[Leaplist] Re: REVISITED: Fedora 9 on Gateway P-7811FX "Centrino 2" notebook

[Bryan J. Smith]

On Fri, 2008-09-05 at 15:00 -0400, Richard F. Ostrow Jr. wrote:
> Bummer. I've confirmed that mine works encrypted, but I haven't been able
> to get my scanner working (If I know the network I'm connecting to, it
> will find it and connect, or it will find anything and connect
> (unencrypted) if it fails to find my network).

Hmmm, can't help you there.  I just see the networks in the
NetworkManager applet for GNOME.  The issue has always just getting the
driver, but once the device was there, it worked.

> Well, a starting place is to look at the switch itself. On my laptop, this
> is located at /sys/bus/pci/drivers/iwl4965/000:04:00.0/rf_kill
> The contents of this file are 0 if the RF_KILL switch is disengaged, or 1
> if RF_KILL is engaged (meaning 1 if no wifi, 0 to enable wifi).
> Here's a script I wrote before I found a kernel module for asus laptops:

You know, this is just pathetic.  I've about had it with Intel and lack
of Linux support, or rather, the false claim of major Linux support.  I
mean, they had a proprietary "management" init script and module for
their proprietary ipw3945 that handled the rfkill switch.

Yet they aren't managing it in the iwlwifi drivers out-of-the-box very
well.  You'd figure this would be primo dos to support in the driver.
Maybe I'm missing something, and that would be on me, but even
NDISWrapper gets that correct without any other user-space support.

Anyhoo, here's mine:  

$ ls -il /sys/bus/pci/drivers/iwl4965/0000:04:00.0/rfkill/rfkill0/
total 0
8816 -rw-r--r-- 1 root root 4096 2008-09-06 10:00 claim
8812 lrwxrwxrwx 1 root root    0 2008-09-06 10:00 device -> ../../../0000:04:00.0
8813 -r--r--r-- 1 root root 4096 2008-09-06 10:00 name
8817 drwxr-xr-x 2 root root    0 2008-09-06 10:00 power
8815 -rw-r--r-- 1 root root 4096 2008-09-06 10:00 state
8810 lrwxrwxrwx 1 root root    0 2008-09-06 10:00 subsystem -> ../../../../../../class/rfkill
8814 -r--r--r-- 1 root root 4096 2008-09-06 10:00 type
8809 -rw-r--r-- 1 root root 4096 2008-09-06 10:00 uevent

Now I've just gotta figure out what some of the less obvious values are
(like state=2):  

$ for i in `find . -mount -type f` ; do echo $i ; cat $i ; done
 ./uevent
RFKILL_NAME=5100AGN
RFKILL_TYPE=wlan
RFKILL_STATE=2
./name
5100AGN
./type
wlan
./state
2
./claim
0
./power/wakeup
(null)

> [code]
> #!/bin/bash
> killswitch=/sys/bus/pci/drivers/iwl4965/000\:04\:00.0/rf_kill
> case `cat $killswitch` in
>    0)
>    /etc/init.d/net.wlan0 stop
>    echo '1' > $killswitch
>    ;;
>    1)
>    echo '0' > $killswitch
>    /etc/init.d/net.wlan0 start
>    ;;
> esac
> [/code]

Again, I've read many things around the iwlwifi driver (virtually
nothing has been on the 5100/5300 series other than, "it works"), and I
don't see any additional details on rfkill support.

> In my case, I found that loading the "asus_laptop" kernel module sets this
> to 0 as soon as it loads, and enables the button to toggle it (via ACPI
> messages). Probably doesn't do the same on yours, but if we're just
> talking about the switch, a simple script can set it to 0 or 1 on demand.

Well, I'm thinking of trying state=0 and state=2, although I'm not quite
sure.

> I'm beginning to look more at game developers than vista on some of
> these... some of these games just plain don't work. Heck, some patches are
> being released that actually screw more things up than they fix (I just
> tried a neverwinter nights 2 patch, going from version 12->13, and now all
> the graphics are screwed up, where they were fine before). I have yet to
> see any DX-10 enhancements myself, though... bioshock on my laptop looks
> just like bioshock on my desktop (with a 7900 GTX on DX-9).

Well, I think it's the XP v. Vista issues, and development houses are
going to side with the XP implementation because it's more popular.  In
fact, I think many times when they patch something for XP, it makes
Vista worse.

As I said, stupid me for even trying to game under Windows.  ;)


-- 
Bryan J  Smith              Professional, Technical Annoyance
mailto:b.j.smith at ieee.org  http://www.linkedin.com/in/bjsmith
-------------------------------------------------------------
           Fission Power:  An Inconvenient Solution


-- 
This message has been scanned for viruses and
dangerous content by MailScanner, and is
believed to be clean.