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;
Subject: Re: [PATCH] xfs: optimise xfs_mod_icount/ifree when delta < 0
To:     Christoph Hellwig <hch@infradead.org>
References: <1572866980-13001-1-git-send-email-zhangshaokun@hisilicon.com>
 <20191104152517.GD10485@infradead.org>
CC:     <linux-xfs@vger.kernel.org>, <linux-kernel@vger.kernel.org>,
        Yang Guo <guoyang2@huawei.com>,
        "Darrick J. Wong" <darrick.wong@oracle.com>
From:   Shaokun Zhang <zhangshaokun@hisilicon.com>
Message-ID: <c2c0ccb7-64c9-4549-6683-0984073ccd0b@hisilicon.com>
Date:   Tue, 5 Nov 2019 11:08:01 +0800
User-Agent: Mozilla/5.0 (Windows NT 6.1; WOW64; rv:45.0) Gecko/20100101
 Thunderbird/45.1.1
MIME-Version: 1.0
In-Reply-To: <20191104152517.GD10485@infradead.org>
Content-Type: text/plain; charset="windows-1252"
Content-Transfer-Encoding: 7bit
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk

Hi Christoph,

On 2019/11/4 23:25, Christoph Hellwig wrote:
> On Mon, Nov 04, 2019 at 07:29:40PM +0800, Shaokun Zhang wrote:
>> From: Yang Guo <guoyang2@huawei.com>
>>
>> percpu_counter_compare will be called by xfs_mod_icount/ifree to check
>> whether the counter less than 0 and it is a expensive function.
>> let's check it only when delta < 0, it will be good for xfs's performance.
> 
> How much overhead do you see?  In the end the compare is just a debug

Thanks your reply, sorry for my not clear description.
__percpu_counter_compare itself is not expensive, but __percpu_counter_sum
called by __percpu_counter_compare is high load, I will list it in next thread.

> check, so if it actually shows up we should remove it entirely.
> 

I'm not sure about it, so I check the delta to do less modification.

Thanks,
Shaokun

>