Created attachment 11685 [details]
kernel config file

Created attachment 11686 [details]
syslog

Related part of the syslog

Reply-To: akpm@linux-foundation.org

On Tue, 5 Jun 2007 06:30:52 -0700
bugme-daemon@bugzilla.kernel.org wrote:

> http://bugzilla.kernel.org/show_bug.cgi?id=8587
> 
>            Summary: disk on promise pdc20378 disabled at boot
>     Kernel Version: 2.6.22-rc4
>             Status: NEW
>           Severity: normal
>              Owner: greg@kroah.com
>          Submitter: jean.luc.coulon@gmail.com
> 
> 
> Most recent kernel where this bug did *NOT* occur: 2.6.22-rc3-git7
> Distribution: Debian sid X86_64
> Hardware Environment: PDC20378 in ASUS A8V Deluxe, Athlon64 3600+
> Software Environment: sata_promise
> Problem Description: disk is disabled
> 
> Steps to reproduce: boot the system
> 

If Jean really meant 2.6.22-rc3-git7 then this is a very recent regression.

Yes, really 2.6.22-rc3-git7


[jean-luc@tangerine] % uname -a
Linux tangerine 2.6.22-rc3-git7-k8-1 #1 PREEMPT Mon Jun 4 13:08:28 CEST 2007
x86_64 GNU/Linux

# CONFIG_BLK_DEV_PDC202XX_OLD is not set
# CONFIG_BLK_DEV_PDC202XX_NEW is not set
CONFIG_BLK_DEV_VIA82CXXX=m


# CONFIG_PATA_PDC_OLD is not set
# CONFIG_PATA_PDC2027X is not set

Hmmm... Which PDC driver do you use?

It is a sata drive.
I use sata_promise

CONFIG_SATA_PROMISE=m


Regards

Jean-Luc

Please try
wget http://www.stardust.webpages.pl/files/patches/libata_revert/libata-1.patch
cat /patch/to/libata-1.patch | patch -p1 -R
make && make test_kernel

Ok, it works that way

J-L

Tejun, it is your patch

commit 464cf177df7727efcc5506322fc5d0c8b896f545
Author: Tejun Heo <htejun@gmail.com>
Date:   Sun May 27 15:10:40 2007 +0200

    libata: always use polling SETXFER

This is because sata_promise doesn't like polling SETXFER.  Under investigation.

Created attachment 11694 [details]
sata_promise-use-TF-issue-for-polling-NODATA

Please test this patch.  Thanks.

With the patch applied, it works fine.

Thanks.  Resolving the bug.