From mboxrd@z Thu Jan  1 00:00:00 1970
From: pbonzini@redhat.com (Paolo Bonzini)
Date: Fri, 12 Nov 2010 13:24:07 +0100
Subject: [ltt-dev] [PATCH 2/2] introduce uatomic_and and uatomic_or
In-Reply-To: <20101112121900.GA29642@Krystal>
References: <1289552247-18130-1-git-send-email-pbonzini@redhat.com>
	<1289552247-18130-3-git-send-email-pbonzini@redhat.com>
	<20101112120846.GB27929@Krystal> <4CDD2F69.7020407@redhat.com>
	<20101112121900.GA29642@Krystal>
Message-ID: <4CDD31E7.1050500@redhat.com>

On 11/12/2010 01:19 PM, Mathieu Desnoyers wrote:
> * Paolo Bonzini (pbonzini at redhat.com) wrote:
>> Return nothing can be useful because it can be optimized on x86 as "lock
>> orl (mem), reg/imm".  However, there are no other return-nothing atomic
>> ops in uatomic_*.h so I decided not to provide this.
>>
>
> How about we start by implementing uatomic_and/uatomic_or that return
> void, and if we ever need uatomic_return_and/uatomic_return_or (see my
> other mail), we add them ?

I am using uatomic_or's return value in my call_rcu-with-futex, but I 
don't need atomic-or-and-exchange really, I can even change it to a load 
followed by an atomic or (it introduces a race but it is benign).  Or I 
can certainly use cmpxchg too.  Updated patches will come.

Should I rename uatomic_add to uatomic_xchg_add too (I prefer this name 
to uatomic_return_add)?

Paolo