PERFORCE change 171334 for review

[Alexander Motin]

Nathan Whitehorn wrote:
> Alexander Motin wrote:
>> @@ -230,7 +229,7 @@
>>          return EIO;
>>      }
>>  
>> -    request->dma = &ch->dma.slot[atadev->unit];
>> +    request->dma = &ch->dma.slot[0];
>>  
>>      if ((error = bus_dmamap_load(request->dma->data_tag,         
>> request->dma->data_map, request->data, request->bytecount,
>>   
> I made an error in this part of the patch -- it should be request->unit,
> not 0, here. Thanks for your work!

It's not your fault, it's my change. It shouldn't be request->unit
there, as only one DMA slot per channel allocated now. It is useless to
allocate more, as rest of ata(4) infrastructure supports only one
request at a time per channel.

-- 
Alexander Motin