diff: work-around for intermodule optimization bug
Fergus Henderson
fjh at cs.mu.OZ.AU
Sun Feb 15 03:18:38 AEDT 1998
On 15-Feb-1998, I wrote:
> When compiling with intermodule optimization enabled,
> private equivalence types in imported modules are visible,
> so previously abstract types may become concrete.
> This happens before semantic checking, so it affects whether
> or not the compiler reports type errors. Clearly it can
> make type errors go away. Less obvious is the fact that it
> can also introduce ambiguity errors into code that would
> otherwise be ambiguity-free.
>
> This bug causes the compiler to fail to bootstrap when compiled
> with options `--intermodule-optimization -O4'.
Hence the following change.
compiler/polymorphism.m:
Add an explicit type qualification, to work around a
bug with intermodule optimization that causes a spuriou
ambiguity errors when the compiler is bootstapped with
`-O4 --intermodule-optimization'.
cvs diff compiler/polymorphism.m
Index: compiler/polymorphism.m
===================================================================
RCS file: /home/mercury1/repository/mercury/compiler/polymorphism.m,v
retrieving revision 1.127
diff -u -r1.127 polymorphism.m
--- polymorphism.m 1998/02/12 01:17:39 1.127
+++ polymorphism.m 1998/02/14 15:55:52
@@ -2246,7 +2246,11 @@
Index0::in, Index::out] is det,
(
Elem = Elem0 - Index0,
- Index is Index0 + 1
+ Index is Index0 + 1,
+ % the following call is a work-around for a compiler
+ % bug with intermodule optimization: it is needed to
+ % resolve a type ambiguity
+ is_pair(Elem)
)),
list__map_foldl(MakeIndex, ClassTypeVars0, ClassTypeVars, First, _),
@@ -2278,6 +2282,9 @@
VarTypes1, VarTypes,
ExtraHeadVars1, ExtraHeadVars,
TypeClassInfoMap1, TypeClassInfoMap).
+
+:- pred is_pair(pair(_, _)::in) is det.
+is_pair(_).
:- pred polymorphism__new_typeclass_info_var(varset, map(var, type),
string, var,
--
Fergus Henderson <fjh at cs.mu.oz.au> | "I have always known that the pursuit
WWW: <http://www.cs.mu.oz.au/~fjh> | of excellence is a lethal habit"
PGP: finger fjh at 128.250.37.3 | -- the last words of T. S. Garp.
More information about the developers
mailing list