Re: [RFC] TIF_NOTIFY_RESUME, arch/*/*/*signal*.c and all such

[Oleg Nesterov <oleg@redhat.com>]

First of all, let me repeat I am useless when it comes to the low-level
or asm code. I can be easily wrong, and I am going to ask the questions.

On 04/28, Al Viro wrote:
>
> It's actually worse than I thought - we can't just lift that check
> to do_notify_resume() and be done with that.  Suppose do_signal() does
> get called on e.g. i386 or arm with !user_mode(regs).  What'll happen next?
>
> We have TIF_SIGPENDING set in thread flags - otherwise we wouldn't get
> there at all.  OK, do_signal() doesn't do anything and returns.  So does
> do_notify_resume().  And we are back into the loop in asm glue, rereading
> the thread flags (still unchanged), checking if anything is to be done
> (yes, it is - TIF_SIGPENDING is still set), calling do_notify_resume(),
> ad infinitum.
>
> Lifting the check into do_notify_resume() will not help at all, obviously.
>
> AFAICS we can get hit by that.

Please look at 29a2e2836ff9ea65a603c89df217f4198973a74f
x86-32: Fix endless loop when processing signals for kernel tasks

> At least i386, arm and mips have
> ret_from_fork going straight to "return from syscall" path, no checks for
> return to user mode done.  And process created by kernel_thread() will
> go there.

Looks like, the patch above fixes that.

But, we call do_notify_resume() first, it would be nice to avoid this
and remove the user_mode() check in do_signal() or lift into
do_notify_resume().

> It's a narrow race, but AFAICS it's not impossible to hit -
> guess the PID of kernel thread to be launched, send it a signal and hit
> the moment before it gets to executing the payload.

Yes. But note that the kernel threads run with all signals ignored.

This is still possible, but a kernel thread should do allow_signal()
and then call kernel_thread() (not kthread_create).



Question. So far I know that on x86 do_notify_resume() && !user_mode()
is only possible on 32bit system, and the possible callers are
ret_from_fork or kernel_execve (if it fails).

Plus, perhaps CONFIG_VM86 makes a difference?

Could you please clarify?

Oleg.