From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S932415Ab2GMFTr (ORCPT ); Fri, 13 Jul 2012 01:19:47 -0400 Received: from mail-gg0-f174.google.com ([209.85.161.174]:53256 "EHLO mail-gg0-f174.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752654Ab2GMFTp (ORCPT ); Fri, 13 Jul 2012 01:19:45 -0400 From: Will Drewry To: linux-kernel@vger.kernel.org, torvalds@linux-foundation.org Cc: qmewlo@gmail.com, eparis@redhat.com, keescook@chromium.org, james.l.morris@oracle.com, hpa@zytor.com, cevans@chromium.org, luto@mit.edu, Will Drewry Subject: [PATCH] x86/vsyscall: allow seccomp filter in vsyscall=emulate Date: Fri, 13 Jul 2012 00:17:52 -0500 Message-Id: <1342156672-7369-1-git-send-email-wad@chromium.org> X-Mailer: git-send-email 1.7.9.5 Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org If a seccomp filter program is installed, older static binaries and distributions with older libc implementations (glibc 2.13 and earlier) that rely on vsyscall use will be terminated regardless of the filter program policy when executing time, gettimeofday, or getcpu. This is only the case when vsyscall emulation is in use (vsyscall=emulate is the default). This patch emulates system call entry inside a vsyscall=emulate trap such that seccomp can properly evaluate the system call. Reported-by: Owen Kibel Signed-off-by: Will Drewry --- arch/x86/kernel/vsyscall_64.c | 29 ++++++++++++++++++++++++++--- 1 file changed, 26 insertions(+), 3 deletions(-) diff --git a/arch/x86/kernel/vsyscall_64.c b/arch/x86/kernel/vsyscall_64.c index 7515cf0..433545f 100644 --- a/arch/x86/kernel/vsyscall_64.c +++ b/arch/x86/kernel/vsyscall_64.c @@ -139,6 +139,14 @@ static int addr_to_vsyscall_nr(unsigned long addr) return nr; } +static int vsyscall_seccomp(struct task_struct *tsk, int syscall_nr) +{ + if (!seccomp_mode(&tsk->seccomp)) + return 0; + task_pt_regs(tsk)->orig_ax = syscall_nr; + return __secure_computing(syscall_nr); +} + static bool write_ok_or_segv(unsigned long ptr, size_t size) { /* @@ -174,6 +182,7 @@ bool emulate_vsyscall(struct pt_regs *regs, unsigned long address) int vsyscall_nr; int prev_sig_on_uaccess_error; long ret; + int skip; /* * No point in checking CS -- the only way to get here is a user mode @@ -205,9 +214,6 @@ bool emulate_vsyscall(struct pt_regs *regs, unsigned long address) } tsk = current; - if (seccomp_mode(&tsk->seccomp)) - do_exit(SIGKILL); - /* * With a real vsyscall, page faults cause SIGSEGV. We want to * preserve that behavior to make writing exploits harder. @@ -222,8 +228,13 @@ bool emulate_vsyscall(struct pt_regs *regs, unsigned long address) * address 0". */ ret = -EFAULT; + skip = 0; switch (vsyscall_nr) { case 0: + skip = vsyscall_seccomp(tsk, __NR_gettimeofday); + if (skip) + break; + if (!write_ok_or_segv(regs->di, sizeof(struct timeval)) || !write_ok_or_segv(regs->si, sizeof(struct timezone))) break; @@ -234,6 +245,10 @@ bool emulate_vsyscall(struct pt_regs *regs, unsigned long address) break; case 1: + skip = vsyscall_seccomp(tsk, __NR_time); + if (skip) + break; + if (!write_ok_or_segv(regs->di, sizeof(time_t))) break; @@ -241,6 +256,10 @@ bool emulate_vsyscall(struct pt_regs *regs, unsigned long address) break; case 2: + skip = vsyscall_seccomp(tsk, __NR_getcpu); + if (skip) + break; + if (!write_ok_or_segv(regs->di, sizeof(unsigned)) || !write_ok_or_segv(regs->si, sizeof(unsigned))) break; @@ -253,6 +272,9 @@ bool emulate_vsyscall(struct pt_regs *regs, unsigned long address) current_thread_info()->sig_on_uaccess_error = prev_sig_on_uaccess_error; + if (skip) + goto do_ret; + if (ret == -EFAULT) { /* Bad news -- userspace fed a bad pointer to a vsyscall. */ warn_bad_vsyscall(KERN_INFO, regs, @@ -271,6 +293,7 @@ bool emulate_vsyscall(struct pt_regs *regs, unsigned long address) regs->ax = ret; +do_ret: /* Emulate a ret instruction. */ regs->ip = caller; regs->sp += 8; -- 1.7.9.5