bpf: Use tracing helpers for lsm programs
Subsystems (7)
Jiri Olsa
reviewerPATCH 1/1 bpf: Use tracing helpers for lsm programs
Currenty lsm uses bpf_tracing_func_proto helpers which do not include stack trace or perf event output. It's useful to have those for bpftrace lsm support [1].
Using tracing_prog_func_proto helpers for lsm programs.
[1] https://github.com/iovisor/bpftrace/pull/1347
--- include/linux/bpf.h | 3 +++ kernel/bpf/bpf_lsm.c | 2 +- kernel/trace/bpf_trace.c | 2 +- 3 files changed, 5 insertions(+), 2 deletions(-)
include/linux/bpf.h +3 -0
@@ -1628,6 +1628,9 @@ extern const struct bpf_func_proto bpf_ringbuf_query_proto;const struct bpf_func_proto *bpf_tracing_func_proto (enum bpf_func_id func_id,const struct bpf_prog *prog); +const struct bpf_func_proto *tracing_prog_func_proto ( +enum bpf_func_id func_id,const struct bpf_prog *prog); +/* Shared helpers among cBPF and eBPF. */ void bpf_user_rnd_init_once (void );u64 bpf_user_rnd_u32 (u64 r1,u64 r2,u64 r3,u64 r4,u64 r5);
kernel/bpf/bpf_lsm.c +1 -1
@@ -49,6 +49,6 @@ const struct bpf_prog_ops lsm_prog_ops = { };const struct bpf_verifier_ops lsm_verifier_ops = { - .get_func_proto = bpf_tracing_func_proto, + .get_func_proto = tracing_prog_func_proto, .is_valid_access = btf_ctx_access, };
kernel/trace/bpf_trace.c +1 -1
@@ -1467,7 +1467,7 @@ raw_tp_prog_func_proto(enum bpf_func_id func_id, const struct bpf_prog *prog) } } -static const struct bpf_func_proto * +const struct bpf_func_proto *tracing_prog_func_proto (enum bpf_func_id func_id,const struct bpf_prog *prog) {switch (func_id) {
Signature · 1 line
2.25.4
Song Liu
maintainerQuoted 6 lines · Jiri Olsa
Currenty lsm uses
bpf_tracing_func_protohelpers which do not include stack trace or perf event output. It's useful to have those for bpftrace lsm support [1].Using
tracing_prog_func_protohelpers for lsm programs.
How about using raw_tp_prog_func_proto?
Thanks,
Song
PS: Please tag the patch with subject prefix "PATCH bpf" for "PATCH bpf-next". I think this one belongs to bpf-next, which means we should wait after the merge window.
Quoted 10 lines · Song Liu
Quoted 6 lines · Jiri Olsa
Currenty lsm uses
bpf_tracing_func_protohelpers which do not include stack trace or perf event output. It's useful to have those for bpftrace lsm support [1].Using
tracing_prog_func_protohelpers for lsm programs.How about using
raw_tp_prog_func_proto?
why?
I think skb/xdp_output is useful for lsm progs too. So I've applied the patch.
Quoted 3 lines
PS: Please tag the patch with subject prefix "PATCH bpf" for "PATCH bpf-next". I think this one belongs to bpf-next, which means we should wait after the merge window.
+1.
Jiri,
pls tag the subject properly.
Song Liu
maintainerQuoted 9 lines · Alexei Starovoitov
Quoted 7 lines · Song Liu
Quoted 5 lines · Jiri Olsa
Currenty lsm uses
bpf_tracing_func_protohelpers which do not include stack trace or perf event output. It's useful to have those for bpftrace lsm support [1].
Quoted 7 lines
Quoted 3 lines
Quoted 1 line
Using
tracing_prog_func_protohelpers for lsm programs.How about using
raw_tp_prog_func_proto?why?
I think skb/xdp_output is useful for lsm progs too. So I've applied the patch.
The only reason I asked was that the commit log only said stack trace and perf event output. No objections allowing other helpers for lsm.
Thanks,
Song
Quoted 8 lines · Alexei Starovoitov
Quoted 7 lines · Song Liu
Quoted 5 lines · Jiri Olsa
Currenty lsm uses
bpf_tracing_func_protohelpers which do not include stack trace or perf event output. It's useful to have those for bpftrace lsm support [1].
Quoted 7 lines
Quoted 3 lines
Quoted 1 line
Using
tracing_prog_func_protohelpers for lsm programs.How about using
raw_tp_prog_func_proto?why?
I think skb/xdp_output is useful for lsm progs too. So I've applied the patch.
right, it's also where d_path will be as well
Quoted 4 lines
Quoted 4 lines
PS: Please tag the patch with subject prefix "PATCH bpf" for "PATCH bpf-next". I think this one belongs to bpf-next, which means we should wait after the merge window.
I must have missed info about that,
thanks for info
Quoted 4 lines
+1.
Jiri,
pls tag the subject properly.
will do, sry
thanks,
jirka
Quoted 21 lines · Jiri Olsa
Quoted 8 lines · Alexei Starovoitov
Quoted 7 lines · Song Liu
Quoted 5 lines · Jiri Olsa
Currenty lsm uses
bpf_tracing_func_protohelpers which do not include stack trace or perf event output. It's useful to have those for bpftrace lsm support [1].Quoted 7 lines
Quoted 3 lines
Quoted 1 line
Using
tracing_prog_func_protohelpers for lsm programs.How about using
raw_tp_prog_func_proto?why?
I think skb/xdp_output is useful for lsm progs too. So I've applied the patch.right, it's also where
d_pathwill be as wellQuoted 1 line
Quoted 6 lines
Quoted 3 lines
Quoted 3 lines
PS: Please tag the patch with subject prefix "PATCH bpf" for "PATCH bpf-next". I think this one belongs to bpf-next, which means we should wait after the merge window.
I must have missed info about that,
thanks for info
Thanks for adding this! LGTM as well.
- KP
Quoted 11 lines
Quoted 5 lines
+1.
Jiri,
pls tag the subject properly.will do, sry
thanks,
jirka