[gentoo-dev] Re: baselayout-1.13 going into ~ARCH soon

[Duncan]

Roy Marples <uberlord@g.o> posted
200611071058.54068.uberlord@g.o, excerpted below, on  Tue, 07 Nov
2006 10:58:53 +0000:

> Actually, before I do that, let me attack this from another angle.
> What do you gain from keeping it mounted as a ramdisk?
> If the answer is performance, well you loose performance at start time as
> you've lost the deptree.
>
> So why would you want to keep it?

Well, as you point out it's mixed from a performance angle.  One reason
I'm beginning to appreciate a bit more now that I have a tmpfs mounted
/tmp as well, however, is the clean start at every boot thing. 
Eliminating even the possibility of trouble from stale or corrupted
carryover has its own >0 benefits.

Back to performance, however, the deptree rebuild isn't as much of a
pull-down as one might expect, for a number of reasons.  Keep in mind that
perhaps the biggest boot-time bottleneck is disk read bandwidth anyway. 
Preloading all those scripts effectively at once therefore has three
overlapping effects.  1)  There's the direct disk read bandwidth, but that
would come in somewhere anyway when the scripts are actually executed.  2)
Having them all loaded at once gives the filesystem elevator scheduler a
bigger chance to do its thing, ordering the reads appropriately for better
efficency.  3) Once it comes time to actually run the scripts, they are
already loaded into cache, so no waiting to read them in then, simply
virtually instant execution.

Thus, while the performance benefit won't be great, it's likely to be
there, and there's the "system hygiene" (I believe that's a decently
accurate label) benefit as well.

So, keeping it at minimum an option is IMO a good thing.  If as proposed
we make it standard, there are other benefits including serious code
simplification as well.  There could well be negatives, especially on
certain archs I'm not familiar with, but for x86, amd64, and probably the
ppc folks, at least, it sounds very reasonable from my perspective.  From
what I've read, the particular challenges faced on MIPS may change the
time effects greatly there (basing this on remarks I recall from the
split-KDE debate, but I haven't the foggiest whether that applies here
or not), so I'd be interested in seeing someone address it from their
perspective.

-- 
Duncan - List replies preferred.   No HTML msgs.
"Every nonfree program has a lord, a master --
and if you use the program, he is your master."  Richard Stallman

-- 
gentoo-dev@g.o mailing list