def TestSynIterInc():
SIZE = 64
# build and run the kernel
prgm = env.Program()
code = prgm.get_stream()
code.add(cal.dcl_output(reg.o0, USAGE=cal.usage.pos))
ones = prgm.acquire_register((1, 1, 1, 1))
counter = prgm.acquire_register()
code.add(cal.mov(counter, ones))
for i in syn_iter(code, 4, step=1, mode=INC):
code.add(cal.iadd(counter, counter, ones))
code.add(cal.mov(reg.o0, counter.x))
domain = (0, 0, SIZE, SIZE)
proc = env.Processor(0)
ext_output=proc.alloc_remote('i', 1, SIZE)
prgm.set_binding(reg.o0, ext_output)
prgm.add(code)
proc.execute(prgm, domain)
passed = True
for i in xrange(0, SIZE):
if ext_output[i] != 5:
passed = False
print "Passed == ", passed
proc.free(ext_output)
return
def cal_nbody():
global pos, vel, mass, prgm, step, proc
step = 0
proc = env.Processor(1)
pos = [
proc.alloc_remote('f', 4, SQRT_BODIES, SQRT_BODIES)
for i in xrange(0, 2)
]
vel = [
proc.alloc_remote('f', 4, SQRT_BODIES, SQRT_BODIES)
for i in xrange(0, 2)
]
pos[0].clear()
vel[0].clear()
pos[1].clear()
vel[1].clear()
for i in xrange(0, N_BODIES):
pos[0][i * 4] = random.uniform(-20.0, 20.0)
pos[0][i * 4 + 1] = random.uniform(-12.0, 12.0)
pos[0][i * 4 + 2] = 0.0
pos[0][i * 4 + 3] = random.uniform(1e9, 1e12)
#vel[0][i * 4] = random.uniform(-1.0, 1.0)
#vel[0][i * 4 + 1] = random.uniform(-1.0, 1.0)
vel[0][i * 4] = 0.0
vel[0][i * 4 + 1] = 0.0
vel[0][i * 4 + 2] = 0.0
vel[0][i * 4 + 3] = 0.0
pos[0][0] = -4.0
pos[0][1] = 0.0
pos[0][2] = 0.0
pos[0][3] = 1e16
ind = 1
pos[0][ind * 4] = 4.0
pos[0][ind * 4 + 1] = 0.0
pos[0][ind * 4 + 2] = 0.0
pos[0][ind * 4 + 3] = 1e16
ind = 2
pos[0][ind * 4] = 0.0
pos[0][ind * 4 + 1] = 4.0
pos[0][ind * 4 + 2] = 0.0
pos[0][ind * 4 + 3] = 1e16
ind = 3
pos[0][ind * 4] = 0.0
pos[0][ind * 4 + 1] = -4.0
pos[0][ind * 4 + 2] = 0.0
pos[0][ind * 4 + 3] = 1e16
code = nbody.cal_nb_generate_2d(prgm, SQRT_BODIES, 0.000002)
prgm += code
prgm.cache_code()
return
def TestRelativeAddressing():
import corepy.arch.cal.platform as env
import corepy.arch.cal.isa as cal
proc = env.Processor(0)
input_mem = proc.alloc_remote('I', 4, 16, 1)
output_mem = proc.alloc_remote('I', 4, 1, 1)
for i in range(16 * 1 * 4):
for j in range(4):
input_mem[i * 4 + j] = i
prgm = env.Program()
code = prgm.get_stream()
cal.set_active_code(code)
cal.dcl_output(o0, USAGE=cal.usage.generic)
cal.dcl_literal(l0, 1, 1, 1, 1)
cal.dcl_literal(l1, 16, 16, 16, 16)
cal.mov(r0, r0('0000'))
cal.mov(r1, r1('0000'))
cal.whileloop()
cal.iadd(r1, r1, g[r0.x])
cal.iadd(r0, r0, l0)
cal.breakc(cal.relop.ge, r0, l1)
cal.endloop()
cal.mov(o0, r1)
prgm.set_binding('g[]', input_mem)
prgm.set_binding('o0', output_mem)
prgm.add(code)
domain = (0, 0, 128, 128)
prgm.print_code()
proc.execute(prgm, domain)
# code.cache_code()
# print code.render_string
if output_mem[0] == 120:
print "Passed relative addressing test"
else:
print "Failed relative addressing test"
proc.free(input_mem)
proc.free(output_mem)
def test_4comp():
proc = env.Processor(0)
prgm = env.Program()
code = prgm.get_stream()
inp = proc.alloc_remote('i', 1, 4, 1)
out = proc.alloc_remote('i', 4, 1, 1)
for i in xrange(0, 4):
inp[i] = i + 1
out[i] = 0
print "inp", inp[0:4]
print "out", out[0:4]
cal.set_active_code(code)
cal.dcl_output(reg.o0, USAGE=cal.usage.generic)
cal.dcl_resource(0, cal.pixtex_type.oned, cal.fmt.float,
UNNORM=True) # positions
r_cnt = prgm.acquire_register()
r = prgm.acquire_registers(4)
cal.mov(r_cnt, r_cnt('0000'))
for i in xrange(0, 4):
cal.sample(0, 0, r[i].x000, r_cnt.x)
cal.add(r_cnt, r_cnt, r_cnt('1111'))
cal.iadd(r[0], r[0], r[1]('0x00'))
cal.iadd(r[0], r[0], r[2]('00x0'))
cal.iadd(r[0], r[0], r[3]('000x'))
cal.iadd(r[0], r[0], r[0])
cal.mov(reg.o0, r[0])
prgm.set_binding(reg.i0, inp)
prgm.set_binding(reg.o0, out)
prgm.add(code)
prgm.print_code()
proc.execute(prgm, (0, 0, 1, 1))
print "inp", inp[0:4]
print "out", out[0:4]
for i in xrange(0, 4):
assert (out[i] == (i + 1) * 2)
return
def TestSynIterIncFloatExtStopExtStart():
SIZE = 64
# build and run the kernel
prgm = env.Program()
code = prgm.get_stream()
code.add(cal.dcl_output(reg.o0, USAGE=cal.usage.pos))
ones = prgm.acquire_register((1, 1, 1, 1))
counter = prgm.acquire_register()
code.add(cal.mov(counter, ones))
stop = prgm.acquire_register((4.0, 4.0, 4.0, 4.0))
start = prgm.acquire_register((2.0, 2.0, 2.0, 2.0))
step = prgm.acquire_register((1.0, 1.0, 1.0, 1.0))
fiter = syn_iter_float(code, stop, step=step, mode=INC)
fiter.set_start_reg(start)
for i in fiter:
code.add(cal.iadd(counter, counter, ones))
code.add(cal.mov(reg.o0, counter.x))
domain = (0, 0, SIZE, SIZE)
proc = env.Processor(0)
ext_output=proc.alloc_remote('i', 1, SIZE, 1)
prgm.set_binding(reg.o0, ext_output)
prgm.add(code)
proc.execute(prgm, domain)
passed = True
for i in xrange(0, SIZE):
if ext_output[i] != 3:
passed = False
print "Passed == ", passed
proc.free(ext_output)
return
def test_foo():
proc = env.Processor(0)
prgm = env.Program()
code = prgm.get_stream()
cal.set_active_code(code)
cb = proc.alloc_remote('i', 1, 4, 1)
out = proc.alloc_remote('i', 4, 1, 1)
gb = proc.alloc_remote('i', 1, 4, 1, True)
for i in xrange(0, 4):
cb[i] = i + 1
out[i] = 42
gb[i] = 67
cal.dcl_output(reg.o0, USAGE=cal.usage.generic)
cal.dcl_cb('cb0[4]')
cal.mov('r0', 'cb0[0]')
cal.mov('r1', 'cb0[1]')
#cal.mov('r2', 'cb0[2]')
#cal.mov('r3', 'cb0[3]')
cal.mov('o0', 'r0')
cal.mov('g[0]', 'r0')
prgm.set_binding('cb0', cb)
prgm.set_binding('o0', out)
prgm.set_binding('g[]', gb)
prgm.add(code)
prgm.print_code()
proc.execute(prgm, (0, 0, 1, 1))
print "cb ", cb[0:4]
print "out", out[0:4]
print "gb ", gb[0:4]
return
cal.breakc(cal.relop.ge, r_count.y, r_limit)
cal.add(r_sum, r_sum, r_sum('1111'))
cal.add(r_count, r_count, r_count('0100'))
cal.endloop()
cal.add(r_count, r_count, r_count('1000'))
cal.endloop()
cal.mov(reg.o0, r_sum)
return code
if __name__ == '__main__':
SIZE = 64
prgm = env.Program()
prgm.add(generate(prgm))
proc = env.Processor(0)
out = proc.alloc_remote('f', 4, SIZE, SIZE)
prgm.set_binding(reg.o0, out)
proc.execute(prgm, (0, 0, SIZE, SIZE))
print out
prgm.print_code()
def ParMD5Transform(parcontext, parblock, blocki):
num = parcontext.number
temp_block = extarray.extarray('I', 16 * num)
ParDecode(num, temp_block, parblock, blocki, 64)
proc = env.Processor(0)
N = int(math.sqrt(num / 4))
#print "N = ", N
def address_4_1d(i, pitch=64):
x = i % N
y = i // 64 * 4
#return x*4 + y*pitch*4*4
return i
def address_4_2d(x, y, pitch=64):
return x * 4 + y * pitch * 4
input_statea = proc.alloc_remote('I', 4, N, N)
input_stateb = proc.alloc_remote('I', 4, N, N)
input_statec = proc.alloc_remote('I', 4, N, N)
input_stated = proc.alloc_remote('I', 4, N, N)
input_block = [proc.alloc_remote('I', 4, N, N) for i in range(16)]
outputa = proc.alloc_remote('I', 4, N, N)
outputb = proc.alloc_remote('I', 4, N, N)
outputc = proc.alloc_remote('I', 4, N, N)
outputd = proc.alloc_remote('I', 4, N, N)
for j in range(N):
for i in range(N):
for k in range(4):
input_statea[address_4_2d(i, j) +
k] = parcontext.statea[k + (i + j * N) * 4]
input_stateb[address_4_2d(i, j) +
k] = parcontext.stateb[k + (i + j * N) * 4]
input_statec[address_4_2d(i, j) +
k] = parcontext.statec[k + (i + j * N) * 4]
input_stated[address_4_2d(i, j) +
k] = parcontext.stated[k + (i + j * N) * 4]
for k in range(N):
for j in range(N):
for l in range(4):
for i in range(16):
input_block[i][address_4_2d(j, k) +
l] = temp_block[i + (j + k * N) * 4 * 16 +
l * 16]
global xcode
if xcode == None:
xcode = env.InstructionStream()
cal.set_active_code(xcode)
S11 = xcode.acquire_register((7, 7, 7, 7))
S12 = xcode.acquire_register((12, 12, 12, 12))
S13 = xcode.acquire_register((17, 17, 17, 17))
S14 = xcode.acquire_register((22, 22, 22, 22))
S21 = xcode.acquire_register((5, 5, 5, 5))
S22 = xcode.acquire_register((9, 9, 9, 9))
S23 = xcode.acquire_register((14, 14, 14, 14))
S24 = xcode.acquire_register((20, 20, 20, 20))
S31 = xcode.acquire_register((4, 4, 4, 4))
S32 = xcode.acquire_register((11, 11, 11, 11))
S33 = xcode.acquire_register((16, 16, 16, 16))
S34 = xcode.acquire_register((23, 23, 23, 23))
S41 = xcode.acquire_register((6, 6, 6, 6))
S42 = xcode.acquire_register((10, 10, 10, 10))
S43 = xcode.acquire_register((15, 15, 15, 15))
S44 = xcode.acquire_register((21, 21, 21, 21))
a = xcode.acquire_register()
b = xcode.acquire_register()
c = xcode.acquire_register()
d = xcode.acquire_register()
x = [xcode.acquire_register() for i in range(16)]
r = xcode.acquire_register()
cal.dcl_resource(0, cal.pixtex_type.twod, cal.fmt.uint,
UNNORM=True) # statea
cal.dcl_resource(1, cal.pixtex_type.twod, cal.fmt.uint,
UNNORM=True) # stateb
cal.dcl_resource(2, cal.pixtex_type.twod, cal.fmt.uint,
UNNORM=True) # statec
cal.dcl_resource(3, cal.pixtex_type.twod, cal.fmt.uint,
UNNORM=True) # stated
for i in range(16):
cal.dcl_resource(i + 4,
cal.pixtex_type.twod,
cal.fmt.uint,
UNNORM=True)
cal.dcl_output(reg.o0, USAGE=cal.usage.generic)
cal.dcl_output(reg.o1, USAGE=cal.usage.generic)
cal.dcl_output(reg.o2, USAGE=cal.usage.generic)
cal.dcl_output(reg.o3, USAGE=cal.usage.generic)
cal.sample(0, 0, a, reg.v0.xy)
cal.sample(1, 0, b, reg.v0.xy)
cal.sample(2, 0, c, reg.v0.xy)
cal.sample(3, 0, d, reg.v0.xy)
for i in range(16):
cal.sample(i + 4, 0, x[i], reg.v0.xy)
# Round 1
FF(a, b, c, d, x[0], S11, 0xd76aa478)
# 1
FF(d, a, b, c, x[1], S12, 0xe8c7b756)
# 2
FF(c, d, a, b, x[2], S13, 0x242070db)
# 3
FF(b, c, d, a, x[3], S14, 0xc1bdceee)
# 4
FF(a, b, c, d, x[4], S11, 0xf57c0faf)
# 5
FF(d, a, b, c, x[5], S12, 0x4787c62a)
# 6
FF(c, d, a, b, x[6], S13, 0xa8304613)
# 7
FF(b, c, d, a, x[7], S14, 0xfd469501)
# 8
FF(a, b, c, d, x[8], S11, 0x698098d8)
# 9
FF(d, a, b, c, x[9], S12, 0x8b44f7af)
# 10
FF(c, d, a, b, x[10], S13, 0xffff5bb1)
# 11
FF(b, c, d, a, x[11], S14, 0x895cd7be)
# 12
FF(a, b, c, d, x[12], S11, 0x6b901122)
# 13
FF(d, a, b, c, x[13], S12, 0xfd987193)
# 14
FF(c, d, a, b, x[14], S13, 0xa679438e)
# 15
FF(b, c, d, a, x[15], S14, 0x49b40821)
# 16
# Round 2
GG(a, b, c, d, x[1], S21, 0xf61e2562)
# 17
GG(d, a, b, c, x[6], S22, 0xc040b340)
# 18
GG(c, d, a, b, x[11], S23, 0x265e5a51)
# 19
GG(b, c, d, a, x[0], S24, 0xe9b6c7aa)
# 20
GG(a, b, c, d, x[5], S21, 0xd62f105d)
# 21
GG(d, a, b, c, x[10], S22, 0x2441453)
# 22
GG(c, d, a, b, x[15], S23, 0xd8a1e681)
# 23
GG(b, c, d, a, x[4], S24, 0xe7d3fbc8)
# 24
GG(a, b, c, d, x[9], S21, 0x21e1cde6)
# 25
GG(d, a, b, c, x[14], S22, 0xc33707d6)
# 26
GG(c, d, a, b, x[3], S23, 0xf4d50d87)
# 27
GG(b, c, d, a, x[8], S24, 0x455a14ed)
# 28
GG(a, b, c, d, x[13], S21, 0xa9e3e905)
# 29
GG(d, a, b, c, x[2], S22, 0xfcefa3f8)
# 30
GG(c, d, a, b, x[7], S23, 0x676f02d9)
# 31
GG(b, c, d, a, x[12], S24, 0x8d2a4c8a)
# 32
# Round 3
HH(a, b, c, d, x[5], S31, 0xfffa3942)
# 33
HH(d, a, b, c, x[8], S32, 0x8771f681)
# 34
HH(c, d, a, b, x[11], S33, 0x6d9d6122)
# 35
HH(b, c, d, a, x[14], S34, 0xfde5380c)
# 36
HH(a, b, c, d, x[1], S31, 0xa4beea44)
# 37
HH(d, a, b, c, x[4], S32, 0x4bdecfa9)
# 38
HH(c, d, a, b, x[7], S33, 0xf6bb4b60)
# 39
HH(b, c, d, a, x[10], S34, 0xbebfbc70)
# 40
HH(a, b, c, d, x[13], S31, 0x289b7ec6)
# 41
HH(d, a, b, c, x[0], S32, 0xeaa127fa)
# 42
HH(c, d, a, b, x[3], S33, 0xd4ef3085)
# 43
HH(b, c, d, a, x[6], S34, 0x4881d05)
# 44
HH(a, b, c, d, x[9], S31, 0xd9d4d039)
# 45
HH(d, a, b, c, x[12], S32, 0xe6db99e5)
# 46
HH(c, d, a, b, x[15], S33, 0x1fa27cf8)
# 47
HH(b, c, d, a, x[2], S34, 0xc4ac5665)
# 48
# Round 4
II(a, b, c, d, x[0], S41, 0xf4292244)
# 49
II(d, a, b, c, x[7], S42, 0x432aff97)
# 50
II(c, d, a, b, x[14], S43, 0xab9423a7)
# 51
II(b, c, d, a, x[5], S44, 0xfc93a039)
# 52
II(a, b, c, d, x[12], S41, 0x655b59c3)
# 53
II(d, a, b, c, x[3], S42, 0x8f0ccc92)
# 54
II(c, d, a, b, x[10], S43, 0xffeff47d)
# 55
II(b, c, d, a, x[1], S44, 0x85845dd1)
# 56
II(a, b, c, d, x[8], S41, 0x6fa87e4f)
# 57
II(d, a, b, c, x[15], S42, 0xfe2ce6e0)
# 58
II(c, d, a, b, x[6], S43, 0xa3014314)
# 59
II(b, c, d, a, x[13], S44, 0x4e0811a1)
# 60
II(a, b, c, d, x[4], S41, 0xf7537e82)
# 61
II(d, a, b, c, x[11], S42, 0xbd3af235)
# 62
II(c, d, a, b, x[2], S43, 0x2ad7d2bb)
# 63
II(b, c, d, a, x[9], S44, 0xeb86d391)
# 64
cal.mov('o0', a)
cal.mov('o1', b)
cal.mov('o2', c)
cal.mov('o3', d)
xcode.release_register(a)
xcode.release_register(b)
xcode.release_register(c)
xcode.release_register(d)
for xi in x:
xcode.release_register(xi)
xcode.set_remote_binding('i0', input_statea)
xcode.set_remote_binding('i1', input_stateb)
xcode.set_remote_binding('i2', input_statec)
xcode.set_remote_binding('i3', input_stated)
for i in range(16): #range(len(input_block)):
xcode.set_remote_binding('i' + str(i + 4), input_block[i])
xcode.set_remote_binding('o0', outputa)
xcode.set_remote_binding('o1', outputb)
xcode.set_remote_binding('o2', outputc)
xcode.set_remote_binding('o3', outputd)
domain = (0, 0, N, N)
global TIME
start_time = time.time()
proc.execute(xcode, domain)
end_time = time.time()
TIME += (end_time - start_time)
for j in range(N):
for i in range(N):
for k in range(4):
parcontext.statea[k + (i + j * N) *
4] += outputa[address_4_2d(i, j) + k]
parcontext.stateb[k + (i + j * N) *
4] += outputb[address_4_2d(i, j) + k]
parcontext.statec[k + (i + j * N) *
4] += outputc[address_4_2d(i, j) + k]
parcontext.stated[k + (i + j * N) *
4] += outputd[address_4_2d(i, j) + k]
proc.free_remote(input_statea)
proc.free_remote(input_stateb)
proc.free_remote(input_statec)
proc.free_remote(input_stated)
for block in input_block:
proc.free_remote(block)
proc.free_remote(outputa)
proc.free_remote(outputb)
proc.free_remote(outputc)
proc.free_remote(outputd)
def MD5Transform(state, block, blocki):
proc = env.Processor(0)
input_state = proc.alloc_remote('I', 4, 1, 1)
input_block = proc.alloc_remote('I', 4, 4, 1)
output = proc.alloc_remote('I', 4, 1, 1)
for i in range(4):
input_state[i] = state[i]
Decode(input_block, block, blocki, 64)
#print map(hex, input_block)
global xcode
if xcode == None:
xcode = env.InstructionStream()
cal.set_active_code(xcode)
S11 = xcode.acquire_register((7, 7, 7, 7))
S12 = xcode.acquire_register((12, 12, 12, 12))
S13 = xcode.acquire_register((17, 17, 17, 17))
S14 = xcode.acquire_register((22, 22, 22, 22))
S21 = xcode.acquire_register((5, 5, 5, 5))
S22 = xcode.acquire_register((9, 9, 9, 9))
S23 = xcode.acquire_register((14, 14, 14, 14))
S24 = xcode.acquire_register((20, 20, 20, 20))
S31 = xcode.acquire_register((4, 4, 4, 4))
S32 = xcode.acquire_register((11, 11, 11, 11))
S33 = xcode.acquire_register((16, 16, 16, 16))
S34 = xcode.acquire_register((23, 23, 23, 23))
S41 = xcode.acquire_register((6, 6, 6, 6))
S42 = xcode.acquire_register((10, 10, 10, 10))
S43 = xcode.acquire_register((15, 15, 15, 15))
S44 = xcode.acquire_register((21, 21, 21, 21))
a = xcode.acquire_register()
b = xcode.acquire_register()
c = xcode.acquire_register()
d = xcode.acquire_register()
x = [xcode.acquire_register() for i in range(16)]
r = xcode.acquire_register()
cal.dcl_cb('cb0[1]')
cal.dcl_cb('cb1[4]')
cal.dcl_output('o0', USAGE=cal.usage.generic)
cal.mov(a, 'cb0[0].x')
cal.mov(b, 'cb0[0].y')
cal.mov(c, 'cb0[0].z')
cal.mov(d, 'cb0[0].w')
for i in range(4):
cal.mov(x[i*4], 'cb1[' + str(i) + '].x')
cal.mov(x[i*4+1], 'cb1[' + str(i) + '].y')
cal.mov(x[i*4+2], 'cb1[' + str(i) + '].z')
cal.mov(x[i*4+3], 'cb1[' + str(i) + '].w')
# Round 1
FF (a, b, c, d, x[ 0], S11, 0xd76aa478); # 1
FF (d, a, b, c, x[ 1], S12, 0xe8c7b756); # 2
FF (c, d, a, b, x[ 2], S13, 0x242070db); # 3
FF (b, c, d, a, x[ 3], S14, 0xc1bdceee); # 4
FF (a, b, c, d, x[ 4], S11, 0xf57c0faf); # 5
FF (d, a, b, c, x[ 5], S12, 0x4787c62a); # 6
FF (c, d, a, b, x[ 6], S13, 0xa8304613); # 7
FF (b, c, d, a, x[ 7], S14, 0xfd469501); # 8
FF (a, b, c, d, x[ 8], S11, 0x698098d8); # 9
FF (d, a, b, c, x[ 9], S12, 0x8b44f7af); # 10
FF (c, d, a, b, x[10], S13, 0xffff5bb1); # 11
FF (b, c, d, a, x[11], S14, 0x895cd7be); # 12
FF (a, b, c, d, x[12], S11, 0x6b901122); # 13
FF (d, a, b, c, x[13], S12, 0xfd987193); # 14
FF (c, d, a, b, x[14], S13, 0xa679438e); # 15
FF (b, c, d, a, x[15], S14, 0x49b40821); # 16
# Round 2
GG (a, b, c, d, x[ 1], S21, 0xf61e2562); # 17
GG (d, a, b, c, x[ 6], S22, 0xc040b340); # 18
GG (c, d, a, b, x[11], S23, 0x265e5a51); # 19
GG (b, c, d, a, x[ 0], S24, 0xe9b6c7aa); # 20
GG (a, b, c, d, x[ 5], S21, 0xd62f105d); # 21
GG (d, a, b, c, x[10], S22, 0x2441453); # 22
GG (c, d, a, b, x[15], S23, 0xd8a1e681); # 23
GG (b, c, d, a, x[ 4], S24, 0xe7d3fbc8); # 24
GG (a, b, c, d, x[ 9], S21, 0x21e1cde6); # 25
GG (d, a, b, c, x[14], S22, 0xc33707d6); # 26
GG (c, d, a, b, x[ 3], S23, 0xf4d50d87); # 27
GG (b, c, d, a, x[ 8], S24, 0x455a14ed); # 28
GG (a, b, c, d, x[13], S21, 0xa9e3e905); # 29
GG (d, a, b, c, x[ 2], S22, 0xfcefa3f8); # 30
GG (c, d, a, b, x[ 7], S23, 0x676f02d9); # 31
GG (b, c, d, a, x[12], S24, 0x8d2a4c8a); # 32
# Round 3
HH (a, b, c, d, x[ 5], S31, 0xfffa3942); # 33
HH (d, a, b, c, x[ 8], S32, 0x8771f681); # 34
HH (c, d, a, b, x[11], S33, 0x6d9d6122); # 35
HH (b, c, d, a, x[14], S34, 0xfde5380c); # 36
HH (a, b, c, d, x[ 1], S31, 0xa4beea44); # 37
HH (d, a, b, c, x[ 4], S32, 0x4bdecfa9); # 38
HH (c, d, a, b, x[ 7], S33, 0xf6bb4b60); # 39
HH (b, c, d, a, x[10], S34, 0xbebfbc70); # 40
HH (a, b, c, d, x[13], S31, 0x289b7ec6); # 41
HH (d, a, b, c, x[ 0], S32, 0xeaa127fa); # 42
HH (c, d, a, b, x[ 3], S33, 0xd4ef3085); # 43
HH (b, c, d, a, x[ 6], S34, 0x4881d05); # 44
HH (a, b, c, d, x[ 9], S31, 0xd9d4d039); # 45
HH (d, a, b, c, x[12], S32, 0xe6db99e5); # 46
HH (c, d, a, b, x[15], S33, 0x1fa27cf8); # 47
HH (b, c, d, a, x[ 2], S34, 0xc4ac5665); # 48
# Round 4
II (a, b, c, d, x[ 0], S41, 0xf4292244); # 49
II (d, a, b, c, x[ 7], S42, 0x432aff97); # 50
II (c, d, a, b, x[14], S43, 0xab9423a7); # 51
II (b, c, d, a, x[ 5], S44, 0xfc93a039); # 52
II (a, b, c, d, x[12], S41, 0x655b59c3); # 53
II (d, a, b, c, x[ 3], S42, 0x8f0ccc92); # 54
II (c, d, a, b, x[10], S43, 0xffeff47d); # 55
II (b, c, d, a, x[ 1], S44, 0x85845dd1); # 56
II (a, b, c, d, x[ 8], S41, 0x6fa87e4f); # 57
II (d, a, b, c, x[15], S42, 0xfe2ce6e0); # 58
II (c, d, a, b, x[ 6], S43, 0xa3014314); # 59
II (b, c, d, a, x[13], S44, 0x4e0811a1); # 60
II (a, b, c, d, x[ 4], S41, 0xf7537e82); # 61
II (d, a, b, c, x[11], S42, 0xbd3af235); # 62
II (c, d, a, b, x[ 2], S43, 0x2ad7d2bb); # 63
II (b, c, d, a, x[ 9], S44, 0xeb86d391); # 64
temp = xcode.acquire_register()
#cal.mov(temp.x___, a.x)
#cal.mov(temp._y__, b.xx)
#cal.mov(temp.__z_, c.xxx)
#cal.mov(temp.___w, d.xxxx)
cal.iadd(temp, a.x000, b('0x00'))
cal.iadd(temp, temp, c('00x0'))
cal.iadd(temp, temp, d('000x'))
cal.mov('o0', temp)
xcode.release_register(a)
xcode.release_register(b)
xcode.release_register(c)
xcode.release_register(d)
for xi in x:
xcode.release_register(xi)
#for i, inst in enumerate(xcode._instructions):
# print inst.render()
xcode.set_remote_binding('cb0', input_state)
xcode.set_remote_binding('cb1', input_block)
xcode.set_remote_binding('o0', output)
domain = (0, 0, 1, 1)
proc.execute(xcode, domain)
state[0] += output[0]
state[1] += output[1]
state[2] += output[2]
state[3] += output[3]
print 'input = ', map(hex, input_state)
print 'output = ', map(hex, output)
proc.free_remote(input_state)
proc.free_remote(input_block)
proc.free_remote(output)
