libbpf: Sanitise internal map names so they are not rejected by the kernel (113e6b7e) · Commits · 戴 / test

The kernel only accepts map names with alphanumeric characters, underscores and periods in their name. However, the auto-generated internal map names used by libbpf takes their prefix from the user-supplied BPF object name, which has no such restriction. This can lead to "Invalid argument" errors when trying to load a BPF program using global variables. Fix this by sanitising the map names, replacing any non-allowed characters with underscores. Fixes: ("bpf, libbpf: support global data/bss/rodata sections") Signed-off-by:

Toke Høiland-Jørgensen <toke@redhat.com> Signed-off-by:

Daniel Borkmann <daniel@iogearbox.net> Link: https://lore.kernel.org/bpf/20200217171701.215215-1-toke@redhat.com

tools/lib/bpf/libbpf.c

+7 −1

Original line number	Diff line number	Diff line
		@@ -24,6 +24,7 @@
		#include <endian.h>
		#include <fcntl.h>
		#include <errno.h>
		#include <ctype.h>
		#include <asm/unistd.h>
		#include <linux/err.h>
		#include <linux/kernel.h>
		@@ -1283,7 +1284,7 @@ static size_t bpf_map_mmap_sz(const struct bpf_map *map)
		static char internal_map_name(struct bpf_object obj,
		enum libbpf_map_type type)
		{
		char map_name[BPF_OBJ_NAME_LEN];
		char map_name[BPF_OBJ_NAME_LEN], *p;
		const char *sfx = libbpf_type_to_btf_name[type];
		int sfx_len = max((size_t)7, strlen(sfx));
		int pfx_len = min((size_t)BPF_OBJ_NAME_LEN - sfx_len - 1,
		@@ -1292,6 +1293,11 @@ static char internal_map_name(struct bpf_object obj,
		snprintf(map_name, sizeof(map_name), "%.s%.s", pfx_len, obj->name,
		sfx_len, libbpf_type_to_btf_name[type]);

		/* sanitise map name to characters allowed by kernel */
		for (p = map_name; *p && p < map_name + sizeof(map_name); p++)
		if (!isalnum(p) && p != '_' && *p != '.')
		*p = '_';

		return strdup(map_name);
		}

Admin message