i965: Add an INTEL_PRECISE_TRIG=1 option to fix SIN/COS output range.

The SIN and COS instructions on Intel hardware can produce values
slightly outside of the [-1.0, 1.0] range for a small set of values.
Obviously, this can break everyone's expectations about trig functions.

According to an internal presentation, the COS instruction can produce
a value up to 1.000027 for inputs in the range (0.08296, 0.09888).  One
suggested workaround is to multiply by 0.99997, scaling down the
amplitude slightly.  Apparently this also minimizes the error function,
reducing the maximum error from 0.00006 to about 0.00003.

When enabled, fixes 16 dEQP precision tests

   dEQP-GLES31.functional.shaders.builtin_functions.precision.
   {cos,sin}.{highp,mediump}_compute.{scalar,vec2,vec4,vec4}.

at the cost of making every sin and cos call more expensive (about
twice the number of cycles on recent hardware).  Enabling this
option has been shown to reduce GPUTest Volplosion performance by
about 10%.

Signed-off-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Matt Turner <mattst88@gmail.com>
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>

1 files changed, 14 insertions, 2 deletions

diff --git a/src/mesa/drivers/dri/i965/brw_vec4_nir.cpp b/src/mesa/drivers/dri/i965/brw_vec4_nir.cpp
index ee6929b..6c8fd06 100644
--- a/src/mesa/drivers/dri/i965/brw_vec4_nir.cpp
+++ b/src/mesa/drivers/dri/i965/brw_vec4_nir.cpp
@@ -1101,12 +1101,24 @@ vec4_visitor::nir_emit_alu(nir_alu_instr *instr)
       break;
 
    case nir_op_fsin:
-      inst = emit_math(SHADER_OPCODE_SIN, dst, op[0]);
+      if (!compiler->precise_trig) {
+         inst = emit_math(SHADER_OPCODE_SIN, dst, op[0]);
+      } else {
+         src_reg tmp = src_reg(this, glsl_type::vec4_type);
+         inst = emit_math(SHADER_OPCODE_SIN, dst_reg(tmp), op[0]);
+         inst = emit(MUL(dst, tmp, brw_imm_f(0.99997)));
+      }
       inst->saturate = instr->dest.saturate;
       break;
 
    case nir_op_fcos:
-      inst = emit_math(SHADER_OPCODE_COS, dst, op[0]);
+      if (!compiler->precise_trig) {
+         inst = emit_math(SHADER_OPCODE_COS, dst, op[0]);
+      } else {
+         src_reg tmp = src_reg(this, glsl_type::vec4_type);
+         inst = emit_math(SHADER_OPCODE_COS, dst_reg(tmp), op[0]);
+         inst = emit(MUL(dst, tmp, brw_imm_f(0.99997)));
+      }
       inst->saturate = instr->dest.saturate;
       break;