public inbox for libc-stable@sourceware.org
 help / color / mirror / Atom feed
From: Aurelien Jarno <aurelien@aurel32.net>
To: libc-stable@sourceware.org
Subject: [2.31 COMMITTED] nptl: Zero-extend arguments to SETXID syscalls [BZ #26248]
Date: Thu, 30 Jul 2020 01:44:39 +0200	[thread overview]
Message-ID: <20200729234439.1760925-1-aurelien@aurel32.net> (raw)

From: "H.J. Lu" <hjl.tools@gmail.com>

nptl has

/* Opcodes and data types for communication with the signal handler to
   change user/group IDs.  */
struct xid_command
{
  int syscall_no;
  long int id[3];
  volatile int cntr;
  volatile int error;
};

 /* This must be last, otherwise the current thread might not have
     permissions to send SIGSETXID syscall to the other threads.  */
  result = INTERNAL_SYSCALL_NCS (cmdp->syscall_no, 3,
                                 cmdp->id[0], cmdp->id[1], cmdp->id[2]);

But the second argument of setgroups syscal is a pointer:

       int setgroups (size_t size, const gid_t *list);

But on x32, pointers passed to syscall must have pointer type so that
they will be zero-extended.  The kernel XID arguments are unsigned and
do not require sign extension.  Change xid_command to

struct xid_command
{
  int syscall_no;
  unsigned long int id[3];
  volatile int cntr;
  volatile int error;
};

so that all arguments are zero-extended.  A testcase is added for x32 and
setgroups returned with EFAULT when running as root without the fix.

(cherry picked from commit 0ad926f34937f7b4843a8b49e5d93199601fe324)
---
 NEWS                 |  1 +
 nptl/Makefile        |  2 +-
 nptl/descr.h         |  8 ++++-
 nptl/tst-setgroups.c | 79 ++++++++++++++++++++++++++++++++++++++++++++
 4 files changed, 88 insertions(+), 2 deletions(-)
 create mode 100644 nptl/tst-setgroups.c

diff --git a/NEWS b/NEWS
index daf4b727cd7..8a857061e11 100644
--- a/NEWS
+++ b/NEWS
@@ -24,6 +24,7 @@ The following bugs are resolved with this release:
   [25933] Off by one error in __strncmp_avx2
   [25966] Incorrect access of __x86_shared_non_temporal_threshold for x32
   [25976] nss_compat: internal_end*ent may clobber errno, hiding ERANGE
+  [26248] Incorrect argument types for INLINE_SETXID_SYSCALL
 
 Security related changes:
 
diff --git a/nptl/Makefile b/nptl/Makefile
index 584e0ffd96e..2df4c9098c7 100644
--- a/nptl/Makefile
+++ b/nptl/Makefile
@@ -332,7 +332,7 @@ tests-internal := tst-rwlock19 tst-rwlock20 \
 		  tst-mutexpi8 tst-mutexpi8-static tst-cancel25
 
 xtests = tst-setuid1 tst-setuid1-static tst-setuid2 \
-	tst-mutexpp1 tst-mutexpp6 tst-mutexpp10
+	tst-mutexpp1 tst-mutexpp6 tst-mutexpp10 tst-setgroups
 
 # This test can run into task limits because of a linux kernel bug
 # and then cause the make process to fail too, see bug 24537.
diff --git a/nptl/descr.h b/nptl/descr.h
index 9dcf480bdfc..5f1f35e9a4a 100644
--- a/nptl/descr.h
+++ b/nptl/descr.h
@@ -94,7 +94,13 @@ struct pthread_unwind_buf
 struct xid_command
 {
   int syscall_no;
-  long int id[3];
+  /* Enforce zero-extension for the pointer argument in
+
+     int setgroups (size_t size, const gid_t *list);
+
+     The kernel XID arguments are unsigned and do not require sign
+     extension.  */
+  unsigned long int id[3];
   volatile int cntr;
   volatile int error; /* -1: no call yet, 0: success seen, >0: error seen.  */
 };
diff --git a/nptl/tst-setgroups.c b/nptl/tst-setgroups.c
new file mode 100644
index 00000000000..ae3c1b1ba0d
--- /dev/null
+++ b/nptl/tst-setgroups.c
@@ -0,0 +1,79 @@
+/* Test setgroups as root and in the presence of threads (Bug 26248)
+   Copyright (C) 2020 Free Software Foundation, Inc.
+   This file is part of the GNU C Library.
+
+   The GNU C Library is free software; you can redistribute it and/or
+   modify it under the terms of the GNU Lesser General Public
+   License as published by the Free Software Foundation; either
+   version 2.1 of the License, or (at your option) any later version.
+
+   The GNU C Library is distributed in the hope that it will be useful,
+   but WITHOUT ANY WARRANTY; without even the implied warranty of
+   MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+   Lesser General Public License for more details.
+
+   You should have received a copy of the GNU Lesser General Public
+   License along with the GNU C Library; if not, see
+   <https://www.gnu.org/licenses/>.  */
+
+#include <stdlib.h>
+#include <limits.h>
+#include <grp.h>
+#include <errno.h>
+#include <error.h>
+#include <support/xthread.h>
+#include <support/support.h>
+#include <support/test-driver.h>
+#include <support/xunistd.h>
+
+/* The purpose of this test is to test the setgroups API as root and in
+   the presence of threads.  Once we create a thread the setgroups
+   implementation must ensure that all threads are set to the same
+   group and this operation should not fail. Lastly we test setgroups
+   with a zero sized group and a bad address and verify we get EPERM.  */
+
+static void *
+start_routine (void *args)
+{
+  return NULL;
+}
+
+static int
+do_test (void)
+{
+  int size;
+  /* NB: Stack address can be at 0xfffXXXXX on 32-bit OSes.  */
+  gid_t list[NGROUPS_MAX];
+  int status = EXIT_SUCCESS;
+
+  pthread_t thread = xpthread_create (NULL, start_routine, NULL);
+
+  size = getgroups (sizeof (list) / sizeof (list[0]), list);
+  if (size < 0)
+    {
+      status = EXIT_FAILURE;
+      error (0, errno, "getgroups failed");
+    }
+  if (setgroups (size, list) < 0)
+    {
+      if (errno == EPERM)
+	status = EXIT_UNSUPPORTED;
+      else
+	{
+	  status = EXIT_FAILURE;
+	  error (0, errno, "setgroups failed");
+	}
+    }
+
+  if (status == EXIT_SUCCESS && setgroups (0, list) < 0)
+    {
+      status = EXIT_FAILURE;
+      error (0, errno, "setgroups failed");
+    }
+
+  xpthread_join (thread);
+
+  return status;
+}
+
+#include <support/test-driver.c>
-- 
2.27.0


                 reply	other threads:[~2020-07-29 23:45 UTC|newest]

Thread overview: [no followups] expand[flat|nested]  mbox.gz  Atom feed

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=20200729234439.1760925-1-aurelien@aurel32.net \
    --to=aurelien@aurel32.net \
    --cc=libc-stable@sourceware.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 read-only IMAP folder(s) and NNTP newsgroup(s).