[parisc-linux] merge activity

Grant Grundler grundler@cup.hp.com
Sat, 02 Dec 2000 21:11:48 -0800


Matthew Wilcox wrote:
> those on the CVS list will have noticed a lot of activity just recently.
> Alan Cox & myself are geting the tree into a state where it can be dropped
> on Linus.

Excellent you guys!

> A number of changes have been made:
> 
> s/STATIC/static/

It's ok that's it dropped but since I added I think you should know
why o put it in, in the first place.

Prumpf criticized this in my first dino code drop claiming one
could just compile with "-Dstatic=" to get the same effect
as "-DSTATIC=". That's not true. data declared static inside
a function will behave differently without "static" storage class.
STATIC serves to differentiate function declarations from local
static data declarations.

Why differentiate? Debuggers. I want to make sure all the symbols
I care about are visible - I'm told under ELF they are. Under SOM
binary format I know they weren't.

> u_long becomes unsigned long

I've been using "ulong" as well. Is less than desirable as well?
Feel free to change if it is.

> old code has mostly been cleaned out
> a lot of #ifdefs have been removed.

Some of the "ifdef 0" I have in "my" code are there for future stuff.
An examples is code related to DELAYED_RESOURCE_CNT in sba_iommu.c.
Fortunately, this wasn't removed.
Haw should I mark this to make sure people understand it's work-in-progress?

thanks,
grant