exec: load_script: don't blindly truncate shebang string
authorOleg Nesterov <[email protected]>
Thu, 3 Jan 2019 23:28:07 +0000 (15:28 -0800)
committerLinus Torvalds <[email protected]>
Fri, 4 Jan 2019 21:13:47 +0000 (13:13 -0800)
load_script() simply truncates bprm->buf and this is very wrong if the
length of shebang string exceeds BINPRM_BUF_SIZE-2.  This can silently
truncate i_arg or (worse) we can execute the wrong binary if buf[2:126]
happens to be the valid executable path.

Change load_script() to return ENOEXEC if it can't find '\n' or zero in
bprm->buf.  Note that '\0' can come from either
prepare_binprm()->memset() or from kernel_read(), we do not care.

Link: http://lkml.kernel.org/r/[email protected]
Signed-off-by: Oleg Nesterov <[email protected]>
Acked-by: Kees Cook <[email protected]>
Acked-by: Michal Hocko <[email protected]>
Cc: Ben Woodard <[email protected]>
Cc: "Eric W. Biederman" <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
fs/binfmt_script.c

index 7cde3f46ad263ab084aafaefa14161902f33d4f1..d0078cbb718b4855dc520dbfae91223d499da81a 100644 (file)
@@ -42,10 +42,14 @@ static int load_script(struct linux_binprm *bprm)
        fput(bprm->file);
        bprm->file = NULL;
 
-       bprm->buf[BINPRM_BUF_SIZE - 1] = '\0';
-       if ((cp = strchr(bprm->buf, '\n')) == NULL)
-               cp = bprm->buf+BINPRM_BUF_SIZE-1;
+       for (cp = bprm->buf+2;; cp++) {
+               if (cp >= bprm->buf + BINPRM_BUF_SIZE)
+                       return -ENOEXEC;
+               if (!*cp || (*cp == '\n'))
+                       break;
+       }
        *cp = '\0';
+
        while (cp > bprm->buf) {
                cp--;
                if ((*cp == ' ') || (*cp == '\t'))