From mboxrd@z Thu Jan 1 00:00:00 1970 From: "zpfalpc23" Subject: =?UTF-8?Q?=E7=AD=94=E5=A4=8D:_=5BXen-devel=5D_Re:_changeset_22?= =?UTF-8?Q?526:7a5ee3800417?= Date: Wed, 16 Mar 2011 18:27:56 +0800 Message-ID: <013601cbe3c4$d36f1a00$7a4d4e00$@com> References: <1299514840.19262.4698.camel@elijah> Mime-Version: 1.0 Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: quoted-printable Return-path: In-Reply-To: Content-Language: zh-cn List-Unsubscribe: , List-Post: List-Help: List-Subscribe: , Sender: xen-devel-bounces@lists.xensource.com Errors-To: xen-devel-bounces@lists.xensource.com To: 'xen devel' Cc: 'George Dunlap' , 'George Dunlap' List-Id: xen-devel@lists.xenproject.org Hi George, I have read you patch 22526 in xen-testing4.0 and xen-unstable = carefully. There is one thing confused me a lot. why the patch for = xen-testing4.0 did not handle the case In ept_set_entry where the = superpage should be splitted. Maybe the race can't happen in this scene. Also When the race of updating p2m table happen, will the normal(be = changed by ept_get_entry) page be POD again? In this situation, what is = it for Xen and VM? Thanks a lot for enlighten me. ----Pengfei zhang -----=E9=82=AE=E4=BB=B6=E5=8E=9F=E4=BB=B6----- =E5=8F=91=E4=BB=B6=E4=BA=BA: dunlapg@gmail.com = [mailto:dunlapg@gmail.com] =E4=BB=A3=E8=A1=A8 George Dunlap =E5=8F=91=E9=80=81=E6=97=B6=E9=97=B4: 2011=E5=B9=B43=E6=9C=8810=E6=97=A5 = 17:43 =E6=94=B6=E4=BB=B6=E4=BA=BA: George Dunlap =E6=8A=84=E9=80=81: =E5=BC=A0=E9=B9=8F=E9=A3=9E; xen devel =E4=B8=BB=E9=A2=98: Re: [Xen-devel] Re: changeset 22526:7a5ee3800417 Oops, forgot to attach the patch... -George On Thu, Mar 10, 2011 at 9:41 AM, George Dunlap = wrote: > Peng fei, > > Can you (1) test this patch, and (2) make sure I've romanized your=20 > name properly? > > Thanks, > -George > > On Wed, Mar 9, 2011 at 1:59 PM, George Dunlap=20 > wrote: >> Peng fei, >> >> I just noticed this patch has a bug -- it doesn't set new_entry.mfn=20 >> if the new mfn is the same as the one already in ept_entry! Standby=20 >> for a new one... >> >> -George >> >> On Mon, Mar 7, 2011 at 4:41 PM, George Dunlap=20 >> wrote: >>> Peng fei, >>> >>> Can you test the attached patch (to xen-unstable)? I don't have=20 >>> EPT-enabled hardware handy... >>> >>> -George >>> >>> On Mon, Mar 7, 2011 at 4:20 PM, George Dunlap = wrote: >>>> Hmm, yeah, that's obviously not right! Let me take a look... >>>> >>>> -George >>>> >>>> On Mon, 2011-03-07 at 09:55 +0000, =E5=BC=A0=E9=B9=8F=E9=A3=9E = wrote: >>>>> Hi, >>>>> Recently, I did a research on the p2m and EPT,and did apply=20 >>>>> your >>>>> patch(22526) to the source code of mine. But there is one place=20 >>>>> confused me: >>>>> >>>>> >>>>> >>>>> ept_entry =3D table + index; >>>>> 1.91 >>>>> 1.92 - ept_entry->emt =3D epte_get_entry_emt(d, gfn, = mfn,=20 >>>>> &ipat, direct_mmio); >>>>> 1.93 - ept_entry->ipat =3D ipat; >>>>> 1.94 - ept_entry->sp =3D i ? 1 : 0; >>>>> 1.95 - ept_entry->avail1 =3D p2mt; >>>>> 1.96 - ept_entry->avail2 =3D 0; >>>>> 1.97 + new_entry.emt =3D epte_get_entry_emt(d, gfn, = mfn,=20 >>>>> &ipat, direct_mmio); >>>>> 1.98 + new_entry.ipat =3D ipat; >>>>> 1.99 + new_entry.sp =3D i ? 1 : 0; >>>>> 1.100 + new_entry.avail1 =3D p2mt; >>>>> 1.101 + new_entry.avail2 =3D 0; >>>>> 1.102 >>>>> 1.103 - if ( ept_entry->mfn =3D=3D mfn_x(mfn) ) >>>>> 1.104 + if ( new_entry.mfn =3D=3D mfn_x(mfn) ) >>>>> 1.105 need_modify_vtd_table =3D 0; >>>>> 1.106 else /* the caller should take care of the=20 >>>>> previous page */ >>>>> 1.107 - ept_entry->mfn =3D mfn_x(mfn); >>>>> 1.108 + new_entry.mfn =3D mfn_x(mfn); >>>>> >>>>> I think, The new_entry.mfn is used without assigned any value. = Maybe there was something else I did not think of. >>>>> Thank you for your kindness! >>>>> >>>>> http://xenbits.xen.org/xen-unstable.hg/rev/7a5ee3800417 >>>> >>>> >>>> >>>> _______________________________________________ >>>> Xen-devel mailing list >>>> Xen-devel@lists.xensource.com >>>> http://lists.xensource.com/xen-devel >>>> >>> >> >