On Sat, Jul 21, 2018 at 12:41:41AM +0200, Wolfram Sang wrote:

quoted

[Ludovic Desroches]
Changes in v3:
 - rebase (cherry-pick was enough)

Thanks for the rebase. I wonder, though, I recall to had more
complicated issues. However...

quoted

 - fix checkpatch errors
 - tests:
   - hangs with a SAMA5D4 (master and slave on different busses) after about
   100 transfers. It's the firs time I do this test.
   - some mismatches with a SAMA5D4 as slave and a SAMA5D2 as master
   I don't know if it's a regression. I don't remember having seen this
   behavior previously.
   I think it's worth taking those patches even if there are some possible
   bugs. It'll allow to get more people using it and so to consolidate the
   slave mode support.

I really want to see those patches go upstream, too. But I am also not a
big fan of delivering the user something with known issues. Especially
not when they affect the main feature to be added. My rationale here is
that someone who is able to fix the issues remaining will also be able
to pick up and apply patches.

Maybe, maybe if it was to be enabled by a special
I2C_AT91_SLAVE_EXPERIMANTEL symbol with lots of explanations. I need to
think twice about that, though.

I understand your point. The experimental mentionning could be a good
trade-off. Let me know once you make up your mind.

Speaking of Kconfig, I think this series needs to place a

	select I2C_SLAVE

somewhere.

Ok I'll update it if we go further with this set of patches.

Regards

Ludovic