On 5 January 2017 at 04:34, Alastair D'Silva <alast...@au1.ibm.com> wrote: > From: Alastair D'Silva <alast...@d-silva.org> > > Some devices change their behaviour based on the state of their input GPIO > lines. > > This patch allows testing of the variable behaviour by providing facilities > for the test to set the state of these GPIO lines. > > Signed-off-by: Alastair D'Silva <alast...@d-silva.org>
I still think this is wrong, I'm afraid. The GPIO line is already connected to something else in the simulation, and injecting a value that didn't come from whatever else it's connected to is likely to confuse things. Tests should work by programming the device that is connected to the other end of this GPIO to make it raise the line. thanks -- PMM
