[Mesa-dev] [PATCH 3/3] nv50/ir: convert slct with boolean result to set

[Karol Herbst]

From: Karol Herbst <karolherbst at gmail.com>

helps mainly feral ported games

changes in shader-db:
total instructions in shared programs : 5730139 -> 5726007 (-0.07%)
total gprs used in shared programs    : 663206 -> 663147 (-0.01%)
total shared used in shared programs  : 548832 -> 548832 (0.00%)
total local used in shared programs   : 20956 -> 20956 (0.00%)
total bytes used in shared programs   : 61212472 -> 61168408 (-0.07%)

                local     shared        gpr       inst      bytes
    helped           0           0          61        1087        1087
      hurt           0           0           4           0           0

Signed-off-by: Karol Herbst <kherbst at redhat.com>
---
 .../nouveau/codegen/nv50_ir_peephole.cpp      | 50 +++++++++++++++++++
 1 file changed, 50 insertions(+)

diff --git a/src/gallium/drivers/nouveau/codegen/nv50_ir_peephole.cpp b/src/gallium/drivers/nouveau/codegen/nv50_ir_peephole.cpp
index 3a4d7e27ef7..3c1ee7f92f5 100644
--- a/src/gallium/drivers/nouveau/codegen/nv50_ir_peephole.cpp
+++ b/src/gallium/drivers/nouveau/codegen/nv50_ir_peephole.cpp
@@ -345,6 +345,8 @@ private:
    void expr(Instruction *, ImmediateValue&, ImmediateValue&);
    void expr(Instruction *, ImmediateValue&, ImmediateValue&, ImmediateValue&);
    void opnd(Instruction *, ImmediateValue&, int s);
+   // 3 srcs where 1st and 2nd are immediates
+   void opnd(Instruction *, ImmediateValue&, ImmediateValue&);
    void opnd3(Instruction *, ImmediateValue&);
 
    void unary(Instruction *, const ImmediateValue&);
@@ -400,6 +402,10 @@ ConstantFolding::visit(BasicBlock *bb)
          opnd(i, src1, 1);
       if (i->srcExists(2) && i->src(2).getImmediate(src2))
          opnd3(i, src2);
+      else
+      if (i->srcExists(2) &&
+          i->src(0).getImmediate(src0) && i->src(1).getImmediate(src1))
+         opnd(i, src0, src1);
    }
    return true;
 }
@@ -902,6 +908,50 @@ ConstantFolding::tryCollapseChainedMULs(Instruction *mul2,
    }
 }
 
+void
+ConstantFolding::opnd(Instruction *i, ImmediateValue &imm0, ImmediateValue &imm1)
+{
+   const Storage &a = imm0.reg;
+   const Storage &b = imm1.reg;
+
+   switch (i->op) {
+   case OP_SLCT: {
+      CmpInstruction *slct = i->asCmp();
+      if (a.data.u32 == 0xffffffff && b.data.u32 == 0x0) {
+         slct->setSrc(0, slct->getSrc(2));
+         slct->setSrc(2, NULL);
+         slct->dType = TYPE_U32;
+         slct->op = OP_SET;
+      }
+      else if (a.data.u32 == 0x3f800000 && b.data.u32 == 0x0) {
+         slct->setSrc(0, slct->getSrc(2));
+         slct->setSrc(2, NULL);
+         slct->dType = TYPE_F32;
+         slct->op = OP_SET;
+      }
+      else if (a.data.u32 == 0x0 && b.data.u32 == 0xffffffff) {
+         slct->swapSources(0, 1);
+         slct->setSrc(0, slct->getSrc(2));
+         slct->setSrc(2, NULL);
+         slct->dType = TYPE_U32;
+         slct->setCondition(inverseCondCode(slct->getCondition()));
+         slct->op = OP_SET;
+      }
+      else if (a.data.u32 == 0x0 && b.data.u32 == 0x3f800000) {
+         slct->swapSources(0, 1);
+         slct->setSrc(0, slct->getSrc(2));
+         slct->setSrc(2, NULL);
+         slct->dType = TYPE_F32;
+         slct->setCondition(inverseCondCode(slct->getCondition()));
+         slct->op = OP_SET;
+      }
+      break;
+   }
+   default:
+      break;
+   }
+}
+
 void
 ConstantFolding::opnd3(Instruction *i, ImmediateValue &imm2)
 {
-- 
2.17.1