nir: Add lrp lowering for doubles in opt_algebraic

Some hardware (i965 on Broadwell generation, for example) does not support
natively the execution of lrp instruction with double arguments.

Add 'lower_flrp64' flag to lower this instruction in that case.

v2:
   - Rename lower_flrp_double to lower_flrp64 (Jason)
   - Fix typo (Jason)
   - Adapt the code to define bit_size information in the opcodes.

Signed-off-by: Samuel Iglesias Gonsálvez <siglesias@igalia.com>
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>

diff --git a/src/compiler/nir/nir.h b/src/compiler/nir/nir.h
index d5eda02..e889964 100644 (file)
--- a/src/compiler/nir/nir.h
+++ b/src/compiler/nir/nir.h
@@ -1625,6 +1625,8 @@ typedef struct nir_shader_compiler_options {
    bool lower_fdiv;
    bool lower_ffma;
    bool lower_flrp32;
+   /** Lowers flrp when it does not support doubles */
+   bool lower_flrp64;
    bool lower_fpow;
    bool lower_fsat;
    bool lower_fsqrt;

diff --git a/src/compiler/nir/nir_opt_algebraic.py b/src/compiler/nir/nir_opt_algebraic.py
index 082bcc1..60ee170 100644 (file)
--- a/src/compiler/nir/nir_opt_algebraic.py
+++ b/src/compiler/nir/nir_opt_algebraic.py
@@ -98,12 +98,15 @@ optimizations = [
    (('~flrp', a, a, b), a),
    (('~flrp', 0.0, a, b), ('fmul', a, b)),
    (('~flrp', a, b, ('b2f', c)), ('bcsel', c, b, a), 'options->lower_flrp32'),
-   (('flrp', a, b, c), ('fadd', ('fmul', c, ('fsub', b, a)), a), 'options->lower_flrp32'),
+   (('flrp@32', a, b, c), ('fadd', ('fmul', c, ('fsub', b, a)), a), 'options->lower_flrp32'),
+   (('flrp@64', a, b, c), ('fadd', ('fmul', c, ('fsub', b, a)), a), 'options->lower_flrp64'),
    (('ffract', a), ('fsub', a, ('ffloor', a)), 'options->lower_ffract'),
    (('~fadd', ('fmul', a, ('fadd', 1.0, ('fneg', ('b2f', c)))), ('fmul', b, ('b2f', c))), ('bcsel', c, b, a), 'options->lower_flrp32'),
-   (('~fadd', ('fmul', a, ('fadd', 1.0, ('fneg',         c ))), ('fmul', b,         c )), ('flrp', a, b, c), '!options->lower_flrp32'),
+   (('~fadd@32', ('fmul', a, ('fadd', 1.0, ('fneg',         c ))), ('fmul', b,         c )), ('flrp', a, b, c), '!options->lower_flrp32'),
+   (('~fadd@64', ('fmul', a, ('fadd', 1.0, ('fneg',         c ))), ('fmul', b,         c )), ('flrp', a, b, c), '!options->lower_flrp64'),
    (('~fadd', a, ('fmul', ('b2f', c), ('fadd', b, ('fneg', a)))), ('bcsel', c, b, a), 'options->lower_flrp32'),
-   (('~fadd', a, ('fmul',         c , ('fadd', b, ('fneg', a)))), ('flrp', a, b, c), '!options->lower_flrp32'),
+   (('~fadd@32', a, ('fmul',         c , ('fadd', b, ('fneg', a)))), ('flrp', a, b, c), '!options->lower_flrp32'),
+   (('~fadd@64', a, ('fmul',         c , ('fadd', b, ('fneg', a)))), ('flrp', a, b, c), '!options->lower_flrp64'),
    (('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