BUGFIX: UDMA-SiS5513 chipset support

BUGFIX: UDMA-SiS5513 chipset support

[Frank Neuber]

[-- Attachment #1: Type: TEXT/PLAIN, Size: 764 bytes --]

Hi,
recently I did an upgrade of my old computer (ASUS SP97-V) to
the kernel-2.4.7.

Problem:
  My system was crashing even when I load the module ide-disk.o
Solution:
  Because of the broken hardware (some UDMA problems with an non
  UDMA-Cabel to the drive) the linux kernel hangs during ide_dma_check.
  Even if UDMA is disabeld in the bios, the kernel detect this drive as
  an udma drive. And this is wrong!!
  My solution was simply to comment out the ide_dma_check in ide.c.
  You can find this patch as attachment.

Frank	

-- 
     _/_/_/_/ _//   _/ Frank Neuber
    _/       _/_/  _/  frank.neuber@gmx.de (private)
   _/_/_/   _/ _/ _/
  _/       _/  _/_/    neuber@opensource-systemberatung.de
 _/       _/    // http://www.opensource-systemberatung.de

[-- Attachment #2: ide-sis.patch --]
[-- Type: TEXT/PLAIN, Size: 421 bytes --]

--- ide.c	Sat Aug 18 19:46:08 2001
+++ ide-sis.c	Sat Aug 18 19:45:51 2001
@@ -3431,7 +3431,7 @@
 			 *   PARANOIA!!!
 			 */
 			(void) (HWIF(drive)->dmaproc(ide_dma_off_quietly, drive));
-			(void) (HWIF(drive)->dmaproc(ide_dma_check, drive));
+			//(void) (HWIF(drive)->dmaproc(ide_dma_check, drive));
 		}
 		drive->dsc_overlap = (drive->next != drive && driver->supports_dsc_overlap);
 		drive->nice1 = 1;

Re: BUGFIX: UDMA-SiS5513 chipset support

[Andre Hedrick]

Frank,

Not the correct solution, you should not set autodma.
This will prevent DMA from being used at INIT; however, you can still
attempt it later.  This patch is an invalid solution, period.

Regards,

Andre Hedrick
CTO ASL, Inc.
Linux ATA Development
-----------------------------------------------------------------------------
ASL, Inc.                                     Tel: (510) 857-0055
38875 Cherry Street                           Fax: (510) 857-0010
Newark, CA 94560                              Web: www.aslab.com

On Sat, 18 Aug 2001, Frank Neuber wrote:

> Hi,
> recently I did an upgrade of my old computer (ASUS SP97-V) to
> the kernel-2.4.7.
> 
> Problem:
>   My system was crashing even when I load the module ide-disk.o
> Solution:
>   Because of the broken hardware (some UDMA problems with an non
>   UDMA-Cabel to the drive) the linux kernel hangs during ide_dma_check.
>   Even if UDMA is disabeld in the bios, the kernel detect this drive as
>   an udma drive. And this is wrong!!
>   My solution was simply to comment out the ide_dma_check in ide.c.
>   You can find this patch as attachment.
> 
> Frank	
> 
> -- 
>      _/_/_/_/ _//   _/ Frank Neuber
>     _/       _/_/  _/  frank.neuber@gmx.de (private)
>    _/_/_/   _/ _/ _/
>   _/       _/  _/_/    neuber@opensource-systemberatung.de
>  _/       _/    // http://www.opensource-systemberatung.de
> 

Re: BUGFIX: UDMA-SiS5513 chipset support

[Frank Neuber]

Hi Hedrick,
I was looking for an option such as "nodma" and I did't found it.
The option "autodma" I had never set. Or should I undefine this in the
source?
Maybe my hdd is broken, the original driver works well with my
ATAPI-CD-DRIVE.

Thanks for your response
 Frank 

On Sat, 18 Aug 2001, Andre Hedrick wrote:
> 
> Not the correct solution, you should not set autodma.
> This will prevent DMA from being used at INIT; however, you can still
> attempt it later.  This patch is an invalid solution, period.

--
     _/_/_/_/ _//   _/ Frank Neuber
    _/       _/_/  _/  frank.neuber@gmx.de (private)
   _/_/_/   _/ _/ _/
  _/       _/  _/_/    neuber@opensource-systemberatung.de
 _/       _/    // http://www.opensource-systemberatung.de

Re: BUGFIX: UDMA-SiS5513 chipset support

[Jussi Laako]

Frank Neuber wrote:
> 
> recently I did an upgrade of my old computer (ASUS SP97-V) to
> the kernel-2.4.7.
> 
> Problem:
>   My system was crashing even when I load the module ide-disk.o
> Solution:
>   Because of the broken hardware (some UDMA problems with an non
>   UDMA-Cabel to the drive) the linux kernel hangs during ide_dma_check.
>   Even if UDMA is disabeld in the bios, the kernel detect this drive as
>   an udma drive. And this is wrong!!
>   My solution was simply to comment out the ide_dma_check in ide.c.
>   You can find this patch as attachment.

Same mobo here, with -ac kernel. I've never had any problems with any kernel
version (IBM/Maxtor/Seagate HDDs). So I'd suggest faulty HDD, not the
controller nor driver.


	- Jussi Laako

-- 
PGP key fingerprint: 161D 6FED 6A92 39E2 EB5B  39DD A4DE 63EB C216 1E4B
Available at PGP keyservers

Re: BUGFIX: UDMA-SiS5513 chipset support

[Frank Neuber]

On Sat, 18 Aug 2001, Jussi Laako wrote:
> 
> Same mobo here, with -ac kernel. I've never had any problems with any kernel
> version (IBM/Maxtor/Seagate HDDs). So I'd suggest faulty HDD, not the
> controller nor driver.
With my old kernel linux-2.0.34 I had no problems with UDMA. Maybe you are
right, my ATAPI-CD-Drive works with UDMA on 2.4.7.

Thanks for your response
 Frank

--
     _/_/_/_/ _//   _/ Frank Neuber
    _/       _/_/  _/  frank.neuber@gmx.de (private)
   _/_/_/   _/ _/ _/
  _/       _/  _/_/    neuber@opensource-systemberatung.de
 _/       _/    // http://www.opensource-systemberatung.de