AMDGPU: Round up kernel argument allocation size
AFAIK the driver's allocation will actually have to round this
up anyway. It is useful to track the rounded up size, so that
the end of the kernel segment is known to be dereferencable so
a wider s_load_dword can be used for a short argument at the end
of the segment.
llvm-svn: 333456
diff --git a/llvm/lib/Target/AMDGPU/AMDGPUMachineFunction.h b/llvm/lib/Target/AMDGPU/AMDGPUMachineFunction.h
index 6f50fca..bcc0e77 100644
--- a/llvm/lib/Target/AMDGPU/AMDGPUMachineFunction.h
+++ b/llvm/lib/Target/AMDGPU/AMDGPUMachineFunction.h
@@ -20,6 +20,7 @@
/// local memory space.
SmallDenseMap<const GlobalValue *, unsigned, 4> LocalMemoryObjects;
+protected:
uint64_t KernArgSize;
unsigned MaxKernArgAlign;
diff --git a/llvm/lib/Target/AMDGPU/AMDGPUSubtarget.cpp b/llvm/lib/Target/AMDGPU/AMDGPUSubtarget.cpp
index a52b113..97fc649 100644
--- a/llvm/lib/Target/AMDGPU/AMDGPUSubtarget.cpp
+++ b/llvm/lib/Target/AMDGPU/AMDGPUSubtarget.cpp
@@ -414,12 +414,16 @@
unsigned SISubtarget::getKernArgSegmentSize(const Function &F,
unsigned ExplicitArgBytes) const {
+ uint64_t TotalSize = ExplicitArgBytes;
unsigned ImplicitBytes = getImplicitArgNumBytes(F);
- if (ImplicitBytes == 0)
- return ExplicitArgBytes;
- unsigned Alignment = getAlignmentForImplicitArgPtr();
- return alignTo(ExplicitArgBytes, Alignment) + ImplicitBytes;
+ if (ImplicitBytes != 0) {
+ unsigned Alignment = getAlignmentForImplicitArgPtr();
+ TotalSize = alignTo(ExplicitArgBytes, Alignment) + ImplicitBytes;
+ }
+
+ // Being able to dereference past the end is useful for emitting scalar loads.
+ return alignTo(TotalSize, 4);
}
unsigned SISubtarget::getOccupancyWithNumSGPRs(unsigned SGPRs) const {
diff --git a/llvm/lib/Target/AMDGPU/SIMachineFunctionInfo.cpp b/llvm/lib/Target/AMDGPU/SIMachineFunctionInfo.cpp
index 495a853..61b6cb3 100644
--- a/llvm/lib/Target/AMDGPU/SIMachineFunctionInfo.cpp
+++ b/llvm/lib/Target/AMDGPU/SIMachineFunctionInfo.cpp
@@ -71,8 +71,11 @@
if (F.hasFnAttribute("amdgpu-implicitarg-ptr"))
ImplicitArgPtr = true;
} else {
- if (F.hasFnAttribute("amdgpu-implicitarg-ptr"))
+ if (F.hasFnAttribute("amdgpu-implicitarg-ptr")) {
KernargSegmentPtr = true;
+ assert(MaxKernArgAlign == 0);
+ MaxKernArgAlign = ST.getAlignmentForImplicitArgPtr();
+ }
}
CallingConv::ID CC = F.getCallingConv();