libbpf: Handle unsupported mmap-based /sys/kernel/btf/vmlinux correctly

libbpf_err_ptr() helpers are meant to return NULL and set errno, if
there is an error. But btf_parse_raw_mmap() is meant to be used
internally and is expected to return ERR_PTR() values. Because of this
mismatch, when libbpf tries to mmap /sys/kernel/btf/vmlinux, we don't
detect the error correctly with IS_ERR() check, and never fallback to
old non-mmap-based way of loading vmlinux BTF.

Fix this by using proper ERR_PTR() returns internally.

Reported-by: Arnaldo Carvalho de Melo <[email protected]>
Reviewed-by: Arnaldo Carvalho de Melo <[email protected]>
Tested-by: Arnaldo Carvalho de Melo <[email protected]>
Fixes: 3c0421c ("libbpf: Use mmap to parse vmlinux BTF from sysfs")
Cc: Lorenz Bauer <[email protected]>
Signed-off-by: Andrii Nakryiko <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Alexei Starovoitov <[email protected]>

Author: anakryiko

tools/lib/bpf/btf.c

@@ -1384,20 +1384,20 @@ static struct btf *btf_parse_raw_mmap(const char *path, struct btf *base_btf)
 
 	fd = open(path, O_RDONLY);
 	if (fd < 0)
-		return libbpf_err_ptr(-errno);
+		return ERR_PTR(-errno);
 
 	if (fstat(fd, &st) < 0) {
 		err = -errno;
 		close(fd);
-		return libbpf_err_ptr(err);
+		return ERR_PTR(err);
 	}
 
 	data = mmap(NULL, st.st_size, PROT_READ, MAP_PRIVATE, fd, 0);
 	err = -errno;
 	close(fd);
 
 	if (data == MAP_FAILED)
-		return libbpf_err_ptr(err);
+		return ERR_PTR(err);
 
 	btf = btf_new(data, st.st_size, base_btf, true);
 	if (IS_ERR(btf))