[PATCH 2/5] MdeModulePkg/EbcDxe: add ARM support

[Pete Batard <pete@akeo.ie>]

From: Ard Biesheuvel <ard.biesheuvel@linaro.org>

* This is a port of the AARCH64 implementation of the EBC runtime to ARM.
* Note that, on its own, this patch only allows running self contained EBC
  applications, such as ones that don't issue calls into the native
  platform, or that aren't being called from native.
* Support for EBC -> native and native -> EBC calls will be added in
  subsequent patches.
* Because EFI32 may not be defined for ARM compilation, this patch also
  alters the EBC Debugger to use MDE_CPU_### macros for address display.

Contributed-under: TianoCore Contribution Agreement 1.0
Signed-off-by: Ard Biesheuvel <ard.biesheuvel@linaro.org>
Signed-off-by: Pete Batard <pete@akeo.ie>
---
 ArmVirtPkg/ArmVirt.dsc.inc                         |   6 +-
 ArmVirtPkg/ArmVirtQemuFvMain.fdf.inc               |  10 +-
 ArmVirtPkg/ArmVirtXen.fdf                          |  10 +-
 MdeModulePkg/MdeModulePkg.dsc                      |   4 +-
 MdeModulePkg/Universal/EbcDxe/Arm/EbcLowLevel.S    | 147 +++++++
 MdeModulePkg/Universal/EbcDxe/Arm/EbcSupport.c     | 470 +++++++++++++++++++++
 MdeModulePkg/Universal/EbcDxe/EbcDebugger.inf      |   6 +-
 .../EbcDxe/EbcDebugger/EdbDisasmSupport.h          |   4 +-
 .../Universal/EbcDxe/EbcDebuggerConfig.inf         |   2 +-
 MdeModulePkg/Universal/EbcDxe/EbcDxe.inf           |   6 +-
 10 files changed, 648 insertions(+), 17 deletions(-)
 create mode 100644 MdeModulePkg/Universal/EbcDxe/Arm/EbcLowLevel.S
 create mode 100644 MdeModulePkg/Universal/EbcDxe/Arm/EbcSupport.c

diff --git a/ArmVirtPkg/ArmVirt.dsc.inc b/ArmVirtPkg/ArmVirt.dsc.inc
index dbd6678accde..3fa016b501c5 100644
--- a/ArmVirtPkg/ArmVirt.dsc.inc
+++ b/ArmVirtPkg/ArmVirt.dsc.inc
@@ -402,6 +402,11 @@ [Components.common]
   MdeModulePkg/Universal/Disk/RamDiskDxe/RamDiskDxe.inf
 
   #
+  # EBC support
+  #
+  MdeModulePkg/Universal/EbcDxe/EbcDxe.inf
+
+  #
   # UEFI application (Shell Embedded Boot Loader)
   #
   ShellPkg/Application/Shell/Shell.inf {
@@ -430,4 +435,3 @@ [Components.AARCH64]
   # ACPI Support
   #
   MdeModulePkg/Universal/Acpi/AcpiTableDxe/AcpiTableDxe.inf
-  MdeModulePkg/Universal/EbcDxe/EbcDxe.inf
diff --git a/ArmVirtPkg/ArmVirtQemuFvMain.fdf.inc b/ArmVirtPkg/ArmVirtQemuFvMain.fdf.inc
index cc5d12aaefea..5145a86a0f33 100644
--- a/ArmVirtPkg/ArmVirtQemuFvMain.fdf.inc
+++ b/ArmVirtPkg/ArmVirtQemuFvMain.fdf.inc
@@ -140,6 +140,11 @@ [FV.FvMain]
   INF MdeModulePkg/Universal/SmbiosDxe/SmbiosDxe.inf
   INF OvmfPkg/SmbiosPlatformDxe/SmbiosPlatformDxe.inf
 
+  #
+  # EBC support
+  #
+  INF MdeModulePkg/Universal/EbcDxe/EbcDxe.inf
+
 !if $(ARCH) == AARCH64
   #
   # ACPI Support
@@ -147,11 +152,6 @@ [FV.FvMain]
   INF MdeModulePkg/Universal/Acpi/AcpiTableDxe/AcpiTableDxe.inf
   INF MdeModulePkg/Universal/Acpi/BootGraphicsResourceTableDxe/BootGraphicsResourceTableDxe.inf
   INF OvmfPkg/AcpiPlatformDxe/QemuFwCfgAcpiPlatformDxe.inf
-
-  #
-  # EBC support
-  #
-  INF MdeModulePkg/Universal/EbcDxe/EbcDxe.inf
 !endif
 
   #
diff --git a/ArmVirtPkg/ArmVirtXen.fdf b/ArmVirtPkg/ArmVirtXen.fdf
index c997251b12b8..ccbd8baa5f82 100644
--- a/ArmVirtPkg/ArmVirtXen.fdf
+++ b/ArmVirtPkg/ArmVirtXen.fdf
@@ -170,6 +170,11 @@ [FV.FvMain]
   INF ShellPkg/Application/Shell/Shell.inf
 
   #
+  # EBC support
+  #
+  INF MdeModulePkg/Universal/EbcDxe/EbcDxe.inf
+
+  #
   # Bds
   #
   INF MdeModulePkg/Universal/DevicePathDxe/DevicePathDxe.inf
@@ -186,11 +191,6 @@ [FV.FvMain]
 !if $(ARCH) == AARCH64
   INF MdeModulePkg/Universal/Acpi/AcpiTableDxe/AcpiTableDxe.inf
   INF ArmVirtPkg/XenAcpiPlatformDxe/XenAcpiPlatformDxe.inf
-
-  #
-  # EBC support
-  #
-  INF MdeModulePkg/Universal/EbcDxe/EbcDxe.inf
 !endif
 
  #
diff --git a/MdeModulePkg/MdeModulePkg.dsc b/MdeModulePkg/MdeModulePkg.dsc
index 5996fe50f680..35094bab9e9a 100644
--- a/MdeModulePkg/MdeModulePkg.dsc
+++ b/MdeModulePkg/MdeModulePkg.dsc
@@ -430,8 +430,10 @@ [Components]
   MdeModulePkg/Library/DxeCapsuleLibFmp/DxeCapsuleLib.inf
   MdeModulePkg/Library/DxeCapsuleLibFmp/DxeRuntimeCapsuleLib.inf
 
-[Components.IA32, Components.X64, Components.IPF, Components.AARCH64]
+[Components.IA32, Components.X64, Components.IPF]
   MdeModulePkg/Universal/Network/UefiPxeBcDxe/UefiPxeBcDxe.inf
+
+[Components.IA32, Components.X64, Components.IPF, Components.ARM, Components.AARCH64]
   MdeModulePkg/Universal/DebugSupportDxe/DebugSupportDxe.inf
   MdeModulePkg/Universal/EbcDxe/EbcDxe.inf
   MdeModulePkg/Universal/EbcDxe/EbcDebugger.inf
diff --git a/MdeModulePkg/Universal/EbcDxe/Arm/EbcLowLevel.S b/MdeModulePkg/Universal/EbcDxe/Arm/EbcLowLevel.S
new file mode 100644
index 000000000000..cb3c11f71673
--- /dev/null
+++ b/MdeModulePkg/Universal/EbcDxe/Arm/EbcLowLevel.S
@@ -0,0 +1,147 @@
+///** @file
+//
+//  This code provides low level routines that support the Virtual Machine
+//  for option ROMs.
+//
+//  Copyright (c) 2016, Linaro, Ltd. All rights reserved.<BR>
+//  Copyright (c) 2015, The Linux Foundation. All rights reserved.<BR>
+//  Copyright (c) 2007 - 2014, Intel Corporation. All rights reserved.<BR>
+//
+//  This program and the accompanying materials
+//  are licensed and made available under the terms and conditions of the BSD License
+//  which accompanies this distribution.  The full text of the license may be found at
+//  http://opensource.org/licenses/bsd-license.php
+//
+//  THE PROGRAM IS DISTRIBUTED UNDER THE BSD LICENSE ON AN "AS IS" BASIS,
+//  WITHOUT WARRANTIES OR REPRESENTATIONS OF ANY KIND, EITHER EXPRESS OR IMPLIED.
+//
+//**/
+
+    .thumb
+    .syntax unified
+
+ASM_GLOBAL ASM_PFX(EbcLLCALLEXNative)
+ASM_GLOBAL ASM_PFX(EbcLLEbcInterpret)
+ASM_GLOBAL ASM_PFX(EbcLLExecuteEbcImageEntryPoint)
+
+INTERWORK_FUNC(EbcLLCALLEXNative)
+INTERWORK_FUNC(EbcLLEbcInterpret)
+INTERWORK_FUNC(EbcLLExecuteEbcImageEntryPoint)
+
+ASM_GLOBAL ASM_PFX(mEbcInstructionBufferTemplate)
+
+//****************************************************************************
+// EbcLLCALLEX
+//
+// This function is called to execute an EBC CALLEX instruction.
+// This instruction requires that we thunk out to external native
+// code. For ARM, we copy the VM stack into the main stack and then pop
+// the first 4 arguments off according to the ARM Procedure Call Standard
+// On return, we restore the stack pointer to its original location.
+//
+//****************************************************************************
+// UINTN EbcLLCALLEXNative(UINTN FuncAddr, UINTN NewStackPointer, VOID *FramePtr)
+ASM_PFX(EbcLLCALLEXNative):
+    mov     ip, r0                  // Preserve r0
+
+    //
+    // If the EBC stack frame is smaller than or equal to 16 bytes, we know there
+    // are no stacked arguments #5 and beyond that we need to copy to the native
+    // stack. In this case, we can perform a tail call which is much more
+    // efficient, since there is no need to touch the native stack at all.
+    //
+    sub     r3, r2, r1              // Length = NewStackPointer - FramePtr
+    cmp     r3, #16
+    bgt     0f
+
+    ldrd    r2, r3, [r1, #8]
+    ldrd    r0, r1, [r1]
+
+    bx      ip
+
+    //
+    // More than 16 bytes: we need to build the full native stack frame and copy
+    // the part of the VM stack exceeding 16 bytes (which may contain stacked
+    // arguments) to the native stack
+    //
+0:  push    {r4, lr}
+    mov     r4, sp
+
+    //
+    // Ensure that the stack pointer remains 8 byte aligned,
+    // even if the size of the VM stack frame is not a multiple of 8
+    //
+    add     r1, r1, #16             // Skip over [potential] reg params
+    tst     r3, #7                  // Multiple of 8?
+    beq     1f
+    ldr     r3, [r2, #-4]!          // No? Then push one word
+    str     r3, [sp, #-8]!          // ... but use two slots
+    b       2f
+
+1:  ldrd    r0, r3, [r2, #-8]!
+    strd    r0, r3, [sp, #-8]!
+2:  cmp     r2, r1
+    bgt     1b
+
+    ldrd    r2, r3, [r1, #-8]
+    ldrd    r0, r1, [r1, #-16]
+
+    blx     ip
+
+    mov     sp, r4
+    pop     {r4, pc}
+
+//****************************************************************************
+// EbcLLEbcInterpret
+//
+// This function is called by the thunk code to handle an Native to EBC call
+// This can handle up to 16 arguments (1-4 on in r0-r3, 5-16 are on the stack)
+// ip contains the Entry point that will be the first argument when
+// EBCInterpret is called.
+//
+//****************************************************************************
+ASM_PFX(EbcLLEbcInterpret):
+    stmdb   sp!, {r4, lr}
+
+    // push the entry point and the address of args #5 - #16 onto the stack
+    add     r4, sp, #8
+    str     ip, [sp, #-8]!
+    str     r4, [sp, #4]
+
+    // call C-code
+    bl      ASM_PFX(EbcInterpret)
+
+    add     sp, sp, #8
+    ldmia   sp!, {r4, pc}
+
+//****************************************************************************
+// EbcLLExecuteEbcImageEntryPoint
+//
+// This function is called by the thunk code to handle the image entry point
+// ip contains the Entry point that will be the third argument when
+// ExecuteEbcImageEntryPoint is called.
+//
+//****************************************************************************
+ASM_PFX(EbcLLExecuteEbcImageEntryPoint):
+    ldr     r2, [ip, #12]
+
+    // tail call to C code
+    b       ASM_PFX(ExecuteEbcImageEntryPoint)
+
+//****************************************************************************
+// mEbcInstructionBufferTemplate
+//****************************************************************************
+    .section    ".rodata", "a"
+    .align      2
+    .arm
+ASM_PFX(mEbcInstructionBufferTemplate):
+    adr     ip, .
+    ldr     pc, 0f
+
+    //
+    // Add a magic code here to help the VM recognize the thunk.
+    //
+    udf     #0xEBC
+
+    .long   0                       // EBC_ENTRYPOINT_SIGNATURE
+0:  .long   0                       // EBC_LL_EBC_ENTRYPOINT_SIGNATURE
diff --git a/MdeModulePkg/Universal/EbcDxe/Arm/EbcSupport.c b/MdeModulePkg/Universal/EbcDxe/Arm/EbcSupport.c
new file mode 100644
index 000000000000..2a42b6c4d0e7
--- /dev/null
+++ b/MdeModulePkg/Universal/EbcDxe/Arm/EbcSupport.c
@@ -0,0 +1,470 @@
+/** @file
+  This module contains EBC support routines that are customized based on
+  the target Arm processor.
+
+Copyright (c) 2016, Linaro, Ltd. All rights reserved.<BR>
+Copyright (c) 2015, The Linux Foundation. All rights reserved.<BR>
+Copyright (c) 2006 - 2014, Intel Corporation. All rights reserved.<BR>
+
+This program and the accompanying materials
+are licensed and made available under the terms and conditions of the BSD License
+which accompanies this distribution.  The full text of the license may be found at
+http://opensource.org/licenses/bsd-license.php
+
+THE PROGRAM IS DISTRIBUTED UNDER THE BSD LICENSE ON AN "AS IS" BASIS,
+WITHOUT WARRANTIES OR REPRESENTATIONS OF ANY KIND, EITHER EXPRESS OR IMPLIED.
+
+**/
+
+#include "EbcInt.h"
+#include "EbcExecute.h"
+#include "EbcDebuggerHook.h"
+
+//
+// Amount of space that is not used in the stack
+//
+#define STACK_REMAIN_SIZE (1024 * 4)
+
+#pragma pack(1)
+typedef struct {
+  UINT32    Instr[2];
+  UINT32    Magic;
+  UINT32    EbcEntryPoint;
+  UINT32    EbcLlEntryPoint;
+} EBC_INSTRUCTION_BUFFER;
+#pragma pack()
+
+extern CONST EBC_INSTRUCTION_BUFFER       mEbcInstructionBufferTemplate;
+
+/**
+  Begin executing an EBC image.
+  This is used for Ebc Thunk call.
+
+  @return The value returned by the EBC application we're going to run.
+
+**/
+UINT64
+EFIAPI
+EbcLLEbcInterpret (
+  VOID
+  );
+
+/**
+  Begin executing an EBC image.
+  This is used for Ebc image entrypoint.
+
+  @return The value returned by the EBC application we're going to run.
+
+**/
+UINT64
+EFIAPI
+EbcLLExecuteEbcImageEntryPoint (
+  VOID
+  );
+
+/**
+  Pushes a 32 bit unsigned value to the VM stack.
+
+  @param VmPtr  The pointer to current VM context.
+  @param Arg    The value to be pushed.
+
+**/
+VOID
+PushU32 (
+  IN VM_CONTEXT *VmPtr,
+  IN UINT32     Arg
+  )
+{
+  //
+  // Advance the VM stack down, and then copy the argument to the stack.
+  // Hope it's aligned.
+  //
+  VmPtr->Gpr[0] -= sizeof (UINT32);
+  *(UINT32 *)(UINTN)VmPtr->Gpr[0] = Arg;
+}
+
+
+/**
+  Begin executing an EBC image.
+
+  This is a thunk function.
+
+  @param  Arg1                  The 1st argument.
+  @param  Arg2                  The 2nd argument.
+  @param  Arg3                  The 3rd argument.
+  @param  Arg4                  The 4th argument.
+  @param  Arg8                  The 8th argument.
+  @param  EntryPoint            The entrypoint of EBC code.
+  @param  Args5_16[]            Array containing arguments #5 to #16.
+
+  @return The value returned by the EBC application we're going to run.
+
+**/
+UINT64
+EFIAPI
+EbcInterpret (
+  IN UINTN      Arg1,
+  IN UINTN      Arg2,
+  IN UINTN      Arg3,
+  IN UINTN      Arg4,
+  IN UINTN      EntryPoint,
+  IN UINTN      Args5_16[]
+  )
+{
+  //
+  // Create a new VM context on the stack
+  //
+  VM_CONTEXT  VmContext;
+  UINTN       Addr;
+  EFI_STATUS  Status;
+  UINTN       StackIndex;
+
+  //
+  // Get the EBC entry point
+  //
+  Addr = EntryPoint;
+
+  //
+  // Now clear out our context
+  //
+  ZeroMem ((VOID *) &VmContext, sizeof (VM_CONTEXT));
+
+  //
+  // Set the VM instruction pointer to the correct location in memory.
+  //
+  VmContext.Ip = (VMIP) Addr;
+
+  //
+  // Initialize the stack pointer for the EBC. Get the current system stack
+  // pointer and adjust it down by the max needed for the interpreter.
+  //
+
+  //
+  // Adjust the VM's stack pointer down.
+  //
+
+  Status = GetEBCStack((EFI_HANDLE)(UINTN)-1, &VmContext.StackPool, &StackIndex);
+  if (EFI_ERROR(Status)) {
+    return Status;
+  }
+  VmContext.StackTop = (UINT8*)VmContext.StackPool + (STACK_REMAIN_SIZE);
+  VmContext.Gpr[0] = (UINT32) ((UINT8*)VmContext.StackPool + STACK_POOL_SIZE);
+  VmContext.HighStackBottom = (UINTN) VmContext.Gpr[0];
+  VmContext.Gpr[0] -= sizeof (UINTN);
+
+  //
+  // Align the stack on a natural boundary.
+  //
+  VmContext.Gpr[0] &= ~(VM_REGISTER)(sizeof (UINTN) - 1);
+
+  //
+  // Put a magic value in the stack gap, then adjust down again.
+  //
+  *(UINTN *) (UINTN) (VmContext.Gpr[0]) = (UINTN) VM_STACK_KEY_VALUE;
+  VmContext.StackMagicPtr             = (UINTN *) (UINTN) VmContext.Gpr[0];
+
+  //
+  // The stack upper to LowStackTop belongs to the VM.
+  //
+  VmContext.LowStackTop   = (UINTN) VmContext.Gpr[0];
+
+  //
+  // For the worst case, assume there are 4 arguments passed in registers, store
+  // them to VM's stack.
+  //
+  PushU32 (&VmContext, (UINT32) Args5_16[11]);
+  PushU32 (&VmContext, (UINT32) Args5_16[10]);
+  PushU32 (&VmContext, (UINT32) Args5_16[9]);
+  PushU32 (&VmContext, (UINT32) Args5_16[8]);
+  PushU32 (&VmContext, (UINT32) Args5_16[7]);
+  PushU32 (&VmContext, (UINT32) Args5_16[6]);
+  PushU32 (&VmContext, (UINT32) Args5_16[5]);
+  PushU32 (&VmContext, (UINT32) Args5_16[4]);
+  PushU32 (&VmContext, (UINT32) Args5_16[3]);
+  PushU32 (&VmContext, (UINT32) Args5_16[2]);
+  PushU32 (&VmContext, (UINT32) Args5_16[1]);
+  PushU32 (&VmContext, (UINT32) Args5_16[0]);
+  PushU32 (&VmContext, (UINT32) Arg4);
+  PushU32 (&VmContext, (UINT32) Arg3);
+  PushU32 (&VmContext, (UINT32) Arg2);
+  PushU32 (&VmContext, (UINT32) Arg1);
+
+  //
+  // Interpreter assumes 64-bit return address is pushed on the stack.
+  // Arm does not do this so pad the stack accordingly.
+  //
+  PushU32 (&VmContext, 0x0UL);
+  PushU32 (&VmContext, 0x0UL);
+  PushU32 (&VmContext, 0x12345678UL);
+  PushU32 (&VmContext, 0x87654321UL);
+
+  //
+  // For Arm, this is where we say our return address is
+  //
+  VmContext.StackRetAddr  = (UINT64) VmContext.Gpr[0];
+
+  //
+  // We need to keep track of where the EBC stack starts. This way, if the EBC
+  // accesses any stack variables above its initial stack setting, then we know
+  // it's accessing variables passed into it, which means the data is on the
+  // VM's stack.
+  // When we're called, on the stack (high to low) we have the parameters, the
+  // return address, then the saved ebp. Save the pointer to the return address.
+  // EBC code knows that's there, so should look above it for function parameters.
+  // The offset is the size of locals (VMContext + Addr + saved ebp).
+  // Note that the interpreter assumes there is a 16 bytes of return address on
+  // the stack too, so adjust accordingly.
+  //  VmContext.HighStackBottom = (UINTN)(Addr + sizeof (VmContext) + sizeof (Addr));
+  //
+
+  //
+  // Begin executing the EBC code
+  //
+  EbcDebuggerHookEbcInterpret (&VmContext);
+  EbcExecute (&VmContext);
+
+  //
+  // Return the value in Gpr[7] unless there was an error
+  //
+  ReturnEBCStack(StackIndex);
+  return (UINT64) VmContext.Gpr[7];
+}
+
+
+/**
+  Begin executing an EBC image.
+
+  @param  ImageHandle      image handle for the EBC application we're executing
+  @param  SystemTable      standard system table passed into an driver's entry
+                           point
+  @param  EntryPoint       The entrypoint of EBC code.
+
+  @return The value returned by the EBC application we're going to run.
+
+**/
+UINT64
+EFIAPI
+ExecuteEbcImageEntryPoint (
+  IN EFI_HANDLE           ImageHandle,
+  IN EFI_SYSTEM_TABLE     *SystemTable,
+  IN UINTN                EntryPoint
+  )
+{
+  //
+  // Create a new VM context on the stack
+  //
+  VM_CONTEXT  VmContext;
+  UINTN       Addr;
+  EFI_STATUS  Status;
+  UINTN       StackIndex;
+
+  //
+  // Get the EBC entry point
+  //
+  Addr = EntryPoint;
+
+  //
+  // Now clear out our context
+  //
+  ZeroMem ((VOID *) &VmContext, sizeof (VM_CONTEXT));
+
+  //
+  // Save the image handle so we can track the thunks created for this image
+  //
+  VmContext.ImageHandle = ImageHandle;
+  VmContext.SystemTable = SystemTable;
+
+  //
+  // Set the VM instruction pointer to the correct location in memory.
+  //
+  VmContext.Ip = (VMIP) Addr;
+
+  //
+  // Initialize the stack pointer for the EBC. Get the current system stack
+  // pointer and adjust it down by the max needed for the interpreter.
+  //
+
+  //
+  // Allocate stack pool
+  //
+  Status = GetEBCStack(ImageHandle, &VmContext.StackPool, &StackIndex);
+  if (EFI_ERROR(Status)) {
+    return Status;
+  }
+  VmContext.StackTop = (UINT8*)VmContext.StackPool + (STACK_REMAIN_SIZE);
+  VmContext.Gpr[0] = (UINT64)(UINTN) ((UINT8*)VmContext.StackPool + STACK_POOL_SIZE);
+  VmContext.HighStackBottom = (UINTN)VmContext.Gpr[0];
+  VmContext.Gpr[0] -= sizeof (UINTN);
+
+  //
+  // Put a magic value in the stack gap, then adjust down again
+  //
+  *(UINTN *) (UINTN) (VmContext.Gpr[0]) = (UINTN) VM_STACK_KEY_VALUE;
+  VmContext.StackMagicPtr             = (UINTN *) (UINTN) VmContext.Gpr[0];
+
+  //
+  // Align the stack on a natural boundary
+  //  VmContext.Gpr[0] &= ~(sizeof(UINTN) - 1);
+  //
+  VmContext.LowStackTop   = (UINTN) VmContext.Gpr[0];
+  VmContext.Gpr[0] -= sizeof (UINTN);
+  *(UINTN *) (UINTN) (VmContext.Gpr[0]) = (UINTN) SystemTable;
+  VmContext.Gpr[0] -= sizeof (UINTN);
+  *(UINTN *) (UINTN) (VmContext.Gpr[0]) = (UINTN) ImageHandle;
+
+  VmContext.Gpr[0] -= 16;
+  VmContext.StackRetAddr  = (UINT64) VmContext.Gpr[0];
+  //
+  // VM pushes 16-bytes for return address. Simulate that here.
+  //
+
+  //
+  // Begin executing the EBC code
+  //
+  EbcDebuggerHookExecuteEbcImageEntryPoint (&VmContext);
+  EbcExecute (&VmContext);
+
+  //
+  // Return the value in Gpr[7] unless there was an error
+  //
+  ReturnEBCStack(StackIndex);
+  return (UINT64) VmContext.Gpr[7];
+}
+
+
+/**
+  Create thunks for an EBC image entry point, or an EBC protocol service.
+
+  @param  ImageHandle           Image handle for the EBC image. If not null, then
+                                we're creating a thunk for an image entry point.
+  @param  EbcEntryPoint         Address of the EBC code that the thunk is to call
+  @param  Thunk                 Returned thunk we create here
+  @param  Flags                 Flags indicating options for creating the thunk
+
+  @retval EFI_SUCCESS           The thunk was created successfully.
+  @retval EFI_INVALID_PARAMETER The parameter of EbcEntryPoint is not 16-bit
+                                aligned.
+  @retval EFI_OUT_OF_RESOURCES  There is not enough memory to created the EBC
+                                Thunk.
+  @retval EFI_BUFFER_TOO_SMALL  EBC_THUNK_SIZE is not larger enough.
+
+**/
+EFI_STATUS
+EbcCreateThunks (
+  IN EFI_HANDLE           ImageHandle,
+  IN VOID                 *EbcEntryPoint,
+  OUT VOID                **Thunk,
+  IN  UINT32              Flags
+  )
+{
+  EBC_INSTRUCTION_BUFFER       *InstructionBuffer;
+
+  //
+  // Check alignment of pointer to EBC code
+  //
+  if ((UINT32) (UINTN) EbcEntryPoint & 0x01) {
+    return EFI_INVALID_PARAMETER;
+  }
+
+  InstructionBuffer = AllocatePool (sizeof (EBC_INSTRUCTION_BUFFER));
+  if (InstructionBuffer == NULL) {
+    return EFI_OUT_OF_RESOURCES;
+  }
+
+  //
+  // Give them the address of our buffer we're going to fix up
+  //
+  *Thunk = InstructionBuffer;
+
+  //
+  // Copy whole thunk instruction buffer template
+  //
+  CopyMem (InstructionBuffer, &mEbcInstructionBufferTemplate,
+    sizeof (EBC_INSTRUCTION_BUFFER));
+
+  //
+  // Patch EbcEntryPoint and EbcLLEbcInterpret
+  //
+  InstructionBuffer->EbcEntryPoint = (UINT32)EbcEntryPoint;
+  if ((Flags & FLAG_THUNK_ENTRY_POINT) != 0) {
+    InstructionBuffer->EbcLlEntryPoint = (UINT32)EbcLLExecuteEbcImageEntryPoint;
+  } else {
+    InstructionBuffer->EbcLlEntryPoint = (UINT32)EbcLLEbcInterpret;
+  }
+
+  //
+  // Add the thunk to the list for this image. Do this last since the add
+  // function flushes the cache for us.
+  //
+  EbcAddImageThunk (ImageHandle, InstructionBuffer,
+    sizeof (EBC_INSTRUCTION_BUFFER));
+
+  return EFI_SUCCESS;
+}
+
+
+/**
+  This function is called to execute an EBC CALLEX instruction.
+  The function check the callee's content to see whether it is common native
+  code or a thunk to another piece of EBC code.
+  If the callee is common native code, use EbcLLCAllEXASM to manipulate,
+  otherwise, set the VM->IP to target EBC code directly to avoid another VM
+  be startup which cost time and stack space.
+
+  @param  VmPtr            Pointer to a VM context.
+  @param  FuncAddr         Callee's address
+  @param  NewStackPointer  New stack pointer after the call
+  @param  FramePtr         New frame pointer after the call
+  @param  Size             The size of call instruction
+
+**/
+VOID
+EbcLLCALLEX (
+  IN VM_CONTEXT   *VmPtr,
+  IN UINTN        FuncAddr,
+  IN UINTN        NewStackPointer,
+  IN VOID         *FramePtr,
+  IN UINT8        Size
+  )
+{
+  CONST EBC_INSTRUCTION_BUFFER *InstructionBuffer;
+
+  //
+  // Processor specific code to check whether the callee is a thunk to EBC.
+  //
+  InstructionBuffer = (EBC_INSTRUCTION_BUFFER *)FuncAddr;
+
+  if (CompareMem (InstructionBuffer, &mEbcInstructionBufferTemplate,
+        sizeof(EBC_INSTRUCTION_BUFFER) - 2 * sizeof (UINT32)) == 0) {
+    //
+    // The callee is a thunk to EBC, adjust the stack pointer down 16 bytes and
+    // put our return address and frame pointer on the VM stack.
+    // Then set the VM's IP to new EBC code.
+    //
+    VmPtr->Gpr[0] -= 8;
+    VmWriteMemN (VmPtr, (UINTN) VmPtr->Gpr[0], (UINTN) FramePtr);
+    VmPtr->FramePtr = (VOID *) (UINTN) VmPtr->Gpr[0];
+    VmPtr->Gpr[0] -= 8;
+    VmWriteMem64 (VmPtr, (UINTN) VmPtr->Gpr[0], (UINT64) (UINTN) (VmPtr->Ip + Size));
+
+    VmPtr->Ip = (VMIP) InstructionBuffer->EbcEntryPoint;
+  } else {
+    //
+    // The callee is not a thunk to EBC, call native code,
+    // and get return value.
+    //
+    // Note that we are not able to distinguish which part of the interval
+    // [NewStackPointer, FramePtr] consists of stacked function arguments for
+    // this call, and which part simply consists of locals in the caller's
+    // stack frame. All we know is that there is an 8 byte gap at the top that
+    // we can ignore.
+    //
+    VmPtr->Gpr[7] = EbcLLCALLEXNative (FuncAddr, NewStackPointer, FramePtr - 8);
+
+    //
+    // Advance the IP.
+    //
+    VmPtr->Ip += Size;
+  }
+}
+
diff --git a/MdeModulePkg/Universal/EbcDxe/EbcDebugger.inf b/MdeModulePkg/Universal/EbcDxe/EbcDebugger.inf
index ce413c0b25f6..b55538304605 100644
--- a/MdeModulePkg/Universal/EbcDxe/EbcDebugger.inf
+++ b/MdeModulePkg/Universal/EbcDxe/EbcDebugger.inf
@@ -25,7 +25,7 @@ [Defines]
 #
 # The following information is for reference only and not required by the build tools.
 #
-#  VALID_ARCHITECTURES           = IA32 X64 IPF AARCH64
+#  VALID_ARCHITECTURES           = IA32 X64 IPF ARM AARCH64
 #
 
 [Sources]
@@ -86,6 +86,10 @@ [Sources.AARCH64]
   AArch64/EbcSupport.c
   AArch64/EbcLowLevel.S
 
+[Sources.ARM]
+  Arm/EbcSupport.c
+  Arm/EbcLowLevel.S
+
 [Packages]
   MdePkg/MdePkg.dec
   MdeModulePkg/MdeModulePkg.dec
diff --git a/MdeModulePkg/Universal/EbcDxe/EbcDebugger/EdbDisasmSupport.h b/MdeModulePkg/Universal/EbcDxe/EbcDebugger/EdbDisasmSupport.h
index af5a7cab99f9..db3b0a1f454f 100644
--- a/MdeModulePkg/Universal/EbcDxe/EbcDebugger/EdbDisasmSupport.h
+++ b/MdeModulePkg/Universal/EbcDxe/EbcDebugger/EdbDisasmSupport.h
@@ -19,10 +19,10 @@ WITHOUT WARRANTIES OR REPRESENTATIONS OF ANY KIND, EITHER EXPRESS OR IMPLIED.
 
 #define EDB_BYTECODE_NUMBER_IN_LINE     5
 
-#ifdef EFI32
+#if defined (MDE_CPU_IA32) || defined (MDE_CPU_ARM)
 #define EDB_PRINT_ADDRESS_FORMAT    L"%08x: "
 #else
-// To use 012l instead of 016l because space is not enough
+// We use 012l instead of 016l due to space constraints
 #define EDB_PRINT_ADDRESS_FORMAT    L"%012lx: "
 #endif
 
diff --git a/MdeModulePkg/Universal/EbcDxe/EbcDebuggerConfig.inf b/MdeModulePkg/Universal/EbcDxe/EbcDebuggerConfig.inf
index 0d931a46f01a..c08ad17dd1c8 100644
--- a/MdeModulePkg/Universal/EbcDxe/EbcDebuggerConfig.inf
+++ b/MdeModulePkg/Universal/EbcDxe/EbcDebuggerConfig.inf
@@ -25,7 +25,7 @@ [Defines]
 #
 # The following information is for reference only and not required by the build tools.
 #
-#  VALID_ARCHITECTURES           = IA32 X64 IPF AARCH64
+#  VALID_ARCHITECTURES           = IA32 X64 IPF ARM AARCH64
 #
 
 [Sources]
diff --git a/MdeModulePkg/Universal/EbcDxe/EbcDxe.inf b/MdeModulePkg/Universal/EbcDxe/EbcDxe.inf
index d11888e25f92..78e058b6cbc6 100644
--- a/MdeModulePkg/Universal/EbcDxe/EbcDxe.inf
+++ b/MdeModulePkg/Universal/EbcDxe/EbcDxe.inf
@@ -29,7 +29,7 @@ [Defines]
 #
 # The following information is for reference only and not required by the build tools.
 #
-#  VALID_ARCHITECTURES           = IA32 X64 IPF AARCH64
+#  VALID_ARCHITECTURES           = IA32 X64 IPF ARM AARCH64
 #
 
 [Sources]
@@ -57,6 +57,10 @@ [Sources.IPF]
   Ipf/EbcSupport.c
   Ipf/EbcLowLevel.s
 
+[Sources.ARM]
+  Arm/EbcSupport.c
+  Arm/EbcLowLevel.S
+
 [Sources.AARCH64]
   AArch64/EbcSupport.c
   AArch64/EbcLowLevel.S
-- 
2.9.3.windows.2