On 11/20/19 21:06, Lendacky, Thomas wrote:
> The flash detection routine will attempt to determine how the flash
> device behaves (e.g. ROM, RAM, Flash). But when SEV-ES is enabled and
> the flash device behaves as a ROM device (meaning it is marked read-only
> by the hypervisor), this check may result in an infinite nested page fault
> because of the attempted write. Since the instruction cannot be emulated
> when SEV-ES is enabled, the RIP is never advanced, resulting in repeated
> nested page faults.
> 
> When SEV-ES is enabled, exit the flash detection early and assume that
> the FD behaves as Flash. This will result in QemuFlashWrite() being called
> to store EFI variables, which will also result in an infinite nested page
> fault when the write is performed. In this case, update QemuFlashWrite()
> to use the VmgMmioWrite function from the VmgExitLib library to have the
> hypervisor perform the write without having to emulate the instruction.

(1) Please include the Bugzilla link in the commit message:

https://bugzilla.tianocore.org/show_bug.cgi?id=2198

> 
> Signed-off-by: Tom Lendacky <thomas.lenda...@amd.com>
> ---
>  OvmfPkg/OvmfPkgIa32.dsc                       |  1 +
>  OvmfPkg/OvmfPkgIa32X64.dsc                    |  1 +
>  OvmfPkg/OvmfPkgX64.dsc                        |  1 +
>  .../FvbServicesRuntimeDxe.inf                 |  2 +
>  .../QemuFlash.c                               | 38 +++++++++++++++++--
>  5 files changed, 40 insertions(+), 3 deletions(-)
> 
> diff --git a/OvmfPkg/OvmfPkgIa32.dsc b/OvmfPkg/OvmfPkgIa32.dsc
> index 56670eefde6b..ff2814c6246e 100644
> --- a/OvmfPkg/OvmfPkgIa32.dsc
> +++ b/OvmfPkg/OvmfPkgIa32.dsc
> @@ -320,6 +320,7 @@ [LibraryClasses.common.DXE_RUNTIME_DRIVER]
>    BaseCryptLib|CryptoPkg/Library/BaseCryptLib/RuntimeCryptLib.inf
>    PciLib|OvmfPkg/Library/DxePciLibI440FxQ35/DxePciLibI440FxQ35.inf
>    QemuFwCfgS3Lib|OvmfPkg/Library/QemuFwCfgS3Lib/DxeQemuFwCfgS3LibFwCfg.inf
> +  VmgExitLib|UefiCpuPkg/Library/VmgExitLib/VmgExitLib.inf
>
>  [LibraryClasses.common.UEFI_DRIVER]
>    PcdLib|MdePkg/Library/DxePcdLib/DxePcdLib.inf
> diff --git a/OvmfPkg/OvmfPkgIa32X64.dsc b/OvmfPkg/OvmfPkgIa32X64.dsc
> index 9897e6889573..212952cfaacd 100644
> --- a/OvmfPkg/OvmfPkgIa32X64.dsc
> +++ b/OvmfPkg/OvmfPkgIa32X64.dsc
> @@ -325,6 +325,7 @@ [LibraryClasses.common.DXE_RUNTIME_DRIVER]
>    BaseCryptLib|CryptoPkg/Library/BaseCryptLib/RuntimeCryptLib.inf
>    PciLib|OvmfPkg/Library/DxePciLibI440FxQ35/DxePciLibI440FxQ35.inf
>    QemuFwCfgS3Lib|OvmfPkg/Library/QemuFwCfgS3Lib/DxeQemuFwCfgS3LibFwCfg.inf
> +  VmgExitLib|UefiCpuPkg/Library/VmgExitLib/VmgExitLib.inf
>
>  [LibraryClasses.common.UEFI_DRIVER]
>    PcdLib|MdePkg/Library/DxePcdLib/DxePcdLib.inf
> diff --git a/OvmfPkg/OvmfPkgX64.dsc b/OvmfPkg/OvmfPkgX64.dsc
> index 59c4f9207fc3..8331fc0b663e 100644
> --- a/OvmfPkg/OvmfPkgX64.dsc
> +++ b/OvmfPkg/OvmfPkgX64.dsc
> @@ -325,6 +325,7 @@ [LibraryClasses.common.DXE_RUNTIME_DRIVER]
>    BaseCryptLib|CryptoPkg/Library/BaseCryptLib/RuntimeCryptLib.inf
>    PciLib|OvmfPkg/Library/DxePciLibI440FxQ35/DxePciLibI440FxQ35.inf
>    QemuFwCfgS3Lib|OvmfPkg/Library/QemuFwCfgS3Lib/DxeQemuFwCfgS3LibFwCfg.inf
> +  VmgExitLib|UefiCpuPkg/Library/VmgExitLib/VmgExitLib.inf
>
>  [LibraryClasses.common.UEFI_DRIVER]
>    PcdLib|MdePkg/Library/DxePcdLib/DxePcdLib.inf
> diff --git a/OvmfPkg/QemuFlashFvbServicesRuntimeDxe/FvbServicesRuntimeDxe.inf 
> b/OvmfPkg/QemuFlashFvbServicesRuntimeDxe/FvbServicesRuntimeDxe.inf
> index ca6326e833ed..0b7741ac07f8 100644
> --- a/OvmfPkg/QemuFlashFvbServicesRuntimeDxe/FvbServicesRuntimeDxe.inf
> +++ b/OvmfPkg/QemuFlashFvbServicesRuntimeDxe/FvbServicesRuntimeDxe.inf
> @@ -38,6 +38,7 @@ [Sources]
>  [Packages]
>    MdePkg/MdePkg.dec
>    MdeModulePkg/MdeModulePkg.dec
> +  UefiCpuPkg/UefiCpuPkg.dec
>    OvmfPkg/OvmfPkg.dec
>
>  [LibraryClasses]
> @@ -52,6 +53,7 @@ [LibraryClasses]
>    UefiBootServicesTableLib
>    UefiDriverEntryPoint
>    UefiRuntimeLib
> +  VmgExitLib
>
>  [Guids]
>    gEfiEventVirtualAddressChangeGuid   # ALWAYS_CONSUMED

These hunks look good.

> diff --git a/OvmfPkg/QemuFlashFvbServicesRuntimeDxe/QemuFlash.c 
> b/OvmfPkg/QemuFlashFvbServicesRuntimeDxe/QemuFlash.c
> index c81c58972bf2..4f3bf690fcad 100644
> --- a/OvmfPkg/QemuFlashFvbServicesRuntimeDxe/QemuFlash.c
> +++ b/OvmfPkg/QemuFlashFvbServicesRuntimeDxe/QemuFlash.c
> @@ -9,7 +9,9 @@
>
>  #include <Library/BaseMemoryLib.h>
>  #include <Library/DebugLib.h>
> +#include <Library/MemEncryptSevLib.h>
>  #include <Library/PcdLib.h>
> +#include <Library/VmgExitLib.h>
>
>  #include "QemuFlash.h"
>
> @@ -80,6 +82,20 @@ QemuFlashDetected (
>
>    DEBUG ((EFI_D_INFO, "QEMU Flash: Attempting flash detection at %p\n", 
> Ptr));
>
> +  if (MemEncryptSevEsIsEnabled()) {
> +    //
> +    // When SEV-ES is enabled, the check below can result in an infinite
> +    // loop with respect to a nested page fault. When the FD behaves as
> +    // a ROM, the nested page table entry is read-only. The check below

(2) I suggest a slight change to the wording above. Please replace

  when the FD behaves as a ROM

with

  when the memslot is mapped read-only

(It's OK to make qemu / kvm references in this code: the name of the
file includes "Qemu".)

> +    // will cause a nested page fault that cannot be emulated, causing
> +    // the instruction to retried over and over. For SEV-ES, assume that
> +    // the FD does not behave as FLASH.

(3) I would recommend

  For SEV-ES, acknowledge that the FD appears as ROM and not as FLASH,
  but report FLASH anyway.

> +    //
> +    DEBUG ((DEBUG_INFO,
> +      "QEMU Flash: SEV-ES enabled, assuming FD behaves as FLASH\n"));
> +    return TRUE;
> +  }
> +
>    OriginalUint8 = *Ptr;
>    *Ptr = CLEAR_STATUS_CMD;
>    ProbeUint8 = *Ptr;
> @@ -147,6 +163,21 @@ QemuFlashRead (
>  }
>
>
> +STATIC
> +VOID
> +QemuFlashPtrWrite (
> +  IN        volatile UINT8                      *Ptr,
> +  IN        UINT8                               Value
> +  )
> +{
> +  if (MemEncryptSevEsIsEnabled()) {
> +    VmgMmioWrite ((UINT8 *) Ptr, &Value, 1);
> +  } else {
> +    *Ptr = Value;
> +  }
> +}
> +
> +
>  /**
>    Write to QEMU Flash
>

(4) This change will break the "FvbServicesSmm.inf" build of this module.

(Because, you have -- correctly -- added the VmgExitLib class to
"FvbServicesRuntimeDxe.inf" only.)

Please:

- create two implementations of the QemuFlashPtrWrite() function, one in
"QemuFlashDxe.c", and another in "QemuFlashSmm.c".

- the former should be the function shown above, the latter should only
perform the direct assignment.

- the '#include <Library/VmgExitLib.h>' directive should also be moved
to "QemuFlashDxe.c".

- the prototype for QemuFlashPtrWrite() should be added to "QemuFlash.h"
please.

> @@ -181,8 +212,9 @@ QemuFlashWrite (
>    //
>    Ptr = QemuFlashPtr (Lba, Offset);
>    for (Loop = 0; Loop < *NumBytes; Loop++) {
> -    *Ptr = WRITE_BYTE_CMD;
> -    *Ptr = Buffer[Loop];
> +    QemuFlashPtrWrite (Ptr, WRITE_BYTE_CMD);
> +    QemuFlashPtrWrite (Ptr, Buffer[Loop]);
> +
>      Ptr++;
>    }
>
> @@ -190,7 +222,7 @@ QemuFlashWrite (
>    // Restore flash to read mode
>    //
>    if (*NumBytes > 0) {
> -    *(Ptr - 1) = READ_ARRAY_CMD;
> +    QemuFlashPtrWrite (Ptr - 1, READ_ARRAY_CMD);
>    }
>
>    return EFI_SUCCESS;
> 

With (1) and (4) fixed, and with (2) and (3) optionally fixed (i.e., if
you agree):

Reviewed-by: Laszlo Ersek <ler...@redhat.com>

Thanks
Laszlo


-=-=-=-=-=-=-=-=-=-=-=-
Groups.io Links: You receive all messages sent to this group.

View/Reply Online (#51054): https://edk2.groups.io/g/devel/message/51054
Mute This Topic: https://groups.io/mt/60973139/21656
Group Owner: devel+ow...@edk2.groups.io
Unsubscribe: https://edk2.groups.io/g/devel/unsub  [arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-

Reply via email to