Mailing-List: contact cygwin-help@sourceware.cygnus.com; run by ezmlm
Sender: cygwin-owner@sources.redhat.com
Date: Sat, 21 Oct 2000 01:18:18 -0400
From: Christopher Faylor <cgf@redhat.com>
To: cygwin@sources.redhat.com
Subject: Re: cygwin on a 386?
Message-ID: <20001021011818.A13632@redhat.com>
Reply-To: cygwin@sources.redhat.com
Mail-Followup-To: cygwin@sources.redhat.com
References: <0056900013040104000002L042*@MHS> <20001019122701.B32595@cygnus.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.3.6i
In-Reply-To: <20001019122701.B32595@cygnus.com>; from cgf@redhat.com on Thu, Oct 19, 2000 at 12:27:01PM -0400

On Fri, 20 Oct 2000 08:58:54 +0200, jurgen.defurne@philips.com wrote:
>On Thu, Oct 19, 2000 at 12:27:01PM -0400, Christopher Faylor wrote:
>>On Thu, Oct 19, 2000 at 08:36:09AM +0200, jurgen.defurne@philips.com wrote:
>>>These instructions have their equivalent since the first 80x86.
>>>	LOCK
>>>	INC	dest
>>>
>>>	LOCK
>>>	XCHG	dest,src
>>>
>>>Of course, these operate at most between a register and memory, not
>>>between memory and memory.
>>
>>Are you answering my question about whether Cygwin works on a 386?
>>Somehow I can't figure this out from your message.  You seem to be
>>instructing me in assembly language, which wasn't what I was asking
>>for.
>
>What I mention is this : I presume that the interlocking mechanism is
>just to isolate a variable from being updated by another task, while
>the current task should do it.  This has been possible since the early
>8086.  What I wondered is why the before mentioned instructions go
>through such great lengths to provide the necessary mechanisms.

The instruction in question is CMPXCHG.  It's only available on a 486
and above:

http://developer.intel.com/design/intarch/techinfo/Pentium/instsum.htm

I'm still waiting for someone to inform me that Cygwin works fine on a
386...

cgf

--
Want to unsubscribe from this list?
Send a message to cygwin-unsubscribe@sourceware.cygnus.com