Re: [PATCH #upstream-fixes] libata: clear PIO pad area

[Sergei Shtylyov <sshtylyov@mvista.com>]

Hello.

On 09/06/2011 09:23 PM, Sergei Shtylyov wrote:

>>>> ata_sff_data_xfer[32]() use pad area if the transfer size isn't
>>>> multiple of transfer size; however, this area wasn't cleared and
>>>> garbage data in pad area could be transferred to the device. Make
>>>> sure the pad area is cleared.

>>>> Signed-off-by: Tejun Heo<tj@kernel.org>
>>>> Cc: Lei Ming<tom.leiming@gmail.com>

>>> And what's the problem with garbage data? Why it's worse than 0s?

>> There were devices which puke on random garbage at the end of CDB and

> Note that ATAPI CDBs are always aligned to 4 bytes, so this path should never be
> hit when sending them.

>> in general sending indeterministic garbage to external device is a bad
>> idea. It increases chance of problems which can be difficult to
>> reproduce and track down while gaining nothing.

> Problems with "don't care" padding bytes? Dunno...

>> It really is a stupid thing to do.

> Well, at least the old code was equally stupid before I patched it. :-)

    Only in ata_sff_data_xfer32(). It was I who removed the padding buffer 
initialization in ata_sff_data_xfer()...

>> Thanks.

WBR, Sergei