crypto: aesni-intel - Fixed build with binutils 2.16

This patch fixes the problem with 2.16 binutils.
Signed-off-by: Aidan O'Mahony <aidan.o.mahony@intel.com>
Signed-off-by: Adrian Hoban <adrian.hoban@intel.com>
Signed-off-by: Gabriele Paoloni <gabriele.paoloni@intel.com>
Signed-off-by: Tadeusz Struk <tadeusz.struk@intel.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>

arch/x86/crypto/aesni-intel_asm.S

@@ -204,9 +204,9 @@ enc:        .octa 0x2
 * arg1, %arg2, %arg3, %r14 are used as a pointer only, not modified
 */
 
-.macro INITIAL_BLOCKS num_initial_blocks TMP1 TMP2 TMP3 TMP4 TMP5 XMM0 XMM1 \
-XMM2 XMM3 XMM4 XMMDst TMP6 TMP7 i i_seq operation
 
+.macro INITIAL_BLOCKS_DEC num_initial_blocks TMP1 TMP2 TMP3 TMP4 TMP5 XMM0 XMM1 \
+XMM2 XMM3 XMM4 XMMDst TMP6 TMP7 i i_seq operation
 	mov	   arg7, %r10           # %r10 = AAD
 	mov	   arg8, %r12           # %r12 = aadLen
 	mov	   %r12, %r11
@@ -228,19 +228,25 @@ _get_AAD_loop2\num_initial_blocks\operation:
 	cmp	   %r11, %r12
 	jne	   _get_AAD_loop2\num_initial_blocks\operation
 _get_AAD_loop2_done\num_initial_blocks\operation:
-	pshufb	   SHUF_MASK(%rip), %xmm\i # byte-reflect the AAD data
+        movdqa     SHUF_MASK(%rip), %xmm14
+	PSHUFB_XMM   %xmm14, %xmm\i # byte-reflect the AAD data
+
 	xor	   %r11, %r11 # initialise the data pointer offset as zero
 
         # start AES for num_initial_blocks blocks
 
 	mov	   %arg5, %rax                      # %rax = *Y0
 	movdqu	   (%rax), \XMM0                    # XMM0 = Y0
-	pshufb	   SHUF_MASK(%rip), \XMM0
-.if \i_seq != 0
+        movdqa     SHUF_MASK(%rip), %xmm14
+	PSHUFB_XMM   %xmm14, \XMM0
+
+.if (\i == 5) || (\i == 6) || (\i == 7)
 .irpc index, \i_seq
 	paddd	   ONE(%rip), \XMM0                 # INCR Y0
 	movdqa	   \XMM0, %xmm\index
-	pshufb	   SHUF_MASK(%rip), %xmm\index      # perform a 16 byte swap
+        movdqa     SHUF_MASK(%rip), %xmm14
+	PSHUFB_XMM   %xmm14, %xmm\index      # perform a 16 byte swap
+
 .endr
 .irpc index, \i_seq
 	pxor	   16*0(%arg1), %xmm\index
@@ -291,10 +297,11 @@ _get_AAD_loop2_done\num_initial_blocks\operation:
 	movdqu	   %xmm\index, (%arg2 , %r11, 1)
 	# write back plaintext/ciphertext for num_initial_blocks
 	add	   $16, %r11
-.if \operation == dec
+
 	movdqa     \TMP1, %xmm\index
-.endif
-	pshufb	   SHUF_MASK(%rip), %xmm\index
+        movdqa     SHUF_MASK(%rip), %xmm14
+	PSHUFB_XMM	   %xmm14, %xmm\index
+
 		# prepare plaintext/ciphertext for GHASH computation
 .endr
 .endif
@@ -327,16 +334,24 @@ _get_AAD_loop2_done\num_initial_blocks\operation:
 */
 	paddd	   ONE(%rip), \XMM0              # INCR Y0
 	movdqa	   \XMM0, \XMM1
-	pshufb	   SHUF_MASK(%rip), \XMM1        # perform a 16 byte swap
+        movdqa     SHUF_MASK(%rip), %xmm14
+	PSHUFB_XMM  %xmm14, \XMM1        # perform a 16 byte swap
+
 	paddd	   ONE(%rip), \XMM0              # INCR Y0
 	movdqa	   \XMM0, \XMM2
-	pshufb	   SHUF_MASK(%rip), \XMM2        # perform a 16 byte swap
+        movdqa     SHUF_MASK(%rip), %xmm14
+	PSHUFB_XMM  %xmm14, \XMM2        # perform a 16 byte swap
+
 	paddd	   ONE(%rip), \XMM0              # INCR Y0
 	movdqa	   \XMM0, \XMM3
-	pshufb	   SHUF_MASK(%rip), \XMM3        # perform a 16 byte swap
+        movdqa     SHUF_MASK(%rip), %xmm14
+	PSHUFB_XMM %xmm14, \XMM3        # perform a 16 byte swap
+
 	paddd	   ONE(%rip), \XMM0              # INCR Y0
 	movdqa	   \XMM0, \XMM4
-	pshufb	   SHUF_MASK(%rip), \XMM4        # perform a 16 byte swap
+        movdqa     SHUF_MASK(%rip), %xmm14
+	PSHUFB_XMM %xmm14, \XMM4        # perform a 16 byte swap
+
 	pxor	   16*0(%arg1), \XMM1
 	pxor	   16*0(%arg1), \XMM2
 	pxor	   16*0(%arg1), \XMM3
@@ -385,41 +400,268 @@ _get_AAD_loop2_done\num_initial_blocks\operation:
 	AESENCLAST \TMP2, \XMM4
 	movdqu	   16*0(%arg3 , %r11 , 1), \TMP1
 	pxor	   \TMP1, \XMM1
-.if \operation == dec
 	movdqu	   \XMM1, 16*0(%arg2 , %r11 , 1)
 	movdqa     \TMP1, \XMM1
-.endif
 	movdqu	   16*1(%arg3 , %r11 , 1), \TMP1
 	pxor	   \TMP1, \XMM2
-.if \operation == dec
 	movdqu	   \XMM2, 16*1(%arg2 , %r11 , 1)
 	movdqa     \TMP1, \XMM2
-.endif
 	movdqu	   16*2(%arg3 , %r11 , 1), \TMP1
 	pxor	   \TMP1, \XMM3
-.if \operation == dec
 	movdqu	   \XMM3, 16*2(%arg2 , %r11 , 1)
 	movdqa     \TMP1, \XMM3
-.endif
 	movdqu	   16*3(%arg3 , %r11 , 1), \TMP1
 	pxor	   \TMP1, \XMM4
-.if \operation == dec
 	movdqu	   \XMM4, 16*3(%arg2 , %r11 , 1)
 	movdqa     \TMP1, \XMM4
-.else
+	add	   $64, %r11
+        movdqa     SHUF_MASK(%rip), %xmm14
+	PSHUFB_XMM %xmm14, \XMM1 # perform a 16 byte swap
+	pxor	   \XMMDst, \XMM1
+# combine GHASHed value with the corresponding ciphertext
+        movdqa     SHUF_MASK(%rip), %xmm14
+	PSHUFB_XMM %xmm14, \XMM2 # perform a 16 byte swap
+        movdqa     SHUF_MASK(%rip), %xmm14
+	PSHUFB_XMM %xmm14, \XMM3 # perform a 16 byte swap
+        movdqa     SHUF_MASK(%rip), %xmm14
+	PSHUFB_XMM %xmm14, \XMM4 # perform a 16 byte swap
+
+_initial_blocks_done\num_initial_blocks\operation:
+
+.endm
+
+
+/*
+* if a = number of total plaintext bytes
+* b = floor(a/16)
+* num_initial_blocks = b mod 4
+* encrypt the initial num_initial_blocks blocks and apply ghash on
+* the ciphertext
+* %r10, %r11, %r12, %rax, %xmm5, %xmm6, %xmm7, %xmm8, %xmm9 registers
+* are clobbered
+* arg1, %arg2, %arg3, %r14 are used as a pointer only, not modified
+*/
+
+
+.macro INITIAL_BLOCKS_ENC num_initial_blocks TMP1 TMP2 TMP3 TMP4 TMP5 XMM0 XMM1 \
+XMM2 XMM3 XMM4 XMMDst TMP6 TMP7 i i_seq operation
+	mov	   arg7, %r10           # %r10 = AAD
+	mov	   arg8, %r12           # %r12 = aadLen
+	mov	   %r12, %r11
+	pxor	   %xmm\i, %xmm\i
+_get_AAD_loop\num_initial_blocks\operation:
+	movd	   (%r10), \TMP1
+	pslldq	   $12, \TMP1
+	psrldq	   $4, %xmm\i
+	pxor	   \TMP1, %xmm\i
+	add	   $4, %r10
+	sub	   $4, %r12
+	jne	   _get_AAD_loop\num_initial_blocks\operation
+	cmp	   $16, %r11
+	je	   _get_AAD_loop2_done\num_initial_blocks\operation
+	mov	   $16, %r12
+_get_AAD_loop2\num_initial_blocks\operation:
+	psrldq	   $4, %xmm\i
+	sub	   $4, %r12
+	cmp	   %r11, %r12
+	jne	   _get_AAD_loop2\num_initial_blocks\operation
+_get_AAD_loop2_done\num_initial_blocks\operation:
+        movdqa     SHUF_MASK(%rip), %xmm14
+	PSHUFB_XMM   %xmm14, %xmm\i # byte-reflect the AAD data
+
+	xor	   %r11, %r11 # initialise the data pointer offset as zero
+
+        # start AES for num_initial_blocks blocks
+
+	mov	   %arg5, %rax                      # %rax = *Y0
+	movdqu	   (%rax), \XMM0                    # XMM0 = Y0
+        movdqa     SHUF_MASK(%rip), %xmm14
+	PSHUFB_XMM   %xmm14, \XMM0
+
+.if (\i == 5) || (\i == 6) || (\i == 7)
+.irpc index, \i_seq
+	paddd	   ONE(%rip), \XMM0                 # INCR Y0
+	movdqa	   \XMM0, %xmm\index
+        movdqa     SHUF_MASK(%rip), %xmm14
+	PSHUFB_XMM   %xmm14, %xmm\index      # perform a 16 byte swap
+
+.endr
+.irpc index, \i_seq
+	pxor	   16*0(%arg1), %xmm\index
+.endr
+.irpc index, \i_seq
+	movaps 0x10(%rdi), \TMP1
+	AESENC     \TMP1, %xmm\index          # Round 1
+.endr
+.irpc index, \i_seq
+	movaps 0x20(%arg1), \TMP1
+	AESENC     \TMP1, %xmm\index          # Round 2
+.endr
+.irpc index, \i_seq
+	movaps 0x30(%arg1), \TMP1
+	AESENC     \TMP1, %xmm\index          # Round 2
+.endr
+.irpc index, \i_seq
+	movaps 0x40(%arg1), \TMP1
+	AESENC     \TMP1, %xmm\index          # Round 2
+.endr
+.irpc index, \i_seq
+	movaps 0x50(%arg1), \TMP1
+	AESENC     \TMP1, %xmm\index          # Round 2
+.endr
+.irpc index, \i_seq
+	movaps 0x60(%arg1), \TMP1
+	AESENC     \TMP1, %xmm\index          # Round 2
+.endr
+.irpc index, \i_seq
+	movaps 0x70(%arg1), \TMP1
+	AESENC     \TMP1, %xmm\index          # Round 2
+.endr
+.irpc index, \i_seq
+	movaps 0x80(%arg1), \TMP1
+	AESENC     \TMP1, %xmm\index          # Round 2
+.endr
+.irpc index, \i_seq
+	movaps 0x90(%arg1), \TMP1
+	AESENC     \TMP1, %xmm\index          # Round 2
+.endr
+.irpc index, \i_seq
+	movaps 0xa0(%arg1), \TMP1
+	AESENCLAST \TMP1, %xmm\index         # Round 10
+.endr
+.irpc index, \i_seq
+	movdqu	   (%arg3 , %r11, 1), \TMP1
+	pxor	   \TMP1, %xmm\index
+	movdqu	   %xmm\index, (%arg2 , %r11, 1)
+	# write back plaintext/ciphertext for num_initial_blocks
+	add	   $16, %r11
+
+        movdqa     SHUF_MASK(%rip), %xmm14
+	PSHUFB_XMM	   %xmm14, %xmm\index
+
+		# prepare plaintext/ciphertext for GHASH computation
+.endr
+.endif
+	GHASH_MUL  %xmm\i, \TMP3, \TMP1, \TMP2, \TMP4, \TMP5, \XMM1
+        # apply GHASH on num_initial_blocks blocks
+
+.if \i == 5
+        pxor       %xmm5, %xmm6
+	GHASH_MUL  %xmm6, \TMP3, \TMP1, \TMP2, \TMP4, \TMP5, \XMM1
+        pxor       %xmm6, %xmm7
+	GHASH_MUL  %xmm7, \TMP3, \TMP1, \TMP2, \TMP4, \TMP5, \XMM1
+        pxor       %xmm7, %xmm8
+	GHASH_MUL  %xmm8, \TMP3, \TMP1, \TMP2, \TMP4, \TMP5, \XMM1
+.elseif \i == 6
+        pxor       %xmm6, %xmm7
+	GHASH_MUL  %xmm7, \TMP3, \TMP1, \TMP2, \TMP4, \TMP5, \XMM1
+        pxor       %xmm7, %xmm8
+	GHASH_MUL  %xmm8, \TMP3, \TMP1, \TMP2, \TMP4, \TMP5, \XMM1
+.elseif \i == 7
+        pxor       %xmm7, %xmm8
+	GHASH_MUL  %xmm8, \TMP3, \TMP1, \TMP2, \TMP4, \TMP5, \XMM1
+.endif
+	cmp	   $64, %r13
+	jl	_initial_blocks_done\num_initial_blocks\operation
+	# no need for precomputed values
+/*
+*
+* Precomputations for HashKey parallel with encryption of first 4 blocks.
+* Haskey_i_k holds XORed values of the low and high parts of the Haskey_i
+*/
+	paddd	   ONE(%rip), \XMM0              # INCR Y0
+	movdqa	   \XMM0, \XMM1
+        movdqa     SHUF_MASK(%rip), %xmm14
+	PSHUFB_XMM  %xmm14, \XMM1        # perform a 16 byte swap
+
+	paddd	   ONE(%rip), \XMM0              # INCR Y0
+	movdqa	   \XMM0, \XMM2
+        movdqa     SHUF_MASK(%rip), %xmm14
+	PSHUFB_XMM  %xmm14, \XMM2        # perform a 16 byte swap
+
+	paddd	   ONE(%rip), \XMM0              # INCR Y0
+	movdqa	   \XMM0, \XMM3
+        movdqa     SHUF_MASK(%rip), %xmm14
+	PSHUFB_XMM %xmm14, \XMM3        # perform a 16 byte swap
+
+	paddd	   ONE(%rip), \XMM0              # INCR Y0
+	movdqa	   \XMM0, \XMM4
+        movdqa     SHUF_MASK(%rip), %xmm14
+	PSHUFB_XMM %xmm14, \XMM4        # perform a 16 byte swap
+
+	pxor	   16*0(%arg1), \XMM1
+	pxor	   16*0(%arg1), \XMM2
+	pxor	   16*0(%arg1), \XMM3
+	pxor	   16*0(%arg1), \XMM4
+	movdqa	   \TMP3, \TMP5
+	pshufd	   $78, \TMP3, \TMP1
+	pxor	   \TMP3, \TMP1
+	movdqa	   \TMP1, HashKey_k(%rsp)
+	GHASH_MUL  \TMP5, \TMP3, \TMP1, \TMP2, \TMP4, \TMP6, \TMP7
+# TMP5 = HashKey^2<<1 (mod poly)
+	movdqa	   \TMP5, HashKey_2(%rsp)
+# HashKey_2 = HashKey^2<<1 (mod poly)
+	pshufd	   $78, \TMP5, \TMP1
+	pxor	   \TMP5, \TMP1
+	movdqa	   \TMP1, HashKey_2_k(%rsp)
+.irpc index, 1234 # do 4 rounds
+	movaps 0x10*\index(%arg1), \TMP1
+	AESENC	   \TMP1, \XMM1
+	AESENC	   \TMP1, \XMM2
+	AESENC	   \TMP1, \XMM3
+	AESENC	   \TMP1, \XMM4
+.endr
+	GHASH_MUL  \TMP5, \TMP3, \TMP1, \TMP2, \TMP4, \TMP6, \TMP7
+# TMP5 = HashKey^3<<1 (mod poly)
+	movdqa	   \TMP5, HashKey_3(%rsp)
+	pshufd	   $78, \TMP5, \TMP1
+	pxor	   \TMP5, \TMP1
+	movdqa	   \TMP1, HashKey_3_k(%rsp)
+.irpc index, 56789 # do next 5 rounds
+	movaps 0x10*\index(%arg1), \TMP1
+	AESENC	   \TMP1, \XMM1
+	AESENC	   \TMP1, \XMM2
+	AESENC	   \TMP1, \XMM3
+	AESENC	   \TMP1, \XMM4
+.endr
+	GHASH_MUL  \TMP5, \TMP3, \TMP1, \TMP2, \TMP4, \TMP6, \TMP7
+# TMP5 = HashKey^3<<1 (mod poly)
+	movdqa	   \TMP5, HashKey_4(%rsp)
+	pshufd	   $78, \TMP5, \TMP1
+	pxor	   \TMP5, \TMP1
+	movdqa	   \TMP1, HashKey_4_k(%rsp)
+	movaps 0xa0(%arg1), \TMP2
+	AESENCLAST \TMP2, \XMM1
+	AESENCLAST \TMP2, \XMM2
+	AESENCLAST \TMP2, \XMM3
+	AESENCLAST \TMP2, \XMM4
+	movdqu	   16*0(%arg3 , %r11 , 1), \TMP1
+	pxor	   \TMP1, \XMM1
+	movdqu	   16*1(%arg3 , %r11 , 1), \TMP1
+	pxor	   \TMP1, \XMM2
+	movdqu	   16*2(%arg3 , %r11 , 1), \TMP1
+	pxor	   \TMP1, \XMM3
+	movdqu	   16*3(%arg3 , %r11 , 1), \TMP1
+	pxor	   \TMP1, \XMM4
 	movdqu     \XMM1, 16*0(%arg2 , %r11 , 1)
 	movdqu     \XMM2, 16*1(%arg2 , %r11 , 1)
 	movdqu     \XMM3, 16*2(%arg2 , %r11 , 1)
 	movdqu     \XMM4, 16*3(%arg2 , %r11 , 1)
-.endif
+
 	add	   $64, %r11
-	pshufb	   SHUF_MASK(%rip), \XMM1 # perform a 16 byte swap
+        movdqa     SHUF_MASK(%rip), %xmm14
+	PSHUFB_XMM %xmm14, \XMM1 # perform a 16 byte swap
 	pxor	   \XMMDst, \XMM1
 # combine GHASHed value with the corresponding ciphertext
-	pshufb	   SHUF_MASK(%rip), \XMM2 # perform a 16 byte swap
-	pshufb	   SHUF_MASK(%rip), \XMM3 # perform a 16 byte swap
-	pshufb	   SHUF_MASK(%rip), \XMM4 # perform a 16 byte swap
+        movdqa     SHUF_MASK(%rip), %xmm14
+	PSHUFB_XMM %xmm14, \XMM2 # perform a 16 byte swap
+        movdqa     SHUF_MASK(%rip), %xmm14
+	PSHUFB_XMM %xmm14, \XMM3 # perform a 16 byte swap
+        movdqa     SHUF_MASK(%rip), %xmm14
+	PSHUFB_XMM %xmm14, \XMM4 # perform a 16 byte swap
+
 _initial_blocks_done\num_initial_blocks\operation:
+
 .endm
 
 /*
@@ -428,7 +670,199 @@ _initial_blocks_done\num_initial_blocks\operation:
 * arg1, %arg2, %arg3 are used as pointers only, not modified
 * %r11 is the data offset value
 */
-.macro GHASH_4_ENCRYPT_4_PARALLEL TMP1 TMP2 TMP3 TMP4 TMP5 \
+.macro GHASH_4_ENCRYPT_4_PARALLEL_ENC TMP1 TMP2 TMP3 TMP4 TMP5 \
+TMP6 XMM0 XMM1 XMM2 XMM3 XMM4 XMM5 XMM6 XMM7 XMM8 operation
+
+	movdqa	  \XMM1, \XMM5
+	movdqa	  \XMM2, \XMM6
+	movdqa	  \XMM3, \XMM7
+	movdqa	  \XMM4, \XMM8
+
+        movdqa    SHUF_MASK(%rip), %xmm15
+        # multiply TMP5 * HashKey using karatsuba
+
+	movdqa	  \XMM5, \TMP4
+	pshufd	  $78, \XMM5, \TMP6
+	pxor	  \XMM5, \TMP6
+	paddd     ONE(%rip), \XMM0		# INCR CNT
+	movdqa	  HashKey_4(%rsp), \TMP5
+	PCLMULQDQ 0x11, \TMP5, \TMP4           # TMP4 = a1*b1
+	movdqa    \XMM0, \XMM1
+	paddd     ONE(%rip), \XMM0		# INCR CNT
+	movdqa    \XMM0, \XMM2
+	paddd     ONE(%rip), \XMM0		# INCR CNT
+	movdqa    \XMM0, \XMM3
+	paddd     ONE(%rip), \XMM0		# INCR CNT
+	movdqa    \XMM0, \XMM4
+	PSHUFB_XMM %xmm15, \XMM1	# perform a 16 byte swap
+	PCLMULQDQ 0x00, \TMP5, \XMM5           # XMM5 = a0*b0
+	PSHUFB_XMM %xmm15, \XMM2	# perform a 16 byte swap
+	PSHUFB_XMM %xmm15, \XMM3	# perform a 16 byte swap
+	PSHUFB_XMM %xmm15, \XMM4	# perform a 16 byte swap
+
+	pxor	  (%arg1), \XMM1
+	pxor	  (%arg1), \XMM2
+	pxor	  (%arg1), \XMM3
+	pxor	  (%arg1), \XMM4
+	movdqa	  HashKey_4_k(%rsp), \TMP5
+	PCLMULQDQ 0x00, \TMP5, \TMP6           # TMP6 = (a1+a0)*(b1+b0)
+	movaps 0x10(%arg1), \TMP1
+	AESENC	  \TMP1, \XMM1              # Round 1
+	AESENC	  \TMP1, \XMM2
+	AESENC	  \TMP1, \XMM3
+	AESENC	  \TMP1, \XMM4
+	movaps 0x20(%arg1), \TMP1
+	AESENC	  \TMP1, \XMM1              # Round 2
+	AESENC	  \TMP1, \XMM2
+	AESENC	  \TMP1, \XMM3
+	AESENC	  \TMP1, \XMM4
+	movdqa	  \XMM6, \TMP1
+	pshufd	  $78, \XMM6, \TMP2
+	pxor	  \XMM6, \TMP2
+	movdqa	  HashKey_3(%rsp), \TMP5
+	PCLMULQDQ 0x11, \TMP5, \TMP1           # TMP1 = a1 * b1
+	movaps 0x30(%arg1), \TMP3
+	AESENC    \TMP3, \XMM1              # Round 3
+	AESENC    \TMP3, \XMM2
+	AESENC    \TMP3, \XMM3
+	AESENC    \TMP3, \XMM4
+	PCLMULQDQ 0x00, \TMP5, \XMM6           # XMM6 = a0*b0
+	movaps 0x40(%arg1), \TMP3
+	AESENC	  \TMP3, \XMM1              # Round 4
+	AESENC	  \TMP3, \XMM2
+	AESENC	  \TMP3, \XMM3
+	AESENC	  \TMP3, \XMM4
+	movdqa	  HashKey_3_k(%rsp), \TMP5
+	PCLMULQDQ 0x00, \TMP5, \TMP2           # TMP2 = (a1+a0)*(b1+b0)
+	movaps 0x50(%arg1), \TMP3
+	AESENC	  \TMP3, \XMM1              # Round 5
+	AESENC	  \TMP3, \XMM2
+	AESENC	  \TMP3, \XMM3
+	AESENC	  \TMP3, \XMM4
+	pxor	  \TMP1, \TMP4
+# accumulate the results in TMP4:XMM5, TMP6 holds the middle part
+	pxor	  \XMM6, \XMM5
+	pxor	  \TMP2, \TMP6
+	movdqa	  \XMM7, \TMP1
+	pshufd	  $78, \XMM7, \TMP2
+	pxor	  \XMM7, \TMP2
+	movdqa	  HashKey_2(%rsp ), \TMP5
+
+        # Multiply TMP5 * HashKey using karatsuba
+
+	PCLMULQDQ 0x11, \TMP5, \TMP1           # TMP1 = a1*b1
+	movaps 0x60(%arg1), \TMP3
+	AESENC	  \TMP3, \XMM1              # Round 6
+	AESENC	  \TMP3, \XMM2
+	AESENC	  \TMP3, \XMM3
+	AESENC	  \TMP3, \XMM4
+	PCLMULQDQ 0x00, \TMP5, \XMM7           # XMM7 = a0*b0
+	movaps 0x70(%arg1), \TMP3
+	AESENC	  \TMP3, \XMM1             # Round 7
+	AESENC	  \TMP3, \XMM2
+	AESENC	  \TMP3, \XMM3
+	AESENC	  \TMP3, \XMM4
+	movdqa	  HashKey_2_k(%rsp), \TMP5
+	PCLMULQDQ 0x00, \TMP5, \TMP2           # TMP2 = (a1+a0)*(b1+b0)
+	movaps 0x80(%arg1), \TMP3
+	AESENC	  \TMP3, \XMM1             # Round 8
+	AESENC	  \TMP3, \XMM2
+	AESENC	  \TMP3, \XMM3
+	AESENC	  \TMP3, \XMM4
+	pxor	  \TMP1, \TMP4
+# accumulate the results in TMP4:XMM5, TMP6 holds the middle part
+	pxor	  \XMM7, \XMM5
+	pxor	  \TMP2, \TMP6
+
+        # Multiply XMM8 * HashKey
+        # XMM8 and TMP5 hold the values for the two operands
+
+	movdqa	  \XMM8, \TMP1
+	pshufd	  $78, \XMM8, \TMP2
+	pxor	  \XMM8, \TMP2
+	movdqa	  HashKey(%rsp), \TMP5
+	PCLMULQDQ 0x11, \TMP5, \TMP1          # TMP1 = a1*b1
+	movaps 0x90(%arg1), \TMP3
+	AESENC	  \TMP3, \XMM1            # Round 9
+	AESENC	  \TMP3, \XMM2
+	AESENC	  \TMP3, \XMM3
+	AESENC	  \TMP3, \XMM4
+	PCLMULQDQ 0x00, \TMP5, \XMM8          # XMM8 = a0*b0
+	movaps 0xa0(%arg1), \TMP3
+	AESENCLAST \TMP3, \XMM1           # Round 10
+	AESENCLAST \TMP3, \XMM2
+	AESENCLAST \TMP3, \XMM3
+	AESENCLAST \TMP3, \XMM4
+	movdqa    HashKey_k(%rsp), \TMP5
+	PCLMULQDQ 0x00, \TMP5, \TMP2          # TMP2 = (a1+a0)*(b1+b0)
+	movdqu	  (%arg3,%r11,1), \TMP3
+	pxor	  \TMP3, \XMM1                 # Ciphertext/Plaintext XOR EK
+	movdqu	  16(%arg3,%r11,1), \TMP3
+	pxor	  \TMP3, \XMM2                 # Ciphertext/Plaintext XOR EK
+	movdqu	  32(%arg3,%r11,1), \TMP3
+	pxor	  \TMP3, \XMM3                 # Ciphertext/Plaintext XOR EK
+	movdqu	  48(%arg3,%r11,1), \TMP3
+	pxor	  \TMP3, \XMM4                 # Ciphertext/Plaintext XOR EK
+        movdqu    \XMM1, (%arg2,%r11,1)        # Write to the ciphertext buffer
+        movdqu    \XMM2, 16(%arg2,%r11,1)      # Write to the ciphertext buffer
+        movdqu    \XMM3, 32(%arg2,%r11,1)      # Write to the ciphertext buffer
+        movdqu    \XMM4, 48(%arg2,%r11,1)      # Write to the ciphertext buffer
+	PSHUFB_XMM %xmm15, \XMM1        # perform a 16 byte swap
+	PSHUFB_XMM %xmm15, \XMM2	# perform a 16 byte swap
+	PSHUFB_XMM %xmm15, \XMM3	# perform a 16 byte swap
+	PSHUFB_XMM %xmm15, \XMM4	# perform a 16 byte swap
+
+	pxor	  \TMP4, \TMP1
+	pxor	  \XMM8, \XMM5
+	pxor	  \TMP6, \TMP2
+	pxor	  \TMP1, \TMP2
+	pxor	  \XMM5, \TMP2
+	movdqa	  \TMP2, \TMP3
+	pslldq	  $8, \TMP3                    # left shift TMP3 2 DWs
+	psrldq	  $8, \TMP2                    # right shift TMP2 2 DWs
+	pxor	  \TMP3, \XMM5
+	pxor	  \TMP2, \TMP1	  # accumulate the results in TMP1:XMM5
+
+        # first phase of reduction
+
+	movdqa    \XMM5, \TMP2
+	movdqa    \XMM5, \TMP3
+	movdqa    \XMM5, \TMP4
+# move XMM5 into TMP2, TMP3, TMP4 in order to perform shifts independently
+	pslld     $31, \TMP2                   # packed right shift << 31
+	pslld     $30, \TMP3                   # packed right shift << 30
+	pslld     $25, \TMP4                   # packed right shift << 25
+	pxor      \TMP3, \TMP2	               # xor the shifted versions
+	pxor      \TMP4, \TMP2
+	movdqa    \TMP2, \TMP5
+	psrldq    $4, \TMP5                    # right shift T5 1 DW
+	pslldq    $12, \TMP2                   # left shift T2 3 DWs
+	pxor      \TMP2, \XMM5
+
+        # second phase of reduction
+
+	movdqa    \XMM5,\TMP2 # make 3 copies of XMM5 into TMP2, TMP3, TMP4
+	movdqa    \XMM5,\TMP3
+	movdqa    \XMM5,\TMP4
+	psrld     $1, \TMP2                    # packed left shift >>1
+	psrld     $2, \TMP3                    # packed left shift >>2
+	psrld     $7, \TMP4                    # packed left shift >>7
+	pxor      \TMP3,\TMP2		       # xor the shifted versions
+	pxor      \TMP4,\TMP2
+	pxor      \TMP5, \TMP2
+	pxor      \TMP2, \XMM5
+	pxor      \TMP1, \XMM5                 # result is in TMP1
+
+	pxor	  \XMM5, \XMM1
+.endm
+
+/*
+* decrypt 4 blocks at a time
+* ghash the 4 previously decrypted ciphertext blocks
+* arg1, %arg2, %arg3 are used as pointers only, not modified
+* %r11 is the data offset value
+*/
+.macro GHASH_4_ENCRYPT_4_PARALLEL_DEC TMP1 TMP2 TMP3 TMP4 TMP5 \
 TMP6 XMM0 XMM1 XMM2 XMM3 XMM4 XMM5 XMM6 XMM7 XMM8 operation
 
 	movdqa	  \XMM1, \XMM5
@@ -436,6 +870,7 @@ TMP6 XMM0 XMM1 XMM2 XMM3 XMM4 XMM5 XMM6 XMM7 XMM8 operation
 	movdqa	  \XMM3, \XMM7
 	movdqa	  \XMM4, \XMM8
 
+        movdqa    SHUF_MASK(%rip), %xmm15
         # multiply TMP5 * HashKey using karatsuba
 
 	movdqa	  \XMM5, \TMP4
@@ -451,11 +886,12 @@ TMP6 XMM0 XMM1 XMM2 XMM3 XMM4 XMM5 XMM6 XMM7 XMM8 operation
 	movdqa    \XMM0, \XMM3
 	paddd     ONE(%rip), \XMM0		# INCR CNT
 	movdqa    \XMM0, \XMM4
-	pshufb	  SHUF_MASK(%rip), \XMM1	# perform a 16 byte swap
+	PSHUFB_XMM %xmm15, \XMM1	# perform a 16 byte swap
 	PCLMULQDQ 0x00, \TMP5, \XMM5           # XMM5 = a0*b0
-	pshufb	  SHUF_MASK(%rip), \XMM2	# perform a 16 byte swap
-	pshufb	  SHUF_MASK(%rip), \XMM3	# perform a 16 byte swap
-	pshufb	  SHUF_MASK(%rip), \XMM4	# perform a 16 byte swap
+	PSHUFB_XMM %xmm15, \XMM2	# perform a 16 byte swap
+	PSHUFB_XMM %xmm15, \XMM3	# perform a 16 byte swap
+	PSHUFB_XMM %xmm15, \XMM4	# perform a 16 byte swap
+
 	pxor	  (%arg1), \XMM1
 	pxor	  (%arg1), \XMM2
 	pxor	  (%arg1), \XMM3
@@ -553,37 +989,24 @@ TMP6 XMM0 XMM1 XMM2 XMM3 XMM4 XMM5 XMM6 XMM7 XMM8 operation
 	PCLMULQDQ 0x00, \TMP5, \TMP2          # TMP2 = (a1+a0)*(b1+b0)
 	movdqu	  (%arg3,%r11,1), \TMP3
 	pxor	  \TMP3, \XMM1                 # Ciphertext/Plaintext XOR EK
-.if \operation == dec
 	movdqu	  \XMM1, (%arg2,%r11,1)        # Write to plaintext buffer
 	movdqa    \TMP3, \XMM1
-.endif
 	movdqu	  16(%arg3,%r11,1), \TMP3
 	pxor	  \TMP3, \XMM2                 # Ciphertext/Plaintext XOR EK
-.if \operation == dec
 	movdqu	  \XMM2, 16(%arg2,%r11,1)      # Write to plaintext buffer
 	movdqa    \TMP3, \XMM2
-.endif
 	movdqu	  32(%arg3,%r11,1), \TMP3
 	pxor	  \TMP3, \XMM3                 # Ciphertext/Plaintext XOR EK
-.if \operation == dec
 	movdqu	  \XMM3, 32(%arg2,%r11,1)      # Write to plaintext buffer
 	movdqa    \TMP3, \XMM3
-.endif
 	movdqu	  48(%arg3,%r11,1), \TMP3
 	pxor	  \TMP3, \XMM4                 # Ciphertext/Plaintext XOR EK
-.if \operation == dec
 	movdqu	  \XMM4, 48(%arg2,%r11,1)      # Write to plaintext buffer
 	movdqa    \TMP3, \XMM4
-.else
-    movdqu    \XMM1, (%arg2,%r11,1)        # Write to the ciphertext buffer
-    movdqu    \XMM2, 16(%arg2,%r11,1)      # Write to the ciphertext buffer
-    movdqu    \XMM3, 32(%arg2,%r11,1)      # Write to the ciphertext buffer
-    movdqu    \XMM4, 48(%arg2,%r11,1)      # Write to the ciphertext buffer
-.endif
-	pshufb	  SHUF_MASK(%rip), \XMM1       # perform a 16 byte swap
-	pshufb	  SHUF_MASK(%rip), \XMM2       # perform a 16 byte swap
-	pshufb	  SHUF_MASK(%rip), \XMM3       # perform a 16 byte swap
-	pshufb	  SHUF_MASK(%rip), \XMM4       # perform a 16 byte sway
+	PSHUFB_XMM %xmm15, \XMM1        # perform a 16 byte swap
+	PSHUFB_XMM %xmm15, \XMM2	# perform a 16 byte swap
+	PSHUFB_XMM %xmm15, \XMM3	# perform a 16 byte swap
+	PSHUFB_XMM %xmm15, \XMM4	# perform a 16 byte swap
 
 	pxor	  \TMP4, \TMP1
 	pxor	  \XMM8, \XMM5
@@ -853,7 +1276,9 @@ ENTRY(aesni_gcm_dec)
 	and	$~63, %rsp                        # align rsp to 64 bytes
 	mov	%arg6, %r12
 	movdqu	(%r12), %xmm13			  # %xmm13 = HashKey
-	pshufb	SHUF_MASK(%rip), %xmm13
+        movdqa  SHUF_MASK(%rip), %xmm2
+	PSHUFB_XMM %xmm2, %xmm13
+
 
 # Precompute HashKey<<1 (mod poly) from the hash key (required for GHASH)
 
@@ -885,22 +1310,22 @@ ENTRY(aesni_gcm_dec)
 	jb _initial_num_blocks_is_1_decrypt
 	je _initial_num_blocks_is_2_decrypt
 _initial_num_blocks_is_3_decrypt:
-	INITIAL_BLOCKS 3, %xmm9, %xmm10, %xmm13, %xmm11, %xmm12, %xmm0, \
+	INITIAL_BLOCKS_DEC 3, %xmm9, %xmm10, %xmm13, %xmm11, %xmm12, %xmm0, \
 %xmm1, %xmm2, %xmm3, %xmm4, %xmm8, %xmm5, %xmm6, 5, 678, dec
 	sub	$48, %r13
 	jmp	_initial_blocks_decrypted
 _initial_num_blocks_is_2_decrypt:
-	INITIAL_BLOCKS	2, %xmm9, %xmm10, %xmm13, %xmm11, %xmm12, %xmm0, \
+	INITIAL_BLOCKS_DEC	2, %xmm9, %xmm10, %xmm13, %xmm11, %xmm12, %xmm0, \
 %xmm1, %xmm2, %xmm3, %xmm4, %xmm8, %xmm5, %xmm6, 6, 78, dec
 	sub	$32, %r13
 	jmp	_initial_blocks_decrypted
 _initial_num_blocks_is_1_decrypt:
-	INITIAL_BLOCKS	1, %xmm9, %xmm10, %xmm13, %xmm11, %xmm12, %xmm0, \
+	INITIAL_BLOCKS_DEC	1, %xmm9, %xmm10, %xmm13, %xmm11, %xmm12, %xmm0, \
 %xmm1, %xmm2, %xmm3, %xmm4, %xmm8, %xmm5, %xmm6, 7, 8, dec
 	sub	$16, %r13
 	jmp	_initial_blocks_decrypted
 _initial_num_blocks_is_0_decrypt:
-	INITIAL_BLOCKS	0, %xmm9, %xmm10, %xmm13, %xmm11, %xmm12, %xmm0, \
+	INITIAL_BLOCKS_DEC	0, %xmm9, %xmm10, %xmm13, %xmm11, %xmm12, %xmm0, \
 %xmm1, %xmm2, %xmm3, %xmm4, %xmm8, %xmm5, %xmm6, 8, 0, dec
 _initial_blocks_decrypted:
 	cmp	$0, %r13
@@ -908,7 +1333,7 @@ _initial_blocks_decrypted:
 	sub	$64, %r13
 	je	_four_cipher_left_decrypt
 _decrypt_by_4:
-	GHASH_4_ENCRYPT_4_PARALLEL	%xmm9, %xmm10, %xmm11, %xmm12, %xmm13, \
+	GHASH_4_ENCRYPT_4_PARALLEL_DEC	%xmm9, %xmm10, %xmm11, %xmm12, %xmm13, \
 %xmm14, %xmm0, %xmm1, %xmm2, %xmm3, %xmm4, %xmm5, %xmm6, %xmm7, %xmm8, dec
 	add	$64, %r11
 	sub	$64, %r13
@@ -924,7 +1349,9 @@ _zero_cipher_left_decrypt:
         # Handle the last <16 byte block seperately
 
 	paddd ONE(%rip), %xmm0         # increment CNT to get Yn
-	pshufb SHUF_MASK(%rip), %xmm0
+        movdqa SHUF_MASK(%rip), %xmm10
+	PSHUFB_XMM %xmm10, %xmm0
+
 	ENCRYPT_SINGLE_BLOCK  %xmm0, %xmm1    # E(K, Yn)
 	sub $16, %r11
 	add %r13, %r11
@@ -934,14 +1361,17 @@ _zero_cipher_left_decrypt:
 # adjust the shuffle mask pointer to be able to shift 16-%r13 bytes
 # (%r13 is the number of bytes in plaintext mod 16)
 	movdqu (%r12), %xmm2           # get the appropriate shuffle mask
-	pshufb %xmm2, %xmm1            # right shift 16-%r13 butes
+	PSHUFB_XMM %xmm2, %xmm1            # right shift 16-%r13 butes
+
 	movdqa  %xmm1, %xmm2
 	pxor %xmm1, %xmm0            # Ciphertext XOR E(K, Yn)
 	movdqu ALL_F-SHIFT_MASK(%r12), %xmm1
 	# get the appropriate mask to mask out top 16-%r13 bytes of %xmm0
 	pand %xmm1, %xmm0            # mask out top 16-%r13 bytes of %xmm0
 	pand    %xmm1, %xmm2
-	pshufb SHUF_MASK(%rip),%xmm2
+        movdqa SHUF_MASK(%rip), %xmm10
+	PSHUFB_XMM %xmm10 ,%xmm2
+
 	pxor %xmm2, %xmm8
 	GHASH_MUL %xmm8, %xmm13, %xmm9, %xmm10, %xmm11, %xmm5, %xmm6
 	          # GHASH computation for the last <16 byte block
@@ -949,13 +1379,13 @@ _zero_cipher_left_decrypt:
 	add $16, %r11
 
         # output %r13 bytes
-	movq	%xmm0, %rax
+	MOVQ_R64_XMM	%xmm0, %rax
 	cmp	$8, %r13
 	jle	_less_than_8_bytes_left_decrypt
 	mov	%rax, (%arg2 , %r11, 1)
 	add	$8, %r11
 	psrldq	$8, %xmm0
-	movq	%xmm0, %rax
+	MOVQ_R64_XMM	%xmm0, %rax
 	sub	$8, %r13
 _less_than_8_bytes_left_decrypt:
 	mov	%al,  (%arg2, %r11, 1)
@@ -968,13 +1398,15 @@ _multiple_of_16_bytes_decrypt:
 	shl	$3, %r12		  # convert into number of bits
 	movd	%r12d, %xmm15		  # len(A) in %xmm15
 	shl	$3, %arg4		  # len(C) in bits (*128)
-	movq	%arg4, %xmm1
+	MOVQ_R64_XMM	%arg4, %xmm1
 	pslldq	$8, %xmm15		  # %xmm15 = len(A)||0x0000000000000000
 	pxor	%xmm1, %xmm15		  # %xmm15 = len(A)||len(C)
 	pxor	%xmm15, %xmm8
 	GHASH_MUL	%xmm8, %xmm13, %xmm9, %xmm10, %xmm11, %xmm5, %xmm6
 	         # final GHASH computation
-	pshufb	SHUF_MASK(%rip), %xmm8
+        movdqa SHUF_MASK(%rip), %xmm10
+	PSHUFB_XMM %xmm10, %xmm8
+
 	mov	%arg5, %rax		  # %rax = *Y0
 	movdqu	(%rax), %xmm0		  # %xmm0 = Y0
 	ENCRYPT_SINGLE_BLOCK	%xmm0,  %xmm1	  # E(K, Y0)
@@ -987,11 +1419,11 @@ _return_T_decrypt:
 	cmp	$12, %r11
 	je	_T_12_decrypt
 _T_8_decrypt:
-	movq	%xmm0, %rax
+	MOVQ_R64_XMM	%xmm0, %rax
 	mov	%rax, (%r10)
 	jmp	_return_T_done_decrypt
 _T_12_decrypt:
-	movq	%xmm0, %rax
+	MOVQ_R64_XMM	%xmm0, %rax
 	mov	%rax, (%r10)
 	psrldq	$8, %xmm0
 	movd	%xmm0, %eax
@@ -1103,7 +1535,9 @@ ENTRY(aesni_gcm_enc)
 	and	$~63, %rsp
 	mov	%arg6, %r12
 	movdqu	(%r12), %xmm13
-	pshufb	SHUF_MASK(%rip), %xmm13
+        movdqa  SHUF_MASK(%rip), %xmm2
+	PSHUFB_XMM %xmm2, %xmm13
+
 
 # precompute HashKey<<1 mod poly from the HashKey (required for GHASH)
 
@@ -1134,22 +1568,22 @@ ENTRY(aesni_gcm_enc)
 	jb	_initial_num_blocks_is_1_encrypt
 	je	_initial_num_blocks_is_2_encrypt
 _initial_num_blocks_is_3_encrypt:
-	INITIAL_BLOCKS	3, %xmm9, %xmm10, %xmm13, %xmm11, %xmm12, %xmm0, \
+	INITIAL_BLOCKS_ENC	3, %xmm9, %xmm10, %xmm13, %xmm11, %xmm12, %xmm0, \
 %xmm1, %xmm2, %xmm3, %xmm4, %xmm8, %xmm5, %xmm6, 5, 678, enc
 	sub	$48, %r13
 	jmp	_initial_blocks_encrypted
 _initial_num_blocks_is_2_encrypt:
-	INITIAL_BLOCKS	2, %xmm9, %xmm10, %xmm13, %xmm11, %xmm12, %xmm0, \
+	INITIAL_BLOCKS_ENC	2, %xmm9, %xmm10, %xmm13, %xmm11, %xmm12, %xmm0, \
 %xmm1, %xmm2, %xmm3, %xmm4, %xmm8, %xmm5, %xmm6, 6, 78, enc
 	sub	$32, %r13
 	jmp	_initial_blocks_encrypted
 _initial_num_blocks_is_1_encrypt:
-	INITIAL_BLOCKS	1, %xmm9, %xmm10, %xmm13, %xmm11, %xmm12, %xmm0, \
+	INITIAL_BLOCKS_ENC	1, %xmm9, %xmm10, %xmm13, %xmm11, %xmm12, %xmm0, \
 %xmm1, %xmm2, %xmm3, %xmm4, %xmm8, %xmm5, %xmm6, 7, 8, enc
 	sub	$16, %r13
 	jmp	_initial_blocks_encrypted
 _initial_num_blocks_is_0_encrypt:
-	INITIAL_BLOCKS	0, %xmm9, %xmm10, %xmm13, %xmm11, %xmm12, %xmm0, \
+	INITIAL_BLOCKS_ENC	0, %xmm9, %xmm10, %xmm13, %xmm11, %xmm12, %xmm0, \
 %xmm1, %xmm2, %xmm3, %xmm4, %xmm8, %xmm5, %xmm6, 8, 0, enc
 _initial_blocks_encrypted:
 
@@ -1160,7 +1594,7 @@ _initial_blocks_encrypted:
 	sub	$64, %r13
 	je	_four_cipher_left_encrypt
 _encrypt_by_4_encrypt:
-	GHASH_4_ENCRYPT_4_PARALLEL	%xmm9, %xmm10, %xmm11, %xmm12, %xmm13, \
+	GHASH_4_ENCRYPT_4_PARALLEL_ENC	%xmm9, %xmm10, %xmm11, %xmm12, %xmm13, \
 %xmm14, %xmm0, %xmm1, %xmm2, %xmm3, %xmm4, %xmm5, %xmm6, %xmm7, %xmm8, enc
 	add	$64, %r11
 	sub	$64, %r13
@@ -1175,7 +1609,9 @@ _zero_cipher_left_encrypt:
 
          # Handle the last <16 Byte block seperately
 	paddd ONE(%rip), %xmm0                # INCR CNT to get Yn
-	pshufb SHUF_MASK(%rip), %xmm0
+        movdqa SHUF_MASK(%rip), %xmm10
+	PSHUFB_XMM %xmm10, %xmm0
+
 	ENCRYPT_SINGLE_BLOCK	%xmm0, %xmm1        # Encrypt(K, Yn)
 	sub $16, %r11
 	add %r13, %r11
@@ -1185,29 +1621,31 @@ _zero_cipher_left_encrypt:
 	# adjust the shuffle mask pointer to be able to shift 16-r13 bytes
 	# (%r13 is the number of bytes in plaintext mod 16)
 	movdqu	(%r12), %xmm2           # get the appropriate shuffle mask
-	pshufb	%xmm2, %xmm1            # shift right 16-r13 byte
+	PSHUFB_XMM	%xmm2, %xmm1            # shift right 16-r13 byte
 	pxor	%xmm1, %xmm0            # Plaintext XOR Encrypt(K, Yn)
 	movdqu	ALL_F-SHIFT_MASK(%r12), %xmm1
 	# get the appropriate mask to mask out top 16-r13 bytes of xmm0
 	pand	%xmm1, %xmm0            # mask out top 16-r13 bytes of xmm0
+        movdqa SHUF_MASK(%rip), %xmm10
+	PSHUFB_XMM %xmm10,%xmm0
 
-	pshufb	SHUF_MASK(%rip),%xmm0
 	pxor	%xmm0, %xmm8
 	GHASH_MUL %xmm8, %xmm13, %xmm9, %xmm10, %xmm11, %xmm5, %xmm6
 	# GHASH computation for the last <16 byte block
 	sub	%r13, %r11
 	add	$16, %r11
-	pshufb SHUF_MASK(%rip), %xmm0
+	PSHUFB_XMM %xmm10, %xmm1
+
 	# shuffle xmm0 back to output as ciphertext
 
         # Output %r13 bytes
-	movq %xmm0, %rax
+	MOVQ_R64_XMM %xmm0, %rax
 	cmp $8, %r13
 	jle _less_than_8_bytes_left_encrypt
 	mov %rax, (%arg2 , %r11, 1)
 	add $8, %r11
 	psrldq $8, %xmm0
-	movq %xmm0, %rax
+	MOVQ_R64_XMM %xmm0, %rax
 	sub $8, %r13
 _less_than_8_bytes_left_encrypt:
 	mov %al,  (%arg2, %r11, 1)
@@ -1220,14 +1658,15 @@ _multiple_of_16_bytes_encrypt:
 	shl	$3, %r12
 	movd	%r12d, %xmm15       # len(A) in %xmm15
 	shl	$3, %arg4               # len(C) in bits (*128)
-	movq	%arg4, %xmm1
+	MOVQ_R64_XMM	%arg4, %xmm1
 	pslldq	$8, %xmm15          # %xmm15 = len(A)||0x0000000000000000
 	pxor	%xmm1, %xmm15       # %xmm15 = len(A)||len(C)
 	pxor	%xmm15, %xmm8
 	GHASH_MUL	%xmm8, %xmm13, %xmm9, %xmm10, %xmm11, %xmm5, %xmm6
 	# final GHASH computation
+        movdqa SHUF_MASK(%rip), %xmm10
+	PSHUFB_XMM %xmm10, %xmm8         # perform a 16 byte swap
 
-	pshufb	SHUF_MASK(%rip), %xmm8         # perform a 16 byte swap
 	mov	%arg5, %rax		       # %rax  = *Y0
 	movdqu	(%rax), %xmm0		       # %xmm0 = Y0
 	ENCRYPT_SINGLE_BLOCK	%xmm0, %xmm15         # Encrypt(K, Y0)
@@ -1240,11 +1679,11 @@ _return_T_encrypt:
 	cmp	$12, %r11
 	je	_T_12_encrypt
 _T_8_encrypt:
-	movq	%xmm0, %rax
+	MOVQ_R64_XMM	%xmm0, %rax
 	mov	%rax, (%r10)
 	jmp	_return_T_done_encrypt
 _T_12_encrypt:
-	movq	%xmm0, %rax
+	MOVQ_R64_XMM	%xmm0, %rax
 	mov	%rax, (%r10)
 	psrldq	$8, %xmm0
 	movd	%xmm0, %eax
@@ -1258,6 +1697,7 @@ _return_T_done_encrypt:
 	pop	%r13
 	pop	%r12
 	ret
+
 #endif