Re: [PATCH 1/2] libata: Add two more columns to the ata_timing table.

[Sergei Shtylyov <sshtylyov@ru.mvista.com>]

Hello.

David Daney wrote:

>>    OK, t6z is yet longer than t9 but putting it into the table seems 
>> pointless anyway as it's fixed at 30 ns, at least for the standard 5 
>> PIO modes (and for other two modes 30 ns would be good anyway).
>>    Though frankly speaking I don't quite understand your care for 
>> this timing, if the ATA standard permits -CE deasserted and data bus 
>> being driven to overlap.
>
> It doesn't matter what the ATA standard permits in this case.  We need 
> to assure that the OCTEON Boot Bus standard is respected.  There are 
> several timing parameters that we have to set based on the documented 
> properties of the device.  Ideally if we try to run bus cycles for 
> non-ATA/CF devices that share the signal lines of the Boot Bus, they 
> should  not interfere with the CF interface.

   Well, your driver is free to handle it on its own, and even respect a 
lesser minimum for PIO5/6. It just doesn't seem practical to add this to 
the table. In fact, about all PATA hardware on market only permits to 
control the active/recovery and the address setup times -- that's why 
the table looks like it looks now.

> David Daney

MBR, Sergei