Re: PIONEER DVD-RW DVRTD08 is disabled if there is no disc present at boot time

[Tejun Heo <tj@kernel.org>]

Robert Hancock wrote:
> linux-ide@momail.e4ward.com wrote:
>> Hello,
>> several Sony Vaio laptops seem to have the PIONEER DVD-RW DVRTD08 DVD/CD
>> drive built in. This drive is available only when a disc is inserted
>> during boot.
>> when no disc is present, dmesg has
>>
>> [    3.715535] ata2: SATA max UDMA/133 abar m2048@0xd3e04000 port
>> 0xd3e04180 irq 218
>> [    4.368166] ata2: SATA link up 1.5 Gbps (SStatus 113 SControl 300)
>> [    4.380063] ata2.00: ATAPI: PIONEER DVD-RW  DVRTD08, 1.00, max UDMA/33
>> [    9.380102] ata2.00: qc timeout (cmd 0xef)
>> [    9.380112] ata2.00: failed to set xfermode (err_mask=0x4)
>> [    9.700093] ata2: SATA link up 1.5 Gbps (SStatus 113 SControl 300)
>> [   19.712097] ata2.00: qc timeout (cmd 0xef)
>> [   19.712105] ata2.00: failed to set xfermode (err_mask=0x4)
>> [   19.712158] ata2: limiting SATA link speed to 1.5 Gbps
>> [   19.712160] ata2.00: limiting speed to UDMA/33:PIO3
>> [   20.032106] ata2: SATA link up 1.5 Gbps (SStatus 113 SControl 310)
>> [   30.040084] ata2.00: qc timeout (cmd 0xef)
>> [   30.040091] ata2.00: failed to set xfermode (err_mask=0x4)
>> [   30.040144] ata2.00: disabled
>> [   30.056102] ata2: hard resetting link
>> [   30.376108] ata2: SATA link up 1.5 Gbps (SStatus 113 SControl 310)
>> [   30.376114] ata2: EH complete
>>
>> when a disc is present, dmesg has
>> [    3.804304] ata2: SATA max UDMA/133 irq_stat 0x00000040, cirq 219
>> [    5.268599] ata2: SATA link up 1.5 Gbps (SStatus 113 SControl 300)
>> [    5.270213] ata2.00: ATAPI: PIONEER DVD-RW  DVRTD08, 1.00, max UDMA/33
>> [    5.272217] ata2.00: configured for UDMA/33
>>
>> You'll find more details at
>> http://ubuntuforums.org/showthread.php?p=6412465#post6412465
>>
>> Having read
>> http://www.mail-archive.com/linux-ide@vger.kernel.org/msg06127.html I
>> guess it is not OK for the device to timeout in response to 0xef.
>> Still I would like to be able to use the drive without having to
>> insert a CD at every boot. Is there any workaround? Can't the driver
>> just treat
>> repeated timeouts like it would treat an abort?
> 
> That case was a bit different in that it was a CompactFlash device
> (which has some different rules, especially older ones), but there was
> also a SATA-PATA bridge chip involved (which seems quite likely the case
> here too since the drive reports UDMA33 maximum). This could be another
> case of the bridge chip being broken. Set Features - Set Transfer Mode
> is a mandatory command that all ATA devices must implement, and the
> device really can't be used properly if it's not successful (for true
> SATA devices it's mostly vestigial, but if there's any actual PATA
> involved, it's required). It's rather bizarre that having a disc
> inserted avoids the problem.

Aiee... these broken devices.  :-(

So, w/ a disk inserted, there's no error whatsoever?  Can you please
post the output of "hdparm -I /dev/sr0"?

-- 
tejun