poor performance hard disk IDE

lib-pata fix is pending. Try the daily ISO from http://cdimage.ubuntu.com/daily-live/current on April 2. Kernel version must be 2.6.20-14 or higher.

confirmed on the 5 april 2007:
00:00.0 Host bridge: Intel Corporation 82855PM Processor to I/O Controller (rev 03)
00:01.0 PCI bridge: Intel Corporation 82855PM Processor to AGP Controller (rev 03)
00:1d.0 USB Controller: Intel Corporation 82801DB/DBL/DBM (ICH4/ICH4-L/ICH4-M) USB UHCI Controller #1 (rev 01)
00:1d.1 USB Controller: Intel Corporation 82801DB/DBL/DBM (ICH4/ICH4-L/ICH4-M) USB UHCI Controller #2 (rev 01)
00:1d.2 USB Controller: Intel Corporation 82801DB/DBL/DBM (ICH4/ICH4-L/ICH4-M) USB UHCI Controller #3 (rev 01)
00:1d.7 USB Controller: Intel Corporation 82801DB/DBM (ICH4/ICH4-M) USB2 EHCI Controller (rev 01)
00:1e.0 PCI bridge: Intel Corporation 82801 Mobile PCI Bridge (rev 81)
00:1f.0 ISA bridge: Intel Corporation 82801DBM (ICH4-M) LPC Interface Bridge (rev 01)
00:1f.1 IDE interface: Intel Corporation 82801DBM (ICH4-M) IDE Controller (rev 01)
00:1f.3 SMBus: Intel Corporation 82801DB/DBL/DBM (ICH4/ICH4-L/ICH4-M) SMBus Controller (rev 01)
00:1f.5 Multimedia audio controller: Intel Corporation 82801DB/DBL/DBM (ICH4/ICH4-L/ICH4-M) AC'97 Audio Controller (rev 01)
00:1f.6 Modem: Intel Corporation 82801DB/DBL/DBM (ICH4/ICH4-L/ICH4-M) AC'97 Modem Controller (rev 01)
01:00.0 VGA compatible controller: ATI Technologies Inc Radeon R250 [Mobility FireGL 9000] (rev 01)
02:00.0 FireWire (IEEE 1394): VIA Technologies, Inc. IEEE 1394 Host Controller (rev 80)
02:01.0 Ethernet controller: Realtek Semiconductor Co., Ltd. RTL-8139/8139C/8139C+ (rev 20)
02:02.0 Network controller: Intel Corporation PRO/Wireless 2200BG Network Connection (rev 05)
02:04.0 CardBus bridge: ENE Technology Inc CB1410 Cardbus Controller

dev/sda:
 Timing cached reads: 808 MB in 2.00 seconds = 403.80 MB/sec
 Timing buffered disk reads: 76 MB in 3.07 seconds = 24.73 MB/sec

kernel version 2.20.14

Confirmed here with latest kernel (and previous)

/dev/sda:
 Timing cached reads: 1356 MB in 2.00 seconds = 677.44 MB/sec
 Timing buffered disk reads: 164 MB in 3.03 seconds = 54.17 MB/sec

/dev/hda:
 Timing cached reads: 1412 MB in 2.00 seconds = 706.09 MB/sec
 Timing buffered disk reads: 204 MB in 3.00 seconds = 67.95 MB/sec

Fix commited???
Another image that will beat the repos?

there was 2.20.14 (which was 11) incidentally which was later upgraded to 2.20.14-12 which is the current one, even if there is a 13 its cool as far as it fixes this problem.

I'm glad it solves the problem in your box, but not here...

$ uname -a
Linux patatilla 2.6.20-14-generic #2 SMP Mon Apr 2 20:37:49 UTC 2007 i686 GNU/Linux

$ apt-cache policy linux-image-2.6.20-14-generic
linux-image-2.6.20-14-generic:
  Instalados: 2.6.20-14.22
  Candidato: 2.6.20-14.22

# hdparm -tT /dev/hda

/dev/hda:
 Timing cached reads: 1474 MB in 2.00 seconds = 737.26 MB/sec
 Timing buffered disk reads: 232 MB in 3.01 seconds = 77.00 MB/sec

# sudo hdparm -tT /dev/sda

/dev/sda:
 Timing cached reads: 1434 MB in 2.00 seconds = 716.55 MB/sec
 Timing buffered disk reads: 164 MB in 3.02 seconds = 54.34 MB/sec

nope, u mis-understood me, the issue is also with me :-

 sudo hdparm -I /dev/sdb |grep dma

 DMA: mdma0 mdma1 mdma2 udma0 udma1 *udma2 udma3 udma4 udma5

   as you can see I am stuck at udma2 & cannot go upto udma5 hence even my reads are not good.

uname -a
Linux ubuntu 2.6.20-14-generic #2 SMP Mon Apr 2 20:37:49 UTC 2007 i686 GNU/Linux

 sudo hdparm -c 1 /dev/sdb
/dev/sdb:
 setting 32-bit IO_support flag to 1
 HDIO_SET_32BIT failed: Invalid argument
 IO_support = 0 (default 16-bit)

 sudo hdparm -tT /dev/sdb

/dev/sdb:
 Timing cached reads: 448 MB in 2.00 seconds = 223.66 MB/sec
 Timing buffered disk reads: 54 MB in 3.02 seconds = 17.90 MB/sec

lspci
00:00.0 Host bridge: Intel Corporation 82845G/GL[Brookdale-G]/GE/PE DRAM Controller/Host-Hub Interface (rev 03)
00:02.0 VGA compatible controller: Intel Corporation 82845G/GL[Brookdale-G]/GE Chipset Integrated Graphics Device (rev 03)
00:1d.0 USB Controller: Intel Corporation 82801DB/DBL/DBM (ICH4/ICH4-L/ICH4-M) USB UHCI Controller #1 (rev 02)
00:1d.1 USB Controller: Intel Corporation 82801DB/DBL/DBM (ICH4/ICH4-L/ICH4-M) USB UHCI Controller #2 (rev 02)
00:1d.2 USB Controller: Intel Corporation 82801DB/DBL/DBM (ICH4/ICH4-L/ICH4-M) USB UHCI Controller #3 (rev 02)
00:1d.7 USB Controller: Intel Corporation 82801DB/DBM (ICH4/ICH4-M) USB2 EHCI Controller (rev 02)
00:1e.0 PCI bridge: Intel Corporation 82801 PCI Bridge (rev 82)
00:1f.0 ISA bridge: Intel Corporation 82801DB/DBL (ICH4/ICH4-L) LPC Interface Bridge (rev 02)
00:1f.1 IDE interface: Intel Corporation 82801DB (ICH4) IDE Controller (rev 02)
00:1f.5 Multimedia audio controller: Intel Corporation 82801DB/DBL/DBM (ICH4/ICH4-L/ICH4-M) AC'97 Audio Controller (rev 02)
03:01.0 Multimedia controller: Philips Semiconductors SAA7130 Video Broadcast Decoder (rev 01)
03:02.0 Ethernet controller: Realtek Semiconductor Co., Ltd. RTL-8139/8139C/8139C+ (rev 10)
03:07.0 Ethernet controller: Silicon Integrated Systems [SiS] SiS900 PCI Fast Ethernet (rev 02)

Hopefully this output is helpful.

hexion, what I meant was if there is another kernel which fixes the issue it would be well worth the download

Sorry for the misunderstood... my english is not very good ;)

Mmmm.. strange, my discs are set at the highest dma mode, and still aren't performing good...

# hdparm -I /dev/sda |grep dma
        DMA: mdma0 mdma1 mdma2 udma0 udma1 udma2 udma3 udma4 udma5 *udma6
# sudo hdparm -I /dev/hda |grep dma
        DMA: mdma0 mdma1 mdma2 udma0 udma1 udma2 udma3 udma4 *udma5
$ lspci | grep IDE
00:0e.0 IDE interface: ALi Corporation M5229 IDE (rev c7)

Sure it will worth it ;)
Regards

lspci |grep IDE

00:1f.1 IDE interface: Intel Corporation 82801DB (ICH4) IDE Controller (rev 02)

I also have this issue.

  hdparm /dev/sda:

    /dev/sda:
    IO_support = 0 (default 16-bit)
    readonly = 0 (off)
    readahead = 256 (on)
    geometry = 12161/255/63, sectors = 195371568, start = 0

  hdparm -I /dev/sda |grep dma:

Continues..

    DMA: mdma0 mdma1 mdma2 udma0 udma1 udma2 udma3 udma4 *udma5

  lspci |grep IDE:
    00:1f.1 IDE interface: Intel Corporation 82801G (ICH7 Family) IDE Controller (rev 02)

Looking at the serial no. generated by the hdd the site shows :-

 http://www.samsung.com/Products/HardDiskDrive/SpinPointPSeries/HardDiskDrive_SpinPointPSeries_SP0802N.htm

Formatted Capacity = 80GB
Ultra ATA 133 compatible (Default : UDMA 100 Mode 5)

 This means my hdd support UDMA 5 then why it is not changeable ?

reading in wikipedia http://en.wikipedia.org/wiki/AT_Attachment it seems AT runs stuff at 16 bits only.
 [quote=wikipedia] Parallel ATA cables transfer data 16 bits at a time (it is a common misconception that they transfer 32 bits of data at a time, mainly because the 40 cable ribbon would appear to allow this). [/quote]

$sudo hdparm -I /dev/sda | grep dma
        DMA: mdma0 mdma1 mdma2 udma0 udma1 udma2 udma3 udma4 *udma5

$ sudo hdparm -tT /dev/sda

/dev/sda:
 Timing cached reads: 1620 MB in 2.00 seconds = 810.49 MB/sec
 Timing buffered disk reads: 28 MB in 3.20 seconds = 8.74 MB/sec

$ uname -a
Linux feisty 2.6.20-15-generic #2 SMP Sun Apr 15 07:36:31 UTC 2007 i686 GNU/Linux

:(

can't help on the topic, since don't have ubuntu installed, but
try to use sdparm for scsi devices (/dev/sd*) ?

sdparm says:

 sudo sdparm /dev/sda
Password:
    /dev/sda: ATA IC25N080ATMR04-0 MO4O
REQUEST SENSE failed, try again with '-v' option

sudo sdparm -v /dev/sda
    /dev/sda: ATA IC25N080ATMR04-0 MO4O
    Request Sense cmd: 03 00 00 00 40 00
request sense: Fixed format, current; Sense key: Illegal Request
 Additional sense: Invalid command operation code
REQUEST SENSE failed

2007/4/27, mator:
> can't help on the topic, since don't have ubuntu installed, but
> try to use sdparm for scsi devices (/dev/sd*) ?
>
> --
> poor performance hard disk IDE
> https://bugs.launchpad.net/bugs/96693
> You received this bug notification because you are a direct subscriber
> of the bug.

Hey, can anyone help me?

Unluckily I cannot. But what is clear is that this is a bug somewhere in between the EIDE drivers and the SCSI emulation.
I'm experiencing a similar problem. I cannot enable the 32bit mode and the multi sector I/O.

What I have done in order to try to have some better performances is to:
1. use tha maximum filesystem read ahead values (-a)
2. use the lookahead drive feature (-A)
3. enable the write cache (-W) (read the manual here)
4. enable the dma (-d)

In the meantime I hope the bug will be fixed.

Hi all,
       I did all the 3 given above, the last one gives :-

 /dev/sdb1:
 setting using_dma to 1 (on)
 HDIO_SET_DMA failed: Inappropriate ioctl for device
shirish@ubuntu ~> sudo hdparm -d1 /dev/sdb2

/dev/sdb2:
 setting using_dma to 1 (on)
 HDIO_SET_DMA failed: Inappropriate ioctl for device

 Guess would have try it to from Live CD or something or am I missing something?

 Anyway here's the output after doing all the above :-

sudo hdparm -tT /dev/sdb

/dev/sdb:
 Timing cached reads: 558 MB in 2.01 seconds = 278.15 MB/sec
 Timing buffered disk reads: 82 MB in 3.03 seconds = 27.08 MB/sec

 Don't see much of a difference from before.

One question: why are you using hdparm on the partition device (sdb1) and not on the disk device (sdb)?
I'm not sure it's the same.
In any case, with hdparm -i /dev/sdb can you see whether the UDMA is being used?

sudo hdparm -i /dev/sdb
Password:

/dev/sdb:

 Model=SAMSUNG SP0802N , FwRev=TK100-24, SerialNo=S00JJ10X390473
 Config={ HardSect NotMFM HdSw>15uSec Fixed DTR>10Mbs }
 RawCHS=16383/16/63, TrkSize=34902, SectSize=554, ECCbytes=4
 BuffType=DualPortCache, BuffSize=2048kB, MaxMultSect=16, MultSect=?16?
 CurCHS=4047/16/255, CurSects=16511760, LBA=yes, LBAsects=156368016
 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 udma3 udma4 udma5
 AdvancedPM=no WriteCache=enabled
 Drive conforms to: ATA/ATAPI-7 T13 1532D revision 0: ATA/ATAPI-1 ATA/ATAPI-2 ATA/ATAPI-3 ATA/ATAPI-4 ATA/ATAPI-5 ATA/ATAPI-6 ATA/ATAPI-7

 * signifies the current active mode

 I did a clean install recently, so this is the default udam2 but then as given above in the comments my hdd supports ATAPI-5

 I thought that was the way, but can do this way also.

sudo hdparm -tT /dev/sdb

/dev/sdb:
 Timing cached reads: 616 MB in 2.01 seconds = 307.18 MB/sec
 Timing buffered disk reads: 46 MB in 3.08 seconds = 14.94 MB/sec

 This is after doing it on the disk device rather than on the partition device. Not much of a difference.

 So there seems to be also a dma issue with the 2.6.20-15 .

Some more notes :-

 I removed the first HDD power link momentarily & made sure that its written as uninstalled the first HDD.& then booted from the CD. So this time it takes the 80 GB hdd as sda

sudo fdisk -l /dev/sda

Disk /dev/sda: 80.0 GB, 80060424192 bytes
255 heads, 63 sectors/track, 9733 cylinders
Units = cylinders of 16065 * 512 = 8225280 bytes

   Device Boot Start End Blocks Id System
/dev/sda1 * 1 1460 11727418+ 83 Linux
/dev/sda2 1461 9484 64452780 83 Linux
/dev/sda3 9485 9733 2000092+ 82 Linux swap / Solaris

sudo hdparm -d1 /dev/sdb

/dev/sdb:
 setting using_dma to 1 (on)
 HDIO_SET_DMA failed: Inappropriate ioctl for device

 Any ideas folks?

what means "Status: Fix Committed"?

I suppose someone has committed him/herself to fix the problem.

I think that was a tongue in cheek answer, it should mean that the fix would be in one of the upcoming kernel changes. Anyway here's my output from gutsy :-

sudo hdparm -tT /dev/sdb

/dev/sdb:
 Timing cached reads: 622 MB in 2.00 seconds = 310.37 MB/sec
 Timing buffered disk reads: 84 MB in 3.03 seconds = 27.77 MB/sec

Then did :-

 /dev/sdb:

 Model=SAMSUNG SP0802N , FwRev=TK100-24, SerialNo=S00JJ10X390473
 Config={ HardSect NotMFM HdSw>15uSec Fixed DTR>10Mbs }
 RawCHS=16383/16/63, TrkSize=34902, SectSize=554, ECCbytes=4
 BuffType=DualPortCache, BuffSize=2048kB, MaxMultSect=16, MultSect=?16?
 CurCHS=4047/16/255, CurSects=16511760, LBA=yes, LBAsects=156368016
 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: ATA/ATAPI-7 T13 1532D revision 0: ATA/ATAPI-1,2,3,4,5,6,7

 What is that MultiSect= ?16? about ?

Then did :-

sudo hdparm -a1 -A1 -d1 /dev/sdb
Password:

/dev/sdb:
 setting fs readahead to 1
 setting using_dma to 1 (on)
 HDIO_SET_DMA failed: Inappropriate ioctl for device
 setting drive read-lookahead to 1 (on)
 readahead = 0 (off)
 look-ahead = 1 (on)

 Now can anybody tell me the difference between readahead & look-ahead ? Which is better ?
 Also the DMA issue is still failing :( or should there be a different flag

 sudo hdparm -tT /dev/sdb

/dev/sdb:
 Timing cached reads: 622 MB in 2.00 seconds = 310.74 MB/sec
 Timing buffered disk reads: 42 MB in 3.09 seconds = 13.60 MB/sec
shirish@ubuntu:~$ sudo hdparm -i /dev/sdb

 As you can see the timing cached reads has improved a tiny bit but buffered disk reads have fallen behind

Lastly, there is this :-

sudo hdparm -i /dev/sdb

/dev/sdb:

 Model=SAMSUNG SP0802N , FwRev=TK100-24, SerialNo=S00JJ10X390473
 Config={ HardSect NotMFM HdSw>15uSec Fixed DTR>10Mbs }
 RawCHS=16383/16/63, TrkSize=34902, SectSize=554, ECCbytes=4
 BuffType=DualPortCache, BuffSize=2048kB, MaxMultSect=16, MultSect=?16?
 CurCHS=4047/16/255, CurSects=16511760, LBA=yes, LBAsects=156368016
 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: ATA/ATAPI-7 T13 1532D revision 0: ATA/ATAPI-1,2,3,4,5,6,7

 * signifies the current active mode

 The kernel is 2.6.22 rc something

 uname -a
Linux ubuntu 2.6.22-6-generic #1 SMP Fri Jun 1 19:24:12 GMT 2007 i686 GNU/Linux

 hdparm shows up as 7.1-2 ubuntu1

 aptitude show hdparm
Package: hdparm
State: installed
Automatically installed: no
Version: 7.1-2ubuntu1
Priority: standard
Section: admin
Maintainer: Ubuntu Core Developers <email address hidden>
Uncompressed Size: 258k
Depends: libc6 (>= 2.5-5)
Suggests: apmd
Replaces: apmd (<= 3.0.2-1.15)
Description: tune hard disk parameters for high performance
 Get/set hard disk parameters for Linux IDE drives. Primary use is for enabling
 irq-unmasking and IDE mult...

Same for me.
# uname -a
Linux bouleau 2.6.20-16-generic #2 SMP Thu Jun 7 20:19:32 UTC 2007 i686 GNU/Linux
# hdparm -i /dev/sda

/dev/sda:

 Model=ST3300620A , FwRev=3.AAE , SerialNo= 9QF33Q66
 Config={ HardSect NotMFM HdSw>15uSec Fixed DTR>10Mbs RotSpdTol>.5% }
 RawCHS=16383/16/63, TrkSize=0, SectSize=0, ECCbytes=4
 BuffType=unknown, BuffSize=16384kB, MaxMultSect=16, MultSect=?16?
 CurCHS=16383/16/63, CurSects=16514064, LBA=yes, LBAsects=268435455
 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 udma3 udma4 udma5
 AdvancedPM=no WriteCache=enabled
 Drive conforms to: Unspecified: ATA/ATAPI-1 ATA/ATAPI-2 ATA/ATAPI-3 ATA/ATAPI-4 ATA/ATAPI-5 ATA/ATAPI-6 ATA/ATAPI-7

 * signifies the current active mode

# hdparm -tT /dev/sda

/dev/sda:
 Timing cached reads: 776 MB in 2.00 seconds = 387.38 MB/sec
 Timing buffered disk reads: 88 MB in 3.04 seconds = 28.93 MB/sec

# hdparm -c1 /dev/sda

/dev/sda:
 setting 32-bit IO_support flag to 1
 HDIO_SET_32BIT failed: Invalid argument
 IO_support = 0 (default 16-bit)

I got the new kernel today 2.6.20-16ubuntu3 and I still don't see any progress
I think it's intel chipset. Does everyone here use a ich southbridge family?

because I have an additional ide card. a cheap silicon i680, and it blows the intel chipset away. Yet they are both being used with the new libata (which causes the sd* instead of hd* right? )

It's still not working in Gutsy with kernel

Linux version 2.6.22-14-generic (buildd@palmer) (gcc version 4.1.3 20070929 (prerelease) (Ubuntu 4.1.2-16ubuntu2)) #1 SMP Sun Oct 14 23:05:12 GMT 2007

Is there a way to turn the SCSI emulation off?

Could someone please blog or write up somewhere public exactly what is going on here. This whole thing of moving /dev/hdx to /dev/sdx just silently appeared and has caused a lot off issues. Firstly some people couldn't boot or mount drives without hacking /etc/fstab (luckily most of my drives were specified using UUIDs.) Secondly the hdparm and optimising performance issue. I have a perfectly good server that I just migrated to gutsy. The IDE drives went from /dev/hda to /dev/sda. But I no longer can turn on DMA nor change it from udma2 to udma5. I am seeing a huge amount of WAIT time in the CPU waiting for disk writes and reads to complete.

Someone must know exactly what has gone on here and should comment about it. It it extremely frustrating to have major decisions made and very little communication. I can't even find out anywhere that technically says what this change was about (is it SCSI emulation or something else)?

Someone will fix somewhere in the future.
The problem is here since months now and has not been fixed in Gutsy either.
I'm actually thinking to move to another distro just for this. Not the bug, the carelessness.
Maybe I'm wrong, but I'd focus resources in kernel issues like this one rather than,say, 3D desktop effects.

Not meaning to throw more wood into the fire but look at this bug report in Debian, specially the last paragraph:

http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=445615

Some people just are not prepared or do not want to increase its distribution userbase.

The same issue is still on Hardy, the kernel has changed to Oct 14 but still the 2.6.22 series things remain same :-

 uname -a

Linux Mugglewille 2.6.22-14-generic #1 SMP Sun Oct 14 23:05:12 GMT 2007 i686 GNU/Linux

hdparm has moved on to become 7.7-1ubuntu1 but no change :(

shirish@Mugglewille:~$ sudo hdparm -a1 -A1 -d1 /dev/sdb

/dev/sdb:
 setting fs readahead to 1
 setting using_dma to 1 (on)
 HDIO_SET_DMA failed: Inappropriate ioctl for device
 setting drive read-lookahead to 1 (on)
 readahead = 0 (off)
 look-ahead = 1 (on)

why readahead is set to off, shouldn't it be set on on ?

Same here for me on Gutsy with my IDE Drive and that SCSI-Emulation.

uname -a

2.6.22-14-generic #1 SMP Sun Oct 14 23:05:12 GMT 2007 i686 GNU/Linux

sudo hdparm -c1 /dev/sda

/dev/sda:
 setting 32-bit IO_support flag to 1
 HDIO_SET_32BIT failed: Invalid argument
 IO_support = 0 (default 16-bit)

It would be really GREAT if this could be solved.

I have a new PC, with SATA controller (JMicron 20360/20363 AHCI Controller (rev 02)) and drive (Hitachi HTS541616J9SA00).
The CPU is a Core 2 Duo, that is an "amd64" family CPU and the distribution is Gutsy up-to-date. Well, the hdparm says:

some1@uqbar:~$ sudo hdparm -c 1 -d 1 /dev/sda

/dev/sda:
 setting 32-bit IO_support flag to 1
 HDIO_SET_32BIT failed: Invalid argument
 setting using_dma to 1 (on)
 HDIO_SET_DMA failed: Inappropriate ioctl for device
 IO_support = 0 (default 16-bit)

In my humble opinion this would be unacceptable as SATA is going the "de facto" standard.
Is there a way to at least circumvent this problem, until a fix (that'd be committed) will come out?
Thanks.

Running kernel is Linux version 2.6.22-14-generic

Could sombody solve this problem? It is very big problem.

And also could you please someone solve the UDMA33 instead of UDMA100 problem

Hello,

I am a fedora user and I am having the same issue.
At some point, I think fedora 4 or 5, hda became sda
and I have had this issue since and still persist on fedora 8 with latest kernel.

I have gone to scsi drives on my servers

[root@dbox ~]# hdparm -tT /dev/sda

/dev/sda:
 Timing cached reads: 318 MB in 2.01 seconds = 158.33 MB/sec
 Timing buffered disk reads: 38 MB in 3.17 seconds = 11.98 MB/sec

Linux dbox 2.6.23.8-63.fc8 #1 SMP Wed Nov 21 18:51:08 EST 2007 i686 athlon i386 GNU/Linux

[root@dbox ~]# hdparm /dev/sda

/dev/sda:
 IO_support = 0 (default 16-bit)
 readonly = 0 (off)
 readahead = 256 (on)
 geometry = 4998/255/63, sectors = 80293248, start = 0
[root@dbox ~]# hdparm -a1 /dev/sda

[root@dbox ~]# hdparm -i /dev/sda

/dev/sda:

 Model=Maxtor 2F040J0 , FwRev=VAM51JJ0, SerialNo=F184DD7E
 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=80293248
 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: ATA/ATAPI-7 T13 1532D revision 0: ATA/ATAPI-1,2,3,4,5,6,7

 * signifies the current active mode

Hi All,

Just a few things:

Dtek, if you are experiencing issues with Fedora, it is best to open a bug report with their bug tracking system, not Ubuntu's.

This bug report was marked "Fix Committed" a while ago. Can the original bug reporter verify this bug has been resolved and we'll go ahead and mark this bug report as "Fix Released".

For those who still have an issue, please verify you have the same hardware as the original bug reporter otherwise open a new bug report. Additionally, if you open a new bug report regarding this issue, please make sure you have tested against the actively developed kernel. The Hardy Heron Alpha series was recently released which contains an updated version of the kernel. You can download and try the new Hardy Heron Alpha release from http://cdimage.ubuntu.com/releases/hardy/ . You should be able to then test the new kernel via the LiveCD. Also make sure your bug report is against the 'linux' source package. Starting with the Hardy release, the kernel source package naming convention changes from 'linux-source-2.6.x' to just 'linux'. Thanks.

Hi all,
  I just did a clean install just couple of days ago. Now on Intrepid but the issue persists, infacts this has become insane.

Default timings

 sudo hdparm -tT /dev/sda

/dev/sda:
 Timing cached reads: 426 MB in 2.00 seconds = 212.75 MB/sec
 Timing buffered disk reads: 86 MB in 3.04 seconds = 28.26 MB/sec

Setting the hdd's for best performance

sudo hdparm -a1 -A1 -d1 /dev/sda

/dev/sda:
 setting fs readahead to 1
 setting using_dma to 1 (on)
 HDIO_SET_DMA failed: Inappropriate ioctl for device
 setting drive read-lookahead to 1 (on)
 HDIO_GET_DMA failed: Inappropriate ioctl for device
 readahead = 0 (off)
 look-ahead = 1 (on)

Again doing the test

sudo hdparm -tT /dev/sda

/dev/sda:
 Timing cached reads: 398 MB in 2.00 seconds = 198.59 MB/sec
 Timing buffered disk reads: 30 MB in 3.13 seconds = 9.58 MB/sec

It has actually has come down :(

This is on kernel 2.6.27-7-generic

Just for info. here are the hard disk details again :-

sudo hdparm -i /dev/sda

/dev/sda:

 Model=SAMSUNG SP0802N , FwRev=TK100-24, SerialNo=S00JJ10X390473
 Config={ HardSect NotMFM HdSw>15uSec Fixed DTR>10Mbs }
 RawCHS=16383/16/63, TrkSize=34902, SectSize=554, ECCbytes=4
 BuffType=DualPortCache, BuffSize=2048kB, MaxMultSect=16, MultSect=?16?
 CurCHS=4047/16/255, CurSects=16511760, LBA=yes, LBAsects=156368016
 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: ATA/ATAPI-7 T13 1532D revision 0: ATA/ATAPI-1,2,3,4,5,6,7

 * signifies the current active mode

The 18 month support period for Feisty Fawn 7.04 has reached it's end of life - http://www.ubuntu.com/news/ubuntu-7.04-end-of-life . As a result we are closing the linux-source-2.6.20 task. However, this will remain open against the actively developed kernel.

@Shrish, can you post the output of 'cat /proc/version_signature' as well as attach your dmesg output. Thanks.

Hi all,
 Here's my stuff.

$ cat /proc/version_signature
Ubuntu 2.6.27-9.19-generic

As far as dmesg is concerned, gimme 5 minutes.
--
          Regards,
          Shirish Agarwal
  This email is licensed under http://creativecommons.org/licenses/by-nc/3.0/
http://flossexperiences.wordpress.com
065C 6D79 A68C E7EA 52B3 8D70 950D 53FB 729A 8B17

Here's the output of cat /proc/version_signature

$ cat /proc/version_signature
Ubuntu 2.6.27-9.19-generic

and attached is the dmesg output

This output is without the second hard disk i.e. the samsung one.

Tomorrow will give you the one after attaching the samsung one.

Hi all,
 Leann Ogasawara

Is there possibility of some of the problem might be due to the libata driver

http://ata.wiki.kernel.org/index.php/Software_status

[quote]

 PATA support

libata is not specific to Serial ATA. It works for Parallel ATA (a.k.a. IDE) as well. Certain SATA controllers such as Promise and SiS include PATA ports on their SATA controllers.

libata has gained full support for PATA, including older chipsets and devices with buggy/problematic designs. Alan Cox has done a lot of work on this. PATA is be fully supported, alongside SATA. libata PATA drivers exist for almost all PATA chipsets.

libata PATA drivers will not be turned on by default for a long time, to avoid conflicts with CONFIG_IDE (the traditional IDE driver).

[/quote]

I got this while I was trying

$ sudo hdparm -c1 /dev/sda

/dev/sda:
 setting 32-bit IO_support flag to 1
 HDIO_SET_32BIT failed: Invalid argument
 IO_support = 0 (default)

This is how my drive stacks up . This is a seagate 160 GiB

$ sudo hdparm -i /dev/sda
[sudo] password for shirish:

/dev/sda:

 Model=ST3160021A , FwRev=8.01 , SerialNo=4JS26D4P
 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=?16?
 CurCHS=4047/16/255, CurSects=16511760, LBA=yes, LBAsects=312581808
 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: ATA/ATAPI-6 T13 1410D revision 2: ATA/ATAPI-1,2,3,4,5,6

 * signifies the current active mode