X-Git-Url: https://git.libre-soc.org/?a=blobdiff_plain;f=src%2Fglsl%2Fnir%2Fnir_opt_algebraic.py;h=cdb19241c43634cd5e071a3829363d81f27d4dbc;hb=bc672e261c5f7ff56cd2b8f6b518ebfdc0163bb7;hp=3ea6320d36c329075456a203bdb25cd4a2af742e;hpb=345c2b288a5f49b82d505eb098847b8e53dde91f;p=mesa.git diff --git a/src/glsl/nir/nir_opt_algebraic.py b/src/glsl/nir/nir_opt_algebraic.py index 3ea6320d36c..cdb19241c43 100644 --- a/src/glsl/nir/nir_opt_algebraic.py +++ b/src/glsl/nir/nir_opt_algebraic.py @@ -56,6 +56,10 @@ optimizations = [ (('iabs', ('ineg', a)), ('iabs', a)), (('fadd', a, 0.0), a), (('iadd', a, 0), a), + (('fadd', ('fmul', a, b), ('fmul', a, c)), ('fmul', a, ('fadd', b, c))), + (('iadd', ('imul', a, b), ('imul', a, c)), ('imul', a, ('iadd', b, c))), + (('fadd', ('fneg', a), a), 0.0), + (('iadd', ('ineg', a), a), 0), (('fmul', a, 0.0), 0.0), (('imul', a, 0), 0), (('fmul', a, 1.0), a), @@ -71,6 +75,9 @@ optimizations = [ (('flrp', a, b, 1.0), b), (('flrp', a, a, b), a), (('flrp', 0.0, a, b), ('fmul', a, b)), + (('flrp', a, b, c), ('fadd', ('fmul', c, ('fsub', b, a)), a), 'options->lower_flrp'), + (('fadd', ('fmul', a, ('fadd', 1.0, ('fneg', c))), ('fmul', b, c)), ('flrp', a, b, c), '!options->lower_flrp'), + (('fadd', a, ('fmul', c, ('fadd', b, ('fneg', a)))), ('flrp', a, b, c), '!options->lower_flrp'), (('ffma', a, b, c), ('fadd', ('fmul', a, b), c), 'options->lower_ffma'), (('fadd', ('fmul', a, b), c), ('ffma', a, b, c), '!options->lower_ffma'), # Comparison simplifications @@ -78,10 +85,6 @@ optimizations = [ (('inot', ('fge', a, b)), ('flt', a, b)), (('inot', ('ilt', a, b)), ('ige', a, b)), (('inot', ('ige', a, b)), ('ilt', a, b)), - (('flt', ('fadd', a, b), 0.0), ('flt', a, ('fneg', b))), - (('fge', ('fadd', a, b), 0.0), ('fge', a, ('fneg', b))), - (('feq', ('fadd', a, b), 0.0), ('feq', a, ('fneg', b))), - (('fne', ('fadd', a, b), 0.0), ('fne', a, ('fneg', b))), (('fge', ('fneg', ('fabs', a)), 0.0), ('feq', a, 0.0)), (('bcsel', ('flt', a, b), a, b), ('fmin', a, b)), (('bcsel', ('flt', a, b), b, a), ('fmax', a, b)), @@ -91,6 +94,18 @@ optimizations = [ (('fsat', a), ('fmin', ('fmax', a, 0.0), 1.0), 'options->lower_fsat'), (('fsat', ('fsat', a)), ('fsat', a)), (('fmin', ('fmax', ('fmin', ('fmax', a, 0.0), 1.0), 0.0), 1.0), ('fmin', ('fmax', a, 0.0), 1.0)), + (('ior', ('flt', a, b), ('flt', a, c)), ('flt', a, ('fmax', b, c))), + (('ior', ('fge', a, b), ('fge', a, c)), ('fge', a, ('fmin', b, c))), + (('slt', a, b), ('b2f', ('flt', a, b)), 'options->lower_scmp'), + (('sge', a, b), ('b2f', ('fge', a, b)), 'options->lower_scmp'), + (('seq', a, b), ('b2f', ('feq', a, b)), 'options->lower_scmp'), + (('sne', a, b), ('b2f', ('fne', a, b)), 'options->lower_scmp'), + # Emulating booleans + (('fmul', ('b2f', a), ('b2f', b)), ('b2f', ('iand', a, b))), + (('fsat', ('fadd', ('b2f', a), ('b2f', b))), ('b2f', ('ior', a, b))), + (('iand', 'a@bool', 1.0), ('b2f', a)), + (('flt', ('fneg', ('b2f', a)), 0), a), # Generated by TGSI KILL_IF. + (('flt', ('fsub', 0.0, ('b2f', a)), 0), a), # Generated by TGSI KILL_IF. # Comparison with the same args. Note that these are not done for # the float versions because NaN always returns false on float # inequalities. @@ -118,7 +133,7 @@ optimizations = [ (('ishr', 0, a), 0), (('ishr', a, 0), a), (('ushr', 0, a), 0), - (('ushr', a, 0), 0), + (('ushr', a, 0), a), # Exponential/logarithmic identities (('fexp2', ('flog2', a)), a), # 2^lg2(a) = a (('fexp', ('flog', a)), a), # e^ln(a) = a @@ -130,6 +145,26 @@ optimizations = [ (('fpow', a, 1.0), a), (('fpow', a, 2.0), ('fmul', a, a)), (('fpow', 2.0, a), ('fexp2', a)), + (('fsqrt', ('fexp2', a)), ('fexp2', ('fmul', 0.5, a))), + (('fsqrt', ('fexp', a)), ('fexp', ('fmul', 0.5, a))), + (('frcp', ('fexp2', a)), ('fexp2', ('fneg', a))), + (('frcp', ('fexp', a)), ('fexp', ('fneg', a))), + (('frsq', ('fexp2', a)), ('fexp2', ('fmul', -0.5, a))), + (('frsq', ('fexp', a)), ('fexp', ('fmul', -0.5, a))), + (('flog2', ('fsqrt', a)), ('fmul', 0.5, ('flog2', a))), + (('flog', ('fsqrt', a)), ('fmul', 0.5, ('flog', a))), + (('flog2', ('frcp', a)), ('fneg', ('flog2', a))), + (('flog', ('frcp', a)), ('fneg', ('flog', a))), + (('flog2', ('frsq', a)), ('fmul', -0.5, ('flog2', a))), + (('flog', ('frsq', a)), ('fmul', -0.5, ('flog', a))), + (('flog2', ('fpow', a, b)), ('fmul', b, ('flog2', a))), + (('flog', ('fpow', a, b)), ('fmul', b, ('flog', a))), + (('fadd', ('flog2', a), ('flog2', b)), ('flog2', ('fmul', a, b))), + (('fadd', ('flog', a), ('flog', b)), ('flog', ('fmul', a, b))), + (('fadd', ('flog2', a), ('fneg', ('flog2', b))), ('flog2', ('fdiv', a, b))), + (('fadd', ('flog', a), ('fneg', ('flog', b))), ('flog', ('fdiv', a, b))), + (('fmul', ('fexp2', a), ('fexp2', b)), ('fexp2', ('fadd', a, b))), + (('fmul', ('fexp', a), ('fexp', b)), ('fexp', ('fadd', a, b))), # Division and reciprocal (('fdiv', 1.0, a), ('frcp', a)), (('frcp', ('frcp', a)), a), @@ -150,18 +185,21 @@ optimizations = [ (('bcsel', a, b, b), b), (('fcsel', a, b, b), b), + # Conversions + (('f2i', ('ftrunc', a)), ('f2i', a)), + (('f2u', ('ftrunc', a)), ('f2u', a)), + # Subtracts - (('fsub', 0.0, ('fsub', 0.0, a)), a), - (('isub', 0, ('isub', 0, a)), a), + (('fsub', a, ('fsub', 0.0, b)), ('fadd', a, b)), + (('isub', a, ('isub', 0, b)), ('iadd', a, b)), + (('fsub', a, b), ('fadd', a, ('fneg', b)), 'options->lower_sub'), + (('isub', a, b), ('iadd', a, ('ineg', b)), 'options->lower_sub'), (('fneg', a), ('fsub', 0.0, a), 'options->lower_negate'), (('ineg', a), ('isub', 0, a), 'options->lower_negate'), (('fadd', a, ('fsub', 0.0, b)), ('fsub', a, b)), (('iadd', a, ('isub', 0, b)), ('isub', a, b)), (('fabs', ('fsub', 0.0, a)), ('fabs', a)), (('iabs', ('isub', 0, a)), ('iabs', a)), - -# This one may not be exact - (('feq', ('fadd', a, b), 0.0), ('feq', a, ('fneg', b))), ] # Add optimizations to handle the case where the result of a ternary is @@ -185,4 +223,17 @@ for op in ['flt', 'fge', 'feq', 'fne', ('bcsel', 'a', (op, 'd', 'b'), (op, 'd', 'c'))), ] +# This section contains "late" optimizations that should be run after the +# regular optimizations have finished. Optimizations should go here if +# they help code generation but do not necessarily produce code that is +# more easily optimizable. +late_optimizations = [ + (('flt', ('fadd', a, b), 0.0), ('flt', a, ('fneg', b))), + (('fge', ('fadd', a, b), 0.0), ('fge', a, ('fneg', b))), + (('feq', ('fadd', a, b), 0.0), ('feq', a, ('fneg', b))), + (('fne', ('fadd', a, b), 0.0), ('fne', a, ('fneg', b))), +] + print nir_algebraic.AlgebraicPass("nir_opt_algebraic", optimizations).render() +print nir_algebraic.AlgebraicPass("nir_opt_algebraic_late", + late_optimizations).render()