On 3 July 2018 at 19:49, Guenter Roeck <li...@roeck-us.net> wrote: > Oddly enough, I now seem to be unable to reproduce the problem either. > Weird, when I wrote the patch it was absolutely necessary. > > The only other change I can think of would be the code to set the > serial baud rate in the boot loader, which I did not have initially. > Is it possible that the hangup could have been caused by that ?
Seems unlikely -- unless you're actually connecting the emulated serial port to a real host serial port, we don't do anything with the baudrate in the cmsdk uart model except complain if the guest hasn't set it to something valid. > Is it worth tracking it down, or are we ok ? It does worry me a little, but if neither you nor I can repro it at the moment there's not a lot we can do about it I think. Looking at the ptimer and cmsdk timer code I think it *should* now implement what the hardware does... If you see the problem again in future let me know and we can have another go at figuring out what's going on. > The series WFM as well. I'll send out a Tested-by: in a minute. Thanks for the testing. -- PMM
