Messages in this thread | | | Subject | Re: [PATCH] Fix broken debug output reserve_region_with_split() | From | Benjamin Herrenschmidt <> | Date | Mon, 20 Oct 2008 14:37:30 +1100 |
| |
On Sat, 2008-10-18 at 08:45 +1100, Benjamin Herrenschmidt wrote: > On Fri, 2008-10-17 at 14:11 -0700, Linus Torvalds wrote: > > > So the default is all the wrong way around. Make the defaults the sane > > ones: 04x for IO, and 08x for MMIO, and then see if anybody wants anything > > else. I seriously doubt they do. The whole ".. but but sparc" argument > > seems to be entirely based on the total mis-conception of truncation that > > has no relevance. > > Well, I definitely want 08x for IO on powerpc and I suspect anything > non-x86 or alpha does as well... So I'm fine but I'd like to keep > the default for IO at 08x unless you really have a strong opinion > against it.
Actually, screw it. 04x is fine, as you said, it's not going to truncate anyway. The reason I wanted 08x on powerpc is that we have legitimate negative-looking values in there due to the way we do PIO on 32-bit and we have most PCI IO looking like it's >64K on 64-but but that's no big deal, ie, we don't actually need the padding for IO space.
So I'm happy with just a padding of 8 for memory which is even good looking for most 64-bit archs in fact, and 4 for IO. I left the ability for archs to pick a different value if they want to tho.
I'll send the patch separately after a test or two with a better mail subject.
Cheers, Ben.
| |