[Subject Prev][Subject Next][Thread Prev][Thread Next][Subject Index][Thread Index]

[LI] Recompiles anyone?



Dear Folks,

In addition to all the talk on the list of recompiling your kernel there's a
question I'd like to ask - some part of the kernel is really written in
assembly (whichever processor) - is that optimized? If not, for example, can
we do anything to optimize there?

To add to all the info, /. had an article from PC World on how to optimize
your Linux box. I've just freshly formated and installed RH6.1 off the PCQ
CD, so can afford to fool around - will report on the process though to the
list. If anyone 's done a recompile with optimizations - what kind of
performance jump have you got? Worth the effort? Does it then make sense to
avoid downloading binaries WHENEVER < Prof G.'s points accepted!> possible?
Thereby, download src, recompile and use? And how often are the makefiles
really helpful? < I'd like to add an incident on what happened at IT.COM
here, but that's going into that article on IT.COM!>

The PC World site is:

http://www.pcworld.com/heres_how/article/0,1400,13702,00.html

ReSourcefully,
M

p.s. if anyone's got any interesting links for really "newbie" recompilers -
please post - will help.

p.p.s the latest Mindcraft results are out (again from /. ) - NT 4.0
outperforms Linux from 1.4 to 2.2 times - methinks something is afoot here.
Thou, I'd like to see a desktop head to head comparision - then we'll
ROCK.....

    Gandalf Mithrandir   /* "Mobilis in mobile." */   gandalf@xxxxxxxxx
/*   Any Sufficiently Advanced Operating System is indistinguishable from Linux   */ 
// Do not meddle with SysAds for they are quick to anger and subtle in their ways //

__________________________________________________
FREE Email for ALL! Sign up at http://www.mail.com

--------------------------------------------------------------------
The Linux India Mailing List Archives are now available.  Please search
the archive at http://lists.linux-india.org/ before posting your question
to avoid repetition and save bandwidth.