From 7a70a2f04d6463013921e9ff1bdad0b02b26b880 Mon Sep 17 00:00:00 2001
From: Jessica Paquette <jpaquette@apple.com>
Date: Wed, 11 Nov 2020 13:21:35 -0800
Subject: [PATCH] [AArch64][GlobalISel] Mark G_FCONSTANT as legal when there is
 full fp16 support

When there is full fp16 support, there is no reason to widen 16-bit
G_FCONSTANTs to 32 bits. Mark them as legal in this case.

Also, we currently import a pattern for materializing a 16-bit 0.0.
Add a testcase showing we select it.

(All other 16-bit G_FCONSTANTS are not yet selected.)

Differential Revision: https://reviews.llvm.org/D89164
---
 .../Target/AArch64/GISel/AArch64LegalizerInfo.cpp  | 15 ++++++++++++---
 .../AArch64/GlobalISel/legalize-fp16-fconstant.mir | 22 ++++++++++++++++++++++
 .../GlobalISel/legalizer-info-validation.mir       |  4 ++--
 .../AArch64/GlobalISel/select-fp16-fconstant.mir   | 17 +++++++++++++++++
 4 files changed, 53 insertions(+), 5 deletions(-)
 create mode 100644 llvm/test/CodeGen/AArch64/GlobalISel/legalize-fp16-fconstant.mir
 create mode 100644 llvm/test/CodeGen/AArch64/GlobalISel/select-fp16-fconstant.mir

diff --git a/llvm/lib/Target/AArch64/GISel/AArch64LegalizerInfo.cpp b/llvm/lib/Target/AArch64/GISel/AArch64LegalizerInfo.cpp
index 8e7186d..efc82ff 100644
--- a/llvm/lib/Target/AArch64/GISel/AArch64LegalizerInfo.cpp
+++ b/llvm/lib/Target/AArch64/GISel/AArch64LegalizerInfo.cpp
@@ -71,6 +71,11 @@ AArch64LegalizerInfo::AArch64LegalizerInfo(const AArch64Subtarget &ST)
     return;
   }
 
+  // Some instructions only support s16 if the subtarget has full 16-bit FP
+  // support.
+  const bool HasFP16 = ST.hasFullFP16();
+  const LLT &MinFPScalar = HasFP16 ? s16 : s32;
+
   getActionDefinitionsBuilder({G_IMPLICIT_DEF, G_FREEZE})
       .legalFor({p0, s1, s8, s16, s32, s64})
       .legalFor(PackedVectorAllTypeList)
@@ -323,9 +328,13 @@ AArch64LegalizerInfo::AArch64LegalizerInfo(const AArch64Subtarget &ST)
       .clampScalar(0, s8, s64)
       .widenScalarToNextPow2(0);
   getActionDefinitionsBuilder(G_FCONSTANT)
-      .legalFor({s32, s64})
-      .clampScalar(0, s32, s64);
-
+      .legalIf([=](const LegalityQuery &Query) {
+        const auto &Ty = Query.Types[0];
+        if (HasFP16 && Ty == s16)
+          return true;
+        return Ty == s32 || Ty == s64;
+      })
+      .clampScalar(0, MinFPScalar, s64);
 
   getActionDefinitionsBuilder({G_ICMP, G_FCMP})
       .legalFor({{s32, s32},
diff --git a/llvm/test/CodeGen/AArch64/GlobalISel/legalize-fp16-fconstant.mir b/llvm/test/CodeGen/AArch64/GlobalISel/legalize-fp16-fconstant.mir
new file mode 100644
index 0000000..685035a
--- /dev/null
+++ b/llvm/test/CodeGen/AArch64/GlobalISel/legalize-fp16-fconstant.mir
@@ -0,0 +1,22 @@
+# NOTE: Assertions have been autogenerated by utils/update_mir_test_checks.py
+# RUN: llc -mtriple aarch64-unknown-unknown -verify-machineinstrs -run-pass=legalizer %s -o - | FileCheck %s --check-prefix=NO-FP16
+# RUN: llc -mtriple aarch64-unknown-unknown -verify-machineinstrs -mattr=+fullfp16 -run-pass=legalizer %s -o - | FileCheck %s --check-prefix=FP16
+
+---
+name:            fp16
+tracksRegLiveness: true
+body: |
+  bb.0:
+    ; NO-FP16-LABEL: name: fp16
+    ; NO-FP16: [[C:%[0-9]+]]:_(s32) = G_FCONSTANT float 0.000000e+00
+    ; NO-FP16: %cst:_(s16) = G_FPTRUNC [[C]](s32)
+    ; NO-FP16: $h0 = COPY %cst(s16)
+    ; NO-FP16: RET_ReallyLR implicit $h0
+    ; FP16-LABEL: name: fp16
+    ; FP16: %cst:_(s16) = G_FCONSTANT half 0xH0000
+    ; FP16: $h0 = COPY %cst(s16)
+    ; FP16: RET_ReallyLR implicit $h0
+    %cst:_(s16) = G_FCONSTANT half 0.0
+    $h0 = COPY %cst
+    RET_ReallyLR implicit $h0
+
diff --git a/llvm/test/CodeGen/AArch64/GlobalISel/legalizer-info-validation.mir b/llvm/test/CodeGen/AArch64/GlobalISel/legalizer-info-validation.mir
index 8ca63a1..a937e33 100644
--- a/llvm/test/CodeGen/AArch64/GlobalISel/legalizer-info-validation.mir
+++ b/llvm/test/CodeGen/AArch64/GlobalISel/legalizer-info-validation.mir
@@ -247,8 +247,8 @@
 # DEBUG-NEXT: .. the first uncovered type index: 1, OK
 # DEBUG-NEXT: .. the first uncovered imm index: 0, OK
 # DEBUG-NEXT: G_FCONSTANT (opcode {{[0-9]+}}): 1 type index, 0 imm indices
-# DEBUG-NEXT: .. the first uncovered type index: 1, OK
-# DEBUG-NEXT: .. the first uncovered imm index: 0, OK
+# DEBUG-NEXT: .. type index coverage check SKIPPED: user-defined predicate detected
+# DEBUG-NEXT: .. imm index coverage check SKIPPED: user-defined predicate detected
 # DEBUG-NEXT: G_VASTART (opcode {{[0-9]+}}): 1 type index, 0 imm indices
 # DEBUG-NEXT: .. the first uncovered type index: 1, OK
 # DEBUG-NEXT: .. the first uncovered imm index: 0, OK
diff --git a/llvm/test/CodeGen/AArch64/GlobalISel/select-fp16-fconstant.mir b/llvm/test/CodeGen/AArch64/GlobalISel/select-fp16-fconstant.mir
new file mode 100644
index 0000000..59ee5bb
--- /dev/null
+++ b/llvm/test/CodeGen/AArch64/GlobalISel/select-fp16-fconstant.mir
@@ -0,0 +1,17 @@
+# NOTE: Assertions have been autogenerated by utils/update_mir_test_checks.py
+# RUN: llc -mtriple=aarch64-unknown-unknown -mattr=+fullfp16 -run-pass=instruction-select -verify-machineinstrs %s -o - | FileCheck %s
+
+---
+name:            positive_zero
+legalized:       true
+regBankSelected: true
+tracksRegLiveness: true
+body:             |
+  bb.0:
+    ; CHECK-LABEL: name: positive_zero
+    ; CHECK: [[FMOVH0_:%[0-9]+]]:fpr16 = FMOVH0
+    ; CHECK: $h0 = COPY [[FMOVH0_]]
+    ; CHECK: RET_ReallyLR implicit $h0
+    %0:fpr(s16) = G_FCONSTANT half 0.0
+    $h0 = COPY %0(s16)
+    RET_ReallyLR implicit $h0
-- 
2.7.4