Re: [gentoo-user] /dev/snd/seq access mode and permission

[Bill Kenworthy <billk@iinet.net.au>]

Try this (my "fix" is in /lib/udev/rules.d/40-gentoo.rules):


moriah ~ # equery f udev
 * Searching for udev ...
 * Contents of sys-fs/udev-171-r6:
/etc
/etc/conf.d
/etc/conf.d/udev
/etc/init.d
...
/lib/udev/rules.d
/lib/udev/rules.d/30-kernel-compat.rules
/lib/udev/rules.d/40-gentoo.rules   
/lib/udev/rules.d/42-qemu-usb.rules 
/lib/udev/rules.d/50-firmware.rules 
/lib/udev/rules.d/50-udev-default.rules
/lib/udev/rules.d/60-cdrom_id.rules 
/lib/udev/rules.d/60-floppy.rules   
/lib/udev/rules.d/60-persistent-alsa.rules
/lib/udev/rules.d/60-persistent-input.rules
...


moriah ~ # grep snd /lib/udev/rules.d/*
/lib/udev/rules.d/40-gentoo.rules:SUBSYSTEM=="snd",     GROUP="audio"
/lib/udev/rules.d/50-udev-default.rules:KERNEL=="seq",                  GROUP="audio", MODE="0660", OPTIONS+="static_node=snd/seq"
/lib/udev/rules.d/50-udev-default.rules:KERNEL=="timer",                GROUP="audio", MODE="0660", OPTIONS+="static_node=snd/timer"
/lib/udev/rules.d/60-persistent-alsa.rules:ENV{ID_SERIAL}=="?*", ENV{ID_IFACE}=="?*", SYMLINK+="snd/by-id/$env{ID_BUS}-$env{ID_SERIAL}-$env{ID_IFACE}"
/lib/udev/rules.d/60-persistent-alsa.rules:ENV{ID_SERIAL}=="?*", ENV{ID_IFACE}=="", SYMLINK+="snd/by-id/$env{ID_BUS}-$env{ID_SERIAL}"
/lib/udev/rules.d/60-persistent-alsa.rules:ENV{ID_PATH}=="?*", SYMLINK+="snd/by-path/$env{ID_PATH}"
moriah ~ # 

BillK


On Wed, 2012-08-22 at 20:35 -0700, Cinder wrote:
> I have tried creating this rule:
> 
> /etc/udev/rules.d/40-seq.rules
>    KERNEL=="snd/seq",      GROUP="audio", MODE="0666"
> 
> ...but I'm not sure about the kernel key pair. I have tried matching "/dev/snd/seq" and just "seq"
> aswell.
> 
> The Bug 406871 for sys-fs/udev-171-r5 looks exactly right, butI have sys-fs/udev-171-r6 installed. I''l check my kernel config and try disabling tmpfs. I have read that it helps real time audio performanc with jack(audio-connection-kit)though.
> 
> I have tried removing all the rules in /etc/udev/rules.d. but the mode and permissions on /dev/snd/seq persist.
> 
> Thanks for everyones help.
> 
> --- marcec@gmx.de wrote:
> 
> From: Marc Joliet <marcec@gmx.de>
> To: gentoo-user@lists.gentoo.org
> Subject: Re: [gentoo-user] /dev/snd/seq access mode and permission
> Date: Fri, 17 Aug 2012 21:20:08 +0200
> 
> Am Fri, 17 Aug 2012 11:40:47 +0200
> schrieb Marc Joliet <marcec@gmx.de>:
> 
> > Am Fri, 17 Aug 2012 01:54:34 -0700
> > schrieb Cinder <cinder@linuxwaves.com>:
> > 
> > > Hi, how do I make changes to permissions and access mode of device nodes persistent? At the moment I have to chown and chmod the /dev/snd/seq node every boot to make it accessible to my user. the other nodes are fine. Here's the output of ls -l /dev/snd/
> > > 
> > > total 0
> > > drwxr-xr-x  2 root root       60 Aug 17 18:44 by-path
> > > crw-rw----+ 1 root audio 116, 12 Aug 17 18:44 controlC0
> > > crw-rw----+ 1 root audio 116, 11 Aug 17 18:44 hwC0D0
> > > crw-rw----+ 1 root audio 116, 10 Aug 17 18:44 hwC0D3
> > > crw-rw----+ 1 root audio 116,  9 Aug 17 18:44 hwC0D4
> > > crw-rw----+ 1 root audio 116,  8 Aug 17 18:44 hwC0D5
> > > crw-rw----+ 1 root audio 116,  7 Aug 17 18:44 pcmC0D0c
> > > crw-rw----+ 1 root audio 116,  6 Aug 17 18:44 pcmC0D0p
> > > crw-rw----+ 1 root audio 116,  5 Aug 17 18:44 pcmC0D1p
> > > crw-rw----+ 1 root audio 116,  4 Aug 17 18:44 pcmC0D3p
> > > crw-rw----+ 1 root audio 116,  3 Aug 17 18:44 pcmC0D7p
> > > crw-rw----+ 1 root audio 116,  2 Aug 17 18:44 pcmC0D8p
> > > crw-------  1 root root  116,  1 Aug 17 18:44 seq
> > > crw-rw----+ 1 root audio 116, 33 Aug 17 18:44 timer
> > > 
> > > I need /dev/snd/seq to look look the others. I can't find the udev rule or configuration that creates these nodes. Many thanks for any consideration.
> > 
> > I have a hack for the same issue in my /etc/local.d/. A comment I put there
> > says this:
> > 
> > # this is caused by using devtmpfs, which creates nodes with root:root and 600;
> > # I believe this is fixed by udev upstream
> > 
> > So devtmpfs creates the device node before udev runs, but udev does not correct
> > the access permissions, which is however fixed by udev upstream (perhaps
> > already in ~arch?). Sadly I do not remember where I read this, but google should
> > be of help there.
> 
> Ah, yes, I did a quick search on b.g.o and found this:
> 
>   https://bugs.gentoo.org/show_bug.cgi?id=406871
> 
> So my comment is wrong, it doesn't have anything to do with devtmpfs, but udev
> upstream did fix it :) .
> 
> HTH