diff --git a/tools/mem_init_gen.py b/tools/mem_init_gen.py
new file mode 100644
index 0000000000000000000000000000000000000000..9fa2cf00d5695f5c5e47ce9f47cb96533f8bcc86
--- /dev/null
+++ b/tools/mem_init_gen.py
@@ -0,0 +1,155 @@
+#!/usr/bin/env python
+##-------------------------------------------------------------------------------
+## CERN BE-CO-HT
+## General Cores
+## https://www.ohwr.org/projects/general-cores
+##-------------------------------------------------------------------------------
+##
+## Python script to generate an init file for block rams. Input file should be
+## raw binary, such as the one generated by GNU binutils "objcopy -O binary".
+##
+## Loosely based on the various init generation tools found in wrpc-sw.
+##
+## Usage:
+## python mem_init_gen.py -of output_format input_file > output_file
+##
+## Supported output formats:
+##
+## bram: custom ascii-encoded binary, one 32-bit value per line, for Xilinx,
+## to be used with the functions in the memory_loader_pkg
+##
+## mif : Altera Memory Initialization File
+##
+## vhd : constant VHDL array of type t_meminit_array, as defined in genram_pkg.
+##
+##-------------------------------------------------------------------------------
+## Copyright CERN 2018
+##-------------------------------------------------------------------------------
+## This Source Code Form is subject to the terms of the Mozilla Public License,
+## version 2.0. If a copy of the MPL was not distributed with this file, You can
+## obtain one at https://mozilla.org/MPL/2.0/.
+##-------------------------------------------------------------------------------
+
+from __future__ import print_function
+
+import argparse
+import binascii
+import datetime
+
+today = datetime.date.today()
+
+parser = argparse.ArgumentParser (
+ description='script to generate an init file for block rams' )
+
+parser.add_argument ( 'in_file' )
+parser.add_argument ( '-i', '--invert', action='store_true', dest='invert',
+ help = 'Invert endianess of input file' )
+parser.add_argument ( '-of', '--oformat', choices = ['BRAM', 'MIF', 'VHD'], type = str.upper,
+ required = False, dest='oformat', default = 'BRAM',
+ help = 'output format (default is BRAM)' )
+parser.add_argument ( '-d', '--depth', type = int, default = 0,
+ required = False, dest='depth',
+ help = 'depth of memory in bytes (default is equal to the '
+ 'size of the input file)' )
+parser.add_argument ( '-w', '--width', type = int,
+ required = False, dest='width', default = 4,
+ help = 'width of memory in bytes (default is 4)' )
+parser.add_argument ( '-p', '--pad', type = int,
+ required = False, dest='pad', default = 0,
+ help = 'byte padding value to use if size argument is larger '
+ 'than input file size (default is 0, max is 255)' )
+parser.add_argument ( '-n', '--name', type = str.lower,
+ required = False, dest='name', default = 'mem_init',
+ help = 'name to use for this block (default is "mem_init")' )
+
+args = parser.parse_args ( )
+
+# check for proper padding value
+if args.pad not in range ( 0, 256 ):
+ parser.error ( 'Padding value must be between 0 and 255' )
+
+# read binary file
+with open ( args.in_file, 'rb' ) as fin:
+ contents = fin.read ( )
+
+# byte value to use for padding
+pad_val = chr ( args.pad & 0xff )
+
+# pad/trim if necessary to get to args.depth
+if args.depth:
+ byte_count = args.depth * args.width
+ if byte_count > len ( contents ):
+ contents += pad_val * ( byte_count - len ( contents ) )
+ else:
+ contents = contents[:byte_count]
+
+# pad if necessary to get to args.width boundary
+contents += pad_val * ( len ( contents ) % args.width )
+
+# convert to 2D list of WIDTH-sized ASCII-encoded (hex format) elements
+# with optional endianess inversion
+if args.invert == True and args.width > 1:
+ words = [ int ( binascii.hexlify ( contents[i:i+args.width][::-1] ), 16 )
+ for i in range ( 0, len ( contents ), args.width ) ]
+else:
+ words = [ int ( binascii.hexlify ( contents[i:i+args.width] ), 16 )
+ for i in range ( 0, len ( contents ), args.width ) ]
+
+# BRAM output
+if args.oformat == 'BRAM':
+ fwidth = args.width * 8
+ for word in words:
+ print ( '{:0{fwidth}b}'.format ( word, fwidth = fwidth ) )
+
+# MIF output
+if args.oformat == 'MIF':
+ print ( 'DEPTH = {};'.format ( len ( words ) ) )
+ print ( 'WIDTH = {};'.format ( args.width ) )
+ print ( 'ADDRESS_RADIX = HEX;' )
+ print ( 'DATA_RADIX = HEX;' )
+ print ( 'CONTENT' )
+ print ( 'BEGIN' )
+ fwidth = args.width * 2
+ for i, word in enumerate ( words ):
+ print ( '{:x} : {:0{fwidth}x};'.format ( i, word, fwidth = fwidth ) )
+ print ( 'END;' )
+
+# VHD output
+if args.oformat == 'VHD':
+ print ( '-' * 80 )
+ print ( '-- Memory initialization file for {}'.format ( args.in_file ) )
+ print ( '--' )
+ print ( '-- This file was automatically generated on {}'.format (
+ today.strftime ( '%A, %B %d %Y') ) )
+ print ( '-- by {} using the following arguments:'.format ( parser.prog ) )
+ print ( '-- {}'.format ( vars(args) ) )
+ print ( '--' )
+ print ( '-- {} is part of OHWR general-cores:'.format ( parser.prog ) )
+ print ( '-- https://www.ohwr.org/projects/general-cores/wiki' )
+ print ( '-' * 80 )
+ print ( )
+ print ( 'library ieee;' )
+ print ( 'use ieee.std_logic_1164.all;' )
+ print ( 'use ieee.numeric_std.all;' )
+ print ( )
+ print ( 'library work;' )
+ print ( 'use work.memory_loader_pkg.all;' )
+ print ( )
+ print ( 'package {}_pkg is'.format ( args.name ) )
+ print ( )
+ print ( ' constant {} : t_meminit_array('.format ( args.name ),
+ '{} downto 0, {} downto 0) := ('.format ( len ( words ) - 1, args.width * 8 - 1 ) )
+ iwidth = len ( str ( len ( words ) ) )
+ fwidth = args.width * 2
+ numcol = 80 / (12 + iwidth + fwidth)
+ for i, word in enumerate ( words ):
+ print ( ' {:{iwidth}d} => x"{:0{fwidth}x}"'.format (
+ i, word, iwidth = iwidth, fwidth = fwidth ), end = '' )
+ if i == len ( words ) - 1:
+ print ( ');' );
+ else:
+ print ( ',', end = '' );
+ if i % numcol == numcol - 1:
+ print ( )
+ print ( )
+ print ( 'end package {}_pkg;'.format ( args.name ) )