Messages in this thread | | | Date | Fri, 20 Sep 2013 11:26:13 +0200 (CEST) | From | Thomas Gleixner <> | Subject | Re: [patch 1/6] hardirq: Make hardirq bits generic |
| |
On Thu, 19 Sep 2013, Geert Uytterhoeven wrote:
> On Thu, Sep 19, 2013 at 7:02 PM, Andreas Schwab <schwab@linux-m68k.org> wrote: > > Thomas Gleixner <tglx@linutronix.de> writes: > >> + /* > >> + * Only the last interrupt leaving the kernel goes through the > >> + * various exception return checks. > >> + */ > >> + cmpl #0, %d0 > > tstl %d0 > > arch/m68k/kernel/built-in.o: In function `bad_inthandler': > (.text+0x2a6): undefined reference to `ret_from_last_interrupt' > > I came up with the quick (whitespace-damaged-gmail) fix below. > Or should we handle the nesting in handle_badint(), too?
Hmm, probably yes. If a badint gets interrupted by a good one, you would fail to go through ret_from_exception.
> --- a/arch/m68k/kernel/entry.S > +++ b/arch/m68k/kernel/entry.S > @@ -313,17 +313,11 @@ user_irqvec_fixup = . + 2 > > ENTRY(bad_inthandler) > SAVE_ALL_INT > - GET_CURRENT(%d0) > - movel %d0,%a1 > - addqb #1,%a1@(TINFO_PREEMPT+1) > > movel %sp,%sp@- > jsr handle_badint > addql #4,%sp > > - movel %curptr@(TASK_STACK),%a1 > - subqb #1,%a1@(TINFO_PREEMPT+1) > - jeq ret_from_last_interrupt > RESTORE_ALL > > However, the resulting kernel hangs (on ARAnyM) after starting userspace: > > | INIT: version 2.86 booting
Hmm.
> I'll have a deeper look when I have some more time...
Thanks a lot!
tglx
| |