From d346b8588c36949695f2b01ca76619e84754dd50 Mon Sep 17 00:00:00 2001
From: Karol Herbst <karolherbst@gmail.com>
Date: Sun, 26 Mar 2017 21:45:58 +0200
Subject: [PATCH] nv50/ir: implement mad post ra folding for nvc0+

changes for GpuTest /test=pixmark_piano /benchmark /no_scorebox /msaa=0
/benchmark_duration_ms=60000 /width=1024 /height=640:

score: 1026 -> 1045

changes for shader-db:
total instructions in shared programs : 3943335 -> 3934925 (-0.21%)
total gprs used in shared programs    : 481563 -> 481563 (0.00%)
total local used in shared programs   : 27469 -> 27469 (0.00%)
total bytes used in shared programs   : 36139384 -> 36061888 (-0.21%)

                local        gpr       inst      bytes
    helped           0           0        3587        3587
      hurt           0           0           0           0

v2: removed TODO
    reorderd to show changes without RA modification
    removed stale debugging print() call
v3: remove predicate checks
    enable only for gf100 ISA

Signed-off-by: Karol Herbst <karolherbst@gmail.com>
Reviewed-by: Ilia Mirkin <imirkin@alum.mit.edu>
---
 .../drivers/nouveau/codegen/nv50_ir_peephole.cpp   | 51 ++++++++++++++++++++--
 1 file changed, 47 insertions(+), 4 deletions(-)

diff --git a/src/gallium/drivers/nouveau/codegen/nv50_ir_peephole.cpp b/src/gallium/drivers/nouveau/codegen/nv50_ir_peephole.cpp
index 6dc4e21..f7293b2 100644
--- a/src/gallium/drivers/nouveau/codegen/nv50_ir_peephole.cpp
+++ b/src/gallium/drivers/nouveau/codegen/nv50_ir_peephole.cpp
@@ -3195,7 +3195,8 @@ class PostRaLoadPropagation : public Pass
 private:
    virtual bool visit(Instruction *);
 
-   void handleMAD(Instruction *);
+   void handleMADforNV50(Instruction *);
+   void handleMADforNVC0(Instruction *);
 };
 
 static bool
@@ -3210,7 +3211,7 @@ post_ra_dead(Instruction *i)
 // Fold Immediate into MAD; must be done after register allocation due to
 // constraint SDST == SSRC2
 void
-PostRaLoadPropagation::handleMAD(Instruction *i)
+PostRaLoadPropagation::handleMADforNV50(Instruction *i)
 {
    if (i->def(0).getFile() != FILE_GPR ||
        i->src(0).getFile() != FILE_GPR ||
@@ -3263,12 +3264,54 @@ PostRaLoadPropagation::handleMAD(Instruction *i)
    }
 }
 
+void
+PostRaLoadPropagation::handleMADforNVC0(Instruction *i)
+{
+   if (i->def(0).getFile() != FILE_GPR ||
+       i->src(0).getFile() != FILE_GPR ||
+       i->src(1).getFile() != FILE_GPR ||
+       i->src(2).getFile() != FILE_GPR ||
+       i->getDef(0)->reg.data.id != i->getSrc(2)->reg.data.id)
+      return;
+
+   // TODO: gm107 can also do this for S32, maybe other chipsets as well
+   if (i->dType != TYPE_F32)
+      return;
+
+   if ((i->src(2).mod | Modifier(NV50_IR_MOD_NEG)) != Modifier(NV50_IR_MOD_NEG))
+      return;
+
+   ImmediateValue val;
+   int s;
+
+   if (i->src(0).getImmediate(val))
+      s = 1;
+   else if (i->src(1).getImmediate(val))
+      s = 0;
+   else
+      return;
+
+   if ((i->src(s).mod | Modifier(NV50_IR_MOD_NEG)) != Modifier(NV50_IR_MOD_NEG))
+      return;
+
+   if (s == 1)
+      i->swapSources(0, 1);
+
+   Instruction *imm = i->getSrc(1)->getInsn();
+   i->setSrc(1, imm->getSrc(0));
+   if (post_ra_dead(imm))
+      delete_Instruction(prog, imm);
+}
+
 bool
 PostRaLoadPropagation::visit(Instruction *i)
 {
    switch (i->op) {
    case OP_MAD:
-      handleMAD(i);
+      if (prog->getTarget()->getChipset() < 0xc0)
+         handleMADforNV50(i);
+      else
+         handleMADforNVC0(i);
       break;
    default:
       break;
@@ -3698,7 +3741,7 @@ bool
 Program::optimizePostRA(int level)
 {
    RUN_PASS(2, FlatteningPass, run);
-   if (getTarget()->getChipset() < 0xc0)
+   if (getTarget()->getChipset() < NVISA_GK20A_CHIPSET)
       RUN_PASS(2, PostRaLoadPropagation, run);
 
    return true;
-- 
2.7.4