[edk2-devel] [PATCH v2 2/6] OvmfPkg/AmdSev: add Grub Firmware Volume Package

[James Bottomley]

On Wed, 2020-11-25 at 09:09 -0800, James Bottomley wrote:
> On Wed, 2020-11-25 at 08:02 -0800, James Bottomley wrote:
> > On Wed, 2020-11-25 at 15:01 +0100, Laszlo Ersek wrote:
> > > This upgrade gave me kernel 5.8.18-100.fc31.x86_64 in the guest
> > > --
> > > and this one does *not* crash. From your boot log below, I see
> > > your guest kernel is 5.5.0; I suggest upgrading it.
> > 
> > Heh, that's easier said than done ... I always make my encrypted
> > images too small to upgrade a kernel easily.  Anyway, after doing
> > the remove and add stuff dance, I finally got it upgraded to the
> > latest debian testing linux-image-5.8.0-3 it's still crashing
> > although with a slightly different traceback.  It looks like there
> > might be something additional in the fedora 5.8 kernel that fixes
> > this.  I'm going to try out upstream kernels next.
> 
> I've got the upstream kernel booting through OVMF with a qemu -kernel
> command line.  I also have a fix: it's not to delete the dummy
> variable which was part of the ancient x86 anti bricking code (which
> is also why arm64 doesn't have the problem).

OK, found the problem: the memory the variable policy itself is stored
in doesn't survive into the linux runtime.  It looks to be pretty
generic except you need a way of detecting the boot time memory is
free.  Because Linux does a remapping, we see this.  I'm not sure how
you'd construct a regression test for something like this, though.

I propose this as the fix ... at least it gets my SEV kernels booting
again.

James

---8>8>8><8<8<8----

>From b33aa8bd8ded4c7cfec67c2d5aa5e23271c75f51 Mon Sep 17 00:00:00 2001
From: James Bottomley <James.Bottomley at HansenPartnership.com>
Date: Wed, 25 Nov 2020 10:10:30 -0800
Subject: [PATCH 1/1] VariablePolicyLib: Fix Runtime panic in
 ValidateSetVariable

The current variable policy is allocated by AllocatePool, which is
boot time only.  This means that if you do any variable setting in the
runtime, the policy has been freed.  Ordinarily this isn't detected
because freed memory is still there, but when you boot the Linux
kernel, it's been remapped so the actual memory no longer exists in
the memory map causing a page fault.

Fix this by making it AllocateRuntimePool.

Signed-off-by: James Bottomley <jejb at linux.ibm.com>
---
 MdeModulePkg/Library/VariablePolicyLib/VariablePolicyLib.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/MdeModulePkg/Library/VariablePolicyLib/VariablePolicyLib.c b/MdeModulePkg/Library/VariablePolicyLib/VariablePolicyLib.c
index 5029ddb96adb..12944ac7ea81 100644
--- a/MdeModulePkg/Library/VariablePolicyLib/VariablePolicyLib.c
+++ b/MdeModulePkg/Library/VariablePolicyLib/VariablePolicyLib.c
@@ -411,7 +411,7 @@ RegisterVariablePolicy (
     }
 
     // Reallocate and copy the table.
-    NewTable = AllocatePool( NewSize );
+    NewTable = AllocateRuntimePool( NewSize );
     if (NewTable == NULL) {
       return EFI_OUT_OF_RESOURCES;
     }
-- 
2.26.2




-=-=-=-=-=-=-=-=-=-=-=-
Groups.io Links: You receive all messages sent to this group.
View/Reply Online (#67994): https://edk2.groups.io/g/devel/message/67994
Mute This Topic: https://groups.io/mt/78455898/1813853
Group Owner: devel+owner at edk2.groups.io
Unsubscribe: https://edk2.groups.io/g/devel/unsub [edk2-devel-archive at redhat.com]
-=-=-=-=-=-=-=-=-=-=-=-