[gnu.gcc] 80386 code generation problem

johnl@ima.ima.isc.com (John R. Levine) (05/30/89)

I have the following problem in GCC 1.34 for the 80386:  If you multiply
two char values producing a short, the compiler aborts.  I see that there is
a table entry particularly for this case, and that on most other machines
the table entry is missing, so the compiler widens the arguments before
multiplying.  I commented out the table entry on the 386, and it works, and
the code is nearly as good as what should have been generated.  Has anyone
made this case work?
-- 
John R. Levine, Segue Software, POB 349, Cambridge MA 02238, +1 617 492 3869
{ bbn | spdcc | decvax | harvard | yale }!ima!johnl, Levine@YALE.something
Massachusetts has 64 licensed drivers who are over 100 years old.  -The Globe