------- Comment #9 from uweigand at gcc dot gnu dot org  2006-01-30 18:32 
-------
This appears to be a pre-existing bug in s390_decompose_address,
that happens to be triggered by this particular glibc code.

The problem is the construct &&label1 - &&label2 used in the
source (b.t.w. is this usage of the GCC &&label extension
valid in the first place?), which causes GCSE to generate RTL
of the form
  (minus (label_ref ..) (label_ref ..))

s390_decompose_address accepts those, because I thought the only
time this construct could ever be generated is from within the
s390 machine-dependent reorg code ...

I'm currently testing a fix.


-- 

uweigand at gcc dot gnu dot org changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
         AssignedTo|unassigned at gcc dot gnu   |uweigand at gcc dot gnu dot
                   |dot org                     |org
             Status|UNCONFIRMED                 |ASSIGNED
     Ever Confirmed|0                           |1
   Last reconfirmed|0000-00-00 00:00:00         |2006-01-30 18:32:04
               date|                            |


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=26018

Reply via email to