Jump to content
3411 posts in this topic

Recommended Posts

:wacko: I think im sinking deeper into his hole! After doing something (randomly trying to delete the file) I somehow managed to get the error boot.plist is not found.......... oh sh*t! how do i get it to rewrite? I don't have the installation cd with me now......
I'll get working on the installation package, give me around an hour from this post.

edit: the kext/networkSelector/nsGUI.app in the latest installation package show a last-modified date of April 21st.

the most recent modification uploaded to the SVN on iwi2200.cpp was on April 17th.

I don't have a 2200 to test, but the installation package in the disk image file on the SVN appears to be up to date.

 

Thanks devoid. Unfortuantely the latest iwi2200 svn DMG that's been there for a while isn't working, and I can't find any earlier versions to revert to. Basically when trying to choose a certain SSID to connect to, it will either randomly start cycling through multiple SSID's (including ones I did not select(?)), or appear to connect but not obtain a DHCP address and no network connectivity. Manually setting the IP in this case doesn't work either. This driver was working well previously, so I'm not sure what happened, something seems to have gotten messed up... I will try to post some logs later tonight when I can get back to my 'tosh if that will help.

 

EDIT: This is only the 10.5 version I am referring to FYI...

@irBOrNE

read the 1st post and build iwi2200.dmg

i'll help you if you can't

 

last log is the worst i've ever seen

at leat look at linux log to know what to search for

can someone post a good tested log for this

the driver is sending a reassoc response

Javalui, I have tried the last 3 versions, and i dont post log cause i cant find anything useful on them.

 

Is supossed we have to try the card starts correctly and then get a kp, post here the pic so you can see the details, is that right?

 

Everytime i can get my card start correctly, it just freeze, dont give any KP :angel:

And when the card start incorrectly, then gives the famous SYS ASSERT.

@irBOrNE

read the 1st post and build iwi2200.dmg

i'll help you if you can't

 

 

OK, I would like to give it a try tonight or tomorrow when I am back at my Leo machine and have some time to walk through learning this procedure (especially now that we have the great tutorial from devoid!) Otherwise, if someone has these tools already installed and would like to go for it ahead of me, feel free to do so, as I don't want to keep others waiting..

900 log. Regular boot, manual kexload. Correct me if I'mwrong, but it looks like it loads well, gets a Microcode HW error, reboots, gets a UCode error, then loops?

 

Going to go test a few more times...

 

900.txt

 

Update: -x boot gives me immediate kp on kextload.

when the driver gets this

IWL_DEBUG_INFO("ALIVE processing complete.\n");

it's ready for 1st scan

this call fires up also after using the switch off/on

i'll add a scan call to this and upload to svn

 

iwi3945 rel 902

iwi4965 rel 903

here are logs from rev903, wiw4965. I have them upon install and restart. No KP but it took a while to start the OS up on restart. It loaded only after like 5 minutes of text for the 4965 scrolling across the screen. It was too fast, but attached is the copy of system.log from the reboot. I honestly don't know how to read this stuff, so I appologize if the log is no good. Hopefully it will help though

 

also, i got an error when compiling the kext. Did anyone else? And if you didn't can you post your kext and ill retest using it?

log_rev903.rtf

log_rev903_after_system_reboot.zip

another 3945 linux log

 

mac80211-10.0.4

iwlwifi-3945-ucode-2.14.1.5

iwlwifi-1.2.25

 

this is very interesting the scan loops with a 20 seconds delay - nop seems to be manual?

can you post this log with associate/disassociate/rfkill actions

 

iwi3945/4965

need log with lines from driver startup

 

scan loop

iwi3945 rel 904

iwi4965 rel 905

 

pere,

can you try other values for iosleep hack? might make the driver more stable

Katched a KP finally.

 

Iwi 3945 last version.

 

Log from reboot. (Have ucode errors).

 

EDIT:

Going to try the iosleep hacks.

 

 

EDIT 2:

 

After raise the iosleep to 4000, get a little bit more, stable, still have time outs on transmit rx on.

 

Log attached.

post-98153-1209596969_thumb.jpg

reboot.txt

Reboot_iosleep_4000.txt

iwi3945 rel 902No KP of freezes, but I take error "en1 not running", and no 1st scan init.iwi3945 rel 904First 2 reboots -> freezes, then norm. Still have "en1 not running", it's something with "if (!netif_running(dev))"? BTW, I take an error when building kexts (902, 904), shot is below. Some script?

system.txt.zip

system.txt.zip

post-153393-1209600330_thumb.jpg

iwi3945 rel 902

No KP of freezes, but I take error "en1 not running", and no 1st scan init.

 

yep, you're right

let's fix it

 

iwi3945 rel 906

iwi4965 rel 907

 

last version for today

 

pere,

i change the call to ieeeopen

this can change driver stability - the drive will only start loading after call enable()

can you test? - i mean hack <_<

i think if you hack the way the driver calls ieeeopen (eg. split commands) you get more stability

 

the password bug is because the "sudo" commands don't work in xcode for 10.5 - try to use your root account

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