2011-04-04 14:58:10

by Matt Fleming

Subject: [PATCH] avr32: init cannot ignore signals sent by force_sig_info()

From: Matt Fleming <[email protected]>

We can delete the code that checks to see if we're sending an ignored
signal to init because force_sig_info() already handles this case.
force_sig_info() will kill init even if the signal handler is SIG_DFL
and the scenario described in the comment where init might "generate
the same exception over and over again" cannot occur (force_sig_info()
clears SIGNAL_UNKILLABLE to ensure that init will die).

Also, the use of is_global_init() is not correct in the multhreaded
case, as Oleg Nesterov explains,

"is_global_init() is not right in theory, /sbin/init can be
multithreaded. And, this doesn't cover the sub-namespace
inits... I'd suggest to check SIGNAL_UNKILLABLE, but looking
closer I think you can simply remove this code."

It seems this code was copied from arch/powerpc in March 2007 in commit

623b0355d5b1 "[AVR32] Clean up exception handling code"

but the code was deleted from arch/powerpc in November 2009 in commit

a0592d42fe3e "powerpc: kill the obsolete code under is_global_init()"

So catch up with powerpc and delete the bogus code.

Signed-off-by: Matt Fleming <[email protected]>
---
arch/avr32/kernel/traps.c | 22 ----------------------
1 files changed, 0 insertions(+), 22 deletions(-)

diff --git a/arch/avr32/kernel/traps.c b/arch/avr32/kernel/traps.c
index b91b204..7aa2575 100644
--- a/arch/avr32/kernel/traps.c
+++ b/arch/avr32/kernel/traps.c
@@ -95,28 +95,6 @@ void _exception(long signr, struct pt_regs *regs, int code,
info.si_code = code;
info.si_addr = (void __user *)addr;
force_sig_info(signr, &info, current);
-
- /*
- * Init gets no signals that it doesn't have a handler for.
- * That's all very well, but if it has caused a synchronous
- * exception and we ignore the resulting signal, it will just
- * generate the same exception over and over again and we get
- * nowhere. Better to kill it and let the kernel panic.
- */
- if (is_global_init(current)) {
- __sighandler_t handler;
-
- spin_lock_irq(&current->sighand->siglock);
- handler = current->sighand->action[signr-1].sa.sa_handler;
- spin_unlock_irq(&current->sighand->siglock);
- if (handler == SIG_DFL) {
- /* init has generated a synchronous exception
- and it doesn't have a handler for the signal */
- printk(KERN_CRIT "init has generated signal %ld "
- "but has no handler for it\n", signr);
- do_exit(signr);
- }
- }
}

asmlinkage void do_nmi(unsigned long ecr, struct pt_regs *regs)
--
1.7.4

2011-04-04 16:05:14

by Oleg Nesterov

Subject: Re: [PATCH] avr32: init cannot ignore signals sent by force_sig_info()

On 04/04, Matt Fleming wrote:
>
> --- a/arch/avr32/kernel/traps.c
> +++ b/arch/avr32/kernel/traps.c
> @@ -95,28 +95,6 @@ void _exception(long signr, struct pt_regs *regs, int code,
> info.si_code = code;
> info.si_addr = (void __user *)addr;
> force_sig_info(signr, &info, current);
> -
> - /*
> - * Init gets no signals that it doesn't have a handler for.
> - * That's all very well, but if it has caused a synchronous
> - * exception and we ignore the resulting signal, it will just
> - * generate the same exception over and over again and we get
> - * nowhere. Better to kill it and let the kernel panic.
> - */
> - if (is_global_init(current)) {
> - __sighandler_t handler;
> -
> - spin_lock_irq(&current->sighand->siglock);
> - handler = current->sighand->action[signr-1].sa.sa_handler;
> - spin_unlock_irq(&current->sighand->siglock);
> - if (handler == SIG_DFL) {
> - /* init has generated a synchronous exception
> - and it doesn't have a handler for the signal */
> - printk(KERN_CRIT "init has generated signal %ld "
> - "but has no handler for it\n", signr);
> - do_exit(signr);
> - }
> - }

Reviewed-by: Oleg Nesterov <[email protected]>

2011-04-13 13:15:26

by Hans-Christian Egtvedt

Subject: Re: [PATCH] avr32: init cannot ignore signals sent by force_sig_info()

On Mon, 2011-04-04 at 15:58 +0100, Matt Fleming wrote:
> From: Matt Fleming <[email protected]>
>
> We can delete the code that checks to see if we're sending an ignored
> signal to init because force_sig_info() already handles this case.
> force_sig_info() will kill init even if the signal handler is SIG_DFL
> and the scenario described in the comment where init might "generate
> the same exception over and over again" cannot occur (force_sig_info()
> clears SIGNAL_UNKILLABLE to ensure that init will die).
>
> Also, the use of is_global_init() is not correct in the multhreaded
> case, as Oleg Nesterov explains,
>
> "is_global_init() is not right in theory, /sbin/init can be
> multithreaded. And, this doesn't cover the sub-namespace
> inits... I'd suggest to check SIGNAL_UNKILLABLE, but looking
> closer I think you can simply remove this code."
>
> It seems this code was copied from arch/powerpc in March 2007 in commit
>
> 623b0355d5b1 "[AVR32] Clean up exception handling code"
>
> but the code was deleted from arch/powerpc in November 2009 in commit
>
> a0592d42fe3e "powerpc: kill the obsolete code under is_global_init()"
>
> So catch up with powerpc and delete the bogus code.
>
> Signed-off-by: Matt Fleming <[email protected]>

Thanks for this cleanup, added to the AVR32 tree.

Signed-off-by: Hans-Christian Egtvedt <[email protected]>

--
Hans-Christian Egtvedt