- From: Frank Mori Hess <fmhess_at_users.sourceforge.net>
- Date: Sun, 14 Nov 2004 21:25:48 -0500
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 On Friday 12 November 2004 11:08 am, Calin A. Culianu wrote: > On Tue, 2 Nov 2004, Frank Mori Hess wrote: > > It's a limitation of comedi at present, you can only have one > > subdevice that supports read commands and one that supports write. In > > addition, your board hardware doesn't support dio commands. > > Is this limitation one that the comedi author(s) feel is justified > given the boards out there? I.E. is it a rare situation when the > board(s) installed in a machine could readily support multiple asynch. > IO on multiple subdevices (esp. in realtime) without stepping on each > other's toes? In other words, would it be terribly hard for someone to > come along some time in the future and modify the comedi internals to > support multiple input commands on multiple subdevices? Or would it > create a buggy and horribly messy situation due to the way the boards > themselves are designed? It is uncommon for a board's hardware to support more than 1 streaming read and 1 write subdevice. I definitely intend to remove the limitation in comedi eventually though, maybe by having a device file for each streaming subdevice using sysfs. - -- Frank -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.4 (GNU/Linux) iD8DBQFBmBOv5vihyNWuA4URArdEAKCChOL0xmAiEqvH13gnxrVThEtr5ACbBoiH IjMlSK0RjM8Jbqy74aQRLPE= =biwU -----END PGP SIGNATURE-----
Received on 2004-11-15Z02:25:48