Bug in lookup_fnfields

[Mark Mitchell]

>>>>> "Martin" == Martin von Loewis <martin@mira.isdn.cs.tu-berlin.de> writes:

    Martin> After asserting that TREE_VALUE only access TREE_LIST
    Martin> nodes, memoize1.C shows a digression in egcs-2.91.15 on
    Martin> i486-pc-linux-gnu. The problem is the line

I've been thinking for a while now that it would be a good idea if all
the TREE_ macros checked the tree code of the thing they're looking
at when gcc is built in a debug configuration.  I've been too lazy to
implement this, though, and I don't know what macro to test to see
whether or not debugging support is requested.  (Do we use NDEBUG, or
something else?)

My plan was:

  o Write a generic macro for checking codes, and complaining 
    about mismatches.

  o Begin using this macro in the TREE_ macros.

    Martin> 		  if (TREE_VALUE (memoized_basetypes) !=
    Martin> these_basetypes) break;

    Martin> as memoized_basetypes really is a TREE_VEC node.

    Martin> Regards, Martin

-- 
Mark Mitchell <mmitchell@usa.net>
http://home.earthlink.net/~mbmitchell
Consulting Services Available