This is the mail archive of the libc-alpha@sourceware.org mailing list for the glibc project.

Index Nav:	[Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav:	[Date Prev] [Date Next]	[Thread Prev] [Thread Next]
Other format:	[Raw text]

Re: [PATCH v2 1/3] Tunables: Add tunables of spin count for pthread adaptive spin mutex

From: kemi <kemi dot wang at intel dot com>
To: Florian Weimer <fweimer at redhat dot com>, Adhemerval Zanella <adhemerval dot zanella at linaro dot org>, Glibc alpha <libc-alpha at sourceware dot org>
Cc: Dave Hansen <dave dot hansen at linux dot intel dot com>, Tim Chen <tim dot c dot chen at intel dot com>, Andi Kleen <andi dot kleen at intel dot com>, Ying Huang <ying dot huang at intel dot com>, Aaron Lu <aaron dot lu at intel dot com>, Lu Aubrey <aubrey dot li at intel dot com>
Date: Mon, 14 May 2018 15:37:11 +0800
Subject: Re: [PATCH v2 1/3] Tunables: Add tunables of spin count for pthread adaptive spin mutex
References: <1524624988-29141-1-git-send-email-kemi.wang@intel.com> <0c66f19d-c0e8-accd-85dd-7e55dd6da1af@redhat.com> <55c818fb-1b7e-47d0-0287-2ea33ce69fd5@intel.com> <3b8bb7bf-68a5-4084-e4dd-bb4fe4411bef@redhat.com> <8dbdd127-01ad-5341-1824-52bd18f1d183@intel.com> <5f863e27-f439-8483-2ee2-14f607f7f69c@redhat.com>


On 2018年05月14日 15:30, Florian Weimer wrote:
> On 05/14/2018 06:03 AM, kemi wrote:
> 
>> So, let's set the default value of spin count as 150, and run some benchmark to test it.
>>
>> What's your idea?
> 
> Picking an arbitrary value for now is good enough, and using benchmarks to tune it in the future certainly makes sense.
> 

Agree.
And will enhance lock benchmark in V3.

>> Two reasons for keeping the tunables here:
>> 1) The overhead of instructions are architecture-specific, so, it is hard or even impossible to have a perfect default value that fits all the architecture well.
>> E.g. The pause instruction in the Skylake platform is 10x expensive than before.
> 
> If we have particularly good numbers for one architecture, we could supply an architecture-specific default.
> 

Agree, and I will do later for x86 architecture.

> Thanks,
> Florian

References:
- Re: [PATCH v2 1/3] Tunables: Add tunables of spin count for pthread adaptive spin mutex
  - From: Florian Weimer
- Re: [PATCH v2 1/3] Tunables: Add tunables of spin count for pthread adaptive spin mutex
  - From: kemi
- Re: [PATCH v2 1/3] Tunables: Add tunables of spin count for pthread adaptive spin mutex
  - From: Florian Weimer
- Re: [PATCH v2 1/3] Tunables: Add tunables of spin count for pthread adaptive spin mutex
  - From: kemi
- Re: [PATCH v2 1/3] Tunables: Add tunables of spin count for pthread adaptive spin mutex
  - From: Florian Weimer

Index Nav:	[Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav:	[Date Prev] [Date Next]	[Thread Prev] [Thread Next]