Re: [PATCH 09/10] Change table chaining layout

[Matt Mackall]

On Mon, Oct 22, 2007 at 11:52:51PM +0100, Alan Cox wrote:
> On Mon, 22 Oct 2007 16:47:07 -0500
> Matt Mackall <mpm@xxxxxxxxxxx> wrote:
> 
> > On Mon, Oct 22, 2007 at 05:21:30PM -0400, Jeff Garzik wrote:
> > > Alan Cox wrote:
> > > >Why can't we just make the list one item longer than the entry count and
> > > >stick a NULL on the end of it like normal people ?
> > > 
> > > Certainly seems safer than the current "let's run off the end of the 
> > > list if anything bad happens" setup...  And I do not think allocating 
> > > n+1 scatterlist entries will have much of a negative impact.
> > 
> > It'll mean m-1 scatterlists fit on a slab.
> 
> Is that really a credible space issue ?

Yes. Especially if m is 2 or 1. A scatterlist on 64-bit x86 looks like
it takes 32 bytes, which means 128 elements fit on a page. One more
spills - ouch!

But maybe chaining means this doesn't matter any more. Maybe we can
even pick a nice moderate sg size and reduce the number of mempools we
need for these things.

-- 
Mathematics is the supreme nostalgia of our time.
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/