linux-kernel - Re: [patch] fs: scale vfsmount refcount (was Re: rcu-walk and dcache scaling tree update and status)

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: <1292225113.18698.22.camel@edumazet-laptop>
Date:	Mon, 13 Dec 2010 08:25:13 +0100
From:	Eric Dumazet <eric.dumazet@...il.com>
To:	Nick Piggin <npiggin@...nel.dk>
Cc:	Linus Torvalds <torvalds@...ux-foundation.org>,
	Andrew Morton <akpm@...ux-foundation.org>,
	Al Viro <viro@...IV.linux.org.uk>,
	Stephen Rothwell <sfr@...b.auug.org.au>,
	linux-fsdevel@...r.kernel.org, linux-kernel@...r.kernel.org
Subject: Re: [patch] fs: scale vfsmount refcount (was Re: rcu-walk and
 dcache scaling tree update and status)

Le lundi 13 décembre 2010 à 13:42 +1100, Nick Piggin a écrit :
> + */
> +static inline void add_mnt_count(struct vfsmount *mnt, int n)

maybe name it __add_mnt_count() (should be called with preempt off)

> +{
> +#ifdef CONFIG_SMP
> +	(*per_cpu_ptr(mnt->mnt_count, smp_processor_id())) += n;

__this_cpu_add(mnt->mnt_count, n);

> +#else
> +	mnt->mnt_count += n;
> +#endif
> +}

and define a preempt safe version :

static inline void __add_mnt_count(struct vfsmount *mnt, int n)
{
#ifdef CONFIG_SMP
	__this_cpu_add(mnt->mnt_count, n);
#else
	mnt->mnt_count += n;
#endif
}

static inline void add_mnt_count(struct vfsmount *mnt, int n)
{
#ifdef CONFIG_SMP
	this_cpu_add(mnt->mnt_count, n);
#else
	preempt_disable();
	mnt->mnt_count += n;
	preempt_enable();
#endif
}

> +
> +static inline void set_mnt_count(struct vfsmount *mnt, int n)
> +{
> +#ifdef CONFIG_SMP


> +	preempt_disable();
> +	(*per_cpu_ptr(mnt->mnt_count, smp_processor_id())) = n;
> +	preempt_enable();

last 3 lines can be replaced by :

	this_cpu_write(mnt->mnt_count, n);

> +#else
> +	mnt->mnt_count = n;
> +#endif
> +}
> +

>  #ifdef CONFIG_SMP
>  	int __percpu *mnt_writers;
> +	int __percpu *mnt_count;
>  #else
>  	int mnt_writers;
> +	int mnt_count;
>  #endif

You could use a struct and one per cpu allocation to use one cache line
for both objects :

struct mnt_counters {
	int writers;
	int count;
};

...

#ifdef CONFIG_SMP
	struct mnt_counters __percpu *mnt_counters;
#else
	struct mnt_counters mnt_counters;
#endif

This would use one pointer instead of two in SMP


--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/