On 01/26/2010 12:56 PM, Mike Frysinger wrote:
> On Monday 25 January 2010 22:51:30 Jie Zhang wrote:
>> On 01/26/2010 12:28 AM, Mike Frysinger wrote:
>>> the bfin.h header should use #ifdef protection naming like all the other
>>> headers, and when it includes local urjtag headers, it should use "..."
>>> not<urjtag/...>.
>>
>> I think this is good. I considered this issue before. But we need also
>> change many existing C files, too. So I prefer do it in a batch for with
>> all other files.
>
> i'm referring to the header files in include/urjtag/.  all the C files in src/
> should be using<urjtag/...>.
>
OK. I will change that when committing the patch.

>> There is a story why we need params in in struct URJ_PART. But I don't
>> know if I'm allowed to tell it.
>
> i dont think there's too much secret background here, but you could send it to
> Robin and i off list first ...

The story is actually about "struct emu_oab *emu_oab" in "struct 
bfin_part_data", which is the real type of params for Blackfin part. 
params in struct URJ_PART itself, as I explained in my last email, just 
provides a place to store part specific data. Thus each part can have 
its own data. For dual-core bf561, each core can have its own status 
recorded in params.

emu_oab was invented to add hooks for another already canceled Blackfin 
variant. Now I think we can fold it into struct bfin_part_data. But I 
will not do this in this patch.


Jie

------------------------------------------------------------------------------
The Planet: dedicated and managed hosting, cloud storage, colocation
Stay online with enterprise data centers and the best network in the business
Choose flexible plans and management services without long-term contracts
Personal 24x7 support from experience hosting pros just a phone call away.
http://p.sf.net/sfu/theplanet-com
_______________________________________________
UrJTAG-development mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/urjtag-development

Reply via email to