Jump to content

NForce4 Serial ATA (DEVELOPMENT)


hamlo
 Share

185 posts in this topic

Recommended Posts

Ur pretty damn lucky...!!Just do us a favour.Put up ur boot log here... and tell us as to how u've setup ur hard disk drive...primary or secondary master? Wats ur cdrom drive installed on ??Is ur optical drive SATA or IDE ?and so on....anything u can think of ;)

Link to comment
Share on other sites

I don't believe in luck. :D

 

So you get the IOATAController blocking bus message? Well, that's SURELY slowing down perfomance since it issues bus soft-resets after every of those messages, effectively halting in-progress commands. So please check it, just to be sure (in Console.app, system.log, check for that exact message. Also it may be interesting to check for filesystem / disk arbitration error messages. )

 

As subramanyam said, I'd like to see your boot log. I'm interested in ApplenForceATA and AppleVIAATA lines, with their IRQs and everything they output to. If you're using a debug enabled driver (like Hamlo's) it'll show register info.

 

Also if you have more than one partition in the same SATA disk (remember: you may lose data...) you can try to copy data around the same data disk. During my tests that causes most failures.

Link to comment
Share on other sites

subramanyam & jape sambo60 is not alone to get working sata with nf4, I did to but when i did repetitive md5 and sha1 checks, I noticed that sometimes only for files greater than 1 GB (approx) I don't get the same checksum.

 

Strangely after a reboot the checksum could be good for one or two times but not definitely.

 

Most weird after a failure in :

 

Disk Utility stopped repairing “SATA” because the following error was encountered:

The underlying task reported failure on exit

1 HFS volume checked

1 volume could not be repaired because of an error

 

After reboot everything was good with disk utility

 

Nov 29 06:49:21: Disk Utility started.

 

Verify and Repair disk “SATA”

Checking HFS Plus volume.

Checking Extents Overflow file.

Checking Catalog file.

Checking Catalog hierarchy.

Checking Extended Attributes file.

Checking volume bitmap.

Checking volume information.

The volume SATA appears to be OK.

Mounting Disk

 

1 HFS volume checked

No repairs were necessary

 

mobo A8N-E with AMD64 X2 3800, ATI X1600, 3x ATA, 1x SATA disks.

10.4.8.dvd+patches+b8 kernel

Link to comment
Share on other sites

Well Robert364 "seems" that he got it working too, but he had IOATAController blocking bus error.

 

AlainToX714, your drive model?

 

For those who are here, can we put our drive models on our signatures so that each other can identifity each HDD more easily? (wow, that sentence was way too complicated for my english level.. :D)

Link to comment
Share on other sites

I don't actually know. I just installed natively off the dvd at tpb (the amd dvd). There is actually no option in the install menus for nf4 sata, but it detected my drives out of the box, or should I say in the box.

 

can you tell me the exact name of the dvd you installed, because i don't know wich one is the "dvd at tpb".

 

Thanks and congrats for making it work and shared with us.

Link to comment
Share on other sites

Maxtor DiamondMax 10 model : 6V320F0 320 GB Sata 3GB HDD

Now THIS is interesting. You have my MoBo model and my HDD model. Maybe I should try your build: 10.4.8.dvd+patches+b8 kernel . It'll take a LONG time to download though.

 

Do you use AppleVIAATA, hamlo's driver, ApplenForceATA? What are your boot messages? Any IOATAController blocking bus?

 

You say you're getting corruption for large files, 100% reproducible? Any freezes?

Link to comment
Share on other sites

Very interesting note for me. Today I received a new wireless usb device in the mail. I tried to use it, and then I started having problems with my disk. Basically, when I have the wireless device plugged in, the IOATAdevice blocking bus errors go crazy and I get problems. I think usb devices plugged in at the same time make bad things happen with sata on nforce4.

Link to comment
Share on other sites

1 thing i wud like to add as well..I reinstalled Osx (10.4.8 with 8.8.1) today morning and on restart removed my cd rom drive and DSL modem and used it.It was amazingly smooth and did not get a single IOATA Blocking error.

 

I copied around files of over 10 gb and the checksum was perfect.

 

 

I then plugged in my modem,installed forcedeth and immediately ,BOOM ,IOATA Blocking bus error....

Link to comment
Share on other sites

What i am running :

- OSX_1048_AMD_SSE3.iso updated with OSX_1048_AMD_SSE3_PPF1, amd_1048_new.txt applied

- SecUpd2006-007Intel.pkg with secupd.txt applied

 

I have unplug for a few days my Sata disk because i don't want it to interfere with PPF1 update

 

Today i plug back my Sata disk to check:

Booting was difficult, i hang for a few times at:

APPLEVIAATA.Driver: VIA 8237 SATA(CMD 0x970 CTR 0xb70 IRQ23 BM0xd808)

but finally pass over

I check immediately with Disk Utility, here is part of the log:

 

Dec 5 07:24:53: Disk Utility started.

 

Verifying volume “SATA”

Checking HFS Plus volume.

The volume SATA needs to be repaired.

......

Error: The underlying task reported failure on exit

Disk Utility stopped verifying “SATA” because the following error was encountered:

The underlying task reported failure on exit

1 HFS volume checked

Volume needs repair

 

Verify and Repair disk “SATA”

......

Invalid volume free block count

(It should be 16502820 instead of 16500351)

Repairing volume.

......

Checking volume information.

The volume SATA was repaired successfully.

Mounting Disk

 

1 HFS volume checked

Repair attempted on 1 volume

1 HFS volume repaired

 

I check in dmesg and see:

 

IOATAController device blocking bus.

jnl: replay_journal: from: 15419904 to: 15568896 (joffset 0x951000)

jnl: journal start/end pointers reset! (jnl 0x240cc98; s 0xee7400 e 0xee7400)

IOATAController device blocking bus.

 

What was working pretty fine for me was:

 

- OSX_1048_AMD_SSE3.iso with amd_1048_rufus.txt, minipatcher.txt, MinipatcherV2.txt applied

- kernel could have been k19-rc12 or b5, i don't remenber.

 

I have been able even to use rsync_hfs and only got for one or two times similar errors as mentionned

before with Disk Utility

 

With small files around 200 MB md5 Checksums were always the sames.

With big files plaese see my precedent post.

Link to comment
Share on other sites

I've described the USB - nForce symptoms YEARS ago, on the first threads that were opened about this. EVERY hardware movement produces more errors. Which is one of the reasons I usally look at the interrupt workloop.

 

Is anobody saying that NEWER kernels (NOT newer OS X versions, but NEWER kernels -- like 10.4.8 newly patched kernels) work BETTER?

Link to comment
Share on other sites

Is anobody saying that NEWER kernels (NOT newer OS X versions, but NEWER kernels -- like 10.4.8 newly patched kernels) work BETTER?
I don't think so. Problem in that driver doesn't handle shared interrupts correctly, you must rewrite this part to give it ability run interrupt handle in primary interrupt context, now it handles in IOWorkLoop context.

 

BTW Hamlo can't provide us with full version of driver now, maybe latter...

Link to comment
Share on other sites

 Share

×
×
  • Create New...