BCM5754/M,BCM5755/M,BCM5787/M,BCM5906/M NIC driver for 10.6 (32/64) [update 22/5]

[adlan]

Originally Posted by scritchy I have device 14e4:167b Broadcomm 5755 onboard nic on an hp xw6600 machine. The kext appears to fail to do anything at all. There is no mention of it in any of the logs in var/log i've tried deleting and reinstalling the kext with both kext helper b7 and kext wizard, and rebuilding kext cache. I've tried using the zip file attachee to this thread, and rebuilding the kext from the source. No difference. I verified the device id w/ dpci manager. Not sure what else to try? it "works" with the hacked 5701 kext floating around, in that it gets an ip and can connect, but the hacked 5701 has horrible upload problems that make it incredibly slow or times out, which is why I arrived here. Any ideas?

try as what Astarael said, try loading the kext manually, and look for any fail message logged. if the kext is loading, it should at least log a build date/time message.

actually there has been a report on the same behavior as this. i'm still investigating what causes this problem and most likely come to a dead end.

btw, do you have a vanilla or distro install?

[Astarael]

I didn't wan to say it to early, but now it seems to be fine... Somehow :P
The kext is loaded since about 24 hours and slow-down event didn't occur. At least so far
While searching BCM in Console I got more results now. These is all I got since I have loaded the kext.

11-04-18 18:13:12	sudo[255]	astarael : TTY=ttys000 ; PWD=/Users/astarael ; USER=root ; COMMAND=/sbin/kextload /System/Library/Extensions/BCM5722D.kext/
11-04-18 18:13:13	kernel	BCM5722D (Build date/time: Mar 25 2011 23:28:49)
11-04-18 18:13:14	kernel	BCM5722D (resetAdapter:145): 4B657654 to B49A89AB in 0 iterations
11-04-18 18:13:14	kernel	my_name_adlan_BCM5722D: Model: BCM5787M NetLink (TM) Gigabit Ethernet
11-04-18 18:13:14	kernel	my_name_adlan_BCM5722D: Loaded successfully
11-04-18 18:13:14	kernel	my_name_adlan_BCM5722D: Ethernet address 00:1e:37:24:13:81
11-04-18 18:13:14	kernel	BCM5722D (resetAdapter:145): 4B657654 to B49A89AB in 0 iterations
11-04-18 18:13:14	kernel	BCM5722D (setMedium:367): Change medium: kIOMediumEthernetAuto
11-04-18 18:13:14	kernel	BCM5722D (setMedium:392): Change medium: kLinkDuplexNegotiate
11-04-18 18:13:14	kernel	BCM5722D (configureLinkAdvertisement:567): advertiseFe: 5E1
11-04-18 18:13:14	kernel	BCM5722D (configureLinkAdvertisement:568): advertiseGe: 300
11-04-18 18:13:14	kernel	BCM5722D (startAutoNegotiation:590): Adv reg: 5E1
11-04-18 22:04:02	kernel	BCM5722D (resetAdapter:145): 4B657654 to B49A89AB in 0 iterations
11-04-18 22:04:02	kernel	BCM5722D (setPowerState:84): Changing power state from 1 to 0
11-04-19 06:01:56	kernel	BCM5722D (setPowerState:84): Changing power state from 0 to 1
11-04-19 06:01:57	kernel	BCM5722D (resetAdapter:145): 4B657654 to B49A89AB in 0 iterations
11-04-19 06:01:57	kernel	BCM5722D (setMedium:367): Change medium: kIOMediumEthernetAuto
11-04-19 06:01:57	kernel	BCM5722D (setMedium:392): Change medium: kLinkDuplexNegotiate
11-04-19 06:01:57	kernel	BCM5722D (configureLinkAdvertisement:567): advertiseFe: 5E1
11-04-19 06:01:57	kernel	BCM5722D (configureLinkAdvertisement:568): advertiseGe: 300
11-04-19 06:01:57	kernel	BCM5722D (startAutoNegotiation:590): Adv reg: 5E1
11-04-19 14:39:59	kernel	BCM5722D (resetAdapter:145): 4B657654 to B49A89AB in 0 iterations
11-04-19 14:39:59	kernel	BCM5722D (setPowerState:84): Changing power state from 1 to 0
11-04-19 14:39:59	kernel	BCM5722D (setPowerState:84): Changing power state from 0 to 1
11-04-19 14:39:59	kernel	BCM5722D (resetAdapter:145): 4B657654 to B49A89AB in 0 iterations
11-04-19 14:39:59	kernel	BCM5722D (setMedium:367): Change medium: kIOMediumEthernetAuto
11-04-19 14:39:59	kernel	BCM5722D (setMedium:392): Change medium: kLinkDuplexNegotiate
11-04-19 14:39:59	kernel	BCM5722D (configureLinkAdvertisement:567): advertiseFe: 5E1
11-04-19 14:39:59	kernel	BCM5722D (configureLinkAdvertisement:568): advertiseGe: 300
11-04-19 14:39:59	kernel	BCM5722D (startAutoNegotiation:590): Adv reg: 5E1
11-04-19 14:52:49	kernel	BCM5722D (resetAdapter:145): 4B657654 to B49A89AB in 0 iterations
11-04-19 14:52:49	kernel	BCM5722D (setPowerState:84): Changing power state from 1 to 0
11-04-19 14:53:03	kernel	BCM5722D (setPowerState:84): Changing power state from 0 to 1
11-04-19 14:53:03	kernel	BCM5722D (resetAdapter:145): 4B657654 to B49A89AB in 0 iterations
11-04-19 14:53:03	kernel	BCM5722D (setMedium:367): Change medium: kIOMediumEthernetAuto
11-04-19 14:53:03	kernel	BCM5722D (setMedium:392): Change medium: kLinkDuplexNegotiate
11-04-19 14:53:03	kernel	BCM5722D (configureLinkAdvertisement:567): advertiseFe: 5E1
11-04-19 14:53:03	kernel	BCM5722D (configureLinkAdvertisement:568): advertiseGe: 300
11-04-19 14:53:03	kernel	BCM5722D (startAutoNegotiation:590): Adv reg: 5E1

PS I hardly ever connect via ethernet, but I tested the drive before and connection was fine. Now I see controller in ifconfig output and system profile so I think it shall work too. Anyway, I will do some tests once I get back home for Easter.

[adlan]

Originally Posted by Astarael I didn't wan to say it to early, but now it seems to be fine... Somehow :P The kext is loaded since about 24 hours and slow-down event didn't occur. At least so far While searching BCM in Console I got more results now. These is all I got since I have loaded the kext. 11-04-18 18:13:12 sudo[255] astarael : TTY=ttys000 ; PWD=/Users/astarael ; USER=root ; COMMAND=/sbin/kextload /System/Library/Extensions/BCM5722D.kext/ 11-04-18 18:13:13 kernel BCM5722D (Build date/time: Mar 25 2011 23:28:49) 11-04-18 18:13:14 kernel BCM5722D (resetAdapter:145): 4B657654 to B49A89AB in 0 iterations 11-04-18 18:13:14 kernel my_name_adlan_BCM5722D: Model: BCM5787M NetLink (TM) Gigabit Ethernet 11-04-18 18:13:14 kernel my_name_adlan_BCM5722D: Loaded successfully 11-04-18 18:13:14 kernel my_name_adlan_BCM5722D: Ethernet address 00:1e:37:24:13:81 11-04-18 18:13:14 kernel BCM5722D (resetAdapter:145): 4B657654 to B49A89AB in 0 iterations 11-04-18 18:13:14 kernel BCM5722D (setMedium:367): Change medium: kIOMediumEthernetAuto .. PS I hardly ever connect via ethernet, but I tested the drive before and connection was fine. Now I see controller in ifconfig output and system profile so I think it shall work too. Anyway, I will do some tests once I get back home for Easter.

that's what i want to hear!

..
11-04-19 14:39:59	kernel	BCM5722D (setPowerState:84): Changing power state from 1 to 0
11-04-19 14:39:59	kernel	BCM5722D (setPowerState:84): Changing power state from 0 to 1
..
11-04-19 14:52:49	kernel	BCM5722D (setPowerState:84): Changing power state from 1 to 0
11-04-19 14:53:03	kernel	BCM5722D (setPowerState:84): Changing power state from 0 to 1

regarding the log message, i'm quite curious of what is happening.. seeing a lot of sudden power state changes and didn't see any link state announcement.

anyway, thank you for testing

[Astarael]

I think these power state changes are here because I used sleep a few times. So no need to worry, also your kext doesn't break sleep

[ronin510]

I've been using your kext for my Lenovo S10 (with a BCM5906). So far, it's been working great!

On your "not implemented" list you have:
BCM5906/M - Isochronous Send Ring

I was wondering what that actually means for us BCM56906 users? Do we lose any common feature/functionality?

Also, I get 6MB/sec when transferring data from the ethernet connection. What do other people get?

[adlan]

Originally Posted by Astarael I think these power state changes are here because I used sleep a few times. So no need to worry, also your kext doesn't break sleep

the small time difference between on and off state is what made me curious.
since you didn't have any problems with it, i'll just ignore the curiosity

[adlan]

Originally Posted by ronin510 I've been using your kext for my Lenovo S10 (with a BCM5906). So far, it's been working great! On your "not implemented" list you have: BCM5906/M - Isochronous Send Ring I was wondering what that actually means for us BCM56906 users? Do we lose any common feature/functionality? Also, I get 6MB/sec when transferring data from the ethernet connection. What do other people get?

hey, glad to hear that!

the isochronous send ring is just a performance enhancement on the transmit side, it's great for data streaming. I don't remember stumbling upon that feature implementation while reading tg3(linux) and if_bge(*bsd) source code though (maybe I missed it, but I'm quite sure about it).

I don't know how much other people got since nobody put it up and I didn't ask for it either. Although I did do some benchmarking using tools like netperf and iperf. Generally I can say its performance is on par with the available driver with only slight difference. If you want the result, I can post it later as I need to dive deep into the disk looking for it .

[michalskie]

i have a lenovo g530

i installed the kext and got a kernel panic .

[adlan]

Originally Posted by michalskie i have a lenovo g530 i installed the kext and got a kernel panic .

hi, sorry for the delay

can you provide me your
- card model
- method of os installation (vanilla/distro(hazard,iatkos,etc)

[derekj0571]

Your kext works flawlessly (So Far)
I just add the HID for the BCM 57778 (1691) and off to the races.
Throughput so far seems good.

If I were you, just add the same to your distro and will include this card.

Oh yes, this is a DVD Retail install of OSX Server 10.6 with Combo Update 10.6.7