On Fri, Apr 03, 2009 at 12:55:36PM -0500, Hugo Villeneuve wrote:
> On Fri, 3 Apr 2009 08:55:41 -0700
> "Mark A. Greer" <[email protected]> wrote:
> 
> > On Tue, Mar 31, 2009 at 01:43:41PM -0700, David Brownell wrote:
> > > On Tuesday 31 March 2009, Mark A. Greer wrote:
> > > > The serial infrastructure is pretty limited WRT this (I have same
> > > > issue on da830 evm--I only want to use uart2).
> > > 
> > > Couldn't you add some kind of "ignore this one" flag to that
> > > infrastructure, and just have DaVinci use it?
> > > 
> > > This issue is hardly new to DaVinci chips, it's a longstanding
> > > annoyance.  Happens on x86 too, ISTR ...
> > 
> > Well, I wasted more of my life than I care to admit trying to better
> > understand (and tweak) the intricacies of the serial subsystem and
> > I'm not going to waste any more.
> > 
> > If someone else wants to do this, they get a gold star.
> > 
> > Hugo, I'm sorry but I'm going to back out most of your patch since
> > pretty much everyone (except you) agrees that misnaming uart2 is
> > unacceptable.
> 
> I don't disagree, I just don't care, which is not the same :)

:)

> First point: Before my patch, if you tried to enable UART0, UART1 and UART2, 
> everything was fine. But if you tried to enable only UART0 and UART2, it 
> failed miserably -> This is a BUG.
> 
> Second point: If you remove what I did, how will you handle the case when I 
> need to enable only UART0 and UART2?

I'll remove setting the flag to 0 but that doesn't really help since
serial8250_register_port() will fail on uart1 and, IIRC, stop there.

I'm not saying there isn't a bug, I'm just saying we're swapping one bug
for another so I'm going back to the original bug.

What happens if you leave uart1 enabled on you board?

> >  If you don't care what uart2 is called, then why not
> > tweak the plat_serial8250_port data?
> 
> Then it means that the serial platform data would need to go into the board 
> setup code (board-sffsdr.c), and not into dm644x.c for example. If this is 
> done like that, I don't mind, but it will need to be done for all the boards 
> for consistency.

If/when my patches are enabled, that data will be pointed to by
soc_info pointer so you can change it in your board before calling
davinci_serial_init().

Mark
--

_______________________________________________
Davinci-linux-open-source mailing list
[email protected]
http://linux.davincidsp.com/mailman/listinfo/davinci-linux-open-source

Reply via email to