From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1753389AbaHURIu (ORCPT ); Thu, 21 Aug 2014 13:08:50 -0400 Received: from mail-ob0-f182.google.com ([209.85.214.182]:48954 "EHLO mail-ob0-f182.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753051AbaHURIs (ORCPT ); Thu, 21 Aug 2014 13:08:48 -0400 MIME-Version: 1.0 In-Reply-To: <1408611405-8943-3-git-send-email-takahiro.akashi@linaro.org> References: <1408611405-8943-1-git-send-email-takahiro.akashi@linaro.org> <1408611405-8943-3-git-send-email-takahiro.akashi@linaro.org> Date: Thu, 21 Aug 2014 12:08:47 -0500 X-Google-Sender-Auth: lba_Rmrgd1wMzq_NU99OQHtacIE Message-ID: Subject: Re: [PATCH v6 2/6] arm64: ptrace: allow tracer to skip a system call From: Kees Cook To: AKASHI Takahiro Cc: Catalin Marinas , Will Deacon , Deepak Saxena , arndb@arndb.de, "linux-arm-kernel@lists.infradead.org" , linaro-kernel@lists.linaro.org, LKML Content-Type: text/plain; charset=UTF-8 Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Thu, Aug 21, 2014 at 3:56 AM, AKASHI Takahiro wrote: > If tracer specifies -1 as a syscall number, this traced system call should > be skipped with a value in x0 used as a return value. > This patch enables this semantics, but there is a restriction here: > > when syscall(-1) is issued by user, tracer cannot skip this system call > and modify a return value at syscall entry. > > In order to ease this flavor, we need to treat whatever value in x0 as > a return value, but this might result in a bogus value being returned, > especially when tracer doesn't do anything at this syscall. > So we always return ENOSYS instead, while we have another chance to change > a return value at syscall exit. > > Please also note: > * syscall entry tracing and syscall exit tracing (ftrace tracepoint and > audit) are always executed, if enabled, even when skipping a system call > (that is, -1). > In this way, we can avoid a potential bug where audit_syscall_entry() > might be called without audit_syscall_exit() at the previous system call > being called, that would cause OOPs in audit_syscall_entry(). > > * syscallno may also be set to -1 if a fatal signal (SIGKILL) is detected > in tracehook_report_syscall_entry(), but since a value set to x0 (ENOSYS) > is not used in this case, we may neglect the case. > > Signed-off-by: AKASHI Takahiro > --- > arch/arm64/include/asm/ptrace.h | 8 ++++++++ > arch/arm64/kernel/entry.S | 4 ++++ > arch/arm64/kernel/ptrace.c | 20 ++++++++++++++++++++ > 3 files changed, 32 insertions(+) > > diff --git a/arch/arm64/include/asm/ptrace.h b/arch/arm64/include/asm/ptrace.h > index 501000f..a58cf62 100644 > --- a/arch/arm64/include/asm/ptrace.h > +++ b/arch/arm64/include/asm/ptrace.h > @@ -65,6 +65,14 @@ > #define COMPAT_PT_TEXT_ADDR 0x10000 > #define COMPAT_PT_DATA_ADDR 0x10004 > #define COMPAT_PT_TEXT_END_ADDR 0x10008 > + > +/* > + * used to skip a system call when tracer changes its number to -1 > + * with ptrace(PTRACE_SET_SYSCALL) > + */ > +#define RET_SKIP_SYSCALL -1 > +#define IS_SKIP_SYSCALL(no) ((int)(no & 0xffffffff) == -1) > + > #ifndef __ASSEMBLY__ > > /* sizeof(struct user) for AArch32 */ > diff --git a/arch/arm64/kernel/entry.S b/arch/arm64/kernel/entry.S > index f0b5e51..fdd6eae 100644 > --- a/arch/arm64/kernel/entry.S > +++ b/arch/arm64/kernel/entry.S > @@ -25,6 +25,7 @@ > #include > #include > #include > +#include > #include > #include > > @@ -671,6 +672,8 @@ ENDPROC(el0_svc) > __sys_trace: > mov x0, sp > bl syscall_trace_enter > + cmp w0, #RET_SKIP_SYSCALL // skip syscall? > + b.eq __sys_trace_return_skipped > adr lr, __sys_trace_return // return address > uxtw scno, w0 // syscall number (possibly new) > mov x1, sp // pointer to regs > @@ -685,6 +688,7 @@ __sys_trace: > > __sys_trace_return: > str x0, [sp] // save returned x0 > +__sys_trace_return_skipped: // x0 already in regs[0] > mov x0, sp > bl syscall_trace_exit > b ret_to_user > diff --git a/arch/arm64/kernel/ptrace.c b/arch/arm64/kernel/ptrace.c > index 8876049..c54dbcc 100644 > --- a/arch/arm64/kernel/ptrace.c > +++ b/arch/arm64/kernel/ptrace.c > @@ -1121,9 +1121,29 @@ static void tracehook_report_syscall(struct pt_regs *regs, > > asmlinkage int syscall_trace_enter(struct pt_regs *regs) > { > + unsigned int saved_syscallno = regs->syscallno; > + > if (test_thread_flag(TIF_SYSCALL_TRACE)) > tracehook_report_syscall(regs, PTRACE_SYSCALL_ENTER); > > + if (IS_SKIP_SYSCALL(regs->syscallno)) { > + /* > + * RESTRICTION: we can't modify a return value of user > + * issued syscall(-1) here. In order to ease this flavor, > + * we need to treat whatever value in x0 as a return value, > + * but this might result in a bogus value being returned. > + */ > + /* > + * NOTE: syscallno may also be set to -1 if fatal signal is > + * detected in tracehook_report_syscall_entry(), but since > + * a value set to x0 here is not used in this case, we may > + * neglect the case. > + */ > + if (!test_thread_flag(TIF_SYSCALL_TRACE) || > + (IS_SKIP_SYSCALL(saved_syscallno))) > + regs->regs[0] = -ENOSYS; > + } > + I don't have a runtime environment yet for arm64, so I can't test this directly myself, so I'm just trying to eyeball this. :) Once the seccomp logic is added here, I don't think using -2 as a special value will work. Doesn't this mean the Oops is possible by the user issuing a "-2" syscall? As in, if TIF_SYSCALL_WORK is set, and the user passed -2 as the syscall, audit will be called only on entry, and then skipped on exit? -Kees > if (test_thread_flag(TIF_SYSCALL_TRACEPOINT)) > trace_sys_enter(regs, regs->syscallno); > > -- > 1.7.9.5 > -- Kees Cook Chrome OS Security From mboxrd@z Thu Jan 1 00:00:00 1970 From: keescook@chromium.org (Kees Cook) Date: Thu, 21 Aug 2014 12:08:47 -0500 Subject: [PATCH v6 2/6] arm64: ptrace: allow tracer to skip a system call In-Reply-To: <1408611405-8943-3-git-send-email-takahiro.akashi@linaro.org> References: <1408611405-8943-1-git-send-email-takahiro.akashi@linaro.org> <1408611405-8943-3-git-send-email-takahiro.akashi@linaro.org> Message-ID: To: linux-arm-kernel@lists.infradead.org List-Id: linux-arm-kernel.lists.infradead.org On Thu, Aug 21, 2014 at 3:56 AM, AKASHI Takahiro wrote: > If tracer specifies -1 as a syscall number, this traced system call should > be skipped with a value in x0 used as a return value. > This patch enables this semantics, but there is a restriction here: > > when syscall(-1) is issued by user, tracer cannot skip this system call > and modify a return value at syscall entry. > > In order to ease this flavor, we need to treat whatever value in x0 as > a return value, but this might result in a bogus value being returned, > especially when tracer doesn't do anything at this syscall. > So we always return ENOSYS instead, while we have another chance to change > a return value at syscall exit. > > Please also note: > * syscall entry tracing and syscall exit tracing (ftrace tracepoint and > audit) are always executed, if enabled, even when skipping a system call > (that is, -1). > In this way, we can avoid a potential bug where audit_syscall_entry() > might be called without audit_syscall_exit() at the previous system call > being called, that would cause OOPs in audit_syscall_entry(). > > * syscallno may also be set to -1 if a fatal signal (SIGKILL) is detected > in tracehook_report_syscall_entry(), but since a value set to x0 (ENOSYS) > is not used in this case, we may neglect the case. > > Signed-off-by: AKASHI Takahiro > --- > arch/arm64/include/asm/ptrace.h | 8 ++++++++ > arch/arm64/kernel/entry.S | 4 ++++ > arch/arm64/kernel/ptrace.c | 20 ++++++++++++++++++++ > 3 files changed, 32 insertions(+) > > diff --git a/arch/arm64/include/asm/ptrace.h b/arch/arm64/include/asm/ptrace.h > index 501000f..a58cf62 100644 > --- a/arch/arm64/include/asm/ptrace.h > +++ b/arch/arm64/include/asm/ptrace.h > @@ -65,6 +65,14 @@ > #define COMPAT_PT_TEXT_ADDR 0x10000 > #define COMPAT_PT_DATA_ADDR 0x10004 > #define COMPAT_PT_TEXT_END_ADDR 0x10008 > + > +/* > + * used to skip a system call when tracer changes its number to -1 > + * with ptrace(PTRACE_SET_SYSCALL) > + */ > +#define RET_SKIP_SYSCALL -1 > +#define IS_SKIP_SYSCALL(no) ((int)(no & 0xffffffff) == -1) > + > #ifndef __ASSEMBLY__ > > /* sizeof(struct user) for AArch32 */ > diff --git a/arch/arm64/kernel/entry.S b/arch/arm64/kernel/entry.S > index f0b5e51..fdd6eae 100644 > --- a/arch/arm64/kernel/entry.S > +++ b/arch/arm64/kernel/entry.S > @@ -25,6 +25,7 @@ > #include > #include > #include > +#include > #include > #include > > @@ -671,6 +672,8 @@ ENDPROC(el0_svc) > __sys_trace: > mov x0, sp > bl syscall_trace_enter > + cmp w0, #RET_SKIP_SYSCALL // skip syscall? > + b.eq __sys_trace_return_skipped > adr lr, __sys_trace_return // return address > uxtw scno, w0 // syscall number (possibly new) > mov x1, sp // pointer to regs > @@ -685,6 +688,7 @@ __sys_trace: > > __sys_trace_return: > str x0, [sp] // save returned x0 > +__sys_trace_return_skipped: // x0 already in regs[0] > mov x0, sp > bl syscall_trace_exit > b ret_to_user > diff --git a/arch/arm64/kernel/ptrace.c b/arch/arm64/kernel/ptrace.c > index 8876049..c54dbcc 100644 > --- a/arch/arm64/kernel/ptrace.c > +++ b/arch/arm64/kernel/ptrace.c > @@ -1121,9 +1121,29 @@ static void tracehook_report_syscall(struct pt_regs *regs, > > asmlinkage int syscall_trace_enter(struct pt_regs *regs) > { > + unsigned int saved_syscallno = regs->syscallno; > + > if (test_thread_flag(TIF_SYSCALL_TRACE)) > tracehook_report_syscall(regs, PTRACE_SYSCALL_ENTER); > > + if (IS_SKIP_SYSCALL(regs->syscallno)) { > + /* > + * RESTRICTION: we can't modify a return value of user > + * issued syscall(-1) here. In order to ease this flavor, > + * we need to treat whatever value in x0 as a return value, > + * but this might result in a bogus value being returned. > + */ > + /* > + * NOTE: syscallno may also be set to -1 if fatal signal is > + * detected in tracehook_report_syscall_entry(), but since > + * a value set to x0 here is not used in this case, we may > + * neglect the case. > + */ > + if (!test_thread_flag(TIF_SYSCALL_TRACE) || > + (IS_SKIP_SYSCALL(saved_syscallno))) > + regs->regs[0] = -ENOSYS; > + } > + I don't have a runtime environment yet for arm64, so I can't test this directly myself, so I'm just trying to eyeball this. :) Once the seccomp logic is added here, I don't think using -2 as a special value will work. Doesn't this mean the Oops is possible by the user issuing a "-2" syscall? As in, if TIF_SYSCALL_WORK is set, and the user passed -2 as the syscall, audit will be called only on entry, and then skipped on exit? -Kees > if (test_thread_flag(TIF_SYSCALL_TRACEPOINT)) > trace_sys_enter(regs, regs->syscallno); > > -- > 1.7.9.5 > -- Kees Cook Chrome OS Security