Have an L1-C5 I am using (500MHz). I am clocking this L1 at 12.288MHz. In the .xn file if I use 400MHz for the system frequency, everything is fine. If I put it at 500MHz, then two things happen.
During build:
1 - system frequency for node "Slave" could not be achived. Using 499998720Hz
2 - reference frequency for node "Slave" could not be achived. Using 99999744Hz
Ok, I guess. As close as it could get. (Going through the calc manually, it seems the above numbers are not correct either. The calculations give 497.66MHz, assuming it rounds down.)
The real problem is during debug, I get a runtime exception during this instruction:
configure_clock_rate(c_spi1, 100, spi_clock_div)
Disassembly:
0x00010448 <configure_clock_rate.coersed..f.7b0.7d.28ui.2cui.2cui.29+150>: ecallf (1r) r4
Error in configure_clock_rate with a non 13MHz osc
-
- Respected Member
- Posts: 283
- Joined: Fri Mar 19, 2010 4:49 am
-
- XCore Addict
- Posts: 169
- Joined: Fri Jan 08, 2010 12:13 am
Hi,
Some clock divider settings on configure clock rate are not valid and will result in an exception. From the library documentation in <install dir>/XMOS/DesktopTools/doc/libs/html/index.html - navigate to File List -> xs1.h
You will see the following
Hope that helps,
Some clock divider settings on configure clock rate are not valid and will result in an exception. From the library documentation in <install dir>/XMOS/DesktopTools/doc/libs/html/index.html - navigate to File List -> xs1.h
You will see the following
So, you can either find the closest one yourself - or if you are trusting then you can use the configure_clock_rate_at_least(...) or configure_clock_rate_at_most(...) functions to find the closest one above or below the required divide (respectively).If the specified rate is not supported by the hardware, an exception is raised.
Hope that helps,
Paul
On two occasions I have been asked, 'Pray, Mr. Babbage, if you put into the machine wrong figures, will the right answers come out?' I am not able rightly to apprehend the kind of confusion of ideas that could provoke such a question.
On two occasions I have been asked, 'Pray, Mr. Babbage, if you put into the machine wrong figures, will the right answers come out?' I am not able rightly to apprehend the kind of confusion of ideas that could provoke such a question.
-
- XCore Addict
- Posts: 165
- Joined: Wed Feb 10, 2010 2:32 pm
@Bearcat
Did you manage to sort this problem out?
Did you manage to sort this problem out?
-
- Respected Member
- Posts: 283
- Joined: Fri Mar 19, 2010 4:49 am
All is fine. Using a non 13MHz freq is ok. A few warning messages from the various tools is the only inconvinence.
I had not discovered the othe XS1 calls. Now I am quite famaliar with them.
Thanks for everyones help.
I had not discovered the othe XS1 calls. Now I am quite famaliar with them.
Thanks for everyones help.