Thanks to all in advance!

Messing around with hdparm only leaves me worse than when I started. When I ran XP it seemed that I had a similar problem with UDMA 100 transfers. Since I am now in a completely windows free environment I can't confirm this.

I have employed the Ultra-DMA Howto of en.tldp.org without success.

I multiplex videos on a dual processor machine. I never even come close to stressing the machine because I can't get the transfer rates. For what I utilize my machine for it is essential that I get this to function and is not just an esoteric nicety

I have checked my bios settings. HDA is on its own proper cable. HDC shares with the cdburner.

Motherboard is a Tyan MP s2460 hosting dual 1.4Ghz athlon processors with 512MB

RedHat 9.0 running 2.4.20-8smp

I have seen numerous postings throughout the web of this problem with no resolution.

linux-2.4.20-8

I have seen where people have commented on ide-mod. The next prints show no ide-mod loaded and that the ide-mod.o does not exist on my system.

[root@localhost ide]# lsmod
Module Size Used by Tainted: P
agpgart 47552 3 (autoclean)
nvidia 1767936 11 (autoclean)
autofs 13396 0 (autoclean) (unused)
tulip 44352 1
sg 37036 0 (autoclean)
sr_mod 17944 0 (autoclean)
ide-scsi 12176 0
scsi_mod 109496 3 [sg sr_mod ide-scsi]
ide-cd 35292 0
cdrom 33920 0 [sr_mod ide-cd]
ohci1394 20680 0 (unused)
ieee1394 51532 0 [ohci1394]
scanner 11228 0 (unused)
keybdev 2912 0 (unused)
mousedev 5560 1
hid 21892 0 (unused)
input 6112 0 [keybdev mousedev hid]
usb-ohci 21896 0 (unused)
usbcore 81440 1 [scanner hid usb-ohci]
ext3 72608 2
jbd 55568 2 [ext3]


[root@localhost ide]# ls
arm ide-disk.c ide-iops.c ide-proc.c Makefile
Config.in ide-dma.c ide-lib.c ide-tape.c pci
ide.c ide-floppy.c ide_modes.h ide-taskfile.c ppc
ide-cd.c ide-geometry.c ide-pnp.c ide-timing.h raid
ide-cd.h ide-io.c ide-probe.c legacy setup-pci.c

Could this be the problem listed above?

[root@localhost ide]# dmesg
Linux version 2.4.20-8smp (bhcompile@stripples.devel.redhat.com) (gcc version 3.2.2 20030222 (Red Hat Linux 3.2.2-5)) #1 SMP Thu Mar 13 16:43:01 EST 2003

****snip****

See the following where the bus is assumed 33 Mhz yet I the Mother board is 100Mhz, HDA is 100 Mhz and HDC shared with cdrw is 133Mhz which hdparm reveals, but dmessage showing the drives only as 33Mhz.

RAMDISK driver initialized: 16 RAM disks of 4096K size 1024 blocksize
Uniform Multi-Platform E-IDE driver Revision: 7.00beta-2.4
ide: Assuming 33MHz system bus speed for PIO modes; override with idebus=xx
AMD7411: IDE controller at PCI slot 00:07.1
AMD7411: chipset revision 1
AMD7411: not 100% native mode: will probe irqs later
ide0: BM-DMA at 0xf000-0xf007, BIOS settings: hdaMA, hdbio
ide1: BM-DMA at 0xf008-0xf00f, BIOS settings: hdcMA, hddMA
hda: QUANTUM FIREBALLP AS60.0, ATA DISK drive
blk: queue c043a1e0, I/O limit 4095Mb (mask 0xffffffff)
hdc: Maxtor 6Y060L0, ATA DISK drive
hdd: CD-RW BCE1610IM, ATAPI CD/DVD-ROM drive
blk: queue c043a660, I/O limit 4095Mb (mask 0xffffffff)
ide0 at 0x1f0-0x1f7,0x3f6 on irq 14
ide1 at 0x170-0x177,0x376 on irq 15
hda: host protected area => 1
hda: 117266688 sectors (60041 MB) w/1902KiB Cache, CHS=116336/16/63, UDMA(33)
hdc: host protected area => 1
hdc: 120103200 sectors (61493 MB) w/2048KiB Cache, CHS=119150/16/63, UDMA(33)
ide-floppy driver 0.99.newide
Partition check:
hda: [PTBL] [7299/255/63] hda1 hda2 < hda5 hda6 hda7 hda8 >
hdc: [EZD] [remap 0->1] [7476/255/63] hdc1 hdc2 < hdc5 >
ide-floppy driver 0.99.newide
md: md driver 0.90.0 MAX_MD_DEVS=256, MD_SB_DISKS=27
md: Autodetecting RAID arrays.
md: autorun ...
md: ... autorun DONE.
**** snip ****

agpgart: Maximum main memory to use for agp memory: 439M
agpgart: Detected AMD 760MP chipset
agpgart: AGP aperture is 64M @ 0xec000000


**************************************************************************************************** ********************************************

This is what hdparm reveals for my drives.

[root@localhost ide]# hdparm -i /dev/hda

/dev/hda:

Model=QUANTUM FIREBALLP AS60.0, FwRev=A1Y.1500, SerialNo=196150175634
Config={ HardSect NotMFM HdSw>15uSec Fixed DTR>10Mbs }
RawCHS=16383/16/63, TrkSize=32256, SectSize=21298, ECCbytes=4
BuffType=DualPortCache, BuffSize=1902kB, MaxMultSect=16, MultSect=16
CurCHS=16383/16/63, CurSects=16514064, LBA=yes, LBAsects=117266688
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 udma3 udma4 udma5
AdvancedPM=no WriteCache=enabled
Drive conforms to: ATA/ATAPI-5 T13 1321D revision 1: 1 2 3 4 5

[root@localhost ide]# hdparm -i /dev/hdc

/dev/hdc:

Model=Maxtor 6Y060L0, FwRev=YAR41VW0, SerialNo=Y2S1HYGE
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=120103200
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 udma3 udma4 udma5 udma6
AdvancedPM=yes: disabled (255) WriteCache=enabled
Drive conforms to: (null): 1 2 3 4 5 6 7

************ Before tuning

[root@localhost ide]# hdparm -Tt /dev/hda

/dev/hda:
Timing buffer-cache reads: 128 MB in 0.52 seconds =246.15 MB/sec
Timing buffered disk reads: 64 MB in 2.51 seconds = 25.50 MB/sec
[root@localhost ide]# hdparm -Tt /dev/hdc

/dev/hdc:
Timing buffer-cache reads: 128 MB in 0.54 seconds =237.04 MB/sec
Timing buffered disk reads: 64 MB in 2.16 seconds = 29.63 MB/sec

*********************

Forcing it to the next level does nothing.

[root@localhost ide]# hdparm -X66 -d1 -u1 -m16 -c3 /dev/hda

/dev/hda:
setting 32-bit IO_support flag to 3
setting multcount to 16
setting unmaskirq to 1 (on)
setting using_dma to 1 (on)
setting xfermode to 66 (UltraDMA mode2)
multcount = 16 (on)
IO_support = 3 (32-bit w/sync)
unmaskirq = 1 (on)
using_dma = 1 (on)

************* After tuning the drive ****************************

[root@localhost ide]# hdparm -Tt /dev/hda

/dev/hda:
Timing buffer-cache reads: 128 MB in 0.52 seconds =246.15 MB/sec
Timing buffered disk reads: 64 MB in 2.50 seconds = 25.60 MB/sec

I have checked all that I know to check in the bios, cabling, and with the drive manufacturers for disk utils to enable higher UDMA modes. neither the quatum nor the Maxtor seem to need this.

I have even tried one post that says the Sound card can be the cause of this exact type of problem. I removed the sound card all together and played with hdparm to no good effect.

Questions:

1. Does "ide: Assuming 33MHz system bus speed for PIO modes; override with idebus=xx" have any affect on me getting the higher udma settings and what exactly does this mean? If yes, how do I pass this to the system?

2. Is "ide-mod" the deciding factor?

3. Is this a known bug that I have not discovered in my research?

4. What solutions, minus recompiling the kernel, could I utilize to correct this? In almost every case this is what has been recommended and in each case has caused more trouble and has not fixed the problem.

5. In each post I see no resolution at the closure of the posts. What is going on with the ide & udma?

It doesn't. I get the same message and then my drive's working as udma4.

I tested my drive and my boyfriend's one and I must tell you that his drive (5600rpm as udma5) gets results very similar to yours. My drive (7200rpm, but older motherboard -> udma4) gets 70/30 in the test. They're not very old.

I'd like to see the test results for one of drives closer to yours (anyone?), but it looks for me rather as a hdparm problem.

Thanks for the (free)dback in such a prompt manner.

Why are my drives only dected as 33Mhz even though they are capable hda of 100 and hdc of 133?

You suspect that hdparm is the issue. Then, should I get another version of it or quit messing with it before I screw something up? LOL

There's also something with kernel (so it looks like kernel-hdparm interaction). In my system there's:
(motherboard supports up to 66MHz).

You get quite good transfer, so I wouldn't mess much in the settings.

Update: as I googled a moment ago, it looks like a kernel driver issue (motherboard support). Probably a new kernel (compiled yourself) would help, but I don't think you'd get really much.

Thanks for the (free)dback!

recompilation of the Kernel was exactly what I had hoped to avoid. It is risky and I need the pc to function.

In my ide folder I am missing ide-mod and it is not a loaded module if you look at the 1st post. I have seen other posts, as I too have been googling, about this module being missing although, I have not found a good resolution out of these posts.

A friend of mine has also reproduced the trouble on his pc's. He also suspects a kernel/module/hdparm issue. No resolution as of yet. Still working on it. Also, because my drives are 5400 rpms, that I should not expect too much more out of them and all this effort may be for nothing.

He also recommended a program call "bonnie" as a better diagnostic tool.

He also states he agrees that ide-mod not being installed might be the source of the problem. He is still investigating.

Possible solution?

Can anyone comment on this ide-2.4.20

http://www.kernel.org/pub/linux/kernel/people/hedrick/

I am going to try this so wish me luck. I will report my findings

your findings Dr Watson??

I was going to try the files this weekend. I had some sensitive information that I am working with on the PC and absolutely must have the PC functioning.

I am working on getting the information off the pc to a safe environment and then I will begin experimentation.

I may have something conclusive on monday or tuesday of next week.

thanks for the reply as i am having a dma issue with my box. I keep getting a "lost interupt" error during boot when ever I enable dma

I went through this earlier. If you look farther down dmesg, you will see udma kick in and then it is fast like it should be. This is the later part of mine.

Now don't ask me why my newer 80GB drive is slower than my older 30GB drive. I suspect the round cable.

I did make my grub look like this though.

That helps the boot speed up a bit till it kicks udma in.

Hope that helps or you get me straight one. I been searching for this too.

Later

The issue you are experiencing may not be related to mine. I am not having an irq issue.

None the less never say never in IT/computers/telecom.

I will post my results.