Andy Green wrote:
> About these fake interrupts, SDIO uses D1 to signal async interrupt...
> does SDIO allow 1-bit data mode?  If you can just select it as a flag to
> the host controller stuff, it will allow to confirm that the problem
> doesn't come from data traffic on D1 being wrongly understood as
> interrupt request actions...

Yeah, I had that on my list of things to try. Just did it. Nope,
doesn't help :-(

I'll repeat it later with the scope watching, though.

We also asked Samsung if they have any silicon errata, but they
say the chip ought to be perfect.

- Werner

Reply via email to