bpf: Fix bug in mmap() implementation for BPF array map (333291ce) · Commits · 戴 / test

mmap() subsystem allows user-space application to memory-map region with initial page offset. This wasn't taken into account in initial implementation of BPF array memory-mapping. This would result in wrong pages, not taking into account requested page shift, being memory-mmaped into user-space. This patch fixes this gap and adds a test for such scenario. Fixes: ("bpf: Add mmap() support for BPF_MAP_TYPE_ARRAY") Signed-off-by:

Andrii Nakryiko <andriin@fb.com> Signed-off-by:

Alexei Starovoitov <ast@kernel.org> Acked-by:

Yonghong Song <yhs@fb.com> Link: https://lore.kernel.org/bpf/20200512235925.3817805-1-andriin@fb.com

kernel/bpf/arraymap.c

+6 −1

Original line number	Diff line number	Diff line
		@@ -486,7 +486,12 @@ static int array_map_mmap(struct bpf_map map, struct vm_area_struct vma)
		if (!(map->map_flags & BPF_F_MMAPABLE))
		return -EINVAL;

		return remap_vmalloc_range(vma, array_map_vmalloc_addr(array), pgoff);
		if (vma->vm_pgoff * PAGE_SIZE + (vma->vm_end - vma->vm_start) >
		PAGE_ALIGN((u64)array->map.max_entries * array->elem_size))
		return -EINVAL;

		return remap_vmalloc_range(vma, array_map_vmalloc_addr(array),
		vma->vm_pgoff + pgoff);
		}

		const struct bpf_map_ops array_map_ops = {

tools/testing/selftests/bpf/prog_tests/mmap.c

+8 −0

Original line number	Diff line number	Diff line
		@@ -217,6 +217,14 @@ void test_mmap(void)

		munmap(tmp2, 4 * page_size);

		/* map all 4 pages, but with pg_off=1 page, should fail */
		tmp1 = mmap(NULL, 4 * page_size, PROT_READ, MAP_SHARED \| MAP_FIXED,
		data_map_fd, page_size /* initial page shift */);
		if (CHECK(tmp1 != MAP_FAILED, "adv_mmap7", "unexpected success")) {
		munmap(tmp1, 4 * page_size);
		goto cleanup;
		}

		tmp1 = mmap(NULL, map_sz, PROT_READ, MAP_SHARED, data_map_fd, 0);
		if (CHECK(tmp1 == MAP_FAILED, "last_mmap", "failed %d\n", errno))
		goto cleanup;

Admin message