On úterý 2. dubna 2019 6:50:17 CEST, Linus Walleij wrote:
On Thu, Mar 7, 2019 at 8:57 PM Jan Kundrát <jan.kundrat@xxxxxxxxx> wrote:
The comment for this dbg_show says that it is supposed to return more
than what the generic code is showing, including de-glitching. That's
wrong because:
- this chip does not support deglitching,
- the code does not print anything extra compared to the generic
handler,
- its behavior is different because it skips unrequested GPIOs; the
generic code prints their names if they're assigned
There is an important difference, though. Previously, dbg_show would
re-check some registers to see if they still match what the regmap
thinks should be in there. This was semi-useful when develpoing the HW
board because it immediately pointed to SPI wiring problem if a CS
connection was missing (0xffs are easy to see). However, I do not think
that this makes much sense -- and one could always do this in some other
way if needed.
Signed-off-by: Jan Kundrát <jan.kundrat@xxxxxxxxx>
Cc: Phil Reid <preid@xxxxxxxxxxxxxxxxx>
Cc: Linus Walleij <linus.walleij@xxxxxxxxxx>
This patch seems to have fallen between the cracks.
Is this something that should still be applied? (Separately.)
Hi Linus, yes, I think that the patch has merit and should be applied.
With kind regards,
Jan
