- From: Frank Mori Hess <fmhess_at_users.sourceforge.net>
- Date: Sun, 21 Dec 2003 13:56:18 -0500
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 On Tuesday 16 December 2003 10:55 am, Nils Juergens wrote: > > The CVS version of adv_pci1710 has one counter subdevice (which actually > works on one of the three counters on the 8254), but the configuration > code was rejected because it does not adhere to the comedi counter > configuration standards (whatever those may be, if i were clearer on > them i would have fixed the code). > > If you look at that code and maybe at whats in 8253.h (some stuff in > there for 8254, too), you should be able to add support for the counter > for your board. > Nils, I think we can enable your counter configuration insn. There are just a few changes that you need to make. First, don't read the counter status back as data[1] in the read instruction. The read instruction should read back the counter value, and thats it. You can define another special configuration instruction specifically for querying the status of a 8253/4 counter. Second, the option to wait (possibly forever) for the counter to load the value you have written to it is unacceptable. Once you add the insn for querying the counter's state, the user can use that to poll the counter until it loads. The config insn needs to be changed to use data[0] as the id number of the configuration instruction (the already used configuration ids are the INSN_CONFIG_* defines in linux/comedi.h). Oh, and document how your configuration instruction works somewhere, for now it could go into comedi/drivers/comedi_counter_unstable.h. - -- Frank -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.0.6 (GNU/Linux) Comment: For info see http://www.gnupg.org iD8DBQE/5ezS5vihyNWuA4URAvcmAJ9rALadOiO0yJnx4+F3pji2lx5UDwCg3sd5 iBw4ZBIHYl3RjF4f1UnRLVs= =en3g -----END PGP SIGNATURE-----
Received on 2003-12-21Z18:56:18