[PATCH bpf-next v3 0/3] selftests/bpf: add coverage for xdp_features in test_progs

[PATCH bpf-next v3 0/3] selftests/bpf: add coverage for xdp_features in test_progs

[Alexis Lothoré (eBPF Foundation)]

Hello,
this small series aims to increase coverage of xdp features in
test_progs. The initial versions proposed to rework test_xdp_features.sh
to make it fit in test_progs, but some discussions in v1 and v2 showed
that the script is still needed as a standalone tool. So this new
revision lets test_xdp_features.sh as-is, and rather adds missing
coverage in existing test (cpu map). The new revision is now also a
follow-up to the update performed by Florian Kauer in [1] for devmap
programs testing.

[1] https://lore.kernel.org/bpf/20240911-devel-koalo-fix-ingress-ifindex-v4-2-5c643ae10258@linutronix.de/

---
Changes in v3:
- Drop xdp_features rework commit
- update xdp_cpumap_attach to extend its coverage
- Link to v2: https://lore.kernel.org/r/20240910-convert_xdp_tests-v2-1-a46367c9d038@bootlin.com

Changes in v2:
- fix endianness management in userspace packet parsing (call htonl on
  constant rather than packet part)

The new test has been run in a local x86 environment and in CI:
 #560/1   xdp_cpumap_attach/CPUMAP with programs in entries:OK
 #560/2   xdp_cpumap_attach/CPUMAP with frags programs in entries:OK
 #560/3   xdp_cpumap_attach/CPUMAP attach with programs in entries on veth:OK
 #560     xdp_cpumap_attach:OK
 Summary: 1/3 PASSED, 0 SKIPPED, 0 FAILED

---
Alexis Lothoré (eBPF Foundation) (3):
      selftests/bpf: fix bpf_map_redirect call for cpu map test
      selftests/bpf: make xdp_cpumap_attach keep redirect prog attached
      selftests/bpf: check program redirect in xdp_cpumap_attach

 .../selftests/bpf/prog_tests/xdp_cpumap_attach.c   | 130 +++++++++++++++++++--
 .../bpf/progs/test_xdp_with_cpumap_helpers.c       |   7 +-
 2 files changed, 129 insertions(+), 8 deletions(-)
---
base-commit: 058d7c3d1691e2e4a4963716ec6c047dff778637
change-id: 20240730-convert_xdp_tests-ccd66bfe33db

Best regards,
-- 
Alexis Lothoré, Bootlin
Embedded Linux and Kernel engineering
https://bootlin.com

[PATCH bpf-next v3 1/3] selftests/bpf: fix bpf_map_redirect call for cpu map test

[Alexis Lothoré (eBPF Foundation)]

xdp_redir_prog currently redirects packets based on the entry at index 1
in cpu_map, but the corresponding test only manipulates the entry at
index 0. This does not really affect the test in its current form since
the program is detached before having the opportunity to execute, but it
needs to be fixed before being able improve the corresponding test (ie,
not only test attach/detach but also the redirect feature)

Fix this XDP program by making it redirect packets based on entry 0 in
cpu_map instead of entry 1.

Signed-off-by: Alexis Lothoré (eBPF Foundation) <alexis.lothore@bootlin.com>
---
Changes in v3:
- new patch
---
 tools/testing/selftests/bpf/progs/test_xdp_with_cpumap_helpers.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/tools/testing/selftests/bpf/progs/test_xdp_with_cpumap_helpers.c b/tools/testing/selftests/bpf/progs/test_xdp_with_cpumap_helpers.c
index 20ec6723df18a6e8c036bf7754fbed83f2d2430b..d848fe96924e32a72e1e0327e3afffeb349b933e 100644
--- a/tools/testing/selftests/bpf/progs/test_xdp_with_cpumap_helpers.c
+++ b/tools/testing/selftests/bpf/progs/test_xdp_with_cpumap_helpers.c
@@ -15,7 +15,7 @@ struct {
 SEC("xdp")
 int xdp_redir_prog(struct xdp_md *ctx)
 {
-	return bpf_redirect_map(&cpu_map, 1, 0);
+	return bpf_redirect_map(&cpu_map, 0, 0);
 }
 
 SEC("xdp")

-- 
2.46.2

[PATCH bpf-next v3 2/3] selftests/bpf: make xdp_cpumap_attach keep redirect prog attached

[Alexis Lothoré (eBPF Foundation)]

Current test only checks attach/detach on cpu map type program, and so
does not check that it can be properly executed, neither that it
redirects correctly.

Update the existing test to extend its coverage:
- keep the redirected program loaded
- try to execute it through bpf_prog_test_run_opts with some dummy
  context

While at it, bring the following minor improvements:
- isolate test interface in its own namespace
- replicate the test on a veth pair

Signed-off-by: Alexis Lothoré (eBPF Foundation) <alexis.lothore@bootlin.com>
---
This change is based on the similar update brought to xdp_devmap_attach
([1]) and then realigns xdp_cpumap_attach with it

[1] https://lore.kernel.org/bpf/20240911-devel-koalo-fix-ingress-ifindex-v4-2-5c643ae10258@linutronix.de/

Changes in v3:
- new patch
---
 .../selftests/bpf/prog_tests/xdp_cpumap_attach.c   | 124 +++++++++++++++++++--
 1 file changed, 117 insertions(+), 7 deletions(-)

diff --git a/tools/testing/selftests/bpf/prog_tests/xdp_cpumap_attach.c b/tools/testing/selftests/bpf/prog_tests/xdp_cpumap_attach.c
index 481626a875d1c3db9c7bfe92c3cca6e967a6d45c..31c225f0239613f6b5adad36b5b0e6e85eeddd9a 100644
--- a/tools/testing/selftests/bpf/prog_tests/xdp_cpumap_attach.c
+++ b/tools/testing/selftests/bpf/prog_tests/xdp_cpumap_attach.c
@@ -2,35 +2,41 @@
 #include <uapi/linux/bpf.h>
 #include <linux/if_link.h>
 #include <test_progs.h>
+#include <network_helpers.h>
 
 #include "test_xdp_with_cpumap_frags_helpers.skel.h"
 #include "test_xdp_with_cpumap_helpers.skel.h"
 
 #define IFINDEX_LO	1
+#define TEST_NS "cpu_attach_ns"
 
 static void test_xdp_with_cpumap_helpers(void)
 {
-	struct test_xdp_with_cpumap_helpers *skel;
+	struct test_xdp_with_cpumap_helpers *skel = NULL;
 	struct bpf_prog_info info = {};
 	__u32 len = sizeof(info);
 	struct bpf_cpumap_val val = {
 		.qsize = 192,
 	};
-	int err, prog_fd, map_fd;
+	int err, prog_fd, prog_redir_fd, map_fd;
+	struct nstoken *nstoken = NULL;
 	__u32 idx = 0;
 
+	SYS(out_close, "ip netns add %s", TEST_NS);
+	nstoken = open_netns(TEST_NS);
+	if (!ASSERT_OK_PTR(nstoken, "open_netns"))
+		goto out_close;
+	SYS(out_close, "ip link set dev lo up");
+
 	skel = test_xdp_with_cpumap_helpers__open_and_load();
 	if (!ASSERT_OK_PTR(skel, "test_xdp_with_cpumap_helpers__open_and_load"))
 		return;
 
-	prog_fd = bpf_program__fd(skel->progs.xdp_redir_prog);
-	err = bpf_xdp_attach(IFINDEX_LO, prog_fd, XDP_FLAGS_SKB_MODE, NULL);
+	prog_redir_fd = bpf_program__fd(skel->progs.xdp_redir_prog);
+	err = bpf_xdp_attach(IFINDEX_LO, prog_redir_fd, XDP_FLAGS_SKB_MODE, NULL);
 	if (!ASSERT_OK(err, "Generic attach of program with 8-byte CPUMAP"))
 		goto out_close;
 
-	err = bpf_xdp_detach(IFINDEX_LO, XDP_FLAGS_SKB_MODE, NULL);
-	ASSERT_OK(err, "XDP program detach");
-
 	prog_fd = bpf_program__fd(skel->progs.xdp_dummy_cm);
 	map_fd = bpf_map__fd(skel->maps.cpu_map);
 	err = bpf_prog_get_info_by_fd(prog_fd, &info, &len);
@@ -45,6 +51,23 @@ static void test_xdp_with_cpumap_helpers(void)
 	ASSERT_OK(err, "Read cpumap entry");
 	ASSERT_EQ(info.id, val.bpf_prog.id, "Match program id to cpumap entry prog_id");
 
+	/* send a packet to trigger any potential bugs in there */
+	char data[10] = {};
+	DECLARE_LIBBPF_OPTS(bpf_test_run_opts, opts,
+			    .data_in = &data,
+			    .data_size_in = 10,
+			    .flags = BPF_F_TEST_XDP_LIVE_FRAMES,
+			    .repeat = 1,
+		);
+	err = bpf_prog_test_run_opts(prog_redir_fd, &opts);
+	ASSERT_OK(err, "XDP test run");
+
+	/* wait for the packets to be flushed */
+	kern_sync_rcu();
+
+	err = bpf_xdp_detach(IFINDEX_LO, XDP_FLAGS_SKB_MODE, NULL);
+	ASSERT_OK(err, "XDP program detach");
+
 	/* can not attach BPF_XDP_CPUMAP program to a device */
 	err = bpf_xdp_attach(IFINDEX_LO, prog_fd, XDP_FLAGS_SKB_MODE, NULL);
 	if (!ASSERT_NEQ(err, 0, "Attach of BPF_XDP_CPUMAP program"))
@@ -65,6 +88,8 @@ static void test_xdp_with_cpumap_helpers(void)
 	ASSERT_NEQ(err, 0, "Add BPF_XDP program with frags to cpumap entry");
 
 out_close:
+	close_netns(nstoken);
+	SYS_NOFAIL("ip netns del %s", TEST_NS);
 	test_xdp_with_cpumap_helpers__destroy(skel);
 }
 
@@ -111,6 +136,88 @@ static void test_xdp_with_cpumap_frags_helpers(void)
 	test_xdp_with_cpumap_frags_helpers__destroy(skel);
 }
 
+static void test_xdp_with_cpumap_helpers_veth(void)
+{
+	int err, cm_fd, cm_fd_redir, map_fd, ifindex_dst, ifindex_src;
+	struct test_xdp_with_cpumap_helpers *skel = NULL;
+	struct bpf_prog_info info = {};
+	struct bpf_cpumap_val val = {
+		.qsize = 192
+	};
+	struct nstoken *nstoken = NULL;
+	__u32 len = sizeof(info);
+	__u32 idx = 0;
+
+	SYS(out_close, "ip netns add %s", TEST_NS);
+	nstoken = open_netns(TEST_NS);
+	if (!ASSERT_OK_PTR(nstoken, "open_netns"))
+		goto out_close;
+
+	SYS(out_close, "ip link add veth_src type veth peer name veth_dst");
+	SYS(out_close, "ip link set dev veth_src up");
+	SYS(out_close, "ip link set dev veth_dst up");
+
+	ifindex_src = if_nametoindex("veth_src");
+	ifindex_dst = if_nametoindex("veth_dst");
+	if (!ASSERT_NEQ(ifindex_src, 0, "val.ifindex") ||
+	    !ASSERT_NEQ(ifindex_dst, 0, "ifindex_dst"))
+		goto out_close;
+
+	skel = test_xdp_with_cpumap_helpers__open_and_load();
+	if (!ASSERT_OK_PTR(skel, "test_xdp_with_cpumap_helpers__open_and_load"))
+		goto out_close;
+
+	cm_fd_redir = bpf_program__fd(skel->progs.xdp_redir_prog);
+	err = bpf_xdp_attach(ifindex_src, cm_fd_redir, XDP_FLAGS_DRV_MODE, NULL);
+	if (!ASSERT_OK(err, "Attach of program with 8-byte cpumap"))
+		goto out_close;
+
+	cm_fd = bpf_program__fd(skel->progs.xdp_dummy_cm);
+	map_fd = bpf_map__fd(skel->maps.cpu_map);
+	err = bpf_prog_get_info_by_fd(cm_fd, &info, &len);
+	if (!ASSERT_OK(err, "bpf_prog_get_info_by_fd"))
+		goto out_close;
+
+	val.bpf_prog.fd = cm_fd;
+	err = bpf_map_update_elem(map_fd, &idx, &val, 0);
+	ASSERT_OK(err, "Add program to cpumap entry");
+
+	err = bpf_map_lookup_elem(map_fd, &idx, &val);
+	ASSERT_OK(err, "Read cpumap entry");
+	ASSERT_EQ(info.id, val.bpf_prog.id, "Match program id to cpumap entry prog_id");
+
+	/* attach dummy to other side to enable reception */
+	cm_fd = bpf_program__fd(skel->progs.xdp_dummy_prog);
+	err = bpf_xdp_attach(ifindex_dst, cm_fd, XDP_FLAGS_DRV_MODE, NULL);
+	if (!ASSERT_OK(err, "Attach of dummy XDP"))
+		goto out_close;
+
+	/* send a packet to trigger any potential bugs in there */
+	char data[10] = {};
+	DECLARE_LIBBPF_OPTS(bpf_test_run_opts, opts,
+			    .data_in = &data,
+			    .data_size_in = 10,
+			    .flags = BPF_F_TEST_XDP_LIVE_FRAMES,
+			    .repeat = 1,
+		);
+	err = bpf_prog_test_run_opts(cm_fd_redir, &opts);
+	ASSERT_OK(err, "XDP test run");
+
+	/* wait for the packets to be flushed */
+	kern_sync_rcu();
+
+	err = bpf_xdp_detach(ifindex_src, XDP_FLAGS_DRV_MODE, NULL);
+	ASSERT_OK(err, "XDP program detach");
+
+	err = bpf_xdp_detach(ifindex_dst, XDP_FLAGS_DRV_MODE, NULL);
+	ASSERT_OK(err, "XDP program detach");
+
+out_close:
+	close_netns(nstoken);
+	SYS_NOFAIL("ip netns del %s", TEST_NS);
+	test_xdp_with_cpumap_helpers__destroy(skel);
+}
+
 void serial_test_xdp_cpumap_attach(void)
 {
 	if (test__start_subtest("CPUMAP with programs in entries"))
@@ -118,4 +225,7 @@ void serial_test_xdp_cpumap_attach(void)
 
 	if (test__start_subtest("CPUMAP with frags programs in entries"))
 		test_xdp_with_cpumap_frags_helpers();
+
+	if (test__start_subtest("CPUMAP attach with programs in entries on veth"))
+		test_xdp_with_cpumap_helpers_veth();
 }

-- 
2.46.2

Re: [PATCH bpf-next v3 2/3] selftests/bpf: make xdp_cpumap_attach keep redirect prog attached

[Martin KaFai Lau]

On 10/9/24 3:12 AM, Alexis Lothoré (eBPF Foundation) wrote:
> Current test only checks attach/detach on cpu map type program, and so
> does not check that it can be properly executed, neither that it
> redirects correctly.
> 
> Update the existing test to extend its coverage:
> - keep the redirected program loaded
> - try to execute it through bpf_prog_test_run_opts with some dummy
>    context
> 
> While at it, bring the following minor improvements:
> - isolate test interface in its own namespace
> - replicate the test on a veth pair
> 
> Signed-off-by: Alexis Lothoré (eBPF Foundation) <alexis.lothore@bootlin.com>
> ---
> This change is based on the similar update brought to xdp_devmap_attach
> ([1]) and then realigns xdp_cpumap_attach with it
> 
> [1] https://lore.kernel.org/bpf/20240911-devel-koalo-fix-ingress-ifindex-v4-2-5c643ae10258@linutronix.de/
> 
> Changes in v3:
> - new patch
> ---
>   .../selftests/bpf/prog_tests/xdp_cpumap_attach.c   | 124 +++++++++++++++++++--
>   1 file changed, 117 insertions(+), 7 deletions(-)
> 
> diff --git a/tools/testing/selftests/bpf/prog_tests/xdp_cpumap_attach.c b/tools/testing/selftests/bpf/prog_tests/xdp_cpumap_attach.c
> index 481626a875d1c3db9c7bfe92c3cca6e967a6d45c..31c225f0239613f6b5adad36b5b0e6e85eeddd9a 100644
> --- a/tools/testing/selftests/bpf/prog_tests/xdp_cpumap_attach.c
> +++ b/tools/testing/selftests/bpf/prog_tests/xdp_cpumap_attach.c
> @@ -2,35 +2,41 @@
>   #include <uapi/linux/bpf.h>
>   #include <linux/if_link.h>
>   #include <test_progs.h>
> +#include <network_helpers.h>
>   
>   #include "test_xdp_with_cpumap_frags_helpers.skel.h"
>   #include "test_xdp_with_cpumap_helpers.skel.h"
>   
>   #define IFINDEX_LO	1
> +#define TEST_NS "cpu_attach_ns"
>   
>   static void test_xdp_with_cpumap_helpers(void)
>   {
> -	struct test_xdp_with_cpumap_helpers *skel;
> +	struct test_xdp_with_cpumap_helpers *skel = NULL;
>   	struct bpf_prog_info info = {};
>   	__u32 len = sizeof(info);
>   	struct bpf_cpumap_val val = {
>   		.qsize = 192,
>   	};
> -	int err, prog_fd, map_fd;
> +	int err, prog_fd, prog_redir_fd, map_fd;
> +	struct nstoken *nstoken = NULL;
>   	__u32 idx = 0;
>   
> +	SYS(out_close, "ip netns add %s", TEST_NS);
> +	nstoken = open_netns(TEST_NS);
> +	if (!ASSERT_OK_PTR(nstoken, "open_netns"))
> +		goto out_close;
> +	SYS(out_close, "ip link set dev lo up");

There is easier helper netns_{new,free} to do all these:
	netns_obj = netns_new(TEST_NS, true);

Only fyi for your future work.

> +
>   	skel = test_xdp_with_cpumap_helpers__open_and_load();
>   	if (!ASSERT_OK_PTR(skel, "test_xdp_with_cpumap_helpers__open_and_load"))
>   		return;
>   
> -	prog_fd = bpf_program__fd(skel->progs.xdp_redir_prog);
> -	err = bpf_xdp_attach(IFINDEX_LO, prog_fd, XDP_FLAGS_SKB_MODE, NULL);
> +	prog_redir_fd = bpf_program__fd(skel->progs.xdp_redir_prog);
> +	err = bpf_xdp_attach(IFINDEX_LO, prog_redir_fd, XDP_FLAGS_SKB_MODE, NULL);
>   	if (!ASSERT_OK(err, "Generic attach of program with 8-byte CPUMAP"))
>   		goto out_close;
>   
> -	err = bpf_xdp_detach(IFINDEX_LO, XDP_FLAGS_SKB_MODE, NULL);
> -	ASSERT_OK(err, "XDP program detach");
> -
>   	prog_fd = bpf_program__fd(skel->progs.xdp_dummy_cm);
>   	map_fd = bpf_map__fd(skel->maps.cpu_map);
>   	err = bpf_prog_get_info_by_fd(prog_fd, &info, &len);
> @@ -45,6 +51,23 @@ static void test_xdp_with_cpumap_helpers(void)
>   	ASSERT_OK(err, "Read cpumap entry");
>   	ASSERT_EQ(info.id, val.bpf_prog.id, "Match program id to cpumap entry prog_id");
>   
> +	/* send a packet to trigger any potential bugs in there */
> +	char data[10] = {};
> +	DECLARE_LIBBPF_OPTS(bpf_test_run_opts, opts,
> +			    .data_in = &data,
> +			    .data_size_in = 10,
> +			    .flags = BPF_F_TEST_XDP_LIVE_FRAMES,
> +			    .repeat = 1,
> +		);
> +	err = bpf_prog_test_run_opts(prog_redir_fd, &opts);
> +	ASSERT_OK(err, "XDP test run");
> +
> +	/* wait for the packets to be flushed */
> +	kern_sync_rcu();
> +
> +	err = bpf_xdp_detach(IFINDEX_LO, XDP_FLAGS_SKB_MODE, NULL);
> +	ASSERT_OK(err, "XDP program detach");
> +
>   	/* can not attach BPF_XDP_CPUMAP program to a device */
>   	err = bpf_xdp_attach(IFINDEX_LO, prog_fd, XDP_FLAGS_SKB_MODE, NULL);
>   	if (!ASSERT_NEQ(err, 0, "Attach of BPF_XDP_CPUMAP program"))
> @@ -65,6 +88,8 @@ static void test_xdp_with_cpumap_helpers(void)
>   	ASSERT_NEQ(err, 0, "Add BPF_XDP program with frags to cpumap entry");
>   
>   out_close:
> +	close_netns(nstoken);
> +	SYS_NOFAIL("ip netns del %s", TEST_NS);
>   	test_xdp_with_cpumap_helpers__destroy(skel);
>   }
>   
> @@ -111,6 +136,88 @@ static void test_xdp_with_cpumap_frags_helpers(void)
>   	test_xdp_with_cpumap_frags_helpers__destroy(skel);
>   }
>   
> +static void test_xdp_with_cpumap_helpers_veth(void)

Unlike the devmap redirect, cpumap redirects pkt to another cpu instead of 
another veth dev. veth test is not very useful here for cpumap, so I take the 
veth test out.

> +{
> +	int err, cm_fd, cm_fd_redir, map_fd, ifindex_dst, ifindex_src;
> +	struct test_xdp_with_cpumap_helpers *skel = NULL;
> +	struct bpf_prog_info info = {};
> +	struct bpf_cpumap_val val = {
> +		.qsize = 192
> +	};
> +	struct nstoken *nstoken = NULL;
> +	__u32 len = sizeof(info);
> +	__u32 idx = 0;
> +
> +	SYS(out_close, "ip netns add %s", TEST_NS);
> +	nstoken = open_netns(TEST_NS);
> +	if (!ASSERT_OK_PTR(nstoken, "open_netns"))
> +		goto out_close;
> +
> +	SYS(out_close, "ip link add veth_src type veth peer name veth_dst");
> +	SYS(out_close, "ip link set dev veth_src up");
> +	SYS(out_close, "ip link set dev veth_dst up");
> +
> +	ifindex_src = if_nametoindex("veth_src");
> +	ifindex_dst = if_nametoindex("veth_dst");
> +	if (!ASSERT_NEQ(ifindex_src, 0, "val.ifindex") ||
> +	    !ASSERT_NEQ(ifindex_dst, 0, "ifindex_dst"))
> +		goto out_close;
> +
> +	skel = test_xdp_with_cpumap_helpers__open_and_load();
> +	if (!ASSERT_OK_PTR(skel, "test_xdp_with_cpumap_helpers__open_and_load"))
> +		goto out_close;
> +
> +	cm_fd_redir = bpf_program__fd(skel->progs.xdp_redir_prog);
> +	err = bpf_xdp_attach(ifindex_src, cm_fd_redir, XDP_FLAGS_DRV_MODE, NULL);
> +	if (!ASSERT_OK(err, "Attach of program with 8-byte cpumap"))
> +		goto out_close;
> +
> +	cm_fd = bpf_program__fd(skel->progs.xdp_dummy_cm);
> +	map_fd = bpf_map__fd(skel->maps.cpu_map);
> +	err = bpf_prog_get_info_by_fd(cm_fd, &info, &len);
> +	if (!ASSERT_OK(err, "bpf_prog_get_info_by_fd"))
> +		goto out_close;
> +
> +	val.bpf_prog.fd = cm_fd;
> +	err = bpf_map_update_elem(map_fd, &idx, &val, 0);
> +	ASSERT_OK(err, "Add program to cpumap entry");
> +
> +	err = bpf_map_lookup_elem(map_fd, &idx, &val);
> +	ASSERT_OK(err, "Read cpumap entry");
> +	ASSERT_EQ(info.id, val.bpf_prog.id, "Match program id to cpumap entry prog_id");
> +
> +	/* attach dummy to other side to enable reception */
> +	cm_fd = bpf_program__fd(skel->progs.xdp_dummy_prog);
> +	err = bpf_xdp_attach(ifindex_dst, cm_fd, XDP_FLAGS_DRV_MODE, NULL);
> +	if (!ASSERT_OK(err, "Attach of dummy XDP"))
> +		goto out_close;
> +
> +	/* send a packet to trigger any potential bugs in there */
> +	char data[10] = {};
> +	DECLARE_LIBBPF_OPTS(bpf_test_run_opts, opts,
> +			    .data_in = &data,
> +			    .data_size_in = 10,
> +			    .flags = BPF_F_TEST_XDP_LIVE_FRAMES,
> +			    .repeat = 1,
> +		);
> +	err = bpf_prog_test_run_opts(cm_fd_redir, &opts);
> +	ASSERT_OK(err, "XDP test run");
> +
> +	/* wait for the packets to be flushed */
> +	kern_sync_rcu();
> +
> +	err = bpf_xdp_detach(ifindex_src, XDP_FLAGS_DRV_MODE, NULL);
> +	ASSERT_OK(err, "XDP program detach");
> +
> +	err = bpf_xdp_detach(ifindex_dst, XDP_FLAGS_DRV_MODE, NULL);
> +	ASSERT_OK(err, "XDP program detach");
> +
> +out_close:
> +	close_netns(nstoken);
> +	SYS_NOFAIL("ip netns del %s", TEST_NS);
> +	test_xdp_with_cpumap_helpers__destroy(skel);
> +}
> +
>   void serial_test_xdp_cpumap_attach(void)

Thanks for improving the existing test by running in its netns, I also removed 
the "serial_". Applied. Thanks.

>   {
>   	if (test__start_subtest("CPUMAP with programs in entries"))
> @@ -118,4 +225,7 @@ void serial_test_xdp_cpumap_attach(void)
>   
>   	if (test__start_subtest("CPUMAP with frags programs in entries"))
>   		test_xdp_with_cpumap_frags_helpers();
> +
> +	if (test__start_subtest("CPUMAP attach with programs in entries on veth"))
> +		test_xdp_with_cpumap_helpers_veth();
>   }
> 

Re: [PATCH bpf-next v3 2/3] selftests/bpf: make xdp_cpumap_attach keep redirect prog attached

[Alexis Lothoré]

Hello Martin,

On 10/11/24 03:23, Martin KaFai Lau wrote:
> On 10/9/24 3:12 AM, Alexis Lothoré (eBPF Foundation) wrote:
>> -    int err, prog_fd, map_fd;
>> +    int err, prog_fd, prog_redir_fd, map_fd;
>> +    struct nstoken *nstoken = NULL;
>>       __u32 idx = 0;
>>   +    SYS(out_close, "ip netns add %s", TEST_NS);
>> +    nstoken = open_netns(TEST_NS);
>> +    if (!ASSERT_OK_PTR(nstoken, "open_netns"))
>> +        goto out_close;
>> +    SYS(out_close, "ip link set dev lo up");
> 
> There is easier helper netns_{new,free} to do all these:
>     netns_obj = netns_new(TEST_NS, true);
> 
> Only fyi for your future work.

ACK, thanks !


-- 
Alexis Lothoré, Bootlin
Embedded Linux and Kernel engineering
https://bootlin.com

[PATCH bpf-next v3 3/3] selftests/bpf: check program redirect in xdp_cpumap_attach

[Alexis Lothoré (eBPF Foundation)]

xdp_cpumap_attach, in its current form, only checks that an xdp cpumap
program can be executed, but not that it performs correctly the cpu
redirect as configured by userspace (bpf_prog_test_run_opts will return
success even if the redirect program returns an error)

Add a check to ensure that the program performs the configured redirect
as well. The check is based on a global variable incremented by a
chained program executed only if the redirect program properly executes.

Signed-off-by: Alexis Lothoré (eBPF Foundation) <alexis.lothore@bootlin.com>
---
Changes in v3:
- new patch
---
 tools/testing/selftests/bpf/prog_tests/xdp_cpumap_attach.c     | 10 ++++++++--
 .../testing/selftests/bpf/progs/test_xdp_with_cpumap_helpers.c |  5 +++++
 2 files changed, 13 insertions(+), 2 deletions(-)

diff --git a/tools/testing/selftests/bpf/prog_tests/xdp_cpumap_attach.c b/tools/testing/selftests/bpf/prog_tests/xdp_cpumap_attach.c
index 31c225f0239613f6b5adad36b5b0e6e85eeddd9a..57d1661dc72aeb152c7cb9c2f63e3b47bf1799d8 100644
--- a/tools/testing/selftests/bpf/prog_tests/xdp_cpumap_attach.c
+++ b/tools/testing/selftests/bpf/prog_tests/xdp_cpumap_attach.c
@@ -62,8 +62,11 @@ static void test_xdp_with_cpumap_helpers(void)
 	err = bpf_prog_test_run_opts(prog_redir_fd, &opts);
 	ASSERT_OK(err, "XDP test run");
 
-	/* wait for the packets to be flushed */
+	/* wait for the packets to be flushed, then check that redirect has been
+	 * performed
+	 */
 	kern_sync_rcu();
+	ASSERT_NEQ(skel->bss->redirect_count, 0, "redirected packets");
 
 	err = bpf_xdp_detach(IFINDEX_LO, XDP_FLAGS_SKB_MODE, NULL);
 	ASSERT_OK(err, "XDP program detach");
@@ -203,8 +206,11 @@ static void test_xdp_with_cpumap_helpers_veth(void)
 	err = bpf_prog_test_run_opts(cm_fd_redir, &opts);
 	ASSERT_OK(err, "XDP test run");
 
-	/* wait for the packets to be flushed */
+	/* wait for the packets to be flushed, then check that redirect has been
+	 * performed
+	 */
 	kern_sync_rcu();
+	ASSERT_NEQ(skel->bss->redirect_count, 0, "redirected packets");
 
 	err = bpf_xdp_detach(ifindex_src, XDP_FLAGS_DRV_MODE, NULL);
 	ASSERT_OK(err, "XDP program detach");
diff --git a/tools/testing/selftests/bpf/progs/test_xdp_with_cpumap_helpers.c b/tools/testing/selftests/bpf/progs/test_xdp_with_cpumap_helpers.c
index d848fe96924e32a72e1e0327e3afffeb349b933e..3619239b01b741dfd81bbebf5d9a62e0cf71e4f4 100644
--- a/tools/testing/selftests/bpf/progs/test_xdp_with_cpumap_helpers.c
+++ b/tools/testing/selftests/bpf/progs/test_xdp_with_cpumap_helpers.c
@@ -12,6 +12,8 @@ struct {
 	__uint(max_entries, 4);
 } cpu_map SEC(".maps");
 
+__u32 redirect_count = 0;
+
 SEC("xdp")
 int xdp_redir_prog(struct xdp_md *ctx)
 {
@@ -27,6 +29,9 @@ int xdp_dummy_prog(struct xdp_md *ctx)
 SEC("xdp/cpumap")
 int xdp_dummy_cm(struct xdp_md *ctx)
 {
+	if (bpf_get_smp_processor_id() == 0)
+		redirect_count++;
+
 	if (ctx->ingress_ifindex == IFINDEX_LO)
 		return XDP_DROP;
 

-- 
2.46.2

Re: [PATCH bpf-next v3 0/3] selftests/bpf: add coverage for xdp_features in test_progs

[patchwork-bot+netdevbpf]

Hello:

This series was applied to bpf/bpf-next.git (net)
by Martin KaFai Lau <martin.lau@kernel.org>:

On Wed, 09 Oct 2024 12:12:06 +0200 you wrote:
> Hello,
> this small series aims to increase coverage of xdp features in
> test_progs. The initial versions proposed to rework test_xdp_features.sh
> to make it fit in test_progs, but some discussions in v1 and v2 showed
> that the script is still needed as a standalone tool. So this new
> revision lets test_xdp_features.sh as-is, and rather adds missing
> coverage in existing test (cpu map). The new revision is now also a
> follow-up to the update performed by Florian Kauer in [1] for devmap
> programs testing.
> 
> [...]

Here is the summary with links:
  - [bpf-next,v3,1/3] selftests/bpf: fix bpf_map_redirect call for cpu map test
    https://git.kernel.org/bpf/bpf-next/c/ac8d16b2d377
  - [bpf-next,v3,2/3] selftests/bpf: make xdp_cpumap_attach keep redirect prog attached
    https://git.kernel.org/bpf/bpf-next/c/d5fbcf46ee82
  - [bpf-next,v3,3/3] selftests/bpf: check program redirect in xdp_cpumap_attach
    https://git.kernel.org/bpf/bpf-next/c/d124d984c8a2

You are awesome, thank you!
-- 
Deet-doot-dot, I am a bot.
https://korg.docs.kernel.org/patchwork/pwbot.html