Re: [PATCH #upstream] libata: enhance command parsing

[Robert Hancock <hancockrwd@gmail.com>]

On 08/22/2009 02:51 PM, Mark Lord wrote:
>> Actually, I think the code+data size of this version is not much
>> different from the previous version. One might be able to come up with
>> a structure that you could encapsulate all the strings into and that
>> you could search through to find the right one, but it'd be tricky to
>> do that without making the structure end up containing a bunch of
>> padding and other wasted space that would add as much data size as you
>> saved in code size. Unless you have a great idea on the subject that
>> I'm not thinking of right after waking up :-)
> ..
>
> I've done lots of this type of thing in the past,
> putting the parsing knowledge into the data structure
> and reducing the code to nearly nothing (at the expense
> of more data than before).
>
> But right now I'm not offering to do it, so no objection
> from me to the current state. :)

Any other thoughts on this one? Jeff?