Re: OMAP4 naming conventions

[Kevin Hilman <khilman@deeprootsystems.com>]

Paul Walmsley <paul@pwsan.com> writes:

> On Tue, 12 May 2009, Kevin Hilman wrote:
>
>> As the OMAP4 patches are coming in, there seems to be a bit of variety
>> in the naming of functions/macros/variables etc.
>> 
>> Could I propose that we just use omap4_* and OMAP4_* instead of
>> OMAP44XX_* or OMAP4XXXX_* etc.
>> 
>> I know that OMAP2 and OMAP3 have a variety of forms here too, but
>> those should probably be cleaned up eventually too.
>> 
>> With proper runtime revision detecting, IMO, we should only really
>> have the OMAP4 prefix, and leave the sub revision handling to runtime
>> code.
>> 
>> Thoughts?
>
> Here are some questions that we should figure out answers to before 
> deciding:
>
> How should macros be named that only apply to specific OMAP4 chips (i.e., 
> what happens if TI repeats a OMAP2420 to 2430 transition)?

I would propose a convention something like:

- OMAP4_* - applies to all OMAP4 chips
- OMAP4xyz_* - applies to the specific 4xyz rev

> How should macros be handled that are only applicable to later ES
> levels?  Tagging ES levels in macros has caught many bugs in the
> OMAP2/3 code.

Since the ES revs would be specific to particular 'xyz' rev, The above
could be extended to use something like OMAP4xyzESn_*

Kevin