ntp/ntpd/refclock_oncore.c

6  * this stuff is worth it, you can buy me a beer in return.   Poul-Henning Kamp
75  * driver not starting up, and it is often because of signal strength being low.
82  * Add back in @@Bl code to do the @@Bj/@@Gj that is in later ONCOREs
85  * before the leap second is to be applied, but in practice at least some of
86  * the receivers turn this indicator on as soon as the message is posted, which
131  *  state machine state) are printed to CLOCKSTATS if that file is enabled
137  * doing an average of 10000 valid 2D and 3D fixes is what the automatic
139  * it seems like it is only using 3D fixes.
256 /* It is possible to test the VP/UT each cycle (@@Ea or equivalent) to
257  * see what mode it is in.  The bits on the M12 are multiplexed with
429  * Understanding the next bit here is not easy unless you have a manual
643 	 * This is a bit complicated, a we dont want to open the same file twice
665 	   Note that the linuxPPS N_PPS file is just like a N_TTY, so we can do
694 	/* for LINUX the PPS device is the result of a line discipline.
748 	 * an old one is found.
894 	/* only set what is legal */
903 	/* If HARDPPS is on, we tell kernel */
915 		/* we know that 'i' is legal from above */
944 	* The first thing we do is see if there is an instance->shmem_fname file (still)
1073 		     "SHMEM (size = %ld) is CONFIGURED and available as %s",
1092  * where N is the unit number viz 127.127.30.N.
1102  *    (0) ONCORE is preinitialized, don't do anything to change it.
1110  *     NB. If a POSITION is specified in the config file with mode=(2,4) [SITE SURVEY]
1111  *	   then this position is set as the INITIAL position of the ONCORE.
1114  * Note that an Oncore UT without a battery backup retains NO information if it is
1116  * For an Oncore VP, there is an eeprom that will contain this data, along with the
1118  * So a UT without Battery Backup is equivalent to doing a HARD RESET on each
1119  *   power cycle, since there is nowhere to store the data.
1130  *	# is comment to end of line
1143  *	   followed by 1-2 fields.  First is a number, the second is 'FT' or 'M'
1144  *	   for feet or meters.	HT is the height above the GPS ellipsoid.
1148  *	There is an optional line, starting with DELAY, followed
1149  *	   by 1 or two fields.	The first is a number (a time) the second is
1151  *	    DELAY  is cable delay, typically a few tens of ns.
1153  *	There is an optional line, starting with OFFSET, followed
1154  *	   by 1 or two fields.	The first is a number (a time) the second is
1156  *	   OFFSET is the offset of the PPS pulse from 0. (only fully implemented
1158  *		offset if the Kernel PLL is in use, but can only do this presently
1160  *		then there is no problem.
1162  *	There is an optional line, with either ASSERT or CLEAR on it, which
1163  *	   determine which transition of the PPS signal is used for timing by the
1164  *	   PPSAPI.  If neither is present, then ASSERT is assumed.
1166  *	   For Flag2, ASSERT=0, and hence is default.
1168  *	There is an optional line, with HARDPPS on it.	Including this line causes
1171  *	   For Flag3, 0 is disabled, and the default.
1175  *	   The file name is the file associated with the shared memory.
1177  *	In shared memory, there is one 'record' for each returned variable.
1182  *	   record is filled once every 15s.
1190  *	There is an optional line with MASK on it followed by one integer field in the
1193  *	   is 10 deg for the VP and 0 deg for all other receivers.
1195  *	There is an optional line with PPSCONTROL on it (only valid for M12 or M12+T
1196  *	   receivers, the option is read, but ignored for all others)
1197  *	   and it is followed by:
1198  *		ON	   Turn PPS on.  This is the default and the default for other
1199  *			       oncore receivers.  The PPS is on even if not tracking
1203  *	  The OFF option is NOT implemented, since the Oncore driver will not work
1376 			instance->traim_in = 1; 	/* so TRAIM alone is YES */
1387 				instance->pps_control = 2;		/* PPS on when satellite is available */
1389 				instance->pps_control = 3;		/* PPS on when TRAIM status is OK */
1604 	 *  times we get here in 0D mode (the 1/15 is in 3D for SHMEM).
1605 	 * This gives good time, which gets better when the SS is done.
1625 	/* Once the Almanac is valid, the M12+T does not produce valid UTC
1715      alternate code for previous 4 lines is
1735 	/* offset in ns, and is positive (late), we subtract */
1746 	/* The following code is necessary, and not just a time_pps_setparams,
1749 	 * that it needs).  We take what is there and ADD what we need.
1750 	 * [[ The results from the time_pps_getcap is unlikely to change so
1752 	 * Unfortunately, there is only ONE set of mode bits in the kernel per
1755 	 * There is still a race condition here where we might mess up someone
1756 	 * elses mode, but if he is being careful too, he should survive.
1887 	oncore_log(instance, LOG_INFO, Msg);	 /* this is long message above */
1998 		cp = "is";
2057  * Nb. @@Ay is not supported for early UT (no plus) model
2074 	oncore_log_f(instance, LOG_INFO, "PPS Offset is set to %ld ns",
2098 	oncore_log_f(instance, LOG_INFO, "Cable delay is set to %ld ns",
2183 	/* check the antenna (did it get unplugged) and almanac (is it ready) for changes. */
2251 	/* Unfortunately, the Gd3 command returns '3' for the M12 v1.3 firmware where it is
2254 	 * then check Ha[130]&0x10 to see if a SS is in progress.
2282 					 * ONCORE_SS_HW.  If the Hardware is capable of doing a Site Survey, then
2468  * It turns on the LEAP indicator when the data is set, and does not,
2472  * @@Bj is only called in June/December.
2686  * sleep(2) is wasteful, and may cause trouble for some OS's, repeating
2688  * been cleared, we reissue the @@Cj that is issued below.
2690  * The first is to get the info, the 2nd is just used as a safe command
2784 		oncore_log(instance, LOG_NOTICE, "Cb: Response is NO ALMANAC");
2825  * This is the Grand Central Station for the Preliminary Initialization.
2832  *   This info is printed, but no longer used.
2884  * Determine the Type from the Model #, this determines #chan and if TRAIM is
2887  * The Information from this routine is NO LONGER USED.
2919 	/* from model number decide which Oncore this is,
3041 	/* If there is Position input in the Config file
3172 		     "Date/Time set to: %d%s%d %2d:%02d:%02d GMT (GMT offset is %s%02d:%02d)",
3232 	 * get the actual leap second -- that is, the one that is
3235 	 * Going through all this for a simple log is probably overkill,
3236 	 * so for fixing bug#1050 the message output is changed to
3317 	 * model number of the attached oncore.  This is not always correct since
3322 	 * This #chan is used by the code rather than that calculated from the model number.
3402  * Note that the ONCORE firmware for the Bj command is wrong at
3405  * data message (page 18, subframe 4) is updated to a date in the
3408  * Since there is a one bit flag, there is no way to tell if it is
3418  * interpret it ourselves, but since this is specific to this receiver
3419  * only, and the above workaround is adequate, we don't bother.
3438 		if (instance->saw_Gj == 0)	/* 0 is dont know if we have Gj */
3467  *  len is length WITH CR/LF.
3508 	/* if MSL is not set, the calculation is meaningless */
3700 	int idx, idy, is, imx, imy;
3739 	is  = lat%60000;
3740 	xs  = is/1000.;
3741 	is  = lon%60000;
3742 	ys  = is/1000.;
4013 		 * Check if we have a SHMEM, and if so try to load whatever is there.
4085 		/* This is obviously a bad date/time... */
4127 			    "Max Message Length so far is %d",