Alias Analysis Improvement Patch

["Sanjiv Kumar Gupta, Noida" <sanjivg at noida dot hcltech dot com>]

Hi,

Attached is the patch based on the paper that
I presented at GCC summit. The patch is enabled
by a new compiler option called -falias-arithmetic.
The patch is tested for x86, sh-elf, and mips-elf. 
I do not have access to an ia64, so could not do 
make check for that but stress-1.17 is clear for ia64-elf.
The code size improvements are given below.

gcc version 3.4 20030609 (experimental)
options#1 -O2 -funroll-all-loops
options#2 -O2 -funroll-all-loops -falias-arithmetic
.text size comparisons for ia64-elf
----------------------------------
filename		options#1	options#2

dct64.lst 		16256 	15408
dogmove.lst 	26555 	26571
g_combat.lst 	11403 	11371
g_func.lst 		50077 	50109
g_misc.lst 		58278 	58214
hack.lst 		78318 	78286
imdecode.lst 	591208 	591192
layer3.lst 		64012 	63428
lpc.lst 		40534 	35430
lpc55encode.lst 	2672 		2656
mdct.lst 		5968 		5520
mkobj.lst 		38919 	38903
physics.lst 	27351 	27335
polyobj.lst 	16781 	16685
q_shared.lst 	25877 	25893
quadrics.lst 	22096 	21936
signature.lst 	7759 		7743
slalom.lst 		27500 	27404
tif_lzw.lst 	30698 	30570
vision.lst 		65881		65865
 
.text size comparisons for sh-elf
---------------------------------
filename		options#1	options#2

dct64.lst 		3680 		3584
layer3.lst 		29232 	29200
linpackc.lst 	17156 	17124
mdct.lst 		1568 		1504
quadrics.lst 	14456 	14488
slalom.lst 		15104 	15040
vision.lst 		23812 	23780

I measured execution timings gains for SH4 but they
are not significant currently. The execution
gains should be better for processors (with no 'base + offset'
addressing) where we have first scheduling pass enabled.

Regards
--Sanjiv

Attachment: PATCH_20030611
Description: Binary data