I recall in the forum, a couple of people reported that if they uncomment line 16 in /etc/init.d/rc.pcmcia:
then rebooted, PCMCIA worked.
I want to find someone who is in this situation, to do a test for me.
Firstly, comment-out that line, if not already, and reboot if you have had to comment it out.
Then look in /sys/class, do you see pcmcia_socket/pcmcia_socket0?
Run this in a terminal:
# elspci -l
(that is letter l on the end)
You should see a line with <yenta_socket> on it, can you please post that entire line here or the forum.
Better yet, send the entire output from running "elspci -l" (highlight text in terminal by holding left-mouse-button down and drag, then click the middle-mouse-button to paste)
Then, if you uncomment that line, then reboot, look again in /sys/class, do you get pcmcia_socket?
Comments:Posted on 11 May 2011, 11:05 by BarryK
I have had some feedback on the forum:
Puppy 4.3.1 has the line "pcmcia-socket-startup" in /etc/init.d/rc.pcmcia, but later puppies have that line commented-out.
'pcmcia-socket-startup' does not normally need to be executed, I think only for older PCMCIA slot interfaces that require firmware (.cis files in /lib/firmware) to be loaded.
'pcmcia-socket-startup' execution is now supposed to be triggered via a udev rule (see /etc/udev/rules.d/85-pcmcia.rules or /lib/udev/rules.d/60-pcmcia.rules). However, the code in /etc/rc.d/rc.sysinit is not replaying the pcmcia kernel uevents.
There is code in there to do the replay, but there seems to have been a change in the layout in /sys. Anyway, after getting feedback from vickyg2003, I can see how to do the replay, and have put a fix in to /etc/rc.d/rc.sysinit.
Posted on 12 May 2011, 9:44 by BarryK
PCMCIA fix confirmed
Forum member Keef has also confirmed that the fix works. This is very good.