lkml.org 
[lkml]   [1999]   [Jul]   [16]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
From
SubjectRe: New kernel/resource.c
Date
On Fri, Jul 16, 1999, Linus Torvalds wrote:

> So instead of using confusing names like "peanut" or "pci_or_isa_or_
> pcmcia_or_who_the_hell_cares_what_electrical_bus_is_outside_the_cpu_
> as_we_cannot_see_it_anyway" , I'm just calling it "PCI". Fair enough? It's
> 99% of the interesting market these days.

Come on now, you're just yanking my chain, right?

I say "pci_io_resource is really not a good name, because these aren't
PCI specific: just call it io_resource". You first stridently claim
that, in fact, they are PCI specific. Then you say, well, not really,
but they might as well be because most interesting stuff is PCI?!?

Call me a stickler, I guess, but it seems fairly sensible to me that
"pci_" be reserved for things that are PCI, since "PCI" is widely
understood to mean a certain thing.

> In particular, you don't have two resource trees for one and the same
> resource. Also, you don't try to have a resource tree based on any
> electrical hierarchy (ie you do NOT try to follow PCI bridges or whatever
> that are largely invisible from a software standpoint).

Ok, I'll summarize my interpretation of where things stand:

- A few weeks ago, I asked for some new functionality in the resource
manager, to handle memory resources and to allow tracking of
hardware separately from arbitrating driver conflicts. I wrote a
patch and we even discussed in advance what I planned to do.
- You put it into 2.3.6.
- I put together some code to enumerate PCI and PnP resources that
didn't mess up anything else and gave me what I wanted, which was
reliable conflict-free hot plug configuration. Martin Mares made
some changes to my PCI stuff and I think it was ready to go: the PnP
stuff is not quite ready, but close.
- You decided to go off and rewrite resource management on your own.
- I said fine, I can work with this too, just let me slide my hardware
tracking info into a separate tree where it won't impact anything
else, since the way I did it in 2.3.6 is now gone.
- You say, tough, this information shouldn't be in the tree to begin
with because it is "irrelevant", and it shouldn't be in a separate
tree either, because having two trees for one resource is "stupid".
Checkmate.

Ok, I yield. I'll shut up and go back to doing something productive.
I hope this leaves no doubts about why I'm not particularly excited
about merging PCMCIA into the kernel tree. I'm not a glutton for
punishment.

-- Dave Hinds

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/

\
 
 \ /
  Last update: 2005-03-22 13:53    [W:0.043 / U:0.152 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site