[jp] Sync problems

Peter Jay Salzman p at dirac.org
Thu Nov 23 10:55:31 EST 2006

On Thu 23 Nov 06,  9:45 AM, Donn Washburn <n5xwb at hal-pc.org> said:
> Peter Jay Salzman wrote:
> >> If you do try it you will be rewarded with a faster more perfectly
> >> configured machine with less chance of a lockup.
> > 
> > Wrong.  Choosing to not build a module does not lead to a "faster more
> > perfectly configured machine with less chance of a lockup".  That's just
> > silly.  Where would you get such an idea?
> > 
> > A module that never gets loaded (including a driver that never gets used)
> > does NOT slow the machine down.  It does NOT lead to a machine that less
> > likely to lock-up.  I suggest you read the Linux Kernel Module Programmer's
> > Guide (which I'm the author of) before making claims like this.
> > 
> > Peter
> Right! because the kernel level instructions work more correctly when 
> configured in the kernel than the default one that the Distro people 
> use.  Like i386 or i586 on a Intel or AMD new chip.  I haven't played 
> much with a 64 bit machine but think of running a 64 bit machine on a 32 
> bit default kernel
> Correct on the no change by using modules - but why by default should 
> this AMD Athlon-xp run correctly with Intel Pentium III instructions?

There are two issues here:

1. Generating instructions that are exclusive to an architecture.
2. Generating optimizations that play on the strengths of the architecture.

Obviously, an executable that contains instructions that are not implemented
will not work.  An executable that contains optimizations for the wrong
platform will be slower, and perhaps, be flaky. 

> The main reason for modules over buildins is custom configuration. 
> Which allows a Distro to setup many different parts of a users machine.


How VBA rounds a number depends on the number's internal representation.
You cannot always predict how it will round when the rounding digit is 5.
If you want a rounding function that rounds according to predictable rules,
you should write your own.
              -- MSDN, on Microsoft VBA's "stochastic" rounding function

Peter Jay Salzman, email: p at dirac.org web: http://www.dirac.org/p    
PGP Fingerprint: B9F1 6CF3 47C4 7CD8 D33E  70A9 A3B9 1945 67EA 951D

More information about the Jpilot mailing list