Author: Richard Plangger <[email protected]>
Branch: vecopt-merge
Changeset: r79683:5190c354f531
Date: 2015-09-18 11:45 +0200
http://bitbucket.org/pypy/pypy/changeset/5190c354f531/
Log: beating the integraiton tests, roughly half work already
diff --git a/rpython/jit/metainterp/optimizeopt/__init__.py
b/rpython/jit/metainterp/optimizeopt/__init__.py
--- a/rpython/jit/metainterp/optimizeopt/__init__.py
+++ b/rpython/jit/metainterp/optimizeopt/__init__.py
@@ -32,9 +32,9 @@
def build_opt_chain(metainterp_sd, enable_opts):
optimizations = []
unroll = 'unroll' in enable_opts # 'enable_opts' is normally a dict
- #if (metainterp_sd.cpu is not None and
- # not metainterp_sd.cpu.supports_guard_gc_type):
- # unroll = False
+ if (metainterp_sd.cpu is not None and
+ not metainterp_sd.cpu.supports_guard_gc_type):
+ unroll = False
for name, opt in unroll_all_opts:
if name in enable_opts:
if opt is not None:
diff --git a/rpython/jit/metainterp/optimizeopt/dependency.py
b/rpython/jit/metainterp/optimizeopt/dependency.py
--- a/rpython/jit/metainterp/optimizeopt/dependency.py
+++ b/rpython/jit/metainterp/optimizeopt/dependency.py
@@ -71,7 +71,11 @@
if exclude_last:
count -= 1
while i < count:
- op = self.path[i].getoperation()
+ node = self.path[i]
+ if node.is_imaginary():
+ i += 1
+ continue
+ op = node.getoperation()
if op.is_guard():
descr = op.getdescr()
if not descr:
@@ -189,9 +193,6 @@
isinstance(descr, compile.CompileLoopVersionDescr)
return False
- # TODO def is_guard_early_exit(self):
- # return self.op.getopnum() == rop.GUARD_EARLY_EXIT
-
def loads_from_complex_object(self):
return rop._ALWAYS_PURE_LAST <= self.op.getopnum() < rop._MALLOC_FIRST
@@ -1057,33 +1058,27 @@
return self.constant - other.constant
def emit_operations(self, opt, result_box=None):
- box = self.var
+ var = self.var
if self.is_identity():
- return box
+ return var
last_op = None
if self.coefficient_mul != 1:
- box_result = box.clonebox()
- last_op = ResOperation(rop.INT_MUL, [box,
ConstInt(self.coefficient_mul)], box_result)
- opt.emit_operation(last_op)
- box = box_result
+ args = [var, ConstInt(self.coefficient_mul)]
+ var = ResOperation(rop.INT_MUL, args)
+ opt.emit_operation(var)
if self.coefficient_div != 1:
- box_result = box.clonebox()
- last_op = ResOperation(rop.INT_FLOORDIV, [box,
ConstInt(self.coefficient_div)], box_result)
- opt.emit_operation(last_op)
- box = box_result
+ args = [var, ConstInt(self.coefficient_div)]
+ var = ResOperation(rop.INT_FLOORDIV, args)
+ opt.emit_operation(var)
if self.constant > 0:
- box_result = box.clonebox()
- last_op = ResOperation(rop.INT_ADD, [box,
ConstInt(self.constant)], box_result)
- opt.emit_operation(last_op)
- box = box_result
+ args = [var, ConstInt(self.constant)]
+ vec = ResOperation(rop.INT_ADD, args)
+ opt.emit_operation(vec)
if self.constant < 0:
- box_result = box.clonebox()
- last_op = ResOperation(rop.INT_SUB, [box,
ConstInt(self.constant)], box_result)
- opt.emit_operation(last_op)
- box = box_result
- if result_box is not None:
- last_op.result = box = result_box
- return box
+ args = [var, ConstInt(self.constant)]
+ var = ResOperation(rop.INT_SUB, args)
+ opt.emit_operation(var)
+ return var
def compare(self, other):
""" Returns if the two are compareable as a first result
diff --git a/rpython/jit/metainterp/optimizeopt/guard.py
b/rpython/jit/metainterp/optimizeopt/guard.py
--- a/rpython/jit/metainterp/optimizeopt/guard.py
+++ b/rpython/jit/metainterp/optimizeopt/guard.py
@@ -133,15 +133,13 @@
def emit_operations(self, opt):
# create trace instructions for the index
- box_lhs = self.emit_varops(opt, self.lhs, self.cmp_op.getarg(0))
- box_rhs = self.emit_varops(opt, self.rhs, self.cmp_op.getarg(1))
- box_result = self.cmp_op.result.clonebox()
+ lhs = self.emit_varops(opt, self.lhs, self.cmp_op.getarg(0))
+ rhs = self.emit_varops(opt, self.rhs, self.cmp_op.getarg(1))
opnum = self.cmp_op.getopnum()
- cmp_op = ResOperation(opnum, [box_lhs, box_rhs], box_result)
+ cmp_op = ResOperation(opnum, [lhs, rhs])
opt.emit_operation(cmp_op)
# emit that actual guard
- guard = self.op.clone()
- guard.setarg(0, box_result)
+ guard = ResOperation(self.op.getopnum(), [cmp_op], self.op.getdescr())
opt.emit_operation(guard)
self.setindex(opt.operation_position()-1)
self.setoperation(guard)
@@ -179,13 +177,12 @@
class GuardStrengthenOpt(object):
""" Note that this optimization is only used in the vector optimizer (yet)
"""
- def __init__(self, index_vars, has_two_labels):
+ def __init__(self, index_vars):
self.index_vars = index_vars
self._newoperations = []
self.strength_reduced = 0 # how many guards could be removed?
self.strongest_guards = {}
self.guards = {}
- self.has_two_labels = has_two_labels
def collect_guard_information(self, loop):
operations = loop.operations
@@ -251,11 +248,11 @@
else:
self.emit_operation(op)
continue
- if op.result:
- index_var = self.index_vars.get(op.result, None)
+ if not op.returns_void():
+ index_var = self.index_vars.get(op, None)
if index_var:
if not index_var.is_identity():
- index_var.emit_operations(self, op.result)
+ index_var.emit_operations(self, op)
continue
self.emit_operation(op)
#
diff --git a/rpython/jit/metainterp/optimizeopt/schedule.py
b/rpython/jit/metainterp/optimizeopt/schedule.py
--- a/rpython/jit/metainterp/optimizeopt/schedule.py
+++ b/rpython/jit/metainterp/optimizeopt/schedule.py
@@ -124,7 +124,6 @@
""" Emit all the operations into the oplist parameter.
Initiates the scheduling. """
assert isinstance(state, SchedulerState)
- import pdb; pdb.set_trace()
while state.has_more():
node = self.next(state)
if node:
diff --git a/rpython/jit/metainterp/optimizeopt/test/test_dependency.py
b/rpython/jit/metainterp/optimizeopt/test/test_dependency.py
--- a/rpython/jit/metainterp/optimizeopt/test/test_dependency.py
+++ b/rpython/jit/metainterp/optimizeopt/test/test_dependency.py
@@ -51,8 +51,10 @@
else:
label = loop.operations[0]
label.setdescr(TargetToken(token))
- loop = VectorLoop(label, loop.operations[0:-1], loop.operations[-1])
+ jump = loop.operations[-1]
+ loop = VectorLoop(label, loop.operations[0:-1], jump)
loop.jump.setdescr(token)
+ # TODO
for op in loop.operations:
if op.getopnum() == rop.GUARD_EARLY_EXIT and op.getdescr() is None:
op.setdescr(ResumeAtLoopHeaderDescr())
@@ -184,10 +186,6 @@
assert not m1.is_adjacent_to(m2)
assert not m2.is_adjacent_to(m1)
- def getmemref(self, idx):
- node = self.last_graph.getnode(idx)
- return self.last_graph.memory_refs[node]
-
class BaseTestDependencyGraph(DependencyBaseTest):
def test_index_var_basic(self):
@@ -264,9 +262,9 @@
jump() # 4:
"""
graph = self.assert_dependencies(ops, full_check=True)
- self.assert_dependent(1,2)
- self.assert_dependent(2,3)
- self.assert_dependent(1,3)
+ self.assert_dependent(graph, 1,2)
+ self.assert_dependent(graph, 2,3)
+ self.assert_dependent(graph, 1,3)
def test_def_use_jump_use_def(self):
ops = """
@@ -417,7 +415,7 @@
jump(p0, i1, i2) # 3:
"""
self.assert_dependencies(ops, full_check=True)
- self.assert_dependent(1,2)
+ self.assert_dependent(graph, 1,2)
def test_setarrayitem_dont_depend_with_memref_info(self):
ops="""
@@ -457,7 +455,7 @@
jump(i24, i19, i21, i3, i4, i5, i22, i7) # 21:
"""
self.assert_dependencies(ops, full_check=False)
- self.assert_dependent(2,12)
+ self.assert_dependent(graph, 2,12)
def test_getfield(self):
trace = """
diff --git a/rpython/jit/metainterp/optimizeopt/test/test_vecopt.py
b/rpython/jit/metainterp/optimizeopt/test/test_vecopt.py
--- a/rpython/jit/metainterp/optimizeopt/test/test_vecopt.py
+++ b/rpython/jit/metainterp/optimizeopt/test/test_vecopt.py
@@ -11,20 +11,31 @@
import rpython.jit.metainterp.optimizeopt.optimizer as optimizeopt
import rpython.jit.metainterp.optimizeopt.virtualize as virtualize
from rpython.jit.metainterp.optimizeopt.dependency import DependencyGraph
-from rpython.jit.metainterp.optimizeopt.vector import (VectorizingOptimizer,
MemoryRef,
- isomorphic, Pair, NotAVectorizeableLoop, NotAProfitableLoop,
GuardStrengthenOpt,
- CostModel, VectorLoop)
-from rpython.jit.metainterp.optimizeopt.schedule import (Scheduler,
SchedulerState)
+from rpython.jit.metainterp.optimizeopt.vector import (VectorizingOptimizer,
+ MemoryRef, isomorphic, Pair, NotAVectorizeableLoop, VectorLoop,
+ NotAProfitableLoop, GuardStrengthenOpt, CostModel, X86_CostModel)
+from rpython.jit.metainterp.optimizeopt.schedule import (Scheduler,
+ SchedulerState, VecScheduleState)
from rpython.jit.metainterp.optimize import InvalidLoop
from rpython.jit.metainterp import compile
from rpython.jit.metainterp.resoperation import rop, ResOperation
+from rpython.jit.metainterp.optimizeopt.version import LoopVersionInfo
class FakeJitDriverStaticData(object):
vec=True
+class FakeLoopInfo(LoopVersionInfo):
+ def __init__(self, loop):
+ self.target_token = loop.label.getdescr()
+ self.label_op = loop.label
+ self.insert_index = -1
+ self.versions = []
+ self.leads_to = {}
+ self.descrs = []
+
class FakeCostModel(CostModel):
- def __init__(self):
- CostModel.__init__(self, 0, 16)
+ def __init__(self, cpu):
+ CostModel.__init__(self, cpu, 16)
def record_cast_int(self): pass
def record_pack_savings(self, pack, times): pass
def record_vector_pack(self, box, index, count): pass
@@ -34,6 +45,19 @@
def profitable(self):
return True
+def index_of_first(opnum, operations, pass_by=0):
+ for i,op in enumerate(operations):
+ if op.getopnum() == opnum:
+ if pass_by == 0:
+ return i
+ else:
+ pass_by -= 1
+ return -1
+
+def find_first_index(loop, opnum, pass_by=0):
+ """ return the first index of the operation having the same opnum or -1 """
+ return index_of_first(opnum, loop.operations, pass_by)
+
ARCH_VEC_REG_SIZE = 16
class VecTestHelper(DependencyBaseTest):
@@ -43,7 +67,11 @@
jitdriver_sd = FakeJitDriverStaticData()
def assert_vectorize(self, loop, expected_loop, call_pure_results=None):
- self._do_optimize_loop(loop)
+ jump = ResOperation(rop.LABEL, loop.jump.getarglist(),
loop.jump.getdescr())
+ compile_data = compile.LoopCompileData(loop.label, jump,
loop.operations)
+ state = self._do_optimize_loop(compile_data)
+ loop.label = state[0].label_op
+ loop.opererations = state[1]
self.assert_equal(loop, expected_loop)
def vectoroptimizer(self, loop):
@@ -56,13 +84,11 @@
def earlyexit(self, loop):
opt = self.vectoroptimizer(loop)
graph = opt.analyse_index_calculations(loop)
- graph.view()
state = SchedulerState(graph)
opt.schedule(state)
return graph.loop
def vectoroptimizer_unrolled(self, loop, unroll_factor = -1):
- loop.snapshot()
opt = self.vectoroptimizer(loop)
opt.linear_find_smallest_type(loop)
if unroll_factor == -1 and opt.smallest_type_bytes == 0:
@@ -71,76 +97,84 @@
unroll_factor = opt.get_unroll_count(ARCH_VEC_REG_SIZE)
print ""
print "unroll factor: ", unroll_factor, opt.smallest_type_bytes
- if opt.loop.find_first_index(rop.GUARD_EARLY_EXIT) == -1:
- idx = loop.find_first_index(rop.LABEL)
- guard = ResOperation(rop.GUARD_EARLY_EXIT, [], None)
- guard.setfailargs([])
- guard.setdescr(compile.ResumeAtLoopHeaderDescr())
- loop.operations.insert(idx+1, guard)
- self.show_dot_graph(DependencyGraph(opt.loop), "original_" +
self.test_name)
- graph = opt.analyse_index_calculations()
+ # TODO if opt.loop.find_first_index(rop.GUARD_EARLY_EXIT) == -1:
+ # idx = loop.find_first_index(rop.LABEL)
+ # guard = ResOperation(rop.GUARD_EARLY_EXIT, [], None)
+ # guard.setfailargs([])
+ # guard.setdescr(compile.ResumeAtLoopHeaderDescr())
+ # loop.operations.insert(idx+1, guard)
+ self.show_dot_graph(DependencyGraph(loop), "original_" +
self.test_name)
+ graph = opt.analyse_index_calculations(loop)
if graph is not None:
- cycle = opt.dependency_graph.cycles()
+ cycle = graph.cycles()
if cycle is not None:
print "CYCLE found %s" % cycle
- self.show_dot_graph(opt.dependency_graph, "early_exit_" +
self.test_name)
+ self.show_dot_graph(graph, "early_exit_" + self.test_name)
assert cycle is None
- loop.operations = opt.schedule(False)
+ state = SchedulerState(graph)
+ opt.schedule(state)
opt.unroll_loop_iterations(loop, unroll_factor)
- opt.loop.operations = opt.get_newoperations()
- self.debug_print_operations(opt.loop)
- opt.clear_newoperations()
+ self.debug_print_operations(loop)
graph = DependencyGraph(loop)
- self.last_graph = graph
- self.show_dot_graph(self.last_graph, self.test_name)
+ self.last_graph = graph # legacy for test_dependency
+ self.show_dot_graph(graph, self.test_name)
+ def gmr(i):
+ return graph.memory_refs[graph.nodes[i]]
+ graph.getmemref = gmr
return opt, graph
def init_packset(self, loop, unroll_factor = -1):
opt, graph = self.vectoroptimizer_unrolled(loop, unroll_factor)
opt.find_adjacent_memory_refs(graph)
- return opt
+ return opt, graph
def extend_packset(self, loop, unroll_factor = -1):
opt, graph = self.vectoroptimizer_unrolled(loop, unroll_factor)
opt.find_adjacent_memory_refs(graph)
opt.extend_packset()
- return opt
+ return opt, graph
def combine_packset(self, loop, unroll_factor = -1):
opt, graph = self.vectoroptimizer_unrolled(loop, unroll_factor)
opt.find_adjacent_memory_refs(graph)
opt.extend_packset()
opt.combine_packset()
- return opt
+ return opt, graph
def schedule(self, loop, unroll_factor = -1, with_guard_opt=False):
+ info = FakeLoopInfo(loop)
+ info.snapshot(loop.operations + [loop.jump], loop.label)
opt, graph = self.vectoroptimizer_unrolled(loop, unroll_factor)
- opt.costmodel = FakeCostModel()
opt.find_adjacent_memory_refs(graph)
opt.extend_packset()
opt.combine_packset()
- opt.schedule(graph, True)
+ costmodel = FakeCostModel(self.cpu)
+ state = VecScheduleState(graph, opt.packset, self.cpu, costmodel)
+ opt.schedule(state)
if with_guard_opt:
- gso = GuardStrengthenOpt(opt.dependency_graph.index_vars,
opt.has_two_labels)
- gso.propagate_all_forward(opt.loop)
+ gso = GuardStrengthenOpt(graph.index_vars)
+ gso.propagate_all_forward(info, loop)
return opt
def vectorize(self, loop, unroll_factor = -1):
- opt = self.vectoroptimizer_unrolled(loop, unroll_factor)
- opt.find_adjacent_memory_refs()
+ info = FakeLoopInfo(loop)
+ info.snapshot(loop.operations + [loop.jump], loop.label)
+ opt, graph = self.vectoroptimizer_unrolled(loop, unroll_factor)
+ opt.find_adjacent_memory_refs(graph)
opt.extend_packset()
opt.combine_packset()
- opt.costmodel.reset_savings()
- opt.schedule(True)
- if not opt.costmodel.profitable():
+ costmodel = X86_CostModel(self.cpu, 0)
+ state = VecScheduleState(graph, opt.packset, self.cpu, costmodel)
+ opt.schedule(state)
+ if not costmodel.profitable():
raise NotAProfitableLoop()
- gso = GuardStrengthenOpt(opt.dependency_graph.index_vars,
opt.has_two_labels)
- gso.propagate_all_forward(opt.loop)
+ gso = GuardStrengthenOpt(graph.index_vars)
+ gso.propagate_all_forward(info, loop)
return opt
def assert_unroll_loop_equals(self, loop, expected_loop, \
unroll_factor = -1):
- vectoroptimizer = self.vectoroptimizer_unrolled(loop, unroll_factor)
+ self.vectoroptimizer_unrolled(loop, unroll_factor)
self.assert_equal(loop, expected_loop)
def assert_pack(self, pack, indices):
@@ -171,23 +205,24 @@
def assert_packset_not_contains_pair(self, packset, x, y):
for pack in packset.packs:
- if pack.left.opidx == x and \
- pack.right.opidx == y:
+ if pack.leftmost(node=True).opidx == x and \
+ pack.rightmost(node=True).opidx == y:
pytest.fail("must not find packset with indices {x},{y}" \
.format(x=x,y=y))
def assert_packset_contains_pair(self, packset, x, y):
for pack in packset.packs:
if isinstance(pack, Pair):
- if pack.left.opidx == x and \
- pack.right.opidx == y:
+ if pack.leftmost(node=True).opidx == x and \
+ pack.rightmost(node=True).opidx == y:
break
else:
pytest.fail("can't find a pack set for indices {x},{y}" \
.format(x=x,y=y))
- def assert_has_memory_ref_at(self, idx):
- node = self.last_graph.nodes[idx]
- assert node in self.last_graph.memory_refs, \
+ def assert_has_memory_ref_at(self, graph, idx):
+ idx -= 1 # label is not in the nodes
+ node = graph.nodes[idx]
+ assert node in graph.memory_refs, \
"operation %s at pos %d has no memory ref!" % \
(node.getoperation(), node.getindex())
@@ -248,7 +283,7 @@
""" it currently rejects pointer arrays """
ops = """
[p0,i0]
- raw_load_r(p0,i0,descr=arraydescr2)
+ getarrayitem_gc_r(p0,i0,descr=arraydescr2)
jump(p0,i0)
"""
self.assert_vectorize(self.parse_loop(ops), self.parse_loop(ops))
@@ -257,9 +292,9 @@
""" it currently rejects pointer arrays """
ops = """
[p0,i0]
- i2 = getarrayitem_gc(p0,i0,descr=floatarraydescr)
+ i2 = getarrayitem_gc_i(p0,i0,descr=arraydescr)
i1 = int_add(i0,1)
- i3 = getarrayitem_gc(p0,i1,descr=floatarraydescr)
+ i3 = getarrayitem_gc_i(p0,i1,descr=arraydescr)
i4 = int_add(i1,1)
jump(p0,i4)
"""
@@ -267,11 +302,12 @@
[p0,i0]
i1 = int_add(i0,1)
i2 = int_add(i0,2)
- i3 = vec_getarrayitem_gc(p0,i0,2,descr=floatarraydescr)
+ v3[2xi64] = vec_getarrayitem_gc_i(p0,i0,descr=arraydescr)
jump(p0,i2)
"""
- vopt = self.vectorize(self.parse_loop(ops),0)
- self.assert_equal(vopt.loop, self.parse_loop(opt))
+ loop = self.parse_loop(ops)
+ vopt = self.vectorize(loop,0)
+ self.assert_equal(loop, self.parse_loop(opt))
def test_vect_unroll_char(self):
""" a 16 byte vector register can hold 16 bytes thus
@@ -306,14 +342,14 @@
[p0,p1,p2,i0]
i4 = int_add(i0, 1)
i5 = int_le(i4, 10)
- guard_true(i5) []
+ guard_true(i5) [p0,p1,p2,i0]
i1 = raw_load_i(p1, i0, descr=floatarraydescr)
i2 = raw_load_i(p2, i0, descr=floatarraydescr)
i3 = int_add(i1,i2)
raw_store(p0, i0, i3, descr=floatarraydescr)
i9 = int_add(i4, 1)
i10 = int_le(i9, 10)
- guard_true(i10) []
+ guard_true(i10) [p0,p1,p2,i4]
i6 = raw_load_i(p1, i4, descr=floatarraydescr)
i7 = raw_load_i(p2, i4, descr=floatarraydescr)
i8 = int_add(i6,i7)
@@ -338,9 +374,9 @@
raw_load_i(p0,i0,descr=arraydescr)
jump(p0,i0)
"""
- vopt = self.vectoroptimizer_unrolled(self.parse_loop(ops),0)
- assert len(vopt.dependency_graph.memory_refs) == 1
- self.assert_has_memory_ref_at(1)
+ vopt, graph = self.vectoroptimizer_unrolled(self.parse_loop(ops),0)
+ assert len(graph.memory_refs) == 1
+ self.assert_has_memory_ref_at(graph, 1)
def test_array_operation_indices_unrolled_1(self):
ops = """
@@ -348,10 +384,10 @@
raw_load_i(p0,i0,descr=chararraydescr)
jump(p0,i0)
"""
- vopt = self.vectoroptimizer_unrolled(self.parse_loop(ops),1)
- assert len(vopt.dependency_graph.memory_refs) == 2
- self.assert_has_memory_ref_at(1)
- self.assert_has_memory_ref_at(2)
+ vopt, graph = self.vectoroptimizer_unrolled(self.parse_loop(ops),1)
+ assert len(graph.memory_refs) == 2
+ self.assert_has_memory_ref_at(graph, 1)
+ self.assert_has_memory_ref_at(graph, 2)
def test_array_operation_indices_unrolled_2(self):
ops = """
@@ -361,20 +397,20 @@
jump(p0,i3,i4)
"""
loop = self.parse_loop(ops)
- vopt = self.vectoroptimizer_unrolled(loop,0)
- assert len(vopt.dependency_graph.memory_refs) == 2
- self.assert_has_memory_ref_at(1)
- self.assert_has_memory_ref_at(2)
+ vopt, graph = self.vectoroptimizer_unrolled(loop,0)
+ assert len(graph.memory_refs) == 2
+ self.assert_has_memory_ref_at(graph, 1)
+ self.assert_has_memory_ref_at(graph, 2)
#
- vopt = self.vectoroptimizer_unrolled(self.parse_loop(ops),1)
- assert len(vopt.dependency_graph.memory_refs) == 4
+ vopt, graph = self.vectoroptimizer_unrolled(self.parse_loop(ops),1)
+ assert len(graph.memory_refs) == 4
for i in [1,2,3,4]:
- self.assert_has_memory_ref_at(i)
+ self.assert_has_memory_ref_at(graph, i)
#
- vopt = self.vectoroptimizer_unrolled(self.parse_loop(ops),3)
- assert len(vopt.dependency_graph.memory_refs) == 8
+ vopt, graph = self.vectoroptimizer_unrolled(self.parse_loop(ops),3)
+ assert len(graph.memory_refs) == 8
for i in [1,2,3,4,5,6,7,8]:
- self.assert_has_memory_ref_at(i)
+ self.assert_has_memory_ref_at(graph, i)
def test_array_memory_ref_adjacent_1(self):
ops = """
@@ -384,12 +420,12 @@
jump(p0,i1)
"""
loop = self.parse_loop(ops)
- vopt = self.vectoroptimizer_unrolled(loop,1)
- vopt.find_adjacent_memory_refs()
- assert len(vopt.dependency_graph.memory_refs) == 2
+ vopt, graph = self.vectoroptimizer_unrolled(loop,1)
+ vopt.find_adjacent_memory_refs(graph)
+ assert len(graph.memory_refs) == 2
- mref1 = self.getmemref(loop.find_first_index(rop.RAW_LOAD))
- mref3 = self.getmemref(loop.find_first_index(rop.RAW_LOAD,1))
+ mref1 = graph.getmemref(find_first_index(loop, rop.RAW_LOAD_I))
+ mref3 = graph.getmemref(find_first_index(loop, rop.RAW_LOAD_I,1))
assert isinstance(mref1, MemoryRef)
assert isinstance(mref3, MemoryRef)
@@ -402,9 +438,9 @@
i3 = raw_load_i(p0,i0,descr=chararraydescr)
jump(p0,i0)
"""
- vopt = self.vectoroptimizer_unrolled(self.parse_loop(ops),0)
- vopt.find_adjacent_memory_refs()
- mref1 = self.getmemref(1)
+ vopt, graph = self.vectoroptimizer_unrolled(self.parse_loop(ops),0)
+ vopt.find_adjacent_memory_refs(graph)
+ mref1 = graph.getmemref(0)
assert isinstance(mref1, MemoryRef)
assert mref1.index_var.coefficient_mul == 1
assert mref1.index_var.constant == 0
@@ -416,9 +452,9 @@
i3 = raw_load_i(p0,i1,descr=chararraydescr)
jump(p0,i1)
"""
- vopt = self.vectoroptimizer_unrolled(self.parse_loop(ops),0)
- vopt.find_adjacent_memory_refs()
- mref1 = self.getmemref(2)
+ vopt, graph = self.vectoroptimizer_unrolled(self.parse_loop(ops),0)
+ vopt.find_adjacent_memory_refs(graph)
+ mref1 = graph.getmemref(1)
assert isinstance(mref1, MemoryRef)
assert mref1.index_var.coefficient_mul == 1
assert mref1.index_var.constant == 1
@@ -430,9 +466,9 @@
i3 = raw_load_i(p0,i1,descr=chararraydescr)
jump(p0,i1)
"""
- vopt = self.vectoroptimizer_unrolled(self.parse_loop(ops),0)
- vopt.find_adjacent_memory_refs()
- mref1 = self.getmemref(2)
+ vopt, graph = self.vectoroptimizer_unrolled(self.parse_loop(ops),0)
+ vopt.find_adjacent_memory_refs(graph)
+ mref1 = graph.getmemref(1)
assert isinstance(mref1, MemoryRef)
assert mref1.index_var.coefficient_mul == 1
assert mref1.index_var.constant == -1
@@ -445,9 +481,9 @@
i3 = raw_load_i(p0,i2,descr=chararraydescr)
jump(p0,i1)
"""
- vopt = self.vectoroptimizer_unrolled(self.parse_loop(ops),0)
- vopt.find_adjacent_memory_refs()
- mref1 = self.getmemref(3)
+ vopt, graph = self.vectoroptimizer_unrolled(self.parse_loop(ops),0)
+ vopt.find_adjacent_memory_refs(graph)
+ mref1 = graph.getmemref(2)
assert isinstance(mref1, MemoryRef)
assert mref1.index_var.coefficient_mul == 3
assert mref1.index_var.constant == 3
@@ -462,9 +498,9 @@
i5 = raw_load_i(p0,i4,descr=chararraydescr)
jump(p0,i4)
"""
- vopt = self.vectoroptimizer_unrolled(self.parse_loop(ops),0)
- vopt.find_adjacent_memory_refs()
- mref1 = self.getmemref(5)
+ vopt, graph = self.vectoroptimizer_unrolled(self.parse_loop(ops),0)
+ vopt.find_adjacent_memory_refs(graph)
+ mref1 = graph.getmemref(4)
assert isinstance(mref1, MemoryRef)
assert mref1.index_var.coefficient_mul == 18
assert mref1.index_var.constant == 48
@@ -480,9 +516,9 @@
i7 = raw_load_i(p0,i6,descr=chararraydescr)
jump(p0,i6)
"""
- vopt = self.vectoroptimizer_unrolled(self.parse_loop(ops),0)
- vopt.find_adjacent_memory_refs()
- mref1 = self.getmemref(7)
+ vopt, graph = self.vectoroptimizer_unrolled(self.parse_loop(ops),0)
+ vopt.find_adjacent_memory_refs(graph)
+ mref1 = graph.getmemref(6)
assert isinstance(mref1, MemoryRef)
assert mref1.index_var.coefficient_mul == 1026
assert mref1.index_var.coefficient_div == 1
@@ -498,9 +534,9 @@
i5 = raw_load_i(p0,i4,descr=chararraydescr)
jump(p0,i4)
"""
- vopt = self.vectoroptimizer_unrolled(self.parse_loop(ops),0)
- vopt.find_adjacent_memory_refs()
- mref1 = self.getmemref(5)
+ vopt, graph = self.vectoroptimizer_unrolled(self.parse_loop(ops),0)
+ vopt.find_adjacent_memory_refs(graph)
+ mref1 = graph.getmemref(4)
assert isinstance(mref1, MemoryRef)
assert mref1.index_var.coefficient_mul == 6
assert mref1.index_var.coefficient_div == 1
@@ -516,16 +552,16 @@
jump(p0,i1,i6)
"""
loop = self.parse_loop(ops)
- vopt = self.vectoroptimizer_unrolled(loop,1)
- vopt.find_adjacent_memory_refs()
+ vopt, graph = self.vectoroptimizer_unrolled(loop,1)
+ vopt.find_adjacent_memory_refs(graph)
- f = lambda x: loop.find_first_index(rop.RAW_LOAD, x)
+ f = lambda x: find_first_index(loop, rop.RAW_LOAD_I, x)
indices = [f(0),f(1),f(2),f(3)]
for i in indices:
- self.assert_has_memory_ref_at(i)
- assert len(vopt.dependency_graph.memory_refs) == 4
+ self.assert_has_memory_ref_at(graph, i+1)
+ assert len(graph.memory_refs) == 4
- mref1, mref3, mref5, mref7 = [self.getmemref(i) for i in indices]
+ mref1, mref3, mref5, mref7 = [graph.getmemref(i) for i in indices]
assert isinstance(mref1, MemoryRef)
assert isinstance(mref3, MemoryRef)
assert isinstance(mref5, MemoryRef)
@@ -545,9 +581,9 @@
i3 = raw_load_i(p0,i2,descr=chararraydescr)
jump(p0,i2)
"""
- vopt = self.vectoroptimizer_unrolled(self.parse_loop(ops),0)
- vopt.find_adjacent_memory_refs()
- mref = self.getmemref(3)
+ vopt, graph = self.vectoroptimizer_unrolled(self.parse_loop(ops),0)
+ vopt.find_adjacent_memory_refs(graph)
+ mref = graph.getmemref(2)
assert mref.index_var.coefficient_div == 16
ops = """
[p0,i0]
@@ -556,9 +592,9 @@
i3 = raw_load_i(p0,i2,descr=chararraydescr)
jump(p0,i2)
"""
- vopt = self.vectoroptimizer_unrolled(self.parse_loop(ops),0)
- vopt.find_adjacent_memory_refs()
- mref = self.getmemref(3)
+ vopt, graph = self.vectoroptimizer_unrolled(self.parse_loop(ops),0)
+ vopt.find_adjacent_memory_refs(graph)
+ mref = graph.getmemref(2)
assert mref.index_var.coefficient_div == 2
assert mref.index_var.constant == 4
ops = """
@@ -571,10 +607,10 @@
i6 = raw_load_i(p0,i5,descr=chararraydescr)
jump(p0,i2)
"""
- vopt = self.vectoroptimizer_unrolled(self.parse_loop(ops),0)
- vopt.find_adjacent_memory_refs()
- mref = self.getmemref(5)
- mref2 = self.getmemref(6)
+ vopt, graph = self.vectoroptimizer_unrolled(self.parse_loop(ops),0)
+ vopt.find_adjacent_memory_refs(graph)
+ mref = graph.getmemref(2)
+ mref2 = graph.getmemref(5)
self.assert_memory_ref_not_adjacent(mref, mref2)
assert mref != mref2
@@ -591,10 +627,10 @@
i7 = raw_load_i(p0,i6,descr=chararraydescr)
jump(p0,i2)
"""
- vopt = self.vectoroptimizer_unrolled(self.parse_loop(ops),0)
- vopt.find_adjacent_memory_refs()
- mref = self.getmemref(6)
- mref2 = self.getmemref(7)
+ vopt, graph = self.vectoroptimizer_unrolled(self.parse_loop(ops),0)
+ vopt.find_adjacent_memory_refs(graph)
+ mref = graph.getmemref(2)
+ mref2 = graph.getmemref(6)
self.assert_memory_ref_not_adjacent(mref, mref2)
assert mref == mref2
@@ -611,10 +647,10 @@
i7 = raw_load_i(p0,i6,descr=chararraydescr)
jump(p0,i2)
"""
- vopt = self.vectoroptimizer_unrolled(self.parse_loop(ops),0)
- vopt.find_adjacent_memory_refs()
- mref = self.getmemref(6)
- mref2 = self.getmemref(7)
+ vopt, graph = self.vectoroptimizer_unrolled(self.parse_loop(ops),0)
+ vopt.find_adjacent_memory_refs(graph)
+ mref = graph.getmemref(2)
+ mref2 = graph.getmemref(6)
self.assert_memory_ref_not_adjacent(mref, mref2)
assert mref != mref2
@@ -622,17 +658,17 @@
def test_packset_init_simple(self):
ops = """
[p0,i0]
- i3 = getarrayitem_raw(p0, i0, descr=chararraydescr)
+ i3 = getarrayitem_raw_i(p0, i0, descr=chararraydescr)
i1 = int_add(i0, 1)
i2 = int_le(i1, 16)
guard_true(i2) [p0, i0]
jump(p0,i1)
"""
loop = self.parse_loop(ops)
- vopt = self.init_packset(loop,1)
+ vopt, graph = self.init_packset(loop,1)
self.assert_independent(4,8)
assert vopt.packset is not None
- assert len(vopt.dependency_graph.memory_refs) == 2
+ assert len(graph.memory_refs) == 2
assert len(vopt.packset.packs) == 1
def test_packset_init_raw_load_not_adjacent_and_adjacent(self):
@@ -642,8 +678,8 @@
jump(p0,i0)
"""
loop = self.parse_loop(ops)
- vopt = self.init_packset(loop,3)
- assert len(vopt.dependency_graph.memory_refs) == 4
+ vopt, graph = self.init_packset(loop,3)
+ assert len(graph.memory_refs) == 4
assert len(vopt.packset.packs) == 0
ops = """
[p0,i0]
@@ -652,8 +688,8 @@
jump(p0,i2)
"""
loop = self.parse_loop(ops)
- vopt = self.init_packset(loop,3)
- assert len(vopt.dependency_graph.memory_refs) == 4
+ vopt, graph = self.init_packset(loop,3)
+ assert len(graph.memory_refs) == 4
assert len(vopt.packset.packs) == 3
for i in range(3):
x = (i+1)*2
@@ -667,24 +703,24 @@
i1 = int_add(i0, 1)
i2 = int_le(i1, 16)
guard_true(i2) [p0, i0]
- i3 = getarrayitem_raw(p0, i1, descr=chararraydescr)
+ i3 = getarrayitem_raw_i(p0, i1, descr=chararraydescr)
jump(p0,i1)
"""
loop = self.parse_loop(ops)
- vopt = self.init_packset(loop,15)
- assert len(vopt.dependency_graph.memory_refs) == 16
+ vopt, graph = self.init_packset(loop,15)
+ assert len(graph.memory_refs) == 16
assert len(vopt.packset.packs) == 15
# assure that memory refs are not adjacent for all
for i in range(15):
for j in range(15):
try:
if i-4 == j or i+4 == j:
- mref1 = self.getmemref(i)
- mref2 = self.getmemref(j)
+ mref1 = graph.getmemref(i)
+ mref2 = graph.getmemref(j)
assert mref1.is_adjacent_to(mref2)
else:
- mref1 = self.getmemref(i)
- mref2 = self.getmemref(j)
+ mref1 = graph.getmemref(i)
+ mref2 = graph.getmemref(j)
assert not mref1.is_adjacent_to(mref2)
except KeyError:
pass
@@ -697,25 +733,20 @@
def test_isomorphic_operations(self):
ops_src = """
[p1,p0,i0]
- i3 = getarrayitem_raw(p0, i0, descr=chararraydescr)
+ i3 = getarrayitem_raw_i(p0, i0, descr=chararraydescr)
i1 = int_add(i0, 1)
i2 = int_le(i1, 16)
- i4 = getarrayitem_raw(p0, i1, descr=chararraydescr)
- i5 = getarrayitem_raw(p1, i1, descr=floatarraydescr)
- i6 = getarrayitem_raw(p0, i1, descr=floatarraydescr)
+ i4 = getarrayitem_raw_i(p0, i1, descr=chararraydescr)
+ f5 = getarrayitem_raw_f(p1, i1, descr=floatarraydescr)
+ f6 = getarrayitem_raw_f(p0, i1, descr=floatarraydescr)
guard_true(i2) [p0, i0]
jump(p1,p0,i1)
"""
loop = self.parse_loop(ops_src)
ops = loop.operations
- assert isomorphic(ops[1], ops[4])
+ assert isomorphic(ops[0], ops[3])
assert not isomorphic(ops[0], ops[1])
assert not isomorphic(ops[0], ops[5])
- # TODO strong assumptions do hold here?
- #assert not isomorphic(ops[4], ops[5])
- #assert not isomorphic(ops[5], ops[6])
- #assert not isomorphic(ops[4], ops[6])
- #assert not isomorphic(ops[1], ops[6])
def test_packset_extend_simple(self):
ops = """
@@ -723,33 +754,33 @@
i1 = int_add(i0, 1)
i2 = int_le(i1, 16)
guard_true(i2) [p0, i0]
- i3 = getarrayitem_raw(p0, i1, descr=chararraydescr)
+ i3 = getarrayitem_raw_i(p0, i1, descr=chararraydescr)
i4 = int_add(i3, 1)
jump(p0,i1)
"""
loop = self.parse_loop(ops)
- vopt = self.extend_packset(loop,1)
- assert len(vopt.dependency_graph.memory_refs) == 2
+ vopt, graph = self.extend_packset(loop,1)
+ assert len(graph.memory_refs) == 2
self.assert_independent(5,10)
assert len(vopt.packset.packs) == 2
- self.assert_packset_empty(vopt.packset, len(loop.operations),
+ self.assert_packset_empty(vopt.packset,
+ len(loop.operations),
[(5,10), (4,9)])
def test_packset_extend_load_modify_store(self):
ops = """
[p0,i0]
- guard_early_exit() []
i1 = int_add(i0, 1)
i2 = int_le(i1, 16)
guard_true(i2) [p0, i0]
- i3 = getarrayitem_raw(p0, i1, descr=chararraydescr)
+ i3 = getarrayitem_raw_i(p0, i1, descr=chararraydescr)
i4 = int_mul(i3, 2)
setarrayitem_raw(p0, i1, i4, descr=chararraydescr)
jump(p0,i1)
"""
loop = self.parse_loop(ops)
- vopt = self.extend_packset(loop,1)
- assert len(vopt.dependency_graph.memory_refs) == 4
+ vopt, graph = self.extend_packset(loop,1)
+ assert len(graph.memory_refs) == 4
self.assert_independent(4,10)
self.assert_independent(5,11)
self.assert_independent(6,12)
@@ -763,15 +794,18 @@
('int',2, [(0,(2,4)),(1,(6,8))]),
('singlefloat',1,[(0,(2,4,6,8))])])
def test_packset_combine_simple(self,descr,packs,packidx):
+ suffix = '_i'
+ if 'float' in descr:
+ suffix = '_f'
ops = """
[p0,i0]
- i3 = getarrayitem_raw(p0, i0, descr={descr}arraydescr)
+ i3 = getarrayitem_raw{suffix}(p0, i0, descr={descr}arraydescr)
i1 = int_add(i0,1)
jump(p0,i1)
- """.format(descr=descr)
+ """.format(descr=descr,suffix=suffix)
loop = self.parse_loop(ops)
- vopt = self.combine_packset(loop,3)
- assert len(vopt.dependency_graph.memory_refs) == 4
+ vopt, graph = self.combine_packset(loop,3)
+ assert len(graph.memory_refs) == 4
assert len(vopt.packset.packs) == packs
for i,t in packidx:
self.assert_pack(vopt.packset.packs[i], t)
@@ -832,7 +866,6 @@
def test_packset_vector_operation(self, op, descr, stride):
ops = """
[p0,p1,p2,i0]
- guard_early_exit() []
i1 = int_add(i0, {stride})
i10 = int_le(i1, 128)
guard_true(i10) []
@@ -864,7 +897,6 @@
def test_schedule_vector_operation(self, op, descr, stride):
ops = """
[p0,p1,p2,i0] # 0
- guard_early_exit() []
i10 = int_le(i0, 128) # 1, 8, 15, 22
guard_true(i10) [p0,p1,p2,i0] # 2, 9, 16, 23
i2 = getarrayitem_raw(p0, i0, descr={descr}arraydescr) # 3, 10, 17, 24
@@ -882,8 +914,8 @@
i11 = int_le(i1, 128)
guard_true(i11) []
i12 = int_add(i1, {stride})
- v1 = vec_getarrayitem_raw(p0, i0, 2, descr={descr}arraydescr)
- v2 = vec_getarrayitem_raw(p1, i0, 2, descr={descr}arraydescr)
+ v1 = vec_getarrayitem_raw(p0, i0, descr={descr}arraydescr)
+ v2 = vec_getarrayitem_raw(p1, i0, descr={descr}arraydescr)
v3 = {op}(v1,v2)
vec_setarrayitem_raw(p2, i0, v3, descr={descr}arraydescr)
jump(p0,p1,p2,i12)
@@ -895,7 +927,6 @@
def test_vschedule_trace_1(self):
ops = """
[i0, i1, i2, i3, i4]
- guard_early_exit() []
i6 = int_mul(i0, 8)
i7 = raw_load(i2, i6, descr=arraydescr)
i8 = raw_load(i3, i6, descr=arraydescr)
@@ -928,7 +959,6 @@
def test_collapse_index_guard_1(self):
ops = """
[p0,i0]
- guard_early_exit() [p0,i0]
i1 = getarrayitem_raw(p0, i0, descr=chararraydescr)
i2 = int_add(i0, 1)
i3 = int_lt(i2, 102)
@@ -949,7 +979,7 @@
{dead_code}
i500 = int_add(i0, 16)
i501 = int_lt(i2, 102)
- i1 = vec_getarrayitem_raw(p0, i0, 16, descr=chararraydescr)
+ v10[16xi8] = vec_getarrayitem_raw(p0, i0, descr=chararraydescr)
jump(p0,i2)
""".format(dead_code=dead_code)
vopt = self.schedule(self.parse_loop(ops),15,with_guard_opt=True)
@@ -958,7 +988,6 @@
def test_too_small_vector(self):
ops = """
[p0,i0]
- guard_early_exit() [p0,i0]
i1 = getarrayitem_raw(p0, 0, descr=chararraydescr) # constant index
i2 = getarrayitem_raw(p0, 1, descr=chararraydescr) # constant index
i4 = int_add(i1, i2)
@@ -976,7 +1005,6 @@
def test_constant_expansion(self):
ops = """
[p0,i0]
- guard_early_exit() [p0,i0]
i1 = getarrayitem_raw(p0, i0, descr=floatarraydescr)
i4 = int_sub(i1, 42)
i3 = int_add(i0,1)
@@ -987,8 +1015,8 @@
opt="""
[p0,i0]
label(p0,i0)
- v3 = vec_int_expand(42, 2)
- label(p0,i0,v3)
+ v3[2xf64] = vec_expand_f(42.0)
+ label(p0,i0,v3[2xf64])
i20 = int_add(i0, 1)
i30 = int_lt(i20, 10)
i2 = int_add(i0, 2)
@@ -996,9 +1024,9 @@
guard_true(i3) [p0,i0]
i4 = int_add(i0, 2)
i5 = int_lt(i2, 10)
- v1 = vec_getarrayitem_raw(p0, i0, 2, descr=floatarraydescr)
- v2 = vec_int_sub(v1, v3)
- jump(p0,i2,v3)
+ v1[2xf64] = vec_getarrayitem_raw(p0, i0, descr=floatarraydescr)
+ v2[2xf64] = vec_int_sub(v1[2xf64], v3[2xf64])
+ jump(p0,i2,v3[2xf64])
"""
vopt = self.vectorize(self.parse_loop(ops),1)
self.assert_equal(vopt.loop, self.parse_loop(opt,add_label=False))
@@ -1006,7 +1034,6 @@
def test_variable_expansion(self):
ops = """
[p0,i0,f3]
- guard_early_exit() [p0,i0]
f1 = getarrayitem_raw(p0, i0, descr=floatarraydescr)
f4 = int_add(f1, f3)
i3 = int_add(i0,1)
@@ -1017,8 +1044,8 @@
opt="""
[p0,i0,f3]
label(p0,i0,f3)
- v3 = vec_float_expand(f3,2)
- label(p0,i0,f3,v3)
+ v3[2xf64] = vec_expand_f(f3)
+ label(p0,i0,f3,v3[2xf64])
i20 = int_add(i0, 1)
i30 = int_lt(i20, 10)
i2 = int_add(i0, 2)
@@ -1026,9 +1053,9 @@
guard_true(i3) [p0,i0,f3]
i4 = int_add(i0, 2)
i5 = int_lt(i2, 10)
- v1 = vec_getarrayitem_raw(p0, i0, 2, descr=floatarraydescr)
- v2 = vec_int_add(v1, v3)
- jump(p0,i2,f3,v3)
+ v1[2xf64] = vec_getarrayitem_raw(p0, i0, descr=floatarraydescr)
+ v2[2xf64] = vec_int_add(v1[2xf64], v3[2xf64])
+ jump(p0,i2,f3,v3[2xf64])
"""
vopt = self.vectorize(self.parse_loop(ops),1)
self.assert_equal(vopt.loop, self.parse_loop(opt, add_label=False))
@@ -1036,7 +1063,6 @@
def test_accumulate_basic(self):
trace = """
[p0, i0, f0]
- guard_early_exit() [p0, i0, f0]
f1 = raw_load(p0, i0, descr=floatarraydescr)
f2 = float_add(f0, f1)
i1 = int_add(i0, 8)
@@ -1063,7 +1089,6 @@
def test_element_f45_in_guard_failargs(self):
ops = """
[p36, i28, p9, i37, p14, f34, p12, p38, f35, p39, i40, i41, p42, i43,
i44, i21, i4, i0, i18]
- guard_early_exit() [p38, p12, p9, p14, p39, i37, i44, f35, i40, p42,
i43, f34, i28, p36, i41]
f45 = raw_load(i21, i44, descr=floatarraydescr)
guard_not_invalidated() [p38, p12, p9, p14, f45, p39, i37, i44, f35,
i40, p42, i43, None, i28, p36, i41]
i46 = int_add(i44, 8)
@@ -1107,7 +1132,6 @@
def test_shrink_vector_size(self):
ops = """
[p0,p1,i1]
- guard_early_exit() []
f1 = getarrayitem_raw(p0, i1, descr=floatarraydescr)
i2 = cast_float_to_singlefloat(f1)
setarrayitem_raw(p1, i1, i2, descr=singlefloatarraydescr)
@@ -1143,7 +1167,6 @@
def test_castup_arith_castdown(self):
ops = """
[p0,p1,p2,i0,i4]
- guard_early_exit() []
i10 = raw_load(p0, i0, descr=singlefloatarraydescr)
i1 = int_add(i0, 4)
i11 = raw_load(p1, i1, descr=singlefloatarraydescr)
@@ -1196,7 +1219,6 @@
def test_truediv_abs_neg_float(self):
ops = """
[f9,p10,i11,p4,i12,p2,p5,p13,i14,p7,i15,p8,i16,f17,i18,i19]
- guard_early_exit() [p8, p7, p5, p4, p2, f9, i12, i11, p10, i15, i14,
p13]
f20 = raw_load(i16, i12, descr=floatarraydescr)
guard_not_invalidated() [p8, p7, p5, p4, p2, f20, None, i12, i11, p10,
i15, i14, p13]
i23 = int_add(i12, 8)
@@ -1216,7 +1238,6 @@
def test_axis_sum(self):
trace = """
[i1, p10, i11, p8, i12, p3, p4, p13, i14, i15, p6, p9, i16, i17, i18,
i19, i20, i21, i22, i23]
- guard_early_exit() [i1, p9, p8, p6, p4, p3, i11, i15, p13, i12, i14,
p10]
f24 = raw_load(i16, i12, descr=floatarraydescr)
guard_not_invalidated() [i1, p9, p8, p6, p4, p3, f24, i11, i15, p13,
i12, i14, p10]
i26 = int_add(i12, 8)
@@ -1246,7 +1267,6 @@
def test_cast_1(self):
trace = """
[i9, i10, p2, p11, i12, i13, p4, p5, p14, i15, p8, i16, p17, i18, i19,
i20, i21, i22, i23]
- guard_early_exit() [p8, p5, p4, p2, p17, i13, i12, i10, i19, p14, p11,
i18, i15, i16, i9]
i24 = raw_load(i20, i16, descr=singlefloatarraydescr)
guard_not_invalidated() [p8, p5, p4, p2, i24, p17, i13, i12, i10, i19,
p14, p11, i18, i15, i16, None]
i27 = int_add(i16, 4)
@@ -1269,7 +1289,6 @@
def test_all_guard(self):
trace = """
[p0, p3, i4, i5, i6, i7]
- guard_early_exit() [p0, p3, i5, i4]
f8 = raw_load(i6, i5, descr=floatarraydescr)
guard_not_invalidated() [p0, f8, p3, i5, i4]
i9 = cast_float_to_int(f8)
@@ -1287,7 +1306,6 @@
def test_max(self):
trace = """
[p3, i4, p2, i5, f6, i7, i8]
- guard_early_exit() [p2, f6, i4, i5, p3]
f9 = raw_load(i7, i5, descr=floatarraydescr)
guard_not_invalidated() [p2, f9, f6, i4, i5, p3]
i10 = float_ge(f6, f9)
@@ -1307,7 +1325,6 @@
def test_abc(self):
trace="""
[p0, p1, p5, p6, p7, p12, p13, i14, i15, i16, i17, i18, i19, i20]
- guard_early_exit() []
debug_merge_point(0, 0, '<code object <module>. file
'/home/rich/proj/da/thesis/bench/user1.py'. line 2> #117 LOAD_NAME')
guard_not_invalidated(descr=<ResumeGuardNotInvalidated object at
0x7fc657d7be20>) [p1, p0, p5, p6, p7, p12, p13]
debug_merge_point(0, 0, '<code object <module>. file
'/home/rich/proj/da/thesis/bench/user1.py'. line 2> #120 LOAD_CONST')
@@ -1354,7 +1371,6 @@
def test_bug1(self):
trace="""
[p0, p1, p6, p7, p11, i83, f57, f61, f65, f70, f78, f81, i48, i56, p46]
- guard_early_exit(descr=<Guard0x7fa392d5c1a0>) [p1, p0, p6, p7, p11,
f81, f78, f70, f65, f61, f57, i83]
guard_not_invalidated(descr=<Guard0x7fa392d5c200>) [p1, p0, p6, p7,
p11, f81, f78, f70, f65, f61, f57, i83]
i91 = int_lt(i83, i48)
guard_true(i91, descr=<Guard0x7fa392d5c260>) [p1, p0, p6, p7, p11,
i48, f81, f78, f70, f65, f61, f57, i83]
@@ -1389,7 +1405,6 @@
def test_1(self):
trace = """
[p0, p1, p6, p7, i13, p14, p15]
- guard_early_exit(descr=<ResumeAtLoopHeaderDescr object at
0x7f89c54cdbe0>) [p1, p0, p6, p7, i13]
guard_not_invalidated(descr=<ResumeGuardNotInvalidated object at
0x7f89c54cdc40>) [p1, p0, p6, p7, i13]
i17 = int_lt(i13, 10000)
guard_true(i17, descr=<ResumeGuardTrueDescr object at 0x7f89c54cdca0>)
[p1, p0, p6, p7, i13]
diff --git a/rpython/jit/metainterp/optimizeopt/vector.py
b/rpython/jit/metainterp/optimizeopt/vector.py
--- a/rpython/jit/metainterp/optimizeopt/vector.py
+++ b/rpython/jit/metainterp/optimizeopt/vector.py
@@ -176,7 +176,6 @@
if vsize == 0 or byte_count == 0 or loop.label.getopnum() != rop.LABEL:
# stop, there is no chance to vectorize this trace
# we cannot optimize normal traces (if there is no label)
- import pdb; pdb. set_trace()
raise NotAVectorizeableLoop()
# find index guards and move to the earliest position
@@ -188,8 +187,6 @@
# unroll
self.unroll_count = self.get_unroll_count(vsize)
self.unroll_loop_iterations(loop, self.unroll_count)
- loop.operations = self.get_newoperations()
- self.clear_newoperations();
# vectorize
graph = DependencyGraph(loop)
@@ -210,8 +207,6 @@
def unroll_loop_iterations(self, loop, unroll_count):
""" Unroll the loop X times. unroll_count + 1 = unroll_factor """
numops = len(loop.operations)
- label_op = loop.label
- jump_op = loop.jump
# use the target token of the label
#target_token = label_op.getdescr()
#if not we_are_translated():
@@ -223,33 +218,32 @@
# jump_op.setdescr(target_token)
#assert jump_op.is_final()
- self.emit_unrolled_operation(label_op)
+ #self.emit_unrolled_operation(label_op)
+
+ #for i in range(0,numops):
+ # op = loop.operations[i].copy()
+ # if op.is_guard():
+ # assert isinstance(op, GuardResOp)
+ # failargs = renamer.rename_failargs(op, clone=True)
+ # snapshot = renamer.rename_rd_snapshot(op.rd_snapshot,
clone=True)
+ # op.setfailargs(failargs)
+ # op.rd_snapshot = snapshot
+ # operations.append(op)
+ # self.emit_unrolled_operation(op)
renamer = Renamer()
- operations = []
- for i in range(1,numops-1):
- op = loop.operations[i].copy()
- if op.is_guard():
- assert isinstance(op, GuardResOp)
- failargs = renamer.rename_failargs(op, clone=True)
- snapshot = renamer.rename_rd_snapshot(op.rd_snapshot,
clone=True)
- op.setfailargs(failargs)
- op.rd_snapshot = snapshot
- operations.append(op)
- self.emit_unrolled_operation(op)
-
+ operations = loop.operations
+ unrolled = []
prohibit_opnums = (rop.GUARD_FUTURE_CONDITION,
- rop.GUARD_EARLY_EXIT,
rop.GUARD_NOT_INVALIDATED)
-
- orig_jump_args = jump_op.getarglist()[:]
+ orig_jump_args = loop.jump.getarglist()[:]
# it is assumed that #label_args == #jump_args
label_arg_count = len(orig_jump_args)
for u in range(unroll_count):
# fill the map with the renaming boxes. keys are boxes from the
label
for i in range(label_arg_count):
- la = label_op.getarg(i)
- ja = jump_op.getarg(i)
+ la = loop.label.getarg(i)
+ ja = loop.jump.getarg(i)
ja = renamer.rename_box(ja)
if la != ja:
renamer.start_renaming(la, ja)
@@ -284,17 +278,18 @@
renamer.rename_failargs(copied_op, clone=True)
copied_op.setfailargs(renamed_failargs)
#
- self.emit_unrolled_operation(copied_op)
+ unrolled.append(copied_op)
# the jump arguments have been changed
# if label(iX) ... jump(i(X+1)) is called, at the next unrolled loop
# must look like this: label(i(X+1)) ... jump(i(X+2))
- args = jump_op.getarglist()
+ args = loop.jump.getarglist()
for i, arg in enumerate(args):
value = renamer.rename_box(arg)
- jump_op.setarg(i, value)
+ loop.jump.setarg(i, value)
#
- self.emit_unrolled_operation(jump_op)
+ #self.emit_unrolled_operation(jump_op)
+ loop.operations = operations + unrolled
def linear_find_smallest_type(self, loop):
# O(#operations)
@@ -456,14 +451,7 @@
fail = True
check[left] = None
check[right] = None
- accum = pack.accum
- if accum:
- self.packset.accum_vars[accum.var] = accum.pos
-
- print " %dx %s " % (len(pack.operations),
- pack.operations[0].op.getopname())
- if accum:
- print " accumulates!"
+ print " ", pack
if fail:
assert False
@@ -537,9 +525,9 @@
modify_later.append(prev_node)
else:
for path in prev_node.iterate_paths(None, backwards=True,
blacklist=True):
- if not path.is_always_pure(exclude_first=True):
- path.set_schedule_priority(90)
+ if not path.is_always_pure():
valid = False
+ else:
if path.last() in zero_deps:
del zero_deps[path.last()]
if not valid:
@@ -559,34 +547,25 @@
for node in zero_deps.keys():
earlyexit.edge_to(node)
- # TODO self.relax_guard_to(guard_node, ee_guard_node)
+ self.mark_guard(guard_node, loop)
if one_valid:
return graph
return None
- def relax_guard_to(self, guard_node, other_node):
- """ Relaxes a guard operation to an earlier guard. """
- # clone this operation object. if the vectorizer is
- # not able to relax guards, it won't leave behind a modified operation
- tgt_op = guard_node.getoperation().clone()
- guard_node.op = tgt_op
-
- op = other_node.getoperation()
- assert isinstance(tgt_op, GuardResOp)
+ def mark_guard(self, node, loop):
+ """ Marks this guard as an early exit! """
+ op = node.getoperation()
assert isinstance(op, GuardResOp)
- olddescr = op.getdescr()
descr = None
- guard_true_false = tgt_op.getopnum() in (rop.GUARD_TRUE,
rop.GUARD_FALSE)
- if guard_true_false:
+ if op.getopnum() in (rop.GUARD_TRUE, rop.GUARD_FALSE):
descr = CompileLoopVersionDescr()
else:
descr = ResumeAtLoopHeaderDescr()
- if olddescr:
- descr.copy_all_attributes_from(olddescr)
+ if op.getdescr():
+ descr.copy_all_attributes_from(op.getdescr())
#
- tgt_op.setdescr(descr)
- tgt_op.setfailargs(op.getfailargs()[:])
-
+ op.setdescr(descr)
+ op.setfailargs(loop.inputargs)
class CostModel(object):
""" Utility to estimate the savings for the new trace loop.
@@ -687,7 +666,6 @@
if forward and origin_pack.is_accumulating():
# in this case the splitted accumulator must
# be combined. This case is not supported
- import pdb; pdb. set_trace()
raise NotAVectorizeableLoop()
#
if self.contains_pair(lnode, rnode):
diff --git a/rpython/jit/metainterp/optimizeopt/version.py
b/rpython/jit/metainterp/optimizeopt/version.py
--- a/rpython/jit/metainterp/optimizeopt/version.py
+++ b/rpython/jit/metainterp/optimizeopt/version.py
@@ -113,26 +113,26 @@
jump.setdescr(token)
-def index_of_first(opnum, operations, pass_by=0):
- """ returns the position of the first operation matching the opnum.
- Or -1 if non is found
- """
- for i,op in enumerate(operations):
- if op.getopnum() == opnum:
- if pass_by == 0:
- return i
- else:
- pass_by -= 1
- return -1
+#def index_of_first(opnum, operations, pass_by=0):
+# """ returns the position of the first operation matching the opnum.
+# Or -1 if non is found
+# """
+# for i,op in enumerate(operations):
+# if op.getopnum() == opnum:
+# if pass_by == 0:
+# return i
+# else:
+# pass_by -= 1
+# return -1
+#
+#def find_first_index(self, opnum, pass_by=0):
+# """ return the first index of the operation having the same opnum or -1
"""
+# return index_of_first(opnum, self.operations, pass_by)
+#
+#def find_first(self, opnum, pass_by=0):
+# index = self.find_first_index(opnum, pass_by)
+# if index != -1:
+# return self.operations[index]
+# return None
-def find_first_index(self, opnum, pass_by=0):
- """ return the first index of the operation having the same opnum or -1 """
- return index_of_first(opnum, self.operations, pass_by)
-def find_first(self, opnum, pass_by=0):
- index = self.find_first_index(opnum, pass_by)
- if index != -1:
- return self.operations[index]
- return None
-
-
_______________________________________________
pypy-commit mailing list
[email protected]
https://mail.python.org/mailman/listinfo/pypy-commit
