Patch for better diagnostic (was: Re: Bug in MI?)

Jason Merrill
Mon Mar 9 00:57:00 GMT 1998

>>>>> Manfred Hollstein <> writes:

>> That can't be right; you can't treat error_mark_node like a TREE_LIST.

> Perhaps, you're wright. But that treatment isn't new due to my patch. I only
> added the the different error message, the while loop was already there.

Agreed.  The old code didn't recognize that we could get an error_mark_node
there.  Do you mind fixing it?


More information about the Gcc-bugs mailing list