linux-kernel - Re: [PATCH] memory-model: fix cheat sheet typo

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Date:   Tue, 10 Apr 2018 23:10:06 +0200
From:   Paolo Bonzini <pbonzini@...hat.com>
To:     paulmck@...ux.vnet.ibm.com
Cc:     linux-kernel@...r.kernel.org,
        Alan Stern <stern@...land.harvard.edu>,
        Andrea Parri <parri.andrea@...il.com>,
        Will Deacon <will.deacon@....com>,
        Peter Zijlstra <peterz@...radead.org>,
        Boqun Feng <boqun.feng@...il.com>,
        Nicholas Piggin <npiggin@...il.com>,
        David Howells <dhowells@...hat.com>,
        Jade Alglave <j.alglave@....ac.uk>,
        Luc Maranget <luc.maranget@...ia.fr>,
        Akira Yokosawa <akiyks@...il.com>
Subject: Re: [PATCH] memory-model: fix cheat sheet typo

On 10/04/2018 22:32, Paul E. McKenney wrote:
> On Mon, Apr 09, 2018 at 11:42:58AM -0700, Paul E. McKenney wrote:
>> On Mon, Apr 09, 2018 at 06:50:15PM +0200, Paolo Bonzini wrote:
>>> "RWM" should be "RMW", and that's more or less the extent to which I
>>> can claim to change the document. :)  In particular, "Self" is not
>>> documented and the difference between "Self" and "SV" is not clear
>>> to me.
>>>
>>> Signed-off-by: Paolo Bonzini <pbonzini@...hat.com>
>>
>> Applied, though without the questions.  ;-)
>>
>> "Self" is for things like smp_load_acquire() and smp_store_release()
>> that order themselves against later and earlier accesses, respectively.
>> This ordering applies to later/earlier access to all variables, not
>> just the one that smp_load_acquire()/smp_store_release() accessed.
>> In contrast, things like smp_mb() order only other accesses, not
>> themselves.  Or at least it is impossible to proves whether or not they
>> order themselves because they are not separately visible to other CPUs.
>>
>> "SV" is "same variable", which applies to pretty much anything that
>> accesses a variable, but not to things like smp_mb() which do not.
>>
>> Does that help?
> 
> On the perhaps naive assumption that silence means assent, how about
> the following patch?

Silence meant "I tried thinking of a patch myself, and hadn't come up
yet with a fully satisfactory one"; that's some kind of assent I guess. :)

Your patch is certainly an improvement!

Thanks,

Paolo

> 							Thanx, Paul
> 
> ------------------------------------------------------------------------
> 
> commit 818e46e8db6cacb099b8640b7f2945a3151c00ab
> Author: Paul E. McKenney <paulmck@...ux.vnet.ibm.com>
> Date:   Tue Apr 10 13:24:19 2018 -0700
> 
>     tools/memory-order: Improve key for SELF and SV
>     
>     The key for "SELF" was missing completely and the key for "SV" was
>     a bit obtuse.  This commit therefore adds a key for "SELF" and improves
>     the one for "SV".
>     
>     Reported-by: Paolo Bonzini <pbonzini@...hat.com>
>     Signed-off-by: Paul E. McKenney <paulmck@...ux.vnet.ibm.com>
> 
> diff --git a/tools/memory-model/Documentation/cheatsheet.txt b/tools/memory-model/Documentation/cheatsheet.txt
> index c0eafdaddfa4..d502993ac7d2 100644
> --- a/tools/memory-model/Documentation/cheatsheet.txt
> +++ b/tools/memory-model/Documentation/cheatsheet.txt
> @@ -26,4 +26,5 @@ Key:	C:	Ordering is cumulative
>  	DR:	Dependent read (address dependency)
>  	DW:	Dependent write (address, data, or control dependency)
>  	RMW:	Atomic read-modify-write operation
> -	SV	Same-variable access
> +	SELF:	Orders self, as opposed to accesses both before and after
> +	SV:	Orders later accesses to the same variable
>