Received-SPF: pass (google.com: best guess record for domain of linux-kernel-owner@vger.kernel.org designates 209.132.180.67 as permitted sender) client-ip=209.132.180.67;
Date:   Thu, 21 Jun 2018 21:19:27 -0400 (EDT)
From:   Mikulas Patocka <mpatocka@redhat.com>
To:     Ingo Molnar <mingo@kernel.org>
cc:     Mike Snitzer <snitzer@redhat.com>,
        Thomas Gleixner <tglx@linutronix.de>,
        Dan Williams <dan.j.williams@intel.com>,
        device-mapper development <dm-devel@redhat.com>,
        X86 ML <x86@kernel.org>, linux-kernel@vger.kernel.org
Subject: Re: [PATCH v2 RESEND] x86: optimize memcpy_flushcache
In-Reply-To: <20180621143140.GA14095@gmail.com>
Message-ID: <alpine.LRH.2.02.1806211224470.19940@file01.intranet.prod.int.rdu2.redhat.com>
References: <20180519052503.325953342@debian.vm> <20180519052631.730455475@debian.vm> <CAPcyv4jpY0x7kZtT+afAKnHKy8Uy1AC_N_QM7RrELSj_0iNrRw@mail.gmail.com> <20180524182013.GA59755@redhat.com> <20180618132306.GA25431@redhat.com>
 <20180621143140.GA14095@gmail.com>
User-Agent: Alpine 2.02 (LRH 1266 2009-07-14)
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk


On Thu, 21 Jun 2018, Ingo Molnar wrote:

> 
> * Mike Snitzer <snitzer@redhat.com> wrote:
> 
> > From: Mikulas Patocka <mpatocka@redhat.com>
> > Subject: [PATCH v2] x86: optimize memcpy_flushcache
> > 
> > In the context of constant short length stores to persistent memory,
> > memcpy_flushcache suffers from a 2% performance degradation compared to
> > explicitly using the "movnti" instruction.
> > 
> > Optimize 4, 8, and 16 byte memcpy_flushcache calls to explicitly use the
> > movnti instruction with inline assembler.
> 
> Linus requested asm optimizations to include actual benchmarks, so it would be 
> nice to describe how this was tested, on what hardware, and what the before/after 
> numbers are.
> 
> Thanks,
> 
> 	Ingo

It was tested on 4-core skylake machine with persistent memory being 
emulated using the memmap kernel option. The dm-writecache target used the 
emulated persistent memory as a cache and sata SSD as a backing device. 
The patch results in 2% improved throughput when writing data using dd.

I don't have access to the machine anymore.

Mikulas