diff options
Diffstat (limited to 'libbb/hash_md5_sha_x86-64.S')
| -rw-r--r-- | libbb/hash_md5_sha_x86-64.S | 472 |
1 files changed, 206 insertions, 266 deletions
diff --git a/libbb/hash_md5_sha_x86-64.S b/libbb/hash_md5_sha_x86-64.S index 466cd9a..3e1c4b4 100644 --- a/libbb/hash_md5_sha_x86-64.S +++ b/libbb/hash_md5_sha_x86-64.S @@ -1,23 +1,27 @@ ### Generated by hash_md5_sha_x86-64.S.sh ### -#if defined(__GNUC__) && defined(__x86_64__) + +#if CONFIG_SHA1_SMALL == 0 && defined(__GNUC__) && defined(__x86_64__) .section .text.sha1_process_block64,"ax",@progbits - .globl sha1_process_block64 - .hidden sha1_process_block64 + .globl sha1_process_block64 + .hidden sha1_process_block64 .type sha1_process_block64, @function + + .balign 8 # allow decoders to fetch at least 4 first insns sha1_process_block64: - pushq %r15 # - pushq %r14 # - pushq %r13 # - pushq %r12 # - pushq %rbp # - pushq %rbx # - pushq %rdi # we need ctx at the end + pushq %r15 # + pushq %r14 # + pushq %r13 # + pushq %r12 # + pushq %rbp # + pushq %rbx # + pushq %rdi # we need ctx at the end #Register and stack use: # eax..edx: a..d # ebp: e # esi,edi: temps # -32+4*n(%rsp),r8...r15: W[0..7,8..15] +# (TODO: actually W[0..7] are used a bit more often, put _thme_ into r8..r15?) movq 4*8(%rdi), %r8 bswapq %r8 @@ -253,7 +257,7 @@ sha1_process_block64: xorl %ecx, %edi # ^d andl %eax, %edi # &b xorl %ecx, %edi # (((c ^ d) & b) ^ d) - leal 0x5A827999(%rdx,%rsi),%edx # e += RCONST + W[n] + leal 0x5A827999(%rdx,%rsi), %edx # e += RCONST + W[n & 15] addl %edi, %edx # e += (((c ^ d) & b) ^ d) movl %ebp, %esi # roll $5, %esi # rotl32(a,5) @@ -270,7 +274,7 @@ sha1_process_block64: xorl %ebx, %edi # ^d andl %ebp, %edi # &b xorl %ebx, %edi # (((c ^ d) & b) ^ d) - leal 0x5A827999(%rcx,%rsi),%ecx # e += RCONST + W[n] + leal 0x5A827999(%rcx,%rsi), %ecx # e += RCONST + W[n & 15] addl %edi, %ecx # e += (((c ^ d) & b) ^ d) movl %edx, %esi # roll $5, %esi # rotl32(a,5) @@ -287,7 +291,7 @@ sha1_process_block64: xorl %eax, %edi # ^d andl %edx, %edi # &b xorl %eax, %edi # (((c ^ d) & b) ^ d) - leal 0x5A827999(%rbx,%rsi),%ebx # e += RCONST + W[n] + leal 0x5A827999(%rbx,%rsi), %ebx # e += RCONST + W[n & 15] addl %edi, %ebx # e += (((c ^ d) & b) ^ d) movl %ecx, %esi # roll $5, %esi # rotl32(a,5) @@ -304,7 +308,7 @@ sha1_process_block64: xorl %ebp, %edi # ^d andl %ecx, %edi # &b xorl %ebp, %edi # (((c ^ d) & b) ^ d) - leal 0x5A827999(%rax,%rsi),%eax # e += RCONST + W[n] + leal 0x5A827999(%rax,%rsi), %eax # e += RCONST + W[n & 15] addl %edi, %eax # e += (((c ^ d) & b) ^ d) movl %ebx, %esi # roll $5, %esi # rotl32(a,5) @@ -320,7 +324,7 @@ sha1_process_block64: movl %ecx, %edi # c xorl %edx, %edi # ^d xorl %ebx, %edi # ^b - leal 0x6ED9EBA1(%rbp,%rsi), %ebp # e += RCONST + mixed_W + leal 0x6ED9EBA1(%rbp,%rsi), %ebp # e += RCONST + W[n & 15] addl %edi, %ebp # e += (c ^ d ^ b) movl %eax, %esi # roll $5, %esi # rotl32(a,5) @@ -336,7 +340,7 @@ sha1_process_block64: movl %ebx, %edi # c xorl %ecx, %edi # ^d xorl %eax, %edi # ^b - leal 0x6ED9EBA1(%rdx,%rsi), %edx # e += RCONST + mixed_W + leal 0x6ED9EBA1(%rdx,%rsi), %edx # e += RCONST + W[n & 15] addl %edi, %edx # e += (c ^ d ^ b) movl %ebp, %esi # roll $5, %esi # rotl32(a,5) @@ -352,7 +356,7 @@ sha1_process_block64: movl %eax, %edi # c xorl %ebx, %edi # ^d xorl %ebp, %edi # ^b - leal 0x6ED9EBA1(%rcx,%rsi), %ecx # e += RCONST + mixed_W + leal 0x6ED9EBA1(%rcx,%rsi), %ecx # e += RCONST + W[n & 15] addl %edi, %ecx # e += (c ^ d ^ b) movl %edx, %esi # roll $5, %esi # rotl32(a,5) @@ -368,135 +372,119 @@ sha1_process_block64: movl %ebp, %edi # c xorl %eax, %edi # ^d xorl %edx, %edi # ^b - leal 0x6ED9EBA1(%rbx,%rsi), %ebx # e += RCONST + mixed_W + leal 0x6ED9EBA1(%rbx,%rsi), %ebx # e += RCONST + W[n & 15] addl %edi, %ebx # e += (c ^ d ^ b) movl %ecx, %esi # roll $5, %esi # rotl32(a,5) addl %esi, %ebx # e += rotl32(a,5) rorl $2, %edx # b = rotl32(b,30) # 24 - movl -32+4*5(%rsp), %esi # W[(n+13) & 15] - xorl -32+4*0(%rsp), %esi # ^W[(n+8) & 15] - xorl %r10d, %esi # ^W[(n+2) & 15] - xorl %r8d, %esi # ^W[n & 15] - roll %esi # - movl %esi, %r8d # store to W[n & 15] + xorl -32+4*5(%rsp), %r8d # W[n & 15] ^= W[(n+13) & 15] + xorl -32+4*0(%rsp), %r8d # ^W[(n+8) & 15] + xorl %r10d, %r8d # ^W[(n+2) & 15] + roll %r8d # movl %edx, %edi # c xorl %ebp, %edi # ^d xorl %ecx, %edi # ^b - leal 0x6ED9EBA1(%rax,%rsi), %eax # e += RCONST + mixed_W + leal 0x6ED9EBA1(%rax,%r8), %eax # e += RCONST + W[n & 15] addl %edi, %eax # e += (c ^ d ^ b) movl %ebx, %esi # roll $5, %esi # rotl32(a,5) addl %esi, %eax # e += rotl32(a,5) rorl $2, %ecx # b = rotl32(b,30) # 25 - movl -32+4*6(%rsp), %esi # W[(n+13) & 15] - xorl -32+4*1(%rsp), %esi # ^W[(n+8) & 15] - xorl %r11d, %esi # ^W[(n+2) & 15] - xorl %r9d, %esi # ^W[n & 15] - roll %esi # - movl %esi, %r9d # store to W[n & 15] + xorl -32+4*6(%rsp), %r9d # W[n & 15] ^= W[(n+13) & 15] + xorl -32+4*1(%rsp), %r9d # ^W[(n+8) & 15] + xorl %r11d, %r9d # ^W[(n+2) & 15] + roll %r9d # movl %ecx, %edi # c xorl %edx, %edi # ^d xorl %ebx, %edi # ^b - leal 0x6ED9EBA1(%rbp,%rsi), %ebp # e += RCONST + mixed_W + leal 0x6ED9EBA1(%rbp,%r9), %ebp # e += RCONST + W[n & 15] addl %edi, %ebp # e += (c ^ d ^ b) movl %eax, %esi # roll $5, %esi # rotl32(a,5) addl %esi, %ebp # e += rotl32(a,5) rorl $2, %ebx # b = rotl32(b,30) # 26 - movl -32+4*7(%rsp), %esi # W[(n+13) & 15] - xorl -32+4*2(%rsp), %esi # ^W[(n+8) & 15] - xorl %r12d, %esi # ^W[(n+2) & 15] - xorl %r10d, %esi # ^W[n & 15] - roll %esi # - movl %esi, %r10d # store to W[n & 15] + xorl -32+4*7(%rsp), %r10d # W[n & 15] ^= W[(n+13) & 15] + xorl -32+4*2(%rsp), %r10d # ^W[(n+8) & 15] + xorl %r12d, %r10d # ^W[(n+2) & 15] + roll %r10d # movl %ebx, %edi # c xorl %ecx, %edi # ^d xorl %eax, %edi # ^b - leal 0x6ED9EBA1(%rdx,%rsi), %edx # e += RCONST + mixed_W + leal 0x6ED9EBA1(%rdx,%r10), %edx # e += RCONST + W[n & 15] addl %edi, %edx # e += (c ^ d ^ b) movl %ebp, %esi # roll $5, %esi # rotl32(a,5) addl %esi, %edx # e += rotl32(a,5) rorl $2, %eax # b = rotl32(b,30) # 27 - movl %r8d, %esi # W[(n+13) & 15] - xorl -32+4*3(%rsp), %esi # ^W[(n+8) & 15] - xorl %r13d, %esi # ^W[(n+2) & 15] - xorl %r11d, %esi # ^W[n & 15] - roll %esi # - movl %esi, %r11d # store to W[n & 15] + xorl %r8d, %r11d # W[n & 15] ^= W[(n+13) & 15] + xorl -32+4*3(%rsp), %r11d # ^W[(n+8) & 15] + xorl %r13d, %r11d # ^W[(n+2) & 15] + roll %r11d # movl %eax, %edi # c xorl %ebx, %edi # ^d xorl %ebp, %edi # ^b - leal 0x6ED9EBA1(%rcx,%rsi), %ecx # e += RCONST + mixed_W + leal 0x6ED9EBA1(%rcx,%r11), %ecx # e += RCONST + W[n & 15] addl %edi, %ecx # e += (c ^ d ^ b) movl %edx, %esi # roll $5, %esi # rotl32(a,5) addl %esi, %ecx # e += rotl32(a,5) rorl $2, %ebp # b = rotl32(b,30) # 28 - movl %r9d, %esi # W[(n+13) & 15] - xorl -32+4*4(%rsp), %esi # ^W[(n+8) & 15] - xorl %r14d, %esi # ^W[(n+2) & 15] - xorl %r12d, %esi # ^W[n & 15] - roll %esi # - movl %esi, %r12d # store to W[n & 15] + xorl %r9d, %r12d # W[n & 15] ^= W[(n+13) & 15] + xorl -32+4*4(%rsp), %r12d # ^W[(n+8) & 15] + xorl %r14d, %r12d # ^W[(n+2) & 15] + roll %r12d # movl %ebp, %edi # c xorl %eax, %edi # ^d xorl %edx, %edi # ^b - leal 0x6ED9EBA1(%rbx,%rsi), %ebx # e += RCONST + mixed_W + leal 0x6ED9EBA1(%rbx,%r12), %ebx # e += RCONST + W[n & 15] addl %edi, %ebx # e += (c ^ d ^ b) movl %ecx, %esi # roll $5, %esi # rotl32(a,5) addl %esi, %ebx # e += rotl32(a,5) rorl $2, %edx # b = rotl32(b,30) # 29 - movl %r10d, %esi # W[(n+13) & 15] - xorl -32+4*5(%rsp), %esi # ^W[(n+8) & 15] - xorl %r15d, %esi # ^W[(n+2) & 15] - xorl %r13d, %esi # ^W[n & 15] - roll %esi # - movl %esi, %r13d # store to W[n & 15] + xorl %r10d, %r13d # W[n & 15] ^= W[(n+13) & 15] + xorl -32+4*5(%rsp), %r13d # ^W[(n+8) & 15] + xorl %r15d, %r13d # ^W[(n+2) & 15] + roll %r13d # movl %edx, %edi # c xorl %ebp, %edi # ^d xorl %ecx, %edi # ^b - leal 0x6ED9EBA1(%rax,%rsi), %eax # e += RCONST + mixed_W + leal 0x6ED9EBA1(%rax,%r13), %eax # e += RCONST + W[n & 15] addl %edi, %eax # e += (c ^ d ^ b) movl %ebx, %esi # roll $5, %esi # rotl32(a,5) addl %esi, %eax # e += rotl32(a,5) rorl $2, %ecx # b = rotl32(b,30) # 30 - movl %r11d, %esi # W[(n+13) & 15] - xorl -32+4*6(%rsp), %esi # ^W[(n+8) & 15] - xorl -32+4*0(%rsp), %esi # ^W[(n+2) & 15] - xorl %r14d, %esi # ^W[n & 15] - roll %esi # - movl %esi, %r14d # store to W[n & 15] + xorl %r11d, %r14d # W[n & 15] ^= W[(n+13) & 15] + xorl -32+4*6(%rsp), %r14d # ^W[(n+8) & 15] + xorl -32+4*0(%rsp), %r14d # ^W[(n+2) & 15] + roll %r14d # movl %ecx, %edi # c xorl %edx, %edi # ^d xorl %ebx, %edi # ^b - leal 0x6ED9EBA1(%rbp,%rsi), %ebp # e += RCONST + mixed_W + leal 0x6ED9EBA1(%rbp,%r14), %ebp # e += RCONST + W[n & 15] addl %edi, %ebp # e += (c ^ d ^ b) movl %eax, %esi # roll $5, %esi # rotl32(a,5) addl %esi, %ebp # e += rotl32(a,5) rorl $2, %ebx # b = rotl32(b,30) # 31 - movl %r12d, %esi # W[(n+13) & 15] - xorl -32+4*7(%rsp), %esi # ^W[(n+8) & 15] - xorl -32+4*1(%rsp), %esi # ^W[(n+2) & 15] - xorl %r15d, %esi # ^W[n & 15] - roll %esi # - movl %esi, %r15d # store to W[n & 15] + xorl %r12d, %r15d # W[n & 15] ^= W[(n+13) & 15] + xorl -32+4*7(%rsp), %r15d # ^W[(n+8) & 15] + xorl -32+4*1(%rsp), %r15d # ^W[(n+2) & 15] + roll %r15d # movl %ebx, %edi # c xorl %ecx, %edi # ^d xorl %eax, %edi # ^b - leal 0x6ED9EBA1(%rdx,%rsi), %edx # e += RCONST + mixed_W + leal 0x6ED9EBA1(%rdx,%r15), %edx # e += RCONST + W[n & 15] addl %edi, %edx # e += (c ^ d ^ b) movl %ebp, %esi # roll $5, %esi # rotl32(a,5) @@ -512,7 +500,7 @@ sha1_process_block64: movl %eax, %edi # c xorl %ebx, %edi # ^d xorl %ebp, %edi # ^b - leal 0x6ED9EBA1(%rcx,%rsi), %ecx # e += RCONST + mixed_W + leal 0x6ED9EBA1(%rcx,%rsi), %ecx # e += RCONST + W[n & 15] addl %edi, %ecx # e += (c ^ d ^ b) movl %edx, %esi # roll $5, %esi # rotl32(a,5) @@ -528,7 +516,7 @@ sha1_process_block64: movl %ebp, %edi # c xorl %eax, %edi # ^d xorl %edx, %edi # ^b - leal 0x6ED9EBA1(%rbx,%rsi), %ebx # e += RCONST + mixed_W + leal 0x6ED9EBA1(%rbx,%rsi), %ebx # e += RCONST + W[n & 15] addl %edi, %ebx # e += (c ^ d ^ b) movl %ecx, %esi # roll $5, %esi # rotl32(a,5) @@ -544,7 +532,7 @@ sha1_process_block64: movl %edx, %edi # c xorl %ebp, %edi # ^d xorl %ecx, %edi # ^b - leal 0x6ED9EBA1(%rax,%rsi), %eax # e += RCONST + mixed_W + leal 0x6ED9EBA1(%rax,%rsi), %eax # e += RCONST + W[n & 15] addl %edi, %eax # e += (c ^ d ^ b) movl %ebx, %esi # roll $5, %esi # rotl32(a,5) @@ -560,7 +548,7 @@ sha1_process_block64: movl %ecx, %edi # c xorl %edx, %edi # ^d xorl %ebx, %edi # ^b - leal 0x6ED9EBA1(%rbp,%rsi), %ebp # e += RCONST + mixed_W + leal 0x6ED9EBA1(%rbp,%rsi), %ebp # e += RCONST + W[n & 15] addl %edi, %ebp # e += (c ^ d ^ b) movl %eax, %esi # roll $5, %esi # rotl32(a,5) @@ -576,7 +564,7 @@ sha1_process_block64: movl %ebx, %edi # c xorl %ecx, %edi # ^d xorl %eax, %edi # ^b - leal 0x6ED9EBA1(%rdx,%rsi), %edx # e += RCONST + mixed_W + leal 0x6ED9EBA1(%rdx,%rsi), %edx # e += RCONST + W[n & 15] addl %edi, %edx # e += (c ^ d ^ b) movl %ebp, %esi # roll $5, %esi # rotl32(a,5) @@ -592,7 +580,7 @@ sha1_process_block64: movl %eax, %edi # c xorl %ebx, %edi # ^d xorl %ebp, %edi # ^b - leal 0x6ED9EBA1(%rcx,%rsi), %ecx # e += RCONST + mixed_W + leal 0x6ED9EBA1(%rcx,%rsi), %ecx # e += RCONST + W[n & 15] addl %edi, %ecx # e += (c ^ d ^ b) movl %edx, %esi # roll $5, %esi # rotl32(a,5) @@ -608,7 +596,7 @@ sha1_process_block64: movl %ebp, %edi # c xorl %eax, %edi # ^d xorl %edx, %edi # ^b - leal 0x6ED9EBA1(%rbx,%rsi), %ebx # e += RCONST + mixed_W + leal 0x6ED9EBA1(%rbx,%rsi), %ebx # e += RCONST + W[n & 15] addl %edi, %ebx # e += (c ^ d ^ b) movl %ecx, %esi # roll $5, %esi # rotl32(a,5) @@ -624,7 +612,7 @@ sha1_process_block64: movl %edx, %edi # c xorl %ebp, %edi # ^d xorl %ecx, %edi # ^b - leal 0x6ED9EBA1(%rax,%rsi), %eax # e += RCONST + mixed_W + leal 0x6ED9EBA1(%rax,%rsi), %eax # e += RCONST + W[n & 15] addl %edi, %eax # e += (c ^ d ^ b) movl %ebx, %esi # roll $5, %esi # rotl32(a,5) @@ -637,14 +625,12 @@ sha1_process_block64: andl %ecx, %esi # si: b & c andl %edx, %edi # di: (b | c) & d orl %esi, %edi # ((b | c) & d) | (b & c) - movl -32+4*5(%rsp), %esi # W[(n+13) & 15] - xorl -32+4*0(%rsp), %esi # ^W[(n+8) & 15] - xorl %r10d, %esi # ^W[(n+2) & 15] - xorl %r8d, %esi # ^W[n & 15] - roll %esi # - movl %esi, %r8d # store to W[n & 15] + xorl -32+4*5(%rsp), %r8d # W[n & 15] ^= W[(n+13) & 15] + xorl -32+4*0(%rsp), %r8d # ^W[(n+8) & 15] + xorl %r10d, %r8d # ^W[(n+2) & 15] + roll %r8d # addl %edi, %ebp # += ((b | c) & d) | (b & c) - leal -0x70e44324(%rbp,%rsi), %ebp # e += RCONST + mixed_W + leal -0x70E44324(%rbp,%r8), %ebp # e += RCONST + W[n & 15] movl %eax, %esi # roll $5, %esi # rotl32(a,5) addl %esi, %ebp # e += rotl32(a,5) @@ -656,14 +642,12 @@ sha1_process_block64: andl %ebx, %esi # si: b & c andl %ecx, %edi # di: (b | c) & d orl %esi, %edi # ((b | c) & d) | (b & c) - movl -32+4*6(%rsp), %esi # W[(n+13) & 15] - xorl -32+4*1(%rsp), %esi # ^W[(n+8) & 15] - xorl %r11d, %esi # ^W[(n+2) & 15] - xorl %r9d, %esi # ^W[n & 15] - roll %esi # - movl %esi, %r9d # store to W[n & 15] + xorl -32+4*6(%rsp), %r9d # W[n & 15] ^= W[(n+13) & 15] + xorl -32+4*1(%rsp), %r9d # ^W[(n+8) & 15] + xorl %r11d, %r9d # ^W[(n+2) & 15] + roll %r9d # addl %edi, %edx # += ((b | c) & d) | (b & c) - leal -0x70e44324(%rdx,%rsi), %edx # e += RCONST + mixed_W + leal -0x70E44324(%rdx,%r9), %edx # e += RCONST + W[n & 15] movl %ebp, %esi # roll $5, %esi # rotl32(a,5) addl %esi, %edx # e += rotl32(a,5) @@ -675,14 +659,12 @@ sha1_process_block64: andl %eax, %esi # si: b & c andl %ebx, %edi # di: (b | c) & d orl %esi, %edi # ((b | c) & d) | (b & c) - movl -32+4*7(%rsp), %esi # W[(n+13) & 15] - xorl -32+4*2(%rsp), %esi # ^W[(n+8) & 15] - xorl %r12d, %esi # ^W[(n+2) & 15] - xorl %r10d, %esi # ^W[n & 15] - roll %esi # - movl %esi, %r10d # store to W[n & 15] + xorl -32+4*7(%rsp), %r10d # W[n & 15] ^= W[(n+13) & 15] + xorl -32+4*2(%rsp), %r10d # ^W[(n+8) & 15] + xorl %r12d, %r10d # ^W[(n+2) & 15] + roll %r10d # addl %edi, %ecx # += ((b | c) & d) | (b & c) - leal -0x70e44324(%rcx,%rsi), %ecx # e += RCONST + mixed_W + leal -0x70E44324(%rcx,%r10), %ecx # e += RCONST + W[n & 15] movl %edx, %esi # roll $5, %esi # rotl32(a,5) addl %esi, %ecx # e += rotl32(a,5) @@ -694,14 +676,12 @@ sha1_process_block64: andl %ebp, %esi # si: b & c andl %eax, %edi # di: (b | c) & d orl %esi, %edi # ((b | c) & d) | (b & c) - movl %r8d, %esi # W[(n+13) & 15] - xorl -32+4*3(%rsp), %esi # ^W[(n+8) & 15] - xorl %r13d, %esi # ^W[(n+2) & 15] - xorl %r11d, %esi # ^W[n & 15] - roll %esi # - movl %esi, %r11d # store to W[n & 15] + xorl %r8d, %r11d # W[n & 15] ^= W[(n+13) & 15] + xorl -32+4*3(%rsp), %r11d # ^W[(n+8) & 15] + xorl %r13d, %r11d # ^W[(n+2) & 15] + roll %r11d # addl %edi, %ebx # += ((b | c) & d) | (b & c) - leal -0x70e44324(%rbx,%rsi), %ebx # e += RCONST + mixed_W + leal -0x70E44324(%rbx,%r11), %ebx # e += RCONST + W[n & 15] movl %ecx, %esi # roll $5, %esi # rotl32(a,5) addl %esi, %ebx # e += rotl32(a,5) @@ -713,14 +693,12 @@ sha1_process_block64: andl %edx, %esi # si: b & c andl %ebp, %edi # di: (b | c) & d orl %esi, %edi # ((b | c) & d) | (b & c) - movl %r9d, %esi # W[(n+13) & 15] - xorl -32+4*4(%rsp), %esi # ^W[(n+8) & 15] - xorl %r14d, %esi # ^W[(n+2) & 15] - xorl %r12d, %esi # ^W[n & 15] - roll %esi # - movl %esi, %r12d # store to W[n & 15] + xorl %r9d, %r12d # W[n & 15] ^= W[(n+13) & 15] + xorl -32+4*4(%rsp), %r12d # ^W[(n+8) & 15] + xorl %r14d, %r12d # ^W[(n+2) & 15] + roll %r12d # addl %edi, %eax # += ((b | c) & d) | (b & c) - leal -0x70e44324(%rax,%rsi), %eax # e += RCONST + mixed_W + leal -0x70E44324(%rax,%r12), %eax # e += RCONST + W[n & 15] movl %ebx, %esi # roll $5, %esi # rotl32(a,5) addl %esi, %eax # e += rotl32(a,5) @@ -732,14 +710,12 @@ sha1_process_block64: andl %ecx, %esi # si: b & c andl %edx, %edi # di: (b | c) & d orl %esi, %edi # ((b | c) & d) | (b & c) - movl %r10d, %esi # W[(n+13) & 15] - xorl -32+4*5(%rsp), %esi # ^W[(n+8) & 15] - xorl %r15d, %esi # ^W[(n+2) & 15] - xorl %r13d, %esi # ^W[n & 15] - roll %esi # - movl %esi, %r13d # store to W[n & 15] + xorl %r10d, %r13d # W[n & 15] ^= W[(n+13) & 15] + xorl -32+4*5(%rsp), %r13d # ^W[(n+8) & 15] + xorl %r15d, %r13d # ^W[(n+2) & 15] + roll %r13d # addl %edi, %ebp # += ((b | c) & d) | (b & c) - leal -0x70e44324(%rbp,%rsi), %ebp # e += RCONST + mixed_W + leal -0x70E44324(%rbp,%r13), %ebp # e += RCONST + W[n & 15] movl %eax, %esi # roll $5, %esi # rotl32(a,5) addl %esi, %ebp # e += rotl32(a,5) @@ -751,14 +727,12 @@ sha1_process_block64: andl %ebx, %esi # si: b & c andl %ecx, %edi # di: (b | c) & d orl %esi, %edi # ((b | c) & d) | (b & c) - movl %r11d, %esi # W[(n+13) & 15] - xorl -32+4*6(%rsp), %esi # ^W[(n+8) & 15] - xorl -32+4*0(%rsp), %esi # ^W[(n+2) & 15] - xorl %r14d, %esi # ^W[n & 15] - roll %esi # - movl %esi, %r14d # store to W[n & 15] + xorl %r11d, %r14d # W[n & 15] ^= W[(n+13) & 15] + xorl -32+4*6(%rsp), %r14d # ^W[(n+8) & 15] + xorl -32+4*0(%rsp), %r14d # ^W[(n+2) & 15] + roll %r14d # addl %edi, %edx # += ((b | c) & d) | (b & c) - leal -0x70e44324(%rdx,%rsi), %edx # e += RCONST + mixed_W + leal -0x70E44324(%rdx,%r14), %edx # e += RCONST + W[n & 15] movl %ebp, %esi # roll $5, %esi # rotl32(a,5) addl %esi, %edx # e += rotl32(a,5) @@ -770,14 +744,12 @@ sha1_process_block64: andl %eax, %esi # si: b & c andl %ebx, %edi # di: (b | c) & d orl %esi, %edi # ((b | c) & d) | (b & c) - movl %r12d, %esi # W[(n+13) & 15] - xorl -32+4*7(%rsp), %esi # ^W[(n+8) & 15] - xorl -32+4*1(%rsp), %esi # ^W[(n+2) & 15] - xorl %r15d, %esi # ^W[n & 15] - roll %esi # - movl %esi, %r15d # store to W[n & 15] + xorl %r12d, %r15d # W[n & 15] ^= W[(n+13) & 15] + xorl -32+4*7(%rsp), %r15d # ^W[(n+8) & 15] + xorl -32+4*1(%rsp), %r15d # ^W[(n+2) & 15] + roll %r15d # addl %edi, %ecx # += ((b | c) & d) | (b & c) - leal -0x70e44324(%rcx,%rsi), %ecx # e += RCONST + mixed_W + leal -0x70E44324(%rcx,%r15), %ecx # e += RCONST + W[n & 15] movl %edx, %esi # roll $5, %esi # rotl32(a,5) addl %esi, %ecx # e += rotl32(a,5) @@ -796,7 +768,7 @@ sha1_process_block64: roll %esi # movl %esi, -32+4*0(%rsp) # store to W[n & 15] addl %edi, %ebx # += ((b | c) & d) | (b & c) - leal -0x70e44324(%rbx,%rsi), %ebx # e += RCONST + mixed_W + leal -0x70E44324(%rbx,%rsi), %ebx # e += RCONST + W[n & 15] movl %ecx, %esi # roll $5, %esi # rotl32(a,5) addl %esi, %ebx # e += rotl32(a,5) @@ -815,7 +787,7 @@ sha1_process_block64: roll %esi # movl %esi, -32+4*1(%rsp) # store to W[n & 15] addl %edi, %eax # += ((b | c) & d) | (b & c) - leal -0x70e44324(%rax,%rsi), %eax # e += RCONST + mixed_W + leal -0x70E44324(%rax,%rsi), %eax # e += RCONST + W[n & 15] movl %ebx, %esi # roll $5, %esi # rotl32(a,5) addl %esi, %eax # e += rotl32(a,5) @@ -834,7 +806,7 @@ sha1_process_block64: roll %esi # movl %esi, -32+4*2(%rsp) # store to W[n & 15] addl %edi, %ebp # += ((b | c) & d) | (b & c) - leal -0x70e44324(%rbp,%rsi), %ebp # e += RCONST + mixed_W + leal -0x70E44324(%rbp,%rsi), %ebp # e += RCONST + W[n & 15] movl %eax, %esi # roll $5, %esi # rotl32(a,5) addl %esi, %ebp # e += rotl32(a,5) @@ -853,7 +825,7 @@ sha1_process_block64: roll %esi # movl %esi, -32+4*3(%rsp) # store to W[n & 15] addl %edi, %edx # += ((b | c) & d) | (b & c) - leal -0x70e44324(%rdx,%rsi), %edx # e += RCONST + mixed_W + leal -0x70E44324(%rdx,%rsi), %edx # e += RCONST + W[n & 15] movl %ebp, %esi # roll $5, %esi # rotl32(a,5) addl %esi, %edx # e += rotl32(a,5) @@ -872,7 +844,7 @@ sha1_process_block64: roll %esi # movl %esi, -32+4*4(%rsp) # store to W[n & 15] addl %edi, %ecx # += ((b | c) & d) | (b & c) - leal -0x70e44324(%rcx,%rsi), %ecx # e += RCONST + mixed_W + leal -0x70E44324(%rcx,%rsi), %ecx # e += RCONST + W[n & 15] movl %edx, %esi # roll $5, %esi # rotl32(a,5) addl %esi, %ecx # e += rotl32(a,5) @@ -891,7 +863,7 @@ sha1_process_block64: roll %esi # movl %esi, -32+4*5(%rsp) # store to W[n & 15] addl %edi, %ebx # += ((b | c) & d) | (b & c) - leal -0x70e44324(%rbx,%rsi), %ebx # e += RCONST + mixed_W + leal -0x70E44324(%rbx,%rsi), %ebx # e += RCONST + W[n & 15] movl %ecx, %esi # roll $5, %esi # rotl32(a,5) addl %esi, %ebx # e += rotl32(a,5) @@ -910,7 +882,7 @@ sha1_process_block64: roll %esi # movl %esi, -32+4*6(%rsp) # store to W[n & 15] addl %edi, %eax # += ((b | c) & d) | (b & c) - leal -0x70e44324(%rax,%rsi), %eax # e += RCONST + mixed_W + leal -0x70E44324(%rax,%rsi), %eax # e += RCONST + W[n & 15] movl %ebx, %esi # roll $5, %esi # rotl32(a,5) addl %esi, %eax # e += rotl32(a,5) @@ -929,7 +901,7 @@ sha1_process_block64: roll %esi # movl %esi, -32+4*7(%rsp) # store to W[n & 15] addl %edi, %ebp # += ((b | c) & d) | (b & c) - leal -0x70e44324(%rbp,%rsi), %ebp # e += RCONST + mixed_W + leal -0x70E44324(%rbp,%rsi), %ebp # e += RCONST + W[n & 15] movl %eax, %esi # roll $5, %esi # rotl32(a,5) addl %esi, %ebp # e += rotl32(a,5) @@ -941,14 +913,12 @@ sha1_process_block64: andl %ebx, %esi # si: b & c andl %ecx, %edi # di: (b | c) & d orl %esi, %edi # ((b | c) & d) | (b & c) - movl -32+4*5(%rsp), %esi # W[(n+13) & 15] - xorl -32+4*0(%rsp), %esi # ^W[(n+8) & 15] - xorl %r10d, %esi # ^W[(n+2) & 15] - xorl %r8d, %esi # ^W[n & 15] - roll %esi # - movl %esi, %r8d # store to W[n & 15] + xorl -32+4*5(%rsp), %r8d # W[n & 15] ^= W[(n+13) & 15] + xorl -32+4*0(%rsp), %r8d # ^W[(n+8) & 15] + xorl %r10d, %r8d # ^W[(n+2) & 15] + roll %r8d # addl %edi, %edx # += ((b | c) & d) | (b & c) - leal -0x70e44324(%rdx,%rsi), %edx # e += RCONST + mixed_W + leal -0x70E44324(%rdx,%r8), %edx # e += RCONST + W[n & 15] movl %ebp, %esi # roll $5, %esi # rotl32(a,5) addl %esi, %edx # e += rotl32(a,5) @@ -960,14 +930,12 @@ sha1_process_block64: andl %eax, %esi # si: b & c andl %ebx, %edi # di: (b | c) & d orl %esi, %edi # ((b | c) & d) | (b & c) - movl -32+4*6(%rsp), %esi # W[(n+13) & 15] - xorl -32+4*1(%rsp), %esi # ^W[(n+8) & 15] - xorl %r11d, %esi # ^W[(n+2) & 15] - xorl %r9d, %esi # ^W[n & 15] - roll %esi # - movl %esi, %r9d # store to W[n & 15] + xorl -32+4*6(%rsp), %r9d # W[n & 15] ^= W[(n+13) & 15] + xorl -32+4*1(%rsp), %r9d # ^W[(n+8) & 15] + xorl %r11d, %r9d # ^W[(n+2) & 15] + roll %r9d # addl %edi, %ecx # += ((b | c) & d) | (b & c) - leal -0x70e44324(%rcx,%rsi), %ecx # e += RCONST + mixed_W + leal -0x70E44324(%rcx,%r9), %ecx # e += RCONST + W[n & 15] movl %edx, %esi # roll $5, %esi # rotl32(a,5) addl %esi, %ecx # e += rotl32(a,5) @@ -979,14 +947,12 @@ sha1_process_block64: andl %ebp, %esi # si: b & c andl %eax, %edi # di: (b | c) & d orl %esi, %edi # ((b | c) & d) | (b & c) - movl -32+4*7(%rsp), %esi # W[(n+13) & 15] - xorl -32+4*2(%rsp), %esi # ^W[(n+8) & 15] - xorl %r12d, %esi # ^W[(n+2) & 15] - xorl %r10d, %esi # ^W[n & 15] - roll %esi # - movl %esi, %r10d # store to W[n & 15] + xorl -32+4*7(%rsp), %r10d # W[n & 15] ^= W[(n+13) & 15] + xorl -32+4*2(%rsp), %r10d # ^W[(n+8) & 15] + xorl %r12d, %r10d # ^W[(n+2) & 15] + roll %r10d # addl %edi, %ebx # += ((b | c) & d) | (b & c) - leal -0x70e44324(%rbx,%rsi), %ebx # e += RCONST + mixed_W + leal -0x70E44324(%rbx,%r10), %ebx # e += RCONST + W[n & 15] movl %ecx, %esi # roll $5, %esi # rotl32(a,5) addl %esi, %ebx # e += rotl32(a,5) @@ -998,77 +964,67 @@ sha1_process_block64: andl %edx, %esi # si: b & c andl %ebp, %edi # di: (b | c) & d orl %esi, %edi # ((b | c) & d) | (b & c) - movl %r8d, %esi # W[(n+13) & 15] - xorl -32+4*3(%rsp), %esi # ^W[(n+8) & 15] - xorl %r13d, %esi # ^W[(n+2) & 15] - xorl %r11d, %esi # ^W[n & 15] - roll %esi # - movl %esi, %r11d # store to W[n & 15] + xorl %r8d, %r11d # W[n & 15] ^= W[(n+13) & 15] + xorl -32+4*3(%rsp), %r11d # ^W[(n+8) & 15] + xorl %r13d, %r11d # ^W[(n+2) & 15] + roll %r11d # addl %edi, %eax # += ((b | c) & d) | (b & c) - leal -0x70e44324(%rax,%rsi), %eax # e += RCONST + mixed_W + leal -0x70E44324(%rax,%r11), %eax # e += RCONST + W[n & 15] movl %ebx, %esi # roll $5, %esi # rotl32(a,5) addl %esi, %eax # e += rotl32(a,5) rorl $2, %ecx # b = rotl32(b,30) # 60 - movl %r9d, %esi # W[(n+13) & 15] - xorl -32+4*4(%rsp), %esi # ^W[(n+8) & 15] - xorl %r14d, %esi # ^W[(n+2) & 15] - xorl %r12d, %esi # ^W[n & 15] - roll %esi # - movl %esi, %r12d # store to W[n & 15] + xorl %r9d, %r12d # W[n & 15] ^= W[(n+13) & 15] + xorl -32+4*4(%rsp), %r12d # ^W[(n+8) & 15] + xorl %r14d, %r12d # ^W[(n+2) & 15] + roll %r12d # movl %ecx, %edi # c xorl %edx, %edi # ^d xorl %ebx, %edi # ^b - leal -0x359d3e2a(%rbp,%rsi), %ebp # e += RCONST + mixed_W + leal -0x359D3E2A(%rbp,%r12), %ebp # e += RCONST + W[n & 15] addl %edi, %ebp # e += (c ^ d ^ b) movl %eax, %esi # roll $5, %esi # rotl32(a,5) addl %esi, %ebp # e += rotl32(a,5) rorl $2, %ebx # b = rotl32(b,30) # 61 - movl %r10d, %esi # W[(n+13) & 15] - xorl -32+4*5(%rsp), %esi # ^W[(n+8) & 15] - xorl %r15d, %esi # ^W[(n+2) & 15] - xorl %r13d, %esi # ^W[n & 15] - roll %esi # - movl %esi, %r13d # store to W[n & 15] + xorl %r10d, %r13d # W[n & 15] ^= W[(n+13) & 15] + xorl -32+4*5(%rsp), %r13d # ^W[(n+8) & 15] + xorl %r15d, %r13d # ^W[(n+2) & 15] + roll %r13d # movl %ebx, %edi # c xorl %ecx, %edi # ^d xorl %eax, %edi # ^b - leal -0x359d3e2a(%rdx,%rsi), %edx # e += RCONST + mixed_W + leal -0x359D3E2A(%rdx,%r13), %edx # e += RCONST + W[n & 15] addl %edi, %edx # e += (c ^ d ^ b) movl %ebp, %esi # roll $5, %esi # rotl32(a,5) addl %esi, %edx # e += rotl32(a,5) rorl $2, %eax # b = rotl32(b,30) # 62 - movl %r11d, %esi # W[(n+13) & 15] - xorl -32+4*6(%rsp), %esi # ^W[(n+8) & 15] - xorl -32+4*0(%rsp), %esi # ^W[(n+2) & 15] - xorl %r14d, %esi # ^W[n & 15] - roll %esi # - movl %esi, %r14d # store to W[n & 15] + xorl %r11d, %r14d # W[n & 15] ^= W[(n+13) & 15] + xorl -32+4*6(%rsp), %r14d # ^W[(n+8) & 15] + xorl -32+4*0(%rsp), %r14d # ^W[(n+2) & 15] + roll %r14d # movl %eax, %edi # c xorl %ebx, %edi # ^d xorl %ebp, %edi # ^b - leal -0x359d3e2a(%rcx,%rsi), %ecx # e += RCONST + mixed_W + leal -0x359D3E2A(%rcx,%r14), %ecx # e += RCONST + W[n & 15] addl %edi, %ecx # e += (c ^ d ^ b) movl %edx, %esi # roll $5, %esi # rotl32(a,5) addl %esi, %ecx # e += rotl32(a,5) rorl $2, %ebp # b = rotl32(b,30) # 63 - movl %r12d, %esi # W[(n+13) & 15] - xorl -32+4*7(%rsp), %esi # ^W[(n+8) & 15] - xorl -32+4*1(%rsp), %esi # ^W[(n+2) & 15] - xorl %r15d, %esi # ^W[n & 15] - roll %esi # - movl %esi, %r15d # store to W[n & 15] + xorl %r12d, %r15d # W[n & 15] ^= W[(n+13) & 15] + xorl -32+4*7(%rsp), %r15d # ^W[(n+8) & 15] + xorl -32+4*1(%rsp), %r15d # ^W[(n+2) & 15] + roll %r15d # movl %ebp, %edi # c xorl %eax, %edi # ^d xorl %edx, %edi # ^b - leal -0x359d3e2a(%rbx,%rsi), %ebx # e += RCONST + mixed_W + leal -0x359D3E2A(%rbx,%r15), %ebx # e += RCONST + W[n & 15] addl %edi, %ebx # e += (c ^ d ^ b) movl %ecx, %esi # roll $5, %esi # rotl32(a,5) @@ -1084,7 +1040,7 @@ sha1_process_block64: movl %edx, %edi # c xorl %ebp, %edi # ^d xorl %ecx, %edi # ^b - leal -0x359d3e2a(%rax,%rsi), %eax # e += RCONST + mixed_W + leal -0x359D3E2A(%rax,%rsi), %eax # e += RCONST + W[n & 15] addl %edi, %eax # e += (c ^ d ^ b) movl %ebx, %esi # roll $5, %esi # rotl32(a,5) @@ -1100,7 +1056,7 @@ sha1_process_block64: movl %ecx, %edi # c xorl %edx, %edi # ^d xorl %ebx, %edi # ^b - leal -0x359d3e2a(%rbp,%rsi), %ebp # e += RCONST + mixed_W + leal -0x359D3E2A(%rbp,%rsi), %ebp # e += RCONST + W[n & 15] addl %edi, %ebp # e += (c ^ d ^ b) movl %eax, %esi # roll $5, %esi # rotl32(a,5) @@ -1116,7 +1072,7 @@ sha1_process_block64: movl %ebx, %edi # c xorl %ecx, %edi # ^d xorl %eax, %edi # ^b - leal -0x359d3e2a(%rdx,%rsi), %edx # e += RCONST + mixed_W + leal -0x359D3E2A(%rdx,%rsi), %edx # e += RCONST + W[n & 15] addl %edi, %edx # e += (c ^ d ^ b) movl %ebp, %esi # roll $5, %esi # rotl32(a,5) @@ -1132,7 +1088,7 @@ sha1_process_block64: movl %eax, %edi # c xorl %ebx, %edi # ^d xorl %ebp, %edi # ^b - leal -0x359d3e2a(%rcx,%rsi), %ecx # e += RCONST + mixed_W + leal -0x359D3E2A(%rcx,%rsi), %ecx # e += RCONST + W[n & 15] addl %edi, %ecx # e += (c ^ d ^ b) movl %edx, %esi # roll $5, %esi # rotl32(a,5) @@ -1148,7 +1104,7 @@ sha1_process_block64: movl %ebp, %edi # c xorl %eax, %edi # ^d xorl %edx, %edi # ^b - leal -0x359d3e2a(%rbx,%rsi), %ebx # e += RCONST + mixed_W + leal -0x359D3E2A(%rbx,%rsi), %ebx # e += RCONST + W[n & 15] addl %edi, %ebx # e += (c ^ d ^ b) movl %ecx, %esi # roll $5, %esi # rotl32(a,5) @@ -1164,7 +1120,7 @@ sha1_process_block64: movl %edx, %edi # c xorl %ebp, %edi # ^d xorl %ecx, %edi # ^b - leal -0x359d3e2a(%rax,%rsi), %eax # e += RCONST + mixed_W + leal -0x359D3E2A(%rax,%rsi), %eax # e += RCONST + W[n & 15] addl %edi, %eax # e += (c ^ d ^ b) movl %ebx, %esi # roll $5, %esi # rotl32(a,5) @@ -1180,7 +1136,7 @@ sha1_process_block64: movl %ecx, %edi # c xorl %edx, %edi # ^d xorl %ebx, %edi # ^b - leal -0x359d3e2a(%rbp,%rsi), %ebp # e += RCONST + mixed_W + leal -0x359D3E2A(%rbp,%rsi), %ebp # e += RCONST + W[n & 15] addl %edi, %ebp # e += (c ^ d ^ b) movl %eax, %esi # roll $5, %esi # rotl32(a,5) @@ -1196,135 +1152,119 @@ sha1_process_block64: movl %ebx, %edi # c xorl %ecx, %edi # ^d xorl %eax, %edi # ^b - leal -0x359d3e2a(%rdx,%rsi), %edx # e += RCONST + mixed_W + leal -0x359D3E2A(%rdx,%rsi), %edx # e += RCONST + W[n & 15] addl %edi, %edx # e += (c ^ d ^ b) movl %ebp, %esi # roll $5, %esi # rotl32(a,5) addl %esi, %edx # e += rotl32(a,5) rorl $2, %eax # b = rotl32(b,30) # 72 - movl -32+4*5(%rsp), %esi # W[(n+13) & 15] - xorl -32+4*0(%rsp), %esi # ^W[(n+8) & 15] - xorl %r10d, %esi # ^W[(n+2) & 15] - xorl %r8d, %esi # ^W[n & 15] - roll %esi # - movl %esi, %r8d # store to W[n & 15] + xorl -32+4*5(%rsp), %r8d # W[n & 15] ^= W[(n+13) & 15] + xorl -32+4*0(%rsp), %r8d # ^W[(n+8) & 15] + xorl %r10d, %r8d # ^W[(n+2) & 15] + roll %r8d # movl %eax, %edi # c xorl %ebx, %edi # ^d xorl %ebp, %edi # ^b - leal -0x359d3e2a(%rcx,%rsi), %ecx # e += RCONST + mixed_W + leal -0x359D3E2A(%rcx,%r8), %ecx # e += RCONST + W[n & 15] addl %edi, %ecx # e += (c ^ d ^ b) movl %edx, %esi # roll $5, %esi # rotl32(a,5) addl %esi, %ecx # e += rotl32(a,5) rorl $2, %ebp # b = rotl32(b,30) # 73 - movl -32+4*6(%rsp), %esi # W[(n+13) & 15] - xorl -32+4*1(%rsp), %esi # ^W[(n+8) & 15] - xorl %r11d, %esi # ^W[(n+2) & 15] - xorl %r9d, %esi # ^W[n & 15] - roll %esi # - movl %esi, %r9d # store to W[n & 15] + xorl -32+4*6(%rsp), %r9d # W[n & 15] ^= W[(n+13) & 15] + xorl -32+4*1(%rsp), %r9d # ^W[(n+8) & 15] + xorl %r11d, %r9d # ^W[(n+2) & 15] + roll %r9d # movl %ebp, %edi # c xorl %eax, %edi # ^d xorl %edx, %edi # ^b - leal -0x359d3e2a(%rbx,%rsi), %ebx # e += RCONST + mixed_W + leal -0x359D3E2A(%rbx,%r9), %ebx # e += RCONST + W[n & 15] addl %edi, %ebx # e += (c ^ d ^ b) movl %ecx, %esi # roll $5, %esi # rotl32(a,5) addl %esi, %ebx # e += rotl32(a,5) rorl $2, %edx # b = rotl32(b,30) # 74 - movl -32+4*7(%rsp), %esi # W[(n+13) & 15] - xorl -32+4*2(%rsp), %esi # ^W[(n+8) & 15] - xorl %r12d, %esi # ^W[(n+2) & 15] - xorl %r10d, %esi # ^W[n & 15] - roll %esi # - movl %esi, %r10d # store to W[n & 15] + xorl -32+4*7(%rsp), %r10d # W[n & 15] ^= W[(n+13) & 15] + xorl -32+4*2(%rsp), %r10d # ^W[(n+8) & 15] + xorl %r12d, %r10d # ^W[(n+2) & 15] + roll %r10d # movl %edx, %edi # c xorl %ebp, %edi # ^d xorl %ecx, %edi # ^b - leal -0x359d3e2a(%rax,%rsi), %eax # e += RCONST + mixed_W + leal -0x359D3E2A(%rax,%r10), %eax # e += RCONST + W[n & 15] addl %edi, %eax # e += (c ^ d ^ b) movl %ebx, %esi # roll $5, %esi # rotl32(a,5) addl %esi, %eax # e += rotl32(a,5) rorl $2, %ecx # b = rotl32(b,30) # 75 - movl %r8d, %esi # W[(n+13) & 15] - xorl -32+4*3(%rsp), %esi # ^W[(n+8) & 15] - xorl %r13d, %esi # ^W[(n+2) & 15] - xorl %r11d, %esi # ^W[n & 15] - roll %esi # - movl %esi, %r11d # store to W[n & 15] + xorl %r8d, %r11d # W[n & 15] ^= W[(n+13) & 15] + xorl -32+4*3(%rsp), %r11d # ^W[(n+8) & 15] + xorl %r13d, %r11d # ^W[(n+2) & 15] + roll %r11d # movl %ecx, %edi # c xorl %edx, %edi # ^d xorl %ebx, %edi # ^b - leal -0x359d3e2a(%rbp,%rsi), %ebp # e += RCONST + mixed_W + leal -0x359D3E2A(%rbp,%r11), %ebp # e += RCONST + W[n & 15] addl %edi, %ebp # e += (c ^ d ^ b) movl %eax, %esi # roll $5, %esi # rotl32(a,5) addl %esi, %ebp # e += rotl32(a,5) rorl $2, %ebx # b = rotl32(b,30) # 76 - movl %r9d, %esi # W[(n+13) & 15] - xorl -32+4*4(%rsp), %esi # ^W[(n+8) & 15] - xorl %r14d, %esi # ^W[(n+2) & 15] - xorl %r12d, %esi # ^W[n & 15] - roll %esi # - movl %esi, %r12d # store to W[n & 15] + xorl %r9d, %r12d # W[n & 15] ^= W[(n+13) & 15] + xorl -32+4*4(%rsp), %r12d # ^W[(n+8) & 15] + xorl %r14d, %r12d # ^W[(n+2) & 15] + roll %r12d # movl %ebx, %edi # c xorl %ecx, %edi # ^d xorl %eax, %edi # ^b - leal -0x359d3e2a(%rdx,%rsi), %edx # e += RCONST + mixed_W + leal -0x359D3E2A(%rdx,%r12), %edx # e += RCONST + W[n & 15] addl %edi, %edx # e += (c ^ d ^ b) movl %ebp, %esi # roll $5, %esi # rotl32(a,5) addl %esi, %edx # e += rotl32(a,5) rorl $2, %eax # b = rotl32(b,30) # 77 - movl %r10d, %esi # W[(n+13) & 15] - xorl -32+4*5(%rsp), %esi # ^W[(n+8) & 15] - xorl %r15d, %esi # ^W[(n+2) & 15] - xorl %r13d, %esi # ^W[n & 15] - roll %esi # - # store to W[n & 15] - unused, not done + xorl %r10d, %r13d # W[n & 15] ^= W[(n+13) & 15] + xorl -32+4*5(%rsp), %r13d # ^W[(n+8) & 15] + xorl %r15d, %r13d # ^W[(n+2) & 15] + roll %r13d # movl %eax, %edi # c xorl %ebx, %edi # ^d xorl %ebp, %edi # ^b - leal -0x359d3e2a(%rcx,%rsi), %ecx # e += RCONST + mixed_W + leal -0x359D3E2A(%rcx,%r13), %ecx # e += RCONST + W[n & 15] addl %edi, %ecx # e += (c ^ d ^ b) movl %edx, %esi # roll $5, %esi # rotl32(a,5) addl %esi, %ecx # e += rotl32(a,5) rorl $2, %ebp # b = rotl32(b,30) # 78 - movl %r11d, %esi # W[(n+13) & 15] - xorl -32+4*6(%rsp), %esi # ^W[(n+8) & 15] - xorl -32+4*0(%rsp), %esi # ^W[(n+2) & 15] - xorl %r14d, %esi # ^W[n & 15] - roll %esi # - # store to W[n & 15] - unused, not done + xorl %r11d, %r14d # W[n & 15] ^= W[(n+13) & 15] + xorl -32+4*6(%rsp), %r14d # ^W[(n+8) & 15] + xorl -32+4*0(%rsp), %r14d # ^W[(n+2) & 15] + roll %r14d # movl %ebp, %edi # c xorl %eax, %edi # ^d xorl %edx, %edi # ^b - leal -0x359d3e2a(%rbx,%rsi), %ebx # e += RCONST + mixed_W + leal -0x359D3E2A(%rbx,%r14), %ebx # e += RCONST + W[n & 15] addl %edi, %ebx # e += (c ^ d ^ b) movl %ecx, %esi # roll $5, %esi # rotl32(a,5) addl %esi, %ebx # e += rotl32(a,5) rorl $2, %edx # b = rotl32(b,30) # 79 - movl %r12d, %esi # W[(n+13) & 15] - xorl -32+4*7(%rsp), %esi # ^W[(n+8) & 15] - xorl -32+4*1(%rsp), %esi # ^W[(n+2) & 15] - xorl %r15d, %esi # ^W[n & 15] - roll %esi # - # store to W[n & 15] - unused, not done + xorl %r12d, %r15d # W[n & 15] ^= W[(n+13) & 15] + xorl -32+4*7(%rsp), %r15d # ^W[(n+8) & 15] + xorl -32+4*1(%rsp), %r15d # ^W[(n+2) & 15] + roll %r15d # movl %edx, %edi # c xorl %ebp, %edi # ^d xorl %ecx, %edi # ^b - leal -0x359d3e2a(%rax,%rsi), %eax # e += RCONST + mixed_W + leal -0x359D3E2A(%rax,%r15), %eax # e += RCONST + W[n & 15] addl %edi, %eax # e += (c ^ d ^ b) movl %ebx, %esi # roll $5, %esi # rotl32(a,5) |