On 07/16/2013 11:15:51 AM, Fabien Chouteau wrote:
On 07/16/2013 05:37 PM, Alexander Graf wrote:
> On 07/16/2013 05:28 PM, Fabien Chouteau wrote:
>> On 07/16/2013 04:06 AM, Scott Wood wrote:
>>> On 07/10/2013 12:10:02 PM, Fabien Chouteau wrote:
>>>> This implementation doesn't include ring priority, TCP/IP
Off-Load, QoS.
>>>>
>>>> Signed-off-by: Fabien Chouteau<chout...@adacore.com>
>>> From the code comments I gather this has been tested on
VxWorks. Has it
>>> been tested on Linux, or anywhere else?
>>>
>> You're right, as I said in the cover letter, this has only been
tested on vxWorks.
>
> Could you please give it a try? IIRC eTSEC support should be in
upstream Linux.
>
I don't have time for that. As I said in the cover letter, I submit
this
patch for those interested in eTSEC, but I won't be able to test/fix
it
for Linux.
Could you please at least document more fully the known limitations,
such as "I'm only interested in 32bits address spaces"?
>>>> + /* ring_base = (etsec->regs[RBASEH].value& 0xF)<< 32; */
>>>> + ring_base += etsec->regs[RBASE0 + ring_nbr].value&
~0x7;
>>>> + start_bd_addr = bd_addr = etsec->regs[RBPTR0 +
ring_nbr].value& ~0x7;
>>> What about RBDBPH (upper bits of physical address)? Likewise for
TX.
>>>
>> I'm only interested in 32bits address spaces, so RBASEH, TBASEH,
RBDBPH or TBDBPH.
>
> Why? I thought e500mc and above can access more than 32bits of
physical address space?
Yes but this is not emulated by QEMU, right? sizeof (hwaddr) for
qemu-system-ppc is 8...
36bit physical is emulated by QEMU. Currently we put CCSR in a place
that would make it difficult to use memory above 4G, but that should
change at some point.
> Oh, but they're always DPAA?
>
I don't understand...
It doesn't matter, because it's not true. We do support 36-bit address
layouts on mpc85xx and mpc86xx, though we don't make it the only
supported config in U-Boot as we do on e500mc+.
-Scott