[llvm] 7ad3649 - AMDGPU: Fix alignment for dynamic allocas

[Matt Arsenault via llvm-commits]

Author: Matt Arsenault
Date: 2020-06-01T13:06:37-04:00
New Revision: 7ad36491cadee1ed91e18215c30328a3d5c104f9

URL: https://github.com/llvm/llvm-project/commit/7ad36491cadee1ed91e18215c30328a3d5c104f9
DIFF: https://github.com/llvm/llvm-project/commit/7ad36491cadee1ed91e18215c30328a3d5c104f9.diff

LOG: AMDGPU: Fix alignment for dynamic allocas

The alignment value also needs to be scaled by the wave size.

Added: 
    

Modified: 
    llvm/lib/Target/AMDGPU/SIISelLowering.cpp
    llvm/test/CodeGen/AMDGPU/non-entry-alloca.ll

Removed: 
    


################################################################################
diff  --git a/llvm/lib/Target/AMDGPU/SIISelLowering.cpp b/llvm/lib/Target/AMDGPU/SIISelLowering.cpp
index bbd3737d2ef0..08effeea1812 100644
--- a/llvm/lib/Target/AMDGPU/SIISelLowering.cpp
+++ b/llvm/lib/Target/AMDGPU/SIISelLowering.cpp
@@ -3126,9 +3126,12 @@ SDValue SITargetLowering::lowerDYNAMIC_STACKALLOCImpl(
 
   unsigned StackAlign = TFL->getStackAlignment();
   Tmp1 = DAG.getNode(Opc, dl, VT, SP, ScaledSize); // Value
-  if (Align > StackAlign)
-    Tmp1 = DAG.getNode(ISD::AND, dl, VT, Tmp1,
-                       DAG.getConstant(-(uint64_t)Align, dl, VT));
+  if (Align > StackAlign) {
+    Tmp1 = DAG.getNode(
+      ISD::AND, dl, VT, Tmp1,
+      DAG.getConstant(-(uint64_t)Align << ST.getWavefrontSizeLog2(), dl, VT));
+  }
+
   Chain = DAG.getCopyToReg(Chain, dl, SPReg, Tmp1);    // Output chain
   Tmp2 = DAG.getCALLSEQ_END(
       Chain, DAG.getIntPtrConstant(0, dl, true),

diff  --git a/llvm/test/CodeGen/AMDGPU/non-entry-alloca.ll b/llvm/test/CodeGen/AMDGPU/non-entry-alloca.ll
index 060d66ae8428..0cd60bd8203e 100644
--- a/llvm/test/CodeGen/AMDGPU/non-entry-alloca.ll
+++ b/llvm/test/CodeGen/AMDGPU/non-entry-alloca.ll
@@ -95,7 +95,7 @@ define amdgpu_kernel void @kernel_non_entry_block_static_alloca_uniformly_reache
 ; GCN-NEXT:    s_cbranch_scc1 BB1_2
 ; GCN-NEXT:  ; %bb.1: ; %bb.0
 ; GCN-NEXT:    s_add_i32 s6, s32, 0x1000
-; GCN-NEXT:    s_andn2_b32 s6, s6, 63
+; GCN-NEXT:    s_and_b32 s6, s6, 0xfffff000
 ; GCN-NEXT:    s_lshl_b32 s7, s7, 2
 ; GCN-NEXT:    s_mov_b32 s32, s6
 ; GCN-NEXT:    v_mov_b32_e32 v2, s6
@@ -223,7 +223,7 @@ define void @func_non_entry_block_static_alloca_align64(i32 addrspace(1)* %out,
 ; GCN-NEXT:    s_cbranch_execz BB3_2
 ; GCN-NEXT:  ; %bb.1: ; %bb.0
 ; GCN-NEXT:    s_add_i32 s6, s32, 0x1000
-; GCN-NEXT:    s_andn2_b32 s6, s6, 63
+; GCN-NEXT:    s_and_b32 s6, s6, 0xfffff000
 ; GCN-NEXT:    v_mov_b32_e32 v2, 0
 ; GCN-NEXT:    v_mov_b32_e32 v5, s6
 ; GCN-NEXT:    v_mov_b32_e32 v6, 1