- From: Frank Mori Hess <fmhess_at_users.sourceforge.net>
- Date: Thu, 29 May 2003 17:43:57 -0500
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 On Thursday 29 May 2003 02:15 pm, Martin Merz wrote: > When i set TRIG_WAKE_EOS the read() - Function comes always back with > "read: broken pipe". So i compiled the driver with "#undef PCIDMA" (in > ni_pcimio.c), i read this in the Mailinglist. This time it seems better. I put some fixes into cvs to try and get TRIG_WAKE_EOS to work with PCIDMA enabled. Would you test that? > > How many scans .... good question. When i set this (Example 1): > cmd->scan_end_src = TRIG_COUNT; > cmd->scan_end_arg = 2; > cmd->stop_src = TRIG_COUNT; > cmd->stop_arg = 1; > i got more than one scans or converts !? ... it's endles. After each > convert the read comes back. I thought after each scan the read() comes > with the analog values and after that with an ret Value 0 !? It's a hardware limitation, you can't ask for less than 2 scans. The driver catches it during the command test now. > > My questions: > 1. Is this the right way to read from the analog channels and the dio's Well, there may be a significant delay between the analog scan completing and you reading the dio lines. Ideally, you would use dio hardware that supported external triggering. The driver support in comedi for dio commands is not very extensive yet however. - -- Frank -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.0.6 (GNU/Linux) Comment: For info see http://www.gnupg.org iD8DBQE+1o0t5vihyNWuA4URAmK1AKCC/XmAHzexubctEFKeewZeNwBZZwCcDqeO 4AmE3AFMpcMQFfdRUY5aQxM= =eVIn -----END PGP SIGNATURE-----
Received on 2003-05-29Z21:43:57