Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Path: utzoo!watmath!clyde!caip!princeton!puvax2!PUCC.BITNET!6101778
From: 6101778@PUCC.BITNET (Angus Christie)
Newsgroups: net.lang.c
Subject: Re: expression sequencing bugs
Message-ID: <1144@PUCC.BITNET>
Date: Thu, 2-Oct-86 21:47:23 EDT
Article-I.D.: PUCC.1144
Posted: Thu Oct  2 21:47:23 1986
Date-Received: Sat, 4-Oct-86 12:00:50 EDT
References: <4197@brl-smoke.ARPA>
Reply-To: 6101778@PUCC.BITNET
Organization: Princeton University Computing Center, Princeton, New Jersey
Lines: 18

>>    /* a = b + b + b */
>>    a = ((b=1),b) + ((b=2),b) + ((b=3),b)
>>
 [with various results for a]
>I would like everyone to test it on as many machines as prossible, to
>see if we can find as least ONE other besides Small-C, that works.
 
MIX-C for the IBM-PC also gives an answer of 6 for all three of the
following:
 
     a = ((b=1),b) + ((b=2),b) + (b=3),b);
     a = (b=1) + (b=2) + (b=3);
     a = (b=1), a += (b=2), a += (b=3);
 
Beware, however, that MIX-C accomplishes this by doing absolutely NO
optimization of any sort.  It doesn't fold constants.  It has no register
variables.  Its execution speed, as a result, leaves something to be
desired.  But then again, what do you expect for only $40.