Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::3:3 as permitted sender) client-ip=2620:137:e000::3:3;
Date:   Thu, 23 Nov 2023 16:24:11 -0500
From:   Kent Overstreet <kent.overstreet@linux.dev>
To:     Qi Zheng <arch0.zheng@gmail.com>
Cc:     linux-mm@kvack.org, linux-kernel@vger.kernel.org,
        Andrew Morton <akpm@linux-foundation.org>,
        Qi Zheng <zhengqi.arch@bytedance.com>,
        Roman Gushchin <roman.gushchin@linux.dev>,
        Dave Chinner <david@fromorbit.com>
Subject: Re: [PATCH 2/7] mm: shrinker: Add a .to_text() method for shrinkers
Message-ID: <20231123212411.s6r5ekvkklvhwfra@moria.home.lan>
References: <20231122232515.177833-1-kent.overstreet@linux.dev>
 <20231122232515.177833-3-kent.overstreet@linux.dev>
 <deed9bb1-02b9-4e89-895b-38a84e5a9408@gmail.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <deed9bb1-02b9-4e89-895b-38a84e5a9408@gmail.com>
Precedence: bulk

On Thu, Nov 23, 2023 at 11:32:59AM +0800, Qi Zheng wrote:
> > +	void (*to_text)(struct seq_buf *, struct shrinker *);
> 
> The "to_text" looks a little strange, how about naming it
> "stat_objects"?

The convention I've been using heavily in bcachefs is
typename_to_text(), or type.to_text(), for debug reports. The
consistency is nice.

> 
> >   	long batch;	/* reclaim batch size, 0 = default */
> >   	int seeks;	/* seeks to recreate an obj */
> > @@ -110,7 +114,6 @@ struct shrinker {
> >   #endif
> >   #ifdef CONFIG_SHRINKER_DEBUG
> >   	int debugfs_id;
> > -	const char *name;
> 
> The name will only be allocated memory when the CONFIG_SHRINKER_DEBUG is
> enabled, otherwise its value is NULL. So you can't move it out and use
> it while CONFIG_SHRINKER_DEBUG is disabled.

Good catch

> > +void shrinkers_to_text(struct seq_buf *out)
> > +{
> > +	struct shrinker *shrinker;
> > +	struct shrinker_by_mem {
> > +		struct shrinker	*shrinker;
> > +		unsigned long	mem;
> 
> The "mem" also looks a little strange, how about naming it
> "freeable"?

The type is just used in this one function, but sure.

> 
> > +	} shrinkers_by_mem[10];
> > +	int i, nr = 0;
> > +
> > +	if (!mutex_trylock(&shrinker_mutex)) {
> > +		seq_buf_puts(out, "(couldn't take shrinker lock)");
> > +		return;
> > +	}
> 
> We now have lockless method (RCU + refcount) to iterate shrinker_list,
> please refer to shrink_slab().

Will do!