[AMDGPU] Fix kernel LDS lowering for constants

[rampitec]

Summary

There is a trivial but severe bug in the recent code collecting
LDS globals used by kernel. It aborts scan on the first constant
without scanning further uses. That leads to LDS overallocation
with multiple kernels in certain cases.

[JonChesterfield]

Ouch. Good to catch that quickly, thanks

llvm/lib/Target/AMDGPU/Utils/AMDGPULDSUtils.cpp

while (!Stack.empty()) {
     }
 
     if (auto *E = dyn_cast<ConstantExpr>(V)) {
       if (F) {
         // Any use which does not end up an instruction disqualifies a
         // variable to be put into a kernel's LDS structure because later
         // we will need to replace only this kernel's uses for which we
         // need to identify a using function.
-        return isUsedOnlyFromFunction(E, F);
+        if (!isUsedOnlyFromFunction(E, F))
+          return false;
       }
       for (const User *U : E->users()) {
         if (Visited.insert(U).second) {
           Stack.push_back(U);
         }
       }
       continue;
     }

llvm/test/CodeGen/AMDGPU/ds_read2.ll

 @sgemm.lA = internal unnamed_addr addrspace(3) global [264 x float] undef, align 4
 @sgemm.lB = internal unnamed_addr addrspace(3) global [776 x float] undef, align 4
 
 define amdgpu_kernel void @sgemm_inner_loop_read2_sequence(float addrspace(1)* %C, i32 %lda, i32 %ldb) #0 {
 ; CI-LABEL: sgemm_inner_loop_read2_sequence:
 ; CI:       ; %bb.0:
 ; CI-NEXT:    s_load_dwordx2 s[4:5], s[0:1], 0x9
 ; CI-NEXT:    s_lshl_b32 s0, s2, 2
-; CI-NEXT:    s_add_i32 s1, s0, 0x8c40
-; CI-NEXT:    s_add_i32 s0, s0, 0x8c80
+; CI-NEXT:    s_add_i32 s1, s0, 0xc20
+; CI-NEXT:    s_addk_i32 s0, 0xc60
+; CI-NEXT:    v_mov_b32_e32 v0, s1
 ; CI-NEXT:    v_mov_b32_e32 v2, s0
 ; CI-NEXT:    v_lshlrev_b32_e32 v8, 2, v1
-; CI-NEXT:    s_mov_b32 s0, 0x8020
-; CI-NEXT:    v_add_i32_e32 v4, vcc, s0, v8
-; CI-NEXT:    s_mov_b32 s0, 0x80a0
-; CI-NEXT:    v_add_i32_e32 v6, vcc, s0, v8
-; CI-NEXT:    v_mov_b32_e32 v0, s1
 ; CI-NEXT:    s_mov_b32 m0, -1
 ; CI-NEXT:    ds_read2_b32 v[0:1], v0 offset1:1
 ; CI-NEXT:    ds_read2_b32 v[2:3], v2 offset1:1
-; CI-NEXT:    ds_read2_b32 v[4:5], v4 offset1:1
-; CI-NEXT:    ds_read2_b32 v[6:7], v6 offset1:1
-; CI-NEXT:    v_add_i32_e32 v8, vcc, 0x8120, v8
+; CI-NEXT:    ds_read2_b32 v[4:5], v8 offset1:1
+; CI-NEXT:    ds_read2_b32 v[6:7], v8 offset0:32 offset1:33
+; CI-NEXT:    ds_read2_b32 v[8:9], v8 offset0:64 offset1:65
 ; CI-NEXT:    s_waitcnt lgkmcnt(0)
 ; CI-NEXT:    v_add_f32_e32 v0, v0, v1
 ; CI-NEXT:    v_add_f32_e32 v0, v0, v2
 ; CI-NEXT:    v_add_f32_e32 v0, v0, v3
-; CI-NEXT:    ds_read2_b32 v[8:9], v8 offset1:1
 ; CI-NEXT:    v_add_f32_e32 v0, v0, v4
 ; CI-NEXT:    v_add_f32_e32 v0, v0, v5
 ; CI-NEXT:    v_add_f32_e32 v0, v0, v6
 ; CI-NEXT:    v_add_f32_e32 v0, v0, v7
-; CI-NEXT:    s_waitcnt lgkmcnt(0)
 ; CI-NEXT:    v_add_f32_e32 v0, v0, v8
 ; CI-NEXT:    s_mov_b32 s7, 0xf000
 ; CI-NEXT:    s_mov_b32 s6, -1
 ; CI-NEXT:    v_add_f32_e32 v0, v0, v9
 ; CI-NEXT:    buffer_store_dword v0, off, s[4:7], 0
 ; CI-NEXT:    s_endpgm
 ;
 ; GFX9-LABEL: sgemm_inner_loop_read2_sequence:
 ; GFX9:       ; %bb.0:
 ; GFX9-NEXT:    s_lshl_b32 s2, s2, 2
-; GFX9-NEXT:    s_add_i32 s3, s2, 0x8c40
-; GFX9-NEXT:    v_lshlrev_b32_e32 v8, 2, v1
-; GFX9-NEXT:    s_add_i32 s2, s2, 0x8c80
+; GFX9-NEXT:    s_add_i32 s3, s2, 0xc20
+; GFX9-NEXT:    s_addk_i32 s2, 0xc60
 ; GFX9-NEXT:    v_mov_b32_e32 v0, s3
 ; GFX9-NEXT:    v_mov_b32_e32 v2, s2
-; GFX9-NEXT:    v_add_u32_e32 v4, 0x8020, v8
-; GFX9-NEXT:    v_add_u32_e32 v6, 0x80a0, v8
+; GFX9-NEXT:    v_lshlrev_b32_e32 v8, 2, v1
 ; GFX9-NEXT:    ds_read2_b32 v[0:1], v0 offset1:1
 ; GFX9-NEXT:    ds_read2_b32 v[2:3], v2 offset1:1
-; GFX9-NEXT:    ds_read2_b32 v[4:5], v4 offset1:1
-; GFX9-NEXT:    ds_read2_b32 v[6:7], v6 offset1:1
-; GFX9-NEXT:    v_add_u32_e32 v8, 0x8120, v8
-; GFX9-NEXT:    s_waitcnt lgkmcnt(3)
+; GFX9-NEXT:    ds_read2_b32 v[4:5], v8 offset1:1
+; GFX9-NEXT:    ds_read2_b32 v[6:7], v8 offset0:32 offset1:33
+; GFX9-NEXT:    ds_read2_b32 v[8:9], v8 offset0:64 offset1:65
+; GFX9-NEXT:    s_waitcnt lgkmcnt(4)
 ; GFX9-NEXT:    v_add_f32_e32 v0, v0, v1
-; GFX9-NEXT:    s_waitcnt lgkmcnt(2)
+; GFX9-NEXT:    s_waitcnt lgkmcnt(3)
 ; GFX9-NEXT:    v_add_f32_e32 v0, v0, v2
 ; GFX9-NEXT:    v_add_f32_e32 v0, v0, v3
-; GFX9-NEXT:    ds_read2_b32 v[8:9], v8 offset1:1
 ; GFX9-NEXT:    s_waitcnt lgkmcnt(2)
 ; GFX9-NEXT:    v_add_f32_e32 v0, v0, v4
 ; GFX9-NEXT:    v_add_f32_e32 v0, v0, v5
 ; GFX9-NEXT:    s_load_dwordx2 s[0:1], s[0:1], 0x24
 ; GFX9-NEXT:    s_waitcnt lgkmcnt(0)
 ; GFX9-NEXT:    v_add_f32_e32 v0, v0, v6
 ; GFX9-NEXT:    v_add_f32_e32 v0, v0, v7
 ; GFX9-NEXT:    v_add_f32_e32 v0, v0, v8

 
 @v2i32_align1 = internal addrspace(3) global [100 x <2 x i32>] undef, align 1
 
 define amdgpu_kernel void @read2_v2i32_align1_odd_offset(<2 x i32> addrspace(1)* %out) {
 ; CI-LABEL: read2_v2i32_align1_odd_offset:
 ; CI:       ; %bb.0: ; %entry
 ; CI-NEXT:    v_mov_b32_e32 v0, 0
 ; CI-NEXT:    s_mov_b32 m0, -1
-; CI-NEXT:    ds_read_u8 v1, v0 offset:37032
-; CI-NEXT:    ds_read_u8 v2, v0 offset:37031
-; CI-NEXT:    ds_read_u8 v3, v0 offset:37030
-; CI-NEXT:    ds_read_u8 v4, v0 offset:37029
-; CI-NEXT:    ds_read_u8 v5, v0 offset:37028
+; CI-NEXT:    ds_read_u8 v1, v0 offset:72
+; CI-NEXT:    ds_read_u8 v2, v0 offset:71
+; CI-NEXT:    ds_read_u8 v3, v0 offset:70
+; CI-NEXT:    ds_read_u8 v4, v0 offset:69
+; CI-NEXT:    ds_read_u8 v5, v0 offset:68
 ; CI-NEXT:    s_waitcnt lgkmcnt(4)
 ; CI-NEXT:    v_lshlrev_b32_e32 v1, 8, v1
 ; CI-NEXT:    s_waitcnt lgkmcnt(3)
 ; CI-NEXT:    v_or_b32_e32 v1, v1, v2
 ; CI-NEXT:    s_waitcnt lgkmcnt(2)
 ; CI-NEXT:    v_lshlrev_b32_e32 v3, 8, v3
 ; CI-NEXT:    s_waitcnt lgkmcnt(1)
 ; CI-NEXT:    v_or_b32_e32 v3, v3, v4
 ; CI-NEXT:    v_lshlrev_b32_e32 v1, 16, v1
 ; CI-NEXT:    v_or_b32_e32 v1, v1, v3
-; CI-NEXT:    ds_read_u8 v2, v0 offset:37027
-; CI-NEXT:    ds_read_u8 v3, v0 offset:37026
-; CI-NEXT:    ds_read_u8 v0, v0 offset:37025
+; CI-NEXT:    ds_read_u8 v2, v0 offset:67
+; CI-NEXT:    ds_read_u8 v3, v0 offset:66
+; CI-NEXT:    ds_read_u8 v0, v0 offset:65
 ; CI-NEXT:    s_load_dwordx2 s[0:1], s[0:1], 0x9
 ; CI-NEXT:    s_mov_b32 s3, 0xf000
 ; CI-NEXT:    s_mov_b32 s2, -1
 ; CI-NEXT:    s_waitcnt lgkmcnt(0)
 ; CI-NEXT:    v_lshlrev_b32_e32 v3, 8, v3
 ; CI-NEXT:    v_or_b32_e32 v0, v3, v0
 ; CI-NEXT:    v_lshlrev_b32_e32 v3, 8, v5
 ; CI-NEXT:    v_or_b32_e32 v2, v3, v2
 ; CI-NEXT:    v_lshlrev_b32_e32 v2, 16, v2
 ; CI-NEXT:    v_or_b32_e32 v0, v2, v0
 ; CI-NEXT:    buffer_store_dwordx2 v[0:1], off, s[0:3], 0
 ; CI-NEXT:    s_endpgm
 ;
 ; GFX9-ALIGNED-LABEL: read2_v2i32_align1_odd_offset:
 ; GFX9-ALIGNED:       ; %bb.0: ; %entry
 ; GFX9-ALIGNED-NEXT:    v_mov_b32_e32 v2, 0
-; GFX9-ALIGNED-NEXT:    ds_read_u8 v0, v2 offset:37025
-; GFX9-ALIGNED-NEXT:    ds_read_u8 v3, v2 offset:37026
-; GFX9-ALIGNED-NEXT:    ds_read_u8 v4, v2 offset:37027
-; GFX9-ALIGNED-NEXT:    ds_read_u8 v5, v2 offset:37028
-; GFX9-ALIGNED-NEXT:    ds_read_u8 v1, v2 offset:37029
-; GFX9-ALIGNED-NEXT:    ds_read_u8 v6, v2 offset:37030
-; GFX9-ALIGNED-NEXT:    ds_read_u8 v7, v2 offset:37031
-; GFX9-ALIGNED-NEXT:    ds_read_u8 v8, v2 offset:37032
+; GFX9-ALIGNED-NEXT:    ds_read_u8 v0, v2 offset:65
+; GFX9-ALIGNED-NEXT:    ds_read_u8 v3, v2 offset:66
+; GFX9-ALIGNED-NEXT:    ds_read_u8 v4, v2 offset:67
+; GFX9-ALIGNED-NEXT:    ds_read_u8 v5, v2 offset:68
+; GFX9-ALIGNED-NEXT:    ds_read_u8 v1, v2 offset:69
+; GFX9-ALIGNED-NEXT:    ds_read_u8 v6, v2 offset:70
+; GFX9-ALIGNED-NEXT:    ds_read_u8 v7, v2 offset:71
+; GFX9-ALIGNED-NEXT:    ds_read_u8 v8, v2 offset:72
 ; GFX9-ALIGNED-NEXT:    s_load_dwordx2 s[0:1], s[0:1], 0x24
 ; GFX9-ALIGNED-NEXT:    s_waitcnt lgkmcnt(0)
 ; GFX9-ALIGNED-NEXT:    v_lshlrev_b32_e32 v3, 8, v3
 ; GFX9-ALIGNED-NEXT:    v_lshlrev_b32_e32 v6, 8, v6
 ; GFX9-ALIGNED-NEXT:    v_or_b32_e32 v1, v6, v1
 ; GFX9-ALIGNED-NEXT:    v_lshlrev_b32_e32 v6, 8, v8
 ; GFX9-ALIGNED-NEXT:    v_or_b32_e32 v0, v3, v0
 ; GFX9-ALIGNED-NEXT:    v_lshlrev_b32_e32 v3, 8, v5
 ; GFX9-ALIGNED-NEXT:    v_or_b32_sdwa v6, v6, v7 dst_sel:WORD_1 dst_unused:UNUSED_PAD src0_sel:DWORD src1_sel:DWORD
 ; GFX9-ALIGNED-NEXT:    v_or_b32_sdwa v3, v3, v4 dst_sel:WORD_1 dst_unused:UNUSED_PAD src0_sel:DWORD src1_sel:DWORD
 ; GFX9-ALIGNED-NEXT:    v_or_b32_e32 v1, v6, v1
 ; GFX9-ALIGNED-NEXT:    v_or_b32_e32 v0, v3, v0
 ; GFX9-ALIGNED-NEXT:    global_store_dwordx2 v2, v[0:1], s[0:1]
 ; GFX9-ALIGNED-NEXT:    s_endpgm
 ;
 ; GFX9-UNALIGNED-LABEL: read2_v2i32_align1_odd_offset:
 ; GFX9-UNALIGNED:       ; %bb.0: ; %entry
-; GFX9-UNALIGNED-NEXT:    v_mov_b32_e32 v0, 0x90a1
+; GFX9-UNALIGNED-NEXT:    v_mov_b32_e32 v0, 0x41
 ; GFX9-UNALIGNED-NEXT:    s_load_dwordx2 s[0:1], s[0:1], 0x24
 ; GFX9-UNALIGNED-NEXT:    ds_read2_b32 v[0:1], v0 offset1:1
 ; GFX9-UNALIGNED-NEXT:    v_mov_b32_e32 v2, 0
 ; GFX9-UNALIGNED-NEXT:    s_waitcnt lgkmcnt(0)
 ; GFX9-UNALIGNED-NEXT:    global_store_dwordx2 v2, v[0:1], s[0:1]
 ; GFX9-UNALIGNED-NEXT:    s_endpgm
 entry:
   %load = load <2 x i32>, <2 x i32> addrspace(3)* bitcast (i8 addrspace(3)* getelementptr (i8, i8 addrspace(3)* bitcast ([100 x <2 x i32>] addrspace(3)* @v2i32_align1 to i8 addrspace(3)*), i32 65) to <2 x i32> addrspace(3)*), align 1

llvm/test/CodeGen/AMDGPU/ds_write2.ll

 @sgemm.lA = internal unnamed_addr addrspace(3) global [264 x float] undef, align 4
 @sgemm.lB = internal unnamed_addr addrspace(3) global [776 x float] undef, align 4
 
 define amdgpu_kernel void @write2_sgemm_sequence(float addrspace(1)* %C, i32 %lda, i32 %ldb, float addrspace(1)* %in) #0 {
 ; CI-LABEL: write2_sgemm_sequence:
 ; CI:       ; %bb.0:
 ; CI-NEXT:    s_load_dwordx2 s[0:1], s[0:1], 0xd
 ; CI-NEXT:    s_lshl_b32 s2, s2, 2
-; CI-NEXT:    s_add_i32 s3, s2, 0x8c40
+; CI-NEXT:    s_add_i32 s3, s2, 0xc20
 ; CI-NEXT:    v_mov_b32_e32 v0, s3
-; CI-NEXT:    s_add_i32 s2, s2, 0x8c80
+; CI-NEXT:    s_addk_i32 s2, 0xc60
 ; CI-NEXT:    s_waitcnt lgkmcnt(0)
 ; CI-NEXT:    s_load_dword s0, s[0:1], 0x0
 ; CI-NEXT:    s_mov_b32 m0, -1
 ; CI-NEXT:    s_waitcnt lgkmcnt(0)
 ; CI-NEXT:    v_mov_b32_e32 v2, s0
 ; CI-NEXT:    v_mov_b32_e32 v3, s0
 ; CI-NEXT:    ds_write2_b32 v0, v2, v3 offset1:1
 ; CI-NEXT:    v_mov_b32_e32 v0, s2
 ; CI-NEXT:    ds_write2_b32 v0, v2, v3 offset1:1
 ; CI-NEXT:    v_lshlrev_b32_e32 v0, 2, v1
-; CI-NEXT:    v_add_i32_e32 v1, vcc, 0x8020, v0
-; CI-NEXT:    ds_write2_b32 v1, v2, v3 offset1:1
-; CI-NEXT:    v_add_i32_e32 v1, vcc, 0x80a0, v0
-; CI-NEXT:    v_add_i32_e32 v0, vcc, 0x8120, v0
-; CI-NEXT:    ds_write2_b32 v1, v2, v3 offset1:1
 ; CI-NEXT:    ds_write2_b32 v0, v2, v3 offset1:1
+; CI-NEXT:    ds_write2_b32 v0, v2, v3 offset0:32 offset1:33
+; CI-NEXT:    ds_write2_b32 v0, v2, v3 offset0:64 offset1:65
 ; CI-NEXT:    s_endpgm
 ;
 ; GFX9-LABEL: write2_sgemm_sequence:
 ; GFX9:       ; %bb.0:
 ; GFX9-NEXT:    s_load_dwordx2 s[0:1], s[0:1], 0x34
 ; GFX9-NEXT:    s_lshl_b32 s2, s2, 2
-; GFX9-NEXT:    s_add_i32 s3, s2, 0x8c40
-; GFX9-NEXT:    s_add_i32 s2, s2, 0x8c80
+; GFX9-NEXT:    s_add_i32 s3, s2, 0xc20
+; GFX9-NEXT:    s_addk_i32 s2, 0xc60
 ; GFX9-NEXT:    v_mov_b32_e32 v0, s3
 ; GFX9-NEXT:    s_waitcnt lgkmcnt(0)
 ; GFX9-NEXT:    s_load_dword s0, s[0:1], 0x0
 ; GFX9-NEXT:    v_mov_b32_e32 v2, s2
 ; GFX9-NEXT:    s_waitcnt lgkmcnt(0)
 ; GFX9-NEXT:    v_mov_b32_e32 v3, s0
 ; GFX9-NEXT:    v_mov_b32_e32 v4, s0
 ; GFX9-NEXT:    ds_write2_b32 v0, v3, v4 offset1:1
 ; GFX9-NEXT:    ds_write2_b32 v2, v3, v4 offset1:1
 ; GFX9-NEXT:    v_lshlrev_b32_e32 v0, 2, v1
-; GFX9-NEXT:    v_add_u32_e32 v1, 0x8020, v0
-; GFX9-NEXT:    ds_write2_b32 v1, v3, v4 offset1:1
-; GFX9-NEXT:    v_add_u32_e32 v1, 0x80a0, v0
-; GFX9-NEXT:    v_add_u32_e32 v0, 0x8120, v0
-; GFX9-NEXT:    ds_write2_b32 v1, v3, v4 offset1:1
 ; GFX9-NEXT:    ds_write2_b32 v0, v3, v4 offset1:1
+; GFX9-NEXT:    ds_write2_b32 v0, v3, v4 offset0:32 offset1:33
+; GFX9-NEXT:    ds_write2_b32 v0, v3, v4 offset0:64 offset1:65
 ; GFX9-NEXT:    s_endpgm
   %x.i = tail call i32 @llvm.amdgcn.workgroup.id.x() #1
   %y.i = tail call i32 @llvm.amdgcn.workitem.id.y() #1
   %val = load float, float addrspace(1)* %in
   %arrayidx44 = getelementptr inbounds [264 x float], [264 x float] addrspace(3)* @sgemm.lA, i32 0, i32 %x.i
   store float %val, float addrspace(3)* %arrayidx44, align 4
   %add47 = add nsw i32 %x.i, 1
   %arrayidx48 = getelementptr inbounds [264 x float], [264 x float] addrspace(3)* @sgemm.lA, i32 0, i32 %add47

 @v2i32_align1 = internal addrspace(3) global [100 x <2 x i32>] undef, align 1
 
 define amdgpu_kernel void @write2_v2i32_align1_odd_offset() {
 ; CI-LABEL: write2_v2i32_align1_odd_offset:
 ; CI:       ; %bb.0: ; %entry
 ; CI-NEXT:    v_mov_b32_e32 v0, 0x7b
 ; CI-NEXT:    v_mov_b32_e32 v1, 0
 ; CI-NEXT:    s_mov_b32 m0, -1
-; CI-NEXT:    ds_write_b8 v1, v0 offset:37025
+; CI-NEXT:    ds_write_b8 v1, v0 offset:65
 ; CI-NEXT:    v_mov_b32_e32 v0, 1
-; CI-NEXT:    ds_write_b8 v1, v0 offset:37030
+; CI-NEXT:    ds_write_b8 v1, v0 offset:70
 ; CI-NEXT:    v_mov_b32_e32 v0, 0xc8
-; CI-NEXT:    ds_write_b8 v1, v0 offset:37029
-; CI-NEXT:    ds_write_b8 v1, v1 offset:37028
-; CI-NEXT:    ds_write_b8 v1, v1 offset:37027
-; CI-NEXT:    ds_write_b8 v1, v1 offset:37026
-; CI-NEXT:    ds_write_b8 v1, v1 offset:37032
-; CI-NEXT:    ds_write_b8 v1, v1 offset:37031
+; CI-NEXT:    ds_write_b8 v1, v0 offset:69
+; CI-NEXT:    ds_write_b8 v1, v1 offset:68
+; CI-NEXT:    ds_write_b8 v1, v1 offset:67
+; CI-NEXT:    ds_write_b8 v1, v1 offset:66
+; CI-NEXT:    ds_write_b8 v1, v1 offset:72
+; CI-NEXT:    ds_write_b8 v1, v1 offset:71
 ; CI-NEXT:    s_endpgm
 ;
 ; GFX9-ALIGNED-LABEL: write2_v2i32_align1_odd_offset:
 ; GFX9-ALIGNED:       ; %bb.0: ; %entry
 ; GFX9-ALIGNED-NEXT:    v_mov_b32_e32 v0, 0x7b
 ; GFX9-ALIGNED-NEXT:    v_mov_b32_e32 v1, 0
-; GFX9-ALIGNED-NEXT:    ds_write_b8 v1, v0 offset:37025
+; GFX9-ALIGNED-NEXT:    ds_write_b8 v1, v0 offset:65
 ; GFX9-ALIGNED-NEXT:    v_mov_b32_e32 v0, 1
-; GFX9-ALIGNED-NEXT:    ds_write_b8 v1, v0 offset:37030
+; GFX9-ALIGNED-NEXT:    ds_write_b8 v1, v0 offset:70
 ; GFX9-ALIGNED-NEXT:    v_mov_b32_e32 v0, 0xc8
-; GFX9-ALIGNED-NEXT:    ds_write_b8 v1, v0 offset:37029
-; GFX9-ALIGNED-NEXT:    ds_write_b8 v1, v1 offset:37028
-; GFX9-ALIGNED-NEXT:    ds_write_b8 v1, v1 offset:37027
-; GFX9-ALIGNED-NEXT:    ds_write_b8 v1, v1 offset:37026
-; GFX9-ALIGNED-NEXT:    ds_write_b8 v1, v1 offset:37032
-; GFX9-ALIGNED-NEXT:    ds_write_b8 v1, v1 offset:37031
+; GFX9-ALIGNED-NEXT:    ds_write_b8 v1, v0 offset:69
+; GFX9-ALIGNED-NEXT:    ds_write_b8 v1, v1 offset:68
+; GFX9-ALIGNED-NEXT:    ds_write_b8 v1, v1 offset:67
+; GFX9-ALIGNED-NEXT:    ds_write_b8 v1, v1 offset:66
+; GFX9-ALIGNED-NEXT:    ds_write_b8 v1, v1 offset:72
+; GFX9-ALIGNED-NEXT:    ds_write_b8 v1, v1 offset:71
 ; GFX9-ALIGNED-NEXT:    s_endpgm
 ;
 ; GFX9-UNALIGNED-LABEL: write2_v2i32_align1_odd_offset:
 ; GFX9-UNALIGNED:       ; %bb.0: ; %entry
-; GFX9-UNALIGNED-NEXT:    v_mov_b32_e32 v0, 0x90a1
+; GFX9-UNALIGNED-NEXT:    v_mov_b32_e32 v0, 0x41
 ; GFX9-UNALIGNED-NEXT:    v_mov_b32_e32 v1, 0x7b
 ; GFX9-UNALIGNED-NEXT:    v_mov_b32_e32 v2, 0x1c8
 ; GFX9-UNALIGNED-NEXT:    ds_write2_b32 v0, v1, v2 offset1:1
 ; GFX9-UNALIGNED-NEXT:    s_endpgm
 entry:
   store <2 x i32> <i32 123, i32 456>, <2 x i32> addrspace(3)* bitcast (i8 addrspace(3)* getelementptr (i8, i8 addrspace(3)* bitcast ([100 x <2 x i32>] addrspace(3)* @v2i32_align1 to i8 addrspace(3)*), i32 65) to <2 x i32> addrspace(3)*), align 1
   ret void
 }

llvm/test/CodeGen/AMDGPU/lower-kernel-lds-constexpr.ll

 ; RUN: opt -S -mtriple=amdgcn-- -amdgpu-lower-module-lds < %s | FileCheck %s
 ; RUN: opt -S -mtriple=amdgcn-- -passes=amdgpu-lower-module-lds < %s | FileCheck %s
 
 ; CHECK: %llvm.amdgcn.kernel.k2.lds.t = type { i32 }
+; CHECK-NOT: %llvm.amdgcn.kernel.k4.lds.t
 
 @lds.1 = internal unnamed_addr addrspace(3) global [2 x i8] undef, align 1
 
 ; Use constant from different kernels
 ;.
 ; CHECK: @lds.1 = internal unnamed_addr addrspace(3) global [2 x i8] undef, align 1
 ; CHECK: @llvm.amdgcn.kernel.k2.lds = internal addrspace(3) global %llvm.amdgcn.kernel.k2.lds.t undef, align 4
 ;.

 ; CHECK-NEXT:    store i64 2, i64* %ptr2, align 1
 ;
   %ptr1 = addrspacecast i64 addrspace(3)* bitcast (i8 addrspace(3)* getelementptr inbounds ([32 x i8], [32 x i8] addrspace(3)* @lds.3, i32 0, i32 16) to i64 addrspace(3)*) to i64*
   store i64 1, i64* %ptr1, align 1
   %ptr2 = addrspacecast i64 addrspace(3)* bitcast (i8 addrspace(3)* getelementptr inbounds ([32 x i8], [32 x i8] addrspace(3)* @lds.3, i32 0, i32 24) to i64 addrspace(3)*) to i64*
   store i64 2, i64* %ptr2, align 1
   ret void
 }
+
+; @lds.1 is used from constant expressions in different kernels.
+; Make sure we do not create a structure for it as we cannot handle it yet.
+define amdgpu_kernel void @k4(i64 %x) {
+; CHECK-LABEL: @k4(
+; CHECK-NEXT:    %ptr = getelementptr inbounds i8, i8* addrspacecast (i8 addrspace(3)* getelementptr inbounds ([2 x i8], [2 x i8] addrspace(3)* @lds.1, i32 0, i32 0) to i8*), i64 %x
+; CHECK-NEXT:    store i8 1, i8* %ptr, align 1
+; CHECK-NEXT:    ret void
+;
+  %ptr = getelementptr inbounds i8, i8* addrspacecast ([2 x i8] addrspace(3)* @lds.1 to i8*), i64 %x
+  store i8 1, i8 addrspace(0)* %ptr, align 1
+  ret void
+}