Uh.. Its not loading the kext on startup, so its not showing up in dmesg...

dmesg will show the kernel message buffer. When the message appears has nothing to do with whether dmesg will show it or not.

More specific instructions:

1. Use kextunload to unload any old version of forcedeth that was loaded on startup.
2. Use kextload to load the new version of forcedeth
3. Check dmesg.

Alright... here's my dmesg. I still don't see anything about forcedeth in there, but whatever.

Don't think this was taken right after the kextload happened. For example, here's what I get.



It's also possible that for some reason, control was never passed to the driver at all. Perhaps another driver for the device was already loaded or something.

I dont get any sort of output like that after I do a kextload :-/

No previous version of forcedeth was loaded? You do have a en0 device present. If no ethernet driver was present, that ought not exist.

I dont see any versions of forcedeth in kextstat.. and I dont see an en0 in ifconfig, although i do see an en0 in System Profiler.. X__x

Ok... when I add my device ID to forcedeth-nockd, it panics.. :-/

EDIT: I was just looking through console and saw that right before OS X panic ed, it actually started to load forcedeth-nocked properly

Fun. -.-

Hmm. Try this version. You'll still need to add in your device id.

Well, I added my devID and it seems to have gotten me online. I've got a green light now for my ethernet in System Preferences and I'm able to load my homepage and the weather widget. I'm gonna go give it a stress-test here..

EDIT: Tried going from Digg.com over to Apple.com/downloads. Got a kernel panic about half way through loading the downloads page... hmm..

EDITx2: Seems to be working now. I was able to download about 75MB worth of software updates without problems... I'll keep an eye on it though..

I feel like an optometrist. Is this better or worse?

I installed it... so far so good... I'll post updates as things progress.

It says "NIC ring full, stalling" in console. I haven't noticed any performance change though. Dunno if this is a bad thing or not..

I think we might've hit the hot spot. I've been able to stream a Stevenote without any problems so far.

Yeah, the last version, I put the disable interrupt voodoo back into doNicPoll(). It's unclear why this is needed, as interrupts ought to be serialized with that function anyway.

So the current list of changes are:

- Changed IOBasicOutputQueue to IOGatedOutputQueue to serialize tx with everything else
- Changed the way the IOWorkLoop is obtained: We now create our own explicitly, and this is now set as the work loop IONetworkController commands use. (previously, they may have been on separate work loops!)

The real question is whether this works better than the really old version... does it?

P.S.: The NIC ring full, stalling thing should be all right. Basically, the amount of packets that are being queued for sending is swamping the buffer of the NIC, so it's telling the OS not to bother it with more requests until there's more room in the buffer.

Much better.

EDIT: Whaaaaaa. Kernel panic when trying to download VLC. x____X

I spoke too soon...

So would you characterize it as a reduction in panics, or just about the same?

Definately less, but they still occur.

Hmm. I believe it's no longer possible to have a race condition in the code, but I could be wrong. I will write a debug version that will make it possible to check for any race conditions. If there are none, then it's probably some other unidentified device driver bug, which would be difficult to fix.

This version ought to report with "such-and-such function violated mutex" if a race condition occurs. Try it and see if any such messages appear.

I'm kinda late to the game, but I've got nothing here...

Downloaded the driver from the first post, copied via usb stick, extracted, chown, chmod, kextload.

I get back that the driver is loaded, but no messages in system.log, only the sudo commands...

From Device Manager in Windows Vista -
NVIDIA nForce Networking Controller
PCI\VEN_10DE&DEV_03EF&SUBSYS_2A5B103C&REV_A2

Any on have any ideas?

Thanks
Cash

It's not directly supported, but you can add the device id to the proper plist. Instructions ought to already be posted somewhere in this thread.

very nice work! thanks for your contributions! i'll give this a shot when wireless is no longer an option

Hi there! I did not know English well (greetings from Russia!), thats why it is difficult to read all topic. And i try to explain my problem.
I download the attached file and follow the steps in instruction.
But only what i can see in terminal, was:
nForce 4 LAN
sending 1 personality to the kernel
Pages on local server dose not loaded. And in a Network panel no Ethernet shows up. If i run diagnostics - near the Ethernet i see red button.
My motherboard is Asus A8N-VM. Information from Device Manager about my networking controller:
{1A3E09BE-1E45-494B-9174-D7385B45BBF5}\NVNET_DEV0269\4&26A698C5&0&01
{1a3e09be-1e45-494b-9174-d7385b45bbf5}\NVNET_DEV0269

Hello there.

planetbeing: I really appreciate your work. Since you included the source-code I was able to recompile the drivers for Leopard. Actually your first one (from the beginning of the thread) kinda worked.
I have a MCP61 chipset (forcedeth 03EF) and it was detected properly by "forcedeth-nock.kext". I was also able to surf the net for a while but whenever pages exceed a size of about 1.5MB my system freezes (without special messages in dmesg by the debug build ).
I would like to try your newer versions, but I always get:



when I try to load the driver. I've also tried recompiling, but this time it doesn't do the trick. On my card "forcedeth-nock.kext" has to be loaded first, otherwise it won't work.

Any ideas why I can't load this extension? The other ones load just fine, but don't really do anything without nock.kext...

Regards,
mcsmart

I have the same message, when try to "kextload forcedeth-nock.kext"

Ok, i find working nock, and now i can see adapter. But he receive bad ip adress. Not like windows. Where is the problem ?

well, I tested the driver and look what I found:

I use a eth hub to share my adsl connection here but my hub is too old. an 10/10 hub. And when the Ethernet cable is on it, leopard keep says "unplugged.." but, when I put it directly into the adsl modem - eth port -(I believe it´s 10/100) tan tannnn lights turns green on network control panel, and message turns to connected!) now, I wish to know, if there is a chance to make the driver to work under the 10/10 port...! otherwise I´ll have to buy a new hub/switch...

perhaps topic is at vacation... new years and that stuff...

planetbeing... Thanks for all your work. I have been reading through the thread from the beginning. You mention that you upload the latest version to the first post.

Could someone please tell me if that link is for tiger or for leopard.

I am going to install this for a uphuck based 10.4.9 v1.4a build of osx. It will be for a realtek 8109 chipset on a ASUS nForce 430 motherboard.

Thank You

Hi.

I've been reading intently to these posts about the nForce4 LAN not working.

I'm in the process of hackintoshing my Asus PC (spec below) which has an nForce 590 (series 5) chipset.

Is there any possibility this will work?

I heard that an Intel 10/100 PCI card will work out-of-box, but I'd prefer to use my on-board one(s).

When checking the Hardware Id's in Device Properties for the nForce NetController I see this.....

' PCI\VEN_10DE&DEV_0373&SUBSYS_CB841043&REV_A2
PCI\VEN_10DE&DEV_0373&SUBSYS_CB841043&REV_A2
PCI\VEN_10DE&DEV_0373&CC_068000
PCI\VEN_10DE&DEV_0373&CC_068 '

I can only conclude that mt Vendor ID is 10DE and my NIC Device ID is 0373.

Is there any help out there?

ASUS M2NE-SLI DELUXE
ATHLON 64 X2 6400 (AM2)
2GB RAM - 250+320gb HDD's
nForce 590 CHIPSET

Guys,

I don't really have the time to answer individual questions on the forum. You may have noticed that I was working on solving one specific issue: Namely, kernel panics under high load. Only Colonel had volunteered to test and without more testers, the effort has stalled.

As for your questions, I really think they have all been addressed earlier in the thread:

1. If nothing comes up in dmesg after loading the kext, it's simply because the driver doesn't recognize the device id of your NIC. You'll have to edit Info.plist in the kext to insert yours. Instructions are everywhere.

2. If it still does not work, it's just not supported. Sorry. I don't know what's wrong. I don't have your hardware. I don't know how to fix it. I can rewrite the driver, porting from a newer version of forcedeth, but that's not guaranteed to solve your problem. It's also a massive effort and may not ever get started or done.

3. Tiger drivers work fine on Leopard, theoretically. I use Leopard, but I cross-compile the drivers to Tiger. They work fine on Leopard.

Also, I get the impression that most people have gotten it working. The only definite issue I know about is simply that if you dual-boot, Windows may leave the NIC in an uninitializeable state. You might try disconnecting the power to the computer for 15 seconds, then rebooting into OS X only. That can be fixed after the kernel panics are resolved.

mcsmart: I may have forgotten to update the Info.plist for nock properly. The dependencies section of the plist is probably the thing that is screwed up. You can fix it by copying it from the Info.plist of just the normal forcedeth or wait for me to do it.

after one week of tests:

1. the driver works here with zephyroth 10.5.1 amd leopard and asus a8n sli se - nforce 4 lan -

2. it only works when I use a 10/100 ethernet port/hub/link, with my old 10/10 it did not work.

3. good work. it´s working very well here now. (with 10/100 hub)

I just made my nforce nic work with driver http://forum.insanelymac.com/index.php?showtopic=73317.
My mobo is the M2N-Sli Deluxe (nForce 570-SLI) and my NIC has the same ID as yours.
Guess you should give it a go

Hi planetbeing and all,

I am trying to track down the lan freezing problem and dual core processors. I can use the latest debug forcedeth.kext for a while but then I do get a kernel panic. Nothing in the dmesg though. Of course everything works just fine if I set cpus=1 in the Boot.plist but I'd like to get away from that.

So I have a question. What's the difference between the normal forcedeth.kext and the forcedeth-nock.kext?

I surmising that the freezes are occurring because of the dual cores but is there any real way to get around this besides the cpus=1 or another network card. My board is an Asus A8N-E and only has one on board nForce 4 LAN.

I've tried both the debugs for forcedeth and forcedeth-nock and the freezing is more or less hit and miss. It does seem that there needs to be alot of traffic occurring before a panic happens.

Thanks for the driver though, I think it's pretty close to working...If you need anything for debugging etc, let me know, I can help out.

I have a 6150/430 motherboard with a single core cpu. The forcedeth-nockd version is the only one that works on my system, but it sends out so much debug info that I had to recompile it with a few IOLog lines commented out. It seems to be working fine for the most part, but I have run into a problem where severe network stalls will make the driver go into an unknown state. I don't get a kernel panic and the interface remains up. I just lose I/O. This hasn't happened very often. I can get it to happen mostly when I access a network drive that's asleep. I don't have time to debug this myself, but if you want me to try out anything to help with the progress of this driver, please let me know.

I have an Nforce 500 chipset on my M2N-E SLI
It has an "nVIDIA nForce4 SLI (CK8-04) - LAN Controller (PHY: Attansic PHY) PCI"
But forcedeth doesn't work here is my log:



Edit:
I tried nockd now same problem i also find that the other terminal has log so here is that log
See it can't vind PHY

my Nvidia nForce card comes up with
DEV_0269
VEN_10DE
so you can add 0x026910DE to the IOPCIPRIMARYMATCH for future releases.
thanks for you efforts.

ok ive got a ga-m55sli-S4 mobo and the network card is a sli based one o.o

the log i got is as follows

Jan 13 04:15:17 kristie-hollys-computer kernel[0]: forcedeth: Found nForce4 LAN with MAC: 0x:xx:x6:xx:xE:xx.
Jan 13 04:15:17 kristie-hollys-computer kernel[0]: forcedeth: Found PHY 0x5040:0x0003 at address 1.
Jan 13 04:15:17 kristie-hollys-computer kernel[0]: com_triton_forcedeth: Ethernet address 0x:xx:x6:xx:xE:xx.
Jan 13 04:15:17 kristie-hollys-computer kernel[0]: forcedeth: Enabling... 1 2 3 4 5 6 7
Jan 13 04:15:17 kristie-hollys-computer kernel[0]: forcedeth: Starting transmit/receive engines
Jan 13 04:15:17 kristie-hollys-computer kernel[0]: forcedeth: Network link down.

anyone used this with nforce 630a??? i need LAN to work

DevID please?

I have installed the driver properly.But it wont work in PPPOE.This is the log:

Jan 31 07:54:52 s-computer kernel[0]: forcedeth: Enabling... 1 2 3 4 5 6 7
Jan 31 07:54:52 s-computer pppd[1675]: pppd 2.4.2 (Apple version 233-13) started by root, uid 501
Jan 31 07:54:52 s-computer kernel[0]: forcedeth: Starting transmit/receive engines
Jan 31 07:54:52 s-computer kernel[0]: forcedeth: Link speed now 100Mbps, code 0x10064.
Jan 31 07:54:52 s-computer pppd[1675]: PPPoE connecting to service '' [access concentrator '']...\n
Jan 31 07:54:52 s-computer kernel[0]: PPPoE inputdata: unexpected data packet on unit = 3
Jan 31 07:54:52 s-computer pppd[1675]: PPPoE connection established.
Jan 31 07:54:52 s-computer pppd[1675]: Connect: ppp0 <--> socket[34:16]
Jan 31 07:54:52 s-computer pppd[1675]: PPPoE hangup
Jan 31 07:54:52 s-computer pppd[1675]: Connection terminated.
Jan 31 07:54:52 s-computer pppd[1675]: PPPoE disconnecting...\n
Jan 31 07:54:52 s-computer pppd[1675]: PPPoE disconnected\n
Jan 31 07:54:52 s-computer kernel[0]: forcedeth: Disabling
Jan 31 07:54:52 s-computer kernel[0]: forcedeth: Enabling... 1 2 3 4 5 6 7
Jan 31 07:54:52 s-computer pppd[1676]: pppd 2.4.2 (Apple version 233-13) started by root, uid 0
Jan 31 07:54:52 s-computer kernel[0]: forcedeth: Starting transmit/receive engines
Jan 31 07:54:52 s-computer kernel[0]: forcedeth: Link speed now 100Mbps, code 0x10064.
Jan 31 07:54:52 s-computer pppd[1676]: local IP address 10.64.64.64
Jan 31 07:54:52 s-computer pppd[1676]: remote IP address 10.112.112.112
Jan 31 07:55:01 s-computer pppd[1665]: PPPoE connection failed, No route to host
Jan 31 07:55:02 s-computer kernel[0]: skgeosx : checking rx_mbuf:0 tx_mbuf:0
Jan 31 07:55:02 s-computer kernel[0]: skgeosx: Disabled!
Jan 31 07:55:02 s-computer /System/Library/CoreServices/mcxd.app/Contents/MacOS/mcxd: MCXD.initializeDirectoryServices: Unable to get the ethernet address

Seems like it cannot get the ethernet address.I dont know if it is wrong on my driver or on my network card.I wish someone could tell me what is going on.

OK, I know this is a pretty old topic, but i have an m2ne-sli mobo with the NVIDIA nForce 500 chipset. I'm sure that the nForce 4 is just the older name for my chipset. I have everything working for os-x except networking. I tried using these drivers and at the end I got a message stating they where allready loaded. I tried unloading them and doing the process again. I was then told the same thing again.

After checking my logs heres what I got:



It looks like it was all going well until it started looking for a PHY?
Anyone got any ideas on how to get this working or is it a new ethernet card for me?
According to device manager my card has the correct vendor and device number so It should all be ok. The OS-X in question is 10.4.8.

Thanks alot.

I've tried all the various permutations of forcedeth and the NVidianetorking kexts I can find on this forum, but no luck on my system.
I've got XFX 680i SLI mobo.
The dev ID for my onboard NIC is 0373 from 10de(NVidia) which I see several other people seem to have working.
In my tail log, I get enabled, but never see Link=100mbps. It just says Link is down.
I'm using ToH Install of Leopard 10.5.1 with Golden Kernel.
Any ideas?

hi,
i 've got ASUS A8N-E mobo,
and forcedeth works for me great, but...
i had to disable LAN in bios before i installed the forcedeth .
after install, i enable LAN, and net goes run!!!!!!!!!!!!!
hope my advice help someone.

Hi all,

this is my first post, so I apologize for any mistake I might do...

I installed OSX on a dc7600 small form factory with HP mobo and this network adapter "INTEL 82566DM Gigabit Ethernet Controller". After days of patching, I managed to have everything working, but the ethernet...
I tried many patches and the forcedeth seems the one getting most near to success, but when I load the kext, the log I get is not looking as good as should be.

I tried it with both 4 kexts in forcedeth/build/release and 2 couldn't be loaded (erroneous permissions in kext) and 2 gave me the same log results


I attached the log here, if anybody can give me an hint, I'd appreciate very much as I'm getting crazy
.

PS: If I was poor on info and need to post more, let me know, please

hEY PAL, IT DOESN'T LOOK LIKE THAT YOU HAVE A NFORCE LAN!

mmm, thanks...

I have to say after I posted the thing I noticed that it's Intel 8256X family controller, therefore I should go with another kext. (like AppleIntel8255X.kext, which doesn't work btw)

Sorry the fuzz and thanks to all of you!

I've been trying unsuccessfully to track down the kernel panic. Nothing I've tried has helped so far (at least consistently - I swear the panic just doesn't happen sometimes for no reason), but I do have one question: why is there no danger of a race condition in the TX buffer between outputPacket and txDone? That's clearly not the cause of the panic, since adding the locks back in doesn't fix it, but I couldn't see offhand why there wouldn't be a conflict. IRQs should be serialized, but don't outputPacket calls come from the outside world, potentially asynchronously?

Anyhow, I'll keep banging my head against it - nothing like losing a few days of my life waiting for reboots

Oh, got it - service isn't being called asynchronously, is it? Well, that explains that... but not the crash I'm trying to increase the granularity of the IOLogging a little with IOSleeps, since I don't know how trustable IOLogs immediately before a panic are (Apple sure goes out of its way to warn you about them), but nothing concrete yet...

Well, as far as I can tell, it's not an obvious race condition of any sort. Spinlocking everything down, going from basic to gated queuing, putting in IOLogs on any sort of conflict... nothing. But the panic only happens when both cores are on. And it's oddly sensitive to context - it happens almost immediately during a normal ftp, for example, but passive mode seems to make it much less likely, and I can scp to and from machines on my immedate network and on my subnet without a hitch (which involves a ton more throughput than ftping from kernel.org, the test case that always crashes). What's different between ftp and scp at the ethernet driver level, and why would it matter only when there are two cores but no obvious asynchronicity? The plot thickens...

Someone posted a fix for the panic with the latest driver and I can't seem to find it now.. you had to edit something in the plist... it's really bugging me now. >_>

You were testing the driver with planetbeing right? Please try to remember what it is, I'd love to turn my other core back one.