[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: libmachuser vs libmach
From: |
Thomas Bushnell, BSG |
Subject: |
Re: libmachuser vs libmach |
Date: |
16 Jun 2002 20:51:29 -0700 |
User-agent: |
Gnus/5.09 (Gnus v5.9.0) Emacs/21.2 |
James Morrison <rocketmail_com@rocketmail.com> writes:
> I've been looking though some of the mach interfaces and libmach
> is mentioned in a few places. This looks to be libmachuser for us,
> is this correct?
libmach existed in some old CMU systems. It included libmachuser, as
well as some additional user-space-only functions. (The CMU systems
had a libmachuser which was, more or less, just like ours.)
> I'm also curious about the machine_slot_data_t type, the comment
> says it is bogus, but the interface for xxx_slot_info uses
> machine_slot_data_t instead of machien_slot_t, should this be fixed?
> Also is there any reason that the function to get the function to
> get more information about a processor is prefixed with xxx?
xxx means that it's deprecated or otherwise discouraged. The fact
that machine_slot_data_t is used only in xxx_slot_info is precisely
why you should believe that it's bogus.
The reason the type is discouraged is that returning a straightforward
struct is a bad idea. (Compare xxx_host_info with the current
host_info RPC.) So there should be a revamped slot_info function too,
(and cpu_control) though that hasn't been done yet.
Thomas