http://d.puremagic.com/issues/show_bug.cgi?id=229

           Summary: Aggregate trips assert
           Product: GDC
           Version: 0.19
          Platform: PC
        OS/Version: Linux
            Status: NEW
          Severity: normal
          Priority: P2
         Component: glue layer
        AssignedTo: dvdfrdmn@users.sf.net
        ReportedBy: larsivar@igesund.net


I am not sure if it is entirely proper to post this here, as I have compiled GDC from svn head with DMD 0.161 support (and I'm new to this GDC game), but the code below (which works with DMD) gives the following assert:

cc1d: ../../gcc/d/d-codegen.cc:2933: virtual void AggLayout::doFields(Array*,
AggregateDeclaration*): Assertion `((field_decl)->decl.mode) != VOIDmode'
failed.
gdctest.d:0: internal compiler error: Aborted

class Foo{ struct Bar { } private Bar bar; }

This is as minimal as I get it, but the non-minimal case actually involved an enum, but since it was the same assert that was tripped, I didn't look further into it.


-- 

http://d.puremagic.com/issues/show_bug.cgi?id=229


dvdfrdmn@users.sf.net changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |FIXED




------- Comment #1 from dvdfrdmn@users.sf.net  2006-06-28 20:36 -------
Fixed in svn.


-- 

d-bugmail@puremagic.com schrieb am 2006-06-27:
> http://d.puremagic.com/issues/show_bug.cgi?id=229

> I am not sure if it is entirely proper to post this here, as I have compiled GDC from svn head with DMD 0.161 support (and I'm new to this GDC game), but the code below (which works with DMD) gives the following assert:
>
> cc1d: ../../gcc/d/d-codegen.cc:2933: virtual void AggLayout::doFields(Array*,
> AggregateDeclaration*): Assertion `((field_decl)->decl.mode) != VOIDmode'
> failed.
> gdctest.d:0: internal compiler error: Aborted
>
> class Foo{ struct Bar { } private Bar bar; }
>
> This is as minimal as I get it, but the non-minimal case actually involved an enum, but since it was the same assert that was tripped, I didn't look further into it.

Added to DStress as http://dstress.kuehne.cn/compile/b/bug_dcodegen_2933_A.d

Thomas