Grant Likely wrote:
>> > Personally, I'm not fond of this approach. There is already some
>> > traction to using the reg-shift property to specify spacing, and I
>> > think it would be appropriate to also define a reg-offset property to
>> > handle the +3 offset and then let the xilinx 16550 nodes use those.
>> Why do we need a reg-offset property when we can just add the offset
>> to the appropriate word(s) in the reg property?
> Primarily because the device creates 32 byte registers starting at 0;
> but they are also big-endian byte accessible so a byte read at offset
> 8 also works.
Probably I misunderstood you: does it give the same result as offset 11?
er; typo; oops. A 32 bit read add offset 0 is the same as a byte read
at offset *3*.
Oh, well... unfortunately, we can't use UPIO_MEM32 "register model" in
8250.c anyway since that makes use of readl()/writel() -- which treat the bus
as bigendian on PPC... anyway, we would need at least a "reg-size" property,
if not new "compatible"...
> reg-offset seems to be a better description of the hardware to me.
Have you considered using the existing "big-endian" property?
No I haven't, but that would work too. I'm happy with that if it
works for you. If the property was defined, then the byte offset to
the first reg would be adjusted by 1^(reg-shift) - 1
You don't mean "xor" by ^, do you? :-O
In fact, it should be <<...
Cheers,
g.
WBR, Sergei
_______________________________________________
Linuxppc-dev mailing list
Linuxppc-dev@ozlabs.org
https://ozlabs.org/mailman/listinfo/linuxppc-dev
