From mboxrd@z Thu Jan  1 00:00:00 1970
Return-Path: <linux-kernel-owner@vger.kernel.org>
Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand
        id S1751562AbdBIGdd (ORCPT <rfc822;w@1wt.eu>);
        Thu, 9 Feb 2017 01:33:33 -0500
Received: from out0-154.mail.aliyun.com ([140.205.0.154]:58464 "EHLO
        out0-154.mail.aliyun.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org
        with ESMTP id S1751114AbdBIGdc (ORCPT
        <rfc822;linux-kernel@vger.kernel.org>);
        Thu, 9 Feb 2017 01:33:32 -0500
X-Alimail-AntiSpam: AC=PASS;BC=-1|-1;BR=01201311R191e4;FP=0|-1|-1|-1|0|-1|-1|-1;HT=e02c03311;MF=hillf.zj@alibaba-inc.com;NM=1;PH=DS;RN=13;SR=0;TI=SMTPD_---.7aHfPSO_1486621984;
Reply-To: "Hillf Danton" <hillf.zj@alibaba-inc.com>
From: "Hillf Danton" <hillf.zj@alibaba-inc.com>
To: "Hillf Danton" <hillf.zj@alibaba-inc.com>,
        "'Shaohua Li'" <shli@fb.com>, <linux-kernel@vger.kernel.org>,
        <linux-mm@kvack.org>
Cc: <Kernel-team@fb.com>, <danielmicay@gmail.com>, <mhocko@suse.com>,
        <minchan@kernel.org>, <hughd@google.com>, <hannes@cmpxchg.org>,
        <riel@redhat.com>, <mgorman@techsingularity.net>,
        <akpm@linux-foundation.org>
References: <cover.1486163864.git.shli@fb.com> <3914c9f53c343357c39cb891210da31aa30ad3a9.1486163864.git.shli@fb.com> <007e01d27eb1$3f98dee0$beca9ca0$@alibaba-inc.com>
In-Reply-To: <007e01d27eb1$3f98dee0$beca9ca0$@alibaba-inc.com>
Subject: Re: [PATCH V2 2/7] mm: move MADV_FREE pages into LRU_INACTIVE_FILE list
Date: Thu, 09 Feb 2017 14:33:03 +0800
Message-ID: <00d601d2829e$5e7930d0$1b6b9270$@alibaba-inc.com>
MIME-Version: 1.0
Content-Type: text/plain;
        charset="us-ascii"
Content-Transfer-Encoding: 7bit
X-Mailer: Microsoft Outlook 14.0
Thread-Index: AQHL8bUnrW28kOshlkGIqYQq7mO9XQMVr0wgAiVuBmOhQ34rMA==
Content-Language: zh-cn
Sender: linux-kernel-owner@vger.kernel.org
List-ID: <linux-kernel.vger.kernel.org>
X-Mailing-List: linux-kernel@vger.kernel.org


On February 04, 2017 2:38 PM Hillf Danton wrote: 
> 
> On February 04, 2017 7:33 AM Shaohua Li wrote:
> > @@ -1404,6 +1401,8 @@ bool madvise_free_huge_pmd(struct mmu_gather *tlb, struct vm_area_struct *vma,
> >  		set_pmd_at(mm, addr, pmd, orig_pmd);
> >  		tlb_remove_pmd_tlb_entry(tlb, pmd, addr);
> >  	}
> > +
> > +	mark_page_lazyfree(page);
> >  	ret = true;
> >  out:
> >  	spin_unlock(ptl);
> 
> <snipped>
> 
> > -void deactivate_page(struct page *page)
> > -{
> > -	if (PageLRU(page) && PageActive(page) && !PageUnevictable(page)) {
> > -		struct pagevec *pvec = &get_cpu_var(lru_deactivate_pvecs);
> > +void mark_page_lazyfree(struct page *page)
> > + {
> > +	if (PageLRU(page) && PageAnon(page) && PageSwapBacked(page) &&
> > +	    !PageUnevictable(page)) {
> > +		struct pagevec *pvec = &get_cpu_var(lru_lazyfree_pvecs);
> >
> >  		get_page(page);
> >  		if (!pagevec_add(pvec, page) || PageCompound(page))
> > -			pagevec_lru_move_fn(pvec, lru_deactivate_fn, NULL);
> > -		put_cpu_var(lru_deactivate_pvecs);
> > +			pagevec_lru_move_fn(pvec, lru_lazyfree_fn, NULL);
> > +		put_cpu_var(lru_lazyfree_pvecs);
> >  	}
> >  }
> 
> You are not adding it but would you please try to fix or avoid flipping
> preempt count with page table lock hold?
> 
preempt_en/disable are embedded in spin_lock/unlock, so please
ignore my noise.

thanks
Hillf

From mboxrd@z Thu Jan  1 00:00:00 1970
Return-Path: <owner-linux-mm@kvack.org>
Received: from mail-pg0-f69.google.com (mail-pg0-f69.google.com [74.125.83.69])
	by kanga.kvack.org (Postfix) with ESMTP id 049FF28089F
	for <linux-mm@kvack.org>; Thu,  9 Feb 2017 01:33:26 -0500 (EST)
Received: by mail-pg0-f69.google.com with SMTP id 204so221873730pge.5
        for <linux-mm@kvack.org>; Wed, 08 Feb 2017 22:33:25 -0800 (PST)
Received: from out4435.biz.mail.alibaba.com (out4435.biz.mail.alibaba.com. [47.88.44.35])
        by mx.google.com with ESMTP id 1si9243286plw.105.2017.02.08.22.33.23
        for <linux-mm@kvack.org>;
        Wed, 08 Feb 2017 22:33:25 -0800 (PST)
Reply-To: "Hillf Danton" <hillf.zj@alibaba-inc.com>
From: "Hillf Danton" <hillf.zj@alibaba-inc.com>
References: <cover.1486163864.git.shli@fb.com> <3914c9f53c343357c39cb891210da31aa30ad3a9.1486163864.git.shli@fb.com> <007e01d27eb1$3f98dee0$beca9ca0$@alibaba-inc.com>
In-Reply-To: <007e01d27eb1$3f98dee0$beca9ca0$@alibaba-inc.com>
Subject: Re: [PATCH V2 2/7] mm: move MADV_FREE pages into LRU_INACTIVE_FILE list
Date: Thu, 09 Feb 2017 14:33:03 +0800
Message-ID: <00d601d2829e$5e7930d0$1b6b9270$@alibaba-inc.com>
MIME-Version: 1.0
Content-Type: text/plain;
	charset="us-ascii"
Content-Transfer-Encoding: 7bit
Content-Language: zh-cn
Sender: owner-linux-mm@kvack.org
List-ID: <linux-mm.kvack.org>
To: Hillf Danton <hillf.zj@alibaba-inc.com>, 'Shaohua Li' <shli@fb.com>, linux-kernel@vger.kernel.org, linux-mm@kvack.org
Cc: Kernel-team@fb.com, danielmicay@gmail.com, mhocko@suse.com, minchan@kernel.org, hughd@google.com, hannes@cmpxchg.org, riel@redhat.com, mgorman@techsingularity.net, akpm@linux-foundation.org


On February 04, 2017 2:38 PM Hillf Danton wrote: 
> 
> On February 04, 2017 7:33 AM Shaohua Li wrote:
> > @@ -1404,6 +1401,8 @@ bool madvise_free_huge_pmd(struct mmu_gather *tlb, struct vm_area_struct *vma,
> >  		set_pmd_at(mm, addr, pmd, orig_pmd);
> >  		tlb_remove_pmd_tlb_entry(tlb, pmd, addr);
> >  	}
> > +
> > +	mark_page_lazyfree(page);
> >  	ret = true;
> >  out:
> >  	spin_unlock(ptl);
> 
> <snipped>
> 
> > -void deactivate_page(struct page *page)
> > -{
> > -	if (PageLRU(page) && PageActive(page) && !PageUnevictable(page)) {
> > -		struct pagevec *pvec = &get_cpu_var(lru_deactivate_pvecs);
> > +void mark_page_lazyfree(struct page *page)
> > + {
> > +	if (PageLRU(page) && PageAnon(page) && PageSwapBacked(page) &&
> > +	    !PageUnevictable(page)) {
> > +		struct pagevec *pvec = &get_cpu_var(lru_lazyfree_pvecs);
> >
> >  		get_page(page);
> >  		if (!pagevec_add(pvec, page) || PageCompound(page))
> > -			pagevec_lru_move_fn(pvec, lru_deactivate_fn, NULL);
> > -		put_cpu_var(lru_deactivate_pvecs);
> > +			pagevec_lru_move_fn(pvec, lru_lazyfree_fn, NULL);
> > +		put_cpu_var(lru_lazyfree_pvecs);
> >  	}
> >  }
> 
> You are not adding it but would you please try to fix or avoid flipping
> preempt count with page table lock hold?
> 
preempt_en/disable are embedded in spin_lock/unlock, so please
ignore my noise.

thanks
Hillf

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org.  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>