# RUN: llc -mtriple=amdgcn -mcpu=gfx1010 -verify-machineinstrs -run-pass si-load-store-opt -o - %s | FileCheck %s -check-prefixes=CHECK,GFX10
# RUN: llc -mtriple=amdgcn -mcpu=gfx1200 -verify-machineinstrs -run-pass si-load-store-opt -o - %s | FileCheck %s -check-prefixes=CHECK,GFX12

# CHECK-LABEL: name: merge_s_buffer_load_x2
# CHECK: S_BUFFER_LOAD_DWORDX2_IMM %0, 0, 0 :: (dereferenceable invariant load (s64), align 4)
name: merge_s_buffer_load_x2
tracksRegLiveness: true
body:             |
  bb.0:
    liveins: $sgpr0_sgpr1_sgpr2_sgpr3

    %0:sgpr_128 = COPY $sgpr0_sgpr1_sgpr2_sgpr3
    %1:sreg_32_xm0_xexec = S_BUFFER_LOAD_DWORD_IMM %0:sgpr_128, 0, 0 :: (dereferenceable invariant load (s32))
    %2:sreg_32_xm0_xexec = S_BUFFER_LOAD_DWORD_IMM %0:sgpr_128, 4, 0 :: (dereferenceable invariant load (s32))

    S_ENDPGM 0
...
---

# CHECK-LABEL: name: merge_s_buffer_load_x1_x2
# CHECK: S_BUFFER_LOAD_DWORD_IMM %0, 0, 0 :: (dereferenceable invariant load (s32))
# CHECK: S_BUFFER_LOAD_DWORDX2_IMM %0, 4, 0 :: (dereferenceable invariant load (s64))
name: merge_s_buffer_load_x1_x2
tracksRegLiveness: true
body:             |
  bb.0:
    liveins: $sgpr0_sgpr1_sgpr2_sgpr3

    %0:sgpr_128 = COPY $sgpr0_sgpr1_sgpr2_sgpr3
    %1:sreg_32_xm0_xexec = S_BUFFER_LOAD_DWORD_IMM %0:sgpr_128, 0, 0 :: (dereferenceable invariant load (s32))
    %2:sgpr_64 = S_BUFFER_LOAD_DWORDX2_IMM %0:sgpr_128, 4, 0 :: (dereferenceable invariant load (s64))

    S_ENDPGM 0
...
---

# CHECK-LABEL: name: merge_s_buffer_load_x2_x1
# GFX10: S_BUFFER_LOAD_DWORDX2_IMM %0, 0, 0 :: (dereferenceable invariant load (s64))
# GFX10: S_BUFFER_LOAD_DWORD_IMM %0, 8, 0 :: (dereferenceable invariant load (s32))
# GFX12: S_BUFFER_LOAD_DWORDX3_IMM %0, 0, 0 :: (dereferenceable invariant load (s96), align 8)
name: merge_s_buffer_load_x2_x1
tracksRegLiveness: true
body:             |
  bb.0:
    liveins: $sgpr0_sgpr1_sgpr2_sgpr3

    %0:sgpr_128 = COPY $sgpr0_sgpr1_sgpr2_sgpr3
    %1:sgpr_64 = S_BUFFER_LOAD_DWORDX2_IMM %0:sgpr_128, 0, 0 :: (dereferenceable invariant load (s64))
    %2:sreg_32_xm0_xexec = S_BUFFER_LOAD_DWORD_IMM %0:sgpr_128, 8, 0 :: (dereferenceable invariant load (s32))

    S_ENDPGM 0
...
---

# CHECK-LABEL: name: merge_s_buffer_load_x4
# CHECK: S_BUFFER_LOAD_DWORDX4_IMM %0, 0, 0 :: (dereferenceable invariant load (s128), align 4)
name: merge_s_buffer_load_x4
tracksRegLiveness: true
body:             |
  bb.0:
    liveins: $sgpr0_sgpr1_sgpr2_sgpr3

    %0:sgpr_128 = COPY $sgpr0_sgpr1_sgpr2_sgpr3
    %1:sreg_32_xm0_xexec = S_BUFFER_LOAD_DWORD_IMM %0:sgpr_128, 0, 0 :: (dereferenceable invariant load (s32))
    %2:sreg_32_xm0_xexec = S_BUFFER_LOAD_DWORD_IMM %0:sgpr_128, 4, 0 :: (dereferenceable invariant load (s32))
    %3:sreg_32_xm0_xexec = S_BUFFER_LOAD_DWORD_IMM %0:sgpr_128, 8, 0 :: (dereferenceable invariant load (s32))
    %4:sreg_32_xm0_xexec = S_BUFFER_LOAD_DWORD_IMM %0:sgpr_128, 12, 0 :: (dereferenceable invariant load (s32))

    S_ENDPGM 0
...
---

# CHECK-LABEL: name: merge_s_buffer_load_x1_x3
# CHECK: S_BUFFER_LOAD_DWORD_IMM %0, 0, 0 :: (dereferenceable invariant load (s32))
# CHECK: S_BUFFER_LOAD_DWORDX3_IMM %0, 4, 0 :: (dereferenceable invariant load (s96), align 16)
name: merge_s_buffer_load_x1_x3
tracksRegLiveness: true
body:             |
  bb.0:
    liveins: $sgpr0_sgpr1_sgpr2_sgpr3

    %0:sgpr_128 = COPY $sgpr0_sgpr1_sgpr2_sgpr3
    %1:sreg_32_xm0_xexec = S_BUFFER_LOAD_DWORD_IMM %0:sgpr_128, 0, 0 :: (dereferenceable invariant load (s32))
    %2:sgpr_96 = S_BUFFER_LOAD_DWORDX3_IMM %0:sgpr_128, 4, 0 :: (dereferenceable invariant load (s96))

    S_ENDPGM 0
...
---

# CHECK-LABEL: name: merge_s_buffer_load_x3_x1
# CHECK: S_BUFFER_LOAD_DWORDX4_IMM %0, 0, 0 :: (dereferenceable invariant load (s128))
name: merge_s_buffer_load_x3_x1
tracksRegLiveness: true
body:             |
  bb.0:
    liveins: $sgpr0_sgpr1_sgpr2_sgpr3

    %0:sgpr_128 = COPY $sgpr0_sgpr1_sgpr2_sgpr3
    %1:sgpr_96 = S_BUFFER_LOAD_DWORDX3_IMM %0:sgpr_128, 0, 0 :: (dereferenceable invariant load (s96))
    %2:sreg_32_xm0_xexec = S_BUFFER_LOAD_DWORD_IMM %0:sgpr_128, 12, 0 :: (dereferenceable invariant load (s32))

    S_ENDPGM 0
...
---

# CHECK-LABEL: name: merge_s_buffer_load_x8
# CHECK: S_BUFFER_LOAD_DWORDX8_IMM %0, 0, 0 :: (dereferenceable invariant load (s256), align 4)
name: merge_s_buffer_load_x8
tracksRegLiveness: true
body:             |
  bb.0:
    liveins: $sgpr0_sgpr1_sgpr2_sgpr3

    %0:sgpr_128 = COPY $sgpr0_sgpr1_sgpr2_sgpr3
    %1:sreg_32_xm0_xexec = S_BUFFER_LOAD_DWORD_IMM %0:sgpr_128, 0, 0 :: (dereferenceable invariant load (s32))
    %2:sreg_32_xm0_xexec = S_BUFFER_LOAD_DWORD_IMM %0:sgpr_128, 4, 0 :: (dereferenceable invariant load (s32))
    %3:sreg_32_xm0_xexec = S_BUFFER_LOAD_DWORD_IMM %0:sgpr_128, 8, 0 :: (dereferenceable invariant load (s32))
    %4:sreg_32_xm0_xexec = S_BUFFER_LOAD_DWORD_IMM %0:sgpr_128, 12, 0 :: (dereferenceable invariant load (s32))
    %5:sreg_32_xm0_xexec = S_BUFFER_LOAD_DWORD_IMM %0:sgpr_128, 16, 0 :: (dereferenceable invariant load (s32))
    %6:sreg_32_xm0_xexec = S_BUFFER_LOAD_DWORD_IMM %0:sgpr_128, 20, 0 :: (dereferenceable invariant load (s32))
    %7:sreg_32_xm0_xexec = S_BUFFER_LOAD_DWORD_IMM %0:sgpr_128, 24, 0 :: (dereferenceable invariant load (s32))
    %8:sreg_32_xm0_xexec = S_BUFFER_LOAD_DWORD_IMM %0:sgpr_128, 28, 0 :: (dereferenceable invariant load (s32))

    S_ENDPGM 0
...
---

# CHECK-LABEL: name: merge_s_buffer_load_x8_reordered
# CHECK: S_BUFFER_LOAD_DWORDX8_IMM %0, 0, 0 :: (dereferenceable invariant load (s256), align 4)
name: merge_s_buffer_load_x8_reordered
tracksRegLiveness: true
body:             |
  bb.0:
    liveins: $sgpr0_sgpr1_sgpr2_sgpr3

    %0:sgpr_128 = COPY $sgpr0_sgpr1_sgpr2_sgpr3
    %1:sreg_32_xm0_xexec = S_BUFFER_LOAD_DWORD_IMM %0:sgpr_128, 20, 0 :: (dereferenceable invariant load (s32))
    %2:sreg_32_xm0_xexec = S_BUFFER_LOAD_DWORD_IMM %0:sgpr_128, 4, 0 :: (dereferenceable invariant load (s32))
    %3:sreg_32_xm0_xexec = S_BUFFER_LOAD_DWORD_IMM %0:sgpr_128, 0, 0 :: (dereferenceable invariant load (s32))
    %4:sreg_32_xm0_xexec = S_BUFFER_LOAD_DWORD_IMM %0:sgpr_128, 28, 0 :: (dereferenceable invariant load (s32))
    %5:sreg_32_xm0_xexec = S_BUFFER_LOAD_DWORD_IMM %0:sgpr_128, 12, 0 :: (dereferenceable invariant load (s32))
    %6:sreg_32_xm0_xexec = S_BUFFER_LOAD_DWORD_IMM %0:sgpr_128, 16, 0 :: (dereferenceable invariant load (s32))
    %7:sreg_32_xm0_xexec = S_BUFFER_LOAD_DWORD_IMM %0:sgpr_128, 8, 0 :: (dereferenceable invariant load (s32))
    %8:sreg_32_xm0_xexec = S_BUFFER_LOAD_DWORD_IMM %0:sgpr_128, 24, 0 :: (dereferenceable invariant load (s32))

    S_ENDPGM 0
...
---

# CHECK-LABEL: name: merge_s_buffer_load_x8_out_of_x2
# CHECK: S_BUFFER_LOAD_DWORDX8_IMM %0, 0, 0 :: (dereferenceable invariant load (s256), align 8)
name: merge_s_buffer_load_x8_out_of_x2
tracksRegLiveness: true
body:             |
  bb.0:
    liveins: $sgpr0_sgpr1_sgpr2_sgpr3

    %0:sgpr_128 = COPY $sgpr0_sgpr1_sgpr2_sgpr3
    %1:sgpr_64 = S_BUFFER_LOAD_DWORDX2_IMM %0:sgpr_128, 16, 0 :: (dereferenceable invariant load (s64))
    %2:sgpr_64 = S_BUFFER_LOAD_DWORDX2_IMM %0:sgpr_128, 8, 0 :: (dereferenceable invariant load (s64))
    %3:sgpr_64 = S_BUFFER_LOAD_DWORDX2_IMM %0:sgpr_128, 0, 0 :: (dereferenceable invariant load (s64))
    %4:sgpr_64 = S_BUFFER_LOAD_DWORDX2_IMM %0:sgpr_128, 24, 0 :: (dereferenceable invariant load (s64))

    S_ENDPGM 0
...
---

# CHECK-LABEL: name: merge_s_buffer_load_x8_out_of_x4
# CHECK: S_BUFFER_LOAD_DWORDX8_IMM %0, 0, 0 :: (dereferenceable invariant load (s256), align 16)
name: merge_s_buffer_load_x8_out_of_x4
tracksRegLiveness: true
body:             |
  bb.0:
    liveins: $sgpr0_sgpr1_sgpr2_sgpr3

    %0:sgpr_128 = COPY $sgpr0_sgpr1_sgpr2_sgpr3
    %1:sgpr_128 = S_BUFFER_LOAD_DWORDX4_IMM %0:sgpr_128, 0, 0 :: (dereferenceable invariant load (s128))
    %2:sgpr_128 = S_BUFFER_LOAD_DWORDX4_IMM %0:sgpr_128, 16, 0 :: (dereferenceable invariant load (s128))

    S_ENDPGM 0
...
---

# CHECK-LABEL: name: merge_s_buffer_load_x8_mixed
# CHECK: S_BUFFER_LOAD_DWORDX8_IMM %0, 0, 0 :: (dereferenceable invariant load (s256), align 16)
name: merge_s_buffer_load_x8_mixed
tracksRegLiveness: true
body:             |
  bb.0:
    liveins: $sgpr0_sgpr1_sgpr2_sgpr3

    %0:sgpr_128 = COPY $sgpr0_sgpr1_sgpr2_sgpr3
    %1:sgpr_128 = S_BUFFER_LOAD_DWORDX4_IMM %0:sgpr_128, 0, 0 :: (dereferenceable invariant load (s128))
    %2:sreg_32_xm0_xexec = S_BUFFER_LOAD_DWORD_IMM %0:sgpr_128, 16, 0 :: (dereferenceable invariant load (s32))
    %3:sgpr_64 = S_BUFFER_LOAD_DWORDX2_IMM %0:sgpr_128, 24, 0 :: (dereferenceable invariant load (s64))
    %4:sreg_32_xm0_xexec = S_BUFFER_LOAD_DWORD_IMM %0:sgpr_128, 20, 0 :: (dereferenceable invariant load (s32))

    S_ENDPGM 0
...
---

# CHECK-LABEL: name: merge_s_buffer_load_sgpr_imm
# CHECK: S_BUFFER_LOAD_DWORDX4_SGPR_IMM %0, %1, 0, 0 :: (dereferenceable invariant load (s128), align 4)
name: merge_s_buffer_load_sgpr_imm
tracksRegLiveness: true
body:             |
  bb.0:
    liveins: $sgpr0_sgpr1_sgpr2_sgpr3, $sgpr4

    %0:sgpr_128 = COPY $sgpr0_sgpr1_sgpr2_sgpr3
    %1:sreg_32 = COPY $sgpr4
    %2:sreg_32_xm0_xexec = S_BUFFER_LOAD_DWORD_SGPR_IMM %0:sgpr_128, %1:sreg_32, 0, 0 :: (dereferenceable invariant load (s32))
    %3:sreg_32_xm0_xexec = S_BUFFER_LOAD_DWORD_SGPR_IMM %0:sgpr_128, %1:sreg_32, 4, 0 :: (dereferenceable invariant load (s32))
    %4:sreg_32_xm0_xexec = S_BUFFER_LOAD_DWORD_SGPR_IMM %0:sgpr_128, %1:sreg_32, 8, 0 :: (dereferenceable invariant load (s32))
    %5:sreg_32_xm0_xexec = S_BUFFER_LOAD_DWORD_SGPR_IMM %0:sgpr_128, %1:sreg_32, 12, 0 :: (dereferenceable invariant load (s32))

    S_ENDPGM 0
...
---

# CHECK-LABEL: name: no_merge_for_different_soffsets
# CHECK: S_BUFFER_LOAD_DWORD_SGPR_IMM %0, %1, 4, 0 :: (dereferenceable invariant load (s32))
# CHECK: S_BUFFER_LOAD_DWORD_SGPR_IMM %0, %2, 8, 0 :: (dereferenceable invariant load (s32))
name: no_merge_for_different_soffsets
tracksRegLiveness: true
body:             |
  bb.0:
    liveins: $sgpr0_sgpr1_sgpr2_sgpr3, $sgpr4, $sgpr5

    %0:sgpr_128 = COPY $sgpr0_sgpr1_sgpr2_sgpr3
    %1:sreg_32 = COPY $sgpr4
    %2:sreg_32 = COPY $sgpr5
    %3:sreg_32_xm0_xexec = S_BUFFER_LOAD_DWORD_SGPR_IMM %0:sgpr_128, %1:sreg_32, 4, 0 :: (dereferenceable invariant load (s32))
    %4:sreg_32_xm0_xexec = S_BUFFER_LOAD_DWORD_SGPR_IMM %0:sgpr_128, %2:sreg_32, 8, 0 :: (dereferenceable invariant load (s32))

    S_ENDPGM 0
...
---

# CHECK-LABEL: name: no_merge_for_non_adjacent_offsets
# CHECK: S_BUFFER_LOAD_DWORD_SGPR_IMM %0, %1, 4, 0 :: (dereferenceable invariant load (s32))
# CHECK: S_BUFFER_LOAD_DWORD_SGPR_IMM %0, %1, 12, 0 :: (dereferenceable invariant load (s32))
name: no_merge_for_non_adjacent_offsets
tracksRegLiveness: true
body:             |
  bb.0:
    liveins: $sgpr0_sgpr1_sgpr2_sgpr3, $sgpr4

    %0:sgpr_128 = COPY $sgpr0_sgpr1_sgpr2_sgpr3
    %1:sreg_32 = COPY $sgpr4
    %2:sreg_32_xm0_xexec = S_BUFFER_LOAD_DWORD_SGPR_IMM %0:sgpr_128, %1:sreg_32, 4, 0 :: (dereferenceable invariant load (s32))
    %3:sreg_32_xm0_xexec = S_BUFFER_LOAD_DWORD_SGPR_IMM %0:sgpr_128, %1:sreg_32, 12, 0 :: (dereferenceable invariant load (s32))

    S_ENDPGM 0
...
---