bpf: mini eBPF library, test stubs and verifier testsuite
1.
the library includes a trivial set of BPF syscall wrappers:
int bpf_create_map(int key_size, int value_size, int max_entries);
int bpf_update_elem(int fd, void *key, void *value);
int bpf_lookup_elem(int fd, void *key, void *value);
int bpf_delete_elem(int fd, void *key);
int bpf_get_next_key(int fd, void *key, void *next_key);
int bpf_prog_load(enum bpf_prog_type prog_type,
const struct sock_filter_int *insns, int insn_len,
const char *license);
bpf_prog_load() stores verifier log into global bpf_log_buf[] array
and BPF_*() macros to build instructions
2.
test stubs configure eBPF infra with 'unspec' map and program types.
These are fake types used by user space testsuite only.
3.
verifier tests valid and invalid programs and expects predefined
error log messages from kernel.
40 tests so far.
$ sudo ./test_verifier
#0 add+sub+mul OK
#1 unreachable OK
#2 unreachable2 OK
#3 out of range jump OK
#4 out of range jump2 OK
#5 test1 ld_imm64 OK
...
Signed-off-by: Alexei Starovoitov <ast@plumgrid.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2014-09-26 15:17:07 +08:00
|
|
|
# kbuild trick to avoid linker error. Can be omitted if a module is built.
|
|
|
|
obj- := dummy.o
|
|
|
|
|
|
|
|
# List of programs to build
|
2016-11-12 02:55:11 +08:00
|
|
|
hostprogs-y := test_lru_dist
|
2014-12-02 07:06:36 +08:00
|
|
|
hostprogs-y += sock_example
|
bpf: add sample usages for persistent maps/progs
This patch adds a couple of stand-alone examples on how BPF_OBJ_PIN
and BPF_OBJ_GET commands can be used.
Example with maps:
# ./fds_example -F /sys/fs/bpf/m -P -m -k 1 -v 42
bpf: map fd:3 (Success)
bpf: pin ret:(0,Success)
bpf: fd:3 u->(1:42) ret:(0,Success)
# ./fds_example -F /sys/fs/bpf/m -G -m -k 1
bpf: get fd:3 (Success)
bpf: fd:3 l->(1):42 ret:(0,Success)
# ./fds_example -F /sys/fs/bpf/m -G -m -k 1 -v 24
bpf: get fd:3 (Success)
bpf: fd:3 u->(1:24) ret:(0,Success)
# ./fds_example -F /sys/fs/bpf/m -G -m -k 1
bpf: get fd:3 (Success)
bpf: fd:3 l->(1):24 ret:(0,Success)
# ./fds_example -F /sys/fs/bpf/m2 -P -m
bpf: map fd:3 (Success)
bpf: pin ret:(0,Success)
# ./fds_example -F /sys/fs/bpf/m2 -G -m -k 1
bpf: get fd:3 (Success)
bpf: fd:3 l->(1):0 ret:(0,Success)
# ./fds_example -F /sys/fs/bpf/m2 -G -m
bpf: get fd:3 (Success)
Example with progs:
# ./fds_example -F /sys/fs/bpf/p -P -p
bpf: prog fd:3 (Success)
bpf: pin ret:(0,Success)
bpf sock:4 <- fd:3 attached ret:(0,Success)
# ./fds_example -F /sys/fs/bpf/p -G -p
bpf: get fd:3 (Success)
bpf: sock:4 <- fd:3 attached ret:(0,Success)
# ./fds_example -F /sys/fs/bpf/p2 -P -p -o ./sockex1_kern.o
bpf: prog fd:5 (Success)
bpf: pin ret:(0,Success)
bpf: sock:3 <- fd:5 attached ret:(0,Success)
# ./fds_example -F /sys/fs/bpf/p2 -G -p
bpf: get fd:3 (Success)
bpf: sock:4 <- fd:3 attached ret:(0,Success)
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: Alexei Starovoitov <ast@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
2015-10-29 21:58:10 +08:00
|
|
|
hostprogs-y += fds_example
|
2014-12-02 07:06:38 +08:00
|
|
|
hostprogs-y += sockex1
|
2014-12-02 07:06:39 +08:00
|
|
|
hostprogs-y += sockex2
|
2015-05-20 07:59:06 +08:00
|
|
|
hostprogs-y += sockex3
|
2015-03-26 03:49:23 +08:00
|
|
|
hostprogs-y += tracex1
|
2015-03-26 03:49:24 +08:00
|
|
|
hostprogs-y += tracex2
|
2015-03-26 03:49:25 +08:00
|
|
|
hostprogs-y += tracex3
|
2015-03-26 03:49:26 +08:00
|
|
|
hostprogs-y += tracex4
|
samples/bpf: bpf_tail_call example for tracing
kprobe example that demonstrates how future seccomp programs may look like.
It attaches to seccomp_phase1() function and tail-calls other BPF programs
depending on syscall number.
Existing optimized classic BPF seccomp programs generated by Chrome look like:
if (sd.nr < 121) {
if (sd.nr < 57) {
if (sd.nr < 22) {
if (sd.nr < 7) {
if (sd.nr < 4) {
if (sd.nr < 1) {
check sys_read
} else {
if (sd.nr < 3) {
check sys_write and sys_open
} else {
check sys_close
}
}
} else {
} else {
} else {
} else {
} else {
}
the future seccomp using native eBPF may look like:
bpf_tail_call(&sd, &syscall_jmp_table, sd.nr);
which is simpler, faster and leaves more room for per-syscall checks.
Usage:
$ sudo ./tracex5
<...>-366 [001] d... 4.870033: : read(fd=1, buf=00007f6d5bebf000, size=771)
<...>-369 [003] d... 4.870066: : mmap
<...>-369 [003] d... 4.870077: : syscall=110 (one of get/set uid/pid/gid)
<...>-369 [003] d... 4.870089: : syscall=107 (one of get/set uid/pid/gid)
sh-369 [000] d... 4.891740: : read(fd=0, buf=00000000023d1000, size=512)
sh-369 [000] d... 4.891747: : write(fd=1, buf=00000000023d3000, size=512)
sh-369 [000] d... 4.891747: : read(fd=1, buf=00000000023d3000, size=512)
Signed-off-by: Alexei Starovoitov <ast@plumgrid.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2015-05-20 07:59:05 +08:00
|
|
|
hostprogs-y += tracex5
|
2015-08-06 15:02:36 +08:00
|
|
|
hostprogs-y += tracex6
|
2016-07-25 20:55:02 +08:00
|
|
|
hostprogs-y += test_probe_write_user
|
2015-10-21 11:02:35 +08:00
|
|
|
hostprogs-y += trace_output
|
2015-06-19 22:00:44 +08:00
|
|
|
hostprogs-y += lathist
|
2016-02-18 11:58:59 +08:00
|
|
|
hostprogs-y += offwaketime
|
2016-03-09 07:07:52 +08:00
|
|
|
hostprogs-y += spintest
|
2016-03-09 07:07:54 +08:00
|
|
|
hostprogs-y += map_perf_test
|
2016-04-07 09:43:31 +08:00
|
|
|
hostprogs-y += test_overhead
|
2016-07-01 01:28:45 +08:00
|
|
|
hostprogs-y += test_cgrp2_array_pin
|
2016-11-23 23:52:30 +08:00
|
|
|
hostprogs-y += test_cgrp2_attach
|
2016-12-02 18:42:32 +08:00
|
|
|
hostprogs-y += test_cgrp2_attach2
|
2016-12-02 00:48:05 +08:00
|
|
|
hostprogs-y += test_cgrp2_sock
|
2016-12-02 00:48:08 +08:00
|
|
|
hostprogs-y += test_cgrp2_sock2
|
Add sample for adding simple drop program to link
Add a sample program that only drops packets at the BPF_PROG_TYPE_XDP_RX
hook of a link. With the drop-only program, observed single core rate is
~20Mpps.
Other tests were run, for instance without the dropcnt increment or
without reading from the packet header, the packet rate was mostly
unchanged.
$ perf record -a samples/bpf/xdp1 $(</sys/class/net/eth0/ifindex)
proto 17: 20403027 drops/s
./pktgen_sample03_burst_single_flow.sh -i $DEV -d $IP -m $MAC -t 4
Running... ctrl^C to stop
Device: eth4@0
Result: OK: 11791017(c11788327+d2689) usec, 59622913 (60byte,0frags)
5056638pps 2427Mb/sec (2427186240bps) errors: 0
Device: eth4@1
Result: OK: 11791012(c11787906+d3106) usec, 60526944 (60byte,0frags)
5133311pps 2463Mb/sec (2463989280bps) errors: 0
Device: eth4@2
Result: OK: 11791019(c11788249+d2769) usec, 59868091 (60byte,0frags)
5077431pps 2437Mb/sec (2437166880bps) errors: 0
Device: eth4@3
Result: OK: 11795039(c11792403+d2636) usec, 59483181 (60byte,0frags)
5043067pps 2420Mb/sec (2420672160bps) errors: 0
perf report --no-children:
26.05% ksoftirqd/0 [mlx4_en] [k] mlx4_en_process_rx_cq
17.84% ksoftirqd/0 [mlx4_en] [k] mlx4_en_alloc_frags
5.52% ksoftirqd/0 [mlx4_en] [k] mlx4_en_free_frag
4.90% swapper [kernel.vmlinux] [k] poll_idle
4.14% ksoftirqd/0 [kernel.vmlinux] [k] get_page_from_freelist
2.78% ksoftirqd/0 [kernel.vmlinux] [k] __free_pages_ok
2.57% ksoftirqd/0 [kernel.vmlinux] [k] bpf_map_lookup_elem
2.51% swapper [mlx4_en] [k] mlx4_en_process_rx_cq
1.94% ksoftirqd/0 [kernel.vmlinux] [k] percpu_array_map_lookup_elem
1.45% swapper [mlx4_en] [k] mlx4_en_alloc_frags
1.35% ksoftirqd/0 [kernel.vmlinux] [k] free_one_page
1.33% swapper [kernel.vmlinux] [k] intel_idle
1.04% ksoftirqd/0 [mlx4_en] [k] 0x000000000001c5c5
0.96% ksoftirqd/0 [mlx4_en] [k] 0x000000000001c58d
0.93% ksoftirqd/0 [mlx4_en] [k] 0x000000000001c6ee
0.92% ksoftirqd/0 [mlx4_en] [k] 0x000000000001c6b9
0.89% ksoftirqd/0 [kernel.vmlinux] [k] __alloc_pages_nodemask
0.83% ksoftirqd/0 [mlx4_en] [k] 0x000000000001c686
0.83% ksoftirqd/0 [mlx4_en] [k] 0x000000000001c5d5
0.78% ksoftirqd/0 [mlx4_en] [k] mlx4_alloc_pages.isra.23
0.77% ksoftirqd/0 [mlx4_en] [k] 0x000000000001c5b4
0.77% ksoftirqd/0 [kernel.vmlinux] [k] net_rx_action
machine specs:
receiver - Intel E5-1630 v3 @ 3.70GHz
sender - Intel E5645 @ 2.40GHz
Mellanox ConnectX-3 @40G
Signed-off-by: Brenden Blanco <bblanco@plumgrid.com>
Acked-by: Alexei Starovoitov <ast@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
2016-07-20 03:16:51 +08:00
|
|
|
hostprogs-y += xdp1
|
2016-07-20 03:16:57 +08:00
|
|
|
hostprogs-y += xdp2
|
2016-08-12 23:57:04 +08:00
|
|
|
hostprogs-y += test_current_task_under_cgroup
|
2016-09-02 09:37:25 +08:00
|
|
|
hostprogs-y += trace_event
|
2016-09-02 09:37:26 +08:00
|
|
|
hostprogs-y += sampleip
|
2016-11-10 07:36:34 +08:00
|
|
|
hostprogs-y += tc_l2_redirect
|
2016-12-01 00:10:11 +08:00
|
|
|
hostprogs-y += lwt_len_hist
|
2016-12-08 07:53:14 +08:00
|
|
|
hostprogs-y += xdp_tx_iptunnel
|
2017-03-23 01:00:35 +08:00
|
|
|
hostprogs-y += test_map_in_map
|
2017-03-23 08:27:36 +08:00
|
|
|
hostprogs-y += per_socket_stats_example
|
2017-07-01 11:02:41 +08:00
|
|
|
hostprogs-y += load_sock_ops
|
bpf: mini eBPF library, test stubs and verifier testsuite
1.
the library includes a trivial set of BPF syscall wrappers:
int bpf_create_map(int key_size, int value_size, int max_entries);
int bpf_update_elem(int fd, void *key, void *value);
int bpf_lookup_elem(int fd, void *key, void *value);
int bpf_delete_elem(int fd, void *key);
int bpf_get_next_key(int fd, void *key, void *next_key);
int bpf_prog_load(enum bpf_prog_type prog_type,
const struct sock_filter_int *insns, int insn_len,
const char *license);
bpf_prog_load() stores verifier log into global bpf_log_buf[] array
and BPF_*() macros to build instructions
2.
test stubs configure eBPF infra with 'unspec' map and program types.
These are fake types used by user space testsuite only.
3.
verifier tests valid and invalid programs and expects predefined
error log messages from kernel.
40 tests so far.
$ sudo ./test_verifier
#0 add+sub+mul OK
#1 unreachable OK
#2 unreachable2 OK
#3 out of range jump OK
#4 out of range jump2 OK
#5 test1 ld_imm64 OK
...
Signed-off-by: Alexei Starovoitov <ast@plumgrid.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2014-09-26 15:17:07 +08:00
|
|
|
|
samples/bpf: Switch over to libbpf
Now that libbpf under tools/lib/bpf/* is synced with the version from
samples/bpf, we can get rid most of the libbpf library here.
Committer notes:
Built it in a docker fedora rawhide container and ran it in the f25 host, seems
to work just like it did before this patch, i.e. the switch to tools/lib/bpf/
doesn't seem to have introduced problems and Joe said he tested it with
all the entries in samples/bpf/ and other code he found:
[root@f5065a7d6272 linux]# make -j4 O=/tmp/build/linux headers_install
<SNIP>
[root@f5065a7d6272 linux]# rm -rf /tmp/build/linux/samples/bpf/
[root@f5065a7d6272 linux]# make -j4 O=/tmp/build/linux samples/bpf/
make[1]: Entering directory '/tmp/build/linux'
CHK include/config/kernel.release
HOSTCC scripts/basic/fixdep
GEN ./Makefile
CHK include/generated/uapi/linux/version.h
Using /git/linux as source for kernel
CHK include/generated/utsrelease.h
HOSTCC scripts/basic/bin2c
HOSTCC arch/x86/tools/relocs_32.o
HOSTCC arch/x86/tools/relocs_64.o
LD samples/bpf/built-in.o
<SNIP>
HOSTCC samples/bpf/fds_example.o
HOSTCC samples/bpf/sockex1_user.o
/git/linux/samples/bpf/fds_example.c: In function 'bpf_prog_create':
/git/linux/samples/bpf/fds_example.c:63:6: warning: passing argument 2 of 'bpf_load_program' discards 'const' qualifier from pointer target type [-Wdiscarded-qualifiers]
insns, insns_cnt, "GPL", 0,
^~~~~
In file included from /git/linux/samples/bpf/libbpf.h:5:0,
from /git/linux/samples/bpf/bpf_load.h:4,
from /git/linux/samples/bpf/fds_example.c:15:
/git/linux/tools/lib/bpf/bpf.h:31:5: note: expected 'struct bpf_insn *' but argument is of type 'const struct bpf_insn *'
int bpf_load_program(enum bpf_prog_type type, struct bpf_insn *insns,
^~~~~~~~~~~~~~~~
HOSTCC samples/bpf/sockex2_user.o
<SNIP>
HOSTCC samples/bpf/xdp_tx_iptunnel_user.o
clang -nostdinc -isystem /usr/lib/gcc/x86_64-redhat-linux/6.2.1/include -I/git/linux/arch/x86/include -I./arch/x86/include/generated/uapi -I./arch/x86/include/generated -I/git/linux/include -I./include -I/git/linux/arch/x86/include/uapi -I/git/linux/include/uapi -I./include/generated/uapi -include /git/linux/include/linux/kconfig.h \
-D__KERNEL__ -D__ASM_SYSREG_H -Wno-unused-value -Wno-pointer-sign \
-Wno-compare-distinct-pointer-types \
-Wno-gnu-variable-sized-type-not-at-end \
-Wno-address-of-packed-member -Wno-tautological-compare \
-O2 -emit-llvm -c /git/linux/samples/bpf/sockex1_kern.c -o -| llc -march=bpf -filetype=obj -o samples/bpf/sockex1_kern.o
HOSTLD samples/bpf/tc_l2_redirect
<SNIP>
HOSTLD samples/bpf/lwt_len_hist
HOSTLD samples/bpf/xdp_tx_iptunnel
make[1]: Leaving directory '/tmp/build/linux'
[root@f5065a7d6272 linux]#
And then, in the host:
[root@jouet bpf]# mount | grep "docker.*devicemapper\/"
/dev/mapper/docker-253:0-1705076-9bd8aa1e0af33adce89ff42090847868ca676932878942be53941a06ec5923f9 on /var/lib/docker/devicemapper/mnt/9bd8aa1e0af33adce89ff42090847868ca676932878942be53941a06ec5923f9 type xfs (rw,relatime,context="system_u:object_r:container_file_t:s0:c73,c276",nouuid,attr2,inode64,sunit=1024,swidth=1024,noquota)
[root@jouet bpf]# cd /var/lib/docker/devicemapper/mnt/9bd8aa1e0af33adce89ff42090847868ca676932878942be53941a06ec5923f9/rootfs/tmp/build/linux/samples/bpf/
[root@jouet bpf]# file offwaketime
offwaketime: ELF 64-bit LSB executable, x86-64, version 1 (SYSV), dynamically linked, interpreter /lib64/ld-linux-x86-64.so.2, for GNU/Linux 2.6.32, BuildID[sha1]=f423d171e0487b2f802b6a792657f0f3c8f6d155, not stripped
[root@jouet bpf]# readelf -SW offwaketime
offwaketime offwaketime_kern.o offwaketime_user.o
[root@jouet bpf]# readelf -SW offwaketime_kern.o
There are 11 section headers, starting at offset 0x700:
Section Headers:
[Nr] Name Type Address Off Size ES Flg Lk Inf Al
[ 0] NULL 0000000000000000 000000 000000 00 0 0 0
[ 1] .strtab STRTAB 0000000000000000 000658 0000a8 00 0 0 1
[ 2] .text PROGBITS 0000000000000000 000040 000000 00 AX 0 0 4
[ 3] kprobe/try_to_wake_up PROGBITS 0000000000000000 000040 0000d8 00 AX 0 0 8
[ 4] .relkprobe/try_to_wake_up REL 0000000000000000 0005a8 000020 10 10 3 8
[ 5] tracepoint/sched/sched_switch PROGBITS 0000000000000000 000118 000318 00 AX 0 0 8
[ 6] .reltracepoint/sched/sched_switch REL 0000000000000000 0005c8 000090 10 10 5 8
[ 7] maps PROGBITS 0000000000000000 000430 000050 00 WA 0 0 4
[ 8] license PROGBITS 0000000000000000 000480 000004 00 WA 0 0 1
[ 9] version PROGBITS 0000000000000000 000484 000004 00 WA 0 0 4
[10] .symtab SYMTAB 0000000000000000 000488 000120 18 1 4 8
Key to Flags:
W (write), A (alloc), X (execute), M (merge), S (strings)
I (info), L (link order), G (group), T (TLS), E (exclude), x (unknown)
O (extra OS processing required) o (OS specific), p (processor specific)
[root@jouet bpf]# ./offwaketime | head -3
qemu-system-x86;entry_SYSCALL_64_fastpath;sys_ppoll;do_sys_poll;poll_schedule_timeout;schedule_hrtimeout_range;schedule_hrtimeout_range_clock;schedule;__schedule;-;try_to_wake_up;hrtimer_wakeup;__hrtimer_run_queues;hrtimer_interrupt;local_apic_timer_interrupt;smp_apic_timer_interrupt;__irqentry_text_start;cpuidle_enter_state;cpuidle_enter;call_cpuidle;cpu_startup_entry;rest_init;start_kernel;x86_64_start_reservations;x86_64_start_kernel;start_cpu;;swapper/0 4
firefox;entry_SYSCALL_64_fastpath;sys_poll;do_sys_poll;poll_schedule_timeout;schedule_hrtimeout_range;schedule_hrtimeout_range_clock;schedule;__schedule;-;try_to_wake_up;pollwake;__wake_up_common;__wake_up_sync_key;pipe_write;__vfs_write;vfs_write;sys_write;entry_SYSCALL_64_fastpath;;Timer 1
swapper/2;start_cpu;start_secondary;cpu_startup_entry;schedule_preempt_disabled;schedule;__schedule;-;---;; 61
[root@jouet bpf]#
Signed-off-by: Joe Stringer <joe@ovn.org>
Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Alexei Starovoitov <ast@fb.com>
Cc: Daniel Borkmann <daniel@iogearbox.net>
Cc: Wang Nan <wangnan0@huawei.com>
Cc: netdev@vger.kernel.org
Link: https://github.com/joestringer/linux/commit/5c40f54a52b1f437123c81e21873f4b4b1f9bd55.patch
Link: http://lkml.kernel.org/n/tip-xr8twtx7sjh5821g8qw47yxk@git.kernel.org
[ Use -I$(srctree)/tools/lib/ to support out of source code tree builds, as noticed by Wang Nan ]
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2016-12-15 06:43:39 +08:00
|
|
|
# Libbpf dependencies
|
2016-12-09 10:46:20 +08:00
|
|
|
LIBBPF := ../../tools/lib/bpf/bpf.o
|
samples/bpf: Switch over to libbpf
Now that libbpf under tools/lib/bpf/* is synced with the version from
samples/bpf, we can get rid most of the libbpf library here.
Committer notes:
Built it in a docker fedora rawhide container and ran it in the f25 host, seems
to work just like it did before this patch, i.e. the switch to tools/lib/bpf/
doesn't seem to have introduced problems and Joe said he tested it with
all the entries in samples/bpf/ and other code he found:
[root@f5065a7d6272 linux]# make -j4 O=/tmp/build/linux headers_install
<SNIP>
[root@f5065a7d6272 linux]# rm -rf /tmp/build/linux/samples/bpf/
[root@f5065a7d6272 linux]# make -j4 O=/tmp/build/linux samples/bpf/
make[1]: Entering directory '/tmp/build/linux'
CHK include/config/kernel.release
HOSTCC scripts/basic/fixdep
GEN ./Makefile
CHK include/generated/uapi/linux/version.h
Using /git/linux as source for kernel
CHK include/generated/utsrelease.h
HOSTCC scripts/basic/bin2c
HOSTCC arch/x86/tools/relocs_32.o
HOSTCC arch/x86/tools/relocs_64.o
LD samples/bpf/built-in.o
<SNIP>
HOSTCC samples/bpf/fds_example.o
HOSTCC samples/bpf/sockex1_user.o
/git/linux/samples/bpf/fds_example.c: In function 'bpf_prog_create':
/git/linux/samples/bpf/fds_example.c:63:6: warning: passing argument 2 of 'bpf_load_program' discards 'const' qualifier from pointer target type [-Wdiscarded-qualifiers]
insns, insns_cnt, "GPL", 0,
^~~~~
In file included from /git/linux/samples/bpf/libbpf.h:5:0,
from /git/linux/samples/bpf/bpf_load.h:4,
from /git/linux/samples/bpf/fds_example.c:15:
/git/linux/tools/lib/bpf/bpf.h:31:5: note: expected 'struct bpf_insn *' but argument is of type 'const struct bpf_insn *'
int bpf_load_program(enum bpf_prog_type type, struct bpf_insn *insns,
^~~~~~~~~~~~~~~~
HOSTCC samples/bpf/sockex2_user.o
<SNIP>
HOSTCC samples/bpf/xdp_tx_iptunnel_user.o
clang -nostdinc -isystem /usr/lib/gcc/x86_64-redhat-linux/6.2.1/include -I/git/linux/arch/x86/include -I./arch/x86/include/generated/uapi -I./arch/x86/include/generated -I/git/linux/include -I./include -I/git/linux/arch/x86/include/uapi -I/git/linux/include/uapi -I./include/generated/uapi -include /git/linux/include/linux/kconfig.h \
-D__KERNEL__ -D__ASM_SYSREG_H -Wno-unused-value -Wno-pointer-sign \
-Wno-compare-distinct-pointer-types \
-Wno-gnu-variable-sized-type-not-at-end \
-Wno-address-of-packed-member -Wno-tautological-compare \
-O2 -emit-llvm -c /git/linux/samples/bpf/sockex1_kern.c -o -| llc -march=bpf -filetype=obj -o samples/bpf/sockex1_kern.o
HOSTLD samples/bpf/tc_l2_redirect
<SNIP>
HOSTLD samples/bpf/lwt_len_hist
HOSTLD samples/bpf/xdp_tx_iptunnel
make[1]: Leaving directory '/tmp/build/linux'
[root@f5065a7d6272 linux]#
And then, in the host:
[root@jouet bpf]# mount | grep "docker.*devicemapper\/"
/dev/mapper/docker-253:0-1705076-9bd8aa1e0af33adce89ff42090847868ca676932878942be53941a06ec5923f9 on /var/lib/docker/devicemapper/mnt/9bd8aa1e0af33adce89ff42090847868ca676932878942be53941a06ec5923f9 type xfs (rw,relatime,context="system_u:object_r:container_file_t:s0:c73,c276",nouuid,attr2,inode64,sunit=1024,swidth=1024,noquota)
[root@jouet bpf]# cd /var/lib/docker/devicemapper/mnt/9bd8aa1e0af33adce89ff42090847868ca676932878942be53941a06ec5923f9/rootfs/tmp/build/linux/samples/bpf/
[root@jouet bpf]# file offwaketime
offwaketime: ELF 64-bit LSB executable, x86-64, version 1 (SYSV), dynamically linked, interpreter /lib64/ld-linux-x86-64.so.2, for GNU/Linux 2.6.32, BuildID[sha1]=f423d171e0487b2f802b6a792657f0f3c8f6d155, not stripped
[root@jouet bpf]# readelf -SW offwaketime
offwaketime offwaketime_kern.o offwaketime_user.o
[root@jouet bpf]# readelf -SW offwaketime_kern.o
There are 11 section headers, starting at offset 0x700:
Section Headers:
[Nr] Name Type Address Off Size ES Flg Lk Inf Al
[ 0] NULL 0000000000000000 000000 000000 00 0 0 0
[ 1] .strtab STRTAB 0000000000000000 000658 0000a8 00 0 0 1
[ 2] .text PROGBITS 0000000000000000 000040 000000 00 AX 0 0 4
[ 3] kprobe/try_to_wake_up PROGBITS 0000000000000000 000040 0000d8 00 AX 0 0 8
[ 4] .relkprobe/try_to_wake_up REL 0000000000000000 0005a8 000020 10 10 3 8
[ 5] tracepoint/sched/sched_switch PROGBITS 0000000000000000 000118 000318 00 AX 0 0 8
[ 6] .reltracepoint/sched/sched_switch REL 0000000000000000 0005c8 000090 10 10 5 8
[ 7] maps PROGBITS 0000000000000000 000430 000050 00 WA 0 0 4
[ 8] license PROGBITS 0000000000000000 000480 000004 00 WA 0 0 1
[ 9] version PROGBITS 0000000000000000 000484 000004 00 WA 0 0 4
[10] .symtab SYMTAB 0000000000000000 000488 000120 18 1 4 8
Key to Flags:
W (write), A (alloc), X (execute), M (merge), S (strings)
I (info), L (link order), G (group), T (TLS), E (exclude), x (unknown)
O (extra OS processing required) o (OS specific), p (processor specific)
[root@jouet bpf]# ./offwaketime | head -3
qemu-system-x86;entry_SYSCALL_64_fastpath;sys_ppoll;do_sys_poll;poll_schedule_timeout;schedule_hrtimeout_range;schedule_hrtimeout_range_clock;schedule;__schedule;-;try_to_wake_up;hrtimer_wakeup;__hrtimer_run_queues;hrtimer_interrupt;local_apic_timer_interrupt;smp_apic_timer_interrupt;__irqentry_text_start;cpuidle_enter_state;cpuidle_enter;call_cpuidle;cpu_startup_entry;rest_init;start_kernel;x86_64_start_reservations;x86_64_start_kernel;start_cpu;;swapper/0 4
firefox;entry_SYSCALL_64_fastpath;sys_poll;do_sys_poll;poll_schedule_timeout;schedule_hrtimeout_range;schedule_hrtimeout_range_clock;schedule;__schedule;-;try_to_wake_up;pollwake;__wake_up_common;__wake_up_sync_key;pipe_write;__vfs_write;vfs_write;sys_write;entry_SYSCALL_64_fastpath;;Timer 1
swapper/2;start_cpu;start_secondary;cpu_startup_entry;schedule_preempt_disabled;schedule;__schedule;-;---;; 61
[root@jouet bpf]#
Signed-off-by: Joe Stringer <joe@ovn.org>
Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Alexei Starovoitov <ast@fb.com>
Cc: Daniel Borkmann <daniel@iogearbox.net>
Cc: Wang Nan <wangnan0@huawei.com>
Cc: netdev@vger.kernel.org
Link: https://github.com/joestringer/linux/commit/5c40f54a52b1f437123c81e21873f4b4b1f9bd55.patch
Link: http://lkml.kernel.org/n/tip-xr8twtx7sjh5821g8qw47yxk@git.kernel.org
[ Use -I$(srctree)/tools/lib/ to support out of source code tree builds, as noticed by Wang Nan ]
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2016-12-15 06:43:39 +08:00
|
|
|
|
|
|
|
test_lru_dist-objs := test_lru_dist.o $(LIBBPF)
|
|
|
|
sock_example-objs := sock_example.o $(LIBBPF)
|
|
|
|
fds_example-objs := bpf_load.o $(LIBBPF) fds_example.o
|
|
|
|
sockex1-objs := bpf_load.o $(LIBBPF) sockex1_user.o
|
|
|
|
sockex2-objs := bpf_load.o $(LIBBPF) sockex2_user.o
|
|
|
|
sockex3-objs := bpf_load.o $(LIBBPF) sockex3_user.o
|
|
|
|
tracex1-objs := bpf_load.o $(LIBBPF) tracex1_user.o
|
|
|
|
tracex2-objs := bpf_load.o $(LIBBPF) tracex2_user.o
|
|
|
|
tracex3-objs := bpf_load.o $(LIBBPF) tracex3_user.o
|
|
|
|
tracex4-objs := bpf_load.o $(LIBBPF) tracex4_user.o
|
|
|
|
tracex5-objs := bpf_load.o $(LIBBPF) tracex5_user.o
|
|
|
|
tracex6-objs := bpf_load.o $(LIBBPF) tracex6_user.o
|
2017-07-01 11:02:41 +08:00
|
|
|
load_sock_ops-objs := bpf_load.o $(LIBBPF) load_sock_ops.o
|
samples/bpf: Switch over to libbpf
Now that libbpf under tools/lib/bpf/* is synced with the version from
samples/bpf, we can get rid most of the libbpf library here.
Committer notes:
Built it in a docker fedora rawhide container and ran it in the f25 host, seems
to work just like it did before this patch, i.e. the switch to tools/lib/bpf/
doesn't seem to have introduced problems and Joe said he tested it with
all the entries in samples/bpf/ and other code he found:
[root@f5065a7d6272 linux]# make -j4 O=/tmp/build/linux headers_install
<SNIP>
[root@f5065a7d6272 linux]# rm -rf /tmp/build/linux/samples/bpf/
[root@f5065a7d6272 linux]# make -j4 O=/tmp/build/linux samples/bpf/
make[1]: Entering directory '/tmp/build/linux'
CHK include/config/kernel.release
HOSTCC scripts/basic/fixdep
GEN ./Makefile
CHK include/generated/uapi/linux/version.h
Using /git/linux as source for kernel
CHK include/generated/utsrelease.h
HOSTCC scripts/basic/bin2c
HOSTCC arch/x86/tools/relocs_32.o
HOSTCC arch/x86/tools/relocs_64.o
LD samples/bpf/built-in.o
<SNIP>
HOSTCC samples/bpf/fds_example.o
HOSTCC samples/bpf/sockex1_user.o
/git/linux/samples/bpf/fds_example.c: In function 'bpf_prog_create':
/git/linux/samples/bpf/fds_example.c:63:6: warning: passing argument 2 of 'bpf_load_program' discards 'const' qualifier from pointer target type [-Wdiscarded-qualifiers]
insns, insns_cnt, "GPL", 0,
^~~~~
In file included from /git/linux/samples/bpf/libbpf.h:5:0,
from /git/linux/samples/bpf/bpf_load.h:4,
from /git/linux/samples/bpf/fds_example.c:15:
/git/linux/tools/lib/bpf/bpf.h:31:5: note: expected 'struct bpf_insn *' but argument is of type 'const struct bpf_insn *'
int bpf_load_program(enum bpf_prog_type type, struct bpf_insn *insns,
^~~~~~~~~~~~~~~~
HOSTCC samples/bpf/sockex2_user.o
<SNIP>
HOSTCC samples/bpf/xdp_tx_iptunnel_user.o
clang -nostdinc -isystem /usr/lib/gcc/x86_64-redhat-linux/6.2.1/include -I/git/linux/arch/x86/include -I./arch/x86/include/generated/uapi -I./arch/x86/include/generated -I/git/linux/include -I./include -I/git/linux/arch/x86/include/uapi -I/git/linux/include/uapi -I./include/generated/uapi -include /git/linux/include/linux/kconfig.h \
-D__KERNEL__ -D__ASM_SYSREG_H -Wno-unused-value -Wno-pointer-sign \
-Wno-compare-distinct-pointer-types \
-Wno-gnu-variable-sized-type-not-at-end \
-Wno-address-of-packed-member -Wno-tautological-compare \
-O2 -emit-llvm -c /git/linux/samples/bpf/sockex1_kern.c -o -| llc -march=bpf -filetype=obj -o samples/bpf/sockex1_kern.o
HOSTLD samples/bpf/tc_l2_redirect
<SNIP>
HOSTLD samples/bpf/lwt_len_hist
HOSTLD samples/bpf/xdp_tx_iptunnel
make[1]: Leaving directory '/tmp/build/linux'
[root@f5065a7d6272 linux]#
And then, in the host:
[root@jouet bpf]# mount | grep "docker.*devicemapper\/"
/dev/mapper/docker-253:0-1705076-9bd8aa1e0af33adce89ff42090847868ca676932878942be53941a06ec5923f9 on /var/lib/docker/devicemapper/mnt/9bd8aa1e0af33adce89ff42090847868ca676932878942be53941a06ec5923f9 type xfs (rw,relatime,context="system_u:object_r:container_file_t:s0:c73,c276",nouuid,attr2,inode64,sunit=1024,swidth=1024,noquota)
[root@jouet bpf]# cd /var/lib/docker/devicemapper/mnt/9bd8aa1e0af33adce89ff42090847868ca676932878942be53941a06ec5923f9/rootfs/tmp/build/linux/samples/bpf/
[root@jouet bpf]# file offwaketime
offwaketime: ELF 64-bit LSB executable, x86-64, version 1 (SYSV), dynamically linked, interpreter /lib64/ld-linux-x86-64.so.2, for GNU/Linux 2.6.32, BuildID[sha1]=f423d171e0487b2f802b6a792657f0f3c8f6d155, not stripped
[root@jouet bpf]# readelf -SW offwaketime
offwaketime offwaketime_kern.o offwaketime_user.o
[root@jouet bpf]# readelf -SW offwaketime_kern.o
There are 11 section headers, starting at offset 0x700:
Section Headers:
[Nr] Name Type Address Off Size ES Flg Lk Inf Al
[ 0] NULL 0000000000000000 000000 000000 00 0 0 0
[ 1] .strtab STRTAB 0000000000000000 000658 0000a8 00 0 0 1
[ 2] .text PROGBITS 0000000000000000 000040 000000 00 AX 0 0 4
[ 3] kprobe/try_to_wake_up PROGBITS 0000000000000000 000040 0000d8 00 AX 0 0 8
[ 4] .relkprobe/try_to_wake_up REL 0000000000000000 0005a8 000020 10 10 3 8
[ 5] tracepoint/sched/sched_switch PROGBITS 0000000000000000 000118 000318 00 AX 0 0 8
[ 6] .reltracepoint/sched/sched_switch REL 0000000000000000 0005c8 000090 10 10 5 8
[ 7] maps PROGBITS 0000000000000000 000430 000050 00 WA 0 0 4
[ 8] license PROGBITS 0000000000000000 000480 000004 00 WA 0 0 1
[ 9] version PROGBITS 0000000000000000 000484 000004 00 WA 0 0 4
[10] .symtab SYMTAB 0000000000000000 000488 000120 18 1 4 8
Key to Flags:
W (write), A (alloc), X (execute), M (merge), S (strings)
I (info), L (link order), G (group), T (TLS), E (exclude), x (unknown)
O (extra OS processing required) o (OS specific), p (processor specific)
[root@jouet bpf]# ./offwaketime | head -3
qemu-system-x86;entry_SYSCALL_64_fastpath;sys_ppoll;do_sys_poll;poll_schedule_timeout;schedule_hrtimeout_range;schedule_hrtimeout_range_clock;schedule;__schedule;-;try_to_wake_up;hrtimer_wakeup;__hrtimer_run_queues;hrtimer_interrupt;local_apic_timer_interrupt;smp_apic_timer_interrupt;__irqentry_text_start;cpuidle_enter_state;cpuidle_enter;call_cpuidle;cpu_startup_entry;rest_init;start_kernel;x86_64_start_reservations;x86_64_start_kernel;start_cpu;;swapper/0 4
firefox;entry_SYSCALL_64_fastpath;sys_poll;do_sys_poll;poll_schedule_timeout;schedule_hrtimeout_range;schedule_hrtimeout_range_clock;schedule;__schedule;-;try_to_wake_up;pollwake;__wake_up_common;__wake_up_sync_key;pipe_write;__vfs_write;vfs_write;sys_write;entry_SYSCALL_64_fastpath;;Timer 1
swapper/2;start_cpu;start_secondary;cpu_startup_entry;schedule_preempt_disabled;schedule;__schedule;-;---;; 61
[root@jouet bpf]#
Signed-off-by: Joe Stringer <joe@ovn.org>
Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Alexei Starovoitov <ast@fb.com>
Cc: Daniel Borkmann <daniel@iogearbox.net>
Cc: Wang Nan <wangnan0@huawei.com>
Cc: netdev@vger.kernel.org
Link: https://github.com/joestringer/linux/commit/5c40f54a52b1f437123c81e21873f4b4b1f9bd55.patch
Link: http://lkml.kernel.org/n/tip-xr8twtx7sjh5821g8qw47yxk@git.kernel.org
[ Use -I$(srctree)/tools/lib/ to support out of source code tree builds, as noticed by Wang Nan ]
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2016-12-15 06:43:39 +08:00
|
|
|
test_probe_write_user-objs := bpf_load.o $(LIBBPF) test_probe_write_user_user.o
|
|
|
|
trace_output-objs := bpf_load.o $(LIBBPF) trace_output_user.o
|
|
|
|
lathist-objs := bpf_load.o $(LIBBPF) lathist_user.o
|
|
|
|
offwaketime-objs := bpf_load.o $(LIBBPF) offwaketime_user.o
|
|
|
|
spintest-objs := bpf_load.o $(LIBBPF) spintest_user.o
|
|
|
|
map_perf_test-objs := bpf_load.o $(LIBBPF) map_perf_test_user.o
|
|
|
|
test_overhead-objs := bpf_load.o $(LIBBPF) test_overhead_user.o
|
|
|
|
test_cgrp2_array_pin-objs := $(LIBBPF) test_cgrp2_array_pin.o
|
|
|
|
test_cgrp2_attach-objs := $(LIBBPF) test_cgrp2_attach.o
|
|
|
|
test_cgrp2_attach2-objs := $(LIBBPF) test_cgrp2_attach2.o cgroup_helpers.o
|
|
|
|
test_cgrp2_sock-objs := $(LIBBPF) test_cgrp2_sock.o
|
|
|
|
test_cgrp2_sock2-objs := bpf_load.o $(LIBBPF) test_cgrp2_sock2.o
|
|
|
|
xdp1-objs := bpf_load.o $(LIBBPF) xdp1_user.o
|
2016-07-20 03:16:57 +08:00
|
|
|
# reuse xdp1 source intentionally
|
samples/bpf: Switch over to libbpf
Now that libbpf under tools/lib/bpf/* is synced with the version from
samples/bpf, we can get rid most of the libbpf library here.
Committer notes:
Built it in a docker fedora rawhide container and ran it in the f25 host, seems
to work just like it did before this patch, i.e. the switch to tools/lib/bpf/
doesn't seem to have introduced problems and Joe said he tested it with
all the entries in samples/bpf/ and other code he found:
[root@f5065a7d6272 linux]# make -j4 O=/tmp/build/linux headers_install
<SNIP>
[root@f5065a7d6272 linux]# rm -rf /tmp/build/linux/samples/bpf/
[root@f5065a7d6272 linux]# make -j4 O=/tmp/build/linux samples/bpf/
make[1]: Entering directory '/tmp/build/linux'
CHK include/config/kernel.release
HOSTCC scripts/basic/fixdep
GEN ./Makefile
CHK include/generated/uapi/linux/version.h
Using /git/linux as source for kernel
CHK include/generated/utsrelease.h
HOSTCC scripts/basic/bin2c
HOSTCC arch/x86/tools/relocs_32.o
HOSTCC arch/x86/tools/relocs_64.o
LD samples/bpf/built-in.o
<SNIP>
HOSTCC samples/bpf/fds_example.o
HOSTCC samples/bpf/sockex1_user.o
/git/linux/samples/bpf/fds_example.c: In function 'bpf_prog_create':
/git/linux/samples/bpf/fds_example.c:63:6: warning: passing argument 2 of 'bpf_load_program' discards 'const' qualifier from pointer target type [-Wdiscarded-qualifiers]
insns, insns_cnt, "GPL", 0,
^~~~~
In file included from /git/linux/samples/bpf/libbpf.h:5:0,
from /git/linux/samples/bpf/bpf_load.h:4,
from /git/linux/samples/bpf/fds_example.c:15:
/git/linux/tools/lib/bpf/bpf.h:31:5: note: expected 'struct bpf_insn *' but argument is of type 'const struct bpf_insn *'
int bpf_load_program(enum bpf_prog_type type, struct bpf_insn *insns,
^~~~~~~~~~~~~~~~
HOSTCC samples/bpf/sockex2_user.o
<SNIP>
HOSTCC samples/bpf/xdp_tx_iptunnel_user.o
clang -nostdinc -isystem /usr/lib/gcc/x86_64-redhat-linux/6.2.1/include -I/git/linux/arch/x86/include -I./arch/x86/include/generated/uapi -I./arch/x86/include/generated -I/git/linux/include -I./include -I/git/linux/arch/x86/include/uapi -I/git/linux/include/uapi -I./include/generated/uapi -include /git/linux/include/linux/kconfig.h \
-D__KERNEL__ -D__ASM_SYSREG_H -Wno-unused-value -Wno-pointer-sign \
-Wno-compare-distinct-pointer-types \
-Wno-gnu-variable-sized-type-not-at-end \
-Wno-address-of-packed-member -Wno-tautological-compare \
-O2 -emit-llvm -c /git/linux/samples/bpf/sockex1_kern.c -o -| llc -march=bpf -filetype=obj -o samples/bpf/sockex1_kern.o
HOSTLD samples/bpf/tc_l2_redirect
<SNIP>
HOSTLD samples/bpf/lwt_len_hist
HOSTLD samples/bpf/xdp_tx_iptunnel
make[1]: Leaving directory '/tmp/build/linux'
[root@f5065a7d6272 linux]#
And then, in the host:
[root@jouet bpf]# mount | grep "docker.*devicemapper\/"
/dev/mapper/docker-253:0-1705076-9bd8aa1e0af33adce89ff42090847868ca676932878942be53941a06ec5923f9 on /var/lib/docker/devicemapper/mnt/9bd8aa1e0af33adce89ff42090847868ca676932878942be53941a06ec5923f9 type xfs (rw,relatime,context="system_u:object_r:container_file_t:s0:c73,c276",nouuid,attr2,inode64,sunit=1024,swidth=1024,noquota)
[root@jouet bpf]# cd /var/lib/docker/devicemapper/mnt/9bd8aa1e0af33adce89ff42090847868ca676932878942be53941a06ec5923f9/rootfs/tmp/build/linux/samples/bpf/
[root@jouet bpf]# file offwaketime
offwaketime: ELF 64-bit LSB executable, x86-64, version 1 (SYSV), dynamically linked, interpreter /lib64/ld-linux-x86-64.so.2, for GNU/Linux 2.6.32, BuildID[sha1]=f423d171e0487b2f802b6a792657f0f3c8f6d155, not stripped
[root@jouet bpf]# readelf -SW offwaketime
offwaketime offwaketime_kern.o offwaketime_user.o
[root@jouet bpf]# readelf -SW offwaketime_kern.o
There are 11 section headers, starting at offset 0x700:
Section Headers:
[Nr] Name Type Address Off Size ES Flg Lk Inf Al
[ 0] NULL 0000000000000000 000000 000000 00 0 0 0
[ 1] .strtab STRTAB 0000000000000000 000658 0000a8 00 0 0 1
[ 2] .text PROGBITS 0000000000000000 000040 000000 00 AX 0 0 4
[ 3] kprobe/try_to_wake_up PROGBITS 0000000000000000 000040 0000d8 00 AX 0 0 8
[ 4] .relkprobe/try_to_wake_up REL 0000000000000000 0005a8 000020 10 10 3 8
[ 5] tracepoint/sched/sched_switch PROGBITS 0000000000000000 000118 000318 00 AX 0 0 8
[ 6] .reltracepoint/sched/sched_switch REL 0000000000000000 0005c8 000090 10 10 5 8
[ 7] maps PROGBITS 0000000000000000 000430 000050 00 WA 0 0 4
[ 8] license PROGBITS 0000000000000000 000480 000004 00 WA 0 0 1
[ 9] version PROGBITS 0000000000000000 000484 000004 00 WA 0 0 4
[10] .symtab SYMTAB 0000000000000000 000488 000120 18 1 4 8
Key to Flags:
W (write), A (alloc), X (execute), M (merge), S (strings)
I (info), L (link order), G (group), T (TLS), E (exclude), x (unknown)
O (extra OS processing required) o (OS specific), p (processor specific)
[root@jouet bpf]# ./offwaketime | head -3
qemu-system-x86;entry_SYSCALL_64_fastpath;sys_ppoll;do_sys_poll;poll_schedule_timeout;schedule_hrtimeout_range;schedule_hrtimeout_range_clock;schedule;__schedule;-;try_to_wake_up;hrtimer_wakeup;__hrtimer_run_queues;hrtimer_interrupt;local_apic_timer_interrupt;smp_apic_timer_interrupt;__irqentry_text_start;cpuidle_enter_state;cpuidle_enter;call_cpuidle;cpu_startup_entry;rest_init;start_kernel;x86_64_start_reservations;x86_64_start_kernel;start_cpu;;swapper/0 4
firefox;entry_SYSCALL_64_fastpath;sys_poll;do_sys_poll;poll_schedule_timeout;schedule_hrtimeout_range;schedule_hrtimeout_range_clock;schedule;__schedule;-;try_to_wake_up;pollwake;__wake_up_common;__wake_up_sync_key;pipe_write;__vfs_write;vfs_write;sys_write;entry_SYSCALL_64_fastpath;;Timer 1
swapper/2;start_cpu;start_secondary;cpu_startup_entry;schedule_preempt_disabled;schedule;__schedule;-;---;; 61
[root@jouet bpf]#
Signed-off-by: Joe Stringer <joe@ovn.org>
Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Alexei Starovoitov <ast@fb.com>
Cc: Daniel Borkmann <daniel@iogearbox.net>
Cc: Wang Nan <wangnan0@huawei.com>
Cc: netdev@vger.kernel.org
Link: https://github.com/joestringer/linux/commit/5c40f54a52b1f437123c81e21873f4b4b1f9bd55.patch
Link: http://lkml.kernel.org/n/tip-xr8twtx7sjh5821g8qw47yxk@git.kernel.org
[ Use -I$(srctree)/tools/lib/ to support out of source code tree builds, as noticed by Wang Nan ]
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2016-12-15 06:43:39 +08:00
|
|
|
xdp2-objs := bpf_load.o $(LIBBPF) xdp1_user.o
|
|
|
|
test_current_task_under_cgroup-objs := bpf_load.o $(LIBBPF) cgroup_helpers.o \
|
2016-08-12 23:57:04 +08:00
|
|
|
test_current_task_under_cgroup_user.o
|
samples/bpf: Switch over to libbpf
Now that libbpf under tools/lib/bpf/* is synced with the version from
samples/bpf, we can get rid most of the libbpf library here.
Committer notes:
Built it in a docker fedora rawhide container and ran it in the f25 host, seems
to work just like it did before this patch, i.e. the switch to tools/lib/bpf/
doesn't seem to have introduced problems and Joe said he tested it with
all the entries in samples/bpf/ and other code he found:
[root@f5065a7d6272 linux]# make -j4 O=/tmp/build/linux headers_install
<SNIP>
[root@f5065a7d6272 linux]# rm -rf /tmp/build/linux/samples/bpf/
[root@f5065a7d6272 linux]# make -j4 O=/tmp/build/linux samples/bpf/
make[1]: Entering directory '/tmp/build/linux'
CHK include/config/kernel.release
HOSTCC scripts/basic/fixdep
GEN ./Makefile
CHK include/generated/uapi/linux/version.h
Using /git/linux as source for kernel
CHK include/generated/utsrelease.h
HOSTCC scripts/basic/bin2c
HOSTCC arch/x86/tools/relocs_32.o
HOSTCC arch/x86/tools/relocs_64.o
LD samples/bpf/built-in.o
<SNIP>
HOSTCC samples/bpf/fds_example.o
HOSTCC samples/bpf/sockex1_user.o
/git/linux/samples/bpf/fds_example.c: In function 'bpf_prog_create':
/git/linux/samples/bpf/fds_example.c:63:6: warning: passing argument 2 of 'bpf_load_program' discards 'const' qualifier from pointer target type [-Wdiscarded-qualifiers]
insns, insns_cnt, "GPL", 0,
^~~~~
In file included from /git/linux/samples/bpf/libbpf.h:5:0,
from /git/linux/samples/bpf/bpf_load.h:4,
from /git/linux/samples/bpf/fds_example.c:15:
/git/linux/tools/lib/bpf/bpf.h:31:5: note: expected 'struct bpf_insn *' but argument is of type 'const struct bpf_insn *'
int bpf_load_program(enum bpf_prog_type type, struct bpf_insn *insns,
^~~~~~~~~~~~~~~~
HOSTCC samples/bpf/sockex2_user.o
<SNIP>
HOSTCC samples/bpf/xdp_tx_iptunnel_user.o
clang -nostdinc -isystem /usr/lib/gcc/x86_64-redhat-linux/6.2.1/include -I/git/linux/arch/x86/include -I./arch/x86/include/generated/uapi -I./arch/x86/include/generated -I/git/linux/include -I./include -I/git/linux/arch/x86/include/uapi -I/git/linux/include/uapi -I./include/generated/uapi -include /git/linux/include/linux/kconfig.h \
-D__KERNEL__ -D__ASM_SYSREG_H -Wno-unused-value -Wno-pointer-sign \
-Wno-compare-distinct-pointer-types \
-Wno-gnu-variable-sized-type-not-at-end \
-Wno-address-of-packed-member -Wno-tautological-compare \
-O2 -emit-llvm -c /git/linux/samples/bpf/sockex1_kern.c -o -| llc -march=bpf -filetype=obj -o samples/bpf/sockex1_kern.o
HOSTLD samples/bpf/tc_l2_redirect
<SNIP>
HOSTLD samples/bpf/lwt_len_hist
HOSTLD samples/bpf/xdp_tx_iptunnel
make[1]: Leaving directory '/tmp/build/linux'
[root@f5065a7d6272 linux]#
And then, in the host:
[root@jouet bpf]# mount | grep "docker.*devicemapper\/"
/dev/mapper/docker-253:0-1705076-9bd8aa1e0af33adce89ff42090847868ca676932878942be53941a06ec5923f9 on /var/lib/docker/devicemapper/mnt/9bd8aa1e0af33adce89ff42090847868ca676932878942be53941a06ec5923f9 type xfs (rw,relatime,context="system_u:object_r:container_file_t:s0:c73,c276",nouuid,attr2,inode64,sunit=1024,swidth=1024,noquota)
[root@jouet bpf]# cd /var/lib/docker/devicemapper/mnt/9bd8aa1e0af33adce89ff42090847868ca676932878942be53941a06ec5923f9/rootfs/tmp/build/linux/samples/bpf/
[root@jouet bpf]# file offwaketime
offwaketime: ELF 64-bit LSB executable, x86-64, version 1 (SYSV), dynamically linked, interpreter /lib64/ld-linux-x86-64.so.2, for GNU/Linux 2.6.32, BuildID[sha1]=f423d171e0487b2f802b6a792657f0f3c8f6d155, not stripped
[root@jouet bpf]# readelf -SW offwaketime
offwaketime offwaketime_kern.o offwaketime_user.o
[root@jouet bpf]# readelf -SW offwaketime_kern.o
There are 11 section headers, starting at offset 0x700:
Section Headers:
[Nr] Name Type Address Off Size ES Flg Lk Inf Al
[ 0] NULL 0000000000000000 000000 000000 00 0 0 0
[ 1] .strtab STRTAB 0000000000000000 000658 0000a8 00 0 0 1
[ 2] .text PROGBITS 0000000000000000 000040 000000 00 AX 0 0 4
[ 3] kprobe/try_to_wake_up PROGBITS 0000000000000000 000040 0000d8 00 AX 0 0 8
[ 4] .relkprobe/try_to_wake_up REL 0000000000000000 0005a8 000020 10 10 3 8
[ 5] tracepoint/sched/sched_switch PROGBITS 0000000000000000 000118 000318 00 AX 0 0 8
[ 6] .reltracepoint/sched/sched_switch REL 0000000000000000 0005c8 000090 10 10 5 8
[ 7] maps PROGBITS 0000000000000000 000430 000050 00 WA 0 0 4
[ 8] license PROGBITS 0000000000000000 000480 000004 00 WA 0 0 1
[ 9] version PROGBITS 0000000000000000 000484 000004 00 WA 0 0 4
[10] .symtab SYMTAB 0000000000000000 000488 000120 18 1 4 8
Key to Flags:
W (write), A (alloc), X (execute), M (merge), S (strings)
I (info), L (link order), G (group), T (TLS), E (exclude), x (unknown)
O (extra OS processing required) o (OS specific), p (processor specific)
[root@jouet bpf]# ./offwaketime | head -3
qemu-system-x86;entry_SYSCALL_64_fastpath;sys_ppoll;do_sys_poll;poll_schedule_timeout;schedule_hrtimeout_range;schedule_hrtimeout_range_clock;schedule;__schedule;-;try_to_wake_up;hrtimer_wakeup;__hrtimer_run_queues;hrtimer_interrupt;local_apic_timer_interrupt;smp_apic_timer_interrupt;__irqentry_text_start;cpuidle_enter_state;cpuidle_enter;call_cpuidle;cpu_startup_entry;rest_init;start_kernel;x86_64_start_reservations;x86_64_start_kernel;start_cpu;;swapper/0 4
firefox;entry_SYSCALL_64_fastpath;sys_poll;do_sys_poll;poll_schedule_timeout;schedule_hrtimeout_range;schedule_hrtimeout_range_clock;schedule;__schedule;-;try_to_wake_up;pollwake;__wake_up_common;__wake_up_sync_key;pipe_write;__vfs_write;vfs_write;sys_write;entry_SYSCALL_64_fastpath;;Timer 1
swapper/2;start_cpu;start_secondary;cpu_startup_entry;schedule_preempt_disabled;schedule;__schedule;-;---;; 61
[root@jouet bpf]#
Signed-off-by: Joe Stringer <joe@ovn.org>
Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Alexei Starovoitov <ast@fb.com>
Cc: Daniel Borkmann <daniel@iogearbox.net>
Cc: Wang Nan <wangnan0@huawei.com>
Cc: netdev@vger.kernel.org
Link: https://github.com/joestringer/linux/commit/5c40f54a52b1f437123c81e21873f4b4b1f9bd55.patch
Link: http://lkml.kernel.org/n/tip-xr8twtx7sjh5821g8qw47yxk@git.kernel.org
[ Use -I$(srctree)/tools/lib/ to support out of source code tree builds, as noticed by Wang Nan ]
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2016-12-15 06:43:39 +08:00
|
|
|
trace_event-objs := bpf_load.o $(LIBBPF) trace_event_user.o
|
|
|
|
sampleip-objs := bpf_load.o $(LIBBPF) sampleip_user.o
|
|
|
|
tc_l2_redirect-objs := bpf_load.o $(LIBBPF) tc_l2_redirect_user.o
|
|
|
|
lwt_len_hist-objs := bpf_load.o $(LIBBPF) lwt_len_hist_user.o
|
|
|
|
xdp_tx_iptunnel-objs := bpf_load.o $(LIBBPF) xdp_tx_iptunnel_user.o
|
2017-03-23 01:00:35 +08:00
|
|
|
test_map_in_map-objs := bpf_load.o $(LIBBPF) test_map_in_map_user.o
|
2017-03-23 08:27:36 +08:00
|
|
|
per_socket_stats_example-objs := $(LIBBPF) cookie_uid_helper_example.o
|
bpf: mini eBPF library, test stubs and verifier testsuite
1.
the library includes a trivial set of BPF syscall wrappers:
int bpf_create_map(int key_size, int value_size, int max_entries);
int bpf_update_elem(int fd, void *key, void *value);
int bpf_lookup_elem(int fd, void *key, void *value);
int bpf_delete_elem(int fd, void *key);
int bpf_get_next_key(int fd, void *key, void *next_key);
int bpf_prog_load(enum bpf_prog_type prog_type,
const struct sock_filter_int *insns, int insn_len,
const char *license);
bpf_prog_load() stores verifier log into global bpf_log_buf[] array
and BPF_*() macros to build instructions
2.
test stubs configure eBPF infra with 'unspec' map and program types.
These are fake types used by user space testsuite only.
3.
verifier tests valid and invalid programs and expects predefined
error log messages from kernel.
40 tests so far.
$ sudo ./test_verifier
#0 add+sub+mul OK
#1 unreachable OK
#2 unreachable2 OK
#3 out of range jump OK
#4 out of range jump2 OK
#5 test1 ld_imm64 OK
...
Signed-off-by: Alexei Starovoitov <ast@plumgrid.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2014-09-26 15:17:07 +08:00
|
|
|
|
|
|
|
# Tell kbuild to always build the programs
|
|
|
|
always := $(hostprogs-y)
|
2014-12-02 07:06:38 +08:00
|
|
|
always += sockex1_kern.o
|
2014-12-02 07:06:39 +08:00
|
|
|
always += sockex2_kern.o
|
2015-05-20 07:59:06 +08:00
|
|
|
always += sockex3_kern.o
|
2015-03-26 03:49:23 +08:00
|
|
|
always += tracex1_kern.o
|
2015-03-26 03:49:24 +08:00
|
|
|
always += tracex2_kern.o
|
2015-03-26 03:49:25 +08:00
|
|
|
always += tracex3_kern.o
|
2015-03-26 03:49:26 +08:00
|
|
|
always += tracex4_kern.o
|
samples/bpf: bpf_tail_call example for tracing
kprobe example that demonstrates how future seccomp programs may look like.
It attaches to seccomp_phase1() function and tail-calls other BPF programs
depending on syscall number.
Existing optimized classic BPF seccomp programs generated by Chrome look like:
if (sd.nr < 121) {
if (sd.nr < 57) {
if (sd.nr < 22) {
if (sd.nr < 7) {
if (sd.nr < 4) {
if (sd.nr < 1) {
check sys_read
} else {
if (sd.nr < 3) {
check sys_write and sys_open
} else {
check sys_close
}
}
} else {
} else {
} else {
} else {
} else {
}
the future seccomp using native eBPF may look like:
bpf_tail_call(&sd, &syscall_jmp_table, sd.nr);
which is simpler, faster and leaves more room for per-syscall checks.
Usage:
$ sudo ./tracex5
<...>-366 [001] d... 4.870033: : read(fd=1, buf=00007f6d5bebf000, size=771)
<...>-369 [003] d... 4.870066: : mmap
<...>-369 [003] d... 4.870077: : syscall=110 (one of get/set uid/pid/gid)
<...>-369 [003] d... 4.870089: : syscall=107 (one of get/set uid/pid/gid)
sh-369 [000] d... 4.891740: : read(fd=0, buf=00000000023d1000, size=512)
sh-369 [000] d... 4.891747: : write(fd=1, buf=00000000023d3000, size=512)
sh-369 [000] d... 4.891747: : read(fd=1, buf=00000000023d3000, size=512)
Signed-off-by: Alexei Starovoitov <ast@plumgrid.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2015-05-20 07:59:05 +08:00
|
|
|
always += tracex5_kern.o
|
2015-08-06 15:02:36 +08:00
|
|
|
always += tracex6_kern.o
|
2016-12-02 00:48:08 +08:00
|
|
|
always += sock_flags_kern.o
|
2016-07-25 20:55:02 +08:00
|
|
|
always += test_probe_write_user_kern.o
|
2015-10-21 11:02:35 +08:00
|
|
|
always += trace_output_kern.o
|
2015-04-02 08:12:13 +08:00
|
|
|
always += tcbpf1_kern.o
|
2016-08-20 02:55:44 +08:00
|
|
|
always += tcbpf2_kern.o
|
2016-11-10 07:36:34 +08:00
|
|
|
always += tc_l2_redirect_kern.o
|
2015-06-19 22:00:44 +08:00
|
|
|
always += lathist_kern.o
|
2016-02-18 11:58:59 +08:00
|
|
|
always += offwaketime_kern.o
|
2016-03-09 07:07:52 +08:00
|
|
|
always += spintest_kern.o
|
2016-03-09 07:07:54 +08:00
|
|
|
always += map_perf_test_kern.o
|
2016-04-07 09:43:31 +08:00
|
|
|
always += test_overhead_tp_kern.o
|
|
|
|
always += test_overhead_kprobe_kern.o
|
2016-05-06 10:49:14 +08:00
|
|
|
always += parse_varlen.o parse_simple.o parse_ldabs.o
|
2016-07-01 01:28:45 +08:00
|
|
|
always += test_cgrp2_tc_kern.o
|
Add sample for adding simple drop program to link
Add a sample program that only drops packets at the BPF_PROG_TYPE_XDP_RX
hook of a link. With the drop-only program, observed single core rate is
~20Mpps.
Other tests were run, for instance without the dropcnt increment or
without reading from the packet header, the packet rate was mostly
unchanged.
$ perf record -a samples/bpf/xdp1 $(</sys/class/net/eth0/ifindex)
proto 17: 20403027 drops/s
./pktgen_sample03_burst_single_flow.sh -i $DEV -d $IP -m $MAC -t 4
Running... ctrl^C to stop
Device: eth4@0
Result: OK: 11791017(c11788327+d2689) usec, 59622913 (60byte,0frags)
5056638pps 2427Mb/sec (2427186240bps) errors: 0
Device: eth4@1
Result: OK: 11791012(c11787906+d3106) usec, 60526944 (60byte,0frags)
5133311pps 2463Mb/sec (2463989280bps) errors: 0
Device: eth4@2
Result: OK: 11791019(c11788249+d2769) usec, 59868091 (60byte,0frags)
5077431pps 2437Mb/sec (2437166880bps) errors: 0
Device: eth4@3
Result: OK: 11795039(c11792403+d2636) usec, 59483181 (60byte,0frags)
5043067pps 2420Mb/sec (2420672160bps) errors: 0
perf report --no-children:
26.05% ksoftirqd/0 [mlx4_en] [k] mlx4_en_process_rx_cq
17.84% ksoftirqd/0 [mlx4_en] [k] mlx4_en_alloc_frags
5.52% ksoftirqd/0 [mlx4_en] [k] mlx4_en_free_frag
4.90% swapper [kernel.vmlinux] [k] poll_idle
4.14% ksoftirqd/0 [kernel.vmlinux] [k] get_page_from_freelist
2.78% ksoftirqd/0 [kernel.vmlinux] [k] __free_pages_ok
2.57% ksoftirqd/0 [kernel.vmlinux] [k] bpf_map_lookup_elem
2.51% swapper [mlx4_en] [k] mlx4_en_process_rx_cq
1.94% ksoftirqd/0 [kernel.vmlinux] [k] percpu_array_map_lookup_elem
1.45% swapper [mlx4_en] [k] mlx4_en_alloc_frags
1.35% ksoftirqd/0 [kernel.vmlinux] [k] free_one_page
1.33% swapper [kernel.vmlinux] [k] intel_idle
1.04% ksoftirqd/0 [mlx4_en] [k] 0x000000000001c5c5
0.96% ksoftirqd/0 [mlx4_en] [k] 0x000000000001c58d
0.93% ksoftirqd/0 [mlx4_en] [k] 0x000000000001c6ee
0.92% ksoftirqd/0 [mlx4_en] [k] 0x000000000001c6b9
0.89% ksoftirqd/0 [kernel.vmlinux] [k] __alloc_pages_nodemask
0.83% ksoftirqd/0 [mlx4_en] [k] 0x000000000001c686
0.83% ksoftirqd/0 [mlx4_en] [k] 0x000000000001c5d5
0.78% ksoftirqd/0 [mlx4_en] [k] mlx4_alloc_pages.isra.23
0.77% ksoftirqd/0 [mlx4_en] [k] 0x000000000001c5b4
0.77% ksoftirqd/0 [kernel.vmlinux] [k] net_rx_action
machine specs:
receiver - Intel E5-1630 v3 @ 3.70GHz
sender - Intel E5645 @ 2.40GHz
Mellanox ConnectX-3 @40G
Signed-off-by: Brenden Blanco <bblanco@plumgrid.com>
Acked-by: Alexei Starovoitov <ast@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
2016-07-20 03:16:51 +08:00
|
|
|
always += xdp1_kern.o
|
2016-07-20 03:16:57 +08:00
|
|
|
always += xdp2_kern.o
|
2016-08-12 23:57:04 +08:00
|
|
|
always += test_current_task_under_cgroup_kern.o
|
2016-09-02 09:37:25 +08:00
|
|
|
always += trace_event_kern.o
|
2016-09-02 09:37:26 +08:00
|
|
|
always += sampleip_kern.o
|
2016-12-01 00:10:11 +08:00
|
|
|
always += lwt_len_hist_kern.o
|
2016-12-08 07:53:14 +08:00
|
|
|
always += xdp_tx_iptunnel_kern.o
|
2017-03-23 01:00:35 +08:00
|
|
|
always += test_map_in_map_kern.o
|
2017-03-23 08:27:36 +08:00
|
|
|
always += cookie_uid_helper_example.o
|
2017-07-01 11:02:43 +08:00
|
|
|
always += tcp_synrto_kern.o
|
2017-07-01 11:02:45 +08:00
|
|
|
always += tcp_rwnd_kern.o
|
2017-07-01 11:02:48 +08:00
|
|
|
always += tcp_bufs_kern.o
|
bpf: mini eBPF library, test stubs and verifier testsuite
1.
the library includes a trivial set of BPF syscall wrappers:
int bpf_create_map(int key_size, int value_size, int max_entries);
int bpf_update_elem(int fd, void *key, void *value);
int bpf_lookup_elem(int fd, void *key, void *value);
int bpf_delete_elem(int fd, void *key);
int bpf_get_next_key(int fd, void *key, void *next_key);
int bpf_prog_load(enum bpf_prog_type prog_type,
const struct sock_filter_int *insns, int insn_len,
const char *license);
bpf_prog_load() stores verifier log into global bpf_log_buf[] array
and BPF_*() macros to build instructions
2.
test stubs configure eBPF infra with 'unspec' map and program types.
These are fake types used by user space testsuite only.
3.
verifier tests valid and invalid programs and expects predefined
error log messages from kernel.
40 tests so far.
$ sudo ./test_verifier
#0 add+sub+mul OK
#1 unreachable OK
#2 unreachable2 OK
#3 out of range jump OK
#4 out of range jump2 OK
#5 test1 ld_imm64 OK
...
Signed-off-by: Alexei Starovoitov <ast@plumgrid.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2014-09-26 15:17:07 +08:00
|
|
|
|
|
|
|
HOSTCFLAGS += -I$(objtree)/usr/include
|
samples/bpf: Switch over to libbpf
Now that libbpf under tools/lib/bpf/* is synced with the version from
samples/bpf, we can get rid most of the libbpf library here.
Committer notes:
Built it in a docker fedora rawhide container and ran it in the f25 host, seems
to work just like it did before this patch, i.e. the switch to tools/lib/bpf/
doesn't seem to have introduced problems and Joe said he tested it with
all the entries in samples/bpf/ and other code he found:
[root@f5065a7d6272 linux]# make -j4 O=/tmp/build/linux headers_install
<SNIP>
[root@f5065a7d6272 linux]# rm -rf /tmp/build/linux/samples/bpf/
[root@f5065a7d6272 linux]# make -j4 O=/tmp/build/linux samples/bpf/
make[1]: Entering directory '/tmp/build/linux'
CHK include/config/kernel.release
HOSTCC scripts/basic/fixdep
GEN ./Makefile
CHK include/generated/uapi/linux/version.h
Using /git/linux as source for kernel
CHK include/generated/utsrelease.h
HOSTCC scripts/basic/bin2c
HOSTCC arch/x86/tools/relocs_32.o
HOSTCC arch/x86/tools/relocs_64.o
LD samples/bpf/built-in.o
<SNIP>
HOSTCC samples/bpf/fds_example.o
HOSTCC samples/bpf/sockex1_user.o
/git/linux/samples/bpf/fds_example.c: In function 'bpf_prog_create':
/git/linux/samples/bpf/fds_example.c:63:6: warning: passing argument 2 of 'bpf_load_program' discards 'const' qualifier from pointer target type [-Wdiscarded-qualifiers]
insns, insns_cnt, "GPL", 0,
^~~~~
In file included from /git/linux/samples/bpf/libbpf.h:5:0,
from /git/linux/samples/bpf/bpf_load.h:4,
from /git/linux/samples/bpf/fds_example.c:15:
/git/linux/tools/lib/bpf/bpf.h:31:5: note: expected 'struct bpf_insn *' but argument is of type 'const struct bpf_insn *'
int bpf_load_program(enum bpf_prog_type type, struct bpf_insn *insns,
^~~~~~~~~~~~~~~~
HOSTCC samples/bpf/sockex2_user.o
<SNIP>
HOSTCC samples/bpf/xdp_tx_iptunnel_user.o
clang -nostdinc -isystem /usr/lib/gcc/x86_64-redhat-linux/6.2.1/include -I/git/linux/arch/x86/include -I./arch/x86/include/generated/uapi -I./arch/x86/include/generated -I/git/linux/include -I./include -I/git/linux/arch/x86/include/uapi -I/git/linux/include/uapi -I./include/generated/uapi -include /git/linux/include/linux/kconfig.h \
-D__KERNEL__ -D__ASM_SYSREG_H -Wno-unused-value -Wno-pointer-sign \
-Wno-compare-distinct-pointer-types \
-Wno-gnu-variable-sized-type-not-at-end \
-Wno-address-of-packed-member -Wno-tautological-compare \
-O2 -emit-llvm -c /git/linux/samples/bpf/sockex1_kern.c -o -| llc -march=bpf -filetype=obj -o samples/bpf/sockex1_kern.o
HOSTLD samples/bpf/tc_l2_redirect
<SNIP>
HOSTLD samples/bpf/lwt_len_hist
HOSTLD samples/bpf/xdp_tx_iptunnel
make[1]: Leaving directory '/tmp/build/linux'
[root@f5065a7d6272 linux]#
And then, in the host:
[root@jouet bpf]# mount | grep "docker.*devicemapper\/"
/dev/mapper/docker-253:0-1705076-9bd8aa1e0af33adce89ff42090847868ca676932878942be53941a06ec5923f9 on /var/lib/docker/devicemapper/mnt/9bd8aa1e0af33adce89ff42090847868ca676932878942be53941a06ec5923f9 type xfs (rw,relatime,context="system_u:object_r:container_file_t:s0:c73,c276",nouuid,attr2,inode64,sunit=1024,swidth=1024,noquota)
[root@jouet bpf]# cd /var/lib/docker/devicemapper/mnt/9bd8aa1e0af33adce89ff42090847868ca676932878942be53941a06ec5923f9/rootfs/tmp/build/linux/samples/bpf/
[root@jouet bpf]# file offwaketime
offwaketime: ELF 64-bit LSB executable, x86-64, version 1 (SYSV), dynamically linked, interpreter /lib64/ld-linux-x86-64.so.2, for GNU/Linux 2.6.32, BuildID[sha1]=f423d171e0487b2f802b6a792657f0f3c8f6d155, not stripped
[root@jouet bpf]# readelf -SW offwaketime
offwaketime offwaketime_kern.o offwaketime_user.o
[root@jouet bpf]# readelf -SW offwaketime_kern.o
There are 11 section headers, starting at offset 0x700:
Section Headers:
[Nr] Name Type Address Off Size ES Flg Lk Inf Al
[ 0] NULL 0000000000000000 000000 000000 00 0 0 0
[ 1] .strtab STRTAB 0000000000000000 000658 0000a8 00 0 0 1
[ 2] .text PROGBITS 0000000000000000 000040 000000 00 AX 0 0 4
[ 3] kprobe/try_to_wake_up PROGBITS 0000000000000000 000040 0000d8 00 AX 0 0 8
[ 4] .relkprobe/try_to_wake_up REL 0000000000000000 0005a8 000020 10 10 3 8
[ 5] tracepoint/sched/sched_switch PROGBITS 0000000000000000 000118 000318 00 AX 0 0 8
[ 6] .reltracepoint/sched/sched_switch REL 0000000000000000 0005c8 000090 10 10 5 8
[ 7] maps PROGBITS 0000000000000000 000430 000050 00 WA 0 0 4
[ 8] license PROGBITS 0000000000000000 000480 000004 00 WA 0 0 1
[ 9] version PROGBITS 0000000000000000 000484 000004 00 WA 0 0 4
[10] .symtab SYMTAB 0000000000000000 000488 000120 18 1 4 8
Key to Flags:
W (write), A (alloc), X (execute), M (merge), S (strings)
I (info), L (link order), G (group), T (TLS), E (exclude), x (unknown)
O (extra OS processing required) o (OS specific), p (processor specific)
[root@jouet bpf]# ./offwaketime | head -3
qemu-system-x86;entry_SYSCALL_64_fastpath;sys_ppoll;do_sys_poll;poll_schedule_timeout;schedule_hrtimeout_range;schedule_hrtimeout_range_clock;schedule;__schedule;-;try_to_wake_up;hrtimer_wakeup;__hrtimer_run_queues;hrtimer_interrupt;local_apic_timer_interrupt;smp_apic_timer_interrupt;__irqentry_text_start;cpuidle_enter_state;cpuidle_enter;call_cpuidle;cpu_startup_entry;rest_init;start_kernel;x86_64_start_reservations;x86_64_start_kernel;start_cpu;;swapper/0 4
firefox;entry_SYSCALL_64_fastpath;sys_poll;do_sys_poll;poll_schedule_timeout;schedule_hrtimeout_range;schedule_hrtimeout_range_clock;schedule;__schedule;-;try_to_wake_up;pollwake;__wake_up_common;__wake_up_sync_key;pipe_write;__vfs_write;vfs_write;sys_write;entry_SYSCALL_64_fastpath;;Timer 1
swapper/2;start_cpu;start_secondary;cpu_startup_entry;schedule_preempt_disabled;schedule;__schedule;-;---;; 61
[root@jouet bpf]#
Signed-off-by: Joe Stringer <joe@ovn.org>
Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Alexei Starovoitov <ast@fb.com>
Cc: Daniel Borkmann <daniel@iogearbox.net>
Cc: Wang Nan <wangnan0@huawei.com>
Cc: netdev@vger.kernel.org
Link: https://github.com/joestringer/linux/commit/5c40f54a52b1f437123c81e21873f4b4b1f9bd55.patch
Link: http://lkml.kernel.org/n/tip-xr8twtx7sjh5821g8qw47yxk@git.kernel.org
[ Use -I$(srctree)/tools/lib/ to support out of source code tree builds, as noticed by Wang Nan ]
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2016-12-15 06:43:39 +08:00
|
|
|
HOSTCFLAGS += -I$(srctree)/tools/lib/
|
2016-11-29 14:07:22 +08:00
|
|
|
HOSTCFLAGS += -I$(srctree)/tools/testing/selftests/bpf/
|
2016-12-09 10:46:19 +08:00
|
|
|
HOSTCFLAGS += -I$(srctree)/tools/lib/ -I$(srctree)/tools/include
|
|
|
|
HOSTCFLAGS += -I$(srctree)/tools/perf
|
2014-12-02 07:06:38 +08:00
|
|
|
|
|
|
|
HOSTCFLAGS_bpf_load.o += -I$(objtree)/usr/include -Wno-unused-variable
|
bpf: add sample usages for persistent maps/progs
This patch adds a couple of stand-alone examples on how BPF_OBJ_PIN
and BPF_OBJ_GET commands can be used.
Example with maps:
# ./fds_example -F /sys/fs/bpf/m -P -m -k 1 -v 42
bpf: map fd:3 (Success)
bpf: pin ret:(0,Success)
bpf: fd:3 u->(1:42) ret:(0,Success)
# ./fds_example -F /sys/fs/bpf/m -G -m -k 1
bpf: get fd:3 (Success)
bpf: fd:3 l->(1):42 ret:(0,Success)
# ./fds_example -F /sys/fs/bpf/m -G -m -k 1 -v 24
bpf: get fd:3 (Success)
bpf: fd:3 u->(1:24) ret:(0,Success)
# ./fds_example -F /sys/fs/bpf/m -G -m -k 1
bpf: get fd:3 (Success)
bpf: fd:3 l->(1):24 ret:(0,Success)
# ./fds_example -F /sys/fs/bpf/m2 -P -m
bpf: map fd:3 (Success)
bpf: pin ret:(0,Success)
# ./fds_example -F /sys/fs/bpf/m2 -G -m -k 1
bpf: get fd:3 (Success)
bpf: fd:3 l->(1):0 ret:(0,Success)
# ./fds_example -F /sys/fs/bpf/m2 -G -m
bpf: get fd:3 (Success)
Example with progs:
# ./fds_example -F /sys/fs/bpf/p -P -p
bpf: prog fd:3 (Success)
bpf: pin ret:(0,Success)
bpf sock:4 <- fd:3 attached ret:(0,Success)
# ./fds_example -F /sys/fs/bpf/p -G -p
bpf: get fd:3 (Success)
bpf: sock:4 <- fd:3 attached ret:(0,Success)
# ./fds_example -F /sys/fs/bpf/p2 -P -p -o ./sockex1_kern.o
bpf: prog fd:5 (Success)
bpf: pin ret:(0,Success)
bpf: sock:3 <- fd:5 attached ret:(0,Success)
# ./fds_example -F /sys/fs/bpf/p2 -G -p
bpf: get fd:3 (Success)
bpf: sock:4 <- fd:3 attached ret:(0,Success)
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: Alexei Starovoitov <ast@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
2015-10-29 21:58:10 +08:00
|
|
|
HOSTLOADLIBES_fds_example += -lelf
|
2014-12-02 07:06:38 +08:00
|
|
|
HOSTLOADLIBES_sockex1 += -lelf
|
2014-12-02 07:06:39 +08:00
|
|
|
HOSTLOADLIBES_sockex2 += -lelf
|
2015-05-20 07:59:06 +08:00
|
|
|
HOSTLOADLIBES_sockex3 += -lelf
|
2015-03-26 03:49:23 +08:00
|
|
|
HOSTLOADLIBES_tracex1 += -lelf
|
2015-03-26 03:49:24 +08:00
|
|
|
HOSTLOADLIBES_tracex2 += -lelf
|
2015-03-26 03:49:25 +08:00
|
|
|
HOSTLOADLIBES_tracex3 += -lelf
|
2015-03-26 03:49:26 +08:00
|
|
|
HOSTLOADLIBES_tracex4 += -lelf -lrt
|
samples/bpf: bpf_tail_call example for tracing
kprobe example that demonstrates how future seccomp programs may look like.
It attaches to seccomp_phase1() function and tail-calls other BPF programs
depending on syscall number.
Existing optimized classic BPF seccomp programs generated by Chrome look like:
if (sd.nr < 121) {
if (sd.nr < 57) {
if (sd.nr < 22) {
if (sd.nr < 7) {
if (sd.nr < 4) {
if (sd.nr < 1) {
check sys_read
} else {
if (sd.nr < 3) {
check sys_write and sys_open
} else {
check sys_close
}
}
} else {
} else {
} else {
} else {
} else {
}
the future seccomp using native eBPF may look like:
bpf_tail_call(&sd, &syscall_jmp_table, sd.nr);
which is simpler, faster and leaves more room for per-syscall checks.
Usage:
$ sudo ./tracex5
<...>-366 [001] d... 4.870033: : read(fd=1, buf=00007f6d5bebf000, size=771)
<...>-369 [003] d... 4.870066: : mmap
<...>-369 [003] d... 4.870077: : syscall=110 (one of get/set uid/pid/gid)
<...>-369 [003] d... 4.870089: : syscall=107 (one of get/set uid/pid/gid)
sh-369 [000] d... 4.891740: : read(fd=0, buf=00000000023d1000, size=512)
sh-369 [000] d... 4.891747: : write(fd=1, buf=00000000023d3000, size=512)
sh-369 [000] d... 4.891747: : read(fd=1, buf=00000000023d3000, size=512)
Signed-off-by: Alexei Starovoitov <ast@plumgrid.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2015-05-20 07:59:05 +08:00
|
|
|
HOSTLOADLIBES_tracex5 += -lelf
|
2015-08-06 15:02:36 +08:00
|
|
|
HOSTLOADLIBES_tracex6 += -lelf
|
2016-12-02 00:48:08 +08:00
|
|
|
HOSTLOADLIBES_test_cgrp2_sock2 += -lelf
|
2017-07-01 11:02:41 +08:00
|
|
|
HOSTLOADLIBES_load_sock_ops += -lelf
|
2016-07-25 20:55:02 +08:00
|
|
|
HOSTLOADLIBES_test_probe_write_user += -lelf
|
2015-10-21 11:02:35 +08:00
|
|
|
HOSTLOADLIBES_trace_output += -lelf -lrt
|
2015-06-19 22:00:44 +08:00
|
|
|
HOSTLOADLIBES_lathist += -lelf
|
2016-02-18 11:58:59 +08:00
|
|
|
HOSTLOADLIBES_offwaketime += -lelf
|
2016-03-09 07:07:52 +08:00
|
|
|
HOSTLOADLIBES_spintest += -lelf
|
2016-03-09 07:07:54 +08:00
|
|
|
HOSTLOADLIBES_map_perf_test += -lelf -lrt
|
2016-04-07 09:43:31 +08:00
|
|
|
HOSTLOADLIBES_test_overhead += -lelf -lrt
|
Add sample for adding simple drop program to link
Add a sample program that only drops packets at the BPF_PROG_TYPE_XDP_RX
hook of a link. With the drop-only program, observed single core rate is
~20Mpps.
Other tests were run, for instance without the dropcnt increment or
without reading from the packet header, the packet rate was mostly
unchanged.
$ perf record -a samples/bpf/xdp1 $(</sys/class/net/eth0/ifindex)
proto 17: 20403027 drops/s
./pktgen_sample03_burst_single_flow.sh -i $DEV -d $IP -m $MAC -t 4
Running... ctrl^C to stop
Device: eth4@0
Result: OK: 11791017(c11788327+d2689) usec, 59622913 (60byte,0frags)
5056638pps 2427Mb/sec (2427186240bps) errors: 0
Device: eth4@1
Result: OK: 11791012(c11787906+d3106) usec, 60526944 (60byte,0frags)
5133311pps 2463Mb/sec (2463989280bps) errors: 0
Device: eth4@2
Result: OK: 11791019(c11788249+d2769) usec, 59868091 (60byte,0frags)
5077431pps 2437Mb/sec (2437166880bps) errors: 0
Device: eth4@3
Result: OK: 11795039(c11792403+d2636) usec, 59483181 (60byte,0frags)
5043067pps 2420Mb/sec (2420672160bps) errors: 0
perf report --no-children:
26.05% ksoftirqd/0 [mlx4_en] [k] mlx4_en_process_rx_cq
17.84% ksoftirqd/0 [mlx4_en] [k] mlx4_en_alloc_frags
5.52% ksoftirqd/0 [mlx4_en] [k] mlx4_en_free_frag
4.90% swapper [kernel.vmlinux] [k] poll_idle
4.14% ksoftirqd/0 [kernel.vmlinux] [k] get_page_from_freelist
2.78% ksoftirqd/0 [kernel.vmlinux] [k] __free_pages_ok
2.57% ksoftirqd/0 [kernel.vmlinux] [k] bpf_map_lookup_elem
2.51% swapper [mlx4_en] [k] mlx4_en_process_rx_cq
1.94% ksoftirqd/0 [kernel.vmlinux] [k] percpu_array_map_lookup_elem
1.45% swapper [mlx4_en] [k] mlx4_en_alloc_frags
1.35% ksoftirqd/0 [kernel.vmlinux] [k] free_one_page
1.33% swapper [kernel.vmlinux] [k] intel_idle
1.04% ksoftirqd/0 [mlx4_en] [k] 0x000000000001c5c5
0.96% ksoftirqd/0 [mlx4_en] [k] 0x000000000001c58d
0.93% ksoftirqd/0 [mlx4_en] [k] 0x000000000001c6ee
0.92% ksoftirqd/0 [mlx4_en] [k] 0x000000000001c6b9
0.89% ksoftirqd/0 [kernel.vmlinux] [k] __alloc_pages_nodemask
0.83% ksoftirqd/0 [mlx4_en] [k] 0x000000000001c686
0.83% ksoftirqd/0 [mlx4_en] [k] 0x000000000001c5d5
0.78% ksoftirqd/0 [mlx4_en] [k] mlx4_alloc_pages.isra.23
0.77% ksoftirqd/0 [mlx4_en] [k] 0x000000000001c5b4
0.77% ksoftirqd/0 [kernel.vmlinux] [k] net_rx_action
machine specs:
receiver - Intel E5-1630 v3 @ 3.70GHz
sender - Intel E5645 @ 2.40GHz
Mellanox ConnectX-3 @40G
Signed-off-by: Brenden Blanco <bblanco@plumgrid.com>
Acked-by: Alexei Starovoitov <ast@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
2016-07-20 03:16:51 +08:00
|
|
|
HOSTLOADLIBES_xdp1 += -lelf
|
2016-07-20 03:16:57 +08:00
|
|
|
HOSTLOADLIBES_xdp2 += -lelf
|
2016-08-12 23:57:04 +08:00
|
|
|
HOSTLOADLIBES_test_current_task_under_cgroup += -lelf
|
2016-09-02 09:37:25 +08:00
|
|
|
HOSTLOADLIBES_trace_event += -lelf
|
2016-09-02 09:37:26 +08:00
|
|
|
HOSTLOADLIBES_sampleip += -lelf
|
2016-11-10 07:36:34 +08:00
|
|
|
HOSTLOADLIBES_tc_l2_redirect += -l elf
|
2016-12-01 00:10:11 +08:00
|
|
|
HOSTLOADLIBES_lwt_len_hist += -l elf
|
2016-12-08 07:53:14 +08:00
|
|
|
HOSTLOADLIBES_xdp_tx_iptunnel += -lelf
|
2017-03-23 01:00:35 +08:00
|
|
|
HOSTLOADLIBES_test_map_in_map += -lelf
|
2014-12-02 07:06:38 +08:00
|
|
|
|
2016-04-28 20:21:14 +08:00
|
|
|
# Allows pointing LLC/CLANG to a LLVM backend with bpf support, redefine on cmdline:
|
|
|
|
# make samples/bpf/ LLC=~/git/llvm/build/bin/llc CLANG=~/git/llvm/build/bin/clang
|
2016-04-28 20:20:53 +08:00
|
|
|
LLC ?= llc
|
2016-04-28 20:21:14 +08:00
|
|
|
CLANG ?= clang
|
2016-04-28 20:20:53 +08:00
|
|
|
|
2016-04-28 20:21:09 +08:00
|
|
|
# Trick to allow make to be run from this directory
|
|
|
|
all:
|
2016-11-22 16:30:26 +08:00
|
|
|
$(MAKE) -C ../../ $(CURDIR)/
|
2016-04-28 20:21:09 +08:00
|
|
|
|
|
|
|
clean:
|
2016-11-22 16:30:26 +08:00
|
|
|
$(MAKE) -C ../../ M=$(CURDIR) clean
|
2016-04-28 20:21:09 +08:00
|
|
|
@rm -f *~
|
|
|
|
|
2017-06-14 07:49:38 +08:00
|
|
|
$(obj)/syscall_nrs.s: $(src)/syscall_nrs.c
|
|
|
|
$(call if_changed_dep,cc_s_c)
|
|
|
|
|
|
|
|
$(obj)/syscall_nrs.h: $(obj)/syscall_nrs.s FORCE
|
|
|
|
$(call filechk,offsets,__SYSCALL_NRS_H__)
|
|
|
|
|
|
|
|
clean-files += syscall_nrs.h
|
|
|
|
|
|
|
|
FORCE:
|
|
|
|
|
|
|
|
|
2016-04-28 20:21:14 +08:00
|
|
|
# Verify LLVM compiler tools are available and bpf target is supported by llc
|
|
|
|
.PHONY: verify_cmds verify_target_bpf $(CLANG) $(LLC)
|
2016-04-28 20:20:58 +08:00
|
|
|
|
2016-04-28 20:21:14 +08:00
|
|
|
verify_cmds: $(CLANG) $(LLC)
|
|
|
|
@for TOOL in $^ ; do \
|
|
|
|
if ! (which -- "$${TOOL}" > /dev/null 2>&1); then \
|
|
|
|
echo "*** ERROR: Cannot find LLVM tool $${TOOL}" ;\
|
|
|
|
exit 1; \
|
|
|
|
else true; fi; \
|
|
|
|
done
|
2016-04-28 20:20:58 +08:00
|
|
|
|
2016-04-28 20:21:14 +08:00
|
|
|
verify_target_bpf: verify_cmds
|
2016-04-28 20:20:58 +08:00
|
|
|
@if ! (${LLC} -march=bpf -mattr=help > /dev/null 2>&1); then \
|
|
|
|
echo "*** ERROR: LLVM (${LLC}) does not support 'bpf' target" ;\
|
|
|
|
echo " NOTICE: LLVM version >= 3.7.1 required" ;\
|
|
|
|
exit 2; \
|
|
|
|
else true; fi
|
|
|
|
|
|
|
|
$(src)/*.c: verify_target_bpf
|
|
|
|
|
2017-06-14 07:49:38 +08:00
|
|
|
$(obj)/tracex5_kern.o: $(obj)/syscall_nrs.h
|
|
|
|
|
2016-04-05 01:01:33 +08:00
|
|
|
# asm/sysreg.h - inline assembly used by it is incompatible with llvm.
|
|
|
|
# But, there is no easy way to fix it, so just exclude it since it is
|
2015-11-13 06:07:46 +08:00
|
|
|
# useless for BPF samples.
|
2015-05-12 12:25:51 +08:00
|
|
|
$(obj)/%.o: $(src)/%.c
|
2017-06-22 04:48:27 +08:00
|
|
|
$(CLANG) $(NOSTDINC_FLAGS) $(LINUXINCLUDE) $(EXTRA_CFLAGS) -I$(obj) \
|
2015-11-13 06:07:46 +08:00
|
|
|
-D__KERNEL__ -D__ASM_SYSREG_H -Wno-unused-value -Wno-pointer-sign \
|
2016-05-06 10:49:14 +08:00
|
|
|
-Wno-compare-distinct-pointer-types \
|
2016-12-02 10:31:12 +08:00
|
|
|
-Wno-gnu-variable-sized-type-not-at-end \
|
|
|
|
-Wno-address-of-packed-member -Wno-tautological-compare \
|
2017-04-24 21:31:06 +08:00
|
|
|
-Wno-unknown-warning-option \
|
2016-04-28 20:20:53 +08:00
|
|
|
-O2 -emit-llvm -c $< -o -| $(LLC) -march=bpf -filetype=obj -o $@
|