AMDGPU: Account for LDS alignment

The current situation isn't great, because the amount of padding
requires is determined by the inverse order of the first encountered
use. We should eventually somehow sort these to minimize wasted space.

Another problem is the alignment of kernel arguments isn't
respected. The group_segment_alignment is always emitted as
the default 16, and typed arguments with higher alignments
or an explicitly set alignment are also ignored.

llvm-svn: 259912

1 files changed, 9 insertions, 4 deletions

diff --git a/llvm/lib/Target/AMDGPU/AMDGPUISelLowering.cpp b/llvm/lib/Target/AMDGPU/AMDGPUISelLowering.cpp
index dfdef5cdf29..00d6d4bad88 100644
--- a/llvm/lib/Target/AMDGPU/AMDGPUISelLowering.cpp
+++ b/llvm/lib/Target/AMDGPU/AMDGPUISelLowering.cpp
@@ -813,11 +813,16 @@ SDValue AMDGPUTargetLowering::LowerGlobalAddress(AMDGPUMachineFunction* MFI,
 
     unsigned Offset;
     if (MFI->LocalMemoryObjects.count(GV) == 0) {
-      uint64_t Size = DL.getTypeAllocSize(GV->getValueType());
-      Offset = MFI->LDSSize;
+      unsigned Align = GV->getAlignment();
+      if (Align == 0)
+        Align = DL.getABITypeAlignment(GV->getValueType());
+
+      /// TODO: We should sort these to minimize wasted space due to alignment
+      /// padding. Currently the padding is decided by the first encountered use
+      /// during lowering.
+      Offset = MFI->LDSSize = alignTo(MFI->LDSSize, Align);
       MFI->LocalMemoryObjects[GV] = Offset;
-      // XXX: Account for alignment?
-      MFI->LDSSize += Size;
+      MFI->LDSSize += DL.getTypeAllocSize(GV->getValueType());
     } else {
       Offset = MFI->LocalMemoryObjects[GV];
     }