On Tue, Apr 03, 2012 at 05:37:30PM -0400, David Miller wrote:
> From: Sam Ravnborg <sam@xxxxxxxxxxxx>
> Date: Tue, 3 Apr 2012 22:44:56 +0200
>
> > I have just tried out v3.4-rc1 on my sparc32 box - and keyboard is not working.
> > I type my username - but nothing happens.
> >
> > I see nothing unusual during booting the kernel.
> > With the same configuration it works with v3.3.
> >
> > As I have no prime suspects I have started to bisect it -
> > but wanted to let you know that there is an issue.
> >
> > bisect will take some time as my build-machine is not exactly
> > quick - so I expect no result until sometime tomorrow.
> > But the good thing is that it is easy to test :-)
>
> Since sparc is one of the few platforms with a serial based keyboard
> I think serial/tty/input layers would be the places to investigate.
>
> Unfortunately I can't help with this since all the machines I have
> access to have USB keyboards and there'd be a loud mob if those
> broke :)
sunzilog was the culprint - so right you were...
I managed to complete the bisect today (getting late here).
It said:
d4e33fac2408d37f7b52e80ca2a89f9fb482914f is first bad commit
commit d4e33fac2408d37f7b52e80ca2a89f9fb482914f
Author: Alan Cox <alan@xxxxxxxxxxxxxxx>
Date: Thu Jan 26 17:44:09 2012 +0000
serial: Kill off NO_IRQ
We transform the offenders into a test of irq <= 0 which will be ok while
the ARM people get their platform sorted. Once that is done (or in a while
if they don't do it anyway) then we will change them all to !irq checks.
For arch specific drivers that are already using NO_IRQ = 0 we just test
against zero so we don't need to re-review them later.
Signed-off-by: Alan Cox <alan@xxxxxxxxxxxxxxx>
Signed-off-by: Greg Kroah-Hartman <gregkh@xxxxxxx>
:040000 040000 e9f4b547c07e96339c1ea04b68c64b10ac576bb5 aa27d6cda463cfacc254cd74a787f68fb4c2f678 M drivers
This was also my prime suspect when I looked at tty/serial.
I assume we have some inconsistency in the interpretation
of irq == 0 somewhere.
In my first post I said that nothing unusual showed up in
the boot log.
But the following line is missing when it does not work:
input: Sun Type 5 keyboard as /devices/root/ffd31a40/ffd31cf0/serio0/input/input0
I will take a look at this tomorrow.
Sam
--
To unsubscribe from this list: send the line "unsubscribe sparclinux" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
