FastTrack 133 "Lite" or why there is no /dev/hda

[Martin Mewes]

Hi folks,

recently I installed Fedore Core 1 on a PC with a FastTrack 133 "Lite" 
on it.

<14:33:12> macmewes@linux:~ $ df
Filesystem           1K-Blöcke   Benutzt Verfügbar Ben% Eingehängt auf
/dev/hdg2             79048292   6264524  68768324   9% /
/dev/hdg1               101086      8337     87530   9% /boot
/dev/hdh1             77348004    375092  73043820   1% /home
none                    775416         0    775416   0% /dev/shm

This is a parallel installation with Windows XP which is on the first 
two harddisks which I would see as hda and hdb if Fedora Kernels came 
with NTFS-Support for mounting them into my system for reading only.

I do not want to start a chit-chat about why this feature is not in 
Fedore-Kernels, but what I wonder about is if Windows XP resides on 
hda and hdb why Fedora (as well as SuSE, RedHat, Knoppix, Debian ...) 
is starting to acutally see the harddisks not with hda but hde?

A "demsg"-extract for you ...

[...]

Uniform Multi-Platform E-IDE driver Revision: 7.00beta4-2.4
ide: Assuming 33MHz system bus speed for PIO modes; override with 
idebus=xx
SiI3112 Serial ATA: IDE controller at PCI slot 00:08.0
SiI3112 Serial ATA: chipset revision 2
SiI3112 Serial ATA: not 100% native mode: will probe irqs later
    ide0: MMIO-DMA , BIOS settings: hda:pio, hdb:pio
    ide1: MMIO-DMA , BIOS settings: hdc:pio, hdd:pio
PDC20276: IDE controller at PCI slot 00:0f.0
PDC20276: chipset revision 1
PDC20276: not 100% native mode: will probe irqs later
    ide2: BM-DMA at 0xc400-0xc407, BIOS settings: hde:pio, hdf:pio
    ide3: BM-DMA at 0xc408-0xc40f, BIOS settings: hdg:pio, hdh:pio
VP_IDE: IDE controller at PCI slot 00:11.1
VP_IDE: chipset revision 6
VP_IDE: not 100% native mode: will probe irqs later
ide: Assuming 33MHz system bus speed for PIO modes; override with 
idebus=xx
VP_IDE: VIA vt8235 (rev 00) IDE UDMA133 controller on pci00:11.1
    ide4: BM-DMA at 0xd400-0xd407, BIOS settings: hdi:pio, hdj:pio
    ide5: BM-DMA at 0xd408-0xd40f, BIOS settings: hdk:DMA, hdl:pio
hda: no response (status = 0xfe)
hdc: no response (status = 0xfe)
hde: IC35L090AVV207-0, ATA DISK drive
hdf: IC35L090AVV207-0, ATA DISK drive
blk: queue c03ff0f8, I/O limit 4095Mb (mask 0xffffffff)
blk: queue c03ff238, I/O limit 4095Mb (mask 0xffffffff)
hdg: IC35L090AVV207-0, ATA DISK drive
hdh: IC35L090AVV207-0, ATA DISK drive
blk: queue c03ff554, I/O limit 4095Mb (mask 0xffffffff)
blk: queue c03ff694, I/O limit 4095Mb (mask 0xffffffff)
hdk: DV-516E, ATAPI CD/DVD-ROM drive
ide2 at 0xb400-0xb407,0xb802 on irq 10
ide3 at 0xbc00-0xbc07,0xc002 on irq 10
ide5 at 0x170-0x177,0x376 on irq 15
hde: attached ide-disk driver.
hde: host protected area => 1
hde: 160836480 sectors (82348 MB) w/1821KiB Cache, CHS=10011/255/63, 
UDMA(100)
hdf: attached ide-disk driver.
hdf: host protected area => 1
hdf: 160836480 sectors (82348 MB) w/1821KiB Cache, CHS=10011/255/63, 
UDMA(100)
hdg: attached ide-disk driver.
hdg: host protected area => 1
hdg: 160836480 sectors (82348 MB) w/1821KiB Cache, CHS=10011/255/63, 
UDMA(100)
hdh: attached ide-disk driver.
hdh: host protected area => 1
hdh: 160836480 sectors (82348 MB) w/1821KiB Cache, CHS=10011/255/63, 
UDMA(100)
Partition check:
 hde: hde1
 hdf: hdf1
 hdg: hdg1 hdg2
 hdh: hdh1 hdh2
ide: late registration of driver.

[...]

Any hints?

bis dahin/kind regards

Martin Mewes

-- 
http://www.mewes.tv/
Official Webmin/Usermin Translation Co-Ordinator 2003/2004
######################################################################

[Alan Cox]

On Sun, Mar 21, 2004 at 02:40:22PM +0100, Martin Mewes wrote:
>  Fedore-Kernels, but what I wonder about is if Windows XP resides on 
>  hda and hdb why Fedora (as well as SuSE, RedHat, Knoppix, Debian ...) 
>  is starting to acutally see the harddisks not with hda but hde? 
hda/b and hdc/d are reserved for the legacy mode IDE ports. If your 
system is set up with no legacy IDE ports then the hda-d ports will never
be used.

Alan

[Martin Mewes]

Hi Alan,

Am Sonntag, 21. März 2004 14:48 schrieb Alan Cox:
>  On Sun, Mar 21, 2004 at 02:40:22PM +0100, Martin Mewes wrote:
>  > Fedore-Kernels, but what I wonder about is if Windows XP resides
>  > on hda and hdb why Fedora (as well as SuSE, RedHat, Knoppix,
>  > Debian ...) is starting to acutally see the harddisks not with
>  > hda but hde?
>
>  hda/b and hdc/d are reserved for the legacy mode IDE ports. If your
>  system is set up with no legacy IDE ports then the hda-d ports will
>  never be used. 
Thanks for this info which leads to the next question.

In this scenario: Is it possible to quicken the boot-process so that 
checking for hd/a-d is skipped?

bis dahin/kind regards

Martin Mewes

-- 
http://www.mewes.tv/
Official Webmin/Usermin Translation Co-Ordinator 2003/2004
######################################################################

[Alan Cox]

On Sun, Mar 21, 2004 at 03:46:53PM +0100, Martin Mewes wrote:
>  In this scenario: Is it possible to quicken the boot-process so that
>
>  checking for hd/a-d is skipped? 
They should only be checked if the BIOS indicates they are present. If
you are getting long pauses you might want to report a bug to the
2.6 kernel IDE maintainer

[Martin Mewes]

Hi Alan,

Am Sonntag, 21. März 2004 16:51 schrieb Alan Cox:
>  On Sun, Mar 21, 2004 at 03:46:53PM +0100, Martin Mewes wrote:
>  > In this scenario: Is it possible to quicken the boot-process so
>  > that checking for hd/a-d is skipped?
>
>  They should only be checked if the BIOS indicates they are present.
>  If you are getting long pauses you might want to report a bug to
>  the 2.6 kernel IDE maintainer 
They are not present in the BIOS because the IDE-Drives are handled by 
the IDE-Raid-Hardware. Looks like something to work on.

Thanks again :-)

bis dahin/kind regards

Martin Mewes

-- 
http://www.mewes.tv/
Official Webmin/Usermin Translation Co-Ordinator 2003/2004
######################################################################

[Chadley Wilson]

On Sun, 2004-03-21 at 18:30, Martin Mewes wrote:
>  Hi Alan,
>  
>  Am Sonntag, 21. März 2004 16:51 schrieb Alan Cox:
>  > On Sun, Mar 21, 2004 at 03:46:53PM +0100, Martin Mewes wrote:
>  > > In this scenario: Is it possible to quicken the boot-process so
>  > > that checking for hd/a-d is skipped?
>  >
>  > They should only be checked if the BIOS indicates they are present.
>  > If you are getting long pauses you might want to report a bug to
>  > the 2.6 kernel IDE maintainer
>  
>  They are not present in the BIOS because the IDE-Drives are handled by 
>  the IDE-Raid-Hardware. Looks like something to work on.
>   
IIRC 
I had four serial ATA drives in a system last week that was running on
the promise controller these drives weren't detected in the bios but but
by the controller, I had to make a driver disk for the first
installation and then I couldn't find the device on /dev/hd- after some
playing they were actually on /dev/sda b c and d.

Could be the same for the fasttrack, I don't know.


Chadley Wilson

Production PLanner / Analyst and Supervisor




A division of Technobyte (Pty) Ltd.    Registration No. 1993/00917/07
Email: chadley(a)pinteq.co.za

Tel: [phone] (Switchboard) or 265 3195 (Direct) 083 342 3730
(Cell)
Fax: + 27 11 265 3073
Web site: www.pinnacle.co.za


WARNING: The information transmitted herewith is intended only for the
person or entity to which it is addressed and may contain confidential
and/or privileged material. Any review, retransmission, dissemination or
other use of, or taking of any action in reliance upon, this information
by persons or entities other than the intended recipient is prohibited.
If you received this in error, please contact the sender and delete all
copies of the material.

[Martin Mewes]

Hi Chadley,

Am Sonntag, 21. März 2004 18:19 schrieb Chadley Wilson:
>  On Sun, 2004-03-21 at 18:30, Martin Mewes wrote:
>  > They are not present in the BIOS because the IDE-Drives are
>  > handled by the IDE-Raid-Hardware. Looks like something to work
>  > on.
>
>  IIRC
>  I had four serial ATA drives in a system last week that was running
>  on the promise controller these drives weren't detected in the bios
>  but but by the controller, I had to make a driver disk for the
>  first installation and then I couldn't find the device on /dev/hd-
>  after some playing they were actually on /dev/sda b c and d. 
Actually this is a Promise-Chip as well ;-)

Following the homepage starting with

http://www.promise.com/support/download/download_eng.asp

which driver would be the closest one?
I cannot find a FastTrack 133 "Lite". On my Motherboard-CD I only find 
drivers for SuSE. Maybe you can share your knowledge?

bis dahin/kind regards

Martin Mewes

-- 
http://www.mewes.tv/
Official Webmin/Usermin Translation Co-Ordinator 2003/2004
######################################################################

[Chadley Wilson]

On Sun, 2004-03-21 at 20:17, Martin Mewes wrote:
>  Hi Chadley,
>  
>  Am Sonntag, 21. März 2004 18:19 schrieb Chadley Wilson:
>  > On Sun, 2004-03-21 at 18:30, Martin Mewes wrote:
>  > > They are not present in the BIOS because the IDE-Drives are
>  > > handled by the IDE-Raid-Hardware. Looks like something to work
>  > > on.
>  >
>  > IIRC
>  > I had four serial ATA drives in a system last week that was running
>  > on the promise controller these drives weren't detected in the bios
>  > but but by the controller, I had to make a driver disk for the
>  > first installation and then I couldn't find the device on /dev/hd-
>  > after some playing they were actually on /dev/sda b c and d.
>  
>  Actually this is a Promise-Chip as well ;-) 
Quite right I couldn't remember
Sorry!

[Martin Mewes]

Hi Chadley,

Am Sonntag, 21. März 2004 19:21 schrieb Chadley Wilson:
>  On Sun, 2004-03-21 at 20:17, Martin Mewes wrote: 
>  > Actually this is a Promise-Chip as well ;-)
>
>  Quite right I couldn't remember Sorry! 
Anyone else some hints for me, before I file this to linux-ide?

bis dahin/kind regards

Martin Mewes

-- 
http://www.mewes.tv/
Official Webmin/Usermin Translation Co-Ordinator 2003/2004
######################################################################

[Alexander Dalloz]

Am Mo, den 22.03.2004 schrieb Martin Mewes um 19:23:

>  Anyone else some hints for me, before I file this to linux-ide?
>  
>  bis dahin/kind regards
>  
>  Martin Mewes 
Could you please again explain what your point is?

You want to have the Promise controller to address to hda up to hdd? For
cosmetic reasons only? Alan answered that already: hda-hdd are reserved
and thus additional controllers address different to not conflict. If
you want to minimize bootup time, causing the system not to check for
disks on then legacy controller ports, just check your motherboard BIOS
settings. Normally each BIOS has a setting where you can enable and
disable the onboard controller channels (primary and secondary).

Alexander


-- 
Alexander Dalloz | Enger, Germany | GPG key 1024D/ED695653 1999-07-13
Fedora GNU/Linux Core 1 (Yarrow) on Athlon CPU kernel 2.4.22-1.2174.nptl
Sirendipity 19:33:07 up 3 days, 4:15, load average: 0.11, 0.19, 0.12 
                   [ Γνωθι σ'αυτον - gnothi seauton ]
             my life is a planetarium - and you are the stars

[Martin Mewes]

Hi,

Am Montag, 22. März 2004 19:38 schrieb Alexander Dalloz:

>  Could you please again explain what your point is? 
The point is, that even if BIOS reports neither Legacy-IDE-HDDs nor 
SATA, Linux sees something on the motherboard and starts looking for 
HDDs beginning with hda.

I rechecked the BIOS and the only thing which is up, is the 
IDE-RAID-Stuff for the Promise Chip.

This is what confuses me somewhat.

bis dahin/kind regards

Martin Mewes

-- 
http://www.mewes.tv/
Official Webmin/Usermin Translation Co-Ordinator 2003/2004
######################################################################

[Alexander Dalloz]

Am Mo, den 22.03.2004 schrieb Martin Mewes um 20:25:
>  Hi,
>  
>  Am Montag, 22. März 2004 19:38 schrieb Alexander Dalloz:
>  
>  > Could you please again explain what your point is?
>  
>  The point is, that even if BIOS reports neither Legacy-IDE-HDDs nor 
>  SATA, Linux sees something on the motherboard and starts looking for 
>  HDDs beginning with hda.
>  
>  I rechecked the BIOS and the only thing which is up, is the 
>  IDE-RAID-Stuff for the Promise Chip.
>  
>  This is what confuses me somewhat. 
Ok, that I can follow and understand confusion.

From your initial posting:

SiI3112 Serial ATA: IDE controller at PCI slot 00:08.0
SiI3112 Serial ATA: chipset revision 2
SiI3112 Serial ATA: not 100% native mode: will probe irqs later
    ide0: MMIO-DMA , BIOS settings: hda:pio, hdb:pio
    ide1: MMIO-DMA , BIOS settings: hdc:pio, hdd:pio

--> Sil3112 is obviously the onboard controller. Seems that your BIOS
can not set this controller to off.

PDC20276: IDE controller at PCI slot 00:0f.0
PDC20276: chipset revision 1
PDC20276: not 100% native mode: will probe irqs later
    ide2: BM-DMA at 0xc400-0xc407, BIOS settings: hde:pio, hdf:pio
    ide3: BM-DMA at 0xc408-0xc40f, BIOS settings: hdg:pio, hdh:pio

--> PDC20276 is the Promise controller.

VP_IDE: IDE controller at PCI slot 00:11.1
VP_IDE: chipset revision 6
VP_IDE: not 100% native mode: will probe irqs later
ide: Assuming 33MHz system bus speed for PIO modes; override with 
idebus=xx
VP_IDE: VIA vt8235 (rev 00) IDE UDMA133 controller on pci00:11.1
    ide4: BM-DMA at 0xd400-0xd407, BIOS settings: hdi:pio, hdj:pio
    ide5: BM-DMA at 0xd408-0xd40f, BIOS settings: hdk:DMA, hdl:pio

--> VP_IDE is a third IDE controller.

All else I can say is that the IDE device naming scheme with the kernel
is fix, so onboard controllers come first as hda - hdd. IIRC the kernel
option "boot off-board controller first" does not change that at all.

>  bis dahin/kind regards
>  
>  Martin Mewes 
Alexander


-- 
Alexander Dalloz | Enger, Germany | GPG key 1024D/ED695653 1999-07-13
Fedora GNU/Linux Core 1 (Yarrow) on Athlon CPU kernel 2.4.22-1.2174.nptl
Sirendipity 21:55:29 up 3 days, 6:37, load average: 0.03, 0.20, 0.16 
                   [ Γνωθι σ'αυτον - gnothi seauton ]
             my life is a planetarium - and you are the stars

[Nicolas Mailhot]

Le lun, 22/03/2004 à 22:04 +0100, Alexander Dalloz a écrit :


>  All else I can say is that the IDE device naming scheme with the
> kernel
>  is fix, so onboard controllers come first as hda - hdd. IIRC the kernel
>  option "boot off-board controller first" does not change that at all. 
This is false btw - I have a system were hda-d are on an addon card
(sii680) and e-h are the built-in controllers (via8235)

(Which is perfect for me btw since the built-in controllers are used by
dvd/cdrom/legacy OS and  ssi680 is the linux raid array)

It did require fun grub stuff like
rootnoverify (hd2,0)
map (hd2) (hd0)
to keep the legacy OS bootable, since it ennumerates the drives the
other way.

Cheers,

-- 
Nicolas Mailhot

[Martin Mewes]

Hi Nicolas,

Nicolas Mailhot <Nicolas.Mailhot(a)laPoste.net&gt; wrote:

> It did require fun grub stuff like
> rootnoverify (hd2,0)
> map (hd2) (hd0)
> to keep the legacy OS bootable, since it ennumerates the drives the
> other way. 
Nice catch. I did not think about this ...

But how to deal with this if you do an initial installation of Linux
as the second OS?

In my case I installed Windows XP on the first two IDEs with a device
driver from diskette in order for Windows XP to see the Promise-Chip.
Fedore Core 1 came on as the second OS and initially the HDDs were
named hde, hdf, hdg, hdh.

I think I can remap the HDDs the way you suggest, but this would not
help in initial installations, does it?


bis dahin - kind regards

Martin Mewes

-- 
http://webmin.mamemu.de/
Official Webmin/Usermin Translation Co-Ordinator 2003/2004
Proud Forte Agent 2.0 Beta Tester

[Nicolas Mailhot]

Le mar, 23/03/2004 à 08:05 +0100, Martin Mewes a écrit :

>  Hi Nicolas,
>  
>  Nicolas Mailhot <Nicolas.Mailhot(a)laPoste.net&gt; wrote:
>  
>  >It did require fun grub stuff like
>  >rootnoverify (hd2,0)
>  >map (hd2) (hd0)
>  >to keep the legacy OS bootable, since it ennumerates the drives the
>  >other way.
>  
>  Nice catch. I did not think about this ...
>  
>  But how to deal with this if you do an initial installation of Linux
>  as the second OS? 
As far as I know grub and linux agree on the drive numbering, so you
just consider linux numbers as the "right" ones and remap drives in grub
before booting in Windows. But I do'nt have the same hardware as you, so
YMMV.

Cheers,

-- 
Nicolas Mailhot