[Mesa-dev] [PATCH 111/133] nir: Add infastructure for generating algebraic transformation passes
Dylan Baker
baker.dylan.c at gmail.com
Tue Jan 6 17:17:37 PST 2015
On Tuesday, January 06, 2015 04:49:04 PM Jason Ekstrand wrote:
> On Tue, Jan 6, 2015 at 4:35 PM, Dylan Baker <baker.dylan.c at gmail.com> wrote:
>
> > I have a couple of comments, but they're just suggestions.
> > On Monday, December 15, 2014 10:12:54 PM Jason Ekstrand wrote:
> > > This commit builds on the nir_search.h infastructure by adds a bit of
> > > python code that makes it stupid easy to write an algebraic
> > transformation
> > > pass. The nir_algebraic.py file contains four python classes that
> > > correspond directly to the datastructures in nir_search.c and allow you
> > to
> > > easily generate the C code to represent them. Given a list of
> > > search-and-replace operations, it can then generate a function that
> > applies
> > > those transformations to a shader.
> > >
> > > The transformations can be specified manually, or they can be specified
> > > using nested tuples. The nested tuples make a neat little language for
> > > specifying expression trees and search-and-replace operations in a very
> > > readable and easy-to-edit fasion.
> > >
> > > The generated code is also fairly efficient. Insteady of blindly calling
> > > nir_replace_instr with every single transformation and on every single
> > > instruction, it uses a switch statement on the instruction opcode to do a
> > > first-order culling and only calls nir_replace_instr if the opcode is
> > known
> > > to match the first opcode in the search expression.
> > > ---
> > > src/glsl/nir/nir_algebraic.py | 249
> > ++++++++++++++++++++++++++++++++++++++++++
> > > 1 file changed, 249 insertions(+)
> > > create mode 100644 src/glsl/nir/nir_algebraic.py
> > >
> > > diff --git a/src/glsl/nir/nir_algebraic.py
> > b/src/glsl/nir/nir_algebraic.py
> > > new file mode 100644
> > > index 0000000..7718a6d
> > > --- /dev/null
> > > +++ b/src/glsl/nir/nir_algebraic.py
> > > @@ -0,0 +1,249 @@
> > > +#! /usr/bin/env python
> > > +#
> > > +# Copyright (C) 2014 Intel Corporation
> > > +#
> > > +# Permission is hereby granted, free of charge, to any person obtaining
> > a
> > > +# copy of this software and associated documentation files (the
> > "Software"),
> > > +# to deal in the Software without restriction, including without
> > limitation
> > > +# the rights to use, copy, modify, merge, publish, distribute,
> > sublicense,
> > > +# and/or sell copies of the Software, and to permit persons to whom the
> > > +# Software is furnished to do so, subject to the following conditions:
> > > +#
> > > +# The above copyright notice and this permission notice (including the
> > next
> > > +# paragraph) shall be included in all copies or substantial portions of
> > the
> > > +# Software.
> > > +#
> > > +# THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
> > EXPRESS OR
> > > +# IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
> > MERCHANTABILITY,
> > > +# FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT
> > SHALL
> > > +# THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR
> > OTHER
> > > +# LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
> > ARISING
> > > +# FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
> > DEALINGS
> > > +# IN THE SOFTWARE.
> > > +#
> > > +# Authors:
> > > +# Jason Ekstrand (jason at jlekstrand.net)
> > > +
> > > +import itertools, struct, sys, mako.template
> >
> > It's idiomatic to have one import per line
> >
>
> Sure. I don't care.
>
>
> >
> > > +
> > > +# Represents a set of variables, each with a unique id
> > > +class VarSet(object):
> > > + def __init__(self):
> > > + self.names = {}
> > > + self.ids = itertools.count()
> > > +
> > > + def __getitem__(self, name):
> > > + if name not in self.names:
> > > + self.names[name] = self.ids.next()
> > > +
> > > + return self.names[name]
> > > +
> > > +class Value(object):
> > > + @staticmethod
> > > + def create(val, name_base, varset):
> > > + if isinstance(val, tuple):
> > > + return Expression(val, name_base, varset)
> > > + elif isinstance(val, Expression):
> > > + return val
> > > + elif isinstance(val, str):
> > (str, unicode)
> >
>
> Shouldn't be a problem. Is that going to help with python3?
No, unfortunately in python2 you have str and unicode which are
subclasses of basestring, in python3 you have str and bytes which do not
descend from a command base class.
It's just a suggestion, since it's a completeness issue
>
>
> > > + return Variable(val, name_base, varset)
> > > + elif isinstance(val, (bool, int, long, float)):
> > > + return Constant(val, name_base)
> > > +
> > > + __template = mako.template.Template("""
> > > +static const ${val.c_type()} ${val.name} = {
> > > + { ${val.type_enum()} },
> > > +% if isinstance(val, Constant):
> > > + { ${hex(val)} /* ${val.value} */ },
> > > +% elif isinstance(val, Variable):
> > > + ${val.index}, /* ${val.var_name} */
> > > +% elif isinstance(val, Expression):
> > > + nir_op_${val.opcode},
> > > + { ${', '.join([src.c_ptr() for src in val.sources])} },
> > > +% endif
> > > +};""")
> > > +
> > > + def __init__(self, name, type_str):
> > > + self.name = name
> > > + self.type_str = type_str
> > > +
> > > + def type_enum(self):
> > > + return "nir_search_value_" + self.type_str
> >
> > You could use the property decorator on these (the one above and two
> > below) to access them as attributes rather than as functions
> >
>
> Why not.
>
>
> >
> > > +
> > > + def c_type(self):
> > > + return "nir_search_" + self.type_str
> > > +
> > > + def c_ptr(self):
> > > + return "&{0}.value".format(self.name)
> > > +
> > > + def render(self):
> > > + return self.__template.render(val=self,
> > > + Constant=Constant,
> > > + Variable=Variable,
> > > + Expression=Expression)
> > > +
> > > +class Constant(Value):
> > > + def __init__(self, val, name):
> > > + Value.__init__(self, name, "constant")
> > > + self.value = val
> > > +
> > > + def __hex__(self):
> > > + # Even if it's an integer, we still need to unpack as an unsigned
> > > + # int. This is because, without C99, we can only assign to the
> > first
> > > + # element of a union in an initializer.
> > > + if isinstance(self.value, (bool)):
> > > + return 'NIR_TRUE' if self.value else 'NIR_FALSE'
> > > + if isinstance(self.value, (int, long)):
> > > + return hex(struct.unpack('I', struct.pack('i', self.value))[0])
> > > + elif isinstance(self.value, float):
> > > + return hex(struct.unpack('I', struct.pack('f', self.value))[0])
> > > + else:
> > > + assert False
> > > +
> > > +class Variable(Value):
> > > + def __init__(self, val, name, varset):
> > > + Value.__init__(self, name, "variable")
> > > + self.var_name = val
> > > + self.index = varset[val]
> > > + self.name = name
> > > +
> > > +class Expression(Value):
> > > + def __init__(self, expr, name_base, varset):
> > > + Value.__init__(self, name_base, "expression")
> > > + assert isinstance(expr, tuple)
> > > +
> > > + self.opcode = expr[0]
> > > + self.sources = [ Value.create(src, "{0}_{1}".format(name_base,
> > i), varset)
> > > + for (i, src) in enumerate(expr[1:]) ]
> > > +
> > > + def render(self):
> > > + srcs = "\n".join([src.render() for src in self.sources])
> > > + return srcs + super(Expression, self).render()
> > > +
> > > +_optimization_ids = itertools.count()
> > > +
> > > +class SearchAndReplace(object):
> > > + def __init__(self, search, replace):
> > > + self.id = _optimization_ids.next()
> > > +
> > > + varset = VarSet()
> > > + if isinstance(search, Expression):
> > > + self.search = search
> > > + else:
> > > + self.search = Expression(search, "search{0}".format(self.id),
> > varset)
> > > +
> > > + if isinstance(replace, Value):
> > > + self.replace = replace
> > > + else:
> > > + self.replace = Value.create(replace, "replace{0}".format(
> > self.id), varset)
> > > +
> > > +_algebraic_pass_template = mako.template.Template("""
> > > +#include "nir.h"
> > > +#include "nir_search.h"
> > > +
> > > +struct transform {
> > > + const nir_search_expression *search;
> > > + const nir_search_value *replace;
> > > +};
> > > +
> > > +% for (opcode, xform_list) in xform_dict.iteritems():
> > > +% for xform in xform_list:
> >
> > mako doesn't care about spaces around it's control flow, you can indent
> > either before or after the % to make the code more readable
> >
> > > + ${xform.search.render()}
> > > + ${xform.replace.render()}
> > > +% endfor
> > > +
> > > +static const struct {
> > > + const nir_search_expression *search;
> > > + const nir_search_value *replace;
> > > +} ${pass_name}_${opcode}_xforms[] = {
> > > +% for xform in xform_list:
> > > + { &${xform.search.name}, ${xform.replace.c_ptr()} },
> > > +% endfor
> > > +};
> > > +% endfor
> > > +
> > > +struct opt_state {
> > > + void *mem_ctx;
> > > + bool progress;
> > > +};
> > > +
> > > +static bool
> > > +${pass_name}_block(nir_block *block, void *void_state)
> > > +{
> > > + struct opt_state *state = void_state;
> > > +
> > > + nir_foreach_instr_safe(block, instr) {
> > > + if (instr->type != nir_instr_type_alu)
> > > + continue;
> > > +
> > > + nir_alu_instr *alu = nir_instr_as_alu(instr);
> > > + if (!alu->dest.dest.is_ssa)
> > > + continue;
> > > +
> > > + switch (alu->op) {
> > > + % for opcode in xform_dict.keys():
> >
> > You could also dedent this by a space or two for readability
> >
> > > + case nir_op_${opcode}:
> > > + for (unsigned i = 0; i <
> > ARRAY_SIZE(${pass_name}_${opcode}_xforms); i++) {
> > > + if (nir_replace_instr(alu,
> > ${pass_name}_${opcode}_xforms[i].search,
> > > +
> > ${pass_name}_${opcode}_xforms[i].replace,
> > > + state->mem_ctx))
> > > + state->progress = true;
> > > + }
> > > + break;
> > > + % endfor
> > > + default:
> > > + break;
> > > + }
> > > + }
> > > +
> > > + return true;
> > > +}
> > > +
> > > +static bool
> > > +${pass_name}_impl(nir_function_impl *impl)
> > > +{
> > > + struct opt_state state;
> > > +
> > > + state.mem_ctx = ralloc_parent(impl);
> > > + state.progress = false;
> > > +
> > > + nir_foreach_block(impl, ${pass_name}_block, &state);
> > > +
> > > + if (state.progress)
> > > + nir_metadata_dirty(impl, nir_metadata_block_index |
> > > + nir_metadata_dominance);
> > > +
> > > + return state.progress;
> > > +}
> > > +
> > > +bool
> > > +${pass_name}(nir_shader *shader)
> > > +{
> > > + bool progress = false;
> > > +
> > > + nir_foreach_overload(shader, overload) {
> > > + if (overload->impl)
> > > + progress |= ${pass_name}_impl(overload->impl);
> > > + }
> > > +
> > > + return progress;
> > > +}
> > > +""")
> > > +
> > > +class AlgebraicPass(object):
> > > + def __init__(self, pass_name, transforms):
> > > + self.xform_dict = {}
> > > + self.pass_name = pass_name
> > > +
> > > + for xform in transforms:
> > > + if not isinstance(xform, SearchAndReplace):
> > > + xform = SearchAndReplace(*xform)
> > > +
> > > + if xform.search.opcode not in self.xform_dict:
> > > + self.xform_dict[xform.search.opcode] = []
> >
> > Rather than doing this consider using collections.defaultdict,
> >
> > > +
> > > + self.xform_dict[xform.search.opcode].append(xform)
> > > +
> > > + def render(self):
> > > + return _algebraic_pass_template.render(pass_name=self.pass_name,
> > > + xform_dict=self.xform_dict)
> > > --
> > > 2.2.0
> > >
> > > _______________________________________________
> > > mesa-dev mailing list
> > > mesa-dev at lists.freedesktop.org
> > > http://lists.freedesktop.org/mailman/listinfo/mesa-dev
> > >
> >
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 473 bytes
Desc: This is a digitally signed message part.
URL: <http://lists.freedesktop.org/archives/mesa-dev/attachments/20150106/db720d27/attachment-0001.sig>
More information about the mesa-dev
mailing list