linux-kernel - Re: [PATCH] eventfd: support delayed wakeup for non-semaphore eventfd to reduce cpu utilization

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

Date:   Tue, 18 Apr 2023 22:12:14 +0800
From:   Wen Yang <wenyang.linux@...mail.com>
To:     Hillf Danton <hdanton@...a.com>
Cc:     Christoph Hellwig <hch@....de>, Fu Wei <wefu@...hat.com>,
        linux-kernel@...r.kernel.org
Subject: Re: [PATCH] eventfd: support delayed wakeup for non-semaphore eventfd
 to reduce cpu utilization


在 2023/4/17 08:07, Hillf Danton 写道:
> On 16 Apr 2023 19:31:55 +0800 Wen Yang <wenyang.linux@...mail.com>
>> For the NON SEMAPHORE eventfd, if it's counter has a nonzero value,
>> then a read(2) returns 8 bytes containing that value, and the counter's
>> value is reset to zero. Therefore, in the NON SEMAPHORE scenario,
>> N event_writes vs ONE event_read is possible.
>>
>> However, the current implementation wakes up the read thread immediately
>> in eventfd_write so that the cpu utilization increases unnecessarily.
>>
>> By adding a configurable delay after eventfd_write, these unnecessary
>> wakeup operations are avoided, thereby reducing cpu utilization.
> If the EPOLLOUT wakeup in eventfd_read() breaks, feel free to specify it
> in the commit message after updating the comment below.
> 	/*
> 	 * Every time that a write(2) is performed on an eventfd, the
> 	 * value of the __u64 being written is added to "count" and a
> 	 * wakeup is performed on "wqh". A read(2) will return the "count"
> 	 * value to userspace, and will reset "count" to zero. The kernel
> 	 * side eventfd_signal() also, adds to the "count" counter and
> 	 * issue a wakeup.
> 	 */

Thanks

We will modify it according to your suggestion and then send v2 later.

--

Best wishes,

Wen