Hello,
Dear Alan Cox,
Dear Jeff Garzik,
Everyone a happy new year!
We are in stable Kernel 2.6.14.5, year 2006.
Since 1997 people asking on several list for a functional PDC202XX Driver.
Since some years I spend hours and hours finding a solution for a stable driver.
(PDC20269/Promise Ultra133 TX2)
There seem to be none.
There are two Solution:
-The IDE Driver (pdc202xx_new) has still problems with "DMA Timeout".
-The Libata Driver (pata_pdc2027x) seems to be still somewhat experimental.
Unfortunatly I'am not a kernel developper else there would be probably already
a solution ;-)
So what is the Solution for the PDC20269 Ultra ATA Controller?
I would mark this bold on the wishlist for 2006 ;-)
Thanx
=========================================================
Anyone wants a small summary of the halfworking solutions
Way 1 - IDE pdc202xx_new --> /dev/hdxx )
------------------------------------------
Just compile your kernel with a module (or built in):
$make menuconfig
-> Device Drivers
-> ATA/ATAPI/MFM/RLL support
-> ATA/ATAPI/MFM/RLL support (IDE [=y])
-> Enhanced IDE/MFM/RLL disk/cdrom/tape/floppy support (BLK_DEV_IDE [=y])
-> PCI IDE chipset support (BLK_DEV_IDEPCI [=y])
-> Generic PCI bus-master DMA support (BLK_DEV_IDEDMA_PCI
->PROMISE PDC202{68|69|70|71|75|76|77} support
$make-kpkg kernel_image
Problems:)
hde: dma_timer_expiry: dma status == 0x60
hde: DMA timeout retry
PDC202XX: Primary channel reset.
hde: timeout waiting for DMA
hde: status error: status=0x58 { DriveReady SeekComplete DataRequest }
ide: failed opcode was: unknown
hde: drive not ready for command
hde: status error: status=0x50 { DriveReady SeekComplete }
ide: failed opcode was: unknown
hde: no DRQ after issuing MULTWRITE_EXT
hde: status error: status=0x50 { DriveReady SeekComplete }
ide: failed opcode was: unknown
hde: no DRQ after issuing MULTWRITE_EXT
hde: status error: status=0x50 { DriveReady SeekComplete }
ide: failed opcode was: unknown
hdf: DMA disabled
PDC202XX: Primary channel reset.
hde: no DRQ after issuing MULTWRITE_EXT
ide2: reset: success
hde: dma_timer_expiry: dma status == 0x20
hde: DMA timeout retry
PDC202XX: Primary channel reset.
hde: timeout waiting for DMA
or similar
Way 2 - libata (emulates SCSI device) --> /dev/sdxx )
-------------------------------------------------------
Get the latest libata patch/kernel
http://www.kernel.org/pub/linux/kernel/people/jgarzik/patchkits/2.6/2.6.15-rc4-libata1.patch.bz2
ftp://ftp.kernel.org/pub/linux/kernel/v2.6/testing/linux-2.6.15-rc5.tar.bz2
....(untar)...
$make menuconfig
-> Device Drivers
-> SCSI device support
-> SCSI device support (SCSI [=m])
-> SCSI low-level drivers
-> Serial ATA (SATA) support (SCSI_SATA [=m])
->SCSI_PATA_PDC2027X
Enable PATA as suggested here:
-http://lkml.org/lkml/2004/8/1/100
-http://seclists.org/lists/linux-kernel/2005/May/0503.html
In include/linux/libata.h
#define ATA_ENABLE_ATAPI /* undefine to disable ATAPI support */
#define ATA_ENABLE_PATA /* define to enable PATA support in some
* low-level drivers */
Compile and pray!
With some harddisk it works without problem (seagate 7200.8)
With others Maxtor Maxline III it's not working at all:
Kernel Panics even at boottime:
ATA1: Abnormal status 0x8 on port 0xD...........
or Lost Interrupts...
or the drivers seems to be loaded (lsmod) but there are no
harddiscs under /prov/scsi/... /dev/sdxx
--
_______________________________________________
The coolest e-mail address on the web and it?s FREE! Sign-up today for Beer Mail @ beer.com.
On Sunday 01 January 2006 17:39, 1qay beer wrote:
> Hello,
> Dear Alan Cox,
> Dear Jeff Garzik,
>
> Everyone a happy new year!
>
> We are in stable Kernel 2.6.14.5, year 2006.
> Since 1997 people asking on several list for a functional PDC202XX Driver.
> Since some years I spend hours and hours finding a solution for a stable
> driver. (PDC20269/Promise Ultra133 TX2)
> There seem to be none.
>
> There are two Solution:
> -The IDE Driver (pdc202xx_new) has still problems with "DMA Timeout".
> -The Libata Driver (pata_pdc2027x) seems to be still somewhat experimental.
>
> Unfortunatly I'am not a kernel developper else there would be probably
> already a solution ;-)
>
> So what is the Solution for the PDC20269 Ultra ATA Controller?
> I would mark this bold on the wishlist for 2006 ;-)
I own this controller and it's been working with the kernel driver since
2002.. I'm not sure what you expect people to do about clearly faulty
hardware. Just buy a new one?
--
Cheers,
Alistair.
'No sense being pessimistic, it probably wouldn't work anyway.'
Third year Computer Science undergraduate.
1F2 55 South Clerk Street, Edinburgh, UK.
On 01/01/2006 09:39 AM, 1qay beer wrote:
> Since 1997 people asking on several list for a functional PDC202XX Driver.
> Since some years I spend hours and hours finding a solution for a stable driver.
> (PDC20269/Promise Ultra133 TX2)
> There seem to be none.
It's been working for me since 2003 as well on a production box, even
through migration to 2.6.
Get a new board.
--
Joshua Kwan
On Sun, Jan 01, 2006 at 01:22:41PM -0800, Joshua Kwan wrote:
> On 01/01/2006 09:39 AM, 1qay beer wrote:
> > Since 1997 people asking on several list for a functional PDC202XX Driver.
> > Since some years I spend hours and hours finding a solution for a stable driver.
> > (PDC20269/Promise Ultra133 TX2)
> > There seem to be none.
>
> It's been working for me since 2003 as well on a production box, even
> through migration to 2.6.
>
> Get a new board.
>
I can confirm this too.
Works 100% OK for me too, in fact has worked for many years already.. currently
using 2.6 series. *knocking wood* :)
-- Pasi K?rkk?inen
^
. .
Linux
/ - \
Choice.of.the
.Next.Generation.
Add my voice to the chorus of people who are having no problems with
their PDC20269's, for several years now (since 2002, I think -- and I
didn't have problems before then, I just didn't have a PDC20269
before).
--
-Barry K. Nathan <[email protected]>
On Sul, 2006-01-01 at 13:39 -0400, 1qay beer wrote:
> Hello,
> Dear Alan Cox,
> Dear Jeff Garzik,
>
> Everyone a happy new year!
Ditto
> -The IDE Driver (pdc202xx_new) has still problems with "DMA Timeout".
The legacy IDE layer is handled by Bartlomiej so you should direct your
enquiries and requests to him and the linux-ide list.
> -The Libata Driver (pata_pdc2027x) seems to be still somewhat experimental.
and while I'm working on libata pata a fair bit the pdc202xx driver is
the excellent work of Albert Lee.
The 20269 has always shown up as a problem for some users but not for
most. Nobody ever really got to the bottom of it to be honest. Please
send Albert your reports about which hardware works and which fails
(<[email protected]>) as it may be very useful. In particular the
drive that works appears to be UDMA 100 and the failing one UDMA 133
Alan
Thanx to everyone.
The Controller Board is fine.
What i noticed is that if you have the controller in a
full PCI Bus Master Slot without IRQ sharing - the Controller works better
than in shared slot.
Generally the first PCI Slot is shared with AGP.
The Second is normaly complet "IRQ share free" and the way to go.
Slot 4/5 is mostly shared (as long you have that many slots).
Slot 3 may be shared with onboard stuff like sound, onboard graphic, usb or others.
The Board is now working without Problems.
What I did is Remove the IDE-Driver from the Kernel and use the Libata Driver,
which now even works with S.M.A.R.T. without Problem.
I removed the UDMA-133 Drive and have now only connected UDMA-100 Drives to the Controller. The UDMA-133 is connected to the Motherboard now.
Mostly Maxtor produced UDMA-133 Drives which are going to fail in half a year anyway ;-(.
Now i use only Seagate Drives which have 5 Years warranty and I never had Problem with them. (too bad maxtor is part of seagate now...I hope they will not fail in their good quality)
Cheers
----- Original Message -----
From: "Alan Cox" <[email protected]>
To: "1qay beer" <[email protected]>
Subject: Re: Wish for 2006 to Alan Cox and Jeff Garzik: A functional Driver for PDC202XX
Date: Mon, 02 Jan 2006 22:35:51 +0000
>
> On Sul, 2006-01-01 at 13:39 -0400, 1qay beer wrote:
> > Hello,
> > Dear Alan Cox,
> > Dear Jeff Garzik,
> >
> > Everyone a happy new year!
>
> Ditto
>
> > -The IDE Driver (pdc202xx_new) has still problems with "DMA Timeout".
>
> The legacy IDE layer is handled by Bartlomiej so you should direct your
> enquiries and requests to him and the linux-ide list.
>
> > -The Libata Driver (pata_pdc2027x) seems to be still somewhat experimental.
>
> and while I'm working on libata pata a fair bit the pdc202xx driver is
> the excellent work of Albert Lee.
>
> The 20269 has always shown up as a problem for some users but not for
> most. Nobody ever really got to the bottom of it to be honest. Please
> send Albert your reports about which hardware works and which fails
> (<[email protected]>) as it may be very useful. In particular the
> drive that works appears to be UDMA 100 and the failing one UDMA 133
>
> Alan
--
_______________________________________________
The coolest e-mail address on the web and it?s FREE! Sign-up today for Beer Mail @ beer.com.