Re: [PATCH] tty: xilinx_uartps: Really fix id assignment

[Michal Simek <michal.simek@xilinx.com>]

On 13. 07. 20 9:11, Helmut Grohne wrote:
> Hi Michal,
> 
> On Fri, Jul 10, 2020 at 01:45:21PM +0200, Michal Simek wrote:
>> Thanks for the patch. I think that it will be very useful to list out
>> all testcases to make sure that we test all that scenarios and finally
>> get over it.
>> We are testing scenarios and likely we are missing some that we are not
>> able catch all issues before we submit a patch.
> 
> Let me try to enumerate those I know:
> 
> uart0    | uart1    | console | remark
> ---------+----------+---------+----------
> serial0  | serial1  | ttyPS0  | regular case
> serial0  | serial1  | ttyPS1  | normal assignment, second console
> serial1  | serial0  | ttyPS0  | -> Jan Kiszka, broken since revert
> disabled | serial0  | ttyPS0  | use only uart1 as serial0
> serial0  | disabled | ttyPS0  | regular case with uart1 disabled
> 
> Out of these, I'm actively using configurations 3 and 4.
> 
> Which of these scenarios do you test already?

For above we are missing also others
serial1 | serial0 | ttyPS1
disabled| serial1 | ttyPS1
serial1 | disables | ttyPS0

All of these above are just not setting any console= on bootargs.
All of these should be tested with earlycon in bootargs to also test
earlyconsole.
All combinations with both uarts should be also tested with
console=ttyPS0,115200 and console=ttyPS1,115200.

Driver supports up to 16 ports. It means also some ttyPS2 and more
should be also tested.

What I found recently in connection to Jan's issue was that testing was
done with bootargs setup which prevent seeing his issue on ultra96.

It means mix of these combinations is tested regularly but not all of
them. Do you see any other combination which is not supported?
When we have list I will ask testing team to test all combinations.

Thanks,
Michal