linux-kernel - Re: [PATCH -next RFC v2 8/8] sbitmap: wake up the number of threads based on required tags

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: <37ecdcd6-3130-fb92-295b-fbb23b53148f@huawei.com>
Date:   Sat, 9 Apr 2022 10:19:32 +0800
From:   "yukuai (C)" <yukuai3@...wei.com>
To:     Bart Van Assche <bvanassche@....org>, <axboe@...nel.dk>,
        <andriy.shevchenko@...ux.intel.com>, <john.garry@...wei.com>,
        <ming.lei@...hat.com>
CC:     <linux-block@...r.kernel.org>, <linux-kernel@...r.kernel.org>,
        <yi.zhang@...wei.com>
Subject: Re: [PATCH -next RFC v2 8/8] sbitmap: wake up the number of threads
 based on required tags

在 2022/04/08 22:31, Bart Van Assche 写道:
> On 4/8/22 00:39, Yu Kuai wrote:
>> Always wake up 'wake_batch' threads will intensify competition and
>> split io won't be issued continuously. Now that how many tags is required
>> is recorded for huge io, it's safe to wake up baed on required tags.
>>
>> Signed-off-by: Yu Kuai <yukuai3@...wei.com>
>> ---
>>   lib/sbitmap.c | 22 +++++++++++++++++++++-
>>   1 file changed, 21 insertions(+), 1 deletion(-)
>>
>> diff --git a/lib/sbitmap.c b/lib/sbitmap.c
>> index 8d01e02ea4b1..eac9fa5c2b4d 100644
>> --- a/lib/sbitmap.c
>> +++ b/lib/sbitmap.c
>> @@ -614,6 +614,26 @@ static inline void sbq_update_preemption(struct 
>> sbitmap_queue *sbq,
>>       WRITE_ONCE(sbq->force_tag_preemption, force);
>>   }
>> +static unsigned int get_wake_nr(struct sbq_wait_state *ws, unsigned 
>> int nr_tags)
> 
> Consider renaming "get_wake_nr()" into "nr_to_wake_up()".
> 
>> +{
>> +    struct sbq_wait *wait;
>> +    struct wait_queue_entry *entry;
>> +    unsigned int nr = 1;
>> +
>> +    spin_lock_irq(&ws->wait.lock);
>> +    list_for_each_entry(entry, &ws->wait.head, entry) {
>> +        wait = container_of(entry, struct sbq_wait, wait);
>> +        if (nr_tags <= wait->nr_tags)
>> +            break;
>> +
>> +        nr++;
>> +        nr_tags -= wait->nr_tags;
>> +    }
>> +    spin_unlock_irq(&ws->wait.lock);
>> +
>> +    return nr;
>> +}
>> +
>>   static bool __sbq_wake_up(struct sbitmap_queue *sbq)
>>   {
>>       struct sbq_wait_state *ws;
>> @@ -648,7 +668,7 @@ static bool __sbq_wake_up(struct sbitmap_queue *sbq)
>>       smp_mb__before_atomic();
>>       atomic_set(&ws->wait_cnt, wake_batch);
>>       sbq_update_preemption(sbq, wake_batch);
>> -    wake_up_nr(&ws->wait, wake_batch);
>> +    wake_up_nr(&ws->wait, get_wake_nr(ws, wake_batch));
>>       return true;
>>   }
> 
> ws->wait.lock is unlocked after the number of threads to wake up has 
> been computed and is locked again by wake_up_nr(). The ws->wait.head 
> list may be modified after get_wake_nr() returns and before wake_up_nr() 
> is called. Isn't that a race condition?
Hi,

That is a race condition, I was hoping that the problem patch 5 fixed
can cover this.

Thanks,
Kuai
> 
> Thanks,
> 
> Bart.
> .
>