[Bug tree-optimization/39251] FAIL: g++.dg/tree-ssa/new1.C scan-tree-dump-not forwprop1 "= .* \+ -"

["ramana at gcc dot gnu dot org" <gcc-bugzilla at gcc dot gnu dot org>]

------- Comment #9 from ramana at gcc dot gnu dot org  2010-01-15 09:05 -------


(In reply to comment #8)


Hi Mark,

Many thanks for looking into this. 

> However, if this is behaving different on ARM from (say) x86, I think that the ABI is a likely
> cause because, as you say, the C++ ABI for these bits is subtly different.  Do
> we know that this is ARM-specific?


I've compared this with the mips port and the x86 port and in both cases the
first bit where there is a difference is after lowering into gimple - i.e there
is no address calculation like this sequence that comes in the ARM port. 


  D.1844_3 = (unsigned int *) D.1824_2;
  D.1845_4 = D.1844_3 + 4;
  D.1846_5 = D.1845_4 + -4;



So yes it does look ARM specific . Also peeking at results on gcc-testresults
doesn't show this failure on x86.

cheers
Ramana 


-- 

ramana at gcc dot gnu dot org changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
      Known to fail|4.5.0                       |4.4.2 4.5.0


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=39251