[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: Where could I upload x32 port bootstrap?

On Sat, 10 Nov 2012, Adam Borowski wrote:
> On Fri, Nov 09, 2012 at 11:27:20PM +0100, Marco d'Itri wrote:
> > On Nov 09, Daniel Schepler <dschepler@gmail.com> wrote:
> > > I've asked a couple people in private mail about this, and haven't
> > > gotten any answer, so I thought I'd ask here for ideas.  Where would
> > > be a good place to upload what I have so far from bootstrapping an x32
> > > port of Debian?
> > Nowhere, until we decide if and how we want to use x32.
> > IIRC there was some agreement that if we decide to support x32 it should 
> > be as a partial architecture.
> That'd make it mostly worthless.  If you need to co-install amd64 packages
> on the same system (but not physical machine!), memory gains are gone.

Can someone give us numbers?  Using VMs or even the bare metal, it should be
possible to gather some statistics about memory usage for x32 versus amd64

1) Desktop running gnome or kde (or both, whatever) with iceweasel open on
the www.debian.org page.

2) Maximum memory usage for a kernel, x.org or libreoffice build run.

3) Memory usage of some common server workload.  E.g. email with
amavisd-new+spamassassin (perl is a memory pig in amd64), or a LAMP stack
with some common web application.

This should give us a pretty clear idea, and could go a long way to prove
the case for x32 as a full arch as far as memory goes.

If someone knows of patological workloads (one that really doesn't benefit
from x32, and one that benefits enormously from x32), those would be
interesting data points as well.

> Speed gains are far better than armel->armhf, at least for i386.  Gains
> compared to amd64 are limited to pointer-heavy code, said to be up to 30%. 
> If Daniel could upload his work somewhere, we'd be able to test this
> ourselves instead of relying on some random benchmarks.


> On the other hand, widespread dumb-ass assumptions about i386/amd64 may
> cause quite a bit of issues: basically any Makefile that talks about "x86"
> is somewhat suspicious.  This is the main reason one may want to oppose
> the inclusion of x32 in Debian, IMHO.

This is no worse than any other new arch.

  "One disk to rule them all, One disk to find them. One disk to bring
  them all and in the darkness grind them. In the Land of Redmond
  where the shadows lie." -- The Silicon Valley Tarot
  Henrique Holschuh

Reply to: