Re: [RFC PATCH 1/3] MIPS: Alchemy: extended DB1200 board support.

[Kevin Hickey <khickey@netlogicmicro.com>]

On Tue, 2009-11-03 at 16:44 +0100, Manuel Lauss wrote:
> On Tue, Nov 3, 2009 at 4:41 PM, Kevin Hickey <khickey@netlogicmicro.com> wrote:
> > On Mon, 2009-11-02 at 21:21 +0100, Manuel Lauss wrote:
> >> Create own directory for DB1200 code and update it with new features.
> >>
> >> - SPI support:
> >>   - tmp121 temperature sensor
> >>   - SPI flash on DB1200
> >> - I2C support
> >>   - NE1619 sensor
> >>   - AT24 eeprom
> >> - I2C/SPI can be selected at boot time via switch S6.8
> >> - Carddetect IRQs for SD cards.
> >> - gen_nand based NAND support.
> >> - hexleds count sleep/wake transitions.
> >>
> >> Signed-off-by: Manuel Lauss <manuel.lauss@gmail.com>
> >> ---
> >
> > The code in this patch all looks good to me.  I don't understand how
> > much value is added by using the hex LEDs for counting sleep/wake
> > transitions.  In our internal builds, we use the hex LEDs for displaying
> > the last interrupt serviced (useful on hangs/crashes and for getting a
> > general sense of what the hardware is working on), the dots blink on
> > timer ticks (often every 100 or 1000 depending on the clock) and the
> > Idle state is shown on LED0.  I don't really have any strong attachment
> > to those usages, but they've served us well.
> 
> I admit it has limited value.. I initially used it to find out how many wakeups
> per second are happening with and without high-speed usb devices attached.
> I left it in because I couldn't come up with any other uses.  Please feel free
> to change it any time.

Fair enough.  If I get some time down the line I'll look into that.  In
the meantime this patch has a lot of value so it should probably just
stay as it is.

Again, if it's not too late:
Acked-by: Kevin Hickey <khickey@netlogicmicro.com>

=Kevin