From mboxrd@z Thu Jan  1 00:00:00 1970
From: Jan Kiszka <jan.kiszka@web.de>
Subject: Re: [PATCH 1/1 v2] KVM: MMU: Optimize guest page table walk
Date: Tue, 26 Apr 2011 09:45:39 +0200
Message-ID: <4DB67823.4010304@web.de>
References: <20110422003222.9d08aee3.takuya.yoshikawa@gmail.com> <20110422003444.5b3a876a.takuya.yoshikawa@gmail.com> <4DB52B1B.5080407@web.de> <4DB6775F.9050207@redhat.com>
Mime-Version: 1.0
Content-Type: multipart/signed; micalg=pgp-sha1;
 protocol="application/pgp-signature";
 boundary="------------enig04FCE2230DA68490037E4E28"
Cc: Takuya Yoshikawa <takuya.yoshikawa@gmail.com>, mtosatti@redhat.com,
	kvm@vger.kernel.org, yoshikawa.takuya@oss.ntt.co.jp,
	xiaoguangrong@cn.fujitsu.com, Joerg.Roedel@amd.com
To: Avi Kivity <avi@redhat.com>
Return-path: <kvm-owner@vger.kernel.org>
Received: from fmmailgate03.web.de ([217.72.192.234]:40128 "EHLO
	fmmailgate03.web.de" rhost-flags-OK-OK-OK-OK) by vger.kernel.org
	with ESMTP id S1755598Ab1DZHqm (ORCPT <rfc822;kvm@vger.kernel.org>);
	Tue, 26 Apr 2011 03:46:42 -0400
In-Reply-To: <4DB6775F.9050207@redhat.com>
Sender: kvm-owner@vger.kernel.org
List-ID: <kvm.vger.kernel.org>

This is an OpenPGP/MIME signed message (RFC 2440 and 3156)
--------------enig04FCE2230DA68490037E4E28
Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: quoted-printable

On 2011-04-26 09:42, Avi Kivity wrote:
> On 04/25/2011 11:04 AM, Jan Kiszka wrote:
>> >  +
>> >  +        ptep_user =3D (pt_element_t __user *)((void *)host_addr +
>> offset);
>> >  +        if (get_user(pte, ptep_user)) {
>>                      ^^^^^^^^^^^^
>> This doesn't work for x86-32: pte is 64 bit, but get_user is only
>> defined up to 32 bit on that platform.
>>
>=20
> I actually considered this, and saw:
>=20
> #ifdef CONFIG_X86_32
> #define __get_user_8(__ret_gu, __val_gu, ptr)                \
>         __get_user_x(X, __ret_gu, __val_gu, ptr)
> #else
> #define __get_user_8(__ret_gu, __val_gu, ptr)                \
>         __get_user_x(8, __ret_gu, __val_gu, ptr)
> #endif
>=20
> #define get_user(x, ptr)                        \
> ({                                    \
>     int __ret_gu;                            \
>     unsigned long __val_gu;                        \
>     __chk_user_ptr(ptr);                        \
>     might_fault();                            \
>     switch (sizeof(*(ptr))) {                    \
>=20
> ...
>=20
>     case 8:                                \
>         __get_user_8(__ret_gu, __val_gu, ptr);            \
>         break;                            \
>=20
> ...
>=20
>     }                                \
>     (x) =3D (__typeof__(*(ptr)))__val_gu;                \
>     __ret_gu;                            \
> })
>=20
> so it should work.  How does it fail?

On x86-32, the above macro resolves to __get_user_X, an undefined symbol.=


>=20
>> Avi, what's your 32-bit buildbot doing? :)
>=20
> I regularly autotest on x86_64, not on i386, sorry.

Good that it's included in my kvm-kmod buildbot.

Jan


--------------enig04FCE2230DA68490037E4E28
Content-Type: application/pgp-signature; name="signature.asc"
Content-Description: OpenPGP digital signature
Content-Disposition: attachment; filename="signature.asc"

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v2.0.15 (GNU/Linux)
Comment: Using GnuPG with SUSE - http://enigmail.mozdev.org/

iEYEARECAAYFAk22eCMACgkQitSsb3rl5xS1cgCfRciicKgDzBvEWlZqTQjeiOHr
P0wAn3UA9GQxqEnVbiCob1ywZdZxEfhR
=6Vmd
-----END PGP SIGNATURE-----

--------------enig04FCE2230DA68490037E4E28--