nouveau: codegen: Use FILE_MEMORY_BUFFER for buffers

Some of the lowering steps we currently do for FILE_MEMORY_GLOBAL only
apply to buffers, making it impossible to use FILE_MEMORY_GLOBAL for
OpenCL global buffers.

This commits changes the buffer code to use FILE_MEMORY_BUFFER at the
ir_from_tgsi and lowering steps, freeing use of FILE_MEMORY_GLOBAL
for use with OpenCL global buffers.

Note that after lowering buffer accesses use the FILE_MEMORY_GLOBAL
register file.

Tested with piglet on a gf119 and a gk107:
./piglit run -o shader -t '.*arb_shader_storage_buffer_object.*' results/shader
[9/9] pass: 9 /
./piglit run -o shader -t '.*arb_compute_shader.*' results/shader
[20/20] skip: 4, pass: 16 |

Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>

src/gallium/drivers/nouveau/codegen/nv50_ir.h

@@ -332,6 +332,7 @@ enum DataFile
    FILE_MEMORY_CONST,
    FILE_SHADER_INPUT,
    FILE_SHADER_OUTPUT,
+   FILE_MEMORY_BUFFER,
    FILE_MEMORY_GLOBAL,
    FILE_MEMORY_SHARED,
    FILE_MEMORY_LOCAL,

src/gallium/drivers/nouveau/codegen/nv50_ir_from_tgsi.cpp

@@ -373,7 +373,7 @@ static nv50_ir::DataFile translateFile(uint file)
    case TGSI_FILE_PREDICATE:       return nv50_ir::FILE_PREDICATE;
    case TGSI_FILE_IMMEDIATE:       return nv50_ir::FILE_IMMEDIATE;
    case TGSI_FILE_SYSTEM_VALUE:    return nv50_ir::FILE_SYSTEM_VALUE;
-   case TGSI_FILE_BUFFER:          return nv50_ir::FILE_MEMORY_GLOBAL;
+   case TGSI_FILE_BUFFER:          return nv50_ir::FILE_MEMORY_BUFFER;
    case TGSI_FILE_MEMORY:          return nv50_ir::FILE_MEMORY_GLOBAL;
    case TGSI_FILE_SAMPLER:
    case TGSI_FILE_NULL:

src/gallium/drivers/nouveau/codegen/nv50_ir_lowering_nvc0.cpp

@@ -1296,13 +1296,14 @@ NVC0LoweringPass::handleATOM(Instruction *atom)
          handleSharedATOMNVE4(atom);
       return true;
    default:
-      assert(atom->src(0).getFile() == FILE_MEMORY_GLOBAL);
+      assert(atom->src(0).getFile() == FILE_MEMORY_BUFFER);
       base = loadBufInfo64(ind, atom->getSrc(0)->reg.fileIndex * 16);
       assert(base->reg.size == 8);
       if (ptr)
          base = bld.mkOp2v(OP_ADD, TYPE_U64, base, base, ptr);
       assert(base->reg.size == 8);
       atom->setIndirect(0, 0, base);
+      atom->getSrc(0)->reg.file = FILE_MEMORY_GLOBAL;
       return true;
    }
    base =
@@ -1889,7 +1890,7 @@ NVC0LoweringPass::handleLDST(Instruction *i)
    } else if (i->src(0).getFile() == FILE_SHADER_OUTPUT) {
       assert(prog->getType() == Program::TYPE_TESSELLATION_CONTROL);
       i->op = OP_VFETCH;
-   } else if (i->src(0).getFile() == FILE_MEMORY_GLOBAL) {
+   } else if (i->src(0).getFile() == FILE_MEMORY_BUFFER) {
       Value *ind = i->getIndirect(0, 1);
       Value *ptr = loadBufInfo64(ind, i->getSrc(0)->reg.fileIndex * 16);
       // XXX come up with a way not to do this for EVERY little access but
@@ -1904,6 +1905,7 @@ NVC0LoweringPass::handleLDST(Instruction *i)
       }
       i->setIndirect(0, 1, NULL);
       i->setIndirect(0, 0, ptr);
+      i->getSrc(0)->reg.file = FILE_MEMORY_GLOBAL;
       bld.mkCmp(OP_SET, CC_GT, TYPE_U32, pred, TYPE_U32, offset, length);
       i->setPredicate(CC_NOT_P, pred);
       if (i->defExists(0)) {
@@ -2241,7 +2243,7 @@ NVC0LoweringPass::visit(Instruction *i)
       break;
    case OP_ATOM:
    {
-      const bool cctl = i->src(0).getFile() == FILE_MEMORY_GLOBAL;
+      const bool cctl = i->src(0).getFile() == FILE_MEMORY_BUFFER;
       handleATOM(i);
       handleCasExch(i, cctl);
    }

src/gallium/drivers/nouveau/codegen/nv50_ir_print.cpp

@@ -460,6 +460,7 @@ int Symbol::print(char *buf, size_t size,
    case FILE_MEMORY_CONST:  c = 'c'; break;
    case FILE_SHADER_INPUT:  c = 'a'; break;
    case FILE_SHADER_OUTPUT: c = 'o'; break;
+   case FILE_MEMORY_BUFFER: c = 'b'; break; // Only used before lowering
    case FILE_MEMORY_GLOBAL: c = 'g'; break;
    case FILE_MEMORY_SHARED: c = 's'; break;
    case FILE_MEMORY_LOCAL:  c = 'l'; break;

src/gallium/drivers/nouveau/codegen/nv50_ir_target_nv50.cpp

@@ -207,6 +207,7 @@ TargetNV50::getFileSize(DataFile file) const
    case FILE_MEMORY_CONST:  return 65536;
    case FILE_SHADER_INPUT:  return 0x200;
    case FILE_SHADER_OUTPUT: return 0x200;
+   case FILE_MEMORY_BUFFER: return 0xffffffff;
    case FILE_MEMORY_GLOBAL: return 0xffffffff;
    case FILE_MEMORY_SHARED: return 16 << 10;
    case FILE_MEMORY_LOCAL:  return 48 << 10;
@@ -406,7 +407,8 @@ TargetNV50::isAccessSupported(DataFile file, DataType ty) const
    if (ty == TYPE_B96 || ty == TYPE_NONE)
       return false;
    if (typeSizeof(ty) > 4)
-      return (file == FILE_MEMORY_LOCAL) || (file == FILE_MEMORY_GLOBAL);
+      return (file == FILE_MEMORY_LOCAL) || (file == FILE_MEMORY_GLOBAL) ||
+             (file == FILE_MEMORY_BUFFER);
    return true;
 }
 
@@ -509,6 +511,7 @@ int TargetNV50::getLatency(const Instruction *i) const
       switch (i->src(0).getFile()) {
       case FILE_MEMORY_LOCAL:
       case FILE_MEMORY_GLOBAL:
+      case FILE_MEMORY_BUFFER:
          return 100; // really 400 to 800
       default:
          return 22;

src/gallium/drivers/nouveau/codegen/nv50_ir_target_nvc0.cpp

@@ -248,6 +248,7 @@ TargetNVC0::getFileSize(DataFile file) const
    case FILE_MEMORY_CONST:  return 65536;
    case FILE_SHADER_INPUT:  return 0x400;
    case FILE_SHADER_OUTPUT: return 0x400;
+   case FILE_MEMORY_BUFFER: return 0xffffffff;
    case FILE_MEMORY_GLOBAL: return 0xffffffff;
    case FILE_MEMORY_SHARED: return 16 << 10;
    case FILE_MEMORY_LOCAL:  return 48 << 10;