From mboxrd@z Thu Jan  1 00:00:00 1970
From: pbonzini@redhat.com (Paolo Bonzini)
Date: Thu, 18 Feb 2010 16:27:01 +0100
Subject: [ltt-dev] [PATCH 12/12] centralize definition of BITS_PER_LONG
In-Reply-To: <20100218134600.GA5825@Krystal>
References: <1266260686-17588-1-git-send-email-pbonzini@redhat.com>
	<1266260686-17588-13-git-send-email-pbonzini@redhat.com>
	<20100218032554.GG11338@Krystal> <4B7CFE10.8070803@redhat.com>
	<20100218134600.GA5825@Krystal>
Message-ID: <4B7D5C45.9020804@redhat.com>

On 02/18/2010 02:46 PM, Mathieu Desnoyers wrote:
>> >
>> >  I32LLP64 is basically only Mingw64 and ia64-hpux.
> And anyway I32LLP64 have 4-byte longs, so they should work anyway with
> these tests.
>
>> >  They'd probably break
>> >  horribly because with 99% likelihood somewhere there is an assumption
>> >  that sizeof (long) == sizeof (void*).
>
> In the atomic ops headers possibly, yes.

You couldn't use atomic op headers to exchange pointers too.  Maybe not 
used by liburcu, but still a strong limitation.

Some time we should also add double-long compare-and-swap, that's very 
useful for lock-free lists.

Paolo