Thread (38 messages) 38 messages, 6 authors, 2017-04-25

Re: [PATCH v3 3/7] kprobes: validate the symbol name length

From: Naveen N. Rao <hidden>
Date: 2017-04-23 15:45:41
Also in: lkml

Excerpts from Masami Hiramatsu's message of April 21, 2017 19:12:
On Wed, 19 Apr 2017 16:38:22 +0000
"Naveen N. Rao" [off-list ref] wrote:
=20
quoted
Excerpts from Masami Hiramatsu's message of April 19, 2017 20:07:
quoted
On Wed, 19 Apr 2017 18:21:02 +0530
"Naveen N. Rao" [off-list ref] wrote:
=20
quoted
When a kprobe is being registered, we use the symbol_name field to
lookup the address where the probe should be placed. Since this is a
user-provided field, let's ensure that the length of the string is
within expected limits.
=20
Would we really need this? Of course it may filter out longer
strings... anyway such name should be rejected by kallsyms.
=20
I felt this would be good to have generically, as kallsyms does many=20
string operations on the symbol name, including an unbounded=20
strchr().
=20
OK, so this is actually for performance reason.
=20
quoted
=20
quoted
=20
[...]
quoted
diff --git a/kernel/kprobes.c b/kernel/kprobes.c
index 6a128f3a7ed1..bb86681c8a10 100644
--- a/kernel/kprobes.c
+++ b/kernel/kprobes.c
@@ -1382,6 +1382,28 @@ bool within_kprobe_blacklist(unsigned long add=
r)
quoted
quoted
quoted
 	return false;
 }
=20
+bool is_valid_kprobe_symbol_name(const char *name)
=20
This just check the length of symbol_name buffer, and can contain
some invalid chars.
=20
Yes, I kept the function name generic incase we would like to do more=20
validation in future, plus it's shorter than=20
is_valid_kprobe_symbol_name_len() ;-)
=20
OK, if this is enough general, we'd better define this in
kernel/kallsyms.c or in kallsyms.h. Of course the function
should be called is_valid_symbol_name(). :-)
I actually think this should be done in kprobes itself. The primary=20
intent is to perform such validation right when we first obtain the=20
input from the user. In this case, however, kallsyms_lookup_name() is=20
also an exported symbol, so I do think some validation there would be=20
good to have as well.
=20
quoted
quoted
quoted
+{
+	size_t sym_len;
+	char *s;
+
+	s =3D strchr(name, ':');
=20
Hmm.. this should be strnchr(). I re-factored the code that moved the=20
strnlen() above this below. I'll fix this.
=20
quoted
quoted
+	if (s) {
+		sym_len =3D strnlen(s+1, KSYM_NAME_LEN);
=20
If you use strnlen() here, you just need to ensure sym_len < KSYM_NAME=
_LEN.
quoted
=20
Hmm.. not sure I follow. Are you saying the check for sym_len <=3D 0 is=20
not needed?
=20
You can check sym_len !=3D 0, but anyway, here we concern about
"longer" string (for performance reason), we can focus on
such case.
(BTW, could you also check the name !=3D NULL at first?)
=20
So, what I think it can be;
=20
if (strnlen(s+1, KSYM_NAME_LEN) =3D=3D KSYM_NAME_LEN ||
    (size_t)(s - name) >=3D MODULE_NAME_LEN)
	return false;
Sure, thanks. I clearly need to refactor this code better!

- Naveen

=
Keyboard shortcuts
hback out one level
jnext message in thread
kprevious message in thread
ldrill in
Escclose help / fold thread tree
?toggle this help