From: Josh Triplett <josh-iaAMLnmF4UmaiuxdJuQwMA@public.gmane.org>
To: Andrew Morton
<akpm-de/tnXTf+JLsfHDXvbKv3WD2FQJk+8+b@public.gmane.org>,
Andy Lutomirski <luto-kltTT9wpgjJwATOyAt5JVQ@public.gmane.org>,
"Eric W. Biederman"
<ebiederm-aS9lmoZGLiVWk0Htik3J/w@public.gmane.org>,
Kees Cook <keescook-F7+t8E8rja9g9hUCZPvPmw@public.gmane.org>,
mtk.manpages-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org,
linux-api-u79uwXL29TY76Z2rM5mHXA@public.gmane.org,
linux-man-u79uwXL29TY76Z2rM5mHXA@public.gmane.org,
linux-kernel-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
Subject: [PATCHv2 2/2] groups: Allow unprivileged processes to use setgroups to drop groups
Date: Sat, 15 Nov 2014 13:02:53 -0800 [thread overview]
Message-ID: <311e7d2982ee36b7c5782dd47e2d589748ef3759.1416085112.git.josh@joshtriplett.org> (raw)
In-Reply-To: <3ccec8a13019b5e8ce7b1d7889677b778b070dc8.1416085112.git.josh-iaAMLnmF4UmaiuxdJuQwMA@public.gmane.org>
Currently, unprivileged processes (without CAP_SETGID) cannot call
setgroups at all. In particular, processes with a set of supplementary
groups cannot further drop permissions without obtaining elevated
permissions first.
Allow unprivileged processes to call setgroups with a subset of their
current groups; only require CAP_SETGID to add a group the process does
not currently have.
Since some privileged programs (such as sudo) allow tests for
non-membership in a group, require no_new_privs to drop group
privileges.
The kernel already maintains the list of supplementary group IDs in
sorted order, and setgroups already needs to sort the new list, so this
just requires a linear comparison of the two sorted lists.
This moves the CAP_SETGID test from setgroups into set_current_groups.
Tested via the following test program:
void run_id(void)
{
pid_t p = fork();
switch (p) {
case -1:
err(1, "fork");
case 0:
execl("/usr/bin/id", "id", NULL);
err(1, "exec");
default:
if (waitpid(p, NULL, 0) < 0)
err(1, "waitpid");
}
}
int main(void)
{
gid_t list1[] = { 1, 2, 3, 4, 5 };
gid_t list2[] = { 2, 3, 4 };
run_id();
if (setgroups(5, list1) < 0)
err(1, "setgroups 1");
run_id();
if (setresgid(1, 1, 1) < 0)
err(1, "setresgid");
if (setresuid(1, 1, 1) < 0)
err(1, "setresuid");
run_id();
if (setgroups(3, list2) < 0)
err(1, "setgroups 2");
run_id();
if (setgroups(5, list1) < 0)
err(1, "setgroups 3");
run_id();
return 0;
}
Without this patch, the test program gets EPERM from the second
setgroups call, after dropping root privileges. With this patch, the
test program successfully drops groups 1 and 5, but then gets EPERM from
the third setgroups call, since that call attempts to add groups the
process does not currently have.
Signed-off-by: Josh Triplett <josh-iaAMLnmF4UmaiuxdJuQwMA@public.gmane.org>
---
v2: Require no_new_privs.
kernel/groups.c | 34 +++++++++++++++++++++++++++++++---
kernel/uid16.c | 2 --
2 files changed, 31 insertions(+), 5 deletions(-)
diff --git a/kernel/groups.c b/kernel/groups.c
index f0667e7..f7fb8dd 100644
--- a/kernel/groups.c
+++ b/kernel/groups.c
@@ -153,6 +153,29 @@ int groups_search(const struct group_info *group_info, kgid_t grp)
return 0;
}
+/* Compare two sorted group lists; return true if the first is a subset of the
+ * second.
+ */
+static bool is_subset(const struct group_info *g1, const struct group_info *g2)
+{
+ unsigned int i, j;
+
+ for (i = 0, j = 0; i < g1->ngroups; i++) {
+ kgid_t gid1 = GROUP_AT(g1, i);
+ kgid_t gid2;
+ for (; j < g2->ngroups; j++) {
+ gid2 = GROUP_AT(g2, j);
+ if (gid_lte(gid1, gid2))
+ break;
+ }
+ if (j >= g2->ngroups || !gid_eq(gid1, gid2))
+ return false;
+ j++;
+ }
+
+ return true;
+}
+
/**
* set_groups_sorted - Change a group subscription in a set of credentials
* @new: The newly prepared set of credentials to alter
@@ -189,11 +212,18 @@ int set_current_groups(struct group_info *group_info)
{
struct cred *new;
+ groups_sort(group_info);
new = prepare_creds();
if (!new)
return -ENOMEM;
+ if (!(ns_capable(current_user_ns(), CAP_SETGID)
+ || (task_no_new_privs(current)
+ && is_subset(group_info, new->group_info)))) {
+ abort_creds(new);
+ return -EPERM;
+ }
- set_groups(new, group_info);
+ set_groups_sorted(new, group_info);
return commit_creds(new);
}
@@ -233,8 +263,6 @@ SYSCALL_DEFINE2(setgroups, int, gidsetsize, gid_t __user *, grouplist)
struct group_info *group_info;
int retval;
- if (!ns_capable(current_user_ns(), CAP_SETGID))
- return -EPERM;
if ((unsigned)gidsetsize > NGROUPS_MAX)
return -EINVAL;
diff --git a/kernel/uid16.c b/kernel/uid16.c
index 602e5bb..b27e167 100644
--- a/kernel/uid16.c
+++ b/kernel/uid16.c
@@ -176,8 +176,6 @@ SYSCALL_DEFINE2(setgroups16, int, gidsetsize, old_gid_t __user *, grouplist)
struct group_info *group_info;
int retval;
- if (!ns_capable(current_user_ns(), CAP_SETGID))
- return -EPERM;
if ((unsigned)gidsetsize > NGROUPS_MAX)
return -EINVAL;
--
2.1.3
--
To unsubscribe from this list: send the line "unsubscribe linux-man" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
next prev parent reply other threads:[~2014-11-15 21:02 UTC|newest]
Thread overview: 3+ messages / expand[flat|nested] mbox.gz Atom feed top
2014-11-15 21:02 [PATCHv2 1/2] groups: Factor out a function to set a pre-sorted group list Josh Triplett
[not found] ` <3ccec8a13019b5e8ce7b1d7889677b778b070dc8.1416085112.git.josh-iaAMLnmF4UmaiuxdJuQwMA@public.gmane.org>
2014-11-15 21:02 ` Josh Triplett [this message]
2014-11-15 21:03 ` [PATCHv2 manpages] getgroups.2: Document unprivileged setgroups calls Josh Triplett
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=311e7d2982ee36b7c5782dd47e2d589748ef3759.1416085112.git.josh@joshtriplett.org \
--to=josh-iaamlnmf4umaiuxdjuqwma@public.gmane.org \
--cc=akpm-de/tnXTf+JLsfHDXvbKv3WD2FQJk+8+b@public.gmane.org \
--cc=ebiederm-aS9lmoZGLiVWk0Htik3J/w@public.gmane.org \
--cc=keescook-F7+t8E8rja9g9hUCZPvPmw@public.gmane.org \
--cc=linux-api-u79uwXL29TY76Z2rM5mHXA@public.gmane.org \
--cc=linux-kernel-u79uwXL29TY76Z2rM5mHXA@public.gmane.org \
--cc=linux-man-u79uwXL29TY76Z2rM5mHXA@public.gmane.org \
--cc=luto-kltTT9wpgjJwATOyAt5JVQ@public.gmane.org \
--cc=mtk.manpages-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).