From: Dmitrii Dolgov <9erthalion6@gmail.com>
To: bpf@vger.kernel.org, ast@kernel.org, andrii@kernel.org,
yhs@fb.com, songliubraving@fb.com
Cc: Dmitrii Dolgov <9erthalion6@gmail.com>
Subject: [RFC PATCH bpf-next 2/2] libbpf: Allow setting bpf_prog priority
Date: Sun, 3 Apr 2022 18:07:18 +0200 [thread overview]
Message-ID: <20220403160718.13730-3-9erthalion6@gmail.com> (raw)
In-Reply-To: <20220403160718.13730-1-9erthalion6@gmail.com>
Introduce prio field in opts structures (bpf_tracepoint_opts,
bpf_link_create_opts, bpf_perf_event_opts) to pass a priority value for
the bpf_prog when a new bpf link is created. This value will be further
used in perf_event_set_bpf_prog to specify in which order to execute
bpf_progs attached to the same tracepoint.
Signed-off-by: Dmitrii Dolgov <9erthalion6@gmail.com>
---
kernel/bpf/syscall.c | 3 ++-
tools/lib/bpf/bpf.c | 1 +
tools/lib/bpf/bpf.h | 1 +
tools/lib/bpf/libbpf.c | 4 +++-
tools/lib/bpf/libbpf.h | 6 ++++--
5 files changed, 11 insertions(+), 4 deletions(-)
diff --git a/kernel/bpf/syscall.c b/kernel/bpf/syscall.c
index 72fb3d2c30a4..629852b35b21 100644
--- a/kernel/bpf/syscall.c
+++ b/kernel/bpf/syscall.c
@@ -3009,7 +3009,8 @@ static int bpf_perf_link_attach(const union bpf_attr *attr, struct bpf_prog *pro
}
event = perf_file->private_data;
- err = perf_event_set_bpf_prog(event, prog, attr->link_create.perf_event.bpf_cookie, 0);
+ err = perf_event_set_bpf_prog(event, prog, attr->link_create.perf_event.bpf_cookie,
+ attr->link_create.perf_event.prio);
if (err) {
bpf_link_cleanup(&link_primer);
goto out_put_file;
diff --git a/tools/lib/bpf/bpf.c b/tools/lib/bpf/bpf.c
index cf27251adb92..029c9809bf9b 100644
--- a/tools/lib/bpf/bpf.c
+++ b/tools/lib/bpf/bpf.c
@@ -851,6 +851,7 @@ int bpf_link_create(int prog_fd, int target_fd,
break;
case BPF_PERF_EVENT:
attr.link_create.perf_event.bpf_cookie = OPTS_GET(opts, perf_event.bpf_cookie, 0);
+ attr.link_create.perf_event.prio = OPTS_GET(opts, perf_event.prio, 0);
if (!OPTS_ZEROED(opts, perf_event))
return libbpf_err(-EINVAL);
break;
diff --git a/tools/lib/bpf/bpf.h b/tools/lib/bpf/bpf.h
index f4b4afb6d4ba..9a8ec9081ba7 100644
--- a/tools/lib/bpf/bpf.h
+++ b/tools/lib/bpf/bpf.h
@@ -412,6 +412,7 @@ struct bpf_link_create_opts {
union {
struct {
__u64 bpf_cookie;
+ __u32 prio;
} perf_event;
struct {
__u32 flags;
diff --git a/tools/lib/bpf/libbpf.c b/tools/lib/bpf/libbpf.c
index 809fe209cdcc..e09c00b53772 100644
--- a/tools/lib/bpf/libbpf.c
+++ b/tools/lib/bpf/libbpf.c
@@ -9912,7 +9912,8 @@ struct bpf_link *bpf_program__attach_perf_event_opts(const struct bpf_program *p
if (kernel_supports(prog->obj, FEAT_PERF_LINK)) {
DECLARE_LIBBPF_OPTS(bpf_link_create_opts, link_opts,
- .perf_event.bpf_cookie = OPTS_GET(opts, bpf_cookie, 0));
+ .perf_event.bpf_cookie = OPTS_GET(opts, bpf_cookie, 0),
+ .perf_event.prio = OPTS_GET(opts, prio, 0));
link_fd = bpf_link_create(prog_fd, pfd, BPF_PERF_EVENT, &link_opts);
if (link_fd < 0) {
@@ -10663,6 +10664,7 @@ struct bpf_link *bpf_program__attach_tracepoint_opts(const struct bpf_program *p
return libbpf_err_ptr(-EINVAL);
pe_opts.bpf_cookie = OPTS_GET(opts, bpf_cookie, 0);
+ pe_opts.prio = OPTS_GET(opts, prio, 0);
pfd = perf_event_open_tracepoint(tp_category, tp_name);
if (pfd < 0) {
diff --git a/tools/lib/bpf/libbpf.h b/tools/lib/bpf/libbpf.h
index 05dde85e19a6..30f1808a4b49 100644
--- a/tools/lib/bpf/libbpf.h
+++ b/tools/lib/bpf/libbpf.h
@@ -394,8 +394,9 @@ struct bpf_perf_event_opts {
size_t sz;
/* custom user-provided value fetchable through bpf_get_attach_cookie() */
__u64 bpf_cookie;
+ __u32 prio;
};
-#define bpf_perf_event_opts__last_field bpf_cookie
+#define bpf_perf_event_opts__last_field prio
LIBBPF_API struct bpf_link *
bpf_program__attach_perf_event(const struct bpf_program *prog, int pfd);
@@ -508,8 +509,9 @@ struct bpf_tracepoint_opts {
size_t sz;
/* custom user-provided value fetchable through bpf_get_attach_cookie() */
__u64 bpf_cookie;
+ __u32 prio;
};
-#define bpf_tracepoint_opts__last_field bpf_cookie
+#define bpf_tracepoint_opts__last_field prio
LIBBPF_API struct bpf_link *
bpf_program__attach_tracepoint(const struct bpf_program *prog,
--
2.32.0
next prev parent reply other threads:[~2022-04-03 16:08 UTC|newest]
Thread overview: 7+ messages / expand[flat|nested] mbox.gz Atom feed top
2022-04-03 16:07 [RFC PATCH bpf-next 0/2] Priorities for bpf progs attached to the same tracepoint Dmitrii Dolgov
2022-04-03 16:07 ` [RFC PATCH bpf-next 1/2] bpf: tracing: Introduce prio field for bpf_prog Dmitrii Dolgov
2022-04-03 16:07 ` Dmitrii Dolgov [this message]
2022-04-04 0:17 ` [RFC PATCH bpf-next 0/2] Priorities for bpf progs attached to the same tracepoint Andrii Nakryiko
2022-04-04 15:29 ` Dmitry Dolgov
2022-04-05 17:20 ` Alexei Starovoitov
2022-04-06 8:12 ` Dmitry Dolgov
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=20220403160718.13730-3-9erthalion6@gmail.com \
--to=9erthalion6@gmail.com \
--cc=andrii@kernel.org \
--cc=ast@kernel.org \
--cc=bpf@vger.kernel.org \
--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).