public inbox for devel@edk2.groups.io
 help / color / mirror / Atom feed
From: "Stefan Berger" <stefanb@linux.ibm.com>
To: "Yao, Jiewen" <jiewen.yao@intel.com>,
	"devel@edk2.groups.io" <devel@edk2.groups.io>,
	"stefanb@linux.vnet.ibm.com" <stefanb@linux.vnet.ibm.com>
Cc: "mhaeuser@posteo.de" <mhaeuser@posteo.de>,
	"spbrogan@outlook.com" <spbrogan@outlook.com>,
	"marcandre.lureau@redhat.com" <marcandre.lureau@redhat.com>,
	"kraxel@redhat.com" <kraxel@redhat.com>
Subject: Re: [edk2-devel] [PATCH v7 0/9] Ovmf: Disable the TPM2 platform hierarchy
Date: Fri, 10 Sep 2021 16:47:46 -0400	[thread overview]
Message-ID: <cb12bd72-185b-5d2d-5995-701918e3c319@linux.ibm.com> (raw)
In-Reply-To: <16A38214549AD34A.16479@groups.io>

On 9/10/21 12:15 PM, Stefan Berger wrote:

>
> On 9/10/21 11:32 AM, Yao, Jiewen wrote:
>> According to the security policy, PP request must be processed before 
>> EndOfDxe.
>>
>> May I know when you trigger PP request?
>
> OVMF has 3 implementations invoking it in 
> PlatformBootManagerAfterConsole():
>
> https://github.com/tianocore/edk2/blob/master/OvmfPkg/Library/PlatformBootManagerLib/BdsPlatform.c#L1517 
>
>
> https://github.com/tianocore/edk2/blob/master/OvmfPkg/Library/PlatformBootManagerLibBhyve/BdsPlatform.c#L1451 
>
>
> https://github.com/tianocore/edk2/blob/master/OvmfPkg/Library/PlatformBootManagerLibGrub/BdsPlatform.c#L1316 
>

Before I post yet another series...:

The problem is that PPI may require interaction with the console, so it 
seems we have to handle it in PlatformBootManagerAfterConsole(). The 
disablement of the TPM 2 platform hierarchy may only occur after that, 
so we have to move this part here after TPM-PPI-Handling from 
BeforeConsole() into AfterConsole() because this is what triggers that 
new code from edk2-platforms to disable that TPM 2 platform hierarchy:

   Status = gBS->InstallProtocolInterface (&Handle,
                   &gEfiDxeSmmReadyToLockProtocolGuid, EFI_NATIVE_INTERFACE,
                   NULL);

And then we have this part here also in BeforeConsole() that has to be 
moved as well into AfterConsole().

   //
   // Dispatch deferred images after EndOfDxe event and ReadyToLock
   // installation.
   //
   EfiBootManagerDispatchDeferredImages ();


This then leads to something like this with the sequence 
(TPM-PPI-handling, gEfiDxeSmmReadyToLockProtocol, 
EfiBootManagerDispatchDeferredImages) needing to stay in that order. 
However, I am not sure know whether one can just move these parts around 
like this.


diff --git a/OvmfPkg/Library/PlatformBootManagerLib/BdsPlatform.c 
b/OvmfPkg/Library/PlatformBootManagerLib/BdsPlatform.c
index 71f63b2448..266d58dfbe 100644
--- a/OvmfPkg/Library/PlatformBootManagerLib/BdsPlatform.c
+++ b/OvmfPkg/Library/PlatformBootManagerLib/BdsPlatform.c
@@ -354,7 +354,6 @@ PlatformBootManagerBeforeConsole (
    VOID
    )
  {
-  EFI_HANDLE    Handle;
    EFI_STATUS    Status;
    UINT16        FrontPageTimeout;
    RETURN_STATUS PcdStatus;
@@ -387,8 +386,10 @@ PlatformBootManagerBeforeConsole (
      SaveS3BootScript ();
    }

+#if 0
    //
    // Prevent further changes to LockBoxes or SMRAM.
+  // Any TPM 2 Physical Presence Interface opcode must be handled BEFORE
    //
    Handle = NULL;
    Status = gBS->InstallProtocolInterface (&Handle,
@@ -401,6 +402,7 @@ PlatformBootManagerBeforeConsole (
    // installation.
    //
    EfiBootManagerDispatchDeferredImages ();
+#endif

    PlatformInitializeConsole (
      XenDetected() ? gXenPlatformConsole : gPlatformConsole);
@@ -437,6 +439,7 @@ PlatformBootManagerBeforeConsole (
    //
    VisitAllInstancesOfProtocol (&gEfiPciIoProtocolGuid, 
ConnectVirtioPciRng,
      NULL);
+
  }


@@ -1474,6 +1477,8 @@ PlatformBootManagerAfterConsole (
    VOID
    )
  {
+  EFI_STATUS                         Status;
+  EFI_HANDLE                         Handle;
    EFI_BOOT_MODE                      BootMode;

    DEBUG ((DEBUG_INFO, "PlatformBootManagerAfterConsole\n"));
@@ -1511,11 +1516,29 @@ PlatformBootManagerAfterConsole (
    //
    PciAcpiInitialization ();

+#if 1
    //
-  // Process TPM PPI request
+  // Process TPM PPI request; this may require interaction via console
    //
    Tcg2PhysicalPresenceLibProcessRequest (NULL);

+  //
+  // Prevent further changes to LockBoxes or SMRAM.
+  // Any TPM 2 Physical Presence Interface opcode must be handled BEFORE
+  //
+  Handle = NULL;
+  Status = gBS->InstallProtocolInterface (&Handle,
+                  &gEfiDxeSmmReadyToLockProtocolGuid, EFI_NATIVE_INTERFACE,
+                  NULL);
+  ASSERT_EFI_ERROR (Status);
+
+  //
+  // Dispatch deferred images after EndOfDxe event and ReadyToLock
+  // installation.
+  //
+  EfiBootManagerDispatchDeferredImages ();
+#endif
+
    //
    // Process QEMU's -kernel command line option
    //


    Stefan


  parent reply	other threads:[~2021-09-10 20:47 UTC|newest]

Thread overview: 22+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-09-09 17:35 [PATCH v7 0/9] Ovmf: Disable the TPM2 platform hierarchy Stefan Berger
2021-09-09 17:35 ` [PATCH v7 1/9] SecurityPkg/TPM: Import PeiDxeTpmPlatformHierarchyLib.c from edk2-platforms Stefan Berger
2021-09-09 17:35 ` [PATCH v7 2/9] SecurityPkg/TPM: Fix bugs in imported PeiDxeTpmPlatformHierarchyLib Stefan Berger
2021-09-09 17:35 ` [PATCH v7 3/9] SecrutiyPkg/Tcg: Import Tcg2PlatformDxe from edk2-platforms Stefan Berger
2021-09-09 17:35 ` [PATCH v7 4/9] SecurityPkg/Tcg: Make Tcg2PlatformDxe buildable Stefan Berger
2021-09-09 17:35 ` [PATCH v7 5/9] SecurityPkg: Introduce new PCD PcdRandomizePlatformHierarchy Stefan Berger
2021-09-09 17:35 ` [PATCH v7 6/9] OvmfPkg: Reference new Tcg2PlatformDxe in the build system for compilation Stefan Berger
2021-09-09 17:35 ` [PATCH v7 7/9] SecurityPkg/Tcg: Import Tcg2PlatformPei from edk2-platforms Stefan Berger
2021-09-09 17:35 ` [PATCH v7 8/9] SecurityPkg/Tcg: Make Tcg2PlatformPei buildable Stefan Berger
2021-09-09 17:35 ` [PATCH v7 9/9] OvmfPkg: Reference new Tcg2PlatformPei in the build system Stefan Berger
2021-09-10 14:24 ` [edk2-devel] [PATCH v7 0/9] Ovmf: Disable the TPM2 platform hierarchy Stefan Berger
2021-09-10 15:32   ` Yao, Jiewen
2021-09-10 16:15     ` Stefan Berger
2021-09-11  2:38       ` Yao, Jiewen
2021-09-11  2:46         ` Yao, Jiewen
2021-09-12  0:42           ` Stefan Berger
2021-09-12  0:45             ` Yao, Jiewen
2021-09-12  1:52               ` Stefan Berger
2021-09-13 14:51           ` Stefan Berger
     [not found]     ` <16A38214549AD34A.16479@groups.io>
2021-09-10 20:47       ` Stefan Berger [this message]
2021-09-13  7:08 ` Yao, Jiewen
     [not found] ` <16A44FFF7B7DEB00.6211@groups.io>
2021-09-13  9:31   ` [edk2-devel] " Yao, Jiewen

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-list from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=cb12bd72-185b-5d2d-5995-701918e3c319@linux.ibm.com \
    --to=devel@edk2.groups.io \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox