Re: Again: UML on s390 (31Bit)

[Ulrich Weigand]

Bodo Stroesser wrote:

>Unfortunately, I guess this will not help. But maybe I'm missing
>something, as I don't even understand, what the effect of the
>attached patch should be.
Have you tried it?

>AFAICS, after each call to do_signal(),
>entry.S will return to user without regs->trap being checked again.
>do_signal() is the only place, where regs->trap is checked, and
>it will be called on return to user exactly once.
It will be called multiple times if *multiple* signals are pending,
and this is exactly the situation in your problem case (some other
signal is pending after the ptrace intercept SIGTRAP was delievered).

>So a practical solution should allow to reset regs->trap while the
>child is on the first or second syscall interception.
This is exactly what this patch is supposed to do: whenever during
a ptrace intercept the PSW is changed (as it presumably is by your
sigreturn implementation), regs->trap is automatically reset.

Bye,
Ulrich

-- 
  Dr. Ulrich Weigand
  Linux on zSeries Development
  Ulrich.Weigand@xxxxxxxxxx
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/