[M3commit] CVS Update: cm3

[Jay Krell]

CVSROOT:	/usr/cvs
Changes by:	jkrell at birch.	10/03/15 16:06:35

Modified files:
	cm3/m3-sys/m3back/src/: Codex86.m3 M3x86.m3 M3x86Rep.i3 
	                        Stackx86.m3 

Log message:
	Remove from one to three instructions from small functions.
	By tracking which non-volatile registers are used and only saving/restoring them.
	Unfortunately, until we are more sophisticated, we have to pay for the size
	for the push instructions, but we nop them out if they aren't needed,
	and only generating the needed pops. (We should look into two and three
	byte nops, but presently I just use one byte nops. The pushes are each
	one instruction. Theoretically, nop is faster than push, though I should check.
	Plus, nop saves us from doing a pop.)
	
	Also a much cleaner fix to the backup(5) vs. patch fix from before. No downside now.
	
	Register allocator should probably favor volatile registers over non-volatile.
	Even a simple PROCEDURE F(a,b:INTEGER) = BEGIN a:=b END F uses ebx. (see test p234, not commited yet)
	
	I noticed an ancient bug in doing this work.
	Functions with >4K frames need to call chkstk or equivalent.
	Specifically, stack pages must be touched in order. call naturally
	touches a page. Or, side motivation: match what the C compiler does. It is correct.
	(See, we also reserve room for sub esp,foo, for 32bit foo, even
	though many functions save save 3 bytes and use an 8bit value and
	some unusual functions can skip the entire thing (if foo = 0))
	
	We can issue errors for such functions and see if any turn up.
	
	See test case p234 (not commited yet).
	
	Use of the new proc_reguse is sprinkled in a bit unnecessarily much here.
	corrupt, set_reg, movop1, load_ind should suffice.