Skip to contentSkip to conversation

bpf: bpftool: fix print error when show bpf map

PATCH Applied Yafang Shao 4 participants 8 messages Mar 20, 2022 – Mar 22, 2022

Subsystems (7)
lore
Share & tools
lore
#0 lore raw

PATCH 1/1 bpf: bpftool: fix print error when show bpf map

If there is no btf_id or frozen, it will not show the pids, but the pids doesn't depends on any one of them.

Below is the result after this change,

$ ./bpftool map show
2: lpm_trie  flags 0x1
	key 8B  value 8B  max_entries 1  memlock 4096B
	pids systemd(1)
3: lpm_trie  flags 0x1
	key 20B  value 8B  max_entries 1  memlock 4096B
	pids systemd(1)

While before this change, the 'pids systemd(1)' can't be displayed.

Fixes 9330986c0300 ("bpf: Add bloom filter map implementation")
Cc Joanne Koong <[email protected]>
---
 tools/bpf/bpftool/map.c | 7 ++-----
 1 file changed, 2 insertions(+), 5 deletions(-)
tools/bpf/bpftool/map.c +2 -5
index f91d9bf9054e..c26378f20831 100644
@@ -620,17 +620,14 @@ static int show_map_close_plain(int fd, struct bpf_map_info *info)
 					    u32_as_hash_field(info->id))
 			printf("\n\tpinned %s", (char *)entry->value);
 	}
-	printf("\n");
 
 	if (frozen_str) {
 		frozen = atoi(frozen_str);
 		free(frozen_str);
 	}
 
-	if (!info->btf_id && !frozen)
-		return 0;
-
-	printf("\t");
+	if (info->btf_id || frozen)
+		printf("\n\t");
 
 	if (info->btf_id)
 		printf("btf_id %d", info->btf_id);
Signature · 1 line
2.17.1
#1 lore raw

PATCH 1/1 bpf: selftests: cleanup RLIMIT_MEMLOCK

Since we have alread switched to memcg-based memory accouting and control, we don't need RLIMIT_MEMLOCK any more.

Cc Roman Gushchin <[email protected]>

---
RLIMIT_MEMLOCK is still used in bpftool and libbpf, but it may be useful for backward compatibility, so I don't cleanup them.

---
 tools/testing/selftests/bpf/bpf_rlimit.h      | 28 -------------------
 .../selftests/bpf/flow_dissector_load.c       |  1 -
 .../selftests/bpf/get_cgroup_id_user.c        |  1 -
 .../selftests/bpf/map_tests/sk_storage_map.c  | 15 ----------
 .../selftests/bpf/test_cgroup_storage.c       |  1 -
 tools/testing/selftests/bpf/test_dev_cgroup.c |  1 -
 tools/testing/selftests/bpf/test_lpm_map.c    |  1 -
 tools/testing/selftests/bpf/test_lru_map.c    |  1 -
 .../selftests/bpf/test_skb_cgroup_id_user.c   |  1 -
 tools/testing/selftests/bpf/test_sock.c       |  1 -
 tools/testing/selftests/bpf/test_sock_addr.c  |  1 -
 tools/testing/selftests/bpf/test_sockmap.c    |  1 -
 tools/testing/selftests/bpf/test_sysctl.c     |  1 -
 tools/testing/selftests/bpf/test_tag.c        |  1 -
 .../bpf/test_tcp_check_syncookie_user.c       |  1 -
 .../selftests/bpf/test_tcpnotify_user.c       |  1 -
 .../testing/selftests/bpf/test_verifier_log.c |  2 --
 tools/testing/selftests/bpf/xdping.c          |  7 -----
 tools/testing/selftests/bpf/xdpxceiver.c      |  5 ----
 19 files changed, 71 deletions(-)
 delete mode 100644 tools/testing/selftests/bpf/bpf_rlimit.h
tools/testing/selftests/bpf/bpf_rlimit.h deleted +0 -28
deleted file mode 100644
index 9dac9b30f8ef..000000000000
@@ -1,28 +0,0 @@
-#include <sys/resource.h>
-#include <stdio.h>
-
-static  __attribute__((constructor)) void bpf_rlimit_ctor(void)
-{
-	struct rlimit rlim_old, rlim_new = {
-		.rlim_cur	= RLIM_INFINITY,
-		.rlim_max	= RLIM_INFINITY,
-	};
-
-	getrlimit(RLIMIT_MEMLOCK, &rlim_old);
-	/* For the sake of running the test cases, we temporarily
-	 * set rlimit to infinity in order for kernel to focus on
-	 * errors from actual test cases and not getting noise
-	 * from hitting memlock limits. The limit is on per-process
-	 * basis and not a global one, hence destructor not really
-	 * needed here.
-	 */
-	if (setrlimit(RLIMIT_MEMLOCK, &rlim_new) < 0) {
-		perror("Unable to lift memlock rlimit");
-		/* Trying out lower limit, but expect potential test
-		 * case failures from this!
-		 */
-		rlim_new.rlim_cur = rlim_old.rlim_cur + (1UL << 20);
-		rlim_new.rlim_max = rlim_old.rlim_max + (1UL << 20);
-		setrlimit(RLIMIT_MEMLOCK, &rlim_new);
-	}
-}
tools/testing/selftests/bpf/flow_dissector_load.c +0 -1
index 87fd1aa323a9..e9df470e3253 100644
@@ -11,7 +11,6 @@
 #include <bpf/bpf.h>
 #include <bpf/libbpf.h>
 
-#include "bpf_rlimit.h"
 #include "flow_dissector_load.h"
 
 const char *cfg_pin_path = "/sys/fs/bpf/flow_dissector";
tools/testing/selftests/bpf/get_cgroup_id_user.c +0 -1
index 3a7b82bd9e94..2c027c7ef49d 100644
@@ -20,7 +20,6 @@
 
 #include "cgroup_helpers.h"
 #include "testing_helpers.h"
-#include "bpf_rlimit.h"
 
 #define CHECK(condition, tag, format...) ({		\
 	int __ret = !!(condition);			\
tools/testing/selftests/bpf/map_tests/sk_storage_map.c +0 -15
index 099eb4dfd4f7..864422241960 100644
@@ -3,7 +3,6 @@
 #include <linux/compiler.h>
 #include <linux/err.h>
 
-#include <sys/resource.h>
 #include <sys/socket.h>
 #include <sys/types.h>
 #include <linux/btf.h>
@@ -395,11 +394,8 @@ static void stop_handler(int signum)
 
 static void test_sk_storage_map_stress_free(void)
 {
-	struct rlimit rlim_old, rlim_new = {};
 	int err;
 
-	getrlimit(RLIMIT_NOFILE, &rlim_old);
-
 	signal(SIGTERM, stop_handler);
 	signal(SIGINT, stop_handler);
 	if (runtime_s > 0) {
@@ -407,14 +403,6 @@ static void test_sk_storage_map_stress_free(void)
 		alarm(runtime_s);
 	}
 
-	if (rlim_old.rlim_cur < nr_sk_threads * nr_sk_per_thread) {
-		rlim_new.rlim_cur = nr_sk_threads * nr_sk_per_thread + 128;
-		rlim_new.rlim_max = rlim_new.rlim_cur + 128;
-		err = setrlimit(RLIMIT_NOFILE, &rlim_new);
-		CHECK(err, "setrlimit(RLIMIT_NOFILE)", "rlim_new:%lu errno:%d",
-		      rlim_new.rlim_cur, errno);
-	}
-
 	err = do_sk_storage_map_stress_free();
 
 	signal(SIGTERM, SIG_DFL);
@@ -424,9 +412,6 @@ static void test_sk_storage_map_stress_free(void)
 		alarm(0);
 	}
 
-	if (rlim_new.rlim_cur)
-		setrlimit(RLIMIT_NOFILE, &rlim_old);
-
 	CHECK(err, "test_sk_storage_map_stress_free", "err:%d\n", err);
 }
 
tools/testing/selftests/bpf/test_cgroup_storage.c +0 -1
index d6a1be4d8020..d8bea0439ec9 100644
@@ -6,7 +6,6 @@
 #include <stdlib.h>
 #include <sys/sysinfo.h>
 
-#include "bpf_rlimit.h"
 #include "cgroup_helpers.h"
 #include "testing_helpers.h"
 
tools/testing/selftests/bpf/test_dev_cgroup.c +0 -1
index c299d3452695..5ac8279f3f93 100644
@@ -15,7 +15,6 @@
 
 #include "cgroup_helpers.h"
 #include "testing_helpers.h"
-#include "bpf_rlimit.h"
 
 #define DEV_CGROUP_PROG "./dev_cgroup.o"
 
tools/testing/selftests/bpf/test_lpm_map.c +0 -1
index baa3e3ecae82..790ff32c735a 100644
@@ -26,7 +26,6 @@
 #include <bpf/bpf.h>
 
 #include "bpf_util.h"
-#include "bpf_rlimit.h"
 
 struct tlpm_node {
 	struct tlpm_node *next;
tools/testing/selftests/bpf/test_lru_map.c +0 -1
index 563bbe18c172..41cea43cd6d5 100644
@@ -18,7 +18,6 @@
 #include <bpf/libbpf.h>
 
 #include "bpf_util.h"
-#include "bpf_rlimit.h"
 #include "../../../include/linux/filter.h"
 
 #define LOCAL_FREE_TARGET	(128)
tools/testing/selftests/bpf/test_skb_cgroup_id_user.c +0 -1
index 4a64306728ab..de7b230023a2 100644
@@ -15,7 +15,6 @@
 #include <bpf/bpf.h>
 #include <bpf/libbpf.h>
 
-#include "bpf_rlimit.h"
 #include "cgroup_helpers.h"
 
 #define CGROUP_PATH		"/skb_cgroup_test"
tools/testing/selftests/bpf/test_sock.c +0 -1
index fe10f8134278..c83e1872d2a6 100644
@@ -14,7 +14,6 @@
 
 #include "cgroup_helpers.h"
 #include <bpf/bpf_endian.h>
-#include "bpf_rlimit.h"
 #include "bpf_util.h"
 
 #define CG_PATH		"/foo"
tools/testing/selftests/bpf/test_sock_addr.c +0 -1
index f3d5d7ac6505..fd047567ee63 100644
@@ -19,7 +19,6 @@
 #include <bpf/libbpf.h>
 
 #include "cgroup_helpers.h"
-#include "bpf_rlimit.h"
 #include "bpf_util.h"
 
 #ifndef ENOTSUPP
tools/testing/selftests/bpf/test_sockmap.c +0 -1
index dfb4f5c0fcb9..3eeebe915903 100644
@@ -37,7 +37,6 @@
 #include <bpf/libbpf.h>
 
 #include "bpf_util.h"
-#include "bpf_rlimit.h"
 #include "cgroup_helpers.h"
 
 int running;
tools/testing/selftests/bpf/test_sysctl.c +0 -1
index 4f6cf833b522..dd6fd170b11a 100644
@@ -14,7 +14,6 @@
 #include <bpf/libbpf.h>
 
 #include <bpf/bpf_endian.h>
-#include "bpf_rlimit.h"
 #include "bpf_util.h"
 #include "cgroup_helpers.h"
 #include "testing_helpers.h"
tools/testing/selftests/bpf/test_tag.c +0 -1
index 0851c42ee31c..4f745de802cf 100644
@@ -20,7 +20,6 @@
 #include <bpf/bpf.h>
 
 #include "../../../include/linux/filter.h"
-#include "bpf_rlimit.h"
 #include "testing_helpers.h"
 
 static struct bpf_insn prog[BPF_MAXINSNS];
tools/testing/selftests/bpf/test_tcp_check_syncookie_user.c +0 -1
index b9e991d43155..894eb0710d6f 100644
@@ -15,7 +15,6 @@
 #include <bpf/bpf.h>
 #include <bpf/libbpf.h>
 
-#include "bpf_rlimit.h"
 #include "cgroup_helpers.h"
 
 static int start_server(const struct sockaddr *addr, socklen_t len)
tools/testing/selftests/bpf/test_tcpnotify_user.c +0 -1
index 4c5114765b23..8284db8b0f13 100644
@@ -19,7 +19,6 @@
 #include <linux/perf_event.h>
 #include <linux/err.h>
 
-#include "bpf_rlimit.h"
 #include "bpf_util.h"
 #include "cgroup_helpers.h"
 
tools/testing/selftests/bpf/test_verifier_log.c +0 -2
index 8d6918c3b4a2..4bca0a7344cc 100644
@@ -11,8 +11,6 @@
 
 #include <bpf/bpf.h>
 
-#include "bpf_rlimit.h"
-
 #define LOG_SIZE (1 << 20)
 
 #define err(str...)	printf("ERROR: " str)
tools/testing/selftests/bpf/xdping.c +0 -7
index c567856fd1bc..bc5eadf2d0f8 100644
@@ -12,7 +12,6 @@
 #include <string.h>
 #include <unistd.h>
 #include <libgen.h>
-#include <sys/resource.h>
 #include <net/if.h>
 #include <sys/types.h>
 #include <sys/socket.h>
@@ -89,7 +88,6 @@ int main(int argc, char **argv)
 {
 	__u32 mode_flags = XDP_FLAGS_DRV_MODE | XDP_FLAGS_SKB_MODE;
 	struct addrinfo *a, hints = { .ai_family = AF_INET };
-	struct rlimit r = {RLIM_INFINITY, RLIM_INFINITY};
 	__u16 count = XDPING_DEFAULT_COUNT;
 	struct pinginfo pinginfo = { 0 };
 	const char *optstr = "c:I:NsS";
@@ -167,11 +165,6 @@ int main(int argc, char **argv)
 		freeaddrinfo(a);
 	}
 
-	if (setrlimit(RLIMIT_MEMLOCK, &r)) {
-		perror("setrlimit(RLIMIT_MEMLOCK)");
-		return 1;
-	}
-
 	snprintf(filename, sizeof(filename), "%s_kern.o", argv[0]);
 
 	if (bpf_prog_test_load(filename, BPF_PROG_TYPE_XDP, &obj, &prog_fd)) {
tools/testing/selftests/bpf/xdpxceiver.c +0 -5
index 5f8296d29e77..5fbaebe89e14 100644
@@ -90,7 +90,6 @@
 #include <string.h>
 #include <stddef.h>
 #include <sys/mman.h>
-#include <sys/resource.h>
 #include <sys/types.h>
 #include <sys/queue.h>
 #include <time.h>
@@ -1448,15 +1447,11 @@ static void ifobject_delete(struct ifobject *ifobj)
 
 int main(int argc, char **argv)
 {
-	struct rlimit _rlim = { RLIM_INFINITY, RLIM_INFINITY };
 	struct pkt_stream *pkt_stream_default;
 	struct ifobject *ifobj_tx, *ifobj_rx;
 	struct test_spec test;
 	u32 i, j;
 
-	if (setrlimit(RLIMIT_MEMLOCK, &_rlim))
-		exit_with_error(errno);
-
 	ifobj_tx = ifobject_create();
 	if (!ifobj_tx)
 		exit_with_error(ENOMEM);
Signature · 1 line
2.17.1
#2 lore raw

Re: [PATCH] bpf: selftests: cleanup RLIMIT_MEMLOCK

Quoted 6 lines · Yafang Shao

Since we have alread switched to memcg-based memory accouting and control, we don't need RLIMIT_MEMLOCK any more.

Quoted 6 lines
Cc Roman Gushchin <[email protected]>

---
RLIMIT_MEMLOCK is still used in bpftool and libbpf, but it may be useful for backward compatibility, so I don't cleanup them.

Hi Yafang!

As I remember, we haven’t cleaned selftests up with the same logic: it’s nice to be able to run the same version of tests on older kernels.

Thanks!

#3 lore raw

Re: [PATCH] bpf: selftests: cleanup RLIMIT_MEMLOCK

Quoted 10 lines · Roman Gushchin
Quoted 10 lines · Yafang Shao

Since we have alread switched to memcg-based memory accouting and control, we don't need RLIMIT_MEMLOCK any more.

Cc Roman Gushchin <[email protected]>
Quoted 8 lines
Quoted 3 lines

---
RLIMIT_MEMLOCK is still used in bpftool and libbpf, but it may be useful for backward compatibility, so I don't cleanup them.

Hi Yafang!

As I remember, we haven’t cleaned selftests up with the same logic: it’s nice to be able to run the same version of tests on older kernels.

It should be fine, at least for test_progs and test_progs-no_alu32. Libbpf now does this automatically if running in "libbpf 1.0" mode.

Yafang, please make sure that all the test binaries you are cleaning up have libbpf_set_strict_mode(LIBBPF_STRICT_ALL) (test_progs does already). You might need to clean up some SEC() definitions, in case we still missed some non-conforming ones, though.

Quoted 1 line

Thanks!

#4 lore raw

Re: [PATCH] bpf: selftests: cleanup RLIMIT_MEMLOCK

Quoted 18 lines · Andrii Nakryiko
Quoted 14 lines · Roman Gushchin
Quoted 13 lines
Quoted 3 lines · Yafang Shao

Since we have alread switched to memcg-based memory accouting and control, we don't need RLIMIT_MEMLOCK any more.

Cc Roman Gushchin <[email protected]>

---
RLIMIT_MEMLOCK is still used in bpftool and libbpf, but it may be useful for backward compatibility, so I don't cleanup them.

Quoted 7 lines
Quoted 4 lines

Hi Yafang!

As I remember, we haven’t cleaned selftests up with the same logic: it’s nice to be able to run the same version of tests on older kernels.

It should be fine, at least for test_progs and test_progs-no_alu32. Libbpf now does this automatically if running in "libbpf 1.0" mode.

Didn’t know this, thanks! Do we link all tests with it?

Quoted 5 lines

Yafang, please make sure that all the test binaries you are cleaning up have libbpf_set_strict_mode(LIBBPF_STRICT_ALL) (test_progs does already). You might need to clean up some SEC() definitions, in case we still missed some non-conforming ones, though.

If so, no objections to the patch from my side.

Thank you!

#5 lore raw

Re: [PATCH] bpf: selftests: cleanup RLIMIT_MEMLOCK

Quoted 21 lines · Roman Gushchin
Quoted 21 lines · Andrii Nakryiko
Quoted 16 lines · Roman Gushchin
Quoted 7 lines
Quoted 3 lines · Yafang Shao

Since we have alread switched to memcg-based memory accouting and control, we don't need RLIMIT_MEMLOCK any more.

Quoted 6 lines
Cc Roman Gushchin <[email protected]>

---
RLIMIT_MEMLOCK is still used in bpftool and libbpf, but it may be useful for backward compatibility, so I don't cleanup them.

Hi Yafang!

Quoted 7 lines
Quoted 5 lines
Quoted 2 lines

As I remember, we haven’t cleaned selftests up with the same logic: it’s nice to be able to run the same version of tests on older kernels.

It should be fine, at least for test_progs and test_progs-no_alu32. Libbpf now does this automatically if running in "libbpf 1.0" mode.

Didn’t know this, thanks! Do we link all tests with it?

Yep, every selftest inevitably relies on libbpf. We just need to make sure to enable that 1.0 mode with libbpf_set_strict_mode() call.

Quoted 10 lines
Quoted 6 lines

Yafang, please make sure that all the test binaries you are cleaning up have libbpf_set_strict_mode(LIBBPF_STRICT_ALL) (test_progs does already). You might need to clean up some SEC() definitions, in case we still missed some non-conforming ones, though.

If so, no objections to the patch from my side.

Thank you!

#6 lore raw

Re: [PATCH] bpf: selftests: cleanup RLIMIT_MEMLOCK

In reply to Andrii Nakryiko
Quoted 21 lines · Andrii Nakryiko
Quoted 10 lines · Roman Gushchin
Quoted 10 lines · Yafang Shao

Since we have alread switched to memcg-based memory accouting and control, we don't need RLIMIT_MEMLOCK any more.

Cc Roman Gushchin <[email protected]>
Quoted 8 lines
Quoted 3 lines

---
RLIMIT_MEMLOCK is still used in bpftool and libbpf, but it may be useful for backward compatibility, so I don't cleanup them.

Hi Yafang!

As I remember, we haven’t cleaned selftests up with the same logic: it’s nice to be able to run the same version of tests on older kernels.

Quoted 8 lines

It should be fine, at least for test_progs and test_progs-no_alu32. Libbpf now does this automatically if running in "libbpf 1.0" mode.

Yafang, please make sure that all the test binaries you are cleaning up have libbpf_set_strict_mode(LIBBPF_STRICT_ALL) (test_progs does already). You might need to clean up some SEC() definitions, in case we still missed some non-conforming ones, though.

Thanks for the suggestion. I will do it.

Signature · 2 lines
Thanks
Yafang
#7 lore raw
In reply to Yafang Shao
This patch was applied to bpf/bpf-next.git (master) by Daniel Borkmann <[email protected]>:

Hello:

This patch was applied to bpf/bpf-next.git (master) by Daniel Borkmann <[email protected]>:

Quoted 12 lines · On Sun, 20 Mar 2022 06:08:14 +0000 you wrote:

If there is no btf_id or frozen, it will not show the pids, but the pids doesn't depends on any one of them.

Below is the result after this change,

$ ./bpftool map show
2: lpm_trie  flags 0x1
	key 8B  value 8B  max_entries 1  memlock 4096B
	pids systemd(1)
3: lpm_trie  flags 0x1
	key 20B  value 8B  max_entries 1  memlock 4096B
	pids systemd(1)
Quoted 1 line

[...]

Here is the summary with links:
- bpf: bpftool: fix print error when show bpf map

    https://git.kernel.org/bpf/bpf-next/c/1824d8ea75f2

You are awesome, thank you!

Signature · 2 lines
Deet-doot-dot, I am a bot.
https://korg.docs.kernel.org/patchwork/pwbot.html