Enabling UDMA 133 on nforce2 chipset

etaym · 05-04-03, 02:23 PM

Hi,

I bought a new box with a chaintech 7NIF2 board that is based on an nforce2 chipset and has a UDMA133 ide controller. The distribution is RH9 with the 2.4.20-8 kernel that comes with it. I don't think the ide controller is getting to it's full potential. The results of hdparm -tT /dev/hda are:
/dev/hda:
Timing buffer-cache reads: 128 MB in 0.30 seconds =426.67 MB/sec
Timing buffered disk reads: 64 MB in 2.13 seconds = 30.05 MB/sec

dmesg gives the following message:

ide: Assuming 33MHz system bus speed for PIO modes; override with idebus=xx
NFORCE2: IDE controller at PCI slot 00:09.0
NFORCE2: chipset revision 162
NFORCE2: not 100% native mode: will probe irqs later
ide0: BM-DMA at 0xf000-0xf007, BIOS settings: hda

MA, hdb

MA
ide1: BM-DMA at 0xf008-0xf00f, BIOS settings: hdc

MA, hdd

MA

hdparm -i /dev/hda shows:
/dev/hda:
Model=Maxtor 6Y080L0, FwRev=YAR41VW0, SerialNo=Y3J1XCDE
Config={ Fixed }
RawCHS=16383/16/63, TrkSize=0, SectSize=0, ECCbytes=57
BuffType=DualPortCache, BuffSize=2048kB, MaxMultSect=16, MultSect=16
CurCHS=16383/16/63, CurSects=16514064, LBA=yes, LBAsects=160086528
IORDY=on/off, tPIO={min:120,w/IORDY:120}, tDMA={min:120,rec:120}
PIO modes: pio0 pio1 pio2 pio3 pio4
DMA modes: mdma0 mdma1 mdma2
UDMA modes: udma0 udma1 *udma2
AdvancedPM=yes: disabled (255) WriteCache=enabled
Drive conforms to: (null): 1 2 3 4 5 6 7

What should I do to get the ide controller running at 133? Does nvidia supplies a driver for that controller? (the rest of the nvidia drivers, gforce, ethernet and sound are installed and working properly)

TIA

Etay Meiri

suexec · 05-04-03, 06:16 PM

I'm getting UDMA 100, and if I had UDMA 133 drives, possibly UDMA 133. Your hdparm thinks that your drive only supports up to UDMA 2 (33MB/s). There was a problem with an earlier release of the nForce2 IDE driver which didn't properly set the 80pin cable mode properly and locked the max speed at UDMA 33. Your RedHat 9 kernel may still be using this older driver. I'm using 2.4.21-rc1 and have no problems (I think it was fixed as of 2.4.21-pre5?)

gambit:~# hdparm -i /dev/hda

/dev/hda:

Model=ST360021A, FwRev=3.05, SerialNo=3HR08HE1
Config={ HardSect NotMFM HdSw>15uSec Fixed DTR>10Mbs RotSpdTol>.5% }
RawCHS=16383/16/63, TrkSize=0, SectSize=0, ECCbytes=4
BuffType=unknown, BuffSize=2048kB, MaxMultSect=16, MultSect=off
CurCHS=16383/16/63, CurSects=16514064, LBA=yes, LBAsects=117231408
IORDY=on/off, tPIO={min:240,w/IORDY:120}, tDMA={min:120,rec:120}
PIO modes: pio0 pio1 pio2 pio3 pio4
DMA modes: mdma0 mdma1 mdma2
UDMA modes: udma0 udma1 udma2 udma3 udma4 *udma5
AdvancedPM=no WriteCache=enabled
Drive conforms to: device does not report version: 1 2 3 4 5

daclink · 05-05-03, 05:57 AM

If you choose to install the 2.4.21pre5 kernel, the ide driver u need is the AMD viper (the nforce2 chipset piggy backs from it).

DaClink

etaym · 05-09-03, 10:19 AM

OK, I upgraded the kernel to 2.4.21-rc1-ac4 and compiled in the AMD VIPER driver. Now dmesg detects the controller as ATA100 (the actual controller is ATA133) and hdparm benchmark improves from 30Mb/s to 47Mb/s. Can I expect any better than that? Is it possible to detect the controller as ATA133?

Thanks,

Etay Meiri

daclink · 05-09-03, 01:33 PM

Sorry... i don't know any more than that.... the system I run only has ATA66 drives in, so i couldn't tell you what happens if i connect something that would take ata133

DaClink

10k · 05-29-03, 12:39 AM

47MB/sec is very fast. You'll be hard pressed to find any hard drive that can sustain transfer faster than that.

cheers,
10k

05-04-03, 02:23 PM	#1
etaym Registered User Join Date: May 2003 Posts: 2	Enabling UDMA 133 on nforce2 chipset Hi, I bought a new box with a chaintech 7NIF2 board that is based on an nforce2 chipset and has a UDMA133 ide controller. The distribution is RH9 with the 2.4.20-8 kernel that comes with it. I don't think the ide controller is getting to it's full potential. The results of hdparm -tT /dev/hda are: /dev/hda: Timing buffer-cache reads: 128 MB in 0.30 seconds =426.67 MB/sec Timing buffered disk reads: 64 MB in 2.13 seconds = 30.05 MB/sec dmesg gives the following message: ide: Assuming 33MHz system bus speed for PIO modes; override with idebus=xx NFORCE2: IDE controller at PCI slot 00:09.0 NFORCE2: chipset revision 162 NFORCE2: not 100% native mode: will probe irqs later ide0: BM-DMA at 0xf000-0xf007, BIOS settings: hdaMA, hdbMA ide1: BM-DMA at 0xf008-0xf00f, BIOS settings: hdcMA, hddMA hdparm -i /dev/hda shows: /dev/hda: Model=Maxtor 6Y080L0, FwRev=YAR41VW0, SerialNo=Y3J1XCDE Config={ Fixed } RawCHS=16383/16/63, TrkSize=0, SectSize=0, ECCbytes=57 BuffType=DualPortCache, BuffSize=2048kB, MaxMultSect=16, MultSect=16 CurCHS=16383/16/63, CurSects=16514064, LBA=yes, LBAsects=160086528 IORDY=on/off, tPIO={min:120,w/IORDY:120}, tDMA={min:120,rec:120} PIO modes: pio0 pio1 pio2 pio3 pio4 DMA modes: mdma0 mdma1 mdma2 UDMA modes: udma0 udma1 *udma2 AdvancedPM=yes: disabled (255) WriteCache=enabled Drive conforms to: (null): 1 2 3 4 5 6 7 What should I do to get the ide controller running at 133? Does nvidia supplies a driver for that controller? (the rest of the nvidia drivers, gforce, ethernet and sound are installed and working properly) TIA Etay Meiri

05-09-03, 10:19 AM	#4
etaym Registered User Join Date: May 2003 Posts: 2	Upgraded to 2.4.21-rc1-ac4 : now ATA100 OK, I upgraded the kernel to 2.4.21-rc1-ac4 and compiled in the AMD VIPER driver. Now dmesg detects the controller as ATA100 (the actual controller is ATA133) and hdparm benchmark improves from 30Mb/s to 47Mb/s. Can I expect any better than that? Is it possible to detect the controller as ATA133? Thanks, Etay Meiri

05-04-03, 06:16 PM	#2
suexec Registered User Join Date: Apr 2003 Posts: 54	I'm getting UDMA 100, and if I had UDMA 133 drives, possibly UDMA 133. Your hdparm thinks that your drive only supports up to UDMA 2 (33MB/s). There was a problem with an earlier release of the nForce2 IDE driver which didn't properly set the 80pin cable mode properly and locked the max speed at UDMA 33. Your RedHat 9 kernel may still be using this older driver. I'm using 2.4.21-rc1 and have no problems (I think it was fixed as of 2.4.21-pre5?) gambit:~# hdparm -i /dev/hda /dev/hda: Model=ST360021A, FwRev=3.05, SerialNo=3HR08HE1 Config={ HardSect NotMFM HdSw>15uSec Fixed DTR>10Mbs RotSpdTol>.5% } RawCHS=16383/16/63, TrkSize=0, SectSize=0, ECCbytes=4 BuffType=unknown, BuffSize=2048kB, MaxMultSect=16, MultSect=off CurCHS=16383/16/63, CurSects=16514064, LBA=yes, LBAsects=117231408 IORDY=on/off, tPIO={min:240,w/IORDY:120}, tDMA={min:120,rec:120} PIO modes: pio0 pio1 pio2 pio3 pio4 DMA modes: mdma0 mdma1 mdma2 UDMA modes: udma0 udma1 udma2 udma3 udma4 *udma5 AdvancedPM=no WriteCache=enabled Drive conforms to: device does not report version: 1 2 3 4 5

05-05-03, 05:57 AM	#3
daclink Registered User Join Date: Apr 2003 Posts: 23	If you choose to install the 2.4.21pre5 kernel, the ide driver u need is the AMD viper (the nforce2 chipset piggy backs from it). DaClink

05-09-03, 01:33 PM	#5
daclink Registered User Join Date: Apr 2003 Posts: 23	Sorry... i don't know any more than that.... the system I run only has ATA66 drives in, so i couldn't tell you what happens if i connect something that would take ata133 DaClink

05-29-03, 12:39 AM	#6
10k Registered User Join Date: Apr 2003 Posts: 13	47MB/sec is very fast. You'll be hard pressed to find any hard drive that can sustain transfer faster than that. cheers, 10k