[PATCH bpf-next v2] libbpf: Allow Golang symbols in uprobe secdef

[Hengqi Chen <hengqi.chen@xxxxxxxxx>]

Golang symbols in ELF files are different from C/C++
which contains special characters like '*', '(' and ')'.
With generics, things get more complicated, there are
symbols like:

  github.com/cilium/ebpf/internal.(*Deque[go.shape.interface { Format(fm
  t.State, int32); TypeName() string;github.com/cilium/ebpf/btf.copy() g
  ithub.com/cilium/ebpf/btf.Type}]).Grow

Matching such symbols using `%m[^\n]` in sscanf, this
excludes newline which typically does not appear in ELF
symbols. This should work in most use-cases and also
work for unicode letters in identifiers. If newline do
show up in ELF symbols, users can still attach to such
symbol by specifying bpf_uprobe_opts::func_name.

A working example can be found at this repo ([0]).

  [0]: https://github.com/chenhengqi/libbpf-go-symbols

Suggested-by: Andrii Nakryiko <andrii@xxxxxxxxxx>
Cc: Arnaldo Carvalho de Melo <acme@xxxxxxxxxx>
Signed-off-by: Hengqi Chen <hengqi.chen@xxxxxxxxx>
---
 tools/lib/bpf/libbpf.c | 22 ++++++++++++++++------
 1 file changed, 16 insertions(+), 6 deletions(-)

diff --git a/tools/lib/bpf/libbpf.c b/tools/lib/bpf/libbpf.c
index b4758e54a815..31b8b252e614 100644
--- a/tools/lib/bpf/libbpf.c
+++ b/tools/lib/bpf/libbpf.c
@@ -11114,7 +11114,7 @@ static int attach_uprobe_multi(const struct bpf_program *prog, long cookie, stru
 
 	*link = NULL;
 
-	n = sscanf(prog->sec_name, "%m[^/]/%m[^:]:%ms",
+	n = sscanf(prog->sec_name, "%m[^/]/%m[^:]:%m[^\n]",
 		   &probe_type, &binary_path, &func_name);
 	switch (n) {
 	case 1:
@@ -11624,14 +11624,14 @@ bpf_program__attach_uprobe_opts(const struct bpf_program *prog, pid_t pid,
 static int attach_uprobe(const struct bpf_program *prog, long cookie, struct bpf_link **link)
 {
 	DECLARE_LIBBPF_OPTS(bpf_uprobe_opts, opts);
-	char *probe_type = NULL, *binary_path = NULL, *func_name = NULL;
-	int n, ret = -EINVAL;
+	char *probe_type = NULL, *binary_path = NULL, *func_name = NULL, *func_off;
+	int n, c, ret = -EINVAL;
 	long offset = 0;
 
 	*link = NULL;
 
-	n = sscanf(prog->sec_name, "%m[^/]/%m[^:]:%m[a-zA-Z0-9_.@]+%li",
-		   &probe_type, &binary_path, &func_name, &offset);
+	n = sscanf(prog->sec_name, "%m[^/]/%m[^:]:%m[^\n]",
+		   &probe_type, &binary_path, &func_name);
 	switch (n) {
 	case 1:
 		/* handle SEC("u[ret]probe") - format is valid, but auto-attach is impossible. */
@@ -11642,7 +11642,17 @@ static int attach_uprobe(const struct bpf_program *prog, long cookie, struct bpf
 			prog->name, prog->sec_name);
 		break;
 	case 3:
-	case 4:
+		/* check if user specifies `+offset`, if yes, this should be
+		 * the last part of the string, make sure sscanf read to EOL
+		 */
+		func_off = strrchr(func_name, '+');
+		if (func_off) {
+			n = sscanf(func_off, "+%li%n", &offset, &c);
+			if (n == 1 && *(func_off + c) == '\0')
+				func_off[0] = '\0';
+			else
+				offset = 0;
+		}
 		opts.retprobe = strcmp(probe_type, "uretprobe") == 0 ||
 				strcmp(probe_type, "uretprobe.s") == 0;
 		if (opts.retprobe && offset != 0) {
-- 
2.34.1