* Re: [PATCH] bpf: support BPF_PROG_QUERY for progs attached to sockmap
@ 2021-10-23 6:05 zhudi (E)
0 siblings, 0 replies; 7+ messages in thread
From: zhudi (E) @ 2021-10-23 6:05 UTC (permalink / raw)
To: Alexei Starovoitov
Cc: David S. Miller, Alexei Starovoitov, Daniel Borkmann,
Andrii Nakryiko, Martin KaFai Lau, Song Liu, Yonghong Song,
John Fastabend, KP Singh, Jakub Sitnicki, bpf, LKML
> On Fri, Oct 22, 2021 at 3:34 AM Di Zhu <zhudi2@huawei.com> wrote:
> >
> > Right now there is no way to query whether BPF programs are
> > attached to a sockmap or not.
> >
> > we can use the standard interface in libbpf to query, such as:
> > bpf_prog_query(mapFd, BPF_SK_SKB_STREAM_PARSER, 0, NULL, ...);
> > the mapFd is the fd of sockmap.
> >
> > Signed-off-by: Di Zhu <zhudi2@huawei.com>
>
> The feature looks fine, but it needs a selftest.
I will add it in my patch, thanks
^ permalink raw reply [flat|nested] 7+ messages in thread
* Re: [PATCH] bpf: support BPF_PROG_QUERY for progs attached to sockmap
@ 2021-10-23 6:01 zhudi (E)
0 siblings, 0 replies; 7+ messages in thread
From: zhudi (E) @ 2021-10-23 6:01 UTC (permalink / raw)
To: John Fastabend, davem, ast, daniel, andrii, kafai,
songliubraving, yhs, kpsingh, jakub
Cc: bpf, linux-kernel
> Di Zhu wrote:
> > Right now there is no way to query whether BPF programs are
> > attached to a sockmap or not.
> >
> > we can use the standard interface in libbpf to query, such as:
> > bpf_prog_query(mapFd, BPF_SK_SKB_STREAM_PARSER, 0, NULL, ...);
> > the mapFd is the fd of sockmap.
> >
> > Signed-off-by: Di Zhu <zhudi2@huawei.com>
> > ---
>
> LGTM, lets add a small test here as well
>
> ./tools/testing/selftests/bpf/prog_tests/sockmap_basic.c
>
> Looks like we can just copy the sk_lookup.c test case which does
> the query tests for BPF_SK_LOOKUP.
Thanks for your advice, I will add it in my patch.
> Also I don't think its required for this series, but a bpftool
> patch to query it would be useful as well if its doesn't just
> work with above.
>
> Thanks!
> John
^ permalink raw reply [flat|nested] 7+ messages in thread
* RE: [PATCH] bpf: support BPF_PROG_QUERY for progs attached to sockmap
2021-10-22 10:33 Di Zhu
2021-10-22 15:25 ` Alexei Starovoitov
@ 2021-10-22 15:46 ` John Fastabend
1 sibling, 0 replies; 7+ messages in thread
From: John Fastabend @ 2021-10-22 15:46 UTC (permalink / raw)
To: Di Zhu, davem, ast, daniel, andrii, kafai, songliubraving, yhs,
john.fastabend, kpsingh, jakub
Cc: bpf, linux-kernel, zhudi2
Di Zhu wrote:
> Right now there is no way to query whether BPF programs are
> attached to a sockmap or not.
>
> we can use the standard interface in libbpf to query, such as:
> bpf_prog_query(mapFd, BPF_SK_SKB_STREAM_PARSER, 0, NULL, ...);
> the mapFd is the fd of sockmap.
>
> Signed-off-by: Di Zhu <zhudi2@huawei.com>
> ---
LGTM, lets add a small test here as well
./tools/testing/selftests/bpf/prog_tests/sockmap_basic.c
Looks like we can just copy the sk_lookup.c test case which does
the query tests for BPF_SK_LOOKUP.
Also I don't think its required for this series, but a bpftool
patch to query it would be useful as well if its doesn't just
work with above.
Thanks!
John
^ permalink raw reply [flat|nested] 7+ messages in thread
* Re: [PATCH] bpf: support BPF_PROG_QUERY for progs attached to sockmap
2021-10-22 10:33 Di Zhu
@ 2021-10-22 15:25 ` Alexei Starovoitov
2021-10-22 15:46 ` John Fastabend
1 sibling, 0 replies; 7+ messages in thread
From: Alexei Starovoitov @ 2021-10-22 15:25 UTC (permalink / raw)
To: Di Zhu
Cc: David S. Miller, Alexei Starovoitov, Daniel Borkmann,
Andrii Nakryiko, Martin KaFai Lau, Song Liu, Yonghong Song,
John Fastabend, KP Singh, Jakub Sitnicki, bpf, LKML
On Fri, Oct 22, 2021 at 3:34 AM Di Zhu <zhudi2@huawei.com> wrote:
>
> Right now there is no way to query whether BPF programs are
> attached to a sockmap or not.
>
> we can use the standard interface in libbpf to query, such as:
> bpf_prog_query(mapFd, BPF_SK_SKB_STREAM_PARSER, 0, NULL, ...);
> the mapFd is the fd of sockmap.
>
> Signed-off-by: Di Zhu <zhudi2@huawei.com>
The feature looks fine, but it needs a selftest.
^ permalink raw reply [flat|nested] 7+ messages in thread
* [PATCH] bpf: support BPF_PROG_QUERY for progs attached to sockmap
@ 2021-10-22 10:33 Di Zhu
2021-10-22 15:25 ` Alexei Starovoitov
2021-10-22 15:46 ` John Fastabend
0 siblings, 2 replies; 7+ messages in thread
From: Di Zhu @ 2021-10-22 10:33 UTC (permalink / raw)
To: davem, ast, daniel, andrii, kafai, songliubraving, yhs,
john.fastabend, kpsingh, jakub
Cc: bpf, linux-kernel, zhudi2
Right now there is no way to query whether BPF programs are
attached to a sockmap or not.
we can use the standard interface in libbpf to query, such as:
bpf_prog_query(mapFd, BPF_SK_SKB_STREAM_PARSER, 0, NULL, ...);
the mapFd is the fd of sockmap.
Signed-off-by: Di Zhu <zhudi2@huawei.com>
---
include/linux/bpf.h | 9 +++++
kernel/bpf/syscall.c | 5 +++
net/core/sock_map.c | 82 ++++++++++++++++++++++++++++++++++++++++----
3 files changed, 89 insertions(+), 7 deletions(-)
diff --git a/include/linux/bpf.h b/include/linux/bpf.h
index d604c8251d88..db7d0e5115b7 100644
--- a/include/linux/bpf.h
+++ b/include/linux/bpf.h
@@ -1961,6 +1961,9 @@ int bpf_prog_test_run_syscall(struct bpf_prog *prog,
int sock_map_get_from_fd(const union bpf_attr *attr, struct bpf_prog *prog);
int sock_map_prog_detach(const union bpf_attr *attr, enum bpf_prog_type ptype);
int sock_map_update_elem_sys(struct bpf_map *map, void *key, void *value, u64 flags);
+int sockmap_bpf_prog_query(const union bpf_attr *attr,
+ union bpf_attr __user *uattr);
+
void sock_map_unhash(struct sock *sk);
void sock_map_close(struct sock *sk, long timeout);
#else
@@ -2014,6 +2017,12 @@ static inline int sock_map_update_elem_sys(struct bpf_map *map, void *key, void
{
return -EOPNOTSUPP;
}
+
+static inline int sockmap_bpf_prog_query(const union bpf_attr *attr,
+ union bpf_attr __user *uattr)
+{
+ return -EINVAL;
+}
#endif /* CONFIG_BPF_SYSCALL */
#endif /* CONFIG_NET && CONFIG_BPF_SYSCALL */
diff --git a/kernel/bpf/syscall.c b/kernel/bpf/syscall.c
index 4e50c0bfdb7d..17faeff8f85f 100644
--- a/kernel/bpf/syscall.c
+++ b/kernel/bpf/syscall.c
@@ -3275,6 +3275,11 @@ static int bpf_prog_query(const union bpf_attr *attr,
case BPF_FLOW_DISSECTOR:
case BPF_SK_LOOKUP:
return netns_bpf_prog_query(attr, uattr);
+ case BPF_SK_SKB_STREAM_PARSER:
+ case BPF_SK_SKB_STREAM_VERDICT:
+ case BPF_SK_MSG_VERDICT:
+ case BPF_SK_SKB_VERDICT:
+ return sockmap_bpf_prog_query(attr, uattr);
default:
return -EINVAL;
}
diff --git a/net/core/sock_map.c b/net/core/sock_map.c
index e252b8ec2b85..269349bd05a8 100644
--- a/net/core/sock_map.c
+++ b/net/core/sock_map.c
@@ -1412,38 +1412,50 @@ static struct sk_psock_progs *sock_map_progs(struct bpf_map *map)
return NULL;
}
-static int sock_map_prog_update(struct bpf_map *map, struct bpf_prog *prog,
- struct bpf_prog *old, u32 which)
+static int sock_map_prog_lookup(struct bpf_map *map, struct bpf_prog **pprog[],
+ u32 which)
{
struct sk_psock_progs *progs = sock_map_progs(map);
- struct bpf_prog **pprog;
if (!progs)
return -EOPNOTSUPP;
switch (which) {
case BPF_SK_MSG_VERDICT:
- pprog = &progs->msg_parser;
+ *pprog = &progs->msg_parser;
break;
#if IS_ENABLED(CONFIG_BPF_STREAM_PARSER)
case BPF_SK_SKB_STREAM_PARSER:
- pprog = &progs->stream_parser;
+ *pprog = &progs->stream_parser;
break;
#endif
case BPF_SK_SKB_STREAM_VERDICT:
if (progs->skb_verdict)
return -EBUSY;
- pprog = &progs->stream_verdict;
+ *pprog = &progs->stream_verdict;
break;
case BPF_SK_SKB_VERDICT:
if (progs->stream_verdict)
return -EBUSY;
- pprog = &progs->skb_verdict;
+ *pprog = &progs->skb_verdict;
break;
default:
return -EOPNOTSUPP;
}
+ return 0;
+}
+
+static int sock_map_prog_update(struct bpf_map *map, struct bpf_prog *prog,
+ struct bpf_prog *old, u32 which)
+{
+ struct bpf_prog **pprog;
+ int ret;
+
+ ret = sock_map_prog_lookup(map, &pprog, which);
+ if (ret)
+ return ret;
+
if (old)
return psock_replace_prog(pprog, prog, old);
@@ -1451,6 +1463,62 @@ static int sock_map_prog_update(struct bpf_map *map, struct bpf_prog *prog,
return 0;
}
+int sockmap_bpf_prog_query(const union bpf_attr *attr,
+ union bpf_attr __user *uattr)
+{
+ __u32 __user *prog_ids = u64_to_user_ptr(attr->query.prog_ids);
+ u32 prog_cnt = 0, flags = 0;
+ u32 ufd = attr->target_fd;
+ struct bpf_prog **pprog;
+ struct bpf_prog *prog;
+ struct bpf_map *map;
+ struct fd f;
+ int ret;
+
+ if (attr->query.query_flags)
+ return -EINVAL;
+
+ if (copy_to_user(&uattr->query.attach_flags, &flags, sizeof(flags)))
+ return -EFAULT;
+
+ f = fdget(ufd);
+ map = __bpf_map_get(f);
+ if (IS_ERR(map))
+ return PTR_ERR(map);
+
+ rcu_read_lock();
+
+ ret = sock_map_prog_lookup(map, &pprog, attr->query.attach_type);
+ if (ret)
+ goto end;
+
+ prog = *pprog;
+ prog_cnt = (!prog) ? 0 : 1;
+ if (copy_to_user(&uattr->query.prog_cnt, &prog_cnt, sizeof(prog_cnt))) {
+ ret = -EFAULT;
+ goto end;
+ }
+
+ if (!attr->query.prog_cnt || !prog_ids || !prog_cnt)
+ goto end;
+
+ prog = bpf_prog_inc_not_zero(prog);
+ if (IS_ERR(prog)) {
+ ret = PTR_ERR(prog);
+ goto end;
+ }
+
+ if (copy_to_user(prog_ids, &prog->aux->id, sizeof(u32)))
+ ret = -EFAULT;
+
+ bpf_prog_put(prog);
+
+end:
+ rcu_read_unlock();
+ fdput(f);
+ return ret;
+}
+
static void sock_map_unlink(struct sock *sk, struct sk_psock_link *link)
{
switch (link->map->map_type) {
--
2.27.0
^ permalink raw reply related [flat|nested] 7+ messages in thread
* Re: [PATCH] bpf: support BPF_PROG_QUERY for progs attached to sockmap
2021-10-19 12:58 Di Zhu
@ 2021-10-22 0:36 ` Alexei Starovoitov
0 siblings, 0 replies; 7+ messages in thread
From: Alexei Starovoitov @ 2021-10-22 0:36 UTC (permalink / raw)
To: Di Zhu
Cc: David S. Miller, Alexei Starovoitov, Daniel Borkmann,
Andrii Nakryiko, Martin KaFai Lau, bpf, LKML
On Tue, Oct 19, 2021 at 5:59 AM Di Zhu <zhudi2@huawei.com> wrote:
> + break;
> +#if IS_ENABLED(CONFIG_BPF_STREAM_PARSER)
> + case BPF_SK_SKB_STREAM_PARSER:
> + *prog = READ_ONCE(progs->skb_parser);
skb_parser?
Please don't submit patches that don't even build.
^ permalink raw reply [flat|nested] 7+ messages in thread
* [PATCH] bpf: support BPF_PROG_QUERY for progs attached to sockmap
@ 2021-10-19 12:58 Di Zhu
2021-10-22 0:36 ` Alexei Starovoitov
0 siblings, 1 reply; 7+ messages in thread
From: Di Zhu @ 2021-10-19 12:58 UTC (permalink / raw)
To: davem, ast, daniel, andrii, kafai; +Cc: bpf, linux-kernel, zhudi2
Right now there is no way to query whether BPF programs are
attached to a sockmap or not.
we can use the standard interface in libbpf to query, such as:
bpf_prog_query(mapFd, BPF_SK_SKB_STREAM_PARSER, 0, NULL, ...);
the mapFd is the fd of sockmap.
Signed-off-by: Di Zhu <zhudi2@huawei.com>
---
include/linux/bpf.h | 8 +++++
kernel/bpf/syscall.c | 4 +++
net/core/sock_map.c | 81 ++++++++++++++++++++++++++++++++++++++++++++
3 files changed, 93 insertions(+)
diff --git a/include/linux/bpf.h b/include/linux/bpf.h
index 1c7fd7c4c6d3..69cf70b077d5 100644
--- a/include/linux/bpf.h
+++ b/include/linux/bpf.h
@@ -1959,6 +1959,8 @@ int bpf_prog_test_run_syscall(struct bpf_prog *prog,
int sock_map_get_from_fd(const union bpf_attr *attr, struct bpf_prog *prog);
int sock_map_prog_detach(const union bpf_attr *attr, enum bpf_prog_type ptype);
int sock_map_update_elem_sys(struct bpf_map *map, void *key, void *value, u64 flags);
+int sockmap_bpf_prog_query(const union bpf_attr *attr,
+ union bpf_attr __user *uattr);
void sock_map_unhash(struct sock *sk);
void sock_map_close(struct sock *sk, long timeout);
#else
@@ -2012,6 +2014,12 @@ static inline int sock_map_update_elem_sys(struct bpf_map *map, void *key, void
{
return -EOPNOTSUPP;
}
+
+static inline int sockmap_bpf_prog_query(const union bpf_attr *attr,
+ union bpf_attr __user *uattr)
+{
+ return -EINVAL;
+}
#endif /* CONFIG_BPF_SYSCALL */
#endif /* CONFIG_NET && CONFIG_BPF_SYSCALL */
diff --git a/kernel/bpf/syscall.c b/kernel/bpf/syscall.c
index 4e50c0bfdb7d..3049b9506583 100644
--- a/kernel/bpf/syscall.c
+++ b/kernel/bpf/syscall.c
@@ -3275,6 +3275,10 @@ static int bpf_prog_query(const union bpf_attr *attr,
case BPF_FLOW_DISSECTOR:
case BPF_SK_LOOKUP:
return netns_bpf_prog_query(attr, uattr);
+ case BPF_SK_SKB_STREAM_PARSER:
+ case BPF_SK_SKB_STREAM_VERDICT:
+ case BPF_SK_MSG_VERDICT:
+ return sockmap_bpf_prog_query(attr, uattr);
default:
return -EINVAL;
}
diff --git a/net/core/sock_map.c b/net/core/sock_map.c
index e252b8ec2b85..32688a1b78c6 100644
--- a/net/core/sock_map.c
+++ b/net/core/sock_map.c
@@ -1451,6 +1451,87 @@ static int sock_map_prog_update(struct bpf_map *map, struct bpf_prog *prog,
return 0;
}
+static int sock_map_prog_lookup(struct bpf_map *map, struct bpf_prog **prog,
+ u32 which)
+{
+ struct sk_psock_progs *progs = sock_map_progs(map);
+
+ if (!progs)
+ return -EOPNOTSUPP;
+
+ switch (which) {
+ case BPF_SK_MSG_VERDICT:
+ *prog = READ_ONCE(progs->msg_parser);
+ break;
+#if IS_ENABLED(CONFIG_BPF_STREAM_PARSER)
+ case BPF_SK_SKB_STREAM_PARSER:
+ *prog = READ_ONCE(progs->skb_parser);
+ break;
+#endif
+ case BPF_SK_SKB_STREAM_VERDICT:
+ *prog = READ_ONCE(progs->skb_verdict);
+ break;
+ default:
+ return -EOPNOTSUPP;
+ }
+
+ return 0;
+}
+
+int sockmap_bpf_prog_query(const union bpf_attr *attr,
+ union bpf_attr __user *uattr)
+{
+ __u32 __user *prog_ids = u64_to_user_ptr(attr->query.prog_ids);
+ u32 prog_cnt = 0, flags = 0;
+ u32 ufd = attr->target_fd;
+ struct bpf_map *map;
+ struct bpf_prog *prog;
+ struct fd f;
+ int ret;
+
+ if (attr->query.query_flags)
+ return -EINVAL;
+
+ if (copy_to_user(&uattr->query.attach_flags, &flags, sizeof(flags)))
+ return -EFAULT;
+
+ f = fdget(ufd);
+ map = __bpf_map_get(f);
+ if (IS_ERR(map))
+ return PTR_ERR(map);
+
+ rcu_read_lock();
+
+ ret = sock_map_prog_lookup(map, &prog, attr->query.attach_type);
+ if (ret)
+ goto end;
+
+ prog_cnt = (!prog) ? 0 : 1;
+ if (copy_to_user(&uattr->query.prog_cnt, &prog_cnt, sizeof(prog_cnt))) {
+ ret = -EFAULT;
+ goto end;
+ }
+
+ if (!attr->query.prog_cnt || !prog_ids || !prog_cnt)
+ goto end;
+
+ prog = bpf_prog_inc_not_zero(prog);
+ if (IS_ERR(prog)) {
+ ret = PTR_ERR(prog);
+ goto end;
+ }
+
+ if (copy_to_user(prog_ids, &prog->aux->id, sizeof(u32)))
+ ret = -EFAULT;
+
+ bpf_prog_put(prog);
+
+end:
+ rcu_read_unlock();
+ fdput(f);
+ return ret;
+}
+
static void sock_map_unlink(struct sock *sk, struct sk_psock_link *link)
{
switch (link->map->map_type) {
--
2.23.0
^ permalink raw reply related [flat|nested] 7+ messages in thread
end of thread, other threads:[~2021-10-23 6:05 UTC | newest]
Thread overview: 7+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2021-10-23 6:05 [PATCH] bpf: support BPF_PROG_QUERY for progs attached to sockmap zhudi (E)
-- strict thread matches above, loose matches on Subject: below --
2021-10-23 6:01 zhudi (E)
2021-10-22 10:33 Di Zhu
2021-10-22 15:25 ` Alexei Starovoitov
2021-10-22 15:46 ` John Fastabend
2021-10-19 12:58 Di Zhu
2021-10-22 0:36 ` Alexei Starovoitov
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).