PR debug/41248 [VTA] deal with Pmode != ptr_mode in expand_debug_expr

[Michael Matz]

Hi,

On Fri, 4 Sep 2009, Steve Ellcey wrote:

> You are right, the genmode failure does not appear to be related to your 
> change, it started with r151350 (Michael Matz's expand from tuple 
> patch).  Michael, have you had any reports of problems with this change?

Nope, but I'm confused about the current state of affairs.  So, you had 
two problems: (1) the assert in expand_debug_stmt triggering, and (2) a 
crash in genmode.

I assume (1) and (2) are proved unrelated meanwhile.  And that 
(1) was fixed by Alexandres patches.  And that (2) was hidden by occurance 
of (1) before that.  Is all of this correct?

I have access to linux ia64 machines, but not to hpux ones.  If the 
miscompilation doesn't depend on 32bit pointers I should be able to 
reproduce it, otherwise I'd need some more help from you.


Ciao,
Michael.