[prev in list] [next in list] [prev in thread] [next in thread]
List: glibc-alpha
Subject: Re: V3 [PATCH] nptl: Zero-extend arguments to SETXID syscalls [BZ #26248]
From: "H.J. Lu via Libc-alpha" <libc-alpha () sourceware ! org>
Date: 2020-07-27 19:17:14
Message-ID: CAMe9rOoGxGd4Z=PNwQ_u5GypBy_tNLmTJa+x7T4e1gYSJK0dcQ () mail ! gmail ! com
[Download RAW message or body]
On Mon, Jul 27, 2020 at 8:49 AM Florian Weimer <fweimer@redhat.com> wrote:
>
> * H. J. Lu via Libc-alpha:
>
> > + /* 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];
>
> Maybe add the missing space before '('?
Fixed.
> Rest looks good to me for glibc 2.32.
>
This is the patch I am checking in.
Thanks.
--
H.J.
["0001-nptl-Zero-extend-arguments-to-SETXID-syscalls-BZ-262.patch" (text/x-patch)]
From e99cc6da913b96266dcd20402fbed4d50e060e3c Mon Sep 17 00:00:00 2001
From: "H.J. Lu" <hjl.tools@gmail.com>
Date: Thu, 16 Jul 2020 03:37:10 -0700
Subject: [PATCH] nptl: Zero-extend arguments to SETXID syscalls [BZ #26248]
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.
---
nptl/Makefile | 2 +-
nptl/descr.h | 8 ++++-
nptl/tst-setgroups.c | 79 ++++++++++++++++++++++++++++++++++++++++++++
3 files changed, 87 insertions(+), 2 deletions(-)
create mode 100644 nptl/tst-setgroups.c
diff --git a/nptl/Makefile b/nptl/Makefile
index 5e62c77853..89569c4f46 100644
--- a/nptl/Makefile
+++ b/nptl/Makefile
@@ -310,7 +310,7 @@ tests-internal := tst-robustpi8 tst-rwlock19 tst-rwlock20 \
tst-setgetname \
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 6a509b6725..d8343ff9a1 100644
--- a/nptl/descr.h
+++ b/nptl/descr.h
@@ -95,7 +95,13 @@ 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 0000000000..ae3c1b1ba0
--- /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>
--
2.26.2
[prev in list] [next in list] [prev in thread] [next in thread]
Configure |
About |
News |
Add a list |
Sponsored by KoreLogic