Fix off-by-1 error in memap's symbol-contained-in-memory-bank check, clean up some linker script stuff (#362)

* Fix off-by-1 error in memap's symbol-contained-in-memory-bank check, clean up some linker script stuff

* Update comment

* One more typo

targets/TARGET_NXP/TARGET_LPC176X/device/TOOLCHAIN_GCC_ARM/LPC1768.ld

@@ -159,7 +159,7 @@ SECTIONS
      * values to stack symbols later */
     .stack_dummy (NOLOAD):
     {
-        *(.stack)
+        . += STACK_SIZE;
     } > RAM
 
     /* Set stack top to end of RAM, and stack limit move down by

tools/cmake/mbed_set_linker_script.cmake

@@ -98,6 +98,7 @@ function(mbed_setup_linker_script mbed_os_target mbed_baremetal_target target_de
                 INTERFACE
                     "-T" "${LINKER_SCRIPT_PATH}"
                 )
+            set_property(TARGET ${TARGET} APPEND PROPERTY INTERFACE_LINK_DEPENDS ${LINKER_SCRIPT_PATH})
         endif()
     endforeach()
 
@@ -150,5 +151,6 @@ function(mbed_set_custom_linker_script target new_linker_script_path)
         PRIVATE
             "-T" "${CUSTOM_LINKER_SCRIPT_PATH}"
     )
+    set_property(TARGET ${target} APPEND PROPERTY LINK_DEPENDS ${CUSTOM_LINKER_SCRIPT_PATH})
 
 endfunction(mbed_set_custom_linker_script)

tools/cmake/mbed_target_functions.cmake

@@ -128,19 +128,14 @@ function(mbed_set_post_build target)
             get_target_property(POST_BUILD_TARGET_LINK_LIBRARIES ${target} LINK_LIBRARIES)
             if("mbed-os" IN_LIST POST_BUILD_TARGET_LINK_LIBRARIES)
                 get_target_property(LINKER_SCRIPT_PATH mbed-os LINKER_SCRIPT_PATH)
-                target_link_options(${target}
-                PRIVATE
-                    "-T" "${LINKER_SCRIPT_PATH}"
-            )
             elseif("mbed-baremetal" IN_LIST POST_BUILD_TARGET_LINK_LIBRARIES)
                 get_target_property(LINKER_SCRIPT_PATH mbed-baremetal LINKER_SCRIPT_PATH)
-                target_link_options(${target}
-                PRIVATE
-                    "-T" "${LINKER_SCRIPT_PATH}"
-            )
             else()
                 message(FATAL_ERROR "Target ${target} used with mbed_set_post_build() but does not link to mbed-os or mbed-baremetal!")
             endif()
+
+            target_link_options(${target} PRIVATE "-T" "${LINKER_SCRIPT_PATH}")
+            set_property(TARGET ${target} APPEND PROPERTY LINK_DEPENDS ${LINKER_SCRIPT_PATH})
         else()
             message(STATUS "${target} uses custom linker script  ${ARGV1}")
             mbed_set_custom_linker_script(${target} ${ARGV1})

tools/python/memap/memap.py

@@ -63,7 +63,6 @@ from collections import defaultdict
 from prettytable import PrettyTable, HEADER
 from jinja2 import FileSystemLoader, StrictUndefined
 from jinja2.environment import Environment
-from future.utils import with_metaclass
 
 
 # Be sure that the tools directory is in the search path
@@ -127,7 +126,7 @@ class _Parser(ABC):
         for banks in self.memory_banks.values():
             for bank_info in banks:
                 if bank_info.contains_addr(symbol_start_addr):
-                    if bank_info.contains_addr(end_addr):
+                    if bank_info.contains_addr(end_addr - 1): # end_addr is the first address past the end of the symbol so we subtract 1 here
                         # Symbol fully inside this memory bank
                         bank_info.used_size += size
 
@@ -146,7 +145,7 @@ class _Parser(ABC):
         """ Adds information about a symbol (e.g. a function or global variable) to the data structures.
 
         Positional arguments:
-        symbol_name - Descriptive name of the symbol, e.g. ".text.some_function"
+        symbol_name - Descriptive name of the symbol, e.g. ".text.some_function" or "*fill*"
         object_name - name of the object file containing the symbol
         start addr - start address of symbol
         size - the size of the symbol being added
@@ -234,7 +233,9 @@ class _GccParser(_Parser):
 
     # Gets the input section name from the line, if it exists.
     # Input section names are always indented 1 space.
-    RE_INPUT_SECTION_NAME = re.compile(r'^ (\.\w+\.?\w*\.?\w*)')  # Note: This allows up to 3 dots... hopefully that's enough...
+    # Note: This allows up to 3 dots... hopefully that's enough...
+    # It can also capture "*fill*" instead of something that looks like a section name.
+    RE_INPUT_SECTION_NAME = re.compile(r'^ ((?:\.\w+\.?\w*\.?\w*)|(?:\*fill\*))')
 
     ALL_SECTIONS = (
         _Parser.SECTIONS
@@ -844,7 +845,7 @@ class MemapParser(object):
 
 def main():
     """Entry Point"""
-    version = '0.4.0'
+    version = '1.0.0'
 
     # Parser handling
     parser = ArgumentParser(