[BusyBox] Re: busybox digest, Vol 1 #258 - 3 msgs

[Mark Whitley]

On Wed, Jan 24, 2001 at 11:50:34AM +0300, Vladimir N. Oleynik wrote:
> Mark Whitley wrote:
> 
> > Hence, the BSS does not exist on the disk, per se, except as a name and a
> > size. The value reported from the 'size' command to show the size of the
> > BSS is the run-time size, not the disk size.  The gain from eliminating
> > big static buffers is that much less memory is used at run-time, at the
> > cost of a slightly larger size on disk. I consider this to be an overall
> > win.
> 
> Well where you have found benefit?  Each utility spends memory only so much,
> how many it is necessary to it.  And on demand.  Stack too memory, and
> strongly limited.  Also you overlook, that strict audit as BSS it is
> allocated nulled is necessary, and it is used in algorithms, and the stack -
> is filled with dust.  All your shifts only will lower speed of performance
> and less size.

My apologies, but I'm having a little difficulty understanding what you're
saying. Are you saying that it would be better to allocate memory off the heap
via malloc(), rather than declare the buffer on the stack at the beginning of
a function?

Mark Whitley
markw at lineo.com