Dan Mick wrote:
> Garrett D'Amore wrote:
>> Dan Mick wrote:
>>> Garrett D'Amore wrote:
>>>> I'd like to gather rough opinions about an idea to "simplify"
>>>> pcmcia. Our pcmcia stack in Solaris is rather complex, and I think
>>>> a lot of this owes to a desire to make it compliant with the
>>>> quasi-portable Card Services API/DDK (specified by the PCMCIA org,
>>>> and largely meant for use with other operating systems; intended to
>>>> facilitate porting pccard drivers, I think.)
>>>
>>> Is the code complexity causing problems? Is this work more
>>> important than new feature support or RFEs/projects on live subsystems?
>>>
>> We have several device that are "live" on pcmcia, and have a lot of
>> complexity introduced to them via this.
>>
>> I strongly suspect there is even "dead" code, which we are keeping
>> around just because of Card Services API specifications. I hate
>> "dead" code.
>>
>> I suppose I could be just like everyone else and avert my eyes, and
>> work on other things.
>
> I hate dead code that's a detriment of some sort. A minute of build
> time, if that's all it is, doesn't warrant much work, and it seems
> like the bug velocity in that code is pretty small; mostly it has to
> do with Cardbus, which we agree we want.
The cardbus implementation would potentially be significantly simpler if
the pcmcia support could be simplified as well. The nexus drivers for
these are somewhat conflated, and pcic has to deal with two very
different style of interfaces.
>
> I dunno; seems like a lot more discussion, even, than it would warrant.
Maybe. This wasn't a high priority item for me, just something I saw as
an area that could potentially easily be cleaned up bit by bit in small
steps if we were allowed to stop supporting the CardServices and
SocketServices DDIs.
- Garrett
