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:   Wed, 13 Jun 2018 23:41:34 +0300
From:   "Kirill A. Shutemov" <kirill.shutemov@linux.intel.com>
To:     Dave Hansen <dave.hansen@intel.com>
Cc:     Ingo Molnar <mingo@redhat.com>, x86@kernel.org,
        Thomas Gleixner <tglx@linutronix.de>,
        "H. Peter Anvin" <hpa@zytor.com>,
        Tom Lendacky <thomas.lendacky@amd.com>,
        Kai Huang <kai.huang@linux.intel.com>,
        Jacob Pan <jacob.jun.pan@linux.intel.com>,
        linux-kernel@vger.kernel.org, linux-mm@kvack.org
Subject: Re: [PATCHv3 03/17] mm/ksm: Do not merge pages with different KeyIDs
Message-ID: <20180613204134.voy7iimrlcywblnx@black.fi.intel.com>
References: <20180612143915.68065-1-kirill.shutemov@linux.intel.com>
 <20180612143915.68065-4-kirill.shutemov@linux.intel.com>
 <63b7e88f-33d6-c5c1-f6cb-1bbb780e2cc4@intel.com>
 <20180613203108.k63fda4hvsqyczw7@black.fi.intel.com>
 <355d6b1f-9188-e93b-9cf1-01ece1879cad@intel.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <355d6b1f-9188-e93b-9cf1-01ece1879cad@intel.com>
User-Agent: NeoMutt/20170714-126-deb55f (1.8.3)
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk

On Wed, Jun 13, 2018 at 08:35:46PM +0000, Dave Hansen wrote:
> On 06/13/2018 01:31 PM, Kirill A. Shutemov wrote:
> >> What actually happens without this patch in place?
> > 
> > One of processes would get the page mapped with wrong KeyID and see
> > garbage. 
> 
> OK, but what about two pages with the same KeyID?  It's actually totally
> possible for KSM to determine that two pages have the same plaintext and
> merge them.  Why don't we do that?

That's exactly what we do :)

> > We setup mapping according to KeyID in vma->vm_page_prot.
> 
> Then why do we bother with page_keyid() and the page_ext stuff?

VMA is not always around.

Using KeyID in vma->vm_page_prot we don't need to change anything in PTE
setup functions. It just works.

-- 
 Kirill A. Shutemov