Darwin nForce4 ATA/SATA controller |
![]() |
Darwin nForce4 ATA/SATA controller |
|
bikedude880
InsanelyMac Legend
|
![]() |
Dec 16 2006, 05:48 AM Post #1
|
![]()
|
Since I have so much free time, I decided to try and write a driver for the nForce series chipsets (nForce/nForce2/nForce3/nForce4) based on the AppleOnboardPCATA kext. This is only a test and I don't expect much to come out of it.
------------------------------------------------------------- MeDevil has successfully figured out the core problem behind the "Device blocking bus" error. ATA/SATA is recognized, drives are UDMA enabled, and will work with most chipsets after the correct device id is added to the plist file (there are already several id's in there.) Download link |
|
tomtefar
InsanelyMac Sage
|
![]() |
Dec 16 2006, 10:49 AM Post #2
|
![]() ![]() ![]() ![]()
|
Will give it a fly-by when I complete the install of my C2D box.
Btw, why not start with Hamlo's driver and try to fix that one?? |
|
macwanabe
InsanelyMac Protégé
|
![]() |
Dec 16 2006, 12:05 PM Post #3
|
![]()
|
Ok installed on Nforce 4 SLI board. It works in the sense that I can see ATA and SATA drives and access them. Speed seems a bit slower. Copy at 43 MB a second. I copied 4 GIG of files to SATA Drive on FAT32 Partition, dmg fail checksum when copied to SATA, data corruption still taking place. Hope this helps.
This post has been edited by macwanabe: Dec 16 2006, 12:24 PM |
|
jape
InsanelyMac Geek
|
![]() |
Dec 16 2006, 12:34 PM Post #4
|
![]() ![]()
|
Source please?
Do you have a nice idea? |
|
|
AlainTox714
InsanelyMac Geek
|
![]() |
Dec 16 2006, 02:22 PM Post #5
|
![]() ![]() ![]()
|
What i am running :
- mobo A8N-E with AMD64 X2 3800, ATI X1600, 3x ATA, 1x SATA disks. - OSX_1048_AMD_SSE3.iso updated with OSX_1048_AMD_SSE3_PPF2, amd_1048_new.txt applied - SecUpd2006-007Intel.pkg with secupd.txt applied Boot fine apparently no problems but when i check in logs please see what i found: kernel[0]: IOATAController device blocking bus. kernel[0]: IOATAController device blocking bus. mDNSResponder: mDNSPlatformRawTime: last_mach_absolute_time 066372A700000013 mDNSResponder: mDNSPlatformRawTime: this_mach_absolute_time 066364F700000013 kernel[0]: IOATAController device blocking bus. kernel[0]: jnl: flushing fs disk buffer returned 0x3c diskarbitrationd[57]: disk1s2 hfs 5F6A1098-CA22-3F01-983F-F60CB5527410 SATA /Volumes/SATA Otherwise with small files <= 650MB the checksum is always the same. With big files at the second try the checksum is diffrent. MD5 (/Volumes/SATA/OPERATING SYSTEM - CD-DVD/Mac OS X 10.4.6 Retail Install DVD (PPC)/Mac OS X Install DVD.dmg) = ced07c1007caab786a185f0f7e148470 MD5 (/Volumes/SATA/OPERATING SYSTEM - CD-DVD/Mac OS X 10.4.6 Retail Install DVD (PPC)/Mac OS X Install DVD.dmg) = ffe439e2db58201747dfc352dc9433b2 |
|
jape
InsanelyMac Geek
|
![]() |
Dec 16 2006, 04:15 PM Post #6
|
![]() ![]()
|
QUOTE why not start with Hamlo's driver and try to fix that one?? Up so far, nobody has really written a kext from scratch (i.e. mine was based on nForceATA, hamlo's latest was a remix of different kexts with some custom pieces, and i dunno about this one). But as I said, I think writing from scratch is useless unless we understand WHAT'S FAILING first. Cause If I were to write a new kext, I would end up copy VIAATA's, "Because that one should have worked ". So I think it's better to just try to edit current avalaible kexts ("try & crash & try again with a different value" procedure) than to lose time rewriting. |
|
bikedude880
InsanelyMac Legend
|
![]() |
Dec 16 2006, 04:38 PM Post #7
|
![]()
|
This one is based off of Apples older nForce kext. It sounds like timing is still off and a couple of access registers are not properly mapped. I will take a look at it when I have the time.
|
|
bikedude880
InsanelyMac Legend
|
![]() |
Dec 16 2006, 04:44 PM Post #8
|
![]()
|
Up so far, nobody has really written a kext from scratch (i.e. mine was based on nForceATA, hamlo's latest was a remix of different kexts with some custom pieces, and i dunno about this one). But as I said, I think writing from scratch is useless unless we understand WHAT'S FAILING first. Cause If I were to write a new kext, I would end up copy VIAATA's, "Because that one should have worked ". So I think it's better to just try to edit current avalaible kexts ("try & crash & try again with a different value" procedure) than to lose time rewriting. I'm currently working at getting the VIA ATA kext working with the SiS chipset and it has roughly the same source as the nForce ATA kext. I will try to output error message to the system log. |
|
subramanyam
InsanelyMac Geek
|
![]() |
Dec 17 2006, 03:17 AM Post #9
|
![]() ![]() ![]()
|
Good to know tat somebody is still trying to make this work....
|
|
bikedude880
InsanelyMac Legend
|
![]() |
Dec 17 2006, 03:28 AM Post #10
|
![]()
|
|
|
AlainTox714
InsanelyMac Geek
|
![]() |
Dec 18 2006, 06:45 AM Post #11
|
![]() ![]() ![]()
|
Please find attached 3 files from FreeBSD sources, not sure it could be helpfull
but in ata-chipset.c there is a lot timing From the man page i post here : http://forum.insanelymac.com/index.php?sho...091&st=120# The currently supported ATA/SATA controller chips are: ....... nVidia: nForce, nForce2, nForce2 MCP, nForce3, nForce3 MCP, nForce3 Pro, nForce4. Good work, i try by myself to understand how it works but my knowledge is inadequate.
Attached File(s)
|
|
subramanyam
InsanelyMac Geek
|
![]() |
Dec 18 2006, 07:57 AM Post #12
|
![]() ![]() ![]()
|
bikedude880...it wud be great if u could actually let us know wat u are trying coz all it requires is one moment of brilliance from one of the members to pull this off...
This post has been edited by subramanyam: Dec 18 2006, 07:58 AM |
|
iMcen
InsanelyMac Geek
|
![]() |
Dec 18 2006, 01:27 PM Post #13
|
![]() ![]() ![]()
|
I've tried the kext, andit gives me lots of read and write errors. Keep up the good work, I would like to test further releases.
|
|
Thico Alves
InsanelyMac Geek
|
![]() |
Dec 18 2006, 04:32 PM Post #14
|
![]() ![]()
|
what about nForce 410/430 controllers?!? I try this one with mine, but it didn't even initializated, and I got still waiting bla bla bla again... AppleVIAATA detected my controller but frozed just after that with a read/write error
|
|
bikedude880
InsanelyMac Legend
|
![]() |
Dec 19 2006, 02:29 AM Post #15
|
![]()
|
Please find attached 3 files from FreeBSD sources, not sure it could be helpfull but in ata-chipset.c there is a lot timing ... The problem with the timing in that file is that they are in the format 0x28269008 and not 0x13 . what about nForce 410/430 controllers?!? I try this one with mine, but it didn't even initializated, and I got still waiting bla bla bla again... AppleVIAATA detected my controller but frozed just after that with a read/write error Working on that. My focus is in other places. Leopard. |
|
evanrodge
InsanelyMac Protégé
|
![]() |
Dec 19 2006, 11:33 PM Post #16
|
![]()
|
I would really like for this to happen so I can use my new dv6000z!! A hefty donation would be in the future of the developer who writes a good, stable driver that gets my laptop working
|
|
subramanyam
InsanelyMac Geek
|
![]() |
Dec 21 2006, 06:01 AM Post #17
|
![]() ![]() ![]()
|
Bikedude880 had mentioned that he intended to use the same source to write the drivers for the Sis and Nforce chipsets.Now that Sis is working,guess this thread deserves a bump...
|
|
Thico Alves
InsanelyMac Geek
|
![]() |
Dec 22 2006, 03:29 AM Post #18
|
![]() ![]()
|
please someone!! make it working on 410/430!!!
I don't know what more to do to get my notebook SATA HD working on OSX . . . |
|
bikedude880
InsanelyMac Legend
|
![]() |
Dec 22 2006, 04:46 AM Post #19
|
![]()
|
please someone!! make it working on 410/430!!! I don't know what more to do to get my notebook SATA HD working on OSX . . . Give it time, these things don't just pop out of people asses Granted I have been a tad busy doing other things, I have not forgotten about this project. I'll take a look at it tonight. |
|
timmyj
InsanelyMac Protégé
|
![]() |
Dec 22 2006, 04:49 AM Post #20
|
![]()
|
doesn't anyone know how to make this driver work?? it's been a work in progress for ages with countless threads and yet still the best solution is this VIAATA kext which just corrupts files. With some of the work you guys are doing, rewriting kernels, getting leopard running, a mere SATA Driver can't be that much of a challenge, anyways I think its great that some people are continually trying and showing support and i thank them for that but still as yet we have no solution. So what's the latest on this?
|
![]() |
|
Lo-Fi Version | Time is now: 3rd September 2010 - 08:58 AM |