From: Jiayuan Chen <jiayuan.chen@linux.dev>
To: bpf@vger.kernel.org
Cc: mrpre@163.com, Jiayuan Chen <jiayuan.chen@linux.dev>,
Alexei Starovoitov <ast@kernel.org>,
Daniel Borkmann <daniel@iogearbox.net>,
Andrii Nakryiko <andrii@kernel.org>,
Martin KaFai Lau <martin.lau@linux.dev>,
Eduard Zingerman <eddyz87@gmail.com>, Song Liu <song@kernel.org>,
Yonghong Song <yonghong.song@linux.dev>,
John Fastabend <john.fastabend@gmail.com>,
KP Singh <kpsingh@kernel.org>,
Stanislav Fomichev <sdf@fomichev.me>, Hao Luo <haoluo@google.com>,
Jiri Olsa <jolsa@kernel.org>,
"David S. Miller" <davem@davemloft.net>,
Eric Dumazet <edumazet@google.com>,
Jakub Kicinski <kuba@kernel.org>, Paolo Abeni <pabeni@redhat.com>,
Simon Horman <horms@kernel.org>,
Jesper Dangaard Brouer <hawk@kernel.org>,
Mykola Lysenko <mykolal@fb.com>, Shuah Khan <shuah@kernel.org>,
Willem de Bruijn <willemb@google.com>,
Jason Xing <kerneljasonxing@gmail.com>,
Anton Protopopov <aspsk@isovalent.com>,
Abhishek Chauhan <quic_abchauha@quicinc.com>,
Jordan Rome <linux@jordanrome.com>,
Martin Kelly <martin.kelly@crowdstrike.com>,
David Lechner <dlechner@baylibre.com>,
linux-kernel@vger.kernel.org, netdev@vger.kernel.org,
linux-kselftest@vger.kernel.org
Subject: [PATCH bpf v2 2/2] selftests/bpf: add perf test for adjust_{head,meta}
Date: Mon, 31 Mar 2025 11:23:45 +0800 [thread overview]
Message-ID: <20250331032354.75808-3-jiayuan.chen@linux.dev> (raw)
In-Reply-To: <20250331032354.75808-1-jiayuan.chen@linux.dev>
We added a memset operation during the adjust operation, which may cause
performance issues.
Therefore, we added perf testing, and testing found that for common header
length operations, memset() operation increased the performance overhead
by 2ns, which is negligible for the net stack.
Before memset
./test_progs -a xdp_adjust_head_perf -v
run adjust head with size 6 cost 56 ns
run adjust head with size 20 cost 56 ns
run adjust head with size 40 cost 56 ns
run adjust head with size 200 cost 56 ns
After memset
./test_progs -a xdp_adjust_head_perf -v
run adjust head with size 6 cost 58 ns
run adjust head with size 20 cost 58 ns
run adjust head with size 40 cost 58 ns
run adjust head with size 200 cost 66 ns
Signed-off-by: Jiayuan Chen <jiayuan.chen@linux.dev>
---
.../selftests/bpf/prog_tests/xdp_perf.c | 52 ++++++++++++++++---
tools/testing/selftests/bpf/progs/xdp_dummy.c | 14 +++++
2 files changed, 59 insertions(+), 7 deletions(-)
diff --git a/tools/testing/selftests/bpf/prog_tests/xdp_perf.c b/tools/testing/selftests/bpf/prog_tests/xdp_perf.c
index ec5369f247cb..1b4260c6e5d7 100644
--- a/tools/testing/selftests/bpf/prog_tests/xdp_perf.c
+++ b/tools/testing/selftests/bpf/prog_tests/xdp_perf.c
@@ -1,10 +1,11 @@
// SPDX-License-Identifier: GPL-2.0
#include <test_progs.h>
+#include <network_helpers.h>
+#include "xdp_dummy.skel.h"
void test_xdp_perf(void)
{
- const char *file = "./xdp_dummy.bpf.o";
- struct bpf_object *obj;
+ struct xdp_dummy *skel;
char in[128], out[128];
int err, prog_fd;
LIBBPF_OPTS(bpf_test_run_opts, topts,
@@ -15,14 +16,51 @@ void test_xdp_perf(void)
.repeat = 1000000,
);
- err = bpf_prog_test_load(file, BPF_PROG_TYPE_XDP, &obj, &prog_fd);
- if (CHECK_FAIL(err))
- return;
-
+ skel = xdp_dummy__open_and_load();
+ prog_fd = bpf_program__fd(skel->progs.xdp_dummy_prog);
err = bpf_prog_test_run_opts(prog_fd, &topts);
ASSERT_OK(err, "test_run");
ASSERT_EQ(topts.retval, XDP_PASS, "test_run retval");
ASSERT_EQ(topts.data_size_out, 128, "test_run data_size_out");
- bpf_object__close(obj);
+ xdp_dummy__destroy(skel);
+}
+
+void test_xdp_adjust_head_perf(void)
+{
+ struct xdp_dummy *skel;
+ int repeat = 9000000;
+ struct xdp_md ctx_in;
+ char data[100];
+ int err, prog_fd;
+ size_t test_header_size[] = {
+ ETH_ALEN,
+ sizeof(struct iphdr),
+ sizeof(struct ipv6hdr),
+ 200,
+ };
+ DECLARE_LIBBPF_OPTS(bpf_test_run_opts, topts,
+ .data_in = &data,
+ .data_size_in = sizeof(data),
+ .repeat = repeat,
+ );
+
+ topts.ctx_in = &ctx_in;
+ topts.ctx_size_in = sizeof(ctx_in);
+ memset(&ctx_in, 0, sizeof(ctx_in));
+ ctx_in.data_meta = 0;
+ ctx_in.data_end = ctx_in.data + sizeof(data);
+
+ skel = xdp_dummy__open_and_load();
+ prog_fd = bpf_program__fd(skel->progs.xdp_dummy_adjust_head);
+
+ for (int i = 0; i < ARRAY_SIZE(test_header_size); i++) {
+ skel->bss->head_size = test_header_size[i];
+ err = bpf_prog_test_run_opts(prog_fd, &topts);
+ ASSERT_OK(err, "test_run");
+ ASSERT_EQ(topts.retval, XDP_PASS, "test_run retval");
+ fprintf(stdout, "run adjust head with size %zd cost %d ns\n",
+ test_header_size[i], topts.duration);
+ }
+ xdp_dummy__destroy(skel);
}
diff --git a/tools/testing/selftests/bpf/progs/xdp_dummy.c b/tools/testing/selftests/bpf/progs/xdp_dummy.c
index d988b2e0cee8..7bebedbbc949 100644
--- a/tools/testing/selftests/bpf/progs/xdp_dummy.c
+++ b/tools/testing/selftests/bpf/progs/xdp_dummy.c
@@ -4,10 +4,24 @@
#include <linux/bpf.h>
#include <bpf/bpf_helpers.h>
+int head_size;
+
SEC("xdp")
int xdp_dummy_prog(struct xdp_md *ctx)
{
return XDP_PASS;
}
+SEC("xdp")
+int xdp_dummy_adjust_head(struct xdp_md *ctx)
+{
+ if (bpf_xdp_adjust_head(ctx, -head_size))
+ return XDP_DROP;
+
+ if (bpf_xdp_adjust_head(ctx, head_size))
+ return XDP_DROP;
+
+ return XDP_PASS;
+}
+
char _license[] SEC("license") = "GPL";
--
2.47.1
next prev parent reply other threads:[~2025-03-31 3:27 UTC|newest]
Thread overview: 11+ messages / expand[flat|nested] mbox.gz Atom feed top
2025-03-31 3:23 [PATCH bpf v2 0/2] bpf, xdp: clean adjust_{head,meta} memory when offset < 0 Jiayuan Chen
2025-03-31 3:23 ` [PATCH bpf v2 1/2] bpf, xdp: clean head/meta when expanding it Jiayuan Chen
2025-04-03 8:17 ` Jesper Dangaard Brouer
2025-04-03 14:24 ` Alexei Starovoitov
2025-04-03 14:32 ` Willem de Bruijn
2025-04-04 0:28 ` Alexei Starovoitov
2025-04-04 0:27 ` Jiayuan Chen
2025-04-04 0:29 ` Alexei Starovoitov
2025-03-31 3:23 ` Jiayuan Chen [this message]
2025-04-03 0:24 ` [PATCH bpf v2 2/2] selftests/bpf: add perf test for adjust_{head,meta} Jakub Kicinski
2025-04-03 9:37 ` Jesper Dangaard Brouer
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=20250331032354.75808-3-jiayuan.chen@linux.dev \
--to=jiayuan.chen@linux.dev \
--cc=andrii@kernel.org \
--cc=aspsk@isovalent.com \
--cc=ast@kernel.org \
--cc=bpf@vger.kernel.org \
--cc=daniel@iogearbox.net \
--cc=davem@davemloft.net \
--cc=dlechner@baylibre.com \
--cc=eddyz87@gmail.com \
--cc=edumazet@google.com \
--cc=haoluo@google.com \
--cc=hawk@kernel.org \
--cc=horms@kernel.org \
--cc=john.fastabend@gmail.com \
--cc=jolsa@kernel.org \
--cc=kerneljasonxing@gmail.com \
--cc=kpsingh@kernel.org \
--cc=kuba@kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-kselftest@vger.kernel.org \
--cc=linux@jordanrome.com \
--cc=martin.kelly@crowdstrike.com \
--cc=martin.lau@linux.dev \
--cc=mrpre@163.com \
--cc=mykolal@fb.com \
--cc=netdev@vger.kernel.org \
--cc=pabeni@redhat.com \
--cc=quic_abchauha@quicinc.com \
--cc=sdf@fomichev.me \
--cc=shuah@kernel.org \
--cc=song@kernel.org \
--cc=willemb@google.com \
--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;
as well as URLs for NNTP newsgroup(s).