[gnu.gcc.bug] additional info to my previous gcc V1.34 m68k bug report

bammi@DSRGSUN.CES.CWRU.EDU (Jwahar R. Bammi) (03/18/89)

I am sorry, but in my previous report about gcc V1.34 and m68k, i may
have given an incomplete explaination ( i just showed it to a friend
and he suggested this followon to explain a little better)

The instruction in question was
>>	cmpb #46,LC0+2 <- bug here for m68000, LC0+2 is not a data alterable

after it goes though gcc-as it gets turned into

	cmpib #46,pc@(<offset>)

and for the m68000, cmpi instruction only data alterable addressing
modes are allowed, and pc@(<offset>) is not.

I hope that clears it up. Sorry about the incomplete info the first
time around.
--
usenet: {decvax,sun}!cwjcc!dsrgsun!bammi	jwahar r. bammi
csnet:       bammi@dsrgsun.ces.CWRU.edu
arpa:        bammi@dsrgsun.ces.CWRU.edu
compuServe:  71515,155

bruce@heather.pooh.com (Bruce Robertson) (03/18/89)

> The instruction in question was
>>	cmpb #46,LC0+2 <- bug here for m68000, LC0+2 is not a data alterable

An intelligent assembler should recognize this case, and generate the
absolute long addressing mode instead of PC relative.