zdebanos commented on PR #3000: URL: https://github.com/apache/nuttx-apps/pull/3000#issuecomment-2686045791
I've managed to prepare the documentation for the cyclictest utility, I've created a [pull request here](https://github.com/apache/nuttx/pull/15913). Concerning the semaphore systemtickhook discussion, I've decided to delete such measurement from this cyclictest, so I mark this as resolved @xiaoxiang781216. I actually did not think twice before implemententing such thing because I found out that measuring the latency would probably too much of a hassle (in the `board_timerhook` function, one would need to measure the first timestamp, so that would require putting the `board_timerhook` function definition in the `cyclictest.c` file, which is not ideal). Since this cyclictest program allows for up to 4 measure-waiting combinations, I've also fixed the calculations. However, I do not have time right now to do proper testing with the POSIX time functions, because it would require project reconfiguration to the tickless mode. The command `cyclictest -p 150 -T /dev/timer1 -m 1 -n 1 -h 20 -D 100 -i 50` (it's mentioned in the documentation) uses the timer device to measure the latencies. Its output is: ``` # Histogram 000000 000000 000001 000000 000002 000000 000003 000000 000004 000000 000005 000000 000006 000000 000007 000000 000008 000000 000009 000000 000010 603045 000011 1395782 000012 000804 000013 000153 000014 000034 000015 000083 000016 000030 000017 000000 000018 000000 000019 000000 # Total: 001999931 # Min Latencies: 00010 # Avg Latencies: 00010 # Max Latencies: 00016 # Histogram Overflows: 00000 ``` Tested on ATSAMV71Q21B 300 MHz, Cortex M7. The units are in us. -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: commits-unsubscr...@nuttx.apache.org For queries about this service, please contact Infrastructure at: us...@infra.apache.org
