linux-kernel - Re: [PATCH -mm -v2 2/6] mm, swap: Add swap readahead hit statistics

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] [day] [month] [year] [list]

Date:   Wed, 12 Jul 2017 10:22:42 +0800
From:   "Huang\, Ying" <ying.huang@...el.com>
To:     Dave Hansen <dave.hansen@...el.com>
Cc:     "Huang\, Ying" <ying.huang@...el.com>,
        Andrew Morton <akpm@...ux-foundation.org>,
        <linux-mm@...ck.org>, <linux-kernel@...r.kernel.org>,
        Minchan Kim <minchan@...nel.org>,
        Rik van Riel <riel@...hat.com>, Shaohua Li <shli@...nel.org>,
        Hugh Dickins <hughd@...gle.com>,
        Fengguang Wu <fengguang.wu@...el.com>,
        Tim Chen <tim.c.chen@...el.com>
Subject: Re: [PATCH -mm -v2 2/6] mm, swap: Add swap readahead hit statistics

Dave Hansen <dave.hansen@...el.com> writes:

> On 06/29/2017 06:44 PM, Huang, Ying wrote:
>>  
>>  static atomic_t swapin_readahead_hits = ATOMIC_INIT(4);
>> +static atomic_long_t swapin_readahead_hits_total = ATOMIC_INIT(0);
>> +static atomic_long_t swapin_readahead_total = ATOMIC_INIT(0);
>>  
>>  void show_swap_cache_info(void)
>>  {
>> @@ -305,8 +307,10 @@ struct page * lookup_swap_cache(swp_entry_t entry)
>>  
>>  	if (page && likely(!PageTransCompound(page))) {
>>  		INC_CACHE_INFO(find_success);
>> -		if (TestClearPageReadahead(page))
>> +		if (TestClearPageReadahead(page)) {
>>  			atomic_inc(&swapin_readahead_hits);
>> +			atomic_long_inc(&swapin_readahead_hits_total);
>> +		}
>>  	}
>
> Adding global atomics that we touch in hot paths seems like poor
> future-proofing.  Are we sure we want to do this and not use some of the
> nice, fancy, percpu counters that we have?

Yes.  It is much better to use percpu counters instead.  Will change it
in the next version.

Best Regards,
Huang, Ying