installing f8/hang at 'loading ata_piix' PROGRESS but now 'ata1.00 exception .... frozen'/FIXED

[John Minson]

as  it turns out the source of all my issues was the dvd drive itself. I 
installed a 9 year old dvd reader and everything works with no boot 
options or bios changes . I dont  know if the dvd writer is broken or 
just sucks . The drive is a SAMSUNG|SH-S202N 20X =-O

Jim Cornette wrote:
> John Minson wrote:
>>
>> after fighting with the 'loading ata_piix driver' issue I managed to 
>> get fc8 installed via a usb dvd drive . the install went fine but now 
>> when I boot off the hd (sata) I eventually get
>>
>> ata1.00 exception emask 0x0 sactt serr 0x0 action 0x2 frozen
>>
>> it then displays 'loading udev' and hangs
>>
>> once again this problem (or some variant there of) has been reported 
>> lots but no clear indication of what to do.
>> I keep seeing references to the 'libpata' module having issues
>>
>
> It's great that the DVD got you further through the process. 
> Apparently the Distribution linked below has a solution and a kernel 
> available to remove the error.
>
> http://bbs.archlinux.org/viewtopic.php?pid=275504
>
> http://article.gmane.org/gmane.linux.ide/26012
>
> According to the comment that led to the above link, it is fixed in 
> the next kernel series 2.6.24.
>
> The next release is up to that version kernel. I'm not sure if later 
> kernels would have a backported fix in it or not. The changelog shows 
> these for a search for ata. The bugs should be found in bugzilla.
>
> Jim
> ---
>
>  rpm -q --changelog kernel-2.6.23.14-107.fc8.i686 |grep ata
> - libata: fix AHCI controller reset (#411171)
> - libata: fix ATAPI tape drives (#394961)
> - libata: allow short SCSI commands for ATAPI devices
> - Fix libata handling of IO ready test (#389971)
> - libata: fix resume on some systems
> - libata: fix pata_serverworks with some drive combinations
> - Fix completely broken sata_sis libata driver (#365331)
> - nfs: fix writeback race causing data corruption
> - direct-io: fix return of stale data after DIO write
> - md/raid5: fix data corruption in some failure cases
> - Use upstream libata DMA disable patch
>   (libata.dma parameter replaces libata.pata_dma)
> - leave boot option as "libata.pata_dma" for now
> - libata: change boot option name to "libata.dma"
> - add option to disable libata PATA DMA
> - Terminate list in ata-piix
> - add Intel ICH8M (Santa Rosa) PCI ID to ata_piix driver
> - As the x86-64 kernel is now relocatable, -kdump can go.
>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: minsonj.vcf
Type: text/x-vcard
Size: 57 bytes
Desc: not available
URL: <http://listman.redhat.com/archives/fedora-list/attachments/20080204/e9772753/attachment-0001.vcf>