On 8/19/2011 11:01, Ken Hornstein wrote:
When ppstest starts, it fires off two copies of my new printf statements inside the z8530tty.c file.Ah, okay, that makes sense! It wouldn't show up unless those events were being watched for.When ntpd starts, at first I thought it wasn't working but then I get a "last message repeated" entry. However it seems inconsistent. Stopping and starting ntpd doesn't show the message.That might be messages left over from when ppstest was running. I am thinking the problem is at the ntpd level (it looks like to me that it's not doing the right magic to enable the PPS reception in the kernel). What version of NetBSD are you running? I see that in the ntpd distribution there is a program called "pps-api"; that one might be worth compiling and trying out.
I'm running NetBSD-current (downloaded about a week ago) so 5.1. The pps-api test code actually works: # ./pps-api /dev/pps1 0.000000000 0.000000000 0 0 0.000000000 1313779702.115506089 0.000000000 1 0 -1313779702.115506172 1313779702.115506089 1313779702.985524433 1 1 0.870018344 1313779703.055525681 1313779702.985524433 2 1 -0.070001248 1313779703.055525681 1313779703.925546044 2 2 0.870020363 1313779703.995570290 1313779703.925546044 3 2 -0.070024246 1313779703.995570290 1313779704.865570673 3 3 0.870000383 1313779704.935566923 1313779704.865570673 4 3 -0.069996250 1313779704.935566923 1313779705.805586323 4 4 0.870019400 1313779705.875586574 1313779705.805586323 5 4 -0.070000251 1313779705.875586574 1313779706.745638990 5 5 0.870052416 1313779706.815637243 1313779706.745638990 6 5 -0.069998253 Very interesting now.