sparc-sun-solaris2.8 bootstrap failure

[Rainer Orth]

Phil Edwards writes:

> After getting a decent lexer, I've now reproduced this during the stage2
> build.  Same backtrace, same NULL pointer.

Indeed: the stage1 cc1 is fine.  Yesterday evening, I could narrow the
breakage down a bit more:

* TZ=GMT -D '20030713 00:00': ok

2003-07-12  Kazu Hirata  <kazu@cs.umass.edu>

	* config/alpha/alpha.c: Fix comment typos.

* TZ=GMT -D '20030714 00:00': broken

2003-07-13  Richard Kenner  <kenner@vlsi1.ultra.nyu.edu>

	* expr.c (expand_expr, case COMPONENT_REF): If reg, copy OP0 to MEM

	Rainer