[tree-ssa] Statement with no effect warning

[Jan Hubicka]

> In message <20031216153946.GL2199@kam.mff.cuni.cz>, Jan Hubicka writes:
>  >> eIn message <20031216143133.GE19896@atrey.karlin.mff.cuni.cz>, Jan Hubicka 
>  >> write
>  >> s:
>  >>  >Hi,
>  >>  >I got bit by the fact that tree-SSA does not emit statement with no
>  >>  >effect warnings currently, as these happens during expansion and we of
>  >>  >course elliminate such statement much earlier.  What is the best place
>  >>  >to re-implement such a warnings?
>  >> So how again did the latest cgraphunit.c changes pass a bootstrap test
>  >> then?  They're failing on i686-pc-linux-gnu.
>  >
>  >Did you disabled checking?
> That was from Diego's tester.  I haven't updated my tree in a couple days
> simply because I'm trying to work through some engineering issues with
> the code to thread to blocks with have PHIs.  Better to wrap my head
> completely around the issue than suck in a bunch of new code which
> may cloud/morph the issues I'm trying to wrap my head around.

Diego's tester runs checking disabled. OK as I've mentioned I will do
both bootstrap each time now to prevent similar situations.

Honza
> 
> 
> jeff
>