On Fri, Jul 16, 2021 at 06:42:05PM -0700, Andrii Nakryiko wrote:
> On Wed, Jul 14, 2021 at 2:45 AM Jiri Olsa <jolsa@xxxxxxxxxx> wrote:
> >
> > From: Alan Maguire <alan.maguire@xxxxxxxxxx>
> >
> > kprobes can be placed on most instructions in a function, not
> > just entry, and ftrace and bpftrace support the function+offset
> > notification for probe placement. Adding parsing of func_name
> > into func+offset to bpf_program__attach_kprobe() allows the
> > user to specify
> >
> > SEC("kprobe/bpf_fentry_test5+0x6")
> >
> > ...for example, and the offset can be passed to perf_event_open_probe()
> > to support kprobe attachment.
> >
> > [jolsa: changed original code to use bpf_program__attach_kprobe_opts
> > and use dynamic allocation in sscanf]
> >
> > Signed-off-by: Alan Maguire <alan.maguire@xxxxxxxxxx>
> > Signed-off-by: Jiri Olsa <jolsa@xxxxxxxxxx>
> > ---
> > tools/lib/bpf/libbpf.c | 24 ++++++++++++++++++++++--
> > 1 file changed, 22 insertions(+), 2 deletions(-)
> >
> > diff --git a/tools/lib/bpf/libbpf.c b/tools/lib/bpf/libbpf.c
> > index d93a6f9408d1..abe6d4842bb0 100644
> > --- a/tools/lib/bpf/libbpf.c
> > +++ b/tools/lib/bpf/libbpf.c
> > @@ -10348,6 +10348,7 @@ static int perf_event_open_probe(bool uprobe, bool retprobe, const char *name,
> >
> > struct bpf_program_attach_kprobe_opts {
> > bool retprobe;
> > + unsigned long offset;
> > };
> >
> > static struct bpf_link*
> > @@ -10360,7 +10361,7 @@ bpf_program__attach_kprobe_opts(struct bpf_program *prog,
> > int pfd, err;
> >
> > pfd = perf_event_open_probe(false /* uprobe */, opts->retprobe, func_name,
> > - 0 /* offset */, -1 /* pid */);
> > + opts->offset, -1 /* pid */);
> > if (pfd < 0) {
> > pr_warn("prog '%s': failed to create %s '%s' perf event: %s\n",
> > prog->name, opts->retprobe ? "kretprobe" : "kprobe", func_name,
> > @@ -10394,12 +10395,31 @@ static struct bpf_link *attach_kprobe(const struct bpf_sec_def *sec,
> > struct bpf_program *prog)
> > {
> > struct bpf_program_attach_kprobe_opts opts;
> > + unsigned long offset = 0;
> > + struct bpf_link *link;
> > const char *func_name;
> > + char *func;
> > + int n, err;
> >
> > func_name = prog->sec_name + sec->len;
> > opts.retprobe = strcmp(sec->sec, "kretprobe/") == 0;
> >
> > - return bpf_program__attach_kprobe_opts(prog, func_name, &opts);
> > + n = sscanf(func_name, "%m[a-zA-Z0-9_.]+%lx", &func, &offset);
>
> could have used %li here to support both +0xabc and +123 forms
ok
>
> > + if (n < 1) {
> > + err = -EINVAL;
> > + pr_warn("kprobe name is invalid: %s\n", func_name);
> > + return libbpf_err_ptr(err);
> > + }
> > + if (opts.retprobe && offset != 0) {
> > + err = -EINVAL;
>
> leaking func here
ugh.. thanks
jirka
>
>
> > + pr_warn("kretprobes do not support offset specification\n");
> > + return libbpf_err_ptr(err);
> > + }
> > +
> > + opts.offset = offset;
> > + link = bpf_program__attach_kprobe_opts(prog, func, &opts);
> > + free(func);
> > + return link;
> > }
> >
> > struct bpf_link *bpf_program__attach_uprobe(struct bpf_program *prog,
> > --
> > 2.31.1
> >
>
