[SM6.9] Enable trivial native vector Dxil Operations plus a few (microsoft#7324)

This enables the generation of native vector DXIL Operations
that are "trivial", meaning they take only a single DXOp Call
instruction to implement as well as a few others that either only took
such a call and some llvm operations or were of particular interest for
other reasons.

This involves allowing the overloads by adding the vector indication in
hctdb, altering the lowering to maintain the vectors instead of
scalarizing them, and a few sundry changes to fix issues along the way.

The "trivial" dxil operations that return a different value from the
overload type had to be moved out of the way and given their own
lowering function so that the main function could generate vectors
conditional on the version and vector type. These will be added in a
later change.

While the long vector supporting intrinsics that weren't given this
treatment will continue to generate scalarized operations, some of them
needed some work as well. The dot product for float vectors longer than
4 had to take the integer fallback path, which required some small
modificaitons and a rename.
Additionally, a heuristic for pow that malfunctioned with too many
elements had to have a limit placed on it.

Since the or()/and()/select() intrinsics translate directly to LLVM ops,
they can have their lowering scalarization removed and what future
scalarization might be needed by the current version can be done by
later passes as with other LLVM operators.

An issue with a special value used to represent unassined dimensions had
to be addressed since new dimensions can exceed that value. It's now
MAX_INT.

Contributes to microsoft#7120, but I'd prefer to leave it open until all
intrinsics are covered

Primary work by @pow2clk

Fixes microsoft#7297 & microsoft#7120

---------

Co-authored-by: Greg Roth <[email protected]>

Author: llvm-beanz

lib/DXIL/DxilOperations.cpp

@@ -96,16 +96,16 @@ const OP::OpCodeProperty OP::m_OpCodeProps[(unsigned)OP::OpCode::NumOpCodes] = {
      "unary",
      Attribute::ReadNone,
      1,
-     {{0x7}},
-     {{0x0}}}, // Overloads: hfd
+     {{0x407}},
+     {{0x7}}}, // Overloads: hfd<hfd
     {OC::Saturate,
      "Saturate",
      OCC::Unary,
      "unary",
      Attribute::ReadNone,
      1,
-     {{0x7}},
-     {{0x0}}}, // Overloads: hfd
+     {{0x407}},
+     {{0x7}}}, // Overloads: hfd<hfd
     {OC::IsNaN,
      "IsNaN",
      OCC::IsSpecialFloat,
@@ -144,112 +144,112 @@ const OP::OpCodeProperty OP::m_OpCodeProps[(unsigned)OP::OpCode::NumOpCodes] = {
      "unary",
      Attribute::ReadNone,
      1,
-     {{0x3}},
-     {{0x0}}}, // Overloads: hf
+     {{0x403}},
+     {{0x3}}}, // Overloads: hf<hf
     {OC::Sin,
      "Sin",
      OCC::Unary,
      "unary",
      Attribute::ReadNone,
      1,
-     {{0x3}},
-     {{0x0}}}, // Overloads: hf
+     {{0x403}},
+     {{0x3}}}, // Overloads: hf<hf
     {OC::Tan,
      "Tan",
      OCC::Unary,
      "unary",
      Attribute::ReadNone,
      1,
-     {{0x3}},
-     {{0x0}}}, // Overloads: hf
+     {{0x403}},
+     {{0x3}}}, // Overloads: hf<hf
     {OC::Acos,
      "Acos",
      OCC::Unary,
      "unary",
      Attribute::ReadNone,
      1,
-     {{0x3}},
-     {{0x0}}}, // Overloads: hf
+     {{0x403}},
+     {{0x3}}}, // Overloads: hf<hf
     {OC::Asin,
      "Asin",
      OCC::Unary,
      "unary",
      Attribute::ReadNone,
      1,
-     {{0x3}},
-     {{0x0}}}, // Overloads: hf
+     {{0x403}},
+     {{0x3}}}, // Overloads: hf<hf
     {OC::Atan,
      "Atan",
      OCC::Unary,
      "unary",
      Attribute::ReadNone,
      1,
-     {{0x3}},
-     {{0x0}}}, // Overloads: hf
+     {{0x403}},
+     {{0x3}}}, // Overloads: hf<hf
     {OC::Hcos,
      "Hcos",
      OCC::Unary,
      "unary",
      Attribute::ReadNone,
      1,
-     {{0x3}},
-     {{0x0}}}, // Overloads: hf
+     {{0x403}},
+     {{0x3}}}, // Overloads: hf<hf
     {OC::Hsin,
      "Hsin",
      OCC::Unary,
      "unary",
      Attribute::ReadNone,
      1,
-     {{0x3}},
-     {{0x0}}}, // Overloads: hf
+     {{0x403}},
+     {{0x3}}}, // Overloads: hf<hf
     {OC::Htan,
      "Htan",
      OCC::Unary,
      "unary",
      Attribute::ReadNone,
      1,
-     {{0x3}},
-     {{0x0}}}, // Overloads: hf
+     {{0x403}},
+     {{0x3}}}, // Overloads: hf<hf
     {OC::Exp,
      "Exp",
      OCC::Unary,
      "unary",
      Attribute::ReadNone,
      1,
-     {{0x3}},
-     {{0x0}}}, // Overloads: hf
+     {{0x403}},
+     {{0x3}}}, // Overloads: hf<hf
     {OC::Frc,
      "Frc",
      OCC::Unary,
      "unary",
      Attribute::ReadNone,
      1,
-     {{0x3}},
-     {{0x0}}}, // Overloads: hf
+     {{0x403}},
+     {{0x3}}}, // Overloads: hf<hf
     {OC::Log,
      "Log",
      OCC::Unary,
      "unary",
      Attribute::ReadNone,
      1,
-     {{0x3}},
-     {{0x0}}}, // Overloads: hf
+     {{0x403}},
+     {{0x3}}}, // Overloads: hf<hf
     {OC::Sqrt,
      "Sqrt",
      OCC::Unary,
      "unary",
      Attribute::ReadNone,
      1,
-     {{0x3}},
-     {{0x0}}}, // Overloads: hf
+     {{0x403}},
+     {{0x3}}}, // Overloads: hf<hf
     {OC::Rsqrt,
      "Rsqrt",
      OCC::Unary,
      "unary",
      Attribute::ReadNone,
      1,
-     {{0x3}},
-     {{0x0}}}, // Overloads: hf
+     {{0x403}},
+     {{0x3}}}, // Overloads: hf<hf
 
     // Unary float - rounding
     {OC::Round_ne,
@@ -258,32 +258,32 @@ const OP::OpCodeProperty OP::m_OpCodeProps[(unsigned)OP::OpCode::NumOpCodes] = {
      "unary",
      Attribute::ReadNone,
      1,
-     {{0x3}},
-     {{0x0}}}, // Overloads: hf
+     {{0x403}},
+     {{0x3}}}, // Overloads: hf<hf
     {OC::Round_ni,
      "Round_ni",
      OCC::Unary,
      "unary",
      Attribute::ReadNone,
      1,
-     {{0x3}},
-     {{0x0}}}, // Overloads: hf
+     {{0x403}},
+     {{0x3}}}, // Overloads: hf<hf
     {OC::Round_pi,
      "Round_pi",
      OCC::Unary,
      "unary",
      Attribute::ReadNone,
      1,
-     {{0x3}},
-     {{0x0}}}, // Overloads: hf
+     {{0x403}},
+     {{0x3}}}, // Overloads: hf<hf
     {OC::Round_z,
      "Round_z",
      OCC::Unary,
      "unary",
      Attribute::ReadNone,
      1,
-     {{0x3}},
-     {{0x0}}}, // Overloads: hf
+     {{0x403}},
+     {{0x3}}}, // Overloads: hf<hf
 
     // Unary int
     {OC::Bfrev,
@@ -292,8 +292,8 @@ const OP::OpCodeProperty OP::m_OpCodeProps[(unsigned)OP::OpCode::NumOpCodes] = {
      "unary",
      Attribute::ReadNone,
      1,
-     {{0xe0}},
-     {{0x0}}}, // Overloads: wil
+     {{0x4e0}},
+     {{0xe0}}}, // Overloads: wil<wil
     {OC::Countbits,
      "Countbits",
      OCC::UnaryBits,
@@ -338,16 +338,16 @@ const OP::OpCodeProperty OP::m_OpCodeProps[(unsigned)OP::OpCode::NumOpCodes] = {
      "binary",
      Attribute::ReadNone,
      1,
-     {{0x7}},
-     {{0x0}}}, // Overloads: hfd
+     {{0x407}},
+     {{0x7}}}, // Overloads: hfd<hfd
     {OC::FMin,
      "FMin",
      OCC::Binary,
      "binary",
      Attribute::ReadNone,
      1,
-     {{0x7}},
-     {{0x0}}}, // Overloads: hfd
+     {{0x407}},
+     {{0x7}}}, // Overloads: hfd<hfd
 
     // Binary int
     {OC::IMax,
@@ -356,16 +356,16 @@ const OP::OpCodeProperty OP::m_OpCodeProps[(unsigned)OP::OpCode::NumOpCodes] = {
      "binary",
      Attribute::ReadNone,
      1,
-     {{0xe0}},
-     {{0x0}}}, // Overloads: wil
+     {{0x4e0}},
+     {{0xe0}}}, // Overloads: wil<wil
     {OC::IMin,
      "IMin",
      OCC::Binary,
      "binary",
      Attribute::ReadNone,
      1,
-     {{0xe0}},
-     {{0x0}}}, // Overloads: wil
+     {{0x4e0}},
+     {{0xe0}}}, // Overloads: wil<wil
 
     // Binary uint
     {OC::UMax,
@@ -374,16 +374,16 @@ const OP::OpCodeProperty OP::m_OpCodeProps[(unsigned)OP::OpCode::NumOpCodes] = {
      "binary",
      Attribute::ReadNone,
      1,
-     {{0xe0}},
-     {{0x0}}}, // Overloads: wil
+     {{0x4e0}},
+     {{0xe0}}}, // Overloads: wil<wil
     {OC::UMin,
      "UMin",
      OCC::Binary,
      "binary",
      Attribute::ReadNone,
      1,
-     {{0xe0}},
-     {{0x0}}}, // Overloads: wil
+     {{0x4e0}},
+     {{0xe0}}}, // Overloads: wil<wil
 
     // Binary int with two outputs
     {OC::IMul,
@@ -438,16 +438,16 @@ const OP::OpCodeProperty OP::m_OpCodeProps[(unsigned)OP::OpCode::NumOpCodes] = {
      "tertiary",
      Attribute::ReadNone,
      1,
-     {{0x7}},
-     {{0x0}}}, // Overloads: hfd
+     {{0x407}},
+     {{0x7}}}, // Overloads: hfd<hfd
     {OC::Fma,
      "Fma",
      OCC::Tertiary,
      "tertiary",
      Attribute::ReadNone,
      1,
-     {{0x4}},
-     {{0x0}}}, // Overloads: d
+     {{0x404}},
+     {{0x4}}}, // Overloads: d<d
 
     // Tertiary int
     {OC::IMad,
@@ -456,8 +456,8 @@ const OP::OpCodeProperty OP::m_OpCodeProps[(unsigned)OP::OpCode::NumOpCodes] = {
      "tertiary",
      Attribute::ReadNone,
      1,
-     {{0xe0}},
-     {{0x0}}}, // Overloads: wil
+     {{0x4e0}},
+     {{0xe0}}}, // Overloads: wil<wil
 
     // Tertiary uint
     {OC::UMad,
@@ -466,8 +466,8 @@ const OP::OpCodeProperty OP::m_OpCodeProps[(unsigned)OP::OpCode::NumOpCodes] = {
      "tertiary",
      Attribute::ReadNone,
      1,
-     {{0xe0}},
-     {{0x0}}}, // Overloads: wil
+     {{0x4e0}},
+     {{0xe0}}}, // Overloads: wil<wil
 
     // Tertiary int
     {OC::Msad,
@@ -764,32 +764,32 @@ const OP::OpCodeProperty OP::m_OpCodeProps[(unsigned)OP::OpCode::NumOpCodes] = {
      "unary",
      Attribute::ReadNone,
      1,
-     {{0x3}},
-     {{0x0}}}, // Overloads: hf
+     {{0x403}},
+     {{0x3}}}, // Overloads: hf<hf
     {OC::DerivCoarseY,
      "DerivCoarseY",
      OCC::Unary,
      "unary",
      Attribute::ReadNone,
      1,
-     {{0x3}},
-     {{0x0}}}, // Overloads: hf
+     {{0x403}},
+     {{0x3}}}, // Overloads: hf<hf
     {OC::DerivFineX,
      "DerivFineX",
      OCC::Unary,
      "unary",
      Attribute::ReadNone,
      1,
-     {{0x3}},
-     {{0x0}}}, // Overloads: hf
+     {{0x403}},
+     {{0x3}}}, // Overloads: hf<hf
     {OC::DerivFineY,
      "DerivFineY",
      OCC::Unary,
      "unary",
      Attribute::ReadNone,
      1,
-     {{0x3}},
-     {{0x0}}}, // Overloads: hf
+     {{0x403}},
+     {{0x3}}}, // Overloads: hf<hf
 
     // Pixel shader
     {OC::EvalSnapped,