: Alsa always uses card0 and does not use card1, even when card0 is a modem
A NOTE has been added to this issue.
======================================================================
<https://bugtrack.alsa-project.org/alsa-bug/view.php?id=359>
======================================================================
Reported By: davidsrsb
Assigned To:
======================================================================
Project: ALSA - driver Issue ID: 359 Category: PCI - intel8x0 Reproducibility: always Severity: major Priority: normal Status: new Distribution: Slackware 10.0 Kernel Version: 2.4.26 ====================================================================== Date Submitted: 07-01-2004 16:26 CEST Last Modified: 01-07-2005 19:18 CET
======================================================================
Summary: Alsa always uses card0 and does not use card1, even
when card0 is a modem
Description:
PC is laptop with AC97 DAA (Si3036 modem) detected as card0 and sound is Intel 8x0 as card1. Alsamixer has to be run with -c1 option to avoid "no mixer elems" error (like bug 338)
KDE has no sound ( same as bug 353)
======================================================================
----------------------------------------------------------------------
jdthood - 01-07-05 19:03
----------------------------------------------------------------------
Quoted 3 lines
The problem is that the current "numbering" concept doesn't match with the idea of hotplug. We need either stop using numbers, or introduce the mangling of physical and logical device numbers.
I agree that a general solution is for sound applications to select cards by type or id rather than by index. However, it remains the fact that some existing applications open card
https://bugtrack.alsa-project.org/alsa-bug/view.php?id=0. So GNU/Linux distributors would like to set things up so that modems and TV cards don't grab index 0.
My proposed solution is to apply the following patch and then use:
options snd-bt87x index=-2
The patch causes negative values of "index" to be treated as bitmasks indicating the permitted indices.
-1 = 0xffffffff = any index
-2 = 0xfffffffe = any index except 0----------------------------------------------------------------------
tiwai - 01-07-05 19:18
----------------------------------------------------------------------
Your hack looks OK to me as a short-term solution. But I'll postpone it after 1.0.8-final release.
(BTW, about this kind of discussion for general implementations, I'd like to do on ML rather than on web. I basically hate the web interface...)
Issue History
Date Modified Username Field Change
======================================================================
07-01-04 16:26 davidsrsb New Issue 07-01-04 16:26 davidsrsb Distribution => Slackware 10.0 07-01-04 16:26 davidsrsb Kernel Version => 2.4.26 07-02-04 14:34 pzad Note Added: 0001374 01-05-05 23:34 jdthood Note Added: 0003090 01-06-05 17:48 phormix Note Added: 0003092 01-07-05 18:10 tiwai Note Added: 0003108 01-07-05 18:28 jdthood Note Added: 0003112 01-07-05 18:59 jdthood Note Added: 0003114 01-07-05 18:59 jdthood Note Added: 0003115 01-07-05 18:59 jdthood Note Deleted: 0003115 01-07-05 19:00 jdthood File Added: core_init_index_jdth20050107-1.patch
01-07-05 19:03 jdthood Note Edited: 0003114
01-07-05 19:18 tiwai Note Added: 0003116
======================================================================
-------------------------------------------------------
The SF.Net email is sponsored by: Beat the post-holiday blues Get a FREE limited edition SourceForge.net t-shirt from ThinkGeek. It's fun and FREE -- well, almost....http://www.thinkgeek.com/sfshirt