[PATCH] D89326: [SVE] Add fatal error when running out of registers for SVE tuple call arguments

[David Sherwood via Phabricator via llvm-commits]

This revision was landed with ongoing or failed builds.
This revision was automatically updated to reflect the committed changes.
Closed by commit rGaf57a0838eba: [SVE] Add fatal error when running out of registers for SVE tuple call arguments (authored by david-arm).

Changed prior to commit:
  https://reviews.llvm.org/D89326?vs=297882&id=298068#toc

Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D89326/new/

https://reviews.llvm.org/D89326

Files:
  llvm/lib/Target/AArch64/AArch64CallingConvention.cpp
  llvm/lib/Target/AArch64/AArch64ISelLowering.cpp
  llvm/test/CodeGen/AArch64/sve-calling-convention-tuples-broken.ll


Index: llvm/test/CodeGen/AArch64/sve-calling-convention-tuples-broken.ll
===================================================================
--- /dev/null
+++ llvm/test/CodeGen/AArch64/sve-calling-convention-tuples-broken.ll
@@ -0,0 +1,23 @@
+; RUN: not --crash llc < %s -mtriple aarch64-linux-gnu -mattr=+sve >/dev/null 2>%t
+; RUN: FileCheck %s < %t
+
+; CHECK: Passing consecutive scalable vector registers unsupported
+
+target datalayout = "e-m:e-i8:8:32-i16:16:32-i64:64-i128:128-n32:64-S128"
+target triple = "aarch64-unknown-linux-gnu"
+
+define float @foo(double* %x0, double* %x1) {
+entry:
+  %0 = call <vscale x 16 x i1> @llvm.aarch64.sve.ptrue.nxv16i1(i32 31)
+  %1 = call <vscale x 2 x i1> @llvm.aarch64.sve.convert.from.svbool.nxv2i1(<vscale x 16 x i1> %0)
+  %2 = call <vscale x 8 x double> @llvm.aarch64.sve.ld4.nxv8f64.nxv2i1(<vscale x 2 x i1> %1, double* %x0)
+  %3 = call <vscale x 8 x double> @llvm.aarch64.sve.ld4.nxv8f64.nxv2i1(<vscale x 2 x i1> %1, double* %x1)
+  %call = call float @callee(float 1.000000e+00, <vscale x 8 x double> %2, <vscale x 8 x double> %3)
+  ret float %call
+}
+
+declare float @callee(float, <vscale x 8 x double>, <vscale x 8 x double>)
+
+declare <vscale x 16 x i1> @llvm.aarch64.sve.ptrue.nxv16i1(i32 immarg)
+declare <vscale x 2 x i1> @llvm.aarch64.sve.convert.from.svbool.nxv2i1(<vscale x 16 x i1>)
+declare <vscale x 8 x double> @llvm.aarch64.sve.ld4.nxv8f64.nxv2i1(<vscale x 2 x i1>, double*)
Index: llvm/lib/Target/AArch64/AArch64ISelLowering.cpp
===================================================================
--- llvm/lib/Target/AArch64/AArch64ISelLowering.cpp
+++ llvm/lib/Target/AArch64/AArch64ISelLowering.cpp
@@ -15566,7 +15566,14 @@
 
 bool AArch64TargetLowering::functionArgumentNeedsConsecutiveRegisters(
     Type *Ty, CallingConv::ID CallConv, bool isVarArg) const {
-  return Ty->isArrayTy();
+  if (Ty->isArrayTy())
+    return true;
+
+  const TypeSize &TySize = Ty->getPrimitiveSizeInBits();
+  if (TySize.isScalable() && TySize.getKnownMinSize() > 128)
+    return true;
+
+  return false;
 }
 
 bool AArch64TargetLowering::shouldNormalizeToSelectSequence(LLVMContext &,
Index: llvm/lib/Target/AArch64/AArch64CallingConvention.cpp
===================================================================
--- llvm/lib/Target/AArch64/AArch64CallingConvention.cpp
+++ llvm/lib/Target/AArch64/AArch64CallingConvention.cpp
@@ -35,6 +35,9 @@
 static const MCPhysReg QRegList[] = {AArch64::Q0, AArch64::Q1, AArch64::Q2,
                                      AArch64::Q3, AArch64::Q4, AArch64::Q5,
                                      AArch64::Q6, AArch64::Q7};
+static const MCPhysReg ZRegList[] = {AArch64::Z0, AArch64::Z1, AArch64::Z2,
+                                     AArch64::Z3, AArch64::Z4, AArch64::Z5,
+                                     AArch64::Z6, AArch64::Z7};
 
 static bool finishStackBlock(SmallVectorImpl<CCValAssign> &PendingMembers,
                              MVT LocVT, ISD::ArgFlagsTy &ArgFlags,
@@ -97,6 +100,8 @@
     RegList = DRegList;
   else if (LocVT.SimpleTy == MVT::f128 || LocVT.is128BitVector())
     RegList = QRegList;
+  else if (LocVT.isScalableVector())
+    RegList = ZRegList;
   else {
     // Not an array we want to split up after all.
     return false;
@@ -141,6 +146,10 @@
     return true;
   }
 
+  if (LocVT.isScalableVector())
+    report_fatal_error(
+        "Passing consecutive scalable vector registers unsupported");
+
   // Mark all regs in the class as unavailable
   for (auto Reg : RegList)
     State.AllocateReg(Reg);


-------------- next part --------------
A non-text attachment was scrubbed...
Name: D89326.298068.patch
Type: text/x-patch
Size: 3555 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20201014/555cdb69/attachment.bin>