On 10/02/18 23:11, David Gibson wrote:
On Sun, Feb 11, 2018 at 09:32:14AM +1100, David Gibson wrote:
On Fri, Feb 09, 2018 at 06:51:35PM +0000, Mark Cave-Ayland wrote:
This allows us to more easily differentiate between the timebase frequency used
to calibrate the MacOS timers and the actual frequency of the hardware clock as
indicated by CUDA_TIMER_FREQ.
Signed-off-by: Mark Cave-Ayland <mark.cave-ayl...@ilande.co.uk>
Reviewed-by: Philippe Mathieu-Daudé <f4...@amsat.org>
Applied, thanks.
Actually, this patch doesn't compile, because you've changed
ti->frequency in a couple of places where you should only be changing
s->frequency. I've fixed it up in my tree.
Ah apologies there - this was a mistake I made quite early on in the
patchset which I must have accidentally brought back in during a
subsequent rebase :(
I've just done a check on patch 7 ("cuda: set timer 1 frequency property
to CUDA_TIMER_FREQ") and confirmed that the cuda.c version in your
ppc-for-2.12 branch matches that in my local branch, so your fixups are
good.
Any thoughts on the last few patches? I know Philippe had a question for
Stefan re: the trace-events patch, however it would be useful for
patches 10/11 to be applied as they complete the transition from CUDA
over to using the mos6522 device (also providing an initial example as
to how it can be used).
ATB,
Mark.
