diff --git a/GNUmakefile b/GNUmakefile
index e043cb63..7b68286d 100755
--- a/GNUmakefile
+++ b/GNUmakefile
@@ -266,6 +266,13 @@ endif  # -DCRYPTOPP_DISABLE_SSSE3
 
 # Begin SunCC
 ifeq ($(SUN_COMPILER),1)
+  COUNT := $(shell $(CXX) $(CXXFLAGS) -E -xarch=sse2 -xdumpmacros /dev/null 2>&1 | $(GREP) -i -c "illegal")
+  ifeq ($(COUNT),0)
+    AES_FLAG = -xarch=sse2 -D__SSE2__=1
+    GCM_FLAG = -xarch=sse2 -D__SSE2__=1
+    SHA_FLAG = -xarch=sse2 -D__SSE2__=1
+    LDFLAGS += -xarch=sse2
+  endif
   COUNT := $(shell $(CXX) $(CXXFLAGS) -E -xarch=ssse3 -xdumpmacros /dev/null 2>&1 | $(GREP) -i -c "illegal")
   ifeq ($(COUNT),0)
     SSSE3_FLAG = -xarch=ssse3 -D__SSSE3__=1
@@ -396,9 +403,12 @@ ifneq ($(IS_PPC32)$(IS_PPC64)$(IS_AIX),000)
     ALTIVEC_FLAG = -mcpu=power4 -maltivec
     ARIA_FLAG = -mcpu=power4 -maltivec
     BLAKE2_FLAG = -mcpu=power4 -maltivec
+    CHAM_FLAG = -mcpu=power4 -maltivec
+    LEA_FLAG = -mcpu=power4 -maltivec
     SIMON_FLAG = -mcpu=power4 -maltivec
-    SIMECK_FLAG = -mcpu=power4 -maltivec
     SPECK_FLAG = -mcpu=power4 -maltivec
+    SIMECK_FLAG = -mcpu=power4 -maltivec
+    SM4_FLAG = -mcpu=power7 -maltivecs
   endif
   # GCC and some compatibles
   HAVE_CRYPTO = $(shell echo | $(CXX) -x c++ $(CXXFLAGS) -mcpu=power8 -maltivec -dM -E - 2>/dev/null | $(GREP) -i -c -E '_ARCH_PWR8|_ARCH_PWR9|__CRYPTO')
@@ -407,9 +417,12 @@ ifneq ($(IS_PPC32)$(IS_PPC64)$(IS_AIX),000)
     AES_FLAG = -mcpu=power8 -maltivec
     GCM_FLAG = -mcpu=power8 -maltivec
     SHA_FLAG = -mcpu=power8 -maltivec
-    SIMECK_FLAG = -mcpu=power8 -maltivec
+    CHAM_FLAG = -mcpu=power8 -maltivec
+    LEA_FLAG = -mcpu=power8 -maltivec
     SIMON_FLAG = -mcpu=power8 -maltivec
     SPECK_FLAG = -mcpu=power8 -maltivec
+    SIMECK_FLAG = -mcpu=power8 -maltivec
+    SM4_FLAG = -mcpu=power8 -maltivec
   endif
   # IBM XL C/C++
   HAVE_ALTIVEC = $(shell $(CXX) $(CXXFLAGS) -qshowmacros -qarch=pwr7 -qaltivec -E adhoc.cpp.proto 2>/dev/null | $(GREP) -i -c '__ALTIVEC__')
@@ -417,9 +430,12 @@ ifneq ($(IS_PPC32)$(IS_PPC64)$(IS_AIX),000)
     ALTIVEC_FLAG = -qarch=pwr7 -qaltivec
     ARIA_FLAG = -qarch=pwr7 -qaltivec
     BLAKE2_FLAG = -qarch=pwr7 -qaltivec
+    CHAM_FLAG = -qarch=pwr7 -qaltivec
+    LEA_FLAG = -qarch=pwr7 -qaltivec
     SIMECK_FLAG = -qarch=pwr7 -qaltivec
     SIMON_FLAG = -qarch=pwr7 -qaltivec
     SPECK_FLAG = -qarch=pwr7 -qaltivec
+    SM4_FLAG = -qarch=pwr7 -qaltivec
   endif
   # IBM XL C/C++
   HAVE_CRYPTO = $(shell $(CXX) $(CXXFLAGS) -qshowmacros -qarch=pwr8 -qaltivec -E adhoc.cpp.proto 2>/dev/null | $(GREP) -i -c -E '_ARCH_PWR8|_ARCH_PWR9|__CRYPTO')
@@ -430,9 +446,12 @@ ifneq ($(IS_PPC32)$(IS_PPC64)$(IS_AIX),000)
     SHA_FLAG = -qarch=pwr8 -qaltivec
     ARIA_FLAG = -qarch=pwr8 -qaltivec
     BLAKE2_FLAG = -qarch=pwr8 -qaltivec
+    CHAM_FLAG = -qarch=pwr8 -qaltivec
+    LEA_FLAG = -qarch=pwr8 -qaltivec
     SIMECK_FLAG = -qarch=pwr8 -qaltivec
     SIMON_FLAG = -qarch=pwr8 -qaltivec
     SPECK_FLAG = -qarch=pwr8 -qaltivec
+    SM4_FLAG = -qarch=pwr8 -qaltivec
   endif
 endif
 
@@ -442,10 +461,6 @@ ifeq ($(XLC_COMPILER),1)
   ifeq ($(findstring -qrtti,$(CXXFLAGS)),)
     CXXFLAGS += -qrtti
   endif
-  # -fPIC causes link errors dues to unknown option
-  ifneq ($(findstring -fPIC,$(CXXFLAGS)),)
-      CXXFLAGS := $(CXXFLAGS:-fPIC=-qpic)
-  endif
   HAVE_BITS=$(shell echo $(CXXFLAGS) | $(GREP) -i -c -E '\-q32|\-q64')
   ifeq ($(IS_PPC64)$(XLC_COMPILER)$(HAVE_BITS),110)
     CXXFLAGS += -q64
@@ -480,6 +495,11 @@ ifeq ($(SUN_COMPILER),1)
   CXXFLAGS := $(subst -fPIC,-KPIC,$(CXXFLAGS))
 endif
 
+# Remove -fPIC if present. IBM XL C/C++ use -qpic
+ifeq ($(XLC_COMPILER),1)
+  CXXFLAGS := $(subst -fPIC,-qpic,$(CXXFLAGS))
+endif
+
 # Add -pipe for everything except IBM XL C/C++, SunCC and ARM.
 # Allow ARM-64 because they seems to have >1 GB of memory
 ifeq ($(XLC_COMPILER)$(SUN_COMPILER)$(IS_ARM32),000)
diff --git a/ppc-simd.h b/ppc-simd.h
index ddfa24ce..c33358a6 100644
--- a/ppc-simd.h
+++ b/ppc-simd.h
@@ -147,7 +147,7 @@ inline T Reverse(const T& src)
 inline uint32x4_p VectorLoadBE(const uint8_t src[16])
 {
 #if defined(CRYPTOPP_XLC_VERSION)
-    return (uint32x4_p)vec_xl_be(0, src);
+    return (uint32x4_p)vec_xl_be(0, (byte*)src);
 #else
 # if defined(CRYPTOPP_LITTLE_ENDIAN)
     return (uint32x4_p)Reverse(vec_vsx_ld(0, src));
@@ -168,7 +168,7 @@ inline uint32x4_p VectorLoadBE(const uint8_t src[16])
 inline uint32x4_p VectorLoadBE(int off, const uint8_t src[16])
 {
 #if defined(CRYPTOPP_XLC_VERSION)
-    return (uint32x4_p)vec_xl_be(off, src);
+    return (uint32x4_p)vec_xl_be(off, (byte*)src);
 #else
 # if defined(CRYPTOPP_LITTLE_ENDIAN)
     return (uint32x4_p)Reverse(vec_vsx_ld(off, src));
@@ -213,7 +213,7 @@ inline uint32x4_p VectorLoad(int off, const byte src[16])
 inline uint32x4_p VectorLoadKey(const byte src[16])
 {
 #if defined(CRYPTOPP_XLC_VERSION)
-    return (uint32x4_p)vec_xl(0, src);
+    return (uint32x4_p)vec_xl(0, (byte*)src);
 #else
     return (uint32x4_p)vec_vsx_ld(0, src);
 #endif
@@ -229,7 +229,7 @@ inline uint32x4_p VectorLoadKey(const byte src[16])
 inline uint32x4_p VectorLoadKey(const word32 src[4])
 {
 #if defined(CRYPTOPP_XLC_VERSION)
-    return (uint32x4_p)vec_xl(0, src);
+    return (uint32x4_p)vec_xl(0, (byte*)src);
 #else
     return (uint32x4_p)vec_vsx_ld(0, src);
 #endif
@@ -246,7 +246,7 @@ inline uint32x4_p VectorLoadKey(const word32 src[4])
 inline uint32x4_p VectorLoadKey(int off, const byte src[16])
 {
 #if defined(CRYPTOPP_XLC_VERSION)
-    return (uint32x4_p)vec_xl(off, src);
+    return (uint32x4_p)vec_xl(off, (byte*)src);
 #else
     return (uint32x4_p)vec_vsx_ld(off, src);
 #endif
@@ -424,7 +424,7 @@ inline T1 VectorShiftLeft(const T1& vec1, const T2& vec2)
 
 #endif  // CRYPTOPP_POWER7_AVAILABLE
 
-#if defined(CRYPTOPP_POWER8_AVAILABLE) || defined(CRYPTOPP_DOXYGEN_PROCESSING)
+#if defined(CRYPTOPP_POWER8_AES_AVAILABLE) || defined(CRYPTOPP_DOXYGEN_PROCESSING)
 
 /// \brief One round of AES encryption
 /// \tparam T1 vector type
@@ -506,6 +506,10 @@ inline T1 VectorDecryptLast(const T1& state, const T2& key)
 #endif
 }
 
+#endif  // CRYPTOPP_POWER8_AES_AVAILABLE
+
+#if defined(CRYPTOPP_POWER8_SHA_AVAILABLE) || defined(CRYPTOPP_DOXYGEN_PROCESSING)
+
 /// \brief SHA256 Sigma functions
 /// \tparam func function
 /// \tparam subfunc sub-function
@@ -546,7 +550,7 @@ inline T VectorSHA512(const T& vec)
 #endif
 }
 
-#endif  // CRYPTOPP_POWER8_AVAILABLE
+#endif  // CRYPTOPP_POWER8_SHA_AVAILABLE
 
 NAMESPACE_END
 
diff --git a/sha-simd.cpp b/sha-simd.cpp
index b2639840..30d6068d 100644
--- a/sha-simd.cpp
+++ b/sha-simd.cpp
@@ -990,7 +990,7 @@ typedef __vector unsigned long long uint64x2_p8;
 uint32x4_p8 VEC_XL_BE(int offset, const uint8_t* data)
 {
 #if defined(CRYPTOPP_XLC_VERSION)
-    return vec_xl_be(offset, data);
+    return (uint32x4_p8)vec_xl_be(offset, (uint8_t*)data);
 #else
     uint32x4_p8 res;
     __asm(" lxvd2x  %x0, %1, %2    \n\t"
@@ -1016,7 +1016,7 @@ template <class T> static inline
 uint32x4_p8 VectorLoad32x4u(const T* data, int offset)
 {
 #if defined(CRYPTOPP_XLC_VERSION)
-    return (uint32x4_p8)vec_xl(offset, data);
+    return (uint32x4_p8)vec_xl(offset, (uint8_t*)data);
 #else
     return (uint32x4_p8)vec_vsx_ld(offset, data);
 #endif
@@ -1324,7 +1324,7 @@ template <class T> static inline
 uint64x2_p8 VectorLoad64x2u(const T* data, int offset)
 {
 #if defined(CRYPTOPP_XLC_VERSION)
-    return (uint64x2_p8)vec_xl(offset, (const uint8_t*)data);
+    return (uint64x2_p8)vec_xl(offset, (uint8_t*)data);
 #else
     return (uint64x2_p8)vec_vsx_ld(offset, (const uint8_t*)data);
 #endif