06.03.2016 23:10, Andy Lutomirski пишет:

On Sun, Mar 6, 2016 at 12:07 PM, Andy Lutomirski [off-list ref] wrote:

quoted

On Mon, Feb 29, 2016 at 1:29 PM, Stas Sergeev [off-list ref] wrote:

quoted

This patch implements the SS_AUTODISARM flag that can be ORed with
SS_ONSTACK when forming ss_flags.
When this flag is set, sigaltstack will be disabled when entering
the signal handler; more precisely, after saving sas to uc_stack.
When leaving the signal handler, the sigaltstack is restored by
uc_stack.
When this flag is used, it is safe to switch from sighandler with
swapcontext(). Without this flag, the subsequent signal will corrupt
the state of the switched-away sighandler.

CC: Ingo Molnar <redacted>
CC: Peter Zijlstra <redacted>
CC: Richard Weinberger <richard-/L3Ra7n9ekc@public.gmane.org>
CC: Andrew Morton <akpm-de/tnXTf+JLsfHDXvbKv3WD2FQJk+8+b@public.gmane.org>
CC: Oleg Nesterov <redacted>
CC: Tejun Heo <redacted>
CC: Heinrich Schuchardt <redacted>
CC: Jason Low <redacted>
CC: Andrea Arcangeli <redacted>
CC: Frederic Weisbecker <redacted>
CC: Konstantin Khlebnikov <redacted>
CC: Josh Triplett <redacted>
CC: "Eric W. Biederman" <ebiederm-aS9lmoZGLiVWk0Htik3J/w@public.gmane.org>
CC: Aleksa Sarai <cyphar-gVpy/LI/lHzQT0dZR+AlfA@public.gmane.org>
CC: "Amanieu d'Antras" <redacted>
CC: Paul Moore <redacted>
CC: Sasha Levin <redacted>
CC: Palmer Dabbelt <redacted>
CC: Vladimir Davydov <redacted>
CC: linux-kernel-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
CC: linux-api-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
CC: Andy Lutomirski <redacted>

Signed-off-by: Stas Sergeev <redacted>
---
  include/linux/sched.h       |  8 ++++++++
  include/linux/signal.h      |  4 +++-
  include/uapi/linux/signal.h |  3 +++
  kernel/fork.c               |  2 +-
  kernel/signal.c             | 23 ++++++++++++-----------
  5 files changed, 27 insertions(+), 13 deletions(-)

diff --git a/include/linux/sched.h b/include/linux/sched.h
index a10494a..26201cd 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h

@@ -1587,6 +1587,7 @@ struct task_struct {

         unsigned long sas_ss_sp;
         size_t sas_ss_size;
+       unsigned sas_ss_flags;

         struct callback_head *task_works;

@@ -2573,6 +2574,13 @@ static inline int sas_ss_flags(unsigned long sp)
         return on_sig_stack(sp) ? SS_ONSTACK : 0;
  }

+static inline void sas_ss_reset(struct task_struct *p)
+{
+       p->sas_ss_sp = 0;
+       p->sas_ss_size = 0;
+       p->sas_ss_flags = SS_DISABLE;
+}
+
  static inline unsigned long sigsp(unsigned long sp, struct ksignal *ksig)
  {
         if (unlikely((ksig->ka.sa.sa_flags & SA_ONSTACK)) && ! sas_ss_flags(sp))

diff --git a/include/linux/signal.h b/include/linux/signal.h
index 92557bb..3fbe814 100644
--- a/include/linux/signal.h
+++ b/include/linux/signal.h

@@ -432,8 +432,10 @@ int __save_altstack(stack_t __user *, unsigned long);
         stack_t __user *__uss = uss; \
         struct task_struct *t = current; \
         put_user_ex((void __user *)t->sas_ss_sp, &__uss->ss_sp); \
-       put_user_ex(sas_ss_flags(sp), &__uss->ss_flags); \
+       put_user_ex(t->sas_ss_flags, &__uss->ss_flags); \
         put_user_ex(t->sas_ss_size, &__uss->ss_size); \
+       if (t->sas_ss_flags & SS_AUTODISARM) \
+               sas_ss_reset(t); \
  } while (0);

  #ifdef CONFIG_PROC_FS

diff --git a/include/uapi/linux/signal.h b/include/uapi/linux/signal.h
index e1bd50c2..4691bc5 100644
--- a/include/uapi/linux/signal.h
+++ b/include/uapi/linux/signal.h

@@ -6,5 +6,8 @@

  #define SS_ONSTACK     1
  #define SS_DISABLE     2
+#define SS_VALMASK     0xf

SS_MODE_MASK, perhaps?

Actually, let's invert that.

#define SS_AUTODISARM (1U << 31)
#define SS_FLAG_BITS SS_AUTODISARM

ss_mode = ss_flags & ~SS_FLAG_BITS;

this way flag bits that are currently undefined will continue to trigger EINVAL.

OK, will do.
Though I am still going to keep SS_AUTODISARM =(1 << 4) -
it doesn't matter from what side we add new flags, does it?