00/14] BPF perf link and user-provided context value Andrii Nakryiko
2021-07-26 16:11 ` [PATCH v2 bpf-next 01/14] bpf: refactor BPF_PROG_RUN into a function Andrii Nakryiko
2021-07-29 16:49 ` Yonghong Song
2021-07-30 4:05 ` Andrii Nakryiko
2021-07-26 16:11 ` [PATCH v2 bpf-next 02/14] bpf: refactor BPF_PROG_RUN_ARRAY family of macros into functions Andrii Nakryiko
2021-07-29 17:04 ` Yonghong Song
2021-07-26 16:12 ` [PATCH v2 bpf-next 03/14] bpf: refactor perf_event_set_bpf_prog() to use struct bpf_prog input Andrii Nakryiko
2021-07-27 8:48 ` Peter Zijlstra
2021-07-29 17:09 ` Yonghong Song
2021-07-26 16:12 ` [PATCH v2 bpf-next 04/14] bpf: implement minimal BPF perf link Andrii Nakryiko
2021-07-27 9:04 ` Peter Zijlstra
2021-07-30 4:23 ` Andrii Nakryiko
2021-07-27 9:12 ` Peter Zijlstra
2021-07-27 20:56 ` Andrii Nakryiko
2021-07-27 15:40 ` Jiri Olsa
2021-07-27 20:56 ` Andrii Nakryiko
2021-07-29 17:35 ` Yonghong Song
2021-07-30 4:16 ` Andrii Nakryiko
2021-07-30 5:42 ` Yonghong Song
2021-07-26 16:12 ` [PATCH v2 bpf-next 05/14] bpf: allow to specify user-provided context value for BPF perf links Andrii Nakryiko
2021-07-27 9:11 ` Peter Zijlstra [this message]
2021-07-27 21:09 ` Andrii Nakryiko
2021-07-28 8:58 ` Peter Zijlstra
2021-07-29 18:00 ` Yonghong Song
2021-07-30 4:31 ` Andrii Nakryiko
2021-07-30 5:49 ` Yonghong Song
2021-07-30 17:48 ` Andrii Nakryiko
2021-07-30 21:34 ` Yonghong Song
2021-07-30 22:06 ` Andrii Nakryiko
2021-07-30 22:28 ` Yonghong Song
2021-07-26 16:12 ` [PATCH v2 bpf-next 06/14] bpf: add bpf_get_user_ctx() BPF helper to access user_ctx value Andrii Nakryiko
2021-07-29 18:17 ` Yonghong Song
2021-07-30 4:49 ` Andrii Nakryiko
2021-07-30 5:53 ` Yonghong Song
2021-07-26 16:12 ` [PATCH v2 bpf-next 07/14] libbpf: re-build libbpf.so when libbpf.map changes Andrii Nakryiko
2021-07-26 16:12 ` [PATCH v2 bpf-next 08/14] libbpf: remove unused bpf_link's destroy operation, but add dealloc Andrii Nakryiko
2021-07-26 16:12 ` [PATCH v2 bpf-next 09/14] libbpf: use BPF perf link when supported by kernel Andrii Nakryiko
2021-07-26 16:12 ` [PATCH v2 bpf-next 10/14] libbpf: add user_ctx support to bpf_link_create() API Andrii Nakryiko
2021-07-26 16:12 ` [PATCH v2 bpf-next 11/14] libbpf: add user_ctx to perf_event, kprobe, uprobe, and tp attach APIs Andrii Nakryiko
2021-07-30 1:11 ` Rafael David Tinoco
2021-07-26 16:12 ` [PATCH v2 bpf-next 12/14] selftests/bpf: test low-level perf BPF link API Andrii Nakryiko
2021-07-26 16:12 ` [PATCH v2 bpf-next 13/14] selftests/bpf: extract uprobe-related helpers into trace_helpers.{c,h} Andrii Nakryiko
2021-07-26 16:12 ` [PATCH v2 bpf-next 14/14] selftests/bpf: add user_ctx selftests for high-level APIs 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=YP/N1HR6GAanBd9m@hirez.programming.kicks-ass.net \
--to=peterz@infradead.org \
--cc=andrii@kernel.org \
--cc=ast@kernel.org \
--cc=bpf@vger.kernel.org \
--cc=daniel@iogearbox.net \
--cc=kernel-team@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).