Re: [PATCH] sata_sil: combined irq + LBT DMA patch for testing

[Thomas Backlund <tmb@mandriva.org>]

Richard Bollinger wrote:
> On 12/3/05, Jeff Garzik <jgarzik@pobox.com> wrote:
>> To make it easy for others to test, since there are merge conflicts,
>> I've combined the two previous sata_sil patches into a single patch.
>>
>> Verified here on my 3112 (Adaptec 1210SA).
>>
>> I'm especially interested to hear from anyone willing to test on a
>> SI 3114 (4-port).
>>
>>
>> The 'sii' branch of
>> rsync://rsync.kernel.org/pub/scm/linux/kernel/git/jgarzik/libata-dev.git
>>
>> contains the following updates:
>>
>>  drivers/scsi/sata_sil.c |  233 +++++++++++++++++++++++++++++++++++++++++++++---
>>  1 files changed, 219 insertions(+), 14 deletions(-)
>>
>> Jeff Garzik:
>>      [libata sata_sil] improved interrupt handling
>>      [libata sata_sil] Greatly improve DMA handling
>>
>> diff --git a/drivers/scsi/sata_sil.c b/drivers/scsi/sata_sil.c
>> index 3609186..9e4630f 100644
>> --- a/drivers/scsi/sata_sil.c
>> +++ b/drivers/scsi/sata_sil.c
>> ...
> 
> Not so well on my Gigabyte GA-K8N Ultra 9.  lspci -v says:

Same for me...
After some more tests i managed to see some error output before the 
freeze, and this is what I also saw:

>> ata1: BUG: SG size underflow
>> ata1: status=0x50 { DriveReady SeekComplete }

and onde by one the raid devices got deactivated until the full freeze...