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, 14 Nov 2018 04:09:47 -0500 (EST)
From:   Pankaj Gupta <pagupta@redhat.com>
To:     Barret Rhoden <brho@google.com>
Cc:     Dan Williams <dan.j.williams@intel.com>,
        David Hildenbrand <david@redhat.com>,
        Dave Jiang <dave.jiang@intel.com>,
        Ross Zwisler <zwisler@kernel.org>,
        Vishal Verma <vishal.l.verma@intel.com>,
        Paolo Bonzini <pbonzini@redhat.com>,
        Radim =?utf-8?B?S3LEjW3DocWZ?= <rkrcmar@redhat.com>,
        Thomas Gleixner <tglx@linutronix.de>,
        Ingo Molnar <mingo@redhat.com>, Borislav Petkov <bp@alien8.de>,
        linux-nvdimm@lists.01.org, linux-kernel@vger.kernel.org,
        "H. Peter Anvin" <hpa@zytor.com>, x86@kernel.org,
        kvm@vger.kernel.org, yu c zhang <yu.c.zhang@intel.com>,
        yi z zhang <yi.z.zhang@intel.com>
Message-ID: <1826878055.33695230.1542186587142.JavaMail.zimbra@redhat.com>
In-Reply-To: <20181113105628.37449f52@gnomeregan.cam.corp.google.com>
References: <20181109203921.178363-1-brho@google.com> <20181109203921.178363-3-brho@google.com> <043a592d-6592-3053-15a0-68cc54a26deb@redhat.com> <286665658.33247363.1542103353780.JavaMail.zimbra@redhat.com> <20181113105628.37449f52@gnomeregan.cam.corp.google.com>
Subject: Re: [PATCH 2/2] kvm: Use huge pages for DAX-backed files
MIME-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: 7bit
Thread-Topic: Use huge pages for DAX-backed files
Thread-Index: zeMs9nesIGx8sF5CHL5ykF2Xzt/5dg==
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk


> > As this patch is dependent on PageReserved patch(which is in progress),
> > just
> > wondering if we are able to test the code path for hugepage with DAX.
> 
> For testing, I used the following patch.  It's not 100%, since it
> intercepts at kvm_is_reserved_pfn(), and not PageReserved() directly.
> The only difference is with kvm_set_pfn_dirty() I think.
> 

Yes, this should be ok.

Thanks,
Pankaj 

> I also have a nasty module that would dump the EPT's and the host page
> table's mappings so I could confirm that the huge pages are being mapped
> correctly.
> 
> -----------------------
> diff --git a/virt/kvm/kvm_main.c b/virt/kvm/kvm_main.c
> index 2679e476b6c3..1b394a0752a0 100644
> --- a/virt/kvm/kvm_main.c
> +++ b/virt/kvm/kvm_main.c
> @@ -148,6 +148,10 @@ __weak int kvm_arch_mmu_notifier_invalidate_range(struct
> kvm *kvm,
>  
>  bool kvm_is_reserved_pfn(kvm_pfn_t pfn)
>  {
> +	// XXX hack
> +	if (is_zone_device_page(pfn_to_page(pfn)))
> +		return false;
> +
>  	if (pfn_valid(pfn))
>  		return PageReserved(pfn_to_page(pfn));
>  
> -----------------------
> 
> Thanks,
> 
> Barret
> 
>