Re: DVD/CD lost after r246713

On 23.02.2013 00:52, Claude Buisson wrote:
> On 02/22/2013 22:46, Alexander Motin wrote:
>> Hi.
>>
>> ATAPI devices on legacy ATA controllers were lost due to command timeout
>> caused by data underrun, caused by r246713.
>>
>> r247165 should fix the issue.
>>
> 
> r247165 applied
> 
> Problem solved (at least at the first reboot)
> 
> Thanks to both of you !!
> 
> BTW, an old intertwining of lines in dmesg is back:
> 
> cd0 at ata1 bus 0 scbus1 target 0 lun 0
> cd0: <SAMSUNG DVD-ROM SD-616T F310> Removable CD-ROM SCSI-0 device
> cd0: 33.300MB/s transfers (UDMA2, ATAPI 12bytes, PIO 65534bytes)
> cd0: Attempt to query device size failed: NOT READY, Medium not present
> ada0 at ata0 bus 0 scbus0 target 0 lun 0
> ada0: <WDC WD3200AAJB-00J3A0 01.03E01> ATA-8 device
> ada0: 100.000MB/s transfers (UDMA5, cd1 at ata1 bus 0 scbus1 target 1 lun 0
>                                    ^
> cd1: <HL-DT-ST CD-RW GCE-8481B C102> Removable CD-ROM SCSI-0 device
> cd1: 33.300MB/s transfers (UDMA2, ATAPI 12bytes, PIO 65534bytes)
> cd1: Attempt to query device size failed: NOT READY, Medium not present
> - tray
> closed
> PIO 8192bytes)
> ^
> ada0: 305245MB (625142448 512 byte sectors: 16H 63S/T 16383C)
> ada1 at ata0 bus 0 scbus0 target 1 lun 0
> ada1: <WDC WD600BB-75CAA0 16.06V16> ATA-5 device
> ada1: 100.000MB/s transfers (UDMA5, PIO 8192bytes)
> ada1: 57220MB (117187500 512 byte sectors: 16H 63S/T 16383C)

They never really went away. ada0 and cd1 in this example are living on
different buses and so have separate locks, so nothing really prevents
them to mix the messages. It rarely happens for disks, but CD tend to
report capacity much longer, increasing mix chances. For USB devices
chances are even higher, because they are often detected when CAM
already gave up waiting and continued booting.

-- 
Alexander Motin