Solution for Qualcomm Atheros AR816x, AR817x and Killer E220x

lgda · October 2, 2014

Hello and thanks for this great works.

I'm pretty new to the "Hackintosh world" so I'm not really an expert (just something to keep in mind to interpret the rest of my post !).

I have a z97x-UD5H motherboard including an Atheros Killer E2201 LAN chip so I tried this driver.

It seems to be working fine but I'm able to crash it (losing network and having to reboot) using the iPerf application with more than 4 threads.

iperf -c achille -P 5 -t 600

It's kind of an extreme situation but I think that's one that can happen to me (TimeMachine + iTunes on a network drive + Movie + ...).

Kernel logs:

Oct  2 20:43:14 localhost kernel[0]: Longterm timer threshold: 1000 ms
Oct  2 20:43:14 localhost kernel[0]: PMAP: PCID enabled
Oct  2 20:43:14 localhost kernel[0]: PMAP: Supervisor Mode Execute Protection enabled
Oct  2 20:43:14 localhost kernel[0]: Darwin Kernel Version 13.4.0: Sun Aug 17 19:50:11 PDT 2014; root:xnu-2422.115.4~1/RELEASE_X86_64
Oct  2 20:43:14 localhost kernel[0]: vm_page_bootstrap: 4047863 free pages and 113673 wired pages
Oct  2 20:43:14 localhost kernel[0]: kext submap [0xffffff7f807a9000 - 0xffffff8000000000], kernel text [0xffffff8000200000 - 0xffffff80007a9000]
Oct  2 20:43:14 localhost kernel[0]: zone leak detection enabled
Oct  2 20:43:14 localhost kernel[0]: "vm_compressor_mode" is 4
Oct  2 20:43:14 localhost kernel[0]: standard timeslicing quantum is 10000 us
Oct  2 20:43:14 localhost kernel[0]: standard background quantum is 2500 us
Oct  2 20:43:14 localhost kernel[0]: mig_table_max_displ = 74
Oct  2 20:43:14 localhost kernel[0]: TSC Deadline Timer supported and enabled
Oct  2 20:43:14 localhost kernel[0]: "name" not a kext
Oct  2 20:43:14 localhost kernel[0]: "Kernel-__SYMTAB" not a kext
Oct  2 20:43:14 localhost kernel[0]: "DriversPackage-2724000" not a kext
Oct  2 20:43:14 localhost kernel[0]: FakeSMCKeyStore: started
Oct  2 20:43:14 localhost kernel[0]: AppleKeyStore starting (BUILT: Aug 17 2014 20:21:39)
Oct  2 20:43:14 localhost kernel[0]: AppleACPICPU: ProcessorId=1 LocalApicId=0 Enabled
Oct  2 20:43:14 localhost kernel[0]: AppleACPICPU: ProcessorId=2 LocalApicId=2 Enabled
Oct  2 20:43:14 localhost kernel[0]: AppleACPICPU: ProcessorId=3 LocalApicId=4 Enabled
Oct  2 20:43:14 localhost kernel[0]: AppleACPICPU: ProcessorId=4 LocalApicId=6 Enabled
Oct  2 20:43:14 localhost kernel[0]: AppleACPICPU: ProcessorId=5 LocalApicId=1 Enabled
Oct  2 20:43:14 localhost kernel[0]: AppleACPICPU: ProcessorId=6 LocalApicId=3 Enabled
Oct  2 20:43:14 localhost kernel[0]: AppleACPICPU: ProcessorId=7 LocalApicId=5 Enabled
Oct  2 20:43:14 localhost kernel[0]: AppleACPICPU: ProcessorId=8 LocalApicId=7 Enabled
Oct  2 20:43:14 localhost kernel[0]: calling mpo_policy_init for TMSafetyNet
Oct  2 20:43:14 localhost kernel[0]: Security policy loaded: Safety net for Time Machine (TMSafetyNet)
Oct  2 20:43:14 localhost kernel[0]: calling mpo_policy_init for Sandbox
Oct  2 20:43:14 localhost kernel[0]: Security policy loaded: Seatbelt sandbox policy (Sandbox)
Oct  2 20:43:14 localhost kernel[0]: calling mpo_policy_init for Quarantine
Oct  2 20:43:14 localhost kernel[0]: Security policy loaded: Quarantine policy (Quarantine)
Oct  2 20:43:14 localhost kernel[0]: Copyright (c) 1982, 1986, 1989, 1991, 1993
Oct  2 20:43:14 localhost kernel[0]: The Regents of the University of California. All rights reserved.
Oct  2 20:43:14 localhost kernel[0]: MAC Framework successfully initialized
Oct  2 20:43:14 localhost kernel[0]: using 16384 buffer headers and 10240 cluster IO buffer headers
Oct  2 20:43:14 localhost kernel[0]: IOAPIC: Version 0x20 Vectors 64:87
Oct  2 20:43:14 localhost kernel[0]: ACPI: sleep states S3 S4 S5
Oct  2 20:43:14 localhost kernel[0]: CPUSensors: CPU family 0x6, model 0x3c, stepping 0x3, cores 4, threads 8, TJmax 100
Oct  2 20:43:14 localhost kernel[0]: CPUSensors: setting platform keys to [j43     ]
Oct  2 20:43:14 localhost kernel[0]: CPUSensors: base CPU multiplier is 40
Oct  2 20:43:14 localhost kernel[0]: CPUSensors: started
Oct  2 20:43:14 localhost kernel[0]: ACPISensors (TZ00): 1 sensor added
Oct  2 20:43:14 localhost kernel[0]: ACPISensors (TZ00): started
Oct  2 20:43:14 localhost kernel[0]: FakeSMC v6.9.1315 Copyright 2014 netkas, slice, usr-sse2, kozlek, navi, THe KiNG, RehabMan. All rights reserved.
Oct  2 20:43:14 localhost kernel[0]: pci (build 20:04:33 Aug 17 2014), flags 0x63008, pfm64 (39 cpu) 0x7f80000000, 0x80000000
Oct  2 20:43:14 localhost kernel[0]: ACPISensors (TZ01): 1 sensor added
Oct  2 20:43:14 localhost kernel[0]: ACPISensors (TZ01): started
Oct  2 20:43:14 localhost kernel[0]: FakeSMC: 13 preconfigured keys added
Oct  2 20:43:14 localhost kernel[0]: [ PCI configuration begin ]
Oct  2 20:43:14 localhost kernel[0]: SMC: successfully initialized
Oct  2 20:43:14 localhost kernel[0]: console relocated to 0x7f89000000
Oct  2 20:43:14 localhost kernel[0]: [ PCI configuration end, bridges 7, devices 14 ]
Oct  2 20:43:14 localhost kernel[0]: mcache: 8 CPU(s), 64 bytes CPU cache line size
Oct  2 20:43:14 localhost kernel[0]: mbinit: done [128 MB total pool size, (85/42) split]
Oct  2 20:43:14 localhost kernel[0]: Pthread support ABORTS when sync kernel primitives misused
Oct  2 20:43:14 localhost kernel[0]: GeForceSensors (pci1): trying to shadow VBIOS...
Oct  2 20:43:14 localhost kernel[0]: rooting via boot-uuid from /chosen: 5661949F-D2BC-398A-B55F-8AC9F5CD7E62
Oct  2 20:43:14 localhost kernel[0]: Waiting on <dict ID="0"><key>IOProviderClass</key><string ID="1">IOResources</string><key>IOResourceMatch</key><string ID="2">boot-uuid-media</string></dict>
Oct  2 20:43:14 localhost kernel[0]: com.apple.AppleFSCompressionTypeZlib kmod start
Oct  2 20:43:14 localhost kernel[0]: com.apple.AppleFSCompressionTypeLZVN kmod start
Oct  2 20:43:14 localhost kernel[0]: com.apple.AppleFSCompressionTypeDataless kmod start
Oct  2 20:43:14 localhost kernel[0]: com.apple.AppleFSCompressionTypeZlib load succeeded
Oct  2 20:43:14 localhost kernel[0]: com.apple.AppleFSCompressionTypeLZVN load succeeded
Oct  2 20:43:14 localhost kernel[0]: com.apple.AppleFSCompressionTypeDataless load succeeded
Oct  2 20:43:14 localhost kernel[0]: Ethernet [AtherosE2200]: TCP/IPv4 segmentation offload enabled.
Oct  2 20:43:14 localhost kernel[0]: Ethernet [AtherosE2200]: TCP/IPv6 segmentation offload enabled.
Oct  2 20:43:14 localhost kernel[0]: Ethernet [AtherosE2200]: TCP/IPv6 checksum offload enabled.
Oct  2 20:43:14 localhost kernel[0]: Ethernet [AtherosE2200]: Killer E2200: (Rev. 16) at 0xffffff81eb2c5000, 74:d4:35:e7:ba:d5
Oct  2 20:43:14 localhost kernel[0]: Ethernet [AtherosE2200]: Version 1.0.0d7 using max interrupt rate 5000.
Oct  2 20:43:14 localhost kernel[0]: Got boot device = IOService:/AppleACPIPlatformExpert/PCI0@0/AppleACPIPCI/SAT0@1F,2/AppleAHCI/SPT5@5/IOAHCIDevice@0/AppleAHCIDiskDriver/IOAHCIBlockStorageDevice/IOBlockStorageDriver/M4-CT256M4SSD2 Media/IOGUIDPartitionScheme/Sans titre@2
Oct  2 20:43:14 localhost kernel[0]: BSD root: disk0s2, major 1, minor 2
Oct  2 20:43:14 localhost kernel[0]: SuperIODevice: found ITE IT8620E on port=0x2e address=0xa30
Oct  2 20:43:14 localhost kernel[0]: IT87xxSensors: started
Oct  2 20:43:14 localhost kernel[0]: GeForceSensors (pci1): VBIOS successfully read from PRAMIN
Oct  2 20:43:14 localhost kernel[0]: hfs: mounted SSD Mac on device root_device
Oct  2 20:43:14 localhost kernel[0]: XCPM: registered
Oct  2 20:43:14 localhost kernel[0]: USBMSC Identifier (non-unique): 0x5e3 0x723 0x9451
Oct  2 20:43:14 Auguste kernel[0]: flow_divert_kctl_disconnect (0): disconnecting group 1
Oct  2 20:43:14 Auguste kernel[0]: Apple16X50ACPI1: Identified Serial Port on ACPI Device=UAR1
Oct  2 20:43:14 Auguste kernel[0]: Previous Shutdown Cause: 5
Oct  2 20:43:14 Auguste kernel[0]: SMC::smcInitHelper ERROR: MMIO regMap == NULL - fall back to old SMC mode
Oct  2 20:43:14 Auguste kernel[0]: init
Oct  2 20:43:14 Auguste kernel[0]: probe
Oct  2 20:43:14 Auguste kernel[0]: start
Oct  2 20:43:14 Auguste kernel[0]: [IOBluetoothHCIController][start] -- completed
Oct  2 20:43:14 Auguste kernel[0]: DSMOS has arrived
Oct  2 20:43:14 Auguste kernel[0]: Apple16X50UARTSync1: Detected 16550AF/C/CF FIFO=16 MaxBaud=115200
Oct  2 20:43:14 Auguste kernel[0]: NVDAStartup: Official
Oct  2 20:43:14 Auguste kernel[0]: NVDAGF100HAL loaded and registered
Oct  2 20:43:14 Auguste kernel[0]: USB Sound assertion in AppleUSBAudioDevice at line 2166
Oct  2 20:43:14 Auguste kernel[0]: hfs: mounted System Test on device disk2s2
Oct  2 20:43:14 Auguste kernel[0]: hfs: mounted Divers on device disk3s2
Oct  2 20:43:14 Auguste kernel[0]: GeForceSensors (pci1): chipset: GF110 (NVC8) bios: 70.10.17.00
Oct  2 20:43:14 Auguste kernel[0]: GeForceSensors (pci1): started
Oct  2 20:43:15 Auguste kernel[0]: hfs: mounted Backup on device disk3s4
Oct  2 20:43:15 Auguste kernel[0]: USB Sound assertion in AppleUSBAudioDevice at line 2166
Oct  2 20:43:16 Auguste kernel[0]: Ethernet [AtherosE2200]: Link up on en0, 1-Gigabit, Full-duplex, Rx/Tx flow-control
Oct  2 20:43:19 Auguste kernel[0]: VM Swap Subsystem is ON
Oct  2 20:43:19 Auguste kernel[0]: ACPI_SMC_PlatformPlugin::start - waitForService(resourceMatching(AppleIntelCPUPowerManagement) timed out
Oct  2 20:43:19 Auguste kernel[0]: WARNING: IOPlatformPluginUtil : getCPUIDInfo: this is an unknown CPU model 0x3c
Oct  2 20:43:19 Auguste kernel[0]: -- power management may be incomplete or unsupported
Oct  2 20:43:20 Auguste kernel[0]: Starting DoubleCommand
Oct  2 20:43:20 Auguste kernel[0]: DoubleCommand does not support jp_plentycom_driver_SteerMouseKB
Oct  2 20:43:20 Auguste kernel[0]: Hijacking keyboard 0 (IOHIDKeyboard)
Oct  2 20:43:20 Auguste kernel[0]: kea FFFFFF7F809010B8: ksea FFFFFF7F8090110A
Oct  2 20:43:20 Auguste kernel[0]: modded kea FFFFFF7F82292488: modded ksea FFFFFF7F82292663
Oct  2 20:43:20 Auguste kernel[0]: Hijacking keyboard 1 (IOHIDKeyboard)
Oct  2 20:43:20 Auguste kernel[0]: kea FFFFFF7F809010B8: ksea FFFFFF7F8090110A
Oct  2 20:43:20 Auguste kernel[0]: modded kea FFFFFF7F82292488: modded ksea FFFFFF7F82292663
Oct  2 20:43:20 Auguste kernel[0]: Hijacking keyboard 2 (IOHIDKeyboard)
Oct  2 20:43:20 Auguste kernel[0]: keyboard not ready?
Oct  2 20:43:20 Auguste kernel[0]: kea 00000000: ksea 00000000
Oct  2 20:43:20 Auguste kernel[0]: modded kea FFFFFF7F82292488: modded ksea FFFFFF7F82292663
Oct  2 20:43:20 Auguste kernel[0]: IOHIDSystem: Seize of AppleHIDKeyboardEventDriver failed.
Oct  2 20:43:20 Auguste kernel[0]: IOHIDSystem: Seize of IOHIDEventDriver failed.
Oct  2 20:43:21 Auguste kernel[0]: AppleKeyStore:Sending lock change 0
Oct  2 20:43:21 Auguste kernel[0]: Sound assertion in AppleHDAEngine at line 578
Oct  2 20:43:24 auguste kernel[0]: SMC::smcReadKeyAction ERROR F3Mn kSMCBadArgumentError(0x89) fKeyHashTable=0x0xffffff8020cf3800
Oct  2 20:43:24 auguste kernel[0]: SMC::smcReadKeyAction ERROR F3Mx kSMCBadArgumentError(0x89) fKeyHashTable=0x0xffffff8020cf3800
Oct  2 20:43:24 auguste kernel[0]: SMC::smcReadKeyAction ERROR F4Mn kSMCBadArgumentError(0x89) fKeyHashTable=0x0xffffff8020cf3800
Oct  2 20:43:24 auguste kernel[0]: SMC::smcReadKeyAction ERROR F4Mx kSMCBadArgumentError(0x89) fKeyHashTable=0x0xffffff8020cf3800
Oct  2 20:43:24 auguste kernel[0]: SMC::smcReadKeyAction ERROR F5Mn kSMCBadArgumentError(0x89) fKeyHashTable=0x0xffffff8020cf3800
Oct  2 20:43:24 auguste kernel[0]: SMC::smcReadKeyAction ERROR F5Mx kSMCBadArgumentError(0x89) fKeyHashTable=0x0xffffff8020cf3800
Oct  2 20:43:29 auguste kernel[0]: CODE SIGNING: cs_invalid_page(0x1000): p=325[GoogleSoftwareUp] final status 0x0, allowing (remove VALID) page
Oct  2 20:43:30 auguste kernel[0]: AppleSRP started.
Oct  2 20:43:32 auguste kernel[0]: utun_ctl_connect: creating interface utun0
Oct  2 20:43:32 auguste kernel[0]: flow_divert_kctl_disconnect (0): disconnecting group 1
Oct  2 20:43:45 auguste kernel[0]: **** [IOBluetoothHCIController][SearchForTransportEventTimeOutHandler] -- Missing Bluetooth Controller Transport!

ifconfig -a

lo0: flags=8049<UP,LOOPBACK,RUNNING,MULTICAST> mtu 16384
	options=3<RXCSUM,TXCSUM>
	inet6 ::1 prefixlen 128 
	inet 127.0.0.1 netmask 0xff000000 
	inet6 fe80::1%lo0 prefixlen 64 scopeid 0x1 
	nd6 options=1<PERFORMNUD>
gif0: flags=8010<POINTOPOINT,MULTICAST> mtu 1280
stf0: flags=0<> mtu 1280
en0: flags=8863<UP,BROADCAST,SMART,RUNNING,SIMPLEX,MULTICAST> mtu 1500
	options=6b<RXCSUM,TXCSUM,VLAN_HWTAGGING,TSO4,TSO6>
	ether 74:d4:35:e7:ba:d5 
	inet6 fe80::76d4:35ff:fee7:bad5%en0 prefixlen 64 scopeid 0x4 
	inet 192.168.1.22 netmask 0xffffff00 broadcast 192.168.1.255
	nd6 options=1<PERFORMNUD>
	media: autoselect (1000baseT <full-duplex>)
	status: active
utun0: flags=8051<UP,POINTOPOINT,RUNNING,MULTICAST> mtu 1380
	inet6 fe80::e55b:3ba8:2e88:6d77%utun0 prefixlen 64 scopeid 0x5 
	inet6 fd62:1101:d34b:fec0:e55b:3ba8:2e88:6d77 prefixlen 64 
	nd6 options=1<PERFORMNUD>

Once again thank you for your work.

I'm really impressed by this community (starting to realize there are some issues with the way Tonymac "works") and the quality of the information we can find out here.

Mieze · October 2, 2014

I have a z97x-UD5H motherboard including an Atheros Killer E2201 LAN chip so I tried this driver.

It seems to be working fine but I'm able to crash it (losing network and having to reboot) using the iPerf application with more than 4 threads.
iperf -c achille -P 5 -t 600

I don't want to embarrass you but I can hardly believe that you are serious about that because it looks more like a bad joke. The driver is working fine, it does not crash but what do you expect to happen when you run a DoS attack against the network stack?

Flooding the network stack with packets will eat up all the available buffer memory so that you loose network connectivity completely. iperf was made for performance tests of networks. It can put the network system under heavy load and when it is used improperly, it doesn't handle these situations gracefully. For a gigabit connection 1 thread is more than enough, 4 threads are sufficient to put a 10Gbit line under full load.

It's kind of an extreme situation but I think that's one that can happen to me (TimeMachine + iTunes on a network drive + Movie + ...).

Nothing could be farther from the truth than that.

Mieze

lgda · October 3, 2014

Thank you for your answer and explanations.

I don't want to embarrass you but I can hardly believe that you are serious about that because it looks more like a bad joke. The driver is working fine, it does not crash but what do you expect to happen when you run a DoS attack against the network stack?

I won't be embarrassed, don't worry about it, and sorry if this test was that stupid.

Should have mentioned I wasn't an expert in network either.

I know I was able to crash the driver without iPerf with the kind of activity I described (ok, there might have been a problem with iTunes trying to copy the whole library instead of just loading it when it occured) and I thought these 5 iPerf threads could have a similar impact on the network.

I tested on my Macbook with 5 threads running without any problem, higher numbers result in segmentation fault but no network loss so it seemed to me there might have been a problem on the driver side and that's why I posted here.

Mieze · October 4, 2014

I tested on my Macbook with 5 threads running without any problem, higher numbers result in segmentation fault but no network loss so it seemed to me there might have been a problem on the driver side and that's why I posted here.

Since 10.8 Apple's network drivers use a private interface to the network stack which supports QoS packet scheduling so that they are able to handle such situations better.

Mieze

Johnny G · October 15, 2014

Tonymacx86 is full of a bunch of tools. They way they answer questions soemtimes is so demeaning. I was banned once for posting a guide they didn't approve, only to find they re-worded a few things, kept 98% of the info and re-posted it! I found out by re-joining which was a PITA because they log your MAC & IP. Had to use my other PC via smartphone mobile hotspot to register. I was furious.

I've used [url="http://www.insanelymac.com/forum/topic/279450-why-insanelymac-does-not-support-tonymacx86/"]#####[/url] many times in the past and present due to the collection of bundled apps and drivers. That is the only thing that site is good for. But, they didn't even ask for your driver or give credit?! Shameful. I was goign to write a nice guide for my current build but to hell with them.

Back to the point, your driver is fantastic. For all intensive purposes, it is working 100%! Currently using the 1.0.0.d7 (aka 1.0.0?) the most current driver. The ALX 1.0.2 driver floating around elsewhere on the 'net was buggy, I could get my z97 hack (killer e2201 NIC) to crash on command during a speedtest from speedtest.net. Also would occasionally crash during VPN, torrenting and large intranet file transfers.

Thanks you very much for putting this together. Please, PM me your paypal address, a donation is in order.

Mieze · October 15, 2014

Thanks you very much for putting this together. Please, PM me your paypal address, a donation is in order.

In case you want to donate, please give the money to insanelymac.com or any organization for the protection of animals, in particular cats. :cat:

Johnny G · October 19, 2014

In case you want to donate, please give the money to insanelymac.com or any organization for the protection of animals, in particular cats.

Mieze, very graceful of you, you are truly a very giving soul

As for animal rights, I went to a local event recently called the Animal Ball. I live on Long Island, NY, USA. The event was a dinner party, open bar and fantastic food and all proceeds went towards Save-A-Pet, an animal shelter where I rescued my 2 little guys from!

My next donation there for the holidays, whether it be food or litter etc., I will double it and donate on your behalf. Paws up! Hehehe.

Any who in regards to this wonderful driver, is it generally assumed at this time with version 1.0.0d7 aka the release version, is working 100% for everyone else too? I have had my OS X box online with internet access without issue for nearly a week now. Torrents, FTP, etc. Over 100GB of files/data have passed through the line, mostly uploading (my original issue) and it is working without fail. Any kernel dumps/logs needed to be contributed on my end let me know.

My NIC on my gigabye ga-z97mx gaming-5 mobo is a Killer E2201.

Zyphrax · October 19, 2014

Hi Mieze,

This driver looks really great. I'm getting a bit frustrated with the log messages caused by ALXEthernet. However I have an Atheros AR8151 chipset (Gigabyte Z77X-UD3H motherboard, I think someone in this topic mentioned it before, but perhaps he had a later motherboard revision).

Do you think your driver would work with the AR8151? I'd be happy to test it for you.

Cheers

P.S. My 5 month old Australian domestic shorthair says hi

fidesachates · November 1, 2014

Hi!

I created an account(and took the tests lol) just to say thanks! Your driver fixed my KP issue that I had almost given up on. A donation is certainly coming to my nearest cat shelter on your behalf!

Aergan · November 4, 2014

Many thanks for providing this driver Mieze.

Under 10.9.5 the driver works very well with my Killer E2201 on a Gigabyte GA-Z97N-Gaming 5 board.

Under 10.10.0, I'm intermittently losing the network connection during use and requires either a reboot or deactive/reactivate of the network connection. Also, I'm unable to get Wake On Lan working but I am also having this issue under Ubuntu 14.04/14.10 which I believe may be related as WOL support is currently broken/disabled in the ALX driver.

When I lose the network connection, it's typically when I am downloading large files via FireFox / Safari, using XBMC/Kodi to stream a 6-8GB 1080p movie over SMB, remoting the machine via Team Viewer or using Garageband to download lessons. Under 10.9.5, I'm not experiencing any of these issues (except I don't think WOL works outside of sleep).

Any advice of what I can maybe do to fix what might be causing this issue?

joe75 · November 5, 2014

I haven't had any problems in betas or final and I download large 10+ gb files @ 10-15MB/s with not one drop ever. Not sure if it will help but maybe try deleting network preference files to rule out any cache/settings problems from upgrading or migrating.

Mieze · November 5, 2014

@Aergan: As there hasn't been any change in the driver programming interface since 10.7, I don't think that it has anything to do with the driver. Maybe your network configuration is messed up which can produce strange errors including complete loss of network connectivity. You might also use the debug version in order to collect some log data and send it to me. Maybe it'll help me to find out what exactly is going on. It might also be a bug in Yosemite's network stack which seems to be not of the quality it should be. See http://www.insanelymac.com/forum/topic/205771-appleintele1000ekext-for-108107106105/page-35?do=findComment&comment=2066893

One more question: are you using any kind of file sharing software?

Please also note that OS X doesn't support WoL from S5 (power off), only WoL from S3 (sleep) is supported and I can confirm that it's working on the Killer NICs.

EDIT: I installed 10.10 on my 2011 iMac this evening. The installation finished, the machine rebooted and got stuck at the spinning wheel. Here is my strong advice: get back to Mavericks and try again next year when Apple hopefully will have managed to fix the issues. At least that is what I'm going to do. :angry:

Mieze

Edited November 5, 2014 by Mieze

madwad · November 5, 2014

Mieze, first of all, many thanks for your hard work, this is absolutely working driver for GA-Z77-DS3H without any crashes even under heavy load.

But... I don't really know if it may be related or not to this driver, but since I installed it onto my hack (10.8.x) replacing obsolete and unstable AtherosE2200Ethernet driver, iMessage stopped working, which previously worked OOB. So I have to ask, if this driver may interfere with iMessage or I have to investigate problem elsewhere? Do you have or anyone here this driver installed with iMessage working? Some little help would be very appreciated, cause I already tried out (all?) workarounds found on the forum/web to get it working, but without success. Still problem with logging in to iMessage...

Week ago I decided to reinstalled my machine to 10.10 Yosemite, I hoped the problem disappears... After one day of trial and errors, I installed this new OS with 100% stable and working HW (sleep/wake, autosleep, net, audio, fusion drive, dual monitors, usb, ...), but the same and identical problem keeps continuing I had before on 10.8.

Just in case, I mention that I'm on Clover with 10.10, I've used Chimera before with 10.8 - same problem with both of these bootloaders, and I use and also used FusionDrive.

@Aergan: Under 10.10.0, I'm intermittently losing the network connection during use and requires either a reboot or deactive/reactivate of the network connection. Also, I'm unable to get Wake On Lan working but I am also having this issue under Ubuntu 14.04/14.10 which I believe may be related as WOL support is currently broken/disabled in the ALX driver.
When I lose the network connection, it's typically when I am downloading large files via FireFox / Safari, using XBMC/Kodi to stream a 6-8GB 1080p movie over SMB, remoting the machine via Team Viewer or using Garageband to download lessons. Under 10.9.5, I'm not experiencing any of these issues (except I don't think WOL works outside of sleep).

After 4 days running hack on 10.10, I can say, the Mieze's driver works under Yosemite without any issue or instability problem, exactly same as under Montain Lion. I guess your problem is elsewhere... I don't use WOL, so I cannot say anything about it.

Thx

Madwad

Aergan · November 5, 2014

@Aergan: As there hasn't been any change in the driver programming interface since 10.7, I don't think that it has anything to do with the driver. Maybe your network configuration is messed up which can produce strange errors including complete loss of network connectivity. You might also use the debug version in order to collect some log data and send it to me. Maybe it'll help me to find out what exactly is going on. It might also be a bug in Yosemite's network stack which seems to be not of the quality it should be. See http://www.insanelymac.com/forum/topic/205771-appleintele1000ekext-for-108107106105/page-35?do=findComment&comment=2066893

One more question: are you using any kind of file sharing software?

Please also note that OS X doesn't support WoL from S5 (power off), only WoL from S3 (sleep) is supported and I can confirm that it's working on the Killer NICs.

EDIT: I installed 10.10 on my 2011 iMac this evening. The installation finished, the machine rebooted and got stuck at the spinning wheel. Here is my strong advice: get back to Mavericks and try again next year when Apple hopefully will have managed to fix the issues. At least that is what I'm going to do.

Mieze

I've tested WOL from S3 state and that works and your explanation makes a lot of sense, many thanks for that.

I've tried both native installation and clover injection, both have the same issue that somewhen during my session it will drop the connection citing "cable unplugged" when physically this is not true. I've tried different cables and gigabit switches just to be sure, but I get no such issue under OSX Mavericks, Ubuntu 14.10 or Windows 8.1.

I'm not running any filesharing facilities / P2P clients, but I am using the following:

Lingo (IRC client)
Microsoft Remote Desktop (not all the time, but a lot of the time)
Twitter
iTunes
Steam
Battle.net
TeamViewer
Dropbox
OneDrive

I rebuilt the kernel caches & repaired permissions last night after moving back to a native kext installation - so far no issues at the moment.

If (and hopefully not 'when') it occurs again, I will load the debug driver as you have suggested.

PS. When I upgraded to Yosemite, I had a similar issue and it turned out to be that enabling 3rd party TRIM support had killed it in my case. I followed Cindori's guide and I was back up and running within 15mins with TRIM support enabled.

Mieze, first of all, many thanks for your hard work, this is absolutely working driver for GA-Z77-DS3H without any crashes even under heavy load.

But... I don't really know if it may be related or not to this driver, but since I installed it onto my hack (10.8.x) replacing obsolete and unstable AtherosE2200Ethernet driver, iMessage stopped working, which previously worked OOB. So I have to ask, if this driver may interfere with iMessage or I have to investigate problem elsewhere? Do you have or anyone here this driver installed with iMessage working? Some little help would be very appreciated, cause I already tried out (all?) workarounds found on the forum/web to get it working, but without success. Still problem with logging in to iMessage...

Week ago I decided to reinstalled my machine to 10.10 Yosemite, I hoped the problem disappears... After one day of trial and errors, I installed this new OS with 100% stable and working HW (sleep/wake, autosleep, net, audio, fusion drive, dual monitors, usb, ...), but the same and identical problem keeps continuing I had before on 10.8.

Just in case, I mention that I'm on Clover with 10.10, I've used Chimera before with 10.8 - same problem with both of these bootloaders, and I use and also used FusionDrive.

After 4 days running hack on 10.10, I can say, the Mieze's driver works under Yosemite without any issue or instability problem, exactly same as under Montain Lion. I guess your problem is elsewhere... I don't use WOL, so I cannot say anything about it.

Thx

Madwad

I had an issue initially with iMessage on Yosemite which I temporarily resolved by filling in details for Hardware UUID, ROM etc. Up until this point it had worked fine under Mavericks but stopped working after it "broke" under Yosemite. After which I had a new prompt within iMessage to contact Apple support with a Customer ID (which I did yesterday) - now that's all validated/activated and iMessage is working for me (read: nothing to do with the network driver at all).

According to Apple Support UK (and take this with a few handfuls of rock-salt), they are validating a lot of Apple ID's that upgrade to iCloud (OSX Yosemite & iOS8) at the moment.

madwad · November 6, 2014

I've tested WOL from S3 state and that works and your explanation makes a lot of sense, many thanks for that.

I've tried both native installation and clover injection, both have the same issue that somewhen during my session it will drop the connection citing "cable unplugged" when physically this is not true. I've tried different cables and gigabit switches just to be sure, but I get no such issue under OSX Mavericks, Ubuntu 14.10 or Windows 8.1.

I'm not running any filesharing facilities / P2P clients, but I am using the following:

Lingo (IRC client)

Microsoft Remote Desktop (not all the time, but a lot of the time)

Twitter

iTunes

Steam

Battle.net

TeamViewer

Dropbox

OneDrive

I rebuilt the kernel caches & repaired permissions last night after moving back to a native kext installation - so far no issues at the moment.

If (and hopefully not 'when') it occurs again, I will load the debug driver as you have suggested.

PS. When I upgraded to Yosemite, I had a similar issue and it turned out to be that enabling 3rd party TRIM support had killed it in my case. I followed Cindori's guide and I was back up and running within 15mins with TRIM support enabled.

I have TRIM support enabled by kext patching directly set in Clover's config, and I have no problem with ethernet even under very heavy load.

I had an issue initially with iMessage on Yosemite which I temporarily resolved by filling in details for Hardware UUID, ROM etc. Up until this point it had worked fine under Mavericks but stopped working after it "broke" under Yosemite. After which I had a new prompt within iMessage to contact Apple support with a Customer ID (which I did yesterday) - now that's all validated/activated and iMessage is working for me (read: nothing to do with the network driver at all).

According to Apple Support UK (and take this with a few handfuls of rock-salt), they are validating a lot of Apple ID's that upgrade to iCloud (OSX Yosemite & iOS8) at the moment.

Thx for your answer, good to know this driver doesn't interfere with iMessage in any way as you said you have it working. I will try to set all needed HW details again this weekend, maybe I forgot something...

Madwad

Aergan · November 6, 2014

I have TRIM support enabled by kext patching directly set in Clover's config, and I have no problem with ethernet even under very heavy load.

I was responding off topic to Yosemite getting stuck after an upgrade, nothing to do with the network driver.

wfaizan · November 11, 2014

I have absolutely no clue how to do this. Anyway you can guide me step by step. Assume I know nothing. Basically because I don't know anything. I've tried to understand it as best as I can, but I have failed. If you can guide me step by step let me know.

So far, I have installed yosemite. And installed power management for i7 haswell. Installed mouse and keyboard drivers. But those were done through MultiB. Which was simple as click and play. Now i am trying to install network drivers. Because it's not through MultiB, I have no clue what to do.

Thanks in advance if you can help.

Mieze · November 12, 2014

@wfaizan: Use KextWizard or any other kext installer tool in order to install the driver.

Mieze

wfaizan · November 12, 2014

Where can I get this "kext" file. I went to github website and on the bottom right it says download ZIP. I have done that and I have it on my MAC. But when I click installation on kext wizard and I browse to the folder I don't know what file to select. I have the Zip and I have the files extracted.

Update: Doh! Feel like an idiot. Didn't realize there were two separate links.

MacKonsti · November 16, 2014

Der Mieze many thanks from me too, your work here on Insanely has always been appreciated.

Your official version 1.0.0 worked on my Gigabyte GA-Z87N-WIFI board (Intel Core i7-4790K) that has Qualcomm Atheros AR8161 Gigabit Ethernet [1969:1091] (rev 10) as reported by lspci tool.

Again, warm thanks!

tanguy_k · November 16, 2014

@Mieze thx a lot for making this open source and available on GitHub

Shmi Anderson · December 10, 2014

Mieze, thank you very much for the kext. I was having problems with a AR8171/75 card and ALXEthernet. My g750jx would instant reboot when under heavy uploading. Your kext works beautifully. Thanks!

Aergan · December 10, 2014

Just wanted to say that my issues with losing network have all been resolved after the 10.10.1 update, definitely not an issue with the driver - many thanks.

dgsga · December 19, 2014

Mieze,

In a word: Excellent!

Many thanks.

Samlant · December 31, 2014

First off- Thanks for this kext!

Though I'm excited to use it and finally have ethernet, I am plagued by the kext reporting that my BIOS has disabled the NIC. I am aware that you know of this issue; I'd simply like to ask if (1) there were any fixes that have been found, and, (2) if not, if you'd like me to supply any debugging information to aid in the research?

As you must have other projects to focus on, I have no problem in waiting this out till the BIOS manufacturer updates their software or someone finds a solution in this next year.

With Unbounding Love, Grace and Peace,

-Sam L'anteigne

Solution for Qualcomm Atheros AR816x, AR817x and Killer E220x

Recommended Posts

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites