Re: New resources - pls, explain :-(

Linus Torvalds (torvalds@transmeta.com)
Thu, 5 Aug 1999 11:26:48 -0700 (PDT)


On Thu, 5 Aug 1999, Linus Torvalds wrote:
>
> There's a planned fix for check_region() and friends coming soon, so this
> will just work again. None of the drivers I needed had a problem with it,
> because they were either pure PCI anddin't bother with check-region at
> all, or they were old ISA-like and didn't show up in the PCI address space
> allocations anyway.
>
> Expect a update later today,

The pre-6 patch is there on ftp.kernel.org now: it has the updated
request_region() etc, which knows about the difference of a driver that
wants to mark the IO ports it is using busy, and just the regular "oh,
this resource exists" kind of thing.

For example, my current /proc/ioports has an entry like this:

b800-b8ff : NCR 53c895
b800-b87f : sym53c8xx
d000-d01f : Intel EtherExpress Pro100
d400-d41f : Intel 82371AB PIIX4 USB
d800-d80f : Intel 82371AB PIIX4 IDE

where the "b800-b8ff" region is the actual PCI resource of the "NCR
53c895" controller, while the "b800-b87f" sub-region is the sym53c8xx
driver that has marked that part of the resource busy (don't ask me why
the driver only wants to look at the first 128 bytes of the 256 byte
decode - I just show them as I see them).

The pre-patch also cleans up the device naming thing - all the device
names are now in the init section, and only the ones we actually need are
copied into the device structures, so that we don't waste precious memory
space by having all device strings available at runtime.

(It still has the vendor strings in non-init memory due to a simple thinko
of mine, fixed after I put the pre-patch out there).

Linus

-
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/