* [PATCH bpf-next v5 0/8] xdp: Avoid unloading xdp prog not attached by sample
@ 2019-02-01 0:19 Maciej Fijalkowski
2019-02-01 0:19 ` [PATCH bpf-next v5 1/8] libbpf: Add a helper for retrieving a map fd for a given name Maciej Fijalkowski
` (8 more replies)
0 siblings, 9 replies; 22+ messages in thread
From: Maciej Fijalkowski @ 2019-02-01 0:19 UTC (permalink / raw)
To: daniel, ast; +Cc: netdev, jakub.kicinski, brouer, john.fastabend
Hi!
This patchset tries to address the situation where:
* user loads a particular xdp sample application that does stats polling
* user loads another sample application on the same interface
* then, user sends SIGINT/SIGTERM to the app that was attached as a first one
* second application ends up with an unloaded xdp program
1st patch contains a helper libbpf function for getting the map fd by a
given map name.
In patch 2 Jesper removes the read_trace_pipe usage from xdp_redirect_cpu which
was a blocker for converting this sample to libbpf usage.
3rd patch updates a bunch of xdp samples to make the use of libbpf.
Patch 4 adjusts RLIMIT_MEMLOCK for two samples touched in this patchset.
In patch 5 extack messages are added for cases where dev_change_xdp_fd returns
with an error so user has an idea what was the reason for not attaching the
xdp program onto interface.
Patch 6 makes the samples behavior similar to what iproute2 does when loading
xdp prog - the "force" flag is introduced.
Patch 7 introduces the libbpf function that will query the driver from
userspace about the currently attached xdp prog id.
Use it in samples that do polling by checking the prog id in signal handler
and comparing it with previously stored one which is the scope of patch 8.
Thanks!
v1->v2:
* add a libbpf helper for getting a prog via relative index
* include xdp_redirect_cpu into conversion
v2->v3: mostly addressing Daniel's/Jesper's comments
* get rid of the helper from v1->v2
* feed the xdp_redirect_cpu with program name instead of number
v3->v4:
* fix help message in xdp_sample_pkts
v4->v5:
* in get_link_xdp_fd, assign prog_id only when libbpf_nl_get_link returned
with 0
* add extack messages in dev_change_xdp_fd
* check the return value of bpf_get_link_xdp_id when exiting from sample progs
Jesper Dangaard Brouer (1):
samples/bpf: xdp_redirect_cpu have not need for read_trace_pipe
Maciej Fijalkowski (7):
libbpf: Add a helper for retrieving a map fd for a given name
samples/bpf: Convert XDP samples to libbpf usage
samples/bpf: Extend RLIMIT_MEMLOCK for xdp_{sample_pkts, router_ipv4}
xdp: Provide extack messages when prog attachment failed
samples/bpf: Add a "force" flag to XDP samples
libbpf: Add a support for getting xdp prog id on ifindex
samples/bpf: Check the prog id before exiting
net/core/dev.c | 12 ++-
samples/bpf/Makefile | 8 +-
samples/bpf/xdp1_user.c | 34 ++++++-
samples/bpf/xdp_adjust_tail_user.c | 38 +++++--
samples/bpf/xdp_redirect_cpu_user.c | 196 +++++++++++++++++++++++++-----------
samples/bpf/xdp_redirect_map_user.c | 106 +++++++++++++++----
samples/bpf/xdp_redirect_user.c | 103 ++++++++++++++++---
samples/bpf/xdp_router_ipv4_user.c | 179 +++++++++++++++++++++++---------
samples/bpf/xdp_rxq_info_user.c | 41 ++++++--
samples/bpf/xdp_sample_pkts_user.c | 81 ++++++++++++---
samples/bpf/xdp_tx_iptunnel_user.c | 71 ++++++++++---
samples/bpf/xdpsock_user.c | 30 +++++-
tools/lib/bpf/libbpf.c | 6 ++
tools/lib/bpf/libbpf.h | 4 +
tools/lib/bpf/libbpf.map | 2 +
tools/lib/bpf/netlink.c | 85 ++++++++++++++++
16 files changed, 796 insertions(+), 200 deletions(-)
--
2.16.1
^ permalink raw reply [flat|nested] 22+ messages in thread
* [PATCH bpf-next v5 1/8] libbpf: Add a helper for retrieving a map fd for a given name
2019-02-01 0:19 [PATCH bpf-next v5 0/8] xdp: Avoid unloading xdp prog not attached by sample Maciej Fijalkowski
@ 2019-02-01 0:19 ` Maciej Fijalkowski
2019-02-01 0:19 ` [PATCH bpf-next v5 2/8] samples/bpf: xdp_redirect_cpu have not need for read_trace_pipe Maciej Fijalkowski
` (7 subsequent siblings)
8 siblings, 0 replies; 22+ messages in thread
From: Maciej Fijalkowski @ 2019-02-01 0:19 UTC (permalink / raw)
To: daniel, ast; +Cc: netdev, jakub.kicinski, brouer, john.fastabend
XDP samples are mostly cooperating with eBPF maps through their file
descriptors. In case of a eBPF program that contains multiple maps it
might be tiresome to iterate through them and call bpf_map__fd for each
one. Add a helper mostly based on bpf_object__find_map_by_name, but
instead of returning the struct bpf_map pointer, return map fd.
Suggested-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Signed-off-by: Maciej Fijalkowski <maciej.fijalkowski@intel.com>
Reviewed-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Acked-by: John Fastabend <john.fastabend@gmail.com>
---
tools/lib/bpf/libbpf.c | 6 ++++++
tools/lib/bpf/libbpf.h | 3 +++
tools/lib/bpf/libbpf.map | 1 +
3 files changed, 10 insertions(+)
diff --git a/tools/lib/bpf/libbpf.c b/tools/lib/bpf/libbpf.c
index 2ccde17957e6..03bc01ca2577 100644
--- a/tools/lib/bpf/libbpf.c
+++ b/tools/lib/bpf/libbpf.c
@@ -2884,6 +2884,12 @@ bpf_object__find_map_by_name(struct bpf_object *obj, const char *name)
return NULL;
}
+int
+bpf_object__find_map_fd_by_name(struct bpf_object *obj, const char *name)
+{
+ return bpf_map__fd(bpf_object__find_map_by_name(obj, name));
+}
+
struct bpf_map *
bpf_object__find_map_by_offset(struct bpf_object *obj, size_t offset)
{
diff --git a/tools/lib/bpf/libbpf.h b/tools/lib/bpf/libbpf.h
index 62ae6cb93da1..931be6f3408c 100644
--- a/tools/lib/bpf/libbpf.h
+++ b/tools/lib/bpf/libbpf.h
@@ -264,6 +264,9 @@ struct bpf_map;
LIBBPF_API struct bpf_map *
bpf_object__find_map_by_name(struct bpf_object *obj, const char *name);
+LIBBPF_API int
+bpf_object__find_map_fd_by_name(struct bpf_object *obj, const char *name);
+
/*
* Get bpf_map through the offset of corresponding struct bpf_map_def
* in the BPF object file.
diff --git a/tools/lib/bpf/libbpf.map b/tools/lib/bpf/libbpf.map
index 266bc95d0142..b183c6c3b990 100644
--- a/tools/lib/bpf/libbpf.map
+++ b/tools/lib/bpf/libbpf.map
@@ -130,4 +130,5 @@ LIBBPF_0.0.2 {
bpf_probe_helper;
bpf_probe_map_type;
bpf_probe_prog_type;
+ bpf_object__find_map_fd_by_name;
} LIBBPF_0.0.1;
--
2.16.1
^ permalink raw reply related [flat|nested] 22+ messages in thread
* [PATCH bpf-next v5 2/8] samples/bpf: xdp_redirect_cpu have not need for read_trace_pipe
2019-02-01 0:19 [PATCH bpf-next v5 0/8] xdp: Avoid unloading xdp prog not attached by sample Maciej Fijalkowski
2019-02-01 0:19 ` [PATCH bpf-next v5 1/8] libbpf: Add a helper for retrieving a map fd for a given name Maciej Fijalkowski
@ 2019-02-01 0:19 ` Maciej Fijalkowski
2019-02-01 0:19 ` [PATCH bpf-next v5 3/8] samples/bpf: Convert XDP samples to libbpf usage Maciej Fijalkowski
` (6 subsequent siblings)
8 siblings, 0 replies; 22+ messages in thread
From: Maciej Fijalkowski @ 2019-02-01 0:19 UTC (permalink / raw)
To: daniel, ast; +Cc: netdev, jakub.kicinski, brouer, john.fastabend
From: Jesper Dangaard Brouer <brouer@redhat.com>
The sample xdp_redirect_cpu is not using helper bpf_trace_printk.
Thus it makes no sense that the --debug option us reading
from /sys/kernel/debug/tracing/trace_pipe via read_trace_pipe.
Simply remove it.
Signed-off-by: Jesper Dangaard Brouer <brouer@redhat.com>
Acked-by: John Fastabend <john.fastabend@gmail.com>
---
samples/bpf/xdp_redirect_cpu_user.c | 10 ----------
1 file changed, 10 deletions(-)
diff --git a/samples/bpf/xdp_redirect_cpu_user.c b/samples/bpf/xdp_redirect_cpu_user.c
index 2d23054aaccf..f141e752ca0a 100644
--- a/samples/bpf/xdp_redirect_cpu_user.c
+++ b/samples/bpf/xdp_redirect_cpu_user.c
@@ -51,7 +51,6 @@ static const struct option long_options[] = {
{"help", no_argument, NULL, 'h' },
{"dev", required_argument, NULL, 'd' },
{"skb-mode", no_argument, NULL, 'S' },
- {"debug", no_argument, NULL, 'D' },
{"sec", required_argument, NULL, 's' },
{"prognum", required_argument, NULL, 'p' },
{"qsize", required_argument, NULL, 'q' },
@@ -563,7 +562,6 @@ int main(int argc, char **argv)
bool use_separators = true;
bool stress_mode = false;
char filename[256];
- bool debug = false;
int added_cpus = 0;
int longindex = 0;
int interval = 2;
@@ -624,9 +622,6 @@ int main(int argc, char **argv)
case 'S':
xdp_flags |= XDP_FLAGS_SKB_MODE;
break;
- case 'D':
- debug = true;
- break;
case 'x':
stress_mode = true;
break;
@@ -688,11 +683,6 @@ int main(int argc, char **argv)
return EXIT_FAIL_XDP;
}
- if (debug) {
- printf("Debug-mode reading trace pipe (fix #define DEBUG)\n");
- read_trace_pipe();
- }
-
stats_poll(interval, use_separators, prog_num, stress_mode);
return EXIT_OK;
}
--
2.16.1
^ permalink raw reply related [flat|nested] 22+ messages in thread
* [PATCH bpf-next v5 3/8] samples/bpf: Convert XDP samples to libbpf usage
2019-02-01 0:19 [PATCH bpf-next v5 0/8] xdp: Avoid unloading xdp prog not attached by sample Maciej Fijalkowski
2019-02-01 0:19 ` [PATCH bpf-next v5 1/8] libbpf: Add a helper for retrieving a map fd for a given name Maciej Fijalkowski
2019-02-01 0:19 ` [PATCH bpf-next v5 2/8] samples/bpf: xdp_redirect_cpu have not need for read_trace_pipe Maciej Fijalkowski
@ 2019-02-01 0:19 ` Maciej Fijalkowski
2019-02-01 0:19 ` [PATCH bpf-next v5 4/8] samples/bpf: Extend RLIMIT_MEMLOCK for xdp_{sample_pkts, router_ipv4} Maciej Fijalkowski
` (5 subsequent siblings)
8 siblings, 0 replies; 22+ messages in thread
From: Maciej Fijalkowski @ 2019-02-01 0:19 UTC (permalink / raw)
To: daniel, ast; +Cc: netdev, jakub.kicinski, brouer, john.fastabend
Some of XDP samples that are attaching the bpf program to the interface
via libbpf's bpf_set_link_xdp_fd are still using the bpf_load.c for
loading and manipulating the ebpf program and maps. Convert them to do
this through libbpf usage and remove bpf_load from the picture.
While at it remove what looks like debug leftover in
xdp_redirect_map_user.c
In xdp_redirect_cpu, change the way that the program to be loaded onto
interface is chosen - user now needs to pass the program's section name
instead of the relative number. In case of typo print out the section
names to choose from.
Signed-off-by: Maciej Fijalkowski <maciej.fijalkowski@intel.com>
Reviewed-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Acked-by: Jesper Dangaard Brouer <brouer@redhat.com>
---
samples/bpf/Makefile | 8 +-
samples/bpf/xdp_redirect_cpu_user.c | 145 +++++++++++++++++++++++++-----------
samples/bpf/xdp_redirect_map_user.c | 47 ++++++++----
samples/bpf/xdp_redirect_user.c | 44 ++++++++---
samples/bpf/xdp_router_ipv4_user.c | 75 +++++++++++++------
samples/bpf/xdp_tx_iptunnel_user.c | 37 ++++++---
6 files changed, 253 insertions(+), 103 deletions(-)
diff --git a/samples/bpf/Makefile b/samples/bpf/Makefile
index db1a91dfa702..a0ef7eddd0b3 100644
--- a/samples/bpf/Makefile
+++ b/samples/bpf/Makefile
@@ -87,18 +87,18 @@ test_cgrp2_sock2-objs := bpf_load.o test_cgrp2_sock2.o
xdp1-objs := xdp1_user.o
# reuse xdp1 source intentionally
xdp2-objs := xdp1_user.o
-xdp_router_ipv4-objs := bpf_load.o xdp_router_ipv4_user.o
+xdp_router_ipv4-objs := xdp_router_ipv4_user.o
test_current_task_under_cgroup-objs := bpf_load.o $(CGROUP_HELPERS) \
test_current_task_under_cgroup_user.o
trace_event-objs := bpf_load.o trace_event_user.o $(TRACE_HELPERS)
sampleip-objs := bpf_load.o sampleip_user.o $(TRACE_HELPERS)
tc_l2_redirect-objs := bpf_load.o tc_l2_redirect_user.o
lwt_len_hist-objs := bpf_load.o lwt_len_hist_user.o
-xdp_tx_iptunnel-objs := bpf_load.o xdp_tx_iptunnel_user.o
+xdp_tx_iptunnel-objs := xdp_tx_iptunnel_user.o
test_map_in_map-objs := bpf_load.o test_map_in_map_user.o
per_socket_stats_example-objs := cookie_uid_helper_example.o
-xdp_redirect-objs := bpf_load.o xdp_redirect_user.o
-xdp_redirect_map-objs := bpf_load.o xdp_redirect_map_user.o
+xdp_redirect-objs := xdp_redirect_user.o
+xdp_redirect_map-objs := xdp_redirect_map_user.o
xdp_redirect_cpu-objs := bpf_load.o xdp_redirect_cpu_user.o
xdp_monitor-objs := bpf_load.o xdp_monitor_user.o
xdp_rxq_info-objs := xdp_rxq_info_user.o
diff --git a/samples/bpf/xdp_redirect_cpu_user.c b/samples/bpf/xdp_redirect_cpu_user.c
index f141e752ca0a..8645ddc2da0e 100644
--- a/samples/bpf/xdp_redirect_cpu_user.c
+++ b/samples/bpf/xdp_redirect_cpu_user.c
@@ -24,12 +24,8 @@ static const char *__doc__ =
/* How many xdp_progs are defined in _kern.c */
#define MAX_PROG 6
-/* Wanted to get rid of bpf_load.h and fake-"libbpf.h" (and instead
- * use bpf/libbpf.h), but cannot as (currently) needed for XDP
- * attaching to a device via bpf_set_link_xdp_fd()
- */
#include <bpf/bpf.h>
-#include "bpf_load.h"
+#include "bpf/libbpf.h"
#include "bpf_util.h"
@@ -38,6 +34,15 @@ static char ifname_buf[IF_NAMESIZE];
static char *ifname;
static __u32 xdp_flags;
+static int cpu_map_fd;
+static int rx_cnt_map_fd;
+static int redirect_err_cnt_map_fd;
+static int cpumap_enqueue_cnt_map_fd;
+static int cpumap_kthread_cnt_map_fd;
+static int cpus_available_map_fd;
+static int cpus_count_map_fd;
+static int cpus_iterator_map_fd;
+static int exception_cnt_map_fd;
/* Exit return codes */
#define EXIT_OK 0
@@ -52,7 +57,7 @@ static const struct option long_options[] = {
{"dev", required_argument, NULL, 'd' },
{"skb-mode", no_argument, NULL, 'S' },
{"sec", required_argument, NULL, 's' },
- {"prognum", required_argument, NULL, 'p' },
+ {"progname", required_argument, NULL, 'p' },
{"qsize", required_argument, NULL, 'q' },
{"cpu", required_argument, NULL, 'c' },
{"stress-mode", no_argument, NULL, 'x' },
@@ -70,7 +75,17 @@ static void int_exit(int sig)
exit(EXIT_OK);
}
-static void usage(char *argv[])
+static void print_avail_progs(struct bpf_object *obj)
+{
+ struct bpf_program *pos;
+
+ bpf_object__for_each_program(pos, obj) {
+ if (bpf_program__is_xdp(pos))
+ printf(" %s\n", bpf_program__title(pos, false));
+ }
+}
+
+static void usage(char *argv[], struct bpf_object *obj)
{
int i;
@@ -88,6 +103,8 @@ static void usage(char *argv[])
long_options[i].val);
printf("\n");
}
+ printf("\n Programs to be used for --progname:\n");
+ print_avail_progs(obj);
printf("\n");
}
@@ -262,7 +279,7 @@ static __u64 calc_errs_pps(struct datarec *r,
static void stats_print(struct stats_record *stats_rec,
struct stats_record *stats_prev,
- int prog_num)
+ char *prog_name)
{
unsigned int nr_cpus = bpf_num_possible_cpus();
double pps = 0, drop = 0, err = 0;
@@ -272,7 +289,7 @@ static void stats_print(struct stats_record *stats_rec,
int i;
/* Header */
- printf("Running XDP/eBPF prog_num:%d\n", prog_num);
+ printf("Running XDP/eBPF prog_name:%s\n", prog_name);
printf("%-15s %-7s %-14s %-11s %-9s\n",
"XDP-cpumap", "CPU:to", "pps", "drop-pps", "extra-info");
@@ -423,20 +440,20 @@ static void stats_collect(struct stats_record *rec)
{
int fd, i;
- fd = map_fd[1]; /* map: rx_cnt */
+ fd = rx_cnt_map_fd;
map_collect_percpu(fd, 0, &rec->rx_cnt);
- fd = map_fd[2]; /* map: redirect_err_cnt */
+ fd = redirect_err_cnt_map_fd;
map_collect_percpu(fd, 1, &rec->redir_err);
- fd = map_fd[3]; /* map: cpumap_enqueue_cnt */
+ fd = cpumap_enqueue_cnt_map_fd;
for (i = 0; i < MAX_CPUS; i++)
map_collect_percpu(fd, i, &rec->enq[i]);
- fd = map_fd[4]; /* map: cpumap_kthread_cnt */
+ fd = cpumap_kthread_cnt_map_fd;
map_collect_percpu(fd, 0, &rec->kthread);
- fd = map_fd[8]; /* map: exception_cnt */
+ fd = exception_cnt_map_fd;
map_collect_percpu(fd, 0, &rec->exception);
}
@@ -461,7 +478,7 @@ static int create_cpu_entry(__u32 cpu, __u32 queue_size,
/* Add a CPU entry to cpumap, as this allocate a cpu entry in
* the kernel for the cpu.
*/
- ret = bpf_map_update_elem(map_fd[0], &cpu, &queue_size, 0);
+ ret = bpf_map_update_elem(cpu_map_fd, &cpu, &queue_size, 0);
if (ret) {
fprintf(stderr, "Create CPU entry failed (err:%d)\n", ret);
exit(EXIT_FAIL_BPF);
@@ -470,23 +487,22 @@ static int create_cpu_entry(__u32 cpu, __u32 queue_size,
/* Inform bpf_prog's that a new CPU is available to select
* from via some control maps.
*/
- /* map_fd[5] = cpus_available */
- ret = bpf_map_update_elem(map_fd[5], &avail_idx, &cpu, 0);
+ ret = bpf_map_update_elem(cpus_available_map_fd, &avail_idx, &cpu, 0);
if (ret) {
fprintf(stderr, "Add to avail CPUs failed\n");
exit(EXIT_FAIL_BPF);
}
/* When not replacing/updating existing entry, bump the count */
- /* map_fd[6] = cpus_count */
- ret = bpf_map_lookup_elem(map_fd[6], &key, &curr_cpus_count);
+ ret = bpf_map_lookup_elem(cpus_count_map_fd, &key, &curr_cpus_count);
if (ret) {
fprintf(stderr, "Failed reading curr cpus_count\n");
exit(EXIT_FAIL_BPF);
}
if (new) {
curr_cpus_count++;
- ret = bpf_map_update_elem(map_fd[6], &key, &curr_cpus_count, 0);
+ ret = bpf_map_update_elem(cpus_count_map_fd, &key,
+ &curr_cpus_count, 0);
if (ret) {
fprintf(stderr, "Failed write curr cpus_count\n");
exit(EXIT_FAIL_BPF);
@@ -509,8 +525,8 @@ static void mark_cpus_unavailable(void)
int ret, i;
for (i = 0; i < MAX_CPUS; i++) {
- /* map_fd[5] = cpus_available */
- ret = bpf_map_update_elem(map_fd[5], &i, &invalid_cpu, 0);
+ ret = bpf_map_update_elem(cpus_available_map_fd, &i,
+ &invalid_cpu, 0);
if (ret) {
fprintf(stderr, "Failed marking CPU unavailable\n");
exit(EXIT_FAIL_BPF);
@@ -530,7 +546,7 @@ static void stress_cpumap(void)
create_cpu_entry(1, 16000, 0, false);
}
-static void stats_poll(int interval, bool use_separators, int prog_num,
+static void stats_poll(int interval, bool use_separators, char *prog_name,
bool stress_mode)
{
struct stats_record *record, *prev;
@@ -546,7 +562,7 @@ static void stats_poll(int interval, bool use_separators, int prog_num,
while (1) {
swap(&prev, &record);
stats_collect(record);
- stats_print(record, prev, prog_num);
+ stats_print(record, prev, prog_name);
sleep(interval);
if (stress_mode)
stress_cpumap();
@@ -556,17 +572,51 @@ static void stats_poll(int interval, bool use_separators, int prog_num,
free_stats_record(prev);
}
+static int init_map_fds(struct bpf_object *obj)
+{
+ cpu_map_fd = bpf_object__find_map_fd_by_name(obj, "cpu_map");
+ rx_cnt_map_fd = bpf_object__find_map_fd_by_name(obj, "rx_cnt");
+ redirect_err_cnt_map_fd =
+ bpf_object__find_map_fd_by_name(obj, "redirect_err_cnt");
+ cpumap_enqueue_cnt_map_fd =
+ bpf_object__find_map_fd_by_name(obj, "cpumap_enqueue_cnt");
+ cpumap_kthread_cnt_map_fd =
+ bpf_object__find_map_fd_by_name(obj, "cpumap_kthread_cnt");
+ cpus_available_map_fd =
+ bpf_object__find_map_fd_by_name(obj, "cpus_available");
+ cpus_count_map_fd = bpf_object__find_map_fd_by_name(obj, "cpus_count");
+ cpus_iterator_map_fd =
+ bpf_object__find_map_fd_by_name(obj, "cpus_iterator");
+ exception_cnt_map_fd =
+ bpf_object__find_map_fd_by_name(obj, "exception_cnt");
+
+ if (cpu_map_fd < 0 || rx_cnt_map_fd < 0 ||
+ redirect_err_cnt_map_fd < 0 || cpumap_enqueue_cnt_map_fd < 0 ||
+ cpumap_kthread_cnt_map_fd < 0 || cpus_available_map_fd < 0 ||
+ cpus_count_map_fd < 0 || cpus_iterator_map_fd < 0 ||
+ exception_cnt_map_fd < 0)
+ return -ENOENT;
+
+ return 0;
+}
+
int main(int argc, char **argv)
{
struct rlimit r = {10 * 1024 * 1024, RLIM_INFINITY};
+ char *prog_name = "xdp_cpu_map5_lb_hash_ip_pairs";
+ struct bpf_prog_load_attr prog_load_attr = {
+ .prog_type = BPF_PROG_TYPE_UNSPEC,
+ };
bool use_separators = true;
bool stress_mode = false;
+ struct bpf_program *prog;
+ struct bpf_object *obj;
char filename[256];
int added_cpus = 0;
int longindex = 0;
int interval = 2;
- int prog_num = 5;
int add_cpu = -1;
+ int prog_fd;
__u32 qsize;
int opt;
@@ -579,22 +629,25 @@ int main(int argc, char **argv)
qsize = 128+64;
snprintf(filename, sizeof(filename), "%s_kern.o", argv[0]);
+ prog_load_attr.file = filename;
if (setrlimit(RLIMIT_MEMLOCK, &r)) {
perror("setrlimit(RLIMIT_MEMLOCK)");
return 1;
}
- if (load_bpf_file(filename)) {
- fprintf(stderr, "ERR in load_bpf_file(): %s", bpf_log_buf);
+ if (bpf_prog_load_xattr(&prog_load_attr, &obj, &prog_fd))
return EXIT_FAIL;
- }
- if (!prog_fd[0]) {
- fprintf(stderr, "ERR: load_bpf_file: %s\n", strerror(errno));
+ if (prog_fd < 0) {
+ fprintf(stderr, "ERR: bpf_prog_load_xattr: %s\n",
+ strerror(errno));
+ return EXIT_FAIL;
+ }
+ if (init_map_fds(obj) < 0) {
+ fprintf(stderr, "bpf_object__find_map_fd_by_name failed\n");
return EXIT_FAIL;
}
-
mark_cpus_unavailable();
/* Parse commands line args */
@@ -630,13 +683,7 @@ int main(int argc, char **argv)
break;
case 'p':
/* Selecting eBPF prog to load */
- prog_num = atoi(optarg);
- if (prog_num < 0 || prog_num >= MAX_PROG) {
- fprintf(stderr,
- "--prognum too large err(%d):%s\n",
- errno, strerror(errno));
- goto error;
- }
+ prog_name = optarg;
break;
case 'c':
/* Add multiple CPUs */
@@ -656,21 +703,21 @@ int main(int argc, char **argv)
case 'h':
error:
default:
- usage(argv);
+ usage(argv, obj);
return EXIT_FAIL_OPTION;
}
}
/* Required option */
if (ifindex == -1) {
fprintf(stderr, "ERR: required option --dev missing\n");
- usage(argv);
+ usage(argv, obj);
return EXIT_FAIL_OPTION;
}
/* Required option */
if (add_cpu == -1) {
fprintf(stderr, "ERR: required option --cpu missing\n");
fprintf(stderr, " Specify multiple --cpu option to add more\n");
- usage(argv);
+ usage(argv, obj);
return EXIT_FAIL_OPTION;
}
@@ -678,11 +725,23 @@ int main(int argc, char **argv)
signal(SIGINT, int_exit);
signal(SIGTERM, int_exit);
- if (bpf_set_link_xdp_fd(ifindex, prog_fd[prog_num], xdp_flags) < 0) {
+ prog = bpf_object__find_program_by_title(obj, prog_name);
+ if (!prog) {
+ fprintf(stderr, "bpf_object__find_program_by_title failed\n");
+ return EXIT_FAIL;
+ }
+
+ prog_fd = bpf_program__fd(prog);
+ if (prog_fd < 0) {
+ fprintf(stderr, "bpf_program__fd failed\n");
+ return EXIT_FAIL;
+ }
+
+ if (bpf_set_link_xdp_fd(ifindex, prog_fd, xdp_flags) < 0) {
fprintf(stderr, "link set xdp fd failed\n");
return EXIT_FAIL_XDP;
}
- stats_poll(interval, use_separators, prog_num, stress_mode);
+ stats_poll(interval, use_separators, prog_name, stress_mode);
return EXIT_OK;
}
diff --git a/samples/bpf/xdp_redirect_map_user.c b/samples/bpf/xdp_redirect_map_user.c
index 4445e76854b5..60d46eea225b 100644
--- a/samples/bpf/xdp_redirect_map_user.c
+++ b/samples/bpf/xdp_redirect_map_user.c
@@ -22,15 +22,16 @@
#include <libgen.h>
#include <sys/resource.h>
-#include "bpf_load.h"
#include "bpf_util.h"
#include <bpf/bpf.h>
+#include "bpf/libbpf.h"
static int ifindex_in;
static int ifindex_out;
static bool ifindex_out_xdp_dummy_attached = true;
static __u32 xdp_flags;
+static int rxcnt_map_fd;
static void int_exit(int sig)
{
@@ -53,7 +54,7 @@ static void poll_stats(int interval, int ifindex)
int i;
sleep(interval);
- assert(bpf_map_lookup_elem(map_fd[1], &key, values) == 0);
+ assert(bpf_map_lookup_elem(rxcnt_map_fd, &key, values) == 0);
for (i = 0; i < nr_cpus; i++)
sum += (values[i] - prev[i]);
if (sum)
@@ -76,9 +77,16 @@ static void usage(const char *prog)
int main(int argc, char **argv)
{
struct rlimit r = {RLIM_INFINITY, RLIM_INFINITY};
+ struct bpf_prog_load_attr prog_load_attr = {
+ .prog_type = BPF_PROG_TYPE_XDP,
+ };
+ struct bpf_program *prog, *dummy_prog;
+ int prog_fd, dummy_prog_fd;
const char *optstr = "SN";
- char filename[256];
+ struct bpf_object *obj;
int ret, opt, key = 0;
+ char filename[256];
+ int tx_port_map_fd;
while ((opt = getopt(argc, argv, optstr)) != -1) {
switch (opt) {
@@ -109,24 +117,40 @@ int main(int argc, char **argv)
printf("input: %d output: %d\n", ifindex_in, ifindex_out);
snprintf(filename, sizeof(filename), "%s_kern.o", argv[0]);
+ prog_load_attr.file = filename;
+
+ if (bpf_prog_load_xattr(&prog_load_attr, &obj, &prog_fd))
+ return 1;
- if (load_bpf_file(filename)) {
- printf("%s", bpf_log_buf);
+ prog = bpf_program__next(NULL, obj);
+ dummy_prog = bpf_program__next(prog, obj);
+ if (!prog || !dummy_prog) {
+ printf("finding a prog in obj file failed\n");
+ return 1;
+ }
+ /* bpf_prog_load_xattr gives us the pointer to first prog's fd,
+ * so we're missing only the fd for dummy prog
+ */
+ dummy_prog_fd = bpf_program__fd(dummy_prog);
+ if (prog_fd < 0 || dummy_prog_fd < 0) {
+ printf("bpf_prog_load_xattr: %s\n", strerror(errno));
return 1;
}
- if (!prog_fd[0]) {
- printf("load_bpf_file: %s\n", strerror(errno));
+ tx_port_map_fd = bpf_object__find_map_fd_by_name(obj, "tx_port");
+ rxcnt_map_fd = bpf_object__find_map_fd_by_name(obj, "rxcnt");
+ if (tx_port_map_fd < 0 || rxcnt_map_fd < 0) {
+ printf("bpf_object__find_map_fd_by_name failed\n");
return 1;
}
- if (bpf_set_link_xdp_fd(ifindex_in, prog_fd[0], xdp_flags) < 0) {
+ if (bpf_set_link_xdp_fd(ifindex_in, prog_fd, xdp_flags) < 0) {
printf("ERROR: link set xdp fd failed on %d\n", ifindex_in);
return 1;
}
/* Loading dummy XDP prog on out-device */
- if (bpf_set_link_xdp_fd(ifindex_out, prog_fd[1],
+ if (bpf_set_link_xdp_fd(ifindex_out, dummy_prog_fd,
(xdp_flags | XDP_FLAGS_UPDATE_IF_NOEXIST)) < 0) {
printf("WARN: link set xdp fd failed on %d\n", ifindex_out);
ifindex_out_xdp_dummy_attached = false;
@@ -135,11 +159,8 @@ int main(int argc, char **argv)
signal(SIGINT, int_exit);
signal(SIGTERM, int_exit);
- printf("map[0] (vports) = %i, map[1] (map) = %i, map[2] (count) = %i\n",
- map_fd[0], map_fd[1], map_fd[2]);
-
/* populate virtual to physical port map */
- ret = bpf_map_update_elem(map_fd[0], &key, &ifindex_out, 0);
+ ret = bpf_map_update_elem(tx_port_map_fd, &key, &ifindex_out, 0);
if (ret) {
perror("bpf_update_elem");
goto out;
diff --git a/samples/bpf/xdp_redirect_user.c b/samples/bpf/xdp_redirect_user.c
index 81a69e36cb78..93404820df68 100644
--- a/samples/bpf/xdp_redirect_user.c
+++ b/samples/bpf/xdp_redirect_user.c
@@ -22,15 +22,16 @@
#include <libgen.h>
#include <sys/resource.h>
-#include "bpf_load.h"
#include "bpf_util.h"
#include <bpf/bpf.h>
+#include "bpf/libbpf.h"
static int ifindex_in;
static int ifindex_out;
static bool ifindex_out_xdp_dummy_attached = true;
static __u32 xdp_flags;
+static int rxcnt_map_fd;
static void int_exit(int sig)
{
@@ -53,7 +54,7 @@ static void poll_stats(int interval, int ifindex)
int i;
sleep(interval);
- assert(bpf_map_lookup_elem(map_fd[1], &key, values) == 0);
+ assert(bpf_map_lookup_elem(rxcnt_map_fd, &key, values) == 0);
for (i = 0; i < nr_cpus; i++)
sum += (values[i] - prev[i]);
if (sum)
@@ -77,9 +78,16 @@ static void usage(const char *prog)
int main(int argc, char **argv)
{
struct rlimit r = {RLIM_INFINITY, RLIM_INFINITY};
+ struct bpf_prog_load_attr prog_load_attr = {
+ .prog_type = BPF_PROG_TYPE_XDP,
+ };
+ struct bpf_program *prog, *dummy_prog;
+ int prog_fd, tx_port_map_fd, opt;
const char *optstr = "SN";
+ struct bpf_object *obj;
char filename[256];
- int ret, opt, key = 0;
+ int dummy_prog_fd;
+ int ret, key = 0;
while ((opt = getopt(argc, argv, optstr)) != -1) {
switch (opt) {
@@ -110,24 +118,40 @@ int main(int argc, char **argv)
printf("input: %d output: %d\n", ifindex_in, ifindex_out);
snprintf(filename, sizeof(filename), "%s_kern.o", argv[0]);
+ prog_load_attr.file = filename;
- if (load_bpf_file(filename)) {
- printf("%s", bpf_log_buf);
+ if (bpf_prog_load_xattr(&prog_load_attr, &obj, &prog_fd))
+ return 1;
+
+ prog = bpf_program__next(NULL, obj);
+ dummy_prog = bpf_program__next(prog, obj);
+ if (!prog || !dummy_prog) {
+ printf("finding a prog in obj file failed\n");
+ return 1;
+ }
+ /* bpf_prog_load_xattr gives us the pointer to first prog's fd,
+ * so we're missing only the fd for dummy prog
+ */
+ dummy_prog_fd = bpf_program__fd(dummy_prog);
+ if (prog_fd < 0 || dummy_prog_fd < 0) {
+ printf("bpf_prog_load_xattr: %s\n", strerror(errno));
return 1;
}
- if (!prog_fd[0]) {
- printf("load_bpf_file: %s\n", strerror(errno));
+ tx_port_map_fd = bpf_object__find_map_fd_by_name(obj, "tx_port");
+ rxcnt_map_fd = bpf_object__find_map_fd_by_name(obj, "rxcnt");
+ if (tx_port_map_fd < 0 || rxcnt_map_fd < 0) {
+ printf("bpf_object__find_map_fd_by_name failed\n");
return 1;
}
- if (bpf_set_link_xdp_fd(ifindex_in, prog_fd[0], xdp_flags) < 0) {
+ if (bpf_set_link_xdp_fd(ifindex_in, prog_fd, xdp_flags) < 0) {
printf("ERROR: link set xdp fd failed on %d\n", ifindex_in);
return 1;
}
/* Loading dummy XDP prog on out-device */
- if (bpf_set_link_xdp_fd(ifindex_out, prog_fd[1],
+ if (bpf_set_link_xdp_fd(ifindex_out, dummy_prog_fd,
(xdp_flags | XDP_FLAGS_UPDATE_IF_NOEXIST)) < 0) {
printf("WARN: link set xdp fd failed on %d\n", ifindex_out);
ifindex_out_xdp_dummy_attached = false;
@@ -137,7 +161,7 @@ int main(int argc, char **argv)
signal(SIGTERM, int_exit);
/* bpf redirect port */
- ret = bpf_map_update_elem(map_fd[0], &key, &ifindex_out, 0);
+ ret = bpf_map_update_elem(tx_port_map_fd, &key, &ifindex_out, 0);
if (ret) {
perror("bpf_update_elem");
goto out;
diff --git a/samples/bpf/xdp_router_ipv4_user.c b/samples/bpf/xdp_router_ipv4_user.c
index b2b4dfa776c8..cea2306f5ab7 100644
--- a/samples/bpf/xdp_router_ipv4_user.c
+++ b/samples/bpf/xdp_router_ipv4_user.c
@@ -15,7 +15,6 @@
#include <string.h>
#include <sys/socket.h>
#include <unistd.h>
-#include "bpf_load.h"
#include <bpf/bpf.h>
#include <arpa/inet.h>
#include <fcntl.h>
@@ -25,11 +24,17 @@
#include <sys/ioctl.h>
#include <sys/syscall.h>
#include "bpf_util.h"
+#include "bpf/libbpf.h"
int sock, sock_arp, flags = 0;
static int total_ifindex;
int *ifindex_list;
char buf[8192];
+static int lpm_map_fd;
+static int rxcnt_map_fd;
+static int arp_table_map_fd;
+static int exact_match_map_fd;
+static int tx_port_map_fd;
static int get_route_table(int rtm_family);
static void int_exit(int sig)
@@ -186,7 +191,8 @@ static void read_route(struct nlmsghdr *nh, int nll)
bpf_set_link_xdp_fd(ifindex_list[i], -1, flags);
exit(0);
}
- assert(bpf_map_update_elem(map_fd[4], &route.iface, &route.iface, 0) == 0);
+ assert(bpf_map_update_elem(tx_port_map_fd,
+ &route.iface, &route.iface, 0) == 0);
if (rtm_family == AF_INET) {
struct trie_value {
__u8 prefix[4];
@@ -207,11 +213,16 @@ static void read_route(struct nlmsghdr *nh, int nll)
direct_entry.arp.dst = 0;
if (route.dst_len == 32) {
if (nh->nlmsg_type == RTM_DELROUTE) {
- assert(bpf_map_delete_elem(map_fd[3], &route.dst) == 0);
+ assert(bpf_map_delete_elem(exact_match_map_fd,
+ &route.dst) == 0);
} else {
- if (bpf_map_lookup_elem(map_fd[2], &route.dst, &direct_entry.arp.mac) == 0)
+ if (bpf_map_lookup_elem(arp_table_map_fd,
+ &route.dst,
+ &direct_entry.arp.mac) == 0)
direct_entry.arp.dst = route.dst;
- assert(bpf_map_update_elem(map_fd[3], &route.dst, &direct_entry, 0) == 0);
+ assert(bpf_map_update_elem(exact_match_map_fd,
+ &route.dst,
+ &direct_entry, 0) == 0);
}
}
for (i = 0; i < 4; i++)
@@ -225,7 +236,7 @@ static void read_route(struct nlmsghdr *nh, int nll)
route.gw, route.dst_len,
route.metric,
route.iface_name);
- if (bpf_map_lookup_elem(map_fd[0], prefix_key,
+ if (bpf_map_lookup_elem(lpm_map_fd, prefix_key,
prefix_value) < 0) {
for (i = 0; i < 4; i++)
prefix_value->prefix[i] = prefix_key->data[i];
@@ -234,7 +245,7 @@ static void read_route(struct nlmsghdr *nh, int nll)
prefix_value->gw = route.gw;
prefix_value->metric = route.metric;
- assert(bpf_map_update_elem(map_fd[0],
+ assert(bpf_map_update_elem(lpm_map_fd,
prefix_key,
prefix_value, 0
) == 0);
@@ -247,7 +258,7 @@ static void read_route(struct nlmsghdr *nh, int nll)
prefix_key->data[2],
prefix_key->data[3],
prefix_key->prefixlen);
- assert(bpf_map_delete_elem(map_fd[0],
+ assert(bpf_map_delete_elem(lpm_map_fd,
prefix_key
) == 0);
/* Rereading the route table to check if
@@ -275,8 +286,7 @@ static void read_route(struct nlmsghdr *nh, int nll)
prefix_value->ifindex = route.iface;
prefix_value->gw = route.gw;
prefix_value->metric = route.metric;
- assert(bpf_map_update_elem(
- map_fd[0],
+ assert(bpf_map_update_elem(lpm_map_fd,
prefix_key,
prefix_value,
0) == 0);
@@ -401,7 +411,8 @@ static void read_arp(struct nlmsghdr *nh, int nll)
arp_entry.mac = atol(mac);
printf("%x\t\t%llx\n", arp_entry.dst, arp_entry.mac);
if (ndm_family == AF_INET) {
- if (bpf_map_lookup_elem(map_fd[3], &arp_entry.dst,
+ if (bpf_map_lookup_elem(exact_match_map_fd,
+ &arp_entry.dst,
&direct_entry) == 0) {
if (nh->nlmsg_type == RTM_DELNEIGH) {
direct_entry.arp.dst = 0;
@@ -410,16 +421,17 @@ static void read_arp(struct nlmsghdr *nh, int nll)
direct_entry.arp.dst = arp_entry.dst;
direct_entry.arp.mac = arp_entry.mac;
}
- assert(bpf_map_update_elem(map_fd[3],
+ assert(bpf_map_update_elem(exact_match_map_fd,
&arp_entry.dst,
&direct_entry, 0
) == 0);
memset(&direct_entry, 0, sizeof(direct_entry));
}
if (nh->nlmsg_type == RTM_DELNEIGH) {
- assert(bpf_map_delete_elem(map_fd[2], &arp_entry.dst) == 0);
+ assert(bpf_map_delete_elem(arp_table_map_fd,
+ &arp_entry.dst) == 0);
} else if (nh->nlmsg_type == RTM_NEWNEIGH) {
- assert(bpf_map_update_elem(map_fd[2],
+ assert(bpf_map_update_elem(arp_table_map_fd,
&arp_entry.dst,
&arp_entry.mac, 0
) == 0);
@@ -553,7 +565,8 @@ static int monitor_route(void)
for (key = 0; key < nr_keys; key++) {
__u64 sum = 0;
- assert(bpf_map_lookup_elem(map_fd[1], &key, values) == 0);
+ assert(bpf_map_lookup_elem(rxcnt_map_fd,
+ &key, values) == 0);
for (i = 0; i < nr_cpus; i++)
sum += (values[i] - prev[key][i]);
if (sum)
@@ -596,11 +609,18 @@ static int monitor_route(void)
int main(int ac, char **argv)
{
+ struct bpf_prog_load_attr prog_load_attr = {
+ .prog_type = BPF_PROG_TYPE_XDP,
+ };
+ struct bpf_object *obj;
char filename[256];
char **ifname_list;
+ int prog_fd;
int i = 1;
snprintf(filename, sizeof(filename), "%s_kern.o", argv[0]);
+ prog_load_attr.file = filename;
+
if (ac < 2) {
printf("usage: %s [-S] Interface name list\n", argv[0]);
return 1;
@@ -614,15 +634,28 @@ int main(int ac, char **argv)
total_ifindex = ac - 1;
ifname_list = (argv + 1);
}
- if (load_bpf_file(filename)) {
- printf("%s", bpf_log_buf);
+
+ if (bpf_prog_load_xattr(&prog_load_attr, &obj, &prog_fd))
return 1;
- }
+
printf("\n**************loading bpf file*********************\n\n\n");
- if (!prog_fd[0]) {
- printf("load_bpf_file: %s\n", strerror(errno));
+ if (!prog_fd) {
+ printf("bpf_prog_load_xattr: %s\n", strerror(errno));
return 1;
}
+
+ lpm_map_fd = bpf_object__find_map_fd_by_name(obj, "lpm_map");
+ rxcnt_map_fd = bpf_object__find_map_fd_by_name(obj, "rxcnt");
+ arp_table_map_fd = bpf_object__find_map_fd_by_name(obj, "arp_table");
+ exact_match_map_fd = bpf_object__find_map_fd_by_name(obj,
+ "exact_match");
+ tx_port_map_fd = bpf_object__find_map_fd_by_name(obj, "tx_port");
+ if (lpm_map_fd < 0 || rxcnt_map_fd < 0 || arp_table_map_fd < 0 ||
+ exact_match_map_fd < 0 || tx_port_map_fd < 0) {
+ printf("bpf_object__find_map_fd_by_name failed\n");
+ return 1;
+ }
+
ifindex_list = (int *)malloc(total_ifindex * sizeof(int *));
for (i = 0; i < total_ifindex; i++) {
ifindex_list[i] = if_nametoindex(ifname_list[i]);
@@ -633,7 +666,7 @@ int main(int ac, char **argv)
}
}
for (i = 0; i < total_ifindex; i++) {
- if (bpf_set_link_xdp_fd(ifindex_list[i], prog_fd[0], flags) < 0) {
+ if (bpf_set_link_xdp_fd(ifindex_list[i], prog_fd, flags) < 0) {
printf("link set xdp fd failed\n");
int recovery_index = i;
diff --git a/samples/bpf/xdp_tx_iptunnel_user.c b/samples/bpf/xdp_tx_iptunnel_user.c
index a4ccc33adac0..5093d8220da5 100644
--- a/samples/bpf/xdp_tx_iptunnel_user.c
+++ b/samples/bpf/xdp_tx_iptunnel_user.c
@@ -17,7 +17,7 @@
#include <netinet/ether.h>
#include <unistd.h>
#include <time.h>
-#include "bpf_load.h"
+#include "bpf/libbpf.h"
#include <bpf/bpf.h>
#include "bpf_util.h"
#include "xdp_tx_iptunnel_common.h"
@@ -26,6 +26,7 @@
static int ifindex = -1;
static __u32 xdp_flags = 0;
+static int rxcnt_map_fd;
static void int_exit(int sig)
{
@@ -53,7 +54,8 @@ static void poll_stats(unsigned int kill_after_s)
for (proto = 0; proto < nr_protos; proto++) {
__u64 sum = 0;
- assert(bpf_map_lookup_elem(map_fd[0], &proto, values) == 0);
+ assert(bpf_map_lookup_elem(rxcnt_map_fd, &proto,
+ values) == 0);
for (i = 0; i < nr_cpus; i++)
sum += (values[i] - prev[proto][i]);
@@ -138,15 +140,19 @@ static int parse_ports(const char *port_str, int *min_port, int *max_port)
int main(int argc, char **argv)
{
+ struct bpf_prog_load_attr prog_load_attr = {
+ .prog_type = BPF_PROG_TYPE_XDP,
+ };
+ struct rlimit r = {RLIM_INFINITY, RLIM_INFINITY};
+ int min_port = 0, max_port = 0, vip2tnl_map_fd;
+ const char *optstr = "i:a:p:s:d:m:T:P:SNh";
unsigned char opt_flags[256] = {};
unsigned int kill_after_s = 0;
- const char *optstr = "i:a:p:s:d:m:T:P:SNh";
- int min_port = 0, max_port = 0;
struct iptnl_info tnl = {};
- struct rlimit r = {RLIM_INFINITY, RLIM_INFINITY};
+ struct bpf_object *obj;
struct vip vip = {};
char filename[256];
- int opt;
+ int opt, prog_fd;
int i;
tnl.family = AF_UNSPEC;
@@ -232,29 +238,36 @@ int main(int argc, char **argv)
}
snprintf(filename, sizeof(filename), "%s_kern.o", argv[0]);
+ prog_load_attr.file = filename;
- if (load_bpf_file(filename)) {
- printf("%s", bpf_log_buf);
+ if (bpf_prog_load_xattr(&prog_load_attr, &obj, &prog_fd))
return 1;
- }
- if (!prog_fd[0]) {
+ if (!prog_fd) {
printf("load_bpf_file: %s\n", strerror(errno));
return 1;
}
+ rxcnt_map_fd = bpf_object__find_map_fd_by_name(obj, "rxcnt");
+ vip2tnl_map_fd = bpf_object__find_map_fd_by_name(obj, "vip2tnl");
+ if (vip2tnl_map_fd < 0 || rxcnt_map_fd < 0) {
+ printf("bpf_object__find_map_fd_by_name failed\n");
+ return 1;
+ }
+
signal(SIGINT, int_exit);
signal(SIGTERM, int_exit);
while (min_port <= max_port) {
vip.dport = htons(min_port++);
- if (bpf_map_update_elem(map_fd[1], &vip, &tnl, BPF_NOEXIST)) {
+ if (bpf_map_update_elem(vip2tnl_map_fd, &vip, &tnl,
+ BPF_NOEXIST)) {
perror("bpf_map_update_elem(&vip2tnl)");
return 1;
}
}
- if (bpf_set_link_xdp_fd(ifindex, prog_fd[0], xdp_flags) < 0) {
+ if (bpf_set_link_xdp_fd(ifindex, prog_fd, xdp_flags) < 0) {
printf("link set xdp fd failed\n");
return 1;
}
--
2.16.1
^ permalink raw reply related [flat|nested] 22+ messages in thread
* [PATCH bpf-next v5 4/8] samples/bpf: Extend RLIMIT_MEMLOCK for xdp_{sample_pkts, router_ipv4}
2019-02-01 0:19 [PATCH bpf-next v5 0/8] xdp: Avoid unloading xdp prog not attached by sample Maciej Fijalkowski
` (2 preceding siblings ...)
2019-02-01 0:19 ` [PATCH bpf-next v5 3/8] samples/bpf: Convert XDP samples to libbpf usage Maciej Fijalkowski
@ 2019-02-01 0:19 ` Maciej Fijalkowski
2019-02-01 0:19 ` [PATCH bpf-next v5 5/8] xdp: Provide extack messages when prog attachment failed Maciej Fijalkowski
` (4 subsequent siblings)
8 siblings, 0 replies; 22+ messages in thread
From: Maciej Fijalkowski @ 2019-02-01 0:19 UTC (permalink / raw)
To: daniel, ast; +Cc: netdev, jakub.kicinski, brouer, john.fastabend
There is a common problem with xdp samples that happens when user wants
to run a particular sample and some bpf program is already loaded. The
default 64kb RLIMIT_MEMLOCK resource limit will cause a following error
(assuming that xdp sample that is failing was converted to libbpf
usage):
libbpf: Error in bpf_object__probe_name():Operation not permitted(1).
Couldn't load basic 'r0 = 0' BPF program.
libbpf: failed to load object './xdp_sample_pkts_kern.o'
Fix it in xdp_sample_pkts and xdp_router_ipv4 by setting RLIMIT_MEMLOCK
to RLIM_INFINITY.
Signed-off-by: Maciej Fijalkowski <maciej.fijalkowski@intel.com>
Reviewed-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Acked-by: Jesper Dangaard Brouer <brouer@redhat.com>
Acked-by: John Fastabend <john.fastabend@gmail.com>
---
samples/bpf/xdp_router_ipv4_user.c | 7 +++++++
samples/bpf/xdp_sample_pkts_user.c | 7 +++++++
2 files changed, 14 insertions(+)
diff --git a/samples/bpf/xdp_router_ipv4_user.c b/samples/bpf/xdp_router_ipv4_user.c
index cea2306f5ab7..c63c6beec7d6 100644
--- a/samples/bpf/xdp_router_ipv4_user.c
+++ b/samples/bpf/xdp_router_ipv4_user.c
@@ -25,6 +25,7 @@
#include <sys/syscall.h>
#include "bpf_util.h"
#include "bpf/libbpf.h"
+#include <sys/resource.h>
int sock, sock_arp, flags = 0;
static int total_ifindex;
@@ -609,6 +610,7 @@ static int monitor_route(void)
int main(int ac, char **argv)
{
+ struct rlimit r = {RLIM_INFINITY, RLIM_INFINITY};
struct bpf_prog_load_attr prog_load_attr = {
.prog_type = BPF_PROG_TYPE_XDP,
};
@@ -635,6 +637,11 @@ int main(int ac, char **argv)
ifname_list = (argv + 1);
}
+ if (setrlimit(RLIMIT_MEMLOCK, &r)) {
+ perror("setrlimit(RLIMIT_MEMLOCK)");
+ return 1;
+ }
+
if (bpf_prog_load_xattr(&prog_load_attr, &obj, &prog_fd))
return 1;
diff --git a/samples/bpf/xdp_sample_pkts_user.c b/samples/bpf/xdp_sample_pkts_user.c
index 8dd87c1eb560..5f5828ee0761 100644
--- a/samples/bpf/xdp_sample_pkts_user.c
+++ b/samples/bpf/xdp_sample_pkts_user.c
@@ -12,6 +12,7 @@
#include <signal.h>
#include <libbpf.h>
#include <bpf/bpf.h>
+#include <sys/resource.h>
#include "perf-sys.h"
#include "trace_helpers.h"
@@ -99,6 +100,7 @@ static void sig_handler(int signo)
int main(int argc, char **argv)
{
+ struct rlimit r = {RLIM_INFINITY, RLIM_INFINITY};
struct bpf_prog_load_attr prog_load_attr = {
.prog_type = BPF_PROG_TYPE_XDP,
};
@@ -114,6 +116,11 @@ int main(int argc, char **argv)
return 1;
}
+ if (setrlimit(RLIMIT_MEMLOCK, &r)) {
+ perror("setrlimit(RLIMIT_MEMLOCK)");
+ return 1;
+ }
+
numcpus = get_nprocs();
if (numcpus > MAX_CPUS)
numcpus = MAX_CPUS;
--
2.16.1
^ permalink raw reply related [flat|nested] 22+ messages in thread
* [PATCH bpf-next v5 5/8] xdp: Provide extack messages when prog attachment failed
2019-02-01 0:19 [PATCH bpf-next v5 0/8] xdp: Avoid unloading xdp prog not attached by sample Maciej Fijalkowski
` (3 preceding siblings ...)
2019-02-01 0:19 ` [PATCH bpf-next v5 4/8] samples/bpf: Extend RLIMIT_MEMLOCK for xdp_{sample_pkts, router_ipv4} Maciej Fijalkowski
@ 2019-02-01 0:19 ` Maciej Fijalkowski
2019-02-01 0:32 ` Jakub Kicinski
` (2 more replies)
2019-02-01 0:19 ` [PATCH bpf-next v5 6/8] samples/bpf: Add a "force" flag to XDP samples Maciej Fijalkowski
` (3 subsequent siblings)
8 siblings, 3 replies; 22+ messages in thread
From: Maciej Fijalkowski @ 2019-02-01 0:19 UTC (permalink / raw)
To: daniel, ast; +Cc: netdev, jakub.kicinski, brouer, john.fastabend
In order to provide more meaningful messages to user when the process of
loading xdp program onto network interface failed, let's add extack
messages within dev_change_xdp_fd.
Suggested-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Signed-off-by: Maciej Fijalkowski <maciej.fijalkowski@intel.com>
---
net/core/dev.c | 12 +++++++++---
1 file changed, 9 insertions(+), 3 deletions(-)
diff --git a/net/core/dev.c b/net/core/dev.c
index 8e276e0192a1..bfa4be42afff 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -7983,8 +7983,10 @@ int dev_change_xdp_fd(struct net_device *dev, struct netlink_ext_ack *extack,
query = flags & XDP_FLAGS_HW_MODE ? XDP_QUERY_PROG_HW : XDP_QUERY_PROG;
bpf_op = bpf_chk = ops->ndo_bpf;
- if (!bpf_op && (flags & (XDP_FLAGS_DRV_MODE | XDP_FLAGS_HW_MODE)))
+ if (!bpf_op && (flags & (XDP_FLAGS_DRV_MODE | XDP_FLAGS_HW_MODE))) {
+ NL_SET_ERR_MSG(extack, "underlying driver does not support XDP in native mode");
return -EOPNOTSUPP;
+ }
if (!bpf_op || (flags & XDP_FLAGS_SKB_MODE))
bpf_op = generic_xdp_install;
if (bpf_op == bpf_chk)
@@ -7992,11 +7994,15 @@ int dev_change_xdp_fd(struct net_device *dev, struct netlink_ext_ack *extack,
if (fd >= 0) {
if (__dev_xdp_query(dev, bpf_chk, XDP_QUERY_PROG) ||
- __dev_xdp_query(dev, bpf_chk, XDP_QUERY_PROG_HW))
+ __dev_xdp_query(dev, bpf_chk, XDP_QUERY_PROG_HW)) {
+ NL_SET_ERR_MSG(extack, "native and generic XDP can't be active at the same time");
return -EEXIST;
+ }
if ((flags & XDP_FLAGS_UPDATE_IF_NOEXIST) &&
- __dev_xdp_query(dev, bpf_op, query))
+ __dev_xdp_query(dev, bpf_op, query)) {
+ NL_SET_ERR_MSG(extack, "XDP program already attached");
return -EBUSY;
+ }
prog = bpf_prog_get_type_dev(fd, BPF_PROG_TYPE_XDP,
bpf_op == ops->ndo_bpf);
--
2.16.1
^ permalink raw reply related [flat|nested] 22+ messages in thread
* [PATCH bpf-next v5 6/8] samples/bpf: Add a "force" flag to XDP samples
2019-02-01 0:19 [PATCH bpf-next v5 0/8] xdp: Avoid unloading xdp prog not attached by sample Maciej Fijalkowski
` (4 preceding siblings ...)
2019-02-01 0:19 ` [PATCH bpf-next v5 5/8] xdp: Provide extack messages when prog attachment failed Maciej Fijalkowski
@ 2019-02-01 0:19 ` Maciej Fijalkowski
2019-02-01 0:19 ` [PATCH bpf-next v5 7/8] libbpf: Add a support for getting xdp prog id on ifindex Maciej Fijalkowski
` (2 subsequent siblings)
8 siblings, 0 replies; 22+ messages in thread
From: Maciej Fijalkowski @ 2019-02-01 0:19 UTC (permalink / raw)
To: daniel, ast; +Cc: netdev, jakub.kicinski, brouer, john.fastabend
Make xdp samples consistent with iproute2 behavior and set the
XDP_FLAGS_UPDATE_IF_NOEXIST by default when setting the xdp program on
interface. Provide an option for user to force the program loading,
which as a result will not include the mentioned flag in
bpf_set_link_xdp_fd call.
Signed-off-by: Maciej Fijalkowski <maciej.fijalkowski@intel.com>
Reviewed-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Acked-by: John Fastabend <john.fastabend@gmail.com>
---
samples/bpf/xdp1_user.c | 10 +++++---
samples/bpf/xdp_adjust_tail_user.c | 8 ++++--
samples/bpf/xdp_redirect_cpu_user.c | 8 ++++--
samples/bpf/xdp_redirect_map_user.c | 10 +++++---
samples/bpf/xdp_redirect_user.c | 10 +++++---
samples/bpf/xdp_router_ipv4_user.c | 50 +++++++++++++++++++++++++++----------
samples/bpf/xdp_rxq_info_user.c | 8 ++++--
samples/bpf/xdp_sample_pkts_user.c | 40 +++++++++++++++++++++++------
samples/bpf/xdp_tx_iptunnel_user.c | 8 ++++--
samples/bpf/xdpsock_user.c | 7 ++++--
10 files changed, 119 insertions(+), 40 deletions(-)
diff --git a/samples/bpf/xdp1_user.c b/samples/bpf/xdp1_user.c
index 8bfda95c77ad..505bce207165 100644
--- a/samples/bpf/xdp1_user.c
+++ b/samples/bpf/xdp1_user.c
@@ -22,7 +22,7 @@
#include "bpf/libbpf.h"
static int ifindex;
-static __u32 xdp_flags;
+static __u32 xdp_flags = XDP_FLAGS_UPDATE_IF_NOEXIST;
static void int_exit(int sig)
{
@@ -63,7 +63,8 @@ static void usage(const char *prog)
"usage: %s [OPTS] IFACE\n\n"
"OPTS:\n"
" -S use skb-mode\n"
- " -N enforce native mode\n",
+ " -N enforce native mode\n"
+ " -F force loading prog\n",
prog);
}
@@ -73,7 +74,7 @@ int main(int argc, char **argv)
struct bpf_prog_load_attr prog_load_attr = {
.prog_type = BPF_PROG_TYPE_XDP,
};
- const char *optstr = "SN";
+ const char *optstr = "FSN";
int prog_fd, map_fd, opt;
struct bpf_object *obj;
struct bpf_map *map;
@@ -87,6 +88,9 @@ int main(int argc, char **argv)
case 'N':
xdp_flags |= XDP_FLAGS_DRV_MODE;
break;
+ case 'F':
+ xdp_flags &= ~XDP_FLAGS_UPDATE_IF_NOEXIST;
+ break;
default:
usage(basename(argv[0]));
return 1;
diff --git a/samples/bpf/xdp_adjust_tail_user.c b/samples/bpf/xdp_adjust_tail_user.c
index 3042ce37dae8..049bddf7778b 100644
--- a/samples/bpf/xdp_adjust_tail_user.c
+++ b/samples/bpf/xdp_adjust_tail_user.c
@@ -24,7 +24,7 @@
#define STATS_INTERVAL_S 2U
static int ifindex = -1;
-static __u32 xdp_flags;
+static __u32 xdp_flags = XDP_FLAGS_UPDATE_IF_NOEXIST;
static void int_exit(int sig)
{
@@ -60,6 +60,7 @@ static void usage(const char *cmd)
printf(" -T <stop-after-X-seconds> Default: 0 (forever)\n");
printf(" -S use skb-mode\n");
printf(" -N enforce native mode\n");
+ printf(" -F force loading prog\n");
printf(" -h Display this help\n");
}
@@ -70,8 +71,8 @@ int main(int argc, char **argv)
.prog_type = BPF_PROG_TYPE_XDP,
};
unsigned char opt_flags[256] = {};
+ const char *optstr = "i:T:SNFh";
unsigned int kill_after_s = 0;
- const char *optstr = "i:T:SNh";
int i, prog_fd, map_fd, opt;
struct bpf_object *obj;
struct bpf_map *map;
@@ -96,6 +97,9 @@ int main(int argc, char **argv)
case 'N':
xdp_flags |= XDP_FLAGS_DRV_MODE;
break;
+ case 'F':
+ xdp_flags &= ~XDP_FLAGS_UPDATE_IF_NOEXIST;
+ break;
default:
usage(argv[0]);
return 1;
diff --git a/samples/bpf/xdp_redirect_cpu_user.c b/samples/bpf/xdp_redirect_cpu_user.c
index 8645ddc2da0e..0224afb55845 100644
--- a/samples/bpf/xdp_redirect_cpu_user.c
+++ b/samples/bpf/xdp_redirect_cpu_user.c
@@ -33,7 +33,7 @@ static int ifindex = -1;
static char ifname_buf[IF_NAMESIZE];
static char *ifname;
-static __u32 xdp_flags;
+static __u32 xdp_flags = XDP_FLAGS_UPDATE_IF_NOEXIST;
static int cpu_map_fd;
static int rx_cnt_map_fd;
static int redirect_err_cnt_map_fd;
@@ -62,6 +62,7 @@ static const struct option long_options[] = {
{"cpu", required_argument, NULL, 'c' },
{"stress-mode", no_argument, NULL, 'x' },
{"no-separators", no_argument, NULL, 'z' },
+ {"force", no_argument, NULL, 'F' },
{0, 0, NULL, 0 }
};
@@ -651,7 +652,7 @@ int main(int argc, char **argv)
mark_cpus_unavailable();
/* Parse commands line args */
- while ((opt = getopt_long(argc, argv, "hSd:",
+ while ((opt = getopt_long(argc, argv, "hSd:s:p:q:c:xzF",
long_options, &longindex)) != -1) {
switch (opt) {
case 'd':
@@ -700,6 +701,9 @@ int main(int argc, char **argv)
case 'q':
qsize = atoi(optarg);
break;
+ case 'F':
+ xdp_flags &= ~XDP_FLAGS_UPDATE_IF_NOEXIST;
+ break;
case 'h':
error:
default:
diff --git a/samples/bpf/xdp_redirect_map_user.c b/samples/bpf/xdp_redirect_map_user.c
index 60d46eea225b..470e1a7e8810 100644
--- a/samples/bpf/xdp_redirect_map_user.c
+++ b/samples/bpf/xdp_redirect_map_user.c
@@ -30,7 +30,7 @@ static int ifindex_in;
static int ifindex_out;
static bool ifindex_out_xdp_dummy_attached = true;
-static __u32 xdp_flags;
+static __u32 xdp_flags = XDP_FLAGS_UPDATE_IF_NOEXIST;
static int rxcnt_map_fd;
static void int_exit(int sig)
@@ -70,7 +70,8 @@ static void usage(const char *prog)
"usage: %s [OPTS] IFINDEX_IN IFINDEX_OUT\n\n"
"OPTS:\n"
" -S use skb-mode\n"
- " -N enforce native mode\n",
+ " -N enforce native mode\n"
+ " -F force loading prog\n",
prog);
}
@@ -82,7 +83,7 @@ int main(int argc, char **argv)
};
struct bpf_program *prog, *dummy_prog;
int prog_fd, dummy_prog_fd;
- const char *optstr = "SN";
+ const char *optstr = "FSN";
struct bpf_object *obj;
int ret, opt, key = 0;
char filename[256];
@@ -96,6 +97,9 @@ int main(int argc, char **argv)
case 'N':
xdp_flags |= XDP_FLAGS_DRV_MODE;
break;
+ case 'F':
+ xdp_flags &= ~XDP_FLAGS_UPDATE_IF_NOEXIST;
+ break;
default:
usage(basename(argv[0]));
return 1;
diff --git a/samples/bpf/xdp_redirect_user.c b/samples/bpf/xdp_redirect_user.c
index 93404820df68..be6058cda97c 100644
--- a/samples/bpf/xdp_redirect_user.c
+++ b/samples/bpf/xdp_redirect_user.c
@@ -30,7 +30,7 @@ static int ifindex_in;
static int ifindex_out;
static bool ifindex_out_xdp_dummy_attached = true;
-static __u32 xdp_flags;
+static __u32 xdp_flags = XDP_FLAGS_UPDATE_IF_NOEXIST;
static int rxcnt_map_fd;
static void int_exit(int sig)
@@ -70,7 +70,8 @@ static void usage(const char *prog)
"usage: %s [OPTS] IFINDEX_IN IFINDEX_OUT\n\n"
"OPTS:\n"
" -S use skb-mode\n"
- " -N enforce native mode\n",
+ " -N enforce native mode\n"
+ " -F force loading prog\n",
prog);
}
@@ -83,7 +84,7 @@ int main(int argc, char **argv)
};
struct bpf_program *prog, *dummy_prog;
int prog_fd, tx_port_map_fd, opt;
- const char *optstr = "SN";
+ const char *optstr = "FSN";
struct bpf_object *obj;
char filename[256];
int dummy_prog_fd;
@@ -97,6 +98,9 @@ int main(int argc, char **argv)
case 'N':
xdp_flags |= XDP_FLAGS_DRV_MODE;
break;
+ case 'F':
+ xdp_flags &= ~XDP_FLAGS_UPDATE_IF_NOEXIST;
+ break;
default:
usage(basename(argv[0]));
return 1;
diff --git a/samples/bpf/xdp_router_ipv4_user.c b/samples/bpf/xdp_router_ipv4_user.c
index c63c6beec7d6..208d6a996478 100644
--- a/samples/bpf/xdp_router_ipv4_user.c
+++ b/samples/bpf/xdp_router_ipv4_user.c
@@ -26,8 +26,9 @@
#include "bpf_util.h"
#include "bpf/libbpf.h"
#include <sys/resource.h>
+#include <libgen.h>
-int sock, sock_arp, flags = 0;
+int sock, sock_arp, flags = XDP_FLAGS_UPDATE_IF_NOEXIST;
static int total_ifindex;
int *ifindex_list;
char buf[8192];
@@ -608,33 +609,56 @@ static int monitor_route(void)
return ret;
}
+static void usage(const char *prog)
+{
+ fprintf(stderr,
+ "%s: %s [OPTS] interface name list\n\n"
+ "OPTS:\n"
+ " -S use skb-mode\n"
+ " -F force loading prog\n",
+ __func__, prog);
+}
+
int main(int ac, char **argv)
{
struct rlimit r = {RLIM_INFINITY, RLIM_INFINITY};
struct bpf_prog_load_attr prog_load_attr = {
.prog_type = BPF_PROG_TYPE_XDP,
};
+ const char *optstr = "SF";
struct bpf_object *obj;
char filename[256];
char **ifname_list;
- int prog_fd;
+ int prog_fd, opt;
int i = 1;
snprintf(filename, sizeof(filename), "%s_kern.o", argv[0]);
prog_load_attr.file = filename;
- if (ac < 2) {
- printf("usage: %s [-S] Interface name list\n", argv[0]);
- return 1;
+ total_ifindex = ac - 1;
+ ifname_list = (argv + 1);
+
+ while ((opt = getopt(ac, argv, optstr)) != -1) {
+ switch (opt) {
+ case 'S':
+ flags |= XDP_FLAGS_SKB_MODE;
+ total_ifindex--;
+ ifname_list++;
+ break;
+ case 'F':
+ flags &= ~XDP_FLAGS_UPDATE_IF_NOEXIST;
+ total_ifindex--;
+ ifname_list++;
+ break;
+ default:
+ usage(basename(argv[0]));
+ return 1;
+ }
}
- if (!strcmp(argv[1], "-S")) {
- flags = XDP_FLAGS_SKB_MODE;
- total_ifindex = ac - 2;
- ifname_list = (argv + 2);
- } else {
- flags = 0;
- total_ifindex = ac - 1;
- ifname_list = (argv + 1);
+
+ if (optind == ac) {
+ usage(basename(argv[0]));
+ return 1;
}
if (setrlimit(RLIMIT_MEMLOCK, &r)) {
diff --git a/samples/bpf/xdp_rxq_info_user.c b/samples/bpf/xdp_rxq_info_user.c
index ef26f882f92f..e7a98c2a440f 100644
--- a/samples/bpf/xdp_rxq_info_user.c
+++ b/samples/bpf/xdp_rxq_info_user.c
@@ -30,7 +30,7 @@ static int ifindex = -1;
static char ifname_buf[IF_NAMESIZE];
static char *ifname;
-static __u32 xdp_flags;
+static __u32 xdp_flags = XDP_FLAGS_UPDATE_IF_NOEXIST;
static struct bpf_map *stats_global_map;
static struct bpf_map *rx_queue_index_map;
@@ -52,6 +52,7 @@ static const struct option long_options[] = {
{"action", required_argument, NULL, 'a' },
{"readmem", no_argument, NULL, 'r' },
{"swapmac", no_argument, NULL, 'm' },
+ {"force", no_argument, NULL, 'F' },
{0, 0, NULL, 0 }
};
@@ -487,7 +488,7 @@ int main(int argc, char **argv)
}
/* Parse commands line args */
- while ((opt = getopt_long(argc, argv, "hSd:",
+ while ((opt = getopt_long(argc, argv, "FhSrmzd:s:a:",
long_options, &longindex)) != -1) {
switch (opt) {
case 'd':
@@ -524,6 +525,9 @@ int main(int argc, char **argv)
case 'm':
cfg_options |= SWAP_MAC;
break;
+ case 'F':
+ xdp_flags &= ~XDP_FLAGS_UPDATE_IF_NOEXIST;
+ break;
case 'h':
error:
default:
diff --git a/samples/bpf/xdp_sample_pkts_user.c b/samples/bpf/xdp_sample_pkts_user.c
index 5f5828ee0761..62f34827c775 100644
--- a/samples/bpf/xdp_sample_pkts_user.c
+++ b/samples/bpf/xdp_sample_pkts_user.c
@@ -13,6 +13,8 @@
#include <libbpf.h>
#include <bpf/bpf.h>
#include <sys/resource.h>
+#include <libgen.h>
+#include <linux/if_link.h>
#include "perf-sys.h"
#include "trace_helpers.h"
@@ -21,12 +23,13 @@
static int pmu_fds[MAX_CPUS], if_idx;
static struct perf_event_mmap_page *headers[MAX_CPUS];
static char *if_name;
+static __u32 xdp_flags = XDP_FLAGS_UPDATE_IF_NOEXIST;
static int do_attach(int idx, int fd, const char *name)
{
int err;
- err = bpf_set_link_xdp_fd(idx, fd, 0);
+ err = bpf_set_link_xdp_fd(idx, fd, xdp_flags);
if (err < 0)
printf("ERROR: failed to attach program to %s\n", name);
@@ -98,21 +101,42 @@ static void sig_handler(int signo)
exit(0);
}
+static void usage(const char *prog)
+{
+ fprintf(stderr,
+ "%s: %s [OPTS] <ifname|ifindex>\n\n"
+ "OPTS:\n"
+ " -F force loading prog\n",
+ __func__, prog);
+}
+
int main(int argc, char **argv)
{
struct rlimit r = {RLIM_INFINITY, RLIM_INFINITY};
struct bpf_prog_load_attr prog_load_attr = {
.prog_type = BPF_PROG_TYPE_XDP,
};
+ const char *optstr = "F";
+ int prog_fd, map_fd, opt;
struct bpf_object *obj;
struct bpf_map *map;
- int prog_fd, map_fd;
char filename[256];
int ret, err, i;
int numcpus;
- if (argc < 2) {
- printf("Usage: %s <ifname>\n", argv[0]);
+ while ((opt = getopt(argc, argv, optstr)) != -1) {
+ switch (opt) {
+ case 'F':
+ xdp_flags &= ~XDP_FLAGS_UPDATE_IF_NOEXIST;
+ break;
+ default:
+ usage(basename(argv[0]));
+ return 1;
+ }
+ }
+
+ if (optind == argc) {
+ usage(basename(argv[0]));
return 1;
}
@@ -143,16 +167,16 @@ int main(int argc, char **argv)
}
map_fd = bpf_map__fd(map);
- if_idx = if_nametoindex(argv[1]);
+ if_idx = if_nametoindex(argv[optind]);
if (!if_idx)
- if_idx = strtoul(argv[1], NULL, 0);
+ if_idx = strtoul(argv[optind], NULL, 0);
if (!if_idx) {
fprintf(stderr, "Invalid ifname\n");
return 1;
}
- if_name = argv[1];
- err = do_attach(if_idx, prog_fd, argv[1]);
+ if_name = argv[optind];
+ err = do_attach(if_idx, prog_fd, if_name);
if (err)
return err;
diff --git a/samples/bpf/xdp_tx_iptunnel_user.c b/samples/bpf/xdp_tx_iptunnel_user.c
index 5093d8220da5..e3de60930d27 100644
--- a/samples/bpf/xdp_tx_iptunnel_user.c
+++ b/samples/bpf/xdp_tx_iptunnel_user.c
@@ -25,7 +25,7 @@
#define STATS_INTERVAL_S 2U
static int ifindex = -1;
-static __u32 xdp_flags = 0;
+static __u32 xdp_flags = XDP_FLAGS_UPDATE_IF_NOEXIST;
static int rxcnt_map_fd;
static void int_exit(int sig)
@@ -83,6 +83,7 @@ static void usage(const char *cmd)
printf(" -P <IP-Protocol> Default is TCP\n");
printf(" -S use skb-mode\n");
printf(" -N enforce native mode\n");
+ printf(" -F Force loading the XDP prog\n");
printf(" -h Display this help\n");
}
@@ -145,7 +146,7 @@ int main(int argc, char **argv)
};
struct rlimit r = {RLIM_INFINITY, RLIM_INFINITY};
int min_port = 0, max_port = 0, vip2tnl_map_fd;
- const char *optstr = "i:a:p:s:d:m:T:P:SNh";
+ const char *optstr = "i:a:p:s:d:m:T:P:FSNh";
unsigned char opt_flags[256] = {};
unsigned int kill_after_s = 0;
struct iptnl_info tnl = {};
@@ -217,6 +218,9 @@ int main(int argc, char **argv)
case 'N':
xdp_flags |= XDP_FLAGS_DRV_MODE;
break;
+ case 'F':
+ xdp_flags &= ~XDP_FLAGS_UPDATE_IF_NOEXIST;
+ break;
default:
usage(argv[0]);
return 1;
diff --git a/samples/bpf/xdpsock_user.c b/samples/bpf/xdpsock_user.c
index 57ecadc58403..188723784768 100644
--- a/samples/bpf/xdpsock_user.c
+++ b/samples/bpf/xdpsock_user.c
@@ -68,7 +68,7 @@ enum benchmark_type {
};
static enum benchmark_type opt_bench = BENCH_RXDROP;
-static u32 opt_xdp_flags;
+static u32 opt_xdp_flags = XDP_FLAGS_UPDATE_IF_NOEXIST;
static const char *opt_if = "";
static int opt_ifindex;
static int opt_queue;
@@ -682,7 +682,7 @@ static void parse_command_line(int argc, char **argv)
opterr = 0;
for (;;) {
- c = getopt_long(argc, argv, "rtli:q:psSNn:cz", long_options,
+ c = getopt_long(argc, argv, "Frtli:q:psSNn:cz", long_options,
&option_index);
if (c == -1)
break;
@@ -725,6 +725,9 @@ static void parse_command_line(int argc, char **argv)
case 'c':
opt_xdp_bind_flags |= XDP_COPY;
break;
+ case 'F':
+ opt_xdp_flags &= ~XDP_FLAGS_UPDATE_IF_NOEXIST;
+ break;
default:
usage(basename(argv[0]));
}
--
2.16.1
^ permalink raw reply related [flat|nested] 22+ messages in thread
* [PATCH bpf-next v5 7/8] libbpf: Add a support for getting xdp prog id on ifindex
2019-02-01 0:19 [PATCH bpf-next v5 0/8] xdp: Avoid unloading xdp prog not attached by sample Maciej Fijalkowski
` (5 preceding siblings ...)
2019-02-01 0:19 ` [PATCH bpf-next v5 6/8] samples/bpf: Add a "force" flag to XDP samples Maciej Fijalkowski
@ 2019-02-01 0:19 ` Maciej Fijalkowski
2019-02-01 21:43 ` Daniel Borkmann
2019-02-01 0:19 ` [PATCH bpf-next v5 8/8] samples/bpf: Check the prog id before exiting Maciej Fijalkowski
2019-02-01 21:23 ` [PATCH bpf-next v5 0/8] xdp: Avoid unloading xdp prog not attached by sample Daniel Borkmann
8 siblings, 1 reply; 22+ messages in thread
From: Maciej Fijalkowski @ 2019-02-01 0:19 UTC (permalink / raw)
To: daniel, ast; +Cc: netdev, jakub.kicinski, brouer, john.fastabend
Since we have a dedicated netlink attributes for xdp setup on a
particular interface, it is now possible to retrieve the program id that
is currently attached to the interface. The use case is targeted for
sample xdp programs, which will store the program id just after loading
bpf program onto iface. On shutdown, the sample will make sure that it
can unload the program by querying again the iface and verifying that
both program id's matches.
Signed-off-by: Maciej Fijalkowski <maciej.fijalkowski@intel.com>
Reviewed-by: Jakub Kicinski <jakub.kicinski@netronome.com>
---
tools/lib/bpf/libbpf.h | 1 +
tools/lib/bpf/libbpf.map | 1 +
tools/lib/bpf/netlink.c | 85 ++++++++++++++++++++++++++++++++++++++++++++++++
3 files changed, 87 insertions(+)
diff --git a/tools/lib/bpf/libbpf.h b/tools/lib/bpf/libbpf.h
index 931be6f3408c..43c77e98df6f 100644
--- a/tools/lib/bpf/libbpf.h
+++ b/tools/lib/bpf/libbpf.h
@@ -317,6 +317,7 @@ LIBBPF_API int bpf_prog_load(const char *file, enum bpf_prog_type type,
struct bpf_object **pobj, int *prog_fd);
LIBBPF_API int bpf_set_link_xdp_fd(int ifindex, int fd, __u32 flags);
+LIBBPF_API int bpf_get_link_xdp_id(int ifindex, __u32 *prog_id, __u32 flags);
enum bpf_perf_event_ret {
LIBBPF_PERF_EVENT_DONE = 0,
diff --git a/tools/lib/bpf/libbpf.map b/tools/lib/bpf/libbpf.map
index b183c6c3b990..d0e023a75d72 100644
--- a/tools/lib/bpf/libbpf.map
+++ b/tools/lib/bpf/libbpf.map
@@ -131,4 +131,5 @@ LIBBPF_0.0.2 {
bpf_probe_map_type;
bpf_probe_prog_type;
bpf_object__find_map_fd_by_name;
+ bpf_get_link_xdp_id;
} LIBBPF_0.0.1;
diff --git a/tools/lib/bpf/netlink.c b/tools/lib/bpf/netlink.c
index 0ce67aea8f3b..ce3ec81b71c0 100644
--- a/tools/lib/bpf/netlink.c
+++ b/tools/lib/bpf/netlink.c
@@ -21,6 +21,12 @@
typedef int (*__dump_nlmsg_t)(struct nlmsghdr *nlmsg, libbpf_dump_nlmsg_t,
void *cookie);
+struct xdp_id_md {
+ int ifindex;
+ __u32 flags;
+ __u32 id;
+};
+
int libbpf_netlink_open(__u32 *nl_pid)
{
struct sockaddr_nl sa;
@@ -196,6 +202,85 @@ static int __dump_link_nlmsg(struct nlmsghdr *nlh,
return dump_link_nlmsg(cookie, ifi, tb);
}
+static unsigned char get_xdp_id_attr(unsigned char mode, __u32 flags)
+{
+ if (mode != XDP_ATTACHED_MULTI)
+ return IFLA_XDP_PROG_ID;
+ if (flags & XDP_FLAGS_DRV_MODE)
+ return IFLA_XDP_DRV_PROG_ID;
+ if (flags & XDP_FLAGS_HW_MODE)
+ return IFLA_XDP_HW_PROG_ID;
+ if (flags & XDP_FLAGS_SKB_MODE)
+ return IFLA_XDP_SKB_PROG_ID;
+
+ return IFLA_XDP_UNSPEC;
+}
+
+static int get_xdp_id(void *cookie, void *msg, struct nlattr **tb)
+{
+ struct nlattr *xdp_tb[IFLA_XDP_MAX + 1];
+ struct xdp_id_md *xdp_id = cookie;
+ struct ifinfomsg *ifinfo = msg;
+ unsigned char mode, xdp_attr;
+ int ret;
+
+ if (xdp_id->ifindex && xdp_id->ifindex != ifinfo->ifi_index)
+ return 0;
+
+ if (!tb[IFLA_XDP])
+ return 0;
+
+ ret = libbpf_nla_parse_nested(xdp_tb, IFLA_XDP_MAX, tb[IFLA_XDP], NULL);
+ if (ret)
+ return ret;
+
+ if (!xdp_tb[IFLA_XDP_ATTACHED])
+ return 0;
+
+ mode = libbpf_nla_getattr_u8(xdp_tb[IFLA_XDP_ATTACHED]);
+ if (mode == XDP_ATTACHED_NONE)
+ return 0;
+
+ xdp_attr = get_xdp_id_attr(mode, xdp_id->flags);
+ if (!xdp_attr || !xdp_tb[xdp_attr])
+ return 0;
+
+ xdp_id->id = libbpf_nla_getattr_u32(xdp_tb[xdp_attr]);
+
+ return 0;
+}
+
+int bpf_get_link_xdp_id(int ifindex, __u32 *prog_id, __u32 flags)
+{
+ struct xdp_id_md xdp_id = {};
+ int sock, ret;
+ __u32 nl_pid;
+ __u32 mask;
+
+ if (flags & ~XDP_FLAGS_MASK)
+ return -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)
+ return -EINVAL;
+
+ sock = libbpf_netlink_open(&nl_pid);
+ if (sock < 0)
+ return sock;
+
+ xdp_id.ifindex = ifindex;
+ xdp_id.flags = flags;
+
+ ret = libbpf_nl_get_link(sock, nl_pid, get_xdp_id, &xdp_id);
+ if (!ret)
+ *prog_id = xdp_id.id;
+
+ close(sock);
+ return ret;
+}
+
int libbpf_nl_get_link(int sock, unsigned int nl_pid,
libbpf_dump_nlmsg_t dump_link_nlmsg, void *cookie)
{
--
2.16.1
^ permalink raw reply related [flat|nested] 22+ messages in thread
* [PATCH bpf-next v5 8/8] samples/bpf: Check the prog id before exiting
2019-02-01 0:19 [PATCH bpf-next v5 0/8] xdp: Avoid unloading xdp prog not attached by sample Maciej Fijalkowski
` (6 preceding siblings ...)
2019-02-01 0:19 ` [PATCH bpf-next v5 7/8] libbpf: Add a support for getting xdp prog id on ifindex Maciej Fijalkowski
@ 2019-02-01 0:19 ` Maciej Fijalkowski
2019-02-01 21:23 ` [PATCH bpf-next v5 0/8] xdp: Avoid unloading xdp prog not attached by sample Daniel Borkmann
8 siblings, 0 replies; 22+ messages in thread
From: Maciej Fijalkowski @ 2019-02-01 0:19 UTC (permalink / raw)
To: daniel, ast; +Cc: netdev, jakub.kicinski, brouer, john.fastabend
Check the program id within the signal handler on polling xdp samples
that were previously converted to libbpf usage. Avoid the situation of
unloading the program that was not attached by sample that is exiting.
Handle also the case where bpf_get_link_xdp_id didn't exit with an error
but the xdp program was not found on an interface.
Reported-by: Michal Papaj <michal.papaj@intel.com>
Reported-by: Jakub Spizewski <jakub.spizewski@intel.com>
Signed-off-by: Maciej Fijalkowski <maciej.fijalkowski@intel.com>
Reviewed-by: Jakub Kicinski <jakub.kicinski@netronome.com>
---
samples/bpf/xdp1_user.c | 24 ++++++++++++++++-
samples/bpf/xdp_adjust_tail_user.c | 30 +++++++++++++++++++---
samples/bpf/xdp_redirect_cpu_user.c | 35 ++++++++++++++++++++-----
samples/bpf/xdp_redirect_map_user.c | 49 ++++++++++++++++++++++++++++++++---
samples/bpf/xdp_redirect_user.c | 49 ++++++++++++++++++++++++++++++++---
samples/bpf/xdp_router_ipv4_user.c | 51 ++++++++++++++++++++++++-------------
samples/bpf/xdp_rxq_info_user.c | 33 ++++++++++++++++++++----
samples/bpf/xdp_sample_pkts_user.c | 34 +++++++++++++++++++++----
samples/bpf/xdp_tx_iptunnel_user.c | 28 +++++++++++++++++---
samples/bpf/xdpsock_user.c | 23 ++++++++++++++++-
10 files changed, 308 insertions(+), 48 deletions(-)
diff --git a/samples/bpf/xdp1_user.c b/samples/bpf/xdp1_user.c
index 505bce207165..6a64e93365e1 100644
--- a/samples/bpf/xdp1_user.c
+++ b/samples/bpf/xdp1_user.c
@@ -23,10 +23,22 @@
static int ifindex;
static __u32 xdp_flags = XDP_FLAGS_UPDATE_IF_NOEXIST;
+static __u32 prog_id;
static void int_exit(int sig)
{
- bpf_set_link_xdp_fd(ifindex, -1, xdp_flags);
+ __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");
+ exit(1);
+ }
+ if (prog_id == curr_prog_id)
+ bpf_set_link_xdp_fd(ifindex, -1, xdp_flags);
+ else if (!curr_prog_id)
+ printf("couldn't find a prog id on a given interface\n");
+ else
+ printf("program on interface changed, not removing\n");
exit(0);
}
@@ -74,11 +86,14 @@ int main(int argc, char **argv)
struct bpf_prog_load_attr prog_load_attr = {
.prog_type = BPF_PROG_TYPE_XDP,
};
+ struct bpf_prog_info info = {};
+ __u32 info_len = sizeof(info);
const char *optstr = "FSN";
int prog_fd, map_fd, opt;
struct bpf_object *obj;
struct bpf_map *map;
char filename[256];
+ int err;
while ((opt = getopt(argc, argv, optstr)) != -1) {
switch (opt) {
@@ -139,6 +154,13 @@ int main(int argc, char **argv)
return 1;
}
+ err = bpf_obj_get_info_by_fd(prog_fd, &info, &info_len);
+ if (err) {
+ printf("can't get prog info - %s\n", strerror(errno));
+ return err;
+ }
+ prog_id = info.id;
+
poll_stats(map_fd, 2);
return 0;
diff --git a/samples/bpf/xdp_adjust_tail_user.c b/samples/bpf/xdp_adjust_tail_user.c
index 049bddf7778b..07e1b9269e49 100644
--- a/samples/bpf/xdp_adjust_tail_user.c
+++ b/samples/bpf/xdp_adjust_tail_user.c
@@ -25,11 +25,24 @@
static int ifindex = -1;
static __u32 xdp_flags = XDP_FLAGS_UPDATE_IF_NOEXIST;
+static __u32 prog_id;
static void int_exit(int sig)
{
- if (ifindex > -1)
- bpf_set_link_xdp_fd(ifindex, -1, xdp_flags);
+ __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");
+ exit(1);
+ }
+ if (prog_id == curr_prog_id)
+ bpf_set_link_xdp_fd(ifindex, -1, xdp_flags);
+ else if (!curr_prog_id)
+ printf("couldn't find a prog id on a given iface\n");
+ else
+ printf("program on interface changed, not removing\n");
+ }
exit(0);
}
@@ -72,11 +85,14 @@ int main(int argc, char **argv)
};
unsigned char opt_flags[256] = {};
const char *optstr = "i:T:SNFh";
+ struct bpf_prog_info info = {};
+ __u32 info_len = sizeof(info);
unsigned int kill_after_s = 0;
int i, prog_fd, map_fd, opt;
struct bpf_object *obj;
struct bpf_map *map;
char filename[256];
+ int err;
for (i = 0; i < strlen(optstr); i++)
if (optstr[i] != 'h' && 'a' <= optstr[i] && optstr[i] <= 'z')
@@ -146,9 +162,15 @@ int main(int argc, char **argv)
return 1;
}
- poll_stats(map_fd, kill_after_s);
+ err = bpf_obj_get_info_by_fd(prog_fd, &info, &info_len);
+ if (err) {
+ printf("can't get prog info - %s\n", strerror(errno));
+ return 1;
+ }
+ prog_id = info.id;
- bpf_set_link_xdp_fd(ifindex, -1, xdp_flags);
+ poll_stats(map_fd, kill_after_s);
+ int_exit(0);
return 0;
}
diff --git a/samples/bpf/xdp_redirect_cpu_user.c b/samples/bpf/xdp_redirect_cpu_user.c
index 0224afb55845..586b294d72d3 100644
--- a/samples/bpf/xdp_redirect_cpu_user.c
+++ b/samples/bpf/xdp_redirect_cpu_user.c
@@ -32,6 +32,7 @@ static const char *__doc__ =
static int ifindex = -1;
static char ifname_buf[IF_NAMESIZE];
static char *ifname;
+static __u32 prog_id;
static __u32 xdp_flags = XDP_FLAGS_UPDATE_IF_NOEXIST;
static int cpu_map_fd;
@@ -68,11 +69,24 @@ static const struct option long_options[] = {
static void int_exit(int sig)
{
- fprintf(stderr,
- "Interrupted: Removing XDP program on ifindex:%d device:%s\n",
- ifindex, ifname);
- if (ifindex > -1)
- bpf_set_link_xdp_fd(ifindex, -1, xdp_flags);
+ __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");
+ 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);
+ } else if (!curr_prog_id) {
+ printf("couldn't find a prog id on a given iface\n");
+ } else {
+ printf("program on interface changed, not removing\n");
+ }
+ }
exit(EXIT_OK);
}
@@ -608,6 +622,8 @@ int main(int argc, char **argv)
struct bpf_prog_load_attr prog_load_attr = {
.prog_type = BPF_PROG_TYPE_UNSPEC,
};
+ struct bpf_prog_info info = {};
+ __u32 info_len = sizeof(info);
bool use_separators = true;
bool stress_mode = false;
struct bpf_program *prog;
@@ -617,9 +633,9 @@ int main(int argc, char **argv)
int longindex = 0;
int interval = 2;
int add_cpu = -1;
+ int opt, err;
int prog_fd;
__u32 qsize;
- int opt;
/* Notice: choosing he queue size is very important with the
* ixgbe driver, because it's driver page recycling trick is
@@ -746,6 +762,13 @@ int main(int argc, char **argv)
return EXIT_FAIL_XDP;
}
+ err = bpf_obj_get_info_by_fd(prog_fd, &info, &info_len);
+ if (err) {
+ printf("can't get prog info - %s\n", strerror(errno));
+ return err;
+ }
+ prog_id = info.id;
+
stats_poll(interval, use_separators, prog_name, stress_mode);
return EXIT_OK;
}
diff --git a/samples/bpf/xdp_redirect_map_user.c b/samples/bpf/xdp_redirect_map_user.c
index 470e1a7e8810..327226be5a06 100644
--- a/samples/bpf/xdp_redirect_map_user.c
+++ b/samples/bpf/xdp_redirect_map_user.c
@@ -29,15 +29,41 @@
static int ifindex_in;
static int ifindex_out;
static bool ifindex_out_xdp_dummy_attached = true;
+static __u32 prog_id;
+static __u32 dummy_prog_id;
static __u32 xdp_flags = XDP_FLAGS_UPDATE_IF_NOEXIST;
static int rxcnt_map_fd;
static void int_exit(int sig)
{
- bpf_set_link_xdp_fd(ifindex_in, -1, xdp_flags);
- if (ifindex_out_xdp_dummy_attached)
- bpf_set_link_xdp_fd(ifindex_out, -1, xdp_flags);
+ __u32 curr_prog_id = 0;
+
+ if (bpf_get_link_xdp_id(ifindex_in, &curr_prog_id, xdp_flags)) {
+ printf("bpf_get_link_xdp_id failed\n");
+ exit(1);
+ }
+ if (prog_id == curr_prog_id)
+ bpf_set_link_xdp_fd(ifindex_in, -1, xdp_flags);
+ else if (!curr_prog_id)
+ printf("couldn't find a prog id on iface IN\n");
+ else
+ printf("program on iface IN changed, not removing\n");
+
+ if (ifindex_out_xdp_dummy_attached) {
+ curr_prog_id = 0;
+ if (bpf_get_link_xdp_id(ifindex_out, &curr_prog_id,
+ xdp_flags)) {
+ printf("bpf_get_link_xdp_id failed\n");
+ exit(1);
+ }
+ if (prog_id == curr_prog_id)
+ bpf_set_link_xdp_fd(ifindex_out, -1, xdp_flags);
+ else if (!curr_prog_id)
+ printf("couldn't find a prog id on iface OUT\n");
+ else
+ printf("program on iface OUT changed, not removing\n");
+ }
exit(0);
}
@@ -82,6 +108,8 @@ int main(int argc, char **argv)
.prog_type = BPF_PROG_TYPE_XDP,
};
struct bpf_program *prog, *dummy_prog;
+ struct bpf_prog_info info = {};
+ __u32 info_len = sizeof(info);
int prog_fd, dummy_prog_fd;
const char *optstr = "FSN";
struct bpf_object *obj;
@@ -153,6 +181,13 @@ int main(int argc, char **argv)
return 1;
}
+ ret = bpf_obj_get_info_by_fd(prog_fd, &info, &info_len);
+ if (ret) {
+ printf("can't get prog info - %s\n", strerror(errno));
+ return ret;
+ }
+ prog_id = info.id;
+
/* Loading dummy XDP prog on out-device */
if (bpf_set_link_xdp_fd(ifindex_out, dummy_prog_fd,
(xdp_flags | XDP_FLAGS_UPDATE_IF_NOEXIST)) < 0) {
@@ -160,6 +195,14 @@ int main(int argc, char **argv)
ifindex_out_xdp_dummy_attached = false;
}
+ memset(&info, 0, sizeof(info));
+ ret = bpf_obj_get_info_by_fd(dummy_prog_fd, &info, &info_len);
+ if (ret) {
+ printf("can't get prog info - %s\n", strerror(errno));
+ return ret;
+ }
+ dummy_prog_id = info.id;
+
signal(SIGINT, int_exit);
signal(SIGTERM, int_exit);
diff --git a/samples/bpf/xdp_redirect_user.c b/samples/bpf/xdp_redirect_user.c
index be6058cda97c..a5d8ad3129ed 100644
--- a/samples/bpf/xdp_redirect_user.c
+++ b/samples/bpf/xdp_redirect_user.c
@@ -29,15 +29,41 @@
static int ifindex_in;
static int ifindex_out;
static bool ifindex_out_xdp_dummy_attached = true;
+static __u32 prog_id;
+static __u32 dummy_prog_id;
static __u32 xdp_flags = XDP_FLAGS_UPDATE_IF_NOEXIST;
static int rxcnt_map_fd;
static void int_exit(int sig)
{
- bpf_set_link_xdp_fd(ifindex_in, -1, xdp_flags);
- if (ifindex_out_xdp_dummy_attached)
- bpf_set_link_xdp_fd(ifindex_out, -1, xdp_flags);
+ __u32 curr_prog_id = 0;
+
+ if (bpf_get_link_xdp_id(ifindex_in, &curr_prog_id, xdp_flags)) {
+ printf("bpf_get_link_xdp_id failed\n");
+ exit(1);
+ }
+ if (prog_id == curr_prog_id)
+ bpf_set_link_xdp_fd(ifindex_in, -1, xdp_flags);
+ else if (!curr_prog_id)
+ printf("couldn't find a prog id on iface IN\n");
+ else
+ printf("program on iface IN changed, not removing\n");
+
+ if (ifindex_out_xdp_dummy_attached) {
+ curr_prog_id = 0;
+ if (bpf_get_link_xdp_id(ifindex_out, &curr_prog_id,
+ xdp_flags)) {
+ printf("bpf_get_link_xdp_id failed\n");
+ exit(1);
+ }
+ if (prog_id == curr_prog_id)
+ bpf_set_link_xdp_fd(ifindex_out, -1, xdp_flags);
+ else if (!curr_prog_id)
+ printf("couldn't find a prog id on iface OUT\n");
+ else
+ printf("program on iface OUT changed, not removing\n");
+ }
exit(0);
}
@@ -84,6 +110,8 @@ int main(int argc, char **argv)
};
struct bpf_program *prog, *dummy_prog;
int prog_fd, tx_port_map_fd, opt;
+ struct bpf_prog_info info = {};
+ __u32 info_len = sizeof(info);
const char *optstr = "FSN";
struct bpf_object *obj;
char filename[256];
@@ -154,6 +182,13 @@ int main(int argc, char **argv)
return 1;
}
+ ret = bpf_obj_get_info_by_fd(prog_fd, &info, &info_len);
+ if (ret) {
+ printf("can't get prog info - %s\n", strerror(errno));
+ return ret;
+ }
+ prog_id = info.id;
+
/* Loading dummy XDP prog on out-device */
if (bpf_set_link_xdp_fd(ifindex_out, dummy_prog_fd,
(xdp_flags | XDP_FLAGS_UPDATE_IF_NOEXIST)) < 0) {
@@ -161,6 +196,14 @@ int main(int argc, char **argv)
ifindex_out_xdp_dummy_attached = false;
}
+ memset(&info, 0, sizeof(info));
+ ret = bpf_obj_get_info_by_fd(prog_fd, &info, &info_len);
+ if (ret) {
+ printf("can't get prog info - %s\n", strerror(errno));
+ return ret;
+ }
+ dummy_prog_id = info.id;
+
signal(SIGINT, int_exit);
signal(SIGTERM, int_exit);
diff --git a/samples/bpf/xdp_router_ipv4_user.c b/samples/bpf/xdp_router_ipv4_user.c
index 208d6a996478..79fe7bc26ab4 100644
--- a/samples/bpf/xdp_router_ipv4_user.c
+++ b/samples/bpf/xdp_router_ipv4_user.c
@@ -30,7 +30,8 @@
int sock, sock_arp, flags = XDP_FLAGS_UPDATE_IF_NOEXIST;
static int total_ifindex;
-int *ifindex_list;
+static int *ifindex_list;
+static __u32 *prog_id_list;
char buf[8192];
static int lpm_map_fd;
static int rxcnt_map_fd;
@@ -41,23 +42,34 @@ static int tx_port_map_fd;
static int get_route_table(int rtm_family);
static void int_exit(int sig)
{
+ __u32 prog_id = 0;
int i = 0;
- for (i = 0; i < total_ifindex; i++)
- bpf_set_link_xdp_fd(ifindex_list[i], -1, flags);
+ 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",
+ ifindex_list[i]);
+ exit(1);
+ }
+ if (prog_id_list[i] == prog_id)
+ bpf_set_link_xdp_fd(ifindex_list[i], -1, flags);
+ else if (!prog_id)
+ printf("couldn't find a prog id on iface %d\n",
+ ifindex_list[i]);
+ else
+ printf("program on iface %d changed, not removing\n",
+ ifindex_list[i]);
+ prog_id = 0;
+ }
exit(0);
}
static void close_and_exit(int sig)
{
- int i = 0;
-
close(sock);
close(sock_arp);
- for (i = 0; i < total_ifindex; i++)
- bpf_set_link_xdp_fd(ifindex_list[i], -1, flags);
- exit(0);
+ int_exit(0);
}
/* Get the mac address of the interface given interface name */
@@ -186,13 +198,8 @@ static void read_route(struct nlmsghdr *nh, int nll)
route.iface_name = alloca(sizeof(char *) * IFNAMSIZ);
route.iface_name = if_indextoname(route.iface, route.iface_name);
route.mac = getmac(route.iface_name);
- if (route.mac == -1) {
- int i = 0;
-
- for (i = 0; i < total_ifindex; i++)
- bpf_set_link_xdp_fd(ifindex_list[i], -1, flags);
- exit(0);
- }
+ if (route.mac == -1)
+ int_exit(0);
assert(bpf_map_update_elem(tx_port_map_fd,
&route.iface, &route.iface, 0) == 0);
if (rtm_family == AF_INET) {
@@ -625,12 +632,14 @@ int main(int ac, char **argv)
struct bpf_prog_load_attr prog_load_attr = {
.prog_type = BPF_PROG_TYPE_XDP,
};
+ struct bpf_prog_info info = {};
+ __u32 info_len = sizeof(info);
const char *optstr = "SF";
struct bpf_object *obj;
char filename[256];
char **ifname_list;
int prog_fd, opt;
- int i = 1;
+ int err, i = 1;
snprintf(filename, sizeof(filename), "%s_kern.o", argv[0]);
prog_load_attr.file = filename;
@@ -687,7 +696,7 @@ int main(int ac, char **argv)
return 1;
}
- ifindex_list = (int *)malloc(total_ifindex * sizeof(int *));
+ ifindex_list = (int *)calloc(total_ifindex, sizeof(int *));
for (i = 0; i < total_ifindex; i++) {
ifindex_list[i] = if_nametoindex(ifname_list[i]);
if (!ifindex_list[i]) {
@@ -696,6 +705,7 @@ int main(int ac, char **argv)
return 1;
}
}
+ 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) {
printf("link set xdp fd failed\n");
@@ -706,6 +716,13 @@ int main(int ac, char **argv)
return 1;
}
+ err = bpf_obj_get_info_by_fd(prog_fd, &info, &info_len);
+ if (err) {
+ printf("can't get prog info - %s\n", strerror(errno));
+ return err;
+ }
+ prog_id_list[i] = info.id;
+ memset(&info, 0, sizeof(info));
printf("Attached to %d\n", ifindex_list[i]);
}
signal(SIGINT, int_exit);
diff --git a/samples/bpf/xdp_rxq_info_user.c b/samples/bpf/xdp_rxq_info_user.c
index e7a98c2a440f..1210f3b170f0 100644
--- a/samples/bpf/xdp_rxq_info_user.c
+++ b/samples/bpf/xdp_rxq_info_user.c
@@ -29,6 +29,7 @@ static const char *__doc__ = " XDP RX-queue info extract example\n\n"
static int ifindex = -1;
static char ifname_buf[IF_NAMESIZE];
static char *ifname;
+static __u32 prog_id;
static __u32 xdp_flags = XDP_FLAGS_UPDATE_IF_NOEXIST;
@@ -58,11 +59,24 @@ static const struct option long_options[] = {
static void int_exit(int sig)
{
- fprintf(stderr,
- "Interrupted: Removing XDP program on ifindex:%d device:%s\n",
- ifindex, ifname);
- if (ifindex > -1)
- bpf_set_link_xdp_fd(ifindex, -1, xdp_flags);
+ __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");
+ 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);
+ } else if (!curr_prog_id) {
+ printf("couldn't find a prog id on a given iface\n");
+ } else {
+ printf("program on interface changed, not removing\n");
+ }
+ }
exit(EXIT_OK);
}
@@ -447,6 +461,8 @@ int main(int argc, char **argv)
struct bpf_prog_load_attr prog_load_attr = {
.prog_type = BPF_PROG_TYPE_XDP,
};
+ struct bpf_prog_info info = {};
+ __u32 info_len = sizeof(info);
int prog_fd, map_fd, opt, err;
bool use_separators = true;
struct config cfg = { 0 };
@@ -580,6 +596,13 @@ int main(int argc, char **argv)
return EXIT_FAIL_XDP;
}
+ err = bpf_obj_get_info_by_fd(prog_fd, &info, &info_len);
+ if (err) {
+ printf("can't get prog info - %s\n", strerror(errno));
+ return err;
+ }
+ prog_id = info.id;
+
stats_poll(interval, action, cfg_options);
return EXIT_OK;
}
diff --git a/samples/bpf/xdp_sample_pkts_user.c b/samples/bpf/xdp_sample_pkts_user.c
index 62f34827c775..dc66345a929a 100644
--- a/samples/bpf/xdp_sample_pkts_user.c
+++ b/samples/bpf/xdp_sample_pkts_user.c
@@ -24,25 +24,49 @@ static int pmu_fds[MAX_CPUS], if_idx;
static struct perf_event_mmap_page *headers[MAX_CPUS];
static char *if_name;
static __u32 xdp_flags = XDP_FLAGS_UPDATE_IF_NOEXIST;
+static __u32 prog_id;
static int do_attach(int idx, int fd, const char *name)
{
+ struct bpf_prog_info info = {};
+ __u32 info_len = sizeof(info);
int err;
err = bpf_set_link_xdp_fd(idx, fd, xdp_flags);
- if (err < 0)
+ if (err < 0) {
printf("ERROR: failed to attach program to %s\n", name);
+ return err;
+ }
+
+ err = bpf_obj_get_info_by_fd(fd, &info, &info_len);
+ if (err) {
+ printf("can't get prog info - %s\n", strerror(errno));
+ return err;
+ }
+ prog_id = info.id;
return err;
}
static int do_detach(int idx, const char *name)
{
- int err;
+ __u32 curr_prog_id = 0;
+ int err = 0;
- err = bpf_set_link_xdp_fd(idx, -1, 0);
- if (err < 0)
- printf("ERROR: failed to detach program from %s\n", name);
+ err = bpf_get_link_xdp_id(idx, &curr_prog_id, 0);
+ if (err) {
+ printf("bpf_get_link_xdp_id failed\n");
+ return err;
+ }
+ if (prog_id == curr_prog_id) {
+ err = bpf_set_link_xdp_fd(idx, -1, 0);
+ if (err < 0)
+ printf("ERROR: failed to detach prog from %s\n", name);
+ } else if (!curr_prog_id) {
+ printf("couldn't find a prog id on a %s\n", name);
+ } else {
+ printf("program on interface changed, not removing\n");
+ }
return err;
}
diff --git a/samples/bpf/xdp_tx_iptunnel_user.c b/samples/bpf/xdp_tx_iptunnel_user.c
index e3de60930d27..4a1511eb7812 100644
--- a/samples/bpf/xdp_tx_iptunnel_user.c
+++ b/samples/bpf/xdp_tx_iptunnel_user.c
@@ -27,11 +27,24 @@
static int ifindex = -1;
static __u32 xdp_flags = XDP_FLAGS_UPDATE_IF_NOEXIST;
static int rxcnt_map_fd;
+static __u32 prog_id;
static void int_exit(int sig)
{
- if (ifindex > -1)
- bpf_set_link_xdp_fd(ifindex, -1, xdp_flags);
+ __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");
+ exit(1);
+ }
+ if (prog_id == curr_prog_id)
+ bpf_set_link_xdp_fd(ifindex, -1, xdp_flags);
+ else if (!curr_prog_id)
+ printf("couldn't find a prog id on a given iface\n");
+ else
+ printf("program on interface changed, not removing\n");
+ }
exit(0);
}
@@ -148,13 +161,15 @@ int main(int argc, char **argv)
int min_port = 0, max_port = 0, vip2tnl_map_fd;
const char *optstr = "i:a:p:s:d:m:T:P:FSNh";
unsigned char opt_flags[256] = {};
+ struct bpf_prog_info info = {};
+ __u32 info_len = sizeof(info);
unsigned int kill_after_s = 0;
struct iptnl_info tnl = {};
struct bpf_object *obj;
struct vip vip = {};
char filename[256];
int opt, prog_fd;
- int i;
+ int i, err;
tnl.family = AF_UNSPEC;
vip.protocol = IPPROTO_TCP;
@@ -276,6 +291,13 @@ int main(int argc, char **argv)
return 1;
}
+ err = bpf_obj_get_info_by_fd(prog_fd, &info, &info_len);
+ if (err) {
+ printf("can't get prog info - %s\n", strerror(errno));
+ return err;
+ }
+ prog_id = info.id;
+
poll_stats(kill_after_s);
bpf_set_link_xdp_fd(ifindex, -1, xdp_flags);
diff --git a/samples/bpf/xdpsock_user.c b/samples/bpf/xdpsock_user.c
index 188723784768..f73055e0191f 100644
--- a/samples/bpf/xdpsock_user.c
+++ b/samples/bpf/xdpsock_user.c
@@ -76,6 +76,7 @@ static int opt_poll;
static int opt_shared_packet_buffer;
static int opt_interval = 1;
static u32 opt_xdp_bind_flags;
+static __u32 prog_id;
struct xdp_umem_uqueue {
u32 cached_prod;
@@ -631,9 +632,20 @@ static void *poller(void *arg)
static void int_exit(int sig)
{
+ __u32 curr_prog_id = 0;
+
(void)sig;
dump_stats();
- bpf_set_link_xdp_fd(opt_ifindex, -1, opt_xdp_flags);
+ if (bpf_get_link_xdp_id(opt_ifindex, &curr_prog_id, opt_xdp_flags)) {
+ printf("bpf_get_link_xdp_id failed\n");
+ exit(EXIT_FAILURE);
+ }
+ if (prog_id == curr_prog_id)
+ bpf_set_link_xdp_fd(opt_ifindex, -1, opt_xdp_flags);
+ else if (!curr_prog_id)
+ printf("couldn't find a prog id on a given interface\n");
+ else
+ printf("program on interface changed, not removing\n");
exit(EXIT_SUCCESS);
}
@@ -907,6 +919,8 @@ int main(int argc, char **argv)
.prog_type = BPF_PROG_TYPE_XDP,
};
int prog_fd, qidconf_map, xsks_map;
+ struct bpf_prog_info info = {};
+ __u32 info_len = sizeof(info);
struct bpf_object *obj;
char xdp_filename[256];
struct bpf_map *map;
@@ -953,6 +967,13 @@ int main(int argc, char **argv)
exit(EXIT_FAILURE);
}
+ ret = bpf_obj_get_info_by_fd(prog_fd, &info, &info_len);
+ if (ret) {
+ printf("can't get prog info - %s\n", strerror(errno));
+ return 1;
+ }
+ prog_id = info.id;
+
ret = bpf_map_update_elem(qidconf_map, &key, &opt_queue, 0);
if (ret) {
fprintf(stderr, "ERROR: bpf_map_update_elem qidconf\n");
--
2.16.1
^ permalink raw reply related [flat|nested] 22+ messages in thread
* Re: [PATCH bpf-next v5 5/8] xdp: Provide extack messages when prog attachment failed
2019-02-01 0:19 ` [PATCH bpf-next v5 5/8] xdp: Provide extack messages when prog attachment failed Maciej Fijalkowski
@ 2019-02-01 0:32 ` Jakub Kicinski
2019-02-01 3:11 ` Jakub Kicinski
2019-02-01 7:04 ` [PATCH bpf-next v5 5/8] xdp: Provide extack messages when prog attachment failed Jesper Dangaard Brouer
2 siblings, 0 replies; 22+ messages in thread
From: Jakub Kicinski @ 2019-02-01 0:32 UTC (permalink / raw)
To: Maciej Fijalkowski; +Cc: daniel, ast, netdev, brouer, john.fastabend
On Fri, 1 Feb 2019 01:19:51 +0100, Maciej Fijalkowski wrote:
> In order to provide more meaningful messages to user when the process of
> loading xdp program onto network interface failed, let's add extack
> messages within dev_change_xdp_fd.
>
> Suggested-by: Jakub Kicinski <jakub.kicinski@netronome.com>
> Signed-off-by: Maciej Fijalkowski <maciej.fijalkowski@intel.com>
Reviewed-by: Jakub Kicinski <jakub.kicinski@netronome.com>
^ permalink raw reply [flat|nested] 22+ messages in thread
* Re: [PATCH bpf-next v5 5/8] xdp: Provide extack messages when prog attachment failed
2019-02-01 0:19 ` [PATCH bpf-next v5 5/8] xdp: Provide extack messages when prog attachment failed Maciej Fijalkowski
2019-02-01 0:32 ` Jakub Kicinski
@ 2019-02-01 3:11 ` Jakub Kicinski
2019-02-01 7:02 ` Co-existing XDP generic and native mode? (Re: [PATCH bpf-next v5 5/8] xdp: Provide extack messages when prog attachment failed) Jesper Dangaard Brouer
2019-02-01 7:04 ` [PATCH bpf-next v5 5/8] xdp: Provide extack messages when prog attachment failed Jesper Dangaard Brouer
2 siblings, 1 reply; 22+ messages in thread
From: Jakub Kicinski @ 2019-02-01 3:11 UTC (permalink / raw)
To: daniel, ast, David Miller
Cc: Maciej Fijalkowski, netdev, brouer, john.fastabend, David Ahern
On Fri, 1 Feb 2019 01:19:51 +0100, Maciej Fijalkowski wrote:
> if (__dev_xdp_query(dev, bpf_chk, XDP_QUERY_PROG) ||
> - __dev_xdp_query(dev, bpf_chk, XDP_QUERY_PROG_HW))
> + __dev_xdp_query(dev, bpf_chk, XDP_QUERY_PROG_HW)) {
> + NL_SET_ERR_MSG(extack, "native and generic XDP can't be active at the same time");
> return -EEXIST;
> + }
This reminds me, since we allowed native/driver and offloaded XDP
programs to coexist in a25717d2b604 ("xdp: support simultaneous
driver and hw XDP attachment") I got an internal feature request
to also allow generic and native mode. Would anyone object to that?
Apart from a touch up to test_offload.py I don't think anything
would care. netlink can already carry multiple IDs, iproute2
understands it, too..
(Obviously as a follow up after this set gets merged.)
^ permalink raw reply [flat|nested] 22+ messages in thread
* Co-existing XDP generic and native mode? (Re: [PATCH bpf-next v5 5/8] xdp: Provide extack messages when prog attachment failed)
2019-02-01 3:11 ` Jakub Kicinski
@ 2019-02-01 7:02 ` Jesper Dangaard Brouer
2019-02-01 18:47 ` Jakub Kicinski
0 siblings, 1 reply; 22+ messages in thread
From: Jesper Dangaard Brouer @ 2019-02-01 7:02 UTC (permalink / raw)
To: Jakub Kicinski
Cc: daniel, ast, David Miller, Maciej Fijalkowski, netdev,
john.fastabend, David Ahern, brouer, Saeed Mahameed
On Thu, 31 Jan 2019 19:11:01 -0800
Jakub Kicinski <jakub.kicinski@netronome.com> wrote:
> On Fri, 1 Feb 2019 01:19:51 +0100, Maciej Fijalkowski wrote:
> > if (__dev_xdp_query(dev, bpf_chk, XDP_QUERY_PROG) ||
> > - __dev_xdp_query(dev, bpf_chk, XDP_QUERY_PROG_HW))
> > + __dev_xdp_query(dev, bpf_chk, XDP_QUERY_PROG_HW)) {
> > + NL_SET_ERR_MSG(extack, "native and generic XDP can't be active at the same time");
> > return -EEXIST;
> > + }
>
> This reminds me, since we allowed native/driver and offloaded XDP
> programs to coexist in a25717d2b604 ("xdp: support simultaneous
> driver and hw XDP attachment") I got an internal feature request
> to also allow generic and native mode. Would anyone object to that?
Well, I will object ;-)
I have two refactor ideas [1] and [2], that depend on not allowing
XDP-native and XDP-generic to co-exist. The general idea is to let
XDP-native use the same fields in net_device->rx[] as XDP-generic given
they (currently) cannot co-exist.
The goal is (1) to move stuff out of driver code, and (2) hopefully
make it easier to implement per RXq XDP progs.
These are only refactor ideas, so if you can argue why your internal
feature request for simultaneous generic and native make more sense,
then I'm open for allowing this ?
[1] https://github.com/xdp-project/xdp-project/blob/master/areas/core/xdp_per_rxq01.org#refactor-idea-move-xdp_rxq_info-to-net_devicenetdev_rx_queue
[2] https://github.com/xdp-project/xdp-project/blob/master/areas/core/xdp_per_rxq01.org#refactor-idea-xdpbpf_prog-into-netdev_rx_queuenet_device
> Apart from a touch up to test_offload.py I don't think anything
> would care. netlink can already carry multiple IDs, iproute2
> understands it, too..
And we did notice you added support for HW+native:
[3] https://github.com/xdp-project/xdp-project/blob/master/areas/core/xdp_per_rxq01.org
--
Best regards,
Jesper Dangaard Brouer
MSc.CS, Principal Kernel Engineer at Red Hat
LinkedIn: http://www.linkedin.com/in/brouer
^ permalink raw reply [flat|nested] 22+ messages in thread
* Re: [PATCH bpf-next v5 5/8] xdp: Provide extack messages when prog attachment failed
2019-02-01 0:19 ` [PATCH bpf-next v5 5/8] xdp: Provide extack messages when prog attachment failed Maciej Fijalkowski
2019-02-01 0:32 ` Jakub Kicinski
2019-02-01 3:11 ` Jakub Kicinski
@ 2019-02-01 7:04 ` Jesper Dangaard Brouer
2 siblings, 0 replies; 22+ messages in thread
From: Jesper Dangaard Brouer @ 2019-02-01 7:04 UTC (permalink / raw)
To: Maciej Fijalkowski
Cc: daniel, ast, netdev, jakub.kicinski, john.fastabend, brouer
On Fri, 1 Feb 2019 01:19:51 +0100
Maciej Fijalkowski <maciejromanfijalkowski@gmail.com> wrote:
> In order to provide more meaningful messages to user when the process of
> loading xdp program onto network interface failed, let's add extack
> messages within dev_change_xdp_fd.
>
> Suggested-by: Jakub Kicinski <jakub.kicinski@netronome.com>
> Signed-off-by: Maciej Fijalkowski <maciej.fijalkowski@intel.com>
Acked-by: Jesper Dangaard Brouer <brouer@redhat.com>
--
Best regards,
Jesper Dangaard Brouer
MSc.CS, Principal Kernel Engineer at Red Hat
LinkedIn: http://www.linkedin.com/in/brouer
^ permalink raw reply [flat|nested] 22+ messages in thread
* Re: Co-existing XDP generic and native mode? (Re: [PATCH bpf-next v5 5/8] xdp: Provide extack messages when prog attachment failed)
2019-02-01 7:02 ` Co-existing XDP generic and native mode? (Re: [PATCH bpf-next v5 5/8] xdp: Provide extack messages when prog attachment failed) Jesper Dangaard Brouer
@ 2019-02-01 18:47 ` Jakub Kicinski
2019-02-01 21:33 ` Daniel Borkmann
0 siblings, 1 reply; 22+ messages in thread
From: Jakub Kicinski @ 2019-02-01 18:47 UTC (permalink / raw)
To: Jesper Dangaard Brouer
Cc: daniel, ast, David Miller, Maciej Fijalkowski, netdev,
john.fastabend, David Ahern, Saeed Mahameed
On Fri, 1 Feb 2019 08:02:36 +0100, Jesper Dangaard Brouer wrote:
> On Thu, 31 Jan 2019 19:11:01 -0800
> Jakub Kicinski <jakub.kicinski@netronome.com> wrote:
>
> > On Fri, 1 Feb 2019 01:19:51 +0100, Maciej Fijalkowski wrote:
> > > if (__dev_xdp_query(dev, bpf_chk, XDP_QUERY_PROG) ||
> > > - __dev_xdp_query(dev, bpf_chk, XDP_QUERY_PROG_HW))
> > > + __dev_xdp_query(dev, bpf_chk, XDP_QUERY_PROG_HW)) {
> > > + NL_SET_ERR_MSG(extack, "native and generic XDP can't be active at the same time");
> > > return -EEXIST;
> > > + }
> >
> > This reminds me, since we allowed native/driver and offloaded XDP
> > programs to coexist in a25717d2b604 ("xdp: support simultaneous
> > driver and hw XDP attachment") I got an internal feature request
> > to also allow generic and native mode. Would anyone object to that?
>
> Well, I will object ;-)
>
> I have two refactor ideas [1] and [2], that depend on not allowing
> XDP-native and XDP-generic to co-exist. The general idea is to let
> XDP-native use the same fields in net_device->rx[] as XDP-generic given
> they (currently) cannot co-exist.
> The goal is (1) to move stuff out of driver code, and (2) hopefully
> make it easier to implement per RXq XDP progs.
You mean you'd use one pointer to keep the prog in the RXQ structure?
Then some from from of an extra flag will be necessary to distinguish?
I.e.:
if (rxq->prog && rxq->is_native)
/* got_prog */
rather than:
if (rxq->native_prog)
/* got_prog */
The cost of this reuse would be a read-only cache line per-q when XDP is
not enabled. Right now drivers have the ability to pack the XDP prog
into a structure which is in cache already, and don't need to bring the
entire RXQ structure out (which is cache line aligned so driver authors
can't do anything to place it cleverly).
No doubt, thought, that if we allow both to be enabled we will have to
bloat the data structures.
> These are only refactor ideas, so if you can argue why your internal
> feature request for simultaneous generic and native make more sense,
> then I'm open for allowing this ?
The request was actually to enable xdpoffload and xdpgeneric at the
same time. I'm happy to have that as another HW offload exclusive
for now :)
> [1] https://github.com/xdp-project/xdp-project/blob/master/areas/core/xdp_per_rxq01.org#refactor-idea-move-xdp_rxq_info-to-net_devicenetdev_rx_queue
>
> [2] https://github.com/xdp-project/xdp-project/blob/master/areas/core/xdp_per_rxq01.org#refactor-idea-xdpbpf_prog-into-netdev_rx_queuenet_device
>
> > Apart from a touch up to test_offload.py I don't think anything
> > would care. netlink can already carry multiple IDs, iproute2
> > understands it, too..
>
> And we did notice you added support for HW+native:
> [3] https://github.com/xdp-project/xdp-project/blob/master/areas/core/xdp_per_rxq01.org
^ permalink raw reply [flat|nested] 22+ messages in thread
* Re: [PATCH bpf-next v5 0/8] xdp: Avoid unloading xdp prog not attached by sample
2019-02-01 0:19 [PATCH bpf-next v5 0/8] xdp: Avoid unloading xdp prog not attached by sample Maciej Fijalkowski
` (7 preceding siblings ...)
2019-02-01 0:19 ` [PATCH bpf-next v5 8/8] samples/bpf: Check the prog id before exiting Maciej Fijalkowski
@ 2019-02-01 21:23 ` Daniel Borkmann
2019-02-01 21:47 ` Maciej Fijałkowski
8 siblings, 1 reply; 22+ messages in thread
From: Daniel Borkmann @ 2019-02-01 21:23 UTC (permalink / raw)
To: Maciej Fijalkowski, ast; +Cc: netdev, jakub.kicinski, brouer, john.fastabend
On 02/01/2019 01:19 AM, Maciej Fijalkowski wrote:
> Hi!
> This patchset tries to address the situation where:
> * user loads a particular xdp sample application that does stats polling
> * user loads another sample application on the same interface
> * then, user sends SIGINT/SIGTERM to the app that was attached as a first one
> * second application ends up with an unloaded xdp program
>
> 1st patch contains a helper libbpf function for getting the map fd by a
> given map name.
> In patch 2 Jesper removes the read_trace_pipe usage from xdp_redirect_cpu which
> was a blocker for converting this sample to libbpf usage.
> 3rd patch updates a bunch of xdp samples to make the use of libbpf.
> Patch 4 adjusts RLIMIT_MEMLOCK for two samples touched in this patchset.
> In patch 5 extack messages are added for cases where dev_change_xdp_fd returns
> with an error so user has an idea what was the reason for not attaching the
> xdp program onto interface.
> Patch 6 makes the samples behavior similar to what iproute2 does when loading
> xdp prog - the "force" flag is introduced.
> Patch 7 introduces the libbpf function that will query the driver from
> userspace about the currently attached xdp prog id.
>
> Use it in samples that do polling by checking the prog id in signal handler
> and comparing it with previously stored one which is the scope of patch 8.
>
> Thanks!
>
> v1->v2:
> * add a libbpf helper for getting a prog via relative index
> * include xdp_redirect_cpu into conversion
>
> v2->v3: mostly addressing Daniel's/Jesper's comments
> * get rid of the helper from v1->v2
> * feed the xdp_redirect_cpu with program name instead of number
>
> v3->v4:
> * fix help message in xdp_sample_pkts
>
> v4->v5:
> * in get_link_xdp_fd, assign prog_id only when libbpf_nl_get_link returned
> with 0
> * add extack messages in dev_change_xdp_fd
> * check the return value of bpf_get_link_xdp_id when exiting from sample progs
Series looks good to me, but doesn't apply cleanly, please rebase.
Thanks,
Daniel
^ permalink raw reply [flat|nested] 22+ messages in thread
* Re: Co-existing XDP generic and native mode? (Re: [PATCH bpf-next v5 5/8] xdp: Provide extack messages when prog attachment failed)
2019-02-01 18:47 ` Jakub Kicinski
@ 2019-02-01 21:33 ` Daniel Borkmann
2019-02-01 21:44 ` Jakub Kicinski
0 siblings, 1 reply; 22+ messages in thread
From: Daniel Borkmann @ 2019-02-01 21:33 UTC (permalink / raw)
To: Jakub Kicinski, Jesper Dangaard Brouer
Cc: ast, David Miller, Maciej Fijalkowski, netdev, john.fastabend,
David Ahern, Saeed Mahameed
On 02/01/2019 07:47 PM, Jakub Kicinski wrote:
[...]
>> These are only refactor ideas, so if you can argue why your internal
>> feature request for simultaneous generic and native make more sense,
>> then I'm open for allowing this ?
>
> The request was actually to enable xdpoffload and xdpgeneric at the
> same time. I'm happy to have that as another HW offload exclusive
> for now :)
The latter is probably fine, though what's the concrete use case? :)
Reason we kept native vs generic separate is mainly so that native XDP
drivers are discouraged to punt missing features to generic hook instead
of properly implementing them in native mode.
Thanks,
Daniel
^ permalink raw reply [flat|nested] 22+ messages in thread
* Re: [PATCH bpf-next v5 7/8] libbpf: Add a support for getting xdp prog id on ifindex
2019-02-01 0:19 ` [PATCH bpf-next v5 7/8] libbpf: Add a support for getting xdp prog id on ifindex Maciej Fijalkowski
@ 2019-02-01 21:43 ` Daniel Borkmann
2019-02-01 21:47 ` Jakub Kicinski
0 siblings, 1 reply; 22+ messages in thread
From: Daniel Borkmann @ 2019-02-01 21:43 UTC (permalink / raw)
To: Maciej Fijalkowski, ast; +Cc: netdev, jakub.kicinski, brouer, john.fastabend
On 02/01/2019 01:19 AM, Maciej Fijalkowski wrote:
> Since we have a dedicated netlink attributes for xdp setup on a
> particular interface, it is now possible to retrieve the program id that
> is currently attached to the interface. The use case is targeted for
> sample xdp programs, which will store the program id just after loading
> bpf program onto iface. On shutdown, the sample will make sure that it
> can unload the program by querying again the iface and verifying that
> both program id's matches.
>
> Signed-off-by: Maciej Fijalkowski <maciej.fijalkowski@intel.com>
> Reviewed-by: Jakub Kicinski <jakub.kicinski@netronome.com>
[...]
> +int bpf_get_link_xdp_id(int ifindex, __u32 *prog_id, __u32 flags)
> +{
> + struct xdp_id_md xdp_id = {};
> + int sock, ret;
> + __u32 nl_pid;
> + __u32 mask;
> +
> + if (flags & ~XDP_FLAGS_MASK)
> + return -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)
> + return -EINVAL;
> +
> + sock = libbpf_netlink_open(&nl_pid);
> + if (sock < 0)
> + return sock;
> +
> + xdp_id.ifindex = ifindex;
> + xdp_id.flags = flags;
> +
> + ret = libbpf_nl_get_link(sock, nl_pid, get_xdp_id, &xdp_id);
> + if (!ret)
> + *prog_id = xdp_id.id;
> +
> + close(sock);
> + return ret;
> +}
Btw, is anyone going to follow-up on XDP_ATTACHED_MULTI support as well
later on?
Thanks,
Daniel
^ permalink raw reply [flat|nested] 22+ messages in thread
* Re: Co-existing XDP generic and native mode? (Re: [PATCH bpf-next v5 5/8] xdp: Provide extack messages when prog attachment failed)
2019-02-01 21:33 ` Daniel Borkmann
@ 2019-02-01 21:44 ` Jakub Kicinski
2019-02-01 23:14 ` Daniel Borkmann
0 siblings, 1 reply; 22+ messages in thread
From: Jakub Kicinski @ 2019-02-01 21:44 UTC (permalink / raw)
To: Daniel Borkmann
Cc: Jesper Dangaard Brouer, ast, David Miller, Maciej Fijalkowski,
netdev, john.fastabend, David Ahern, Saeed Mahameed
On Fri, 1 Feb 2019 22:33:22 +0100, Daniel Borkmann wrote:
> On 02/01/2019 07:47 PM, Jakub Kicinski wrote:
> >> These are only refactor ideas, so if you can argue why your internal
> >> feature request for simultaneous generic and native make more sense,
> >> then I'm open for allowing this ?
> >
> > The request was actually to enable xdpoffload and xdpgeneric at the
> > same time. I'm happy to have that as another HW offload exclusive
> > for now :)
>
> The latter is probably fine, though what's the concrete use case? :)
I think it was more of a "I expected this to work, since driver+offload
worked" than a feature request. Looking back at it it was filed as bug
I converted it to feature.
> Reason we kept native vs generic separate is mainly so that native XDP
> drivers are discouraged to punt missing features to generic hook instead
> of properly implementing them in native mode.
Agreed, although one could make a counter argument that the
performance should be a strong enough incentive and we shouldn't stop
people for experimenting and prototyping :) The code change looks
simple enough:
diff --git a/net/core/dev.c b/net/core/dev.c
index 8e276e0192a1..ce4880e5e95d 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -7976,11 +7976,13 @@ int dev_change_xdp_fd(struct net_device *dev, struct netlink_ext_ack *extack,
enum bpf_netdev_command query;
struct bpf_prog *prog = NULL;
bpf_op_t bpf_op, bpf_chk;
+ bool offload;
int err;
ASSERT_RTNL();
- query = flags & XDP_FLAGS_HW_MODE ? XDP_QUERY_PROG_HW : XDP_QUERY_PROG;
+ offload = flags & XDP_FLAGS_HW_MODE;
+ query = offload ? XDP_QUERY_PROG_HW : XDP_QUERY_PROG;
bpf_op = bpf_chk = ops->ndo_bpf;
if (!bpf_op && (flags & (XDP_FLAGS_DRV_MODE | XDP_FLAGS_HW_MODE)))
@@ -7991,8 +7993,7 @@ int dev_change_xdp_fd(struct net_device *dev, struct netlink_ext_ack *extack,
bpf_chk = generic_xdp_install;
if (fd >= 0) {
- if (__dev_xdp_query(dev, bpf_chk, XDP_QUERY_PROG) ||
- __dev_xdp_query(dev, bpf_chk, XDP_QUERY_PROG_HW))
+ if (!offload && __dev_xdp_query(dev, bpf_chk, XDP_QUERY_PROG))
return -EEXIST;
if ((flags & XDP_FLAGS_UPDATE_IF_NOEXIST) &&
__dev_xdp_query(dev, bpf_op, query))
@@ -8003,8 +8004,7 @@ int dev_change_xdp_fd(struct net_device *dev, struct netlink_ext_ack *extack,
if (IS_ERR(prog))
return PTR_ERR(prog);
- if (!(flags & XDP_FLAGS_HW_MODE) &&
- bpf_prog_is_dev_bound(prog->aux)) {
+ if (!offload && bpf_prog_is_dev_bound(prog->aux)) {
NL_SET_ERR_MSG(extack, "using device-bound program without HW_MODE flag is not supported");
bpf_prog_put(prog);
return -EINVAL;
Do you think we shouldn't do it?
^ permalink raw reply related [flat|nested] 22+ messages in thread
* Re: [PATCH bpf-next v5 0/8] xdp: Avoid unloading xdp prog not attached by sample
2019-02-01 21:23 ` [PATCH bpf-next v5 0/8] xdp: Avoid unloading xdp prog not attached by sample Daniel Borkmann
@ 2019-02-01 21:47 ` Maciej Fijałkowski
0 siblings, 0 replies; 22+ messages in thread
From: Maciej Fijałkowski @ 2019-02-01 21:47 UTC (permalink / raw)
To: Daniel Borkmann; +Cc: ast, netdev, jakub.kicinski, brouer, john.fastabend
On Fri, 1 Feb 2019 22:23:45 +0100
Daniel Borkmann <daniel@iogearbox.net> wrote:
> On 02/01/2019 01:19 AM, Maciej Fijalkowski wrote:
> > Hi!
> > This patchset tries to address the situation where:
> > * user loads a particular xdp sample application that does stats polling
> > * user loads another sample application on the same interface
> > * then, user sends SIGINT/SIGTERM to the app that was attached as a first
> > one
> > * second application ends up with an unloaded xdp program
> >
> > 1st patch contains a helper libbpf function for getting the map fd by a
> > given map name.
> > In patch 2 Jesper removes the read_trace_pipe usage from xdp_redirect_cpu
> > which was a blocker for converting this sample to libbpf usage.
> > 3rd patch updates a bunch of xdp samples to make the use of libbpf.
> > Patch 4 adjusts RLIMIT_MEMLOCK for two samples touched in this patchset.
> > In patch 5 extack messages are added for cases where dev_change_xdp_fd
> > returns with an error so user has an idea what was the reason for not
> > attaching the xdp program onto interface.
> > Patch 6 makes the samples behavior similar to what iproute2 does when
> > loading xdp prog - the "force" flag is introduced.
> > Patch 7 introduces the libbpf function that will query the driver from
> > userspace about the currently attached xdp prog id.
> >
> > Use it in samples that do polling by checking the prog id in signal handler
> > and comparing it with previously stored one which is the scope of patch 8.
> >
> > Thanks!
> >
> > v1->v2:
> > * add a libbpf helper for getting a prog via relative index
> > * include xdp_redirect_cpu into conversion
> >
> > v2->v3: mostly addressing Daniel's/Jesper's comments
> > * get rid of the helper from v1->v2
> > * feed the xdp_redirect_cpu with program name instead of number
> >
> > v3->v4:
> > * fix help message in xdp_sample_pkts
> >
> > v4->v5:
> > * in get_link_xdp_fd, assign prog_id only when libbpf_nl_get_link returned
> > with 0
> > * add extack messages in dev_change_xdp_fd
> > * check the return value of bpf_get_link_xdp_id when exiting from sample
> > progs
>
> Series looks good to me, but doesn't apply cleanly, please rebase.
>
> Thanks,
> Daniel
Sure, sending v6 in a minute.
^ permalink raw reply [flat|nested] 22+ messages in thread
* Re: [PATCH bpf-next v5 7/8] libbpf: Add a support for getting xdp prog id on ifindex
2019-02-01 21:43 ` Daniel Borkmann
@ 2019-02-01 21:47 ` Jakub Kicinski
2019-02-01 23:00 ` Daniel Borkmann
0 siblings, 1 reply; 22+ messages in thread
From: Jakub Kicinski @ 2019-02-01 21:47 UTC (permalink / raw)
To: Daniel Borkmann; +Cc: Maciej Fijalkowski, ast, netdev, brouer, john.fastabend
On Fri, 1 Feb 2019 22:43:39 +0100, Daniel Borkmann wrote:
> On 02/01/2019 01:19 AM, Maciej Fijalkowski wrote:
> > Since we have a dedicated netlink attributes for xdp setup on a
> > particular interface, it is now possible to retrieve the program id that
> > is currently attached to the interface. The use case is targeted for
> > sample xdp programs, which will store the program id just after loading
> > bpf program onto iface. On shutdown, the sample will make sure that it
> > can unload the program by querying again the iface and verifying that
> > both program id's matches.
> >
> > Signed-off-by: Maciej Fijalkowski <maciej.fijalkowski@intel.com>
> > Reviewed-by: Jakub Kicinski <jakub.kicinski@netronome.com>
> [...]
> > +int bpf_get_link_xdp_id(int ifindex, __u32 *prog_id, __u32 flags)
> > +{
> > + struct xdp_id_md xdp_id = {};
> > + int sock, ret;
> > + __u32 nl_pid;
> > + __u32 mask;
> > +
> > + if (flags & ~XDP_FLAGS_MASK)
> > + return -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)
> > + return -EINVAL;
> > +
> > + sock = libbpf_netlink_open(&nl_pid);
> > + if (sock < 0)
> > + return sock;
> > +
> > + xdp_id.ifindex = ifindex;
> > + xdp_id.flags = flags;
> > +
> > + ret = libbpf_nl_get_link(sock, nl_pid, get_xdp_id, &xdp_id);
> > + if (!ret)
> > + *prog_id = xdp_id.id;
> > +
> > + close(sock);
> > + return ret;
> > +}
>
> Btw, is anyone going to follow-up on XDP_ATTACHED_MULTI support as well
> later on?
I haven't tested to be honest, but I think Maciek got that right -
get_xdp_id_attr() should return IFLA_XDP_PROG_ID or a mode-specific
attr based on flags. And there is a check that only flag is set.
Or do you mean retrieving all program ids with one dump?
^ permalink raw reply [flat|nested] 22+ messages in thread
* Re: [PATCH bpf-next v5 7/8] libbpf: Add a support for getting xdp prog id on ifindex
2019-02-01 21:47 ` Jakub Kicinski
@ 2019-02-01 23:00 ` Daniel Borkmann
0 siblings, 0 replies; 22+ messages in thread
From: Daniel Borkmann @ 2019-02-01 23:00 UTC (permalink / raw)
To: Jakub Kicinski; +Cc: Maciej Fijalkowski, ast, netdev, brouer, john.fastabend
On 02/01/2019 10:47 PM, Jakub Kicinski wrote:
> On Fri, 1 Feb 2019 22:43:39 +0100, Daniel Borkmann wrote:
>> On 02/01/2019 01:19 AM, Maciej Fijalkowski wrote:
>>> Since we have a dedicated netlink attributes for xdp setup on a
>>> particular interface, it is now possible to retrieve the program id that
>>> is currently attached to the interface. The use case is targeted for
>>> sample xdp programs, which will store the program id just after loading
>>> bpf program onto iface. On shutdown, the sample will make sure that it
>>> can unload the program by querying again the iface and verifying that
>>> both program id's matches.
>>>
>>> Signed-off-by: Maciej Fijalkowski <maciej.fijalkowski@intel.com>
>>> Reviewed-by: Jakub Kicinski <jakub.kicinski@netronome.com>
>> [...]
>>> +int bpf_get_link_xdp_id(int ifindex, __u32 *prog_id, __u32 flags)
>>> +{
>>> + struct xdp_id_md xdp_id = {};
>>> + int sock, ret;
>>> + __u32 nl_pid;
>>> + __u32 mask;
>>> +
>>> + if (flags & ~XDP_FLAGS_MASK)
>>> + return -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)
>>> + return -EINVAL;
>>> +
>>> + sock = libbpf_netlink_open(&nl_pid);
>>> + if (sock < 0)
>>> + return sock;
>>> +
>>> + xdp_id.ifindex = ifindex;
>>> + xdp_id.flags = flags;
>>> +
>>> + ret = libbpf_nl_get_link(sock, nl_pid, get_xdp_id, &xdp_id);
>>> + if (!ret)
>>> + *prog_id = xdp_id.id;
>>> +
>>> + close(sock);
>>> + return ret;
>>> +}
>>
>> Btw, is anyone going to follow-up on XDP_ATTACHED_MULTI support as well
>> later on?
>
> I haven't tested to be honest, but I think Maciek got that right -
> get_xdp_id_attr() should return IFLA_XDP_PROG_ID or a mode-specific
> attr based on flags. And there is a check that only flag is set.
>
> Or do you mean retrieving all program ids with one dump?
Yeah was thinking about the latter, but agree it's fine and probably
cleaner this way here.
^ permalink raw reply [flat|nested] 22+ messages in thread
* Re: Co-existing XDP generic and native mode? (Re: [PATCH bpf-next v5 5/8] xdp: Provide extack messages when prog attachment failed)
2019-02-01 21:44 ` Jakub Kicinski
@ 2019-02-01 23:14 ` Daniel Borkmann
0 siblings, 0 replies; 22+ messages in thread
From: Daniel Borkmann @ 2019-02-01 23:14 UTC (permalink / raw)
To: Jakub Kicinski
Cc: Jesper Dangaard Brouer, ast, David Miller, Maciej Fijalkowski,
netdev, john.fastabend, David Ahern, Saeed Mahameed
On 02/01/2019 10:44 PM, Jakub Kicinski wrote:
[...]
> diff --git a/net/core/dev.c b/net/core/dev.c
> index 8e276e0192a1..ce4880e5e95d 100644
> --- a/net/core/dev.c
> +++ b/net/core/dev.c
> @@ -7976,11 +7976,13 @@ int dev_change_xdp_fd(struct net_device *dev, struct netlink_ext_ack *extack,
> enum bpf_netdev_command query;
> struct bpf_prog *prog = NULL;
> bpf_op_t bpf_op, bpf_chk;
> + bool offload;
> int err;
>
> ASSERT_RTNL();
>
> - query = flags & XDP_FLAGS_HW_MODE ? XDP_QUERY_PROG_HW : XDP_QUERY_PROG;
> + offload = flags & XDP_FLAGS_HW_MODE;
> + query = offload ? XDP_QUERY_PROG_HW : XDP_QUERY_PROG;
>
> bpf_op = bpf_chk = ops->ndo_bpf;
> if (!bpf_op && (flags & (XDP_FLAGS_DRV_MODE | XDP_FLAGS_HW_MODE)))
> @@ -7991,8 +7993,7 @@ int dev_change_xdp_fd(struct net_device *dev, struct netlink_ext_ack *extack,
> bpf_chk = generic_xdp_install;
>
> if (fd >= 0) {
> - if (__dev_xdp_query(dev, bpf_chk, XDP_QUERY_PROG) ||
> - __dev_xdp_query(dev, bpf_chk, XDP_QUERY_PROG_HW))
> + if (!offload && __dev_xdp_query(dev, bpf_chk, XDP_QUERY_PROG))
> return -EEXIST;
> if ((flags & XDP_FLAGS_UPDATE_IF_NOEXIST) &&
> __dev_xdp_query(dev, bpf_op, query))
> @@ -8003,8 +8004,7 @@ int dev_change_xdp_fd(struct net_device *dev, struct netlink_ext_ack *extack,
> if (IS_ERR(prog))
> return PTR_ERR(prog);
>
> - if (!(flags & XDP_FLAGS_HW_MODE) &&
> - bpf_prog_is_dev_bound(prog->aux)) {
> + if (!offload && bpf_prog_is_dev_bound(prog->aux)) {
> NL_SET_ERR_MSG(extack, "using device-bound program without HW_MODE flag is not supported");
> bpf_prog_put(prog);
> return -EINVAL;
>
> Do you think we shouldn't do it?
Yeah that looks good to me, lets do it. At least this would allow for
prototyping in combination with HW mode till full support of a feature
has landed properly in native mode.
Thanks,
Daniel
^ permalink raw reply [flat|nested] 22+ messages in thread
end of thread, other threads:[~2019-02-01 23:14 UTC | newest]
Thread overview: 22+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2019-02-01 0:19 [PATCH bpf-next v5 0/8] xdp: Avoid unloading xdp prog not attached by sample Maciej Fijalkowski
2019-02-01 0:19 ` [PATCH bpf-next v5 1/8] libbpf: Add a helper for retrieving a map fd for a given name Maciej Fijalkowski
2019-02-01 0:19 ` [PATCH bpf-next v5 2/8] samples/bpf: xdp_redirect_cpu have not need for read_trace_pipe Maciej Fijalkowski
2019-02-01 0:19 ` [PATCH bpf-next v5 3/8] samples/bpf: Convert XDP samples to libbpf usage Maciej Fijalkowski
2019-02-01 0:19 ` [PATCH bpf-next v5 4/8] samples/bpf: Extend RLIMIT_MEMLOCK for xdp_{sample_pkts, router_ipv4} Maciej Fijalkowski
2019-02-01 0:19 ` [PATCH bpf-next v5 5/8] xdp: Provide extack messages when prog attachment failed Maciej Fijalkowski
2019-02-01 0:32 ` Jakub Kicinski
2019-02-01 3:11 ` Jakub Kicinski
2019-02-01 7:02 ` Co-existing XDP generic and native mode? (Re: [PATCH bpf-next v5 5/8] xdp: Provide extack messages when prog attachment failed) Jesper Dangaard Brouer
2019-02-01 18:47 ` Jakub Kicinski
2019-02-01 21:33 ` Daniel Borkmann
2019-02-01 21:44 ` Jakub Kicinski
2019-02-01 23:14 ` Daniel Borkmann
2019-02-01 7:04 ` [PATCH bpf-next v5 5/8] xdp: Provide extack messages when prog attachment failed Jesper Dangaard Brouer
2019-02-01 0:19 ` [PATCH bpf-next v5 6/8] samples/bpf: Add a "force" flag to XDP samples Maciej Fijalkowski
2019-02-01 0:19 ` [PATCH bpf-next v5 7/8] libbpf: Add a support for getting xdp prog id on ifindex Maciej Fijalkowski
2019-02-01 21:43 ` Daniel Borkmann
2019-02-01 21:47 ` Jakub Kicinski
2019-02-01 23:00 ` Daniel Borkmann
2019-02-01 0:19 ` [PATCH bpf-next v5 8/8] samples/bpf: Check the prog id before exiting Maciej Fijalkowski
2019-02-01 21:23 ` [PATCH bpf-next v5 0/8] xdp: Avoid unloading xdp prog not attached by sample Daniel Borkmann
2019-02-01 21:47 ` Maciej Fijałkowski
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).