Re: [PATCH] Use uint32_t instead of unsigned32_t.

[Justus Winter]

Quoting Flávio Cruz (2016-04-05 01:05:43)
> On 5 April 2016 at 00:59, Justus Winter <4winter@informatik.uni-hamburg.de>
> wrote:
> 
> > Quoting Justus Winter (2016-04-05 00:46:11)
> > > Quoting Flávio Cruz (2016-04-05 00:39:48)
> > > > > This broke your test suite:
> > > > Should work with the new patch. Thanks for the report.
> > >
> > > It does indeed.
> >
> > Furthermore, I was able to build a seemingly working Hurd package
> > using the fixed MIG, but I see these kind of warnings that never
> > appeared before (and neither have I seen the polymorphic type being
> > used):
> >
> > fsServer.c: In function ‘_Xfile_getcontrol’:
> > fsServer.c:1086:22: warning: large integer implicitly truncated to
> > unsigned type [-Woverflow]
> >    /* msgt_name = */  MACH_MSG_TYPE_POLYMORPHIC,
> >                       ^
> 
> 
> Mig was using the polymorphic type before but was outputting '-1' instead
> of MACH_MSG_TYPE_POLYMORPHIC.
> The other patch to change the definition of mach_msg_type_name_t to
> unsigned char will fix this. I'm preparing an updated version.

Oh good :)

Justus