Added AES+GHASH implementation using AES-NI opcodes; also ARM-Thumb assembly for faster Montgomery multiplication on Cortex-M0+. Added selection functions for "default" implementations.

conf/Unix.mk

@@ -53,7 +53,7 @@ LDDLLOUT = -o
 
 # Static linker.
 LD = gcc
-LDFLAGS =
+LDFLAGS = 
 LDOUT = -o 
 
 # C# compiler; we assume usage of Mono.

conf/Unix32.mk

@@ -0,0 +1,12 @@
+# Example configuration file for compiling on a Unix-like system with
+# GCC, targeting a 32-bit output. Moreover, it enables the "LOMUL" setting
+# to make the code select the "small" integer implementations (i15, m15,
+# ctmul32...), which is not necessarily a good idea for performance, but
+# handy for tests.
+
+include conf/Unix.mk
+
+BUILD = build32
+CFLAGS = -W -Wall -Os -fPIC -m32 -DBR_LOMUL
+LDFLAGS = -m32
+LDDLLFLAGS = -shared -m32

conf/UnixClang.mk

@@ -0,0 +1,9 @@
+# Example configuration file for compiling on a Unix-like system with
+# clang as compiler instead of gcc.
+
+# We are on a Unix system so we assume a Single Unix compatible 'make'
+# utility, and Unix defaults.
+include conf/Unix.mk
+
+BUILD = bclang
+CC = clang

conf/samd20.mk

@@ -10,7 +10,7 @@ BUILD = samd20
 
 # C compiler, linker, and static library builder.
 CC = arm-none-eabi-gcc
-CFLAGS = -W -Wall -Os -mthumb -ffunction-sections -fdata-sections -mcpu=cortex-m0plus
+CFLAGS = -W -Wall -Os -mthumb -ffunction-sections -fdata-sections -mcpu=cortex-m0plus -DBR_ARMEL_CORTEXM_GCC
 LD = arm-none-eabi-gcc
 AR = arm-none-eabi-ar