diff --git a/arch/arc/bsp/fatal_error.c b/arch/arc/bsp/fatal_error.c
index 4590d8c369e..44ccc17e795 100644
--- a/arch/arc/bsp/fatal_error.c
+++ b/arch/arc/bsp/fatal_error.c
@@ -39,7 +39,7 @@ This module provides the _SysFatalErrorHandler() routine for ARCv2 BSPs.
 
 #include <cputype.h>
 #include <nanokernel.h>
-#include <nanokernel/cpu.h>
+#include <arch/cpu.h>
 #include <toolchain.h>
 #include <sections.h>
 #include "board.h"
diff --git a/arch/arc/bsp/reset.s b/arch/arc/bsp/reset.s
index 6f711a3d910..8d1f1390218 100644
--- a/arch/arc/bsp/reset.s
+++ b/arch/arc/bsp/reset.s
@@ -40,7 +40,7 @@ Reset handler that prepares the system for running C code.
 #include <board.h>
 #include <toolchain.h>
 #include <sections.h>
-#include <nanokernel/cpu.h>
+#include <arch/cpu.h>
 
 #define _RAM_END (CONFIG_RAM_START + CONFIG_RAM_SIZE)
 
diff --git a/arch/arc/core/context.c b/arch/arc/core/context.c
index cc0f0da31b1..cec380c02ce 100644
--- a/arch/arc/core/context.c
+++ b/arch/arc/core/context.c
@@ -37,7 +37,7 @@
  */
 
 #include <nanokernel.h>
-#include <nanokernel/cpu.h>
+#include <arch/cpu.h>
 #include <toolchain.h>
 #include <nanok.h>
 #include <nanocontextentry.h>
diff --git a/arch/arc/core/cpu_idle.s b/arch/arc/core/cpu_idle.s
index 1bd4f0d7806..ef547a5edc8 100644
--- a/arch/arc/core/cpu_idle.s
+++ b/arch/arc/core/cpu_idle.s
@@ -41,7 +41,7 @@ CPU power management routines.
 #include <offsets.h>
 #include <toolchain.h>
 #include <sections.h>
-#include <nanokernel/cpu.h>
+#include <arch/cpu.h>
 
 GTEXT(nano_cpu_idle)
 GTEXT(nano_cpu_atomic_idle)
diff --git a/arch/arc/core/fast_irq.s b/arch/arc/core/fast_irq.s
index 37929e709c8..3a6969f3b3f 100644
--- a/arch/arc/core/fast_irq.s
+++ b/arch/arc/core/fast_irq.s
@@ -42,7 +42,7 @@ See isr_wrapper.s for details.
 #include <nanok.h>
 #include <offsets.h>
 #include <toolchain.h>
-#include <nanokernel/cpu.h>
+#include <arch/cpu.h>
 #include "swap_macros.h"
 
 GTEXT(_firq_enter)
diff --git a/arch/arc/core/fatal.c b/arch/arc/core/fatal.c
index bfe192533f1..791a45d17e2 100644
--- a/arch/arc/core/fatal.c
+++ b/arch/arc/core/fatal.c
@@ -39,7 +39,7 @@ ARCv2 CPUs.
 #include <nanok.h>
 #include <offsets.h>
 #include <toolchain.h>
-#include <nanokernel/cpu.h>
+#include <arch/cpu.h>
 
 #ifdef CONFIG_PRINTK
 #include <misc/printk.h>
diff --git a/arch/arc/core/fault.c b/arch/arc/core/fault.c
index dc5c2898679..fa07768c2e7 100644
--- a/arch/arc/core/fault.c
+++ b/arch/arc/core/fault.c
@@ -40,7 +40,7 @@
 
 #include <cputype.h>
 #include <nanokernel.h>
-#include <nanokernel/cpu.h>
+#include <arch/cpu.h>
 #include <nanok.h>
 
 #ifdef CONFIG_PRINTK
diff --git a/arch/arc/core/fault_s.s b/arch/arc/core/fault_s.s
index fd5136388e6..c167bc582b3 100644
--- a/arch/arc/core/fault_s.s
+++ b/arch/arc/core/fault_s.s
@@ -39,7 +39,7 @@
 
 #include <toolchain.h>
 #include <sections.h>
-#include <nanokernel/cpu.h>
+#include <arch/cpu.h>
 #include "swap_macros.h"
 
 GTEXT(_Fault)
diff --git a/arch/arc/core/irq_lock.s b/arch/arc/core/irq_lock.s
index de7da469a60..721b9b98a5d 100644
--- a/arch/arc/core/irq_lock.s
+++ b/arch/arc/core/irq_lock.s
@@ -38,7 +38,7 @@
 #include <nanok.h>
 #include <offsets.h>
 #include <toolchain.h>
-#include <nanokernel/cpu.h>
+#include <arch/cpu.h>
 
 /*******************************************************************************
 *
diff --git a/arch/arc/core/irq_manage.c b/arch/arc/core/irq_manage.c
index 1a736ec9ae7..83a358497ba 100644
--- a/arch/arc/core/irq_manage.c
+++ b/arch/arc/core/irq_manage.c
@@ -45,7 +45,7 @@
  */
 
 #include <nanokernel.h>
-#include <nanokernel/cpu.h>
+#include <arch/cpu.h>
 #include <misc/__assert.h>
 #include <toolchain.h>
 #include <sections.h>
diff --git a/arch/arc/core/isr_wrapper.s b/arch/arc/core/isr_wrapper.s
index b2202004c52..97e6740073c 100644
--- a/arch/arc/core/isr_wrapper.s
+++ b/arch/arc/core/isr_wrapper.s
@@ -44,7 +44,7 @@ a parameter.
 #include <sections.h>
 #include <sw_isr_table.h>
 #include <nanok.h>
-#include <nanokernel/cpu.h>
+#include <arch/cpu.h>
 
 GTEXT(_isr_enter)
 GTEXT(_isr_demux)
diff --git a/arch/arc/core/regular_irq.s b/arch/arc/core/regular_irq.s
index 4ea21f991fe..ab89a4aa8aa 100644
--- a/arch/arc/core/regular_irq.s
+++ b/arch/arc/core/regular_irq.s
@@ -43,7 +43,7 @@ See isr_wrapper.s for details.
 #include <nanok.h>
 #include <offsets.h>
 #include <toolchain.h>
-#include <nanokernel/cpu.h>
+#include <arch/cpu.h>
 #include "swap_macros.h"
 
 GTEXT(_rirq_enter)
diff --git a/arch/arc/core/swap.s b/arch/arc/core/swap.s
index a83a534c438..75c8f1be304 100644
--- a/arch/arc/core/swap.s
+++ b/arch/arc/core/swap.s
@@ -43,7 +43,7 @@ See isr_wrapper.s for details.
 #include <nanok.h>
 #include <offsets.h>
 #include <toolchain.h>
-#include <nanokernel/cpu.h>
+#include <arch/cpu.h>
 #include <v2/irq.h>
 #include "swap_macros.h"
 
diff --git a/arch/arc/core/swap_macros.h b/arch/arc/core/swap_macros.h
index 1f42f2a0002..2f4d55a24a5 100644
--- a/arch/arc/core/swap_macros.h
+++ b/arch/arc/core/swap_macros.h
@@ -36,7 +36,7 @@
 #include <nanok.h>
 #include <offsets.h>
 #include <toolchain.h>
-#include <nanokernel/cpu.h>
+#include <arch/cpu.h>
 
 #ifdef _ASMLANGUAGE
 
diff --git a/arch/arc/include/nanok.h b/arch/arc/include/nanok.h
index 567898e0f49..4847f23d128 100644
--- a/arch/arc/include/nanok.h
+++ b/arch/arc/include/nanok.h
@@ -50,7 +50,7 @@ extern "C" {
 
 #include <toolchain.h>
 #include <sections.h>
-#include <nanokernel/cpu.h>
+#include <arch/cpu.h>
 
 #ifndef _ASMLANGUAGE
 #include <stdint.h>
diff --git a/arch/arc/include/v2/cache.h b/arch/arc/include/v2/cache.h
index 5a258232b2b..b7167c22b3a 100644
--- a/arch/arc/include/v2/cache.h
+++ b/arch/arc/include/v2/cache.h
@@ -39,7 +39,7 @@
 #ifndef _ARCV2_CACHE__H_
 #define _ARCV2_CACHE__H_
 
-#include <nanokernel/cpu.h>
+#include <arch/cpu.h>
 
 #ifndef _ASMLANGUAGE
 
diff --git a/arch/arc/interrupt_controller/arcv2_irq_unit.c b/arch/arc/interrupt_controller/arcv2_irq_unit.c
index c49c8d22469..dc062c8de0e 100644
--- a/arch/arc/interrupt_controller/arcv2_irq_unit.c
+++ b/arch/arc/interrupt_controller/arcv2_irq_unit.c
@@ -40,7 +40,7 @@
  */
 
 #include <nanokernel.h>
-#include <nanokernel/cpu.h>
+#include <arch/cpu.h>
 #include <board.h>
 
 /*
diff --git a/arch/arc/serial/nsim_uart.c b/arch/arc/serial/nsim_uart.c
index 98bfbc1250c..1fca9f9013d 100644
--- a/arch/arc/serial/nsim_uart.c
+++ b/arch/arc/serial/nsim_uart.c
@@ -29,7 +29,7 @@
  */
 
 #include <nanokernel.h>
-#include <nanokernel/cpu.h>
+#include <arch/cpu.h>
 #include <sections.h>
 #include <misc/__assert.h>
 #include <stdint.h>
diff --git a/arch/arc/timer/arcv2_timer0.c b/arch/arc/timer/arcv2_timer0.c
index 680c9197bf6..e9c43f0f141 100644
--- a/arch/arc/timer/arcv2_timer0.c
+++ b/arch/arc/timer/arcv2_timer0.c
@@ -40,11 +40,11 @@ The ARCv2 processor timer provides a 32-bit incrementing, wrap-to-zero counter.
 */
 
 #include <nanokernel.h>
-#include <nanokernel/cpu.h>
+#include <arch/cpu.h>
 #include <toolchain.h>
 #include <sections.h>
 #include <misc/__assert.h>
-#include <nanokernel/arc/v2/aux_regs.h>
+#include <arch/arc/v2/aux_regs.h>
 #include <clock_vars.h>
 #include <drivers/system_timer.h>
 
diff --git a/arch/arm/bsp/CortexM/nmi.c b/arch/arm/bsp/CortexM/nmi.c
index 80f44d71707..87d55792851 100644
--- a/arch/arm/bsp/CortexM/nmi.c
+++ b/arch/arm/bsp/CortexM/nmi.c
@@ -38,7 +38,7 @@ run time handler.
 */
 
 #include <nanokernel.h>
-#include <nanokernel/cpu.h>
+#include <arch/cpu.h>
 #include <misc/printk.h>
 #include <toolchain.h>
 #include <sections.h>
diff --git a/arch/arm/bsp/CortexM/reset.s b/arch/arm/bsp/CortexM/reset.s
index de13fabbefe..c90d226bffb 100644
--- a/arch/arm/bsp/CortexM/reset.s
+++ b/arch/arm/bsp/CortexM/reset.s
@@ -40,7 +40,7 @@ Reset handler that prepares the system for running C code.
 #include <board.h>
 #include <toolchain.h>
 #include <sections.h>
-#include <nanokernel/cpu.h>
+#include <arch/cpu.h>
 #include "vector_table.h"
 
 _ASM_FILE_PROLOGUE
diff --git a/arch/arm/bsp/CortexM/scb.c b/arch/arm/bsp/CortexM/scb.c
index 5792a74410b..7ef417dc89c 100644
--- a/arch/arm/bsp/CortexM/scb.c
+++ b/arch/arm/bsp/CortexM/scb.c
@@ -39,7 +39,7 @@ definitions and more complex routines, if needed.
 */
 
 #include <nanokernel.h>
-#include <nanokernel/cpu.h>
+#include <arch/cpu.h>
 #include <misc/util.h>
 
 #define SCB_AIRCR_VECTKEY_EN_W 0x05FA
diff --git a/arch/arm/bsp/CortexM/scs.c b/arch/arm/bsp/CortexM/scs.c
index 0369155ff19..40753c018e3 100644
--- a/arch/arm/bsp/CortexM/scs.c
+++ b/arch/arm/bsp/CortexM/scs.c
@@ -38,7 +38,7 @@ definitions and more complex routines, if needed.
 */
 
 #include <nanokernel.h>
-#include <nanokernel/cpu.h>
+#include <arch/cpu.h>
 #include <toolchain.h>
 #include <sections.h>
 
diff --git a/arch/arm/bsp/sysFatalErrorHandler.c b/arch/arm/bsp/sysFatalErrorHandler.c
index 68bb5bf0fc5..8a6b3da7ca7 100644
--- a/arch/arm/bsp/sysFatalErrorHandler.c
+++ b/arch/arm/bsp/sysFatalErrorHandler.c
@@ -39,7 +39,7 @@ This module provides the _SysFatalErrorHandler() routine for Cortex-M BSPs.
 
 #include <cputype.h>
 #include <nanokernel.h>
-#include <nanokernel/cpu.h>
+#include <arch/cpu.h>
 #include <toolchain.h>
 #include <sections.h>
 #include "board.h"
diff --git a/arch/arm/core/basepri.s b/arch/arm/core/basepri.s
index 2f07ed943f2..6e5cc2f956e 100644
--- a/arch/arm/core/basepri.s
+++ b/arch/arm/core/basepri.s
@@ -52,7 +52,7 @@ unlocked. This achieves two purposes:
 
 #include <toolchain.h>
 #include <sections.h>
-#include <nanokernel/cpu.h>
+#include <arch/cpu.h>
 
 _ASM_FILE_PROLOGUE
 
diff --git a/arch/arm/core/cpu_idle.s b/arch/arm/core/cpu_idle.s
index 2a9b4880d9e..ecb5ea54887 100644
--- a/arch/arm/core/cpu_idle.s
+++ b/arch/arm/core/cpu_idle.s
@@ -39,7 +39,7 @@ DESCRIPTION
 #include <offsets.h>
 #include <toolchain.h>
 #include <sections.h>
-#include <nanokernel/cpu.h>
+#include <arch/cpu.h>
 #ifdef CONFIG_TICKLESS_IDLE
 #include <nanok.h>
 #endif
diff --git a/arch/arm/core/exc_exit.s b/arch/arm/core/exc_exit.s
index eb970fb0095..4ab8c5af764 100644
--- a/arch/arm/core/exc_exit.s
+++ b/arch/arm/core/exc_exit.s
@@ -43,7 +43,7 @@ wrapped around by _isr_wrapper()).
 #include <nanok.h>
 #include <offsets.h>
 #include <toolchain.h>
-#include <nanokernel/cpu.h>
+#include <arch/cpu.h>
 
 _ASM_FILE_PROLOGUE
 
diff --git a/arch/arm/core/fault.c b/arch/arm/core/fault.c
index a36b1705884..a320e913a40 100644
--- a/arch/arm/core/fault.c
+++ b/arch/arm/core/fault.c
@@ -40,7 +40,7 @@ Common fault handler for ARM Cortex-M processors.
 
 #include <cputype.h>
 #include <nanokernel.h>
-#include <nanokernel/cpu.h>
+#include <arch/cpu.h>
 #include <nanok.h>
 
 #ifdef CONFIG_PRINTK
diff --git a/arch/arm/core/fault_s.s b/arch/arm/core/fault_s.s
index d9fd9f7bed4..59a8b80857e 100644
--- a/arch/arm/core/fault_s.s
+++ b/arch/arm/core/fault_s.s
@@ -39,7 +39,7 @@ Fault handlers for ARM Cortex-M processors.
 
 #include <toolchain.h>
 #include <sections.h>
-#include <nanokernel/cpu.h>
+#include <arch/cpu.h>
 
 _ASM_FILE_PROLOGUE
 
diff --git a/arch/arm/core/gdb_stub.s b/arch/arm/core/gdb_stub.s
index d6d872a59bb..97391a66095 100644
--- a/arch/arm/core/gdb_stub.s
+++ b/arch/arm/core/gdb_stub.s
@@ -47,7 +47,7 @@ anymore.
 #include <toolchain.h>
 #include <sections.h>
 #include <nanok.h>
-#include <nanokernel/cpu.h>
+#include <arch/cpu.h>
 
 _ASM_FILE_PROLOGUE
 
diff --git a/arch/arm/core/gdb_stub_irq_vector_table.c b/arch/arm/core/gdb_stub_irq_vector_table.c
index 1f0f79dbd7f..1b3c78ba3a9 100644
--- a/arch/arm/core/gdb_stub_irq_vector_table.c
+++ b/arch/arm/core/gdb_stub_irq_vector_table.c
@@ -39,7 +39,7 @@ user-installed ISRs.
 
 #include <toolchain.h>
 #include <sections.h>
-#include <nanokernel/cpu.h>
+#include <arch/cpu.h>
 
 typedef void (*vth)(void); /* Vector Table Handler */
 
diff --git a/arch/arm/core/irq_init.c b/arch/arm/core/irq_init.c
index 19a5a75acdc..f72f26ce0e1 100644
--- a/arch/arm/core/irq_init.c
+++ b/arch/arm/core/irq_init.c
@@ -43,7 +43,7 @@ the PendSV exception.
 #include <toolchain.h>
 #include <sections.h>
 #include <nanokernel.h>
-#include <nanokernel/cpu.h>
+#include <arch/cpu.h>
 
 /*******************************************************************************
 *
diff --git a/arch/arm/core/irq_manage.c b/arch/arm/core/irq_manage.c
index 4cba0ad5c28..af582d07e3a 100644
--- a/arch/arm/core/irq_manage.c
+++ b/arch/arm/core/irq_manage.c
@@ -38,7 +38,7 @@ SW_ISR_TABLE_DYNAMIC has to be enabled for connecting ISRs at runtime.
 */
 
 #include <nanokernel.h>
-#include <nanokernel/cpu.h>
+#include <arch/cpu.h>
 #include <misc/__assert.h>
 #include <toolchain.h>
 #include <sections.h>
diff --git a/arch/arm/core/isr_wrapper.s b/arch/arm/core/isr_wrapper.s
index cf520b528a3..1365b1f5299 100644
--- a/arch/arm/core/isr_wrapper.s
+++ b/arch/arm/core/isr_wrapper.s
@@ -44,7 +44,7 @@ a parameter.
 #include <sections.h>
 #include <sw_isr_table.h>
 #include <nanok.h>
-#include <nanokernel/cpu.h>
+#include <arch/cpu.h>
 
 _ASM_FILE_PROLOGUE
 
diff --git a/arch/arm/core/nano_fiber_abort.c b/arch/arm/core/nano_fiber_abort.c
index a78abd636c9..77aa154042b 100644
--- a/arch/arm/core/nano_fiber_abort.c
+++ b/arch/arm/core/nano_fiber_abort.c
@@ -49,7 +49,7 @@ the PendSV exception.
 #include <toolchain.h>
 #include <sections.h>
 #include <nanokernel.h>
-#include <nanokernel/cpu.h>
+#include <arch/cpu.h>
 
 extern void _nano_fiber_swap(void);
 
diff --git a/arch/arm/core/nanocontext.c b/arch/arm/core/nanocontext.c
index ce4e1061851..8d0c8b3ad99 100644
--- a/arch/arm/core/nanocontext.c
+++ b/arch/arm/core/nanocontext.c
@@ -37,7 +37,7 @@ architecture.
 */
 
 #include <nanokernel.h>
-#include <nanokernel/cpu.h>
+#include <arch/cpu.h>
 #include <toolchain.h>
 #include <nanok.h>
 #include <nanocontextentry.h>
diff --git a/arch/arm/core/nanofatal.c b/arch/arm/core/nanofatal.c
index 072ba630711..a6ac8d1e3db 100644
--- a/arch/arm/core/nanofatal.c
+++ b/arch/arm/core/nanofatal.c
@@ -42,7 +42,7 @@ This module provides the _NanoFatalErrorHandler() routine for ARM Cortex-M.
 
 #include <cputype.h>
 #include <nanokernel.h>
-#include <nanokernel/cpu.h>
+#include <arch/cpu.h>
 #include <nanok.h>
 
 #ifdef CONFIG_PRINTK
diff --git a/arch/arm/core/swap.s b/arch/arm/core/swap.s
index b312f5fbf40..3f11a86f90c 100644
--- a/arch/arm/core/swap.s
+++ b/arch/arm/core/swap.s
@@ -41,7 +41,7 @@ on ARM Cortex-M3/M4 CPUs.
 #include <nanok.h>
 #include <offsets.h>
 #include <toolchain.h>
-#include <nanokernel/cpu.h>
+#include <arch/cpu.h>
 
 _ASM_FILE_PROLOGUE
 
diff --git a/arch/arm/fsl_frdm_k64f/linker.cmd b/arch/arm/fsl_frdm_k64f/linker.cmd
index 3b6f69de511..f084712ab61 100644
--- a/arch/arm/fsl_frdm_k64f/linker.cmd
+++ b/arch/arm/fsl_frdm_k64f/linker.cmd
@@ -57,4 +57,4 @@ This is the linker script for both standard images and XIP images.
 #define SRAM_START  0x20000000  /* 192K in SRAM space */
 #define SRAM_SIZE   192K
 
-#include <nanokernel/arm/CortexM/scripts/linker.cmd>
+#include <arch/arm/CortexM/scripts/linker.cmd>
diff --git a/arch/arm/fsl_frdm_k64f/system.c b/arch/arm/fsl_frdm_k64f/system.c
index ae2821deefd..c900aff682c 100644
--- a/arch/arm/fsl_frdm_k64f/system.c
+++ b/arch/arm/fsl_frdm_k64f/system.c
@@ -37,7 +37,7 @@ for the fsl_frdm_k64f BSP.
 */
 
 #include <nanokernel.h>
-#include <nanokernel/cpu.h>
+#include <arch/cpu.h>
 #include <cputype.h>
 #include <board.h>
 #include <drivers/k20_mcg.h>
diff --git a/arch/arm/include/CortexM/exc.h b/arch/arm/include/CortexM/exc.h
index 5c313104d33..e2644b8219f 100644
--- a/arch/arm/include/CortexM/exc.h
+++ b/arch/arm/include/CortexM/exc.h
@@ -38,7 +38,7 @@ Exception/interrupt context helpers.
 #ifndef _ARM_CORTEXM_ISR__H_
 #define _ARM_CORTEXM_ISR__H_
 
-#include <nanokernel/cpu.h>
+#include <arch/cpu.h>
 #include <asm_inline.h>
 
 #ifdef _ASMLANGUAGE
diff --git a/arch/arm/include/nanok.h b/arch/arm/include/nanok.h
index bb54727a295..22f650752d0 100644
--- a/arch/arm/include/nanok.h
+++ b/arch/arm/include/nanok.h
@@ -50,7 +50,7 @@ extern "C" {
 
 #include <toolchain.h>
 #include <sections.h>
-#include <nanokernel/cpu.h>
+#include <arch/cpu.h>
 
 #ifndef _ASMLANGUAGE
 #include <stdint.h>
diff --git a/arch/arm/ti_lm3s6965/linker.cmd b/arch/arm/ti_lm3s6965/linker.cmd
index bb8abfb57c9..83b3f4f823f 100644
--- a/arch/arm/ti_lm3s6965/linker.cmd
+++ b/arch/arm/ti_lm3s6965/linker.cmd
@@ -35,4 +35,4 @@
 #define FLASH_SIZE  256K
 #define SRAM_SIZE   64K
 
-#include <nanokernel/arm/CortexM/scripts/linker.cmd>
+#include <arch/arm/CortexM/scripts/linker.cmd>
diff --git a/arch/arm/ti_lm3s6965/system.c b/arch/arm/ti_lm3s6965/system.c
index 5d285e76e65..bf016b83079 100644
--- a/arch/arm/ti_lm3s6965/system.c
+++ b/arch/arm/ti_lm3s6965/system.c
@@ -37,7 +37,7 @@ for the ti_lm3s6965 BSP.
 */
 
 #include <nanokernel.h>
-#include <nanokernel/cpu.h>
+#include <arch/cpu.h>
 #include <cputype.h>
 #include <board.h>
 #include <drivers/uart.h>
diff --git a/arch/arm/timer/systick.c b/arch/arm/timer/systick.c
index 59190f4497b..83f18d41471 100644
--- a/arch/arm/timer/systick.c
+++ b/arch/arm/timer/systick.c
@@ -52,7 +52,7 @@ conjunction with a microkernel.
 */
 
 #include <nanokernel.h>
-#include <nanokernel/cpu.h>
+#include <arch/cpu.h>
 #include <toolchain.h>
 #include <sections.h>
 #include <misc/__assert.h>
diff --git a/arch/arm/timer/systick_gdb.s b/arch/arm/timer/systick_gdb.s
index 8f00f96ee85..016981030bd 100644
--- a/arch/arm/timer/systick_gdb.s
+++ b/arch/arm/timer/systick_gdb.s
@@ -42,7 +42,7 @@ correct state of the thread that was interrupted.
 
 #include <toolchain.h>
 #include <sections.h>
-#include <nanokernel/cpu.h>
+#include <arch/cpu.h>
 
 _ASM_FILE_PROLOGUE
 
diff --git a/arch/x86/bsp/cache.c b/arch/x86/bsp/cache.c
index 375fabf2056..200d15b7b5f 100644
--- a/arch/x86/bsp/cache.c
+++ b/arch/x86/bsp/cache.c
@@ -35,7 +35,7 @@ This module contains functions for manipulation caches.
 */
 
 #include <nanokernel.h>
-#include <nanokernel/cpu.h>
+#include <arch/cpu.h>
 #include <misc/util.h>
 
 #ifdef CONFIG_CLFLUSH_INSTRUCTION_SUPPORTED
diff --git a/arch/x86/bsp/cache_s.s b/arch/x86/bsp/cache_s.s
index 05e8b7df040..0f010109f95 100644
--- a/arch/x86/bsp/cache_s.s
+++ b/arch/x86/bsp/cache_s.s
@@ -37,7 +37,7 @@ This module contains functions for manipulating caches.
 #ifndef CONFIG_CLFLUSH_INSTRUCTION_SUPPORTED
 
 #define _ASMLANGUAGE
-#include <nanokernel/x86/asm.h>
+#include <arch/x86/asm.h>
 
 	/* externs (internal APIs) */
 	GTEXT(_SysCacheFlush)
diff --git a/arch/x86/bsp/crt0.s b/arch/x86/bsp/crt0.s
index 5ed3c29ce46..eda4accfc6b 100644
--- a/arch/x86/bsp/crt0.s
+++ b/arch/x86/bsp/crt0.s
@@ -44,7 +44,7 @@ been left in place should future booting scenarios arise which require its use.
 
 #define _ASMLANGUAGE
 
-#include <nanokernel/x86/asm.h>
+#include <arch/x86/asm.h>
 
 	/* exports (private APIs) */
 
diff --git a/arch/x86/bsp/driver_static_irq_stubs.s b/arch/x86/bsp/driver_static_irq_stubs.s
index 1f59ada8b67..69fde727cfd 100644
--- a/arch/x86/bsp/driver_static_irq_stubs.s
+++ b/arch/x86/bsp/driver_static_irq_stubs.s
@@ -40,7 +40,7 @@ by x86 BSPs.
 
 #ifndef CONFIG_DYNAMIC_INT_STUBS
 
-#include <nanokernel/x86/asm.h>
+#include <arch/x86/asm.h>
 #include <drivers/ioapic.h>
 #include <drivers/loapic.h>
 #include <drivers/pic.h>
diff --git a/arch/x86/bsp/i8259Boi.s b/arch/x86/bsp/i8259Boi.s
index c0762d8f104..f46a4e0b64e 100644
--- a/arch/x86/bsp/i8259Boi.s
+++ b/arch/x86/bsp/i8259Boi.s
@@ -47,8 +47,8 @@ a real IRQ has been inserted.
 /* includes */
 #define _ASMLANGUAGE
 
-#include <nanokernel/cpu.h>
-#include <nanokernel/x86/asm.h>
+#include <arch/cpu.h>
+#include <arch/x86/asm.h>
 
 #include <drivers/pic.h>
 #include <board.h>
diff --git a/arch/x86/bsp/rand32-timestamp.c b/arch/x86/bsp/rand32-timestamp.c
index a3f9927cab1..e6eaa5c6cf0 100644
--- a/arch/x86/bsp/rand32-timestamp.c
+++ b/arch/x86/bsp/rand32-timestamp.c
@@ -39,7 +39,7 @@ number generator.
 */
 
 #include <nanokernel.h>
-#include <nanokernel/cpu.h>
+#include <arch/cpu.h>
 #include <drivers/rand32.h>
 
 /*******************************************************************************
diff --git a/arch/x86/bsp/sysFatalErrorHandler.c b/arch/x86/bsp/sysFatalErrorHandler.c
index 2fe5ef844b5..6f0a34426c5 100644
--- a/arch/x86/bsp/sysFatalErrorHandler.c
+++ b/arch/x86/bsp/sysFatalErrorHandler.c
@@ -40,7 +40,7 @@ supported BSPs.
 
 #include <cputype.h>
 #include <nanokernel.h>
-#include <nanokernel/cpu.h>
+#include <arch/cpu.h>
 #include <toolchain.h>
 #include <sections.h>
 
diff --git a/arch/x86/bsp/systemApic.c b/arch/x86/bsp/systemApic.c
index da98c969c5e..e42bcef89e8 100644
--- a/arch/x86/bsp/systemApic.c
+++ b/arch/x86/bsp/systemApic.c
@@ -39,7 +39,7 @@ for the atom_n28xx variant of generic_pc BSP.
 #include <misc/__assert.h>
 #include "board.h"
 #include <nanokernel.h>
-#include <nanokernel/cpu.h>
+#include <arch/cpu.h>
 #include <drivers/ioapic.h>
 #include <drivers/loapic.h>
 
diff --git a/arch/x86/bsp/systemPic.c b/arch/x86/bsp/systemPic.c
index a9f96a9310f..58d3c418d5d 100644
--- a/arch/x86/bsp/systemPic.c
+++ b/arch/x86/bsp/systemPic.c
@@ -38,7 +38,7 @@ for the pentium4 and minuteia variants of the generic_pc BSP.
 
 #include "board.h"
 #include <nanokernel.h>
-#include <nanokernel/cpu.h>
+#include <arch/cpu.h>
 #include <drivers/pic.h>
 
 /* Handle possible stray or spurious interrupts on the master and slave PICs */
diff --git a/arch/x86/core/cpuhalt.s b/arch/x86/core/cpuhalt.s
index b78fb23df73..d68b98041be 100644
--- a/arch/x86/core/cpuhalt.s
+++ b/arch/x86/core/cpuhalt.s
@@ -51,7 +51,7 @@ by nano_cpu_atomic_idle().
 
 #define _ASMLANGUAGE
 
-#include <nanokernel/x86/asm.h>
+#include <arch/x86/asm.h>
 
 	/* exports (external APIs) */
 
diff --git a/arch/x86/core/excstub.s b/arch/x86/core/excstub.s
index dc938841f88..2bb33cd7fc5 100644
--- a/arch/x86/core/excstub.s
+++ b/arch/x86/core/excstub.s
@@ -41,8 +41,8 @@ and exiting a C exception handler.
 #define _ASMLANGUAGE
 
 #include <nanok.h>
-#include <nanokernel/x86/asm.h>
-#include <nanokernel/x86/arch.h> /* For MK_ISR_NAME */
+#include <arch/x86/asm.h>
+#include <arch/x86/arch.h> /* For MK_ISR_NAME */
 #include <offsets.h>	/* nanokernel structure offset definitions */
 
 
diff --git a/arch/x86/core/ffs.s b/arch/x86/core/ffs.s
index 9c8eb05e80f..c42338f63f3 100644
--- a/arch/x86/core/ffs.s
+++ b/arch/x86/core/ffs.s
@@ -42,7 +42,7 @@ are defined in arch.h.
 
 #define _ASMLANGUAGE
 
-#include <nanokernel/x86/asm.h>
+#include <arch/x86/asm.h>
 #include <offsets.h>	/* nanokernel structure offset definitions */
 
 
diff --git a/arch/x86/core/gdt.c b/arch/x86/core/gdt.c
index 984b16cca90..aa3550d2295 100644
--- a/arch/x86/core/gdt.c
+++ b/arch/x86/core/gdt.c
@@ -43,7 +43,7 @@ for the IA-32 architecture.
 #include <sections.h>
 
 #include <nanok.h>
-#include <nanokernel/cpu.h>
+#include <arch/cpu.h>
 #include <gdt.h>
 
 /* defines */
diff --git a/arch/x86/core/intboiexit.s b/arch/x86/core/intboiexit.s
index 17b6dc12392..931c5e031c7 100644
--- a/arch/x86/core/intboiexit.s
+++ b/arch/x86/core/intboiexit.s
@@ -52,7 +52,7 @@ _IntBoiExit() is present.
 */
 
 #define _ASMLANGUAGE
-#include <nanokernel/x86/asm.h>
+#include <arch/x86/asm.h>
 #include <offsets.h>		/* nanokernel structure offset definitions */
 
 
diff --git a/arch/x86/core/intconnect.c b/arch/x86/core/intconnect.c
index eead3b97f10..2b7f1c22e6e 100644
--- a/arch/x86/core/intconnect.c
+++ b/arch/x86/core/intconnect.c
@@ -90,7 +90,7 @@ The _INT_STUB_SIZE macro is defined in include/nanokernel/x86/arch.h.
 #ifndef CONFIG_NO_ISRS
 
 #include <nanokernel.h>
-#include <nanokernel/cpu.h>
+#include <arch/cpu.h>
 #include <nanok.h>
 
 /* the _idt_base_address symbol is generated via a linker script */
diff --git a/arch/x86/core/intstub.s b/arch/x86/core/intstub.s
index 35b9826f3b6..62985638ae8 100644
--- a/arch/x86/core/intstub.s
+++ b/arch/x86/core/intstub.s
@@ -43,9 +43,9 @@ entering and exiting a C interrupt handler.
 #ifndef CONFIG_NO_ISRS
 
 #include <nanok.h>
-#include <nanokernel/x86/asm.h>
+#include <arch/x86/asm.h>
 #include <offsets.h>	/* nanokernel structure offset definitions */
-#include <nanokernel/cpu.h>	/* _NANO_ERR_SPURIOUS_INT */
+#include <arch/cpu.h>	/* _NANO_ERR_SPURIOUS_INT */
 
 
 
diff --git a/arch/x86/core/msr.s b/arch/x86/core/msr.s
index 75f587a260f..24fb792e86d 100644
--- a/arch/x86/core/msr.s
+++ b/arch/x86/core/msr.s
@@ -38,7 +38,7 @@ utilities.
 
 #define _ASMLANGUAGE
 
-#include <nanokernel/x86/asm.h>
+#include <arch/x86/asm.h>
 
 	/* exports (internal APIs) */
 
diff --git a/arch/x86/core/nanoatomic.s b/arch/x86/core/nanoatomic.s
index ceb21039641..6194decfcf5 100644
--- a/arch/x86/core/nanoatomic.s
+++ b/arch/x86/core/nanoatomic.s
@@ -49,7 +49,7 @@ as there is no requirement for this capability.
 
 #define _ASMLANGUAGE
 
-#include <nanokernel/x86/asm.h>
+#include <arch/x86/asm.h>
 
 	/* exports (public APIs) */
 
diff --git a/arch/x86/core/nanoatomic_nolock.c b/arch/x86/core/nanoatomic_nolock.c
index fc73ed64fc1..32b50ba7629 100644
--- a/arch/x86/core/nanoatomic_nolock.c
+++ b/arch/x86/core/nanoatomic_nolock.c
@@ -50,7 +50,7 @@ as there is no requirement for this capability.
 #if defined(CONFIG_LOCK_INSTRUCTION_UNSUPPORTED)
 
 #include <nanokernel.h>
-#include <nanokernel/cpu.h>
+#include <arch/cpu.h>
 
 /*******************************************************************************
 *
diff --git a/arch/x86/core/nanofatal.c b/arch/x86/core/nanofatal.c
index bb2d210edd0..d53bb11536e 100644
--- a/arch/x86/core/nanofatal.c
+++ b/arch/x86/core/nanofatal.c
@@ -41,7 +41,7 @@ This module provides the _NanoFatalErrorHandler() routine.
 
 #include <cputype.h>
 #include <nanokernel.h>
-#include <nanokernel/cpu.h>
+#include <arch/cpu.h>
 #include <nanok.h>
 #include <misc/printk.h>
 
diff --git a/arch/x86/core/swap.s b/arch/x86/core/swap.s
index 629bde0dc22..0f59e2f480e 100644
--- a/arch/x86/core/swap.s
+++ b/arch/x86/core/swap.s
@@ -44,7 +44,7 @@ save frame on the stack.
 #define _ASMLANGUAGE
 
 #include <nanok.h>
-#include <nanokernel/x86/asm.h>
+#include <arch/x86/asm.h>
 #include <offsets.h>	/* nanokernel structure offset definitions */
 
 	/* exports (internal APIs) */
diff --git a/arch/x86/core/unaligned.s b/arch/x86/core/unaligned.s
index cc6bcec924c..98dc2edf7b8 100644
--- a/arch/x86/core/unaligned.s
+++ b/arch/x86/core/unaligned.s
@@ -42,7 +42,7 @@ performed byte-by-byte.
 
 #ifdef CONFIG_UNALIGNED_WRITE_UNSUPPORTED
 
-#include <nanokernel/x86/asm.h>
+#include <arch/x86/asm.h>
 
 	/* exports (internal APIs) */
 
diff --git a/arch/x86/generic_pc/linker.cmd b/arch/x86/generic_pc/linker.cmd
index dd0b0ac305f..d6f567899e9 100644
--- a/arch/x86/generic_pc/linker.cmd
+++ b/arch/x86/generic_pc/linker.cmd
@@ -65,9 +65,9 @@ MEMORY
     IDT_LIST        : ORIGIN = 2K, LENGTH = 2K
     }
 
-#include <nanokernel/x86/linker-common-sections.h>
+#include <arch/x86/linker-common-sections.h>
 
 /* start adding bsp specific linker sections here */
 
 /* no sections should appear after linker-epilog.h */
-#include <nanokernel/x86/linker-epilog.h>
+#include <arch/x86/linker-epilog.h>
diff --git a/arch/x86/generic_pc/system.c b/arch/x86/generic_pc/system.c
index b671fe4b85c..e41e6925c12 100644
--- a/arch/x86/generic_pc/system.c
+++ b/arch/x86/generic_pc/system.c
@@ -38,7 +38,7 @@ for the generic_pc BSP.
 
 #include <cputype.h>
 #include <nanokernel.h>
-#include <nanokernel/cpu.h>
+#include <arch/cpu.h>
 #include "board.h"
 #include <drivers/uart.h>
 #include <drivers/pic.h>
diff --git a/arch/x86/include/asmPrv.h b/arch/x86/include/asmPrv.h
index bdbdb1ee0ca..c425a42b61d 100644
--- a/arch/x86/include/asmPrv.h
+++ b/arch/x86/include/asmPrv.h
@@ -33,7 +33,7 @@
 #ifndef __INCsysX86AsmPrvh
 #define __INCsysX86AsmPrvh
 
-#include <nanokernel/x86/asm.h>
+#include <arch/x86/asm.h>
 
 #ifdef __cplusplus
 extern "C" {
diff --git a/arch/x86/include/gdt.h b/arch/x86/include/gdt.h
index 2ab341e3fb0..d231c76d240 100644
--- a/arch/x86/include/gdt.h
+++ b/arch/x86/include/gdt.h
@@ -41,7 +41,7 @@ IA-32 architecture.
 
 /* includes */
 
-#include <nanokernel/x86/arch.h>
+#include <arch/x86/arch.h>
 
 #ifndef _ASMLANGUAGE
 
diff --git a/arch/x86/include/nanok.h b/arch/x86/include/nanok.h
index 97345bbf61e..8f7c9f991be 100644
--- a/arch/x86/include/nanok.h
+++ b/arch/x86/include/nanok.h
@@ -55,7 +55,7 @@ offsets.o module.
 
 #ifndef _ASMLANGUAGE
 #include <nanokernel.h>		   /* public nanokernel API */
-#include <nanokernel/x86/arch.h> /* private IA-32 definitions */
+#include <arch/x86/arch.h> /* private IA-32 definitions */
 #include <stdint.h>
 /*
  * This pulls in the code shared with the IDT generator that actually
diff --git a/arch/x86/interrupt_controller/ioApicIntr.c b/arch/x86/interrupt_controller/ioApicIntr.c
index 1b93a9c3784..1c2052c4516 100644
--- a/arch/x86/interrupt_controller/ioApicIntr.c
+++ b/arch/x86/interrupt_controller/ioApicIntr.c
@@ -78,7 +78,7 @@ SEE ALSO: loApicIntr.c
 /* includes */
 
 #include <nanokernel.h>
-#include <nanokernel/cpu.h>
+#include <arch/cpu.h>
 
 #include "board.h"
 
diff --git a/arch/x86/interrupt_controller/loApicIntr.c b/arch/x86/interrupt_controller/loApicIntr.c
index c2e6b4b63fd..c80d60a1153 100644
--- a/arch/x86/interrupt_controller/loApicIntr.c
+++ b/arch/x86/interrupt_controller/loApicIntr.c
@@ -106,7 +106,7 @@ INCLUDE FILES: loapic.h
 /* includes */
 
 #include <nanokernel.h>
-#include <nanokernel/cpu.h>
+#include <arch/cpu.h>
 #include <stdint.h>
 
 #include "board.h"
diff --git a/arch/x86/quark/linker.cmd b/arch/x86/quark/linker.cmd
index dd0b0ac305f..d6f567899e9 100644
--- a/arch/x86/quark/linker.cmd
+++ b/arch/x86/quark/linker.cmd
@@ -65,9 +65,9 @@ MEMORY
     IDT_LIST        : ORIGIN = 2K, LENGTH = 2K
     }
 
-#include <nanokernel/x86/linker-common-sections.h>
+#include <arch/x86/linker-common-sections.h>
 
 /* start adding bsp specific linker sections here */
 
 /* no sections should appear after linker-epilog.h */
-#include <nanokernel/x86/linker-epilog.h>
+#include <arch/x86/linker-epilog.h>
diff --git a/arch/x86/quark/system.c b/arch/x86/quark/system.c
index d49ec4c1ce8..280798a1969 100644
--- a/arch/x86/quark/system.c
+++ b/arch/x86/quark/system.c
@@ -41,7 +41,7 @@ Handlers for the secondary serial port have not been added.
 
 #include <cputype.h>
 #include <nanokernel.h>
-#include <nanokernel/cpu.h>
+#include <arch/cpu.h>
 #include <misc/printk.h>
 #include <misc/__assert.h>
 #include "board.h"
diff --git a/arch/x86/timer/hpet.c b/arch/x86/timer/hpet.c
index acad0cd936a..31e7646ca98 100644
--- a/arch/x86/timer/hpet.c
+++ b/arch/x86/timer/hpet.c
@@ -65,7 +65,7 @@ In a nanokernel-only system this device driver omits more complex capabilities
 */
 
 #include <nanokernel.h>
-#include <nanokernel/cpu.h>
+#include <arch/cpu.h>
 #include <toolchain.h>
 #include <sections.h>
 #include <clock_vars.h>
diff --git a/arch/x86/timer/loApicTimer.c b/arch/x86/timer/loApicTimer.c
index 476b9400a81..5d18d7eb376 100644
--- a/arch/x86/timer/loApicTimer.c
+++ b/arch/x86/timer/loApicTimer.c
@@ -46,7 +46,7 @@ After reset, the timer is initialized to zero.
 /* includes */
 
 #include <nanokernel.h>
-#include <nanokernel/cpu.h>
+#include <arch/cpu.h>
 #include <toolchain.h>
 #include <sections.h>
 #include <clock_vars.h>
diff --git a/drivers/bluetooth/uart.c b/drivers/bluetooth/uart.c
index f3a11e2b63c..1236b304b2c 100644
--- a/drivers/bluetooth/uart.c
+++ b/drivers/bluetooth/uart.c
@@ -34,7 +34,7 @@
 #include <stddef.h>
 
 #include <nanokernel.h>
-#include <nanokernel/cpu.h>
+#include <arch/cpu.h>
 
 #include <board.h>
 #include <drivers/uart.h>
diff --git a/drivers/console/uart_console.c b/drivers/console/uart_console.c
index ab7a285379b..9243c1e676f 100644
--- a/drivers/console/uart_console.c
+++ b/drivers/console/uart_console.c
@@ -38,7 +38,7 @@
 */
 
 #include <nanokernel.h>
-#include <nanokernel/cpu.h>
+#include <arch/cpu.h>
 
 #include <stdio.h>
 #include <stdint.h>
diff --git a/drivers/interrupt_controller/i8259.c b/drivers/interrupt_controller/i8259.c
index 6b624f73c47..2f9857a5b72 100644
--- a/drivers/interrupt_controller/i8259.c
+++ b/drivers/interrupt_controller/i8259.c
@@ -86,7 +86,7 @@ the last level acknowledged and serviced.
 /* includes */
 
 #include <nanokernel.h>
-#include <nanokernel/cpu.h>
+#include <arch/cpu.h>
 #include <toolchain.h>
 #include <sections.h>
 
diff --git a/drivers/pci/pci.c b/drivers/pci/pci.c
index 0427727816d..d86e7e3c210 100644
--- a/drivers/pci/pci.c
+++ b/drivers/pci/pci.c
@@ -116,7 +116,7 @@ restarting from 0.
 */
 
 #include <nanokernel.h>
-#include <nanokernel/cpu.h>
+#include <arch/cpu.h>
 #include <misc/printk.h>
 #include <toolchain.h>
 #include <sections.h>
diff --git a/drivers/pci/pci_config.c b/drivers/pci/pci_config.c
index f07e09a593c..35b6a23e37c 100644
--- a/drivers/pci/pci_config.c
+++ b/drivers/pci/pci_config.c
@@ -40,7 +40,7 @@ This module implements the PCI config space access functions
 /* includes */
 
 #include <nanokernel.h>
-#include <nanokernel/cpu.h>
+#include <arch/cpu.h>
 
 #include <pci/pci_mgr.h>
 #include <string.h>
diff --git a/drivers/pci/pci_interface.c b/drivers/pci/pci_interface.c
index e83866c17ef..a69d72d1f25 100644
--- a/drivers/pci/pci_interface.c
+++ b/drivers/pci/pci_interface.c
@@ -40,7 +40,7 @@ This module implements the PCI H/W access functions.
 /* includes */
 
 #include <nanokernel.h>
-#include <nanokernel/cpu.h>
+#include <arch/cpu.h>
 
 #include <pci/pci_mgr.h>
 #include <string.h>
diff --git a/drivers/serial/k20UartDrv.c b/drivers/serial/k20UartDrv.c
index 834508a0f03..47fb2dd14e8 100644
--- a/drivers/serial/k20UartDrv.c
+++ b/drivers/serial/k20UartDrv.c
@@ -43,7 +43,7 @@ INCLUDE FILES: drivers/serial/k20_uart.h
 /* includes */
 
 #include <nanokernel.h>
-#include <nanokernel/cpu.h>
+#include <arch/cpu.h>
 #include <stdint.h>
 
 #include <board.h>
diff --git a/drivers/serial/ns16550.c b/drivers/serial/ns16550.c
index faa5e3f0c52..21da327ee19 100644
--- a/drivers/serial/ns16550.c
+++ b/drivers/serial/ns16550.c
@@ -59,7 +59,7 @@ INCLUDE FILES: drivers/uart.h
 /* includes */
 
 #include <nanokernel.h>
-#include <nanokernel/cpu.h>
+#include <arch/cpu.h>
 #include <stdint.h>
 
 #include <board.h>
diff --git a/drivers/serial/stellarisUartDrv.c b/drivers/serial/stellarisUartDrv.c
index 51bf877a953..75cee5f1e0c 100644
--- a/drivers/serial/stellarisUartDrv.c
+++ b/drivers/serial/stellarisUartDrv.c
@@ -40,7 +40,7 @@ and STDOUT_CONSOLE APIs.
 */
 
 #include <nanokernel.h>
-#include <nanokernel/cpu.h>
+#include <arch/cpu.h>
 #include <misc/__assert.h>
 #include <board.h>
 #include <drivers/uart.h>
diff --git a/drivers/timer/i8253.c b/drivers/timer/i8253.c
index c66b85a6eee..3d3055c2459 100644
--- a/drivers/timer/i8253.c
+++ b/drivers/timer/i8253.c
@@ -52,7 +52,7 @@ directly invoke the VIOAPIC APIs to configure/unmask the IRQ.
 /* includes */
 
 #include <nanokernel.h>
-#include <nanokernel/cpu.h>
+#include <arch/cpu.h>
 #include <toolchain.h>
 #include <sections.h>
 #include <limits.h>
diff --git a/include/nanokernel/arc/arch.h b/include/arch/arc/arch.h
similarity index 87%
rename from include/nanokernel/arc/arch.h
rename to include/arch/arc/arch.h
index a80e30c2186..03bb3964a92 100644
--- a/include/nanokernel/arc/arch.h
+++ b/include/arch/arc/arch.h
@@ -50,14 +50,14 @@ extern "C" {
 #endif
 
 #ifdef CONFIG_CPU_ARCV2
-#include <nanokernel/arc/v2/init.h>
-#include <nanokernel/arc/v2/exc.h>
-#include <nanokernel/arc/v2/irq.h>
-#include <nanokernel/arc/v2/ffs.h>
-#include <nanokernel/arc/v2/error.h>
-#include <nanokernel/arc/v2/misc.h>
-#include <nanokernel/arc/v2/aux_regs.h>
-#include <nanokernel/arc/v2/arcv2_irq_unit.h>
+#include <arch/arc/v2/init.h>
+#include <arch/arc/v2/exc.h>
+#include <arch/arc/v2/irq.h>
+#include <arch/arc/v2/ffs.h>
+#include <arch/arc/v2/error.h>
+#include <arch/arc/v2/misc.h>
+#include <arch/arc/v2/aux_regs.h>
+#include <arch/arc/v2/arcv2_irq_unit.h>
 #endif
 
 #ifdef __cplusplus
diff --git a/include/nanokernel/arc/arcprc.h b/include/arch/arc/arcprc.h
similarity index 100%
rename from include/nanokernel/arc/arcprc.h
rename to include/arch/arc/arcprc.h
diff --git a/include/nanokernel/arc/v2/arcv2_irq_unit.h b/include/arch/arc/v2/arcv2_irq_unit.h
similarity index 100%
rename from include/nanokernel/arc/v2/arcv2_irq_unit.h
rename to include/arch/arc/v2/arcv2_irq_unit.h
diff --git a/include/nanokernel/arc/v2/aux_regs.h b/include/arch/arc/v2/aux_regs.h
similarity index 100%
rename from include/nanokernel/arc/v2/aux_regs.h
rename to include/arch/arc/v2/aux_regs.h
diff --git a/include/nanokernel/arc/v2/error.h b/include/arch/arc/v2/error.h
similarity index 98%
rename from include/nanokernel/arc/v2/error.h
rename to include/arch/arc/v2/error.h
index 62c95181d14..b7265c8ec58 100644
--- a/include/nanokernel/arc/v2/error.h
+++ b/include/arch/arc/v2/error.h
@@ -38,7 +38,7 @@ ARC-specific nanokernel error handling interface. Included by ARC/arch.h.
 #ifndef _ARCH_ARC_V2_ERROR_H_
 #define _ARCH_ARC_V2_ERROR_H_
 
-#include <nanokernel/arc/v2/exc.h>
+#include <arch/arc/v2/exc.h>
 
 #ifndef _ASMLANGUAGE
 extern FUNC_NORETURN void _NanoFatalErrorHandler(unsigned int,
diff --git a/include/nanokernel/arc/v2/exc.h b/include/arch/arc/v2/exc.h
similarity index 100%
rename from include/nanokernel/arc/v2/exc.h
rename to include/arch/arc/v2/exc.h
diff --git a/include/nanokernel/arc/v2/ffs.h b/include/arch/arc/v2/ffs.h
similarity index 100%
rename from include/nanokernel/arc/v2/ffs.h
rename to include/arch/arc/v2/ffs.h
diff --git a/include/nanokernel/arc/v2/init.h b/include/arch/arc/v2/init.h
similarity index 100%
rename from include/nanokernel/arc/v2/init.h
rename to include/arch/arc/v2/init.h
diff --git a/include/nanokernel/arc/v2/irq.h b/include/arch/arc/v2/irq.h
similarity index 98%
rename from include/nanokernel/arc/v2/irq.h
rename to include/arch/arc/v2/irq.h
index 70954b50d7c..f2ecb156e1f 100644
--- a/include/nanokernel/arc/v2/irq.h
+++ b/include/arch/arc/v2/irq.h
@@ -38,7 +38,7 @@
 #ifndef _ARCH_ARC_V2_IRQ__H_
 #define _ARCH_ARC_V2_IRQ__H_
 
-#include <nanokernel/arc/v2/aux_regs.h>
+#include <arch/arc/v2/aux_regs.h>
 
 #ifdef _ASMLANGUAGE
 GTEXT(_irq_exit);
diff --git a/include/nanokernel/arc/v2/misc.h b/include/arch/arc/v2/misc.h
similarity index 100%
rename from include/nanokernel/arc/v2/misc.h
rename to include/arch/arc/v2/misc.h
diff --git a/include/nanokernel/arm/ARMprc.h b/include/arch/arm/ARMprc.h
similarity index 100%
rename from include/nanokernel/arm/ARMprc.h
rename to include/arch/arm/ARMprc.h
diff --git a/include/nanokernel/arm/CortexM/asm_inline.h b/include/arch/arm/CortexM/asm_inline.h
similarity index 94%
rename from include/nanokernel/arm/CortexM/asm_inline.h
rename to include/arch/arm/CortexM/asm_inline.h
index 21bcc7113ac..4d38ae22faa 100644
--- a/include/nanokernel/arm/CortexM/asm_inline.h
+++ b/include/arch/arm/CortexM/asm_inline.h
@@ -39,9 +39,9 @@
  */
 
 #if defined(__GNUC__)
-#include <nanokernel/arm/CortexM/asm_inline_gcc.h>
+#include <arch/arm/CortexM/asm_inline_gcc.h>
 #else
-#include <nanokernel/arm/CortexM/asm_inline_other.h>
+#include <arch/arm/CortexM/asm_inline_other.h>
 #endif
 
 #endif /* _ASM_INLINE_PUBLIC_H */
diff --git a/include/nanokernel/arm/CortexM/asm_inline_gcc.h b/include/arch/arm/CortexM/asm_inline_gcc.h
similarity index 99%
rename from include/nanokernel/arm/CortexM/asm_inline_gcc.h
rename to include/arch/arm/CortexM/asm_inline_gcc.h
index 885604d2bd8..d03ddf777a8 100644
--- a/include/nanokernel/arm/CortexM/asm_inline_gcc.h
+++ b/include/arch/arm/CortexM/asm_inline_gcc.h
@@ -50,7 +50,7 @@
 
 #else /* !_ASMLANGUAGE */
 #include <stdint.h>
-#include <nanokernel/arm/CortexM/nvic.h>
+#include <arch/arm/CortexM/nvic.h>
 
 /*******************************************************************************
 *
diff --git a/include/nanokernel/arm/CortexM/error.h b/include/arch/arm/CortexM/error.h
similarity index 98%
rename from include/nanokernel/arm/CortexM/error.h
rename to include/arch/arm/CortexM/error.h
index a21768fe075..d79743ea870 100644
--- a/include/nanokernel/arm/CortexM/error.h
+++ b/include/arch/arm/CortexM/error.h
@@ -38,7 +38,7 @@ ARM-specific nanokernel error handling interface. Included by ARM/arch.h.
 #ifndef _ARCH_ARM_CORTEXM_ERROR_H_
 #define _ARCH_ARM_CORTEXM_ERROR_H_
 
-#include <nanokernel/arm/CortexM/exc.h>
+#include <arch/arm/CortexM/exc.h>
 
 #ifndef _ASMLANGUAGE
 extern FUNC_NORETURN void _NanoFatalErrorHandler(unsigned int,
diff --git a/include/nanokernel/arm/CortexM/exc.h b/include/arch/arm/CortexM/exc.h
similarity index 100%
rename from include/nanokernel/arm/CortexM/exc.h
rename to include/arch/arm/CortexM/exc.h
diff --git a/include/nanokernel/arm/CortexM/ffs.h b/include/arch/arm/CortexM/ffs.h
similarity index 100%
rename from include/nanokernel/arm/CortexM/ffs.h
rename to include/arch/arm/CortexM/ffs.h
diff --git a/include/nanokernel/arm/CortexM/gdb_stub.h b/include/arch/arm/CortexM/gdb_stub.h
similarity index 100%
rename from include/nanokernel/arm/CortexM/gdb_stub.h
rename to include/arch/arm/CortexM/gdb_stub.h
diff --git a/include/nanokernel/arm/CortexM/init.h b/include/arch/arm/CortexM/init.h
similarity index 100%
rename from include/nanokernel/arm/CortexM/init.h
rename to include/arch/arm/CortexM/init.h
diff --git a/include/nanokernel/arm/CortexM/irq.h b/include/arch/arm/CortexM/irq.h
similarity index 98%
rename from include/nanokernel/arm/CortexM/irq.h
rename to include/arch/arm/CortexM/irq.h
index 33ea1bea042..99693efd2b4 100644
--- a/include/nanokernel/arm/CortexM/irq.h
+++ b/include/arch/arm/CortexM/irq.h
@@ -38,7 +38,7 @@ ARM-specific nanokernel interrupt handling interface. Included by ARM/arch.h.
 #ifndef _ARCH_ARM_CORTEXM_IRQ_H_
 #define _ARCH_ARM_CORTEXM_IRQ_H_
 
-#include <nanokernel/arm/CortexM/nvic.h>
+#include <arch/arm/CortexM/nvic.h>
 
 #ifdef _ASMLANGUAGE
 GTEXT(_IntExit);
diff --git a/include/nanokernel/arm/CortexM/memory_map-m0.h b/include/arch/arm/CortexM/memory_map-m0.h
similarity index 100%
rename from include/nanokernel/arm/CortexM/memory_map-m0.h
rename to include/arch/arm/CortexM/memory_map-m0.h
diff --git a/include/nanokernel/arm/CortexM/memory_map-m3.h b/include/arch/arm/CortexM/memory_map-m3.h
similarity index 100%
rename from include/nanokernel/arm/CortexM/memory_map-m3.h
rename to include/arch/arm/CortexM/memory_map-m3.h
diff --git a/include/nanokernel/arm/CortexM/memory_map.h b/include/arch/arm/CortexM/memory_map.h
similarity index 97%
rename from include/nanokernel/arm/CortexM/memory_map.h
rename to include/arch/arm/CortexM/memory_map.h
index 68816af9f73..213cee6f794 100644
--- a/include/nanokernel/arm/CortexM/memory_map.h
+++ b/include/arch/arm/CortexM/memory_map.h
@@ -73,9 +73,9 @@ processors.
 /* 0xe0000000 -> 0xffffffff is different between M3 and M0 */
 
 #if defined(CONFIG_CPU_CORTEXM3)
-#include <nanokernel/arm/CortexM/memory_map-m3.h>
+#include <arch/arm/CortexM/memory_map-m3.h>
 #elif defined(CONFIG_CPU_CORTEXM0)
-#include <nanokernel/arm/CortexM/memory_map-m0.h>
+#include <arch/arm/CortexM/memory_map-m0.h>
 #else
 #error Unknown CPU
 #endif
diff --git a/include/nanokernel/arm/CortexM/misc.h b/include/arch/arm/CortexM/misc.h
similarity index 100%
rename from include/nanokernel/arm/CortexM/misc.h
rename to include/arch/arm/CortexM/misc.h
diff --git a/include/nanokernel/arm/CortexM/nvic.h b/include/arch/arm/CortexM/nvic.h
similarity index 99%
rename from include/nanokernel/arm/CortexM/nvic.h
rename to include/arch/arm/CortexM/nvic.h
index 4127c99c4e1..86cabb727f1 100644
--- a/include/nanokernel/arm/CortexM/nvic.h
+++ b/include/arch/arm/CortexM/nvic.h
@@ -52,7 +52,7 @@ Supports up to 240 IRQs and 256 priority levels.
 #define _NVIC_H_
 
 #include <misc/__assert.h>
-#include <nanokernel/arm/CortexM/scs.h>
+#include <arch/arm/CortexM/scs.h>
 
 /* for assembler, only works with constants */
 #define _EXC_PRIO(pri) (((pri) << (8 - CONFIG_NUM_IRQ_PRIO_BITS)) & 0xff)
diff --git a/include/nanokernel/arm/CortexM/scb.h b/include/arch/arm/CortexM/scb.h
similarity index 99%
rename from include/nanokernel/arm/CortexM/scb.h
rename to include/arch/arm/CortexM/scb.h
index e5c1ef6b567..7268f4015c7 100644
--- a/include/nanokernel/arm/CortexM/scb.h
+++ b/include/arch/arm/CortexM/scb.h
@@ -64,9 +64,9 @@ registers is the way to implement it.
 #else
 
 #include <nanokernel.h>
-#include <nanokernel/cpu.h>
+#include <arch/cpu.h>
 #include <misc/__assert.h>
-#include <nanokernel/arm/CortexM/scs.h>
+#include <arch/arm/CortexM/scs.h>
 #include <misc/util.h>
 #include <stdint.h>
 
diff --git a/include/nanokernel/arm/CortexM/scripts/linker.cmd b/include/arch/arm/CortexM/scripts/linker.cmd
similarity index 100%
rename from include/nanokernel/arm/CortexM/scripts/linker.cmd
rename to include/arch/arm/CortexM/scripts/linker.cmd
diff --git a/include/nanokernel/arm/CortexM/scs.h b/include/arch/arm/CortexM/scs.h
similarity index 99%
rename from include/nanokernel/arm/CortexM/scs.h
rename to include/arch/arm/CortexM/scs.h
index ac92f1fba4d..a37bcea84ab 100644
--- a/include/nanokernel/arm/CortexM/scs.h
+++ b/include/arch/arm/CortexM/scs.h
@@ -73,12 +73,12 @@ still considered part of the NVIC and an API for it is provided in nvic.h.
 
 #ifdef _ASMLANGUAGE
 
-#include <nanokernel/arm/CortexM/memory_map.h>
+#include <arch/arm/CortexM/memory_map.h>
 
 #else /* _ASMLANGUAGE */
 
 #include <nanokernel.h>
-#include <nanokernel/cpu.h>
+#include <arch/cpu.h>
 #include <stdint.h>
 
 union __ictr {
diff --git a/include/nanokernel/arm/arch.h b/include/arch/arm/arch.h
similarity index 80%
rename from include/nanokernel/arm/arch.h
rename to include/arch/arm/arch.h
index 7a6d5d98bd9..5e9b7b0ce51 100644
--- a/include/nanokernel/arm/arch.h
+++ b/include/arch/arm/arch.h
@@ -50,18 +50,18 @@ extern "C" {
 #endif
 
 #ifdef CONFIG_CPU_CORTEXM
-#include <nanokernel/arm/CortexM/init.h>
-#include <nanokernel/arm/CortexM/exc.h>
-#include <nanokernel/arm/CortexM/irq.h>
-#include <nanokernel/arm/CortexM/ffs.h>
-#include <nanokernel/arm/CortexM/error.h>
-#include <nanokernel/arm/CortexM/misc.h>
-#include <nanokernel/arm/CortexM/scs.h>
-#include <nanokernel/arm/CortexM/scb.h>
-#include <nanokernel/arm/CortexM/nvic.h>
-#include <nanokernel/arm/CortexM/memory_map.h>
-#include <nanokernel/arm/CortexM/gdb_stub.h>
-#include <nanokernel/arm/CortexM/asm_inline.h>
+#include <arch/arm/CortexM/init.h>
+#include <arch/arm/CortexM/exc.h>
+#include <arch/arm/CortexM/irq.h>
+#include <arch/arm/CortexM/ffs.h>
+#include <arch/arm/CortexM/error.h>
+#include <arch/arm/CortexM/misc.h>
+#include <arch/arm/CortexM/scs.h>
+#include <arch/arm/CortexM/scb.h>
+#include <arch/arm/CortexM/nvic.h>
+#include <arch/arm/CortexM/memory_map.h>
+#include <arch/arm/CortexM/gdb_stub.h>
+#include <arch/arm/CortexM/asm_inline.h>
 #endif
 
 #ifdef __cplusplus
diff --git a/include/nanokernel/cpu.h b/include/arch/cpu.h
similarity index 90%
rename from include/nanokernel/cpu.h
rename to include/arch/cpu.h
index 43487df4560..46dec21832e 100644
--- a/include/nanokernel/cpu.h
+++ b/include/arch/cpu.h
@@ -30,17 +30,17 @@
  * POSSIBILITY OF SUCH DAMAGE.
  */
 
-#ifndef __NANOKERNELCPU_H__
-#define __NANOKERNELCPU_H__
+#ifndef __ARCHCPU_H__
+#define __ARCHCPU_H__
 
 #if defined(VXMICRO_ARCH_x86)
-#include <nanokernel/x86/arch.h>
+#include <arch/x86/arch.h>
 #elif defined(VXMICRO_ARCH_arm)
-#include <nanokernel/arm/arch.h>
+#include <arch/arm/arch.h>
 #elif defined(VXMICRO_ARCH_arc)
-#include <nanokernel/arc/arch.h>
+#include <arch/arc/arch.h>
 #else
 #error "Unknown VXMICRO_ARCH"
 #endif
 
-#endif /* __NANOKERNELCPU_H__ */
+#endif /* __ARCHCPU_H__ */
diff --git a/include/nanokernel/private.h b/include/arch/private.h
similarity index 100%
rename from include/nanokernel/private.h
rename to include/arch/private.h
diff --git a/include/nanokernel/x86/Intelprc.h b/include/arch/x86/Intelprc.h
similarity index 96%
rename from include/nanokernel/x86/Intelprc.h
rename to include/arch/x86/Intelprc.h
index 7e47a4c9267..67886421ed9 100644
--- a/include/nanokernel/x86/Intelprc.h
+++ b/include/arch/x86/Intelprc.h
@@ -52,8 +52,8 @@ extern "C"
 #include <stdint.h>
 #include <toolchain.h>
 #include <misc/util.h>
-#include <nanokernel/x86/addr_types.h>
-#include <nanokernel/x86/asm_inline.h>
+#include <arch/x86/addr_types.h>
+#include <arch/x86/asm_inline.h>
 #include <drivers/system_timer.h> /* timer_driver() needed by kernel_main.c */
 
 /*
diff --git a/include/nanokernel/x86/addr_types.h b/include/arch/x86/addr_types.h
similarity index 100%
rename from include/nanokernel/x86/addr_types.h
rename to include/arch/x86/addr_types.h
diff --git a/include/nanokernel/x86/arch.h b/include/arch/x86/arch.h
similarity index 99%
rename from include/nanokernel/x86/arch.h
rename to include/arch/x86/arch.h
index 4e927fbb415..89a4715468f 100644
--- a/include/nanokernel/x86/arch.h
+++ b/include/arch/x86/arch.h
@@ -42,7 +42,7 @@ by the generic nanokernel interface header (nanokernel.h)
 /* WARNING: must include nanokernel.h before this file */
 
 #ifndef _ASMLANGUAGE
-#include <nanokernel/x86/Intelprc.h>
+#include <arch/x86/Intelprc.h>
 #endif
 
 /*
diff --git a/include/nanokernel/x86/asm.h b/include/arch/x86/asm.h
similarity index 100%
rename from include/nanokernel/x86/asm.h
rename to include/arch/x86/asm.h
diff --git a/include/nanokernel/x86/asm_inline.h b/include/arch/x86/asm_inline.h
similarity index 95%
rename from include/nanokernel/x86/asm_inline.h
rename to include/arch/x86/asm_inline.h
index 22a3303c715..f31129eedf0 100644
--- a/include/nanokernel/x86/asm_inline.h
+++ b/include/arch/x86/asm_inline.h
@@ -39,9 +39,9 @@
  */
 
 #if defined(__GNUC__)
-#include <nanokernel/x86/asm_inline_gcc.h>
+#include <arch/x86/asm_inline_gcc.h>
 #else
-#include <nanokernel/x86/asm_inline_other.h>
+#include <arch/x86/asm_inline_other.h>
 #endif
 
 #endif /* _ASM_INLINE_PUBLIC_H */
diff --git a/include/nanokernel/x86/asm_inline_gcc.h b/include/arch/x86/asm_inline_gcc.h
similarity index 100%
rename from include/nanokernel/x86/asm_inline_gcc.h
rename to include/arch/x86/asm_inline_gcc.h
diff --git a/include/nanokernel/x86/linker-common-sections.h b/include/arch/x86/linker-common-sections.h
similarity index 100%
rename from include/nanokernel/x86/linker-common-sections.h
rename to include/arch/x86/linker-common-sections.h
diff --git a/include/nanokernel/x86/linker-defs-arch.h b/include/arch/x86/linker-defs-arch.h
similarity index 100%
rename from include/nanokernel/x86/linker-defs-arch.h
rename to include/arch/x86/linker-defs-arch.h
diff --git a/include/nanokernel/x86/linker-epilog.h b/include/arch/x86/linker-epilog.h
similarity index 100%
rename from include/nanokernel/x86/linker-epilog.h
rename to include/arch/x86/linker-epilog.h
diff --git a/include/nanokernel/x86/segselect.h b/include/arch/x86/segselect.h
similarity index 100%
rename from include/nanokernel/x86/segselect.h
rename to include/arch/x86/segselect.h
diff --git a/include/cputype.h b/include/cputype.h
index fb7a5058a62..1a507ba43c2 100644
--- a/include/cputype.h
+++ b/include/cputype.h
@@ -37,11 +37,11 @@
 #define _CPUTYPE_H
 
 #if defined(VXMICRO_ARCH_x86)
-#include <nanokernel/x86/Intelprc.h>
+#include <arch/x86/Intelprc.h>
 #elif defined(VXMICRO_ARCH_arm)
-#include <nanokernel/arm/ARMprc.h>
+#include <arch/arm/ARMprc.h>
 #elif defined(VXMICRO_ARCH_arc)
-#include <nanokernel/arc/arcprc.h>
+#include <arch/arc/arcprc.h>
 #else
 #error "Unknown VXMICRO_ARCH type"
 #endif
diff --git a/include/linker-defs.h b/include/linker-defs.h
index d8108c8f932..d2e8e207770 100644
--- a/include/linker-defs.h
+++ b/include/linker-defs.h
@@ -46,7 +46,7 @@ This file may be included by:
 
 /* include platform dependent linker-defs */
 #ifdef VXMICRO_ARCH_x86
-#include <nanokernel/x86/linker-defs-arch.h>
+#include <arch/x86/linker-defs-arch.h>
 #elif defined(VXMICRO_ARCH_arm)
 /* Nothing yet to include */
 #elif defined(VXMICRO_ARCH_arc)
diff --git a/include/nanokernel.h b/include/nanokernel.h
index 0fec5f3f2ee..7ab13fbaad5 100644
--- a/include/nanokernel.h
+++ b/include/nanokernel.h
@@ -36,7 +36,7 @@
 #include <stdint.h>
 #include <toolchain.h> /* compiler specific configuration options */
 
-#include <nanokernel/private.h>
+#include <arch/private.h>
 #include <kernel_version.h>
 #include <clock_vars.h>
 #include <drivers/rand32.h>
diff --git a/kernel/microkernel/cmdPkt.c b/kernel/microkernel/cmdPkt.c
index 8d634509f49..8a60e7fbf31 100644
--- a/kernel/microkernel/cmdPkt.c
+++ b/kernel/microkernel/cmdPkt.c
@@ -55,7 +55,7 @@ that have an ISR component should use their own command packet set.
 /* includes */
 
 #include <nanokernel.h>
-#include <nanokernel/cpu.h>
+#include <arch/cpu.h>
 #include <microkernel/cmdPkt.h>
 #include <minik.h>
 #include <sections.h>
diff --git a/kernel/microkernel/global.c b/kernel/microkernel/global.c
index 339487fde67..26030859e7f 100644
--- a/kernel/microkernel/global.c
+++ b/kernel/microkernel/global.c
@@ -37,7 +37,7 @@
 #include <sections.h>
 
 #ifdef CONFIG_BOOT_TIME_MEASUREMENT
-#include <nanokernel/cpu.h>
+#include <arch/cpu.h>
 #endif
 
 extern void init_node(void);        /* defined by sysgen */
diff --git a/kernel/microkernel/idle.c b/kernel/microkernel/idle.c
index 624b7410cc9..68752e692c7 100644
--- a/kernel/microkernel/idle.c
+++ b/kernel/microkernel/idle.c
@@ -38,7 +38,7 @@ task, depending on how the kernel is configured.
 
 #include <minik.h>
 #include <nanok.h>
-#include <nanokernel/cpu.h>
+#include <arch/cpu.h>
 #include <toolchain.h>
 #include <sections.h>
 #include <microkernel.h>
@@ -283,7 +283,7 @@ unsigned char _sys_power_save_flag = 1;
 
 #include <cputype.h>
 #include <nanokernel.h>
-#include <nanokernel/cpu.h>
+#include <arch/cpu.h>
 #include <microkernel/k_types.h>
 #ifdef CONFIG_ADVANCED_IDLE
 #include <advidle.h>
diff --git a/kernel/microkernel/k_irq.c b/kernel/microkernel/k_irq.c
index e18e1cc0709..f18148cc193 100644
--- a/kernel/microkernel/k_irq.c
+++ b/kernel/microkernel/k_irq.c
@@ -98,7 +98,7 @@ NANO_CPU_INT_STUB_DECL(irq_obj_mem_stub[MAX_TASK_IRQS]);
 #define RELEASE_VECTOR(v) _IntVecMarkFree(v)
 
 #elif defined(CONFIG_CPU_CORTEXM3)
-#include <nanokernel/cpu.h>
+#include <arch/cpu.h>
 #define IRQ_STUB
 #define RELEASE_VECTOR(v) irq_disconnect(v)
 #else
diff --git a/kernel/microkernel/k_task.c b/kernel/microkernel/k_task.c
index 17f7bdbffa5..b88af5927b3 100644
--- a/kernel/microkernel/k_task.c
+++ b/kernel/microkernel/k_task.c
@@ -33,7 +33,7 @@
 
 #include <microkernel.h>
 #include <nanokernel.h>
-#include <nanokernel/cpu.h>
+#include <arch/cpu.h>
 #include <toolchain.h>
 #include <sections.h>
 
diff --git a/kernel/microkernel/ticker.c b/kernel/microkernel/ticker.c
index 55a1b151a86..8123f6a41a5 100644
--- a/kernel/microkernel/ticker.c
+++ b/kernel/microkernel/ticker.c
@@ -36,7 +36,7 @@ This module implements the microkernel's tick event handler.
 */
 
 #include <nanokernel.h>
-#include <nanokernel/cpu.h>
+#include <arch/cpu.h>
 
 #include <minik.h>
 #include <drivers/system_timer.h>
diff --git a/kernel/nanokernel/nano_init.c b/kernel/nanokernel/nano_init.c
index c21fee44697..92cfe9663ee 100644
--- a/kernel/nanokernel/nano_init.c
+++ b/kernel/nanokernel/nano_init.c
@@ -38,7 +38,7 @@ This module contains routines that are used to initialize the nanokernel.
 #include <offsets.h>
 #include <cputype.h>
 #include <nanokernel.h>
-#include <nanokernel/cpu.h>
+#include <arch/cpu.h>
 #include <misc/printk.h>
 #include <drivers/rand32.h>
 #include <sections.h>
diff --git a/kernel/nanokernel/string_s.c b/kernel/nanokernel/string_s.c
index 554037bc434..01cf3b7ad09 100644
--- a/kernel/nanokernel/string_s.c
+++ b/kernel/nanokernel/string_s.c
@@ -38,7 +38,7 @@ required by Security Development Lifecycle
 
 /* includes */
 #include <nanokernel.h>
-#include <nanokernel/cpu.h>
+#include <arch/cpu.h>
 #include <toolchain.h>
 #include <sections.h>
 #include <string_s.h>
diff --git a/samples/bluetooth/test_bluetooth/bluetooth.c b/samples/bluetooth/test_bluetooth/bluetooth.c
index 450b11944b8..4bc71c86fe9 100644
--- a/samples/bluetooth/test_bluetooth/bluetooth.c
+++ b/samples/bluetooth/test_bluetooth/bluetooth.c
@@ -34,7 +34,7 @@
 #include <tc_util.h>
 
 #include <nanokernel.h>
-#include <nanokernel/cpu.h>
+#include <arch/cpu.h>
 
 #include <bluetooth/bluetooth.h>
 
diff --git a/samples/microkernel/apps/hello_world/src/hello.c b/samples/microkernel/apps/hello_world/src/hello.c
index 79daad61213..95a93e54850 100644
--- a/samples/microkernel/apps/hello_world/src/hello.c
+++ b/samples/microkernel/apps/hello_world/src/hello.c
@@ -98,7 +98,7 @@ void taskB(void)
  */
 
 #include <nanokernel.h>
-#include <nanokernel/cpu.h>
+#include <arch/cpu.h>
 
 /* specify delay between greetings (in ms); compute equivalent in ticks */
 
diff --git a/samples/microkernel/apps/philosophers/src/phil_fiber.c b/samples/microkernel/apps/philosophers/src/phil_fiber.c
index 823c3d817e6..895cd537bcc 100644
--- a/samples/microkernel/apps/philosophers/src/phil_fiber.c
+++ b/samples/microkernel/apps/philosophers/src/phil_fiber.c
@@ -39,7 +39,7 @@
   #include "phil.h"
 #endif /*  CONFIG_NANOKERNEL */
 
-#include <nanokernel/cpu.h>	/* irq_lock/irq_unlock */
+#include <arch/cpu.h>	/* irq_lock/irq_unlock */
 
 /* defines */
 
diff --git a/samples/microkernel/benchmark/boot_time/src/boot_time.c b/samples/microkernel/benchmark/boot_time/src/boot_time.c
index a0e4dcd2a76..36ce060060b 100644
--- a/samples/microkernel/benchmark/boot_time/src/boot_time.c
+++ b/samples/microkernel/benchmark/boot_time/src/boot_time.c
@@ -47,7 +47,7 @@ Measure boot time for both nanokernel and microkernel project which includes
 #include <vxmicro.h>
 #include <version.h>
 #endif
-#include <nanokernel/cpu.h>
+#include <arch/cpu.h>
 #include <tc_util.h>
 
 /* externs */
diff --git a/samples/microkernel/benchmark/footprint/src/microkernel_footprint.c b/samples/microkernel/benchmark/footprint/src/microkernel_footprint.c
index ccab74bf265..46a365090a4 100644
--- a/samples/microkernel/benchmark/footprint/src/microkernel_footprint.c
+++ b/samples/microkernel/benchmark/footprint/src/microkernel_footprint.c
@@ -43,7 +43,7 @@
 
 /* includes */
 
-#include <nanokernel/cpu.h>
+#include <arch/cpu.h>
 #include <misc/printk.h>
 #include <stdio.h>
 
diff --git a/samples/microkernel/benchmark/latency_measure/src/main.c b/samples/microkernel/benchmark/latency_measure/src/main.c
index d3f59bf5edd..1245d8a7b26 100644
--- a/samples/microkernel/benchmark/latency_measure/src/main.c
+++ b/samples/microkernel/benchmark/latency_measure/src/main.c
@@ -39,7 +39,7 @@
 #include "utils.h"
 #include <tc_util.h>
 
-#include <nanokernel/cpu.h>
+#include <arch/cpu.h>
 
 uint32_t tm_off; /* time necessary to read the time */
 int errorCount = 0; /* track number of errors */
diff --git a/samples/microkernel/benchmark/latency_measure/src/micro_int_to_task.c b/samples/microkernel/benchmark/latency_measure/src/micro_int_to_task.c
index 7dc89770d3d..4400b950724 100644
--- a/samples/microkernel/benchmark/latency_measure/src/micro_int_to_task.c
+++ b/samples/microkernel/benchmark/latency_measure/src/micro_int_to_task.c
@@ -40,7 +40,7 @@
 #include "timestamp.h"
 #include "utils.h"
 
-#include <nanokernel/cpu.h>
+#include <arch/cpu.h>
 
 static volatile int flagVar = 0;
 
diff --git a/samples/microkernel/benchmark/latency_measure/src/micro_int_to_task_evt.c b/samples/microkernel/benchmark/latency_measure/src/micro_int_to_task_evt.c
index f704fdaf7b5..45783535d04 100644
--- a/samples/microkernel/benchmark/latency_measure/src/micro_int_to_task_evt.c
+++ b/samples/microkernel/benchmark/latency_measure/src/micro_int_to_task_evt.c
@@ -44,7 +44,7 @@
 #include "timestamp.h"
 #include "utils.h"
 
-#include <nanokernel/cpu.h>
+#include <arch/cpu.h>
 
 static uint32_t timestamp = 0;
 
diff --git a/samples/microkernel/benchmark/latency_measure/src/micro_sema_lock_release.c b/samples/microkernel/benchmark/latency_measure/src/micro_sema_lock_release.c
index 523fec4b477..d981ff4d1aa 100644
--- a/samples/microkernel/benchmark/latency_measure/src/micro_sema_lock_release.c
+++ b/samples/microkernel/benchmark/latency_measure/src/micro_sema_lock_release.c
@@ -43,7 +43,7 @@
 #include "timestamp.h"
 #include "utils.h"
 
-#include <nanokernel/cpu.h>
+#include <arch/cpu.h>
 
 /* the number of semaphores used in lock/unlock test */
 #define N_TEST_SEMA (SEMAEND - SEMASTART)
diff --git a/samples/microkernel/benchmark/latency_measure/src/nano_ctx_switch.c b/samples/microkernel/benchmark/latency_measure/src/nano_ctx_switch.c
index 8b43c5e18fb..f3a2be29966 100644
--- a/samples/microkernel/benchmark/latency_measure/src/nano_ctx_switch.c
+++ b/samples/microkernel/benchmark/latency_measure/src/nano_ctx_switch.c
@@ -43,7 +43,7 @@
 #include "timestamp.h"
 #include "utils.h"
 
-#include <nanokernel/cpu.h>
+#include <arch/cpu.h>
 
 /* number of context switches */
 #define NCTXSWITCH   10000
diff --git a/samples/microkernel/benchmark/latency_measure/src/nano_int.c b/samples/microkernel/benchmark/latency_measure/src/nano_int.c
index 2d6322784ea..0fb862894a5 100644
--- a/samples/microkernel/benchmark/latency_measure/src/nano_int.c
+++ b/samples/microkernel/benchmark/latency_measure/src/nano_int.c
@@ -39,7 +39,7 @@
 #include "timestamp.h"
 #include "utils.h"
 
-#include <nanokernel/cpu.h>
+#include <arch/cpu.h>
 
 #define STACKSIZE 2000
 
diff --git a/samples/microkernel/benchmark/latency_measure/src/nano_int_lock_unlock.c b/samples/microkernel/benchmark/latency_measure/src/nano_int_lock_unlock.c
index 1a9d6cae52b..d8390158e7b 100644
--- a/samples/microkernel/benchmark/latency_measure/src/nano_int_lock_unlock.c
+++ b/samples/microkernel/benchmark/latency_measure/src/nano_int_lock_unlock.c
@@ -42,7 +42,7 @@
 #include "timestamp.h"
 #include "utils.h"
 
-#include <nanokernel/cpu.h>
+#include <arch/cpu.h>
 
 /* total number of interrupt lock/unlock cycles */
 #define NTESTS 100000
diff --git a/samples/microkernel/benchmark/latency_measure/src/nano_int_to_fiber.c b/samples/microkernel/benchmark/latency_measure/src/nano_int_to_fiber.c
index 47d2efb985a..2b7e210fbb8 100644
--- a/samples/microkernel/benchmark/latency_measure/src/nano_int_to_fiber.c
+++ b/samples/microkernel/benchmark/latency_measure/src/nano_int_to_fiber.c
@@ -39,7 +39,7 @@
 #include "timestamp.h"
 #include "utils.h"
 
-#include <nanokernel/cpu.h>
+#include <arch/cpu.h>
 
 #define STACKSIZE 2000
 
diff --git a/samples/microkernel/benchmark/latency_measure/src/nano_int_to_fiber_sem.c b/samples/microkernel/benchmark/latency_measure/src/nano_int_to_fiber_sem.c
index c0d59cf1d88..2070a0798fa 100644
--- a/samples/microkernel/benchmark/latency_measure/src/nano_int_to_fiber_sem.c
+++ b/samples/microkernel/benchmark/latency_measure/src/nano_int_to_fiber_sem.c
@@ -48,7 +48,7 @@
 #include "timestamp.h"
 #include "utils.h"
 
-#include <nanokernel/cpu.h>
+#include <arch/cpu.h>
 
 #define STACKSIZE 2000
 
diff --git a/samples/microkernel/benchmark/latency_measure/src/raise_int.c b/samples/microkernel/benchmark/latency_measure/src/raise_int.c
index 90468e346e7..6d6594da63c 100644
--- a/samples/microkernel/benchmark/latency_measure/src/raise_int.c
+++ b/samples/microkernel/benchmark/latency_measure/src/raise_int.c
@@ -1406,7 +1406,7 @@ void raiseInt(uint8_t id)
 #endif /* Intel */
 
 #if defined(CONFIG_CPU_CORTEXM3)
-#include <nanokernel/cpu.h>
+#include <arch/cpu.h>
 /*******************************************************************************
  *
  * raiseInt - generate a software interrupt
diff --git a/samples/microkernel/benchmark/latency_measure/src/utils.c b/samples/microkernel/benchmark/latency_measure/src/utils.c
index 920ae2aa66d..d377d2e5f3e 100644
--- a/samples/microkernel/benchmark/latency_measure/src/utils.c
+++ b/samples/microkernel/benchmark/latency_measure/src/utils.c
@@ -46,7 +46,7 @@
 
 #endif /*  CONFIG_NANOKERNEL */
 
-#include <nanokernel/cpu.h>
+#include <arch/cpu.h>
 
 static uint8_t vector; /* the interrupt vector we allocate */
 
diff --git a/samples/microkernel/test/test_events/src/events.c b/samples/microkernel/test/test_events/src/events.c
index 9c24ee0d6f8..040f203a83b 100644
--- a/samples/microkernel/test/test_events/src/events.c
+++ b/samples/microkernel/test/test_events/src/events.c
@@ -42,7 +42,7 @@ This modules tests the following event APIs:
 
 #include <tc_util.h>
 #include <vxmicro.h>
-#include <nanokernel/cpu.h>
+#include <arch/cpu.h>
 #include <toolchain.h>
 
 /* test uses 1 software IRQs */
diff --git a/samples/microkernel/test/test_events/src/test_fiber.c b/samples/microkernel/test/test_events/src/test_fiber.c
index 61db5233e91..b9eaa3ed25a 100644
--- a/samples/microkernel/test/test_events/src/test_fiber.c
+++ b/samples/microkernel/test/test_events/src/test_fiber.c
@@ -38,7 +38,7 @@ event signaling
 
 /* includes */
 #include <nanokernel.h>
-#include <nanokernel/cpu.h>
+#include <arch/cpu.h>
 #include <vxmicro.h>
 
 /* defines */
diff --git a/samples/microkernel/test/test_fp_sharing/src/main.c b/samples/microkernel/test/test_fp_sharing/src/main.c
index 2ab4ffc1185..a27dd4603ec 100644
--- a/samples/microkernel/test/test_fp_sharing/src/main.c
+++ b/samples/microkernel/test/test_fp_sharing/src/main.c
@@ -87,7 +87,7 @@ x87 FPU registers are being saved/restored.
 #include <float_regs_x86_other.h>
 #endif /* __GNUC__ */
 
-#include <nanokernel/cpu.h>
+#include <arch/cpu.h>
 #include <tc_util.h>
 #include "float_context.h"
 #include <stddef.h>
diff --git a/samples/microkernel/test/test_pipe/src/pipe.c b/samples/microkernel/test/test_pipe/src/pipe.c
index 256b208e880..ec239b32b9e 100644
--- a/samples/microkernel/test/test_pipe/src/pipe.c
+++ b/samples/microkernel/test/test_pipe/src/pipe.c
@@ -44,7 +44,7 @@ The following target pipe routine does not yet have a test case:
 /* includes */
 
 #include <nanokernel.h>
-#include <nanokernel/cpu.h>
+#include <arch/cpu.h>
 #include <tc_util.h>
 #include <vxmicro.h>
 #include <misc/util.h>
diff --git a/samples/microkernel/test/test_pool/src/pool.c b/samples/microkernel/test/test_pool/src/pool.c
index 2a00c704baf..220a1d025d6 100644
--- a/samples/microkernel/test/test_pool/src/pool.c
+++ b/samples/microkernel/test/test_pool/src/pool.c
@@ -42,7 +42,7 @@ This modules tests the following memory pool routines:
 
 #include <tc_util.h>
 #include <nanokernel.h>
-#include <nanokernel/cpu.h>
+#include <arch/cpu.h>
 #include <vxmicro.h>
 #include <misc/util.h>
 
diff --git a/samples/microkernel/test/test_sema/src/main.c b/samples/microkernel/test/test_sema/src/main.c
index 49b96aa6fa6..215c4c20ee4 100644
--- a/samples/microkernel/test/test_sema/src/main.c
+++ b/samples/microkernel/test/test_sema/src/main.c
@@ -45,7 +45,7 @@ then announces the result of the test.
 /* includes */
 
 #include <nanokernel.h>
-#include <nanokernel/cpu.h>
+#include <arch/cpu.h>
 #include <vxmicro.h>
 #include <tc_util.h>
 
diff --git a/samples/microkernel/test/test_sema/src/sema.c b/samples/microkernel/test/test_sema/src/sema.c
index bc525504852..4cf1433b565 100644
--- a/samples/microkernel/test/test_sema/src/sema.c
+++ b/samples/microkernel/test/test_sema/src/sema.c
@@ -44,7 +44,7 @@ This modules tests the following semaphore routines:
 /* includes */
 
 #include <nanokernel.h>
-#include <nanokernel/cpu.h>
+#include <arch/cpu.h>
 #include <vxmicro.h>
 #include <tc_util.h>
 
diff --git a/samples/microkernel/test/test_sema/src/test_fiber.c b/samples/microkernel/test/test_sema/src/test_fiber.c
index 24d44f7da71..7ffcaaea2f5 100644
--- a/samples/microkernel/test/test_sema/src/test_fiber.c
+++ b/samples/microkernel/test/test_sema/src/test_fiber.c
@@ -38,7 +38,7 @@ semaphore signaling
 
 /* includes */
 #include <nanokernel.h>
-#include <nanokernel/cpu.h>
+#include <arch/cpu.h>
 #include <vxmicro.h>
 
 /* defines */
diff --git a/samples/microkernel/test/test_stackprot/src/stackprot.c b/samples/microkernel/test/test_stackprot/src/stackprot.c
index 7dee7568f9c..c049eed2fd1 100644
--- a/samples/microkernel/test/test_stackprot/src/stackprot.c
+++ b/samples/microkernel/test/test_stackprot/src/stackprot.c
@@ -55,7 +55,7 @@ DESCRIPTION
 /* locals */
 
 #else
-#include <nanokernel/cpu.h>
+#include <arch/cpu.h>
 
 /* defines */
 #define STACKSIZE               1024
diff --git a/samples/microkernel/test/test_static_idt/src/static_idt.c b/samples/microkernel/test/test_static_idt/src/static_idt.c
index c66f234e44a..f13a88a481a 100644
--- a/samples/microkernel/test/test_static_idt/src/static_idt.c
+++ b/samples/microkernel/test/test_static_idt/src/static_idt.c
@@ -40,7 +40,7 @@ Ensures interrupt and exception stubs are installed correctly.
 #include <tc_util.h>
 
 #include <nanokernel.h>
-#include <nanokernel/cpu.h>
+#include <arch/cpu.h>
 #include <nanok.h>
 #if defined(__GNUC__)
 #include <test_asm_inline_gcc.h>
diff --git a/samples/microkernel/test/test_static_idt/src/test_stubs.s b/samples/microkernel/test/test_static_idt/src/test_stubs.s
index cbb5df69e31..b33d7612c32 100644
--- a/samples/microkernel/test/test_static_idt/src/test_stubs.s
+++ b/samples/microkernel/test/test_static_idt/src/test_stubs.s
@@ -42,9 +42,9 @@ testing.
 
 /* IA-32 specific */
 
-#include <nanokernel/cpu.h>
+#include <arch/cpu.h>
 #include <nanok.h>
-#include <nanokernel/x86/asm.h>
+#include <arch/x86/asm.h>
 #include <asmPrv.h>
 #include <asm_inline.h>
 
diff --git a/samples/microkernel/test/test_task/src/task.c b/samples/microkernel/test/test_task/src/task.c
index 76319c462ae..55a67cb458e 100644
--- a/samples/microkernel/test/test_task/src/task.c
+++ b/samples/microkernel/test/test_task/src/task.c
@@ -42,7 +42,7 @@ This module tests the following task APIs:
 
 #include <tc_util.h>
 #include <vxmicro.h>
-#include <nanokernel/cpu.h>
+#include <arch/cpu.h>
 
 /* test uses 1 software IRQs */
 #define NUM_SW_IRQS 1
diff --git a/samples/microkernel/test/test_task_irq/src/main.c b/samples/microkernel/test/test_task_irq/src/main.c
index 3d54d2253a3..963ed1a0f66 100644
--- a/samples/microkernel/test/test_task_irq/src/main.c
+++ b/samples/microkernel/test/test_task_irq/src/main.c
@@ -37,7 +37,7 @@
 
 #include <tc_util.h>
 #include <nanokernel.h>
-#include <nanokernel/cpu.h>
+#include <arch/cpu.h>
 #include <vxmicro.h>
 
 /* One of the task IRQ objects will not be allocated */
diff --git a/samples/microkernel/test/test_task_irq/src/raise_int.c b/samples/microkernel/test/test_task_irq/src/raise_int.c
index 89afb9c3f3b..e7c661e5f57 100644
--- a/samples/microkernel/test/test_task_irq/src/raise_int.c
+++ b/samples/microkernel/test/test_task_irq/src/raise_int.c
@@ -1406,7 +1406,7 @@ void raiseInt(uint8_t id)
 #endif /* Intel */
 
 #if defined(CONFIG_CPU_CORTEXM3)
-#include <nanokernel/cpu.h>
+#include <arch/cpu.h>
 /*******************************************************************************
  *
  * raiseInt - generate a software interrupt
diff --git a/samples/microkernel/test/test_tickless/src/test_tickless.c b/samples/microkernel/test/test_tickless/src/test_tickless.c
index 512ad9cc13e..12993d146b2 100644
--- a/samples/microkernel/test/test_tickless/src/test_tickless.c
+++ b/samples/microkernel/test/test_tickless/src/test_tickless.c
@@ -38,7 +38,7 @@ Unit test for tickless idle feature.
 #include <vxmicro.h>
 
 #include <misc/printk.h>
-#include <nanokernel/cpu.h>
+#include <arch/cpu.h>
 #include <tc_util.h>
 
 #define SLEEP_TICKS 10
diff --git a/samples/nanokernel/benchmark/footprint/src/nanokernel_footprint.c b/samples/nanokernel/benchmark/footprint/src/nanokernel_footprint.c
index 4bdc38befe8..6396f0c2a34 100644
--- a/samples/nanokernel/benchmark/footprint/src/nanokernel_footprint.c
+++ b/samples/nanokernel/benchmark/footprint/src/nanokernel_footprint.c
@@ -40,7 +40,7 @@ volatile int i = 0;		/* counter used by background task */
 /* includes */
 
 #include <misc/printk.h>
-#include <nanokernel/cpu.h>
+#include <arch/cpu.h>
 #include <stdio.h>
 
 #ifdef TEST_reg
diff --git a/samples/nanokernel/benchmark/sys_kernel/src/syskernel.c b/samples/nanokernel/benchmark/sys_kernel/src/syskernel.c
index 5d76067c556..d401f81cc52 100644
--- a/samples/nanokernel/benchmark/sys_kernel/src/syskernel.c
+++ b/samples/nanokernel/benchmark/sys_kernel/src/syskernel.c
@@ -34,7 +34,7 @@
 #include <vxmicro.h>
 #else
 #include <nanokernel.h>
-#include <nanokernel/cpu.h>
+#include <arch/cpu.h>
 #endif
 
 #include "syskernel.h"
diff --git a/samples/nanokernel/test/test_arm_m3_irq_vector_table/src/main.c b/samples/nanokernel/test/test_arm_m3_irq_vector_table/src/main.c
index 7744fbf669b..138bce98ac3 100644
--- a/samples/nanokernel/test/test_arm_m3_irq_vector_table/src/main.c
+++ b/samples/nanokernel/test/test_arm_m3_irq_vector_table/src/main.c
@@ -45,7 +45,7 @@ The ISRs are installed at build time, directly in the vector table.
   #error project can only run on Cortex-M3
 #endif
 
-#include <nanokernel/cpu.h>
+#include <arch/cpu.h>
 #include <tc_util.h>
 #include <sections.h>
 
diff --git a/samples/nanokernel/test/test_context/src/context.c b/samples/nanokernel/test/test_context/src/context.c
index b6f8fee4cb7..24598c12679 100644
--- a/samples/nanokernel/test/test_context/src/context.c
+++ b/samples/nanokernel/test/test_context/src/context.c
@@ -45,7 +45,7 @@ This module tests the following CPU and context related routines:
 
 #include <tc_util.h>
 #include <nanok.h>
-#include <nanokernel/cpu.h>
+#include <arch/cpu.h>
 
 /* test uses 1 software IRQs */
 #define NUM_SW_IRQS 1
diff --git a/samples/nanokernel/test/test_fifo/src/fifo.c b/samples/nanokernel/test/test_fifo/src/fifo.c
index 2113b1f546e..56b22a3521b 100644
--- a/samples/nanokernel/test/test_fifo/src/fifo.c
+++ b/samples/nanokernel/test/test_fifo/src/fifo.c
@@ -64,7 +64,7 @@ in ISR context.
 /* includes */
 
 #include <tc_util.h>
-#include <nanokernel/cpu.h>
+#include <arch/cpu.h>
 
 /* test uses 2 software IRQs */
 #define NUM_SW_IRQS 2
diff --git a/samples/nanokernel/test/test_lifo/src/lifo.c b/samples/nanokernel/test/test_lifo/src/lifo.c
index 92bc4945854..d6763ca2afa 100644
--- a/samples/nanokernel/test/test_lifo/src/lifo.c
+++ b/samples/nanokernel/test/test_lifo/src/lifo.c
@@ -55,7 +55,7 @@ These scenarios will be tested using a combinations of tasks, fibers and ISRs.
 /* includes */
 
 #include <tc_util.h>
-#include <nanokernel/cpu.h>
+#include <arch/cpu.h>
 
 /* test uses 2 software IRQs */
 #define NUM_SW_IRQS 2
diff --git a/samples/nanokernel/test/test_sema/src/sema.c b/samples/nanokernel/test/test_sema/src/sema.c
index d597f0fc885..bd3cfe385de 100644
--- a/samples/nanokernel/test/test_sema/src/sema.c
+++ b/samples/nanokernel/test/test_sema/src/sema.c
@@ -53,7 +53,7 @@ Scenario #3:
 /* includes */
 
 #include <tc_util.h>
-#include <nanokernel/cpu.h>
+#include <arch/cpu.h>
 
 /* test uses 2 software IRQs */
 #define NUM_SW_IRQS 2
diff --git a/samples/nanokernel/test/test_stack/src/stack.c b/samples/nanokernel/test/test_stack/src/stack.c
index b4a1289bbde..d608c92a41b 100644
--- a/samples/nanokernel/test/test_stack/src/stack.c
+++ b/samples/nanokernel/test/test_stack/src/stack.c
@@ -63,7 +63,7 @@ these are run in ISR context.
 /* includes */
 
 #include <tc_util.h>
-#include <nanokernel/cpu.h>
+#include <arch/cpu.h>
 
 /* test uses 2 software IRQs */
 #define NUM_SW_IRQS 2
diff --git a/samples/nanokernel/test/test_timer/src/timer.c b/samples/nanokernel/test/test_timer/src/timer.c
index 978e96c8099..d5e9b4c1b8f 100644
--- a/samples/nanokernel/test/test_timer/src/timer.c
+++ b/samples/nanokernel/test/test_timer/src/timer.c
@@ -42,7 +42,7 @@ This module tests the following timer related routines:
 /* includes */
 
 #include <tc_util.h>
-#include <nanokernel/cpu.h>
+#include <arch/cpu.h>
 
 /* defines */