Brijesh and Ray I don’t disagree the disadvantage of dynamic PCD. However, dynamic PCD is widely used in today's solution.
If developers know clearly when this PCD value is finalized, then it won't be a problem. Otherwise, I prefer we drop the dynamic PCD support in EDKII infrastructure. That would be a huge change!!! Please note this is already *version 12*. We really need move forward. I suggest we keep it as is and check in. If we find problem later, we can resolve the specific issue. Thank you Yao Jiewen > -----Original Message----- > From: Brijesh Singh <brijesh.si...@amd.com> > Sent: Friday, November 12, 2021 1:08 AM > To: Ni, Ray <ray...@intel.com>; devel@edk2.groups.io > Cc: brijesh.si...@amd.com; James Bottomley <j...@linux.ibm.com>; Xu, Min M > <min.m...@intel.com>; Yao, Jiewen <jiewen....@intel.com>; Tom Lendacky > <thomas.lenda...@amd.com>; Justen, Jordan L <jordan.l.jus...@intel.com>; > Ard Biesheuvel <ardb+tianoc...@kernel.org>; Erdem Aktas > <erdemak...@google.com>; Michael Roth <michael.r...@amd.com>; Gerd > Hoffmann <kra...@redhat.com>; Kinney, Michael D > <michael.d.kin...@intel.com>; Liming Gao <gaolim...@byosoft.com.cn>; Liu, > Zhiguang <zhiguang....@intel.com>; Kumar, Rahul1 <rahul1.ku...@intel.com>; > Dong, Eric <eric.d...@intel.com> > Subject: Re: [PATCH v12 20/32] MdePkg: Define > ConfidentialComputingGuestAttr > > Hi Ray, > > On 11/11/21 8:00 AM, Ni, Ray wrote: > > I don't prefer to use a dynamic PCD for passing data. > > > > Because developers don't know when this PCD value is finalized (PCD always > has a default value). > > > > If the value is determined in PEI and consumed in DXE, HOB is a better > > choice. > > If the value is determined in PEI and consumed in PEI, PPI is a better > > choice. > (you can use PPI depex) > > If the value is determined in DXE and consumed in DXE, Protocol is a better > choice. (You can use Protocol depex) > > > > I wish you had raised your concerns early to avoid going in this PCD > direction. The PCD approach was discussed some time back. Both the > SEV-SNP and TDX patches are dependent on it. > > Having said so, if your preference is not to use the PCD, then it can be > done after SNP and TDX patches are merged. > > Jiewen/Min/Gerd thoughts? > > thanks > > > -----Original Message----- > > From: Brijesh Singh <brijesh.si...@amd.com> > > Sent: Thursday, November 11, 2021 6:15 AM > > To: devel@edk2.groups.io > > Cc: James Bottomley <j...@linux.ibm.com>; Xu, Min M > <min.m...@intel.com>; Yao, Jiewen <jiewen....@intel.com>; Tom Lendacky > <thomas.lenda...@amd.com>; Justen, Jordan L <jordan.l.jus...@intel.com>; > Ard Biesheuvel <ardb+tianoc...@kernel.org>; Erdem Aktas > <erdemak...@google.com>; Michael Roth <michael.r...@amd.com>; Gerd > Hoffmann <kra...@redhat.com>; Kinney, Michael D > <michael.d.kin...@intel.com>; Liming Gao <gaolim...@byosoft.com.cn>; Liu, > Zhiguang <zhiguang....@intel.com>; Ni, Ray <ray...@intel.com>; Kumar, Rahul1 > <rahul1.ku...@intel.com>; Dong, Eric <eric.d...@intel.com>; Brijesh Singh > <brijesh.si...@amd.com>; Michael Roth <michael.r...@amd.com> > > Subject: [PATCH v12 20/32] MdePkg: Define ConfidentialComputingGuestAttr > > > > BZ: > https://nam11.safelinks.protection.outlook.com/?url=https%3A%2F%2Fbugzilla. > tianocore.org%2Fshow_bug.cgi%3Fid%3D3275&data=04%7C01%7Cbrijesh. > singh%40amd.com%7C40483d937fa84cebe69908d9a51b9afd%7C3dd8961fe488 > 4e608e11a82d994e183d%7C0%7C0%7C637722360254370524%7CUnknown%7 > CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJ > XVCI6Mn0%3D%7C1000&sdata=EtT7bzDCZxsMA9sTCqISftp62QbezdoSf4k2 > eCfZsws%3D&reserved=0 > > > > While initializing APs, the MpInitLib may need to know whether the guest is > running with active AMD SEV or Intel TDX memory encryption. > > > > Add a new ConfidentialComputingGuestAttr PCD that can be used to query the > memory encryption attribute. > > > > Cc: Michael D Kinney <michael.d.kin...@intel.com> > > Cc: Liming Gao <gaolim...@byosoft.com.cn> > > Cc: Zhiguang Liu <zhiguang....@intel.com> > > Cc: Michael Roth <michael.r...@amd.com> > > Cc: Ray Ni <ray...@intel.com> > > Cc: Rahul Kumar <rahul1.ku...@intel.com> > > Cc: Eric Dong <eric.d...@intel.com> > > Cc: James Bottomley <j...@linux.ibm.com> > > Cc: Min Xu <min.m...@intel.com> > > Cc: Jiewen Yao <jiewen....@intel.com> > > Cc: Tom Lendacky <thomas.lenda...@amd.com> > > Cc: Jordan Justen <jordan.l.jus...@intel.com> > > Cc: Ard Biesheuvel <ardb+tianoc...@kernel.org> > > Cc: Erdem Aktas <erdemak...@google.com> > > Cc: Gerd Hoffmann <kra...@redhat.com> > > Suggested-by: Jiewen Yao <jiewen....@intel.com> > > Acked-by: Gerd Hoffmann <kra...@redhat.com> > > Signed-off-by: Brijesh Singh <brijesh.si...@amd.com> > > --- > > MdePkg/MdePkg.dec | 4 +++ > > .../Include/ConfidentialComputingGuestAttr.h | 25 +++++++++++++++++++ > > 2 files changed, 29 insertions(+) > > create mode 100644 MdePkg/Include/ConfidentialComputingGuestAttr.h > > > > diff --git a/MdePkg/MdePkg.dec b/MdePkg/MdePkg.dec index > 8b18415b107a..cd903c35d2ff 100644 > > --- a/MdePkg/MdePkg.dec > > +++ b/MdePkg/MdePkg.dec > > @@ -2396,5 +2396,9 @@ [PcdsFixedAtBuild, PcdsPatchableInModule, > PcdsDynamic, PcdsDynamicEx] > > # @Prompt FSB Clock. > > gEfiMdePkgTokenSpaceGuid.PcdFSBClock|200000000|UINT32|0x0000000c > > > > + ## This dynamic PCD indicates the memory encryption attribute of the > > guest. > > + # @Prompt Memory encryption attribute > > + > gEfiMdePkgTokenSpaceGuid.PcdConfidentialComputingGuestAttr|0|UINT64|0x > > + 0000002e > > + > > [UserExtensions.TianoCore."ExtraFiles"] > > MdePkgExtra.uni > > diff --git a/MdePkg/Include/ConfidentialComputingGuestAttr.h > b/MdePkg/Include/ConfidentialComputingGuestAttr.h > > new file mode 100644 > > index 000000000000..495b0df0ac33 > > --- /dev/null > > +++ b/MdePkg/Include/ConfidentialComputingGuestAttr.h > > @@ -0,0 +1,25 @@ > > +/** @file > > +Definitions for Confidential Computing Attribute > > + > > +Copyright (c) 2021 AMD Inc. All rights reserved.<BR> > > +SPDX-License-Identifier: BSD-2-Clause-Patent > > + > > +**/ > > + > > +#ifndef CONFIDENTIAL_COMPUTING_GUEST_ATTR_H_ > > +#define CONFIDENTIAL_COMPUTING_GUEST_ATTR_H_ > > + > > +typedef enum { > > + /* The guest is running with memory encryption disabled. */ > > + CCAttrNotEncrypted = 0, > > + > > + /* The guest is running with AMD SEV memory encryption enabled. */ > > + CCAttrAmdSev = 0x100, > > + CCAttrAmdSevEs = 0x101, > > + CCAttrAmdSevSnp = 0x102, > > + > > + /* The guest is running with Intel TDX memory encryption enabled. */ > > + CCAttrIntelTdx = 0x200, > > +} CONFIDENTIAL_COMPUTING_GUEST_ATTR; > > + > > +#endif > > -- > > 2.25.1 > > -=-=-=-=-=-=-=-=-=-=-=- Groups.io Links: You receive all messages sent to this group. View/Reply Online (#83674): https://edk2.groups.io/g/devel/message/83674 Mute This Topic: https://groups.io/mt/86969144/21656 Group Owner: devel+ow...@edk2.groups.io Unsubscribe: https://edk2.groups.io/g/devel/unsub [arch...@mail-archive.com] -=-=-=-=-=-=-=-=-=-=-=-
