* [PATCH bpf-next 0/4] libbpf: streamline netlink-based XDP APIs
@ 2022-01-20 6:14 Andrii Nakryiko
2022-01-20 6:14 ` [PATCH bpf-next 1/4] libbpf: streamline low-level " Andrii Nakryiko
` (3 more replies)
0 siblings, 4 replies; 6+ messages in thread
From: Andrii Nakryiko @ 2022-01-20 6:14 UTC (permalink / raw)
To: bpf, ast, daniel; +Cc: andrii, kernel-team
Revamp existing low-level XDP APIs provided by libbpf to follow more
consistent naming (new APIs follow bpf_tc_xxx() approach where it makes
sense) and be extensible without ABI breakages (OPTS-based). See patch #1 for
details, remaining patches switch bpftool, selftests/bpf and samples/bpf to
new APIs.
Andrii Nakryiko (4):
libbpf: streamline low-level XDP APIs
bpftool: use new API for attaching XDP program
selftests/bpf: switch to new libbpf XDP APIs
samples/bpf: adapt samples/bpf to bpf_xdp_xxx() APIs
samples/bpf/xdp1_user.c | 8 +-
samples/bpf/xdp_adjust_tail_user.c | 8 +-
samples/bpf/xdp_fwd_user.c | 4 +-
samples/bpf/xdp_router_ipv4_user.c | 10 +-
samples/bpf/xdp_rxq_info_user.c | 8 +-
samples/bpf/xdp_sample_pkts_user.c | 8 +-
samples/bpf/xdp_sample_user.c | 9 +-
samples/bpf/xdp_tx_iptunnel_user.c | 10 +-
samples/bpf/xdpsock_ctrl_proc.c | 2 +-
samples/bpf/xdpsock_user.c | 10 +-
samples/bpf/xsk_fwd.c | 4 +-
tools/bpf/bpftool/net.c | 2 +-
tools/lib/bpf/libbpf.h | 29 +++++
tools/lib/bpf/libbpf.map | 4 +
tools/lib/bpf/netlink.c | 117 +++++++++++++-----
.../selftests/bpf/prog_tests/xdp_attach.c | 29 ++---
.../bpf/prog_tests/xdp_cpumap_attach.c | 8 +-
.../bpf/prog_tests/xdp_devmap_attach.c | 8 +-
.../selftests/bpf/prog_tests/xdp_info.c | 14 +--
.../selftests/bpf/prog_tests/xdp_link.c | 26 ++--
.../selftests/bpf/xdp_redirect_multi.c | 8 +-
tools/testing/selftests/bpf/xdping.c | 4 +-
22 files changed, 205 insertions(+), 125 deletions(-)
--
2.30.2
^ permalink raw reply [flat|nested] 6+ messages in thread
* [PATCH bpf-next 1/4] libbpf: streamline low-level XDP APIs
2022-01-20 6:14 [PATCH bpf-next 0/4] libbpf: streamline netlink-based XDP APIs Andrii Nakryiko
@ 2022-01-20 6:14 ` Andrii Nakryiko
2022-01-20 12:00 ` Toke Høiland-Jørgensen
2022-01-20 6:14 ` [PATCH bpf-next 2/4] bpftool: use new API for attaching XDP program Andrii Nakryiko
` (2 subsequent siblings)
3 siblings, 1 reply; 6+ messages in thread
From: Andrii Nakryiko @ 2022-01-20 6:14 UTC (permalink / raw)
To: bpf, ast, daniel; +Cc: andrii, kernel-team
Introduce 4 new netlink-based XDP APIs for attaching, detaching, and
querying XDP programs:
- bpf_xdp_attach;
- bpf_xdp_detach;
- bpf_xdp_query;
- bpf_xdp_query_id.
These APIs replace bpf_set_link_xdp_fd, bpf_set_link_xdp_fd_opts,
bpf_get_link_xdp_id, and bpf_get_link_xdp_info APIs ([0]). The latter
don't follow a consistent naming pattern and some of them use
non-extensible approaches (e.g., struct xdp_link_info which can't be
modified without breaking libbpf ABI).
The approach I took with these low-level XDP APIs is similar to what we
did with low-level TC APIs. There is a nice duality of bpf_tc_attach vs
bpf_xdp_attach, and so on. I left bpf_xdp_attach() to support detaching
when -1 is specified for prog_fd for generality and convenience, but
bpf_xdp_detach() is preferred due to clearer naming and associated
semantics. Both bpf_xdp_attach() and bpf_xdp_detach() accept the same
opts struct allowing to specify expected old_prog_fd.
While doing the refactoring, I noticed that old APIs require users to
specify opts with old_fd == -1 to declare "don't care about already
attached XDP prog fd" condition. Otherwise, FD 0 is assumed, which is
essentially never an intended behavior. So I made this behavior
consistent with other kernel and libbpf APIs, in which zero FD means "no
FD". This seems to be more in line with the latest thinking in BPF land
and should cause less user confusion, hopefully.
For querying, I left two APIs, both more generic bpf_xdp_query()
allowing to query multiple IDs and attach mode, but also
a specialization of it, bpf_xdp_query_id(), which returns only requested
prog_id. Uses of prog_id returning bpf_get_link_xdp_id() were so
prevalent across selftests and samples, that it seemed a very common use
case and using bpf_xdp_query() for doing it felt very cumbersome with
a highly branches if/else chain based on flags and attach mode.
Old APIs are scheduled for deprecation in libbpf 0.8 release.
[0] Closes: https://github.com/libbpf/libbpf/issues/309
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
---
tools/lib/bpf/libbpf.h | 29 ++++++++++
tools/lib/bpf/libbpf.map | 4 ++
tools/lib/bpf/netlink.c | 117 ++++++++++++++++++++++++++++-----------
3 files changed, 117 insertions(+), 33 deletions(-)
diff --git a/tools/lib/bpf/libbpf.h b/tools/lib/bpf/libbpf.h
index 9728551501ae..94670066de62 100644
--- a/tools/lib/bpf/libbpf.h
+++ b/tools/lib/bpf/libbpf.h
@@ -833,13 +833,42 @@ struct bpf_xdp_set_link_opts {
};
#define bpf_xdp_set_link_opts__last_field old_fd
+LIBBPF_DEPRECATED_SINCE(0, 8, "use bpf_xdp_attach() instead")
LIBBPF_API int bpf_set_link_xdp_fd(int ifindex, int fd, __u32 flags);
+LIBBPF_DEPRECATED_SINCE(0, 8, "use bpf_xdp_attach() instead")
LIBBPF_API int bpf_set_link_xdp_fd_opts(int ifindex, int fd, __u32 flags,
const struct bpf_xdp_set_link_opts *opts);
+LIBBPF_DEPRECATED_SINCE(0, 8, "use bpf_xdp_query_id() instead")
LIBBPF_API int bpf_get_link_xdp_id(int ifindex, __u32 *prog_id, __u32 flags);
+LIBBPF_DEPRECATED_SINCE(0, 8, "use bpf_xdp_query() instead")
LIBBPF_API int bpf_get_link_xdp_info(int ifindex, struct xdp_link_info *info,
size_t info_size, __u32 flags);
+struct bpf_xdp_attach_opts {
+ size_t sz;
+ int old_prog_fd;
+ size_t :0;
+};
+#define bpf_xdp_attach_opts__last_field old_prog_fd
+
+struct bpf_xdp_query_opts {
+ size_t sz;
+ __u32 prog_id; /* output */
+ __u32 drv_prog_id; /* output */
+ __u32 hw_prog_id; /* output */
+ __u32 skb_prog_id; /* output */
+ __u8 attach_mode; /* output */
+ size_t :0;
+};
+#define bpf_xdp_query_opts__last_field attach_mode
+
+LIBBPF_API int bpf_xdp_attach(int ifindex, int prog_fd, __u32 flags,
+ const struct bpf_xdp_attach_opts *opts);
+LIBBPF_API int bpf_xdp_detach(int ifindex, __u32 flags,
+ const struct bpf_xdp_attach_opts *opts);
+LIBBPF_API int bpf_xdp_query(int ifindex, int flags, struct bpf_xdp_query_opts *opts);
+LIBBPF_API int bpf_xdp_query_id(int ifindex, int flags, __u32 *prog_id);
+
/* TC related API */
enum bpf_tc_attach_point {
BPF_TC_INGRESS = 1 << 0,
diff --git a/tools/lib/bpf/libbpf.map b/tools/lib/bpf/libbpf.map
index 8262cfca2240..e10f0822845a 100644
--- a/tools/lib/bpf/libbpf.map
+++ b/tools/lib/bpf/libbpf.map
@@ -428,6 +428,10 @@ LIBBPF_0.7.0 {
bpf_program__log_level;
bpf_program__set_log_buf;
bpf_program__set_log_level;
+ bpf_xdp_attach;
+ bpf_xdp_detach;
+ bpf_xdp_query;
+ bpf_xdp_query_id;
libbpf_probe_bpf_helper;
libbpf_probe_bpf_map_type;
libbpf_probe_bpf_prog_type;
diff --git a/tools/lib/bpf/netlink.c b/tools/lib/bpf/netlink.c
index 39f25e09b51e..c39c37f99d5c 100644
--- a/tools/lib/bpf/netlink.c
+++ b/tools/lib/bpf/netlink.c
@@ -217,6 +217,28 @@ static int __bpf_set_link_xdp_fd_replace(int ifindex, int fd, int old_fd,
return libbpf_netlink_send_recv(&req, NULL, NULL, NULL);
}
+int bpf_xdp_attach(int ifindex, int prog_fd, __u32 flags, const struct bpf_xdp_attach_opts *opts)
+{
+ int old_prog_fd, err;
+
+ if (!OPTS_VALID(opts, bpf_xdp_attach_opts))
+ return libbpf_err(-EINVAL);
+
+ old_prog_fd = OPTS_GET(opts, old_prog_fd, 0);
+ if (old_prog_fd)
+ flags |= XDP_FLAGS_REPLACE;
+ else
+ old_prog_fd = -1;
+
+ err = __bpf_set_link_xdp_fd_replace(ifindex, prog_fd, old_prog_fd, flags);
+ return libbpf_err(err);
+}
+
+int bpf_xdp_detach(int ifindex, __u32 flags, const struct bpf_xdp_attach_opts *opts)
+{
+ return bpf_xdp_attach(ifindex, -1, flags, opts);
+}
+
int bpf_set_link_xdp_fd_opts(int ifindex, int fd, __u32 flags,
const struct bpf_xdp_set_link_opts *opts)
{
@@ -303,69 +325,98 @@ static int get_xdp_info(void *cookie, void *msg, struct nlattr **tb)
return 0;
}
-int bpf_get_link_xdp_info(int ifindex, struct xdp_link_info *info,
- size_t info_size, __u32 flags)
+int bpf_xdp_query(int ifindex, int xdp_flags, struct bpf_xdp_query_opts *opts)
{
- struct xdp_id_md xdp_id = {};
- __u32 mask;
- int ret;
struct libbpf_nla_req req = {
.nh.nlmsg_len = NLMSG_LENGTH(sizeof(struct ifinfomsg)),
.nh.nlmsg_type = RTM_GETLINK,
.nh.nlmsg_flags = NLM_F_DUMP | NLM_F_REQUEST,
.ifinfo.ifi_family = AF_PACKET,
};
+ struct xdp_id_md xdp_id = {};
+ int err;
- if (flags & ~XDP_FLAGS_MASK || !info_size)
+ if (!OPTS_VALID(opts, bpf_xdp_query_opts))
+ return libbpf_err(-EINVAL);
+
+ if (xdp_flags & ~XDP_FLAGS_MASK)
return libbpf_err(-EINVAL);
/* Check whether the single {HW,DRV,SKB} mode is set */
- flags &= (XDP_FLAGS_SKB_MODE | XDP_FLAGS_DRV_MODE | XDP_FLAGS_HW_MODE);
- mask = flags - 1;
- if (flags && flags & mask)
+ xdp_flags &= XDP_FLAGS_SKB_MODE | XDP_FLAGS_DRV_MODE | XDP_FLAGS_HW_MODE;
+ if (xdp_flags & (xdp_flags - 1))
return libbpf_err(-EINVAL);
xdp_id.ifindex = ifindex;
- xdp_id.flags = flags;
+ xdp_id.flags = xdp_flags;
- ret = libbpf_netlink_send_recv(&req, __dump_link_nlmsg,
+ err = libbpf_netlink_send_recv(&req, __dump_link_nlmsg,
get_xdp_info, &xdp_id);
- if (!ret) {
- size_t sz = min(info_size, sizeof(xdp_id.info));
+ if (err)
+ return libbpf_err(err);
- memcpy(info, &xdp_id.info, sz);
- memset((void *) info + sz, 0, info_size - sz);
- }
+ OPTS_SET(opts, prog_id, xdp_id.info.prog_id);
+ OPTS_SET(opts, drv_prog_id, xdp_id.info.drv_prog_id);
+ OPTS_SET(opts, hw_prog_id, xdp_id.info.hw_prog_id);
+ OPTS_SET(opts, skb_prog_id, xdp_id.info.skb_prog_id);
+ OPTS_SET(opts, attach_mode, xdp_id.info.attach_mode);
- return libbpf_err(ret);
+ return 0;
}
-static __u32 get_xdp_id(struct xdp_link_info *info, __u32 flags)
+int bpf_get_link_xdp_info(int ifindex, struct xdp_link_info *info,
+ size_t info_size, __u32 flags)
{
- flags &= XDP_FLAGS_MODES;
+ LIBBPF_OPTS(bpf_xdp_query_opts, opts);
+ size_t sz;
+ int err;
+
+ if (!info_size)
+ return libbpf_err(-EINVAL);
- if (info->attach_mode != XDP_ATTACHED_MULTI && !flags)
- return info->prog_id;
- if (flags & XDP_FLAGS_DRV_MODE)
- return info->drv_prog_id;
- if (flags & XDP_FLAGS_HW_MODE)
- return info->hw_prog_id;
- if (flags & XDP_FLAGS_SKB_MODE)
- return info->skb_prog_id;
+ err = bpf_xdp_query(ifindex, flags, &opts);
+ if (err)
+ return libbpf_err(err);
+
+ /* struct xdp_link_info field layout matches struct bpf_xdp_query_opts
+ * layout after sz field
+ */
+ sz = min(info_size, offsetofend(struct xdp_link_info, attach_mode));
+ memcpy(info, &opts.prog_id, sz);
+ memset((void *)info + sz, 0, info_size - sz);
return 0;
}
-int bpf_get_link_xdp_id(int ifindex, __u32 *prog_id, __u32 flags)
+int bpf_xdp_query_id(int ifindex, int flags, __u32 *prog_id)
{
- struct xdp_link_info info;
+ LIBBPF_OPTS(bpf_xdp_query_opts, opts);
int ret;
- ret = bpf_get_link_xdp_info(ifindex, &info, sizeof(info), flags);
- if (!ret)
- *prog_id = get_xdp_id(&info, flags);
+ ret = bpf_xdp_query(ifindex, flags, &opts);
+ if (ret)
+ return libbpf_err(ret);
+
+ flags &= XDP_FLAGS_MODES;
- return libbpf_err(ret);
+ if (opts.attach_mode != XDP_ATTACHED_MULTI && !flags)
+ *prog_id = opts.prog_id;
+ else if (flags & XDP_FLAGS_DRV_MODE)
+ *prog_id = opts.drv_prog_id;
+ else if (flags & XDP_FLAGS_HW_MODE)
+ *prog_id = opts.hw_prog_id;
+ else if (flags & XDP_FLAGS_SKB_MODE)
+ *prog_id = opts.skb_prog_id;
+ else
+ *prog_id = 0;
+
+ return 0;
+}
+
+
+int bpf_get_link_xdp_id(int ifindex, __u32 *prog_id, __u32 flags)
+{
+ return bpf_xdp_query_id(ifindex, flags, prog_id);
}
typedef int (*qdisc_config_t)(struct libbpf_nla_req *req);
--
2.30.2
^ permalink raw reply related [flat|nested] 6+ messages in thread
* [PATCH bpf-next 2/4] bpftool: use new API for attaching XDP program
2022-01-20 6:14 [PATCH bpf-next 0/4] libbpf: streamline netlink-based XDP APIs Andrii Nakryiko
2022-01-20 6:14 ` [PATCH bpf-next 1/4] libbpf: streamline low-level " Andrii Nakryiko
@ 2022-01-20 6:14 ` Andrii Nakryiko
2022-01-20 6:14 ` [PATCH bpf-next 3/4] selftests/bpf: switch to new libbpf XDP APIs Andrii Nakryiko
2022-01-20 6:14 ` [PATCH bpf-next 4/4] samples/bpf: adapt samples/bpf to bpf_xdp_xxx() APIs Andrii Nakryiko
3 siblings, 0 replies; 6+ messages in thread
From: Andrii Nakryiko @ 2022-01-20 6:14 UTC (permalink / raw)
To: bpf, ast, daniel; +Cc: andrii, kernel-team
Switch to new bpf_xdp_attach() API to avoid deprecation warnings.
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
---
tools/bpf/bpftool/net.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/tools/bpf/bpftool/net.c b/tools/bpf/bpftool/net.c
index 649053704bd7..526a332c48e6 100644
--- a/tools/bpf/bpftool/net.c
+++ b/tools/bpf/bpftool/net.c
@@ -551,7 +551,7 @@ static int do_attach_detach_xdp(int progfd, enum net_attach_type attach_type,
if (attach_type == NET_ATTACH_TYPE_XDP_OFFLOAD)
flags |= XDP_FLAGS_HW_MODE;
- return bpf_set_link_xdp_fd(ifindex, progfd, flags);
+ return bpf_xdp_attach(ifindex, progfd, flags, NULL);
}
static int do_attach(int argc, char **argv)
--
2.30.2
^ permalink raw reply related [flat|nested] 6+ messages in thread
* [PATCH bpf-next 3/4] selftests/bpf: switch to new libbpf XDP APIs
2022-01-20 6:14 [PATCH bpf-next 0/4] libbpf: streamline netlink-based XDP APIs Andrii Nakryiko
2022-01-20 6:14 ` [PATCH bpf-next 1/4] libbpf: streamline low-level " Andrii Nakryiko
2022-01-20 6:14 ` [PATCH bpf-next 2/4] bpftool: use new API for attaching XDP program Andrii Nakryiko
@ 2022-01-20 6:14 ` Andrii Nakryiko
2022-01-20 6:14 ` [PATCH bpf-next 4/4] samples/bpf: adapt samples/bpf to bpf_xdp_xxx() APIs Andrii Nakryiko
3 siblings, 0 replies; 6+ messages in thread
From: Andrii Nakryiko @ 2022-01-20 6:14 UTC (permalink / raw)
To: bpf, ast, daniel; +Cc: andrii, kernel-team
Switch to using new bpf_xdp_*() APIs across all selftests. Take
advantage of a more straightforward and user-friendly semantics of
old_prog_fd (0 means "don't care") in few places.
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
---
.../selftests/bpf/prog_tests/xdp_attach.c | 29 +++++++++----------
.../bpf/prog_tests/xdp_cpumap_attach.c | 8 ++---
.../bpf/prog_tests/xdp_devmap_attach.c | 8 ++---
.../selftests/bpf/prog_tests/xdp_info.c | 14 ++++-----
.../selftests/bpf/prog_tests/xdp_link.c | 26 ++++++++---------
.../selftests/bpf/xdp_redirect_multi.c | 8 ++---
tools/testing/selftests/bpf/xdping.c | 4 +--
7 files changed, 47 insertions(+), 50 deletions(-)
diff --git a/tools/testing/selftests/bpf/prog_tests/xdp_attach.c b/tools/testing/selftests/bpf/prog_tests/xdp_attach.c
index c6fa390e3aa1..62aa3edda5e6 100644
--- a/tools/testing/selftests/bpf/prog_tests/xdp_attach.c
+++ b/tools/testing/selftests/bpf/prog_tests/xdp_attach.c
@@ -11,8 +11,7 @@ void serial_test_xdp_attach(void)
const char *file = "./test_xdp.o";
struct bpf_prog_info info = {};
int err, fd1, fd2, fd3;
- DECLARE_LIBBPF_OPTS(bpf_xdp_set_link_opts, opts,
- .old_fd = -1);
+ LIBBPF_OPTS(bpf_xdp_attach_opts, opts);
len = sizeof(info);
@@ -38,49 +37,47 @@ void serial_test_xdp_attach(void)
if (CHECK_FAIL(err))
goto out_2;
- err = bpf_set_link_xdp_fd_opts(IFINDEX_LO, fd1, XDP_FLAGS_REPLACE,
- &opts);
+ err = bpf_xdp_attach(IFINDEX_LO, fd1, XDP_FLAGS_REPLACE, &opts);
if (CHECK(err, "load_ok", "initial load failed"))
goto out_close;
- err = bpf_get_link_xdp_id(IFINDEX_LO, &id0, 0);
+ err = bpf_xdp_query_id(IFINDEX_LO, 0, &id0);
if (CHECK(err || id0 != id1, "id1_check",
"loaded prog id %u != id1 %u, err %d", id0, id1, err))
goto out_close;
- err = bpf_set_link_xdp_fd_opts(IFINDEX_LO, fd2, XDP_FLAGS_REPLACE,
- &opts);
+ err = bpf_xdp_attach(IFINDEX_LO, fd2, XDP_FLAGS_REPLACE, &opts);
if (CHECK(!err, "load_fail", "load with expected id didn't fail"))
goto out;
- opts.old_fd = fd1;
- err = bpf_set_link_xdp_fd_opts(IFINDEX_LO, fd2, 0, &opts);
+ opts.old_prog_fd = fd1;
+ err = bpf_xdp_attach(IFINDEX_LO, fd2, 0, &opts);
if (CHECK(err, "replace_ok", "replace valid old_fd failed"))
goto out;
- err = bpf_get_link_xdp_id(IFINDEX_LO, &id0, 0);
+ err = bpf_xdp_query_id(IFINDEX_LO, 0, &id0);
if (CHECK(err || id0 != id2, "id2_check",
"loaded prog id %u != id2 %u, err %d", id0, id2, err))
goto out_close;
- err = bpf_set_link_xdp_fd_opts(IFINDEX_LO, fd3, 0, &opts);
+ err = bpf_xdp_attach(IFINDEX_LO, fd3, 0, &opts);
if (CHECK(!err, "replace_fail", "replace invalid old_fd didn't fail"))
goto out;
- err = bpf_set_link_xdp_fd_opts(IFINDEX_LO, -1, 0, &opts);
+ err = bpf_xdp_detach(IFINDEX_LO, 0, &opts);
if (CHECK(!err, "remove_fail", "remove invalid old_fd didn't fail"))
goto out;
- opts.old_fd = fd2;
- err = bpf_set_link_xdp_fd_opts(IFINDEX_LO, -1, 0, &opts);
+ opts.old_prog_fd = fd2;
+ err = bpf_xdp_detach(IFINDEX_LO, 0, &opts);
if (CHECK(err, "remove_ok", "remove valid old_fd failed"))
goto out;
- err = bpf_get_link_xdp_id(IFINDEX_LO, &id0, 0);
+ err = bpf_xdp_query_id(IFINDEX_LO, 0, &id0);
if (CHECK(err || id0 != 0, "unload_check",
"loaded prog id %u != 0, err %d", id0, err))
goto out_close;
out:
- bpf_set_link_xdp_fd(IFINDEX_LO, -1, 0);
+ bpf_xdp_detach(IFINDEX_LO, 0, NULL);
out_close:
bpf_object__close(obj3);
out_2:
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 fd812bd43600..abe9d9f988ec 100644
--- a/tools/testing/selftests/bpf/prog_tests/xdp_cpumap_attach.c
+++ b/tools/testing/selftests/bpf/prog_tests/xdp_cpumap_attach.c
@@ -23,11 +23,11 @@ void serial_test_xdp_cpumap_attach(void)
return;
prog_fd = bpf_program__fd(skel->progs.xdp_redir_prog);
- err = bpf_set_link_xdp_fd(IFINDEX_LO, prog_fd, XDP_FLAGS_SKB_MODE);
+ err = bpf_xdp_attach(IFINDEX_LO, prog_fd, XDP_FLAGS_SKB_MODE, NULL);
if (!ASSERT_OK(err, "Generic attach of program with 8-byte CPUMAP"))
goto out_close;
- err = bpf_set_link_xdp_fd(IFINDEX_LO, -1, XDP_FLAGS_SKB_MODE);
+ 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);
@@ -45,9 +45,9 @@ void serial_test_xdp_cpumap_attach(void)
ASSERT_EQ(info.id, val.bpf_prog.id, "Match program id to cpumap entry prog_id");
/* can not attach BPF_XDP_CPUMAP program to a device */
- err = bpf_set_link_xdp_fd(IFINDEX_LO, prog_fd, XDP_FLAGS_SKB_MODE);
+ err = bpf_xdp_attach(IFINDEX_LO, prog_fd, XDP_FLAGS_SKB_MODE, NULL);
if (!ASSERT_NEQ(err, 0, "Attach of BPF_XDP_CPUMAP program"))
- bpf_set_link_xdp_fd(IFINDEX_LO, -1, XDP_FLAGS_SKB_MODE);
+ bpf_xdp_detach(IFINDEX_LO, XDP_FLAGS_SKB_MODE, NULL);
val.qsize = 192;
val.bpf_prog.fd = bpf_program__fd(skel->progs.xdp_dummy_prog);
diff --git a/tools/testing/selftests/bpf/prog_tests/xdp_devmap_attach.c b/tools/testing/selftests/bpf/prog_tests/xdp_devmap_attach.c
index 3079d5568f8f..fc1a40c3193b 100644
--- a/tools/testing/selftests/bpf/prog_tests/xdp_devmap_attach.c
+++ b/tools/testing/selftests/bpf/prog_tests/xdp_devmap_attach.c
@@ -25,11 +25,11 @@ static void test_xdp_with_devmap_helpers(void)
return;
dm_fd = bpf_program__fd(skel->progs.xdp_redir_prog);
- err = bpf_set_link_xdp_fd(IFINDEX_LO, dm_fd, XDP_FLAGS_SKB_MODE);
+ err = bpf_xdp_attach(IFINDEX_LO, dm_fd, XDP_FLAGS_SKB_MODE, NULL);
if (!ASSERT_OK(err, "Generic attach of program with 8-byte devmap"))
goto out_close;
- err = bpf_set_link_xdp_fd(IFINDEX_LO, -1, XDP_FLAGS_SKB_MODE);
+ err = bpf_xdp_detach(IFINDEX_LO, XDP_FLAGS_SKB_MODE, NULL);
ASSERT_OK(err, "XDP program detach");
dm_fd = bpf_program__fd(skel->progs.xdp_dummy_dm);
@@ -47,9 +47,9 @@ static void test_xdp_with_devmap_helpers(void)
ASSERT_EQ(info.id, val.bpf_prog.id, "Match program id to devmap entry prog_id");
/* can not attach BPF_XDP_DEVMAP program to a device */
- err = bpf_set_link_xdp_fd(IFINDEX_LO, dm_fd, XDP_FLAGS_SKB_MODE);
+ err = bpf_xdp_attach(IFINDEX_LO, dm_fd, XDP_FLAGS_SKB_MODE, NULL);
if (!ASSERT_NEQ(err, 0, "Attach of BPF_XDP_DEVMAP program"))
- bpf_set_link_xdp_fd(IFINDEX_LO, -1, XDP_FLAGS_SKB_MODE);
+ bpf_xdp_detach(IFINDEX_LO, XDP_FLAGS_SKB_MODE, NULL);
val.ifindex = 1;
val.bpf_prog.fd = bpf_program__fd(skel->progs.xdp_dummy_prog);
diff --git a/tools/testing/selftests/bpf/prog_tests/xdp_info.c b/tools/testing/selftests/bpf/prog_tests/xdp_info.c
index abe48e82e1dc..0d01ff6cb91a 100644
--- a/tools/testing/selftests/bpf/prog_tests/xdp_info.c
+++ b/tools/testing/selftests/bpf/prog_tests/xdp_info.c
@@ -14,13 +14,13 @@ void serial_test_xdp_info(void)
/* Get prog_id for XDP_ATTACHED_NONE mode */
- err = bpf_get_link_xdp_id(IFINDEX_LO, &prog_id, 0);
+ err = bpf_xdp_query_id(IFINDEX_LO, 0, &prog_id);
if (CHECK(err, "get_xdp_none", "errno=%d\n", errno))
return;
if (CHECK(prog_id, "prog_id_none", "unexpected prog_id=%u\n", prog_id))
return;
- err = bpf_get_link_xdp_id(IFINDEX_LO, &prog_id, XDP_FLAGS_SKB_MODE);
+ err = bpf_xdp_query_id(IFINDEX_LO, XDP_FLAGS_SKB_MODE, &prog_id);
if (CHECK(err, "get_xdp_none_skb", "errno=%d\n", errno))
return;
if (CHECK(prog_id, "prog_id_none_skb", "unexpected prog_id=%u\n",
@@ -37,32 +37,32 @@ void serial_test_xdp_info(void)
if (CHECK(err, "get_prog_info", "errno=%d\n", errno))
goto out_close;
- err = bpf_set_link_xdp_fd(IFINDEX_LO, prog_fd, XDP_FLAGS_SKB_MODE);
+ err = bpf_xdp_attach(IFINDEX_LO, prog_fd, XDP_FLAGS_SKB_MODE, NULL);
if (CHECK(err, "set_xdp_skb", "errno=%d\n", errno))
goto out_close;
/* Get prog_id for single prog mode */
- err = bpf_get_link_xdp_id(IFINDEX_LO, &prog_id, 0);
+ err = bpf_xdp_query_id(IFINDEX_LO, 0, &prog_id);
if (CHECK(err, "get_xdp", "errno=%d\n", errno))
goto out;
if (CHECK(prog_id != info.id, "prog_id", "prog_id not available\n"))
goto out;
- err = bpf_get_link_xdp_id(IFINDEX_LO, &prog_id, XDP_FLAGS_SKB_MODE);
+ err = bpf_xdp_query_id(IFINDEX_LO, XDP_FLAGS_SKB_MODE, &prog_id);
if (CHECK(err, "get_xdp_skb", "errno=%d\n", errno))
goto out;
if (CHECK(prog_id != info.id, "prog_id_skb", "prog_id not available\n"))
goto out;
- err = bpf_get_link_xdp_id(IFINDEX_LO, &prog_id, XDP_FLAGS_DRV_MODE);
+ err = bpf_xdp_query_id(IFINDEX_LO, XDP_FLAGS_DRV_MODE, &prog_id);
if (CHECK(err, "get_xdp_drv", "errno=%d\n", errno))
goto out;
if (CHECK(prog_id, "prog_id_drv", "unexpected prog_id=%u\n", prog_id))
goto out;
out:
- bpf_set_link_xdp_fd(IFINDEX_LO, -1, 0);
+ bpf_xdp_detach(IFINDEX_LO, 0, NULL);
out_close:
bpf_object__close(obj);
}
diff --git a/tools/testing/selftests/bpf/prog_tests/xdp_link.c b/tools/testing/selftests/bpf/prog_tests/xdp_link.c
index 983ab0b47d30..0c5e4ea8eaae 100644
--- a/tools/testing/selftests/bpf/prog_tests/xdp_link.c
+++ b/tools/testing/selftests/bpf/prog_tests/xdp_link.c
@@ -9,8 +9,8 @@
void serial_test_xdp_link(void)
{
__u32 duration = 0, id1, id2, id0 = 0, prog_fd1, prog_fd2, err;
- DECLARE_LIBBPF_OPTS(bpf_xdp_set_link_opts, opts, .old_fd = -1);
struct test_xdp_link *skel1 = NULL, *skel2 = NULL;
+ LIBBPF_OPTS(bpf_xdp_attach_opts, opts);
struct bpf_link_info link_info;
struct bpf_prog_info prog_info;
struct bpf_link *link;
@@ -40,12 +40,12 @@ void serial_test_xdp_link(void)
id2 = prog_info.id;
/* set initial prog attachment */
- err = bpf_set_link_xdp_fd_opts(IFINDEX_LO, prog_fd1, XDP_FLAGS_REPLACE, &opts);
+ err = bpf_xdp_attach(IFINDEX_LO, prog_fd1, XDP_FLAGS_REPLACE, &opts);
if (CHECK(err, "fd_attach", "initial prog attach failed: %d\n", err))
goto cleanup;
/* validate prog ID */
- err = bpf_get_link_xdp_id(IFINDEX_LO, &id0, 0);
+ err = bpf_xdp_query_id(IFINDEX_LO, 0, &id0);
CHECK(err || id0 != id1, "id1_check",
"loaded prog id %u != id1 %u, err %d", id0, id1, err);
@@ -54,14 +54,14 @@ void serial_test_xdp_link(void)
if (!ASSERT_ERR_PTR(link, "link_attach_should_fail")) {
bpf_link__destroy(link);
/* best-effort detach prog */
- opts.old_fd = prog_fd1;
- bpf_set_link_xdp_fd_opts(IFINDEX_LO, -1, XDP_FLAGS_REPLACE, &opts);
+ opts.old_prog_fd = prog_fd1;
+ bpf_xdp_detach(IFINDEX_LO, XDP_FLAGS_REPLACE, &opts);
goto cleanup;
}
/* detach BPF program */
- opts.old_fd = prog_fd1;
- err = bpf_set_link_xdp_fd_opts(IFINDEX_LO, -1, XDP_FLAGS_REPLACE, &opts);
+ opts.old_prog_fd = prog_fd1;
+ err = bpf_xdp_detach(IFINDEX_LO, XDP_FLAGS_REPLACE, &opts);
if (CHECK(err, "prog_detach", "failed %d\n", err))
goto cleanup;
@@ -72,24 +72,24 @@ void serial_test_xdp_link(void)
skel1->links.xdp_handler = link;
/* validate prog ID */
- err = bpf_get_link_xdp_id(IFINDEX_LO, &id0, 0);
+ err = bpf_xdp_query_id(IFINDEX_LO, 0, &id0);
if (CHECK(err || id0 != id1, "id1_check",
"loaded prog id %u != id1 %u, err %d", id0, id1, err))
goto cleanup;
/* BPF prog attach is not allowed to replace BPF link */
- opts.old_fd = prog_fd1;
- err = bpf_set_link_xdp_fd_opts(IFINDEX_LO, prog_fd2, XDP_FLAGS_REPLACE, &opts);
+ opts.old_prog_fd = prog_fd1;
+ err = bpf_xdp_attach(IFINDEX_LO, prog_fd2, XDP_FLAGS_REPLACE, &opts);
if (CHECK(!err, "prog_attach_fail", "unexpected success\n"))
goto cleanup;
/* Can't force-update when BPF link is active */
- err = bpf_set_link_xdp_fd(IFINDEX_LO, prog_fd2, 0);
+ err = bpf_xdp_attach(IFINDEX_LO, prog_fd2, 0, NULL);
if (CHECK(!err, "prog_update_fail", "unexpected success\n"))
goto cleanup;
/* Can't force-detach when BPF link is active */
- err = bpf_set_link_xdp_fd(IFINDEX_LO, -1, 0);
+ err = bpf_xdp_detach(IFINDEX_LO, 0, NULL);
if (CHECK(!err, "prog_detach_fail", "unexpected success\n"))
goto cleanup;
@@ -109,7 +109,7 @@ void serial_test_xdp_link(void)
goto cleanup;
skel2->links.xdp_handler = link;
- err = bpf_get_link_xdp_id(IFINDEX_LO, &id0, 0);
+ err = bpf_xdp_query_id(IFINDEX_LO, 0, &id0);
if (CHECK(err || id0 != id2, "id2_check",
"loaded prog id %u != id2 %u, err %d", id0, id1, err))
goto cleanup;
diff --git a/tools/testing/selftests/bpf/xdp_redirect_multi.c b/tools/testing/selftests/bpf/xdp_redirect_multi.c
index 51c8224b4ccc..aaedbf4955c3 100644
--- a/tools/testing/selftests/bpf/xdp_redirect_multi.c
+++ b/tools/testing/selftests/bpf/xdp_redirect_multi.c
@@ -32,12 +32,12 @@ static void int_exit(int sig)
int i;
for (i = 0; ifaces[i] > 0; i++) {
- if (bpf_get_link_xdp_id(ifaces[i], &prog_id, xdp_flags)) {
- printf("bpf_get_link_xdp_id failed\n");
+ if (bpf_xdp_query_id(ifaces[i], xdp_flags, &prog_id)) {
+ printf("bpf_xdp_query_id failed\n");
exit(1);
}
if (prog_id)
- bpf_set_link_xdp_fd(ifaces[i], -1, xdp_flags);
+ bpf_xdp_detach(ifaces[i], xdp_flags, NULL);
}
exit(0);
@@ -210,7 +210,7 @@ int main(int argc, char **argv)
}
/* bind prog_fd to each interface */
- ret = bpf_set_link_xdp_fd(ifindex, prog_fd, xdp_flags);
+ ret = bpf_xdp_attach(ifindex, prog_fd, xdp_flags, NULL);
if (ret) {
printf("Set xdp fd failed on %d\n", ifindex);
goto err_out;
diff --git a/tools/testing/selftests/bpf/xdping.c b/tools/testing/selftests/bpf/xdping.c
index baa870a759a2..c567856fd1bc 100644
--- a/tools/testing/selftests/bpf/xdping.c
+++ b/tools/testing/selftests/bpf/xdping.c
@@ -29,7 +29,7 @@ static __u32 xdp_flags = XDP_FLAGS_UPDATE_IF_NOEXIST;
static void cleanup(int sig)
{
- bpf_set_link_xdp_fd(ifindex, -1, xdp_flags);
+ bpf_xdp_detach(ifindex, xdp_flags, NULL);
if (sig)
exit(1);
}
@@ -203,7 +203,7 @@ int main(int argc, char **argv)
printf("XDP setup disrupts network connectivity, hit Ctrl+C to quit\n");
- if (bpf_set_link_xdp_fd(ifindex, prog_fd, xdp_flags) < 0) {
+ if (bpf_xdp_attach(ifindex, prog_fd, xdp_flags, NULL) < 0) {
fprintf(stderr, "Link set xdp fd failed for %s\n", ifname);
goto done;
}
--
2.30.2
^ permalink raw reply related [flat|nested] 6+ messages in thread
* [PATCH bpf-next 4/4] samples/bpf: adapt samples/bpf to bpf_xdp_xxx() APIs
2022-01-20 6:14 [PATCH bpf-next 0/4] libbpf: streamline netlink-based XDP APIs Andrii Nakryiko
` (2 preceding siblings ...)
2022-01-20 6:14 ` [PATCH bpf-next 3/4] selftests/bpf: switch to new libbpf XDP APIs Andrii Nakryiko
@ 2022-01-20 6:14 ` Andrii Nakryiko
3 siblings, 0 replies; 6+ messages in thread
From: Andrii Nakryiko @ 2022-01-20 6:14 UTC (permalink / raw)
To: bpf, ast, daniel; +Cc: andrii, kernel-team
Use new bpf_xdp_*() APIs across all XDP-related BPF samples.
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
---
samples/bpf/xdp1_user.c | 8 ++++----
samples/bpf/xdp_adjust_tail_user.c | 8 ++++----
samples/bpf/xdp_fwd_user.c | 4 ++--
samples/bpf/xdp_router_ipv4_user.c | 10 +++++-----
samples/bpf/xdp_rxq_info_user.c | 8 ++++----
samples/bpf/xdp_sample_pkts_user.c | 8 ++++----
samples/bpf/xdp_sample_user.c | 9 ++++-----
samples/bpf/xdp_tx_iptunnel_user.c | 10 +++++-----
samples/bpf/xdpsock_ctrl_proc.c | 2 +-
samples/bpf/xdpsock_user.c | 10 +++++-----
samples/bpf/xsk_fwd.c | 4 ++--
11 files changed, 40 insertions(+), 41 deletions(-)
diff --git a/samples/bpf/xdp1_user.c b/samples/bpf/xdp1_user.c
index 8675fa5273df..3ec8ad9c1750 100644
--- a/samples/bpf/xdp1_user.c
+++ b/samples/bpf/xdp1_user.c
@@ -26,12 +26,12 @@ static void int_exit(int sig)
{
__u32 curr_prog_id = 0;
- if (bpf_get_link_xdp_id(ifindex, &curr_prog_id, xdp_flags)) {
- printf("bpf_get_link_xdp_id failed\n");
+ if (bpf_xdp_query_id(ifindex, xdp_flags, &curr_prog_id)) {
+ printf("bpf_xdp_query_id failed\n");
exit(1);
}
if (prog_id == curr_prog_id)
- bpf_set_link_xdp_fd(ifindex, -1, xdp_flags);
+ bpf_xdp_detach(ifindex, xdp_flags, NULL);
else if (!curr_prog_id)
printf("couldn't find a prog id on a given interface\n");
else
@@ -143,7 +143,7 @@ int main(int argc, char **argv)
signal(SIGINT, int_exit);
signal(SIGTERM, int_exit);
- if (bpf_set_link_xdp_fd(ifindex, prog_fd, xdp_flags) < 0) {
+ if (bpf_xdp_attach(ifindex, prog_fd, xdp_flags, NULL) < 0) {
printf("link set xdp fd failed\n");
return 1;
}
diff --git a/samples/bpf/xdp_adjust_tail_user.c b/samples/bpf/xdp_adjust_tail_user.c
index a70b094c8ec5..6c61d5f570fb 100644
--- a/samples/bpf/xdp_adjust_tail_user.c
+++ b/samples/bpf/xdp_adjust_tail_user.c
@@ -34,12 +34,12 @@ static void int_exit(int sig)
__u32 curr_prog_id = 0;
if (ifindex > -1) {
- if (bpf_get_link_xdp_id(ifindex, &curr_prog_id, xdp_flags)) {
- printf("bpf_get_link_xdp_id failed\n");
+ if (bpf_xdp_query_id(ifindex, xdp_flags, &curr_prog_id)) {
+ printf("bpf_xdp_query_id failed\n");
exit(1);
}
if (prog_id == curr_prog_id)
- bpf_set_link_xdp_fd(ifindex, -1, xdp_flags);
+ bpf_xdp_detach(ifindex, xdp_flags, NULL);
else if (!curr_prog_id)
printf("couldn't find a prog id on a given iface\n");
else
@@ -173,7 +173,7 @@ int main(int argc, char **argv)
signal(SIGINT, int_exit);
signal(SIGTERM, int_exit);
- if (bpf_set_link_xdp_fd(ifindex, prog_fd, xdp_flags) < 0) {
+ if (bpf_xdp_attach(ifindex, prog_fd, xdp_flags, NULL) < 0) {
printf("link set xdp fd failed\n");
return 1;
}
diff --git a/samples/bpf/xdp_fwd_user.c b/samples/bpf/xdp_fwd_user.c
index 4ad896782f77..79ccd9891924 100644
--- a/samples/bpf/xdp_fwd_user.c
+++ b/samples/bpf/xdp_fwd_user.c
@@ -33,7 +33,7 @@ static int do_attach(int idx, int prog_fd, int map_fd, const char *name)
{
int err;
- err = bpf_set_link_xdp_fd(idx, prog_fd, xdp_flags);
+ err = bpf_xdp_attach(idx, prog_fd, xdp_flags, NULL);
if (err < 0) {
printf("ERROR: failed to attach program to %s\n", name);
return err;
@@ -51,7 +51,7 @@ static int do_detach(int idx, const char *name)
{
int err;
- err = bpf_set_link_xdp_fd(idx, -1, xdp_flags);
+ err = bpf_xdp_detach(idx, xdp_flags, NULL);
if (err < 0)
printf("ERROR: failed to detach program from %s\n", name);
diff --git a/samples/bpf/xdp_router_ipv4_user.c b/samples/bpf/xdp_router_ipv4_user.c
index cfaf7e50e431..2d565ba54b8c 100644
--- a/samples/bpf/xdp_router_ipv4_user.c
+++ b/samples/bpf/xdp_router_ipv4_user.c
@@ -43,13 +43,13 @@ static void int_exit(int sig)
int i = 0;
for (i = 0; i < total_ifindex; i++) {
- if (bpf_get_link_xdp_id(ifindex_list[i], &prog_id, flags)) {
- printf("bpf_get_link_xdp_id on iface %d failed\n",
+ if (bpf_xdp_query_id(ifindex_list[i], flags, &prog_id)) {
+ printf("bpf_xdp_query_id on iface %d failed\n",
ifindex_list[i]);
exit(1);
}
if (prog_id_list[i] == prog_id)
- bpf_set_link_xdp_fd(ifindex_list[i], -1, flags);
+ bpf_xdp_detach(ifindex_list[i], flags, NULL);
else if (!prog_id)
printf("couldn't find a prog id on iface %d\n",
ifindex_list[i]);
@@ -716,12 +716,12 @@ int main(int ac, char **argv)
}
prog_id_list = (__u32 *)calloc(total_ifindex, sizeof(__u32 *));
for (i = 0; i < total_ifindex; i++) {
- if (bpf_set_link_xdp_fd(ifindex_list[i], prog_fd, flags) < 0) {
+ if (bpf_xdp_attach(ifindex_list[i], prog_fd, flags, NULL) < 0) {
printf("link set xdp fd failed\n");
int recovery_index = i;
for (i = 0; i < recovery_index; i++)
- bpf_set_link_xdp_fd(ifindex_list[i], -1, flags);
+ bpf_xdp_detach(ifindex_list[i], flags, NULL);
return 1;
}
diff --git a/samples/bpf/xdp_rxq_info_user.c b/samples/bpf/xdp_rxq_info_user.c
index 4033f345aa29..fb2532d13aac 100644
--- a/samples/bpf/xdp_rxq_info_user.c
+++ b/samples/bpf/xdp_rxq_info_user.c
@@ -62,15 +62,15 @@ static void int_exit(int sig)
__u32 curr_prog_id = 0;
if (ifindex > -1) {
- if (bpf_get_link_xdp_id(ifindex, &curr_prog_id, xdp_flags)) {
- printf("bpf_get_link_xdp_id failed\n");
+ if (bpf_xdp_query_id(ifindex, xdp_flags, &curr_prog_id)) {
+ printf("bpf_xdp_query_id failed\n");
exit(EXIT_FAIL);
}
if (prog_id == curr_prog_id) {
fprintf(stderr,
"Interrupted: Removing XDP program on ifindex:%d device:%s\n",
ifindex, ifname);
- bpf_set_link_xdp_fd(ifindex, -1, xdp_flags);
+ bpf_xdp_detach(ifindex, xdp_flags, NULL);
} else if (!curr_prog_id) {
printf("couldn't find a prog id on a given iface\n");
} else {
@@ -582,7 +582,7 @@ int main(int argc, char **argv)
signal(SIGINT, int_exit);
signal(SIGTERM, int_exit);
- if (bpf_set_link_xdp_fd(ifindex, prog_fd, xdp_flags) < 0) {
+ if (bpf_xdp_attach(ifindex, prog_fd, xdp_flags, NULL) < 0) {
fprintf(stderr, "link set xdp fd failed\n");
return EXIT_FAIL_XDP;
}
diff --git a/samples/bpf/xdp_sample_pkts_user.c b/samples/bpf/xdp_sample_pkts_user.c
index 587eacb49103..0a2b3e997aed 100644
--- a/samples/bpf/xdp_sample_pkts_user.c
+++ b/samples/bpf/xdp_sample_pkts_user.c
@@ -30,7 +30,7 @@ static int do_attach(int idx, int fd, const char *name)
__u32 info_len = sizeof(info);
int err;
- err = bpf_set_link_xdp_fd(idx, fd, xdp_flags);
+ err = bpf_xdp_attach(idx, fd, xdp_flags, NULL);
if (err < 0) {
printf("ERROR: failed to attach program to %s\n", name);
return err;
@@ -51,13 +51,13 @@ static int do_detach(int idx, const char *name)
__u32 curr_prog_id = 0;
int err = 0;
- err = bpf_get_link_xdp_id(idx, &curr_prog_id, xdp_flags);
+ err = bpf_xdp_query_id(idx, xdp_flags, &curr_prog_id);
if (err) {
- printf("bpf_get_link_xdp_id failed\n");
+ printf("bpf_xdp_query_id failed\n");
return err;
}
if (prog_id == curr_prog_id) {
- err = bpf_set_link_xdp_fd(idx, -1, xdp_flags);
+ err = bpf_xdp_detach(idx, xdp_flags, NULL);
if (err < 0)
printf("ERROR: failed to detach prog from %s\n", name);
} else if (!curr_prog_id) {
diff --git a/samples/bpf/xdp_sample_user.c b/samples/bpf/xdp_sample_user.c
index 8740838e7767..ae70a7943d85 100644
--- a/samples/bpf/xdp_sample_user.c
+++ b/samples/bpf/xdp_sample_user.c
@@ -1265,7 +1265,7 @@ static int __sample_remove_xdp(int ifindex, __u32 prog_id, int xdp_flags)
int ret;
if (prog_id) {
- ret = bpf_get_link_xdp_id(ifindex, &cur_prog_id, xdp_flags);
+ ret = bpf_xdp_query_id(ifindex, xdp_flags, &cur_prog_id);
if (ret < 0)
return -errno;
@@ -1278,7 +1278,7 @@ static int __sample_remove_xdp(int ifindex, __u32 prog_id, int xdp_flags)
}
}
- return bpf_set_link_xdp_fd(ifindex, -1, xdp_flags);
+ return bpf_xdp_detach(ifindex, xdp_flags, NULL);
}
int sample_install_xdp(struct bpf_program *xdp_prog, int ifindex, bool generic,
@@ -1295,8 +1295,7 @@ int sample_install_xdp(struct bpf_program *xdp_prog, int ifindex, bool generic,
xdp_flags |= !force ? XDP_FLAGS_UPDATE_IF_NOEXIST : 0;
xdp_flags |= generic ? XDP_FLAGS_SKB_MODE : XDP_FLAGS_DRV_MODE;
- ret = bpf_set_link_xdp_fd(ifindex, bpf_program__fd(xdp_prog),
- xdp_flags);
+ ret = bpf_xdp_attach(ifindex, bpf_program__fd(xdp_prog), xdp_flags, NULL);
if (ret < 0) {
ret = -errno;
fprintf(stderr,
@@ -1308,7 +1307,7 @@ int sample_install_xdp(struct bpf_program *xdp_prog, int ifindex, bool generic,
return ret;
}
- ret = bpf_get_link_xdp_id(ifindex, &prog_id, xdp_flags);
+ ret = bpf_xdp_query_id(ifindex, xdp_flags, &prog_id);
if (ret < 0) {
ret = -errno;
fprintf(stderr,
diff --git a/samples/bpf/xdp_tx_iptunnel_user.c b/samples/bpf/xdp_tx_iptunnel_user.c
index 1d4f305d02aa..7370c03c96fc 100644
--- a/samples/bpf/xdp_tx_iptunnel_user.c
+++ b/samples/bpf/xdp_tx_iptunnel_user.c
@@ -32,12 +32,12 @@ static void int_exit(int sig)
__u32 curr_prog_id = 0;
if (ifindex > -1) {
- if (bpf_get_link_xdp_id(ifindex, &curr_prog_id, xdp_flags)) {
- printf("bpf_get_link_xdp_id failed\n");
+ if (bpf_xdp_query_id(ifindex, xdp_flags, &curr_prog_id)) {
+ printf("bpf_xdp_query_id failed\n");
exit(1);
}
if (prog_id == curr_prog_id)
- bpf_set_link_xdp_fd(ifindex, -1, xdp_flags);
+ bpf_xdp_detach(ifindex, xdp_flags, NULL);
else if (!curr_prog_id)
printf("couldn't find a prog id on a given iface\n");
else
@@ -288,7 +288,7 @@ int main(int argc, char **argv)
}
}
- if (bpf_set_link_xdp_fd(ifindex, prog_fd, xdp_flags) < 0) {
+ if (bpf_xdp_attach(ifindex, prog_fd, xdp_flags, NULL) < 0) {
printf("link set xdp fd failed\n");
return 1;
}
@@ -302,7 +302,7 @@ int main(int argc, char **argv)
poll_stats(kill_after_s);
- bpf_set_link_xdp_fd(ifindex, -1, xdp_flags);
+ bpf_xdp_detach(ifindex, xdp_flags, NULL);
return 0;
}
diff --git a/samples/bpf/xdpsock_ctrl_proc.c b/samples/bpf/xdpsock_ctrl_proc.c
index cc4408797ab7..28b5f2a9fa08 100644
--- a/samples/bpf/xdpsock_ctrl_proc.c
+++ b/samples/bpf/xdpsock_ctrl_proc.c
@@ -173,7 +173,7 @@ main(int argc, char **argv)
unlink(SOCKET_NAME);
/* Unset fd for given ifindex */
- err = bpf_set_link_xdp_fd(ifindex, -1, 0);
+ err = bpf_xdp_detach(ifindex, 0, NULL);
if (err) {
fprintf(stderr, "Error when unsetting bpf prog_fd for ifindex(%d)\n", ifindex);
return err;
diff --git a/samples/bpf/xdpsock_user.c b/samples/bpf/xdpsock_user.c
index aa50864e4415..19288a2bbc75 100644
--- a/samples/bpf/xdpsock_user.c
+++ b/samples/bpf/xdpsock_user.c
@@ -571,13 +571,13 @@ static void remove_xdp_program(void)
{
u32 curr_prog_id = 0;
- if (bpf_get_link_xdp_id(opt_ifindex, &curr_prog_id, opt_xdp_flags)) {
- printf("bpf_get_link_xdp_id failed\n");
+ if (bpf_xdp_query_id(opt_ifindex, opt_xdp_flags, &curr_prog_id)) {
+ printf("bpf_xdp_query_id failed\n");
exit(EXIT_FAILURE);
}
if (prog_id == curr_prog_id)
- bpf_set_link_xdp_fd(opt_ifindex, -1, opt_xdp_flags);
+ bpf_xdp_detach(opt_ifindex, opt_xdp_flags, NULL);
else if (!curr_prog_id)
printf("couldn't find a prog id on a given interface\n");
else
@@ -1027,7 +1027,7 @@ static struct xsk_socket_info *xsk_configure_socket(struct xsk_umem_info *umem,
if (ret)
exit_with_error(-ret);
- ret = bpf_get_link_xdp_id(opt_ifindex, &prog_id, opt_xdp_flags);
+ ret = bpf_xdp_query_id(opt_ifindex, opt_xdp_flags, &prog_id);
if (ret)
exit_with_error(-ret);
@@ -1760,7 +1760,7 @@ static void load_xdp_program(char **argv, struct bpf_object **obj)
exit(EXIT_FAILURE);
}
- if (bpf_set_link_xdp_fd(opt_ifindex, prog_fd, opt_xdp_flags) < 0) {
+ if (bpf_xdp_attach(opt_ifindex, prog_fd, opt_xdp_flags, NULL) < 0) {
fprintf(stderr, "ERROR: link set xdp fd failed\n");
exit(EXIT_FAILURE);
}
diff --git a/samples/bpf/xsk_fwd.c b/samples/bpf/xsk_fwd.c
index 52e7c4ffd228..2220509588a0 100644
--- a/samples/bpf/xsk_fwd.c
+++ b/samples/bpf/xsk_fwd.c
@@ -974,8 +974,8 @@ static void remove_xdp_program(void)
int i;
for (i = 0 ; i < n_ports; i++)
- bpf_set_link_xdp_fd(if_nametoindex(port_params[i].iface), -1,
- port_params[i].xsk_cfg.xdp_flags);
+ bpf_xdp_detach(if_nametoindex(port_params[i].iface),
+ port_params[i].xsk_cfg.xdp_flags, NULL);
}
int main(int argc, char **argv)
--
2.30.2
^ permalink raw reply related [flat|nested] 6+ messages in thread
* Re: [PATCH bpf-next 1/4] libbpf: streamline low-level XDP APIs
2022-01-20 6:14 ` [PATCH bpf-next 1/4] libbpf: streamline low-level " Andrii Nakryiko
@ 2022-01-20 12:00 ` Toke Høiland-Jørgensen
0 siblings, 0 replies; 6+ messages in thread
From: Toke Høiland-Jørgensen @ 2022-01-20 12:00 UTC (permalink / raw)
To: Andrii Nakryiko, bpf, ast, daniel; +Cc: andrii, kernel-team
Andrii Nakryiko <andrii@kernel.org> writes:
> Introduce 4 new netlink-based XDP APIs for attaching, detaching, and
> querying XDP programs:
> - bpf_xdp_attach;
> - bpf_xdp_detach;
> - bpf_xdp_query;
> - bpf_xdp_query_id.
>
> These APIs replace bpf_set_link_xdp_fd, bpf_set_link_xdp_fd_opts,
> bpf_get_link_xdp_id, and bpf_get_link_xdp_info APIs ([0]). The latter
> don't follow a consistent naming pattern and some of them use
> non-extensible approaches (e.g., struct xdp_link_info which can't be
> modified without breaking libbpf ABI).
>
> The approach I took with these low-level XDP APIs is similar to what we
> did with low-level TC APIs. There is a nice duality of bpf_tc_attach vs
> bpf_xdp_attach, and so on. I left bpf_xdp_attach() to support detaching
> when -1 is specified for prog_fd for generality and convenience, but
> bpf_xdp_detach() is preferred due to clearer naming and associated
> semantics. Both bpf_xdp_attach() and bpf_xdp_detach() accept the same
> opts struct allowing to specify expected old_prog_fd.
>
> While doing the refactoring, I noticed that old APIs require users to
> specify opts with old_fd == -1 to declare "don't care about already
> attached XDP prog fd" condition. Otherwise, FD 0 is assumed, which is
> essentially never an intended behavior. So I made this behavior
> consistent with other kernel and libbpf APIs, in which zero FD means "no
> FD". This seems to be more in line with the latest thinking in BPF land
> and should cause less user confusion, hopefully.
>
> For querying, I left two APIs, both more generic bpf_xdp_query()
> allowing to query multiple IDs and attach mode, but also
> a specialization of it, bpf_xdp_query_id(), which returns only requested
> prog_id. Uses of prog_id returning bpf_get_link_xdp_id() were so
> prevalent across selftests and samples, that it seemed a very common use
> case and using bpf_xdp_query() for doing it felt very cumbersome with
> a highly branches if/else chain based on flags and attach mode.
>
> Old APIs are scheduled for deprecation in libbpf 0.8 release.
>
> [0] Closes: https://github.com/libbpf/libbpf/issues/309
>
> Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Acked-by: Toke Høiland-Jørgensen <toke@redhat.com>
^ permalink raw reply [flat|nested] 6+ messages in thread
end of thread, other threads:[~2022-01-20 12:00 UTC | newest]
Thread overview: 6+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2022-01-20 6:14 [PATCH bpf-next 0/4] libbpf: streamline netlink-based XDP APIs Andrii Nakryiko
2022-01-20 6:14 ` [PATCH bpf-next 1/4] libbpf: streamline low-level " Andrii Nakryiko
2022-01-20 12:00 ` Toke Høiland-Jørgensen
2022-01-20 6:14 ` [PATCH bpf-next 2/4] bpftool: use new API for attaching XDP program Andrii Nakryiko
2022-01-20 6:14 ` [PATCH bpf-next 3/4] selftests/bpf: switch to new libbpf XDP APIs Andrii Nakryiko
2022-01-20 6:14 ` [PATCH bpf-next 4/4] samples/bpf: adapt samples/bpf to bpf_xdp_xxx() APIs Andrii Nakryiko
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.