Jump to content

Intel Wireless driver

jalavoui

By jalavoui,
in LAN and Wireless

 Share
3992 posts in this topic

Recommended Posts

jalavoui
jalavoui
Posted
  • Author
Posted
when pressing the combination Fn+F2 (where is located the antenna), the load was diminished to normal without any hang up of the system.

looks like you manage to turn the radio on :P

maybe others can try the same

can you post the log with more on/off retries of networkselector?

 

it's better to check the en(x) - most computers only have 2 network cards, so the device for the wireless card should be en1.

 

Sep 3 18:33:19 EvilCartman kernel[0]: iwi3945: RFKILL base status: 0x0

go into windows, power the card on and reboot to mac os

if you get:RFKILL base status: 0x1

it will mean the card is powered on

the driver loads better this way

i need to find howto hack this deam thing...

auqs10
auqs10
Posted
Posted

Hi Jalavoui, here are the logs archived as ZIP files (before NS, exactly after logs cleaning and restart, also those recorded after several turnings on/off with NS coupled with Fn+F2). What I have observed is that the CPU load to 100% is achieved each time the Network Selector activates option 1 (Turning on/off). When turning off the card the load falls to 3-5% (normal mode) imediatly after pressing Fn+F2. Instead, when turning on the card the load falls to ~50% and as you can see in the screen capture attached, the Network Selector loads the CPU.I forgot to tell that takes about 20 sec till automatically CPU goes to a normal load about 3-5%, when Network Selector with option 1 turns on the card.

Evil_Cartman
Evil_Cartman
Posted
Posted
go into windows, power the card on and reboot to mac os

if you get:RFKILL base status: 0x1

it will mean the card is powered on

the driver loads better this way

You know, my notebook sets wireless adapter to off on each boot. So after logging to windows I must press h/w switch to make wifi working, but reboot sets it back to off. Windows driver itself cannot to turn adapter on.

 

Right now I've pressed h/w switch once on the mac os boot and there is no any changes - networkSelector on first launch shows option (1) as "Turn card off" - looks like card is on. But when I try to "Turn card on" it stuck with 100% cpu as I've described above. Can't do anything better with rev 502, sorry.

jalavoui
jalavoui
Posted
  • Author
Posted

iwi3945 rev

this is a scan send command test. it should work better for cards who can get rfkill= 0x1

if rewrite the radio hack to not hang up - i hope ;)

rename this file from .zip to .dmg

 

 

Evil_Cartman,

don't press the swicth too soon. wait for mac os to boot and iwi3945 driver to finish launch before press it.

iwi3945.dmg

Evil_Cartman
Evil_Cartman
Posted
Posted
Evil_Cartman,

don't press the swicth too soon. wait for mac os to boot and iwi3945 driver to finish launch before press it.

Think I've found proper way to get rfkill=0x1

I'm pressing h/w switch after networkSelector starts - is that exactly what you mean telling me about driver launch? ;)

Anyway, your nightly build stops hanging my computer for any rfkill value, and that is the best news for me.

dmesg.txt

ioreg.txt

system.txt

dragonomous
dragonomous
Posted
Posted

I'm kinda new getting OS X on an Intel machine, but I generally have success in figuring out how to get the hardware I need installed. I'm currently working on an HP Pavilion dv6000 series laptop and of course, I can't get the Intel PRO/Wireless 3945a/b/g card to work. I've tried installing the patched BIOS update and I've tried installing iwi3945.

 

The BIOS update didn't do anything noticeable except made my mouse freakout randomly.

 

iwi3945 got my card to be recognized with the correct MAC address, but it was not recognized as an Airport card and when using the network selector tool, nothing seemed to work. I'd select an option and it wouldn't do anything.

 

I know lots of these posts all say the same, "I can't get it to work" but I can't seem to find any other possible solutions. Let me know if you need any information and thanks for your help guys!

kevinosx
kevinosx
Posted
Posted
I know lots of these posts all say the same, "I can't get it to work" but I can't seem to find any other possible solutions. Let me know if you need any information and thanks for your help guys!

I know the feeling, I've been waiting months for these wonderful gentlemen to finish hacking the driver (great work by the way)

 

The driver is not working as of yet so there is no way to get it to "work" until they are finished, trust me, you will know when it's done.

jalavoui
jalavoui
Posted
  • Author
Posted

iwi3945 rev 503

http://code.google.com/p/iwidarwin/

- test it with rfkill=0x1 - can someone explain this to other iwi3945 users?

- firmware updated to iwlwifi-3945-ucode-2.14.1.5

- rx_on fix

antoni as send me a Automator script for iwi3945. can you post it here and explain?

i can use a hand on the rfkill hack. i don't have this card, so if someone can find a solutiuon please help.

 

iwi4965

- i received this log from linux. i'll put it here so others can use it

 

 

 

about logs:

 

before posting logs look at them first.

a good log should start with this line:

 

iwi3945: disable 0 led 1 mode 0

 

because this is the point when the driver is first loaded and it's much easy for developers to understand what the driver is doing

 

the getlogs.command might fail to get this if the system.log is too short. to fix this:

- open console, click "logs"

- select the system.log file

- if you don't find the first log line, use reload and choose a diferent file size until you get it

- copy the log from console and paste it in textedit

- post the "good" log file here

 

i'm unable to post pictures, so check this example.

 

thanks

 

 

justjoe,

can you modify the getogs.command to build logs only if the "first line" exists?

linux_4965_log.txt

good_logs.rtfd.zip

auqs10
auqs10
Posted
Posted

Hi again, still not working, but the 503 version has something solved about loading the CPU. When turning on the card about 30 sec the Network Selector loads around 50% the CPU then goes down to 3-5% as normaly. Here are the logs after several turnings on and off (even with shotcuts Fn+F2).

Evil_Cartman
Evil_Cartman
Posted
Posted
iwi3945 rev 503

...

justjoe,

can you modify the getogs.command to build logs only if the "first line" exists?

What should I do if my logs do not contains that "first line"?

I've searched over and over and there is definitely no such line as you described.

Here's my logs from current boot.

dmesg.txt

ioreg.txt

system.txt

jalavoui
jalavoui
Posted
  • Author
Posted

iwi4965 rev 504

changed old code from iwi3945 - this should allow the driver to do something.

 

iwi3945 rev 505

changed ipw_sw_reset.

try this release first from http://code.google.com/p/iwidarwin/

 

after testing it try this one - it includes a old firmware version and can be usefull to find out why interrupts are missing. rename the .zip to .dmg

 

please post separate logs for each version.

iwi3945.dmg

auqs10
auqs10
Posted
Posted

Jalavoui, unfortunately there are only bad news with both versions 505 and the old one uploaded above in your post. The both versions give panic error on system launch exactly after the two lines concerning the ucode

iwi3945: SW RF KILL supported in EEPROM

iwi3945: HW RF KILL supported in EEPROM

panic (cpu0 ..........

Thank you so much for all your efforts to solve this issue.

masterzeta
masterzeta
Posted
Posted

Hi there, I have an 2100 card and I cant get it to work, everytime I boot I get a kernel panic unless I unistall the driver, I know that some people got this card runnig and I wonder what do I need to change. I'm using a 8.8.1 kernel in osx 10.4.8.

 

By booting in verbose mode i see that the card gets recognized and everything, but just after iwi2100: card MAC address is (mycardmac) I got a kernel panic.

 

Also the messages seem to enter a loop before the kernel panic, a loop from tablezise1 to f/w initialization conplete: succes. this repeats at least twice and then I get the kernel panic.

 

I would like to know if the current driver is broken and if so, where I can get an older release.

 

Also, if you need extended logs just tell me how to get them and I'll post them. I would like to contribute with the driver's development, at least as a guinea pig :D:P

jalavoui
jalavoui
Posted
  • Author
Posted

iwi3945 rev 506

rx_on function fixes.

 

about kernel panics:

if you get it, reboot, press F8

type -stype sh/etc/rc this will load the driver

take a picture of the screen and post it here

 

thanks

 

iwi3945: SW RF KILL supported in EEPROMiwi3945: HW RF KILL supported in EEPROM

 

i've add debug msgs for this in this release

 

iwi3945 rev 507

same as 506 with more debug for nic_init()

auqs10
auqs10
Posted
Posted

Hi Jalavoui, as I told you I have downloaded Kubuntu 7.04 Feisty Live & Install CD. In fact, this message is sent using Konqueror Web Browser launched with active Intel WireLess 3945a/b/g under live system booted from CD. What should I look for to help you solve the driver for OSX?I will try though the new version immediately.

Evil_Cartman
Evil_Cartman
Posted
Posted

Excuse me for 2-day silence - had to (re)compile damn dvd project.

There are two more lines before the panic. I will try to get a screen capture, because the logs are empty.

I have the same errors with rev 507 and I wrote down last log lines :tomato:

SW RF KILL supported in EEPROM
HW RF KILL supported in EEPROM
ipw_rx_queue_alloc
ipw_rx_queue_replenish

- kernel panic here.

Guest
This topic is now closed to further replies.
 Share
Go to topic listing
×
×
  • Create New...