"Fossies" - the Fresh Open Source Software Archive

Member "PPSkit-2.1.7/TODO" (17 Apr 2001, 2402 Bytes) of package /linux/misc/old/PPSkit-2.1.7.tar.gz:

As a special service "Fossies" has tried to format the requested text file into HTML format (style: standard) with prefixed line numbers. Alternatively you can here view or download the uninterpreted source code file.

    1 (Emacs: -*- indented-text -*-)
    2 Things to be done ($Date: 2001/04/17 12:17:10 $):
    4         0) Reset some PPS variables when setting the time...
    5 	   Really think about the mess with the line discipline...
    7 	1) Prove that the new code is correct and that it does not
    8            break anything.  Check if the Linux timekeeping code has
    9            any benefit from the new code.  Most likely the precision
   10            with the new code should be around few microseconds while
   11            the method using select() is limited to a few (10)
   12            milliseconds in kernel 2.0. [5us are definitely possible]
   14         2) Make sure that do_gettimeoffset returns a consistent time
   15            offset (it performs internal calibrations occasionally that
   16            also take time).  The offset should be as precise as
   17            possible. [looks very good]
   19 	3) Extend the patch for architecture other than i386.  As I
   20 	   only have a Intel PC, I can't test other platforms.
   21 	   Feedback is welcome!  [partially done, but completely
   22 	   untested by me]
   24 	4) Get the time of the pulse more exactly (must be done in the
   25 	   serial driver's interrupt routine); that is earlier and
   26 	   with a fixed time offset (currently possibly variable).
   27 	   [done in 0.7.0]
   29 	   At least the propagation delay should be determined (a
   30 	   tricky job).  I'm no longer sure whether it is really
   31 	   needed... [Settable with PPS API]
   33 	5) Discussions with several interested users have shown that I
   34 	   should write a NTP/PPS MINI-HOWTO (still to be done, but
   35 	   have a look at README)...
   37 	   [I preferred to make a hyperlinked HTML version from the
   38 	   old NTP-FAQ instead]
   40 	   In addition I wrote another document (``NTP FAQ'') using
   41 	   the DocBook SGML DTD [released]
   43 	6) Improve Linux support in xntp (NTP software for UNIX) even
   44 	   more.  Frank Kardel added some support for the parse
   45 	   drivers [temporary hack] My Meinberg PZF535 failed
   46 	   completely with the nanokernel code (STA_NANO being set)
   47 	   (magic -127 second offset) while ntpd-4.0.92c seems to work
   48 	   with the nanokernel when STA_NANO is off. [completely clueless]
   50 	7) Make sure the multiple CPUs don't mess up the time when
   51            interpolating time using the TSC registers local to the
   52            CPUs. [Linux kernel since 2.2.16 synchronizes TSC counters
   53            during boot, requiring identical CPU frequencies or option
   54            ``notsc'']