[llvm] r275893 - [NVPTX] Force minimum alignment of 4 for byval arguments of device-side functions.

[Artem Belevich via llvm-commits]

Author: tra
Date: Mon Jul 18 14:54:56 2016
New Revision: 275893

URL: http://llvm.org/viewvc/llvm-project?rev=275893&view=rev
Log:
[NVPTX] Force minimum alignment of 4 for byval arguments of device-side functions.

Taking address of a byval variable in PTX is legal, but currently runs
into miscompilation by ptxas on sm_50+ (NVIDIA issue 1789042).
Work around the issue by enforcing minimum alignment on byval arguments
of device functions.

The change is a no-op on SASS level for sm_3x where ptxas already aligns
local copy by at least 4.

Differential Revision: https://reviews.llvm.org/D22428

Modified:
    llvm/trunk/lib/Target/NVPTX/NVPTXAsmPrinter.cpp
    llvm/trunk/lib/Target/NVPTX/NVPTXISelLowering.cpp
    llvm/trunk/test/CodeGen/NVPTX/param-align.ll

Modified: llvm/trunk/lib/Target/NVPTX/NVPTXAsmPrinter.cpp
URL: http://llvm.org/viewvc/llvm-project/llvm/trunk/lib/Target/NVPTX/NVPTXAsmPrinter.cpp?rev=275893&r1=275892&r2=275893&view=diff
==============================================================================
--- llvm/trunk/lib/Target/NVPTX/NVPTXAsmPrinter.cpp (original)
+++ llvm/trunk/lib/Target/NVPTX/NVPTXAsmPrinter.cpp Mon Jul 18 14:54:56 2016
@@ -1589,7 +1589,19 @@ void NVPTXAsmPrinter::emitFunctionParamL
       unsigned align = PAL.getParamAlignment(paramIndex + 1);
       if (align == 0)
         align = DL.getABITypeAlignment(ETy);
-
+      // Work around a bug in ptxas. When PTX code takes address of
+      // byval parameter with alignment < 4, ptxas generates code to
+      // spill argument into memory. Alas on sm_50+ ptxas generates
+      // SASS code that fails with misaligned access. To work around
+      // the problem, make sure that we align byval parameters by at
+      // least 4. Matching change must be made in LowerCall() where we
+      // prepare parameters for the call.
+      //
+      // TODO: this will need to be undone when we get to support multi-TU
+      // device-side compilation as it breaks ABI compatibility with nvcc.
+      // Hopefully ptxas bug is fixed by then.
+      if (!isKernelFunc && align < 4)
+        align = 4;
       unsigned sz = DL.getTypeAllocSize(ETy);
       O << "\t.param .align " << align << " .b8 ";
       printParamName(I, paramIndex, O);

Modified: llvm/trunk/lib/Target/NVPTX/NVPTXISelLowering.cpp
URL: http://llvm.org/viewvc/llvm-project/llvm/trunk/lib/Target/NVPTX/NVPTXISelLowering.cpp?rev=275893&r1=275892&r2=275893&view=diff
==============================================================================
--- llvm/trunk/lib/Target/NVPTX/NVPTXISelLowering.cpp (original)
+++ llvm/trunk/lib/Target/NVPTX/NVPTXISelLowering.cpp Mon Jul 18 14:54:56 2016
@@ -1072,6 +1072,7 @@ SDValue NVPTXTargetLowering::LowerCall(T
   MachineFunction &MF = DAG.getMachineFunction();
   const Function *F = MF.getFunction();
   auto &DL = MF.getDataLayout();
+  bool isKernel = llvm::isKernelFunction(*F);
 
   SDValue tempChain = Chain;
   Chain = DAG.getCALLSEQ_START(Chain,
@@ -1337,11 +1338,15 @@ SDValue NVPTXTargetLowering::LowerCall(T
     // The ByValAlign in the Outs[OIdx].Flags is alway set at this point,
     // so we don't need to worry about natural alignment or not.
     // See TargetLowering::LowerCallTo().
-    SDValue DeclareParamOps[] = {
-      Chain, DAG.getConstant(Outs[OIdx].Flags.getByValAlign(), dl, MVT::i32),
-      DAG.getConstant(paramCount, dl, MVT::i32),
-      DAG.getConstant(sz, dl, MVT::i32), InFlag
-    };
+
+    // Enforce minumum alignment of 4 to work around ptxas miscompile
+    // for sm_50+. See corresponding alignment adjustment in
+    // emitFunctionParamList() for details.
+    if (!isKernel && ArgAlign < 4)
+      ArgAlign = 4;
+    SDValue DeclareParamOps[] = {Chain, DAG.getConstant(ArgAlign, dl, MVT::i32),
+                                 DAG.getConstant(paramCount, dl, MVT::i32),
+                                 DAG.getConstant(sz, dl, MVT::i32), InFlag};
     Chain = DAG.getNode(NVPTXISD::DeclareParam, dl, DeclareParamVTs,
                         DeclareParamOps);
     InFlag = Chain.getValue(1);

Modified: llvm/trunk/test/CodeGen/NVPTX/param-align.ll
URL: http://llvm.org/viewvc/llvm-project/llvm/trunk/test/CodeGen/NVPTX/param-align.ll?rev=275893&r1=275892&r2=275893&view=diff
==============================================================================
--- llvm/trunk/test/CodeGen/NVPTX/param-align.ll (original)
+++ llvm/trunk/test/CodeGen/NVPTX/param-align.ll Mon Jul 18 14:54:56 2016
@@ -23,3 +23,11 @@ define ptx_device void @t3(%struct.float
 ; CHECK: .param .align 4 .b8 t3_param_0[8]
   ret void
 }
+
+;;; Need at least 4-byte alignment in order to avoid miscompilation by
+;;; ptxas for sm_50+
+define ptx_device void @t4(i8* byval %x) {
+; CHECK: .func t4
+; CHECK: .param .align 4 .b8 t4_param_0[1]
+  ret void
+}