[Coco] c.comp and c.pass1 / c.pass2 data sizes.

[gene heskett]

On Saturday, April 28, 2012 09:01:59 PM Stephen H. Fischer did opine:

> Hi,
> 
> Gene or anybody:
> 
> Suppose you have the largest program that can be compiled by the "C"
> Compiler.
> 
> c.comp now gives a out of memory error that before reportedly the
> program could be compiled. (Previous author statement, new system
> environment now in use.)
> 
> My questions are:
> 
> Given the data sizes for c.comp and c.pass1, is additional memory
> requested as needed up to the 64K limit?
> 
> Can c.pass1 / c.pass2 compile a larger program than c.comp?
> 
> Note, I have NO previous c.comp experience.
> 
> SHF

I have a copy of c.comp, but have never tried it on the biggest package I 
ever built, which was rzsz-3.3.6, around 34k of executable for either.  My 
guess is that is was split originally due to the memory requirements while 
building larger programs.  I used cpass1 and cpass2 to build rzsz & kept 
the intermediate files in /r0, the ramdisk 'myram' that I wrote.

Unforch, this is 2012, 20 years down the log, so a lot of that has faded 
off into the inevitable distance of what is now a 77 YO mind.  So all I can 
say is to make sure you have ansifront-012, then the latest c.prep19c I 
think it is now, and c.pass1 & c.pass2, in that order, and see how that 
works. cc.2.5.2 or newer as the exec script makes it easy to adjust what 
you are using as its a script itself, and you may even be able to modify 
the amount of memory each of the subprocesses gets to do its thing.

Sorry I can't be more specific.
Cheers, Gene
-- 
"There are four boxes to be used in defense of liberty:
 soap, ballot, jury, and ammo. Please use in that order."
-Ed Howdershelt (Author)
My web page: <http://coyoteden.dyndns-free.com:85/gene>
User n.:
	A programmer who will believe anything you tell him.