[APPROVE] make window-configuration saving of frame position optional

[Nix]

On 29 Oct 2006, Michael Sperber stated:
> [Nix, I'm not sure you're a committer.  Let me know if you're not, and
> I'll commit your patch.]

I'm sure I'm not a committer :)

> Nix <nix at esperi.org.uk> writes:
>
>> Unfortunately, another feature was added at the same time: frame
>> positions are now saved in window configurations,
>
> Hm, I don't remember this detail, but I do remember I tried to follow
> the essence of what the old code did pretty slavishly.  But never

That code was rather... fugly, with a lot of #iffed out portions and
so forth.

> mind, your reasoning is sound.  (Generally, window-configuration
> restore happens way too often.)

What I'm worried about here is the saving. Does querying the frame
position and size involve a round-trip to the X server? 'cos if it does
we're roundtripping to hell and back.

(I actually expect that it doesn't on the basis that I'd have seen the
network traffic... it's probably locally cached.)


(Tracking GC leak now: current suspect; horrible heap fragmentation
combined with the lack of an upper bound on the size of new heap
allocations (the comments say there is such a bound but they're
wrong). The new GC uses a new allocator, which nothing else uses, so
that allocator is a definite source of potential bugs.)

-- 
`When we are born we have plenty of Hydrogen but as we age our
 Hydrogen pool becomes depleted.'