linux-kernel - Re: [RFC][PATCH 2/5] refcount: Use atomic_*

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]

Message-ID: <20211208205646.GB16608@worktop.programming.kicks-ass.net>
Date:   Wed, 8 Dec 2021 21:56:46 +0100
From:   Peter Zijlstra <peterz@...radead.org>
To:     will@...nel.org, boqun.feng@...il.com
Cc:     linux-kernel@...r.kernel.org, x86@...nel.org, mark.rutland@....com,
        elver@...gle.com, keescook@...omium.org, hch@...radead.org,
        torvalds@...ux-foundation.org
Subject: Re: [RFC][PATCH 2/5] refcount: Use atomic_*_ofl()

On Wed, Dec 08, 2021 at 08:19:35PM +0100, Peter Zijlstra wrote:
> On Wed, Dec 08, 2021 at 07:36:57PM +0100, Peter Zijlstra wrote:
> > Use the shiny new atomic_*_ofl() functions in order to have better
> > code-gen.
> > 
> 
> *sigh*, so I forgot to adjust the other primitives to the slightly
> tweaked range of these new functions..

Looks like it all just works. The add/sub stuff needs to check before
and after anyway, and it saturing slightly earlier isn't a problem.

The only difference I found was, so I'll just fold that.

--- a/include/linux/refcount.h
+++ b/include/linux/refcount.h
@@ -346,7 +346,7 @@ static inline void __refcount_dec(refcou
 	if (oldp)
 		*oldp = old;
 
-	if (unlikely(old <= 1))
+	if (unlikely(old - 1 <= 0))
 		refcount_warn_saturate(r, REFCOUNT_DEC_LEAK);
 }