| Message ID | 20220518205924.399291-2-benjamin.tissoires@redhat.com (mailing list archive) |
|---|---|
| State | Accepted |
| Commit | 979497674e63666a99fd7d242dba53a5ca5d628b |
| Delegated to: | BPF |
| Headers | show |
| Series | Introduce eBPF support for HID devices | expand |
On Wed, May 18, 2022 at 1:59 PM Benjamin Tissoires <benjamin.tissoires@redhat.com> wrote: > > Tracing and syscall BPF program types are very convenient to add BPF > capabilities to subsystem otherwise not BPF capable. > When we add kfuncs capabilities to those program types, we can add > BPF features to subsystems without having to touch BPF core. > > Signed-off-by: Benjamin Tissoires <benjamin.tissoires@redhat.com> Applied this patch, since Yosry's work needs it too.
diff --git a/kernel/bpf/btf.c b/kernel/bpf/btf.c index 2f0b0440131c..7bccaa4646e5 100644 --- a/kernel/bpf/btf.c +++ b/kernel/bpf/btf.c @@ -202,6 +202,8 @@ enum btf_kfunc_hook { BTF_KFUNC_HOOK_XDP, BTF_KFUNC_HOOK_TC, BTF_KFUNC_HOOK_STRUCT_OPS, + BTF_KFUNC_HOOK_TRACING, + BTF_KFUNC_HOOK_SYSCALL, BTF_KFUNC_HOOK_MAX, }; @@ -7110,6 +7112,10 @@ static int bpf_prog_type_to_kfunc_hook(enum bpf_prog_type prog_type) return BTF_KFUNC_HOOK_TC; case BPF_PROG_TYPE_STRUCT_OPS: return BTF_KFUNC_HOOK_STRUCT_OPS; + case BPF_PROG_TYPE_TRACING: + return BTF_KFUNC_HOOK_TRACING; + case BPF_PROG_TYPE_SYSCALL: + return BTF_KFUNC_HOOK_SYSCALL; default: return BTF_KFUNC_HOOK_MAX; }
Tracing and syscall BPF program types are very convenient to add BPF capabilities to subsystem otherwise not BPF capable. When we add kfuncs capabilities to those program types, we can add BPF features to subsystems without having to touch BPF core. Signed-off-by: Benjamin Tissoires <benjamin.tissoires@redhat.com> --- changes in v5: - also add syscalls new in v4: - I think this is where I need to add my new kfuncs, though in the end I need to be able to change the incoming data, so maybe only fmod_ret is the one we need to be able to be RW. --- kernel/bpf/btf.c | 6 ++++++ 1 file changed, 6 insertions(+)