diff mbox series

arm64: ptrace: Fix seccomp of traced syscall -1 (NO_SYSCALL)

Message ID 90edd33b-6353-1228-791f-0336d94d5f8c@majoroak.me.uk (mailing list archive)
State New, archived
Headers show
Series arm64: ptrace: Fix seccomp of traced syscall -1 (NO_SYSCALL) | expand

Commit Message

Timothy Baldwin Jan. 18, 2021, 2:58 a.m. UTC 
From c047f549699d31ed91d5ac0cadbcf76a02cd801e Mon Sep 17 00:00:00 2001
From: Timothy E Baldwin<T.E.Baldwin99@members.leeds.ac.uk>
Date: Sat, 16 Jan 2021 15:18:54 +0000
Subject: [PATCH] arm64: ptrace: Fix seccomp of traced syscall -1 (NO_SYSCALL)

Since commit f086f67485c5 ("arm64: ptrace: add support for syscall
emulation"), if system call number -1 is called and the process is being
traced with PTRACE_SYSCALL, for example by strace, the seccomp check is
skipped and -ENOSYS is returned unconditionally (unless altered by the
tracer) rather than carrying out action specified in the seccomp filter.

The consequence of this is that it is not possible to reliably strace
a seccomp based implementation of a foreign system call interface in
which r7/x8 is permitted to be -1 on entry to a system call.

Also trace_sys_enter and audit_syscall_entry are skipped if a system
call is skipped.

Fix by removing the in_syscall(regs) check restoring the previous behaviour
which is like AArch32, x86 (which uses generic code) and everything else.

Fixes: f086f67485c5 ("arm64: ptrace: add support for syscall emulation")
Signed-off-by: Timothy E Baldwin<T.E.Baldwin99@members.leeds.ac.uk>
Cc: Sudeep Holla<sudeep.holla@arm.com>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Catalin Marinas<catalin.marinas@arm.com>
Cc: Will Deacon<will.deacon@arm.com>
Cc: Kees Cook<keescook@chromium.org>
Cc:stable@vger.kernel.org
---
  arch/arm64/kernel/ptrace.c | 2 +-
  1 file changed, 1 insertion(+), 1 deletion(-)

Comments

Sudeep Holla Feb. 24, 2021, 2:49 p.m. UTC | #1 
On Mon, Jan 18, 2021 at 02:58:58AM +0000, Timothy Baldwin wrote:
> From c047f549699d31ed91d5ac0cadbcf76a02cd801e Mon Sep 17 00:00:00 2001
> From: Timothy E Baldwin<T.E.Baldwin99@members.leeds.ac.uk>
> Date: Sat, 16 Jan 2021 15:18:54 +0000
> Subject: [PATCH] arm64: ptrace: Fix seccomp of traced syscall -1 (NO_SYSCALL)
> 
> Since commit f086f67485c5 ("arm64: ptrace: add support for syscall
> emulation"), if system call number -1 is called and the process is being
> traced with PTRACE_SYSCALL, for example by strace, the seccomp check is
> skipped and -ENOSYS is returned unconditionally (unless altered by the
> tracer) rather than carrying out action specified in the seccomp filter.
> 
> The consequence of this is that it is not possible to reliably strace
> a seccomp based implementation of a foreign system call interface in
> which r7/x8 is permitted to be -1 on entry to a system call.
> 
> Also trace_sys_enter and audit_syscall_entry are skipped if a system
> call is skipped.
> 
> Fix by removing the in_syscall(regs) check restoring the previous behaviour
> which is like AArch32, x86 (which uses generic code) and everything else.
> 

Ah, my fault. At the time of timing this I didn't test with seccomp and
also for some reason IIRC I had assumed the flags SYSCALL_{EMU,TRACE}
and seccomp calls are mutually exclusive and can't happen together.

FWIW,
Reviewed-by: Sudeep Holla <sudeep.holla@arm.com>

Also I ran some minimal tests I have, so
Tested-by: Sudeep Holla <sudeep.holla@arm.com>

I have also asked Haibo Xu <Haibo.Xu@arm.com> who help me testing back then
to test again.


> Fixes: f086f67485c5 ("arm64: ptrace: add support for syscall emulation")
> Signed-off-by: Timothy E Baldwin<T.E.Baldwin99@members.leeds.ac.uk>
> Cc: Sudeep Holla<sudeep.holla@arm.com>
> Cc: Oleg Nesterov <oleg@redhat.com>
> Cc: Catalin Marinas<catalin.marinas@arm.com>
> Cc: Will Deacon<will.deacon@arm.com>
> Cc: Kees Cook<keescook@chromium.org>
> Cc:stable@vger.kernel.org
> ---
>  arch/arm64/kernel/ptrace.c | 2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)
> 
> diff --git a/arch/arm64/kernel/ptrace.c b/arch/arm64/kernel/ptrace.c
> index 8ac487c84e37..1d75471979cb 100644
> --- a/arch/arm64/kernel/ptrace.c
> +++ b/arch/arm64/kernel/ptrace.c
> @@ -1796,7 +1796,7 @@ int syscall_trace_enter(struct pt_regs *regs)
>  	if (flags & (_TIF_SYSCALL_EMU | _TIF_SYSCALL_TRACE)) {
>  		tracehook_report_syscall(regs, PTRACE_SYSCALL_ENTER);
> -		if (!in_syscall(regs) || (flags & _TIF_SYSCALL_EMU))
> +		if (flags & _TIF_SYSCALL_EMU)
>  			return NO_SYSCALL;
>  	}
> -- 
> 2.27.0
> 
> The specific implementation of a seccomp based foreign system call interface
> is my port of RISC OS to Linux, in the spirit User Mode Linux:
> https://github.com/TimothyEBaldwin/RISC_OS_Linux_Binary
> 
> 
> 
> _______________________________________________
> linux-arm-kernel mailing list
> linux-arm-kernel@lists.infradead.org
> http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
Kees Cook Feb. 24, 2021, 10:48 p.m. UTC | #2 
On Wed, Feb 24, 2021 at 02:49:20PM +0000, Sudeep Holla wrote:
> On Mon, Jan 18, 2021 at 02:58:58AM +0000, Timothy Baldwin wrote:
> > From c047f549699d31ed91d5ac0cadbcf76a02cd801e Mon Sep 17 00:00:00 2001
> > From: Timothy E Baldwin<T.E.Baldwin99@members.leeds.ac.uk>
> > Date: Sat, 16 Jan 2021 15:18:54 +0000
> > Subject: [PATCH] arm64: ptrace: Fix seccomp of traced syscall -1 (NO_SYSCALL)
> > 
> > Since commit f086f67485c5 ("arm64: ptrace: add support for syscall
> > emulation"), if system call number -1 is called and the process is being
> > traced with PTRACE_SYSCALL, for example by strace, the seccomp check is
> > skipped and -ENOSYS is returned unconditionally (unless altered by the
> > tracer) rather than carrying out action specified in the seccomp filter.
> > 
> > The consequence of this is that it is not possible to reliably strace
> > a seccomp based implementation of a foreign system call interface in
> > which r7/x8 is permitted to be -1 on entry to a system call.
> > 
> > Also trace_sys_enter and audit_syscall_entry are skipped if a system
> > call is skipped.
> > 
> > Fix by removing the in_syscall(regs) check restoring the previous behaviour
> > which is like AArch32, x86 (which uses generic code) and everything else.
> > 
> 
> Ah, my fault. At the time of timing this I didn't test with seccomp and
> also for some reason IIRC I had assumed the flags SYSCALL_{EMU,TRACE}
> and seccomp calls are mutually exclusive and can't happen together.
> 
> FWIW,
> Reviewed-by: Sudeep Holla <sudeep.holla@arm.com>
> 
> Also I ran some minimal tests I have, so
> Tested-by: Sudeep Holla <sudeep.holla@arm.com>
> 
> I have also asked Haibo Xu <Haibo.Xu@arm.com> who help me testing back then
> to test again.

Thanks for catching and fixing this! Does this pass the seccomp selftests?

Reviewed-by: Kees Cook <keescook@chromium.org>

Will, do you want to take this? I don't usually put the arch-specific
seccomp bits through the seccomp tree.

-Kees

> 
> 
> > Fixes: f086f67485c5 ("arm64: ptrace: add support for syscall emulation")
> > Signed-off-by: Timothy E Baldwin<T.E.Baldwin99@members.leeds.ac.uk>
> > Cc: Sudeep Holla<sudeep.holla@arm.com>
> > Cc: Oleg Nesterov <oleg@redhat.com>
> > Cc: Catalin Marinas<catalin.marinas@arm.com>
> > Cc: Will Deacon<will.deacon@arm.com>
> > Cc: Kees Cook<keescook@chromium.org>
> > Cc:stable@vger.kernel.org
> > ---
> >  arch/arm64/kernel/ptrace.c | 2 +-
> >  1 file changed, 1 insertion(+), 1 deletion(-)
> > 
> > diff --git a/arch/arm64/kernel/ptrace.c b/arch/arm64/kernel/ptrace.c
> > index 8ac487c84e37..1d75471979cb 100644
> > --- a/arch/arm64/kernel/ptrace.c
> > +++ b/arch/arm64/kernel/ptrace.c
> > @@ -1796,7 +1796,7 @@ int syscall_trace_enter(struct pt_regs *regs)
> >  	if (flags & (_TIF_SYSCALL_EMU | _TIF_SYSCALL_TRACE)) {
> >  		tracehook_report_syscall(regs, PTRACE_SYSCALL_ENTER);
> > -		if (!in_syscall(regs) || (flags & _TIF_SYSCALL_EMU))
> > +		if (flags & _TIF_SYSCALL_EMU)
> >  			return NO_SYSCALL;
> >  	}
> > -- 
> > 2.27.0
> > 
> > The specific implementation of a seccomp based foreign system call interface
> > is my port of RISC OS to Linux, in the spirit User Mode Linux:
> > https://github.com/TimothyEBaldwin/RISC_OS_Linux_Binary
> > 
> > 
> > 
> > _______________________________________________
> > linux-arm-kernel mailing list
> > linux-arm-kernel@lists.infradead.org
> > http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
> 
> -- 
> Regards,
> Sudeep
Will Deacon Feb. 25, 2021, 10:12 a.m. UTC | #3 
On Wed, Feb 24, 2021 at 02:48:05PM -0800, Kees Cook wrote:
> On Wed, Feb 24, 2021 at 02:49:20PM +0000, Sudeep Holla wrote:
> > On Mon, Jan 18, 2021 at 02:58:58AM +0000, Timothy Baldwin wrote:
> > > From c047f549699d31ed91d5ac0cadbcf76a02cd801e Mon Sep 17 00:00:00 2001
> > > From: Timothy E Baldwin<T.E.Baldwin99@members.leeds.ac.uk>
> > > Date: Sat, 16 Jan 2021 15:18:54 +0000
> > > Subject: [PATCH] arm64: ptrace: Fix seccomp of traced syscall -1 (NO_SYSCALL)
> > > 
> > > Since commit f086f67485c5 ("arm64: ptrace: add support for syscall
> > > emulation"), if system call number -1 is called and the process is being
> > > traced with PTRACE_SYSCALL, for example by strace, the seccomp check is
> > > skipped and -ENOSYS is returned unconditionally (unless altered by the
> > > tracer) rather than carrying out action specified in the seccomp filter.
> > > 
> > > The consequence of this is that it is not possible to reliably strace
> > > a seccomp based implementation of a foreign system call interface in
> > > which r7/x8 is permitted to be -1 on entry to a system call.
> > > 
> > > Also trace_sys_enter and audit_syscall_entry are skipped if a system
> > > call is skipped.
> > > 
> > > Fix by removing the in_syscall(regs) check restoring the previous behaviour
> > > which is like AArch32, x86 (which uses generic code) and everything else.
> > > 
> > 
> > Ah, my fault. At the time of timing this I didn't test with seccomp and
> > also for some reason IIRC I had assumed the flags SYSCALL_{EMU,TRACE}
> > and seccomp calls are mutually exclusive and can't happen together.
> > 
> > FWIW,
> > Reviewed-by: Sudeep Holla <sudeep.holla@arm.com>
> > 
> > Also I ran some minimal tests I have, so
> > Tested-by: Sudeep Holla <sudeep.holla@arm.com>
> > 
> > I have also asked Haibo Xu <Haibo.Xu@arm.com> who help me testing back then
> > to test again.
> 
> Thanks for catching and fixing this! Does this pass the seccomp selftests?
> 
> Reviewed-by: Kees Cook <keescook@chromium.org>
> 
> Will, do you want to take this? I don't usually put the arch-specific
> seccomp bits through the seccomp tree.

Sure, I'll add it to the fixes pile. Thanks!

Will
Will Deacon Feb. 25, 2021, 11:02 a.m. UTC | #4 
On Mon, 18 Jan 2021 02:58:58 +0000, Timothy Baldwin wrote:
>  From c047f549699d31ed91d5ac0cadbcf76a02cd801e Mon Sep 17 00:00:00 2001
> From: Timothy E Baldwin<T.E.Baldwin99@members.leeds.ac.uk>
> Date: Sat, 16 Jan 2021 15:18:54 +0000
> Subject: [PATCH] arm64: ptrace: Fix seccomp of traced syscall -1 (NO_SYSCALL)
> 
> Since commit f086f67485c5 ("arm64: ptrace: add support for syscall
> emulation"), if system call number -1 is called and the process is being
> traced with PTRACE_SYSCALL, for example by strace, the seccomp check is
> skipped and -ENOSYS is returned unconditionally (unless altered by the
> tracer) rather than carrying out action specified in the seccomp filter.
> 
> [...]

(Note that I had to fix up whitespace errors in order for git am to apply
 this, so please check the resulting commit).

Applied to arm64 (for-next/fixes), thanks!

[1/1] arm64: ptrace: Fix seccomp of traced syscall -1 (NO_SYSCALL)
      https://git.kernel.org/arm64/c/df84fe947089

Cheers,
diff mbox series

Patch

diff --git a/arch/arm64/kernel/ptrace.c b/arch/arm64/kernel/ptrace.c
index 8ac487c84e37..1d75471979cb 100644
--- a/arch/arm64/kernel/ptrace.c
+++ b/arch/arm64/kernel/ptrace.c
@@ -1796,7 +1796,7 @@  int syscall_trace_enter(struct pt_regs *regs)
  
  	if (flags & (_TIF_SYSCALL_EMU | _TIF_SYSCALL_TRACE)) {
  		tracehook_report_syscall(regs, PTRACE_SYSCALL_ENTER);
-		if (!in_syscall(regs) || (flags & _TIF_SYSCALL_EMU))
+		if (flags & _TIF_SYSCALL_EMU)
  			return NO_SYSCALL;
  	}