Bug 8343

Summary:	broadcom wifi devices using wl driver, are not detected as wireless devices
Product:	Mageia	Reporter:	Luc Menut <lmenut>
Component:	RPM Packages	Assignee:	Colin Guthrie <mageia>
Status:	RESOLVED FIXED	QA Contact:
Severity:	normal
Priority:	release_blocker	CC:	mageia, tmb
Version:	Cauldron
Target Milestone:	---
Hardware:	All
OS:	Linux
Whiteboard:
Source RPM:	initscripts-9.34-30.mga3.src.rpm	CVE:
Status comment:

Description Luc Menut 2012-12-09 22:01:08 CET

Description of problem:
In network-functions, is_wireless_device() detects if a card is a wireless device using [-d "/sys/class/net/$1/phy80211" ] since kernel 3.6 (/sys/class/net/$1/wireless was removed in 3.6.*)
http://svnweb.mageia.org/packages?view=revision&revision=309529

Sadly, testing /sys/class/net/$1/phy80211 doesn't work with all the wireless devices. e.g. the wl driver doesn't use phy80211, so that broadcom wireless card using wl are not detected as wireless devices.
As /sys/class/net/$1/wireless is restored with kernel 3.7,
http://git.kernel.org/?p=linux/kernel/git/torvalds/linux.git;a=commit;h=38c1a01cf10c6e4049b4ffbd4a6af655df2a46e1

I suggest that we test both /sys/class/net/$1/wireless or /sys/class/net/$1/phy80211 :
is_wireless_device ()
{
    [ -d "/sys/class/net/$1/wireless" -o -d "/sys/class/net/$1/phy80211" ]
}


Version-Release number of selected component (if applicable):
initscripts-9.34-30.mga3.src.rpm
broadcom-wl-5.100.82.112-5.mga3.nonfree.src.rpm
kmod-broadcom-wl-5.100.82.112-49.mga3.nonfree.src.rpm

How reproducible:
always

Luc Menut 2012-12-09 22:02:54 CET

Priority: Normal => release_blocker
CC: (none) => mageia, tmb

Comment 1 Manuel Hiebel 2012-12-09 22:05:12 CET

dup of bug 7828 ? (but you have more details)

Comment 2 Thomas Backlund 2012-12-09 22:24:25 CET

fix (checking both paths) added in initscripts-9.41-2.mga3, will be part of mga3 beta1

Status: NEW => RESOLVED
Resolution: (none) => FIXED

Comment 3 Luc Menut 2012-12-09 22:33:40 CET

(In reply to comment #1)
> dup of bug 7828 ? (but you have more details)

I don't think so, at least not for the initial report (#7828c1), which concerns kernel 3.5 and was reported before that blino has switched the test to check phy80211 instead of wireless.

(In reply to comment #2)
> fix (checking both paths) added in initscripts-9.41-2.mga3, will be part of
> mga3 beta1

Thanks

Comment 4 Colin Guthrie 2012-12-10 00:29:37 CET

FWIW, I hadn't yet pushed newer initscripts to cauldron (was intending to but kept forgetting :s). It's been in testing for a while and I've been using it for ages quite happily. Might have a few minor network+network managers issues, but nothing too big I think. I remembered today (along with an unfinished dracut upgrade (jeez, I suck at the moment), but figured I'd best wait to after beta1. May as well push it now tho'. I'll hold off on the dracut upgrade as I know of at least one transitional issue I still need to fix.