Re: [Patch, gfortran] common blocks getting mangled in nested modules

[Paul Thomas <paulthomas2 at wanadoo dot fr>]

Tobi,

> What if one uses different modules which contain the same common block, will
> they not interfere?
Yes, they are correctly determined to be ambiguous.

gfortran - Error: Name 'a' at (1) is an ambiguous reference to 'a' from 
module 'm1'

ifc - Error 119 at (11:collision.f90) : More than one program entity 
with this name is accessible

Standard - 11.3.2: ".... Except for these cases, the local name for any 
entity given accessibility by a USE statement shall differ from the 
local names of all other entities accessible to the scoping unit through 
USE statements and otherwise."

However this:

module m1
 common /x/ a
end module m1
module m2
 common /x/ a
end module m2

program collision
 use m1
 use m2, only: b=>a
 a = 1
 print *, b
end program collision

is not being compiled correctly, since it produces a undefined reference 
to `__m1__a'.  It arrives at I believe this is symptomatic of the same 
malady that is exposed by use+host associated common variables that we 
have discussed off-list and that, I hope, will be fixed by a later patch 
in the series.

Cheers

Paul T