Add AES-NI optimized code derived from latest OpenSSL upstream.

Add AES instruction set detection.
Add missing license headers to a few files.

Makefile.in

@@ -35,8 +35,8 @@ CRYPTO_SRCS = crypto/aes/crypto_aes.c crypto/scrypt/crypto_scrypt-nosse.c \
 CRYPTO_HDRS = crypto/crypto_utils.h crypto/scrypt/crypto_scrypt.h \
 	crypto/scrypt/sha256.h crypto/scrypt/crypto_aesctr.h crypto/aes/crypto_aes.h \
 	crypto/sha2_utils.h crypto/sha3_utils.h $(MAINHDRS)
-CRYPTO_ASM_SRCS = crypto/aes/vpaes-x86_64.s
+CRYPTO_ASM_SRCS = crypto/aes/vpaes-x86_64.s crypto/aes/aesni-x86_64.s
-CRYPTO_ASM_OBJS = crypto/aes/vpaes-x86_64.o
+CRYPTO_ASM_OBJS = crypto/aes/vpaes-x86_64.o crypto/aes/aesni-x86_64.o
 CRYPTO_ASM_HDRS = crypto/aes/crypto_aes.h
 CRYPTO_COMPAT_SRCS = crypto/old/sha2_utils_old.c crypto/old/sha3_utils_old.c
 CRYPTO_COMPAT_HDRS = crypto/old/sha2_utils_old.h crypto/old/sha3_utils_old.h

crypto/aes/crypto_aes.c

@@ -67,6 +67,8 @@
 extern uint64_t lzma_crc64(const uint8_t *buf, size_t size, uint64_t crc);
 extern int vpaes_set_encrypt_key(const unsigned char *userKey, int bits, AES_KEY *key);
 extern void vpaes_encrypt(const unsigned char *in, unsigned char *out, const AES_KEY *key);
+extern int aesni_set_encrypt_key(const unsigned char *userKey, int bits, AES_KEY *key);
+extern void aesni_encrypt(const unsigned char *in, unsigned char *out, const AES_KEY *key);
 
 setkey_func_ptr enc_setkey;
 encrypt_func_ptr enc_encrypt;
@@ -78,7 +80,11 @@ aes_module_init(processor_info_t *pc)
 	enc_encrypt = AES_encrypt;
 
 	if (pc->proc_type == PROC_X64_INTEL || pc->proc_type == PROC_X64_AMD) {
-		if (pc->sse_level >= 3 && pc->sse_sub_level >= 1) {
+		if (pc->aes_avail) {
+			enc_setkey = aesni_set_encrypt_key;
+			enc_encrypt = aesni_encrypt;
+
+		} else if (pc->sse_level >= 3 && pc->sse_sub_level >= 1) {
 			enc_setkey = vpaes_set_encrypt_key;
 			enc_encrypt = vpaes_encrypt;
 		}

crypto/scrypt/crypto_aesctr.c

@@ -1,3 +1,23 @@
+/*
+ * This file is a part of Pcompress, a chunked parallel multi-
+ * algorithm lossless compression and decompression program.
+ *
+ * Copyright (C) 2012 Moinak Ghosh. All rights reserved.
+ * Use is subject to license terms.
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU Lesser General Public
+ * License as published by the Free Software Foundation; either
+ * version 3 of the License, or (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
+ * Lesser General Public License for more details.
+ *
+ * moinakg@belenix.org, http://moinakg.wordpress.com/
+ */
+
 /*-
  * Copyright 2007-2009 Colin Percival
  * All rights reserved.

crypto/scrypt/crypto_aesctr.h

@@ -1,3 +1,23 @@
+/*
+ * This file is a part of Pcompress, a chunked parallel multi-
+ * algorithm lossless compression and decompression program.
+ *
+ * Copyright (C) 2012 Moinak Ghosh. All rights reserved.
+ * Use is subject to license terms.
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU Lesser General Public
+ * License as published by the Free Software Foundation; either
+ * version 3 of the License, or (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
+ * Lesser General Public License for more details.
+ *
+ * moinakg@belenix.org, http://moinakg.wordpress.com/
+ */
+
 /*-
  * Copyright 2009 Colin Percival
  * All rights reserved.

utils/cpuid.c

@@ -1,3 +1,23 @@
+/*
+ * This file is a part of Pcompress, a chunked parallel multi-
+ * algorithm lossless compression and decompression program.
+ *
+ * Copyright (C) 2012 Moinak Ghosh. All rights reserved.
+ * Use is subject to license terms.
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU Lesser General Public
+ * License as published by the Free Software Foundation; either
+ * version 3 of the License, or (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
+ * Lesser General Public License for more details.
+ *
+ * moinakg@belenix.org, http://moinakg.wordpress.com/
+ */
+
 /*
  * Copyright 2008  Veselin Georgiev,
  * anrieffNOSPAM @ mgail_DOT.com (convert to gmail)
@@ -30,12 +50,13 @@
 
 #ifdef	__x86_64__
 
-#define SSE4_1_FLAG     0x080000
+#define	SSE4_1_FLAG	0x080000
-#define SSE4_2_FLAG     0x100000
+#define	SSE4_2_FLAG	0x100000
-#define SSE3_FLAG       0x1
+#define	SSE3_FLAG	0x1
-#define SSSE3_FLAG      0x200
+#define	SSSE3_FLAG	0x200
-#define AVX_FLAG        0x10000000
+#define	AVX_FLAG		0x10000000
-#define XOP_FLAG        0x800
+#define	XOP_FLAG		0x800
+#define	AES_FLAG		0x2000000
 
 void
 exec_cpuid(uint32_t *regs)
@@ -150,6 +171,10 @@ cpuid_basic_identify(processor_info_t *pc)
 			pc->avx_level = 1;
 		}
 
+		if (raw.basic_cpuid[1][2] & AES_FLAG) {
+			pc->aes_avail = 1;
+		}
+
 		if (raw.ext_cpuid[1][2] & XOP_FLAG) {
 			pc->xop_avail = 1;
 		}

utils/cpuid.h

@@ -1,3 +1,23 @@
+/*
+ * This file is a part of Pcompress, a chunked parallel multi-
+ * algorithm lossless compression and decompression program.
+ *
+ * Copyright (C) 2012 Moinak Ghosh. All rights reserved.
+ * Use is subject to license terms.
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU Lesser General Public
+ * License as published by the Free Software Foundation; either
+ * version 3 of the License, or (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
+ * Lesser General Public License for more details.
+ *
+ * moinakg@belenix.org, http://moinakg.wordpress.com/
+ */
+
 /*
  * Copyright 2008  Veselin Georgiev,
  * anrieffNOSPAM @ mgail_DOT.com (convert to gmail)
@@ -46,6 +66,7 @@ typedef struct {
 	int sse_sub_level;
 	int avx_level;
 	int xop_avail;
+	int aes_avail;
 	proc_type_t proc_type;
 } processor_info_t;
 

utils/heapq.c

@@ -1,3 +1,23 @@
+/*
+ * This file is a part of Pcompress, a chunked parallel multi-
+ * algorithm lossless compression and decompression program.
+ *
+ * Copyright (C) 2012 Moinak Ghosh. All rights reserved.
+ * Use is subject to license terms.
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU Lesser General Public
+ * License as published by the Free Software Foundation; either
+ * version 3 of the License, or (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
+ * Lesser General Public License for more details.
+ *
+ * moinakg@belenix.org, http://moinakg.wordpress.com/
+ */
+
 /*
  * Functions for a rudimentary fast min-heap implementation.
  * Derived from Python's _heapqmodule.c by way of drastic simplification

utils/heapq.h

@@ -1,3 +1,23 @@
+/*
+ * This file is a part of Pcompress, a chunked parallel multi-
+ * algorithm lossless compression and decompression program.
+ *
+ * Copyright (C) 2012 Moinak Ghosh. All rights reserved.
+ * Use is subject to license terms.
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU Lesser General Public
+ * License as published by the Free Software Foundation; either
+ * version 3 of the License, or (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
+ * Lesser General Public License for more details.
+ *
+ * moinakg@belenix.org, http://moinakg.wordpress.com/
+ */
+
 #ifndef __HEAPQ_H_
 
 #define __TYPE int64_t

utils/sse_level.c

@@ -1,3 +1,23 @@
+/*
+ * This file is a part of Pcompress, a chunked parallel multi-
+ * algorithm lossless compression and decompression program.
+ *
+ * Copyright (C) 2012 Moinak Ghosh. All rights reserved.
+ * Use is subject to license terms.
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU Lesser General Public
+ * License as published by the Free Software Foundation; either
+ * version 3 of the License, or (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
+ * Lesser General Public License for more details.
+ *
+ * moinakg@belenix.org, http://moinakg.wordpress.com/
+ */
+
 #include <stdio.h>
 #include <utils.h>
 #include <cpuid.h>