[PATCH 1/4] signals/sigaltstack: If SS_AUTODISARM, bypass on_sig_stack

If a signal stack is set up with SS_AUTODISARM, then the kernel
inherently avoids incorrectly resetting the signal stack if signals
recurse: the signal stack will be reset on the first signal
delivery.  This means that we don't need check the stack pointer
when delivering signals if SS_AUTODISARM is set.

This will make segmented x86 programs more robust: currently there's
a hole that could be triggered if ESP/RSP appears to point to the
signal stack but actually doesn't due to a nonzero SS base.

Signed-off-by: Stas Sergeev <redacted>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Cc: Aleksa Sarai <redacted>
Cc: Amanieu d'Antras <redacted>
Cc: Andrea Arcangeli <redacted>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Andy Lutomirski <luto@amacapital.net>
Cc: Borislav Petkov <bp@alien8.de>
Cc: Brian Gerst <redacted>
Cc: Denys Vlasenko <redacted>
Cc: Eric W. Biederman <redacted>
Cc: Frederic Weisbecker <redacted>
Cc: H. Peter Anvin <hpa@zytor.com>
Cc: Heinrich Schuchardt <redacted>
Cc: Jason Low <redacted>
Cc: Josh Triplett <josh@joshtriplett.org>
Cc: Konstantin Khlebnikov <redacted>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Palmer Dabbelt <palmer@dabbelt.com>
Cc: Paul Moore <redacted>
Cc: Pavel Emelyanov <redacted>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Richard Weinberger <richard@nod.at>
Cc: Sasha Levin <redacted>
Cc: Shuah Khan <redacted>
Cc: Tejun Heo <tj@kernel.org>
Cc: Thomas Gleixner <redacted>
Cc: Vladimir Davydov <redacted>
Cc: linux-api@vger.kernel.org
Cc: linux-kernel@vger.kernel.org
Signed-off-by: Andy Lutomirski <luto@kernel.org>
---
 include/linux/sched.h | 12 ++++++++++++
 1 file changed, 12 insertions(+)

diff --git a/include/linux/sched.h b/include/linux/sched.h
index 2950c5cd3005..8f03a93348b9 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h

@@ -2576,6 +2576,18 @@ static inline int kill_cad_pid(int sig, int priv)
  */
 static inline int on_sig_stack(unsigned long sp)
 {
+	/*
+	 * If the signal stack is AUTODISARM then, by construction, we
+	 * can't be on the signal stack unless user code deliberately set
+	 * SS_AUTODISARM when we were already on the it.
+	 *
+	 * This improve reliability: if user state gets corrupted such that
+	 * the stack pointer points very close to the end of the signal stack,
+	 * then this check will enable the signal to be handled anyway.
+	 */
+	if (current->sas_ss_flags & SS_AUTODISARM)
+		return 0;
+
 #ifdef CONFIG_STACK_GROWSUP
 	return sp >= current->sas_ss_sp &&
 		sp - current->sas_ss_sp < current->sas_ss_size;

-- 
2.5.5

* Andy Lutomirski [off-list ref] wrote:

If a signal stack is set up with SS_AUTODISARM, then the kernel
inherently avoids incorrectly resetting the signal stack if signals
recurse: the signal stack will be reset on the first signal
delivery.  This means that we don't need check the stack pointer
when delivering signals if SS_AUTODISARM is set.

This will make segmented x86 programs more robust: currently there's
a hole that could be triggered if ESP/RSP appears to point to the
signal stack but actually doesn't due to a nonzero SS base.

Signed-off-by: Stas Sergeev <redacted>

Presuably that SOB from Stas is stray, as there's no matching From: line?
I've removed it.

Thanks,

	Ingo

On May 3, 2016 11:32 PM, "Ingo Molnar" [off-list ref] wrote:

* Andy Lutomirski [off-list ref] wrote:

quoted

If a signal stack is set up with SS_AUTODISARM, then the kernel
inherently avoids incorrectly resetting the signal stack if signals
recurse: the signal stack will be reset on the first signal
delivery.  This means that we don't need check the stack pointer
when delivering signals if SS_AUTODISARM is set.

This will make segmented x86 programs more robust: currently there's
a hole that could be triggered if ESP/RSP appears to point to the
signal stack but actually doesn't due to a nonzero SS base.

Signed-off-by: Stas Sergeev <redacted>

Presuably that SOB from Stas is stray, as there's no matching From: line?
I've removed it.

Yes.  It was a cut-and-paste-o -- I meant to change it to cc.

Thanks,

        Ingo

03.05.2016 20:31, Andy Lutomirski пишет:

"on the it" -> "on it".

Anyway, I am a bit puzzled with this patch.
You say "unless user code deliberately set
SS_AUTODISARM when we were already on the it"
so what happens in case it actually does?

Without your patch: if user sets up the same sas - no stack switch.
if user sets up different sas - stack switch on nested signal.

With your patch: stack switch in any case, so if user
set up same sas - stack corruption by nested signal.

Or am I missing the intention?

On May 7, 2016 7:38 AM, "Stas Sergeev" [off-list ref] wrote:

03.05.2016 20:31, Andy Lutomirski пишет:

quoted

If a signal stack is set up with SS_AUTODISARM, then the kernel
inherently avoids incorrectly resetting the signal stack if signals
recurse: the signal stack will be reset on the first signal
delivery.  This means that we don't need check the stack pointer
when delivering signals if SS_AUTODISARM is set.

This will make segmented x86 programs more robust: currently there's
a hole that could be triggered if ESP/RSP appears to point to the
signal stack but actually doesn't due to a nonzero SS base.

Signed-off-by: Stas Sergeev <redacted>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Cc: Aleksa Sarai <redacted>
Cc: Amanieu d'Antras <redacted>
Cc: Andrea Arcangeli <redacted>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Andy Lutomirski <luto@amacapital.net>
Cc: Borislav Petkov <bp@alien8.de>
Cc: Brian Gerst <redacted>
Cc: Denys Vlasenko <redacted>
Cc: Eric W. Biederman <redacted>
Cc: Frederic Weisbecker <redacted>
Cc: H. Peter Anvin <hpa@zytor.com>
Cc: Heinrich Schuchardt <redacted>
Cc: Jason Low <redacted>
Cc: Josh Triplett <josh@joshtriplett.org>
Cc: Konstantin Khlebnikov <redacted>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Palmer Dabbelt <palmer@dabbelt.com>
Cc: Paul Moore <redacted>
Cc: Pavel Emelyanov <redacted>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Richard Weinberger <richard@nod.at>
Cc: Sasha Levin <redacted>
Cc: Shuah Khan <redacted>
Cc: Tejun Heo <tj@kernel.org>
Cc: Thomas Gleixner <redacted>
Cc: Vladimir Davydov <redacted>
Cc: linux-api@vger.kernel.org
Cc: linux-kernel@vger.kernel.org
Signed-off-by: Andy Lutomirski <luto@kernel.org>
---
  include/linux/sched.h | 12 ++++++++++++
  1 file changed, 12 insertions(+)

diff --git a/include/linux/sched.h b/include/linux/sched.h
index 2950c5cd3005..8f03a93348b9 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h

@@ -2576,6 +2576,18 @@ static inline int kill_cad_pid(int sig, int priv)
   */
  static inline int on_sig_stack(unsigned long sp)
  {
+       /*
+        * If the signal stack is AUTODISARM then, by construction, we
+        * can't be on the signal stack unless user code deliberately set
+        * SS_AUTODISARM when we were already on the it.

"on the it" -> "on it".

Anyway, I am a bit puzzled with this patch.
You say "unless user code deliberately set

SS_AUTODISARM when we were already on the it"
so what happens in case it actually does?

Stack corruption.  Don't do that.

Without your patch: if user sets up the same sas - no stack switch.
if user sets up different sas - stack switch on nested signal.

With your patch: stack switch in any case, so if user
set up same sas - stack corruption by nested signal.

Or am I missing the intention?

The intention is to make everything completely explicit.  With
SS_AUTODISARM, the kernel knows directly whether you're on the signal
stack, and there should be no need to look at sp.  If you set
SS_AUTODISARM and get a signal, the signal stack gets disarmed.  If
you take a nested signal, it's delivered normally.  When you return
all the way out, the signal stack is re-armed.

For DOSEMU, this means that no 16-bit register state can possibly
cause a signal to be delivered wrong, because the register state when
a signal is raised won't affect delivery, which seems like a good
thing to me.

If this behavior would be problematic for you, can you explain why?

09.05.2016 04:32, Andy Lutomirski пишет:

On May 7, 2016 7:38 AM, "Stas Sergeev" [off-list ref] wrote:

quoted

03.05.2016 20:31, Andy Lutomirski пишет:

quoted

If a signal stack is set up with SS_AUTODISARM, then the kernel
inherently avoids incorrectly resetting the signal stack if signals
recurse: the signal stack will be reset on the first signal
delivery.  This means that we don't need check the stack pointer
when delivering signals if SS_AUTODISARM is set.

This will make segmented x86 programs more robust: currently there's
a hole that could be triggered if ESP/RSP appears to point to the
signal stack but actually doesn't due to a nonzero SS base.

Signed-off-by: Stas Sergeev <redacted>
Cc: Al Viro <viro-RmSDqhL/yNMiFSDQTTA3OLVCufUGDwFn@public.gmane.org>
Cc: Aleksa Sarai <cyphar-gVpy/LI/lHzQT0dZR+AlfA@public.gmane.org>
Cc: Amanieu d'Antras <redacted>
Cc: Andrea Arcangeli <redacted>
Cc: Andrew Morton <akpm-de/tnXTf+JLsfHDXvbKv3WD2FQJk+8+b@public.gmane.org>
Cc: Andy Lutomirski <redacted>
Cc: Borislav Petkov <redacted>
Cc: Brian Gerst <redacted>
Cc: Denys Vlasenko <redacted>
Cc: Eric W. Biederman <ebiederm-aS9lmoZGLiVWk0Htik3J/w@public.gmane.org>
Cc: Frederic Weisbecker <redacted>
Cc: H. Peter Anvin <redacted>
Cc: Heinrich Schuchardt <redacted>
Cc: Jason Low <redacted>
Cc: Josh Triplett <redacted>
Cc: Konstantin Khlebnikov <redacted>
Cc: Linus Torvalds <torvalds-de/tnXTf+JLsfHDXvbKv3WD2FQJk+8+b@public.gmane.org>
Cc: Oleg Nesterov <redacted>
Cc: Palmer Dabbelt <redacted>
Cc: Paul Moore <redacted>
Cc: Pavel Emelyanov <redacted>
Cc: Peter Zijlstra <redacted>
Cc: Richard Weinberger <richard-/L3Ra7n9ekc@public.gmane.org>
Cc: Sasha Levin <redacted>
Cc: Shuah Khan <redacted>
Cc: Tejun Heo <redacted>
Cc: Thomas Gleixner <redacted>
Cc: Vladimir Davydov <redacted>
Cc: linux-api-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
Cc: linux-kernel-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
Signed-off-by: Andy Lutomirski <redacted>
---
   include/linux/sched.h | 12 ++++++++++++
   1 file changed, 12 insertions(+)

diff --git a/include/linux/sched.h b/include/linux/sched.h
index 2950c5cd3005..8f03a93348b9 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h

@@ -2576,6 +2576,18 @@ static inline int kill_cad_pid(int sig, int priv)
    */
   static inline int on_sig_stack(unsigned long sp)
   {
+       /*
+        * If the signal stack is AUTODISARM then, by construction, we
+        * can't be on the signal stack unless user code deliberately set
+        * SS_AUTODISARM when we were already on the it.

"on the it" -> "on it".

Anyway, I am a bit puzzled with this patch.
You say "unless user code deliberately set

SS_AUTODISARM when we were already on the it"
so what happens in case it actually does?

Stack corruption.  Don't do that.

Only after your change, I have to admit. :)

quoted

Without your patch: if user sets up the same sas - no stack switch.
if user sets up different sas - stack switch on nested signal.

With your patch: stack switch in any case, so if user
set up same sas - stack corruption by nested signal.

Or am I missing the intention?

The intention is to make everything completely explicit.  With
SS_AUTODISARM, the kernel knows directly whether you're on the signal
stack, and there should be no need to look at sp.  If you set
SS_AUTODISARM and get a signal, the signal stack gets disarmed.  If
you take a nested signal, it's delivered normally.  When you return
all the way out, the signal stack is re-armed.

For DOSEMU, this means that no 16-bit register state can possibly
cause a signal to be delivered wrong, because the register state when
a signal is raised won't affect delivery, which seems like a good
thing to me.

Yes, but doesn't affect dosemu1 which doesn't use SS_AUTODISARM.
So IMHO the SS check should still be added, even if not for dosemu2.

If this behavior would be problematic for you, can you explain why?

Only theoretically: if someone sets SS_AUTODISARM inside a
sighandler. Since this doesn't give EPERM, I wouldn't deliberately
make it a broken scenario (esp if it wasn't before the particular change).
Ideally it would give EPERM, but we can't, so doesn't matter much.
I just wanted to warn about the possible regression.

On May 8, 2016 7:05 PM, "Stas Sergeev" [off-list ref] wrote:

09.05.2016 04:32, Andy Lutomirski пишет:

quoted

On May 7, 2016 7:38 AM, "Stas Sergeev" [off-list ref] wrote:

quoted

03.05.2016 20:31, Andy Lutomirski пишет:

quoted

If a signal stack is set up with SS_AUTODISARM, then the kernel
inherently avoids incorrectly resetting the signal stack if signals
recurse: the signal stack will be reset on the first signal
delivery.  This means that we don't need check the stack pointer
when delivering signals if SS_AUTODISARM is set.

This will make segmented x86 programs more robust: currently there's
a hole that could be triggered if ESP/RSP appears to point to the
signal stack but actually doesn't due to a nonzero SS base.

Signed-off-by: Stas Sergeev <redacted>
Cc: Al Viro <viro-RmSDqhL/yNMiFSDQTTA3OLVCufUGDwFn@public.gmane.org>
Cc: Aleksa Sarai <cyphar-gVpy/LI/lHzQT0dZR+AlfA@public.gmane.org>
Cc: Amanieu d'Antras <redacted>
Cc: Andrea Arcangeli <redacted>
Cc: Andrew Morton <akpm-de/tnXTf+JLsfHDXvbKv3WD2FQJk+8+b@public.gmane.org>
Cc: Andy Lutomirski <redacted>
Cc: Borislav Petkov <redacted>
Cc: Brian Gerst <redacted>
Cc: Denys Vlasenko <redacted>
Cc: Eric W. Biederman <ebiederm-aS9lmoZGLiVWk0Htik3J/w@public.gmane.org>
Cc: Frederic Weisbecker <redacted>
Cc: H. Peter Anvin <redacted>
Cc: Heinrich Schuchardt <redacted>
Cc: Jason Low <redacted>
Cc: Josh Triplett <redacted>
Cc: Konstantin Khlebnikov <redacted>
Cc: Linus Torvalds <torvalds-de/tnXTf+JLsfHDXvbKv3WD2FQJk+8+b@public.gmane.org>
Cc: Oleg Nesterov <redacted>
Cc: Palmer Dabbelt <redacted>
Cc: Paul Moore <redacted>
Cc: Pavel Emelyanov <redacted>
Cc: Peter Zijlstra <redacted>
Cc: Richard Weinberger <richard-/L3Ra7n9ekc@public.gmane.org>
Cc: Sasha Levin <redacted>
Cc: Shuah Khan <redacted>
Cc: Tejun Heo <redacted>
Cc: Thomas Gleixner <redacted>
Cc: Vladimir Davydov <redacted>
Cc: linux-api-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
Cc: linux-kernel-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
Signed-off-by: Andy Lutomirski <redacted>
---
   include/linux/sched.h | 12 ++++++++++++
   1 file changed, 12 insertions(+)

diff --git a/include/linux/sched.h b/include/linux/sched.h
index 2950c5cd3005..8f03a93348b9 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h

@@ -2576,6 +2576,18 @@ static inline int kill_cad_pid(int sig, int priv)
    */
   static inline int on_sig_stack(unsigned long sp)
   {
+       /*
+        * If the signal stack is AUTODISARM then, by construction, we
+        * can't be on the signal stack unless user code deliberately set
+        * SS_AUTODISARM when we were already on the it.

"on the it" -> "on it".

Anyway, I am a bit puzzled with this patch.
You say "unless user code deliberately set

SS_AUTODISARM when we were already on the it"
so what happens in case it actually does?

Stack corruption.  Don't do that.

Only after your change, I have to admit. :)

quoted

quoted

Without your patch: if user sets up the same sas - no stack switch.
if user sets up different sas - stack switch on nested signal.

With your patch: stack switch in any case, so if user
set up same sas - stack corruption by nested signal.

Or am I missing the intention?

The intention is to make everything completely explicit.  With
SS_AUTODISARM, the kernel knows directly whether you're on the signal
stack, and there should be no need to look at sp.  If you set
SS_AUTODISARM and get a signal, the signal stack gets disarmed.  If
you take a nested signal, it's delivered normally.  When you return
all the way out, the signal stack is re-armed.

For DOSEMU, this means that no 16-bit register state can possibly
cause a signal to be delivered wrong, because the register state when
a signal is raised won't affect delivery, which seems like a good
thing to me.

Yes, but doesn't affect dosemu1 which doesn't use SS_AUTODISARM.
So IMHO the SS check should still be added, even if not for dosemu2.

quoted

If this behavior would be problematic for you, can you explain why?

Only theoretically: if someone sets SS_AUTODISARM inside a
sighandler. Since this doesn't give EPERM, I wouldn't deliberately
make it a broken scenario (esp if it wasn't before the particular change).
Ideally it would give EPERM, but we can't, so doesn't matter much.
I just wanted to warn about the possible regression.

I suppose we could return an error if you are on the sigstack when
setting SS_AUTODISARM, although I was hoping to avoid yet more special
cases.

--
To unsubscribe from this list: send the line "unsubscribe linux-api" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

14.05.2016 07:18, Andy Lutomirski пишет:

On May 8, 2016 7:05 PM, "Stas Sergeev" [off-list ref] wrote:

quoted

09.05.2016 04:32, Andy Lutomirski пишет:

quoted

On May 7, 2016 7:38 AM, "Stas Sergeev" [off-list ref] wrote:

quoted

03.05.2016 20:31, Andy Lutomirski пишет:

quoted

If a signal stack is set up with SS_AUTODISARM, then the kernel
inherently avoids incorrectly resetting the signal stack if signals
recurse: the signal stack will be reset on the first signal
delivery.  This means that we don't need check the stack pointer
when delivering signals if SS_AUTODISARM is set.

This will make segmented x86 programs more robust: currently there's
a hole that could be triggered if ESP/RSP appears to point to the
signal stack but actually doesn't due to a nonzero SS base.

Signed-off-by: Stas Sergeev <redacted>
Cc: Al Viro <viro-RmSDqhL/yNMiFSDQTTA3OLVCufUGDwFn@public.gmane.org>
Cc: Aleksa Sarai <cyphar-gVpy/LI/lHzQT0dZR+AlfA@public.gmane.org>
Cc: Amanieu d'Antras <redacted>
Cc: Andrea Arcangeli <redacted>
Cc: Andrew Morton <akpm-de/tnXTf+JLsfHDXvbKv3WD2FQJk+8+b@public.gmane.org>
Cc: Andy Lutomirski <redacted>
Cc: Borislav Petkov <redacted>
Cc: Brian Gerst <redacted>
Cc: Denys Vlasenko <redacted>
Cc: Eric W. Biederman <ebiederm-aS9lmoZGLiVWk0Htik3J/w@public.gmane.org>
Cc: Frederic Weisbecker <redacted>
Cc: H. Peter Anvin <redacted>
Cc: Heinrich Schuchardt <redacted>
Cc: Jason Low <redacted>
Cc: Josh Triplett <redacted>
Cc: Konstantin Khlebnikov <redacted>
Cc: Linus Torvalds <torvalds-de/tnXTf+JLsfHDXvbKv3WD2FQJk+8+b@public.gmane.org>
Cc: Oleg Nesterov <redacted>
Cc: Palmer Dabbelt <redacted>
Cc: Paul Moore <redacted>
Cc: Pavel Emelyanov <redacted>
Cc: Peter Zijlstra <redacted>
Cc: Richard Weinberger <richard-/L3Ra7n9ekc@public.gmane.org>
Cc: Sasha Levin <redacted>
Cc: Shuah Khan <redacted>
Cc: Tejun Heo <redacted>
Cc: Thomas Gleixner <redacted>
Cc: Vladimir Davydov <redacted>
Cc: linux-api-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
Cc: linux-kernel-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
Signed-off-by: Andy Lutomirski <redacted>
---
    include/linux/sched.h | 12 ++++++++++++
    1 file changed, 12 insertions(+)

diff --git a/include/linux/sched.h b/include/linux/sched.h
index 2950c5cd3005..8f03a93348b9 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h

@@ -2576,6 +2576,18 @@ static inline int kill_cad_pid(int sig, int priv)
     */
    static inline int on_sig_stack(unsigned long sp)
    {
+       /*
+        * If the signal stack is AUTODISARM then, by construction, we
+        * can't be on the signal stack unless user code deliberately set
+        * SS_AUTODISARM when we were already on the it.

"on the it" -> "on it".

Anyway, I am a bit puzzled with this patch.
You say "unless user code deliberately set

SS_AUTODISARM when we were already on the it"
so what happens in case it actually does?

Stack corruption.  Don't do that.

Only after your change, I have to admit. :)

quoted

quoted

Without your patch: if user sets up the same sas - no stack switch.
if user sets up different sas - stack switch on nested signal.

With your patch: stack switch in any case, so if user
set up same sas - stack corruption by nested signal.

Or am I missing the intention?

The intention is to make everything completely explicit.  With
SS_AUTODISARM, the kernel knows directly whether you're on the signal
stack, and there should be no need to look at sp.  If you set
SS_AUTODISARM and get a signal, the signal stack gets disarmed.  If
you take a nested signal, it's delivered normally.  When you return
all the way out, the signal stack is re-armed.

For DOSEMU, this means that no 16-bit register state can possibly
cause a signal to be delivered wrong, because the register state when
a signal is raised won't affect delivery, which seems like a good
thing to me.

Yes, but doesn't affect dosemu1 which doesn't use SS_AUTODISARM.
So IMHO the SS check should still be added, even if not for dosemu2.

quoted

If this behavior would be problematic for you, can you explain why?

Only theoretically: if someone sets SS_AUTODISARM inside a
sighandler. Since this doesn't give EPERM, I wouldn't deliberately
make it a broken scenario (esp if it wasn't before the particular change).
Ideally it would give EPERM, but we can't, so doesn't matter much.
I just wanted to warn about the possible regression.

I suppose we could return an error if you are on the sigstack when
setting SS_AUTODISARM, although I was hoping to avoid yet more special
cases.

Hmm.
How about extending the generic check then?
Currently it is roughly:
if (on_sig_stack(sp)) return -EPERM;

and we could do:
if (on_sig_stack(sp) || on_new_sas(new_sas, sp)) return -EPERM;

Looks like it will close the potential hole opened by your commit
without introducing the special case for SS_AUTODISARM.
What do you think?

On May 14, 2016 4:18 AM, "Stas Sergeev" [off-list ref] wrote:

14.05.2016 07:18, Andy Lutomirski пишет:

quoted

On May 8, 2016 7:05 PM, "Stas Sergeev" [off-list ref] wrote:

quoted

09.05.2016 04:32, Andy Lutomirski пишет:

quoted

On May 7, 2016 7:38 AM, "Stas Sergeev" [off-list ref] wrote:

quoted

03.05.2016 20:31, Andy Lutomirski пишет:

quoted

If a signal stack is set up with SS_AUTODISARM, then the kernel
inherently avoids incorrectly resetting the signal stack if signals
recurse: the signal stack will be reset on the first signal
delivery.  This means that we don't need check the stack pointer
when delivering signals if SS_AUTODISARM is set.

This will make segmented x86 programs more robust: currently there's
a hole that could be triggered if ESP/RSP appears to point to the
signal stack but actually doesn't due to a nonzero SS base.

Signed-off-by: Stas Sergeev <redacted>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Cc: Aleksa Sarai <redacted>
Cc: Amanieu d'Antras <redacted>
Cc: Andrea Arcangeli <redacted>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Andy Lutomirski <luto@amacapital.net>
Cc: Borislav Petkov <bp@alien8.de>
Cc: Brian Gerst <redacted>
Cc: Denys Vlasenko <redacted>
Cc: Eric W. Biederman <redacted>
Cc: Frederic Weisbecker <redacted>
Cc: H. Peter Anvin <hpa@zytor.com>
Cc: Heinrich Schuchardt <redacted>
Cc: Jason Low <redacted>
Cc: Josh Triplett <josh@joshtriplett.org>
Cc: Konstantin Khlebnikov <redacted>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Palmer Dabbelt <palmer@dabbelt.com>
Cc: Paul Moore <redacted>
Cc: Pavel Emelyanov <redacted>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Richard Weinberger <richard@nod.at>
Cc: Sasha Levin <redacted>
Cc: Shuah Khan <redacted>
Cc: Tejun Heo <tj@kernel.org>
Cc: Thomas Gleixner <redacted>
Cc: Vladimir Davydov <redacted>
Cc: linux-api@vger.kernel.org
Cc: linux-kernel@vger.kernel.org
Signed-off-by: Andy Lutomirski <luto@kernel.org>
---
    include/linux/sched.h | 12 ++++++++++++
    1 file changed, 12 insertions(+)

diff --git a/include/linux/sched.h b/include/linux/sched.h
index 2950c5cd3005..8f03a93348b9 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h

@@ -2576,6 +2576,18 @@ static inline int kill_cad_pid(int sig, int priv)
     */
    static inline int on_sig_stack(unsigned long sp)
    {
+       /*
+        * If the signal stack is AUTODISARM then, by construction, we
+        * can't be on the signal stack unless user code deliberately set
+        * SS_AUTODISARM when we were already on the it.

"on the it" -> "on it".

Anyway, I am a bit puzzled with this patch.
You say "unless user code deliberately set

SS_AUTODISARM when we were already on the it"
so what happens in case it actually does?

Stack corruption.  Don't do that.

Only after your change, I have to admit. :)

quoted

quoted

Without your patch: if user sets up the same sas - no stack switch.
if user sets up different sas - stack switch on nested signal.

With your patch: stack switch in any case, so if user
set up same sas - stack corruption by nested signal.

Or am I missing the intention?

The intention is to make everything completely explicit.  With
SS_AUTODISARM, the kernel knows directly whether you're on the signal
stack, and there should be no need to look at sp.  If you set
SS_AUTODISARM and get a signal, the signal stack gets disarmed.  If
you take a nested signal, it's delivered normally.  When you return
all the way out, the signal stack is re-armed.

For DOSEMU, this means that no 16-bit register state can possibly
cause a signal to be delivered wrong, because the register state when
a signal is raised won't affect delivery, which seems like a good
thing to me.

Yes, but doesn't affect dosemu1 which doesn't use SS_AUTODISARM.
So IMHO the SS check should still be added, even if not for dosemu2.

quoted

If this behavior would be problematic for you, can you explain why?

Only theoretically: if someone sets SS_AUTODISARM inside a
sighandler. Since this doesn't give EPERM, I wouldn't deliberately
make it a broken scenario (esp if it wasn't before the particular change).
Ideally it would give EPERM, but we can't, so doesn't matter much.
I just wanted to warn about the possible regression.

I suppose we could return an error if you are on the sigstack when
setting SS_AUTODISARM, although I was hoping to avoid yet more special
cases.

Hmm.
How about extending the generic check then?
Currently it is roughly:
if (on_sig_stack(sp)) return -EPERM;

and we could do:
if (on_sig_stack(sp) || on_new_sas(new_sas, sp)) return -EPERM;

Looks like it will close the potential hole opened by your commit
without introducing the special case for SS_AUTODISARM.
What do you think?

It's still a wee bit ugly.  Also, doesn't that change existing
behavior for the existing non-AUTODISARM case?  Also, we'd have to
make sure that sigreturn doesn't trigger this check.

My inclination would be leave it alone.