Received-SPF: pass (google.com: domain of linux-kernel+bounces-152902-linux.lists.archive=gmail.com@vger.kernel.org designates 147.75.80.249 as permitted sender) client-ip=147.75.80.249;
Message-ID: <e1a06ea3-57b2-4562-895b-a2fb5d5667cc@suse.cz>
Date: Mon, 22 Apr 2024 09:49:04 +0200
Precedence: bulk
MIME-Version: 1.0
User-Agent: Mozilla Thunderbird
Subject: Re: [PATCH v3 1/2] slub: introduce count_partial_free_approx()
To: David Rientjes <rientjes@google.com>,
 Jianfeng Wang <jianfeng.w.wang@oracle.com>
Cc: linux-mm@kvack.org, linux-kernel@vger.kernel.org, cl@linux.com,
 akpm@linux-foundation.org, penberg@kernel.org
References: <20240419175611.47413-1-jianfeng.w.wang@oracle.com>
 <20240419175611.47413-2-jianfeng.w.wang@oracle.com>
 <3e5d2937-76ab-546b-9ce8-7e7140424278@google.com>
Content-Language: en-US
From: Vlastimil Babka <vbabka@suse.cz>
Autocrypt: addr=vbabka@suse.cz; keydata=
 xsFNBFZdmxYBEADsw/SiUSjB0dM+vSh95UkgcHjzEVBlby/Fg+g42O7LAEkCYXi/vvq31JTB
 KxRWDHX0R2tgpFDXHnzZcQywawu8eSq0LxzxFNYMvtB7sV1pxYwej2qx9B75qW2plBs+7+YB
 87tMFA+u+L4Z5xAzIimfLD5EKC56kJ1CsXlM8S/LHcmdD9Ctkn3trYDNnat0eoAcfPIP2OZ+
 9oe9IF/R28zmh0ifLXyJQQz5ofdj4bPf8ecEW0rhcqHfTD8k4yK0xxt3xW+6Exqp9n9bydiy
 tcSAw/TahjW6yrA+6JhSBv1v2tIm+itQc073zjSX8OFL51qQVzRFr7H2UQG33lw2QrvHRXqD
 Ot7ViKam7v0Ho9wEWiQOOZlHItOOXFphWb2yq3nzrKe45oWoSgkxKb97MVsQ+q2SYjJRBBH4
 8qKhphADYxkIP6yut/eaj9ImvRUZZRi0DTc8xfnvHGTjKbJzC2xpFcY0DQbZzuwsIZ8OPJCc
 LM4S7mT25NE5kUTG/TKQCk922vRdGVMoLA7dIQrgXnRXtyT61sg8PG4wcfOnuWf8577aXP1x
 6mzw3/jh3F+oSBHb/GcLC7mvWreJifUL2gEdssGfXhGWBo6zLS3qhgtwjay0Jl+kza1lo+Cv
 BB2T79D4WGdDuVa4eOrQ02TxqGN7G0Biz5ZLRSFzQSQwLn8fbwARAQABzSBWbGFzdGltaWwg
 QmFia2EgPHZiYWJrYUBzdXNlLmN6PsLBlAQTAQoAPgIbAwULCQgHAwUVCgkICwUWAgMBAAIe
 AQIXgBYhBKlA1DSZLC6OmRA9UCJPp+fMgqZkBQJkBREIBQkRadznAAoJECJPp+fMgqZkNxIQ
 ALZRqwdUGzqL2aeSavbum/VF/+td+nZfuH0xeWiO2w8mG0+nPd5j9ujYeHcUP1edE7uQrjOC
 Gs9sm8+W1xYnbClMJTsXiAV88D2btFUdU1mCXURAL9wWZ8Jsmz5ZH2V6AUszvNezsS/VIT87
 AmTtj31TLDGwdxaZTSYLwAOOOtyqafOEq+gJB30RxTRE3h3G1zpO7OM9K6ysLdAlwAGYWgJJ
 V4JqGsQ/lyEtxxFpUCjb5Pztp7cQxhlkil0oBYHkudiG8j1U3DG8iC6rnB4yJaLphKx57NuQ
 PIY0Bccg+r9gIQ4XeSK2PQhdXdy3UWBr913ZQ9AI2usid3s5vabo4iBvpJNFLgUmxFnr73SJ
 KsRh/2OBsg1XXF/wRQGBO9vRuJUAbnaIVcmGOUogdBVS9Sun/Sy4GNA++KtFZK95U7J417/J
 Hub2xV6Ehc7UGW6fIvIQmzJ3zaTEfuriU1P8ayfddrAgZb25JnOW7L1zdYL8rXiezOyYZ8Fm
 ZyXjzWdO0RpxcUEp6GsJr11Bc4F3aae9OZtwtLL/jxc7y6pUugB00PodgnQ6CMcfR/HjXlae
 h2VS3zl9+tQWHu6s1R58t5BuMS2FNA58wU/IazImc/ZQA+slDBfhRDGYlExjg19UXWe/gMcl
 De3P1kxYPgZdGE2eZpRLIbt+rYnqQKy8UxlszsBNBFsZNTUBCACfQfpSsWJZyi+SHoRdVyX5
 J6rI7okc4+b571a7RXD5UhS9dlVRVVAtrU9ANSLqPTQKGVxHrqD39XSw8hxK61pw8p90pg4G
 /N3iuWEvyt+t0SxDDkClnGsDyRhlUyEWYFEoBrrCizbmahOUwqkJbNMfzj5Y7n7OIJOxNRkB
 IBOjPdF26dMP69BwePQao1M8Acrrex9sAHYjQGyVmReRjVEtv9iG4DoTsnIR3amKVk6si4Ea
 X/mrapJqSCcBUVYUFH8M7bsm4CSxier5ofy8jTEa/CfvkqpKThTMCQPNZKY7hke5qEq1CBk2
 wxhX48ZrJEFf1v3NuV3OimgsF2odzieNABEBAAHCwXwEGAEKACYCGwwWIQSpQNQ0mSwujpkQ
 PVAiT6fnzIKmZAUCZAUSmwUJDK5EZgAKCRAiT6fnzIKmZOJGEACOKABgo9wJXsbWhGWYO7mD
 8R8mUyJHqbvaz+yTLnvRwfe/VwafFfDMx5GYVYzMY9TWpA8psFTKTUIIQmx2scYsRBUwm5VI
 EurRWKqENcDRjyo+ol59j0FViYysjQQeobXBDDE31t5SBg++veI6tXfpco/UiKEsDswL1WAr
 tEAZaruo7254TyH+gydURl2wJuzo/aZ7Y7PpqaODbYv727Dvm5eX64HCyyAH0s6sOCyGF5/p
 eIhrOn24oBf67KtdAN3H9JoFNUVTYJc1VJU3R1JtVdgwEdr+NEciEfYl0O19VpLE/PZxP4wX
 PWnhf5WjdoNI1Xec+RcJ5p/pSel0jnvBX8L2cmniYnmI883NhtGZsEWj++wyKiS4NranDFlA
 HdDM3b4lUth1pTtABKQ1YuTvehj7EfoWD3bv9kuGZGPrAeFNiHPdOT7DaXKeHpW9homgtBxj
 8aX/UkSvEGJKUEbFL9cVa5tzyialGkSiZJNkWgeHe+jEcfRT6pJZOJidSCdzvJpbdJmm+eED
 w9XOLH1IIWh7RURU7G1iOfEfmImFeC3cbbS73LQEFGe1urxvIH5K/7vX+FkNcr9ujwWuPE9b
 1C2o4i/yZPLXIVy387EjA6GZMqvQUFuSTs/GeBcv0NjIQi8867H3uLjz+mQy63fAitsDwLmR
 EP+ylKVEKb0Q2A==
In-Reply-To: <3e5d2937-76ab-546b-9ce8-7e7140424278@google.com>
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 7bit

On 4/20/24 2:18 AM, David Rientjes wrote:
> On Fri, 19 Apr 2024, Jianfeng Wang wrote:
> 
>> diff --git a/mm/slub.c b/mm/slub.c
>> index 1bb2a93cf7b6..993cbbdd2b6c 100644
>> --- a/mm/slub.c
>> +++ b/mm/slub.c
>> @@ -3213,6 +3213,43 @@ static inline bool free_debug_processing(struct kmem_cache *s,
>>  #endif /* CONFIG_SLUB_DEBUG */
>>  
>>  #if defined(CONFIG_SLUB_DEBUG) || defined(SLAB_SUPPORTS_SYSFS)
>> +#define MAX_PARTIAL_TO_SCAN 10000
>> +
>> +static unsigned long count_partial_free_approx(struct kmem_cache_node *n)
>> +{
>> +	unsigned long flags;
>> +	unsigned long x = 0;
>> +	struct slab *slab;
>> +
>> +	spin_lock_irqsave(&n->list_lock, flags);
>> +	if (n->nr_partial <= MAX_PARTIAL_TO_SCAN) {
>> +		list_for_each_entry(slab, &n->partial, slab_list)
>> +			x += slab->objects - slab->inuse;
>> +	} else {
>> +		/*
>> +		 * For a long list, approximate the total count of objects in
>> +		 * it to meet the limit on the number of slabs to scan.
>> +		 * Scan from both the list's head and tail for better accuracy.
>> +		 */
>> +		unsigned long scanned = 0;
>> +
>> +		list_for_each_entry(slab, &n->partial, slab_list) {
>> +			x += slab->objects - slab->inuse;
>> +			if (++scanned == MAX_PARTIAL_TO_SCAN / 2)
>> +				break;
>> +		}
>> +		list_for_each_entry_reverse(slab, &n->partial, slab_list) {
>> +			x += slab->objects - slab->inuse;
>> +			if (++scanned == MAX_PARTIAL_TO_SCAN)
>> +				break;
>> +		}
>> +		x = mult_frac(x, n->nr_partial, scanned);
>> +		x = min(x, node_nr_objs(n));
>> +	}
>> +	spin_unlock_irqrestore(&n->list_lock, flags);
>> +	return x;
>> +}
> 
> Creative :)
> 
> The default value of MAX_PARTIAL_TO_SCAN seems to work well in practice 
> while being large enough to bias for actual values?
> 
> I can't think of a better way to avoid the disruption that very long 
> partial lists cause.  If the actual value is needed, it will need to be 
> read from the sysfs file for that slab cache.
> 
> It does beg the question of whether we want to extend slabinfo to indicate 
> that some fields are approximations, however.  Adding a suffix such as 
> " : approx" to a slab cache line may be helpful if the disparity in the 
> estimates would actually make a difference in practice.

I'm afraid that changing the layout of /proc/slabinfo has a much higher
chance of breaking some consumer, than the imprecision due to approximation
has. So I would rather not risk it.

> I have a hard time believing that this approximation will not be "close 
> enough" for all practical purposes, given that the value could very well 
> substantially change the instant after the iteration is done anyway.
> 
> So for that reason, this sounds good to me!
> 
> Acked-by: David Rientjes <rientjes@google.com>
> 
>> +
>>  static unsigned long count_partial(struct kmem_cache_node *n,
>>  					int (*get_count)(struct slab *))
>>  {
>> @@ -7089,7 +7126,7 @@ void get_slabinfo(struct kmem_cache *s, struct slabinfo *sinfo)
>>  	for_each_kmem_cache_node(s, node, n) {
>>  		nr_slabs += node_nr_slabs(n);
>>  		nr_objs += node_nr_objs(n);
>> -		nr_free += count_partial(n, count_free);
>> +		nr_free += count_partial_free_approx(n);
>>  	}
>>  
>>  	sinfo->active_objs = nr_objs - nr_free;