Intel Wireless 2100, 2200bg, 2915bg, 3945abg, 4965agn

[jalavoui]

last version as code to fix the en1 bug

the driver is expected to detect witch enx you're using

some msgs in log still show en1 - ignore it

[mlazy]

Jalavoui,

 

I'm stuck here. Cold boot, and the first lines I get about the driver are at line 327:

 

Apr 23 13:51:24 Sony kernel[0]: os Channel 187 [5.2GHz] is restricted -- skipping.

Apr 23 13:51:24 Sony kernel[0]: iwl3945: U iwl3945_init_geos Channel 188 [5.2GHz] is restricted -- skipping.

Apr 23 13:51:24 Sony kernel[0]: iwl3945: U iwl3945_init_geos Channel 189 [5.2GHz] is restricted -- skipping.

Apr 23 13:51:24 Sony kernel[0]: iwl3945: U iwl3945_init_geos Channel 192 [5.2GHz] is restricted -- skipping.

Apr 23 13:51:24 Sony kernel[0]: iwl3945: U iwl3945_init_geos Channel 196 [5.2GHz] is restricted -- skipping.

Apr 23 13:51:24 Sony kernel[0]: e 0x14

 

 

No mesages at all about startup. Attached is a log from the power button to the desktop and then some. I guess someone else needs to test this build.....

 

805.txt

 

 

Gonna go try your last suggestion now.

[jalavoui]

you seem to be letting the driver autoload

if using kextload do you get inital drive log lines?

 

im think the problem is:

Apr 23 13:52:04 Sony kernel[0]: goto system preferences -> networks and press apply if you keep seeing this

maybe you forgot to follow it

this msg is show ever 2 seconds - the driver won't start if enx isn't add to sys prefs - nets

but if you get inital errors on driver load this will also fail

 

anyone else to post logs?

[Miguel Madrid]

I'll not able to test today as I have a family party (My sister birthday)...

 

 

but I think the startup issue could be related to when the kext is load...

 

As I said before, a reboot is needed, because the first time the kext is loaded don't work in the same way as second time...

 

I think we should find the correct way to test, or jalavoui will not be able to get info from the logs.

 

As far as I see it could be:

 

- Update svn, build with xcode. Repair permision on kext

- load the kext

- Goto system preferences and apply (not sure of this)

- reboot

- load the kext againg

 

Other important thing is the harware switch... I think it must be on during the computer bios startup... too many options...

 

I'll read you later.... enjoy the evening...

[jalavoui]

@#$ iwi cards

[Doctadre]

Alright I think jala/mlazy are up to some good here. First off the bat, I think the timer issue is back, I guess its trying to regulate the temp of the card before it can load???. But I did get a few good lines

 

Apr 23 13:20:14 doctadres-mac-pro kernel[0]: iwl3945: U iwl3945_rx_handle r = 7, i = 6, REPLY_ADD_STA, 0x18

Apr 23 13:20:14 doctadres-mac-pro kernel[0]: iwl3945: U iwl3945_rx_reply_add_sta Received REPLY_ADD_STA: 0x01

Apr 23 13:20:14 doctadres-mac-pro kernel[0]: iwl3945: U iwl3945_send_cmd_sync Wait REPLY_ADD_STA Condition: 49ms left

Apr 23 13:20:14 doctadres-mac-pro kernel[0]: iwl3945: U iwl3945_send_add_station REPLY_ADD_STA PASSED

 

Then later I get:

 

Apr 23 13:20:14 doctadres-mac-pro kernel[0]: darwin_iwi3945: Ethernet address 00:19:d2:be:49:86

Apr 23 13:20:14 doctadres-mac-pro kernel[0]: darwin_iwi3945::enable()

Apr 23 13:20:14 doctadres-mac-pro kernel[0]: ifconfig going up

Apr 23 13:20:14 doctadres-mac-pro kernel[0]: outputPacket called by someone

Apr 23 13:20:14 doctadres-mac-pro kernel[0]: outputPaccket2 called

 

Then it tells me thermal timer is not needed, somewhere we need to comment out the calls to these functions. You have two separate functions that may be in conflict with each other. I need a bigger partition any one else game, or am I just completely off the marker?

 

Kext below

systemlog805.txt

[righteye]

Build the Revision 805

First time it freezes my system:

second boot KP

third boot this are the logs:

 

i'll be back tomorrow with more logs.

[mlazy]

Hey,

 

Manual kextload goes right to it as well:

 

Apr 23 14:20:18 Sony sudo[417]: super : TTY=ttys001 ; PWD=/Users/super/Desktop/iwi3945/805/build/Debug ; USER=root ; COMMAND=/sbin/kextload -t iwi3945.kext

Apr 23 14:20:23 Sony kernel[0]: 945/805/compatibility.cpp:6520

Apr 23 14:20:23 Sony kernel[0]: iwl3945: U iwl3945_rx_handle r = 18, i = 17, REPLY_3945_RX, 0x1b

Apr 23 14:20:23 Sony kernel[0]: iwl3945: U iwl3945_rx_reply_rx Rssi -85 noise -127 qual 40 sig_avg 0 noise_diff 0

Apr 23 14:20:23 Sony kernel[0]: ieee80211_rx_irqsafe @ /Users/super/Desktop/iwi3945/805/compatibility.cpp:1841

 

I've got to figure out what I'm doing wrong here. I have a pretty good record testing with you here over the past 3 weeks.

 

manual_kextload.txt

[jalavoui]

at least some conclusion:

- the driver works fine for some cards

- if you don't get a kp them the card fail to start and is doing nothing

- the panic is at ieee80211_scan_completed() this means the driver complets the scan job ok

need to find out why

[Cxddev]

iwi4965-osx5.1 uploaded to svn...

 

jalavoui - can you check it in case I baulked the repository (again!)

 

BTW the version is a bit out of date, but it should build ok.

[jalavoui]

ok

we'll share some bugs here

 

it's ok for building

you can post iwi4965.kext here for testing

first tip:

change ieee_open,checkfirstup

 

i hope you find a way to start scaning - i'm getting crazy about this

 

 

iwi3945 rel 807

this might be the way to scan

only works if you manage to start the driver without errors

if you get a kp - post the picture

[pere]

My log from 805...Always frozen except safe boot, manual load

Safe_boot_manual_load.txt

[righteye]

Logs 806

Guys you move fast again ...

after a KP ... a few reboots and 7 minutes waiting for each reboot ... the driver loaded

have a look when the driver loads ... allot better

 

[pere]

Logs 807.

No freezes.

manual_load.txt

reboot_log.txt

[mlazy]

807 manual kextload: no KP, sees new enet, starts scan.

 

I'll reboot into -x mode and manually load again to confirm results.

 

807 built from Jalavoui's source.

 

iwi3945r807.kext.zip

 

Still getting some errors on my card version. booted -x, manual load. looks interesting, very interesting..

 

807syslog.txt

[pere]

@Javalui: "check if you can get this in log: SCAN_COMPLETE_NOTIFICATION

you only get this if you manage to load the driver without errors"

 

807 revisited... i turned off my laptop to check the card restarting from cold...

Got this:

23/04/08 21:57:57 kernel ieee80211_scan_completed on manual load log. Is this an advance?.

manual_load.txt

reboot_log.txt

[jalavoui]

iwi3945 rel 808

more scan hacks

all this work is to try to get SCAN_COMPLETE_NOTIFICATION in logs

this only works if you get it without errors

maybe i'm missing something - but i'm near to find

[Doctadre]

Damn fast release, here are my 807 logs, didnt look over them much, the card is cooking tho. I get the same things earlier as well as this:

 

Apr 23 15:13:58 doctadres-mac-pro kernel[0]: iwl3945: U iwl3945_alive_start RFKILL status: 0x1

Apr 23 15:13:58 doctadres-mac-pro kernel[0]: iwl3945: U iwl3945_init_channel_map Channel map already initialized.

Apr 23 15:13:58 doctadres-mac-pro kernel[0]: iwl3945: U iwl3945_init_geos Geography modes already initialized

 

Soon after I get this error:

 

Apr 23 15:13:58 doctadres-mac-pro kernel[0]: error iwl3945: Microcode HW error detected. Restarting.

 

and then

 

Apr 23 15:13:58 doctadres-mac-pro kernel[0]: iwl3945: U iwl3945_irq_handle_error Restarting adapter due to uCode error.

 

This seems to loop and every time the temp of the card seems to drive up. Gonna try to boot 808, I wont have time to comb it tho, I have a few patients i need to check up on

systemlog807.txt

[mlazy]

808:

 

iwi3945r808.kext.zip

 

Jalvoui, what does this error mean right when I kextload:

 

Apr 23 16:26:09 Sony sudo[374]: super : TTY=ttys000 ; PWD=/Users/super/Desktop/iwi3945/808/build/Debug ; USER=root ; COMMAND=/sbin/kextload -t iwi3945.kext

Apr 23 16:26:14 Sony kernel[0]: 5_rx_handle r = 3, i = 2, REPLY_ERROR, 0x02

Apr 23 16:26:14 Sony kernel[0]: error iwl3945: Error Reply type 0x00000000 cmd UNKNOWN (0x3C) seq 0x022C ser 0x0000003C

[pere]

Logs for 808...

No SCAN_COMPLETE_NOTIFICATION...

manual_load.txt

Reboot_log.txt

[mlazy]

808 logs:

 

SSHHHHH... don't tell Jalavoui:

 

Apr 23 16:35:00 Sony kernel[0]: iwl3945: U iwl3945_bg_scan_completed SCAN complete scan

 

!!!!!!!!

 

 

808syslog.txt

[righteye]

Revision 807

Yes ... you really fast

No matter what ... i get same logs on this revision

I was playing with the Fn+F2 ... you can see that.

 

[jalavoui]

iwi3945 rel 809

the driver needs to be configure before scan

let's see what logs show from this

the hacks are in check_firstup()

 

can someone put iwi2200.dmg from svn in iwidarwin downloads?

also need a working iwi2200.dmg for os x 10.5

 

i have one more release to post

waiting for logs...

[Ruud]

to cxddev, I just build the iwi4965.kext, installed and all went ok, en1 showd in network.

I can see the errors now in system.log, i will now figure out how to create good log files.

 

No Kp's after install

 

update: hope these logs are ok

dmesg.txt

ioreg.txt

system.txt

[adamisnt]

Hi all,

 

I have been lurking for a while, but have been unable to help testing since I have a 4965 card. It looks like a version has been submitted, so I would like to help.

 

I have plenty of experience developing in windows (.NET), but unfortunately nothing on the mac. Could somebody repoint me to the post explaining how to test and how to create logs - I would really appreciate it. I have been looking for it for the last 30 minutes.

 

Jalavoui - you are the man. I can't believe you are doing all of this without even owning the card. Much respect to you.