Quoting Ariadne Conill:
"In several other operating systems, it is a hard requirement that the first argument to execve(2) be the name of a program, thus prohibiting a scenario where argc < 1. POSIX 2017 also recommends this behaviour, but it is not an explicit requirement[1]:
The argument arg0 should point to a filename string that is associated with the process being started by one of the exec functions. ... Interestingly, Michael Kerrisk opened an issue about this in 2008[2], but there was no consensus to support fixing this issue then. Hopefully now that CVE-2021-4034 shows practical exploitative use[3] of this bug in a shellcode, we can reconsider."
An examination of existing[4] users of execve(..., NULL, NULL) shows mostly test code, or example rootkit code. While rejecting a NULL argv would be preferred, it looks like the main cause of userspace confusion is an assumption that argc >= 1, and buggy programs may skip argv[0] when iterating. To protect against userspace bugs of this nature, insert an extra NULL pointer in argv when argc == 0, so that argv[1] != envp[0].
Note that this is only done in the argc == 0 case because some userspace programs expect to find envp at exactly argv[argc]. The overlap of these two misguided assumptions is believed to be zero.
[1] https://pubs.opengroup.org/onlinepubs/9699919799/functions/exec.html [2] https://bugzilla.kernel.org/show_bug.cgi?id=8408 [3] https://www.qualys.com/2022/01/25/cve-2021-4034/pwnkit.txt [4] https://codesearch.debian.net/search?q=execve%5C+*%5C%28%5B%5E%2C%5D%2B%2C+*...
Reported-by: Ariadne Conill ariadne@dereferenced.org Reported-by: Michael Kerrisk mtk.manpages@gmail.com Cc: Matthew Wilcox willy@infradead.org Cc: Christian Brauner brauner@kernel.org Cc: Rich Felker dalias@libc.org Cc: Eric Biederman ebiederm@xmission.com Cc: Alexander Viro viro@zeniv.linux.org.uk Cc: linux-fsdevel@vger.kernel.org Cc: stable@vger.kernel.org Signed-off-by: Kees Cook keescook@chromium.org --- fs/binfmt_elf.c | 10 +++++++++- fs/exec.c | 7 ++++++- 2 files changed, 15 insertions(+), 2 deletions(-)
diff --git a/fs/binfmt_elf.c b/fs/binfmt_elf.c index 605017eb9349..e456c48658ad 100644 --- a/fs/binfmt_elf.c +++ b/fs/binfmt_elf.c @@ -297,7 +297,8 @@ create_elf_tables(struct linux_binprm *bprm, const struct elfhdr *exec, ei_index = elf_info - (elf_addr_t *)mm->saved_auxv; sp = STACK_ADD(p, ei_index);
- items = (argc + 1) + (envc + 1) + 1; + /* Make room for extra pointer when argc == 0. See below. */ + items = (min(argc, 1) + 1) + (envc + 1) + 1; bprm->p = STACK_ROUND(sp, items);
/* Point sp at the lowest address on the stack */ @@ -326,6 +327,13 @@ create_elf_tables(struct linux_binprm *bprm, const struct elfhdr *exec,
/* Populate list of argv pointers back to argv strings. */ p = mm->arg_end = mm->arg_start; + /* + * Include an extra NULL pointer in argv when argc == 0 so + * that argv[1] != envp[0] to help userspace programs from + * mishandling argc == 0. See fs/exec.c bprm_stack_limits(). + */ + if (argc == 0 && put_user(0, sp++)) + return -EFAULT; while (argc-- > 0) { size_t len; if (put_user((elf_addr_t)p, sp++)) diff --git a/fs/exec.c b/fs/exec.c index 79f2c9483302..0b36384e55b1 100644 --- a/fs/exec.c +++ b/fs/exec.c @@ -495,8 +495,13 @@ static int bprm_stack_limits(struct linux_binprm *bprm) * the stack. They aren't stored until much later when we can't * signal to the parent that the child has run out of stack space. * Instead, calculate it here so it's possible to fail gracefully. + * + * In the case of argc < 1, make sure there is a NULL pointer gap + * between argv and envp to ensure confused userspace programs don't + * start processing from argv[1], thinking argc can never be 0, + * to block them from walking envp by accident. See fs/binfmt_elf.c. */ - ptr_size = (bprm->argc + bprm->envc) * sizeof(void *); + ptr_size = (min(bprm->argc, 1) + bprm->envc) * sizeof(void *); if (limit <= ptr_size) return -E2BIG; limit -= ptr_size;