[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [RFC] hurd/hurd_types.defs
From: |
Thomas Bushnell, BSG |
Subject: |
Re: [RFC] hurd/hurd_types.defs |
Date: |
02 May 2001 00:40:14 -0700 |
User-agent: |
Gnus/5.0803 (Gnus v5.8.3) Emacs/20.7 |
Erik Verbruggen <ejv@cs.kun.nl> writes:
> Um, well, we might want to update this. Or something. Or just drop the
> network part. Just out of curiosity: Mach did work on Alpha, no? How did
> they do this 32/64 bit stuff there?
We are certainly not going to drop the network transparency part.
The MiG type "int" is specified as MACH_MSG_TYPE_INTEGER_32 (see
<mach/std_types.defs>). MiG assumes that its types will also be C
types, and the C type "int" on Alpha is a 64 bit type.
That means that on the Alpha, an int MiG type only holds 32 bits, even
if the machine is using 64 bit registers now and then to hold them.
It would be better if we dropped the C type "int" entirely, and
changed to using the MiG type "int32" or "int64", which can then have
C typedefs to the right native types, and the whole thing would be
clearer.
Thomas
- [RFC] hurd/hurd_types.defs, Neal H Walfield, 2001/05/01
- Re: [RFC] hurd/hurd_types.defs, Thomas Bushnell, BSG, 2001/05/01
- Re: [RFC] hurd/hurd_types.defs, Erik Verbruggen, 2001/05/02
- Re: [RFC] hurd/hurd_types.defs, Neal H Walfield, 2001/05/03
- Re: [RFC] hurd/hurd_types.defs, Roland McGrath, 2001/05/04
- Re: [RFC] hurd/hurd_types.defs, Thomas Bushnell, BSG, 2001/05/04
- Re: [RFC] hurd/hurd_types.defs, Neal H Walfield, 2001/05/07
- Re: [RFC] hurd/hurd_types.defs, Thomas Bushnell, BSG, 2001/05/07