linux-kernel - Re: [LKP] [mm] 9bc8039e71: will-it-scale.per_thread

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, 30 Jan 2019 21:54:15 -0500
From:   Waiman Long <longman@...hat.com>
To:     Tim Chen <tim.c.chen@...ux.intel.com>,
        "Wang, Kemi" <kemi.wang@...el.com>,
        Linus Torvalds <torvalds@...ux-foundation.org>,
        "vbabka@...e.cz" <vbabka@...e.cz>,
        Davidlohr Bueso <dave@...olabs.net>
Cc:     "yang.shi@...ux.alibaba.com" <yang.shi@...ux.alibaba.com>,
        Linux Kernel Mailing List <linux-kernel@...r.kernel.org>,
        Matthew Wilcox <willy@...radead.org>,
        "mhocko@...nel.org" <mhocko@...nel.org>,
        Colin King <colin.king@...onical.com>,
        Andrew Morton <akpm@...ux-foundation.org>,
        "ldufour@...ux.vnet.ibm.com" <ldufour@...ux.vnet.ibm.com>,
        "lkp@...org" <lkp@...org>,
        "kirill.shutemov@...ux.intel.com" <kirill.shutemov@...ux.intel.com>
Subject: Re: [LKP] [mm] 9bc8039e71: will-it-scale.per_thread_ops -64.1%
 regression

On 01/30/2019 07:06 PM, Tim Chen wrote:
> On 12/27/18 6:55 PM, Waiman Long wrote:
>> On 12/27/2018 08:31 PM, Wang, Kemi wrote:
>>> Hi, Waiman
>>>    Did you post that patch? Let's see if it helps.
>> I did post the patch a while ago. I will need to rebase it to a new
>> baseline. Will do that in a week or 2.
>>
>> -Longman
>>
> Waiman,
>
> In a large scale OLTP benchmark, we are also seeing a 1% loss in performance
> from patch 9bc8039e71. We'll be interested to try out your patch to
> restore some optimistic spinning on reader held sem.
>
> Thanks.
>
> Tim

I am reworking the rwsem code and reader-optimistic spinning code isn't
ready with the reworked code yet. I need a bit more time to work on that
and certainly keep you posted once it is ready for review.

Cheers,
Longman