[PATCH] R600: Call EmitFunctionHeader() in the AsmPrinter to populate the ELF symbol table
Tom Stellard
thomas.stellard at amd.com
Fri Sep 26 14:07:05 PDT 2014
On 09/26/2014 04:58 PM, Matt Arsenault wrote:
>
> On Sep 26, 2014, at 12:17 PM, Tom Stellard <thomas.stellard at amd.com> wrote:
>
>> ---
>> lib/Target/R600/AMDGPUAsmPrinter.cpp | 2 +-
>> test/CodeGen/R600/128bit-kernel-args.ll | 8 +-
>> test/CodeGen/R600/32-bit-local-address-space.ll | 24 ++---
>> test/CodeGen/R600/64bit-kernel-args.ll | 2 +-
>> test/CodeGen/R600/add.ll | 16 +--
>> test/CodeGen/R600/add_i64.ll | 12 +--
>> test/CodeGen/R600/address-space.ll | 2 +-
>> test/CodeGen/R600/and.ll | 28 ++---
>> test/CodeGen/R600/anyext.ll | 2 +-
>> test/CodeGen/R600/array-ptr-calc-i32.ll | 2 +-
>> test/CodeGen/R600/array-ptr-calc-i64.ll | 2 +-
>> test/CodeGen/R600/atomic_cmp_swap_local.ll | 10 +-
>> test/CodeGen/R600/atomic_load_add.ll | 8 +-
>> test/CodeGen/R600/atomic_load_sub.ll | 8 +-
>> test/CodeGen/R600/basic-branch.ll | 2 +-
>> test/CodeGen/R600/basic-loop.ll | 2 +-
>> test/CodeGen/R600/bfe_uint.ll | 4 +-
>> test/CodeGen/R600/bfi_int.ll | 6 +-
>> test/CodeGen/R600/bitcast.ll | 8 +-
>> test/CodeGen/R600/build_vector.ll | 8 +-
>> test/CodeGen/R600/call_fs.ll | 4 +-
>> test/CodeGen/R600/cayman-loop-bug.ll | 2 +-
>> test/CodeGen/R600/cf-stack-bug.ll | 8 +-
>> test/CodeGen/R600/codegen-prepare-addrmode-sext.ll | 2 +-
>> test/CodeGen/R600/combine_vloads.ll | 2 +-
>> test/CodeGen/R600/complex-folding.ll | 2 +-
>> test/CodeGen/R600/concat_vectors.ll | 62 +++++------
>> test/CodeGen/R600/copy-illegal-type.ll | 18 ++--
>> test/CodeGen/R600/ctlz_zero_undef.ll | 8 +-
>> test/CodeGen/R600/ctpop.ll | 30 +++---
>> test/CodeGen/R600/ctpop64.ll | 14 +--
>> test/CodeGen/R600/cttz_zero_undef.ll | 8 +-
>> test/CodeGen/R600/cvt_f32_ubyte.ll | 18 ++--
>> .../R600/dagcombiner-bug-illegal-vec4-int-to-fp.ll | 4 +-
>> test/CodeGen/R600/default-fp-mode.ll | 2 +-
>> .../CodeGen/R600/disconnected-predset-break-bug.ll | 2 +-
>> test/CodeGen/R600/dot4-folding.ll | 2 +-
>> .../ds-negative-offset-addressing-mode-loop.ll | 2 +-
>> test/CodeGen/R600/extload.ll | 20 ++--
>> test/CodeGen/R600/extract_vector_elt_i16.ll | 4 +-
>> test/CodeGen/R600/fabs.f64.ll | 16 +--
>> test/CodeGen/R600/fabs.ll | 14 +--
>> test/CodeGen/R600/fadd.ll | 8 +-
>> test/CodeGen/R600/fadd64.ll | 2 +-
>> test/CodeGen/R600/fceil.ll | 12 +--
>> test/CodeGen/R600/fceil64.ll | 12 +--
>> test/CodeGen/R600/fcmp.ll | 4 +-
>> test/CodeGen/R600/fcmp64.ll | 12 +--
>> test/CodeGen/R600/fconst64.ll | 2 +-
>> test/CodeGen/R600/fcopysign.f32.ll | 6 +-
>> test/CodeGen/R600/fcopysign.f64.ll | 6 +-
>> test/CodeGen/R600/fdiv.ll | 6 +-
>> test/CodeGen/R600/fdiv64.ll | 2 +-
>> test/CodeGen/R600/fetch-limits.r600.ll | 2 +-
>> test/CodeGen/R600/fetch-limits.r700+.ll | 2 +-
>> test/CodeGen/R600/ffloor.ll | 12 +--
>> test/CodeGen/R600/flat-address-space.ll | 14 +--
>> test/CodeGen/R600/fma.f64.ll | 6 +-
>> test/CodeGen/R600/fma.ll | 6 +-
>> test/CodeGen/R600/fmul.ll | 10 +-
>> test/CodeGen/R600/fmul64.ll | 6 +-
>> test/CodeGen/R600/fmuladd.ll | 4 +-
>> test/CodeGen/R600/fneg-fabs.f64.ll | 14 +--
>> test/CodeGen/R600/fneg-fabs.ll | 16 +--
>> test/CodeGen/R600/fneg.f64.ll | 10 +-
>> test/CodeGen/R600/fneg.ll | 10 +-
>> test/CodeGen/R600/fp16_to_fp.ll | 4 +-
>> test/CodeGen/R600/fp32_to_fp16.ll | 2 +-
>> test/CodeGen/R600/fp64_to_sint.ll | 6 +-
>> test/CodeGen/R600/fp_to_sint.ll | 12 +--
>> test/CodeGen/R600/fp_to_uint.f64.ll | 2 +-
>> test/CodeGen/R600/fp_to_uint.ll | 12 +--
>> test/CodeGen/R600/fpext.ll | 2 +-
>> test/CodeGen/R600/fptrunc.ll | 2 +-
>> test/CodeGen/R600/frem.ll | 8 +-
>> test/CodeGen/R600/fsqrt.ll | 4 +-
>> test/CodeGen/R600/fsub.ll | 10 +-
>> test/CodeGen/R600/fsub64.ll | 2 +-
>> test/CodeGen/R600/ftrunc.ll | 12 +--
>> test/CodeGen/R600/gep-address-space.ll | 8 +-
>> test/CodeGen/R600/global_atomics.ll | 8 +-
>> test/CodeGen/R600/gv-const-addrspace-fail.ll | 8 +-
>> test/CodeGen/R600/gv-const-addrspace.ll | 8 +-
>> test/CodeGen/R600/half.ll | 12 +--
>> test/CodeGen/R600/icmp64.ll | 20 ++--
>> test/CodeGen/R600/imm.ll | 42 ++++----
>> test/CodeGen/R600/indirect-private-64.ll | 8 +-
>> test/CodeGen/R600/infinite-loop.ll | 2 +-
>> test/CodeGen/R600/input-mods.ll | 4 +-
>> test/CodeGen/R600/insert_vector_elt.ll | 48 ++++-----
>> test/CodeGen/R600/kcache-fold.ll | 4 +-
>> test/CodeGen/R600/kernel-args.ll | 116 ++++++++++-----------
>> test/CodeGen/R600/lds-oqap-crash.ll | 2 +-
>> test/CodeGen/R600/lds-output-queue.ll | 4 +-
>> test/CodeGen/R600/lds-size.ll | 2 +-
>> test/CodeGen/R600/legalizedag-bug-expand-setcc.ll | 2 +-
>> test/CodeGen/R600/literals.ll | 8 +-
>> test/CodeGen/R600/llvm.AMDGPU.abs.ll | 6 +-
>> test/CodeGen/R600/llvm.AMDGPU.barrier.global.ll | 2 +-
>> test/CodeGen/R600/llvm.AMDGPU.barrier.local.ll | 2 +-
>> test/CodeGen/R600/llvm.AMDGPU.bfe.i32.ll | 72 ++++++-------
>> test/CodeGen/R600/llvm.AMDGPU.bfe.u32.ll | 88 ++++++++--------
>> test/CodeGen/R600/llvm.AMDGPU.bfi.ll | 8 +-
>> test/CodeGen/R600/llvm.AMDGPU.bfm.ll | 8 +-
>> test/CodeGen/R600/llvm.AMDGPU.brev.ll | 4 +-
>> test/CodeGen/R600/llvm.AMDGPU.clamp.ll | 4 +-
>> test/CodeGen/R600/llvm.AMDGPU.cube.ll | 2 +-
>> test/CodeGen/R600/llvm.AMDGPU.cvt_f32_ubyte.ll | 8 +-
>> test/CodeGen/R600/llvm.AMDGPU.div_fixup.ll | 4 +-
>> test/CodeGen/R600/llvm.AMDGPU.div_fmas.ll | 4 +-
>> test/CodeGen/R600/llvm.AMDGPU.fract.ll | 4 +-
>> test/CodeGen/R600/llvm.AMDGPU.imad24.ll | 2 +-
>> test/CodeGen/R600/llvm.AMDGPU.imax.ll | 4 +-
>> test/CodeGen/R600/llvm.AMDGPU.imin.ll | 4 +-
>> test/CodeGen/R600/llvm.AMDGPU.imul24.ll | 2 +-
>> test/CodeGen/R600/llvm.AMDGPU.kill.ll | 2 +-
>> test/CodeGen/R600/llvm.AMDGPU.ldexp.ll | 4 +-
>> test/CodeGen/R600/llvm.AMDGPU.legacy.rsq.ll | 2 +-
>> test/CodeGen/R600/llvm.AMDGPU.rcp.f64.ll | 6 +-
>> test/CodeGen/R600/llvm.AMDGPU.rcp.ll | 6 +-
>> test/CodeGen/R600/llvm.AMDGPU.rsq.clamped.f64.ll | 2 +-
>> test/CodeGen/R600/llvm.AMDGPU.rsq.clamped.ll | 2 +-
>> test/CodeGen/R600/llvm.AMDGPU.rsq.ll | 6 +-
>> test/CodeGen/R600/llvm.AMDGPU.trig_preop.ll | 4 +-
>> test/CodeGen/R600/llvm.AMDGPU.trunc.ll | 4 +-
>> test/CodeGen/R600/llvm.AMDGPU.umad24.ll | 2 +-
>> test/CodeGen/R600/llvm.AMDGPU.umax.ll | 6 +-
>> test/CodeGen/R600/llvm.AMDGPU.umin.ll | 6 +-
>> test/CodeGen/R600/llvm.AMDGPU.umul24.ll | 2 +-
>> test/CodeGen/R600/llvm.SI.gather4.ll | 70 ++++++-------
>> test/CodeGen/R600/llvm.SI.getlod.ll | 6 +-
>> test/CodeGen/R600/llvm.SI.image.ll | 6 +-
>> test/CodeGen/R600/llvm.SI.image.sample.ll | 40 +++----
>> test/CodeGen/R600/llvm.SI.image.sample.o.ll | 40 +++----
>> test/CodeGen/R600/llvm.SI.load.dword.ll | 2 +-
>> test/CodeGen/R600/llvm.SI.sample-masked.ll | 14 +--
>> test/CodeGen/R600/llvm.SI.sample.ll | 2 +-
>> test/CodeGen/R600/llvm.SI.sendmsg.ll | 2 +-
>> test/CodeGen/R600/llvm.SI.tbuffer.store.ll | 8 +-
>> test/CodeGen/R600/llvm.amdgpu.kilp.ll | 2 +-
>> test/CodeGen/R600/llvm.amdgpu.lrp.ll | 2 +-
>> test/CodeGen/R600/llvm.exp2.ll | 6 +-
>> test/CodeGen/R600/llvm.floor.ll | 12 +--
>> test/CodeGen/R600/llvm.log2.ll | 6 +-
>> test/CodeGen/R600/llvm.memcpy.ll | 18 ++--
>> test/CodeGen/R600/llvm.rint.f64.ll | 6 +-
>> test/CodeGen/R600/llvm.rint.ll | 8 +-
>> test/CodeGen/R600/llvm.round.ll | 2 +-
>> test/CodeGen/R600/llvm.sin.ll | 8 +-
>> test/CodeGen/R600/llvm.sqrt.ll | 12 +--
>> test/CodeGen/R600/llvm.trunc.ll | 2 +-
>> test/CodeGen/R600/load-i1.ll | 20 ++--
>> test/CodeGen/R600/load.ll | 86 +++++++--------
>> test/CodeGen/R600/load.vec.ll | 8 +-
>> test/CodeGen/R600/load64.ll | 6 +-
>> test/CodeGen/R600/local-64.ll | 32 +++---
>> test/CodeGen/R600/local-atomics.ll | 108 +++++++++----------
>> test/CodeGen/R600/local-atomics64.ll | 100 +++++++++---------
>> test/CodeGen/R600/local-memory-two-objects.ll | 2 +-
>> test/CodeGen/R600/local-memory.ll | 2 +-
>> test/CodeGen/R600/loop-idiom.ll | 16 +--
>> test/CodeGen/R600/mad-sub.ll | 14 +--
>> test/CodeGen/R600/mad_int24.ll | 2 +-
>> test/CodeGen/R600/mad_uint24.ll | 8 +-
>> test/CodeGen/R600/max-literals.ll | 4 +-
>> test/CodeGen/R600/missing-store.ll | 2 +-
>> test/CodeGen/R600/mubuf.ll | 24 ++---
>> test/CodeGen/R600/mul.ll | 26 ++---
>> test/CodeGen/R600/mul_int24.ll | 2 +-
>> test/CodeGen/R600/mul_uint24.ll | 8 +-
>> .../R600/no-initializer-constant-addrspace.ll | 4 +-
>> test/CodeGen/R600/operand-spacing.ll | 2 +-
>> test/CodeGen/R600/or.ll | 36 +++----
>> test/CodeGen/R600/packetizer.ll | 2 +-
>> test/CodeGen/R600/predicate-dp4.ll | 2 +-
>> test/CodeGen/R600/predicates.ll | 8 +-
>> test/CodeGen/R600/private-memory.ll | 14 +--
>> test/CodeGen/R600/r600-encoding.ll | 4 +-
>> test/CodeGen/R600/register-count-comments.ll | 4 +-
>> test/CodeGen/R600/reorder-stores.ll | 8 +-
>> test/CodeGen/R600/rotl.i64.ll | 4 +-
>> test/CodeGen/R600/rotl.ll | 6 +-
>> test/CodeGen/R600/rotr.i64.ll | 8 +-
>> test/CodeGen/R600/rotr.ll | 6 +-
>> test/CodeGen/R600/rsq.ll | 6 +-
>> test/CodeGen/R600/saddo.ll | 10 +-
>> test/CodeGen/R600/salu-to-valu.ll | 10 +-
>> test/CodeGen/R600/scalar_to_vector.ll | 4 +-
>> test/CodeGen/R600/schedule-global-loads.ll | 4 +-
>> test/CodeGen/R600/schedule-kernel-arg-loads.ll | 2 +-
>> .../R600/schedule-vs-if-nested-loop-failure.ll | 2 +-
>> test/CodeGen/R600/sdiv.ll | 6 +-
>> test/CodeGen/R600/sdivrem24.ll | 24 ++---
>> test/CodeGen/R600/select-i1.ll | 2 +-
>> test/CodeGen/R600/select-vectors.ll | 22 ++--
>> test/CodeGen/R600/select.ll | 2 +-
>> test/CodeGen/R600/select64.ll | 8 +-
>> test/CodeGen/R600/selectcc-opt.ll | 8 +-
>> test/CodeGen/R600/selectcc.ll | 2 +-
>> test/CodeGen/R600/set-dx10.ll | 24 ++---
>> test/CodeGen/R600/setcc-equivalent.ll | 4 +-
>> test/CodeGen/R600/setcc-opt.ll | 2 +-
>> test/CodeGen/R600/setcc.ll | 52 ++++-----
>> test/CodeGen/R600/setcc64.ll | 48 ++++-----
>> test/CodeGen/R600/seto.ll | 2 +-
>> test/CodeGen/R600/setuo.ll | 2 +-
>> test/CodeGen/R600/sext-in-reg.ll | 64 ++++++------
>> test/CodeGen/R600/sgpr-control-flow.ll | 4 +-
>> test/CodeGen/R600/sgpr-copy-duplicate-operand.ll | 2 +-
>> test/CodeGen/R600/sgpr-copy.ll | 12 +--
>> test/CodeGen/R600/shared-op-cycle.ll | 2 +-
>> test/CodeGen/R600/shl.ll | 20 ++--
>> test/CodeGen/R600/shl_add_constant.ll | 10 +-
>> test/CodeGen/R600/shl_add_ptr.ll | 32 +++---
>> test/CodeGen/R600/si-annotate-cf-assertion.ll | 2 +-
>> test/CodeGen/R600/si-lod-bias.ll | 2 +-
>> test/CodeGen/R600/si-sgpr-spill.ll | 4 +-
>> test/CodeGen/R600/si-vector-hang.ll | 2 +-
>> test/CodeGen/R600/sign_extend.ll | 12 +--
>> .../R600/simplify-demanded-bits-build-pair.ll | 2 +-
>> test/CodeGen/R600/sint_to_fp.ll | 12 +--
>> test/CodeGen/R600/sint_to_fp64.ll | 6 +-
>> test/CodeGen/R600/smrd.ll | 14 +--
>> test/CodeGen/R600/split-scalar-i64-add.ll | 4 +-
>> test/CodeGen/R600/sra.ll | 24 ++---
>> test/CodeGen/R600/srl.ll | 20 ++--
>> test/CodeGen/R600/ssubo.ll | 10 +-
>> test/CodeGen/R600/store-v3i32.ll | 2 +-
>> test/CodeGen/R600/store-v3i64.ll | 8 +-
>> test/CodeGen/R600/store-vector-ptrs.ll | 2 +-
>> test/CodeGen/R600/store.ll | 98 ++++++++---------
>> test/CodeGen/R600/store.r600.ll | 4 +-
>> test/CodeGen/R600/structurize.ll | 2 +-
>> test/CodeGen/R600/structurize1.ll | 2 +-
>> test/CodeGen/R600/sub.ll | 8 +-
>> test/CodeGen/R600/swizzle-export.ll | 4 +-
>> test/CodeGen/R600/trunc-store-i1.ll | 6 +-
>> .../R600/trunc-vector-store-assertion-failure.ll | 2 +-
>> test/CodeGen/R600/trunc.ll | 12 +--
>> test/CodeGen/R600/uaddo.ll | 10 +-
>> test/CodeGen/R600/udiv.ll | 10 +-
>> test/CodeGen/R600/udivrem.ll | 6 +-
>> test/CodeGen/R600/udivrem24.ll | 24 ++---
>> test/CodeGen/R600/udivrem64.ll | 4 +-
>> test/CodeGen/R600/uint_to_fp.f64.ll | 6 +-
>> test/CodeGen/R600/uint_to_fp.ll | 10 +-
>> test/CodeGen/R600/unaligned-load-store.ll | 16 +--
>> .../R600/unhandled-loop-condition-assertion.ll | 6 +-
>> test/CodeGen/R600/unroll.ll | 2 +-
>> test/CodeGen/R600/unsupported-cc.ll | 20 ++--
>> test/CodeGen/R600/urem.ll | 8 +-
>> test/CodeGen/R600/use-sgpr-multiple-times.ll | 16 +--
>> test/CodeGen/R600/usubo.ll | 10 +-
>> test/CodeGen/R600/v1i64-kernel-arg.ll | 4 +-
>> test/CodeGen/R600/v_cndmask.ll | 4 +-
>> test/CodeGen/R600/vector-alloca.ll | 6 +-
>> test/CodeGen/R600/vertex-fetch-encoding.ll | 6 +-
>> test/CodeGen/R600/vop-shrink.ll | 4 +-
>> test/CodeGen/R600/vselect.ll | 14 +--
>> test/CodeGen/R600/vselect64.ll | 2 +-
>> test/CodeGen/R600/vtx-fetch-branch.ll | 2 +-
>> test/CodeGen/R600/vtx-schedule.ll | 2 +-
>> test/CodeGen/R600/wait.ll | 2 +-
>> test/CodeGen/R600/work-item-intrinsics.ll | 30 +++---
>> test/CodeGen/R600/wrong-transalu-pos-fix.ll | 2 +-
>> test/CodeGen/R600/xor.ll | 30 +++---
>> test/CodeGen/R600/zero_extend.ll | 8 +-
>> 267 files changed, 1511 insertions(+), 1511 deletions(-)
>>
>> diff --git a/lib/Target/R600/AMDGPUAsmPrinter.cpp b/lib/Target/R600/AMDGPUAsmPrinter.cpp
>> index 07ef6a7..1a0a7db 100644
>> --- a/lib/Target/R600/AMDGPUAsmPrinter.cpp
>> +++ b/lib/Target/R600/AMDGPUAsmPrinter.cpp
>> @@ -99,7 +99,7 @@ void AMDGPUAsmPrinter::EmitEndOfAsmFile(Module &M) {
>> bool AMDGPUAsmPrinter::runOnMachineFunction(MachineFunction &MF) {
>> SetupMachineFunction(MF);
>>
>> - OutStreamer.emitRawComment(Twine('@') + MF.getName() + Twine(':'));
>> + EmitFunctionHeader();
>
> Why does this need to get rid of the @ requiring all of the test updates?
>
Old output:
@v_cnd_nan_nosgpr:
New output:
v_cnd_nan_nosgpr: ; @v_cnd_nan_nosgpr
So, we either need to remove @ at the beginning, or the : at the end.
I guess we could also just leave the emitRawComment call there and the
tests would still work. I don't have a strong preference one way or the
other.
-Tom
>>
>> MCContext &Context = getObjFileLowering().getContext();
>> const MCSectionELF *ConfigSection = Context.getELFSection(".AMDGPU.config",
>> diff --git a/test/CodeGen/R600/128bit-kernel-args.ll b/test/CodeGen/R600/128bit-kernel-args.ll
>> index 3c4fcf7..9bd6b05 100644
>> --- a/test/CodeGen/R600/128bit-kernel-args.ll
>> +++ b/test/CodeGen/R600/128bit-kernel-args.ll
>> @@ -1,12 +1,12 @@
>> ; RUN: llc < %s -march=r600 -mcpu=redwood | FileCheck %s --check-prefix=R600-CHECK
>> ; RUN: llc < %s -march=r600 -mcpu=SI -verify-machineinstrs | FileCheck %s --check-prefix=SI-CHECK
>>
>> -; R600-CHECK: @v4i32_kernel_arg
>> +; R600-CHECK: v4i32_kernel_arg
>> ; R600-CHECK-DAG: MOV {{[* ]*}}T[[GPR:[0-9]]].X, KC0[3].Y
>> ; R600-CHECK-DAG: MOV {{[* ]*}}T[[GPR]].Y, KC0[3].Z
>> ; R600-CHECK-DAG: MOV {{[* ]*}}T[[GPR]].Z, KC0[3].W
>> ; R600-CHECK-DAG: MOV {{[* ]*}}T[[GPR]].W, KC0[4].X
>> -; SI-CHECK: @v4i32_kernel_arg
>> +; SI-CHECK: v4i32_kernel_arg
>> ; SI-CHECK: BUFFER_STORE_DWORDX4
>> define void @v4i32_kernel_arg(<4 x i32> addrspace(1)* %out, <4 x i32> %in) {
>> entry:
>> @@ -14,12 +14,12 @@ entry:
>> ret void
>> }
>>
>> -; R600-CHECK: @v4f32_kernel_arg
>> +; R600-CHECK: v4f32_kernel_arg
>> ; R600-CHECK-DAG: MOV {{[* ]*}}T[[GPR:[0-9]]].X, KC0[3].Y
>> ; R600-CHECK-DAG: MOV {{[* ]*}}T[[GPR]].Y, KC0[3].Z
>> ; R600-CHECK-DAG: MOV {{[* ]*}}T[[GPR]].Z, KC0[3].W
>> ; R600-CHECK-DAG: MOV {{[* ]*}}T[[GPR]].W, KC0[4].X
>> -; SI-CHECK: @v4f32_kernel_arg
>> +; SI-CHECK: v4f32_kernel_arg
>> ; SI-CHECK: BUFFER_STORE_DWORDX4
>> define void @v4f32_kernel_args(<4 x float> addrspace(1)* %out, <4 x float> %in) {
>> entry:
>> diff --git a/test/CodeGen/R600/32-bit-local-address-space.ll b/test/CodeGen/R600/32-bit-local-address-space.ll
>> index e13d719..7b189cc 100644
>> --- a/test/CodeGen/R600/32-bit-local-address-space.ll
>> +++ b/test/CodeGen/R600/32-bit-local-address-space.ll
>> @@ -10,7 +10,7 @@
>> ; Instructions with B32, U32, and I32 in their name take 32-bit operands, while
>> ; instructions with B64, U64, and I64 take 64-bit operands.
>>
>> -; FUNC-LABEL: @local_address_load
>> +; FUNC-LABEL: local_address_load
>> ; CHECK: V_MOV_B32_e{{32|64}} [[PTR:v[0-9]]]
>> ; CHECK: DS_READ_B32 v{{[0-9]+}}, [[PTR]]
>> define void @local_address_load(i32 addrspace(1)* %out, i32 addrspace(3)* %in) {
>> @@ -20,7 +20,7 @@ entry:
>> ret void
>> }
>>
>> -; FUNC-LABEL: @local_address_gep
>> +; FUNC-LABEL: local_address_gep
>> ; CHECK: S_ADD_I32 [[SPTR:s[0-9]]]
>> ; CHECK: V_MOV_B32_e32 [[VPTR:v[0-9]+]], [[SPTR]]
>> ; CHECK: DS_READ_B32 [[VPTR]]
>> @@ -32,7 +32,7 @@ entry:
>> ret void
>> }
>>
>> -; FUNC-LABEL: @local_address_gep_const_offset
>> +; FUNC-LABEL: local_address_gep_const_offset
>> ; CHECK: V_MOV_B32_e32 [[VPTR:v[0-9]+]], s{{[0-9]+}}
>> ; CHECK: DS_READ_B32 v{{[0-9]+}}, [[VPTR]], 0x4,
>> define void @local_address_gep_const_offset(i32 addrspace(1)* %out, i32 addrspace(3)* %in) {
>> @@ -44,7 +44,7 @@ entry:
>> }
>>
>> ; Offset too large, can't fold into 16-bit immediate offset.
>> -; FUNC-LABEL: @local_address_gep_large_const_offset
>> +; FUNC-LABEL: local_address_gep_large_const_offset
>> ; CHECK: S_ADD_I32 [[SPTR:s[0-9]]], s{{[0-9]+}}, 0x10004
>> ; CHECK: V_MOV_B32_e32 [[VPTR:v[0-9]+]], [[SPTR]]
>> ; CHECK: DS_READ_B32 [[VPTR]]
>> @@ -56,7 +56,7 @@ entry:
>> ret void
>> }
>>
>> -; FUNC-LABEL: @null_32bit_lds_ptr:
>> +; FUNC-LABEL: null_32bit_lds_ptr:
>> ; CHECK: V_CMP_NE_I32
>> ; CHECK-NOT: V_CMP_NE_I32
>> ; CHECK: V_CNDMASK_B32
>> @@ -67,7 +67,7 @@ define void @null_32bit_lds_ptr(i32 addrspace(1)* %out, i32 addrspace(3)* %lds)
>> ret void
>> }
>>
>> -; FUNC-LABEL: @mul_32bit_ptr:
>> +; FUNC-LABEL: mul_32bit_ptr:
>> ; CHECK: V_MUL_LO_I32
>> ; CHECK-NEXT: V_ADD_I32_e32
>> ; CHECK-NEXT: DS_READ_B32
>> @@ -80,7 +80,7 @@ define void @mul_32bit_ptr(float addrspace(1)* %out, [3 x float] addrspace(3)* %
>>
>> @g_lds = addrspace(3) global float zeroinitializer, align 4
>>
>> -; FUNC-LABEL: @infer_ptr_alignment_global_offset:
>> +; FUNC-LABEL: infer_ptr_alignment_global_offset:
>> ; CHECK: V_MOV_B32_e32 [[REG:v[0-9]+]], 0
>> ; CHECK: DS_READ_B32 v{{[0-9]+}}, [[REG]]
>> define void @infer_ptr_alignment_global_offset(float addrspace(1)* %out, i32 %tid) {
>> @@ -93,21 +93,21 @@ define void @infer_ptr_alignment_global_offset(float addrspace(1)* %out, i32 %ti
>> @ptr = addrspace(3) global i32 addrspace(3)* null
>> @dst = addrspace(3) global [16384 x i32] zeroinitializer
>>
>> -; FUNC-LABEL: @global_ptr:
>> +; FUNC-LABEL: global_ptr:
>> ; CHECK: DS_WRITE_B32
>> define void @global_ptr() nounwind {
>> store i32 addrspace(3)* getelementptr ([16384 x i32] addrspace(3)* @dst, i32 0, i32 16), i32 addrspace(3)* addrspace(3)* @ptr
>> ret void
>> }
>>
>> -; FUNC-LABEL: @local_address_store
>> +; FUNC-LABEL: local_address_store
>> ; CHECK: DS_WRITE_B32
>> define void @local_address_store(i32 addrspace(3)* %out, i32 %val) {
>> store i32 %val, i32 addrspace(3)* %out
>> ret void
>> }
>>
>> -; FUNC-LABEL: @local_address_gep_store
>> +; FUNC-LABEL: local_address_gep_store
>> ; CHECK: S_ADD_I32 [[SADDR:s[0-9]+]],
>> ; CHECK: V_MOV_B32_e32 [[ADDR:v[0-9]+]], [[SADDR]]
>> ; CHECK: DS_WRITE_B32 [[ADDR]], v{{[0-9]+}},
>> @@ -117,7 +117,7 @@ define void @local_address_gep_store(i32 addrspace(3)* %out, i32, i32 %val, i32
>> ret void
>> }
>>
>> -; FUNC-LABEL: @local_address_gep_const_offset_store
>> +; FUNC-LABEL: local_address_gep_const_offset_store
>> ; CHECK: V_MOV_B32_e32 [[VPTR:v[0-9]+]], s{{[0-9]+}}
>> ; CHECK: V_MOV_B32_e32 [[VAL:v[0-9]+]], s{{[0-9]+}}
>> ; CHECK: DS_WRITE_B32 [[VPTR]], [[VAL]], 0x4
>> @@ -128,7 +128,7 @@ define void @local_address_gep_const_offset_store(i32 addrspace(3)* %out, i32 %v
>> }
>>
>> ; Offset too large, can't fold into 16-bit immediate offset.
>> -; FUNC-LABEL: @local_address_gep_large_const_offset_store
>> +; FUNC-LABEL: local_address_gep_large_const_offset_store
>> ; CHECK: S_ADD_I32 [[SPTR:s[0-9]]], s{{[0-9]+}}, 0x10004
>> ; CHECK: V_MOV_B32_e32 [[VPTR:v[0-9]+]], [[SPTR]]
>> ; CHECK: DS_WRITE_B32 [[VPTR]], v{{[0-9]+}}, 0
>> diff --git a/test/CodeGen/R600/64bit-kernel-args.ll b/test/CodeGen/R600/64bit-kernel-args.ll
>> index 2d82c1e..7255b52 100644
>> --- a/test/CodeGen/R600/64bit-kernel-args.ll
>> +++ b/test/CodeGen/R600/64bit-kernel-args.ll
>> @@ -1,6 +1,6 @@
>> ; RUN: llc < %s -march=r600 -mcpu=tahiti -verify-machineinstrs | FileCheck %s --check-prefix=SI-CHECK
>>
>> -; SI-CHECK: @f64_kernel_arg
>> +; SI-CHECK: f64_kernel_arg
>> ; SI-CHECK-DAG: S_LOAD_DWORDX2 s[{{[0-9]:[0-9]}}], s[0:1], 0x9
>> ; SI-CHECK-DAG: S_LOAD_DWORDX2 s[{{[0-9]:[0-9]}}], s[0:1], 0xb
>> ; SI-CHECK: BUFFER_STORE_DWORDX2
>> diff --git a/test/CodeGen/R600/add.ll b/test/CodeGen/R600/add.ll
>> index 8cf43d1..a20ada5 100644
>> --- a/test/CodeGen/R600/add.ll
>> +++ b/test/CodeGen/R600/add.ll
>> @@ -1,7 +1,7 @@
>> ; RUN: llc < %s -march=r600 -mcpu=redwood | FileCheck --check-prefix=EG-CHECK --check-prefix=FUNC %s
>> ; RUN: llc < %s -march=r600 -mcpu=verde -verify-machineinstrs | FileCheck --check-prefix=SI-CHECK --check-prefix=FUNC %s
>>
>> -;FUNC-LABEL: @test1:
>> +;FUNC-LABEL: test1:
>> ;EG-CHECK: ADD_INT {{[* ]*}}T{{[0-9]+\.[XYZW], T[0-9]+\.[XYZW], T[0-9]+\.[XYZW]}}
>>
>> ;SI-CHECK: V_ADD_I32_e32 [[REG:v[0-9]+]], {{v[0-9]+, v[0-9]+}}
>> @@ -16,7 +16,7 @@ define void @test1(i32 addrspace(1)* %out, i32 addrspace(1)* %in) {
>> ret void
>> }
>>
>> -;FUNC-LABEL: @test2:
>> +;FUNC-LABEL: test2:
>> ;EG-CHECK: ADD_INT {{[* ]*}}T{{[0-9]+\.[XYZW], T[0-9]+\.[XYZW], T[0-9]+\.[XYZW]}}
>> ;EG-CHECK: ADD_INT {{[* ]*}}T{{[0-9]+\.[XYZW], T[0-9]+\.[XYZW], T[0-9]+\.[XYZW]}}
>>
>> @@ -32,7 +32,7 @@ define void @test2(<2 x i32> addrspace(1)* %out, <2 x i32> addrspace(1)* %in) {
>> ret void
>> }
>>
>> -;FUNC-LABEL: @test4:
>> +;FUNC-LABEL: test4:
>> ;EG-CHECK: ADD_INT {{[* ]*}}T{{[0-9]+\.[XYZW], T[0-9]+\.[XYZW], T[0-9]+\.[XYZW]}}
>> ;EG-CHECK: ADD_INT {{[* ]*}}T{{[0-9]+\.[XYZW], T[0-9]+\.[XYZW], T[0-9]+\.[XYZW]}}
>> ;EG-CHECK: ADD_INT {{[* ]*}}T{{[0-9]+\.[XYZW], T[0-9]+\.[XYZW], T[0-9]+\.[XYZW]}}
>> @@ -52,7 +52,7 @@ define void @test4(<4 x i32> addrspace(1)* %out, <4 x i32> addrspace(1)* %in) {
>> ret void
>> }
>>
>> -; FUNC-LABEL: @test8
>> +; FUNC-LABEL: test8
>> ; EG-CHECK: ADD_INT
>> ; EG-CHECK: ADD_INT
>> ; EG-CHECK: ADD_INT
>> @@ -76,7 +76,7 @@ entry:
>> ret void
>> }
>>
>> -; FUNC-LABEL: @test16
>> +; FUNC-LABEL: test16
>> ; EG-CHECK: ADD_INT
>> ; EG-CHECK: ADD_INT
>> ; EG-CHECK: ADD_INT
>> @@ -116,7 +116,7 @@ entry:
>> ret void
>> }
>>
>> -; FUNC-LABEL: @add64
>> +; FUNC-LABEL: add64
>> ; SI-CHECK: S_ADD_U32
>> ; SI-CHECK: S_ADDC_U32
>> define void @add64(i64 addrspace(1)* %out, i64 %a, i64 %b) {
>> @@ -131,7 +131,7 @@ entry:
>> ; %0 will be stored in a VGPR, so the comiler will be forced to copy %a
>> ; to a VGPR before doing the add.
>>
>> -; FUNC-LABEL: @add64_sgpr_vgpr
>> +; FUNC-LABEL: add64_sgpr_vgpr
>> ; SI-CHECK-NOT: V_ADDC_U32_e32 s
>> define void @add64_sgpr_vgpr(i64 addrspace(1)* %out, i64 %a, i64 addrspace(1)* %in) {
>> entry:
>> @@ -142,7 +142,7 @@ entry:
>> }
>>
>> ; Test i64 add inside a branch.
>> -; FUNC-LABEL: @add64_in_branch
>> +; FUNC-LABEL: add64_in_branch
>> ; SI-CHECK: S_ADD_U32
>> ; SI-CHECK: S_ADDC_U32
>> define void @add64_in_branch(i64 addrspace(1)* %out, i64 addrspace(1)* %in, i64 %a, i64 %b, i64 %c) {
>> diff --git a/test/CodeGen/R600/add_i64.ll b/test/CodeGen/R600/add_i64.ll
>> index 5be969c..4e7b3a2 100644
>> --- a/test/CodeGen/R600/add_i64.ll
>> +++ b/test/CodeGen/R600/add_i64.ll
>> @@ -3,7 +3,7 @@
>>
>> declare i32 @llvm.r600.read.tidig.x() readnone
>>
>> -; SI-LABEL: @test_i64_vreg:
>> +; SI-LABEL: test_i64_vreg:
>> ; SI: V_ADD_I32
>> ; SI: V_ADDC_U32
>> define void @test_i64_vreg(i64 addrspace(1)* noalias %out, i64 addrspace(1)* noalias %inA, i64 addrspace(1)* noalias %inB) {
>> @@ -18,7 +18,7 @@ define void @test_i64_vreg(i64 addrspace(1)* noalias %out, i64 addrspace(1)* noa
>> }
>>
>> ; Check that the SGPR add operand is correctly moved to a VGPR.
>> -; SI-LABEL: @sgpr_operand:
>> +; SI-LABEL: sgpr_operand:
>> ; SI: V_ADD_I32
>> ; SI: V_ADDC_U32
>> define void @sgpr_operand(i64 addrspace(1)* noalias %out, i64 addrspace(1)* noalias %in, i64 addrspace(1)* noalias %in_bar, i64 %a) {
>> @@ -31,7 +31,7 @@ define void @sgpr_operand(i64 addrspace(1)* noalias %out, i64 addrspace(1)* noal
>> ; Swap the arguments. Check that the SGPR -> VGPR copy works with the
>> ; SGPR as other operand.
>> ;
>> -; SI-LABEL: @sgpr_operand_reversed:
>> +; SI-LABEL: sgpr_operand_reversed:
>> ; SI: V_ADD_I32
>> ; SI: V_ADDC_U32
>> define void @sgpr_operand_reversed(i64 addrspace(1)* noalias %out, i64 addrspace(1)* noalias %in, i64 %a) {
>> @@ -42,7 +42,7 @@ define void @sgpr_operand_reversed(i64 addrspace(1)* noalias %out, i64 addrspace
>> }
>>
>>
>> -; SI-LABEL: @test_v2i64_sreg:
>> +; SI-LABEL: test_v2i64_sreg:
>> ; SI: S_ADD_U32
>> ; SI: S_ADDC_U32
>> ; SI: S_ADD_U32
>> @@ -53,7 +53,7 @@ define void @test_v2i64_sreg(<2 x i64> addrspace(1)* noalias %out, <2 x i64> %a,
>> ret void
>> }
>>
>> -; SI-LABEL: @test_v2i64_vreg:
>> +; SI-LABEL: test_v2i64_vreg:
>> ; SI: V_ADD_I32
>> ; SI: V_ADDC_U32
>> ; SI: V_ADD_I32
>> @@ -69,7 +69,7 @@ define void @test_v2i64_vreg(<2 x i64> addrspace(1)* noalias %out, <2 x i64> add
>> ret void
>> }
>>
>> -; SI-LABEL: @trunc_i64_add_to_i32
>> +; SI-LABEL: trunc_i64_add_to_i32
>> ; SI: S_LOAD_DWORD s[[SREG0:[0-9]+]]
>> ; SI: S_LOAD_DWORD s[[SREG1:[0-9]+]]
>> ; SI: S_ADD_I32 [[SRESULT:s[0-9]+]], s[[SREG1]], s[[SREG0]]
>> diff --git a/test/CodeGen/R600/address-space.ll b/test/CodeGen/R600/address-space.ll
>> index 7f52472..5c27949 100644
>> --- a/test/CodeGen/R600/address-space.ll
>> +++ b/test/CodeGen/R600/address-space.ll
>> @@ -7,7 +7,7 @@
>> ; FIXME: Extra V_MOV from SGPR to VGPR for second read. The address is
>> ; already in a VGPR after the first read.
>>
>> -; CHECK-LABEL: @do_as_ptr_calcs:
>> +; CHECK-LABEL: do_as_ptr_calcs:
>> ; CHECK: S_LOAD_DWORD [[SREG1:s[0-9]+]],
>> ; CHECK: V_MOV_B32_e32 [[VREG1:v[0-9]+]], [[SREG1]]
>> ; CHECK-DAG: DS_READ_B32 v{{[0-9]+}}, [[VREG1]], 0xc
>> diff --git a/test/CodeGen/R600/and.ll b/test/CodeGen/R600/and.ll
>> index ccb97e3..3ec5397 100644
>> --- a/test/CodeGen/R600/and.ll
>> +++ b/test/CodeGen/R600/and.ll
>> @@ -1,7 +1,7 @@
>> ; RUN: llc -march=r600 -mcpu=redwood < %s | FileCheck -check-prefix=EG -check-prefix=FUNC %s
>> ; RUN: llc -march=r600 -mcpu=verde -verify-machineinstrs < %s | FileCheck -check-prefix=SI -check-prefix=FUNC %s
>>
>> -; FUNC-LABEL: @test2
>> +; FUNC-LABEL: test2
>> ; EG: AND_INT {{\*? *}}T{{[0-9]+\.[XYZW], T[0-9]+\.[XYZW], T[0-9]+\.[XYZW]}}
>> ; EG: AND_INT {{\*? *}}T{{[0-9]+\.[XYZW], T[0-9]+\.[XYZW], T[0-9]+\.[XYZW]}}
>>
>> @@ -17,7 +17,7 @@ define void @test2(<2 x i32> addrspace(1)* %out, <2 x i32> addrspace(1)* %in) {
>> ret void
>> }
>>
>> -; FUNC-LABEL: @test4
>> +; FUNC-LABEL: test4
>> ; EG: AND_INT {{\** *}}T{{[0-9]+\.[XYZW], T[0-9]+\.[XYZW], T[0-9]+\.[XYZW]}}
>> ; EG: AND_INT {{\** *}}T{{[0-9]+\.[XYZW], T[0-9]+\.[XYZW], T[0-9]+\.[XYZW]}}
>> ; EG: AND_INT {{\** *}}T{{[0-9]+\.[XYZW], T[0-9]+\.[XYZW], T[0-9]+\.[XYZW]}}
>> @@ -37,7 +37,7 @@ define void @test4(<4 x i32> addrspace(1)* %out, <4 x i32> addrspace(1)* %in) {
>> ret void
>> }
>>
>> -; FUNC-LABEL: @s_and_i32
>> +; FUNC-LABEL: s_and_i32
>> ; SI: S_AND_B32
>> define void @s_and_i32(i32 addrspace(1)* %out, i32 %a, i32 %b) {
>> %and = and i32 %a, %b
>> @@ -45,7 +45,7 @@ define void @s_and_i32(i32 addrspace(1)* %out, i32 %a, i32 %b) {
>> ret void
>> }
>>
>> -; FUNC-LABEL: @s_and_constant_i32
>> +; FUNC-LABEL: s_and_constant_i32
>> ; SI: S_AND_B32 s{{[0-9]+}}, s{{[0-9]+}}, 0x12d687
>> define void @s_and_constant_i32(i32 addrspace(1)* %out, i32 %a) {
>> %and = and i32 %a, 1234567
>> @@ -53,7 +53,7 @@ define void @s_and_constant_i32(i32 addrspace(1)* %out, i32 %a) {
>> ret void
>> }
>>
>> -; FUNC-LABEL: @v_and_i32
>> +; FUNC-LABEL: v_and_i32
>> ; SI: V_AND_B32
>> define void @v_and_i32(i32 addrspace(1)* %out, i32 addrspace(1)* %aptr, i32 addrspace(1)* %bptr) {
>> %a = load i32 addrspace(1)* %aptr, align 4
>> @@ -63,7 +63,7 @@ define void @v_and_i32(i32 addrspace(1)* %out, i32 addrspace(1)* %aptr, i32 addr
>> ret void
>> }
>>
>> -; FUNC-LABEL: @v_and_constant_i32
>> +; FUNC-LABEL: v_and_constant_i32
>> ; SI: V_AND_B32
>> define void @v_and_constant_i32(i32 addrspace(1)* %out, i32 addrspace(1)* %aptr) {
>> %a = load i32 addrspace(1)* %aptr, align 4
>> @@ -72,7 +72,7 @@ define void @v_and_constant_i32(i32 addrspace(1)* %out, i32 addrspace(1)* %aptr)
>> ret void
>> }
>>
>> -; FUNC-LABEL: @s_and_i64
>> +; FUNC-LABEL: s_and_i64
>> ; SI: S_AND_B64
>> define void @s_and_i64(i64 addrspace(1)* %out, i64 %a, i64 %b) {
>> %and = and i64 %a, %b
>> @@ -81,7 +81,7 @@ define void @s_and_i64(i64 addrspace(1)* %out, i64 %a, i64 %b) {
>> }
>>
>> ; FIXME: Should use SGPRs
>> -; FUNC-LABEL: @s_and_i1
>> +; FUNC-LABEL: s_and_i1
>> ; SI: V_AND_B32
>> define void @s_and_i1(i1 addrspace(1)* %out, i1 %a, i1 %b) {
>> %and = and i1 %a, %b
>> @@ -89,7 +89,7 @@ define void @s_and_i1(i1 addrspace(1)* %out, i1 %a, i1 %b) {
>> ret void
>> }
>>
>> -; FUNC-LABEL: @s_and_constant_i64
>> +; FUNC-LABEL: s_and_constant_i64
>> ; SI: S_AND_B64
>> define void @s_and_constant_i64(i64 addrspace(1)* %out, i64 %a) {
>> %and = and i64 %a, 281474976710655
>> @@ -97,7 +97,7 @@ define void @s_and_constant_i64(i64 addrspace(1)* %out, i64 %a) {
>> ret void
>> }
>>
>> -; FUNC-LABEL: @v_and_i64
>> +; FUNC-LABEL: v_and_i64
>> ; SI: V_AND_B32
>> ; SI: V_AND_B32
>> define void @v_and_i64(i64 addrspace(1)* %out, i64 addrspace(1)* %aptr, i64 addrspace(1)* %bptr) {
>> @@ -108,7 +108,7 @@ define void @v_and_i64(i64 addrspace(1)* %out, i64 addrspace(1)* %aptr, i64 addr
>> ret void
>> }
>>
>> -; FUNC-LABEL: @v_and_i64_br
>> +; FUNC-LABEL: v_and_i64_br
>> ; SI: V_AND_B32
>> ; SI: V_AND_B32
>> define void @v_and_i64_br(i64 addrspace(1)* %out, i64 addrspace(1)* %aptr, i64 addrspace(1)* %bptr, i32 %cond) {
>> @@ -128,7 +128,7 @@ endif:
>> ret void
>> }
>>
>> -; FUNC-LABEL: @v_and_constant_i64
>> +; FUNC-LABEL: v_and_constant_i64
>> ; SI: V_AND_B32_e32 {{v[0-9]+}}, {{s[0-9]+}}, {{v[0-9]+}}
>> ; SI: V_AND_B32_e32 {{v[0-9]+}}, {{s[0-9]+}}, {{v[0-9]+}}
>> define void @v_and_constant_i64(i64 addrspace(1)* %out, i64 addrspace(1)* %aptr) {
>> @@ -139,7 +139,7 @@ define void @v_and_constant_i64(i64 addrspace(1)* %out, i64 addrspace(1)* %aptr)
>> }
>>
>> ; FIXME: Replace and 0 with mov 0
>> -; FUNC-LABEL: @v_and_inline_imm_i64
>> +; FUNC-LABEL: v_and_inline_imm_i64
>> ; SI: V_AND_B32_e32 {{v[0-9]+}}, 64, {{v[0-9]+}}
>> ; SI: V_AND_B32_e32 {{v[0-9]+}}, 0, {{v[0-9]+}}
>> define void @v_and_inline_imm_i64(i64 addrspace(1)* %out, i64 addrspace(1)* %aptr) {
>> @@ -149,7 +149,7 @@ define void @v_and_inline_imm_i64(i64 addrspace(1)* %out, i64 addrspace(1)* %apt
>> ret void
>> }
>>
>> -; FUNC-LABEL: @s_and_inline_imm_i64
>> +; FUNC-LABEL: s_and_inline_imm_i64
>> ; SI: S_AND_B64 s{{\[[0-9]+:[0-9]+\]}}, s{{\[[0-9]+:[0-9]+\]}}, 64
>> define void @s_and_inline_imm_i64(i64 addrspace(1)* %out, i64 addrspace(1)* %aptr, i64 %a) {
>> %and = and i64 %a, 64
>> diff --git a/test/CodeGen/R600/anyext.ll b/test/CodeGen/R600/anyext.ll
>> index bbe5d0a..9e21738 100644
>> --- a/test/CodeGen/R600/anyext.ll
>> +++ b/test/CodeGen/R600/anyext.ll
>> @@ -1,6 +1,6 @@
>> ; RUN: llc < %s -march=r600 -mcpu=verde -verify-machineinstrs | FileCheck %s
>>
>> -; CHECK-LABEL: @anyext_i1_i32
>> +; CHECK-LABEL: anyext_i1_i32
>> ; CHECK: V_CNDMASK_B32_e64
>> define void @anyext_i1_i32(i32 addrspace(1)* %out, i32 %cond) {
>> entry:
>> diff --git a/test/CodeGen/R600/array-ptr-calc-i32.ll b/test/CodeGen/R600/array-ptr-calc-i32.ll
>> index fd5bd52..818be0d 100644
>> --- a/test/CodeGen/R600/array-ptr-calc-i32.ll
>> +++ b/test/CodeGen/R600/array-ptr-calc-i32.ll
>> @@ -9,7 +9,7 @@ declare void @llvm.AMDGPU.barrier.local() nounwind noduplicate
>> ; 64-bit pointer add. This should work since private pointers should
>> ; be 32-bits.
>>
>> -; SI-LABEL: @test_private_array_ptr_calc:
>> +; SI-LABEL: test_private_array_ptr_calc:
>>
>> ; FIXME: We end up with zero argument for ADD, because
>> ; SIRegisterInfo::eliminateFrameIndex() blindly replaces the frame index
>> diff --git a/test/CodeGen/R600/array-ptr-calc-i64.ll b/test/CodeGen/R600/array-ptr-calc-i64.ll
>> index 2962471..e93d73a 100644
>> --- a/test/CodeGen/R600/array-ptr-calc-i64.ll
>> +++ b/test/CodeGen/R600/array-ptr-calc-i64.ll
>> @@ -2,7 +2,7 @@
>>
>> declare i32 @llvm.SI.tid() readnone
>>
>> -; SI-LABEL: @test_array_ptr_calc
>> +; SI-LABEL: test_array_ptr_calc
>> ; SI: V_MUL_LO_I32
>> ; SI: V_MUL_HI_I32
>> define void @test_array_ptr_calc(i32 addrspace(1)* noalias %out, [1025 x i32] addrspace(1)* noalias %inA, i32 addrspace(1)* noalias %inB) {
>> diff --git a/test/CodeGen/R600/atomic_cmp_swap_local.ll b/test/CodeGen/R600/atomic_cmp_swap_local.ll
>> index 9f89774..dc8796d 100644
>> --- a/test/CodeGen/R600/atomic_cmp_swap_local.ll
>> +++ b/test/CodeGen/R600/atomic_cmp_swap_local.ll
>> @@ -1,7 +1,7 @@
>> ; RUN: llc -march=r600 -mcpu=SI -verify-machineinstrs < %s | FileCheck -check-prefix=SI -check-prefix=FUNC %s
>> ; RUN: llc -march=r600 -mcpu=bonaire -verify-machineinstrs < %s | FileCheck -check-prefix=CI -check-prefix=FUNC %s
>>
>> -; FUNC-LABEL: @lds_atomic_cmpxchg_ret_i32_offset:
>> +; FUNC-LABEL: lds_atomic_cmpxchg_ret_i32_offset:
>> ; SI: S_LOAD_DWORD [[PTR:s[0-9]+]], s{{\[[0-9]+:[0-9]+\]}}, 0xb
>> ; SI: S_LOAD_DWORD [[SWAP:s[0-9]+]], s{{\[[0-9]+:[0-9]+\]}}, 0xc
>> ; SI-DAG: V_MOV_B32_e32 [[VCMP:v[0-9]+]], 7
>> @@ -17,7 +17,7 @@ define void @lds_atomic_cmpxchg_ret_i32_offset(i32 addrspace(1)* %out, i32 addrs
>> ret void
>> }
>>
>> -; FUNC-LABEL: @lds_atomic_cmpxchg_ret_i64_offset:
>> +; FUNC-LABEL: lds_atomic_cmpxchg_ret_i64_offset:
>> ; SI: S_LOAD_DWORD [[PTR:s[0-9]+]], s{{\[[0-9]+:[0-9]+\]}}, 0xb
>> ; SI: S_LOAD_DWORDX2 s{{\[}}[[LOSWAP:[0-9]+]]:[[HISWAP:[0-9]+]]{{\]}}, s{{\[[0-9]+:[0-9]+\]}}, 0xd
>> ; SI: S_MOV_B64 s{{\[}}[[LOSCMP:[0-9]+]]:[[HISCMP:[0-9]+]]{{\]}}, 7
>> @@ -37,7 +37,7 @@ define void @lds_atomic_cmpxchg_ret_i64_offset(i64 addrspace(1)* %out, i64 addrs
>> ret void
>> }
>>
>> -; FUNC-LABEL: @lds_atomic_cmpxchg_ret_i32_bad_si_offset
>> +; FUNC-LABEL: lds_atomic_cmpxchg_ret_i32_bad_si_offset
>> ; SI: DS_CMPST_RTN_B32 v{{[0-9]+}}, v{{[0-9]+}}, v{{[0-9]+}}, v{{[0-9]+}}, 0x0
>> ; CI: DS_CMPST_RTN_B32 v{{[0-9]+}}, v{{[0-9]+}}, v{{[0-9]+}}, v{{[0-9]+}}, 0x10
>> ; SI: S_ENDPGM
>> @@ -51,7 +51,7 @@ define void @lds_atomic_cmpxchg_ret_i32_bad_si_offset(i32 addrspace(1)* %out, i3
>> ret void
>> }
>>
>> -; FUNC-LABEL: @lds_atomic_cmpxchg_noret_i32_offset:
>> +; FUNC-LABEL: lds_atomic_cmpxchg_noret_i32_offset:
>> ; SI: S_LOAD_DWORD [[PTR:s[0-9]+]], s{{\[[0-9]+:[0-9]+\]}}, 0x9
>> ; SI: S_LOAD_DWORD [[SWAP:s[0-9]+]], s{{\[[0-9]+:[0-9]+\]}}, 0xa
>> ; SI-DAG: V_MOV_B32_e32 [[VCMP:v[0-9]+]], 7
>> @@ -66,7 +66,7 @@ define void @lds_atomic_cmpxchg_noret_i32_offset(i32 addrspace(3)* %ptr, i32 %sw
>> ret void
>> }
>>
>> -; FUNC-LABEL: @lds_atomic_cmpxchg_noret_i64_offset:
>> +; FUNC-LABEL: lds_atomic_cmpxchg_noret_i64_offset:
>> ; SI: S_LOAD_DWORD [[PTR:s[0-9]+]], s{{\[[0-9]+:[0-9]+\]}}, 0x9
>> ; SI: S_LOAD_DWORDX2 s{{\[}}[[LOSWAP:[0-9]+]]:[[HISWAP:[0-9]+]]{{\]}}, s{{\[[0-9]+:[0-9]+\]}}, 0xb
>> ; SI: S_MOV_B64 s{{\[}}[[LOSCMP:[0-9]+]]:[[HISCMP:[0-9]+]]{{\]}}, 7
>> diff --git a/test/CodeGen/R600/atomic_load_add.ll b/test/CodeGen/R600/atomic_load_add.ll
>> index 6dfb8c7..6c39d1e 100644
>> --- a/test/CodeGen/R600/atomic_load_add.ll
>> +++ b/test/CodeGen/R600/atomic_load_add.ll
>> @@ -1,7 +1,7 @@
>> ; RUN: llc -march=r600 -mcpu=SI -verify-machineinstrs < %s | FileCheck %s -check-prefix=SI -check-prefix=FUNC
>> ; RUN: llc -march=r600 -mcpu=redwood < %s | FileCheck -check-prefix=R600 -check-prefix=FUNC %s
>>
>> -; FUNC-LABEL: @atomic_add_local
>> +; FUNC-LABEL: atomic_add_local
>> ; R600: LDS_ADD *
>> ; SI: DS_ADD_U32
>> define void @atomic_add_local(i32 addrspace(3)* %local) {
>> @@ -9,7 +9,7 @@ define void @atomic_add_local(i32 addrspace(3)* %local) {
>> ret void
>> }
>>
>> -; FUNC-LABEL: @atomic_add_local_const_offset
>> +; FUNC-LABEL: atomic_add_local_const_offset
>> ; R600: LDS_ADD *
>> ; SI: DS_ADD_U32 v{{[0-9]+}}, v{{[0-9]+}}, 0x10
>> define void @atomic_add_local_const_offset(i32 addrspace(3)* %local) {
>> @@ -18,7 +18,7 @@ define void @atomic_add_local_const_offset(i32 addrspace(3)* %local) {
>> ret void
>> }
>>
>> -; FUNC-LABEL: @atomic_add_ret_local
>> +; FUNC-LABEL: atomic_add_ret_local
>> ; R600: LDS_ADD_RET *
>> ; SI: DS_ADD_RTN_U32
>> define void @atomic_add_ret_local(i32 addrspace(1)* %out, i32 addrspace(3)* %local) {
>> @@ -27,7 +27,7 @@ define void @atomic_add_ret_local(i32 addrspace(1)* %out, i32 addrspace(3)* %loc
>> ret void
>> }
>>
>> -; FUNC-LABEL: @atomic_add_ret_local_const_offset
>> +; FUNC-LABEL: atomic_add_ret_local_const_offset
>> ; R600: LDS_ADD_RET *
>> ; SI: DS_ADD_RTN_U32 v{{[0-9]+}}, v{{[0-9]+}}, v{{[0-9]+}}, 0x14
>> define void @atomic_add_ret_local_const_offset(i32 addrspace(1)* %out, i32 addrspace(3)* %local) {
>> diff --git a/test/CodeGen/R600/atomic_load_sub.ll b/test/CodeGen/R600/atomic_load_sub.ll
>> index e5937f9..c302f6e 100644
>> --- a/test/CodeGen/R600/atomic_load_sub.ll
>> +++ b/test/CodeGen/R600/atomic_load_sub.ll
>> @@ -1,7 +1,7 @@
>> ; RUN: llc -march=r600 -mcpu=redwood < %s | FileCheck -check-prefix=R600 -check-prefix=FUNC %s
>> ; RUN: llc -march=r600 -mcpu=SI -verify-machineinstrs < %s | FileCheck -check-prefix=SI -check-prefix=FUNC %s
>>
>> -; FUNC-LABEL: @atomic_sub_local
>> +; FUNC-LABEL: atomic_sub_local
>> ; R600: LDS_SUB *
>> ; SI: DS_SUB_U32
>> define void @atomic_sub_local(i32 addrspace(3)* %local) {
>> @@ -9,7 +9,7 @@ define void @atomic_sub_local(i32 addrspace(3)* %local) {
>> ret void
>> }
>>
>> -; FUNC-LABEL: @atomic_sub_local_const_offset
>> +; FUNC-LABEL: atomic_sub_local_const_offset
>> ; R600: LDS_SUB *
>> ; SI: DS_SUB_U32 v{{[0-9]+}}, v{{[0-9]+}}, 0x10
>> define void @atomic_sub_local_const_offset(i32 addrspace(3)* %local) {
>> @@ -18,7 +18,7 @@ define void @atomic_sub_local_const_offset(i32 addrspace(3)* %local) {
>> ret void
>> }
>>
>> -; FUNC-LABEL: @atomic_sub_ret_local
>> +; FUNC-LABEL: atomic_sub_ret_local
>> ; R600: LDS_SUB_RET *
>> ; SI: DS_SUB_RTN_U32
>> define void @atomic_sub_ret_local(i32 addrspace(1)* %out, i32 addrspace(3)* %local) {
>> @@ -27,7 +27,7 @@ define void @atomic_sub_ret_local(i32 addrspace(1)* %out, i32 addrspace(3)* %loc
>> ret void
>> }
>>
>> -; FUNC-LABEL: @atomic_sub_ret_local_const_offset
>> +; FUNC-LABEL: atomic_sub_ret_local_const_offset
>> ; R600: LDS_SUB_RET *
>> ; SI: DS_SUB_RTN_U32 v{{[0-9]+}}, v{{[0-9]+}}, v{{[0-9]+}}, 0x14
>> define void @atomic_sub_ret_local_const_offset(i32 addrspace(1)* %out, i32 addrspace(3)* %local) {
>> diff --git a/test/CodeGen/R600/basic-branch.ll b/test/CodeGen/R600/basic-branch.ll
>> index d084132..29f8481 100644
>> --- a/test/CodeGen/R600/basic-branch.ll
>> +++ b/test/CodeGen/R600/basic-branch.ll
>> @@ -1,7 +1,7 @@
>> ; XFAIL: *
>> ; RUN: llc -O0 -march=r600 -mcpu=SI -verify-machineinstrs < %s | FileCheck %s
>>
>> -; CHECK-LABEL: @test_branch(
>> +; CHECK-LABEL: test_branch(
>> define void @test_branch(i32 addrspace(1)* noalias %out, i32 addrspace(1)* noalias %in, i32 %val) nounwind {
>> %cmp = icmp ne i32 %val, 0
>> br i1 %cmp, label %store, label %end
>> diff --git a/test/CodeGen/R600/basic-loop.ll b/test/CodeGen/R600/basic-loop.ll
>> index 6d0ff07..7d8c71e 100644
>> --- a/test/CodeGen/R600/basic-loop.ll
>> +++ b/test/CodeGen/R600/basic-loop.ll
>> @@ -1,7 +1,7 @@
>> ; XFAIL: *
>> ; RUN: llc -O0 -verify-machineinstrs -march=r600 -mcpu=SI < %s | FileCheck %s
>>
>> -; CHECK-LABEL: @test_loop:
>> +; CHECK-LABEL: test_loop:
>> define void @test_loop(i32 addrspace(1)* noalias %out, i32 addrspace(1)* noalias %in, i32 %val) nounwind {
>> entry:
>> br label %loop.body
>> diff --git a/test/CodeGen/R600/bfe_uint.ll b/test/CodeGen/R600/bfe_uint.ll
>> index fe466e6..5c1153f 100644
>> --- a/test/CodeGen/R600/bfe_uint.ll
>> +++ b/test/CodeGen/R600/bfe_uint.ll
>> @@ -2,7 +2,7 @@
>>
>> ; XFAIL: *
>>
>> -; CHECK: @bfe_def
>> +; CHECK: bfe_def
>> ; CHECK: BFE_UINT
>> define void @bfe_def(i32 addrspace(1)* %out, i32 %x) {
>> entry:
>> @@ -17,7 +17,7 @@ entry:
>> ; implmented with a LSHR instruction, which is better, because LSHR has less
>> ; operands and requires less constants.
>>
>> -; CHECK: @bfe_shift
>> +; CHECK: bfe_shift
>> ; CHECK-NOT: BFE_UINT
>> define void @bfe_shift(i32 addrspace(1)* %out, i32 %x) {
>> entry:
>> diff --git a/test/CodeGen/R600/bfi_int.ll b/test/CodeGen/R600/bfi_int.ll
>> index 107158d..7f2059d 100644
>> --- a/test/CodeGen/R600/bfi_int.ll
>> +++ b/test/CodeGen/R600/bfi_int.ll
>> @@ -4,7 +4,7 @@
>> ; BFI_INT Definition pattern from ISA docs
>> ; (y & x) | (z & ~x)
>> ;
>> -; R600-CHECK: @bfi_def
>> +; R600-CHECK: bfi_def
>> ; R600-CHECK: BFI_INT
>> ; SI-CHECK: @bfi_def
>> ; SI-CHECK: V_BFI_B32
>> @@ -20,7 +20,7 @@ entry:
>>
>> ; SHA-256 Ch function
>> ; z ^ (x & (y ^ z))
>> -; R600-CHECK: @bfi_sha256_ch
>> +; R600-CHECK: bfi_sha256_ch
>> ; R600-CHECK: BFI_INT
>> ; SI-CHECK: @bfi_sha256_ch
>> ; SI-CHECK: V_BFI_B32
>> @@ -35,7 +35,7 @@ entry:
>>
>> ; SHA-256 Ma function
>> ; ((x & z) | (y & (x | z)))
>> -; R600-CHECK: @bfi_sha256_ma
>> +; R600-CHECK: bfi_sha256_ma
>> ; R600-CHECK: XOR_INT * [[DST:T[0-9]+\.[XYZW]]], KC0[2].Z, KC0[2].W
>> ; R600-CHECK: BFI_INT * {{T[0-9]+\.[XYZW]}}, {{[[DST]]|PV\.[XYZW]}}, KC0[3].X, KC0[2].W
>> ; SI-CHECK: V_XOR_B32_e32 [[DST:v[0-9]+]], {{s[0-9]+, v[0-9]+}}
>> diff --git a/test/CodeGen/R600/bitcast.ll b/test/CodeGen/R600/bitcast.ll
>> index 0be79e6..51e8f83 100644
>> --- a/test/CodeGen/R600/bitcast.ll
>> +++ b/test/CodeGen/R600/bitcast.ll
>> @@ -4,7 +4,7 @@
>>
>> declare void @llvm.SI.export(i32, i32, i32, i32, i32, float, float, float, float)
>>
>> -; FUNC-LABEL: @v32i8_to_v8i32
>> +; FUNC-LABEL: v32i8_to_v8i32
>> ; SI: S_ENDPGM
>> define void @v32i8_to_v8i32(<32 x i8> addrspace(2)* inreg) #0 {
>> entry:
>> @@ -17,7 +17,7 @@ entry:
>> ret void
>> }
>>
>> -; FUNC-LABEL: @i8ptr_v16i8ptr
>> +; FUNC-LABEL: i8ptr_v16i8ptr
>> ; SI: S_ENDPGM
>> define void @i8ptr_v16i8ptr(<16 x i8> addrspace(1)* %out, i8 addrspace(1)* %in) {
>> entry:
>> @@ -55,7 +55,7 @@ define void @i32_to_v4i8(<4 x i8> addrspace(1)* %out, i32 addrspace(1)* %in) nou
>> ret void
>> }
>>
>> -; FUNC-LABEL: @bitcast_v2i32_to_f64
>> +; FUNC-LABEL: bitcast_v2i32_to_f64
>> ; SI: S_ENDPGM
>> define void @bitcast_v2i32_to_f64(double addrspace(1)* %out, <2 x i32> addrspace(1)* %in) {
>> %val = load <2 x i32> addrspace(1)* %in, align 8
>> @@ -65,7 +65,7 @@ define void @bitcast_v2i32_to_f64(double addrspace(1)* %out, <2 x i32> addrspace
>> ret void
>> }
>>
>> -; FUNC-LABEL: @bitcast_f64_to_v2i32
>> +; FUNC-LABEL: bitcast_f64_to_v2i32
>> ; SI: S_ENDPGM
>> define void @bitcast_f64_to_v2i32(<2 x i32> addrspace(1)* %out, double addrspace(1)* %in) {
>> %val = load double addrspace(1)* %in, align 8
>> diff --git a/test/CodeGen/R600/build_vector.ll b/test/CodeGen/R600/build_vector.ll
>> index 8179de1..cf31a2b 100644
>> --- a/test/CodeGen/R600/build_vector.ll
>> +++ b/test/CodeGen/R600/build_vector.ll
>> @@ -1,11 +1,11 @@
>> ; RUN: llc < %s -march=r600 -mcpu=redwood | FileCheck %s --check-prefix=R600-CHECK
>> ; RUN: llc < %s -march=r600 -mcpu=SI -verify-machineinstrs | FileCheck %s --check-prefix=SI-CHECK
>>
>> -; R600-CHECK: @build_vector2
>> +; R600-CHECK: build_vector2
>> ; R600-CHECK: MOV
>> ; R600-CHECK: MOV
>> ; R600-CHECK-NOT: MOV
>> -; SI-CHECK: @build_vector2
>> +; SI-CHECK: build_vector2
>> ; SI-CHECK-DAG: V_MOV_B32_e32 v[[X:[0-9]]], 5
>> ; SI-CHECK-DAG: V_MOV_B32_e32 v[[Y:[0-9]]], 6
>> ; SI-CHECK: BUFFER_STORE_DWORDX2 v{{\[}}[[X]]:[[Y]]{{\]}}
>> @@ -15,13 +15,13 @@ entry:
>> ret void
>> }
>>
>> -; R600-CHECK: @build_vector4
>> +; R600-CHECK: build_vector4
>> ; R600-CHECK: MOV
>> ; R600-CHECK: MOV
>> ; R600-CHECK: MOV
>> ; R600-CHECK: MOV
>> ; R600-CHECK-NOT: MOV
>> -; SI-CHECK: @build_vector4
>> +; SI-CHECK: build_vector4
>> ; SI-CHECK-DAG: V_MOV_B32_e32 v[[X:[0-9]]], 5
>> ; SI-CHECK-DAG: V_MOV_B32_e32 v[[Y:[0-9]]], 6
>> ; SI-CHECK-DAG: V_MOV_B32_e32 v[[Z:[0-9]]], 7
>> diff --git a/test/CodeGen/R600/call_fs.ll b/test/CodeGen/R600/call_fs.ll
>> index f7c4e5b..c4673b3 100644
>> --- a/test/CodeGen/R600/call_fs.ll
>> +++ b/test/CodeGen/R600/call_fs.ll
>> @@ -2,10 +2,10 @@
>> ; RUN: llc < %s -march=r600 -mcpu=redwood -show-mc-encoding -o - | FileCheck --check-prefix=EG-CHECK %s
>> ; RUN: llc < %s -march=r600 -mcpu=rv710 -show-mc-encoding -o - | FileCheck --check-prefix=R600-CHECK %s
>>
>> -; EG-CHECK: @call_fs
>> +; EG-CHECK: call_fs
>> ; EG-CHECK: .long 257
>> ; EG-CHECK: CALL_FS ; encoding: [0x00,0x00,0x00,0x00,0x00,0x00,0xc0,0x84]
>> -; R600-CHECK: @call_fs
>> +; R600-CHECK: call_fs
>> ; R600-CHECK: .long 257
>> ; R600-CHECK:CALL_FS ; encoding: [0x00,0x00,0x00,0x00,0x00,0x00,0x80,0x89]
>>
>> diff --git a/test/CodeGen/R600/cayman-loop-bug.ll b/test/CodeGen/R600/cayman-loop-bug.ll
>> index a873528..47a3420 100644
>> --- a/test/CodeGen/R600/cayman-loop-bug.ll
>> +++ b/test/CodeGen/R600/cayman-loop-bug.ll
>> @@ -1,6 +1,6 @@
>> ; RUN: llc < %s -march=r600 -mcpu=cayman | FileCheck %s
>>
>> -; CHECK-LABEL: @main
>> +; CHECK-LABEL: main
>> ; CHECK: LOOP_START_DX10
>> ; CHECK: ALU_PUSH_BEFORE
>> ; CHECK: LOOP_START_DX10
>> diff --git a/test/CodeGen/R600/cf-stack-bug.ll b/test/CodeGen/R600/cf-stack-bug.ll
>> index c3a4612..216a3bc 100644
>> --- a/test/CodeGen/R600/cf-stack-bug.ll
>> +++ b/test/CodeGen/R600/cf-stack-bug.ll
>> @@ -17,7 +17,7 @@
>> ; BUG64-NOT: Applying bug work-around
>> ; BUG32-NOT: Applying bug work-around
>> ; NOBUG-NOT: Applying bug work-around
>> -; FUNC-LABEL: @nested3
>> +; FUNC-LABEL: nested3
>> define void @nested3(i32 addrspace(1)* %out, i32 %cond) {
>> entry:
>> %0 = icmp sgt i32 %cond, 0
>> @@ -50,7 +50,7 @@ end:
>> ; BUG64: Applying bug work-around
>> ; BUG32-NOT: Applying bug work-around
>> ; NOBUG-NOT: Applying bug work-around
>> -; FUNC-LABEL: @nested4
>> +; FUNC-LABEL: nested4
>> define void @nested4(i32 addrspace(1)* %out, i32 %cond) {
>> entry:
>> %0 = icmp sgt i32 %cond, 0
>> @@ -91,7 +91,7 @@ end:
>> ; BUG64: Applying bug work-around
>> ; BUG32-NOT: Applying bug work-around
>> ; NOBUG-NOT: Applying bug work-around
>> -; FUNC-LABEL: @nested7
>> +; FUNC-LABEL: nested7
>> define void @nested7(i32 addrspace(1)* %out, i32 %cond) {
>> entry:
>> %0 = icmp sgt i32 %cond, 0
>> @@ -156,7 +156,7 @@ end:
>> ; BUG64: Applying bug work-around
>> ; BUG32: Applying bug work-around
>> ; NOBUG-NOT: Applying bug work-around
>> -; FUNC-LABEL: @nested8
>> +; FUNC-LABEL: nested8
>> define void @nested8(i32 addrspace(1)* %out, i32 %cond) {
>> entry:
>> %0 = icmp sgt i32 %cond, 0
>> diff --git a/test/CodeGen/R600/codegen-prepare-addrmode-sext.ll b/test/CodeGen/R600/codegen-prepare-addrmode-sext.ll
>> index f5d731d..825316e 100644
>> --- a/test/CodeGen/R600/codegen-prepare-addrmode-sext.ll
>> +++ b/test/CodeGen/R600/codegen-prepare-addrmode-sext.ll
>> @@ -4,7 +4,7 @@
>> target datalayout = "e-p:32:32-p1:64:64-p2:64:64-p3:32:32-p4:32:32-p5:64:64-i64:64-v16:16-v24:32-v32:32-v48:64-v96:128-v192:256-v256:256-v512:512-v1024:1024-v2048:2048-n32:64"
>> target triple = "r600--"
>>
>> -; FUNC-LABEL: @test
>> +; FUNC-LABEL: test
>> ; OPT: mul nsw i32
>> ; OPT-NEXT: sext
>> ; SI-LLC: S_MUL_I32
>> diff --git a/test/CodeGen/R600/combine_vloads.ll b/test/CodeGen/R600/combine_vloads.ll
>> index f8ec712..585dfa9 100644
>> --- a/test/CodeGen/R600/combine_vloads.ll
>> +++ b/test/CodeGen/R600/combine_vloads.ll
>> @@ -9,7 +9,7 @@
>>
>>
>> ; 128-bit loads instead of many 8-bit
>> -; EG-LABEL: @combine_vloads:
>> +; EG-LABEL: combine_vloads:
>> ; EG: VTX_READ_128
>> ; EG: VTX_READ_128
>> define void @combine_vloads(<8 x i8> addrspace(1)* nocapture %src, <8 x i8> addrspace(1)* nocapture %result) nounwind {
>> diff --git a/test/CodeGen/R600/complex-folding.ll b/test/CodeGen/R600/complex-folding.ll
>> index 99f0d99..a2964e5 100644
>> --- a/test/CodeGen/R600/complex-folding.ll
>> +++ b/test/CodeGen/R600/complex-folding.ll
>> @@ -1,6 +1,6 @@
>> ;RUN: llc < %s -march=r600 -mcpu=redwood | FileCheck %s
>>
>> -; CHECK: @main
>> +; CHECK: main
>> ; CHECK-NOT: MOV
>> define void @main(<4 x float> inreg %reg0) #0 {
>> entry:
>> diff --git a/test/CodeGen/R600/concat_vectors.ll b/test/CodeGen/R600/concat_vectors.ll
>> index 753282b..a11ec8e 100644
>> --- a/test/CodeGen/R600/concat_vectors.ll
>> +++ b/test/CodeGen/R600/concat_vectors.ll
>> @@ -1,6 +1,6 @@
>> ; RUN: llc -march=r600 -mcpu=SI -verify-machineinstrs < %s | FileCheck -check-prefix=SI -check-prefix=FUNC %s
>>
>> -; FUNC-LABEL: @test_concat_v1i32
>> +; FUNC-LABEL: test_concat_v1i32
>> ; 0x80f000 is the high 32 bits of the resource descriptor used by MUBUF
>> ; instructions that access scratch memory. Bit 23, which is the add_tid_enable
>> ; bit, is only set for scratch access, so we can check for the absence of this
>> @@ -13,7 +13,7 @@ define void @test_concat_v1i32(<2 x i32> addrspace(1)* %out, <1 x i32> %a, <1 x
>> ret void
>> }
>>
>> -; FUNC-LABEL: @test_concat_v2i32
>> +; FUNC-LABEL: test_concat_v2i32
>> ; SI-NOT: S_MOV_B32 s{{[0-9]}}, 0x80f000
>> ; SI-NOT: MOVREL
>> define void @test_concat_v2i32(<4 x i32> addrspace(1)* %out, <2 x i32> %a, <2 x i32> %b) nounwind {
>> @@ -22,7 +22,7 @@ define void @test_concat_v2i32(<4 x i32> addrspace(1)* %out, <2 x i32> %a, <2 x
>> ret void
>> }
>>
>> -; FUNC-LABEL: @test_concat_v4i32
>> +; FUNC-LABEL: test_concat_v4i32
>> ; SI-NOT: S_MOV_B32 s{{[0-9]}}, 0x80f000
>> ; SI-NOT: MOVREL
>> define void @test_concat_v4i32(<8 x i32> addrspace(1)* %out, <4 x i32> %a, <4 x i32> %b) nounwind {
>> @@ -31,7 +31,7 @@ define void @test_concat_v4i32(<8 x i32> addrspace(1)* %out, <4 x i32> %a, <4 x
>> ret void
>> }
>>
>> -; FUNC-LABEL: @test_concat_v8i32
>> +; FUNC-LABEL: test_concat_v8i32
>> ; SI-NOT: S_MOV_B32 s{{[0-9]}}, 0x80f000
>> ; SI-NOT: MOVREL
>> define void @test_concat_v8i32(<16 x i32> addrspace(1)* %out, <8 x i32> %a, <8 x i32> %b) nounwind {
>> @@ -40,7 +40,7 @@ define void @test_concat_v8i32(<16 x i32> addrspace(1)* %out, <8 x i32> %a, <8 x
>> ret void
>> }
>>
>> -; FUNC-LABEL: @test_concat_v16i32
>> +; FUNC-LABEL: test_concat_v16i32
>> ; SI-NOT: S_MOV_B32 s{{[0-9]}}, 0x80f000
>> ; SI-NOT: MOVREL
>> define void @test_concat_v16i32(<32 x i32> addrspace(1)* %out, <16 x i32> %a, <16 x i32> %b) nounwind {
>> @@ -49,7 +49,7 @@ define void @test_concat_v16i32(<32 x i32> addrspace(1)* %out, <16 x i32> %a, <1
>> ret void
>> }
>>
>> -; FUNC-LABEL: @test_concat_v1f32
>> +; FUNC-LABEL: test_concat_v1f32
>> ; SI-NOT: S_MOV_B32 s{{[0-9]}}, 0x80f000
>> ; SI-NOT: MOVREL
>> define void @test_concat_v1f32(<2 x float> addrspace(1)* %out, <1 x float> %a, <1 x float> %b) nounwind {
>> @@ -58,7 +58,7 @@ define void @test_concat_v1f32(<2 x float> addrspace(1)* %out, <1 x float> %a, <
>> ret void
>> }
>>
>> -; FUNC-LABEL: @test_concat_v2f32
>> +; FUNC-LABEL: test_concat_v2f32
>> ; SI-NOT: S_MOV_B32 s{{[0-9]}}, 0x80f000
>> ; SI-NOT: MOVREL
>> define void @test_concat_v2f32(<4 x float> addrspace(1)* %out, <2 x float> %a, <2 x float> %b) nounwind {
>> @@ -67,7 +67,7 @@ define void @test_concat_v2f32(<4 x float> addrspace(1)* %out, <2 x float> %a, <
>> ret void
>> }
>>
>> -; FUNC-LABEL: @test_concat_v4f32
>> +; FUNC-LABEL: test_concat_v4f32
>> ; SI-NOT: S_MOV_B32 s{{[0-9]}}, 0x80f000
>> ; SI-NOT: MOVREL
>> define void @test_concat_v4f32(<8 x float> addrspace(1)* %out, <4 x float> %a, <4 x float> %b) nounwind {
>> @@ -76,7 +76,7 @@ define void @test_concat_v4f32(<8 x float> addrspace(1)* %out, <4 x float> %a, <
>> ret void
>> }
>>
>> -; FUNC-LABEL: @test_concat_v8f32
>> +; FUNC-LABEL: test_concat_v8f32
>> ; SI-NOT: S_MOV_B32 s{{[0-9]}}, 0x80f000
>> ; SI-NOT: MOVREL
>> define void @test_concat_v8f32(<16 x float> addrspace(1)* %out, <8 x float> %a, <8 x float> %b) nounwind {
>> @@ -85,7 +85,7 @@ define void @test_concat_v8f32(<16 x float> addrspace(1)* %out, <8 x float> %a,
>> ret void
>> }
>>
>> -; FUNC-LABEL: @test_concat_v16f32
>> +; FUNC-LABEL: test_concat_v16f32
>> ; SI-NOT: S_MOV_B32 s{{[0-9]}}, 0x80f000
>> ; SI-NOT: MOVREL
>> define void @test_concat_v16f32(<32 x float> addrspace(1)* %out, <16 x float> %a, <16 x float> %b) nounwind {
>> @@ -94,7 +94,7 @@ define void @test_concat_v16f32(<32 x float> addrspace(1)* %out, <16 x float> %a
>> ret void
>> }
>>
>> -; FUNC-LABEL: @test_concat_v1i64
>> +; FUNC-LABEL: test_concat_v1i64
>> ; SI-NOT: S_MOV_B32 s{{[0-9]}}, 0x80f000
>> ; SI-NOT: MOVREL
>> define void @test_concat_v1i64(<2 x double> addrspace(1)* %out, <1 x double> %a, <1 x double> %b) nounwind {
>> @@ -103,7 +103,7 @@ define void @test_concat_v1i64(<2 x double> addrspace(1)* %out, <1 x double> %a,
>> ret void
>> }
>>
>> -; FUNC-LABEL: @test_concat_v2i64
>> +; FUNC-LABEL: test_concat_v2i64
>> ; SI-NOT: S_MOV_B32 s{{[0-9]}}, 0x80f000
>> ; SI-NOT: MOVREL
>> define void @test_concat_v2i64(<4 x double> addrspace(1)* %out, <2 x double> %a, <2 x double> %b) nounwind {
>> @@ -112,7 +112,7 @@ define void @test_concat_v2i64(<4 x double> addrspace(1)* %out, <2 x double> %a,
>> ret void
>> }
>>
>> -; FUNC-LABEL: @test_concat_v4i64
>> +; FUNC-LABEL: test_concat_v4i64
>> ; SI-NOT: S_MOV_B32 s{{[0-9]}}, 0x80f000
>> ; SI-NOT: MOVREL
>> define void @test_concat_v4i64(<8 x double> addrspace(1)* %out, <4 x double> %a, <4 x double> %b) nounwind {
>> @@ -121,7 +121,7 @@ define void @test_concat_v4i64(<8 x double> addrspace(1)* %out, <4 x double> %a,
>> ret void
>> }
>>
>> -; FUNC-LABEL: @test_concat_v8i64
>> +; FUNC-LABEL: test_concat_v8i64
>> ; SI-NOT: S_MOV_B32 s{{[0-9]}}, 0x80f000
>> ; SI-NOT: MOVREL
>> define void @test_concat_v8i64(<16 x double> addrspace(1)* %out, <8 x double> %a, <8 x double> %b) nounwind {
>> @@ -130,7 +130,7 @@ define void @test_concat_v8i64(<16 x double> addrspace(1)* %out, <8 x double> %a
>> ret void
>> }
>>
>> -; FUNC-LABEL: @test_concat_v16i64
>> +; FUNC-LABEL: test_concat_v16i64
>> ; SI-NOT: S_MOV_B32 s{{[0-9]}}, 0x80f000
>> ; SI-NOT: MOVREL
>> define void @test_concat_v16i64(<32 x double> addrspace(1)* %out, <16 x double> %a, <16 x double> %b) nounwind {
>> @@ -139,7 +139,7 @@ define void @test_concat_v16i64(<32 x double> addrspace(1)* %out, <16 x double>
>> ret void
>> }
>>
>> -; FUNC-LABEL: @test_concat_v1f64
>> +; FUNC-LABEL: test_concat_v1f64
>> ; SI-NOT: S_MOV_B32 s{{[0-9]}}, 0x80f000
>> ; SI-NOT: MOVREL
>> define void @test_concat_v1f64(<2 x double> addrspace(1)* %out, <1 x double> %a, <1 x double> %b) nounwind {
>> @@ -148,7 +148,7 @@ define void @test_concat_v1f64(<2 x double> addrspace(1)* %out, <1 x double> %a,
>> ret void
>> }
>>
>> -; FUNC-LABEL: @test_concat_v2f64
>> +; FUNC-LABEL: test_concat_v2f64
>> ; SI-NOT: S_MOV_B32 s{{[0-9]}}, 0x80f000
>> ; SI-NOT: MOVREL
>> define void @test_concat_v2f64(<4 x double> addrspace(1)* %out, <2 x double> %a, <2 x double> %b) nounwind {
>> @@ -157,7 +157,7 @@ define void @test_concat_v2f64(<4 x double> addrspace(1)* %out, <2 x double> %a,
>> ret void
>> }
>>
>> -; FUNC-LABEL: @test_concat_v4f64
>> +; FUNC-LABEL: test_concat_v4f64
>> ; SI-NOT: S_MOV_B32 s{{[0-9]}}, 0x80f000
>> ; SI-NOT: MOVREL
>> define void @test_concat_v4f64(<8 x double> addrspace(1)* %out, <4 x double> %a, <4 x double> %b) nounwind {
>> @@ -166,7 +166,7 @@ define void @test_concat_v4f64(<8 x double> addrspace(1)* %out, <4 x double> %a,
>> ret void
>> }
>>
>> -; FUNC-LABEL: @test_concat_v8f64
>> +; FUNC-LABEL: test_concat_v8f64
>> ; SI-NOT: S_MOV_B32 s{{[0-9]}}, 0x80f000
>> ; SI-NOT: MOVREL
>> define void @test_concat_v8f64(<16 x double> addrspace(1)* %out, <8 x double> %a, <8 x double> %b) nounwind {
>> @@ -175,7 +175,7 @@ define void @test_concat_v8f64(<16 x double> addrspace(1)* %out, <8 x double> %a
>> ret void
>> }
>>
>> -; FUNC-LABEL: @test_concat_v16f64
>> +; FUNC-LABEL: test_concat_v16f64
>> ; SI-NOT: S_MOV_B32 s{{[0-9]}}, 0x80f000
>> ; SI-NOT: MOVREL
>> define void @test_concat_v16f64(<32 x double> addrspace(1)* %out, <16 x double> %a, <16 x double> %b) nounwind {
>> @@ -184,7 +184,7 @@ define void @test_concat_v16f64(<32 x double> addrspace(1)* %out, <16 x double>
>> ret void
>> }
>>
>> -; FUNC-LABEL: @test_concat_v1i1
>> +; FUNC-LABEL: test_concat_v1i1
>> ; SI-NOT: S_MOV_B32 s{{[0-9]}}, 0x80f000
>> ; SI-NOT: MOVREL
>> define void @test_concat_v1i1(<2 x i1> addrspace(1)* %out, <1 x i1> %a, <1 x i1> %b) nounwind {
>> @@ -193,7 +193,7 @@ define void @test_concat_v1i1(<2 x i1> addrspace(1)* %out, <1 x i1> %a, <1 x i1>
>> ret void
>> }
>>
>> -; FUNC-LABEL: @test_concat_v2i1
>> +; FUNC-LABEL: test_concat_v2i1
>> ; SI-NOT: S_MOV_B32 s{{[0-9]}}, 0x80f000
>> ; SI-NOT: MOVREL
>> define void @test_concat_v2i1(<4 x i1> addrspace(1)* %out, <2 x i1> %a, <2 x i1> %b) nounwind {
>> @@ -202,7 +202,7 @@ define void @test_concat_v2i1(<4 x i1> addrspace(1)* %out, <2 x i1> %a, <2 x i1>
>> ret void
>> }
>>
>> -; FUNC-LABEL: @test_concat_v4i1
>> +; FUNC-LABEL: test_concat_v4i1
>> ; SI-NOT: S_MOV_B32 s{{[0-9]}}, 0x80f000
>> ; SI-NOT: MOVREL
>> define void @test_concat_v4i1(<8 x i1> addrspace(1)* %out, <4 x i1> %a, <4 x i1> %b) nounwind {
>> @@ -211,7 +211,7 @@ define void @test_concat_v4i1(<8 x i1> addrspace(1)* %out, <4 x i1> %a, <4 x i1>
>> ret void
>> }
>>
>> -; FUNC-LABEL: @test_concat_v8i1
>> +; FUNC-LABEL: test_concat_v8i1
>> ; SI-NOT: S_MOV_B32 s{{[0-9]}}, 0x80f000
>> ; SI-NOT: MOVREL
>> define void @test_concat_v8i1(<16 x i1> addrspace(1)* %out, <8 x i1> %a, <8 x i1> %b) nounwind {
>> @@ -220,7 +220,7 @@ define void @test_concat_v8i1(<16 x i1> addrspace(1)* %out, <8 x i1> %a, <8 x i1
>> ret void
>> }
>>
>> -; FUNC-LABEL: @test_concat_v16i1
>> +; FUNC-LABEL: test_concat_v16i1
>> ; SI-NOT: S_MOV_B32 s{{[0-9]}}, 0x80f000
>> ; SI-NOT: MOVREL
>> define void @test_concat_v16i1(<32 x i1> addrspace(1)* %out, <16 x i1> %a, <16 x i1> %b) nounwind {
>> @@ -229,7 +229,7 @@ define void @test_concat_v16i1(<32 x i1> addrspace(1)* %out, <16 x i1> %a, <16 x
>> ret void
>> }
>>
>> -; FUNC-LABEL: @test_concat_v32i1
>> +; FUNC-LABEL: test_concat_v32i1
>> ; SI-NOT: S_MOV_B32 s{{[0-9]}}, 0x80f000
>> ; SI-NOT: MOVREL
>> define void @test_concat_v32i1(<64 x i1> addrspace(1)* %out, <32 x i1> %a, <32 x i1> %b) nounwind {
>> @@ -238,7 +238,7 @@ define void @test_concat_v32i1(<64 x i1> addrspace(1)* %out, <32 x i1> %a, <32 x
>> ret void
>> }
>>
>> -; FUNC-LABEL: @test_concat_v1i16
>> +; FUNC-LABEL: test_concat_v1i16
>> ; SI-NOT: S_MOV_B32 s{{[0-9]}}, 0x80f000
>> ; SI-NOT: MOVREL
>> define void @test_concat_v1i16(<2 x i16> addrspace(1)* %out, <1 x i16> %a, <1 x i16> %b) nounwind {
>> @@ -247,7 +247,7 @@ define void @test_concat_v1i16(<2 x i16> addrspace(1)* %out, <1 x i16> %a, <1 x
>> ret void
>> }
>>
>> -; FUNC-LABEL: @test_concat_v2i16
>> +; FUNC-LABEL: test_concat_v2i16
>> ; SI-NOT: S_MOV_B32 s{{[0-9]}}, 0x80f000
>> ; SI-NOT: MOVREL
>> define void @test_concat_v2i16(<4 x i16> addrspace(1)* %out, <2 x i16> %a, <2 x i16> %b) nounwind {
>> @@ -256,7 +256,7 @@ define void @test_concat_v2i16(<4 x i16> addrspace(1)* %out, <2 x i16> %a, <2 x
>> ret void
>> }
>>
>> -; FUNC-LABEL: @test_concat_v4i16
>> +; FUNC-LABEL: test_concat_v4i16
>> ; SI-NOT: S_MOV_B32 s{{[0-9]}}, 0x80f000
>> ; SI-NOT: MOVREL
>> define void @test_concat_v4i16(<8 x i16> addrspace(1)* %out, <4 x i16> %a, <4 x i16> %b) nounwind {
>> @@ -265,7 +265,7 @@ define void @test_concat_v4i16(<8 x i16> addrspace(1)* %out, <4 x i16> %a, <4 x
>> ret void
>> }
>>
>> -; FUNC-LABEL: @test_concat_v8i16
>> +; FUNC-LABEL: test_concat_v8i16
>> ; SI-NOT: S_MOV_B32 s{{[0-9]}}, 0x80f000
>> ; SI-NOT: MOVREL
>> define void @test_concat_v8i16(<16 x i16> addrspace(1)* %out, <8 x i16> %a, <8 x i16> %b) nounwind {
>> @@ -274,7 +274,7 @@ define void @test_concat_v8i16(<16 x i16> addrspace(1)* %out, <8 x i16> %a, <8 x
>> ret void
>> }
>>
>> -; FUNC-LABEL: @test_concat_v16i16
>> +; FUNC-LABEL: test_concat_v16i16
>> ; SI-NOT: S_MOV_B32 s{{[0-9]}}, 0x80f000
>> ; SI-NOT: MOVREL
>> define void @test_concat_v16i16(<32 x i16> addrspace(1)* %out, <16 x i16> %a, <16 x i16> %b) nounwind {
>> diff --git a/test/CodeGen/R600/copy-illegal-type.ll b/test/CodeGen/R600/copy-illegal-type.ll
>> index f7c2321..84c7024 100644
>> --- a/test/CodeGen/R600/copy-illegal-type.ll
>> +++ b/test/CodeGen/R600/copy-illegal-type.ll
>> @@ -1,6 +1,6 @@
>> ; RUN: llc -march=r600 -mcpu=tahiti < %s | FileCheck -check-prefix=SI -check-prefix=FUNC %s
>>
>> -; FUNC-LABEL: @test_copy_v4i8
>> +; FUNC-LABEL: test_copy_v4i8
>> ; SI: BUFFER_LOAD_DWORD [[REG:v[0-9]+]]
>> ; SI: BUFFER_STORE_DWORD [[REG]]
>> ; SI: S_ENDPGM
>> @@ -10,7 +10,7 @@ define void @test_copy_v4i8(<4 x i8> addrspace(1)* %out, <4 x i8> addrspace(1)*
>> ret void
>> }
>>
>> -; FUNC-LABEL: @test_copy_v4i8_x2
>> +; FUNC-LABEL: test_copy_v4i8_x2
>> ; SI: BUFFER_LOAD_DWORD [[REG:v[0-9]+]]
>> ; SI: BUFFER_STORE_DWORD [[REG]]
>> ; SI: BUFFER_STORE_DWORD [[REG]]
>> @@ -22,7 +22,7 @@ define void @test_copy_v4i8_x2(<4 x i8> addrspace(1)* %out0, <4 x i8> addrspace(
>> ret void
>> }
>>
>> -; FUNC-LABEL: @test_copy_v4i8_x3
>> +; FUNC-LABEL: test_copy_v4i8_x3
>> ; SI: BUFFER_LOAD_DWORD [[REG:v[0-9]+]]
>> ; SI: BUFFER_STORE_DWORD [[REG]]
>> ; SI: BUFFER_STORE_DWORD [[REG]]
>> @@ -36,7 +36,7 @@ define void @test_copy_v4i8_x3(<4 x i8> addrspace(1)* %out0, <4 x i8> addrspace(
>> ret void
>> }
>>
>> -; FUNC-LABEL: @test_copy_v4i8_x4
>> +; FUNC-LABEL: test_copy_v4i8_x4
>> ; SI: BUFFER_LOAD_DWORD [[REG:v[0-9]+]]
>> ; SI: BUFFER_STORE_DWORD [[REG]]
>> ; SI: BUFFER_STORE_DWORD [[REG]]
>> @@ -52,7 +52,7 @@ define void @test_copy_v4i8_x4(<4 x i8> addrspace(1)* %out0, <4 x i8> addrspace(
>> ret void
>> }
>>
>> -; FUNC-LABEL: @test_copy_v4i8_extra_use
>> +; FUNC-LABEL: test_copy_v4i8_extra_use
>> ; SI: BUFFER_LOAD_UBYTE
>> ; SI: BUFFER_LOAD_UBYTE
>> ; SI: BUFFER_LOAD_UBYTE
>> @@ -88,7 +88,7 @@ define void @test_copy_v4i8_extra_use(<4 x i8> addrspace(1)* %out0, <4 x i8> add
>> ret void
>> }
>>
>> -; FUNC-LABEL: @test_copy_v4i8_x2_extra_use
>> +; FUNC-LABEL: test_copy_v4i8_x2_extra_use
>> ; SI: BUFFER_LOAD_UBYTE
>> ; SI: BUFFER_LOAD_UBYTE
>> ; SI: BUFFER_LOAD_UBYTE
>> @@ -127,7 +127,7 @@ define void @test_copy_v4i8_x2_extra_use(<4 x i8> addrspace(1)* %out0, <4 x i8>
>> ret void
>> }
>>
>> -; FUNC-LABEL: @test_copy_v3i8
>> +; FUNC-LABEL: test_copy_v3i8
>> ; SI-NOT: BFE
>> ; SI-NOT: BFI
>> ; SI: S_ENDPGM
>> @@ -137,7 +137,7 @@ define void @test_copy_v3i8(<3 x i8> addrspace(1)* %out, <3 x i8> addrspace(1)*
>> ret void
>> }
>>
>> -; FUNC-LABEL: @test_copy_v4i8_volatile_load
>> +; FUNC-LABEL: test_copy_v4i8_volatile_load
>> ; SI: BUFFER_LOAD_UBYTE
>> ; SI: BUFFER_LOAD_UBYTE
>> ; SI: BUFFER_LOAD_UBYTE
>> @@ -149,7 +149,7 @@ define void @test_copy_v4i8_volatile_load(<4 x i8> addrspace(1)* %out, <4 x i8>
>> ret void
>> }
>>
>> -; FUNC-LABEL: @test_copy_v4i8_volatile_store
>> +; FUNC-LABEL: test_copy_v4i8_volatile_store
>> ; SI: BUFFER_LOAD_UBYTE
>> ; SI: BUFFER_LOAD_UBYTE
>> ; SI: BUFFER_LOAD_UBYTE
>> diff --git a/test/CodeGen/R600/ctlz_zero_undef.ll b/test/CodeGen/R600/ctlz_zero_undef.ll
>> index 1340ef9..f04f70b 100644
>> --- a/test/CodeGen/R600/ctlz_zero_undef.ll
>> +++ b/test/CodeGen/R600/ctlz_zero_undef.ll
>> @@ -5,7 +5,7 @@ declare i32 @llvm.ctlz.i32(i32, i1) nounwind readnone
>> declare <2 x i32> @llvm.ctlz.v2i32(<2 x i32>, i1) nounwind readnone
>> declare <4 x i32> @llvm.ctlz.v4i32(<4 x i32>, i1) nounwind readnone
>>
>> -; FUNC-LABEL: @s_ctlz_zero_undef_i32:
>> +; FUNC-LABEL: s_ctlz_zero_undef_i32:
>> ; SI: S_LOAD_DWORD [[VAL:s[0-9]+]],
>> ; SI: S_FLBIT_I32_B32 [[SRESULT:s[0-9]+]], [[VAL]]
>> ; SI: V_MOV_B32_e32 [[VRESULT:v[0-9]+]], [[SRESULT]]
>> @@ -19,7 +19,7 @@ define void @s_ctlz_zero_undef_i32(i32 addrspace(1)* noalias %out, i32 %val) nou
>> ret void
>> }
>>
>> -; FUNC-LABEL: @v_ctlz_zero_undef_i32:
>> +; FUNC-LABEL: v_ctlz_zero_undef_i32:
>> ; SI: BUFFER_LOAD_DWORD [[VAL:v[0-9]+]],
>> ; SI: V_FFBH_U32_e32 [[RESULT:v[0-9]+]], [[VAL]]
>> ; SI: BUFFER_STORE_DWORD [[RESULT]],
>> @@ -33,7 +33,7 @@ define void @v_ctlz_zero_undef_i32(i32 addrspace(1)* noalias %out, i32 addrspace
>> ret void
>> }
>>
>> -; FUNC-LABEL: @v_ctlz_zero_undef_v2i32:
>> +; FUNC-LABEL: v_ctlz_zero_undef_v2i32:
>> ; SI: BUFFER_LOAD_DWORDX2
>> ; SI: V_FFBH_U32_e32
>> ; SI: V_FFBH_U32_e32
>> @@ -49,7 +49,7 @@ define void @v_ctlz_zero_undef_v2i32(<2 x i32> addrspace(1)* noalias %out, <2 x
>> ret void
>> }
>>
>> -; FUNC-LABEL: @v_ctlz_zero_undef_v4i32:
>> +; FUNC-LABEL: v_ctlz_zero_undef_v4i32:
>> ; SI: BUFFER_LOAD_DWORDX4
>> ; SI: V_FFBH_U32_e32
>> ; SI: V_FFBH_U32_e32
>> diff --git a/test/CodeGen/R600/ctpop.ll b/test/CodeGen/R600/ctpop.ll
>> index 61d4943..11f13a9 100644
>> --- a/test/CodeGen/R600/ctpop.ll
>> +++ b/test/CodeGen/R600/ctpop.ll
>> @@ -7,7 +7,7 @@ declare <4 x i32> @llvm.ctpop.v4i32(<4 x i32>) nounwind readnone
>> declare <8 x i32> @llvm.ctpop.v8i32(<8 x i32>) nounwind readnone
>> declare <16 x i32> @llvm.ctpop.v16i32(<16 x i32>) nounwind readnone
>>
>> -; FUNC-LABEL: @s_ctpop_i32:
>> +; FUNC-LABEL: s_ctpop_i32:
>> ; SI: S_LOAD_DWORD [[SVAL:s[0-9]+]],
>> ; SI: S_BCNT1_I32_B32 [[SRESULT:s[0-9]+]], [[SVAL]]
>> ; SI: V_MOV_B32_e32 [[VRESULT:v[0-9]+]], [[SRESULT]]
>> @@ -22,7 +22,7 @@ define void @s_ctpop_i32(i32 addrspace(1)* noalias %out, i32 %val) nounwind {
>> }
>>
>> ; XXX - Why 0 in register?
>> -; FUNC-LABEL: @v_ctpop_i32:
>> +; FUNC-LABEL: v_ctpop_i32:
>> ; SI: BUFFER_LOAD_DWORD [[VAL:v[0-9]+]],
>> ; SI: V_MOV_B32_e32 [[VZERO:v[0-9]+]], 0
>> ; SI: V_BCNT_U32_B32_e32 [[RESULT:v[0-9]+]], [[VAL]], [[VZERO]]
>> @@ -37,7 +37,7 @@ define void @v_ctpop_i32(i32 addrspace(1)* noalias %out, i32 addrspace(1)* noali
>> ret void
>> }
>>
>> -; FUNC-LABEL: @v_ctpop_add_chain_i32
>> +; FUNC-LABEL: v_ctpop_add_chain_i32
>> ; SI: BUFFER_LOAD_DWORD [[VAL0:v[0-9]+]],
>> ; SI: BUFFER_LOAD_DWORD [[VAL1:v[0-9]+]],
>> ; SI: V_MOV_B32_e32 [[VZERO:v[0-9]+]], 0
>> @@ -58,7 +58,7 @@ define void @v_ctpop_add_chain_i32(i32 addrspace(1)* noalias %out, i32 addrspace
>> ret void
>> }
>>
>> -; FUNC-LABEL: @v_ctpop_add_sgpr_i32
>> +; FUNC-LABEL: v_ctpop_add_sgpr_i32
>> ; SI: BUFFER_LOAD_DWORD [[VAL0:v[0-9]+]],
>> ; SI-NEXT: S_WAITCNT
>> ; SI-NEXT: V_BCNT_U32_B32_e64 [[RESULT:v[0-9]+]], [[VAL0]], s{{[0-9]+}}
>> @@ -72,7 +72,7 @@ define void @v_ctpop_add_sgpr_i32(i32 addrspace(1)* noalias %out, i32 addrspace(
>> ret void
>> }
>>
>> -; FUNC-LABEL: @v_ctpop_v2i32:
>> +; FUNC-LABEL: v_ctpop_v2i32:
>> ; SI: V_BCNT_U32_B32_e32
>> ; SI: V_BCNT_U32_B32_e32
>> ; SI: S_ENDPGM
>> @@ -86,7 +86,7 @@ define void @v_ctpop_v2i32(<2 x i32> addrspace(1)* noalias %out, <2 x i32> addrs
>> ret void
>> }
>>
>> -; FUNC-LABEL: @v_ctpop_v4i32:
>> +; FUNC-LABEL: v_ctpop_v4i32:
>> ; SI: V_BCNT_U32_B32_e32
>> ; SI: V_BCNT_U32_B32_e32
>> ; SI: V_BCNT_U32_B32_e32
>> @@ -104,7 +104,7 @@ define void @v_ctpop_v4i32(<4 x i32> addrspace(1)* noalias %out, <4 x i32> addrs
>> ret void
>> }
>>
>> -; FUNC-LABEL: @v_ctpop_v8i32:
>> +; FUNC-LABEL: v_ctpop_v8i32:
>> ; SI: V_BCNT_U32_B32_e32
>> ; SI: V_BCNT_U32_B32_e32
>> ; SI: V_BCNT_U32_B32_e32
>> @@ -130,7 +130,7 @@ define void @v_ctpop_v8i32(<8 x i32> addrspace(1)* noalias %out, <8 x i32> addrs
>> ret void
>> }
>>
>> -; FUNC-LABEL: @v_ctpop_v16i32:
>> +; FUNC-LABEL: v_ctpop_v16i32:
>> ; SI: V_BCNT_U32_B32_e32
>> ; SI: V_BCNT_U32_B32_e32
>> ; SI: V_BCNT_U32_B32_e32
>> @@ -172,7 +172,7 @@ define void @v_ctpop_v16i32(<16 x i32> addrspace(1)* noalias %out, <16 x i32> ad
>> ret void
>> }
>>
>> -; FUNC-LABEL: @v_ctpop_i32_add_inline_constant:
>> +; FUNC-LABEL: v_ctpop_i32_add_inline_constant:
>> ; SI: BUFFER_LOAD_DWORD [[VAL:v[0-9]+]],
>> ; SI: V_BCNT_U32_B32_e64 [[RESULT:v[0-9]+]], [[VAL]], 4
>> ; SI: BUFFER_STORE_DWORD [[RESULT]],
>> @@ -187,7 +187,7 @@ define void @v_ctpop_i32_add_inline_constant(i32 addrspace(1)* noalias %out, i32
>> ret void
>> }
>>
>> -; FUNC-LABEL: @v_ctpop_i32_add_inline_constant_inv:
>> +; FUNC-LABEL: v_ctpop_i32_add_inline_constant_inv:
>> ; SI: BUFFER_LOAD_DWORD [[VAL:v[0-9]+]],
>> ; SI: V_BCNT_U32_B32_e64 [[RESULT:v[0-9]+]], [[VAL]], 4
>> ; SI: BUFFER_STORE_DWORD [[RESULT]],
>> @@ -202,7 +202,7 @@ define void @v_ctpop_i32_add_inline_constant_inv(i32 addrspace(1)* noalias %out,
>> ret void
>> }
>>
>> -; FUNC-LABEL: @v_ctpop_i32_add_literal:
>> +; FUNC-LABEL: v_ctpop_i32_add_literal:
>> ; SI: BUFFER_LOAD_DWORD [[VAL:v[0-9]+]],
>> ; SI: V_MOV_B32_e32 [[LIT:v[0-9]+]], 0x1869f
>> ; SI: V_BCNT_U32_B32_e32 [[RESULT:v[0-9]+]], [[VAL]], [[LIT]]
>> @@ -216,7 +216,7 @@ define void @v_ctpop_i32_add_literal(i32 addrspace(1)* noalias %out, i32 addrspa
>> ret void
>> }
>>
>> -; FUNC-LABEL: @v_ctpop_i32_add_var:
>> +; FUNC-LABEL: v_ctpop_i32_add_var:
>> ; SI-DAG: BUFFER_LOAD_DWORD [[VAL:v[0-9]+]],
>> ; SI-DAG: S_LOAD_DWORD [[VAR:s[0-9]+]],
>> ; SI: V_BCNT_U32_B32_e64 [[RESULT:v[0-9]+]], [[VAL]], [[VAR]]
>> @@ -232,7 +232,7 @@ define void @v_ctpop_i32_add_var(i32 addrspace(1)* noalias %out, i32 addrspace(1
>> ret void
>> }
>>
>> -; FUNC-LABEL: @v_ctpop_i32_add_var_inv:
>> +; FUNC-LABEL: v_ctpop_i32_add_var_inv:
>> ; SI-DAG: BUFFER_LOAD_DWORD [[VAL:v[0-9]+]],
>> ; SI-DAG: S_LOAD_DWORD [[VAR:s[0-9]+]],
>> ; SI: V_BCNT_U32_B32_e64 [[RESULT:v[0-9]+]], [[VAL]], [[VAR]]
>> @@ -248,7 +248,7 @@ define void @v_ctpop_i32_add_var_inv(i32 addrspace(1)* noalias %out, i32 addrspa
>> ret void
>> }
>>
>> -; FUNC-LABEL: @v_ctpop_i32_add_vvar_inv
>> +; FUNC-LABEL: v_ctpop_i32_add_vvar_inv
>> ; SI-DAG: BUFFER_LOAD_DWORD [[VAL:v[0-9]+]], s[{{[0-9]+:[0-9]+}}], {{0$}}
>> ; SI-DAG: BUFFER_LOAD_DWORD [[VAR:v[0-9]+]], s[{{[0-9]+:[0-9]+}}], 0 offset:0x10
>> ; SI: V_BCNT_U32_B32_e32 [[RESULT:v[0-9]+]], [[VAL]], [[VAR]]
>> @@ -269,7 +269,7 @@ define void @v_ctpop_i32_add_vvar_inv(i32 addrspace(1)* noalias %out, i32 addrsp
>> ; FIXME: We currently disallow SALU instructions in all branches,
>> ; but there are some cases when the should be allowed.
>>
>> -; FUNC-LABEL: @ctpop_i32_in_br
>> +; FUNC-LABEL: ctpop_i32_in_br
>> ; SI: S_LOAD_DWORD [[VAL:s[0-9]+]], s[{{[0-9]+:[0-9]+}}], 0xd
>> ; SI: S_BCNT1_I32_B32 [[SRESULT:s[0-9]+]], [[VAL]]
>> ; SI: V_MOV_B32_e32 [[RESULT]], [[SRESULT]]
>> diff --git a/test/CodeGen/R600/ctpop64.ll b/test/CodeGen/R600/ctpop64.ll
>> index 76091c5..dd7f08c 100644
>> --- a/test/CodeGen/R600/ctpop64.ll
>> +++ b/test/CodeGen/R600/ctpop64.ll
>> @@ -6,7 +6,7 @@ declare <4 x i64> @llvm.ctpop.v4i64(<4 x i64>) nounwind readnone
>> declare <8 x i64> @llvm.ctpop.v8i64(<8 x i64>) nounwind readnone
>> declare <16 x i64> @llvm.ctpop.v16i64(<16 x i64>) nounwind readnone
>>
>> -; FUNC-LABEL: @s_ctpop_i64:
>> +; FUNC-LABEL: s_ctpop_i64:
>> ; SI: S_LOAD_DWORDX2 [[SVAL:s\[[0-9]+:[0-9]+\]]], s{{\[[0-9]+:[0-9]+\]}}, 0xb
>> ; SI: S_BCNT1_I32_B64 [[SRESULT:s[0-9]+]], [[SVAL]]
>> ; SI: V_MOV_B32_e32 [[VRESULT:v[0-9]+]], [[SRESULT]]
>> @@ -19,7 +19,7 @@ define void @s_ctpop_i64(i32 addrspace(1)* noalias %out, i64 %val) nounwind {
>> ret void
>> }
>>
>> -; FUNC-LABEL: @v_ctpop_i64:
>> +; FUNC-LABEL: v_ctpop_i64:
>> ; SI: BUFFER_LOAD_DWORDX2 v{{\[}}[[LOVAL:[0-9]+]]:[[HIVAL:[0-9]+]]{{\]}},
>> ; SI: V_MOV_B32_e32 [[VZERO:v[0-9]+]], 0
>> ; SI: V_BCNT_U32_B32_e32 [[MIDRESULT:v[0-9]+]], v[[LOVAL]], [[VZERO]]
>> @@ -34,7 +34,7 @@ define void @v_ctpop_i64(i32 addrspace(1)* noalias %out, i64 addrspace(1)* noali
>> ret void
>> }
>>
>> -; FUNC-LABEL: @s_ctpop_v2i64:
>> +; FUNC-LABEL: s_ctpop_v2i64:
>> ; SI: S_BCNT1_I32_B64
>> ; SI: S_BCNT1_I32_B64
>> ; SI: S_ENDPGM
>> @@ -45,7 +45,7 @@ define void @s_ctpop_v2i64(<2 x i32> addrspace(1)* noalias %out, <2 x i64> %val)
>> ret void
>> }
>>
>> -; FUNC-LABEL: @s_ctpop_v4i64:
>> +; FUNC-LABEL: s_ctpop_v4i64:
>> ; SI: S_BCNT1_I32_B64
>> ; SI: S_BCNT1_I32_B64
>> ; SI: S_BCNT1_I32_B64
>> @@ -58,7 +58,7 @@ define void @s_ctpop_v4i64(<4 x i32> addrspace(1)* noalias %out, <4 x i64> %val)
>> ret void
>> }
>>
>> -; FUNC-LABEL: @v_ctpop_v2i64:
>> +; FUNC-LABEL: v_ctpop_v2i64:
>> ; SI: V_BCNT_U32_B32
>> ; SI: V_BCNT_U32_B32
>> ; SI: V_BCNT_U32_B32
>> @@ -72,7 +72,7 @@ define void @v_ctpop_v2i64(<2 x i32> addrspace(1)* noalias %out, <2 x i64> addrs
>> ret void
>> }
>>
>> -; FUNC-LABEL: @v_ctpop_v4i64:
>> +; FUNC-LABEL: v_ctpop_v4i64:
>> ; SI: V_BCNT_U32_B32
>> ; SI: V_BCNT_U32_B32
>> ; SI: V_BCNT_U32_B32
>> @@ -93,7 +93,7 @@ define void @v_ctpop_v4i64(<4 x i32> addrspace(1)* noalias %out, <4 x i64> addrs
>> ; FIXME: We currently disallow SALU instructions in all branches,
>> ; but there are some cases when the should be allowed.
>>
>> -; FUNC-LABEL: @ctpop_i64_in_br
>> +; FUNC-LABEL: ctpop_i64_in_br
>> ; SI: S_LOAD_DWORDX2 s{{\[}}[[LOVAL:[0-9]+]]:[[HIVAL:[0-9]+]]{{\]}}, s[{{[0-9]+:[0-9]+}}], 0xd
>> ; SI: S_BCNT1_I32_B64 [[RESULT:s[0-9]+]], {{s\[}}[[LOVAL]]:[[HIVAL]]{{\]}}
>> ; SI: V_MOV_B32_e32 v[[VLO:[0-9]+]], [[RESULT]]
>> diff --git a/test/CodeGen/R600/cttz_zero_undef.ll b/test/CodeGen/R600/cttz_zero_undef.ll
>> index 9c4a355..74b8b90 100644
>> --- a/test/CodeGen/R600/cttz_zero_undef.ll
>> +++ b/test/CodeGen/R600/cttz_zero_undef.ll
>> @@ -5,7 +5,7 @@ declare i32 @llvm.cttz.i32(i32, i1) nounwind readnone
>> declare <2 x i32> @llvm.cttz.v2i32(<2 x i32>, i1) nounwind readnone
>> declare <4 x i32> @llvm.cttz.v4i32(<4 x i32>, i1) nounwind readnone
>>
>> -; FUNC-LABEL: @s_cttz_zero_undef_i32:
>> +; FUNC-LABEL: s_cttz_zero_undef_i32:
>> ; SI: S_LOAD_DWORD [[VAL:s[0-9]+]],
>> ; SI: S_FF1_I32_B32 [[SRESULT:s[0-9]+]], [[VAL]]
>> ; SI: V_MOV_B32_e32 [[VRESULT:v[0-9]+]], [[SRESULT]]
>> @@ -19,7 +19,7 @@ define void @s_cttz_zero_undef_i32(i32 addrspace(1)* noalias %out, i32 %val) nou
>> ret void
>> }
>>
>> -; FUNC-LABEL: @v_cttz_zero_undef_i32:
>> +; FUNC-LABEL: v_cttz_zero_undef_i32:
>> ; SI: BUFFER_LOAD_DWORD [[VAL:v[0-9]+]],
>> ; SI: V_FFBL_B32_e32 [[RESULT:v[0-9]+]], [[VAL]]
>> ; SI: BUFFER_STORE_DWORD [[RESULT]],
>> @@ -33,7 +33,7 @@ define void @v_cttz_zero_undef_i32(i32 addrspace(1)* noalias %out, i32 addrspace
>> ret void
>> }
>>
>> -; FUNC-LABEL: @v_cttz_zero_undef_v2i32:
>> +; FUNC-LABEL: v_cttz_zero_undef_v2i32:
>> ; SI: BUFFER_LOAD_DWORDX2
>> ; SI: V_FFBL_B32_e32
>> ; SI: V_FFBL_B32_e32
>> @@ -49,7 +49,7 @@ define void @v_cttz_zero_undef_v2i32(<2 x i32> addrspace(1)* noalias %out, <2 x
>> ret void
>> }
>>
>> -; FUNC-LABEL: @v_cttz_zero_undef_v4i32:
>> +; FUNC-LABEL: v_cttz_zero_undef_v4i32:
>> ; SI: BUFFER_LOAD_DWORDX4
>> ; SI: V_FFBL_B32_e32
>> ; SI: V_FFBL_B32_e32
>> diff --git a/test/CodeGen/R600/cvt_f32_ubyte.ll b/test/CodeGen/R600/cvt_f32_ubyte.ll
>> index 3e667b1..259c0e4 100644
>> --- a/test/CodeGen/R600/cvt_f32_ubyte.ll
>> +++ b/test/CodeGen/R600/cvt_f32_ubyte.ll
>> @@ -1,6 +1,6 @@
>> ; RUN: llc -march=r600 -mcpu=SI -verify-machineinstrs < %s | FileCheck -check-prefix=SI %s
>>
>> -; SI-LABEL: @load_i8_to_f32:
>> +; SI-LABEL: load_i8_to_f32:
>> ; SI: BUFFER_LOAD_UBYTE [[LOADREG:v[0-9]+]],
>> ; SI-NOT: BFE
>> ; SI-NOT: LSHR
>> @@ -13,7 +13,7 @@ define void @load_i8_to_f32(float addrspace(1)* noalias %out, i8 addrspace(1)* n
>> ret void
>> }
>>
>> -; SI-LABEL: @load_v2i8_to_v2f32:
>> +; SI-LABEL: load_v2i8_to_v2f32:
>> ; SI: BUFFER_LOAD_USHORT [[LOADREG:v[0-9]+]],
>> ; SI-NOT: BFE
>> ; SI-NOT: LSHR
>> @@ -28,7 +28,7 @@ define void @load_v2i8_to_v2f32(<2 x float> addrspace(1)* noalias %out, <2 x i8>
>> ret void
>> }
>>
>> -; SI-LABEL: @load_v3i8_to_v3f32:
>> +; SI-LABEL: load_v3i8_to_v3f32:
>> ; SI-NOT: BFE
>> ; SI-NOT: V_CVT_F32_UBYTE3_e32
>> ; SI-DAG: V_CVT_F32_UBYTE2_e32
>> @@ -42,7 +42,7 @@ define void @load_v3i8_to_v3f32(<3 x float> addrspace(1)* noalias %out, <3 x i8>
>> ret void
>> }
>>
>> -; SI-LABEL: @load_v4i8_to_v4f32:
>> +; SI-LABEL: load_v4i8_to_v4f32:
>> ; We can't use BUFFER_LOAD_DWORD here, because the load is byte aligned, and
>> ; BUFFER_LOAD_DWORD requires dword alignment.
>> ; SI: BUFFER_LOAD_USHORT
>> @@ -66,7 +66,7 @@ define void @load_v4i8_to_v4f32(<4 x float> addrspace(1)* noalias %out, <4 x i8>
>> ; for each component, but computeKnownBits doesn't handle vectors very
>> ; well.
>>
>> -; SI-LABEL: @load_v4i8_to_v4f32_2_uses:
>> +; SI-LABEL: load_v4i8_to_v4f32_2_uses:
>> ; SI: BUFFER_LOAD_UBYTE
>> ; SI: BUFFER_LOAD_UBYTE
>> ; SI: BUFFER_LOAD_UBYTE
>> @@ -93,7 +93,7 @@ define void @load_v4i8_to_v4f32_2_uses(<4 x float> addrspace(1)* noalias %out, <
>> }
>>
>> ; Make sure this doesn't crash.
>> -; SI-LABEL: @load_v7i8_to_v7f32:
>> +; SI-LABEL: load_v7i8_to_v7f32:
>> ; SI: S_ENDPGM
>> define void @load_v7i8_to_v7f32(<7 x float> addrspace(1)* noalias %out, <7 x i8> addrspace(1)* noalias %in) nounwind {
>> %load = load <7 x i8> addrspace(1)* %in, align 1
>> @@ -102,7 +102,7 @@ define void @load_v7i8_to_v7f32(<7 x float> addrspace(1)* noalias %out, <7 x i8>
>> ret void
>> }
>>
>> -; SI-LABEL: @load_v8i8_to_v8f32:
>> +; SI-LABEL: load_v8i8_to_v8f32:
>> ; SI: BUFFER_LOAD_DWORDX2 v{{\[}}[[LOLOAD:[0-9]+]]:[[HILOAD:[0-9]+]]{{\]}},
>> ; SI-NOT: BFE
>> ; SI-NOT: LSHR
>> @@ -131,7 +131,7 @@ define void @load_v8i8_to_v8f32(<8 x float> addrspace(1)* noalias %out, <8 x i8>
>> ret void
>> }
>>
>> -; SI-LABEL: @i8_zext_inreg_i32_to_f32:
>> +; SI-LABEL: i8_zext_inreg_i32_to_f32:
>> ; SI: BUFFER_LOAD_DWORD [[LOADREG:v[0-9]+]],
>> ; SI: V_ADD_I32_e32 [[ADD:v[0-9]+]], 2, [[LOADREG]]
>> ; SI-NEXT: V_CVT_F32_UBYTE0_e32 [[CONV:v[0-9]+]], [[ADD]]
>> @@ -145,7 +145,7 @@ define void @i8_zext_inreg_i32_to_f32(float addrspace(1)* noalias %out, i32 addr
>> ret void
>> }
>>
>> -; SI-LABEL: @i8_zext_inreg_hi1_to_f32:
>> +; SI-LABEL: i8_zext_inreg_hi1_to_f32:
>> define void @i8_zext_inreg_hi1_to_f32(float addrspace(1)* noalias %out, i32 addrspace(1)* noalias %in) nounwind {
>> %load = load i32 addrspace(1)* %in, align 4
>> %inreg = and i32 %load, 65280
>> diff --git a/test/CodeGen/R600/dagcombiner-bug-illegal-vec4-int-to-fp.ll b/test/CodeGen/R600/dagcombiner-bug-illegal-vec4-int-to-fp.ll
>> index 6607c12..e8a8353 100644
>> --- a/test/CodeGen/R600/dagcombiner-bug-illegal-vec4-int-to-fp.ll
>> +++ b/test/CodeGen/R600/dagcombiner-bug-illegal-vec4-int-to-fp.ll
>> @@ -7,7 +7,7 @@
>> ; ISD::UINT_TO_FP and ISD::SINT_TO_FP opcodes.
>>
>>
>> -; CHECK: @sint
>> +; CHECK: sint
>> ; CHECK: INT_TO_FLT * T{{[0-9]+\.[XYZW], T[0-9]+\.[XYZW]}}
>>
>> define void @sint(<4 x float> addrspace(1)* %out, i32 addrspace(1)* %in) {
>> @@ -21,7 +21,7 @@ entry:
>> ret void
>> }
>>
>> -;CHECK: @uint
>> +;CHECK: uint
>> ;CHECK: UINT_TO_FLT * T{{[0-9]+\.[XYZW], T[0-9]+\.[XYZW]}}
>>
>> define void @uint(<4 x float> addrspace(1)* %out, i32 addrspace(1)* %in) {
>> diff --git a/test/CodeGen/R600/default-fp-mode.ll b/test/CodeGen/R600/default-fp-mode.ll
>> index b24a7a2..13de2b0 100644
>> --- a/test/CodeGen/R600/default-fp-mode.ll
>> +++ b/test/CodeGen/R600/default-fp-mode.ll
>> @@ -6,7 +6,7 @@
>> ; RUN: llc -march=r600 -mcpu=SI -mattr=-fp32-denormals < %s | FileCheck -check-prefix=DEFAULT -check-prefix=FUNC %s
>> ; RUN: llc -march=r600 -mcpu=SI -mattr=+fp64-denormals < %s | FileCheck -check-prefix=DEFAULT -check-prefix=FUNC %s
>>
>> -; FUNC-LABEL: @test_kernel
>> +; FUNC-LABEL: test_kernel
>>
>> ; DEFAULT: FloatMode: 192
>> ; DEFAULT: IeeeMode: 0
>> diff --git a/test/CodeGen/R600/disconnected-predset-break-bug.ll b/test/CodeGen/R600/disconnected-predset-break-bug.ll
>> index 012c17b..f6ba0ca 100644
>> --- a/test/CodeGen/R600/disconnected-predset-break-bug.ll
>> +++ b/test/CodeGen/R600/disconnected-predset-break-bug.ll
>> @@ -4,7 +4,7 @@
>> ; result. This tests that there are no instructions between the PRED_SET*
>> ; and the PREDICATE_BREAK in this loop.
>>
>> -; CHECK: @loop_ge
>> +; CHECK: loop_ge
>> ; CHECK: LOOP_START_DX10
>> ; CHECK: ALU_PUSH_BEFORE
>> ; CHECK-NEXT: JUMP
>> diff --git a/test/CodeGen/R600/dot4-folding.ll b/test/CodeGen/R600/dot4-folding.ll
>> index 3e8330f..37ce689 100644
>> --- a/test/CodeGen/R600/dot4-folding.ll
>> +++ b/test/CodeGen/R600/dot4-folding.ll
>> @@ -2,7 +2,7 @@
>>
>> ; Exactly one constant vector can be folded into dot4, which means exactly
>> ; 4 MOV instructions
>> -; CHECK: @main
>> +; CHECK: main
>> ; CHECK: MOV
>> ; CHECK: MOV
>> ; CHECK: MOV
>> diff --git a/test/CodeGen/R600/ds-negative-offset-addressing-mode-loop.ll b/test/CodeGen/R600/ds-negative-offset-addressing-mode-loop.ll
>> index 231cf3d..73b130e 100644
>> --- a/test/CodeGen/R600/ds-negative-offset-addressing-mode-loop.ll
>> +++ b/test/CodeGen/R600/ds-negative-offset-addressing-mode-loop.ll
>> @@ -5,7 +5,7 @@ declare i32 @llvm.r600.read.tidig.x() #0
>> declare void @llvm.AMDGPU.barrier.local() #1
>>
>> ; Function Attrs: nounwind
>> -; CHECK-LABEL: @signed_ds_offset_addressing_loop
>> +; CHECK-LABEL: signed_ds_offset_addressing_loop
>> ; CHECK: BB0_1:
>> ; CHECK: V_ADD_I32_e32 [[VADDR:v[0-9]+]],
>> ; SI-DAG: DS_READ_B32 v{{[0-9]+}}, [[VADDR]], 0x0
>> diff --git a/test/CodeGen/R600/extload.ll b/test/CodeGen/R600/extload.ll
>> index 9725bbf..1569d2e 100644
>> --- a/test/CodeGen/R600/extload.ll
>> +++ b/test/CodeGen/R600/extload.ll
>> @@ -1,7 +1,7 @@
>> ; RUN: llc -march=r600 -mcpu=cypress < %s | FileCheck -check-prefix=EG -check-prefix=FUNC %s
>> ; RUN: llc -march=r600 -mcpu=SI -verify-machineinstrs< %s | FileCheck -check-prefix=SI -check-prefix=FUNC %s
>>
>> -; FUNC-LABEL: @anyext_load_i8:
>> +; FUNC-LABEL: anyext_load_i8:
>> ; EG: AND_INT
>> ; EG: 255
>> define void @anyext_load_i8(i8 addrspace(1)* nocapture noalias %out, i8 addrspace(1)* nocapture noalias %src) nounwind {
>> @@ -13,7 +13,7 @@ define void @anyext_load_i8(i8 addrspace(1)* nocapture noalias %out, i8 addrspac
>> ret void
>> }
>>
>> -; FUNC-LABEL: @anyext_load_i16:
>> +; FUNC-LABEL: anyext_load_i16:
>> ; EG: AND_INT
>> ; EG: AND_INT
>> ; EG-DAG: 65535
>> @@ -27,7 +27,7 @@ define void @anyext_load_i16(i16 addrspace(1)* nocapture noalias %out, i16 addrs
>> ret void
>> }
>>
>> -; FUNC-LABEL: @anyext_load_lds_i8:
>> +; FUNC-LABEL: anyext_load_lds_i8:
>> ; EG: AND_INT
>> ; EG: 255
>> define void @anyext_load_lds_i8(i8 addrspace(3)* nocapture noalias %out, i8 addrspace(3)* nocapture noalias %src) nounwind {
>> @@ -39,7 +39,7 @@ define void @anyext_load_lds_i8(i8 addrspace(3)* nocapture noalias %out, i8 addr
>> ret void
>> }
>>
>> -; FUNC-LABEL: @anyext_load_lds_i16:
>> +; FUNC-LABEL: anyext_load_lds_i16:
>> ; EG: AND_INT
>> ; EG: AND_INT
>> ; EG-DAG: 65535
>> @@ -53,7 +53,7 @@ define void @anyext_load_lds_i16(i16 addrspace(3)* nocapture noalias %out, i16 a
>> ret void
>> }
>>
>> -; FUNC-LABEL: @sextload_global_i8_to_i64
>> +; FUNC-LABEL: sextload_global_i8_to_i64
>> ; SI: BUFFER_LOAD_SBYTE [[LOAD:v[0-9]+]],
>> ; SI: V_ASHRREV_I32_e32 v{{[0-9]+}}, 31, [[LOAD]]
>> ; SI: BUFFER_STORE_DWORDX2
>> @@ -64,7 +64,7 @@ define void @sextload_global_i8_to_i64(i64 addrspace(1)* %out, i8 addrspace(1)*
>> ret void
>> }
>>
>> -; FUNC-LABEL: @sextload_global_i16_to_i64
>> +; FUNC-LABEL: sextload_global_i16_to_i64
>> ; SI: BUFFER_LOAD_SSHORT [[LOAD:v[0-9]+]],
>> ; SI: V_ASHRREV_I32_e32 v{{[0-9]+}}, 31, [[LOAD]]
>> ; SI: BUFFER_STORE_DWORDX2
>> @@ -75,7 +75,7 @@ define void @sextload_global_i16_to_i64(i64 addrspace(1)* %out, i16 addrspace(1)
>> ret void
>> }
>>
>> -; FUNC-LABEL: @sextload_global_i32_to_i64
>> +; FUNC-LABEL: sextload_global_i32_to_i64
>> ; SI: BUFFER_LOAD_DWORD [[LOAD:v[0-9]+]],
>> ; SI: V_ASHRREV_I32_e32 v{{[0-9]+}}, 31, [[LOAD]]
>> ; SI: BUFFER_STORE_DWORDX2
>> @@ -86,7 +86,7 @@ define void @sextload_global_i32_to_i64(i64 addrspace(1)* %out, i32 addrspace(1)
>> ret void
>> }
>>
>> -; FUNC-LABEL: @zextload_global_i8_to_i64
>> +; FUNC-LABEL: zextload_global_i8_to_i64
>> ; SI-DAG: S_MOV_B32 [[ZERO:s[0-9]+]], 0{{$}}
>> ; SI-DAG: BUFFER_LOAD_UBYTE [[LOAD:v[0-9]+]],
>> ; SI: V_MOV_B32_e32 {{v[0-9]+}}, [[ZERO]]
>> @@ -98,7 +98,7 @@ define void @zextload_global_i8_to_i64(i64 addrspace(1)* %out, i8 addrspace(1)*
>> ret void
>> }
>>
>> -; FUNC-LABEL: @zextload_global_i16_to_i64
>> +; FUNC-LABEL: zextload_global_i16_to_i64
>> ; SI-DAG: S_MOV_B32 [[ZERO:s[0-9]+]], 0{{$}}
>> ; SI-DAG: BUFFER_LOAD_USHORT [[LOAD:v[0-9]+]],
>> ; SI: V_MOV_B32_e32 {{v[0-9]+}}, [[ZERO]]
>> @@ -110,7 +110,7 @@ define void @zextload_global_i16_to_i64(i64 addrspace(1)* %out, i16 addrspace(1)
>> ret void
>> }
>>
>> -; FUNC-LABEL: @zextload_global_i32_to_i64
>> +; FUNC-LABEL: zextload_global_i32_to_i64
>> ; SI-DAG: S_MOV_B32 [[ZERO:s[0-9]+]], 0{{$}}
>> ; SI-DAG: BUFFER_LOAD_DWORD [[LOAD:v[0-9]+]],
>> ; SI: V_MOV_B32_e32 {{v[0-9]+}}, [[ZERO]]
>> diff --git a/test/CodeGen/R600/extract_vector_elt_i16.ll b/test/CodeGen/R600/extract_vector_elt_i16.ll
>> index 26b8f2c..d670f05 100644
>> --- a/test/CodeGen/R600/extract_vector_elt_i16.ll
>> +++ b/test/CodeGen/R600/extract_vector_elt_i16.ll
>> @@ -1,6 +1,6 @@
>> ; RUN: llc -march=r600 -mcpu=SI -verify-machineinstrs< %s | FileCheck -check-prefix=SI -check-prefix=FUNC %s
>>
>> -; FUNC-LABEL: @extract_vector_elt_v2i16
>> +; FUNC-LABEL: extract_vector_elt_v2i16
>> ; SI: BUFFER_LOAD_USHORT
>> ; SI: BUFFER_LOAD_USHORT
>> ; SI: BUFFER_STORE_SHORT
>> @@ -14,7 +14,7 @@ define void @extract_vector_elt_v2i16(i16 addrspace(1)* %out, <2 x i16> %foo) no
>> ret void
>> }
>>
>> -; FUNC-LABEL: @extract_vector_elt_v4i16
>> +; FUNC-LABEL: extract_vector_elt_v4i16
>> ; SI: BUFFER_LOAD_USHORT
>> ; SI: BUFFER_LOAD_USHORT
>> ; SI: BUFFER_STORE_SHORT
>> diff --git a/test/CodeGen/R600/fabs.f64.ll b/test/CodeGen/R600/fabs.f64.ll
>> index 1e1bdf4..c56ff18 100644
>> --- a/test/CodeGen/R600/fabs.f64.ll
>> +++ b/test/CodeGen/R600/fabs.f64.ll
>> @@ -7,7 +7,7 @@ declare double @llvm.fabs.f64(double) readnone
>> declare <2 x double> @llvm.fabs.v2f64(<2 x double>) readnone
>> declare <4 x double> @llvm.fabs.v4f64(<4 x double>) readnone
>>
>> -; FUNC-LABEL: @v_fabs_f64
>> +; FUNC-LABEL: v_fabs_f64:
>> ; SI: V_AND_B32
>> ; SI: S_ENDPGM
>> define void @v_fabs_f64(double addrspace(1)* %out, double addrspace(1)* %in) {
>> @@ -20,7 +20,7 @@ define void @v_fabs_f64(double addrspace(1)* %out, double addrspace(1)* %in) {
>> ret void
>> }
>>
>> -; FUNC-LABEL: @fabs_f64
>> +; FUNC-LABEL: fabs_f64:
>> ; SI: V_AND_B32
>> ; SI-NOT: V_AND_B32
>> ; SI: S_ENDPGM
>> @@ -30,7 +30,7 @@ define void @fabs_f64(double addrspace(1)* %out, double %in) {
>> ret void
>> }
>>
>> -; FUNC-LABEL: @fabs_v2f64
>> +; FUNC-LABEL: fabs_v2f64:
>> ; SI: V_AND_B32
>> ; SI: V_AND_B32
>> ; SI: S_ENDPGM
>> @@ -40,7 +40,7 @@ define void @fabs_v2f64(<2 x double> addrspace(1)* %out, <2 x double> %in) {
>> ret void
>> }
>>
>> -; FUNC-LABEL: @fabs_v4f64
>> +; FUNC-LABEL: fabs_v4f64:
>> ; SI: V_AND_B32
>> ; SI: V_AND_B32
>> ; SI: V_AND_B32
>> @@ -52,7 +52,7 @@ define void @fabs_v4f64(<4 x double> addrspace(1)* %out, <4 x double> %in) {
>> ret void
>> }
>>
>> -; SI-LABEL: @fabs_fold_f64
>> +; SI-LABEL: fabs_fold_f64:
>> ; SI: S_LOAD_DWORDX2 [[ABS_VALUE:s\[[0-9]+:[0-9]+\]]], {{s\[[0-9]+:[0-9]+\]}}, 0xb
>> ; SI-NOT: AND
>> ; SI: V_MUL_F64 {{v\[[0-9]+:[0-9]+\]}}, |[[ABS_VALUE]]|, {{v\[[0-9]+:[0-9]+\]}}
>> @@ -64,7 +64,7 @@ define void @fabs_fold_f64(double addrspace(1)* %out, double %in0, double %in1)
>> ret void
>> }
>>
>> -; SI-LABEL: @fabs_fn_fold_f64
>> +; SI-LABEL: fabs_fn_fold_f64:
>> ; SI: S_LOAD_DWORDX2 [[ABS_VALUE:s\[[0-9]+:[0-9]+\]]], {{s\[[0-9]+:[0-9]+\]}}, 0xb
>> ; SI-NOT: AND
>> ; SI: V_MUL_F64 {{v\[[0-9]+:[0-9]+\]}}, |[[ABS_VALUE]]|, {{v\[[0-9]+:[0-9]+\]}}
>> @@ -76,7 +76,7 @@ define void @fabs_fn_fold_f64(double addrspace(1)* %out, double %in0, double %in
>> ret void
>> }
>>
>> -; FUNC-LABEL: @fabs_free_f64
>> +; FUNC-LABEL: fabs_free_f64:
>> ; SI: V_AND_B32
>> ; SI: S_ENDPGM
>> define void @fabs_free_f64(double addrspace(1)* %out, i64 %in) {
>> @@ -86,7 +86,7 @@ define void @fabs_free_f64(double addrspace(1)* %out, i64 %in) {
>> ret void
>> }
>>
>> -; FUNC-LABEL: @fabs_fn_free_f64
>> +; FUNC-LABEL: fabs_fn_free_f64:
>> ; SI: V_AND_B32
>> ; SI: S_ENDPGM
>> define void @fabs_fn_free_f64(double addrspace(1)* %out, i64 %in) {
>> diff --git a/test/CodeGen/R600/fabs.ll b/test/CodeGen/R600/fabs.ll
>> index 610c419..eb8c9e4 100644
>> --- a/test/CodeGen/R600/fabs.ll
>> +++ b/test/CodeGen/R600/fabs.ll
>> @@ -6,7 +6,7 @@
>> ; (fabs (f32 bitcast (i32 a))) => (f32 bitcast (and (i32 a), 0x7FFFFFFF))
>> ; unless isFabsFree returns true
>>
>> -; FUNC-LABEL: @fabs_fn_free
>> +; FUNC-LABEL: fabs_fn_free
>> ; R600-NOT: AND
>> ; R600: |PV.{{[XYZW]}}|
>>
>> @@ -19,7 +19,7 @@ define void @fabs_fn_free(float addrspace(1)* %out, i32 %in) {
>> ret void
>> }
>>
>> -; FUNC-LABEL: @fabs_free
>> +; FUNC-LABEL: fabs_free
>> ; R600-NOT: AND
>> ; R600: |PV.{{[XYZW]}}|
>>
>> @@ -32,7 +32,7 @@ define void @fabs_free(float addrspace(1)* %out, i32 %in) {
>> ret void
>> }
>>
>> -; FUNC-LABEL: @fabs_f32
>> +; FUNC-LABEL: fabs_f32
>> ; R600: |{{(PV|T[0-9])\.[XYZW]}}|
>>
>> ; SI: V_AND_B32
>> @@ -42,7 +42,7 @@ define void @fabs_f32(float addrspace(1)* %out, float %in) {
>> ret void
>> }
>>
>> -; FUNC-LABEL: @fabs_v2f32
>> +; FUNC-LABEL: fabs_v2f32
>> ; R600: |{{(PV|T[0-9])\.[XYZW]}}|
>> ; R600: |{{(PV|T[0-9])\.[XYZW]}}|
>>
>> @@ -54,7 +54,7 @@ define void @fabs_v2f32(<2 x float> addrspace(1)* %out, <2 x float> %in) {
>> ret void
>> }
>>
>> -; FUNC-LABEL: @fabs_v4
>> +; FUNC-LABEL: fabs_v4
>> ; R600: |{{(PV|T[0-9])\.[XYZW]}}|
>> ; R600: |{{(PV|T[0-9])\.[XYZW]}}|
>> ; R600: |{{(PV|T[0-9])\.[XYZW]}}|
>> @@ -70,7 +70,7 @@ define void @fabs_v4f32(<4 x float> addrspace(1)* %out, <4 x float> %in) {
>> ret void
>> }
>>
>> -; SI-LABEL: @fabs_fn_fold
>> +; SI-LABEL: fabs_fn_fold
>> ; SI: S_LOAD_DWORD [[ABS_VALUE:s[0-9]+]], s[{{[0-9]+:[0-9]+}}], 0xb
>> ; SI-NOT: AND
>> ; SI: V_MUL_F32_e64 v{{[0-9]+}}, |[[ABS_VALUE]]|, v{{[0-9]+}}
>> @@ -81,7 +81,7 @@ define void @fabs_fn_fold(float addrspace(1)* %out, float %in0, float %in1) {
>> ret void
>> }
>>
>> -; SI-LABEL: @fabs_fold
>> +; SI-LABEL: fabs_fold
>> ; SI: S_LOAD_DWORD [[ABS_VALUE:s[0-9]+]], s[{{[0-9]+:[0-9]+}}], 0xb
>> ; SI-NOT: AND
>> ; SI: V_MUL_F32_e64 v{{[0-9]+}}, |[[ABS_VALUE]]|, v{{[0-9]+}}
>> diff --git a/test/CodeGen/R600/fadd.ll b/test/CodeGen/R600/fadd.ll
>> index 3a87c89..cc8d91d 100644
>> --- a/test/CodeGen/R600/fadd.ll
>> +++ b/test/CodeGen/R600/fadd.ll
>> @@ -1,7 +1,7 @@
>> ; RUN: llc -march=r600 -mcpu=redwood < %s | FileCheck %s -check-prefix=R600 -check-prefix=FUNC
>> ; RUN: llc -march=r600 -mcpu=SI -verify-machineinstrs < %s | FileCheck %s -check-prefix=SI -check-prefix=FUNC
>>
>> -; FUNC-LABEL: @fadd_f32
>> +; FUNC-LABEL: fadd_f32
>> ; R600: ADD {{\** *}}T{{[0-9]+\.[XYZW]}}, KC0[2].Z, KC0[2].W
>> ; SI: V_ADD_F32
>> define void @fadd_f32(float addrspace(1)* %out, float %a, float %b) {
>> @@ -10,7 +10,7 @@ define void @fadd_f32(float addrspace(1)* %out, float %a, float %b) {
>> ret void
>> }
>>
>> -; FUNC-LABEL: @fadd_v2f32
>> +; FUNC-LABEL: fadd_v2f32
>> ; R600-DAG: ADD {{\** *}}T{{[0-9]\.[XYZW]}}, KC0[3].X, KC0[3].Z
>> ; R600-DAG: ADD {{\** *}}T{{[0-9]\.[XYZW]}}, KC0[2].W, KC0[3].Y
>> ; SI: V_ADD_F32
>> @@ -21,7 +21,7 @@ define void @fadd_v2f32(<2 x float> addrspace(1)* %out, <2 x float> %a, <2 x flo
>> ret void
>> }
>>
>> -; FUNC-LABEL: @fadd_v4f32
>> +; FUNC-LABEL: fadd_v4f32
>> ; R600: ADD {{\** *}}T{{[0-9]+\.[XYZW], T[0-9]+\.[XYZW]}}
>> ; R600: ADD {{\** *}}T{{[0-9]+\.[XYZW], T[0-9]+\.[XYZW]}}
>> ; R600: ADD {{\** *}}T{{[0-9]+\.[XYZW], T[0-9]+\.[XYZW]}}
>> @@ -39,7 +39,7 @@ define void @fadd_v4f32(<4 x float> addrspace(1)* %out, <4 x float> addrspace(1)
>> ret void
>> }
>>
>> -; FUNC-LABEL: @fadd_v8f32
>> +; FUNC-LABEL: fadd_v8f32
>> ; R600: ADD
>> ; R600: ADD
>> ; R600: ADD
>> diff --git a/test/CodeGen/R600/fadd64.ll b/test/CodeGen/R600/fadd64.ll
>> index 48cd3cf..e6f4e3e 100644
>> --- a/test/CodeGen/R600/fadd64.ll
>> +++ b/test/CodeGen/R600/fadd64.ll
>> @@ -1,6 +1,6 @@
>> ; RUN: llc < %s -march=r600 -mcpu=tahiti -verify-machineinstrs | FileCheck %s
>>
>> -; CHECK: @fadd_f64
>> +; CHECK: fadd_f64
>> ; CHECK: V_ADD_F64 {{v[[0-9]+:[0-9]+]}}, {{v[[0-9]+:[0-9]+]}}, {{v[[0-9]+:[0-9]+]}}
>>
>> define void @fadd_f64(double addrspace(1)* %out, double addrspace(1)* %in1,
>> diff --git a/test/CodeGen/R600/fceil.ll b/test/CodeGen/R600/fceil.ll
>> index 458363a..6ed716b 100644
>> --- a/test/CodeGen/R600/fceil.ll
>> +++ b/test/CodeGen/R600/fceil.ll
>> @@ -8,7 +8,7 @@ declare <4 x float> @llvm.ceil.v4f32(<4 x float>) nounwind readnone
>> declare <8 x float> @llvm.ceil.v8f32(<8 x float>) nounwind readnone
>> declare <16 x float> @llvm.ceil.v16f32(<16 x float>) nounwind readnone
>>
>> -; FUNC-LABEL: @fceil_f32:
>> +; FUNC-LABEL: fceil_f32:
>> ; SI: V_CEIL_F32_e32
>> ; EG: MEM_RAT_CACHELESS STORE_RAW [[RESULT:T[0-9]+\.[XYZW]]]
>> ; EG: CEIL {{\*? *}}[[RESULT]]
>> @@ -18,7 +18,7 @@ define void @fceil_f32(float addrspace(1)* %out, float %x) {
>> ret void
>> }
>>
>> -; FUNC-LABEL: @fceil_v2f32:
>> +; FUNC-LABEL: fceil_v2f32:
>> ; SI: V_CEIL_F32_e32
>> ; SI: V_CEIL_F32_e32
>> ; EG: MEM_RAT_CACHELESS STORE_RAW [[RESULT:T[0-9]+]]{{\.[XYZW]}}
>> @@ -30,7 +30,7 @@ define void @fceil_v2f32(<2 x float> addrspace(1)* %out, <2 x float> %x) {
>> ret void
>> }
>>
>> -; FUNC-LABEL: @fceil_v3f32:
>> +; FUNC-LABEL: fceil_v3f32:
>> ; FIXME-SI: V_CEIL_F32_e32
>> ; FIXME-SI: V_CEIL_F32_e32
>> ; FIXME-SI: V_CEIL_F32_e32
>> @@ -46,7 +46,7 @@ define void @fceil_v3f32(<3 x float> addrspace(1)* %out, <3 x float> %x) {
>> ret void
>> }
>>
>> -; FUNC-LABEL: @fceil_v4f32:
>> +; FUNC-LABEL: fceil_v4f32:
>> ; SI: V_CEIL_F32_e32
>> ; SI: V_CEIL_F32_e32
>> ; SI: V_CEIL_F32_e32
>> @@ -62,7 +62,7 @@ define void @fceil_v4f32(<4 x float> addrspace(1)* %out, <4 x float> %x) {
>> ret void
>> }
>>
>> -; FUNC-LABEL: @fceil_v8f32:
>> +; FUNC-LABEL: fceil_v8f32:
>> ; SI: V_CEIL_F32_e32
>> ; SI: V_CEIL_F32_e32
>> ; SI: V_CEIL_F32_e32
>> @@ -87,7 +87,7 @@ define void @fceil_v8f32(<8 x float> addrspace(1)* %out, <8 x float> %x) {
>> ret void
>> }
>>
>> -; FUNC-LABEL: @fceil_v16f32:
>> +; FUNC-LABEL: fceil_v16f32:
>> ; SI: V_CEIL_F32_e32
>> ; SI: V_CEIL_F32_e32
>> ; SI: V_CEIL_F32_e32
>> diff --git a/test/CodeGen/R600/fceil64.ll b/test/CodeGen/R600/fceil64.ll
>> index fa1740a..4534c02 100644
>> --- a/test/CodeGen/R600/fceil64.ll
>> +++ b/test/CodeGen/R600/fceil64.ll
>> @@ -8,7 +8,7 @@ declare <4 x double> @llvm.ceil.v4f64(<4 x double>) nounwind readnone
>> declare <8 x double> @llvm.ceil.v8f64(<8 x double>) nounwind readnone
>> declare <16 x double> @llvm.ceil.v16f64(<16 x double>) nounwind readnone
>>
>> -; FUNC-LABEL: @fceil_f64:
>> +; FUNC-LABEL: fceil_f64:
>> ; CI: V_CEIL_F64_e32
>> ; SI: S_BFE_I32 [[SEXP:s[0-9]+]], {{s[0-9]+}}, 0xb0014
>> ; SI: S_ADD_I32 s{{[0-9]+}}, [[SEXP]], 0xfffffc01
>> @@ -34,7 +34,7 @@ define void @fceil_f64(double addrspace(1)* %out, double %x) {
>> ret void
>> }
>>
>> -; FUNC-LABEL: @fceil_v2f64:
>> +; FUNC-LABEL: fceil_v2f64:
>> ; CI: V_CEIL_F64_e32
>> ; CI: V_CEIL_F64_e32
>> define void @fceil_v2f64(<2 x double> addrspace(1)* %out, <2 x double> %x) {
>> @@ -43,7 +43,7 @@ define void @fceil_v2f64(<2 x double> addrspace(1)* %out, <2 x double> %x) {
>> ret void
>> }
>>
>> -; FIXME-FUNC-LABEL: @fceil_v3f64:
>> +; FIXME-FUNC-LABEL: fceil_v3f64:
>> ; FIXME-CI: V_CEIL_F64_e32
>> ; FIXME-CI: V_CEIL_F64_e32
>> ; FIXME-CI: V_CEIL_F64_e32
>> @@ -53,7 +53,7 @@ define void @fceil_v2f64(<2 x double> addrspace(1)* %out, <2 x double> %x) {
>> ; ret void
>> ; }
>>
>> -; FUNC-LABEL: @fceil_v4f64:
>> +; FUNC-LABEL: fceil_v4f64:
>> ; CI: V_CEIL_F64_e32
>> ; CI: V_CEIL_F64_e32
>> ; CI: V_CEIL_F64_e32
>> @@ -64,7 +64,7 @@ define void @fceil_v4f64(<4 x double> addrspace(1)* %out, <4 x double> %x) {
>> ret void
>> }
>>
>> -; FUNC-LABEL: @fceil_v8f64:
>> +; FUNC-LABEL: fceil_v8f64:
>> ; CI: V_CEIL_F64_e32
>> ; CI: V_CEIL_F64_e32
>> ; CI: V_CEIL_F64_e32
>> @@ -79,7 +79,7 @@ define void @fceil_v8f64(<8 x double> addrspace(1)* %out, <8 x double> %x) {
>> ret void
>> }
>>
>> -; FUNC-LABEL: @fceil_v16f64:
>> +; FUNC-LABEL: fceil_v16f64:
>> ; CI: V_CEIL_F64_e32
>> ; CI: V_CEIL_F64_e32
>> ; CI: V_CEIL_F64_e32
>> diff --git a/test/CodeGen/R600/fcmp.ll b/test/CodeGen/R600/fcmp.ll
>> index c76a758..5de69a3 100644
>> --- a/test/CodeGen/R600/fcmp.ll
>> +++ b/test/CodeGen/R600/fcmp.ll
>> @@ -1,6 +1,6 @@
>> ; RUN: llc < %s -march=r600 -mcpu=redwood | FileCheck %s
>>
>> -; CHECK: @fcmp_sext
>> +; CHECK: fcmp_sext
>> ; CHECK: SETE_DX10 T{{[0-9]+\.[XYZW], T[0-9]+\.[XYZW], T[0-9]+\.[XYZW]}}
>>
>> define void @fcmp_sext(i32 addrspace(1)* %out, float addrspace(1)* %in) {
>> @@ -18,7 +18,7 @@ entry:
>> ; SET*_DX10 instruction. Previously we were lowering this to:
>> ; SET* + FP_TO_SINT
>>
>> -; CHECK: @fcmp_br
>> +; CHECK: fcmp_br
>> ; CHECK: SET{{[N]*}}E_DX10 * T{{[0-9]+\.[XYZW],}}
>> ; CHECK-NEXT {{[0-9]+(5.0}}
>>
>> diff --git a/test/CodeGen/R600/fcmp64.ll b/test/CodeGen/R600/fcmp64.ll
>> index 8cbe9f6..1216b87 100644
>> --- a/test/CodeGen/R600/fcmp64.ll
>> +++ b/test/CodeGen/R600/fcmp64.ll
>> @@ -1,6 +1,6 @@
>> ; RUN: llc < %s -march=r600 -mcpu=tahiti -verify-machineinstrs | FileCheck %s
>>
>> -; CHECK: @flt_f64
>> +; CHECK: flt_f64
>> ; CHECK: V_CMP_LT_F64_e64 {{s[[0-9]+:[0-9]+], v[[0-9]+:[0-9]+], v[[0-9]+:[0-9]+]}}
>>
>> define void @flt_f64(double addrspace(1)* %out, double addrspace(1)* %in1,
>> @@ -13,7 +13,7 @@ define void @flt_f64(double addrspace(1)* %out, double addrspace(1)* %in1,
>> ret void
>> }
>>
>> -; CHECK: @fle_f64
>> +; CHECK: fle_f64
>> ; CHECK: V_CMP_LE_F64_e64 {{s[[0-9]+:[0-9]+], v[[0-9]+:[0-9]+], v[[0-9]+:[0-9]+]}}
>>
>> define void @fle_f64(double addrspace(1)* %out, double addrspace(1)* %in1,
>> @@ -26,7 +26,7 @@ define void @fle_f64(double addrspace(1)* %out, double addrspace(1)* %in1,
>> ret void
>> }
>>
>> -; CHECK: @fgt_f64
>> +; CHECK: fgt_f64
>> ; CHECK: V_CMP_GT_F64_e64 {{s[[0-9]+:[0-9]+], v[[0-9]+:[0-9]+], v[[0-9]+:[0-9]+]}}
>>
>> define void @fgt_f64(double addrspace(1)* %out, double addrspace(1)* %in1,
>> @@ -39,7 +39,7 @@ define void @fgt_f64(double addrspace(1)* %out, double addrspace(1)* %in1,
>> ret void
>> }
>>
>> -; CHECK: @fge_f64
>> +; CHECK: fge_f64
>> ; CHECK: V_CMP_GE_F64_e64 {{s[[0-9]+:[0-9]+], v[[0-9]+:[0-9]+], v[[0-9]+:[0-9]+]}}
>>
>> define void @fge_f64(double addrspace(1)* %out, double addrspace(1)* %in1,
>> @@ -52,7 +52,7 @@ define void @fge_f64(double addrspace(1)* %out, double addrspace(1)* %in1,
>> ret void
>> }
>>
>> -; CHECK: @fne_f64
>> +; CHECK: fne_f64
>> ; CHECK: V_CMP_NEQ_F64_e32 vcc, {{v[[0-9]+:[0-9]+], v[[0-9]+:[0-9]+]}}
>>
>> define void @fne_f64(double addrspace(1)* %out, double addrspace(1)* %in1,
>> @@ -65,7 +65,7 @@ define void @fne_f64(double addrspace(1)* %out, double addrspace(1)* %in1,
>> ret void
>> }
>>
>> -; CHECK: @feq_f64
>> +; CHECK: feq_f64
>> ; CHECK: V_CMP_EQ_F64_e64 {{s[[0-9]+:[0-9]+], v[[0-9]+:[0-9]+], v[[0-9]+:[0-9]+]}}
>>
>> define void @feq_f64(double addrspace(1)* %out, double addrspace(1)* %in1,
>> diff --git a/test/CodeGen/R600/fconst64.ll b/test/CodeGen/R600/fconst64.ll
>> index 9c3a7e3..49eeb2e 100644
>> --- a/test/CodeGen/R600/fconst64.ll
>> +++ b/test/CodeGen/R600/fconst64.ll
>> @@ -1,6 +1,6 @@
>> ; RUN: llc < %s -march=r600 -mcpu=tahiti -verify-machineinstrs | FileCheck %s
>>
>> -; CHECK: @fconst_f64
>> +; CHECK: fconst_f64
>> ; CHECK-DAG: S_MOV_B32 {{s[0-9]+}}, 0x40140000
>> ; CHECK-DAG: S_MOV_B32 {{s[0-9]+}}, 0
>>
>> diff --git a/test/CodeGen/R600/fcopysign.f32.ll b/test/CodeGen/R600/fcopysign.f32.ll
>> index 15ee40e..c7031a8 100644
>> --- a/test/CodeGen/R600/fcopysign.f32.ll
>> +++ b/test/CodeGen/R600/fcopysign.f32.ll
>> @@ -7,7 +7,7 @@ declare <2 x float> @llvm.copysign.v2f32(<2 x float>, <2 x float>) nounwind read
>> declare <4 x float> @llvm.copysign.v4f32(<4 x float>, <4 x float>) nounwind readnone
>>
>> ; Try to identify arg based on higher address.
>> -; FUNC-LABEL: @test_copysign_f32:
>> +; FUNC-LABEL: test_copysign_f32:
>> ; SI: S_LOAD_DWORD [[SMAG:s[0-9]+]], {{.*}} 0xb
>> ; SI: S_LOAD_DWORD [[SSIGN:s[0-9]+]], {{.*}} 0xc
>> ; SI-DAG: V_MOV_B32_e32 [[VSIGN:v[0-9]+]], [[SSIGN]]
>> @@ -24,7 +24,7 @@ define void @test_copysign_f32(float addrspace(1)* %out, float %mag, float %sign
>> ret void
>> }
>>
>> -; FUNC-LABEL: @test_copysign_v2f32:
>> +; FUNC-LABEL: test_copysign_v2f32:
>> ; SI: S_ENDPGM
>>
>> ; EG: BFI_INT
>> @@ -35,7 +35,7 @@ define void @test_copysign_v2f32(<2 x float> addrspace(1)* %out, <2 x float> %ma
>> ret void
>> }
>>
>> -; FUNC-LABEL: @test_copysign_v4f32:
>> +; FUNC-LABEL: test_copysign_v4f32:
>> ; SI: S_ENDPGM
>>
>> ; EG: BFI_INT
>> diff --git a/test/CodeGen/R600/fcopysign.f64.ll b/test/CodeGen/R600/fcopysign.f64.ll
>> index c72329c..86e50a1 100644
>> --- a/test/CodeGen/R600/fcopysign.f64.ll
>> +++ b/test/CodeGen/R600/fcopysign.f64.ll
>> @@ -4,7 +4,7 @@ declare double @llvm.copysign.f64(double, double) nounwind readnone
>> declare <2 x double> @llvm.copysign.v2f64(<2 x double>, <2 x double>) nounwind readnone
>> declare <4 x double> @llvm.copysign.v4f64(<4 x double>, <4 x double>) nounwind readnone
>>
>> -; FUNC-LABEL: @test_copysign_f64:
>> +; FUNC-LABEL: test_copysign_f64:
>> ; SI-DAG: S_LOAD_DWORDX2 s{{\[}}[[SMAG_LO:[0-9]+]]:[[SMAG_HI:[0-9]+]]{{\]}}, s{{\[[0-9]+:[0-9]+\]}}, 0xb
>> ; SI-DAG: S_LOAD_DWORDX2 s{{\[}}[[SSIGN_LO:[0-9]+]]:[[SSIGN_HI:[0-9]+]]{{\]}}, s{{\[[0-9]+:[0-9]+\]}}, 0xd
>> ; SI-DAG: V_MOV_B32_e32 v[[VSIGN_HI:[0-9]+]], s[[SSIGN_HI]]
>> @@ -20,7 +20,7 @@ define void @test_copysign_f64(double addrspace(1)* %out, double %mag, double %s
>> ret void
>> }
>>
>> -; FUNC-LABEL: @test_copysign_v2f64:
>> +; FUNC-LABEL: test_copysign_v2f64:
>> ; SI: S_ENDPGM
>> define void @test_copysign_v2f64(<2 x double> addrspace(1)* %out, <2 x double> %mag, <2 x double> %sign) nounwind {
>> %result = call <2 x double> @llvm.copysign.v2f64(<2 x double> %mag, <2 x double> %sign)
>> @@ -28,7 +28,7 @@ define void @test_copysign_v2f64(<2 x double> addrspace(1)* %out, <2 x double> %
>> ret void
>> }
>>
>> -; FUNC-LABEL: @test_copysign_v4f64:
>> +; FUNC-LABEL: test_copysign_v4f64:
>> ; SI: S_ENDPGM
>> define void @test_copysign_v4f64(<4 x double> addrspace(1)* %out, <4 x double> %mag, <4 x double> %sign) nounwind {
>> %result = call <4 x double> @llvm.copysign.v4f64(<4 x double> %mag, <4 x double> %sign)
>> diff --git a/test/CodeGen/R600/fdiv.ll b/test/CodeGen/R600/fdiv.ll
>> index 20db65c..a72b0e8 100644
>> --- a/test/CodeGen/R600/fdiv.ll
>> +++ b/test/CodeGen/R600/fdiv.ll
>> @@ -5,7 +5,7 @@
>> ; scheduler is scheduling the RECIP_IEEE and MUL_IEEE instructions in separate
>> ; instruction groups.
>>
>> -; FUNC-LABEL: @fdiv_f32
>> +; FUNC-LABEL: fdiv_f32
>> ; R600-DAG: RECIP_IEEE * T{{[0-9]+\.[XYZW]}}, KC0[3].Z
>> ; R600-DAG: RECIP_IEEE * T{{[0-9]+\.[XYZW]}}, KC0[3].Y
>> ; R600-DAG: MUL_IEEE {{\** *}}T{{[0-9]+\.[XYZW]}}, KC0[3].X, PS
>> @@ -22,7 +22,7 @@ entry:
>>
>>
>>
>> -; FUNC-LABEL: @fdiv_v2f32
>> +; FUNC-LABEL: fdiv_v2f32
>> ; R600-DAG: RECIP_IEEE * T{{[0-9]+\.[XYZW]}}, KC0[3].Z
>> ; R600-DAG: RECIP_IEEE * T{{[0-9]+\.[XYZW]}}, KC0[3].Y
>> ; R600-DAG: MUL_IEEE {{\** *}}T{{[0-9]+\.[XYZW]}}, KC0[3].X, PS
>> @@ -39,7 +39,7 @@ entry:
>> ret void
>> }
>>
>> -; FUNC-LABEL: @fdiv_v4f32
>> +; FUNC-LABEL: fdiv_v4f32
>> ; R600-DAG: RECIP_IEEE * T{{[0-9]+\.[XYZW], T[0-9]+\.[XYZW]}}
>> ; R600-DAG: RECIP_IEEE * T{{[0-9]+\.[XYZW], T[0-9]+\.[XYZW]}}
>> ; R600-DAG: RECIP_IEEE * T{{[0-9]+\.[XYZW], T[0-9]+\.[XYZW]}}
>> diff --git a/test/CodeGen/R600/fdiv64.ll b/test/CodeGen/R600/fdiv64.ll
>> index 79b5c8b..944718a 100644
>> --- a/test/CodeGen/R600/fdiv64.ll
>> +++ b/test/CodeGen/R600/fdiv64.ll
>> @@ -1,6 +1,6 @@
>> ; RUN: llc < %s -march=r600 -mcpu=tahiti -verify-machineinstrs | FileCheck %s
>>
>> -; CHECK: @fdiv_f64
>> +; CHECK: fdiv_f64
>> ; CHECK: V_RCP_F64_e32 {{v\[[0-9]+:[0-9]+\]}}
>> ; CHECK: V_MUL_F64 {{v\[[0-9]+:[0-9]+\]}}, {{v\[[0-9]+:[0-9]+\]}}, {{v\[[0-9]+:[0-9]+\]}}
>>
>> diff --git a/test/CodeGen/R600/fetch-limits.r600.ll b/test/CodeGen/R600/fetch-limits.r600.ll
>> index f78d1d9..dd8f8c6 100644
>> --- a/test/CodeGen/R600/fetch-limits.r600.ll
>> +++ b/test/CodeGen/R600/fetch-limits.r600.ll
>> @@ -3,7 +3,7 @@
>> ; RUN: llc < %s -march=r600 -mcpu=rv670 | FileCheck %s
>>
>> ; R600 supports 8 fetches in a clause
>> -; CHECK: @fetch_limits_r600
>> +; CHECK: fetch_limits_r600
>> ; CHECK: Fetch clause
>> ; CHECK: Fetch clause
>>
>> diff --git a/test/CodeGen/R600/fetch-limits.r700+.ll b/test/CodeGen/R600/fetch-limits.r700+.ll
>> index 1a8a43f..6e95510 100644
>> --- a/test/CodeGen/R600/fetch-limits.r700+.ll
>> +++ b/test/CodeGen/R600/fetch-limits.r700+.ll
>> @@ -12,7 +12,7 @@
>> ; RUN: llc < %s -march=r600 -mcpu=cayman | FileCheck %s
>>
>> ; r700+ supports 16 fetches in a clause
>> -; CHECK: @fetch_limits_r700
>> +; CHECK: fetch_limits_r700
>> ; CHECK: Fetch clause
>> ; CHECK: Fetch clause
>>
>> diff --git a/test/CodeGen/R600/ffloor.ll b/test/CodeGen/R600/ffloor.ll
>> index 1a98b75..22b5df7 100644
>> --- a/test/CodeGen/R600/ffloor.ll
>> +++ b/test/CodeGen/R600/ffloor.ll
>> @@ -8,7 +8,7 @@ declare <4 x double> @llvm.floor.v4f64(<4 x double>) nounwind readnone
>> declare <8 x double> @llvm.floor.v8f64(<8 x double>) nounwind readnone
>> declare <16 x double> @llvm.floor.v16f64(<16 x double>) nounwind readnone
>>
>> -; FUNC-LABEL: @ffloor_f64:
>> +; FUNC-LABEL: ffloor_f64:
>> ; CI: V_FLOOR_F64_e32
>>
>> ; SI: S_BFE_I32 [[SEXP:s[0-9]+]], {{s[0-9]+}}, 0xb0014
>> @@ -35,7 +35,7 @@ define void @ffloor_f64(double addrspace(1)* %out, double %x) {
>> ret void
>> }
>>
>> -; FUNC-LABEL: @ffloor_v2f64:
>> +; FUNC-LABEL: ffloor_v2f64:
>> ; CI: V_FLOOR_F64_e32
>> ; CI: V_FLOOR_F64_e32
>> define void @ffloor_v2f64(<2 x double> addrspace(1)* %out, <2 x double> %x) {
>> @@ -44,7 +44,7 @@ define void @ffloor_v2f64(<2 x double> addrspace(1)* %out, <2 x double> %x) {
>> ret void
>> }
>>
>> -; FIXME-FUNC-LABEL: @ffloor_v3f64:
>> +; FIXME-FUNC-LABEL: ffloor_v3f64:
>> ; FIXME-CI: V_FLOOR_F64_e32
>> ; FIXME-CI: V_FLOOR_F64_e32
>> ; FIXME-CI: V_FLOOR_F64_e32
>> @@ -54,7 +54,7 @@ define void @ffloor_v2f64(<2 x double> addrspace(1)* %out, <2 x double> %x) {
>> ; ret void
>> ; }
>>
>> -; FUNC-LABEL: @ffloor_v4f64:
>> +; FUNC-LABEL: ffloor_v4f64:
>> ; CI: V_FLOOR_F64_e32
>> ; CI: V_FLOOR_F64_e32
>> ; CI: V_FLOOR_F64_e32
>> @@ -65,7 +65,7 @@ define void @ffloor_v4f64(<4 x double> addrspace(1)* %out, <4 x double> %x) {
>> ret void
>> }
>>
>> -; FUNC-LABEL: @ffloor_v8f64:
>> +; FUNC-LABEL: ffloor_v8f64:
>> ; CI: V_FLOOR_F64_e32
>> ; CI: V_FLOOR_F64_e32
>> ; CI: V_FLOOR_F64_e32
>> @@ -80,7 +80,7 @@ define void @ffloor_v8f64(<8 x double> addrspace(1)* %out, <8 x double> %x) {
>> ret void
>> }
>>
>> -; FUNC-LABEL: @ffloor_v16f64:
>> +; FUNC-LABEL: ffloor_v16f64:
>> ; CI: V_FLOOR_F64_e32
>> ; CI: V_FLOOR_F64_e32
>> ; CI: V_FLOOR_F64_e32
>> diff --git a/test/CodeGen/R600/flat-address-space.ll b/test/CodeGen/R600/flat-address-space.ll
>> index 3f32e4d..229622f 100644
>> --- a/test/CodeGen/R600/flat-address-space.ll
>> +++ b/test/CodeGen/R600/flat-address-space.ll
>> @@ -5,7 +5,7 @@
>> ; specialize away generic pointer accesses.
>>
>>
>> -; CHECK-LABEL: @branch_use_flat_i32:
>> +; CHECK-LABEL: branch_use_flat_i32:
>> ; CHECK: FLAT_STORE_DWORD {{v[0-9]+}}, {{v\[[0-9]+:[0-9]+\]}}, [M0, FLAT_SCRATCH]
>> ; CHECK: S_ENDPGM
>> define void @branch_use_flat_i32(i32 addrspace(1)* noalias %out, i32 addrspace(1)* %gptr, i32 addrspace(3)* %lptr, i32 %x, i32 %c) #0 {
>> @@ -34,7 +34,7 @@ end:
>> ; These testcases might become useless when there are optimizations to
>> ; remove generic pointers.
>>
>> -; CHECK-LABEL: @store_flat_i32:
>> +; CHECK-LABEL: store_flat_i32:
>> ; CHECK: V_MOV_B32_e32 v[[DATA:[0-9]+]], {{s[0-9]+}}
>> ; CHECK: V_MOV_B32_e32 v[[LO_VREG:[0-9]+]], {{s[0-9]+}}
>> ; CHECK: V_MOV_B32_e32 v[[HI_VREG:[0-9]+]], {{s[0-9]+}}
>> @@ -45,7 +45,7 @@ define void @store_flat_i32(i32 addrspace(1)* %gptr, i32 %x) #0 {
>> ret void
>> }
>>
>> -; CHECK-LABEL: @store_flat_i64:
>> +; CHECK-LABEL: store_flat_i64:
>> ; CHECK: FLAT_STORE_DWORDX2
>> define void @store_flat_i64(i64 addrspace(1)* %gptr, i64 %x) #0 {
>> %fptr = addrspacecast i64 addrspace(1)* %gptr to i64 addrspace(4)*
>> @@ -53,7 +53,7 @@ define void @store_flat_i64(i64 addrspace(1)* %gptr, i64 %x) #0 {
>> ret void
>> }
>>
>> -; CHECK-LABEL: @store_flat_v4i32:
>> +; CHECK-LABEL: store_flat_v4i32:
>> ; CHECK: FLAT_STORE_DWORDX4
>> define void @store_flat_v4i32(<4 x i32> addrspace(1)* %gptr, <4 x i32> %x) #0 {
>> %fptr = addrspacecast <4 x i32> addrspace(1)* %gptr to <4 x i32> addrspace(4)*
>> @@ -61,7 +61,7 @@ define void @store_flat_v4i32(<4 x i32> addrspace(1)* %gptr, <4 x i32> %x) #0 {
>> ret void
>> }
>>
>> -; CHECK-LABEL: @store_flat_trunc_i16:
>> +; CHECK-LABEL: store_flat_trunc_i16:
>> ; CHECK: FLAT_STORE_SHORT
>> define void @store_flat_trunc_i16(i16 addrspace(1)* %gptr, i32 %x) #0 {
>> %fptr = addrspacecast i16 addrspace(1)* %gptr to i16 addrspace(4)*
>> @@ -70,7 +70,7 @@ define void @store_flat_trunc_i16(i16 addrspace(1)* %gptr, i32 %x) #0 {
>> ret void
>> }
>>
>> -; CHECK-LABEL: @store_flat_trunc_i8:
>> +; CHECK-LABEL: store_flat_trunc_i8:
>> ; CHECK: FLAT_STORE_BYTE
>> define void @store_flat_trunc_i8(i8 addrspace(1)* %gptr, i32 %x) #0 {
>> %fptr = addrspacecast i8 addrspace(1)* %gptr to i8 addrspace(4)*
>> @@ -154,7 +154,7 @@ define void @zextload_flat_i16(i32 addrspace(1)* noalias %out, i16 addrspace(1)*
>> ; scratch allocations again.
>>
>> ; Check for prologue initializing special SGPRs pointing to scratch.
>> -; CHECK-LABEL: @store_flat_scratch:
>> +; CHECK-LABEL: store_flat_scratch:
>> ; CHECK: S_MOVK_I32 flat_scratch_lo, 0
>> ; CHECK-NO-PROMOTE: S_MOVK_I32 flat_scratch_hi, 40
>> ; CHECK-PROMOTE: S_MOVK_I32 flat_scratch_hi, 0
>> diff --git a/test/CodeGen/R600/fma.f64.ll b/test/CodeGen/R600/fma.f64.ll
>> index b4b9450..82c0865 100644
>> --- a/test/CodeGen/R600/fma.f64.ll
>> +++ b/test/CodeGen/R600/fma.f64.ll
>> @@ -5,7 +5,7 @@ declare <2 x double> @llvm.fma.v2f64(<2 x double>, <2 x double>, <2 x double>) n
>> declare <4 x double> @llvm.fma.v4f64(<4 x double>, <4 x double>, <4 x double>) nounwind readnone
>>
>>
>> -; FUNC-LABEL: @fma_f64
>> +; FUNC-LABEL: fma_f64
>> ; SI: V_FMA_F64 {{v\[[0-9]+:[0-9]+\], v\[[0-9]+:[0-9]+\], v\[[0-9]+:[0-9]+\], v\[[0-9]+:[0-9]+\]}}
>> define void @fma_f64(double addrspace(1)* %out, double addrspace(1)* %in1,
>> double addrspace(1)* %in2, double addrspace(1)* %in3) {
>> @@ -17,7 +17,7 @@ define void @fma_f64(double addrspace(1)* %out, double addrspace(1)* %in1,
>> ret void
>> }
>>
>> -; FUNC-LABEL: @fma_v2f64
>> +; FUNC-LABEL: fma_v2f64
>> ; SI: V_FMA_F64
>> ; SI: V_FMA_F64
>> define void @fma_v2f64(<2 x double> addrspace(1)* %out, <2 x double> addrspace(1)* %in1,
>> @@ -30,7 +30,7 @@ define void @fma_v2f64(<2 x double> addrspace(1)* %out, <2 x double> addrspace(1
>> ret void
>> }
>>
>> -; FUNC-LABEL: @fma_v4f64
>> +; FUNC-LABEL: fma_v4f64
>> ; SI: V_FMA_F64
>> ; SI: V_FMA_F64
>> ; SI: V_FMA_F64
>> diff --git a/test/CodeGen/R600/fma.ll b/test/CodeGen/R600/fma.ll
>> index eec8eb7..66b7b88 100644
>> --- a/test/CodeGen/R600/fma.ll
>> +++ b/test/CodeGen/R600/fma.ll
>> @@ -5,7 +5,7 @@ declare float @llvm.fma.f32(float, float, float) nounwind readnone
>> declare <2 x float> @llvm.fma.v2f32(<2 x float>, <2 x float>, <2 x float>) nounwind readnone
>> declare <4 x float> @llvm.fma.v4f32(<4 x float>, <4 x float>, <4 x float>) nounwind readnone
>>
>> -; FUNC-LABEL: @fma_f32
>> +; FUNC-LABEL: fma_f32
>> ; SI: V_FMA_F32 {{v[0-9]+, v[0-9]+, v[0-9]+, v[0-9]+}}
>> define void @fma_f32(float addrspace(1)* %out, float addrspace(1)* %in1,
>> float addrspace(1)* %in2, float addrspace(1)* %in3) {
>> @@ -17,7 +17,7 @@ define void @fma_f32(float addrspace(1)* %out, float addrspace(1)* %in1,
>> ret void
>> }
>>
>> -; FUNC-LABEL: @fma_v2f32
>> +; FUNC-LABEL: fma_v2f32
>> ; SI: V_FMA_F32
>> ; SI: V_FMA_F32
>> define void @fma_v2f32(<2 x float> addrspace(1)* %out, <2 x float> addrspace(1)* %in1,
>> @@ -30,7 +30,7 @@ define void @fma_v2f32(<2 x float> addrspace(1)* %out, <2 x float> addrspace(1)*
>> ret void
>> }
>>
>> -; FUNC-LABEL: @fma_v4f32
>> +; FUNC-LABEL: fma_v4f32
>> ; SI: V_FMA_F32
>> ; SI: V_FMA_F32
>> ; SI: V_FMA_F32
>> diff --git a/test/CodeGen/R600/fmul.ll b/test/CodeGen/R600/fmul.ll
>> index 3c88bea..eb86adf 100644
>> --- a/test/CodeGen/R600/fmul.ll
>> +++ b/test/CodeGen/R600/fmul.ll
>> @@ -2,7 +2,7 @@
>> ; RUN: llc -march=r600 -mcpu=redwood < %s | FileCheck -check-prefix=R600 -check-prefix=FUNC %s
>>
>>
>> -; FUNC-LABEL: @fmul_f32
>> +; FUNC-LABEL: fmul_f32
>> ; R600: MUL_IEEE {{\** *}}{{T[0-9]+\.[XYZW]}}, KC0[2].Z, KC0[2].W
>>
>> ; SI: V_MUL_F32
>> @@ -17,7 +17,7 @@ declare float @llvm.R600.load.input(i32) readnone
>>
>> declare void @llvm.AMDGPU.store.output(float, i32)
>>
>> -; FUNC-LABEL: @fmul_v2f32
>> +; FUNC-LABEL: fmul_v2f32
>> ; R600: MUL_IEEE {{\** *}}T{{[0-9]+\.[XYZW]}}
>> ; R600: MUL_IEEE {{\** *}}T{{[0-9]+\.[XYZW]}}
>>
>> @@ -30,7 +30,7 @@ entry:
>> ret void
>> }
>>
>> -; FUNC-LABEL: @fmul_v4f32
>> +; FUNC-LABEL: fmul_v4f32
>> ; R600: MUL_IEEE {{\** *}}T{{[0-9]+\.[XYZW], T[0-9]+\.[XYZW]}}
>> ; R600: MUL_IEEE {{\** *}}T{{[0-9]+\.[XYZW], T[0-9]+\.[XYZW]}}
>> ; R600: MUL_IEEE {{\** *}}T{{[0-9]+\.[XYZW], T[0-9]+\.[XYZW]}}
>> @@ -49,7 +49,7 @@ define void @fmul_v4f32(<4 x float> addrspace(1)* %out, <4 x float> addrspace(1)
>> ret void
>> }
>>
>> -; FUNC-LABEL: @test_mul_2_k
>> +; FUNC-LABEL: test_mul_2_k
>> ; SI: V_MUL_F32
>> ; SI-NOT: V_MUL_F32
>> ; SI: S_ENDPGM
>> @@ -60,7 +60,7 @@ define void @test_mul_2_k(float addrspace(1)* %out, float %x) #0 {
>> ret void
>> }
>>
>> -; FUNC-LABEL: @test_mul_2_k_inv
>> +; FUNC-LABEL: test_mul_2_k_inv
>> ; SI: V_MUL_F32
>> ; SI-NOT: V_MUL_F32
>> ; SI-NOT: V_MAD_F32
>> diff --git a/test/CodeGen/R600/fmul64.ll b/test/CodeGen/R600/fmul64.ll
>> index e8491f9..7b737c2 100644
>> --- a/test/CodeGen/R600/fmul64.ll
>> +++ b/test/CodeGen/R600/fmul64.ll
>> @@ -1,6 +1,6 @@
>> ; RUN: llc -march=r600 -mcpu=tahiti -verify-machineinstrs < %s | FileCheck -check-prefix=FUNC -check-prefix=SI %s
>>
>> -; FUNC-LABEL: @fmul_f64
>> +; FUNC-LABEL: fmul_f64
>> ; SI: V_MUL_F64 {{v\[[0-9]+:[0-9]+\], v\[[0-9]+:[0-9]+\], v\[[0-9]+:[0-9]+\]}}
>> define void @fmul_f64(double addrspace(1)* %out, double addrspace(1)* %in1,
>> double addrspace(1)* %in2) {
>> @@ -11,7 +11,7 @@ define void @fmul_f64(double addrspace(1)* %out, double addrspace(1)* %in1,
>> ret void
>> }
>>
>> -; FUNC-LABEL: @fmul_v2f64
>> +; FUNC-LABEL: fmul_v2f64
>> ; SI: V_MUL_F64 {{v\[[0-9]+:[0-9]+\], v\[[0-9]+:[0-9]+\], v\[[0-9]+:[0-9]+\]}}
>> ; SI: V_MUL_F64 {{v\[[0-9]+:[0-9]+\], v\[[0-9]+:[0-9]+\], v\[[0-9]+:[0-9]+\]}}
>> define void @fmul_v2f64(<2 x double> addrspace(1)* %out, <2 x double> addrspace(1)* %in1,
>> @@ -23,7 +23,7 @@ define void @fmul_v2f64(<2 x double> addrspace(1)* %out, <2 x double> addrspace(
>> ret void
>> }
>>
>> -; FUNC-LABEL: @fmul_v4f64
>> +; FUNC-LABEL: fmul_v4f64
>> ; SI: V_MUL_F64 {{v\[[0-9]+:[0-9]+\], v\[[0-9]+:[0-9]+\], v\[[0-9]+:[0-9]+\]}}
>> ; SI: V_MUL_F64 {{v\[[0-9]+:[0-9]+\], v\[[0-9]+:[0-9]+\], v\[[0-9]+:[0-9]+\]}}
>> ; SI: V_MUL_F64 {{v\[[0-9]+:[0-9]+\], v\[[0-9]+:[0-9]+\], v\[[0-9]+:[0-9]+\]}}
>> diff --git a/test/CodeGen/R600/fmuladd.ll b/test/CodeGen/R600/fmuladd.ll
>> index 48944f6..b59ad52 100644
>> --- a/test/CodeGen/R600/fmuladd.ll
>> +++ b/test/CodeGen/R600/fmuladd.ll
>> @@ -1,6 +1,6 @@
>> ; RUN: llc < %s -march=r600 -mcpu=SI -verify-machineinstrs | FileCheck %s
>>
>> -; CHECK: @fmuladd_f32
>> +; CHECK: fmuladd_f32
>> ; CHECK: V_MAD_F32 {{v[0-9]+, v[0-9]+, v[0-9]+, v[0-9]+}}
>>
>> define void @fmuladd_f32(float addrspace(1)* %out, float addrspace(1)* %in1,
>> @@ -15,7 +15,7 @@ define void @fmuladd_f32(float addrspace(1)* %out, float addrspace(1)* %in1,
>>
>> declare float @llvm.fmuladd.f32(float, float, float)
>>
>> -; CHECK: @fmuladd_f64
>> +; CHECK: fmuladd_f64
>> ; CHECK: V_FMA_F64 {{v\[[0-9]+:[0-9]+\], v\[[0-9]+:[0-9]+\], v\[[0-9]+:[0-9]+\], v\[[0-9]+:[0-9]+\]}}
>>
>> define void @fmuladd_f64(double addrspace(1)* %out, double addrspace(1)* %in1,
>> diff --git a/test/CodeGen/R600/fneg-fabs.f64.ll b/test/CodeGen/R600/fneg-fabs.f64.ll
>> index 1442ca4..020789c 100644
>> --- a/test/CodeGen/R600/fneg-fabs.f64.ll
>> +++ b/test/CodeGen/R600/fneg-fabs.f64.ll
>> @@ -3,7 +3,7 @@
>> ; FIXME: Check something here. Currently it seems fabs + fneg aren't
>> ; into 2 modifiers, although theoretically that should work.
>>
>> -; FUNC-LABEL: @fneg_fabs_fadd_f64
>> +; FUNC-LABEL: fneg_fabs_fadd_f64
>> ; SI: V_MOV_B32_e32 [[IMMREG:v[0-9]+]], 0x7fffffff
>> ; SI: V_AND_B32_e32 v[[FABS:[0-9]+]], {{s[0-9]+}}, [[IMMREG]]
>> ; SI: V_ADD_F64 {{v\[[0-9]+:[0-9]+\]}}, {{v\[[0-9]+:[0-9]+\]}}, -v{{\[[0-9]+}}:[[FABS]]{{\]}}
>> @@ -25,7 +25,7 @@ define void @v_fneg_fabs_fadd_f64(double addrspace(1)* %out, double addrspace(1)
>> ret void
>> }
>>
>> -; FUNC-LABEL: @fneg_fabs_fmul_f64
>> +; FUNC-LABEL: fneg_fabs_fmul_f64
>> ; SI: V_MUL_F64 {{v\[[0-9]+:[0-9]+\]}}, {{s\[[0-9]+:[0-9]+\]}}, -|{{v\[[0-9]+:[0-9]+\]}}|
>> define void @fneg_fabs_fmul_f64(double addrspace(1)* %out, double %x, double %y) {
>> %fabs = call double @llvm.fabs.f64(double %x)
>> @@ -35,7 +35,7 @@ define void @fneg_fabs_fmul_f64(double addrspace(1)* %out, double %x, double %y)
>> ret void
>> }
>>
>> -; FUNC-LABEL: @fneg_fabs_free_f64
>> +; FUNC-LABEL: fneg_fabs_free_f64
>> define void @fneg_fabs_free_f64(double addrspace(1)* %out, i64 %in) {
>> %bc = bitcast i64 %in to double
>> %fabs = call double @llvm.fabs.f64(double %bc)
>> @@ -44,7 +44,7 @@ define void @fneg_fabs_free_f64(double addrspace(1)* %out, i64 %in) {
>> ret void
>> }
>>
>> -; FUNC-LABEL: @fneg_fabs_fn_free_f64
>> +; FUNC-LABEL: fneg_fabs_fn_free_f64
>> ; SI: V_MOV_B32_e32 [[IMMREG:v[0-9]+]], 0x80000000
>> ; SI: V_OR_B32_e32 v{{[0-9]+}}, s{{[0-9]+}}, [[IMMREG]]
>> define void @fneg_fabs_fn_free_f64(double addrspace(1)* %out, i64 %in) {
>> @@ -55,7 +55,7 @@ define void @fneg_fabs_fn_free_f64(double addrspace(1)* %out, i64 %in) {
>> ret void
>> }
>>
>> -; FUNC-LABEL: @fneg_fabs_f64
>> +; FUNC-LABEL: fneg_fabs_f64
>> ; SI: S_LOAD_DWORDX2
>> ; SI: S_LOAD_DWORDX2 s{{\[}}[[LO_X:[0-9]+]]:[[HI_X:[0-9]+]]{{\]}}
>> ; SI: V_MOV_B32_e32 [[IMMREG:v[0-9]+]], 0x80000000
>> @@ -69,7 +69,7 @@ define void @fneg_fabs_f64(double addrspace(1)* %out, double %in) {
>> ret void
>> }
>>
>> -; FUNC-LABEL: @fneg_fabs_v2f64
>> +; FUNC-LABEL: fneg_fabs_v2f64
>> ; SI: V_MOV_B32_e32 [[IMMREG:v[0-9]+]], 0x80000000
>> ; SI-NOT: 0x80000000
>> ; SI: V_OR_B32_e32 v{{[0-9]+}}, s{{[0-9]+}}, [[IMMREG]]
>> @@ -81,7 +81,7 @@ define void @fneg_fabs_v2f64(<2 x double> addrspace(1)* %out, <2 x double> %in)
>> ret void
>> }
>>
>> -; FUNC-LABEL: @fneg_fabs_v4f64
>> +; FUNC-LABEL: fneg_fabs_v4f64
>> ; SI: V_MOV_B32_e32 [[IMMREG:v[0-9]+]], 0x80000000
>> ; SI-NOT: 0x80000000
>> ; SI: V_OR_B32_e32 v{{[0-9]+}}, s{{[0-9]+}}, [[IMMREG]]
>> diff --git a/test/CodeGen/R600/fneg-fabs.ll b/test/CodeGen/R600/fneg-fabs.ll
>> index 28cdd06..941dcb8 100644
>> --- a/test/CodeGen/R600/fneg-fabs.ll
>> +++ b/test/CodeGen/R600/fneg-fabs.ll
>> @@ -1,7 +1,7 @@
>> ; RUN: llc -march=r600 -mcpu=SI -verify-machineinstrs < %s | FileCheck -check-prefix=SI -check-prefix=FUNC %s
>> ; RUN: llc -march=r600 -mcpu=redwood < %s | FileCheck -check-prefix=R600 -check-prefix=FUNC %s
>>
>> -; FUNC-LABEL: @fneg_fabs_fadd_f32
>> +; FUNC-LABEL: fneg_fabs_fadd_f32
>> ; SI-NOT: AND
>> ; SI: V_SUB_F32_e64 {{v[0-9]+}}, {{s[0-9]+}}, |{{v[0-9]+}}|
>> define void @fneg_fabs_fadd_f32(float addrspace(1)* %out, float %x, float %y) {
>> @@ -12,7 +12,7 @@ define void @fneg_fabs_fadd_f32(float addrspace(1)* %out, float %x, float %y) {
>> ret void
>> }
>>
>> -; FUNC-LABEL: @fneg_fabs_fmul_f32
>> +; FUNC-LABEL: fneg_fabs_fmul_f32
>> ; SI-NOT: AND
>> ; SI: V_MUL_F32_e64 {{v[0-9]+}}, {{s[0-9]+}}, -|{{v[0-9]+}}|
>> ; SI-NOT: AND
>> @@ -28,7 +28,7 @@ define void @fneg_fabs_fmul_f32(float addrspace(1)* %out, float %x, float %y) {
>> ; (fabs (f32 bitcast (i32 a))) => (f32 bitcast (and (i32 a), 0x7FFFFFFF))
>> ; unless isFabsFree returns true
>>
>> -; FUNC-LABEL: @fneg_fabs_free_f32
>> +; FUNC-LABEL: fneg_fabs_free_f32
>> ; R600-NOT: AND
>> ; R600: |PV.{{[XYZW]}}|
>> ; R600: -PV
>> @@ -43,7 +43,7 @@ define void @fneg_fabs_free_f32(float addrspace(1)* %out, i32 %in) {
>> ret void
>> }
>>
>> -; FUNC-LABEL: @fneg_fabs_fn_free_f32
>> +; FUNC-LABEL: fneg_fabs_fn_free_f32
>> ; R600-NOT: AND
>> ; R600: |PV.{{[XYZW]}}|
>> ; R600: -PV
>> @@ -58,7 +58,7 @@ define void @fneg_fabs_fn_free_f32(float addrspace(1)* %out, i32 %in) {
>> ret void
>> }
>>
>> -; FUNC-LABEL: @fneg_fabs_f32
>> +; FUNC-LABEL: fneg_fabs_f32
>> ; SI: V_MOV_B32_e32 [[IMMREG:v[0-9]+]], 0x80000000
>> ; SI: V_OR_B32_e32 v{{[0-9]+}}, s{{[0-9]+}}, [[IMMREG]]
>> define void @fneg_fabs_f32(float addrspace(1)* %out, float %in) {
>> @@ -68,7 +68,7 @@ define void @fneg_fabs_f32(float addrspace(1)* %out, float %in) {
>> ret void
>> }
>>
>> -; FUNC-LABEL: @v_fneg_fabs_f32
>> +; FUNC-LABEL: v_fneg_fabs_f32
>> ; SI: V_OR_B32_e32 v{{[0-9]+}}, 0x80000000, v{{[0-9]+}}
>> define void @v_fneg_fabs_f32(float addrspace(1)* %out, float addrspace(1)* %in) {
>> %val = load float addrspace(1)* %in, align 4
>> @@ -78,7 +78,7 @@ define void @v_fneg_fabs_f32(float addrspace(1)* %out, float addrspace(1)* %in)
>> ret void
>> }
>>
>> -; FUNC-LABEL: @fneg_fabs_v2f32
>> +; FUNC-LABEL: fneg_fabs_v2f32
>> ; R600: |{{(PV|T[0-9])\.[XYZW]}}|
>> ; R600: -PV
>> ; R600: |{{(PV|T[0-9])\.[XYZW]}}|
>> @@ -97,7 +97,7 @@ define void @fneg_fabs_v2f32(<2 x float> addrspace(1)* %out, <2 x float> %in) {
>> }
>>
>> ; FIXME: SGPR should be used directly for first src operand.
>> -; FUNC-LABEL: @fneg_fabs_v4f32
>> +; FUNC-LABEL: fneg_fabs_v4f32
>> ; SI: V_MOV_B32_e32 [[IMMREG:v[0-9]+]], 0x80000000
>> ; SI-NOT: 0x80000000
>> ; SI: V_OR_B32_e32 v{{[0-9]+}}, v{{[0-9]+}}, [[IMMREG]]
>> diff --git a/test/CodeGen/R600/fneg.f64.ll b/test/CodeGen/R600/fneg.f64.ll
>> index f0b341b..f4e8593 100644
>> --- a/test/CodeGen/R600/fneg.f64.ll
>> +++ b/test/CodeGen/R600/fneg.f64.ll
>> @@ -1,6 +1,6 @@
>> ; RUN: llc -march=r600 -mcpu=SI -verify-machineinstrs < %s | FileCheck -check-prefix=SI -check-prefix=FUNC %s
>>
>> -; FUNC-LABEL: @fneg_f64
>> +; FUNC-LABEL: fneg_f64
>> ; SI: V_XOR_B32
>> define void @fneg_f64(double addrspace(1)* %out, double %in) {
>> %fneg = fsub double -0.000000e+00, %in
>> @@ -8,7 +8,7 @@ define void @fneg_f64(double addrspace(1)* %out, double %in) {
>> ret void
>> }
>>
>> -; FUNC-LABEL: @fneg_v2f64
>> +; FUNC-LABEL: fneg_v2f64
>> ; SI: V_XOR_B32
>> ; SI: V_XOR_B32
>> define void @fneg_v2f64(<2 x double> addrspace(1)* nocapture %out, <2 x double> %in) {
>> @@ -17,7 +17,7 @@ define void @fneg_v2f64(<2 x double> addrspace(1)* nocapture %out, <2 x double>
>> ret void
>> }
>>
>> -; FUNC-LABEL: @fneg_v4f64
>> +; FUNC-LABEL: fneg_v4f64
>> ; R600: -PV
>> ; R600: -T
>> ; R600: -PV
>> @@ -37,7 +37,7 @@ define void @fneg_v4f64(<4 x double> addrspace(1)* nocapture %out, <4 x double>
>> ; (fneg (f64 bitcast (i64 a))) => (f64 bitcast (xor (i64 a), 0x80000000))
>> ; unless the target returns true for isNegFree()
>>
>> -; FUNC-LABEL: @fneg_free_f64
>> +; FUNC-LABEL: fneg_free_f64
>> ; FIXME: Unnecessary copy to VGPRs
>> ; SI: V_ADD_F64 {{v\[[0-9]+:[0-9]+\]}}, {{v\[[0-9]+:[0-9]+\]}}, -{{v\[[0-9]+:[0-9]+\]}}, 0, 0
>> define void @fneg_free_f64(double addrspace(1)* %out, i64 %in) {
>> @@ -47,7 +47,7 @@ define void @fneg_free_f64(double addrspace(1)* %out, i64 %in) {
>> ret void
>> }
>>
>> -; SI-LABEL: @fneg_fold
>> +; SI-LABEL: fneg_fold
>> ; SI: S_LOAD_DWORDX2 [[NEG_VALUE:s\[[0-9]+:[0-9]+\]]], {{s\[[0-9]+:[0-9]+\]}}, 0xb
>> ; SI-NOT: XOR
>> ; SI: V_MUL_F64 {{v\[[0-9]+:[0-9]+\]}}, -[[NEG_VALUE]], [[NEG_VALUE]]
>> diff --git a/test/CodeGen/R600/fneg.ll b/test/CodeGen/R600/fneg.ll
>> index 8631301..9e1f539 100644
>> --- a/test/CodeGen/R600/fneg.ll
>> +++ b/test/CodeGen/R600/fneg.ll
>> @@ -1,7 +1,7 @@
>> ; RUN: llc -march=r600 -mcpu=SI -verify-machineinstrs < %s | FileCheck -check-prefix=SI -check-prefix=FUNC %s
>> ; RUN: llc -march=r600 -mcpu=redwood < %s | FileCheck -check-prefix=R600 -check-prefix=FUNC %s
>>
>> -; FUNC-LABEL: @fneg_f32
>> +; FUNC-LABEL: fneg_f32
>> ; R600: -PV
>>
>> ; SI: V_XOR_B32
>> @@ -11,7 +11,7 @@ define void @fneg_f32(float addrspace(1)* %out, float %in) {
>> ret void
>> }
>>
>> -; FUNC-LABEL: @fneg_v2f32
>> +; FUNC-LABEL: fneg_v2f32
>> ; R600: -PV
>> ; R600: -PV
>>
>> @@ -23,7 +23,7 @@ define void @fneg_v2f32(<2 x float> addrspace(1)* nocapture %out, <2 x float> %i
>> ret void
>> }
>>
>> -; FUNC-LABEL: @fneg_v4f32
>> +; FUNC-LABEL: fneg_v4f32
>> ; R600: -PV
>> ; R600: -T
>> ; R600: -PV
>> @@ -43,7 +43,7 @@ define void @fneg_v4f32(<4 x float> addrspace(1)* nocapture %out, <4 x float> %i
>> ; (fneg (f32 bitcast (i32 a))) => (f32 bitcast (xor (i32 a), 0x80000000))
>> ; unless the target returns true for isNegFree()
>>
>> -; FUNC-LABEL: @fneg_free_f32
>> +; FUNC-LABEL: fneg_free_f32
>> ; R600-NOT: XOR
>> ; R600: -KC0[2].Z
>>
>> @@ -56,7 +56,7 @@ define void @fneg_free_f32(float addrspace(1)* %out, i32 %in) {
>> ret void
>> }
>>
>> -; FUNC-LABEL: @fneg_fold
>> +; FUNC-LABEL: fneg_fold
>> ; SI: S_LOAD_DWORD [[NEG_VALUE:s[0-9]+]], s[{{[0-9]+:[0-9]+}}], 0xb
>> ; SI-NOT: XOR
>> ; SI: V_MUL_F32_e64 v{{[0-9]+}}, -[[NEG_VALUE]], [[NEG_VALUE]]
>> diff --git a/test/CodeGen/R600/fp16_to_fp.ll b/test/CodeGen/R600/fp16_to_fp.ll
>> index 777eadc..c5ad09b 100644
>> --- a/test/CodeGen/R600/fp16_to_fp.ll
>> +++ b/test/CodeGen/R600/fp16_to_fp.ll
>> @@ -3,7 +3,7 @@
>> declare float @llvm.convert.from.fp16.f32(i16) nounwind readnone
>> declare double @llvm.convert.from.fp16.f64(i16) nounwind readnone
>>
>> -; SI-LABEL: @test_convert_fp16_to_fp32:
>> +; SI-LABEL: test_convert_fp16_to_fp32:
>> ; SI: BUFFER_LOAD_USHORT [[VAL:v[0-9]+]]
>> ; SI: V_CVT_F32_F16_e32 [[RESULT:v[0-9]+]], [[VAL]]
>> ; SI: BUFFER_STORE_DWORD [[RESULT]]
>> @@ -15,7 +15,7 @@ define void @test_convert_fp16_to_fp32(float addrspace(1)* noalias %out, i16 add
>> }
>>
>>
>> -; SI-LABEL: @test_convert_fp16_to_fp64:
>> +; SI-LABEL: test_convert_fp16_to_fp64:
>> ; SI: BUFFER_LOAD_USHORT [[VAL:v[0-9]+]]
>> ; SI: V_CVT_F32_F16_e32 [[RESULT32:v[0-9]+]], [[VAL]]
>> ; SI: V_CVT_F64_F32_e32 [[RESULT:v\[[0-9]+:[0-9]+\]]], [[RESULT32]]
>> diff --git a/test/CodeGen/R600/fp32_to_fp16.ll b/test/CodeGen/R600/fp32_to_fp16.ll
>> index 6b5ff00..705f492 100644
>> --- a/test/CodeGen/R600/fp32_to_fp16.ll
>> +++ b/test/CodeGen/R600/fp32_to_fp16.ll
>> @@ -2,7 +2,7 @@
>>
>> declare i16 @llvm.convert.to.fp16.f32(float) nounwind readnone
>>
>> -; SI-LABEL: @test_convert_fp32_to_fp16:
>> +; SI-LABEL: test_convert_fp32_to_fp16:
>> ; SI: BUFFER_LOAD_DWORD [[VAL:v[0-9]+]]
>> ; SI: V_CVT_F16_F32_e32 [[RESULT:v[0-9]+]], [[VAL]]
>> ; SI: BUFFER_STORE_SHORT [[RESULT]]
>> diff --git a/test/CodeGen/R600/fp64_to_sint.ll b/test/CodeGen/R600/fp64_to_sint.ll
>> index 221b32d..3366c5d 100644
>> --- a/test/CodeGen/R600/fp64_to_sint.ll
>> +++ b/test/CodeGen/R600/fp64_to_sint.ll
>> @@ -1,6 +1,6 @@
>> ; RUN: llc -march=r600 -mcpu=SI -verify-machineinstrs < %s | FileCheck -check-prefix=SI -check-prefix=FUNC %s
>>
>> -; FUNC-LABEL: @fp_to_sint_f64_i32
>> +; FUNC-LABEL: fp_to_sint_f64_i32
>> ; SI: V_CVT_I32_F64_e32
>> define void @fp_to_sint_f64_i32(i32 addrspace(1)* %out, double %in) {
>> %result = fptosi double %in to i32
>> @@ -8,7 +8,7 @@ define void @fp_to_sint_f64_i32(i32 addrspace(1)* %out, double %in) {
>> ret void
>> }
>>
>> -; FUNC-LABEL: @fp_to_sint_v2f64_v2i32
>> +; FUNC-LABEL: fp_to_sint_v2f64_v2i32
>> ; SI: V_CVT_I32_F64_e32
>> ; SI: V_CVT_I32_F64_e32
>> define void @fp_to_sint_v2f64_v2i32(<2 x i32> addrspace(1)* %out, <2 x double> %in) {
>> @@ -17,7 +17,7 @@ define void @fp_to_sint_v2f64_v2i32(<2 x i32> addrspace(1)* %out, <2 x double> %
>> ret void
>> }
>>
>> -; FUNC-LABEL: @fp_to_sint_v4f64_v4i32
>> +; FUNC-LABEL: fp_to_sint_v4f64_v4i32
>> ; SI: V_CVT_I32_F64_e32
>> ; SI: V_CVT_I32_F64_e32
>> ; SI: V_CVT_I32_F64_e32
>> diff --git a/test/CodeGen/R600/fp_to_sint.ll b/test/CodeGen/R600/fp_to_sint.ll
>> index 322a55b..6fd4384 100644
>> --- a/test/CodeGen/R600/fp_to_sint.ll
>> +++ b/test/CodeGen/R600/fp_to_sint.ll
>> @@ -1,7 +1,7 @@
>> ; RUN: llc -march=r600 -mcpu=redwood < %s | FileCheck %s --check-prefix=EG --check-prefix=FUNC
>> ; RUN: llc -march=r600 -mcpu=SI -verify-machineinstrs < %s | FileCheck %s --check-prefix=SI --check-prefix=FUNC
>>
>> -; FUNC-LABEL: @fp_to_sint_i32
>> +; FUNC-LABEL: fp_to_sint_i32
>> ; EG: FLT_TO_INT {{\** *}}T{{[0-9]+\.[XYZW], PV\.[XYZW]}}
>> ; SI: V_CVT_I32_F32_e32
>> ; SI: S_ENDPGM
>> @@ -11,7 +11,7 @@ define void @fp_to_sint_i32 (i32 addrspace(1)* %out, float %in) {
>> ret void
>> }
>>
>> -; FUNC-LABEL: @fp_to_sint_v2i32
>> +; FUNC-LABEL: fp_to_sint_v2i32
>> ; EG: FLT_TO_INT {{\** *}}T{{[0-9]+\.[XYZW], PV\.[XYZW]}}
>> ; EG: FLT_TO_INT {{\** *}}T{{[0-9]+\.[XYZW], PV\.[XYZW]}}
>> ; SI: V_CVT_I32_F32_e32
>> @@ -22,7 +22,7 @@ define void @fp_to_sint_v2i32(<2 x i32> addrspace(1)* %out, <2 x float> %in) {
>> ret void
>> }
>>
>> -; FUNC-LABEL: @fp_to_sint_v4i32
>> +; FUNC-LABEL: fp_to_sint_v4i32
>> ; EG: FLT_TO_INT {{\** *}}T{{[0-9]+\.[XYZW], PV\.[XYZW]}}
>> ; EG: FLT_TO_INT {{\** *}}T{{[0-9]+\.[XYZW]}}
>> ; EG: FLT_TO_INT {{\** *}}T{{[0-9]+\.[XYZW], PV\.[XYZW]}}
>> @@ -38,7 +38,7 @@ define void @fp_to_sint_v4i32(<4 x i32> addrspace(1)* %out, <4 x float> addrspac
>> ret void
>> }
>>
>> -; FUNC-LABEL: @fp_to_sint_i64
>> +; FUNC-LABEL: fp_to_sint_i64
>>
>> ; EG-DAG: AND_INT
>> ; EG-DAG: LSHR
>> @@ -71,7 +71,7 @@ entry:
>> ret void
>> }
>>
>> -; FUNC: @fp_to_sint_v2i64
>> +; FUNC: fp_to_sint_v2i64
>> ; EG-DAG: AND_INT
>> ; EG-DAG: LSHR
>> ; EG-DAG: SUB_INT
>> @@ -122,7 +122,7 @@ define void @fp_to_sint_v2i64(<2 x i64> addrspace(1)* %out, <2 x float> %x) {
>> ret void
>> }
>>
>> -; FUNC: @fp_to_sint_v4i64
>> +; FUNC: fp_to_sint_v4i64
>> ; EG-DAG: AND_INT
>> ; EG-DAG: LSHR
>> ; EG-DAG: SUB_INT
>> diff --git a/test/CodeGen/R600/fp_to_uint.f64.ll b/test/CodeGen/R600/fp_to_uint.f64.ll
>> index bf607ce..26db352 100644
>> --- a/test/CodeGen/R600/fp_to_uint.f64.ll
>> +++ b/test/CodeGen/R600/fp_to_uint.f64.ll
>> @@ -1,6 +1,6 @@
>> ; RUN: llc -march=r600 -mcpu=SI -verify-machineinstrs < %s | FileCheck -check-prefix=SI %s
>>
>> -; SI-LABEL: @fp_to_uint_i32_f64
>> +; SI-LABEL: fp_to_uint_i32_f64
>> ; SI: V_CVT_U32_F64_e32
>> define void @fp_to_uint_i32_f64(i32 addrspace(1)* %out, double %in) {
>> %cast = fptoui double %in to i32
>> diff --git a/test/CodeGen/R600/fp_to_uint.ll b/test/CodeGen/R600/fp_to_uint.ll
>> index 6e71823..552c915 100644
>> --- a/test/CodeGen/R600/fp_to_uint.ll
>> +++ b/test/CodeGen/R600/fp_to_uint.ll
>> @@ -1,7 +1,7 @@
>> ; RUN: llc -march=r600 -mcpu=redwood < %s | FileCheck %s --check-prefix=EG --check-prefix=FUNC
>> ; RUN: llc -march=r600 -mcpu=SI -verify-machineinstrs < %s | FileCheck %s --check-prefix=SI --check-prefix=FUNC
>>
>> -; FUNC-LABEL: @fp_to_uint_i32
>> +; FUNC-LABEL: fp_to_uint_i32
>> ; EG: FLT_TO_UINT {{\** *}}T{{[0-9]+\.[XYZW], PV\.[XYZW]}}
>> ; SI: V_CVT_U32_F32_e32
>> ; SI: S_ENDPGM
>> @@ -11,7 +11,7 @@ define void @fp_to_uint_i32 (i32 addrspace(1)* %out, float %in) {
>> ret void
>> }
>>
>> -; FUNC-LABEL: @fp_to_uint_v2i32
>> +; FUNC-LABEL: fp_to_uint_v2i32
>> ; EG: FLT_TO_UINT {{\** *}}T{{[0-9]+\.[XYZW], PV\.[XYZW]}}
>> ; EG: FLT_TO_UINT {{\** *}}T{{[0-9]+\.[XYZW], T[0-9]+\.[XYZW]}}
>> ; SI: V_CVT_U32_F32_e32
>> @@ -23,7 +23,7 @@ define void @fp_to_uint_v2i32(<2 x i32> addrspace(1)* %out, <2 x float> %in) {
>> ret void
>> }
>>
>> -; FUNC-LABEL: @fp_to_uint_v4i32
>> +; FUNC-LABEL: fp_to_uint_v4i32
>> ; EG: FLT_TO_UINT {{\** *}}T{{[0-9]+\.[XYZW], PV\.[XYZW]}}
>> ; EG: FLT_TO_UINT {{\** *}}T{{[0-9]+\.[XYZW], T[0-9]+\.[XYZW]}}
>> ; EG: FLT_TO_UINT {{\** *}}T{{[0-9]+\.[XYZW], PV\.[XYZW]}}
>> @@ -40,7 +40,7 @@ define void @fp_to_uint_v4i32(<4 x i32> addrspace(1)* %out, <4 x float> addrspac
>> ret void
>> }
>>
>> -; FUNC: @fp_to_uint_i64
>> +; FUNC: fp_to_uint_i64
>> ; EG-DAG: AND_INT
>> ; EG-DAG: LSHR
>> ; EG-DAG: SUB_INT
>> @@ -70,7 +70,7 @@ define void @fp_to_uint_i64(i64 addrspace(1)* %out, float %x) {
>> ret void
>> }
>>
>> -; FUNC: @fp_to_uint_v2i64
>> +; FUNC: fp_to_uint_v2i64
>> ; EG-DAG: AND_INT
>> ; EG-DAG: LSHR
>> ; EG-DAG: SUB_INT
>> @@ -121,7 +121,7 @@ define void @fp_to_uint_v2i64(<2 x i64> addrspace(1)* %out, <2 x float> %x) {
>> ret void
>> }
>>
>> -; FUNC: @fp_to_uint_v4i64
>> +; FUNC: fp_to_uint_v4i64
>> ; EG-DAG: AND_INT
>> ; EG-DAG: LSHR
>> ; EG-DAG: SUB_INT
>> diff --git a/test/CodeGen/R600/fpext.ll b/test/CodeGen/R600/fpext.ll
>> index 143ee79..3f23409 100644
>> --- a/test/CodeGen/R600/fpext.ll
>> +++ b/test/CodeGen/R600/fpext.ll
>> @@ -1,6 +1,6 @@
>> ; RUN: llc < %s -march=r600 -mcpu=SI -verify-machineinstrs | FileCheck %s --check-prefix=CHECK
>>
>> -; CHECK: @fpext
>> +; CHECK: fpext
>> ; CHECK: V_CVT_F64_F32_e32
>> define void @fpext(double addrspace(1)* %out, float %in) {
>> %result = fpext float %in to double
>> diff --git a/test/CodeGen/R600/fptrunc.ll b/test/CodeGen/R600/fptrunc.ll
>> index 20a8c00..1427e93 100644
>> --- a/test/CodeGen/R600/fptrunc.ll
>> +++ b/test/CodeGen/R600/fptrunc.ll
>> @@ -1,6 +1,6 @@
>> ; RUN: llc < %s -march=r600 -mcpu=SI -verify-machineinstrs | FileCheck %s --check-prefix=CHECK
>>
>> -; CHECK: @fptrunc
>> +; CHECK: fptrunc
>> ; CHECK: V_CVT_F32_F64_e32
>> define void @fptrunc(float addrspace(1)* %out, double %in) {
>> %result = fptrunc double %in to float
>> diff --git a/test/CodeGen/R600/frem.ll b/test/CodeGen/R600/frem.ll
>> index d415384..a41eeac 100644
>> --- a/test/CodeGen/R600/frem.ll
>> +++ b/test/CodeGen/R600/frem.ll
>> @@ -1,6 +1,6 @@
>> ; RUN: llc -march=r600 -mcpu=SI -enable-misched < %s | FileCheck -check-prefix=SI -check-prefix=FUNC %s
>>
>> -; FUNC-LABEL: @frem_f32:
>> +; FUNC-LABEL: frem_f32:
>> ; SI-DAG: BUFFER_LOAD_DWORD [[X:v[0-9]+]], {{.*$}}
>> ; SI-DAG: BUFFER_LOAD_DWORD [[Y:v[0-9]+]], {{.*}} offset:0x10
>> ; SI-DAG: V_CMP
>> @@ -21,7 +21,7 @@ define void @frem_f32(float addrspace(1)* %out, float addrspace(1)* %in1,
>> ret void
>> }
>>
>> -; FUNC-LABEL: @unsafe_frem_f32:
>> +; FUNC-LABEL: unsafe_frem_f32:
>> ; SI: BUFFER_LOAD_DWORD [[Y:v[0-9]+]], {{.*}} offset:0x10
>> ; SI: BUFFER_LOAD_DWORD [[X:v[0-9]+]], {{.*}}
>> ; SI: V_RCP_F32_e32 [[INVY:v[0-9]+]], [[Y]]
>> @@ -43,7 +43,7 @@ define void @unsafe_frem_f32(float addrspace(1)* %out, float addrspace(1)* %in1,
>> ; TODO: This should check something when f64 fdiv is implemented
>> ; correctly
>>
>> -; FUNC-LABEL: @frem_f64:
>> +; FUNC-LABEL: frem_f64:
>> ; SI: S_ENDPGM
>> define void @frem_f64(double addrspace(1)* %out, double addrspace(1)* %in1,
>> double addrspace(1)* %in2) #0 {
>> @@ -54,7 +54,7 @@ define void @frem_f64(double addrspace(1)* %out, double addrspace(1)* %in1,
>> ret void
>> }
>>
>> -; FUNC-LABEL: @unsafe_frem_f64:
>> +; FUNC-LABEL: unsafe_frem_f64:
>> ; SI: V_RCP_F64_e32
>> ; SI: V_MUL_F64
>> ; SI: V_BFE_I32
>> diff --git a/test/CodeGen/R600/fsqrt.ll b/test/CodeGen/R600/fsqrt.ll
>> index ae50b17..7662fad 100644
>> --- a/test/CodeGen/R600/fsqrt.ll
>> +++ b/test/CodeGen/R600/fsqrt.ll
>> @@ -1,6 +1,6 @@
>> ; RUN: llc < %s -march=r600 -mcpu=tahiti -verify-machineinstrs | FileCheck %s
>>
>> -; CHECK: @fsqrt_f32
>> +; CHECK: fsqrt_f32
>> ; CHECK: V_SQRT_F32_e32 {{v[0-9]+, v[0-9]+}}
>>
>> define void @fsqrt_f32(float addrspace(1)* %out, float addrspace(1)* %in) {
>> @@ -10,7 +10,7 @@ define void @fsqrt_f32(float addrspace(1)* %out, float addrspace(1)* %in) {
>> ret void
>> }
>>
>> -; CHECK: @fsqrt_f64
>> +; CHECK: fsqrt_f64
>> ; CHECK: V_SQRT_F64_e32 {{v\[[0-9]+:[0-9]+\], v\[[0-9]+:[0-9]+\]}}
>>
>> define void @fsqrt_f64(double addrspace(1)* %out, double addrspace(1)* %in) {
>> diff --git a/test/CodeGen/R600/fsub.ll b/test/CodeGen/R600/fsub.ll
>> index 22c6268..b0511cc 100644
>> --- a/test/CodeGen/R600/fsub.ll
>> +++ b/test/CodeGen/R600/fsub.ll
>> @@ -2,7 +2,7 @@
>> ; RUN: llc -march=r600 -mcpu=SI -verify-machineinstrs < %s | FileCheck -check-prefix=SI -check-prefix=FUNC %s
>>
>>
>> -; FUNC-LABEL: @v_fsub_f32
>> +; FUNC-LABEL: v_fsub_f32
>> ; SI: V_SUBREV_F32_e32 {{v[0-9]+}}, {{v[0-9]+}}, {{v[0-9]+}}
>> define void @v_fsub_f32(float addrspace(1)* %out, float addrspace(1)* %in) {
>> %b_ptr = getelementptr float addrspace(1)* %in, i32 1
>> @@ -13,7 +13,7 @@ define void @v_fsub_f32(float addrspace(1)* %out, float addrspace(1)* %in) {
>> ret void
>> }
>>
>> -; FUNC-LABEL: @s_fsub_f32
>> +; FUNC-LABEL: s_fsub_f32
>> ; R600: ADD {{\** *}}T{{[0-9]+\.[XYZW]}}, KC0[2].Z, -KC0[2].W
>>
>> ; SI: V_SUB_F32_e32 {{v[0-9]+}}, {{s[0-9]+}}, {{v[0-9]+}}
>> @@ -27,7 +27,7 @@ declare float @llvm.R600.load.input(i32) readnone
>>
>> declare void @llvm.AMDGPU.store.output(float, i32)
>>
>> -; FUNC-LABEL: @fsub_v2f32
>> +; FUNC-LABEL: fsub_v2f32
>> ; R600-DAG: ADD {{\** *}}T{{[0-9]+\.[XYZW]}}, KC0[3].X, -KC0[3].Z
>> ; R600-DAG: ADD {{\** *}}T{{[0-9]+\.[XYZW]}}, KC0[2].W, -KC0[3].Y
>>
>> @@ -40,7 +40,7 @@ define void @fsub_v2f32(<2 x float> addrspace(1)* %out, <2 x float> %a, <2 x flo
>> ret void
>> }
>>
>> -; FUNC-LABEL: @v_fsub_v4f32
>> +; FUNC-LABEL: v_fsub_v4f32
>> ; R600: ADD {{\** *}}T{{[0-9]+\.[XYZW], T[0-9]+\.[XYZW], -T[0-9]+\.[XYZW]}}
>> ; R600: ADD {{\** *}}T{{[0-9]+\.[XYZW], T[0-9]+\.[XYZW], -T[0-9]+\.[XYZW]}}
>> ; R600: ADD {{\** *}}T{{[0-9]+\.[XYZW], T[0-9]+\.[XYZW], -T[0-9]+\.[XYZW]}}
>> @@ -61,7 +61,7 @@ define void @v_fsub_v4f32(<4 x float> addrspace(1)* %out, <4 x float> addrspace(
>>
>> ; FIXME: Should be using SGPR directly for first operand
>>
>> -; FUNC-LABEL: @s_fsub_v4f32
>> +; FUNC-LABEL: s_fsub_v4f32
>> ; SI: V_SUBREV_F32_e32 {{v[0-9]+}}, {{v[0-9]+}}, {{v[0-9]+}}
>> ; SI: V_SUBREV_F32_e32 {{v[0-9]+}}, {{v[0-9]+}}, {{v[0-9]+}}
>> ; SI: V_SUBREV_F32_e32 {{v[0-9]+}}, {{v[0-9]+}}, {{v[0-9]+}}
>> diff --git a/test/CodeGen/R600/fsub64.ll b/test/CodeGen/R600/fsub64.ll
>> index f5e5708..90396cd 100644
>> --- a/test/CodeGen/R600/fsub64.ll
>> +++ b/test/CodeGen/R600/fsub64.ll
>> @@ -1,6 +1,6 @@
>> ; RUN: llc -march=r600 -mcpu=tahiti -verify-machineinstrs < %s | FileCheck -check-prefix=SI %s
>>
>> -; SI-LABEL: @fsub_f64:
>> +; SI-LABEL: fsub_f64:
>> ; SI: V_ADD_F64 {{v\[[0-9]+:[0-9]+\], v\[[0-9]+:[0-9]+\], -v\[[0-9]+:[0-9]+\]}}
>> define void @fsub_f64(double addrspace(1)* %out, double addrspace(1)* %in1,
>> double addrspace(1)* %in2) {
>> diff --git a/test/CodeGen/R600/ftrunc.ll b/test/CodeGen/R600/ftrunc.ll
>> index 0d7d467..79ccd01 100644
>> --- a/test/CodeGen/R600/ftrunc.ll
>> +++ b/test/CodeGen/R600/ftrunc.ll
>> @@ -8,7 +8,7 @@ declare <4 x float> @llvm.trunc.v4f32(<4 x float>) nounwind readnone
>> declare <8 x float> @llvm.trunc.v8f32(<8 x float>) nounwind readnone
>> declare <16 x float> @llvm.trunc.v16f32(<16 x float>) nounwind readnone
>>
>> -; FUNC-LABEL: @ftrunc_f32:
>> +; FUNC-LABEL: ftrunc_f32:
>> ; EG: TRUNC
>> ; SI: V_TRUNC_F32_e32
>> define void @ftrunc_f32(float addrspace(1)* %out, float %x) {
>> @@ -17,7 +17,7 @@ define void @ftrunc_f32(float addrspace(1)* %out, float %x) {
>> ret void
>> }
>>
>> -; FUNC-LABEL: @ftrunc_v2f32:
>> +; FUNC-LABEL: ftrunc_v2f32:
>> ; EG: TRUNC
>> ; EG: TRUNC
>> ; SI: V_TRUNC_F32_e32
>> @@ -28,7 +28,7 @@ define void @ftrunc_v2f32(<2 x float> addrspace(1)* %out, <2 x float> %x) {
>> ret void
>> }
>>
>> -; FIXME-FUNC-LABEL: @ftrunc_v3f32:
>> +; FIXME-FUNC-LABEL: ftrunc_v3f32:
>> ; FIXME-EG: TRUNC
>> ; FIXME-EG: TRUNC
>> ; FIXME-EG: TRUNC
>> @@ -41,7 +41,7 @@ define void @ftrunc_v2f32(<2 x float> addrspace(1)* %out, <2 x float> %x) {
>> ; ret void
>> ; }
>>
>> -; FUNC-LABEL: @ftrunc_v4f32:
>> +; FUNC-LABEL: ftrunc_v4f32:
>> ; EG: TRUNC
>> ; EG: TRUNC
>> ; EG: TRUNC
>> @@ -56,7 +56,7 @@ define void @ftrunc_v4f32(<4 x float> addrspace(1)* %out, <4 x float> %x) {
>> ret void
>> }
>>
>> -; FUNC-LABEL: @ftrunc_v8f32:
>> +; FUNC-LABEL: ftrunc_v8f32:
>> ; EG: TRUNC
>> ; EG: TRUNC
>> ; EG: TRUNC
>> @@ -79,7 +79,7 @@ define void @ftrunc_v8f32(<8 x float> addrspace(1)* %out, <8 x float> %x) {
>> ret void
>> }
>>
>> -; FUNC-LABEL: @ftrunc_v16f32:
>> +; FUNC-LABEL: ftrunc_v16f32:
>> ; EG: TRUNC
>> ; EG: TRUNC
>> ; EG: TRUNC
>> diff --git a/test/CodeGen/R600/gep-address-space.ll b/test/CodeGen/R600/gep-address-space.ll
>> index cd698f4..cfeb24d 100644
>> --- a/test/CodeGen/R600/gep-address-space.ll
>> +++ b/test/CodeGen/R600/gep-address-space.ll
>> @@ -2,7 +2,7 @@
>> ; RUN: llc -march=r600 -mcpu=bonaire -verify-machineinstrs< %s | FileCheck --check-prefix=CI --check-prefix=CHECK %s
>>
>> define void @use_gep_address_space([1024 x i32] addrspace(3)* %array) nounwind {
>> -; CHECK-LABEL: @use_gep_address_space:
>> +; CHECK-LABEL: use_gep_address_space:
>> ; CHECK: V_MOV_B32_e32 [[PTR:v[0-9]+]], s{{[0-9]+}}
>> ; CHECK: DS_WRITE_B32 [[PTR]], v{{[0-9]+}}, 0x40
>> %p = getelementptr [1024 x i32] addrspace(3)* %array, i16 0, i16 16
>> @@ -11,7 +11,7 @@ define void @use_gep_address_space([1024 x i32] addrspace(3)* %array) nounwind {
>> }
>>
>> define void @use_gep_address_space_large_offset([1024 x i32] addrspace(3)* %array) nounwind {
>> -; CHECK-LABEL: @use_gep_address_space_large_offset:
>> +; CHECK-LABEL: use_gep_address_space_large_offset:
>> ; The LDS offset will be 65536 bytes, which is larger than the size of LDS on
>> ; SI, which is why it is being OR'd with the base pointer.
>> ; SI: S_OR_B32
>> @@ -23,7 +23,7 @@ define void @use_gep_address_space_large_offset([1024 x i32] addrspace(3)* %arra
>> }
>>
>> define void @gep_as_vector_v4(<4 x [1024 x i32] addrspace(3)*> %array) nounwind {
>> -; CHECK-LABEL: @gep_as_vector_v4:
>> +; CHECK-LABEL: gep_as_vector_v4:
>> ; CHECK: S_ADD_I32
>> ; CHECK: S_ADD_I32
>> ; CHECK: S_ADD_I32
>> @@ -41,7 +41,7 @@ define void @gep_as_vector_v4(<4 x [1024 x i32] addrspace(3)*> %array) nounwind
>> }
>>
>> define void @gep_as_vector_v2(<2 x [1024 x i32] addrspace(3)*> %array) nounwind {
>> -; CHECK-LABEL: @gep_as_vector_v2:
>> +; CHECK-LABEL: gep_as_vector_v2:
>> ; CHECK: S_ADD_I32
>> ; CHECK: S_ADD_I32
>> %p = getelementptr <2 x [1024 x i32] addrspace(3)*> %array, <2 x i16> zeroinitializer, <2 x i16> <i16 16, i16 16>
>> diff --git a/test/CodeGen/R600/global_atomics.ll b/test/CodeGen/R600/global_atomics.ll
>> index 665913f..2b81c8f 100644
>> --- a/test/CodeGen/R600/global_atomics.ll
>> +++ b/test/CodeGen/R600/global_atomics.ll
>> @@ -1,6 +1,6 @@
>> ; RUN: llc < %s -march=r600 -mcpu=SI -verify-machineinstrs | FileCheck --check-prefix=SI --check-prefix=FUNC %s
>>
>> -; FUNC-LABEL: @atomic_load_i32_offset
>> +; FUNC-LABEL: atomic_load_i32_offset
>> ; SI: BUFFER_ATOMIC_ADD v{{[0-9]+}}, s[{{[0-9]+}}:{{[0-9]+}}], 0{{$}}
>> define void @atomic_load_i32_offset(i32 addrspace(1)* %out, i32 %in) {
>> entry:
>> @@ -8,7 +8,7 @@ entry:
>> ret void
>> }
>>
>> -; FUNC-LABEL: @atomic_load_i32_ret_offset
>> +; FUNC-LABEL: atomic_load_i32_ret_offset
>> ; SI: BUFFER_ATOMIC_ADD [[RET:v[0-9]+]], s[{{[0-9]+}}:{{[0-9]+}}], 0 glc
>> ; SI: BUFFER_STORE_DWORD [[RET]]
>> define void @atomic_load_i32_ret_offset(i32 addrspace(1)* %out, i32 addrspace(1)* %out2, i32 %in) {
>> @@ -18,7 +18,7 @@ entry:
>> ret void
>> }
>>
>> -; FUNC-LABEL: @atomic_load_i32_addr64
>> +; FUNC-LABEL: atomic_load_i32_addr64
>> ; SI: BUFFER_ATOMIC_ADD v{{[0-9]+}}, v[{{[0-9]+}}:{{[0-9]+}}], s[{{[0-9]+}}:{{[0-9]+}}], 0 addr64{{$}}
>> define void @atomic_load_i32_addr64(i32 addrspace(1)* %out, i32 %in, i64 %index) {
>> entry:
>> @@ -27,7 +27,7 @@ entry:
>> ret void
>> }
>>
>> -; FUNC-LABEL: @atomic_load_i32_ret_addr64
>> +; FUNC-LABEL: atomic_load_i32_ret_addr64
>> ; SI: BUFFER_ATOMIC_ADD [[RET:v[0-9]+]], v[{{[0-9]+}}:{{[0-9]+}}], s[{{[0-9]+}}:{{[0-9]+}}], 0 addr64 glc{{$}}
>> ; SI: BUFFER_STORE_DWORD [[RET]]
>> define void @atomic_load_i32_ret_addr64(i32 addrspace(1)* %out, i32 addrspace(1)* %out2, i32 %in, i64 %index) {
>> diff --git a/test/CodeGen/R600/gv-const-addrspace-fail.ll b/test/CodeGen/R600/gv-const-addrspace-fail.ll
>> index 7ad55ed..96efe9d 100644
>> --- a/test/CodeGen/R600/gv-const-addrspace-fail.ll
>> +++ b/test/CodeGen/R600/gv-const-addrspace-fail.ll
>> @@ -4,7 +4,7 @@
>>
>> @a = internal addrspace(2) constant [1 x i8] [ i8 7 ], align 1
>>
>> -; FUNC-LABEL: @test_i8
>> +; FUNC-LABEL: test_i8
>> ; EG: CF_END
>> ; SI: BUFFER_STORE_BYTE
>> ; SI: S_ENDPGM
>> @@ -17,7 +17,7 @@ define void @test_i8( i32 %s, i8 addrspace(1)* %out) #3 {
>>
>> @b = internal addrspace(2) constant [1 x i16] [ i16 7 ], align 2
>>
>> -; FUNC-LABEL: @test_i16
>> +; FUNC-LABEL: test_i16
>> ; EG: CF_END
>> ; SI: BUFFER_STORE_SHORT
>> ; SI: S_ENDPGM
>> @@ -33,7 +33,7 @@ define void @test_i16( i32 %s, i16 addrspace(1)* %out) #3 {
>> ; The illegal i8s aren't handled
>> @struct_bar_gv = internal addrspace(2) constant [1 x %struct.bar] [ %struct.bar { float 16.0, [5 x i8] [i8 0, i8 1, i8 2, i8 3, i8 4] } ]
>>
>> -; FUNC-LABEL: @struct_bar_gv_load
>> +; FUNC-LABEL: struct_bar_gv_load
>> define void @struct_bar_gv_load(i8 addrspace(1)* %out, i32 %index) {
>> %gep = getelementptr inbounds [1 x %struct.bar] addrspace(2)* @struct_bar_gv, i32 0, i32 0, i32 1, i32 %index
>> %load = load i8 addrspace(2)* %gep, align 1
>> @@ -48,7 +48,7 @@ define void @struct_bar_gv_load(i8 addrspace(1)* %out, i32 %index) {
>> <4 x i32> <i32 9, i32 10, i32 11, i32 12>,
>> <4 x i32> <i32 13, i32 14, i32 15, i32 16> ]
>>
>> -; FUNC-LABEL: @array_vector_gv_load
>> +; FUNC-LABEL: array_vector_gv_load
>> define void @array_vector_gv_load(<4 x i32> addrspace(1)* %out, i32 %index) {
>> %gep = getelementptr inbounds [4 x <4 x i32>] addrspace(2)* @array_vector_gv, i32 0, i32 %index
>> %load = load <4 x i32> addrspace(2)* %gep, align 16
>> diff --git a/test/CodeGen/R600/gv-const-addrspace.ll b/test/CodeGen/R600/gv-const-addrspace.ll
>> index e0ac317..70596ec 100644
>> --- a/test/CodeGen/R600/gv-const-addrspace.ll
>> +++ b/test/CodeGen/R600/gv-const-addrspace.ll
>> @@ -6,7 +6,7 @@
>>
>> @float_gv = internal unnamed_addr addrspace(2) constant [5 x float] [float 0.0, float 1.0, float 2.0, float 3.0, float 4.0], align 4
>>
>> -; FUNC-LABEL: @float
>> +; FUNC-LABEL: float
>> ; FIXME: We should be using S_LOAD_DWORD here.
>> ; SI: BUFFER_LOAD_DWORD
>>
>> @@ -27,7 +27,7 @@ entry:
>>
>> @i32_gv = internal unnamed_addr addrspace(2) constant [5 x i32] [i32 0, i32 1, i32 2, i32 3, i32 4], align 4
>>
>> -; FUNC-LABEL: @i32
>> +; FUNC-LABEL: i32
>>
>> ; FIXME: We should be using S_LOAD_DWORD here.
>> ; SI: BUFFER_LOAD_DWORD
>> @@ -52,7 +52,7 @@ entry:
>>
>> @struct_foo_gv = internal unnamed_addr addrspace(2) constant [1 x %struct.foo] [ %struct.foo { float 16.0, [5 x i32] [i32 0, i32 1, i32 2, i32 3, i32 4] } ]
>>
>> -; FUNC-LABEL: @struct_foo_gv_load
>> +; FUNC-LABEL: struct_foo_gv_load
>> ; SI: S_LOAD_DWORD
>>
>> define void @struct_foo_gv_load(i32 addrspace(1)* %out, i32 %index) {
>> @@ -67,7 +67,7 @@ define void @struct_foo_gv_load(i32 addrspace(1)* %out, i32 %index) {
>> <1 x i32> <i32 3>,
>> <1 x i32> <i32 4> ]
>>
>> -; FUNC-LABEL: @array_v1_gv_load
>> +; FUNC-LABEL: array_v1_gv_load
>> ; FIXME: We should be using S_LOAD_DWORD here.
>> ; SI: BUFFER_LOAD_DWORD
>> define void @array_v1_gv_load(<1 x i32> addrspace(1)* %out, i32 %index) {
>> diff --git a/test/CodeGen/R600/half.ll b/test/CodeGen/R600/half.ll
>> index 42aa4fa..471b9bf 100644
>> --- a/test/CodeGen/R600/half.ll
>> +++ b/test/CodeGen/R600/half.ll
>> @@ -1,7 +1,7 @@
>> ; RUN: llc < %s -march=r600 -mcpu=SI | FileCheck %s
>>
>> define void @test_load_store(half addrspace(1)* %in, half addrspace(1)* %out) {
>> -; CHECK-LABEL: @test_load_store
>> +; CHECK-LABEL: test_load_store
>> ; CHECK: BUFFER_LOAD_USHORT [[TMP:v[0-9]+]]
>> ; CHECK: BUFFER_STORE_SHORT [[TMP]]
>> %val = load half addrspace(1)* %in
>> @@ -10,7 +10,7 @@ define void @test_load_store(half addrspace(1)* %in, half addrspace(1)* %out) {
>> }
>>
>> define void @test_bitcast_from_half(half addrspace(1)* %in, i16 addrspace(1)* %out) {
>> -; CHECK-LABEL: @test_bitcast_from_half
>> +; CHECK-LABEL: test_bitcast_from_half
>> ; CHECK: BUFFER_LOAD_USHORT [[TMP:v[0-9]+]]
>> ; CHECK: BUFFER_STORE_SHORT [[TMP]]
>> %val = load half addrspace(1) * %in
>> @@ -20,7 +20,7 @@ define void @test_bitcast_from_half(half addrspace(1)* %in, i16 addrspace(1)* %o
>> }
>>
>> define void @test_bitcast_to_half(half addrspace(1)* %out, i16 addrspace(1)* %in) {
>> -; CHECK-LABEL: @test_bitcast_to_half
>> +; CHECK-LABEL: test_bitcast_to_half
>> ; CHECK: BUFFER_LOAD_USHORT [[TMP:v[0-9]+]]
>> ; CHECK: BUFFER_STORE_SHORT [[TMP]]
>> %val = load i16 addrspace(1)* %in
>> @@ -30,7 +30,7 @@ define void @test_bitcast_to_half(half addrspace(1)* %out, i16 addrspace(1)* %in
>> }
>>
>> define void @test_extend32(half addrspace(1)* %in, float addrspace(1)* %out) {
>> -; CHECK-LABEL: @test_extend32
>> +; CHECK-LABEL: test_extend32
>> ; CHECK: V_CVT_F32_F16_e32
>>
>> %val16 = load half addrspace(1)* %in
>> @@ -40,7 +40,7 @@ define void @test_extend32(half addrspace(1)* %in, float addrspace(1)* %out) {
>> }
>>
>> define void @test_extend64(half addrspace(1)* %in, double addrspace(1)* %out) {
>> -; CHECK-LABEL: @test_extend64
>> +; CHECK-LABEL: test_extend64
>> ; CHECK: V_CVT_F32_F16_e32
>> ; CHECK: V_CVT_F64_F32_e32
>>
>> @@ -51,7 +51,7 @@ define void @test_extend64(half addrspace(1)* %in, double addrspace(1)* %out) {
>> }
>>
>> define void @test_trunc32(float addrspace(1)* %in, half addrspace(1)* %out) {
>> -; CHECK-LABEL: @test_trunc32
>> +; CHECK-LABEL: test_trunc32
>> ; CHECK: V_CVT_F16_F32_e32
>>
>> %val32 = load float addrspace(1)* %in
>> diff --git a/test/CodeGen/R600/icmp64.ll b/test/CodeGen/R600/icmp64.ll
>> index c9e62ff..c869539 100644
>> --- a/test/CodeGen/R600/icmp64.ll
>> +++ b/test/CodeGen/R600/icmp64.ll
>> @@ -1,6 +1,6 @@
>> ; RUN: llc -march=r600 -mcpu=SI -verify-machineinstrs < %s | FileCheck -check-prefix=SI %s
>>
>> -; SI-LABEL: @test_i64_eq:
>> +; SI-LABEL: test_i64_eq:
>> ; SI: V_CMP_EQ_I64
>> define void @test_i64_eq(i32 addrspace(1)* %out, i64 %a, i64 %b) nounwind {
>> %cmp = icmp eq i64 %a, %b
>> @@ -9,7 +9,7 @@ define void @test_i64_eq(i32 addrspace(1)* %out, i64 %a, i64 %b) nounwind {
>> ret void
>> }
>>
>> -; SI-LABEL: @test_i64_ne:
>> +; SI-LABEL: test_i64_ne:
>> ; SI: V_CMP_NE_I64
>> define void @test_i64_ne(i32 addrspace(1)* %out, i64 %a, i64 %b) nounwind {
>> %cmp = icmp ne i64 %a, %b
>> @@ -18,7 +18,7 @@ define void @test_i64_ne(i32 addrspace(1)* %out, i64 %a, i64 %b) nounwind {
>> ret void
>> }
>>
>> -; SI-LABEL: @test_i64_slt:
>> +; SI-LABEL: test_i64_slt:
>> ; SI: V_CMP_LT_I64
>> define void @test_i64_slt(i32 addrspace(1)* %out, i64 %a, i64 %b) nounwind {
>> %cmp = icmp slt i64 %a, %b
>> @@ -27,7 +27,7 @@ define void @test_i64_slt(i32 addrspace(1)* %out, i64 %a, i64 %b) nounwind {
>> ret void
>> }
>>
>> -; SI-LABEL: @test_i64_ult:
>> +; SI-LABEL: test_i64_ult:
>> ; SI: V_CMP_LT_U64
>> define void @test_i64_ult(i32 addrspace(1)* %out, i64 %a, i64 %b) nounwind {
>> %cmp = icmp ult i64 %a, %b
>> @@ -36,7 +36,7 @@ define void @test_i64_ult(i32 addrspace(1)* %out, i64 %a, i64 %b) nounwind {
>> ret void
>> }
>>
>> -; SI-LABEL: @test_i64_sle:
>> +; SI-LABEL: test_i64_sle:
>> ; SI: V_CMP_LE_I64
>> define void @test_i64_sle(i32 addrspace(1)* %out, i64 %a, i64 %b) nounwind {
>> %cmp = icmp sle i64 %a, %b
>> @@ -45,7 +45,7 @@ define void @test_i64_sle(i32 addrspace(1)* %out, i64 %a, i64 %b) nounwind {
>> ret void
>> }
>>
>> -; SI-LABEL: @test_i64_ule:
>> +; SI-LABEL: test_i64_ule:
>> ; SI: V_CMP_LE_U64
>> define void @test_i64_ule(i32 addrspace(1)* %out, i64 %a, i64 %b) nounwind {
>> %cmp = icmp ule i64 %a, %b
>> @@ -54,7 +54,7 @@ define void @test_i64_ule(i32 addrspace(1)* %out, i64 %a, i64 %b) nounwind {
>> ret void
>> }
>>
>> -; SI-LABEL: @test_i64_sgt:
>> +; SI-LABEL: test_i64_sgt:
>> ; SI: V_CMP_GT_I64
>> define void @test_i64_sgt(i32 addrspace(1)* %out, i64 %a, i64 %b) nounwind {
>> %cmp = icmp sgt i64 %a, %b
>> @@ -63,7 +63,7 @@ define void @test_i64_sgt(i32 addrspace(1)* %out, i64 %a, i64 %b) nounwind {
>> ret void
>> }
>>
>> -; SI-LABEL: @test_i64_ugt:
>> +; SI-LABEL: test_i64_ugt:
>> ; SI: V_CMP_GT_U64
>> define void @test_i64_ugt(i32 addrspace(1)* %out, i64 %a, i64 %b) nounwind {
>> %cmp = icmp ugt i64 %a, %b
>> @@ -72,7 +72,7 @@ define void @test_i64_ugt(i32 addrspace(1)* %out, i64 %a, i64 %b) nounwind {
>> ret void
>> }
>>
>> -; SI-LABEL: @test_i64_sge:
>> +; SI-LABEL: test_i64_sge:
>> ; SI: V_CMP_GE_I64
>> define void @test_i64_sge(i32 addrspace(1)* %out, i64 %a, i64 %b) nounwind {
>> %cmp = icmp sge i64 %a, %b
>> @@ -81,7 +81,7 @@ define void @test_i64_sge(i32 addrspace(1)* %out, i64 %a, i64 %b) nounwind {
>> ret void
>> }
>>
>> -; SI-LABEL: @test_i64_uge:
>> +; SI-LABEL: test_i64_uge:
>> ; SI: V_CMP_GE_U64
>> define void @test_i64_uge(i32 addrspace(1)* %out, i64 %a, i64 %b) nounwind {
>> %cmp = icmp uge i64 %a, %b
>> diff --git a/test/CodeGen/R600/imm.ll b/test/CodeGen/R600/imm.ll
>> index e1d680c..0ae8150 100644
>> --- a/test/CodeGen/R600/imm.ll
>> +++ b/test/CodeGen/R600/imm.ll
>> @@ -1,7 +1,7 @@
>> ; RUN: llc -march=r600 -mcpu=verde -verify-machineinstrs < %s | FileCheck %s
>>
>> ; Use a 64-bit value with lo bits that can be represented as an inline constant
>> -; CHECK-LABEL: @i64_imm_inline_lo
>> +; CHECK-LABEL: i64_imm_inline_lo
>> ; CHECK: S_MOV_B32 [[LO:s[0-9]+]], 5
>> ; CHECK: V_MOV_B32_e32 v[[LO_VGPR:[0-9]+]], [[LO]]
>> ; CHECK: BUFFER_STORE_DWORDX2 v{{\[}}[[LO_VGPR]]:
>> @@ -12,7 +12,7 @@ entry:
>> }
>>
>> ; Use a 64-bit value with hi bits that can be represented as an inline constant
>> -; CHECK-LABEL: @i64_imm_inline_hi
>> +; CHECK-LABEL: i64_imm_inline_hi
>> ; CHECK: S_MOV_B32 [[HI:s[0-9]+]], 5
>> ; CHECK: V_MOV_B32_e32 v[[HI_VGPR:[0-9]+]], [[HI]]
>> ; CHECK: BUFFER_STORE_DWORDX2 v{{\[[0-9]+:}}[[HI_VGPR]]
>> @@ -22,7 +22,7 @@ entry:
>> ret void
>> }
>>
>> -; CHECK-LABEL: @store_inline_imm_0.0_f32
>> +; CHECK-LABEL: store_inline_imm_0.0_f32
>> ; CHECK: V_MOV_B32_e32 [[REG:v[0-9]+]], 0{{$}}
>> ; CHECK-NEXT: BUFFER_STORE_DWORD [[REG]]
>> define void @store_inline_imm_0.0_f32(float addrspace(1)* %out) {
>> @@ -30,7 +30,7 @@ define void @store_inline_imm_0.0_f32(float addrspace(1)* %out) {
>> ret void
>> }
>>
>> -; CHECK-LABEL: @store_inline_imm_0.5_f32
>> +; CHECK-LABEL: store_inline_imm_0.5_f32
>> ; CHECK: V_MOV_B32_e32 [[REG:v[0-9]+]], 0.5{{$}}
>> ; CHECK-NEXT: BUFFER_STORE_DWORD [[REG]]
>> define void @store_inline_imm_0.5_f32(float addrspace(1)* %out) {
>> @@ -38,7 +38,7 @@ define void @store_inline_imm_0.5_f32(float addrspace(1)* %out) {
>> ret void
>> }
>>
>> -; CHECK-LABEL: @store_inline_imm_m_0.5_f32
>> +; CHECK-LABEL: store_inline_imm_m_0.5_f32
>> ; CHECK: V_MOV_B32_e32 [[REG:v[0-9]+]], -0.5{{$}}
>> ; CHECK-NEXT: BUFFER_STORE_DWORD [[REG]]
>> define void @store_inline_imm_m_0.5_f32(float addrspace(1)* %out) {
>> @@ -46,7 +46,7 @@ define void @store_inline_imm_m_0.5_f32(float addrspace(1)* %out) {
>> ret void
>> }
>>
>> -; CHECK-LABEL: @store_inline_imm_1.0_f32
>> +; CHECK-LABEL: store_inline_imm_1.0_f32
>> ; CHECK: V_MOV_B32_e32 [[REG:v[0-9]+]], 1.0{{$}}
>> ; CHECK-NEXT: BUFFER_STORE_DWORD [[REG]]
>> define void @store_inline_imm_1.0_f32(float addrspace(1)* %out) {
>> @@ -54,7 +54,7 @@ define void @store_inline_imm_1.0_f32(float addrspace(1)* %out) {
>> ret void
>> }
>>
>> -; CHECK-LABEL: @store_inline_imm_m_1.0_f32
>> +; CHECK-LABEL: store_inline_imm_m_1.0_f32
>> ; CHECK: V_MOV_B32_e32 [[REG:v[0-9]+]], -1.0{{$}}
>> ; CHECK-NEXT: BUFFER_STORE_DWORD [[REG]]
>> define void @store_inline_imm_m_1.0_f32(float addrspace(1)* %out) {
>> @@ -62,7 +62,7 @@ define void @store_inline_imm_m_1.0_f32(float addrspace(1)* %out) {
>> ret void
>> }
>>
>> -; CHECK-LABEL: @store_inline_imm_2.0_f32
>> +; CHECK-LABEL: store_inline_imm_2.0_f32
>> ; CHECK: V_MOV_B32_e32 [[REG:v[0-9]+]], 2.0{{$}}
>> ; CHECK-NEXT: BUFFER_STORE_DWORD [[REG]]
>> define void @store_inline_imm_2.0_f32(float addrspace(1)* %out) {
>> @@ -70,7 +70,7 @@ define void @store_inline_imm_2.0_f32(float addrspace(1)* %out) {
>> ret void
>> }
>>
>> -; CHECK-LABEL: @store_inline_imm_m_2.0_f32
>> +; CHECK-LABEL: store_inline_imm_m_2.0_f32
>> ; CHECK: V_MOV_B32_e32 [[REG:v[0-9]+]], -2.0{{$}}
>> ; CHECK-NEXT: BUFFER_STORE_DWORD [[REG]]
>> define void @store_inline_imm_m_2.0_f32(float addrspace(1)* %out) {
>> @@ -78,7 +78,7 @@ define void @store_inline_imm_m_2.0_f32(float addrspace(1)* %out) {
>> ret void
>> }
>>
>> -; CHECK-LABEL: @store_inline_imm_4.0_f32
>> +; CHECK-LABEL: store_inline_imm_4.0_f32
>> ; CHECK: V_MOV_B32_e32 [[REG:v[0-9]+]], 4.0{{$}}
>> ; CHECK-NEXT: BUFFER_STORE_DWORD [[REG]]
>> define void @store_inline_imm_4.0_f32(float addrspace(1)* %out) {
>> @@ -86,7 +86,7 @@ define void @store_inline_imm_4.0_f32(float addrspace(1)* %out) {
>> ret void
>> }
>>
>> -; CHECK-LABEL: @store_inline_imm_m_4.0_f32
>> +; CHECK-LABEL: store_inline_imm_m_4.0_f32
>> ; CHECK: V_MOV_B32_e32 [[REG:v[0-9]+]], -4.0{{$}}
>> ; CHECK-NEXT: BUFFER_STORE_DWORD [[REG]]
>> define void @store_inline_imm_m_4.0_f32(float addrspace(1)* %out) {
>> @@ -94,7 +94,7 @@ define void @store_inline_imm_m_4.0_f32(float addrspace(1)* %out) {
>> ret void
>> }
>>
>> -; CHECK-LABEL: @store_literal_imm_f32
>> +; CHECK-LABEL: store_literal_imm_f32
>> ; CHECK: V_MOV_B32_e32 [[REG:v[0-9]+]], 0x45800000
>> ; CHECK-NEXT: BUFFER_STORE_DWORD [[REG]]
>> define void @store_literal_imm_f32(float addrspace(1)* %out) {
>> @@ -102,7 +102,7 @@ define void @store_literal_imm_f32(float addrspace(1)* %out) {
>> ret void
>> }
>>
>> -; CHECK-LABEL: @add_inline_imm_0.0_f32
>> +; CHECK-LABEL: add_inline_imm_0.0_f32
>> ; CHECK: S_LOAD_DWORD [[VAL:s[0-9]+]]
>> ; CHECK: V_ADD_F32_e64 [[REG:v[0-9]+]], [[VAL]], 0.0,
>> ; CHECK-NEXT: BUFFER_STORE_DWORD [[REG]]
>> @@ -112,7 +112,7 @@ define void @add_inline_imm_0.0_f32(float addrspace(1)* %out, float %x) {
>> ret void
>> }
>>
>> -; CHECK-LABEL: @add_inline_imm_0.5_f32
>> +; CHECK-LABEL: add_inline_imm_0.5_f32
>> ; CHECK: S_LOAD_DWORD [[VAL:s[0-9]+]]
>> ; CHECK: V_ADD_F32_e64 [[REG:v[0-9]+]], [[VAL]], 0.5,
>> ; CHECK-NEXT: BUFFER_STORE_DWORD [[REG]]
>> @@ -122,7 +122,7 @@ define void @add_inline_imm_0.5_f32(float addrspace(1)* %out, float %x) {
>> ret void
>> }
>>
>> -; CHECK-LABEL: @add_inline_imm_neg_0.5_f32
>> +; CHECK-LABEL: add_inline_imm_neg_0.5_f32
>> ; CHECK: S_LOAD_DWORD [[VAL:s[0-9]+]]
>> ; CHECK: V_ADD_F32_e64 [[REG:v[0-9]+]], [[VAL]], -0.5,
>> ; CHECK-NEXT: BUFFER_STORE_DWORD [[REG]]
>> @@ -132,7 +132,7 @@ define void @add_inline_imm_neg_0.5_f32(float addrspace(1)* %out, float %x) {
>> ret void
>> }
>>
>> -; CHECK-LABEL: @add_inline_imm_1.0_f32
>> +; CHECK-LABEL: add_inline_imm_1.0_f32
>> ; CHECK: S_LOAD_DWORD [[VAL:s[0-9]+]]
>> ; CHECK: V_ADD_F32_e64 [[REG:v[0-9]+]], [[VAL]], 1.0,
>> ; CHECK-NEXT: BUFFER_STORE_DWORD [[REG]]
>> @@ -142,7 +142,7 @@ define void @add_inline_imm_1.0_f32(float addrspace(1)* %out, float %x) {
>> ret void
>> }
>>
>> -; CHECK-LABEL: @add_inline_imm_neg_1.0_f32
>> +; CHECK-LABEL: add_inline_imm_neg_1.0_f32
>> ; CHECK: S_LOAD_DWORD [[VAL:s[0-9]+]]
>> ; CHECK: V_ADD_F32_e64 [[REG:v[0-9]+]], [[VAL]], -1.0,
>> ; CHECK-NEXT: BUFFER_STORE_DWORD [[REG]]
>> @@ -152,7 +152,7 @@ define void @add_inline_imm_neg_1.0_f32(float addrspace(1)* %out, float %x) {
>> ret void
>> }
>>
>> -; CHECK-LABEL: @add_inline_imm_2.0_f32
>> +; CHECK-LABEL: add_inline_imm_2.0_f32
>> ; CHECK: S_LOAD_DWORD [[VAL:s[0-9]+]]
>> ; CHECK: V_ADD_F32_e64 [[REG:v[0-9]+]], [[VAL]], 2.0,
>> ; CHECK-NEXT: BUFFER_STORE_DWORD [[REG]]
>> @@ -162,7 +162,7 @@ define void @add_inline_imm_2.0_f32(float addrspace(1)* %out, float %x) {
>> ret void
>> }
>>
>> -; CHECK-LABEL: @add_inline_imm_neg_2.0_f32
>> +; CHECK-LABEL: add_inline_imm_neg_2.0_f32
>> ; CHECK: S_LOAD_DWORD [[VAL:s[0-9]+]]
>> ; CHECK: V_ADD_F32_e64 [[REG:v[0-9]+]], [[VAL]], -2.0,
>> ; CHECK-NEXT: BUFFER_STORE_DWORD [[REG]]
>> @@ -172,7 +172,7 @@ define void @add_inline_imm_neg_2.0_f32(float addrspace(1)* %out, float %x) {
>> ret void
>> }
>>
>> -; CHECK-LABEL: @add_inline_imm_4.0_f32
>> +; CHECK-LABEL: add_inline_imm_4.0_f32
>> ; CHECK: S_LOAD_DWORD [[VAL:s[0-9]+]]
>> ; CHECK: V_ADD_F32_e64 [[REG:v[0-9]+]], [[VAL]], 4.0,
>> ; CHECK-NEXT: BUFFER_STORE_DWORD [[REG]]
>> @@ -182,7 +182,7 @@ define void @add_inline_imm_4.0_f32(float addrspace(1)* %out, float %x) {
>> ret void
>> }
>>
>> -; CHECK-LABEL: @add_inline_imm_neg_4.0_f32
>> +; CHECK-LABEL: add_inline_imm_neg_4.0_f32
>> ; CHECK: S_LOAD_DWORD [[VAL:s[0-9]+]]
>> ; CHECK: V_ADD_F32_e64 [[REG:v[0-9]+]], [[VAL]], -4.0,
>> ; CHECK-NEXT: BUFFER_STORE_DWORD [[REG]]
>> diff --git a/test/CodeGen/R600/indirect-private-64.ll b/test/CodeGen/R600/indirect-private-64.ll
>> index 5747434..92c115d 100644
>> --- a/test/CodeGen/R600/indirect-private-64.ll
>> +++ b/test/CodeGen/R600/indirect-private-64.ll
>> @@ -4,7 +4,7 @@
>>
>> declare void @llvm.AMDGPU.barrier.local() noduplicate nounwind
>>
>> -; SI-LABEL: @private_access_f64_alloca:
>> +; SI-LABEL: private_access_f64_alloca:
>>
>> ; SI-ALLOCA: BUFFER_STORE_DWORDX2
>> ; SI-ALLOCA: BUFFER_LOAD_DWORDX2
>> @@ -22,7 +22,7 @@ define void @private_access_f64_alloca(double addrspace(1)* noalias %out, double
>> ret void
>> }
>>
>> -; SI-LABEL: @private_access_v2f64_alloca:
>> +; SI-LABEL: private_access_v2f64_alloca:
>>
>> ; SI-ALLOCA: BUFFER_STORE_DWORDX4
>> ; SI-ALLOCA: BUFFER_LOAD_DWORDX4
>> @@ -46,7 +46,7 @@ define void @private_access_v2f64_alloca(<2 x double> addrspace(1)* noalias %out
>> ret void
>> }
>>
>> -; SI-LABEL: @private_access_i64_alloca:
>> +; SI-LABEL: private_access_i64_alloca:
>>
>> ; SI-ALLOCA: BUFFER_STORE_DWORDX2
>> ; SI-ALLOCA: BUFFER_LOAD_DWORDX2
>> @@ -64,7 +64,7 @@ define void @private_access_i64_alloca(i64 addrspace(1)* noalias %out, i64 addrs
>> ret void
>> }
>>
>> -; SI-LABEL: @private_access_v2i64_alloca:
>> +; SI-LABEL: private_access_v2i64_alloca:
>>
>> ; SI-ALLOCA: BUFFER_STORE_DWORDX4
>> ; SI-ALLOCA: BUFFER_LOAD_DWORDX4
>> diff --git a/test/CodeGen/R600/infinite-loop.ll b/test/CodeGen/R600/infinite-loop.ll
>> index 68ffaae..851eaa4 100644
>> --- a/test/CodeGen/R600/infinite-loop.ll
>> +++ b/test/CodeGen/R600/infinite-loop.ll
>> @@ -1,6 +1,6 @@
>> ; RUN: llc -march=r600 -mcpu=SI -verify-machineinstrs < %s | FileCheck -check-prefix=SI %s
>>
>> -; SI-LABEL: @infinite_loop:
>> +; SI-LABEL: infinite_loop:
>> ; SI: V_MOV_B32_e32 [[REG:v[0-9]+]], 0x3e7
>> ; SI: BB0_1:
>> ; SI: BUFFER_STORE_DWORD [[REG]]
>> diff --git a/test/CodeGen/R600/input-mods.ll b/test/CodeGen/R600/input-mods.ll
>> index 13bfbab..3aabe8b 100644
>> --- a/test/CodeGen/R600/input-mods.ll
>> +++ b/test/CodeGen/R600/input-mods.ll
>> @@ -1,9 +1,9 @@
>> ;RUN: llc < %s -march=r600 -mcpu=redwood | FileCheck %s --check-prefix=EG-CHECK
>> ;RUN: llc < %s -march=r600 -mcpu=cayman | FileCheck %s --check-prefix=CM-CHECK
>>
>> -;EG-CHECK-LABEL: @test
>> +;EG-CHECK-LABEL: test
>> ;EG-CHECK: EXP_IEEE *
>> -;CM-CHECK-LABEL: @test
>> +;CM-CHECK-LABEL: test
>> ;CM-CHECK: EXP_IEEE T{{[0-9]+}}.X, -|T{{[0-9]+}}.X|
>> ;CM-CHECK: EXP_IEEE T{{[0-9]+}}.Y (MASKED), -|T{{[0-9]+}}.X|
>> ;CM-CHECK: EXP_IEEE T{{[0-9]+}}.Z (MASKED), -|T{{[0-9]+}}.X|
>> diff --git a/test/CodeGen/R600/insert_vector_elt.ll b/test/CodeGen/R600/insert_vector_elt.ll
>> index 18769ed..ee80a56 100644
>> --- a/test/CodeGen/R600/insert_vector_elt.ll
>> +++ b/test/CodeGen/R600/insert_vector_elt.ll
>> @@ -8,7 +8,7 @@
>> ; FIXME: Why is the constant moved into the intermediate register and
>> ; not just directly into the vector component?
>>
>> -; SI-LABEL: @insertelement_v4f32_0:
>> +; SI-LABEL: insertelement_v4f32_0:
>> ; S_LOAD_DWORDX4 s{{[}}[[LOW_REG:[0-9]+]]:
>> ; V_MOV_B32_e32
>> ; V_MOV_B32_e32 [[CONSTREG:v[0-9]+]], 5.000000e+00
>> @@ -20,35 +20,35 @@ define void @insertelement_v4f32_0(<4 x float> addrspace(1)* %out, <4 x float> %
>> ret void
>> }
>>
>> -; SI-LABEL: @insertelement_v4f32_1:
>> +; SI-LABEL: insertelement_v4f32_1:
>> define void @insertelement_v4f32_1(<4 x float> addrspace(1)* %out, <4 x float> %a) nounwind {
>> %vecins = insertelement <4 x float> %a, float 5.000000e+00, i32 1
>> store <4 x float> %vecins, <4 x float> addrspace(1)* %out, align 16
>> ret void
>> }
>>
>> -; SI-LABEL: @insertelement_v4f32_2:
>> +; SI-LABEL: insertelement_v4f32_2:
>> define void @insertelement_v4f32_2(<4 x float> addrspace(1)* %out, <4 x float> %a) nounwind {
>> %vecins = insertelement <4 x float> %a, float 5.000000e+00, i32 2
>> store <4 x float> %vecins, <4 x float> addrspace(1)* %out, align 16
>> ret void
>> }
>>
>> -; SI-LABEL: @insertelement_v4f32_3:
>> +; SI-LABEL: insertelement_v4f32_3:
>> define void @insertelement_v4f32_3(<4 x float> addrspace(1)* %out, <4 x float> %a) nounwind {
>> %vecins = insertelement <4 x float> %a, float 5.000000e+00, i32 3
>> store <4 x float> %vecins, <4 x float> addrspace(1)* %out, align 16
>> ret void
>> }
>>
>> -; SI-LABEL: @insertelement_v4i32_0:
>> +; SI-LABEL: insertelement_v4i32_0:
>> define void @insertelement_v4i32_0(<4 x i32> addrspace(1)* %out, <4 x i32> %a) nounwind {
>> %vecins = insertelement <4 x i32> %a, i32 999, i32 0
>> store <4 x i32> %vecins, <4 x i32> addrspace(1)* %out, align 16
>> ret void
>> }
>>
>> -; SI-LABEL: @dynamic_insertelement_v2f32:
>> +; SI-LABEL: dynamic_insertelement_v2f32:
>> ; SI: V_MOV_B32_e32 [[CONST:v[0-9]+]], 0x40a00000
>> ; SI: V_MOVRELD_B32_e32 v[[LOW_RESULT_REG:[0-9]+]], [[CONST]]
>> ; SI: BUFFER_STORE_DWORDX2 {{v\[}}[[LOW_RESULT_REG]]:
>> @@ -58,7 +58,7 @@ define void @dynamic_insertelement_v2f32(<2 x float> addrspace(1)* %out, <2 x fl
>> ret void
>> }
>>
>> -; SI-LABEL: @dynamic_insertelement_v4f32:
>> +; SI-LABEL: dynamic_insertelement_v4f32:
>> ; SI: V_MOV_B32_e32 [[CONST:v[0-9]+]], 0x40a00000
>> ; SI: V_MOVRELD_B32_e32 v[[LOW_RESULT_REG:[0-9]+]], [[CONST]]
>> ; SI: BUFFER_STORE_DWORDX4 {{v\[}}[[LOW_RESULT_REG]]:
>> @@ -68,7 +68,7 @@ define void @dynamic_insertelement_v4f32(<4 x float> addrspace(1)* %out, <4 x fl
>> ret void
>> }
>>
>> -; SI-LABEL: @dynamic_insertelement_v8f32:
>> +; SI-LABEL: dynamic_insertelement_v8f32:
>> ; FIXMESI: BUFFER_STORE_DWORDX4
>> ; FIXMESI: BUFFER_STORE_DWORDX4
>> define void @dynamic_insertelement_v8f32(<8 x float> addrspace(1)* %out, <8 x float> %a, i32 %b) nounwind {
>> @@ -77,7 +77,7 @@ define void @dynamic_insertelement_v8f32(<8 x float> addrspace(1)* %out, <8 x fl
>> ret void
>> }
>>
>> -; SI-LABEL: @dynamic_insertelement_v16f32:
>> +; SI-LABEL: dynamic_insertelement_v16f32:
>> ; FIXMESI: BUFFER_STORE_DWORDX4
>> ; FIXMESI: BUFFER_STORE_DWORDX4
>> ; FIXMESI: BUFFER_STORE_DWORDX4
>> @@ -88,7 +88,7 @@ define void @dynamic_insertelement_v16f32(<16 x float> addrspace(1)* %out, <16 x
>> ret void
>> }
>>
>> -; SI-LABEL: @dynamic_insertelement_v2i32:
>> +; SI-LABEL: dynamic_insertelement_v2i32:
>> ; SI: BUFFER_STORE_DWORDX2
>> define void @dynamic_insertelement_v2i32(<2 x i32> addrspace(1)* %out, <2 x i32> %a, i32 %b) nounwind {
>> %vecins = insertelement <2 x i32> %a, i32 5, i32 %b
>> @@ -96,7 +96,7 @@ define void @dynamic_insertelement_v2i32(<2 x i32> addrspace(1)* %out, <2 x i32>
>> ret void
>> }
>>
>> -; SI-LABEL: @dynamic_insertelement_v4i32:
>> +; SI-LABEL: dynamic_insertelement_v4i32:
>> ; SI: BUFFER_STORE_DWORDX4
>> define void @dynamic_insertelement_v4i32(<4 x i32> addrspace(1)* %out, <4 x i32> %a, i32 %b) nounwind {
>> %vecins = insertelement <4 x i32> %a, i32 5, i32 %b
>> @@ -104,7 +104,7 @@ define void @dynamic_insertelement_v4i32(<4 x i32> addrspace(1)* %out, <4 x i32>
>> ret void
>> }
>>
>> -; SI-LABEL: @dynamic_insertelement_v8i32:
>> +; SI-LABEL: dynamic_insertelement_v8i32:
>> ; FIXMESI: BUFFER_STORE_DWORDX4
>> ; FIXMESI: BUFFER_STORE_DWORDX4
>> define void @dynamic_insertelement_v8i32(<8 x i32> addrspace(1)* %out, <8 x i32> %a, i32 %b) nounwind {
>> @@ -113,7 +113,7 @@ define void @dynamic_insertelement_v8i32(<8 x i32> addrspace(1)* %out, <8 x i32>
>> ret void
>> }
>>
>> -; SI-LABEL: @dynamic_insertelement_v16i32:
>> +; SI-LABEL: dynamic_insertelement_v16i32:
>> ; FIXMESI: BUFFER_STORE_DWORDX4
>> ; FIXMESI: BUFFER_STORE_DWORDX4
>> ; FIXMESI: BUFFER_STORE_DWORDX4
>> @@ -125,7 +125,7 @@ define void @dynamic_insertelement_v16i32(<16 x i32> addrspace(1)* %out, <16 x i
>> }
>>
>>
>> -; SI-LABEL: @dynamic_insertelement_v2i16:
>> +; SI-LABEL: dynamic_insertelement_v2i16:
>> ; FIXMESI: BUFFER_STORE_DWORDX2
>> define void @dynamic_insertelement_v2i16(<2 x i16> addrspace(1)* %out, <2 x i16> %a, i32 %b) nounwind {
>> %vecins = insertelement <2 x i16> %a, i16 5, i32 %b
>> @@ -133,7 +133,7 @@ define void @dynamic_insertelement_v2i16(<2 x i16> addrspace(1)* %out, <2 x i16>
>> ret void
>> }
>>
>> -; SI-LABEL: @dynamic_insertelement_v4i16:
>> +; SI-LABEL: dynamic_insertelement_v4i16:
>> ; FIXMESI: BUFFER_STORE_DWORDX4
>> define void @dynamic_insertelement_v4i16(<4 x i16> addrspace(1)* %out, <4 x i16> %a, i32 %b) nounwind {
>> %vecins = insertelement <4 x i16> %a, i16 5, i32 %b
>> @@ -142,7 +142,7 @@ define void @dynamic_insertelement_v4i16(<4 x i16> addrspace(1)* %out, <4 x i16>
>> }
>>
>>
>> -; SI-LABEL: @dynamic_insertelement_v2i8:
>> +; SI-LABEL: dynamic_insertelement_v2i8:
>> ; FIXMESI: BUFFER_STORE_USHORT
>> define void @dynamic_insertelement_v2i8(<2 x i8> addrspace(1)* %out, <2 x i8> %a, i32 %b) nounwind {
>> %vecins = insertelement <2 x i8> %a, i8 5, i32 %b
>> @@ -150,7 +150,7 @@ define void @dynamic_insertelement_v2i8(<2 x i8> addrspace(1)* %out, <2 x i8> %a
>> ret void
>> }
>>
>> -; SI-LABEL: @dynamic_insertelement_v4i8:
>> +; SI-LABEL: dynamic_insertelement_v4i8:
>> ; FIXMESI: BUFFER_STORE_DWORD
>> define void @dynamic_insertelement_v4i8(<4 x i8> addrspace(1)* %out, <4 x i8> %a, i32 %b) nounwind {
>> %vecins = insertelement <4 x i8> %a, i8 5, i32 %b
>> @@ -158,7 +158,7 @@ define void @dynamic_insertelement_v4i8(<4 x i8> addrspace(1)* %out, <4 x i8> %a
>> ret void
>> }
>>
>> -; SI-LABEL: @dynamic_insertelement_v8i8:
>> +; SI-LABEL: dynamic_insertelement_v8i8:
>> ; FIXMESI: BUFFER_STORE_DWORDX2
>> define void @dynamic_insertelement_v8i8(<8 x i8> addrspace(1)* %out, <8 x i8> %a, i32 %b) nounwind {
>> %vecins = insertelement <8 x i8> %a, i8 5, i32 %b
>> @@ -166,7 +166,7 @@ define void @dynamic_insertelement_v8i8(<8 x i8> addrspace(1)* %out, <8 x i8> %a
>> ret void
>> }
>>
>> -; SI-LABEL: @dynamic_insertelement_v16i8:
>> +; SI-LABEL: dynamic_insertelement_v16i8:
>> ; FIXMESI: BUFFER_STORE_DWORDX4
>> define void @dynamic_insertelement_v16i8(<16 x i8> addrspace(1)* %out, <16 x i8> %a, i32 %b) nounwind {
>> %vecins = insertelement <16 x i8> %a, i8 5, i32 %b
>> @@ -176,7 +176,7 @@ define void @dynamic_insertelement_v16i8(<16 x i8> addrspace(1)* %out, <16 x i8>
>>
>> ; This test requires handling INSERT_SUBREG in SIFixSGPRCopies. Check that
>> ; the compiler doesn't crash.
>> -; SI-LABEL: @insert_split_bb
>> +; SI-LABEL: insert_split_bb
>> define void @insert_split_bb(<2 x i32> addrspace(1)* %out, i32 addrspace(1)* %in, i32 %a, i32 %b) {
>> entry:
>> %0 = insertelement <2 x i32> undef, i32 %a, i32 0
>> @@ -200,7 +200,7 @@ endif:
>> ret void
>> }
>>
>> -; SI-LABEL: @dynamic_insertelement_v2f64:
>> +; SI-LABEL: dynamic_insertelement_v2f64:
>> ; SI: BUFFER_STORE_DWORDX2
>> ; SI: BUFFER_STORE_DWORDX2
>> ; SI: BUFFER_STORE_DWORDX2
>> @@ -212,7 +212,7 @@ define void @dynamic_insertelement_v2f64(<2 x double> addrspace(1)* %out, <2 x d
>> ret void
>> }
>>
>> -; SI-LABEL: @dynamic_insertelement_v2i64:
>> +; SI-LABEL: dynamic_insertelement_v2i64:
>> ; SI: BUFFER_STORE_DWORDX2
>> ; SI: BUFFER_STORE_DWORDX2
>> ; SI: S_ENDPGM
>> @@ -222,7 +222,7 @@ define void @dynamic_insertelement_v2i64(<2 x i64> addrspace(1)* %out, <2 x i64>
>> ret void
>> }
>>
>> -; SI-LABEL: @dynamic_insertelement_v4f64:
>> +; SI-LABEL: dynamic_insertelement_v4f64:
>> ; SI: BUFFER_STORE_DWORDX2
>> ; SI: BUFFER_STORE_DWORDX2
>> ; SI: BUFFER_STORE_DWORDX2
>> @@ -234,7 +234,7 @@ define void @dynamic_insertelement_v4f64(<4 x double> addrspace(1)* %out, <4 x d
>> ret void
>> }
>>
>> -; SI-LABEL: @dynamic_insertelement_v8f64:
>> +; SI-LABEL: dynamic_insertelement_v8f64:
>> ; SI: BUFFER_STORE_DWORDX2
>> ; SI: BUFFER_STORE_DWORDX2
>> ; SI: BUFFER_STORE_DWORDX2
>> diff --git a/test/CodeGen/R600/kcache-fold.ll b/test/CodeGen/R600/kcache-fold.ll
>> index 0baa3cd..20568ae 100644
>> --- a/test/CodeGen/R600/kcache-fold.ll
>> +++ b/test/CodeGen/R600/kcache-fold.ll
>> @@ -1,6 +1,6 @@
>> ;RUN: llc < %s -march=r600 -mcpu=redwood | FileCheck %s
>>
>> -; CHECK: @main1
>> +; CHECK: main1
>> ; CHECK: MOV * T{{[0-9]+\.[XYZW], KC0}}
>> define void @main1() {
>> main_body:
>> @@ -48,7 +48,7 @@ main_body:
>> ret void
>> }
>>
>> -; CHECK: @main2
>> +; CHECK: main2
>> ; CHECK-NOT: MOV
>> define void @main2() {
>> main_body:
>> diff --git a/test/CodeGen/R600/kernel-args.ll b/test/CodeGen/R600/kernel-args.ll
>> index d0750de..48e8809 100644
>> --- a/test/CodeGen/R600/kernel-args.ll
>> +++ b/test/CodeGen/R600/kernel-args.ll
>> @@ -2,9 +2,9 @@
>> ; RUN: llc < %s -march=r600 -mcpu=cayman | FileCheck %s --check-prefix=EG-CHECK
>> ; RUN: llc < %s -march=r600 -mcpu=SI -verify-machineinstrs | FileCheck %s --check-prefix=SI-CHECK
>>
>> -; EG-CHECK-LABEL: @i8_arg
>> +; EG-CHECK-LABEL: i8_arg
>> ; EG-CHECK: MOV {{[ *]*}}T{{[0-9]+\.[XYZW]}}, KC0[2].Z
>> -; SI-CHECK-LABEL: @i8_arg
>> +; SI-CHECK-LABEL: i8_arg
>> ; SI-CHECK: BUFFER_LOAD_UBYTE
>>
>> define void @i8_arg(i32 addrspace(1)* nocapture %out, i8 %in) nounwind {
>> @@ -14,9 +14,9 @@ entry:
>> ret void
>> }
>>
>> -; EG-CHECK-LABEL: @i8_zext_arg
>> +; EG-CHECK-LABEL: i8_zext_arg
>> ; EG-CHECK: MOV {{[ *]*}}T{{[0-9]+\.[XYZW]}}, KC0[2].Z
>> -; SI-CHECK-LABEL: @i8_zext_arg
>> +; SI-CHECK-LABEL: i8_zext_arg
>> ; SI-CHECK: S_LOAD_DWORD s{{[0-9]}}, s[0:1], 0xb
>>
>> define void @i8_zext_arg(i32 addrspace(1)* nocapture %out, i8 zeroext %in) nounwind {
>> @@ -26,9 +26,9 @@ entry:
>> ret void
>> }
>>
>> -; EG-CHECK-LABEL: @i8_sext_arg
>> +; EG-CHECK-LABEL: i8_sext_arg
>> ; EG-CHECK: MOV {{[ *]*}}T{{[0-9]+\.[XYZW]}}, KC0[2].Z
>> -; SI-CHECK-LABEL: @i8_sext_arg
>> +; SI-CHECK-LABEL: i8_sext_arg
>> ; SI-CHECK: S_LOAD_DWORD s{{[0-9]}}, s[0:1], 0xb
>>
>> define void @i8_sext_arg(i32 addrspace(1)* nocapture %out, i8 signext %in) nounwind {
>> @@ -38,9 +38,9 @@ entry:
>> ret void
>> }
>>
>> -; EG-CHECK-LABEL: @i16_arg
>> +; EG-CHECK-LABEL: i16_arg
>> ; EG-CHECK: MOV {{[ *]*}}T{{[0-9]+\.[XYZW]}}, KC0[2].Z
>> -; SI-CHECK-LABEL: @i16_arg
>> +; SI-CHECK-LABEL: i16_arg
>> ; SI-CHECK: BUFFER_LOAD_USHORT
>>
>> define void @i16_arg(i32 addrspace(1)* nocapture %out, i16 %in) nounwind {
>> @@ -50,9 +50,9 @@ entry:
>> ret void
>> }
>>
>> -; EG-CHECK-LABEL: @i16_zext_arg
>> +; EG-CHECK-LABEL: i16_zext_arg
>> ; EG-CHECK: MOV {{[ *]*}}T{{[0-9]+\.[XYZW]}}, KC0[2].Z
>> -; SI-CHECK-LABEL: @i16_zext_arg
>> +; SI-CHECK-LABEL: i16_zext_arg
>> ; SI-CHECK: S_LOAD_DWORD s{{[0-9]}}, s[0:1], 0xb
>>
>> define void @i16_zext_arg(i32 addrspace(1)* nocapture %out, i16 zeroext %in) nounwind {
>> @@ -62,9 +62,9 @@ entry:
>> ret void
>> }
>>
>> -; EG-CHECK-LABEL: @i16_sext_arg
>> +; EG-CHECK-LABEL: i16_sext_arg
>> ; EG-CHECK: MOV {{[ *]*}}T{{[0-9]+\.[XYZW]}}, KC0[2].Z
>> -; SI-CHECK-LABEL: @i16_sext_arg
>> +; SI-CHECK-LABEL: i16_sext_arg
>> ; SI-CHECK: S_LOAD_DWORD s{{[0-9]}}, s[0:1], 0xb
>>
>> define void @i16_sext_arg(i32 addrspace(1)* nocapture %out, i16 signext %in) nounwind {
>> @@ -74,9 +74,9 @@ entry:
>> ret void
>> }
>>
>> -; EG-CHECK-LABEL: @i32_arg
>> +; EG-CHECK-LABEL: i32_arg
>> ; EG-CHECK: T{{[0-9]\.[XYZW]}}, KC0[2].Z
>> -; SI-CHECK-LABEL: @i32_arg
>> +; SI-CHECK-LABEL: i32_arg
>> ; S_LOAD_DWORD s{{[0-9]}}, s[0:1], 0xb
>> define void @i32_arg(i32 addrspace(1)* nocapture %out, i32 %in) nounwind {
>> entry:
>> @@ -84,9 +84,9 @@ entry:
>> ret void
>> }
>>
>> -; EG-CHECK-LABEL: @f32_arg
>> +; EG-CHECK-LABEL: f32_arg
>> ; EG-CHECK: T{{[0-9]\.[XYZW]}}, KC0[2].Z
>> -; SI-CHECK-LABEL: @f32_arg
>> +; SI-CHECK-LABEL: f32_arg
>> ; S_LOAD_DWORD s{{[0-9]}}, s[0:1], 0xb
>> define void @f32_arg(float addrspace(1)* nocapture %out, float %in) nounwind {
>> entry:
>> @@ -94,10 +94,10 @@ entry:
>> ret void
>> }
>>
>> -; EG-CHECK-LABEL: @v2i8_arg
>> +; EG-CHECK-LABEL: v2i8_arg
>> ; EG-CHECK: VTX_READ_8
>> ; EG-CHECK: VTX_READ_8
>> -; SI-CHECK-LABEL: @v2i8_arg
>> +; SI-CHECK-LABEL: v2i8_arg
>> ; SI-CHECK: BUFFER_LOAD_UBYTE
>> ; SI-CHECK: BUFFER_LOAD_UBYTE
>> define void @v2i8_arg(<2 x i8> addrspace(1)* %out, <2 x i8> %in) {
>> @@ -106,10 +106,10 @@ entry:
>> ret void
>> }
>>
>> -; EG-CHECK-LABEL: @v2i16_arg
>> +; EG-CHECK-LABEL: v2i16_arg
>> ; EG-CHECK: VTX_READ_16
>> ; EG-CHECK: VTX_READ_16
>> -; SI-CHECK-LABEL: @v2i16_arg
>> +; SI-CHECK-LABEL: v2i16_arg
>> ; SI-CHECK-DAG: BUFFER_LOAD_USHORT
>> ; SI-CHECK-DAG: BUFFER_LOAD_USHORT
>> define void @v2i16_arg(<2 x i16> addrspace(1)* %out, <2 x i16> %in) {
>> @@ -118,10 +118,10 @@ entry:
>> ret void
>> }
>>
>> -; EG-CHECK-LABEL: @v2i32_arg
>> +; EG-CHECK-LABEL: v2i32_arg
>> ; EG-CHECK-DAG: T{{[0-9]\.[XYZW]}}, KC0[3].X
>> ; EG-CHECK-DAG: T{{[0-9]\.[XYZW]}}, KC0[2].W
>> -; SI-CHECK-LABEL: @v2i32_arg
>> +; SI-CHECK-LABEL: v2i32_arg
>> ; SI-CHECK: S_LOAD_DWORDX2 s{{\[[0-9]:[0-9]\]}}, s[0:1], 0xb
>> define void @v2i32_arg(<2 x i32> addrspace(1)* nocapture %out, <2 x i32> %in) nounwind {
>> entry:
>> @@ -129,10 +129,10 @@ entry:
>> ret void
>> }
>>
>> -; EG-CHECK-LABEL: @v2f32_arg
>> +; EG-CHECK-LABEL: v2f32_arg
>> ; EG-CHECK-DAG: T{{[0-9]\.[XYZW]}}, KC0[3].X
>> ; EG-CHECK-DAG: T{{[0-9]\.[XYZW]}}, KC0[2].W
>> -; SI-CHECK-LABEL: @v2f32_arg
>> +; SI-CHECK-LABEL: v2f32_arg
>> ; SI-CHECK: S_LOAD_DWORDX2 s{{\[[0-9]:[0-9]\]}}, s[0:1], 0xb
>> define void @v2f32_arg(<2 x float> addrspace(1)* nocapture %out, <2 x float> %in) nounwind {
>> entry:
>> @@ -140,32 +140,32 @@ entry:
>> ret void
>> }
>>
>> -; EG-CHECK-LABEL: @v3i8_arg
>> +; EG-CHECK-LABEL: v3i8_arg
>> ; VTX_READ_8 T{{[0-9]}}.X, T{{[0-9]}}.X, 40
>> ; VTX_READ_8 T{{[0-9]}}.X, T{{[0-9]}}.X, 41
>> ; VTX_READ_8 T{{[0-9]}}.X, T{{[0-9]}}.X, 42
>> -; SI-CHECK-LABEL: @v3i8_arg
>> +; SI-CHECK-LABEL: v3i8_arg
>> define void @v3i8_arg(<3 x i8> addrspace(1)* nocapture %out, <3 x i8> %in) nounwind {
>> entry:
>> store <3 x i8> %in, <3 x i8> addrspace(1)* %out, align 4
>> ret void
>> }
>>
>> -; EG-CHECK-LABEL: @v3i16_arg
>> +; EG-CHECK-LABEL: v3i16_arg
>> ; VTX_READ_16 T{{[0-9]}}.X, T{{[0-9]}}.X, 44
>> ; VTX_READ_16 T{{[0-9]}}.X, T{{[0-9]}}.X, 46
>> ; VTX_READ_16 T{{[0-9]}}.X, T{{[0-9]}}.X, 48
>> -; SI-CHECK-LABEL: @v3i16_arg
>> +; SI-CHECK-LABEL: v3i16_arg
>> define void @v3i16_arg(<3 x i16> addrspace(1)* nocapture %out, <3 x i16> %in) nounwind {
>> entry:
>> store <3 x i16> %in, <3 x i16> addrspace(1)* %out, align 4
>> ret void
>> }
>> -; EG-CHECK-LABEL: @v3i32_arg
>> +; EG-CHECK-LABEL: v3i32_arg
>> ; EG-CHECK-DAG: T{{[0-9]\.[XYZW]}}, KC0[3].Y
>> ; EG-CHECK-DAG: T{{[0-9]\.[XYZW]}}, KC0[3].Z
>> ; EG-CHECK-DAG: T{{[0-9]\.[XYZW]}}, KC0[3].W
>> -; SI-CHECK-LABEL: @v3i32_arg
>> +; SI-CHECK-LABEL: v3i32_arg
>> ; SI-CHECK: S_LOAD_DWORDX4 s{{\[[0-9]:[0-9]+\]}}, s[0:1], 0xd
>> define void @v3i32_arg(<3 x i32> addrspace(1)* nocapture %out, <3 x i32> %in) nounwind {
>> entry:
>> @@ -173,11 +173,11 @@ entry:
>> ret void
>> }
>>
>> -; EG-CHECK-LABEL: @v3f32_arg
>> +; EG-CHECK-LABEL: v3f32_arg
>> ; EG-CHECK-DAG: T{{[0-9]\.[XYZW]}}, KC0[3].Y
>> ; EG-CHECK-DAG: T{{[0-9]\.[XYZW]}}, KC0[3].Z
>> ; EG-CHECK-DAG: T{{[0-9]\.[XYZW]}}, KC0[3].W
>> -; SI-CHECK-LABEL: @v3f32_arg
>> +; SI-CHECK-LABEL: v3f32_arg
>> ; SI-CHECK: S_LOAD_DWORDX4 s{{\[[0-9]:[0-9]+\]}}, s[0:1], 0xd
>> define void @v3f32_arg(<3 x float> addrspace(1)* nocapture %out, <3 x float> %in) nounwind {
>> entry:
>> @@ -185,12 +185,12 @@ entry:
>> ret void
>> }
>>
>> -; EG-CHECK-LABEL: @v4i8_arg
>> +; EG-CHECK-LABEL: v4i8_arg
>> ; EG-CHECK: VTX_READ_8
>> ; EG-CHECK: VTX_READ_8
>> ; EG-CHECK: VTX_READ_8
>> ; EG-CHECK: VTX_READ_8
>> -; SI-CHECK-LABEL: @v4i8_arg
>> +; SI-CHECK-LABEL: v4i8_arg
>> ; SI-CHECK: BUFFER_LOAD_UBYTE
>> ; SI-CHECK: BUFFER_LOAD_UBYTE
>> ; SI-CHECK: BUFFER_LOAD_UBYTE
>> @@ -201,12 +201,12 @@ entry:
>> ret void
>> }
>>
>> -; EG-CHECK-LABEL: @v4i16_arg
>> +; EG-CHECK-LABEL: v4i16_arg
>> ; EG-CHECK: VTX_READ_16
>> ; EG-CHECK: VTX_READ_16
>> ; EG-CHECK: VTX_READ_16
>> ; EG-CHECK: VTX_READ_16
>> -; SI-CHECK-LABEL: @v4i16_arg
>> +; SI-CHECK-LABEL: v4i16_arg
>> ; SI-CHECK: BUFFER_LOAD_USHORT
>> ; SI-CHECK: BUFFER_LOAD_USHORT
>> ; SI-CHECK: BUFFER_LOAD_USHORT
>> @@ -217,12 +217,12 @@ entry:
>> ret void
>> }
>>
>> -; EG-CHECK-LABEL: @v4i32_arg
>> +; EG-CHECK-LABEL: v4i32_arg
>> ; EG-CHECK-DAG: T{{[0-9]\.[XYZW]}}, KC0[3].Y
>> ; EG-CHECK-DAG: T{{[0-9]\.[XYZW]}}, KC0[3].Z
>> ; EG-CHECK-DAG: T{{[0-9]\.[XYZW]}}, KC0[3].W
>> ; EG-CHECK-DAG: T{{[0-9]\.[XYZW]}}, KC0[4].X
>> -; SI-CHECK-LABEL: @v4i32_arg
>> +; SI-CHECK-LABEL: v4i32_arg
>> ; SI-CHECK: S_LOAD_DWORDX4 s{{\[[0-9]:[0-9]\]}}, s[0:1], 0xd
>> define void @v4i32_arg(<4 x i32> addrspace(1)* nocapture %out, <4 x i32> %in) nounwind {
>> entry:
>> @@ -230,12 +230,12 @@ entry:
>> ret void
>> }
>>
>> -; EG-CHECK-LABEL: @v4f32_arg
>> +; EG-CHECK-LABEL: v4f32_arg
>> ; EG-CHECK-DAG: T{{[0-9]\.[XYZW]}}, KC0[3].Y
>> ; EG-CHECK-DAG: T{{[0-9]\.[XYZW]}}, KC0[3].Z
>> ; EG-CHECK-DAG: T{{[0-9]\.[XYZW]}}, KC0[3].W
>> ; EG-CHECK-DAG: T{{[0-9]\.[XYZW]}}, KC0[4].X
>> -; SI-CHECK-LABEL: @v4f32_arg
>> +; SI-CHECK-LABEL: v4f32_arg
>> ; SI-CHECK: S_LOAD_DWORDX4 s{{\[[0-9]:[0-9]\]}}, s[0:1], 0xd
>> define void @v4f32_arg(<4 x float> addrspace(1)* nocapture %out, <4 x float> %in) nounwind {
>> entry:
>> @@ -243,7 +243,7 @@ entry:
>> ret void
>> }
>>
>> -; EG-CHECK-LABEL: @v8i8_arg
>> +; EG-CHECK-LABEL: v8i8_arg
>> ; EG-CHECK: VTX_READ_8
>> ; EG-CHECK: VTX_READ_8
>> ; EG-CHECK: VTX_READ_8
>> @@ -252,7 +252,7 @@ entry:
>> ; EG-CHECK: VTX_READ_8
>> ; EG-CHECK: VTX_READ_8
>> ; EG-CHECK: VTX_READ_8
>> -; SI-CHECK-LABEL: @v8i8_arg
>> +; SI-CHECK-LABEL: v8i8_arg
>> ; SI-CHECK: BUFFER_LOAD_UBYTE
>> ; SI-CHECK: BUFFER_LOAD_UBYTE
>> ; SI-CHECK: BUFFER_LOAD_UBYTE
>> @@ -266,7 +266,7 @@ entry:
>> ret void
>> }
>>
>> -; EG-CHECK-LABEL: @v8i16_arg
>> +; EG-CHECK-LABEL: v8i16_arg
>> ; EG-CHECK: VTX_READ_16
>> ; EG-CHECK: VTX_READ_16
>> ; EG-CHECK: VTX_READ_16
>> @@ -275,7 +275,7 @@ entry:
>> ; EG-CHECK: VTX_READ_16
>> ; EG-CHECK: VTX_READ_16
>> ; EG-CHECK: VTX_READ_16
>> -; SI-CHECK-LABEL: @v8i16_arg
>> +; SI-CHECK-LABEL: v8i16_arg
>> ; SI-CHECK: BUFFER_LOAD_USHORT
>> ; SI-CHECK: BUFFER_LOAD_USHORT
>> ; SI-CHECK: BUFFER_LOAD_USHORT
>> @@ -290,7 +290,7 @@ entry:
>> ret void
>> }
>>
>> -; EG-CHECK-LABEL: @v8i32_arg
>> +; EG-CHECK-LABEL: v8i32_arg
>> ; EG-CHECK-DAG: T{{[0-9]\.[XYZW]}}, KC0[4].Y
>> ; EG-CHECK-DAG: T{{[0-9]\.[XYZW]}}, KC0[4].Z
>> ; EG-CHECK-DAG: T{{[0-9]\.[XYZW]}}, KC0[4].W
>> @@ -299,7 +299,7 @@ entry:
>> ; EG-CHECK-DAG: T{{[0-9]\.[XYZW]}}, KC0[5].Z
>> ; EG-CHECK-DAG: T{{[0-9]\.[XYZW]}}, KC0[5].W
>> ; EG-CHECK-DAG: T{{[0-9]\.[XYZW]}}, KC0[6].X
>> -; SI-CHECK-LABEL: @v8i32_arg
>> +; SI-CHECK-LABEL: v8i32_arg
>> ; SI-CHECK: S_LOAD_DWORDX8 s{{\[[0-9]:[0-9]+\]}}, s[0:1], 0x11
>> define void @v8i32_arg(<8 x i32> addrspace(1)* nocapture %out, <8 x i32> %in) nounwind {
>> entry:
>> @@ -307,7 +307,7 @@ entry:
>> ret void
>> }
>>
>> -; EG-CHECK-LABEL: @v8f32_arg
>> +; EG-CHECK-LABEL: v8f32_arg
>> ; EG-CHECK-DAG: T{{[0-9]\.[XYZW]}}, KC0[4].Y
>> ; EG-CHECK-DAG: T{{[0-9]\.[XYZW]}}, KC0[4].Z
>> ; EG-CHECK-DAG: T{{[0-9]\.[XYZW]}}, KC0[4].W
>> @@ -316,7 +316,7 @@ entry:
>> ; EG-CHECK-DAG: T{{[0-9]\.[XYZW]}}, KC0[5].Z
>> ; EG-CHECK-DAG: T{{[0-9]\.[XYZW]}}, KC0[5].W
>> ; EG-CHECK-DAG: T{{[0-9]\.[XYZW]}}, KC0[6].X
>> -; SI-CHECK-LABEL: @v8f32_arg
>> +; SI-CHECK-LABEL: v8f32_arg
>> ; SI-CHECK: S_LOAD_DWORDX8 s{{\[[0-9]:[0-9]+\]}}, s[0:1], 0x11
>> define void @v8f32_arg(<8 x float> addrspace(1)* nocapture %out, <8 x float> %in) nounwind {
>> entry:
>> @@ -324,7 +324,7 @@ entry:
>> ret void
>> }
>>
>> -; EG-CHECK-LABEL: @v16i8_arg
>> +; EG-CHECK-LABEL: v16i8_arg
>> ; EG-CHECK: VTX_READ_8
>> ; EG-CHECK: VTX_READ_8
>> ; EG-CHECK: VTX_READ_8
>> @@ -341,7 +341,7 @@ entry:
>> ; EG-CHECK: VTX_READ_8
>> ; EG-CHECK: VTX_READ_8
>> ; EG-CHECK: VTX_READ_8
>> -; SI-CHECK-LABEL: @v16i8_arg
>> +; SI-CHECK-LABEL: v16i8_arg
>> ; SI-CHECK: BUFFER_LOAD_UBYTE
>> ; SI-CHECK: BUFFER_LOAD_UBYTE
>> ; SI-CHECK: BUFFER_LOAD_UBYTE
>> @@ -364,7 +364,7 @@ entry:
>> ret void
>> }
>>
>> -; EG-CHECK-LABEL: @v16i16_arg
>> +; EG-CHECK-LABEL: v16i16_arg
>> ; EG-CHECK: VTX_READ_16
>> ; EG-CHECK: VTX_READ_16
>> ; EG-CHECK: VTX_READ_16
>> @@ -381,7 +381,7 @@ entry:
>> ; EG-CHECK: VTX_READ_16
>> ; EG-CHECK: VTX_READ_16
>> ; EG-CHECK: VTX_READ_16
>> -; SI-CHECK-LABEL: @v16i16_arg
>> +; SI-CHECK-LABEL: v16i16_arg
>> ; SI-CHECK: BUFFER_LOAD_USHORT
>> ; SI-CHECK: BUFFER_LOAD_USHORT
>> ; SI-CHECK: BUFFER_LOAD_USHORT
>> @@ -404,7 +404,7 @@ entry:
>> ret void
>> }
>>
>> -; EG-CHECK-LABEL: @v16i32_arg
>> +; EG-CHECK-LABEL: v16i32_arg
>> ; EG-CHECK-DAG: T{{[0-9]\.[XYZW]}}, KC0[6].Y
>> ; EG-CHECK-DAG: T{{[0-9]\.[XYZW]}}, KC0[6].Z
>> ; EG-CHECK-DAG: T{{[0-9]\.[XYZW]}}, KC0[6].W
>> @@ -421,7 +421,7 @@ entry:
>> ; EG-CHECK-DAG: T{{[0-9]\.[XYZW]}}, KC0[9].Z
>> ; EG-CHECK-DAG: T{{[0-9]\.[XYZW]}}, KC0[9].W
>> ; EG-CHECK-DAG: T{{[0-9]\.[XYZW]}}, KC0[10].X
>> -; SI-CHECK-LABEL: @v16i32_arg
>> +; SI-CHECK-LABEL: v16i32_arg
>> ; SI-CHECK: S_LOAD_DWORDX16 s{{\[[0-9]:[0-9]+\]}}, s[0:1], 0x19
>> define void @v16i32_arg(<16 x i32> addrspace(1)* nocapture %out, <16 x i32> %in) nounwind {
>> entry:
>> @@ -429,7 +429,7 @@ entry:
>> ret void
>> }
>>
>> -; EG-CHECK-LABEL: @v16f32_arg
>> +; EG-CHECK-LABEL: v16f32_arg
>> ; EG-CHECK-DAG: T{{[0-9]\.[XYZW]}}, KC0[6].Y
>> ; EG-CHECK-DAG: T{{[0-9]\.[XYZW]}}, KC0[6].Z
>> ; EG-CHECK-DAG: T{{[0-9]\.[XYZW]}}, KC0[6].W
>> @@ -446,7 +446,7 @@ entry:
>> ; EG-CHECK-DAG: T{{[0-9]\.[XYZW]}}, KC0[9].Z
>> ; EG-CHECK-DAG: T{{[0-9]\.[XYZW]}}, KC0[9].W
>> ; EG-CHECK-DAG: T{{[0-9]\.[XYZW]}}, KC0[10].X
>> -; SI-CHECK-LABEL: @v16f32_arg
>> +; SI-CHECK-LABEL: v16f32_arg
>> ; SI-CHECK: S_LOAD_DWORDX16 s{{\[[0-9]:[0-9]+\]}}, s[0:1], 0x19
>> define void @v16f32_arg(<16 x float> addrspace(1)* nocapture %out, <16 x float> %in) nounwind {
>> entry:
>> @@ -454,7 +454,7 @@ entry:
>> ret void
>> }
>>
>> -; FUNC-LABEL: @kernel_arg_i64
>> +; FUNC-LABEL: kernel_arg_i64
>> ; SI: S_LOAD_DWORDX2
>> ; SI: S_LOAD_DWORDX2
>> ; SI: BUFFER_STORE_DWORDX2
>> @@ -463,7 +463,7 @@ define void @kernel_arg_i64(i64 addrspace(1)* %out, i64 %a) nounwind {
>> ret void
>> }
>>
>> -; XFUNC-LABEL: @kernel_arg_v1i64
>> +; XFUNC-LABEL: kernel_arg_v1i64
>> ; XSI: S_LOAD_DWORDX2
>> ; XSI: S_LOAD_DWORDX2
>> ; XSI: BUFFER_STORE_DWORDX2
>> diff --git a/test/CodeGen/R600/lds-oqap-crash.ll b/test/CodeGen/R600/lds-oqap-crash.ll
>> index 7959150..932ba6c 100644
>> --- a/test/CodeGen/R600/lds-oqap-crash.ll
>> +++ b/test/CodeGen/R600/lds-oqap-crash.ll
>> @@ -9,7 +9,7 @@
>> ; because the LDS instructions are pseudo instructions and the OQAP
>> ; reads and writes are bundled together in the same instruction.
>>
>> -; CHECK: @lds_crash
>> +; CHECK: lds_crash
>> define void @lds_crash(i32 addrspace(1)* %out, i32 addrspace(3)* %in, i32 %a, i32 %b, i32 %c) {
>> entry:
>> %0 = load i32 addrspace(3)* %in
>> diff --git a/test/CodeGen/R600/lds-output-queue.ll b/test/CodeGen/R600/lds-output-queue.ll
>> index d5dc061..39de160 100644
>> --- a/test/CodeGen/R600/lds-output-queue.ll
>> +++ b/test/CodeGen/R600/lds-output-queue.ll
>> @@ -3,7 +3,7 @@
>> ; This test checks that the lds input queue will is empty at the end of
>> ; the ALU clause.
>>
>> -; CHECK-LABEL: @lds_input_queue
>> +; CHECK-LABEL: lds_input_queue
>> ; CHECK: LDS_READ_RET * OQAP
>> ; CHECK-NOT: ALU clause
>> ; CHECK: MOV * T{{[0-9]\.[XYZW]}}, OQAP
>> @@ -84,7 +84,7 @@ declare void @llvm.AMDGPU.barrier.local()
>> ; analysis, we should be able to keep these instructions sparate before
>> ; scheduling.
>> ;
>> -; CHECK-LABEL: @local_global_alias
>> +; CHECK-LABEL: local_global_alias
>> ; CHECK: LDS_READ_RET
>> ; CHECK-NOT: ALU clause
>> ; CHECK: MOV * T{{[0-9]\.[XYZW]}}, OQAP
>> diff --git a/test/CodeGen/R600/lds-size.ll b/test/CodeGen/R600/lds-size.ll
>> index 9182e25..0cda80b 100644
>> --- a/test/CodeGen/R600/lds-size.ll
>> +++ b/test/CodeGen/R600/lds-size.ll
>> @@ -3,7 +3,7 @@
>> ; This test makes sure we do not double count global values when they are
>> ; used in different basic blocks.
>>
>> -; CHECK-LABEL: @test
>> +; CHECK-LABEL: test
>> ; CHECK: .long 166120
>> ; CHECK-NEXT: .long 1
>> @lds = internal unnamed_addr addrspace(3) global i32 zeroinitializer, align 4
>> diff --git a/test/CodeGen/R600/legalizedag-bug-expand-setcc.ll b/test/CodeGen/R600/legalizedag-bug-expand-setcc.ll
>> index 1aae7f9..6100e86 100644
>> --- a/test/CodeGen/R600/legalizedag-bug-expand-setcc.ll
>> +++ b/test/CodeGen/R600/legalizedag-bug-expand-setcc.ll
>> @@ -8,7 +8,7 @@
>> ; instructions, when only one is needed.
>> ;
>>
>> -; CHECK: @setcc_expand
>> +; CHECK: setcc_expand
>> ; CHECK: SET
>> ; CHECK-NOT: CND
>> define void @setcc_expand(i32 addrspace(1)* %out, i32 %in) {
>> diff --git a/test/CodeGen/R600/literals.ll b/test/CodeGen/R600/literals.ll
>> index 47191e0..771f3b5 100644
>> --- a/test/CodeGen/R600/literals.ll
>> +++ b/test/CodeGen/R600/literals.ll
>> @@ -6,7 +6,7 @@
>> ; or
>> ; ADD_INT literal.x KC0[2].Z, 5
>>
>> -; CHECK: @i32_literal
>> +; CHECK: i32_literal
>> ; CHECK: ADD_INT {{\** *}}T{{[0-9]\.[XYZW]}}, KC0[2].Z, literal.x
>> ; CHECK-NEXT: LSHR
>> ; CHECK-NEXT: 5
>> @@ -23,7 +23,7 @@ entry:
>> ; or
>> ; ADD literal.x KC0[2].Z, 5.0
>>
>> -; CHECK: @float_literal
>> +; CHECK: float_literal
>> ; CHECK: ADD {{\** *}}T{{[0-9]\.[XYZW]}}, KC0[2].Z, literal.x
>> ; CHECK-NEXT: LSHR
>> ; CHECK-NEXT: 1084227584(5.0
>> @@ -35,7 +35,7 @@ entry:
>> }
>>
>> ; Make sure inline literals are folded into REG_SEQUENCE instructions.
>> -; CHECK: @inline_literal_reg_sequence
>> +; CHECK: inline_literal_reg_sequence
>> ; CHECK: MOV {{\** *}}T[[GPR:[0-9]]].X, 0.0
>> ; CHECK-NEXT: MOV {{\** *}}T[[GPR]].Y, 0.0
>> ; CHECK-NEXT: MOV {{\** *}}T[[GPR]].Z, 0.0
>> @@ -47,7 +47,7 @@ entry:
>> ret void
>> }
>>
>> -; CHECK: @inline_literal_dot4
>> +; CHECK: inline_literal_dot4
>> ; CHECK: DOT4 T[[GPR:[0-9]]].X, 1.0
>> ; CHECK-NEXT: DOT4 T[[GPR]].Y (MASKED), 1.0
>> ; CHECK-NEXT: DOT4 T[[GPR]].Z (MASKED), 1.0
>> diff --git a/test/CodeGen/R600/llvm.AMDGPU.abs.ll b/test/CodeGen/R600/llvm.AMDGPU.abs.ll
>> index a0a47b7..6557a1c 100644
>> --- a/test/CodeGen/R600/llvm.AMDGPU.abs.ll
>> +++ b/test/CodeGen/R600/llvm.AMDGPU.abs.ll
>> @@ -6,7 +6,7 @@ declare i32 @llvm.AMDGPU.abs(i32) nounwind readnone
>> ; Legacy name
>> declare i32 @llvm.AMDIL.abs.i32(i32) nounwind readnone
>>
>> -; FUNC-LABEL: @s_abs_i32
>> +; FUNC-LABEL: s_abs_i32
>> ; SI: S_SUB_I32
>> ; SI: S_MAX_I32
>> ; SI: S_ENDPGM
>> @@ -19,7 +19,7 @@ define void @s_abs_i32(i32 addrspace(1)* %out, i32 %src) nounwind {
>> ret void
>> }
>>
>> -; FUNC-LABEL: @v_abs_i32
>> +; FUNC-LABEL: v_abs_i32
>> ; SI: V_SUB_I32_e32
>> ; SI: V_MAX_I32_e32
>> ; SI: S_ENDPGM
>> @@ -33,7 +33,7 @@ define void @v_abs_i32(i32 addrspace(1)* %out, i32 addrspace(1)* %src) nounwind
>> ret void
>> }
>>
>> -; FUNC-LABEL: @abs_i32_legacy_amdil
>> +; FUNC-LABEL: abs_i32_legacy_amdil
>> ; SI: V_SUB_I32_e32
>> ; SI: V_MAX_I32_e32
>> ; SI: S_ENDPGM
>> diff --git a/test/CodeGen/R600/llvm.AMDGPU.barrier.global.ll b/test/CodeGen/R600/llvm.AMDGPU.barrier.global.ll
>> index 47f5255..b65e680 100644
>> --- a/test/CodeGen/R600/llvm.AMDGPU.barrier.global.ll
>> +++ b/test/CodeGen/R600/llvm.AMDGPU.barrier.global.ll
>> @@ -1,7 +1,7 @@
>> ; RUN: llc -march=r600 -mcpu=SI < %s | FileCheck -check-prefix=SI -check-prefix=FUNC %s
>> ; RUN: llc -march=r600 -mcpu=redwood < %s | FileCheck -check-prefix=EG -check-prefix=FUNC %s
>>
>> -; FUNC-LABEL: @test_barrier_global
>> +; FUNC-LABEL: test_barrier_global
>> ; EG: GROUP_BARRIER
>> ; SI: S_BARRIER
>>
>> diff --git a/test/CodeGen/R600/llvm.AMDGPU.barrier.local.ll b/test/CodeGen/R600/llvm.AMDGPU.barrier.local.ll
>> index 7203675..5e5f92c 100644
>> --- a/test/CodeGen/R600/llvm.AMDGPU.barrier.local.ll
>> +++ b/test/CodeGen/R600/llvm.AMDGPU.barrier.local.ll
>> @@ -1,7 +1,7 @@
>> ; RUN: llc -march=r600 -mcpu=SI < %s | FileCheck -check-prefix=SI -check-prefix=FUNC %s
>> ; RUN: llc -march=r600 -mcpu=redwood < %s | FileCheck -check-prefix=EG -check-prefix=FUNC %s
>>
>> -; FUNC-LABEL: @test_barrier_local
>> +; FUNC-LABEL: test_barrier_local
>> ; EG: GROUP_BARRIER
>> ; SI: S_BARRIER
>>
>> diff --git a/test/CodeGen/R600/llvm.AMDGPU.bfe.i32.ll b/test/CodeGen/R600/llvm.AMDGPU.bfe.i32.ll
>> index eb50942..a141e6d 100644
>> --- a/test/CodeGen/R600/llvm.AMDGPU.bfe.i32.ll
>> +++ b/test/CodeGen/R600/llvm.AMDGPU.bfe.i32.ll
>> @@ -3,7 +3,7 @@
>>
>> declare i32 @llvm.AMDGPU.bfe.i32(i32, i32, i32) nounwind readnone
>>
>> -; FUNC-LABEL: @bfe_i32_arg_arg_arg
>> +; FUNC-LABEL: bfe_i32_arg_arg_arg
>> ; SI: V_BFE_I32
>> ; EG: BFE_INT
>> ; EG: encoding: [{{[x0-9a-f]+,[x0-9a-f]+,[x0-9a-f]+,[x0-9a-f]+,[x0-9a-f]+}},0xac
>> @@ -13,7 +13,7 @@ define void @bfe_i32_arg_arg_arg(i32 addrspace(1)* %out, i32 %src0, i32 %src1, i
>> ret void
>> }
>>
>> -; FUNC-LABEL: @bfe_i32_arg_arg_imm
>> +; FUNC-LABEL: bfe_i32_arg_arg_imm
>> ; SI: V_BFE_I32
>> ; EG: BFE_INT
>> define void @bfe_i32_arg_arg_imm(i32 addrspace(1)* %out, i32 %src0, i32 %src1) nounwind {
>> @@ -22,7 +22,7 @@ define void @bfe_i32_arg_arg_imm(i32 addrspace(1)* %out, i32 %src0, i32 %src1) n
>> ret void
>> }
>>
>> -; FUNC-LABEL: @bfe_i32_arg_imm_arg
>> +; FUNC-LABEL: bfe_i32_arg_imm_arg
>> ; SI: V_BFE_I32
>> ; EG: BFE_INT
>> define void @bfe_i32_arg_imm_arg(i32 addrspace(1)* %out, i32 %src0, i32 %src2) nounwind {
>> @@ -31,7 +31,7 @@ define void @bfe_i32_arg_imm_arg(i32 addrspace(1)* %out, i32 %src0, i32 %src2) n
>> ret void
>> }
>>
>> -; FUNC-LABEL: @bfe_i32_imm_arg_arg
>> +; FUNC-LABEL: bfe_i32_imm_arg_arg
>> ; SI: V_BFE_I32
>> ; EG: BFE_INT
>> define void @bfe_i32_imm_arg_arg(i32 addrspace(1)* %out, i32 %src1, i32 %src2) nounwind {
>> @@ -40,7 +40,7 @@ define void @bfe_i32_imm_arg_arg(i32 addrspace(1)* %out, i32 %src1, i32 %src2) n
>> ret void
>> }
>>
>> -; FUNC-LABEL: @v_bfe_print_arg
>> +; FUNC-LABEL: v_bfe_print_arg
>> ; SI: V_BFE_I32 v{{[0-9]+}}, v{{[0-9]+}}, 2, 8
>> define void @v_bfe_print_arg(i32 addrspace(1)* %out, i32 addrspace(1)* %src0) nounwind {
>> %load = load i32 addrspace(1)* %src0, align 4
>> @@ -49,7 +49,7 @@ define void @v_bfe_print_arg(i32 addrspace(1)* %out, i32 addrspace(1)* %src0) no
>> ret void
>> }
>>
>> -; FUNC-LABEL: @bfe_i32_arg_0_width_reg_offset
>> +; FUNC-LABEL: bfe_i32_arg_0_width_reg_offset
>> ; SI-NOT: BFE
>> ; SI: S_ENDPGM
>> ; EG-NOT: BFE
>> @@ -59,7 +59,7 @@ define void @bfe_i32_arg_0_width_reg_offset(i32 addrspace(1)* %out, i32 %src0, i
>> ret void
>> }
>>
>> -; FUNC-LABEL: @bfe_i32_arg_0_width_imm_offset
>> +; FUNC-LABEL: bfe_i32_arg_0_width_imm_offset
>> ; SI-NOT: BFE
>> ; SI: S_ENDPGM
>> ; EG-NOT: BFE
>> @@ -69,7 +69,7 @@ define void @bfe_i32_arg_0_width_imm_offset(i32 addrspace(1)* %out, i32 %src0, i
>> ret void
>> }
>>
>> -; FUNC-LABEL: @bfe_i32_test_6
>> +; FUNC-LABEL: bfe_i32_test_6
>> ; SI: V_LSHLREV_B32_e32 v{{[0-9]+}}, 31, v{{[0-9]+}}
>> ; SI: V_ASHRREV_I32_e32 v{{[0-9]+}}, 1, v{{[0-9]+}}
>> ; SI: S_ENDPGM
>> @@ -81,7 +81,7 @@ define void @bfe_i32_test_6(i32 addrspace(1)* %out, i32 addrspace(1)* %in) nounw
>> ret void
>> }
>>
>> -; FUNC-LABEL: @bfe_i32_test_7
>> +; FUNC-LABEL: bfe_i32_test_7
>> ; SI-NOT: SHL
>> ; SI-NOT: BFE
>> ; SI: V_MOV_B32_e32 [[VREG:v[0-9]+]], 0
>> @@ -96,7 +96,7 @@ define void @bfe_i32_test_7(i32 addrspace(1)* %out, i32 addrspace(1)* %in) nounw
>> }
>>
>> ; FIXME: The shifts should be 1 BFE
>> -; FUNC-LABEL: @bfe_i32_test_8
>> +; FUNC-LABEL: bfe_i32_test_8
>> ; SI: BUFFER_LOAD_DWORD
>> ; SI: V_BFE_I32 v{{[0-9]+}}, v{{[0-9]+}}, 0, 1
>> ; SI: S_ENDPGM
>> @@ -108,7 +108,7 @@ define void @bfe_i32_test_8(i32 addrspace(1)* %out, i32 addrspace(1)* %in) nounw
>> ret void
>> }
>>
>> -; FUNC-LABEL: @bfe_i32_test_9
>> +; FUNC-LABEL: bfe_i32_test_9
>> ; SI-NOT: BFE
>> ; SI: V_ASHRREV_I32_e32 v{{[0-9]+}}, 31, v{{[0-9]+}}
>> ; SI-NOT: BFE
>> @@ -120,7 +120,7 @@ define void @bfe_i32_test_9(i32 addrspace(1)* %out, i32 addrspace(1)* %in) nounw
>> ret void
>> }
>>
>> -; FUNC-LABEL: @bfe_i32_test_10
>> +; FUNC-LABEL: bfe_i32_test_10
>> ; SI-NOT: BFE
>> ; SI: V_ASHRREV_I32_e32 v{{[0-9]+}}, 1, v{{[0-9]+}}
>> ; SI-NOT: BFE
>> @@ -132,7 +132,7 @@ define void @bfe_i32_test_10(i32 addrspace(1)* %out, i32 addrspace(1)* %in) noun
>> ret void
>> }
>>
>> -; FUNC-LABEL: @bfe_i32_test_11
>> +; FUNC-LABEL: bfe_i32_test_11
>> ; SI-NOT: BFE
>> ; SI: V_ASHRREV_I32_e32 v{{[0-9]+}}, 8, v{{[0-9]+}}
>> ; SI-NOT: BFE
>> @@ -144,7 +144,7 @@ define void @bfe_i32_test_11(i32 addrspace(1)* %out, i32 addrspace(1)* %in) noun
>> ret void
>> }
>>
>> -; FUNC-LABEL: @bfe_i32_test_12
>> +; FUNC-LABEL: bfe_i32_test_12
>> ; SI-NOT: BFE
>> ; SI: V_ASHRREV_I32_e32 v{{[0-9]+}}, 24, v{{[0-9]+}}
>> ; SI-NOT: BFE
>> @@ -156,7 +156,7 @@ define void @bfe_i32_test_12(i32 addrspace(1)* %out, i32 addrspace(1)* %in) noun
>> ret void
>> }
>>
>> -; FUNC-LABEL: @bfe_i32_test_13
>> +; FUNC-LABEL: bfe_i32_test_13
>> ; SI: V_ASHRREV_I32_e32 {{v[0-9]+}}, 31, {{v[0-9]+}}
>> ; SI-NOT: BFE
>> ; SI: S_ENDPGM
>> @@ -167,7 +167,7 @@ define void @bfe_i32_test_13(i32 addrspace(1)* %out, i32 addrspace(1)* %in) noun
>> store i32 %bfe, i32 addrspace(1)* %out, align 4 ret void
>> }
>>
>> -; FUNC-LABEL: @bfe_i32_test_14
>> +; FUNC-LABEL: bfe_i32_test_14
>> ; SI-NOT: LSHR
>> ; SI-NOT: BFE
>> ; SI: S_ENDPGM
>> @@ -178,7 +178,7 @@ define void @bfe_i32_test_14(i32 addrspace(1)* %out, i32 addrspace(1)* %in) noun
>> store i32 %bfe, i32 addrspace(1)* %out, align 4 ret void
>> }
>>
>> -; FUNC-LABEL: @bfe_i32_constant_fold_test_0
>> +; FUNC-LABEL: bfe_i32_constant_fold_test_0
>> ; SI-NOT: BFE
>> ; SI: V_MOV_B32_e32 [[VREG:v[0-9]+]], 0
>> ; SI: BUFFER_STORE_DWORD [[VREG]],
>> @@ -190,7 +190,7 @@ define void @bfe_i32_constant_fold_test_0(i32 addrspace(1)* %out) nounwind {
>> ret void
>> }
>>
>> -; FUNC-LABEL: @bfe_i32_constant_fold_test_1
>> +; FUNC-LABEL: bfe_i32_constant_fold_test_1
>> ; SI-NOT: BFE
>> ; SI: V_MOV_B32_e32 [[VREG:v[0-9]+]], 0
>> ; SI: BUFFER_STORE_DWORD [[VREG]],
>> @@ -202,7 +202,7 @@ define void @bfe_i32_constant_fold_test_1(i32 addrspace(1)* %out) nounwind {
>> ret void
>> }
>>
>> -; FUNC-LABEL: @bfe_i32_constant_fold_test_2
>> +; FUNC-LABEL: bfe_i32_constant_fold_test_2
>> ; SI-NOT: BFE
>> ; SI: V_MOV_B32_e32 [[VREG:v[0-9]+]], 0
>> ; SI: BUFFER_STORE_DWORD [[VREG]],
>> @@ -214,7 +214,7 @@ define void @bfe_i32_constant_fold_test_2(i32 addrspace(1)* %out) nounwind {
>> ret void
>> }
>>
>> -; FUNC-LABEL: @bfe_i32_constant_fold_test_3
>> +; FUNC-LABEL: bfe_i32_constant_fold_test_3
>> ; SI-NOT: BFE
>> ; SI: V_MOV_B32_e32 [[VREG:v[0-9]+]], -1
>> ; SI: BUFFER_STORE_DWORD [[VREG]],
>> @@ -226,7 +226,7 @@ define void @bfe_i32_constant_fold_test_3(i32 addrspace(1)* %out) nounwind {
>> ret void
>> }
>>
>> -; FUNC-LABEL: @bfe_i32_constant_fold_test_4
>> +; FUNC-LABEL: bfe_i32_constant_fold_test_4
>> ; SI-NOT: BFE
>> ; SI: V_MOV_B32_e32 [[VREG:v[0-9]+]], -1
>> ; SI: BUFFER_STORE_DWORD [[VREG]],
>> @@ -238,7 +238,7 @@ define void @bfe_i32_constant_fold_test_4(i32 addrspace(1)* %out) nounwind {
>> ret void
>> }
>>
>> -; FUNC-LABEL: @bfe_i32_constant_fold_test_5
>> +; FUNC-LABEL: bfe_i32_constant_fold_test_5
>> ; SI-NOT: BFE
>> ; SI: V_MOV_B32_e32 [[VREG:v[0-9]+]], -1
>> ; SI: BUFFER_STORE_DWORD [[VREG]],
>> @@ -250,7 +250,7 @@ define void @bfe_i32_constant_fold_test_5(i32 addrspace(1)* %out) nounwind {
>> ret void
>> }
>>
>> -; FUNC-LABEL: @bfe_i32_constant_fold_test_6
>> +; FUNC-LABEL: bfe_i32_constant_fold_test_6
>> ; SI-NOT: BFE
>> ; SI: V_MOV_B32_e32 [[VREG:v[0-9]+]], 0xffffff80
>> ; SI: BUFFER_STORE_DWORD [[VREG]],
>> @@ -262,7 +262,7 @@ define void @bfe_i32_constant_fold_test_6(i32 addrspace(1)* %out) nounwind {
>> ret void
>> }
>>
>> -; FUNC-LABEL: @bfe_i32_constant_fold_test_7
>> +; FUNC-LABEL: bfe_i32_constant_fold_test_7
>> ; SI-NOT: BFE
>> ; SI: V_MOV_B32_e32 [[VREG:v[0-9]+]], 0x7f
>> ; SI: BUFFER_STORE_DWORD [[VREG]],
>> @@ -274,7 +274,7 @@ define void @bfe_i32_constant_fold_test_7(i32 addrspace(1)* %out) nounwind {
>> ret void
>> }
>>
>> -; FUNC-LABEL: @bfe_i32_constant_fold_test_8
>> +; FUNC-LABEL: bfe_i32_constant_fold_test_8
>> ; SI-NOT: BFE
>> ; SI: V_MOV_B32_e32 [[VREG:v[0-9]+]], 1
>> ; SI: BUFFER_STORE_DWORD [[VREG]],
>> @@ -286,7 +286,7 @@ define void @bfe_i32_constant_fold_test_8(i32 addrspace(1)* %out) nounwind {
>> ret void
>> }
>>
>> -; FUNC-LABEL: @bfe_i32_constant_fold_test_9
>> +; FUNC-LABEL: bfe_i32_constant_fold_test_9
>> ; SI-NOT: BFE
>> ; SI: V_MOV_B32_e32 [[VREG:v[0-9]+]], 1
>> ; SI: BUFFER_STORE_DWORD [[VREG]],
>> @@ -298,7 +298,7 @@ define void @bfe_i32_constant_fold_test_9(i32 addrspace(1)* %out) nounwind {
>> ret void
>> }
>>
>> -; FUNC-LABEL: @bfe_i32_constant_fold_test_10
>> +; FUNC-LABEL: bfe_i32_constant_fold_test_10
>> ; SI-NOT: BFE
>> ; SI: V_MOV_B32_e32 [[VREG:v[0-9]+]], 0
>> ; SI: BUFFER_STORE_DWORD [[VREG]],
>> @@ -310,7 +310,7 @@ define void @bfe_i32_constant_fold_test_10(i32 addrspace(1)* %out) nounwind {
>> ret void
>> }
>>
>> -; FUNC-LABEL: @bfe_i32_constant_fold_test_11
>> +; FUNC-LABEL: bfe_i32_constant_fold_test_11
>> ; SI-NOT: BFE
>> ; SI: V_MOV_B32_e32 [[VREG:v[0-9]+]], -6
>> ; SI: BUFFER_STORE_DWORD [[VREG]],
>> @@ -322,7 +322,7 @@ define void @bfe_i32_constant_fold_test_11(i32 addrspace(1)* %out) nounwind {
>> ret void
>> }
>>
>> -; FUNC-LABEL: @bfe_i32_constant_fold_test_12
>> +; FUNC-LABEL: bfe_i32_constant_fold_test_12
>> ; SI-NOT: BFE
>> ; SI: V_MOV_B32_e32 [[VREG:v[0-9]+]], 0
>> ; SI: BUFFER_STORE_DWORD [[VREG]],
>> @@ -334,7 +334,7 @@ define void @bfe_i32_constant_fold_test_12(i32 addrspace(1)* %out) nounwind {
>> ret void
>> }
>>
>> -; FUNC-LABEL: @bfe_i32_constant_fold_test_13
>> +; FUNC-LABEL: bfe_i32_constant_fold_test_13
>> ; SI-NOT: BFE
>> ; SI: V_MOV_B32_e32 [[VREG:v[0-9]+]], 1
>> ; SI: BUFFER_STORE_DWORD [[VREG]],
>> @@ -346,7 +346,7 @@ define void @bfe_i32_constant_fold_test_13(i32 addrspace(1)* %out) nounwind {
>> ret void
>> }
>>
>> -; FUNC-LABEL: @bfe_i32_constant_fold_test_14
>> +; FUNC-LABEL: bfe_i32_constant_fold_test_14
>> ; SI-NOT: BFE
>> ; SI: V_MOV_B32_e32 [[VREG:v[0-9]+]], 40
>> ; SI: BUFFER_STORE_DWORD [[VREG]],
>> @@ -358,7 +358,7 @@ define void @bfe_i32_constant_fold_test_14(i32 addrspace(1)* %out) nounwind {
>> ret void
>> }
>>
>> -; FUNC-LABEL: @bfe_i32_constant_fold_test_15
>> +; FUNC-LABEL: bfe_i32_constant_fold_test_15
>> ; SI-NOT: BFE
>> ; SI: V_MOV_B32_e32 [[VREG:v[0-9]+]], 10
>> ; SI: BUFFER_STORE_DWORD [[VREG]],
>> @@ -370,7 +370,7 @@ define void @bfe_i32_constant_fold_test_15(i32 addrspace(1)* %out) nounwind {
>> ret void
>> }
>>
>> -; FUNC-LABEL: @bfe_i32_constant_fold_test_16
>> +; FUNC-LABEL: bfe_i32_constant_fold_test_16
>> ; SI-NOT: BFE
>> ; SI: V_MOV_B32_e32 [[VREG:v[0-9]+]], -1
>> ; SI: BUFFER_STORE_DWORD [[VREG]],
>> @@ -382,7 +382,7 @@ define void @bfe_i32_constant_fold_test_16(i32 addrspace(1)* %out) nounwind {
>> ret void
>> }
>>
>> -; FUNC-LABEL: @bfe_i32_constant_fold_test_17
>> +; FUNC-LABEL: bfe_i32_constant_fold_test_17
>> ; SI-NOT: BFE
>> ; SI: V_MOV_B32_e32 [[VREG:v[0-9]+]], 0x7f
>> ; SI: BUFFER_STORE_DWORD [[VREG]],
>> @@ -394,7 +394,7 @@ define void @bfe_i32_constant_fold_test_17(i32 addrspace(1)* %out) nounwind {
>> ret void
>> }
>>
>> -; FUNC-LABEL: @bfe_i32_constant_fold_test_18
>> +; FUNC-LABEL: bfe_i32_constant_fold_test_18
>> ; SI-NOT: BFE
>> ; SI: V_MOV_B32_e32 [[VREG:v[0-9]+]], 0
>> ; SI: BUFFER_STORE_DWORD [[VREG]],
>> @@ -408,7 +408,7 @@ define void @bfe_i32_constant_fold_test_18(i32 addrspace(1)* %out) nounwind {
>>
>> ; XXX - This should really be a single BFE, but the sext_inreg of the
>> ; extended type i24 is never custom lowered.
>> -; FUNC-LABEL: @bfe_sext_in_reg_i24
>> +; FUNC-LABEL: bfe_sext_in_reg_i24
>> ; SI: BUFFER_LOAD_DWORD [[LOAD:v[0-9]+]],
>> ; SI: V_LSHLREV_B32_e32 {{v[0-9]+}}, 8, {{v[0-9]+}}
>> ; SI: V_ASHRREV_I32_e32 {{v[0-9]+}}, 8, {{v[0-9]+}}
>> diff --git a/test/CodeGen/R600/llvm.AMDGPU.bfe.u32.ll b/test/CodeGen/R600/llvm.AMDGPU.bfe.u32.ll
>> index 1a62253..505c704 100644
>> --- a/test/CodeGen/R600/llvm.AMDGPU.bfe.u32.ll
>> +++ b/test/CodeGen/R600/llvm.AMDGPU.bfe.u32.ll
>> @@ -3,7 +3,7 @@
>>
>> declare i32 @llvm.AMDGPU.bfe.u32(i32, i32, i32) nounwind readnone
>>
>> -; FUNC-LABEL: @bfe_u32_arg_arg_arg
>> +; FUNC-LABEL: bfe_u32_arg_arg_arg
>> ; SI: V_BFE_U32
>> ; EG: BFE_UINT
>> define void @bfe_u32_arg_arg_arg(i32 addrspace(1)* %out, i32 %src0, i32 %src1, i32 %src2) nounwind {
>> @@ -12,7 +12,7 @@ define void @bfe_u32_arg_arg_arg(i32 addrspace(1)* %out, i32 %src0, i32 %src1, i
>> ret void
>> }
>>
>> -; FUNC-LABEL: @bfe_u32_arg_arg_imm
>> +; FUNC-LABEL: bfe_u32_arg_arg_imm
>> ; SI: V_BFE_U32
>> ; EG: BFE_UINT
>> define void @bfe_u32_arg_arg_imm(i32 addrspace(1)* %out, i32 %src0, i32 %src1) nounwind {
>> @@ -21,7 +21,7 @@ define void @bfe_u32_arg_arg_imm(i32 addrspace(1)* %out, i32 %src0, i32 %src1) n
>> ret void
>> }
>>
>> -; FUNC-LABEL: @bfe_u32_arg_imm_arg
>> +; FUNC-LABEL: bfe_u32_arg_imm_arg
>> ; SI: V_BFE_U32
>> ; EG: BFE_UINT
>> define void @bfe_u32_arg_imm_arg(i32 addrspace(1)* %out, i32 %src0, i32 %src2) nounwind {
>> @@ -30,7 +30,7 @@ define void @bfe_u32_arg_imm_arg(i32 addrspace(1)* %out, i32 %src0, i32 %src2) n
>> ret void
>> }
>>
>> -; FUNC-LABEL: @bfe_u32_imm_arg_arg
>> +; FUNC-LABEL: bfe_u32_imm_arg_arg
>> ; SI: V_BFE_U32
>> ; EG: BFE_UINT
>> define void @bfe_u32_imm_arg_arg(i32 addrspace(1)* %out, i32 %src1, i32 %src2) nounwind {
>> @@ -39,7 +39,7 @@ define void @bfe_u32_imm_arg_arg(i32 addrspace(1)* %out, i32 %src1, i32 %src2) n
>> ret void
>> }
>>
>> -; FUNC-LABEL: @bfe_u32_arg_0_width_reg_offset
>> +; FUNC-LABEL: bfe_u32_arg_0_width_reg_offset
>> ; SI-NOT: BFE
>> ; SI: S_ENDPGM
>> ; EG-NOT: BFE
>> @@ -49,7 +49,7 @@ define void @bfe_u32_arg_0_width_reg_offset(i32 addrspace(1)* %out, i32 %src0, i
>> ret void
>> }
>>
>> -; FUNC-LABEL: @bfe_u32_arg_0_width_imm_offset
>> +; FUNC-LABEL: bfe_u32_arg_0_width_imm_offset
>> ; SI-NOT: BFE
>> ; SI: S_ENDPGM
>> ; EG-NOT: BFE
>> @@ -59,7 +59,7 @@ define void @bfe_u32_arg_0_width_imm_offset(i32 addrspace(1)* %out, i32 %src0, i
>> ret void
>> }
>>
>> -; FUNC-LABEL: @bfe_u32_zextload_i8
>> +; FUNC-LABEL: bfe_u32_zextload_i8
>> ; SI: BUFFER_LOAD_UBYTE
>> ; SI-NOT: BFE
>> ; SI: S_ENDPGM
>> @@ -71,7 +71,7 @@ define void @bfe_u32_zextload_i8(i32 addrspace(1)* %out, i8 addrspace(1)* %in) n
>> ret void
>> }
>>
>> -; FUNC-LABEL: @bfe_u32_zext_in_reg_i8
>> +; FUNC-LABEL: bfe_u32_zext_in_reg_i8
>> ; SI: BUFFER_LOAD_DWORD
>> ; SI: V_ADD_I32
>> ; SI-NEXT: V_AND_B32_e32
>> @@ -86,7 +86,7 @@ define void @bfe_u32_zext_in_reg_i8(i32 addrspace(1)* %out, i32 addrspace(1)* %i
>> ret void
>> }
>>
>> -; FUNC-LABEL: @bfe_u32_zext_in_reg_i16
>> +; FUNC-LABEL: bfe_u32_zext_in_reg_i16
>> ; SI: BUFFER_LOAD_DWORD
>> ; SI: V_ADD_I32
>> ; SI-NEXT: V_AND_B32_e32
>> @@ -101,7 +101,7 @@ define void @bfe_u32_zext_in_reg_i16(i32 addrspace(1)* %out, i32 addrspace(1)* %
>> ret void
>> }
>>
>> -; FUNC-LABEL: @bfe_u32_zext_in_reg_i8_offset_1
>> +; FUNC-LABEL: bfe_u32_zext_in_reg_i8_offset_1
>> ; SI: BUFFER_LOAD_DWORD
>> ; SI: V_ADD_I32
>> ; SI: BFE
>> @@ -115,7 +115,7 @@ define void @bfe_u32_zext_in_reg_i8_offset_1(i32 addrspace(1)* %out, i32 addrspa
>> ret void
>> }
>>
>> -; FUNC-LABEL: @bfe_u32_zext_in_reg_i8_offset_3
>> +; FUNC-LABEL: bfe_u32_zext_in_reg_i8_offset_3
>> ; SI: BUFFER_LOAD_DWORD
>> ; SI: V_ADD_I32
>> ; SI-NEXT: V_AND_B32_e32 {{v[0-9]+}}, 0xf8
>> @@ -130,7 +130,7 @@ define void @bfe_u32_zext_in_reg_i8_offset_3(i32 addrspace(1)* %out, i32 addrspa
>> ret void
>> }
>>
>> -; FUNC-LABEL: @bfe_u32_zext_in_reg_i8_offset_7
>> +; FUNC-LABEL: bfe_u32_zext_in_reg_i8_offset_7
>> ; SI: BUFFER_LOAD_DWORD
>> ; SI: V_ADD_I32
>> ; SI-NEXT: V_AND_B32_e32 {{v[0-9]+}}, 0x80
>> @@ -145,7 +145,7 @@ define void @bfe_u32_zext_in_reg_i8_offset_7(i32 addrspace(1)* %out, i32 addrspa
>> ret void
>> }
>>
>> -; FUNC-LABEL: @bfe_u32_zext_in_reg_i16_offset_8
>> +; FUNC-LABEL: bfe_u32_zext_in_reg_i16_offset_8
>> ; SI: BUFFER_LOAD_DWORD
>> ; SI: V_ADD_I32
>> ; SI-NEXT: BFE
>> @@ -159,7 +159,7 @@ define void @bfe_u32_zext_in_reg_i16_offset_8(i32 addrspace(1)* %out, i32 addrsp
>> ret void
>> }
>>
>> -; FUNC-LABEL: @bfe_u32_test_1
>> +; FUNC-LABEL: bfe_u32_test_1
>> ; SI: BUFFER_LOAD_DWORD
>> ; SI: V_AND_B32_e32 {{v[0-9]+}}, 1, {{v[0-9]+}}
>> ; SI: S_ENDPGM
>> @@ -187,7 +187,7 @@ define void @bfe_u32_test_3(i32 addrspace(1)* %out, i32 addrspace(1)* %in) nounw
>> ret void
>> }
>>
>> -; FUNC-LABEL: @bfe_u32_test_4
>> +; FUNC-LABEL: bfe_u32_test_4
>> ; SI-NOT: LSHL
>> ; SI-NOT: SHR
>> ; SI-NOT: BFE
>> @@ -203,7 +203,7 @@ define void @bfe_u32_test_4(i32 addrspace(1)* %out, i32 addrspace(1)* %in) nounw
>> ret void
>> }
>>
>> -; FUNC-LABEL: @bfe_u32_test_5
>> +; FUNC-LABEL: bfe_u32_test_5
>> ; SI: BUFFER_LOAD_DWORD
>> ; SI-NOT: LSHL
>> ; SI-NOT: SHR
>> @@ -218,7 +218,7 @@ define void @bfe_u32_test_5(i32 addrspace(1)* %out, i32 addrspace(1)* %in) nounw
>> ret void
>> }
>>
>> -; FUNC-LABEL: @bfe_u32_test_6
>> +; FUNC-LABEL: bfe_u32_test_6
>> ; SI: V_LSHLREV_B32_e32 v{{[0-9]+}}, 31, v{{[0-9]+}}
>> ; SI: V_LSHRREV_B32_e32 v{{[0-9]+}}, 1, v{{[0-9]+}}
>> ; SI: S_ENDPGM
>> @@ -230,7 +230,7 @@ define void @bfe_u32_test_6(i32 addrspace(1)* %out, i32 addrspace(1)* %in) nounw
>> ret void
>> }
>>
>> -; FUNC-LABEL: @bfe_u32_test_7
>> +; FUNC-LABEL: bfe_u32_test_7
>> ; SI: V_LSHLREV_B32_e32 v{{[0-9]+}}, 31, v{{[0-9]+}}
>> ; SI-NOT: BFE
>> ; SI: S_ENDPGM
>> @@ -242,7 +242,7 @@ define void @bfe_u32_test_7(i32 addrspace(1)* %out, i32 addrspace(1)* %in) nounw
>> ret void
>> }
>>
>> -; FUNC-LABEL: @bfe_u32_test_8
>> +; FUNC-LABEL: bfe_u32_test_8
>> ; SI-NOT: BFE
>> ; SI: V_AND_B32_e32 {{v[0-9]+}}, 1, {{v[0-9]+}}
>> ; SI-NOT: BFE
>> @@ -255,7 +255,7 @@ define void @bfe_u32_test_8(i32 addrspace(1)* %out, i32 addrspace(1)* %in) nounw
>> ret void
>> }
>>
>> -; FUNC-LABEL: @bfe_u32_test_9
>> +; FUNC-LABEL: bfe_u32_test_9
>> ; SI-NOT: BFE
>> ; SI: V_LSHRREV_B32_e32 v{{[0-9]+}}, 31, v{{[0-9]+}}
>> ; SI-NOT: BFE
>> @@ -267,7 +267,7 @@ define void @bfe_u32_test_9(i32 addrspace(1)* %out, i32 addrspace(1)* %in) nounw
>> ret void
>> }
>>
>> -; FUNC-LABEL: @bfe_u32_test_10
>> +; FUNC-LABEL: bfe_u32_test_10
>> ; SI-NOT: BFE
>> ; SI: V_LSHRREV_B32_e32 v{{[0-9]+}}, 1, v{{[0-9]+}}
>> ; SI-NOT: BFE
>> @@ -279,7 +279,7 @@ define void @bfe_u32_test_10(i32 addrspace(1)* %out, i32 addrspace(1)* %in) noun
>> ret void
>> }
>>
>> -; FUNC-LABEL: @bfe_u32_test_11
>> +; FUNC-LABEL: bfe_u32_test_11
>> ; SI-NOT: BFE
>> ; SI: V_LSHRREV_B32_e32 v{{[0-9]+}}, 8, v{{[0-9]+}}
>> ; SI-NOT: BFE
>> @@ -291,7 +291,7 @@ define void @bfe_u32_test_11(i32 addrspace(1)* %out, i32 addrspace(1)* %in) noun
>> ret void
>> }
>>
>> -; FUNC-LABEL: @bfe_u32_test_12
>> +; FUNC-LABEL: bfe_u32_test_12
>> ; SI-NOT: BFE
>> ; SI: V_LSHRREV_B32_e32 v{{[0-9]+}}, 24, v{{[0-9]+}}
>> ; SI-NOT: BFE
>> @@ -303,7 +303,7 @@ define void @bfe_u32_test_12(i32 addrspace(1)* %out, i32 addrspace(1)* %in) noun
>> ret void
>> }
>>
>> -; FUNC-LABEL: @bfe_u32_test_13
>> +; FUNC-LABEL: bfe_u32_test_13
>> ; V_ASHRREV_U32_e32 {{v[0-9]+}}, 31, {{v[0-9]+}}
>> ; SI-NOT: BFE
>> ; SI: S_ENDPGM
>> @@ -314,7 +314,7 @@ define void @bfe_u32_test_13(i32 addrspace(1)* %out, i32 addrspace(1)* %in) noun
>> store i32 %bfe, i32 addrspace(1)* %out, align 4 ret void
>> }
>>
>> -; FUNC-LABEL: @bfe_u32_test_14
>> +; FUNC-LABEL: bfe_u32_test_14
>> ; SI-NOT: LSHR
>> ; SI-NOT: BFE
>> ; SI: S_ENDPGM
>> @@ -325,7 +325,7 @@ define void @bfe_u32_test_14(i32 addrspace(1)* %out, i32 addrspace(1)* %in) noun
>> store i32 %bfe, i32 addrspace(1)* %out, align 4 ret void
>> }
>>
>> -; FUNC-LABEL: @bfe_u32_constant_fold_test_0
>> +; FUNC-LABEL: bfe_u32_constant_fold_test_0
>> ; SI-NOT: BFE
>> ; SI: V_MOV_B32_e32 [[VREG:v[0-9]+]], 0
>> ; SI: BUFFER_STORE_DWORD [[VREG]],
>> @@ -337,7 +337,7 @@ define void @bfe_u32_constant_fold_test_0(i32 addrspace(1)* %out) nounwind {
>> ret void
>> }
>>
>> -; FUNC-LABEL: @bfe_u32_constant_fold_test_1
>> +; FUNC-LABEL: bfe_u32_constant_fold_test_1
>> ; SI-NOT: BFE
>> ; SI: V_MOV_B32_e32 [[VREG:v[0-9]+]], 0
>> ; SI: BUFFER_STORE_DWORD [[VREG]],
>> @@ -349,7 +349,7 @@ define void @bfe_u32_constant_fold_test_1(i32 addrspace(1)* %out) nounwind {
>> ret void
>> }
>>
>> -; FUNC-LABEL: @bfe_u32_constant_fold_test_2
>> +; FUNC-LABEL: bfe_u32_constant_fold_test_2
>> ; SI-NOT: BFE
>> ; SI: V_MOV_B32_e32 [[VREG:v[0-9]+]], 0
>> ; SI: BUFFER_STORE_DWORD [[VREG]],
>> @@ -361,7 +361,7 @@ define void @bfe_u32_constant_fold_test_2(i32 addrspace(1)* %out) nounwind {
>> ret void
>> }
>>
>> -; FUNC-LABEL: @bfe_u32_constant_fold_test_3
>> +; FUNC-LABEL: bfe_u32_constant_fold_test_3
>> ; SI-NOT: BFE
>> ; SI: V_MOV_B32_e32 [[VREG:v[0-9]+]], 1
>> ; SI: BUFFER_STORE_DWORD [[VREG]],
>> @@ -373,7 +373,7 @@ define void @bfe_u32_constant_fold_test_3(i32 addrspace(1)* %out) nounwind {
>> ret void
>> }
>>
>> -; FUNC-LABEL: @bfe_u32_constant_fold_test_4
>> +; FUNC-LABEL: bfe_u32_constant_fold_test_4
>> ; SI-NOT: BFE
>> ; SI: V_MOV_B32_e32 [[VREG:v[0-9]+]], -1
>> ; SI: BUFFER_STORE_DWORD [[VREG]],
>> @@ -385,7 +385,7 @@ define void @bfe_u32_constant_fold_test_4(i32 addrspace(1)* %out) nounwind {
>> ret void
>> }
>>
>> -; FUNC-LABEL: @bfe_u32_constant_fold_test_5
>> +; FUNC-LABEL: bfe_u32_constant_fold_test_5
>> ; SI-NOT: BFE
>> ; SI: V_MOV_B32_e32 [[VREG:v[0-9]+]], 1
>> ; SI: BUFFER_STORE_DWORD [[VREG]],
>> @@ -397,7 +397,7 @@ define void @bfe_u32_constant_fold_test_5(i32 addrspace(1)* %out) nounwind {
>> ret void
>> }
>>
>> -; FUNC-LABEL: @bfe_u32_constant_fold_test_6
>> +; FUNC-LABEL: bfe_u32_constant_fold_test_6
>> ; SI-NOT: BFE
>> ; SI: V_MOV_B32_e32 [[VREG:v[0-9]+]], 0x80
>> ; SI: BUFFER_STORE_DWORD [[VREG]],
>> @@ -409,7 +409,7 @@ define void @bfe_u32_constant_fold_test_6(i32 addrspace(1)* %out) nounwind {
>> ret void
>> }
>>
>> -; FUNC-LABEL: @bfe_u32_constant_fold_test_7
>> +; FUNC-LABEL: bfe_u32_constant_fold_test_7
>> ; SI-NOT: BFE
>> ; SI: V_MOV_B32_e32 [[VREG:v[0-9]+]], 0x7f
>> ; SI: BUFFER_STORE_DWORD [[VREG]],
>> @@ -421,7 +421,7 @@ define void @bfe_u32_constant_fold_test_7(i32 addrspace(1)* %out) nounwind {
>> ret void
>> }
>>
>> -; FUNC-LABEL: @bfe_u32_constant_fold_test_8
>> +; FUNC-LABEL: bfe_u32_constant_fold_test_8
>> ; SI-NOT: BFE
>> ; SI: V_MOV_B32_e32 [[VREG:v[0-9]+]], 1
>> ; SI: BUFFER_STORE_DWORD [[VREG]],
>> @@ -433,7 +433,7 @@ define void @bfe_u32_constant_fold_test_8(i32 addrspace(1)* %out) nounwind {
>> ret void
>> }
>>
>> -; FUNC-LABEL: @bfe_u32_constant_fold_test_9
>> +; FUNC-LABEL: bfe_u32_constant_fold_test_9
>> ; SI-NOT: BFE
>> ; SI: V_MOV_B32_e32 [[VREG:v[0-9]+]], 1
>> ; SI: BUFFER_STORE_DWORD [[VREG]],
>> @@ -445,7 +445,7 @@ define void @bfe_u32_constant_fold_test_9(i32 addrspace(1)* %out) nounwind {
>> ret void
>> }
>>
>> -; FUNC-LABEL: @bfe_u32_constant_fold_test_10
>> +; FUNC-LABEL: bfe_u32_constant_fold_test_10
>> ; SI-NOT: BFE
>> ; SI: V_MOV_B32_e32 [[VREG:v[0-9]+]], 0
>> ; SI: BUFFER_STORE_DWORD [[VREG]],
>> @@ -457,7 +457,7 @@ define void @bfe_u32_constant_fold_test_10(i32 addrspace(1)* %out) nounwind {
>> ret void
>> }
>>
>> -; FUNC-LABEL: @bfe_u32_constant_fold_test_11
>> +; FUNC-LABEL: bfe_u32_constant_fold_test_11
>> ; SI-NOT: BFE
>> ; SI: V_MOV_B32_e32 [[VREG:v[0-9]+]], 10
>> ; SI: BUFFER_STORE_DWORD [[VREG]],
>> @@ -469,7 +469,7 @@ define void @bfe_u32_constant_fold_test_11(i32 addrspace(1)* %out) nounwind {
>> ret void
>> }
>>
>> -; FUNC-LABEL: @bfe_u32_constant_fold_test_12
>> +; FUNC-LABEL: bfe_u32_constant_fold_test_12
>> ; SI-NOT: BFE
>> ; SI: V_MOV_B32_e32 [[VREG:v[0-9]+]], 0
>> ; SI: BUFFER_STORE_DWORD [[VREG]],
>> @@ -481,7 +481,7 @@ define void @bfe_u32_constant_fold_test_12(i32 addrspace(1)* %out) nounwind {
>> ret void
>> }
>>
>> -; FUNC-LABEL: @bfe_u32_constant_fold_test_13
>> +; FUNC-LABEL: bfe_u32_constant_fold_test_13
>> ; SI-NOT: BFE
>> ; SI: V_MOV_B32_e32 [[VREG:v[0-9]+]], 1
>> ; SI: BUFFER_STORE_DWORD [[VREG]],
>> @@ -493,7 +493,7 @@ define void @bfe_u32_constant_fold_test_13(i32 addrspace(1)* %out) nounwind {
>> ret void
>> }
>>
>> -; FUNC-LABEL: @bfe_u32_constant_fold_test_14
>> +; FUNC-LABEL: bfe_u32_constant_fold_test_14
>> ; SI-NOT: BFE
>> ; SI: V_MOV_B32_e32 [[VREG:v[0-9]+]], 40
>> ; SI: BUFFER_STORE_DWORD [[VREG]],
>> @@ -505,7 +505,7 @@ define void @bfe_u32_constant_fold_test_14(i32 addrspace(1)* %out) nounwind {
>> ret void
>> }
>>
>> -; FUNC-LABEL: @bfe_u32_constant_fold_test_15
>> +; FUNC-LABEL: bfe_u32_constant_fold_test_15
>> ; SI-NOT: BFE
>> ; SI: V_MOV_B32_e32 [[VREG:v[0-9]+]], 10
>> ; SI: BUFFER_STORE_DWORD [[VREG]],
>> @@ -517,7 +517,7 @@ define void @bfe_u32_constant_fold_test_15(i32 addrspace(1)* %out) nounwind {
>> ret void
>> }
>>
>> -; FUNC-LABEL: @bfe_u32_constant_fold_test_16
>> +; FUNC-LABEL: bfe_u32_constant_fold_test_16
>> ; SI-NOT: BFE
>> ; SI: V_MOV_B32_e32 [[VREG:v[0-9]+]], 0x7f
>> ; SI: BUFFER_STORE_DWORD [[VREG]],
>> @@ -529,7 +529,7 @@ define void @bfe_u32_constant_fold_test_16(i32 addrspace(1)* %out) nounwind {
>> ret void
>> }
>>
>> -; FUNC-LABEL: @bfe_u32_constant_fold_test_17
>> +; FUNC-LABEL: bfe_u32_constant_fold_test_17
>> ; SI-NOT: BFE
>> ; SI: V_MOV_B32_e32 [[VREG:v[0-9]+]], 0x7f
>> ; SI: BUFFER_STORE_DWORD [[VREG]],
>> @@ -541,7 +541,7 @@ define void @bfe_u32_constant_fold_test_17(i32 addrspace(1)* %out) nounwind {
>> ret void
>> }
>>
>> -; FUNC-LABEL: @bfe_u32_constant_fold_test_18
>> +; FUNC-LABEL: bfe_u32_constant_fold_test_18
>> ; SI-NOT: BFE
>> ; SI: V_MOV_B32_e32 [[VREG:v[0-9]+]], 0
>> ; SI: BUFFER_STORE_DWORD [[VREG]],
>> diff --git a/test/CodeGen/R600/llvm.AMDGPU.bfi.ll b/test/CodeGen/R600/llvm.AMDGPU.bfi.ll
>> index e1de45b..3afbc77 100644
>> --- a/test/CodeGen/R600/llvm.AMDGPU.bfi.ll
>> +++ b/test/CodeGen/R600/llvm.AMDGPU.bfi.ll
>> @@ -3,7 +3,7 @@
>>
>> declare i32 @llvm.AMDGPU.bfi(i32, i32, i32) nounwind readnone
>>
>> -; FUNC-LABEL: @bfi_arg_arg_arg
>> +; FUNC-LABEL: bfi_arg_arg_arg
>> ; SI: V_BFI_B32
>> ; EG: BFI_INT
>> define void @bfi_arg_arg_arg(i32 addrspace(1)* %out, i32 %src0, i32 %src1, i32 %src2) nounwind {
>> @@ -12,7 +12,7 @@ define void @bfi_arg_arg_arg(i32 addrspace(1)* %out, i32 %src0, i32 %src1, i32 %
>> ret void
>> }
>>
>> -; FUNC-LABEL: @bfi_arg_arg_imm
>> +; FUNC-LABEL: bfi_arg_arg_imm
>> ; SI: V_BFI_B32
>> ; EG: BFI_INT
>> define void @bfi_arg_arg_imm(i32 addrspace(1)* %out, i32 %src0, i32 %src1) nounwind {
>> @@ -21,7 +21,7 @@ define void @bfi_arg_arg_imm(i32 addrspace(1)* %out, i32 %src0, i32 %src1) nounw
>> ret void
>> }
>>
>> -; FUNC-LABEL: @bfi_arg_imm_arg
>> +; FUNC-LABEL: bfi_arg_imm_arg
>> ; SI: V_BFI_B32
>> ; EG: BFI_INT
>> define void @bfi_arg_imm_arg(i32 addrspace(1)* %out, i32 %src0, i32 %src2) nounwind {
>> @@ -30,7 +30,7 @@ define void @bfi_arg_imm_arg(i32 addrspace(1)* %out, i32 %src0, i32 %src2) nounw
>> ret void
>> }
>>
>> -; FUNC-LABEL: @bfi_imm_arg_arg
>> +; FUNC-LABEL: bfi_imm_arg_arg
>> ; SI: V_BFI_B32
>> ; EG: BFI_INT
>> define void @bfi_imm_arg_arg(i32 addrspace(1)* %out, i32 %src1, i32 %src2) nounwind {
>> diff --git a/test/CodeGen/R600/llvm.AMDGPU.bfm.ll b/test/CodeGen/R600/llvm.AMDGPU.bfm.ll
>> index ef8721e..68938a5 100644
>> --- a/test/CodeGen/R600/llvm.AMDGPU.bfm.ll
>> +++ b/test/CodeGen/R600/llvm.AMDGPU.bfm.ll
>> @@ -3,7 +3,7 @@
>>
>> declare i32 @llvm.AMDGPU.bfm(i32, i32) nounwind readnone
>>
>> -; FUNC-LABEL: @bfm_arg_arg
>> +; FUNC-LABEL: bfm_arg_arg
>> ; SI: V_BFM
>> ; EG: BFM_INT
>> define void @bfm_arg_arg(i32 addrspace(1)* %out, i32 %src0, i32 %src1) nounwind {
>> @@ -12,7 +12,7 @@ define void @bfm_arg_arg(i32 addrspace(1)* %out, i32 %src0, i32 %src1) nounwind
>> ret void
>> }
>>
>> -; FUNC-LABEL: @bfm_arg_imm
>> +; FUNC-LABEL: bfm_arg_imm
>> ; SI: V_BFM
>> ; EG: BFM_INT
>> define void @bfm_arg_imm(i32 addrspace(1)* %out, i32 %src0) nounwind {
>> @@ -21,7 +21,7 @@ define void @bfm_arg_imm(i32 addrspace(1)* %out, i32 %src0) nounwind {
>> ret void
>> }
>>
>> -; FUNC-LABEL: @bfm_imm_arg
>> +; FUNC-LABEL: bfm_imm_arg
>> ; SI: V_BFM
>> ; EG: BFM_INT
>> define void @bfm_imm_arg(i32 addrspace(1)* %out, i32 %src1) nounwind {
>> @@ -30,7 +30,7 @@ define void @bfm_imm_arg(i32 addrspace(1)* %out, i32 %src1) nounwind {
>> ret void
>> }
>>
>> -; FUNC-LABEL: @bfm_imm_imm
>> +; FUNC-LABEL: bfm_imm_imm
>> ; SI: V_BFM
>> ; EG: BFM_INT
>> define void @bfm_imm_imm(i32 addrspace(1)* %out) nounwind {
>> diff --git a/test/CodeGen/R600/llvm.AMDGPU.brev.ll b/test/CodeGen/R600/llvm.AMDGPU.brev.ll
>> index 68a5ad0..e1ffc38 100644
>> --- a/test/CodeGen/R600/llvm.AMDGPU.brev.ll
>> +++ b/test/CodeGen/R600/llvm.AMDGPU.brev.ll
>> @@ -2,7 +2,7 @@
>>
>> declare i32 @llvm.AMDGPU.brev(i32) nounwind readnone
>>
>> -; FUNC-LABEL: @s_brev_i32:
>> +; FUNC-LABEL: s_brev_i32:
>> ; SI: S_LOAD_DWORD [[VAL:s[0-9]+]],
>> ; SI: S_BREV_B32 [[SRESULT:s[0-9]+]], [[VAL]]
>> ; SI: V_MOV_B32_e32 [[VRESULT:v[0-9]+]], [[SRESULT]]
>> @@ -14,7 +14,7 @@ define void @s_brev_i32(i32 addrspace(1)* noalias %out, i32 %val) nounwind {
>> ret void
>> }
>>
>> -; FUNC-LABEL: @v_brev_i32:
>> +; FUNC-LABEL: v_brev_i32:
>> ; SI: BUFFER_LOAD_DWORD [[VAL:v[0-9]+]],
>> ; SI: V_BFREV_B32_e32 [[RESULT:v[0-9]+]], [[VAL]]
>> ; SI: BUFFER_STORE_DWORD [[RESULT]],
>> diff --git a/test/CodeGen/R600/llvm.AMDGPU.clamp.ll b/test/CodeGen/R600/llvm.AMDGPU.clamp.ll
>> index cafca08..1a5633c 100644
>> --- a/test/CodeGen/R600/llvm.AMDGPU.clamp.ll
>> +++ b/test/CodeGen/R600/llvm.AMDGPU.clamp.ll
>> @@ -4,7 +4,7 @@
>> declare float @llvm.AMDGPU.clamp.f32(float, float, float) nounwind readnone
>> declare float @llvm.AMDIL.clamp.f32(float, float, float) nounwind readnone
>>
>> -; FUNC-LABEL: @clamp_0_1_f32
>> +; FUNC-LABEL: clamp_0_1_f32
>> ; SI: S_LOAD_DWORD [[ARG:s[0-9]+]],
>> ; SI: V_ADD_F32_e64 [[RESULT:v[0-9]+]], 0, [[ARG]], 1, 0
>> ; SI: BUFFER_STORE_DWORD [[RESULT]]
>> @@ -17,7 +17,7 @@ define void @clamp_0_1_f32(float addrspace(1)* %out, float %src) nounwind {
>> ret void
>> }
>>
>> -; FUNC-LABEL: @clamp_0_1_amdil_legacy_f32
>> +; FUNC-LABEL: clamp_0_1_amdil_legacy_f32
>> ; SI: S_LOAD_DWORD [[ARG:s[0-9]+]],
>> ; SI: V_ADD_F32_e64 [[RESULT:v[0-9]+]], 0, [[ARG]], 1, 0
>> ; SI: BUFFER_STORE_DWORD [[RESULT]]
>> diff --git a/test/CodeGen/R600/llvm.AMDGPU.cube.ll b/test/CodeGen/R600/llvm.AMDGPU.cube.ll
>> index 110bbfd..71a5f65 100644
>> --- a/test/CodeGen/R600/llvm.AMDGPU.cube.ll
>> +++ b/test/CodeGen/R600/llvm.AMDGPU.cube.ll
>> @@ -1,7 +1,7 @@
>>
>> ; RUN: llc < %s -march=r600 -mcpu=redwood | FileCheck %s
>>
>> -; CHECK: @cube
>> +; CHECK: cube
>> ; CHECK: CUBE T{{[0-9]}}.X
>> ; CHECK: CUBE T{{[0-9]}}.Y
>> ; CHECK: CUBE T{{[0-9]}}.Z
>> diff --git a/test/CodeGen/R600/llvm.AMDGPU.cvt_f32_ubyte.ll b/test/CodeGen/R600/llvm.AMDGPU.cvt_f32_ubyte.ll
>> index 6facb47..beb81c4 100644
>> --- a/test/CodeGen/R600/llvm.AMDGPU.cvt_f32_ubyte.ll
>> +++ b/test/CodeGen/R600/llvm.AMDGPU.cvt_f32_ubyte.ll
>> @@ -5,7 +5,7 @@ declare float @llvm.AMDGPU.cvt.f32.ubyte1(i32) nounwind readnone
>> declare float @llvm.AMDGPU.cvt.f32.ubyte2(i32) nounwind readnone
>> declare float @llvm.AMDGPU.cvt.f32.ubyte3(i32) nounwind readnone
>>
>> -; SI-LABEL: @test_unpack_byte0_to_float:
>> +; SI-LABEL: test_unpack_byte0_to_float:
>> ; SI: V_CVT_F32_UBYTE0
>> define void @test_unpack_byte0_to_float(float addrspace(1)* %out, i32 addrspace(1)* %in) nounwind {
>> %val = load i32 addrspace(1)* %in, align 4
>> @@ -14,7 +14,7 @@ define void @test_unpack_byte0_to_float(float addrspace(1)* %out, i32 addrspace(
>> ret void
>> }
>>
>> -; SI-LABEL: @test_unpack_byte1_to_float:
>> +; SI-LABEL: test_unpack_byte1_to_float:
>> ; SI: V_CVT_F32_UBYTE1
>> define void @test_unpack_byte1_to_float(float addrspace(1)* %out, i32 addrspace(1)* %in) nounwind {
>> %val = load i32 addrspace(1)* %in, align 4
>> @@ -23,7 +23,7 @@ define void @test_unpack_byte1_to_float(float addrspace(1)* %out, i32 addrspace(
>> ret void
>> }
>>
>> -; SI-LABEL: @test_unpack_byte2_to_float:
>> +; SI-LABEL: test_unpack_byte2_to_float:
>> ; SI: V_CVT_F32_UBYTE2
>> define void @test_unpack_byte2_to_float(float addrspace(1)* %out, i32 addrspace(1)* %in) nounwind {
>> %val = load i32 addrspace(1)* %in, align 4
>> @@ -32,7 +32,7 @@ define void @test_unpack_byte2_to_float(float addrspace(1)* %out, i32 addrspace(
>> ret void
>> }
>>
>> -; SI-LABEL: @test_unpack_byte3_to_float:
>> +; SI-LABEL: test_unpack_byte3_to_float:
>> ; SI: V_CVT_F32_UBYTE3
>> define void @test_unpack_byte3_to_float(float addrspace(1)* %out, i32 addrspace(1)* %in) nounwind {
>> %val = load i32 addrspace(1)* %in, align 4
>> diff --git a/test/CodeGen/R600/llvm.AMDGPU.div_fixup.ll b/test/CodeGen/R600/llvm.AMDGPU.div_fixup.ll
>> index d0eeeab..79d8f95 100644
>> --- a/test/CodeGen/R600/llvm.AMDGPU.div_fixup.ll
>> +++ b/test/CodeGen/R600/llvm.AMDGPU.div_fixup.ll
>> @@ -3,7 +3,7 @@
>> declare float @llvm.AMDGPU.div.fixup.f32(float, float, float) nounwind readnone
>> declare double @llvm.AMDGPU.div.fixup.f64(double, double, double) nounwind readnone
>>
>> -; SI-LABEL: @test_div_fixup_f32:
>> +; SI-LABEL: test_div_fixup_f32:
>> ; SI-DAG: S_LOAD_DWORD [[SA:s[0-9]+]], s{{\[[0-9]+:[0-9]+\]}}, 0xb
>> ; SI-DAG: S_LOAD_DWORD [[SC:s[0-9]+]], s{{\[[0-9]+:[0-9]+\]}}, 0xd
>> ; SI-DAG: S_LOAD_DWORD [[SB:s[0-9]+]], s{{\[[0-9]+:[0-9]+\]}}, 0xc
>> @@ -18,7 +18,7 @@ define void @test_div_fixup_f32(float addrspace(1)* %out, float %a, float %b, fl
>> ret void
>> }
>>
>> -; SI-LABEL: @test_div_fixup_f64:
>> +; SI-LABEL: test_div_fixup_f64:
>> ; SI: V_DIV_FIXUP_F64
>> define void @test_div_fixup_f64(double addrspace(1)* %out, double %a, double %b, double %c) nounwind {
>> %result = call double @llvm.AMDGPU.div.fixup.f64(double %a, double %b, double %c) nounwind readnone
>> diff --git a/test/CodeGen/R600/llvm.AMDGPU.div_fmas.ll b/test/CodeGen/R600/llvm.AMDGPU.div_fmas.ll
>> index 40b441d..7e89c15 100644
>> --- a/test/CodeGen/R600/llvm.AMDGPU.div_fmas.ll
>> +++ b/test/CodeGen/R600/llvm.AMDGPU.div_fmas.ll
>> @@ -3,7 +3,7 @@
>> declare float @llvm.AMDGPU.div.fmas.f32(float, float, float) nounwind readnone
>> declare double @llvm.AMDGPU.div.fmas.f64(double, double, double) nounwind readnone
>>
>> -; SI-LABEL: @test_div_fmas_f32:
>> +; SI-LABEL: test_div_fmas_f32:
>> ; SI-DAG: S_LOAD_DWORD [[SA:s[0-9]+]], s{{\[[0-9]+:[0-9]+\]}}, 0xb
>> ; SI-DAG: S_LOAD_DWORD [[SC:s[0-9]+]], s{{\[[0-9]+:[0-9]+\]}}, 0xd
>> ; SI-DAG: S_LOAD_DWORD [[SB:s[0-9]+]], s{{\[[0-9]+:[0-9]+\]}}, 0xc
>> @@ -18,7 +18,7 @@ define void @test_div_fmas_f32(float addrspace(1)* %out, float %a, float %b, flo
>> ret void
>> }
>>
>> -; SI-LABEL: @test_div_fmas_f64:
>> +; SI-LABEL: test_div_fmas_f64:
>> ; SI: V_DIV_FMAS_F64
>> define void @test_div_fmas_f64(double addrspace(1)* %out, double %a, double %b, double %c) nounwind {
>> %result = call double @llvm.AMDGPU.div.fmas.f64(double %a, double %b, double %c) nounwind readnone
>> diff --git a/test/CodeGen/R600/llvm.AMDGPU.fract.ll b/test/CodeGen/R600/llvm.AMDGPU.fract.ll
>> index 72ec1c5..394087e 100644
>> --- a/test/CodeGen/R600/llvm.AMDGPU.fract.ll
>> +++ b/test/CodeGen/R600/llvm.AMDGPU.fract.ll
>> @@ -6,7 +6,7 @@ declare float @llvm.AMDGPU.fract.f32(float) nounwind readnone
>> ; Legacy name
>> declare float @llvm.AMDIL.fraction.f32(float) nounwind readnone
>>
>> -; FUNC-LABEL: @fract_f32
>> +; FUNC-LABEL: fract_f32
>> ; SI: V_FRACT_F32
>> ; EG: FRACT
>> define void @fract_f32(float addrspace(1)* %out, float addrspace(1)* %src) nounwind {
>> @@ -16,7 +16,7 @@ define void @fract_f32(float addrspace(1)* %out, float addrspace(1)* %src) nounw
>> ret void
>> }
>>
>> -; FUNC-LABEL: @fract_f32_legacy_amdil
>> +; FUNC-LABEL: fract_f32_legacy_amdil
>> ; SI: V_FRACT_F32
>> ; EG: FRACT
>> define void @fract_f32_legacy_amdil(float addrspace(1)* %out, float addrspace(1)* %src) nounwind {
>> diff --git a/test/CodeGen/R600/llvm.AMDGPU.imad24.ll b/test/CodeGen/R600/llvm.AMDGPU.imad24.ll
>> index 95795ea..b20473a 100644
>> --- a/test/CodeGen/R600/llvm.AMDGPU.imad24.ll
>> +++ b/test/CodeGen/R600/llvm.AMDGPU.imad24.ll
>> @@ -8,7 +8,7 @@
>>
>> declare i32 @llvm.AMDGPU.imad24(i32, i32, i32) nounwind readnone
>>
>> -; FUNC-LABEL: @test_imad24
>> +; FUNC-LABEL: test_imad24
>> ; SI: V_MAD_I32_I24
>> ; CM: MULADD_INT24
>> ; R600: MULLO_INT
>> diff --git a/test/CodeGen/R600/llvm.AMDGPU.imax.ll b/test/CodeGen/R600/llvm.AMDGPU.imax.ll
>> index 01c9f43..f239270 100644
>> --- a/test/CodeGen/R600/llvm.AMDGPU.imax.ll
>> +++ b/test/CodeGen/R600/llvm.AMDGPU.imax.ll
>> @@ -1,6 +1,6 @@
>> ; RUN: llc < %s -march=r600 -mcpu=verde -verify-machineinstrs | FileCheck -check-prefix=SI %s
>>
>> -; SI-LABEL: @vector_imax
>> +; SI-LABEL: vector_imax
>> ; SI: V_MAX_I32_e32
>> define void @vector_imax(i32 %p0, i32 %p1, i32 addrspace(1)* %in) #0 {
>> main_body:
>> @@ -11,7 +11,7 @@ main_body:
>> ret void
>> }
>>
>> -; SI-LABEL: @scalar_imax
>> +; SI-LABEL: scalar_imax
>> ; SI: S_MAX_I32
>> define void @scalar_imax(i32 %p0, i32 %p1) #0 {
>> entry:
>> diff --git a/test/CodeGen/R600/llvm.AMDGPU.imin.ll b/test/CodeGen/R600/llvm.AMDGPU.imin.ll
>> index 565bf34..78213ca 100644
>> --- a/test/CodeGen/R600/llvm.AMDGPU.imin.ll
>> +++ b/test/CodeGen/R600/llvm.AMDGPU.imin.ll
>> @@ -1,6 +1,6 @@
>> ; RUN: llc < %s -march=r600 -mcpu=verde -verify-machineinstrs | FileCheck -check-prefix=SI %s
>>
>> -; SI-LABEL: @vector_imin
>> +; SI-LABEL: vector_imin
>> ; SI: V_MIN_I32_e32
>> define void @vector_imin(i32 %p0, i32 %p1, i32 addrspace(1)* %in) #0 {
>> main_body:
>> @@ -11,7 +11,7 @@ main_body:
>> ret void
>> }
>>
>> -; SI-LABEL: @scalar_imin
>> +; SI-LABEL: scalar_imin
>> ; SI: S_MIN_I32
>> define void @scalar_imin(i32 %p0, i32 %p1) #0 {
>> entry:
>> diff --git a/test/CodeGen/R600/llvm.AMDGPU.imul24.ll b/test/CodeGen/R600/llvm.AMDGPU.imul24.ll
>> index 8ee3520..cc54111 100644
>> --- a/test/CodeGen/R600/llvm.AMDGPU.imul24.ll
>> +++ b/test/CodeGen/R600/llvm.AMDGPU.imul24.ll
>> @@ -4,7 +4,7 @@
>>
>> declare i32 @llvm.AMDGPU.imul24(i32, i32) nounwind readnone
>>
>> -; FUNC-LABEL: @test_imul24
>> +; FUNC-LABEL: test_imul24
>> ; SI: V_MUL_I32_I24
>> ; CM: MUL_INT24
>> ; R600: MULLO_INT
>> diff --git a/test/CodeGen/R600/llvm.AMDGPU.kill.ll b/test/CodeGen/R600/llvm.AMDGPU.kill.ll
>> index 1f82ffb..abe8e7c 100644
>> --- a/test/CodeGen/R600/llvm.AMDGPU.kill.ll
>> +++ b/test/CodeGen/R600/llvm.AMDGPU.kill.ll
>> @@ -1,6 +1,6 @@
>> ; RUN: llc -march=r600 -mcpu=verde -verify-machineinstrs < %s | FileCheck -check-prefix=SI %s
>>
>> -; SI-LABEL: @kill_gs_const
>> +; SI-LABEL: kill_gs_const
>> ; SI-NOT: V_CMPX_LE_F32
>> ; SI: S_MOV_B64 exec, 0
>>
>> diff --git a/test/CodeGen/R600/llvm.AMDGPU.ldexp.ll b/test/CodeGen/R600/llvm.AMDGPU.ldexp.ll
>> index d59426c..8f833a1 100644
>> --- a/test/CodeGen/R600/llvm.AMDGPU.ldexp.ll
>> +++ b/test/CodeGen/R600/llvm.AMDGPU.ldexp.ll
>> @@ -3,7 +3,7 @@
>> declare float @llvm.AMDGPU.ldexp.f32(float, i32) nounwind readnone
>> declare double @llvm.AMDGPU.ldexp.f64(double, i32) nounwind readnone
>>
>> -; SI-LABEL: @test_ldexp_f32:
>> +; SI-LABEL: test_ldexp_f32:
>> ; SI: V_LDEXP_F32
>> ; SI: S_ENDPGM
>> define void @test_ldexp_f32(float addrspace(1)* %out, float %a, i32 %b) nounwind {
>> @@ -12,7 +12,7 @@ define void @test_ldexp_f32(float addrspace(1)* %out, float %a, i32 %b) nounwind
>> ret void
>> }
>>
>> -; SI-LABEL: @test_ldexp_f64:
>> +; SI-LABEL: test_ldexp_f64:
>> ; SI: V_LDEXP_F64
>> ; SI: S_ENDPGM
>> define void @test_ldexp_f64(double addrspace(1)* %out, double %a, i32 %b) nounwind {
>> diff --git a/test/CodeGen/R600/llvm.AMDGPU.legacy.rsq.ll b/test/CodeGen/R600/llvm.AMDGPU.legacy.rsq.ll
>> index 51964ee..8cc6b3c 100644
>> --- a/test/CodeGen/R600/llvm.AMDGPU.legacy.rsq.ll
>> +++ b/test/CodeGen/R600/llvm.AMDGPU.legacy.rsq.ll
>> @@ -3,7 +3,7 @@
>>
>> declare float @llvm.AMDGPU.legacy.rsq(float) nounwind readnone
>>
>> -; FUNC-LABEL: @rsq_legacy_f32
>> +; FUNC-LABEL: rsq_legacy_f32
>> ; SI: V_RSQ_LEGACY_F32_e32
>> ; EG: RECIPSQRT_IEEE
>> define void @rsq_legacy_f32(float addrspace(1)* %out, float %src) nounwind {
>> diff --git a/test/CodeGen/R600/llvm.AMDGPU.rcp.f64.ll b/test/CodeGen/R600/llvm.AMDGPU.rcp.f64.ll
>> index b5dda0c..b95b2c7 100644
>> --- a/test/CodeGen/R600/llvm.AMDGPU.rcp.f64.ll
>> +++ b/test/CodeGen/R600/llvm.AMDGPU.rcp.f64.ll
>> @@ -3,7 +3,7 @@
>> declare double @llvm.AMDGPU.rcp.f64(double) nounwind readnone
>> declare double @llvm.sqrt.f64(double) nounwind readnone
>>
>> -; FUNC-LABEL: @rcp_f64
>> +; FUNC-LABEL: rcp_f64
>> ; SI: V_RCP_F64_e32
>> define void @rcp_f64(double addrspace(1)* %out, double %src) nounwind {
>> %rcp = call double @llvm.AMDGPU.rcp.f64(double %src) nounwind readnone
>> @@ -11,7 +11,7 @@ define void @rcp_f64(double addrspace(1)* %out, double %src) nounwind {
>> ret void
>> }
>>
>> -; FUNC-LABEL: @rcp_pat_f64
>> +; FUNC-LABEL: rcp_pat_f64
>> ; SI: V_RCP_F64_e32
>> define void @rcp_pat_f64(double addrspace(1)* %out, double %src) nounwind {
>> %rcp = fdiv double 1.0, %src
>> @@ -19,7 +19,7 @@ define void @rcp_pat_f64(double addrspace(1)* %out, double %src) nounwind {
>> ret void
>> }
>>
>> -; FUNC-LABEL: @rsq_rcp_pat_f64
>> +; FUNC-LABEL: rsq_rcp_pat_f64
>> ; SI-UNSAFE: V_RSQ_F64_e32
>> ; SI-SAFE-NOT: V_RSQ_F64_e32
>> define void @rsq_rcp_pat_f64(double addrspace(1)* %out, double %src) nounwind {
>> diff --git a/test/CodeGen/R600/llvm.AMDGPU.rcp.ll b/test/CodeGen/R600/llvm.AMDGPU.rcp.ll
>> index df6c3bb..065a7b5 100644
>> --- a/test/CodeGen/R600/llvm.AMDGPU.rcp.ll
>> +++ b/test/CodeGen/R600/llvm.AMDGPU.rcp.ll
>> @@ -10,7 +10,7 @@ declare double @llvm.AMDGPU.rcp.f64(double) nounwind readnone
>>
>> declare float @llvm.sqrt.f32(float) nounwind readnone
>>
>> -; FUNC-LABEL: @rcp_f32
>> +; FUNC-LABEL: rcp_f32
>> ; SI: V_RCP_F32_e32
>> ; EG: RECIP_IEEE
>> define void @rcp_f32(float addrspace(1)* %out, float %src) nounwind {
>> @@ -20,7 +20,7 @@ define void @rcp_f32(float addrspace(1)* %out, float %src) nounwind {
>> }
>>
>> ; FIXME: Evergreen only ever does unsafe fp math.
>> -; FUNC-LABEL: @rcp_pat_f32
>> +; FUNC-LABEL: rcp_pat_f32
>>
>> ; SI-SAFE: V_RCP_F32_e32
>> ; XSI-SAFE-SPDENORM-NOT: V_RCP_F32_e32
>> @@ -33,7 +33,7 @@ define void @rcp_pat_f32(float addrspace(1)* %out, float %src) nounwind {
>> ret void
>> }
>>
>> -; FUNC-LABEL: @rsq_rcp_pat_f32
>> +; FUNC-LABEL: rsq_rcp_pat_f32
>> ; SI-UNSAFE: V_RSQ_F32_e32
>> ; SI-SAFE: V_SQRT_F32_e32
>> ; SI-SAFE: V_RCP_F32_e32
>> diff --git a/test/CodeGen/R600/llvm.AMDGPU.rsq.clamped.f64.ll b/test/CodeGen/R600/llvm.AMDGPU.rsq.clamped.f64.ll
>> index 100d6ff..4660951 100644
>> --- a/test/CodeGen/R600/llvm.AMDGPU.rsq.clamped.f64.ll
>> +++ b/test/CodeGen/R600/llvm.AMDGPU.rsq.clamped.f64.ll
>> @@ -2,7 +2,7 @@
>>
>> declare double @llvm.AMDGPU.rsq.clamped.f64(double) nounwind readnone
>>
>> -; FUNC-LABEL: @rsq_clamped_f64
>> +; FUNC-LABEL: rsq_clamped_f64
>> ; SI: V_RSQ_CLAMP_F64_e32
>> define void @rsq_clamped_f64(double addrspace(1)* %out, double %src) nounwind {
>> %rsq_clamped = call double @llvm.AMDGPU.rsq.clamped.f64(double %src) nounwind readnone
>> diff --git a/test/CodeGen/R600/llvm.AMDGPU.rsq.clamped.ll b/test/CodeGen/R600/llvm.AMDGPU.rsq.clamped.ll
>> index 683df73..6937b4b 100644
>> --- a/test/CodeGen/R600/llvm.AMDGPU.rsq.clamped.ll
>> +++ b/test/CodeGen/R600/llvm.AMDGPU.rsq.clamped.ll
>> @@ -4,7 +4,7 @@
>>
>> declare float @llvm.AMDGPU.rsq.clamped.f32(float) nounwind readnone
>>
>> -; FUNC-LABEL: @rsq_clamped_f32
>> +; FUNC-LABEL: rsq_clamped_f32
>> ; SI: V_RSQ_CLAMP_F32_e32
>> ; EG: RECIPSQRT_CLAMPED
>> define void @rsq_clamped_f32(float addrspace(1)* %out, float %src) nounwind {
>> diff --git a/test/CodeGen/R600/llvm.AMDGPU.rsq.ll b/test/CodeGen/R600/llvm.AMDGPU.rsq.ll
>> index 449b3af..38b8c95 100644
>> --- a/test/CodeGen/R600/llvm.AMDGPU.rsq.ll
>> +++ b/test/CodeGen/R600/llvm.AMDGPU.rsq.ll
>> @@ -3,7 +3,7 @@
>>
>> declare float @llvm.AMDGPU.rsq.f32(float) nounwind readnone
>>
>> -; FUNC-LABEL: @rsq_f32
>> +; FUNC-LABEL: rsq_f32
>> ; SI: V_RSQ_F32_e32 {{v[0-9]+}}, {{s[0-9]+}}
>> ; EG: RECIPSQRT_IEEE
>> define void @rsq_f32(float addrspace(1)* %out, float %src) nounwind {
>> @@ -13,7 +13,7 @@ define void @rsq_f32(float addrspace(1)* %out, float %src) nounwind {
>> }
>>
>> ; TODO: Really these should be constant folded
>> -; FUNC-LABEL: @rsq_f32_constant_4.0
>> +; FUNC-LABEL: rsq_f32_constant_4.0
>> ; SI: V_RSQ_F32_e32 {{v[0-9]+}}, 4.0
>> ; EG: RECIPSQRT_IEEE
>> define void @rsq_f32_constant_4.0(float addrspace(1)* %out) nounwind {
>> @@ -22,7 +22,7 @@ define void @rsq_f32_constant_4.0(float addrspace(1)* %out) nounwind {
>> ret void
>> }
>>
>> -; FUNC-LABEL: @rsq_f32_constant_100.0
>> +; FUNC-LABEL: rsq_f32_constant_100.0
>> ; SI: V_RSQ_F32_e32 {{v[0-9]+}}, 0x42c80000
>> ; EG: RECIPSQRT_IEEE
>> define void @rsq_f32_constant_100.0(float addrspace(1)* %out) nounwind {
>> diff --git a/test/CodeGen/R600/llvm.AMDGPU.trig_preop.ll b/test/CodeGen/R600/llvm.AMDGPU.trig_preop.ll
>> index 1c736d4..aa32dc4 100644
>> --- a/test/CodeGen/R600/llvm.AMDGPU.trig_preop.ll
>> +++ b/test/CodeGen/R600/llvm.AMDGPU.trig_preop.ll
>> @@ -2,7 +2,7 @@
>>
>> declare double @llvm.AMDGPU.trig.preop.f64(double, i32) nounwind readnone
>>
>> -; SI-LABEL: @test_trig_preop_f64:
>> +; SI-LABEL: test_trig_preop_f64:
>> ; SI-DAG: BUFFER_LOAD_DWORD [[SEG:v[0-9]+]]
>> ; SI-DAG: BUFFER_LOAD_DWORDX2 [[SRC:v\[[0-9]+:[0-9]+\]]],
>> ; SI: V_TRIG_PREOP_F64 [[RESULT:v\[[0-9]+:[0-9]+\]]], [[SRC]], [[SEG]]
>> @@ -16,7 +16,7 @@ define void @test_trig_preop_f64(double addrspace(1)* %out, double addrspace(1)*
>> ret void
>> }
>>
>> -; SI-LABEL: @test_trig_preop_f64_imm_segment:
>> +; SI-LABEL: test_trig_preop_f64_imm_segment:
>> ; SI: BUFFER_LOAD_DWORDX2 [[SRC:v\[[0-9]+:[0-9]+\]]],
>> ; SI: V_TRIG_PREOP_F64 [[RESULT:v\[[0-9]+:[0-9]+\]]], [[SRC]], 7
>> ; SI: BUFFER_STORE_DWORDX2 [[RESULT]],
>> diff --git a/test/CodeGen/R600/llvm.AMDGPU.trunc.ll b/test/CodeGen/R600/llvm.AMDGPU.trunc.ll
>> index e6bb2c4..9f58e93 100644
>> --- a/test/CodeGen/R600/llvm.AMDGPU.trunc.ll
>> +++ b/test/CodeGen/R600/llvm.AMDGPU.trunc.ll
>> @@ -1,9 +1,9 @@
>> ; RUN: llc < %s -march=r600 -mcpu=redwood | FileCheck --check-prefix=R600-CHECK %s
>> ; RUN: llc < %s -march=r600 -mcpu=verde -verify-machineinstrs | FileCheck --check-prefix=SI-CHECK %s
>>
>> -; R600-CHECK: @amdgpu_trunc
>> +; R600-CHECK: amdgpu_trunc
>> ; R600-CHECK: TRUNC T{{[0-9]+\.[XYZW]}}, KC0[2].Z
>> -; SI-CHECK: @amdgpu_trunc
>> +; SI-CHECK: amdgpu_trunc
>> ; SI-CHECK: V_TRUNC_F32
>>
>> define void @amdgpu_trunc(float addrspace(1)* %out, float %x) {
>> diff --git a/test/CodeGen/R600/llvm.AMDGPU.umad24.ll b/test/CodeGen/R600/llvm.AMDGPU.umad24.ll
>> index afdfb18..5b01b3c 100644
>> --- a/test/CodeGen/R600/llvm.AMDGPU.umad24.ll
>> +++ b/test/CodeGen/R600/llvm.AMDGPU.umad24.ll
>> @@ -6,7 +6,7 @@
>>
>> declare i32 @llvm.AMDGPU.umad24(i32, i32, i32) nounwind readnone
>>
>> -; FUNC-LABEL: @test_umad24
>> +; FUNC-LABEL: test_umad24
>> ; SI: V_MAD_U32_U24
>> ; EG: MULADD_UINT24
>> ; R600: MULLO_UINT
>> diff --git a/test/CodeGen/R600/llvm.AMDGPU.umax.ll b/test/CodeGen/R600/llvm.AMDGPU.umax.ll
>> index 1b8da2e..ca2a5d4 100644
>> --- a/test/CodeGen/R600/llvm.AMDGPU.umax.ll
>> +++ b/test/CodeGen/R600/llvm.AMDGPU.umax.ll
>> @@ -1,6 +1,6 @@
>> ; RUN: llc < %s -march=r600 -mcpu=verde -verify-machineinstrs | FileCheck -check-prefix=SI %s
>>
>> -; SI-LABEL: @vector_umax
>> +; SI-LABEL: vector_umax
>> ; SI: V_MAX_U32_e32
>> define void @vector_umax(i32 %p0, i32 %p1, i32 addrspace(1)* %in) #0 {
>> main_body:
>> @@ -11,7 +11,7 @@ main_body:
>> ret void
>> }
>>
>> -; SI-LABEL: @scalar_umax
>> +; SI-LABEL: scalar_umax
>> ; SI: S_MAX_U32
>> define void @scalar_umax(i32 %p0, i32 %p1) #0 {
>> entry:
>> @@ -21,7 +21,7 @@ entry:
>> ret void
>> }
>>
>> -; SI-LABEL: @trunc_zext_umax
>> +; SI-LABEL: trunc_zext_umax
>> ; SI: BUFFER_LOAD_UBYTE [[VREG:v[0-9]+]],
>> ; SI: V_MAX_U32_e32 [[RESULT:v[0-9]+]], 0, [[VREG]]
>> ; SI-NOT: AND
>> diff --git a/test/CodeGen/R600/llvm.AMDGPU.umin.ll b/test/CodeGen/R600/llvm.AMDGPU.umin.ll
>> index 08397f8..8db70bb 100644
>> --- a/test/CodeGen/R600/llvm.AMDGPU.umin.ll
>> +++ b/test/CodeGen/R600/llvm.AMDGPU.umin.ll
>> @@ -1,6 +1,6 @@
>> ; RUN: llc < %s -march=r600 -mcpu=verde -verify-machineinstrs | FileCheck -check-prefix=SI %s
>>
>> -; SI-LABEL: @vector_umin
>> +; SI-LABEL: vector_umin
>> ; SI: V_MIN_U32_e32
>> define void @vector_umin(i32 %p0, i32 %p1, i32 addrspace(1)* %in) #0 {
>> main_body:
>> @@ -11,7 +11,7 @@ main_body:
>> ret void
>> }
>>
>> -; SI-LABEL: @scalar_umin
>> +; SI-LABEL: scalar_umin
>> ; SI: S_MIN_U32
>> define void @scalar_umin(i32 %p0, i32 %p1) #0 {
>> entry:
>> @@ -21,7 +21,7 @@ entry:
>> ret void
>> }
>>
>> -; SI-LABEL: @trunc_zext_umin
>> +; SI-LABEL: trunc_zext_umin
>> ; SI: BUFFER_LOAD_UBYTE [[VREG:v[0-9]+]],
>> ; SI: V_MIN_U32_e32 [[RESULT:v[0-9]+]], 0, [[VREG]]
>> ; SI-NOT: AND
>> diff --git a/test/CodeGen/R600/llvm.AMDGPU.umul24.ll b/test/CodeGen/R600/llvm.AMDGPU.umul24.ll
>> index 72a3602..2192fd9 100644
>> --- a/test/CodeGen/R600/llvm.AMDGPU.umul24.ll
>> +++ b/test/CodeGen/R600/llvm.AMDGPU.umul24.ll
>> @@ -6,7 +6,7 @@
>>
>> declare i32 @llvm.AMDGPU.umul24(i32, i32) nounwind readnone
>>
>> -; FUNC-LABEL: @test_umul24
>> +; FUNC-LABEL: test_umul24
>> ; SI: V_MUL_U32_U24
>> ; R600: MUL_UINT24
>> ; R600: MULLO_UINT
>> diff --git a/test/CodeGen/R600/llvm.SI.gather4.ll b/test/CodeGen/R600/llvm.SI.gather4.ll
>> index 8402faa..cf03ede 100644
>> --- a/test/CodeGen/R600/llvm.SI.gather4.ll
>> +++ b/test/CodeGen/R600/llvm.SI.gather4.ll
>> @@ -1,6 +1,6 @@
>> ;RUN: llc < %s -march=r600 -mcpu=verde -verify-machineinstrs | FileCheck %s
>>
>> -;CHECK-LABEL: @gather4_v2
>> +;CHECK-LABEL: gather4_v2:
>> ;CHECK: IMAGE_GATHER4 {{v\[[0-9]+:[0-9]+\]}}, 1, 0, 0, -1, 0, 0, 0, 0, {{v\[[0-9]+:[0-9]+\]}}, {{s\[[0-9]+:[0-9]+\]}}, {{s\[[0-9]+:[0-9]+\]}}
>> define void @gather4_v2() #0 {
>> main_body:
>> @@ -13,7 +13,7 @@ main_body:
>> ret void
>> }
>>
>> -;CHECK-LABEL: @gather4
>> +;CHECK-LABEL: gather4:
>> ;CHECK: IMAGE_GATHER4 {{v\[[0-9]+:[0-9]+\]}}, 1, 0, 0, -1, 0, 0, 0, 0, {{v\[[0-9]+:[0-9]+\]}}, {{s\[[0-9]+:[0-9]+\]}}, {{s\[[0-9]+:[0-9]+\]}}
>> define void @gather4() #0 {
>> main_body:
>> @@ -26,7 +26,7 @@ main_body:
>> ret void
>> }
>>
>> -;CHECK-LABEL: @gather4_cl
>> +;CHECK-LABEL: gather4_cl:
>> ;CHECK: IMAGE_GATHER4_CL {{v\[[0-9]+:[0-9]+\]}}, 1, 0, 0, -1, 0, 0, 0, 0, {{v\[[0-9]+:[0-9]+\]}}, {{s\[[0-9]+:[0-9]+\]}}, {{s\[[0-9]+:[0-9]+\]}}
>> define void @gather4_cl() #0 {
>> main_body:
>> @@ -39,7 +39,7 @@ main_body:
>> ret void
>> }
>>
>> -;CHECK-LABEL: @gather4_l
>> +;CHECK-LABEL: gather4_l:
>> ;CHECK: IMAGE_GATHER4_L {{v\[[0-9]+:[0-9]+\]}}, 1, 0, 0, -1, 0, 0, 0, 0, {{v\[[0-9]+:[0-9]+\]}}, {{s\[[0-9]+:[0-9]+\]}}, {{s\[[0-9]+:[0-9]+\]}}
>> define void @gather4_l() #0 {
>> main_body:
>> @@ -52,7 +52,7 @@ main_body:
>> ret void
>> }
>>
>> -;CHECK-LABEL: @gather4_b
>> +;CHECK-LABEL: gather4_b:
>> ;CHECK: IMAGE_GATHER4_B {{v\[[0-9]+:[0-9]+\]}}, 1, 0, 0, -1, 0, 0, 0, 0, {{v\[[0-9]+:[0-9]+\]}}, {{s\[[0-9]+:[0-9]+\]}}, {{s\[[0-9]+:[0-9]+\]}}
>> define void @gather4_b() #0 {
>> main_body:
>> @@ -65,7 +65,7 @@ main_body:
>> ret void
>> }
>>
>> -;CHECK-LABEL: @gather4_b_cl
>> +;CHECK-LABEL: gather4_b_cl:
>> ;CHECK: IMAGE_GATHER4_B_CL {{v\[[0-9]+:[0-9]+\]}}, 1, 0, 0, -1, 0, 0, 0, 0, {{v\[[0-9]+:[0-9]+\]}}, {{s\[[0-9]+:[0-9]+\]}}, {{s\[[0-9]+:[0-9]+\]}}
>> define void @gather4_b_cl() #0 {
>> main_body:
>> @@ -78,7 +78,7 @@ main_body:
>> ret void
>> }
>>
>> -;CHECK-LABEL: @gather4_b_cl_v8
>> +;CHECK-LABEL: gather4_b_cl_v8:
>> ;CHECK: IMAGE_GATHER4_B_CL {{v\[[0-9]+:[0-9]+\]}}, 1, 0, 0, -1, 0, 0, 0, 0, {{v\[[0-9]+:[0-9]+\]}}, {{s\[[0-9]+:[0-9]+\]}}, {{s\[[0-9]+:[0-9]+\]}}
>> define void @gather4_b_cl_v8() #0 {
>> main_body:
>> @@ -91,7 +91,7 @@ main_body:
>> ret void
>> }
>>
>> -;CHECK-LABEL: @gather4_lz_v2
>> +;CHECK-LABEL: gather4_lz_v2:
>> ;CHECK: IMAGE_GATHER4_LZ {{v\[[0-9]+:[0-9]+\]}}, 1, 0, 0, -1, 0, 0, 0, 0, {{v\[[0-9]+:[0-9]+\]}}, {{s\[[0-9]+:[0-9]+\]}}, {{s\[[0-9]+:[0-9]+\]}}
>> define void @gather4_lz_v2() #0 {
>> main_body:
>> @@ -104,7 +104,7 @@ main_body:
>> ret void
>> }
>>
>> -;CHECK-LABEL: @gather4_lz
>> +;CHECK-LABEL: gather4_lz:
>> ;CHECK: IMAGE_GATHER4_LZ {{v\[[0-9]+:[0-9]+\]}}, 1, 0, 0, -1, 0, 0, 0, 0, {{v\[[0-9]+:[0-9]+\]}}, {{s\[[0-9]+:[0-9]+\]}}, {{s\[[0-9]+:[0-9]+\]}}
>> define void @gather4_lz() #0 {
>> main_body:
>> @@ -119,7 +119,7 @@ main_body:
>>
>>
>>
>> -;CHECK-LABEL: @gather4_o
>> +;CHECK-LABEL: gather4_o:
>> ;CHECK: IMAGE_GATHER4_O {{v\[[0-9]+:[0-9]+\]}}, 1, 0, 0, -1, 0, 0, 0, 0, {{v\[[0-9]+:[0-9]+\]}}, {{s\[[0-9]+:[0-9]+\]}}, {{s\[[0-9]+:[0-9]+\]}}
>> define void @gather4_o() #0 {
>> main_body:
>> @@ -132,7 +132,7 @@ main_body:
>> ret void
>> }
>>
>> -;CHECK-LABEL: @gather4_cl_o
>> +;CHECK-LABEL: gather4_cl_o:
>> ;CHECK: IMAGE_GATHER4_CL_O {{v\[[0-9]+:[0-9]+\]}}, 1, 0, 0, -1, 0, 0, 0, 0, {{v\[[0-9]+:[0-9]+\]}}, {{s\[[0-9]+:[0-9]+\]}}, {{s\[[0-9]+:[0-9]+\]}}
>> define void @gather4_cl_o() #0 {
>> main_body:
>> @@ -145,7 +145,7 @@ main_body:
>> ret void
>> }
>>
>> -;CHECK-LABEL: @gather4_cl_o_v8
>> +;CHECK-LABEL: gather4_cl_o_v8:
>> ;CHECK: IMAGE_GATHER4_CL_O {{v\[[0-9]+:[0-9]+\]}}, 1, 0, 0, -1, 0, 0, 0, 0, {{v\[[0-9]+:[0-9]+\]}}, {{s\[[0-9]+:[0-9]+\]}}, {{s\[[0-9]+:[0-9]+\]}}
>> define void @gather4_cl_o_v8() #0 {
>> main_body:
>> @@ -158,7 +158,7 @@ main_body:
>> ret void
>> }
>>
>> -;CHECK-LABEL: @gather4_l_o
>> +;CHECK-LABEL: gather4_l_o:
>> ;CHECK: IMAGE_GATHER4_L_O {{v\[[0-9]+:[0-9]+\]}}, 1, 0, 0, -1, 0, 0, 0, 0, {{v\[[0-9]+:[0-9]+\]}}, {{s\[[0-9]+:[0-9]+\]}}, {{s\[[0-9]+:[0-9]+\]}}
>> define void @gather4_l_o() #0 {
>> main_body:
>> @@ -171,7 +171,7 @@ main_body:
>> ret void
>> }
>>
>> -;CHECK-LABEL: @gather4_l_o_v8
>> +;CHECK-LABEL: gather4_l_o_v8:
>> ;CHECK: IMAGE_GATHER4_L_O {{v\[[0-9]+:[0-9]+\]}}, 1, 0, 0, -1, 0, 0, 0, 0, {{v\[[0-9]+:[0-9]+\]}}, {{s\[[0-9]+:[0-9]+\]}}, {{s\[[0-9]+:[0-9]+\]}}
>> define void @gather4_l_o_v8() #0 {
>> main_body:
>> @@ -184,7 +184,7 @@ main_body:
>> ret void
>> }
>>
>> -;CHECK-LABEL: @gather4_b_o
>> +;CHECK-LABEL: gather4_b_o:
>> ;CHECK: IMAGE_GATHER4_B_O {{v\[[0-9]+:[0-9]+\]}}, 1, 0, 0, -1, 0, 0, 0, 0, {{v\[[0-9]+:[0-9]+\]}}, {{s\[[0-9]+:[0-9]+\]}}, {{s\[[0-9]+:[0-9]+\]}}
>> define void @gather4_b_o() #0 {
>> main_body:
>> @@ -197,7 +197,7 @@ main_body:
>> ret void
>> }
>>
>> -;CHECK-LABEL: @gather4_b_o_v8
>> +;CHECK-LABEL: gather4_b_o_v8:
>> ;CHECK: IMAGE_GATHER4_B_O {{v\[[0-9]+:[0-9]+\]}}, 1, 0, 0, -1, 0, 0, 0, 0, {{v\[[0-9]+:[0-9]+\]}}, {{s\[[0-9]+:[0-9]+\]}}, {{s\[[0-9]+:[0-9]+\]}}
>> define void @gather4_b_o_v8() #0 {
>> main_body:
>> @@ -210,7 +210,7 @@ main_body:
>> ret void
>> }
>>
>> -;CHECK-LABEL: @gather4_b_cl_o
>> +;CHECK-LABEL: gather4_b_cl_o:
>> ;CHECK: IMAGE_GATHER4_B_CL_O {{v\[[0-9]+:[0-9]+\]}}, 1, 0, 0, -1, 0, 0, 0, 0, {{v\[[0-9]+:[0-9]+\]}}, {{s\[[0-9]+:[0-9]+\]}}, {{s\[[0-9]+:[0-9]+\]}}
>> define void @gather4_b_cl_o() #0 {
>> main_body:
>> @@ -223,7 +223,7 @@ main_body:
>> ret void
>> }
>>
>> -;CHECK-LABEL: @gather4_lz_o
>> +;CHECK-LABEL: gather4_lz_o:
>> ;CHECK: IMAGE_GATHER4_LZ_O {{v\[[0-9]+:[0-9]+\]}}, 1, 0, 0, -1, 0, 0, 0, 0, {{v\[[0-9]+:[0-9]+\]}}, {{s\[[0-9]+:[0-9]+\]}}, {{s\[[0-9]+:[0-9]+\]}}
>> define void @gather4_lz_o() #0 {
>> main_body:
>> @@ -238,7 +238,7 @@ main_body:
>>
>>
>>
>> -;CHECK-LABEL: @gather4_c
>> +;CHECK-LABEL: gather4_c:
>> ;CHECK: IMAGE_GATHER4_C {{v\[[0-9]+:[0-9]+\]}}, 1, 0, 0, -1, 0, 0, 0, 0, {{v\[[0-9]+:[0-9]+\]}}, {{s\[[0-9]+:[0-9]+\]}}, {{s\[[0-9]+:[0-9]+\]}}
>> define void @gather4_c() #0 {
>> main_body:
>> @@ -251,7 +251,7 @@ main_body:
>> ret void
>> }
>>
>> -;CHECK-LABEL: @gather4_c_cl
>> +;CHECK-LABEL: gather4_c_cl:
>> ;CHECK: IMAGE_GATHER4_C_CL {{v\[[0-9]+:[0-9]+\]}}, 1, 0, 0, -1, 0, 0, 0, 0, {{v\[[0-9]+:[0-9]+\]}}, {{s\[[0-9]+:[0-9]+\]}}, {{s\[[0-9]+:[0-9]+\]}}
>> define void @gather4_c_cl() #0 {
>> main_body:
>> @@ -264,7 +264,7 @@ main_body:
>> ret void
>> }
>>
>> -;CHECK-LABEL: @gather4_c_cl_v8
>> +;CHECK-LABEL: gather4_c_cl_v8:
>> ;CHECK: IMAGE_GATHER4_C_CL {{v\[[0-9]+:[0-9]+\]}}, 1, 0, 0, -1, 0, 0, 0, 0, {{v\[[0-9]+:[0-9]+\]}}, {{s\[[0-9]+:[0-9]+\]}}, {{s\[[0-9]+:[0-9]+\]}}
>> define void @gather4_c_cl_v8() #0 {
>> main_body:
>> @@ -277,7 +277,7 @@ main_body:
>> ret void
>> }
>>
>> -;CHECK-LABEL: @gather4_c_l
>> +;CHECK-LABEL: gather4_c_l:
>> ;CHECK: IMAGE_GATHER4_C_L {{v\[[0-9]+:[0-9]+\]}}, 1, 0, 0, -1, 0, 0, 0, 0, {{v\[[0-9]+:[0-9]+\]}}, {{s\[[0-9]+:[0-9]+\]}}, {{s\[[0-9]+:[0-9]+\]}}
>> define void @gather4_c_l() #0 {
>> main_body:
>> @@ -290,7 +290,7 @@ main_body:
>> ret void
>> }
>>
>> -;CHECK-LABEL: @gather4_c_l_v8
>> +;CHECK-LABEL: gather4_c_l_v8:
>> ;CHECK: IMAGE_GATHER4_C_L {{v\[[0-9]+:[0-9]+\]}}, 1, 0, 0, -1, 0, 0, 0, 0, {{v\[[0-9]+:[0-9]+\]}}, {{s\[[0-9]+:[0-9]+\]}}, {{s\[[0-9]+:[0-9]+\]}}
>> define void @gather4_c_l_v8() #0 {
>> main_body:
>> @@ -303,7 +303,7 @@ main_body:
>> ret void
>> }
>>
>> -;CHECK-LABEL: @gather4_c_b
>> +;CHECK-LABEL: gather4_c_b:
>> ;CHECK: IMAGE_GATHER4_C_B {{v\[[0-9]+:[0-9]+\]}}, 1, 0, 0, -1, 0, 0, 0, 0, {{v\[[0-9]+:[0-9]+\]}}, {{s\[[0-9]+:[0-9]+\]}}, {{s\[[0-9]+:[0-9]+\]}}
>> define void @gather4_c_b() #0 {
>> main_body:
>> @@ -316,7 +316,7 @@ main_body:
>> ret void
>> }
>>
>> -;CHECK-LABEL: @gather4_c_b_v8
>> +;CHECK-LABEL: gather4_c_b_v8:
>> ;CHECK: IMAGE_GATHER4_C_B {{v\[[0-9]+:[0-9]+\]}}, 1, 0, 0, -1, 0, 0, 0, 0, {{v\[[0-9]+:[0-9]+\]}}, {{s\[[0-9]+:[0-9]+\]}}, {{s\[[0-9]+:[0-9]+\]}}
>> define void @gather4_c_b_v8() #0 {
>> main_body:
>> @@ -329,7 +329,7 @@ main_body:
>> ret void
>> }
>>
>> -;CHECK-LABEL: @gather4_c_b_cl
>> +;CHECK-LABEL: gather4_c_b_cl:
>> ;CHECK: IMAGE_GATHER4_C_B_CL {{v\[[0-9]+:[0-9]+\]}}, 1, 0, 0, -1, 0, 0, 0, 0, {{v\[[0-9]+:[0-9]+\]}}, {{s\[[0-9]+:[0-9]+\]}}, {{s\[[0-9]+:[0-9]+\]}}
>> define void @gather4_c_b_cl() #0 {
>> main_body:
>> @@ -342,7 +342,7 @@ main_body:
>> ret void
>> }
>>
>> -;CHECK-LABEL: @gather4_c_lz
>> +;CHECK-LABEL: gather4_c_lz:
>> ;CHECK: IMAGE_GATHER4_C_LZ {{v\[[0-9]+:[0-9]+\]}}, 1, 0, 0, -1, 0, 0, 0, 0, {{v\[[0-9]+:[0-9]+\]}}, {{s\[[0-9]+:[0-9]+\]}}, {{s\[[0-9]+:[0-9]+\]}}
>> define void @gather4_c_lz() #0 {
>> main_body:
>> @@ -357,7 +357,7 @@ main_body:
>>
>>
>>
>> -;CHECK-LABEL: @gather4_c_o
>> +;CHECK-LABEL: gather4_c_o:
>> ;CHECK: IMAGE_GATHER4_C_O {{v\[[0-9]+:[0-9]+\]}}, 1, 0, 0, -1, 0, 0, 0, 0, {{v\[[0-9]+:[0-9]+\]}}, {{s\[[0-9]+:[0-9]+\]}}, {{s\[[0-9]+:[0-9]+\]}}
>> define void @gather4_c_o() #0 {
>> main_body:
>> @@ -370,7 +370,7 @@ main_body:
>> ret void
>> }
>>
>> -;CHECK-LABEL: @gather4_c_o_v8
>> +;CHECK-LABEL: gather4_c_o_v8:
>> ;CHECK: IMAGE_GATHER4_C_O {{v\[[0-9]+:[0-9]+\]}}, 1, 0, 0, -1, 0, 0, 0, 0, {{v\[[0-9]+:[0-9]+\]}}, {{s\[[0-9]+:[0-9]+\]}}, {{s\[[0-9]+:[0-9]+\]}}
>> define void @gather4_c_o_v8() #0 {
>> main_body:
>> @@ -383,7 +383,7 @@ main_body:
>> ret void
>> }
>>
>> -;CHECK-LABEL: @gather4_c_cl_o
>> +;CHECK-LABEL: gather4_c_cl_o:
>> ;CHECK: IMAGE_GATHER4_C_CL_O {{v\[[0-9]+:[0-9]+\]}}, 1, 0, 0, -1, 0, 0, 0, 0, {{v\[[0-9]+:[0-9]+\]}}, {{s\[[0-9]+:[0-9]+\]}}, {{s\[[0-9]+:[0-9]+\]}}
>> define void @gather4_c_cl_o() #0 {
>> main_body:
>> @@ -396,7 +396,7 @@ main_body:
>> ret void
>> }
>>
>> -;CHECK-LABEL: @gather4_c_l_o
>> +;CHECK-LABEL: gather4_c_l_o:
>> ;CHECK: IMAGE_GATHER4_C_L_O {{v\[[0-9]+:[0-9]+\]}}, 1, 0, 0, -1, 0, 0, 0, 0, {{v\[[0-9]+:[0-9]+\]}}, {{s\[[0-9]+:[0-9]+\]}}, {{s\[[0-9]+:[0-9]+\]}}
>> define void @gather4_c_l_o() #0 {
>> main_body:
>> @@ -409,7 +409,7 @@ main_body:
>> ret void
>> }
>>
>> -;CHECK-LABEL: @gather4_c_b_o
>> +;CHECK-LABEL: gather4_c_b_o:
>> ;CHECK: IMAGE_GATHER4_C_B_O {{v\[[0-9]+:[0-9]+\]}}, 1, 0, 0, -1, 0, 0, 0, 0, {{v\[[0-9]+:[0-9]+\]}}, {{s\[[0-9]+:[0-9]+\]}}, {{s\[[0-9]+:[0-9]+\]}}
>> define void @gather4_c_b_o() #0 {
>> main_body:
>> @@ -422,7 +422,7 @@ main_body:
>> ret void
>> }
>>
>> -;CHECK-LABEL: @gather4_c_b_cl_o
>> +;CHECK-LABEL: gather4_c_b_cl_o:
>> ;CHECK: IMAGE_GATHER4_C_B_CL_O {{v\[[0-9]+:[0-9]+\]}}, 1, 0, 0, -1, 0, 0, 0, 0, {{v\[[0-9]+:[0-9]+\]}}, {{s\[[0-9]+:[0-9]+\]}}, {{s\[[0-9]+:[0-9]+\]}}
>> define void @gather4_c_b_cl_o() #0 {
>> main_body:
>> @@ -435,7 +435,7 @@ main_body:
>> ret void
>> }
>>
>> -;CHECK-LABEL: @gather4_c_lz_o
>> +;CHECK-LABEL: gather4_c_lz_o:
>> ;CHECK: IMAGE_GATHER4_C_LZ_O {{v\[[0-9]+:[0-9]+\]}}, 1, 0, 0, -1, 0, 0, 0, 0, {{v\[[0-9]+:[0-9]+\]}}, {{s\[[0-9]+:[0-9]+\]}}, {{s\[[0-9]+:[0-9]+\]}}
>> define void @gather4_c_lz_o() #0 {
>> main_body:
>> @@ -448,7 +448,7 @@ main_body:
>> ret void
>> }
>>
>> -;CHECK-LABEL: @gather4_c_lz_o_v8
>> +;CHECK-LABEL: gather4_c_lz_o_v8:
>> ;CHECK: IMAGE_GATHER4_C_LZ_O {{v\[[0-9]+:[0-9]+\]}}, 1, 0, 0, -1, 0, 0, 0, 0, {{v\[[0-9]+:[0-9]+\]}}, {{s\[[0-9]+:[0-9]+\]}}, {{s\[[0-9]+:[0-9]+\]}}
>> define void @gather4_c_lz_o_v8() #0 {
>> main_body:
>> diff --git a/test/CodeGen/R600/llvm.SI.getlod.ll b/test/CodeGen/R600/llvm.SI.getlod.ll
>> index a7a17ec..4cd704c 100644
>> --- a/test/CodeGen/R600/llvm.SI.getlod.ll
>> +++ b/test/CodeGen/R600/llvm.SI.getlod.ll
>> @@ -1,6 +1,6 @@
>> ;RUN: llc < %s -march=r600 -mcpu=verde -verify-machineinstrs | FileCheck %s
>>
>> -;CHECK-LABEL: @getlod
>> +;CHECK-LABEL: getlod
>> ;CHECK: IMAGE_GET_LOD {{v\[[0-9]+:[0-9]+\]}}, 3, 0, 0, -1, 0, 0, 0, 0, {{v[0-9]+}}, {{s\[[0-9]+:[0-9]+\]}}, {{s\[[0-9]+:[0-9]+\]}}
>> define void @getlod() #0 {
>> main_body:
>> @@ -11,7 +11,7 @@ main_body:
>> ret void
>> }
>>
>> -;CHECK-LABEL: @getlod_v2
>> +;CHECK-LABEL: getlod_v2
>> ;CHECK: IMAGE_GET_LOD {{v\[[0-9]+:[0-9]+\]}}, 3, 0, 0, -1, 0, 0, 0, 0, {{v\[[0-9]+:[0-9]+\]}}, {{s\[[0-9]+:[0-9]+\]}}, {{s\[[0-9]+:[0-9]+\]}}
>> define void @getlod_v2() #0 {
>> main_body:
>> @@ -22,7 +22,7 @@ main_body:
>> ret void
>> }
>>
>> -;CHECK-LABEL: @getlod_v4
>> +;CHECK-LABEL: getlod_v4
>> ;CHECK: IMAGE_GET_LOD {{v\[[0-9]+:[0-9]+\]}}, 3, 0, 0, -1, 0, 0, 0, 0, {{v\[[0-9]+:[0-9]+\]}}, {{s\[[0-9]+:[0-9]+\]}}, {{s\[[0-9]+:[0-9]+\]}}
>> define void @getlod_v4() #0 {
>> main_body:
>> diff --git a/test/CodeGen/R600/llvm.SI.image.ll b/test/CodeGen/R600/llvm.SI.image.ll
>> index eac0b8e..7a7e2da 100644
>> --- a/test/CodeGen/R600/llvm.SI.image.ll
>> +++ b/test/CodeGen/R600/llvm.SI.image.ll
>> @@ -1,6 +1,6 @@
>> ;RUN: llc < %s -march=r600 -mcpu=verde -verify-machineinstrs | FileCheck %s
>>
>> -;CHECK-LABEL: @image_load
>> +;CHECK-LABEL: image_load
>> ;CHECK: IMAGE_LOAD {{v\[[0-9]+:[0-9]+\]}}, 15, 0, 0, 0, 0, 0, 0, 0, {{v\[[0-9]+:[0-9]+\]}}, {{s\[[0-9]+:[0-9]+\]}}
>> define void @image_load() #0 {
>> main_body:
>> @@ -13,7 +13,7 @@ main_body:
>> ret void
>> }
>>
>> -;CHECK-LABEL: @image_load_mip
>> +;CHECK-LABEL: image_load_mip
>> ;CHECK: IMAGE_LOAD_MIP {{v\[[0-9]+:[0-9]+\]}}, 15, 0, 0, 0, 0, 0, 0, 0, {{v\[[0-9]+:[0-9]+\]}}, {{s\[[0-9]+:[0-9]+\]}}
>> define void @image_load_mip() #0 {
>> main_body:
>> @@ -26,7 +26,7 @@ main_body:
>> ret void
>> }
>>
>> -;CHECK-LABEL: @getresinfo
>> +;CHECK-LABEL: getresinfo
>> ;CHECK: IMAGE_GET_RESINFO {{v\[[0-9]+:[0-9]+\]}}, 15, 0, 0, 0, 0, 0, 0, 0, {{v[0-9]+}}, {{s\[[0-9]+:[0-9]+\]}}
>> define void @getresinfo() #0 {
>> main_body:
>> diff --git a/test/CodeGen/R600/llvm.SI.image.sample.ll b/test/CodeGen/R600/llvm.SI.image.sample.ll
>> index 14dff7e..2b2d175 100644
>> --- a/test/CodeGen/R600/llvm.SI.image.sample.ll
>> +++ b/test/CodeGen/R600/llvm.SI.image.sample.ll
>> @@ -1,6 +1,6 @@
>> ;RUN: llc < %s -march=r600 -mcpu=verde -verify-machineinstrs | FileCheck %s
>>
>> -;CHECK-LABEL: @sample
>> +;CHECK-LABEL: sample:
>> ;CHECK: IMAGE_SAMPLE {{v\[[0-9]+:[0-9]+\]}}, 15, 0, 0, 0, 0, 0, 0, 0, {{v\[[0-9]+:[0-9]+\]}}, {{s\[[0-9]+:[0-9]+\]}}, {{s\[[0-9]+:[0-9]+\]}}
>> define void @sample() #0 {
>> main_body:
>> @@ -13,7 +13,7 @@ main_body:
>> ret void
>> }
>>
>> -;CHECK-LABEL: @sample_cl
>> +;CHECK-LABEL: sample_cl:
>> ;CHECK: IMAGE_SAMPLE_CL {{v\[[0-9]+:[0-9]+\]}}, 15, 0, 0, 0, 0, 0, 0, 0, {{v\[[0-9]+:[0-9]+\]}}, {{s\[[0-9]+:[0-9]+\]}}, {{s\[[0-9]+:[0-9]+\]}}
>> define void @sample_cl() #0 {
>> main_body:
>> @@ -26,7 +26,7 @@ main_body:
>> ret void
>> }
>>
>> -;CHECK-LABEL: @sample_d
>> +;CHECK-LABEL: sample_d:
>> ;CHECK: IMAGE_SAMPLE_D {{v\[[0-9]+:[0-9]+\]}}, 15, 0, 0, 0, 0, 0, 0, 0, {{v\[[0-9]+:[0-9]+\]}}, {{s\[[0-9]+:[0-9]+\]}}, {{s\[[0-9]+:[0-9]+\]}}
>> define void @sample_d() #0 {
>> main_body:
>> @@ -39,7 +39,7 @@ main_body:
>> ret void
>> }
>>
>> -;CHECK-LABEL: @sample_d_cl
>> +;CHECK-LABEL: sample_d_cl:
>> ;CHECK: IMAGE_SAMPLE_D_CL {{v\[[0-9]+:[0-9]+\]}}, 15, 0, 0, 0, 0, 0, 0, 0, {{v\[[0-9]+:[0-9]+\]}}, {{s\[[0-9]+:[0-9]+\]}}, {{s\[[0-9]+:[0-9]+\]}}
>> define void @sample_d_cl() #0 {
>> main_body:
>> @@ -52,7 +52,7 @@ main_body:
>> ret void
>> }
>>
>> -;CHECK-LABEL: @sample_l
>> +;CHECK-LABEL: sample_l:
>> ;CHECK: IMAGE_SAMPLE_L {{v\[[0-9]+:[0-9]+\]}}, 15, 0, 0, 0, 0, 0, 0, 0, {{v\[[0-9]+:[0-9]+\]}}, {{s\[[0-9]+:[0-9]+\]}}, {{s\[[0-9]+:[0-9]+\]}}
>> define void @sample_l() #0 {
>> main_body:
>> @@ -65,7 +65,7 @@ main_body:
>> ret void
>> }
>>
>> -;CHECK-LABEL: @sample_b
>> +;CHECK-LABEL: sample_b:
>> ;CHECK: IMAGE_SAMPLE_B {{v\[[0-9]+:[0-9]+\]}}, 15, 0, 0, 0, 0, 0, 0, 0, {{v\[[0-9]+:[0-9]+\]}}, {{s\[[0-9]+:[0-9]+\]}}, {{s\[[0-9]+:[0-9]+\]}}
>> define void @sample_b() #0 {
>> main_body:
>> @@ -78,7 +78,7 @@ main_body:
>> ret void
>> }
>>
>> -;CHECK-LABEL: @sample_b_cl
>> +;CHECK-LABEL: sample_b_cl:
>> ;CHECK: IMAGE_SAMPLE_B_CL {{v\[[0-9]+:[0-9]+\]}}, 15, 0, 0, 0, 0, 0, 0, 0, {{v\[[0-9]+:[0-9]+\]}}, {{s\[[0-9]+:[0-9]+\]}}, {{s\[[0-9]+:[0-9]+\]}}
>> define void @sample_b_cl() #0 {
>> main_body:
>> @@ -91,7 +91,7 @@ main_body:
>> ret void
>> }
>>
>> -;CHECK-LABEL: @sample_lz
>> +;CHECK-LABEL: sample_lz:
>> ;CHECK: IMAGE_SAMPLE_LZ {{v\[[0-9]+:[0-9]+\]}}, 15, 0, 0, 0, 0, 0, 0, 0, {{v\[[0-9]+:[0-9]+\]}}, {{s\[[0-9]+:[0-9]+\]}}, {{s\[[0-9]+:[0-9]+\]}}
>> define void @sample_lz() #0 {
>> main_body:
>> @@ -104,7 +104,7 @@ main_body:
>> ret void
>> }
>>
>> -;CHECK-LABEL: @sample_cd
>> +;CHECK-LABEL: sample_cd:
>> ;CHECK: IMAGE_SAMPLE_CD {{v\[[0-9]+:[0-9]+\]}}, 15, 0, 0, 0, 0, 0, 0, 0, {{v\[[0-9]+:[0-9]+\]}}, {{s\[[0-9]+:[0-9]+\]}}, {{s\[[0-9]+:[0-9]+\]}}
>> define void @sample_cd() #0 {
>> main_body:
>> @@ -117,7 +117,7 @@ main_body:
>> ret void
>> }
>>
>> -;CHECK-LABEL: @sample_cd_cl
>> +;CHECK-LABEL: sample_cd_cl:
>> ;CHECK: IMAGE_SAMPLE_CD_CL {{v\[[0-9]+:[0-9]+\]}}, 15, 0, 0, 0, 0, 0, 0, 0, {{v\[[0-9]+:[0-9]+\]}}, {{s\[[0-9]+:[0-9]+\]}}, {{s\[[0-9]+:[0-9]+\]}}
>> define void @sample_cd_cl() #0 {
>> main_body:
>> @@ -130,7 +130,7 @@ main_body:
>> ret void
>> }
>>
>> -;CHECK-LABEL: @sample_c
>> +;CHECK-LABEL: sample_c:
>> ;CHECK: IMAGE_SAMPLE_C {{v\[[0-9]+:[0-9]+\]}}, 15, 0, 0, 0, 0, 0, 0, 0, {{v\[[0-9]+:[0-9]+\]}}, {{s\[[0-9]+:[0-9]+\]}}, {{s\[[0-9]+:[0-9]+\]}}
>> define void @sample_c() #0 {
>> main_body:
>> @@ -143,7 +143,7 @@ main_body:
>> ret void
>> }
>>
>> -;CHECK-LABEL: @sample_c_cl
>> +;CHECK-LABEL: sample_c_cl:
>> ;CHECK: IMAGE_SAMPLE_C_CL {{v\[[0-9]+:[0-9]+\]}}, 15, 0, 0, 0, 0, 0, 0, 0, {{v\[[0-9]+:[0-9]+\]}}, {{s\[[0-9]+:[0-9]+\]}}, {{s\[[0-9]+:[0-9]+\]}}
>> define void @sample_c_cl() #0 {
>> main_body:
>> @@ -156,7 +156,7 @@ main_body:
>> ret void
>> }
>>
>> -;CHECK-LABEL: @sample_c_d
>> +;CHECK-LABEL: sample_c_d:
>> ;CHECK: IMAGE_SAMPLE_C_D {{v\[[0-9]+:[0-9]+\]}}, 15, 0, 0, 0, 0, 0, 0, 0, {{v\[[0-9]+:[0-9]+\]}}, {{s\[[0-9]+:[0-9]+\]}}, {{s\[[0-9]+:[0-9]+\]}}
>> define void @sample_c_d() #0 {
>> main_body:
>> @@ -169,7 +169,7 @@ main_body:
>> ret void
>> }
>>
>> -;CHECK-LABEL: @sample_c_d_cl
>> +;CHECK-LABEL: sample_c_d_cl:
>> ;CHECK: IMAGE_SAMPLE_C_D_CL {{v\[[0-9]+:[0-9]+\]}}, 15, 0, 0, 0, 0, 0, 0, 0, {{v\[[0-9]+:[0-9]+\]}}, {{s\[[0-9]+:[0-9]+\]}}, {{s\[[0-9]+:[0-9]+\]}}
>> define void @sample_c_d_cl() #0 {
>> main_body:
>> @@ -182,7 +182,7 @@ main_body:
>> ret void
>> }
>>
>> -;CHECK-LABEL: @sample_c_l
>> +;CHECK-LABEL: sample_c_l:
>> ;CHECK: IMAGE_SAMPLE_C_L {{v\[[0-9]+:[0-9]+\]}}, 15, 0, 0, 0, 0, 0, 0, 0, {{v\[[0-9]+:[0-9]+\]}}, {{s\[[0-9]+:[0-9]+\]}}, {{s\[[0-9]+:[0-9]+\]}}
>> define void @sample_c_l() #0 {
>> main_body:
>> @@ -195,7 +195,7 @@ main_body:
>> ret void
>> }
>>
>> -;CHECK-LABEL: @sample_c_b
>> +;CHECK-LABEL: sample_c_b:
>> ;CHECK: IMAGE_SAMPLE_C_B {{v\[[0-9]+:[0-9]+\]}}, 15, 0, 0, 0, 0, 0, 0, 0, {{v\[[0-9]+:[0-9]+\]}}, {{s\[[0-9]+:[0-9]+\]}}, {{s\[[0-9]+:[0-9]+\]}}
>> define void @sample_c_b() #0 {
>> main_body:
>> @@ -208,7 +208,7 @@ main_body:
>> ret void
>> }
>>
>> -;CHECK-LABEL: @sample_c_b_cl
>> +;CHECK-LABEL: sample_c_b_cl:
>> ;CHECK: IMAGE_SAMPLE_C_B_CL {{v\[[0-9]+:[0-9]+\]}}, 15, 0, 0, 0, 0, 0, 0, 0, {{v\[[0-9]+:[0-9]+\]}}, {{s\[[0-9]+:[0-9]+\]}}, {{s\[[0-9]+:[0-9]+\]}}
>> define void @sample_c_b_cl() #0 {
>> main_body:
>> @@ -221,7 +221,7 @@ main_body:
>> ret void
>> }
>>
>> -;CHECK-LABEL: @sample_c_lz
>> +;CHECK-LABEL: sample_c_lz:
>> ;CHECK: IMAGE_SAMPLE_C_LZ {{v\[[0-9]+:[0-9]+\]}}, 15, 0, 0, 0, 0, 0, 0, 0, {{v\[[0-9]+:[0-9]+\]}}, {{s\[[0-9]+:[0-9]+\]}}, {{s\[[0-9]+:[0-9]+\]}}
>> define void @sample_c_lz() #0 {
>> main_body:
>> @@ -234,7 +234,7 @@ main_body:
>> ret void
>> }
>>
>> -;CHECK-LABEL: @sample_c_cd
>> +;CHECK-LABEL: sample_c_cd:
>> ;CHECK: IMAGE_SAMPLE_C_CD {{v\[[0-9]+:[0-9]+\]}}, 15, 0, 0, 0, 0, 0, 0, 0, {{v\[[0-9]+:[0-9]+\]}}, {{s\[[0-9]+:[0-9]+\]}}, {{s\[[0-9]+:[0-9]+\]}}
>> define void @sample_c_cd() #0 {
>> main_body:
>> @@ -247,7 +247,7 @@ main_body:
>> ret void
>> }
>>
>> -;CHECK-LABEL: @sample_c_cd_cl
>> +;CHECK-LABEL: sample_c_cd_cl:
>> ;CHECK: IMAGE_SAMPLE_C_CD_CL {{v\[[0-9]+:[0-9]+\]}}, 15, 0, 0, 0, 0, 0, 0, 0, {{v\[[0-9]+:[0-9]+\]}}, {{s\[[0-9]+:[0-9]+\]}}, {{s\[[0-9]+:[0-9]+\]}}
>> define void @sample_c_cd_cl() #0 {
>> main_body:
>> diff --git a/test/CodeGen/R600/llvm.SI.image.sample.o.ll b/test/CodeGen/R600/llvm.SI.image.sample.o.ll
>> index ed3ef91..c4234bb 100644
>> --- a/test/CodeGen/R600/llvm.SI.image.sample.o.ll
>> +++ b/test/CodeGen/R600/llvm.SI.image.sample.o.ll
>> @@ -1,6 +1,6 @@
>> ;RUN: llc < %s -march=r600 -mcpu=verde -verify-machineinstrs | FileCheck %s
>>
>> -;CHECK-LABEL: @sample
>> +;CHECK-LABEL: sample:
>> ;CHECK: IMAGE_SAMPLE_O {{v\[[0-9]+:[0-9]+\]}}, 15, 0, 0, 0, 0, 0, 0, 0, {{v\[[0-9]+:[0-9]+\]}}, {{s\[[0-9]+:[0-9]+\]}}, {{s\[[0-9]+:[0-9]+\]}}
>> define void @sample() #0 {
>> main_body:
>> @@ -13,7 +13,7 @@ main_body:
>> ret void
>> }
>>
>> -;CHECK-LABEL: @sample_cl
>> +;CHECK-LABEL: sample_cl:
>> ;CHECK: IMAGE_SAMPLE_CL_O {{v\[[0-9]+:[0-9]+\]}}, 15, 0, 0, 0, 0, 0, 0, 0, {{v\[[0-9]+:[0-9]+\]}}, {{s\[[0-9]+:[0-9]+\]}}, {{s\[[0-9]+:[0-9]+\]}}
>> define void @sample_cl() #0 {
>> main_body:
>> @@ -26,7 +26,7 @@ main_body:
>> ret void
>> }
>>
>> -;CHECK-LABEL: @sample_d
>> +;CHECK-LABEL: sample_d:
>> ;CHECK: IMAGE_SAMPLE_D_O {{v\[[0-9]+:[0-9]+\]}}, 15, 0, 0, 0, 0, 0, 0, 0, {{v\[[0-9]+:[0-9]+\]}}, {{s\[[0-9]+:[0-9]+\]}}, {{s\[[0-9]+:[0-9]+\]}}
>> define void @sample_d() #0 {
>> main_body:
>> @@ -39,7 +39,7 @@ main_body:
>> ret void
>> }
>>
>> -;CHECK-LABEL: @sample_d_cl
>> +;CHECK-LABEL: sample_d_cl:
>> ;CHECK: IMAGE_SAMPLE_D_CL_O {{v\[[0-9]+:[0-9]+\]}}, 15, 0, 0, 0, 0, 0, 0, 0, {{v\[[0-9]+:[0-9]+\]}}, {{s\[[0-9]+:[0-9]+\]}}, {{s\[[0-9]+:[0-9]+\]}}
>> define void @sample_d_cl() #0 {
>> main_body:
>> @@ -52,7 +52,7 @@ main_body:
>> ret void
>> }
>>
>> -;CHECK-LABEL: @sample_l
>> +;CHECK-LABEL: sample_l:
>> ;CHECK: IMAGE_SAMPLE_L_O {{v\[[0-9]+:[0-9]+\]}}, 15, 0, 0, 0, 0, 0, 0, 0, {{v\[[0-9]+:[0-9]+\]}}, {{s\[[0-9]+:[0-9]+\]}}, {{s\[[0-9]+:[0-9]+\]}}
>> define void @sample_l() #0 {
>> main_body:
>> @@ -65,7 +65,7 @@ main_body:
>> ret void
>> }
>>
>> -;CHECK-LABEL: @sample_b
>> +;CHECK-LABEL: sample_b:
>> ;CHECK: IMAGE_SAMPLE_B_O {{v\[[0-9]+:[0-9]+\]}}, 15, 0, 0, 0, 0, 0, 0, 0, {{v\[[0-9]+:[0-9]+\]}}, {{s\[[0-9]+:[0-9]+\]}}, {{s\[[0-9]+:[0-9]+\]}}
>> define void @sample_b() #0 {
>> main_body:
>> @@ -78,7 +78,7 @@ main_body:
>> ret void
>> }
>>
>> -;CHECK-LABEL: @sample_b_cl
>> +;CHECK-LABEL: sample_b_cl:
>> ;CHECK: IMAGE_SAMPLE_B_CL_O {{v\[[0-9]+:[0-9]+\]}}, 15, 0, 0, 0, 0, 0, 0, 0, {{v\[[0-9]+:[0-9]+\]}}, {{s\[[0-9]+:[0-9]+\]}}, {{s\[[0-9]+:[0-9]+\]}}
>> define void @sample_b_cl() #0 {
>> main_body:
>> @@ -91,7 +91,7 @@ main_body:
>> ret void
>> }
>>
>> -;CHECK-LABEL: @sample_lz
>> +;CHECK-LABEL: sample_lz:
>> ;CHECK: IMAGE_SAMPLE_LZ_O {{v\[[0-9]+:[0-9]+\]}}, 15, 0, 0, 0, 0, 0, 0, 0, {{v\[[0-9]+:[0-9]+\]}}, {{s\[[0-9]+:[0-9]+\]}}, {{s\[[0-9]+:[0-9]+\]}}
>> define void @sample_lz() #0 {
>> main_body:
>> @@ -104,7 +104,7 @@ main_body:
>> ret void
>> }
>>
>> -;CHECK-LABEL: @sample_cd
>> +;CHECK-LABEL: sample_cd:
>> ;CHECK: IMAGE_SAMPLE_CD_O {{v\[[0-9]+:[0-9]+\]}}, 15, 0, 0, 0, 0, 0, 0, 0, {{v\[[0-9]+:[0-9]+\]}}, {{s\[[0-9]+:[0-9]+\]}}, {{s\[[0-9]+:[0-9]+\]}}
>> define void @sample_cd() #0 {
>> main_body:
>> @@ -117,7 +117,7 @@ main_body:
>> ret void
>> }
>>
>> -;CHECK-LABEL: @sample_cd_cl
>> +;CHECK-LABEL: sample_cd_cl:
>> ;CHECK: IMAGE_SAMPLE_CD_CL_O {{v\[[0-9]+:[0-9]+\]}}, 15, 0, 0, 0, 0, 0, 0, 0, {{v\[[0-9]+:[0-9]+\]}}, {{s\[[0-9]+:[0-9]+\]}}, {{s\[[0-9]+:[0-9]+\]}}
>> define void @sample_cd_cl() #0 {
>> main_body:
>> @@ -130,7 +130,7 @@ main_body:
>> ret void
>> }
>>
>> -;CHECK-LABEL: @sample_c
>> +;CHECK-LABEL: sample_c:
>> ;CHECK: IMAGE_SAMPLE_C_O {{v\[[0-9]+:[0-9]+\]}}, 15, 0, 0, 0, 0, 0, 0, 0, {{v\[[0-9]+:[0-9]+\]}}, {{s\[[0-9]+:[0-9]+\]}}, {{s\[[0-9]+:[0-9]+\]}}
>> define void @sample_c() #0 {
>> main_body:
>> @@ -143,7 +143,7 @@ main_body:
>> ret void
>> }
>>
>> -;CHECK-LABEL: @sample_c_cl
>> +;CHECK-LABEL: sample_c_cl:
>> ;CHECK: IMAGE_SAMPLE_C_CL_O {{v\[[0-9]+:[0-9]+\]}}, 15, 0, 0, 0, 0, 0, 0, 0, {{v\[[0-9]+:[0-9]+\]}}, {{s\[[0-9]+:[0-9]+\]}}, {{s\[[0-9]+:[0-9]+\]}}
>> define void @sample_c_cl() #0 {
>> main_body:
>> @@ -156,7 +156,7 @@ main_body:
>> ret void
>> }
>>
>> -;CHECK-LABEL: @sample_c_d
>> +;CHECK-LABEL: sample_c_d:
>> ;CHECK: IMAGE_SAMPLE_C_D_O {{v\[[0-9]+:[0-9]+\]}}, 15, 0, 0, 0, 0, 0, 0, 0, {{v\[[0-9]+:[0-9]+\]}}, {{s\[[0-9]+:[0-9]+\]}}, {{s\[[0-9]+:[0-9]+\]}}
>> define void @sample_c_d() #0 {
>> main_body:
>> @@ -169,7 +169,7 @@ main_body:
>> ret void
>> }
>>
>> -;CHECK-LABEL: @sample_c_d_cl
>> +;CHECK-LABEL: sample_c_d_cl:
>> ;CHECK: IMAGE_SAMPLE_C_D_CL_O {{v\[[0-9]+:[0-9]+\]}}, 15, 0, 0, 0, 0, 0, 0, 0, {{v\[[0-9]+:[0-9]+\]}}, {{s\[[0-9]+:[0-9]+\]}}, {{s\[[0-9]+:[0-9]+\]}}
>> define void @sample_c_d_cl() #0 {
>> main_body:
>> @@ -182,7 +182,7 @@ main_body:
>> ret void
>> }
>>
>> -;CHECK-LABEL: @sample_c_l
>> +;CHECK-LABEL: sample_c_l:
>> ;CHECK: IMAGE_SAMPLE_C_L_O {{v\[[0-9]+:[0-9]+\]}}, 15, 0, 0, 0, 0, 0, 0, 0, {{v\[[0-9]+:[0-9]+\]}}, {{s\[[0-9]+:[0-9]+\]}}, {{s\[[0-9]+:[0-9]+\]}}
>> define void @sample_c_l() #0 {
>> main_body:
>> @@ -195,7 +195,7 @@ main_body:
>> ret void
>> }
>>
>> -;CHECK-LABEL: @sample_c_b
>> +;CHECK-LABEL: sample_c_b:
>> ;CHECK: IMAGE_SAMPLE_C_B_O {{v\[[0-9]+:[0-9]+\]}}, 15, 0, 0, 0, 0, 0, 0, 0, {{v\[[0-9]+:[0-9]+\]}}, {{s\[[0-9]+:[0-9]+\]}}, {{s\[[0-9]+:[0-9]+\]}}
>> define void @sample_c_b() #0 {
>> main_body:
>> @@ -208,7 +208,7 @@ main_body:
>> ret void
>> }
>>
>> -;CHECK-LABEL: @sample_c_b_cl
>> +;CHECK-LABEL: sample_c_b_cl:
>> ;CHECK: IMAGE_SAMPLE_C_B_CL_O {{v\[[0-9]+:[0-9]+\]}}, 15, 0, 0, 0, 0, 0, 0, 0, {{v\[[0-9]+:[0-9]+\]}}, {{s\[[0-9]+:[0-9]+\]}}, {{s\[[0-9]+:[0-9]+\]}}
>> define void @sample_c_b_cl() #0 {
>> main_body:
>> @@ -221,7 +221,7 @@ main_body:
>> ret void
>> }
>>
>> -;CHECK-LABEL: @sample_c_lz
>> +;CHECK-LABEL: sample_c_lz:
>> ;CHECK: IMAGE_SAMPLE_C_LZ_O {{v\[[0-9]+:[0-9]+\]}}, 15, 0, 0, 0, 0, 0, 0, 0, {{v\[[0-9]+:[0-9]+\]}}, {{s\[[0-9]+:[0-9]+\]}}, {{s\[[0-9]+:[0-9]+\]}}
>> define void @sample_c_lz() #0 {
>> main_body:
>> @@ -234,7 +234,7 @@ main_body:
>> ret void
>> }
>>
>> -;CHECK-LABEL: @sample_c_cd
>> +;CHECK-LABEL: sample_c_cd:
>> ;CHECK: IMAGE_SAMPLE_C_CD_O {{v\[[0-9]+:[0-9]+\]}}, 15, 0, 0, 0, 0, 0, 0, 0, {{v\[[0-9]+:[0-9]+\]}}, {{s\[[0-9]+:[0-9]+\]}}, {{s\[[0-9]+:[0-9]+\]}}
>> define void @sample_c_cd() #0 {
>> main_body:
>> @@ -247,7 +247,7 @@ main_body:
>> ret void
>> }
>>
>> -;CHECK-LABEL: @sample_c_cd_cl
>> +;CHECK-LABEL: sample_c_cd_cl:
>> ;CHECK: IMAGE_SAMPLE_C_CD_CL_O {{v\[[0-9]+:[0-9]+\]}}, 15, 0, 0, 0, 0, 0, 0, 0, {{v\[[0-9]+:[0-9]+\]}}, {{s\[[0-9]+:[0-9]+\]}}, {{s\[[0-9]+:[0-9]+\]}}
>> define void @sample_c_cd_cl() #0 {
>> main_body:
>> diff --git a/test/CodeGen/R600/llvm.SI.load.dword.ll b/test/CodeGen/R600/llvm.SI.load.dword.ll
>> index a622775..518ca20 100644
>> --- a/test/CodeGen/R600/llvm.SI.load.dword.ll
>> +++ b/test/CodeGen/R600/llvm.SI.load.dword.ll
>> @@ -3,7 +3,7 @@
>> ; Example of a simple geometry shader loading vertex attributes from the
>> ; ESGS ring buffer
>>
>> -; CHECK-LABEL: @main
>> +; CHECK-LABEL: main
>> ; CHECK: BUFFER_LOAD_DWORD
>> ; CHECK: BUFFER_LOAD_DWORD
>> ; CHECK: BUFFER_LOAD_DWORD
>> diff --git a/test/CodeGen/R600/llvm.SI.sample-masked.ll b/test/CodeGen/R600/llvm.SI.sample-masked.ll
>> index 445359a..64ebdaf 100644
>> --- a/test/CodeGen/R600/llvm.SI.sample-masked.ll
>> +++ b/test/CodeGen/R600/llvm.SI.sample-masked.ll
>> @@ -1,6 +1,6 @@
>> ;RUN: llc < %s -march=r600 -mcpu=verde | FileCheck %s
>>
>> -; CHECK-LABEL: @v1
>> +; CHECK-LABEL: v1
>> ; CHECK: IMAGE_SAMPLE {{v\[[0-9]+:[0-9]+\]}}, 13
>> define void @v1(i32 %a1) #0 {
>> entry:
>> @@ -13,7 +13,7 @@ entry:
>> ret void
>> }
>>
>> -; CHECK-LABEL: @v2
>> +; CHECK-LABEL: v2
>> ; CHECK: IMAGE_SAMPLE {{v\[[0-9]+:[0-9]+\]}}, 11
>> define void @v2(i32 %a1) #0 {
>> entry:
>> @@ -26,7 +26,7 @@ entry:
>> ret void
>> }
>>
>> -; CHECK-LABEL: @v3
>> +; CHECK-LABEL: v3
>> ; CHECK: IMAGE_SAMPLE {{v\[[0-9]+:[0-9]+\]}}, 14
>> define void @v3(i32 %a1) #0 {
>> entry:
>> @@ -39,7 +39,7 @@ entry:
>> ret void
>> }
>>
>> -; CHECK-LABEL: @v4
>> +; CHECK-LABEL: v4
>> ; CHECK: IMAGE_SAMPLE {{v\[[0-9]+:[0-9]+\]}}, 7
>> define void @v4(i32 %a1) #0 {
>> entry:
>> @@ -52,7 +52,7 @@ entry:
>> ret void
>> }
>>
>> -; CHECK-LABEL: @v5
>> +; CHECK-LABEL: v5
>> ; CHECK: IMAGE_SAMPLE {{v\[[0-9]+:[0-9]+\]}}, 10
>> define void @v5(i32 %a1) #0 {
>> entry:
>> @@ -64,7 +64,7 @@ entry:
>> ret void
>> }
>>
>> -; CHECK-LABEL: @v6
>> +; CHECK-LABEL: v6
>> ; CHECK: IMAGE_SAMPLE {{v\[[0-9]+:[0-9]+\]}}, 6
>> define void @v6(i32 %a1) #0 {
>> entry:
>> @@ -76,7 +76,7 @@ entry:
>> ret void
>> }
>>
>> -; CHECK-LABEL: @v7
>> +; CHECK-LABEL: v7
>> ; CHECK: IMAGE_SAMPLE {{v\[[0-9]+:[0-9]+\]}}, 9
>> define void @v7(i32 %a1) #0 {
>> entry:
>> diff --git a/test/CodeGen/R600/llvm.SI.sample.ll b/test/CodeGen/R600/llvm.SI.sample.ll
>> index 24e8f64..18f30cb 100644
>> --- a/test/CodeGen/R600/llvm.SI.sample.ll
>> +++ b/test/CodeGen/R600/llvm.SI.sample.ll
>> @@ -135,7 +135,7 @@ define void @test(i32 %a1, i32 %a2, i32 %a3, i32 %a4) #0 {
>> ret void
>> }
>>
>> -; CHECK: @v1
>> +; CHECK: v1
>> ; CHECK: IMAGE_SAMPLE {{v\[[0-9]+:[0-9]+\]}}, 15
>> define void @v1(i32 %a1) #0 {
>> entry:
>> diff --git a/test/CodeGen/R600/llvm.SI.sendmsg.ll b/test/CodeGen/R600/llvm.SI.sendmsg.ll
>> index 581d422..d6d1051 100644
>> --- a/test/CodeGen/R600/llvm.SI.sendmsg.ll
>> +++ b/test/CodeGen/R600/llvm.SI.sendmsg.ll
>> @@ -1,6 +1,6 @@
>> ;RUN: llc < %s -march=r600 -mcpu=verde -verify-machineinstrs | FileCheck %s
>>
>> -; CHECK-LABEL: @main
>> +; CHECK-LABEL: main
>> ; CHECK: S_SENDMSG Gs(emit stream 0)
>> ; CHECK: S_SENDMSG Gs(cut stream 1)
>> ; CHECK: S_SENDMSG Gs(emit-cut stream 2)
>> diff --git a/test/CodeGen/R600/llvm.SI.tbuffer.store.ll b/test/CodeGen/R600/llvm.SI.tbuffer.store.ll
>> index 740581a..7f4ddcf 100644
>> --- a/test/CodeGen/R600/llvm.SI.tbuffer.store.ll
>> +++ b/test/CodeGen/R600/llvm.SI.tbuffer.store.ll
>> @@ -1,6 +1,6 @@
>> ;RUN: llc < %s -march=r600 -mcpu=verde -verify-machineinstrs | FileCheck %s
>>
>> -;CHECK-LABEL: @test1
>> +;CHECK-LABEL: test1
>> ;CHECK: TBUFFER_STORE_FORMAT_XYZW {{v\[[0-9]+:[0-9]+\]}}, 0x20, -1, 0, -1, 0, 14, 4, {{v[0-9]+}}, {{s\[[0-9]+:[0-9]+\]}}, -1, 0, 0
>> define void @test1(i32 %a1, i32 %vaddr) #0 {
>> %vdata = insertelement <4 x i32> undef, i32 %a1, i32 0
>> @@ -10,7 +10,7 @@ define void @test1(i32 %a1, i32 %vaddr) #0 {
>> ret void
>> }
>>
>> -;CHECK-LABEL: @test2
>> +;CHECK-LABEL: test2
>> ;CHECK: TBUFFER_STORE_FORMAT_XYZ {{v\[[0-9]+:[0-9]+\]}}, 0x18, -1, 0, -1, 0, 13, 4, {{v[0-9]+}}, {{s\[[0-9]+:[0-9]+\]}}, -1, 0, 0
>> define void @test2(i32 %a1, i32 %vaddr) #0 {
>> %vdata = insertelement <4 x i32> undef, i32 %a1, i32 0
>> @@ -20,7 +20,7 @@ define void @test2(i32 %a1, i32 %vaddr) #0 {
>> ret void
>> }
>>
>> -;CHECK-LABEL: @test3
>> +;CHECK-LABEL: test3
>> ;CHECK: TBUFFER_STORE_FORMAT_XY {{v\[[0-9]+:[0-9]+\]}}, 0x10, -1, 0, -1, 0, 11, 4, {{v[0-9]+}}, {{s\[[0-9]+:[0-9]+\]}}, -1, 0, 0
>> define void @test3(i32 %a1, i32 %vaddr) #0 {
>> %vdata = insertelement <2 x i32> undef, i32 %a1, i32 0
>> @@ -30,7 +30,7 @@ define void @test3(i32 %a1, i32 %vaddr) #0 {
>> ret void
>> }
>>
>> -;CHECK-LABEL: @test4
>> +;CHECK-LABEL: test4
>> ;CHECK: TBUFFER_STORE_FORMAT_X {{v[0-9]+}}, 0x8, -1, 0, -1, 0, 4, 4, {{v[0-9]+}}, {{s\[[0-9]+:[0-9]+\]}}, -1, 0, 0
>> define void @test4(i32 %vdata, i32 %vaddr) #0 {
>> call void @llvm.SI.tbuffer.store.i32(<16 x i8> undef, i32 %vdata,
>> diff --git a/test/CodeGen/R600/llvm.amdgpu.kilp.ll b/test/CodeGen/R600/llvm.amdgpu.kilp.ll
>> index 1b8b1bf..ad56461 100644
>> --- a/test/CodeGen/R600/llvm.amdgpu.kilp.ll
>> +++ b/test/CodeGen/R600/llvm.amdgpu.kilp.ll
>> @@ -1,6 +1,6 @@
>> ; RUN: llc -march=r600 -mcpu=verde -verify-machineinstrs < %s | FileCheck -check-prefix=SI %s
>>
>> -; SI-LABEL: @kilp_gs_const
>> +; SI-LABEL: kilp_gs_const
>> ; SI: S_MOV_B64 exec, 0
>> define void @kilp_gs_const() #0 {
>> main_body:
>> diff --git a/test/CodeGen/R600/llvm.amdgpu.lrp.ll b/test/CodeGen/R600/llvm.amdgpu.lrp.ll
>> index c493a01..0cc8974 100644
>> --- a/test/CodeGen/R600/llvm.amdgpu.lrp.ll
>> +++ b/test/CodeGen/R600/llvm.amdgpu.lrp.ll
>> @@ -2,7 +2,7 @@
>>
>> declare float @llvm.AMDGPU.lrp(float, float, float) nounwind readnone
>>
>> -; FUNC-LABEL: @test_lrp
>> +; FUNC-LABEL: test_lrp
>> ; SI: V_SUB_F32
>> ; SI: V_MAD_F32
>> define void @test_lrp(float addrspace(1)* %out, float %src0, float %src1, float %src2) nounwind {
>> diff --git a/test/CodeGen/R600/llvm.exp2.ll b/test/CodeGen/R600/llvm.exp2.ll
>> index 119d5ef..1ef7d23 100644
>> --- a/test/CodeGen/R600/llvm.exp2.ll
>> +++ b/test/CodeGen/R600/llvm.exp2.ll
>> @@ -2,7 +2,7 @@
>> ;RUN: llc < %s -march=r600 -mcpu=cayman | FileCheck %s --check-prefix=CM-CHECK --check-prefix=FUNC
>> ;RUN: llc < %s -march=r600 -mcpu=SI | FileCheck %s --check-prefix=SI-CHECK --check-prefix=FUNC
>>
>> -;FUNC-LABEL: @test
>> +;FUNC-LABEL: test
>> ;EG-CHECK: EXP_IEEE
>> ;CM-CHECK-DAG: EXP_IEEE T{{[0-9]+\.[XYZW]}} (MASKED)
>> ;CM-CHECK-DAG: EXP_IEEE T{{[0-9]+\.[XYZW]}} (MASKED)
>> @@ -17,7 +17,7 @@ entry:
>> ret void
>> }
>>
>> -;FUNC-LABEL: @testv2
>> +;FUNC-LABEL: testv2
>> ;EG-CHECK: EXP_IEEE
>> ;EG-CHECK: EXP_IEEE
>> ; FIXME: We should be able to merge these packets together on Cayman so we
>> @@ -40,7 +40,7 @@ entry:
>> ret void
>> }
>>
>> -;FUNC-LABEL: @testv4
>> +;FUNC-LABEL: testv4
>> ;EG-CHECK: EXP_IEEE
>> ;EG-CHECK: EXP_IEEE
>> ;EG-CHECK: EXP_IEEE
>> diff --git a/test/CodeGen/R600/llvm.floor.ll b/test/CodeGen/R600/llvm.floor.ll
>> index f7071cd..89db7b9 100644
>> --- a/test/CodeGen/R600/llvm.floor.ll
>> +++ b/test/CodeGen/R600/llvm.floor.ll
>> @@ -1,9 +1,9 @@
>> ; RUN: llc < %s -march=r600 -mcpu=redwood | FileCheck %s --check-prefix=R600-CHECK
>> ; RUN: llc < %s -march=r600 -mcpu=SI -verify-machineinstrs | FileCheck %s --check-prefix=SI-CHECK
>>
>> -; R600-CHECK: @f32
>> +; R600-CHECK: f32
>> ; R600-CHECK: FLOOR
>> -; SI-CHECK: @f32
>> +; SI-CHECK: f32
>> ; SI-CHECK: V_FLOOR_F32_e32
>> define void @f32(float addrspace(1)* %out, float %in) {
>> entry:
>> @@ -12,10 +12,10 @@ entry:
>> ret void
>> }
>>
>> -; R600-CHECK: @v2f32
>> +; R600-CHECK: v2f32
>> ; R600-CHECK: FLOOR
>> ; R600-CHECK: FLOOR
>> -; SI-CHECK: @v2f32
>> +; SI-CHECK: v2f32
>> ; SI-CHECK: V_FLOOR_F32_e32
>> ; SI-CHECK: V_FLOOR_F32_e32
>> define void @v2f32(<2 x float> addrspace(1)* %out, <2 x float> %in) {
>> @@ -25,12 +25,12 @@ entry:
>> ret void
>> }
>>
>> -; R600-CHECK: @v4f32
>> +; R600-CHECK: v4f32
>> ; R600-CHECK: FLOOR
>> ; R600-CHECK: FLOOR
>> ; R600-CHECK: FLOOR
>> ; R600-CHECK: FLOOR
>> -; SI-CHECK: @v4f32
>> +; SI-CHECK: v4f32
>> ; SI-CHECK: V_FLOOR_F32_e32
>> ; SI-CHECK: V_FLOOR_F32_e32
>> ; SI-CHECK: V_FLOOR_F32_e32
>> diff --git a/test/CodeGen/R600/llvm.log2.ll b/test/CodeGen/R600/llvm.log2.ll
>> index 4cba2d4..63c0855 100644
>> --- a/test/CodeGen/R600/llvm.log2.ll
>> +++ b/test/CodeGen/R600/llvm.log2.ll
>> @@ -2,7 +2,7 @@
>> ;RUN: llc < %s -march=r600 -mcpu=cayman | FileCheck %s --check-prefix=CM-CHECK --check-prefix=FUNC
>> ;RUN: llc < %s -march=r600 -mcpu=SI | FileCheck %s --check-prefix=SI-CHECK --check-prefix=FUNC
>>
>> -;FUNC-LABEL: @test
>> +;FUNC-LABEL: test
>> ;EG-CHECK: LOG_IEEE
>> ;CM-CHECK-DAG: LOG_IEEE T{{[0-9]+\.[XYZW]}} (MASKED)
>> ;CM-CHECK-DAG: LOG_IEEE T{{[0-9]+\.[XYZW]}} (MASKED)
>> @@ -17,7 +17,7 @@ entry:
>> ret void
>> }
>>
>> -;FUNC-LABEL: @testv2
>> +;FUNC-LABEL: testv2
>> ;EG-CHECK: LOG_IEEE
>> ;EG-CHECK: LOG_IEEE
>> ; FIXME: We should be able to merge these packets together on Cayman so we
>> @@ -40,7 +40,7 @@ entry:
>> ret void
>> }
>>
>> -;FUNC-LABEL: @testv4
>> +;FUNC-LABEL: testv4
>> ;EG-CHECK: LOG_IEEE
>> ;EG-CHECK: LOG_IEEE
>> ;EG-CHECK: LOG_IEEE
>> diff --git a/test/CodeGen/R600/llvm.memcpy.ll b/test/CodeGen/R600/llvm.memcpy.ll
>> index 68a4050..81661e7 100644
>> --- a/test/CodeGen/R600/llvm.memcpy.ll
>> +++ b/test/CodeGen/R600/llvm.memcpy.ll
>> @@ -4,7 +4,7 @@ declare void @llvm.memcpy.p3i8.p3i8.i32(i8 addrspace(3)* nocapture, i8 addrspace
>> declare void @llvm.memcpy.p1i8.p1i8.i64(i8 addrspace(1)* nocapture, i8 addrspace(1)* nocapture, i64, i32, i1) nounwind
>>
>>
>> -; FUNC-LABEL: @test_small_memcpy_i64_lds_to_lds_align1
>> +; FUNC-LABEL: test_small_memcpy_i64_lds_to_lds_align1
>> ; SI: DS_READ_U8
>> ; SI: DS_WRITE_B8
>> ; SI: DS_READ_U8
>> @@ -85,7 +85,7 @@ define void @test_small_memcpy_i64_lds_to_lds_align1(i64 addrspace(3)* noalias %
>> ret void
>> }
>>
>> -; FUNC-LABEL: @test_small_memcpy_i64_lds_to_lds_align2
>> +; FUNC-LABEL: test_small_memcpy_i64_lds_to_lds_align2
>> ; SI: DS_READ_U16
>> ; SI: DS_READ_U16
>> ; SI: DS_READ_U16
>> @@ -130,7 +130,7 @@ define void @test_small_memcpy_i64_lds_to_lds_align2(i64 addrspace(3)* noalias %
>> ret void
>> }
>>
>> -; FUNC-LABEL: @test_small_memcpy_i64_lds_to_lds_align4
>> +; FUNC-LABEL: test_small_memcpy_i64_lds_to_lds_align4
>> ; SI-DAG: DS_READ_B32
>> ; SI-DAG: DS_WRITE_B32
>>
>> @@ -167,7 +167,7 @@ define void @test_small_memcpy_i64_lds_to_lds_align4(i64 addrspace(3)* noalias %
>> }
>>
>> ; FIXME: Use 64-bit ops
>> -; FUNC-LABEL: @test_small_memcpy_i64_lds_to_lds_align8
>> +; FUNC-LABEL: test_small_memcpy_i64_lds_to_lds_align8
>>
>> ; SI-DAG: DS_READ_B32
>> ; SI-DAG: DS_WRITE_B32
>> @@ -204,7 +204,7 @@ define void @test_small_memcpy_i64_lds_to_lds_align8(i64 addrspace(3)* noalias %
>> ret void
>> }
>>
>> -; FUNC-LABEL: @test_small_memcpy_i64_global_to_global_align1
>> +; FUNC-LABEL: test_small_memcpy_i64_global_to_global_align1
>> ; SI-DAG: BUFFER_LOAD_UBYTE
>> ; SI-DAG: BUFFER_STORE_BYTE
>> ; SI-DAG: BUFFER_LOAD_UBYTE
>> @@ -281,7 +281,7 @@ define void @test_small_memcpy_i64_global_to_global_align1(i64 addrspace(1)* noa
>> ret void
>> }
>>
>> -; FUNC-LABEL: @test_small_memcpy_i64_global_to_global_align2
>> +; FUNC-LABEL: test_small_memcpy_i64_global_to_global_align2
>> ; SI-DAG: BUFFER_LOAD_USHORT
>> ; SI-DAG: BUFFER_LOAD_USHORT
>> ; SI-DAG: BUFFER_LOAD_USHORT
>> @@ -324,7 +324,7 @@ define void @test_small_memcpy_i64_global_to_global_align2(i64 addrspace(1)* noa
>> ret void
>> }
>>
>> -; FUNC-LABEL: @test_small_memcpy_i64_global_to_global_align4
>> +; FUNC-LABEL: test_small_memcpy_i64_global_to_global_align4
>> ; SI: BUFFER_LOAD_DWORDX4
>> ; SI: BUFFER_LOAD_DWORDX4
>> ; SI: BUFFER_STORE_DWORDX4
>> @@ -337,7 +337,7 @@ define void @test_small_memcpy_i64_global_to_global_align4(i64 addrspace(1)* noa
>> ret void
>> }
>>
>> -; FUNC-LABEL: @test_small_memcpy_i64_global_to_global_align8
>> +; FUNC-LABEL: test_small_memcpy_i64_global_to_global_align8
>> ; SI: BUFFER_LOAD_DWORDX4
>> ; SI: BUFFER_LOAD_DWORDX4
>> ; SI: BUFFER_STORE_DWORDX4
>> @@ -350,7 +350,7 @@ define void @test_small_memcpy_i64_global_to_global_align8(i64 addrspace(1)* noa
>> ret void
>> }
>>
>> -; FUNC-LABEL: @test_small_memcpy_i64_global_to_global_align16
>> +; FUNC-LABEL: test_small_memcpy_i64_global_to_global_align16
>> ; SI: BUFFER_LOAD_DWORDX4
>> ; SI: BUFFER_LOAD_DWORDX4
>> ; SI: BUFFER_STORE_DWORDX4
>> diff --git a/test/CodeGen/R600/llvm.rint.f64.ll b/test/CodeGen/R600/llvm.rint.f64.ll
>> index 3e2884b..5b62238 100644
>> --- a/test/CodeGen/R600/llvm.rint.f64.ll
>> +++ b/test/CodeGen/R600/llvm.rint.f64.ll
>> @@ -1,7 +1,7 @@
>> ; RUN: llc -march=r600 -mcpu=bonaire -verify-machineinstrs < %s | FileCheck -check-prefix=CI -check-prefix=FUNC %s
>> ; RUN: llc -march=r600 -mcpu=SI -verify-machineinstrs < %s | FileCheck -check-prefix=SI -check-prefix=FUNC %s
>>
>> -; FUNC-LABEL: @rint_f64
>> +; FUNC-LABEL: rint_f64
>> ; CI: V_RNDNE_F64_e32
>>
>> ; SI-DAG: V_ADD_F64
>> @@ -17,7 +17,7 @@ entry:
>> ret void
>> }
>>
>> -; FUNC-LABEL: @rint_v2f64
>> +; FUNC-LABEL: rint_v2f64
>> ; CI: V_RNDNE_F64_e32
>> ; CI: V_RNDNE_F64_e32
>> define void @rint_v2f64(<2 x double> addrspace(1)* %out, <2 x double> %in) {
>> @@ -27,7 +27,7 @@ entry:
>> ret void
>> }
>>
>> -; FUNC-LABEL: @rint_v4f64
>> +; FUNC-LABEL: rint_v4f64
>> ; CI: V_RNDNE_F64_e32
>> ; CI: V_RNDNE_F64_e32
>> ; CI: V_RNDNE_F64_e32
>> diff --git a/test/CodeGen/R600/llvm.rint.ll b/test/CodeGen/R600/llvm.rint.ll
>> index 209bb43..3b1086a 100644
>> --- a/test/CodeGen/R600/llvm.rint.ll
>> +++ b/test/CodeGen/R600/llvm.rint.ll
>> @@ -1,7 +1,7 @@
>> ; RUN: llc -march=r600 -mcpu=redwood < %s | FileCheck %s -check-prefix=R600 -check-prefix=FUNC
>> ; RUN: llc -march=r600 -mcpu=SI -verify-machineinstrs < %s | FileCheck -check-prefix=SI -check-prefix=FUNC %s
>>
>> -; FUNC-LABEL: @rint_f32
>> +; FUNC-LABEL: rint_f32
>> ; R600: RNDNE
>>
>> ; SI: V_RNDNE_F32_e32
>> @@ -12,7 +12,7 @@ entry:
>> ret void
>> }
>>
>> -; FUNC-LABEL: @rint_v2f32
>> +; FUNC-LABEL: rint_v2f32
>> ; R600: RNDNE
>> ; R600: RNDNE
>>
>> @@ -25,7 +25,7 @@ entry:
>> ret void
>> }
>>
>> -; FUNC-LABEL: @rint_v4f32
>> +; FUNC-LABEL: rint_v4f32
>> ; R600: RNDNE
>> ; R600: RNDNE
>> ; R600: RNDNE
>> @@ -42,7 +42,7 @@ entry:
>> ret void
>> }
>>
>> -; FUNC-LABEL: @legacy_amdil_round_nearest_f32
>> +; FUNC-LABEL: legacy_amdil_round_nearest_f32
>> ; R600: RNDNE
>>
>> ; SI: V_RNDNE_F32_e32
>> diff --git a/test/CodeGen/R600/llvm.round.ll b/test/CodeGen/R600/llvm.round.ll
>> index 5700427..1dd1cb8 100644
>> --- a/test/CodeGen/R600/llvm.round.ll
>> +++ b/test/CodeGen/R600/llvm.round.ll
>> @@ -1,6 +1,6 @@
>> ; RUN: llc -march=r600 -mcpu=redwood < %s | FileCheck %s --check-prefix=R600 --check-prefix=FUNC
>>
>> -; FUNC-LABEL: @f32
>> +; FUNC-LABEL: f32
>> ; R600: FRACT {{.*}}, [[ARG:KC[0-9]\[[0-9]+\]\.[XYZW]]]
>> ; R600-DAG: ADD {{.*}}, -0.5
>> ; R600-DAG: CEIL {{.*}} [[ARG]]
>> diff --git a/test/CodeGen/R600/llvm.sin.ll b/test/CodeGen/R600/llvm.sin.ll
>> index 50ddb14..dc17722 100644
>> --- a/test/CodeGen/R600/llvm.sin.ll
>> +++ b/test/CodeGen/R600/llvm.sin.ll
>> @@ -19,7 +19,7 @@ define void @sin_f32(float addrspace(1)* %out, float %x) #1 {
>> ret void
>> }
>>
>> -; FUNC-LABEL: @sin_3x_f32
>> +; FUNC-LABEL: sin_3x_f32
>> ; SI-UNSAFE-NOT: V_ADD_F32
>> ; SI-UNSAFE: 0x3ef47644
>> ; SI-UNSAFE: V_MUL_F32
>> @@ -35,7 +35,7 @@ define void @sin_3x_f32(float addrspace(1)* %out, float %x) #1 {
>> ret void
>> }
>>
>> -; FUNC-LABEL: @sin_2x_f32
>> +; FUNC-LABEL: sin_2x_f32
>> ; SI-UNSAFE-NOT: V_ADD_F32
>> ; SI-UNSAFE: 0x3ea2f983
>> ; SI-UNSAFE: V_MUL_F32
>> @@ -51,7 +51,7 @@ define void @sin_2x_f32(float addrspace(1)* %out, float %x) #1 {
>> ret void
>> }
>>
>> -; FUNC-LABEL: @test_2sin_f32
>> +; FUNC-LABEL: test_2sin_f32
>> ; SI-UNSAFE: 0x3ea2f983
>> ; SI-UNSAFE: V_MUL_F32
>> ; SI-SAFE: V_ADD_F32
>> @@ -66,7 +66,7 @@ define void @test_2sin_f32(float addrspace(1)* %out, float %x) #1 {
>> ret void
>> }
>>
>> -; FUNC-LABEL: @sin_v4f32
>> +; FUNC-LABEL: sin_v4f32
>> ; EG: SIN * T{{[0-9]+\.[XYZW], PV\.[XYZW]}}
>> ; EG: SIN * T{{[0-9]+\.[XYZW], PV\.[XYZW]}}
>> ; EG: SIN * T{{[0-9]+\.[XYZW], PV\.[XYZW]}}
>> diff --git a/test/CodeGen/R600/llvm.sqrt.ll b/test/CodeGen/R600/llvm.sqrt.ll
>> index 4eee37f..b065926 100644
>> --- a/test/CodeGen/R600/llvm.sqrt.ll
>> +++ b/test/CodeGen/R600/llvm.sqrt.ll
>> @@ -1,10 +1,10 @@
>> ; RUN: llc < %s -march=r600 --mcpu=redwood | FileCheck %s --check-prefix=R600-CHECK
>> ; RUN: llc < %s -march=r600 --mcpu=SI -verify-machineinstrs| FileCheck %s --check-prefix=SI-CHECK
>>
>> -; R600-CHECK-LABEL: @sqrt_f32
>> +; R600-CHECK-LABEL: sqrt_f32
>> ; R600-CHECK: RECIPSQRT_CLAMPED * T{{[0-9]\.[XYZW]}}, KC0[2].Z
>> ; R600-CHECK: MUL NON-IEEE T{{[0-9]\.[XYZW]}}, KC0[2].Z, PS
>> -; SI-CHECK-LABEL: @sqrt_f32
>> +; SI-CHECK-LABEL: sqrt_f32
>> ; SI-CHECK: V_SQRT_F32_e32
>> define void @sqrt_f32(float addrspace(1)* %out, float %in) {
>> entry:
>> @@ -13,12 +13,12 @@ entry:
>> ret void
>> }
>>
>> -; R600-CHECK-LABEL: @sqrt_v2f32
>> +; R600-CHECK-LABEL: sqrt_v2f32
>> ; R600-CHECK-DAG: RECIPSQRT_CLAMPED * T{{[0-9]\.[XYZW]}}, KC0[2].W
>> ; R600-CHECK-DAG: MUL NON-IEEE T{{[0-9]\.[XYZW]}}, KC0[2].W, PS
>> ; R600-CHECK-DAG: RECIPSQRT_CLAMPED * T{{[0-9]\.[XYZW]}}, KC0[3].X
>> ; R600-CHECK-DAG: MUL NON-IEEE T{{[0-9]\.[XYZW]}}, KC0[3].X, PS
>> -; SI-CHECK-LABEL: @sqrt_v2f32
>> +; SI-CHECK-LABEL: sqrt_v2f32
>> ; SI-CHECK: V_SQRT_F32_e32
>> ; SI-CHECK: V_SQRT_F32_e32
>> define void @sqrt_v2f32(<2 x float> addrspace(1)* %out, <2 x float> %in) {
>> @@ -28,7 +28,7 @@ entry:
>> ret void
>> }
>>
>> -; R600-CHECK-LABEL: @sqrt_v4f32
>> +; R600-CHECK-LABEL: sqrt_v4f32
>> ; R600-CHECK-DAG: RECIPSQRT_CLAMPED * T{{[0-9]\.[XYZW]}}, KC0[3].Y
>> ; R600-CHECK-DAG: MUL NON-IEEE T{{[0-9]\.[XYZW]}}, KC0[3].Y, PS
>> ; R600-CHECK-DAG: RECIPSQRT_CLAMPED * T{{[0-9]\.[XYZW]}}, KC0[3].Z
>> @@ -37,7 +37,7 @@ entry:
>> ; R600-CHECK-DAG: MUL NON-IEEE T{{[0-9]\.[XYZW]}}, KC0[3].W, PS
>> ; R600-CHECK-DAG: RECIPSQRT_CLAMPED * T{{[0-9]\.[XYZW]}}, KC0[4].X
>> ; R600-CHECK-DAG: MUL NON-IEEE T{{[0-9]\.[XYZW]}}, KC0[4].X, PS
>> -; SI-CHECK-LABEL: @sqrt_v4f32
>> +; SI-CHECK-LABEL: sqrt_v4f32
>> ; SI-CHECK: V_SQRT_F32_e32
>> ; SI-CHECK: V_SQRT_F32_e32
>> ; SI-CHECK: V_SQRT_F32_e32
>> diff --git a/test/CodeGen/R600/llvm.trunc.ll b/test/CodeGen/R600/llvm.trunc.ll
>> index fa6fb99..1e0fb7e 100644
>> --- a/test/CodeGen/R600/llvm.trunc.ll
>> +++ b/test/CodeGen/R600/llvm.trunc.ll
>> @@ -1,6 +1,6 @@
>> ; RUN: llc < %s -march=r600 -mcpu=redwood | FileCheck %s
>>
>> -; CHECK-LABEL: @trunc_f32
>> +; CHECK-LABEL: trunc_f32
>> ; CHECK: TRUNC
>>
>> define void @trunc_f32(float addrspace(1)* %out, float %in) {
>> diff --git a/test/CodeGen/R600/load-i1.ll b/test/CodeGen/R600/load-i1.ll
>> index 9ba81b8..7b6f2a2 100644
>> --- a/test/CodeGen/R600/load-i1.ll
>> +++ b/test/CodeGen/R600/load-i1.ll
>> @@ -1,7 +1,7 @@
>> ; RUN: llc -march=r600 -mcpu=SI -verify-machineinstrs< %s | FileCheck -check-prefix=SI %s
>>
>>
>> -; SI-LABEL: @global_copy_i1_to_i1
>> +; SI-LABEL: global_copy_i1_to_i1
>> ; SI: BUFFER_LOAD_UBYTE
>> ; SI: V_AND_B32_e32 v{{[0-9]+}}, 1
>> ; SI: BUFFER_STORE_BYTE
>> @@ -12,7 +12,7 @@ define void @global_copy_i1_to_i1(i1 addrspace(1)* %out, i1 addrspace(1)* %in) n
>> ret void
>> }
>>
>> -; SI-LABEL: @global_sextload_i1_to_i32
>> +; SI-LABEL: global_sextload_i1_to_i32
>> ; XSI: BUFFER_LOAD_BYTE
>> ; SI: BUFFER_STORE_DWORD
>> ; SI: S_ENDPGM
>> @@ -23,7 +23,7 @@ define void @global_sextload_i1_to_i32(i32 addrspace(1)* %out, i1 addrspace(1)*
>> ret void
>> }
>>
>> -; SI-LABEL: @global_zextload_i1_to_i32
>> +; SI-LABEL: global_zextload_i1_to_i32
>> ; SI: BUFFER_LOAD_UBYTE
>> ; SI: BUFFER_STORE_DWORD
>> ; SI: S_ENDPGM
>> @@ -34,7 +34,7 @@ define void @global_zextload_i1_to_i32(i32 addrspace(1)* %out, i1 addrspace(1)*
>> ret void
>> }
>>
>> -; SI-LABEL: @global_sextload_i1_to_i64
>> +; SI-LABEL: global_sextload_i1_to_i64
>> ; XSI: BUFFER_LOAD_BYTE
>> ; SI: BUFFER_STORE_DWORDX2
>> ; SI: S_ENDPGM
>> @@ -45,7 +45,7 @@ define void @global_sextload_i1_to_i64(i64 addrspace(1)* %out, i1 addrspace(1)*
>> ret void
>> }
>>
>> -; SI-LABEL: @global_zextload_i1_to_i64
>> +; SI-LABEL: global_zextload_i1_to_i64
>> ; SI: BUFFER_LOAD_UBYTE
>> ; SI: BUFFER_STORE_DWORDX2
>> ; SI: S_ENDPGM
>> @@ -56,7 +56,7 @@ define void @global_zextload_i1_to_i64(i64 addrspace(1)* %out, i1 addrspace(1)*
>> ret void
>> }
>>
>> -; SI-LABEL: @i1_arg
>> +; SI-LABEL: i1_arg
>> ; SI: BUFFER_LOAD_UBYTE
>> ; SI: V_AND_B32_e32
>> ; SI: BUFFER_STORE_BYTE
>> @@ -66,7 +66,7 @@ define void @i1_arg(i1 addrspace(1)* %out, i1 %x) nounwind {
>> ret void
>> }
>>
>> -; SI-LABEL: @i1_arg_zext_i32
>> +; SI-LABEL: i1_arg_zext_i32
>> ; SI: BUFFER_LOAD_UBYTE
>> ; SI: BUFFER_STORE_DWORD
>> ; SI: S_ENDPGM
>> @@ -76,7 +76,7 @@ define void @i1_arg_zext_i32(i32 addrspace(1)* %out, i1 %x) nounwind {
>> ret void
>> }
>>
>> -; SI-LABEL: @i1_arg_zext_i64
>> +; SI-LABEL: i1_arg_zext_i64
>> ; SI: BUFFER_LOAD_UBYTE
>> ; SI: BUFFER_STORE_DWORDX2
>> ; SI: S_ENDPGM
>> @@ -86,7 +86,7 @@ define void @i1_arg_zext_i64(i64 addrspace(1)* %out, i1 %x) nounwind {
>> ret void
>> }
>>
>> -; SI-LABEL: @i1_arg_sext_i32
>> +; SI-LABEL: i1_arg_sext_i32
>> ; XSI: BUFFER_LOAD_BYTE
>> ; SI: BUFFER_STORE_DWORD
>> ; SI: S_ENDPGM
>> @@ -96,7 +96,7 @@ define void @i1_arg_sext_i32(i32 addrspace(1)* %out, i1 %x) nounwind {
>> ret void
>> }
>>
>> -; SI-LABEL: @i1_arg_sext_i64
>> +; SI-LABEL: i1_arg_sext_i64
>> ; XSI: BUFFER_LOAD_BYTE
>> ; SI: BUFFER_STORE_DWORDX2
>> ; SI: S_ENDPGM
>> diff --git a/test/CodeGen/R600/load.ll b/test/CodeGen/R600/load.ll
>> index 8905fbd..a6aa46e 100644
>> --- a/test/CodeGen/R600/load.ll
>> +++ b/test/CodeGen/R600/load.ll
>> @@ -7,7 +7,7 @@
>> ;===------------------------------------------------------------------------===;
>>
>> ; Load an i8 value from the global address space.
>> -; FUNC-LABEL: @load_i8
>> +; FUNC-LABEL: load_i8
>> ; R600-CHECK: VTX_READ_8 T{{[0-9]+\.X, T[0-9]+\.X}}
>>
>> ; SI-CHECK: BUFFER_LOAD_UBYTE v{{[0-9]+}},
>> @@ -18,7 +18,7 @@ define void @load_i8(i32 addrspace(1)* %out, i8 addrspace(1)* %in) {
>> ret void
>> }
>>
>> -; FUNC-LABEL: @load_i8_sext
>> +; FUNC-LABEL: load_i8_sext
>> ; R600-CHECK: VTX_READ_8 [[DST:T[0-9]\.[XYZW]]], [[DST]]
>> ; R600-CHECK: LSHL {{[* ]*}}T{{[0-9]}}.[[LSHL_CHAN:[XYZW]]], [[DST]]
>> ; R600-CHECK: 24
>> @@ -33,7 +33,7 @@ entry:
>> ret void
>> }
>>
>> -; FUNC-LABEL: @load_v2i8
>> +; FUNC-LABEL: load_v2i8
>> ; R600-CHECK: VTX_READ_8
>> ; R600-CHECK: VTX_READ_8
>> ; SI-CHECK: BUFFER_LOAD_UBYTE
>> @@ -46,7 +46,7 @@ entry:
>> ret void
>> }
>>
>> -; FUNC-LABEL: @load_v2i8_sext
>> +; FUNC-LABEL: load_v2i8_sext
>> ; R600-CHECK-DAG: VTX_READ_8 [[DST_X:T[0-9]\.[XYZW]]], [[DST_X]]
>> ; R600-CHECK-DAG: VTX_READ_8 [[DST_Y:T[0-9]\.[XYZW]]], [[DST_Y]]
>> ; R600-CHECK-DAG: LSHL {{[* ]*}}T{{[0-9]}}.[[LSHL_X_CHAN:[XYZW]]], [[DST_X]]
>> @@ -67,7 +67,7 @@ entry:
>> ret void
>> }
>>
>> -; FUNC-LABEL: @load_v4i8
>> +; FUNC-LABEL: load_v4i8
>> ; R600-CHECK: VTX_READ_8
>> ; R600-CHECK: VTX_READ_8
>> ; R600-CHECK: VTX_READ_8
>> @@ -84,7 +84,7 @@ entry:
>> ret void
>> }
>>
>> -; FUNC-LABEL: @load_v4i8_sext
>> +; FUNC-LABEL: load_v4i8_sext
>> ; R600-CHECK-DAG: VTX_READ_8 [[DST_X:T[0-9]\.[XYZW]]], [[DST_X]]
>> ; R600-CHECK-DAG: VTX_READ_8 [[DST_Y:T[0-9]\.[XYZW]]], [[DST_Y]]
>> ; R600-CHECK-DAG: VTX_READ_8 [[DST_Z:T[0-9]\.[XYZW]]], [[DST_Z]]
>> @@ -118,7 +118,7 @@ entry:
>> }
>>
>> ; Load an i16 value from the global address space.
>> -; FUNC-LABEL: @load_i16
>> +; FUNC-LABEL: load_i16
>> ; R600-CHECK: VTX_READ_16 T{{[0-9]+\.X, T[0-9]+\.X}}
>> ; SI-CHECK: BUFFER_LOAD_USHORT
>> define void @load_i16(i32 addrspace(1)* %out, i16 addrspace(1)* %in) {
>> @@ -129,7 +129,7 @@ entry:
>> ret void
>> }
>>
>> -; FUNC-LABEL: @load_i16_sext
>> +; FUNC-LABEL: load_i16_sext
>> ; R600-CHECK: VTX_READ_16 [[DST:T[0-9]\.[XYZW]]], [[DST]]
>> ; R600-CHECK: LSHL {{[* ]*}}T{{[0-9]}}.[[LSHL_CHAN:[XYZW]]], [[DST]]
>> ; R600-CHECK: 16
>> @@ -144,7 +144,7 @@ entry:
>> ret void
>> }
>>
>> -; FUNC-LABEL: @load_v2i16
>> +; FUNC-LABEL: load_v2i16
>> ; R600-CHECK: VTX_READ_16
>> ; R600-CHECK: VTX_READ_16
>> ; SI-CHECK: BUFFER_LOAD_USHORT
>> @@ -157,7 +157,7 @@ entry:
>> ret void
>> }
>>
>> -; FUNC-LABEL: @load_v2i16_sext
>> +; FUNC-LABEL: load_v2i16_sext
>> ; R600-CHECK-DAG: VTX_READ_16 [[DST_X:T[0-9]\.[XYZW]]], [[DST_X]]
>> ; R600-CHECK-DAG: VTX_READ_16 [[DST_Y:T[0-9]\.[XYZW]]], [[DST_Y]]
>> ; R600-CHECK-DAG: LSHL {{[* ]*}}T{{[0-9]}}.[[LSHL_X_CHAN:[XYZW]]], [[DST_X]]
>> @@ -178,7 +178,7 @@ entry:
>> ret void
>> }
>>
>> -; FUNC-LABEL: @load_v4i16
>> +; FUNC-LABEL: load_v4i16
>> ; R600-CHECK: VTX_READ_16
>> ; R600-CHECK: VTX_READ_16
>> ; R600-CHECK: VTX_READ_16
>> @@ -195,7 +195,7 @@ entry:
>> ret void
>> }
>>
>> -; FUNC-LABEL: @load_v4i16_sext
>> +; FUNC-LABEL: load_v4i16_sext
>> ; R600-CHECK-DAG: VTX_READ_16 [[DST_X:T[0-9]\.[XYZW]]], [[DST_X]]
>> ; R600-CHECK-DAG: VTX_READ_16 [[DST_Y:T[0-9]\.[XYZW]]], [[DST_Y]]
>> ; R600-CHECK-DAG: VTX_READ_16 [[DST_Z:T[0-9]\.[XYZW]]], [[DST_Z]]
>> @@ -229,7 +229,7 @@ entry:
>> }
>>
>> ; load an i32 value from the global address space.
>> -; FUNC-LABEL: @load_i32
>> +; FUNC-LABEL: load_i32
>> ; R600-CHECK: VTX_READ_32 T{{[0-9]+}}.X, T{{[0-9]+}}.X, 0
>>
>> ; SI-CHECK: BUFFER_LOAD_DWORD v{{[0-9]+}}
>> @@ -241,7 +241,7 @@ entry:
>> }
>>
>> ; load a f32 value from the global address space.
>> -; FUNC-LABEL: @load_f32
>> +; FUNC-LABEL: load_f32
>> ; R600-CHECK: VTX_READ_32 T{{[0-9]+}}.X, T{{[0-9]+}}.X, 0
>>
>> ; SI-CHECK: BUFFER_LOAD_DWORD v{{[0-9]+}}
>> @@ -253,7 +253,7 @@ entry:
>> }
>>
>> ; load a v2f32 value from the global address space
>> -; FUNC-LABEL: @load_v2f32
>> +; FUNC-LABEL: load_v2f32
>> ; R600-CHECK: MEM_RAT
>> ; R600-CHECK: VTX_READ_64
>> ; SI-CHECK: BUFFER_LOAD_DWORDX2
>> @@ -264,7 +264,7 @@ entry:
>> ret void
>> }
>>
>> -; FUNC-LABEL: @load_i64
>> +; FUNC-LABEL: load_i64
>> ; R600-CHECK: VTX_READ_64
>> ; SI-CHECK: BUFFER_LOAD_DWORDX2
>> define void @load_i64(i64 addrspace(1)* %out, i64 addrspace(1)* %in) {
>> @@ -274,7 +274,7 @@ entry:
>> ret void
>> }
>>
>> -; FUNC-LABEL: @load_i64_sext
>> +; FUNC-LABEL: load_i64_sext
>> ; R600-CHECK: MEM_RAT
>> ; R600-CHECK: MEM_RAT
>> ; R600-CHECK: ASHR {{[* ]*}}T{{[0-9]\.[XYZW]}}, T{{[0-9]\.[XYZW]}}, literal.x
>> @@ -289,7 +289,7 @@ entry:
>> ret void
>> }
>>
>> -; FUNC-LABEL: @load_i64_zext
>> +; FUNC-LABEL: load_i64_zext
>> ; R600-CHECK: MEM_RAT
>> ; R600-CHECK: MEM_RAT
>> define void @load_i64_zext(i64 addrspace(1)* %out, i32 addrspace(1)* %in) {
>> @@ -300,7 +300,7 @@ entry:
>> ret void
>> }
>>
>> -; FUNC-LABEL: @load_v8i32
>> +; FUNC-LABEL: load_v8i32
>> ; R600-CHECK: VTX_READ_128
>> ; R600-CHECK: VTX_READ_128
>> ; XXX: We should be using DWORDX4 instructions on SI.
>> @@ -319,7 +319,7 @@ entry:
>> ret void
>> }
>>
>> -; FUNC-LABEL: @load_v16i32
>> +; FUNC-LABEL: load_v16i32
>> ; R600-CHECK: VTX_READ_128
>> ; R600-CHECK: VTX_READ_128
>> ; R600-CHECK: VTX_READ_128
>> @@ -353,7 +353,7 @@ entry:
>> ;===------------------------------------------------------------------------===;
>>
>> ; Load a sign-extended i8 value
>> -; FUNC-LABEL: @load_const_i8_sext
>> +; FUNC-LABEL: load_const_i8_sext
>> ; R600-CHECK: VTX_READ_8 [[DST:T[0-9]\.[XYZW]]], [[DST]]
>> ; R600-CHECK: LSHL {{[* ]*}}T{{[0-9]}}.[[LSHL_CHAN:[XYZW]]], [[DST]]
>> ; R600-CHECK: 24
>> @@ -369,7 +369,7 @@ entry:
>> }
>>
>> ; Load an aligned i8 value
>> -; FUNC-LABEL: @load_const_i8_aligned
>> +; FUNC-LABEL: load_const_i8_aligned
>> ; R600-CHECK: VTX_READ_8 T{{[0-9]+\.X, T[0-9]+\.X}}
>> ; SI-CHECK: BUFFER_LOAD_UBYTE v{{[0-9]+}},
>> define void @load_const_i8_aligned(i32 addrspace(1)* %out, i8 addrspace(2)* %in) {
>> @@ -381,7 +381,7 @@ entry:
>> }
>>
>> ; Load an un-aligned i8 value
>> -; FUNC-LABEL: @load_const_i8_unaligned
>> +; FUNC-LABEL: load_const_i8_unaligned
>> ; R600-CHECK: VTX_READ_8 T{{[0-9]+\.X, T[0-9]+\.X}}
>> ; SI-CHECK: BUFFER_LOAD_UBYTE v{{[0-9]+}},
>> define void @load_const_i8_unaligned(i32 addrspace(1)* %out, i8 addrspace(2)* %in) {
>> @@ -394,7 +394,7 @@ entry:
>> }
>>
>> ; Load a sign-extended i16 value
>> -; FUNC-LABEL: @load_const_i16_sext
>> +; FUNC-LABEL: load_const_i16_sext
>> ; R600-CHECK: VTX_READ_16 [[DST:T[0-9]\.[XYZW]]], [[DST]]
>> ; R600-CHECK: LSHL {{[* ]*}}T{{[0-9]}}.[[LSHL_CHAN:[XYZW]]], [[DST]]
>> ; R600-CHECK: 16
>> @@ -410,7 +410,7 @@ entry:
>> }
>>
>> ; Load an aligned i16 value
>> -; FUNC-LABEL: @load_const_i16_aligned
>> +; FUNC-LABEL: load_const_i16_aligned
>> ; R600-CHECK: VTX_READ_16 T{{[0-9]+\.X, T[0-9]+\.X}}
>> ; SI-CHECK: BUFFER_LOAD_USHORT
>> define void @load_const_i16_aligned(i32 addrspace(1)* %out, i16 addrspace(2)* %in) {
>> @@ -422,7 +422,7 @@ entry:
>> }
>>
>> ; Load an un-aligned i16 value
>> -; FUNC-LABEL: @load_const_i16_unaligned
>> +; FUNC-LABEL: load_const_i16_unaligned
>> ; R600-CHECK: VTX_READ_16 T{{[0-9]+\.X, T[0-9]+\.X}}
>> ; SI-CHECK: BUFFER_LOAD_USHORT
>> define void @load_const_i16_unaligned(i32 addrspace(1)* %out, i16 addrspace(2)* %in) {
>> @@ -435,7 +435,7 @@ entry:
>> }
>>
>> ; Load an i32 value from the constant address space.
>> -; FUNC-LABEL: @load_const_addrspace_i32
>> +; FUNC-LABEL: load_const_addrspace_i32
>> ; R600-CHECK: VTX_READ_32 T{{[0-9]+}}.X, T{{[0-9]+}}.X, 0
>>
>> ; SI-CHECK: S_LOAD_DWORD s{{[0-9]+}}
>> @@ -447,7 +447,7 @@ entry:
>> }
>>
>> ; Load a f32 value from the constant address space.
>> -; FUNC-LABEL: @load_const_addrspace_f32
>> +; FUNC-LABEL: load_const_addrspace_f32
>> ; R600-CHECK: VTX_READ_32 T{{[0-9]+}}.X, T{{[0-9]+}}.X, 0
>>
>> ; SI-CHECK: S_LOAD_DWORD s{{[0-9]+}}
>> @@ -462,7 +462,7 @@ define void @load_const_addrspace_f32(float addrspace(1)* %out, float addrspace(
>> ;===------------------------------------------------------------------------===;
>>
>> ; Load an i8 value from the local address space.
>> -; FUNC-LABEL: @load_i8_local
>> +; FUNC-LABEL: load_i8_local
>> ; R600-CHECK: LDS_UBYTE_READ_RET
>> ; SI-CHECK-NOT: S_WQM_B64
>> ; SI-CHECK: S_MOV_B32 m0
>> @@ -474,7 +474,7 @@ define void @load_i8_local(i32 addrspace(1)* %out, i8 addrspace(3)* %in) {
>> ret void
>> }
>>
>> -; FUNC-LABEL: @load_i8_sext_local
>> +; FUNC-LABEL: load_i8_sext_local
>> ; R600-CHECK: LDS_UBYTE_READ_RET
>> ; R600-CHECK: ASHR
>> ; SI-CHECK-NOT: S_WQM_B64
>> @@ -488,7 +488,7 @@ entry:
>> ret void
>> }
>>
>> -; FUNC-LABEL: @load_v2i8_local
>> +; FUNC-LABEL: load_v2i8_local
>> ; R600-CHECK: LDS_UBYTE_READ_RET
>> ; R600-CHECK: LDS_UBYTE_READ_RET
>> ; SI-CHECK-NOT: S_WQM_B64
>> @@ -503,7 +503,7 @@ entry:
>> ret void
>> }
>>
>> -; FUNC-LABEL: @load_v2i8_sext_local
>> +; FUNC-LABEL: load_v2i8_sext_local
>> ; R600-CHECK-DAG: LDS_UBYTE_READ_RET
>> ; R600-CHECK-DAG: LDS_UBYTE_READ_RET
>> ; R600-CHECK-DAG: ASHR
>> @@ -520,7 +520,7 @@ entry:
>> ret void
>> }
>>
>> -; FUNC-LABEL: @load_v4i8_local
>> +; FUNC-LABEL: load_v4i8_local
>> ; R600-CHECK: LDS_UBYTE_READ_RET
>> ; R600-CHECK: LDS_UBYTE_READ_RET
>> ; R600-CHECK: LDS_UBYTE_READ_RET
>> @@ -539,7 +539,7 @@ entry:
>> ret void
>> }
>>
>> -; FUNC-LABEL: @load_v4i8_sext_local
>> +; FUNC-LABEL: load_v4i8_sext_local
>> ; R600-CHECK-DAG: LDS_UBYTE_READ_RET
>> ; R600-CHECK-DAG: LDS_UBYTE_READ_RET
>> ; R600-CHECK-DAG: LDS_UBYTE_READ_RET
>> @@ -563,7 +563,7 @@ entry:
>> }
>>
>> ; Load an i16 value from the local address space.
>> -; FUNC-LABEL: @load_i16_local
>> +; FUNC-LABEL: load_i16_local
>> ; R600-CHECK: LDS_USHORT_READ_RET
>> ; SI-CHECK-NOT: S_WQM_B64
>> ; SI-CHECK: S_MOV_B32 m0
>> @@ -576,7 +576,7 @@ entry:
>> ret void
>> }
>>
>> -; FUNC-LABEL: @load_i16_sext_local
>> +; FUNC-LABEL: load_i16_sext_local
>> ; R600-CHECK: LDS_USHORT_READ_RET
>> ; R600-CHECK: ASHR
>> ; SI-CHECK-NOT: S_WQM_B64
>> @@ -590,7 +590,7 @@ entry:
>> ret void
>> }
>>
>> -; FUNC-LABEL: @load_v2i16_local
>> +; FUNC-LABEL: load_v2i16_local
>> ; R600-CHECK: LDS_USHORT_READ_RET
>> ; R600-CHECK: LDS_USHORT_READ_RET
>> ; SI-CHECK-NOT: S_WQM_B64
>> @@ -605,7 +605,7 @@ entry:
>> ret void
>> }
>>
>> -; FUNC-LABEL: @load_v2i16_sext_local
>> +; FUNC-LABEL: load_v2i16_sext_local
>> ; R600-CHECK-DAG: LDS_USHORT_READ_RET
>> ; R600-CHECK-DAG: LDS_USHORT_READ_RET
>> ; R600-CHECK-DAG: ASHR
>> @@ -622,7 +622,7 @@ entry:
>> ret void
>> }
>>
>> -; FUNC-LABEL: @load_v4i16_local
>> +; FUNC-LABEL: load_v4i16_local
>> ; R600-CHECK: LDS_USHORT_READ_RET
>> ; R600-CHECK: LDS_USHORT_READ_RET
>> ; R600-CHECK: LDS_USHORT_READ_RET
>> @@ -641,7 +641,7 @@ entry:
>> ret void
>> }
>>
>> -; FUNC-LABEL: @load_v4i16_sext_local
>> +; FUNC-LABEL: load_v4i16_sext_local
>> ; R600-CHECK-DAG: LDS_USHORT_READ_RET
>> ; R600-CHECK-DAG: LDS_USHORT_READ_RET
>> ; R600-CHECK-DAG: LDS_USHORT_READ_RET
>> @@ -665,7 +665,7 @@ entry:
>> }
>>
>> ; load an i32 value from the local address space.
>> -; FUNC-LABEL: @load_i32_local
>> +; FUNC-LABEL: load_i32_local
>> ; R600-CHECK: LDS_READ_RET
>> ; SI-CHECK-NOT: S_WQM_B64
>> ; SI-CHECK: S_MOV_B32 m0
>> @@ -678,7 +678,7 @@ entry:
>> }
>>
>> ; load a f32 value from the local address space.
>> -; FUNC-LABEL: @load_f32_local
>> +; FUNC-LABEL: load_f32_local
>> ; R600-CHECK: LDS_READ_RET
>> ; SI-CHECK: S_MOV_B32 m0
>> ; SI-CHECK: DS_READ_B32
>> @@ -690,7 +690,7 @@ entry:
>> }
>>
>> ; load a v2f32 value from the local address space
>> -; FUNC-LABEL: @load_v2f32_local
>> +; FUNC-LABEL: load_v2f32_local
>> ; R600-CHECK: LDS_READ_RET
>> ; R600-CHECK: LDS_READ_RET
>> ; SI-CHECK: S_MOV_B32 m0
>> diff --git a/test/CodeGen/R600/load.vec.ll b/test/CodeGen/R600/load.vec.ll
>> index 81a6310..eae1359 100644
>> --- a/test/CodeGen/R600/load.vec.ll
>> +++ b/test/CodeGen/R600/load.vec.ll
>> @@ -2,9 +2,9 @@
>> ; RUN: llc < %s -march=r600 -mcpu=SI -verify-machineinstrs | FileCheck --check-prefix=SI-CHECK %s
>>
>> ; load a v2i32 value from the global address space.
>> -; EG-CHECK: @load_v2i32
>> +; EG-CHECK: load_v2i32
>> ; EG-CHECK: VTX_READ_64 T{{[0-9]+}}.XY, T{{[0-9]+}}.X, 0
>> -; SI-CHECK: @load_v2i32
>> +; SI-CHECK: load_v2i32
>> ; SI-CHECK: BUFFER_LOAD_DWORDX2 v[{{[0-9]+:[0-9]+}}]
>> define void @load_v2i32(<2 x i32> addrspace(1)* %out, <2 x i32> addrspace(1)* %in) {
>> %a = load <2 x i32> addrspace(1) * %in
>> @@ -13,9 +13,9 @@ define void @load_v2i32(<2 x i32> addrspace(1)* %out, <2 x i32> addrspace(1)* %i
>> }
>>
>> ; load a v4i32 value from the global address space.
>> -; EG-CHECK: @load_v4i32
>> +; EG-CHECK: load_v4i32
>> ; EG-CHECK: VTX_READ_128 T{{[0-9]+}}.XYZW, T{{[0-9]+}}.X, 0
>> -; SI-CHECK: @load_v4i32
>> +; SI-CHECK: load_v4i32
>> ; SI-CHECK: BUFFER_LOAD_DWORDX4 v[{{[0-9]+:[0-9]+}}]
>> define void @load_v4i32(<4 x i32> addrspace(1)* %out, <4 x i32> addrspace(1)* %in) {
>> %a = load <4 x i32> addrspace(1) * %in
>> diff --git a/test/CodeGen/R600/load64.ll b/test/CodeGen/R600/load64.ll
>> index a117557..f287e85 100644
>> --- a/test/CodeGen/R600/load64.ll
>> +++ b/test/CodeGen/R600/load64.ll
>> @@ -1,7 +1,7 @@
>> ; RUN: llc < %s -march=r600 -mcpu=tahiti -verify-machineinstrs | FileCheck %s
>>
>> ; load a f64 value from the global address space.
>> -; CHECK-LABEL: @load_f64:
>> +; CHECK-LABEL: load_f64:
>> ; CHECK: BUFFER_LOAD_DWORDX2 v[{{[0-9]+:[0-9]+}}]
>> ; CHECK: BUFFER_STORE_DWORDX2 v[{{[0-9]+:[0-9]+}}]
>> define void @load_f64(double addrspace(1)* %out, double addrspace(1)* %in) {
>> @@ -10,7 +10,7 @@ define void @load_f64(double addrspace(1)* %out, double addrspace(1)* %in) {
>> ret void
>> }
>>
>> -; CHECK-LABEL: @load_i64:
>> +; CHECK-LABEL: load_i64:
>> ; CHECK: BUFFER_LOAD_DWORDX2 v[{{[0-9]+:[0-9]+}}]
>> ; CHECK: BUFFER_STORE_DWORDX2 v[{{[0-9]+:[0-9]+}}]
>> define void @load_i64(i64 addrspace(1)* %out, i64 addrspace(1)* %in) {
>> @@ -20,7 +20,7 @@ define void @load_i64(i64 addrspace(1)* %out, i64 addrspace(1)* %in) {
>> }
>>
>> ; Load a f64 value from the constant address space.
>> -; CHECK-LABEL: @load_const_addrspace_f64:
>> +; CHECK-LABEL: load_const_addrspace_f64:
>> ; CHECK: S_LOAD_DWORDX2 s[{{[0-9]+:[0-9]+}}]
>> ; CHECK: BUFFER_STORE_DWORDX2 v[{{[0-9]+:[0-9]+}}]
>> define void @load_const_addrspace_f64(double addrspace(1)* %out, double addrspace(2)* %in) {
>> diff --git a/test/CodeGen/R600/local-64.ll b/test/CodeGen/R600/local-64.ll
>> index ef48eef..3993e29 100644
>> --- a/test/CodeGen/R600/local-64.ll
>> +++ b/test/CodeGen/R600/local-64.ll
>> @@ -1,7 +1,7 @@
>> ; RUN: llc -march=r600 -mcpu=SI -verify-machineinstrs< %s | FileCheck --check-prefix=SI --check-prefix=BOTH %s
>> ; RUN: llc -march=r600 -mcpu=bonaire -verify-machineinstrs< %s | FileCheck --check-prefix=CI --check-prefix=BOTH %s
>>
>> -; BOTH-LABEL: @local_i32_load
>> +; BOTH-LABEL: local_i32_load
>> ; BOTH: DS_READ_B32 [[REG:v[0-9]+]], v{{[0-9]+}}, 0x1c, [M0]
>> ; BOTH: BUFFER_STORE_DWORD [[REG]],
>> define void @local_i32_load(i32 addrspace(1)* %out, i32 addrspace(3)* %in) nounwind {
>> @@ -11,7 +11,7 @@ define void @local_i32_load(i32 addrspace(1)* %out, i32 addrspace(3)* %in) nounw
>> ret void
>> }
>>
>> -; BOTH-LABEL: @local_i32_load_0_offset
>> +; BOTH-LABEL: local_i32_load_0_offset
>> ; BOTH: DS_READ_B32 [[REG:v[0-9]+]], v{{[0-9]+}}, 0x0, [M0]
>> ; BOTH: BUFFER_STORE_DWORD [[REG]],
>> define void @local_i32_load_0_offset(i32 addrspace(1)* %out, i32 addrspace(3)* %in) nounwind {
>> @@ -20,7 +20,7 @@ define void @local_i32_load_0_offset(i32 addrspace(1)* %out, i32 addrspace(3)* %
>> ret void
>> }
>>
>> -; BOTH-LABEL: @local_i8_load_i16_max_offset
>> +; BOTH-LABEL: local_i8_load_i16_max_offset
>> ; BOTH-NOT: ADD
>> ; BOTH: DS_READ_U8 [[REG:v[0-9]+]], {{v[0-9]+}}, 0xffff, [M0]
>> ; BOTH: BUFFER_STORE_BYTE [[REG]],
>> @@ -31,7 +31,7 @@ define void @local_i8_load_i16_max_offset(i8 addrspace(1)* %out, i8 addrspace(3)
>> ret void
>> }
>>
>> -; BOTH-LABEL: @local_i8_load_over_i16_max_offset
>> +; BOTH-LABEL: local_i8_load_over_i16_max_offset
>> ; The LDS offset will be 65536 bytes, which is larger than the size of LDS on
>> ; SI, which is why it is being OR'd with the base pointer.
>> ; SI: S_OR_B32 [[ADDR:s[0-9]+]], s{{[0-9]+}}, 0x10000
>> @@ -46,7 +46,7 @@ define void @local_i8_load_over_i16_max_offset(i8 addrspace(1)* %out, i8 addrspa
>> ret void
>> }
>>
>> -; BOTH-LABEL: @local_i64_load
>> +; BOTH-LABEL: local_i64_load
>> ; BOTH-NOT: ADD
>> ; BOTH: DS_READ_B64 [[REG:v[[0-9]+:[0-9]+]]], v{{[0-9]+}}, 0x38, [M0]
>> ; BOTH: BUFFER_STORE_DWORDX2 [[REG]],
>> @@ -57,7 +57,7 @@ define void @local_i64_load(i64 addrspace(1)* %out, i64 addrspace(3)* %in) nounw
>> ret void
>> }
>>
>> -; BOTH-LABEL: @local_i64_load_0_offset
>> +; BOTH-LABEL: local_i64_load_0_offset
>> ; BOTH: DS_READ_B64 [[REG:v\[[0-9]+:[0-9]+\]]], v{{[0-9]+}}, 0x0, [M0]
>> ; BOTH: BUFFER_STORE_DWORDX2 [[REG]],
>> define void @local_i64_load_0_offset(i64 addrspace(1)* %out, i64 addrspace(3)* %in) nounwind {
>> @@ -66,7 +66,7 @@ define void @local_i64_load_0_offset(i64 addrspace(1)* %out, i64 addrspace(3)* %
>> ret void
>> }
>>
>> -; BOTH-LABEL: @local_f64_load
>> +; BOTH-LABEL: local_f64_load
>> ; BOTH-NOT: ADD
>> ; BOTH: DS_READ_B64 [[REG:v[[0-9]+:[0-9]+]]], v{{[0-9]+}}, 0x38, [M0]
>> ; BOTH: BUFFER_STORE_DWORDX2 [[REG]],
>> @@ -77,7 +77,7 @@ define void @local_f64_load(double addrspace(1)* %out, double addrspace(3)* %in)
>> ret void
>> }
>>
>> -; BOTH-LABEL: @local_f64_load_0_offset
>> +; BOTH-LABEL: local_f64_load_0_offset
>> ; BOTH: DS_READ_B64 [[REG:v\[[0-9]+:[0-9]+\]]], v{{[0-9]+}}, 0x0, [M0]
>> ; BOTH: BUFFER_STORE_DWORDX2 [[REG]],
>> define void @local_f64_load_0_offset(double addrspace(1)* %out, double addrspace(3)* %in) nounwind {
>> @@ -86,7 +86,7 @@ define void @local_f64_load_0_offset(double addrspace(1)* %out, double addrspace
>> ret void
>> }
>>
>> -; BOTH-LABEL: @local_i64_store
>> +; BOTH-LABEL: local_i64_store
>> ; BOTH-NOT: ADD
>> ; BOTH: DS_WRITE_B64 v{{[0-9]+}}, {{v\[[0-9]+:[0-9]+\]}}, 0x38 [M0]
>> define void @local_i64_store(i64 addrspace(3)* %out) nounwind {
>> @@ -95,7 +95,7 @@ define void @local_i64_store(i64 addrspace(3)* %out) nounwind {
>> ret void
>> }
>>
>> -; BOTH-LABEL: @local_i64_store_0_offset
>> +; BOTH-LABEL: local_i64_store_0_offset
>> ; BOTH-NOT: ADD
>> ; BOTH: DS_WRITE_B64 v{{[0-9]+}}, {{v\[[0-9]+:[0-9]+\]}}, 0x0 [M0]
>> define void @local_i64_store_0_offset(i64 addrspace(3)* %out) nounwind {
>> @@ -103,7 +103,7 @@ define void @local_i64_store_0_offset(i64 addrspace(3)* %out) nounwind {
>> ret void
>> }
>>
>> -; BOTH-LABEL: @local_f64_store
>> +; BOTH-LABEL: local_f64_store
>> ; BOTH-NOT: ADD
>> ; BOTH: DS_WRITE_B64 v{{[0-9]+}}, {{v\[[0-9]+:[0-9]+\]}}, 0x38 [M0]
>> define void @local_f64_store(double addrspace(3)* %out) nounwind {
>> @@ -112,14 +112,14 @@ define void @local_f64_store(double addrspace(3)* %out) nounwind {
>> ret void
>> }
>>
>> -; BOTH-LABEL: @local_f64_store_0_offset
>> +; BOTH-LABEL: local_f64_store_0_offset
>> ; BOTH: DS_WRITE_B64 v{{[0-9]+}}, {{v\[[0-9]+:[0-9]+\]}}, 0x0 [M0]
>> define void @local_f64_store_0_offset(double addrspace(3)* %out) nounwind {
>> store double 20.0, double addrspace(3)* %out, align 8
>> ret void
>> }
>>
>> -; BOTH-LABEL: @local_v2i64_store
>> +; BOTH-LABEL: local_v2i64_store
>> ; BOTH-NOT: ADD
>> ; BOTH-DAG: DS_WRITE_B64 v{{[0-9]+}}, {{v\[[0-9]+:[0-9]+\]}}, 0x78 [M0]
>> ; BOTH-DAG: DS_WRITE_B64 v{{[0-9]+}}, {{v\[[0-9]+:[0-9]+\]}}, 0x70 [M0]
>> @@ -129,7 +129,7 @@ define void @local_v2i64_store(<2 x i64> addrspace(3)* %out) nounwind {
>> ret void
>> }
>>
>> -; BOTH-LABEL: @local_v2i64_store_0_offset
>> +; BOTH-LABEL: local_v2i64_store_0_offset
>> ; BOTH-NOT: ADD
>> ; BOTH-DAG: DS_WRITE_B64 v{{[0-9]+}}, {{v\[[0-9]+:[0-9]+\]}}, 0x8 [M0]
>> ; BOTH-DAG: DS_WRITE_B64 v{{[0-9]+}}, {{v\[[0-9]+:[0-9]+\]}}, 0x0 [M0]
>> @@ -138,7 +138,7 @@ define void @local_v2i64_store_0_offset(<2 x i64> addrspace(3)* %out) nounwind {
>> ret void
>> }
>>
>> -; BOTH-LABEL: @local_v4i64_store
>> +; BOTH-LABEL: local_v4i64_store
>> ; BOTH-NOT: ADD
>> ; BOTH-DAG: DS_WRITE_B64 v{{[0-9]+}}, {{v\[[0-9]+:[0-9]+\]}}, 0xf8 [M0]
>> ; BOTH-DAG: DS_WRITE_B64 v{{[0-9]+}}, {{v\[[0-9]+:[0-9]+\]}}, 0xf0 [M0]
>> @@ -150,7 +150,7 @@ define void @local_v4i64_store(<4 x i64> addrspace(3)* %out) nounwind {
>> ret void
>> }
>>
>> -; BOTH-LABEL: @local_v4i64_store_0_offset
>> +; BOTH-LABEL: local_v4i64_store_0_offset
>> ; BOTH-NOT: ADD
>> ; BOTH-DAG: DS_WRITE_B64 v{{[0-9]+}}, {{v\[[0-9]+:[0-9]+\]}}, 0x18 [M0]
>> ; BOTH-DAG: DS_WRITE_B64 v{{[0-9]+}}, {{v\[[0-9]+:[0-9]+\]}}, 0x10 [M0]
>> diff --git a/test/CodeGen/R600/local-atomics.ll b/test/CodeGen/R600/local-atomics.ll
>> index 65802e9..5a9784c 100644
>> --- a/test/CodeGen/R600/local-atomics.ll
>> +++ b/test/CodeGen/R600/local-atomics.ll
>> @@ -2,7 +2,7 @@
>> ; RUN: llc -march=r600 -mcpu=bonaire -verify-machineinstrs < %s | FileCheck -check-prefix=CI -check-prefix=FUNC %s
>> ; RUN: llc -march=r600 -mcpu=redwood -verify-machineinstrs < %s | FileCheck -check-prefix=EG -check-prefix=FUNC %s
>>
>> -; FUNC-LABEL: @lds_atomic_xchg_ret_i32:
>> +; FUNC-LABEL: lds_atomic_xchg_ret_i32:
>> ; EG: LDS_WRXCHG_RET *
>> ; SI: S_LOAD_DWORD [[SPTR:s[0-9]+]],
>> ; SI: V_MOV_B32_e32 [[DATA:v[0-9]+]], 4
>> @@ -16,7 +16,7 @@ define void @lds_atomic_xchg_ret_i32(i32 addrspace(1)* %out, i32 addrspace(3)* %
>> ret void
>> }
>>
>> -; FUNC-LABEL: @lds_atomic_xchg_ret_i32_offset:
>> +; FUNC-LABEL: lds_atomic_xchg_ret_i32_offset:
>> ; EG: LDS_WRXCHG_RET *
>> ; SI: DS_WRXCHG_RTN_B32 v{{[0-9]+}}, v{{[0-9]+}}, v{{[0-9]+}}, 0x10
>> ; SI: S_ENDPGM
>> @@ -28,7 +28,7 @@ define void @lds_atomic_xchg_ret_i32_offset(i32 addrspace(1)* %out, i32 addrspac
>> }
>>
>> ; XXX - Is it really necessary to load 4 into VGPR?
>> -; FUNC-LABEL: @lds_atomic_add_ret_i32:
>> +; FUNC-LABEL: lds_atomic_add_ret_i32:
>> ; EG: LDS_ADD_RET *
>> ; SI: S_LOAD_DWORD [[SPTR:s[0-9]+]],
>> ; SI: V_MOV_B32_e32 [[DATA:v[0-9]+]], 4
>> @@ -42,7 +42,7 @@ define void @lds_atomic_add_ret_i32(i32 addrspace(1)* %out, i32 addrspace(3)* %p
>> ret void
>> }
>>
>> -; FUNC-LABEL: @lds_atomic_add_ret_i32_offset:
>> +; FUNC-LABEL: lds_atomic_add_ret_i32_offset:
>> ; EG: LDS_ADD_RET *
>> ; SI: DS_ADD_RTN_U32 v{{[0-9]+}}, v{{[0-9]+}}, v{{[0-9]+}}, 0x10
>> ; SI: S_ENDPGM
>> @@ -53,7 +53,7 @@ define void @lds_atomic_add_ret_i32_offset(i32 addrspace(1)* %out, i32 addrspace
>> ret void
>> }
>>
>> -; FUNC-LABEL: @lds_atomic_add_ret_i32_bad_si_offset
>> +; FUNC-LABEL: lds_atomic_add_ret_i32_bad_si_offset
>> ; EG: LDS_ADD_RET *
>> ; SI: DS_ADD_RTN_U32 v{{[0-9]+}}, v{{[0-9]+}}, v{{[0-9]+}}, 0x0
>> ; CI: DS_ADD_RTN_U32 v{{[0-9]+}}, v{{[0-9]+}}, v{{[0-9]+}}, 0x10
>> @@ -67,7 +67,7 @@ define void @lds_atomic_add_ret_i32_bad_si_offset(i32 addrspace(1)* %out, i32 ad
>> ret void
>> }
>>
>> -; FUNC-LABEL: @lds_atomic_inc_ret_i32:
>> +; FUNC-LABEL: lds_atomic_inc_ret_i32:
>> ; EG: LDS_ADD_RET *
>> ; SI: S_MOV_B32 [[SNEGONE:s[0-9]+]], -1
>> ; SI: V_MOV_B32_e32 [[NEGONE:v[0-9]+]], [[SNEGONE]]
>> @@ -79,7 +79,7 @@ define void @lds_atomic_inc_ret_i32(i32 addrspace(1)* %out, i32 addrspace(3)* %p
>> ret void
>> }
>>
>> -; FUNC-LABEL: @lds_atomic_inc_ret_i32_offset:
>> +; FUNC-LABEL: lds_atomic_inc_ret_i32_offset:
>> ; EG: LDS_ADD_RET *
>> ; SI: S_MOV_B32 [[SNEGONE:s[0-9]+]], -1
>> ; SI: V_MOV_B32_e32 [[NEGONE:v[0-9]+]], [[SNEGONE]]
>> @@ -92,7 +92,7 @@ define void @lds_atomic_inc_ret_i32_offset(i32 addrspace(1)* %out, i32 addrspace
>> ret void
>> }
>>
>> -; FUNC-LABEL: @lds_atomic_inc_ret_i32_bad_si_offset:
>> +; FUNC-LABEL: lds_atomic_inc_ret_i32_bad_si_offset:
>> ; EG: LDS_ADD_RET *
>> ; SI: DS_INC_RTN_U32 v{{[0-9]+}}, v{{[0-9]+}}, v{{[0-9]+}}, 0x0
>> ; CI: DS_INC_RTN_U32 v{{[0-9]+}}, v{{[0-9]+}}, v{{[0-9]+}}, 0x10
>> @@ -106,7 +106,7 @@ define void @lds_atomic_inc_ret_i32_bad_si_offset(i32 addrspace(1)* %out, i32 ad
>> ret void
>> }
>>
>> -; FUNC-LABEL: @lds_atomic_sub_ret_i32:
>> +; FUNC-LABEL: lds_atomic_sub_ret_i32:
>> ; EG: LDS_SUB_RET *
>> ; SI: DS_SUB_RTN_U32
>> ; SI: S_ENDPGM
>> @@ -116,7 +116,7 @@ define void @lds_atomic_sub_ret_i32(i32 addrspace(1)* %out, i32 addrspace(3)* %p
>> ret void
>> }
>>
>> -; FUNC-LABEL: @lds_atomic_sub_ret_i32_offset:
>> +; FUNC-LABEL: lds_atomic_sub_ret_i32_offset:
>> ; EG: LDS_SUB_RET *
>> ; SI: DS_SUB_RTN_U32 v{{[0-9]+}}, v{{[0-9]+}}, v{{[0-9]+}}, 0x10
>> ; SI: S_ENDPGM
>> @@ -127,7 +127,7 @@ define void @lds_atomic_sub_ret_i32_offset(i32 addrspace(1)* %out, i32 addrspace
>> ret void
>> }
>>
>> -; FUNC-LABEL: @lds_atomic_dec_ret_i32:
>> +; FUNC-LABEL: lds_atomic_dec_ret_i32:
>> ; EG: LDS_SUB_RET *
>> ; SI: S_MOV_B32 [[SNEGONE:s[0-9]+]], -1
>> ; SI: V_MOV_B32_e32 [[NEGONE:v[0-9]+]], [[SNEGONE]]
>> @@ -139,7 +139,7 @@ define void @lds_atomic_dec_ret_i32(i32 addrspace(1)* %out, i32 addrspace(3)* %p
>> ret void
>> }
>>
>> -; FUNC-LABEL: @lds_atomic_dec_ret_i32_offset:
>> +; FUNC-LABEL: lds_atomic_dec_ret_i32_offset:
>> ; EG: LDS_SUB_RET *
>> ; SI: S_MOV_B32 [[SNEGONE:s[0-9]+]], -1
>> ; SI: V_MOV_B32_e32 [[NEGONE:v[0-9]+]], [[SNEGONE]]
>> @@ -152,7 +152,7 @@ define void @lds_atomic_dec_ret_i32_offset(i32 addrspace(1)* %out, i32 addrspace
>> ret void
>> }
>>
>> -; FUNC-LABEL: @lds_atomic_and_ret_i32:
>> +; FUNC-LABEL: lds_atomic_and_ret_i32:
>> ; EG: LDS_AND_RET *
>> ; SI: DS_AND_RTN_B32
>> ; SI: S_ENDPGM
>> @@ -162,7 +162,7 @@ define void @lds_atomic_and_ret_i32(i32 addrspace(1)* %out, i32 addrspace(3)* %p
>> ret void
>> }
>>
>> -; FUNC-LABEL: @lds_atomic_and_ret_i32_offset:
>> +; FUNC-LABEL: lds_atomic_and_ret_i32_offset:
>> ; EG: LDS_AND_RET *
>> ; SI: DS_AND_RTN_B32 v{{[0-9]+}}, v{{[0-9]+}}, v{{[0-9]+}}, 0x10
>> ; SI: S_ENDPGM
>> @@ -173,7 +173,7 @@ define void @lds_atomic_and_ret_i32_offset(i32 addrspace(1)* %out, i32 addrspace
>> ret void
>> }
>>
>> -; FUNC-LABEL: @lds_atomic_or_ret_i32:
>> +; FUNC-LABEL: lds_atomic_or_ret_i32:
>> ; EG: LDS_OR_RET *
>> ; SI: DS_OR_RTN_B32
>> ; SI: S_ENDPGM
>> @@ -183,7 +183,7 @@ define void @lds_atomic_or_ret_i32(i32 addrspace(1)* %out, i32 addrspace(3)* %pt
>> ret void
>> }
>>
>> -; FUNC-LABEL: @lds_atomic_or_ret_i32_offset:
>> +; FUNC-LABEL: lds_atomic_or_ret_i32_offset:
>> ; EG: LDS_OR_RET *
>> ; SI: DS_OR_RTN_B32 v{{[0-9]+}}, v{{[0-9]+}}, v{{[0-9]+}}, 0x10
>> ; SI: S_ENDPGM
>> @@ -194,7 +194,7 @@ define void @lds_atomic_or_ret_i32_offset(i32 addrspace(1)* %out, i32 addrspace(
>> ret void
>> }
>>
>> -; FUNC-LABEL: @lds_atomic_xor_ret_i32:
>> +; FUNC-LABEL: lds_atomic_xor_ret_i32:
>> ; EG: LDS_XOR_RET *
>> ; SI: DS_XOR_RTN_B32
>> ; SI: S_ENDPGM
>> @@ -204,7 +204,7 @@ define void @lds_atomic_xor_ret_i32(i32 addrspace(1)* %out, i32 addrspace(3)* %p
>> ret void
>> }
>>
>> -; FUNC-LABEL: @lds_atomic_xor_ret_i32_offset:
>> +; FUNC-LABEL: lds_atomic_xor_ret_i32_offset:
>> ; EG: LDS_XOR_RET *
>> ; SI: DS_XOR_RTN_B32 v{{[0-9]+}}, v{{[0-9]+}}, v{{[0-9]+}}, 0x10
>> ; SI: S_ENDPGM
>> @@ -216,14 +216,14 @@ define void @lds_atomic_xor_ret_i32_offset(i32 addrspace(1)* %out, i32 addrspace
>> }
>>
>> ; FIXME: There is no atomic nand instr
>> -; XFUNC-LABEL: @lds_atomic_nand_ret_i32:uction, so we somehow need to expand this.
>> +; XFUNC-LABEL: lds_atomic_nand_ret_i32:uction, so we somehow need to expand this.
>> ; define void @lds_atomic_nand_ret_i32(i32 addrspace(1)* %out, i32 addrspace(3)* %ptr) nounwind {
>> ; %result = atomicrmw nand i32 addrspace(3)* %ptr, i32 4 seq_cst
>> ; store i32 %result, i32 addrspace(1)* %out, align 4
>> ; ret void
>> ; }
>>
>> -; FUNC-LABEL: @lds_atomic_min_ret_i32:
>> +; FUNC-LABEL: lds_atomic_min_ret_i32:
>> ; EG: LDS_MIN_INT_RET *
>> ; SI: DS_MIN_RTN_I32
>> ; SI: S_ENDPGM
>> @@ -233,7 +233,7 @@ define void @lds_atomic_min_ret_i32(i32 addrspace(1)* %out, i32 addrspace(3)* %p
>> ret void
>> }
>>
>> -; FUNC-LABEL: @lds_atomic_min_ret_i32_offset:
>> +; FUNC-LABEL: lds_atomic_min_ret_i32_offset:
>> ; EG: LDS_MIN_INT_RET *
>> ; SI: DS_MIN_RTN_I32 v{{[0-9]+}}, v{{[0-9]+}}, v{{[0-9]+}}, 0x10
>> ; SI: S_ENDPGM
>> @@ -244,7 +244,7 @@ define void @lds_atomic_min_ret_i32_offset(i32 addrspace(1)* %out, i32 addrspace
>> ret void
>> }
>>
>> -; FUNC-LABEL: @lds_atomic_max_ret_i32:
>> +; FUNC-LABEL: lds_atomic_max_ret_i32:
>> ; EG: LDS_MAX_INT_RET *
>> ; SI: DS_MAX_RTN_I32
>> ; SI: S_ENDPGM
>> @@ -254,7 +254,7 @@ define void @lds_atomic_max_ret_i32(i32 addrspace(1)* %out, i32 addrspace(3)* %p
>> ret void
>> }
>>
>> -; FUNC-LABEL: @lds_atomic_max_ret_i32_offset:
>> +; FUNC-LABEL: lds_atomic_max_ret_i32_offset:
>> ; EG: LDS_MAX_INT_RET *
>> ; SI: DS_MAX_RTN_I32 v{{[0-9]+}}, v{{[0-9]+}}, v{{[0-9]+}}, 0x10
>> ; SI: S_ENDPGM
>> @@ -265,7 +265,7 @@ define void @lds_atomic_max_ret_i32_offset(i32 addrspace(1)* %out, i32 addrspace
>> ret void
>> }
>>
>> -; FUNC-LABEL: @lds_atomic_umin_ret_i32:
>> +; FUNC-LABEL: lds_atomic_umin_ret_i32:
>> ; EG: LDS_MIN_UINT_RET *
>> ; SI: DS_MIN_RTN_U32
>> ; SI: S_ENDPGM
>> @@ -275,7 +275,7 @@ define void @lds_atomic_umin_ret_i32(i32 addrspace(1)* %out, i32 addrspace(3)* %
>> ret void
>> }
>>
>> -; FUNC-LABEL: @lds_atomic_umin_ret_i32_offset:
>> +; FUNC-LABEL: lds_atomic_umin_ret_i32_offset:
>> ; EG: LDS_MIN_UINT_RET *
>> ; SI: DS_MIN_RTN_U32 v{{[0-9]+}}, v{{[0-9]+}}, v{{[0-9]+}}, 0x10
>> ; SI: S_ENDPGM
>> @@ -286,7 +286,7 @@ define void @lds_atomic_umin_ret_i32_offset(i32 addrspace(1)* %out, i32 addrspac
>> ret void
>> }
>>
>> -; FUNC-LABEL: @lds_atomic_umax_ret_i32:
>> +; FUNC-LABEL: lds_atomic_umax_ret_i32:
>> ; EG: LDS_MAX_UINT_RET *
>> ; SI: DS_MAX_RTN_U32
>> ; SI: S_ENDPGM
>> @@ -296,7 +296,7 @@ define void @lds_atomic_umax_ret_i32(i32 addrspace(1)* %out, i32 addrspace(3)* %
>> ret void
>> }
>>
>> -; FUNC-LABEL: @lds_atomic_umax_ret_i32_offset:
>> +; FUNC-LABEL: lds_atomic_umax_ret_i32_offset:
>> ; EG: LDS_MAX_UINT_RET *
>> ; SI: DS_MAX_RTN_U32 v{{[0-9]+}}, v{{[0-9]+}}, v{{[0-9]+}}, 0x10
>> ; SI: S_ENDPGM
>> @@ -307,7 +307,7 @@ define void @lds_atomic_umax_ret_i32_offset(i32 addrspace(1)* %out, i32 addrspac
>> ret void
>> }
>>
>> -; FUNC-LABEL: @lds_atomic_xchg_noret_i32:
>> +; FUNC-LABEL: lds_atomic_xchg_noret_i32:
>> ; SI: S_LOAD_DWORD [[SPTR:s[0-9]+]],
>> ; SI: V_MOV_B32_e32 [[DATA:v[0-9]+]], 4
>> ; SI: V_MOV_B32_e32 [[VPTR:v[0-9]+]], [[SPTR]]
>> @@ -318,7 +318,7 @@ define void @lds_atomic_xchg_noret_i32(i32 addrspace(3)* %ptr) nounwind {
>> ret void
>> }
>>
>> -; FUNC-LABEL: @lds_atomic_xchg_noret_i32_offset:
>> +; FUNC-LABEL: lds_atomic_xchg_noret_i32_offset:
>> ; SI: DS_WRXCHG_RTN_B32 v{{[0-9]+}}, v{{[0-9]+}}, v{{[0-9]+}}, 0x10
>> ; SI: S_ENDPGM
>> define void @lds_atomic_xchg_noret_i32_offset(i32 addrspace(3)* %ptr) nounwind {
>> @@ -328,7 +328,7 @@ define void @lds_atomic_xchg_noret_i32_offset(i32 addrspace(3)* %ptr) nounwind {
>> }
>>
>> ; XXX - Is it really necessary to load 4 into VGPR?
>> -; FUNC-LABEL: @lds_atomic_add_noret_i32:
>> +; FUNC-LABEL: lds_atomic_add_noret_i32:
>> ; SI: S_LOAD_DWORD [[SPTR:s[0-9]+]],
>> ; SI: V_MOV_B32_e32 [[DATA:v[0-9]+]], 4
>> ; SI: V_MOV_B32_e32 [[VPTR:v[0-9]+]], [[SPTR]]
>> @@ -339,7 +339,7 @@ define void @lds_atomic_add_noret_i32(i32 addrspace(3)* %ptr) nounwind {
>> ret void
>> }
>>
>> -; FUNC-LABEL: @lds_atomic_add_noret_i32_offset:
>> +; FUNC-LABEL: lds_atomic_add_noret_i32_offset:
>> ; SI: DS_ADD_U32 v{{[0-9]+}}, v{{[0-9]+}}, 0x10
>> ; SI: S_ENDPGM
>> define void @lds_atomic_add_noret_i32_offset(i32 addrspace(3)* %ptr) nounwind {
>> @@ -348,7 +348,7 @@ define void @lds_atomic_add_noret_i32_offset(i32 addrspace(3)* %ptr) nounwind {
>> ret void
>> }
>>
>> -; FUNC-LABEL: @lds_atomic_add_noret_i32_bad_si_offset
>> +; FUNC-LABEL: lds_atomic_add_noret_i32_bad_si_offset
>> ; SI: DS_ADD_U32 v{{[0-9]+}}, v{{[0-9]+}}, 0x0
>> ; CI: DS_ADD_U32 v{{[0-9]+}}, v{{[0-9]+}}, 0x10
>> ; SI: S_ENDPGM
>> @@ -360,7 +360,7 @@ define void @lds_atomic_add_noret_i32_bad_si_offset(i32 addrspace(3)* %ptr, i32
>> ret void
>> }
>>
>> -; FUNC-LABEL: @lds_atomic_inc_noret_i32:
>> +; FUNC-LABEL: lds_atomic_inc_noret_i32:
>> ; SI: S_MOV_B32 [[SNEGONE:s[0-9]+]], -1
>> ; SI: V_MOV_B32_e32 [[NEGONE:v[0-9]+]], [[SNEGONE]]
>> ; SI: DS_INC_U32 v{{[0-9]+}}, [[NEGONE]], 0x0
>> @@ -370,7 +370,7 @@ define void @lds_atomic_inc_noret_i32(i32 addrspace(3)* %ptr) nounwind {
>> ret void
>> }
>>
>> -; FUNC-LABEL: @lds_atomic_inc_noret_i32_offset:
>> +; FUNC-LABEL: lds_atomic_inc_noret_i32_offset:
>> ; SI: S_MOV_B32 [[SNEGONE:s[0-9]+]], -1
>> ; SI: V_MOV_B32_e32 [[NEGONE:v[0-9]+]], [[SNEGONE]]
>> ; SI: DS_INC_U32 v{{[0-9]+}}, [[NEGONE]], 0x10
>> @@ -381,7 +381,7 @@ define void @lds_atomic_inc_noret_i32_offset(i32 addrspace(3)* %ptr) nounwind {
>> ret void
>> }
>>
>> -; FUNC-LABEL: @lds_atomic_inc_noret_i32_bad_si_offset:
>> +; FUNC-LABEL: lds_atomic_inc_noret_i32_bad_si_offset:
>> ; SI: DS_INC_U32 v{{[0-9]+}}, v{{[0-9]+}}, 0x0
>> ; CI: DS_INC_U32 v{{[0-9]+}}, v{{[0-9]+}}, 0x10
>> ; SI: S_ENDPGM
>> @@ -393,7 +393,7 @@ define void @lds_atomic_inc_noret_i32_bad_si_offset(i32 addrspace(3)* %ptr, i32
>> ret void
>> }
>>
>> -; FUNC-LABEL: @lds_atomic_sub_noret_i32:
>> +; FUNC-LABEL: lds_atomic_sub_noret_i32:
>> ; SI: DS_SUB_U32
>> ; SI: S_ENDPGM
>> define void @lds_atomic_sub_noret_i32(i32 addrspace(3)* %ptr) nounwind {
>> @@ -401,7 +401,7 @@ define void @lds_atomic_sub_noret_i32(i32 addrspace(3)* %ptr) nounwind {
>> ret void
>> }
>>
>> -; FUNC-LABEL: @lds_atomic_sub_noret_i32_offset:
>> +; FUNC-LABEL: lds_atomic_sub_noret_i32_offset:
>> ; SI: DS_SUB_U32 v{{[0-9]+}}, v{{[0-9]+}}, 0x10
>> ; SI: S_ENDPGM
>> define void @lds_atomic_sub_noret_i32_offset(i32 addrspace(3)* %ptr) nounwind {
>> @@ -410,7 +410,7 @@ define void @lds_atomic_sub_noret_i32_offset(i32 addrspace(3)* %ptr) nounwind {
>> ret void
>> }
>>
>> -; FUNC-LABEL: @lds_atomic_dec_noret_i32:
>> +; FUNC-LABEL: lds_atomic_dec_noret_i32:
>> ; SI: S_MOV_B32 [[SNEGONE:s[0-9]+]], -1
>> ; SI: V_MOV_B32_e32 [[NEGONE:v[0-9]+]], [[SNEGONE]]
>> ; SI: DS_DEC_U32 v{{[0-9]+}}, [[NEGONE]], 0x0
>> @@ -420,7 +420,7 @@ define void @lds_atomic_dec_noret_i32(i32 addrspace(3)* %ptr) nounwind {
>> ret void
>> }
>>
>> -; FUNC-LABEL: @lds_atomic_dec_noret_i32_offset:
>> +; FUNC-LABEL: lds_atomic_dec_noret_i32_offset:
>> ; SI: S_MOV_B32 [[SNEGONE:s[0-9]+]], -1
>> ; SI: V_MOV_B32_e32 [[NEGONE:v[0-9]+]], [[SNEGONE]]
>> ; SI: DS_DEC_U32 v{{[0-9]+}}, [[NEGONE]], 0x10
>> @@ -431,7 +431,7 @@ define void @lds_atomic_dec_noret_i32_offset(i32 addrspace(3)* %ptr) nounwind {
>> ret void
>> }
>>
>> -; FUNC-LABEL: @lds_atomic_and_noret_i32:
>> +; FUNC-LABEL: lds_atomic_and_noret_i32:
>> ; SI: DS_AND_B32
>> ; SI: S_ENDPGM
>> define void @lds_atomic_and_noret_i32(i32 addrspace(3)* %ptr) nounwind {
>> @@ -439,7 +439,7 @@ define void @lds_atomic_and_noret_i32(i32 addrspace(3)* %ptr) nounwind {
>> ret void
>> }
>>
>> -; FUNC-LABEL: @lds_atomic_and_noret_i32_offset:
>> +; FUNC-LABEL: lds_atomic_and_noret_i32_offset:
>> ; SI: DS_AND_B32 v{{[0-9]+}}, v{{[0-9]+}}, 0x10
>> ; SI: S_ENDPGM
>> define void @lds_atomic_and_noret_i32_offset(i32 addrspace(3)* %ptr) nounwind {
>> @@ -448,7 +448,7 @@ define void @lds_atomic_and_noret_i32_offset(i32 addrspace(3)* %ptr) nounwind {
>> ret void
>> }
>>
>> -; FUNC-LABEL: @lds_atomic_or_noret_i32:
>> +; FUNC-LABEL: lds_atomic_or_noret_i32:
>> ; SI: DS_OR_B32
>> ; SI: S_ENDPGM
>> define void @lds_atomic_or_noret_i32(i32 addrspace(3)* %ptr) nounwind {
>> @@ -456,7 +456,7 @@ define void @lds_atomic_or_noret_i32(i32 addrspace(3)* %ptr) nounwind {
>> ret void
>> }
>>
>> -; FUNC-LABEL: @lds_atomic_or_noret_i32_offset:
>> +; FUNC-LABEL: lds_atomic_or_noret_i32_offset:
>> ; SI: DS_OR_B32 v{{[0-9]+}}, v{{[0-9]+}}, 0x10
>> ; SI: S_ENDPGM
>> define void @lds_atomic_or_noret_i32_offset(i32 addrspace(3)* %ptr) nounwind {
>> @@ -465,7 +465,7 @@ define void @lds_atomic_or_noret_i32_offset(i32 addrspace(3)* %ptr) nounwind {
>> ret void
>> }
>>
>> -; FUNC-LABEL: @lds_atomic_xor_noret_i32:
>> +; FUNC-LABEL: lds_atomic_xor_noret_i32:
>> ; SI: DS_XOR_B32
>> ; SI: S_ENDPGM
>> define void @lds_atomic_xor_noret_i32(i32 addrspace(3)* %ptr) nounwind {
>> @@ -473,7 +473,7 @@ define void @lds_atomic_xor_noret_i32(i32 addrspace(3)* %ptr) nounwind {
>> ret void
>> }
>>
>> -; FUNC-LABEL: @lds_atomic_xor_noret_i32_offset:
>> +; FUNC-LABEL: lds_atomic_xor_noret_i32_offset:
>> ; SI: DS_XOR_B32 v{{[0-9]+}}, v{{[0-9]+}}, 0x10
>> ; SI: S_ENDPGM
>> define void @lds_atomic_xor_noret_i32_offset(i32 addrspace(3)* %ptr) nounwind {
>> @@ -483,13 +483,13 @@ define void @lds_atomic_xor_noret_i32_offset(i32 addrspace(3)* %ptr) nounwind {
>> }
>>
>> ; FIXME: There is no atomic nand instr
>> -; XFUNC-LABEL: @lds_atomic_nand_noret_i32:uction, so we somehow need to expand this.
>> +; XFUNC-LABEL: lds_atomic_nand_noret_i32:uction, so we somehow need to expand this.
>> ; define void @lds_atomic_nand_noret_i32(i32 addrspace(3)* %ptr) nounwind {
>> ; %result = atomicrmw nand i32 addrspace(3)* %ptr, i32 4 seq_cst
>> ; ret void
>> ; }
>>
>> -; FUNC-LABEL: @lds_atomic_min_noret_i32:
>> +; FUNC-LABEL: lds_atomic_min_noret_i32:
>> ; SI: DS_MIN_I32
>> ; SI: S_ENDPGM
>> define void @lds_atomic_min_noret_i32(i32 addrspace(3)* %ptr) nounwind {
>> @@ -497,7 +497,7 @@ define void @lds_atomic_min_noret_i32(i32 addrspace(3)* %ptr) nounwind {
>> ret void
>> }
>>
>> -; FUNC-LABEL: @lds_atomic_min_noret_i32_offset:
>> +; FUNC-LABEL: lds_atomic_min_noret_i32_offset:
>> ; SI: DS_MIN_I32 v{{[0-9]+}}, v{{[0-9]+}}, 0x10
>> ; SI: S_ENDPGM
>> define void @lds_atomic_min_noret_i32_offset(i32 addrspace(3)* %ptr) nounwind {
>> @@ -506,7 +506,7 @@ define void @lds_atomic_min_noret_i32_offset(i32 addrspace(3)* %ptr) nounwind {
>> ret void
>> }
>>
>> -; FUNC-LABEL: @lds_atomic_max_noret_i32:
>> +; FUNC-LABEL: lds_atomic_max_noret_i32:
>> ; SI: DS_MAX_I32
>> ; SI: S_ENDPGM
>> define void @lds_atomic_max_noret_i32(i32 addrspace(3)* %ptr) nounwind {
>> @@ -514,7 +514,7 @@ define void @lds_atomic_max_noret_i32(i32 addrspace(3)* %ptr) nounwind {
>> ret void
>> }
>>
>> -; FUNC-LABEL: @lds_atomic_max_noret_i32_offset:
>> +; FUNC-LABEL: lds_atomic_max_noret_i32_offset:
>> ; SI: DS_MAX_I32 v{{[0-9]+}}, v{{[0-9]+}}, 0x10
>> ; SI: S_ENDPGM
>> define void @lds_atomic_max_noret_i32_offset(i32 addrspace(3)* %ptr) nounwind {
>> @@ -523,7 +523,7 @@ define void @lds_atomic_max_noret_i32_offset(i32 addrspace(3)* %ptr) nounwind {
>> ret void
>> }
>>
>> -; FUNC-LABEL: @lds_atomic_umin_noret_i32:
>> +; FUNC-LABEL: lds_atomic_umin_noret_i32:
>> ; SI: DS_MIN_U32
>> ; SI: S_ENDPGM
>> define void @lds_atomic_umin_noret_i32(i32 addrspace(3)* %ptr) nounwind {
>> @@ -531,7 +531,7 @@ define void @lds_atomic_umin_noret_i32(i32 addrspace(3)* %ptr) nounwind {
>> ret void
>> }
>>
>> -; FUNC-LABEL: @lds_atomic_umin_noret_i32_offset:
>> +; FUNC-LABEL: lds_atomic_umin_noret_i32_offset:
>> ; SI: DS_MIN_U32 v{{[0-9]+}}, v{{[0-9]+}}, 0x10
>> ; SI: S_ENDPGM
>> define void @lds_atomic_umin_noret_i32_offset(i32 addrspace(3)* %ptr) nounwind {
>> @@ -540,7 +540,7 @@ define void @lds_atomic_umin_noret_i32_offset(i32 addrspace(3)* %ptr) nounwind {
>> ret void
>> }
>>
>> -; FUNC-LABEL: @lds_atomic_umax_noret_i32:
>> +; FUNC-LABEL: lds_atomic_umax_noret_i32:
>> ; SI: DS_MAX_U32
>> ; SI: S_ENDPGM
>> define void @lds_atomic_umax_noret_i32(i32 addrspace(3)* %ptr) nounwind {
>> @@ -548,7 +548,7 @@ define void @lds_atomic_umax_noret_i32(i32 addrspace(3)* %ptr) nounwind {
>> ret void
>> }
>>
>> -; FUNC-LABEL: @lds_atomic_umax_noret_i32_offset:
>> +; FUNC-LABEL: lds_atomic_umax_noret_i32_offset:
>> ; SI: DS_MAX_U32 v{{[0-9]+}}, v{{[0-9]+}}, 0x10
>> ; SI: S_ENDPGM
>> define void @lds_atomic_umax_noret_i32_offset(i32 addrspace(3)* %ptr) nounwind {
>> diff --git a/test/CodeGen/R600/local-atomics64.ll b/test/CodeGen/R600/local-atomics64.ll
>> index 975fed2..4999bb8 100644
>> --- a/test/CodeGen/R600/local-atomics64.ll
>> +++ b/test/CodeGen/R600/local-atomics64.ll
>> @@ -1,6 +1,6 @@
>> ; RUN: llc -march=r600 -mcpu=SI -verify-machineinstrs < %s | FileCheck -check-prefix=SI %s
>>
>> -; FUNC-LABEL: @lds_atomic_xchg_ret_i64:
>> +; FUNC-LABEL: lds_atomic_xchg_ret_i64:
>> ; SI: DS_WRXCHG_RTN_B64
>> ; SI: S_ENDPGM
>> define void @lds_atomic_xchg_ret_i64(i64 addrspace(1)* %out, i64 addrspace(3)* %ptr) nounwind {
>> @@ -9,7 +9,7 @@ define void @lds_atomic_xchg_ret_i64(i64 addrspace(1)* %out, i64 addrspace(3)* %
>> ret void
>> }
>>
>> -; FUNC-LABEL: @lds_atomic_xchg_ret_i64_offset:
>> +; FUNC-LABEL: lds_atomic_xchg_ret_i64_offset:
>> ; SI: DS_WRXCHG_RTN_B64 {{.*}} 0x20
>> ; SI: S_ENDPGM
>> define void @lds_atomic_xchg_ret_i64_offset(i64 addrspace(1)* %out, i64 addrspace(3)* %ptr) nounwind {
>> @@ -19,7 +19,7 @@ define void @lds_atomic_xchg_ret_i64_offset(i64 addrspace(1)* %out, i64 addrspac
>> ret void
>> }
>>
>> -; FUNC-LABEL: @lds_atomic_add_ret_i64:
>> +; FUNC-LABEL: lds_atomic_add_ret_i64:
>> ; SI: DS_ADD_RTN_U64
>> ; SI: S_ENDPGM
>> define void @lds_atomic_add_ret_i64(i64 addrspace(1)* %out, i64 addrspace(3)* %ptr) nounwind {
>> @@ -28,7 +28,7 @@ define void @lds_atomic_add_ret_i64(i64 addrspace(1)* %out, i64 addrspace(3)* %p
>> ret void
>> }
>>
>> -; FUNC-LABEL: @lds_atomic_add_ret_i64_offset:
>> +; FUNC-LABEL: lds_atomic_add_ret_i64_offset:
>> ; SI: S_LOAD_DWORD [[PTR:s[0-9]+]], s{{\[[0-9]+:[0-9]+\]}}, 0xb
>> ; SI: S_MOV_B64 s{{\[}}[[LOSDATA:[0-9]+]]:[[HISDATA:[0-9]+]]{{\]}}, 9
>> ; SI-DAG: V_MOV_B32_e32 v[[LOVDATA:[0-9]+]], s[[LOSDATA]]
>> @@ -44,7 +44,7 @@ define void @lds_atomic_add_ret_i64_offset(i64 addrspace(1)* %out, i64 addrspace
>> ret void
>> }
>>
>> -; FUNC-LABEL: @lds_atomic_inc_ret_i64:
>> +; FUNC-LABEL: lds_atomic_inc_ret_i64:
>> ; SI: S_MOV_B64 s{{\[}}[[LOSDATA:[0-9]+]]:[[HISDATA:[0-9]+]]{{\]}}, -1
>> ; SI-DAG: V_MOV_B32_e32 v[[LOVDATA:[0-9]+]], s[[LOSDATA]]
>> ; SI-DAG: V_MOV_B32_e32 v[[HIVDATA:[0-9]+]], s[[HISDATA]]
>> @@ -57,7 +57,7 @@ define void @lds_atomic_inc_ret_i64(i64 addrspace(1)* %out, i64 addrspace(3)* %p
>> ret void
>> }
>>
>> -; FUNC-LABEL: @lds_atomic_inc_ret_i64_offset:
>> +; FUNC-LABEL: lds_atomic_inc_ret_i64_offset:
>> ; SI: DS_INC_RTN_U64 {{.*}} 0x20
>> ; SI: S_ENDPGM
>> define void @lds_atomic_inc_ret_i64_offset(i64 addrspace(1)* %out, i64 addrspace(3)* %ptr) nounwind {
>> @@ -67,7 +67,7 @@ define void @lds_atomic_inc_ret_i64_offset(i64 addrspace(1)* %out, i64 addrspace
>> ret void
>> }
>>
>> -; FUNC-LABEL: @lds_atomic_sub_ret_i64:
>> +; FUNC-LABEL: lds_atomic_sub_ret_i64:
>> ; SI: DS_SUB_RTN_U64
>> ; SI: S_ENDPGM
>> define void @lds_atomic_sub_ret_i64(i64 addrspace(1)* %out, i64 addrspace(3)* %ptr) nounwind {
>> @@ -76,7 +76,7 @@ define void @lds_atomic_sub_ret_i64(i64 addrspace(1)* %out, i64 addrspace(3)* %p
>> ret void
>> }
>>
>> -; FUNC-LABEL: @lds_atomic_sub_ret_i64_offset:
>> +; FUNC-LABEL: lds_atomic_sub_ret_i64_offset:
>> ; SI: DS_SUB_RTN_U64 {{.*}} 0x20
>> ; SI: S_ENDPGM
>> define void @lds_atomic_sub_ret_i64_offset(i64 addrspace(1)* %out, i64 addrspace(3)* %ptr) nounwind {
>> @@ -86,7 +86,7 @@ define void @lds_atomic_sub_ret_i64_offset(i64 addrspace(1)* %out, i64 addrspace
>> ret void
>> }
>>
>> -; FUNC-LABEL: @lds_atomic_dec_ret_i64:
>> +; FUNC-LABEL: lds_atomic_dec_ret_i64:
>> ; SI: S_MOV_B64 s{{\[}}[[LOSDATA:[0-9]+]]:[[HISDATA:[0-9]+]]{{\]}}, -1
>> ; SI-DAG: V_MOV_B32_e32 v[[LOVDATA:[0-9]+]], s[[LOSDATA]]
>> ; SI-DAG: V_MOV_B32_e32 v[[HIVDATA:[0-9]+]], s[[HISDATA]]
>> @@ -99,7 +99,7 @@ define void @lds_atomic_dec_ret_i64(i64 addrspace(1)* %out, i64 addrspace(3)* %p
>> ret void
>> }
>>
>> -; FUNC-LABEL: @lds_atomic_dec_ret_i64_offset:
>> +; FUNC-LABEL: lds_atomic_dec_ret_i64_offset:
>> ; SI: DS_DEC_RTN_U64 {{.*}} 0x20
>> ; SI: S_ENDPGM
>> define void @lds_atomic_dec_ret_i64_offset(i64 addrspace(1)* %out, i64 addrspace(3)* %ptr) nounwind {
>> @@ -109,7 +109,7 @@ define void @lds_atomic_dec_ret_i64_offset(i64 addrspace(1)* %out, i64 addrspace
>> ret void
>> }
>>
>> -; FUNC-LABEL: @lds_atomic_and_ret_i64:
>> +; FUNC-LABEL: lds_atomic_and_ret_i64:
>> ; SI: DS_AND_RTN_B64
>> ; SI: S_ENDPGM
>> define void @lds_atomic_and_ret_i64(i64 addrspace(1)* %out, i64 addrspace(3)* %ptr) nounwind {
>> @@ -118,7 +118,7 @@ define void @lds_atomic_and_ret_i64(i64 addrspace(1)* %out, i64 addrspace(3)* %p
>> ret void
>> }
>>
>> -; FUNC-LABEL: @lds_atomic_and_ret_i64_offset:
>> +; FUNC-LABEL: lds_atomic_and_ret_i64_offset:
>> ; SI: DS_AND_RTN_B64 {{.*}} 0x20
>> ; SI: S_ENDPGM
>> define void @lds_atomic_and_ret_i64_offset(i64 addrspace(1)* %out, i64 addrspace(3)* %ptr) nounwind {
>> @@ -128,7 +128,7 @@ define void @lds_atomic_and_ret_i64_offset(i64 addrspace(1)* %out, i64 addrspace
>> ret void
>> }
>>
>> -; FUNC-LABEL: @lds_atomic_or_ret_i64:
>> +; FUNC-LABEL: lds_atomic_or_ret_i64:
>> ; SI: DS_OR_RTN_B64
>> ; SI: S_ENDPGM
>> define void @lds_atomic_or_ret_i64(i64 addrspace(1)* %out, i64 addrspace(3)* %ptr) nounwind {
>> @@ -137,7 +137,7 @@ define void @lds_atomic_or_ret_i64(i64 addrspace(1)* %out, i64 addrspace(3)* %pt
>> ret void
>> }
>>
>> -; FUNC-LABEL: @lds_atomic_or_ret_i64_offset:
>> +; FUNC-LABEL: lds_atomic_or_ret_i64_offset:
>> ; SI: DS_OR_RTN_B64 {{.*}} 0x20
>> ; SI: S_ENDPGM
>> define void @lds_atomic_or_ret_i64_offset(i64 addrspace(1)* %out, i64 addrspace(3)* %ptr) nounwind {
>> @@ -147,7 +147,7 @@ define void @lds_atomic_or_ret_i64_offset(i64 addrspace(1)* %out, i64 addrspace(
>> ret void
>> }
>>
>> -; FUNC-LABEL: @lds_atomic_xor_ret_i64:
>> +; FUNC-LABEL: lds_atomic_xor_ret_i64:
>> ; SI: DS_XOR_RTN_B64
>> ; SI: S_ENDPGM
>> define void @lds_atomic_xor_ret_i64(i64 addrspace(1)* %out, i64 addrspace(3)* %ptr) nounwind {
>> @@ -156,7 +156,7 @@ define void @lds_atomic_xor_ret_i64(i64 addrspace(1)* %out, i64 addrspace(3)* %p
>> ret void
>> }
>>
>> -; FUNC-LABEL: @lds_atomic_xor_ret_i64_offset:
>> +; FUNC-LABEL: lds_atomic_xor_ret_i64_offset:
>> ; SI: DS_XOR_RTN_B64 {{.*}} 0x20
>> ; SI: S_ENDPGM
>> define void @lds_atomic_xor_ret_i64_offset(i64 addrspace(1)* %out, i64 addrspace(3)* %ptr) nounwind {
>> @@ -167,14 +167,14 @@ define void @lds_atomic_xor_ret_i64_offset(i64 addrspace(1)* %out, i64 addrspace
>> }
>>
>> ; FIXME: There is no atomic nand instr
>> -; XFUNC-LABEL: @lds_atomic_nand_ret_i64:uction, so we somehow need to expand this.
>> +; XFUNC-LABEL: lds_atomic_nand_ret_i64:uction, so we somehow need to expand this.
>> ; define void @lds_atomic_nand_ret_i64(i64 addrspace(1)* %out, i64 addrspace(3)* %ptr) nounwind {
>> ; %result = atomicrmw nand i64 addrspace(3)* %ptr, i32 4 seq_cst
>> ; store i64 %result, i64 addrspace(1)* %out, align 8
>> ; ret void
>> ; }
>>
>> -; FUNC-LABEL: @lds_atomic_min_ret_i64:
>> +; FUNC-LABEL: lds_atomic_min_ret_i64:
>> ; SI: DS_MIN_RTN_I64
>> ; SI: S_ENDPGM
>> define void @lds_atomic_min_ret_i64(i64 addrspace(1)* %out, i64 addrspace(3)* %ptr) nounwind {
>> @@ -183,7 +183,7 @@ define void @lds_atomic_min_ret_i64(i64 addrspace(1)* %out, i64 addrspace(3)* %p
>> ret void
>> }
>>
>> -; FUNC-LABEL: @lds_atomic_min_ret_i64_offset:
>> +; FUNC-LABEL: lds_atomic_min_ret_i64_offset:
>> ; SI: DS_MIN_RTN_I64 {{.*}} 0x20
>> ; SI: S_ENDPGM
>> define void @lds_atomic_min_ret_i64_offset(i64 addrspace(1)* %out, i64 addrspace(3)* %ptr) nounwind {
>> @@ -193,7 +193,7 @@ define void @lds_atomic_min_ret_i64_offset(i64 addrspace(1)* %out, i64 addrspace
>> ret void
>> }
>>
>> -; FUNC-LABEL: @lds_atomic_max_ret_i64:
>> +; FUNC-LABEL: lds_atomic_max_ret_i64:
>> ; SI: DS_MAX_RTN_I64
>> ; SI: S_ENDPGM
>> define void @lds_atomic_max_ret_i64(i64 addrspace(1)* %out, i64 addrspace(3)* %ptr) nounwind {
>> @@ -202,7 +202,7 @@ define void @lds_atomic_max_ret_i64(i64 addrspace(1)* %out, i64 addrspace(3)* %p
>> ret void
>> }
>>
>> -; FUNC-LABEL: @lds_atomic_max_ret_i64_offset:
>> +; FUNC-LABEL: lds_atomic_max_ret_i64_offset:
>> ; SI: DS_MAX_RTN_I64 {{.*}} 0x20
>> ; SI: S_ENDPGM
>> define void @lds_atomic_max_ret_i64_offset(i64 addrspace(1)* %out, i64 addrspace(3)* %ptr) nounwind {
>> @@ -212,7 +212,7 @@ define void @lds_atomic_max_ret_i64_offset(i64 addrspace(1)* %out, i64 addrspace
>> ret void
>> }
>>
>> -; FUNC-LABEL: @lds_atomic_umin_ret_i64:
>> +; FUNC-LABEL: lds_atomic_umin_ret_i64:
>> ; SI: DS_MIN_RTN_U64
>> ; SI: S_ENDPGM
>> define void @lds_atomic_umin_ret_i64(i64 addrspace(1)* %out, i64 addrspace(3)* %ptr) nounwind {
>> @@ -221,7 +221,7 @@ define void @lds_atomic_umin_ret_i64(i64 addrspace(1)* %out, i64 addrspace(3)* %
>> ret void
>> }
>>
>> -; FUNC-LABEL: @lds_atomic_umin_ret_i64_offset:
>> +; FUNC-LABEL: lds_atomic_umin_ret_i64_offset:
>> ; SI: DS_MIN_RTN_U64 {{.*}} 0x20
>> ; SI: S_ENDPGM
>> define void @lds_atomic_umin_ret_i64_offset(i64 addrspace(1)* %out, i64 addrspace(3)* %ptr) nounwind {
>> @@ -231,7 +231,7 @@ define void @lds_atomic_umin_ret_i64_offset(i64 addrspace(1)* %out, i64 addrspac
>> ret void
>> }
>>
>> -; FUNC-LABEL: @lds_atomic_umax_ret_i64:
>> +; FUNC-LABEL: lds_atomic_umax_ret_i64:
>> ; SI: DS_MAX_RTN_U64
>> ; SI: S_ENDPGM
>> define void @lds_atomic_umax_ret_i64(i64 addrspace(1)* %out, i64 addrspace(3)* %ptr) nounwind {
>> @@ -240,7 +240,7 @@ define void @lds_atomic_umax_ret_i64(i64 addrspace(1)* %out, i64 addrspace(3)* %
>> ret void
>> }
>>
>> -; FUNC-LABEL: @lds_atomic_umax_ret_i64_offset:
>> +; FUNC-LABEL: lds_atomic_umax_ret_i64_offset:
>> ; SI: DS_MAX_RTN_U64 {{.*}} 0x20
>> ; SI: S_ENDPGM
>> define void @lds_atomic_umax_ret_i64_offset(i64 addrspace(1)* %out, i64 addrspace(3)* %ptr) nounwind {
>> @@ -250,7 +250,7 @@ define void @lds_atomic_umax_ret_i64_offset(i64 addrspace(1)* %out, i64 addrspac
>> ret void
>> }
>>
>> -; FUNC-LABEL: @lds_atomic_xchg_noret_i64:
>> +; FUNC-LABEL: lds_atomic_xchg_noret_i64:
>> ; SI: DS_WRXCHG_RTN_B64
>> ; SI: S_ENDPGM
>> define void @lds_atomic_xchg_noret_i64(i64 addrspace(3)* %ptr) nounwind {
>> @@ -258,7 +258,7 @@ define void @lds_atomic_xchg_noret_i64(i64 addrspace(3)* %ptr) nounwind {
>> ret void
>> }
>>
>> -; FUNC-LABEL: @lds_atomic_xchg_noret_i64_offset:
>> +; FUNC-LABEL: lds_atomic_xchg_noret_i64_offset:
>> ; SI: DS_WRXCHG_RTN_B64 {{.*}} 0x20
>> ; SI: S_ENDPGM
>> define void @lds_atomic_xchg_noret_i64_offset(i64 addrspace(3)* %ptr) nounwind {
>> @@ -267,7 +267,7 @@ define void @lds_atomic_xchg_noret_i64_offset(i64 addrspace(3)* %ptr) nounwind {
>> ret void
>> }
>>
>> -; FUNC-LABEL: @lds_atomic_add_noret_i64:
>> +; FUNC-LABEL: lds_atomic_add_noret_i64:
>> ; SI: DS_ADD_U64
>> ; SI: S_ENDPGM
>> define void @lds_atomic_add_noret_i64(i64 addrspace(3)* %ptr) nounwind {
>> @@ -275,7 +275,7 @@ define void @lds_atomic_add_noret_i64(i64 addrspace(3)* %ptr) nounwind {
>> ret void
>> }
>>
>> -; FUNC-LABEL: @lds_atomic_add_noret_i64_offset:
>> +; FUNC-LABEL: lds_atomic_add_noret_i64_offset:
>> ; SI: S_LOAD_DWORD [[PTR:s[0-9]+]], s{{\[[0-9]+:[0-9]+\]}}, 0x9
>> ; SI: S_MOV_B64 s{{\[}}[[LOSDATA:[0-9]+]]:[[HISDATA:[0-9]+]]{{\]}}, 9
>> ; SI-DAG: V_MOV_B32_e32 v[[LOVDATA:[0-9]+]], s[[LOSDATA]]
>> @@ -289,7 +289,7 @@ define void @lds_atomic_add_noret_i64_offset(i64 addrspace(3)* %ptr) nounwind {
>> ret void
>> }
>>
>> -; FUNC-LABEL: @lds_atomic_inc_noret_i64:
>> +; FUNC-LABEL: lds_atomic_inc_noret_i64:
>> ; SI: S_MOV_B64 s{{\[}}[[LOSDATA:[0-9]+]]:[[HISDATA:[0-9]+]]{{\]}}, -1
>> ; SI-DAG: V_MOV_B32_e32 v[[LOVDATA:[0-9]+]], s[[LOSDATA]]
>> ; SI-DAG: V_MOV_B32_e32 v[[HIVDATA:[0-9]+]], s[[HISDATA]]
>> @@ -300,7 +300,7 @@ define void @lds_atomic_inc_noret_i64(i64 addrspace(3)* %ptr) nounwind {
>> ret void
>> }
>>
>> -; FUNC-LABEL: @lds_atomic_inc_noret_i64_offset:
>> +; FUNC-LABEL: lds_atomic_inc_noret_i64_offset:
>> ; SI: DS_INC_U64 {{.*}} 0x20
>> ; SI: S_ENDPGM
>> define void @lds_atomic_inc_noret_i64_offset(i64 addrspace(3)* %ptr) nounwind {
>> @@ -309,7 +309,7 @@ define void @lds_atomic_inc_noret_i64_offset(i64 addrspace(3)* %ptr) nounwind {
>> ret void
>> }
>>
>> -; FUNC-LABEL: @lds_atomic_sub_noret_i64:
>> +; FUNC-LABEL: lds_atomic_sub_noret_i64:
>> ; SI: DS_SUB_U64
>> ; SI: S_ENDPGM
>> define void @lds_atomic_sub_noret_i64(i64 addrspace(3)* %ptr) nounwind {
>> @@ -317,7 +317,7 @@ define void @lds_atomic_sub_noret_i64(i64 addrspace(3)* %ptr) nounwind {
>> ret void
>> }
>>
>> -; FUNC-LABEL: @lds_atomic_sub_noret_i64_offset:
>> +; FUNC-LABEL: lds_atomic_sub_noret_i64_offset:
>> ; SI: DS_SUB_U64 {{.*}} 0x20
>> ; SI: S_ENDPGM
>> define void @lds_atomic_sub_noret_i64_offset(i64 addrspace(3)* %ptr) nounwind {
>> @@ -326,7 +326,7 @@ define void @lds_atomic_sub_noret_i64_offset(i64 addrspace(3)* %ptr) nounwind {
>> ret void
>> }
>>
>> -; FUNC-LABEL: @lds_atomic_dec_noret_i64:
>> +; FUNC-LABEL: lds_atomic_dec_noret_i64:
>> ; SI: S_MOV_B64 s{{\[}}[[LOSDATA:[0-9]+]]:[[HISDATA:[0-9]+]]{{\]}}, -1
>> ; SI-DAG: V_MOV_B32_e32 v[[LOVDATA:[0-9]+]], s[[LOSDATA]]
>> ; SI-DAG: V_MOV_B32_e32 v[[HIVDATA:[0-9]+]], s[[HISDATA]]
>> @@ -337,7 +337,7 @@ define void @lds_atomic_dec_noret_i64(i64 addrspace(3)* %ptr) nounwind {
>> ret void
>> }
>>
>> -; FUNC-LABEL: @lds_atomic_dec_noret_i64_offset:
>> +; FUNC-LABEL: lds_atomic_dec_noret_i64_offset:
>> ; SI: DS_DEC_U64 {{.*}} 0x20
>> ; SI: S_ENDPGM
>> define void @lds_atomic_dec_noret_i64_offset(i64 addrspace(3)* %ptr) nounwind {
>> @@ -346,7 +346,7 @@ define void @lds_atomic_dec_noret_i64_offset(i64 addrspace(3)* %ptr) nounwind {
>> ret void
>> }
>>
>> -; FUNC-LABEL: @lds_atomic_and_noret_i64:
>> +; FUNC-LABEL: lds_atomic_and_noret_i64:
>> ; SI: DS_AND_B64
>> ; SI: S_ENDPGM
>> define void @lds_atomic_and_noret_i64(i64 addrspace(3)* %ptr) nounwind {
>> @@ -354,7 +354,7 @@ define void @lds_atomic_and_noret_i64(i64 addrspace(3)* %ptr) nounwind {
>> ret void
>> }
>>
>> -; FUNC-LABEL: @lds_atomic_and_noret_i64_offset:
>> +; FUNC-LABEL: lds_atomic_and_noret_i64_offset:
>> ; SI: DS_AND_B64 {{.*}} 0x20
>> ; SI: S_ENDPGM
>> define void @lds_atomic_and_noret_i64_offset(i64 addrspace(3)* %ptr) nounwind {
>> @@ -363,7 +363,7 @@ define void @lds_atomic_and_noret_i64_offset(i64 addrspace(3)* %ptr) nounwind {
>> ret void
>> }
>>
>> -; FUNC-LABEL: @lds_atomic_or_noret_i64:
>> +; FUNC-LABEL: lds_atomic_or_noret_i64:
>> ; SI: DS_OR_B64
>> ; SI: S_ENDPGM
>> define void @lds_atomic_or_noret_i64(i64 addrspace(3)* %ptr) nounwind {
>> @@ -371,7 +371,7 @@ define void @lds_atomic_or_noret_i64(i64 addrspace(3)* %ptr) nounwind {
>> ret void
>> }
>>
>> -; FUNC-LABEL: @lds_atomic_or_noret_i64_offset:
>> +; FUNC-LABEL: lds_atomic_or_noret_i64_offset:
>> ; SI: DS_OR_B64 {{.*}} 0x20
>> ; SI: S_ENDPGM
>> define void @lds_atomic_or_noret_i64_offset(i64 addrspace(3)* %ptr) nounwind {
>> @@ -380,7 +380,7 @@ define void @lds_atomic_or_noret_i64_offset(i64 addrspace(3)* %ptr) nounwind {
>> ret void
>> }
>>
>> -; FUNC-LABEL: @lds_atomic_xor_noret_i64:
>> +; FUNC-LABEL: lds_atomic_xor_noret_i64:
>> ; SI: DS_XOR_B64
>> ; SI: S_ENDPGM
>> define void @lds_atomic_xor_noret_i64(i64 addrspace(3)* %ptr) nounwind {
>> @@ -388,7 +388,7 @@ define void @lds_atomic_xor_noret_i64(i64 addrspace(3)* %ptr) nounwind {
>> ret void
>> }
>>
>> -; FUNC-LABEL: @lds_atomic_xor_noret_i64_offset:
>> +; FUNC-LABEL: lds_atomic_xor_noret_i64_offset:
>> ; SI: DS_XOR_B64 {{.*}} 0x20
>> ; SI: S_ENDPGM
>> define void @lds_atomic_xor_noret_i64_offset(i64 addrspace(3)* %ptr) nounwind {
>> @@ -398,13 +398,13 @@ define void @lds_atomic_xor_noret_i64_offset(i64 addrspace(3)* %ptr) nounwind {
>> }
>>
>> ; FIXME: There is no atomic nand instr
>> -; XFUNC-LABEL: @lds_atomic_nand_noret_i64:uction, so we somehow need to expand this.
>> +; XFUNC-LABEL: lds_atomic_nand_noret_i64:uction, so we somehow need to expand this.
>> ; define void @lds_atomic_nand_noret_i64(i64 addrspace(3)* %ptr) nounwind {
>> ; %result = atomicrmw nand i64 addrspace(3)* %ptr, i32 4 seq_cst
>> ; ret void
>> ; }
>>
>> -; FUNC-LABEL: @lds_atomic_min_noret_i64:
>> +; FUNC-LABEL: lds_atomic_min_noret_i64:
>> ; SI: DS_MIN_I64
>> ; SI: S_ENDPGM
>> define void @lds_atomic_min_noret_i64(i64 addrspace(3)* %ptr) nounwind {
>> @@ -412,7 +412,7 @@ define void @lds_atomic_min_noret_i64(i64 addrspace(3)* %ptr) nounwind {
>> ret void
>> }
>>
>> -; FUNC-LABEL: @lds_atomic_min_noret_i64_offset:
>> +; FUNC-LABEL: lds_atomic_min_noret_i64_offset:
>> ; SI: DS_MIN_I64 {{.*}} 0x20
>> ; SI: S_ENDPGM
>> define void @lds_atomic_min_noret_i64_offset(i64 addrspace(3)* %ptr) nounwind {
>> @@ -421,7 +421,7 @@ define void @lds_atomic_min_noret_i64_offset(i64 addrspace(3)* %ptr) nounwind {
>> ret void
>> }
>>
>> -; FUNC-LABEL: @lds_atomic_max_noret_i64:
>> +; FUNC-LABEL: lds_atomic_max_noret_i64:
>> ; SI: DS_MAX_I64
>> ; SI: S_ENDPGM
>> define void @lds_atomic_max_noret_i64(i64 addrspace(3)* %ptr) nounwind {
>> @@ -429,7 +429,7 @@ define void @lds_atomic_max_noret_i64(i64 addrspace(3)* %ptr) nounwind {
>> ret void
>> }
>>
>> -; FUNC-LABEL: @lds_atomic_max_noret_i64_offset:
>> +; FUNC-LABEL: lds_atomic_max_noret_i64_offset:
>> ; SI: DS_MAX_I64 {{.*}} 0x20
>> ; SI: S_ENDPGM
>> define void @lds_atomic_max_noret_i64_offset(i64 addrspace(3)* %ptr) nounwind {
>> @@ -438,7 +438,7 @@ define void @lds_atomic_max_noret_i64_offset(i64 addrspace(3)* %ptr) nounwind {
>> ret void
>> }
>>
>> -; FUNC-LABEL: @lds_atomic_umin_noret_i64:
>> +; FUNC-LABEL: lds_atomic_umin_noret_i64:
>> ; SI: DS_MIN_U64
>> ; SI: S_ENDPGM
>> define void @lds_atomic_umin_noret_i64(i64 addrspace(3)* %ptr) nounwind {
>> @@ -446,7 +446,7 @@ define void @lds_atomic_umin_noret_i64(i64 addrspace(3)* %ptr) nounwind {
>> ret void
>> }
>>
>> -; FUNC-LABEL: @lds_atomic_umin_noret_i64_offset:
>> +; FUNC-LABEL: lds_atomic_umin_noret_i64_offset:
>> ; SI: DS_MIN_U64 {{.*}} 0x20
>> ; SI: S_ENDPGM
>> define void @lds_atomic_umin_noret_i64_offset(i64 addrspace(3)* %ptr) nounwind {
>> @@ -455,7 +455,7 @@ define void @lds_atomic_umin_noret_i64_offset(i64 addrspace(3)* %ptr) nounwind {
>> ret void
>> }
>>
>> -; FUNC-LABEL: @lds_atomic_umax_noret_i64:
>> +; FUNC-LABEL: lds_atomic_umax_noret_i64:
>> ; SI: DS_MAX_U64
>> ; SI: S_ENDPGM
>> define void @lds_atomic_umax_noret_i64(i64 addrspace(3)* %ptr) nounwind {
>> @@ -463,7 +463,7 @@ define void @lds_atomic_umax_noret_i64(i64 addrspace(3)* %ptr) nounwind {
>> ret void
>> }
>>
>> -; FUNC-LABEL: @lds_atomic_umax_noret_i64_offset:
>> +; FUNC-LABEL: lds_atomic_umax_noret_i64_offset:
>> ; SI: DS_MAX_U64 {{.*}} 0x20
>> ; SI: S_ENDPGM
>> define void @lds_atomic_umax_noret_i64_offset(i64 addrspace(3)* %ptr) nounwind {
>> diff --git a/test/CodeGen/R600/local-memory-two-objects.ll b/test/CodeGen/R600/local-memory-two-objects.ll
>> index f0f6824..9ac927e 100644
>> --- a/test/CodeGen/R600/local-memory-two-objects.ll
>> +++ b/test/CodeGen/R600/local-memory-two-objects.ll
>> @@ -5,7 +5,7 @@
>> @local_memory_two_objects.local_mem0 = internal unnamed_addr addrspace(3) global [4 x i32] zeroinitializer, align 4
>> @local_memory_two_objects.local_mem1 = internal unnamed_addr addrspace(3) global [4 x i32] zeroinitializer, align 4
>>
>> -; EG-CHECK: @local_memory_two_objects
>> +; EG-CHECK: local_memory_two_objects
>>
>> ; Check that the LDS size emitted correctly
>> ; EG-CHECK: .long 166120
>> diff --git a/test/CodeGen/R600/local-memory.ll b/test/CodeGen/R600/local-memory.ll
>> index ece8940..c437dec 100644
>> --- a/test/CodeGen/R600/local-memory.ll
>> +++ b/test/CodeGen/R600/local-memory.ll
>> @@ -4,7 +4,7 @@
>>
>> @local_memory.local_mem = internal unnamed_addr addrspace(3) global [128 x i32] zeroinitializer, align 4
>>
>> -; FUNC-LABEL: @local_memory
>> +; FUNC-LABEL: local_memory
>>
>> ; Check that the LDS size emitted correctly
>> ; EG: .long 166120
>> diff --git a/test/CodeGen/R600/loop-idiom.ll b/test/CodeGen/R600/loop-idiom.ll
>> index 128f661..9029f48 100644
>> --- a/test/CodeGen/R600/loop-idiom.ll
>> +++ b/test/CodeGen/R600/loop-idiom.ll
>> @@ -9,9 +9,9 @@ target triple = "r600--"
>> ; Make sure loop-idiom doesn't create memcpy or memset. There are no library
>> ; implementations of these for R600.
>>
>> -; FUNC: @no_memcpy
>> -; R600-NOT: @llvm.memcpy
>> -; SI-NOT: @llvm.memcpy
>> +; FUNC: no_memcpy
>> +; R600-NOT: llvm.memcpy
>> +; SI-NOT: llvm.memcpy
>> define void @no_memcpy(i8 addrspace(3)* %in, i32 %size) {
>> entry:
>> %dest = alloca i8, i32 32
>> @@ -31,11 +31,11 @@ for.end:
>> ret void
>> }
>>
>> -; FUNC: @no_memset
>> -; R600-NOT: @llvm.memset
>> -; R600-NOT: @memset_pattern16
>> -; SI-NOT: @llvm.memset
>> -; SI-NOT: @memset_pattern16
>> +; FUNC: no_memset
>> +; R600-NOT: llvm.memset
>> +; R600-NOT: memset_pattern16
>> +; SI-NOT: llvm.memset
>> +; SI-NOT: memset_pattern16
>> define void @no_memset(i32 %size) {
>> entry:
>> %dest = alloca i8, i32 32
>> diff --git a/test/CodeGen/R600/mad-sub.ll b/test/CodeGen/R600/mad-sub.ll
>> index 08069e2..041b2b2 100644
>> --- a/test/CodeGen/R600/mad-sub.ll
>> +++ b/test/CodeGen/R600/mad-sub.ll
>> @@ -3,7 +3,7 @@
>> declare i32 @llvm.r600.read.tidig.x() #0
>> declare float @llvm.fabs.f32(float) #0
>>
>> -; FUNC-LABEL: @mad_sub_f32
>> +; FUNC-LABEL: mad_sub_f32
>> ; SI: BUFFER_LOAD_DWORD [[REGA:v[0-9]+]]
>> ; SI: BUFFER_LOAD_DWORD [[REGB:v[0-9]+]]
>> ; SI: BUFFER_LOAD_DWORD [[REGC:v[0-9]+]]
>> @@ -27,7 +27,7 @@ define void @mad_sub_f32(float addrspace(1)* noalias nocapture %out, float addrs
>> ret void
>> }
>>
>> -; FUNC-LABEL: @mad_sub_inv_f32
>> +; FUNC-LABEL: mad_sub_inv_f32
>> ; SI: BUFFER_LOAD_DWORD [[REGA:v[0-9]+]]
>> ; SI: BUFFER_LOAD_DWORD [[REGB:v[0-9]+]]
>> ; SI: BUFFER_LOAD_DWORD [[REGC:v[0-9]+]]
>> @@ -51,7 +51,7 @@ define void @mad_sub_inv_f32(float addrspace(1)* noalias nocapture %out, float a
>> ret void
>> }
>>
>> -; FUNC-LABEL: @mad_sub_f64
>> +; FUNC-LABEL: mad_sub_f64
>> ; SI: V_MUL_F64
>> ; SI: V_ADD_F64
>> define void @mad_sub_f64(double addrspace(1)* noalias nocapture %out, double addrspace(1)* noalias nocapture readonly %ptr) #1 {
>> @@ -72,7 +72,7 @@ define void @mad_sub_f64(double addrspace(1)* noalias nocapture %out, double add
>> ret void
>> }
>>
>> -; FUNC-LABEL: @mad_sub_fabs_f32
>> +; FUNC-LABEL: mad_sub_fabs_f32
>> ; SI: BUFFER_LOAD_DWORD [[REGA:v[0-9]+]]
>> ; SI: BUFFER_LOAD_DWORD [[REGB:v[0-9]+]]
>> ; SI: BUFFER_LOAD_DWORD [[REGC:v[0-9]+]]
>> @@ -97,7 +97,7 @@ define void @mad_sub_fabs_f32(float addrspace(1)* noalias nocapture %out, float
>> ret void
>> }
>>
>> -; FUNC-LABEL: @mad_sub_fabs_inv_f32
>> +; FUNC-LABEL: mad_sub_fabs_inv_f32
>> ; SI: BUFFER_LOAD_DWORD [[REGA:v[0-9]+]]
>> ; SI: BUFFER_LOAD_DWORD [[REGB:v[0-9]+]]
>> ; SI: BUFFER_LOAD_DWORD [[REGC:v[0-9]+]]
>> @@ -122,7 +122,7 @@ define void @mad_sub_fabs_inv_f32(float addrspace(1)* noalias nocapture %out, fl
>> ret void
>> }
>>
>> -; FUNC-LABEL: @neg_neg_mad_f32
>> +; FUNC-LABEL: neg_neg_mad_f32
>> ; SI: V_MAD_F32 {{v[0-9]+}}, {{v[0-9]+}}, {{v[0-9]+}}, {{v[0-9]+}}
>> define void @neg_neg_mad_f32(float addrspace(1)* noalias nocapture %out, float addrspace(1)* noalias nocapture readonly %ptr) #1 {
>> %tid = tail call i32 @llvm.r600.read.tidig.x() #0
>> @@ -144,7 +144,7 @@ define void @neg_neg_mad_f32(float addrspace(1)* noalias nocapture %out, float a
>> ret void
>> }
>>
>> -; FUNC-LABEL: @mad_fabs_sub_f32
>> +; FUNC-LABEL: mad_fabs_sub_f32
>> ; SI: BUFFER_LOAD_DWORD [[REGA:v[0-9]+]]
>> ; SI: BUFFER_LOAD_DWORD [[REGB:v[0-9]+]]
>> ; SI: BUFFER_LOAD_DWORD [[REGC:v[0-9]+]]
>> diff --git a/test/CodeGen/R600/mad_int24.ll b/test/CodeGen/R600/mad_int24.ll
>> index abb5290..eb7c7cd 100644
>> --- a/test/CodeGen/R600/mad_int24.ll
>> +++ b/test/CodeGen/R600/mad_int24.ll
>> @@ -2,7 +2,7 @@
>> ; RUN: llc < %s -march=r600 -mcpu=cayman | FileCheck %s --check-prefix=CM --check-prefix=FUNC
>> ; RUN: llc < %s -march=r600 -mcpu=SI -verify-machineinstrs | FileCheck %s --check-prefix=SI --check-prefix=FUNC
>>
>> -; FUNC-LABEL: @i32_mad24
>> +; FUNC-LABEL: i32_mad24
>> ; Signed 24-bit multiply is not supported on pre-Cayman GPUs.
>> ; EG: MULLO_INT
>> ; Make sure we aren't masking the inputs.
>> diff --git a/test/CodeGen/R600/mad_uint24.ll b/test/CodeGen/R600/mad_uint24.ll
>> index 0f0893b..6149dd9 100644
>> --- a/test/CodeGen/R600/mad_uint24.ll
>> +++ b/test/CodeGen/R600/mad_uint24.ll
>> @@ -2,7 +2,7 @@
>> ; RUN: llc < %s -march=r600 -mcpu=cayman | FileCheck %s --check-prefix=EG --check-prefix=FUNC
>> ; RUN: llc < %s -march=r600 -mcpu=SI -verify-machineinstrs | FileCheck %s --check-prefix=SI --check-prefix=FUNC
>>
>> -; FUNC-LABEL: @u32_mad24
>> +; FUNC-LABEL: u32_mad24
>> ; EG: MULADD_UINT24
>> ; SI: V_MAD_U32_U24
>>
>> @@ -18,7 +18,7 @@ entry:
>> ret void
>> }
>>
>> -; FUNC-LABEL: @i16_mad24
>> +; FUNC-LABEL: i16_mad24
>> ; The order of A and B does not matter.
>> ; EG: MULADD_UINT24 {{[* ]*}}T{{[0-9]}}.[[MAD_CHAN:[XYZW]]]
>> ; The result must be sign-extended
>> @@ -36,7 +36,7 @@ entry:
>> ret void
>> }
>>
>> -; FUNC-LABEL: @i8_mad24
>> +; FUNC-LABEL: i8_mad24
>> ; EG: MULADD_UINT24 {{[* ]*}}T{{[0-9]}}.[[MAD_CHAN:[XYZW]]]
>> ; The result must be sign-extended
>> ; EG: BFE_INT {{[* ]*}}T{{[0-9]\.[XYZW]}}, PV.[[MAD_CHAN]], 0.0, literal.x
>> @@ -60,7 +60,7 @@ entry:
>> ; 24-bit mad pattern wasn't being matched.
>>
>> ; Check that the select instruction is not deleted.
>> -; FUNC-LABEL: @i24_i32_i32_mad
>> +; FUNC-LABEL: i24_i32_i32_mad
>> ; EG: CNDE_INT
>> ; SI: V_CNDMASK
>> define void @i24_i32_i32_mad(i32 addrspace(1)* %out, i32 %a, i32 %b, i32 %c, i32 %d) {
>> diff --git a/test/CodeGen/R600/max-literals.ll b/test/CodeGen/R600/max-literals.ll
>> index 65a6d2b..96a0193 100644
>> --- a/test/CodeGen/R600/max-literals.ll
>> +++ b/test/CodeGen/R600/max-literals.ll
>> @@ -1,6 +1,6 @@
>> ;RUN: llc < %s -march=r600 -mcpu=redwood | FileCheck %s
>>
>> -; CHECK: @main
>> +; CHECK: main
>> ; CHECK: ADD *
>>
>> define void @main(<4 x float> inreg %reg0, <4 x float> inreg %reg1, <4 x float> inreg %reg2) #0 {
>> @@ -29,7 +29,7 @@ main_body:
>> ret void
>> }
>>
>> -; CHECK: @main
>> +; CHECK: main
>> ; CHECK-NOT: ADD *
>>
>> define void @main2(<4 x float> inreg %reg0, <4 x float> inreg %reg1, <4 x float> inreg %reg2) #0 {
>> diff --git a/test/CodeGen/R600/missing-store.ll b/test/CodeGen/R600/missing-store.ll
>> index ab1f4d0..8b3fb90 100644
>> --- a/test/CodeGen/R600/missing-store.ll
>> +++ b/test/CodeGen/R600/missing-store.ll
>> @@ -5,7 +5,7 @@
>> ; Make sure when the load from %ptr2 is folded the chain isn't lost,
>> ; resulting in losing the store to gptr
>>
>> -; FUNC-LABEL: @missing_store_reduced
>> +; FUNC-LABEL: missing_store_reduced
>> ; SI: DS_READ_B64
>> ; SI: BUFFER_STORE_DWORD
>> ; SI: BUFFER_LOAD_DWORD
>> diff --git a/test/CodeGen/R600/mubuf.ll b/test/CodeGen/R600/mubuf.ll
>> index 7bffead..3f93a8f 100644
>> --- a/test/CodeGen/R600/mubuf.ll
>> +++ b/test/CodeGen/R600/mubuf.ll
>> @@ -7,7 +7,7 @@ declare i32 @llvm.r600.read.tidig.x() readnone
>> ;;;==========================================================================;;;
>>
>> ; MUBUF load with an immediate byte offset that fits into 12-bits
>> -; CHECK-LABEL: @mubuf_load0
>> +; CHECK-LABEL: mubuf_load0
>> ; CHECK: BUFFER_LOAD_DWORD v{{[0-9]}}, s[{{[0-9]:[0-9]}}], 0 offset:0x4 ; encoding: [0x04,0x00,0x30,0xe0
>> define void @mubuf_load0(i32 addrspace(1)* %out, i32 addrspace(1)* %in) {
>> entry:
>> @@ -18,7 +18,7 @@ entry:
>> }
>>
>> ; MUBUF load with the largest possible immediate offset
>> -; CHECK-LABEL: @mubuf_load1
>> +; CHECK-LABEL: mubuf_load1
>> ; CHECK: BUFFER_LOAD_UBYTE v{{[0-9]}}, s[{{[0-9]:[0-9]}}], 0 offset:0xfff ; encoding: [0xff,0x0f,0x20,0xe0
>> define void @mubuf_load1(i8 addrspace(1)* %out, i8 addrspace(1)* %in) {
>> entry:
>> @@ -29,7 +29,7 @@ entry:
>> }
>>
>> ; MUBUF load with an immediate byte offset that doesn't fit into 12-bits
>> -; CHECK-LABEL: @mubuf_load2
>> +; CHECK-LABEL: mubuf_load2
>> ; CHECK: BUFFER_LOAD_DWORD v{{[0-9]}}, v[{{[0-9]+:[0-9]+}}], s[{{[0-9]+:[0-9]+}}], 0 addr64 ; encoding: [0x00,0x80,0x30,0xe0
>> define void @mubuf_load2(i32 addrspace(1)* %out, i32 addrspace(1)* %in) {
>> entry:
>> @@ -40,7 +40,7 @@ entry:
>> }
>>
>> ; MUBUF load with a 12-bit immediate offset and a register offset
>> -; CHECK-LABEL: @mubuf_load3
>> +; CHECK-LABEL: mubuf_load3
>> ; CHECK-NOT: ADD
>> ; CHECK: BUFFER_LOAD_DWORD v{{[0-9]}}, v[{{[0-9]+:[0-9]+}}], s[{{[0-9]+:[0-9]+}}], 0 addr64 offset:0x4 ; encoding: [0x04,0x80,0x30,0xe0
>> define void @mubuf_load3(i32 addrspace(1)* %out, i32 addrspace(1)* %in, i64 %offset) {
>> @@ -57,7 +57,7 @@ entry:
>> ;;;==========================================================================;;;
>>
>> ; MUBUF store with an immediate byte offset that fits into 12-bits
>> -; CHECK-LABEL: @mubuf_store0
>> +; CHECK-LABEL: mubuf_store0
>> ; CHECK: BUFFER_STORE_DWORD v{{[0-9]}}, s[{{[0-9]:[0-9]}}], 0 offset:0x4 ; encoding: [0x04,0x00,0x70,0xe0
>> define void @mubuf_store0(i32 addrspace(1)* %out) {
>> entry:
>> @@ -67,7 +67,7 @@ entry:
>> }
>>
>> ; MUBUF store with the largest possible immediate offset
>> -; CHECK-LABEL: @mubuf_store1
>> +; CHECK-LABEL: mubuf_store1
>> ; CHECK: BUFFER_STORE_BYTE v{{[0-9]}}, s[{{[0-9]:[0-9]}}], 0 offset:0xfff ; encoding: [0xff,0x0f,0x60,0xe0
>>
>> define void @mubuf_store1(i8 addrspace(1)* %out) {
>> @@ -78,7 +78,7 @@ entry:
>> }
>>
>> ; MUBUF store with an immediate byte offset that doesn't fit into 12-bits
>> -; CHECK-LABEL: @mubuf_store2
>> +; CHECK-LABEL: mubuf_store2
>> ; CHECK: BUFFER_STORE_DWORD v{{[0-9]}}, v[{{[0-9]+:[0-9]+}}], s[{{[0-9]:[0-9]}}], 0 addr64 ; encoding: [0x00,0x80,0x70,0xe0
>> define void @mubuf_store2(i32 addrspace(1)* %out) {
>> entry:
>> @@ -88,7 +88,7 @@ entry:
>> }
>>
>> ; MUBUF store with a 12-bit immediate offset and a register offset
>> -; CHECK-LABEL: @mubuf_store3
>> +; CHECK-LABEL: mubuf_store3
>> ; CHECK-NOT: ADD
>> ; CHECK: BUFFER_STORE_DWORD v{{[0-9]}}, v[{{[0-9]:[0-9]}}], s[{{[0-9]:[0-9]}}], 0 addr64 offset:0x4 ; encoding: [0x04,0x80,0x70,0xe0
>> define void @mubuf_store3(i32 addrspace(1)* %out, i64 %offset) {
>> @@ -99,14 +99,14 @@ entry:
>> ret void
>> }
>>
>> -; CHECK-LABEL: @store_sgpr_ptr
>> +; CHECK-LABEL: store_sgpr_ptr
>> ; CHECK: BUFFER_STORE_DWORD v{{[0-9]+}}, s{{\[[0-9]+:[0-9]+\]}}, 0
>> define void @store_sgpr_ptr(i32 addrspace(1)* %out) #0 {
>> store i32 99, i32 addrspace(1)* %out, align 4
>> ret void
>> }
>>
>> -; CHECK-LABEL: @store_sgpr_ptr_offset
>> +; CHECK-LABEL: store_sgpr_ptr_offset
>> ; CHECK: BUFFER_STORE_DWORD v{{[0-9]+}}, s{{\[[0-9]+:[0-9]+\]}}, 0 offset:0x28
>> define void @store_sgpr_ptr_offset(i32 addrspace(1)* %out) #0 {
>> %out.gep = getelementptr i32 addrspace(1)* %out, i32 10
>> @@ -114,7 +114,7 @@ define void @store_sgpr_ptr_offset(i32 addrspace(1)* %out) #0 {
>> ret void
>> }
>>
>> -; CHECK-LABEL: @store_sgpr_ptr_large_offset
>> +; CHECK-LABEL: store_sgpr_ptr_large_offset
>> ; CHECK: BUFFER_STORE_DWORD v{{[0-9]+}}, v{{\[[0-9]+:[0-9]+\]}}, s{{\[[0-9]+:[0-9]+\]}}, 0 addr64
>> define void @store_sgpr_ptr_large_offset(i32 addrspace(1)* %out) #0 {
>> %out.gep = getelementptr i32 addrspace(1)* %out, i32 32768
>> @@ -122,7 +122,7 @@ define void @store_sgpr_ptr_large_offset(i32 addrspace(1)* %out) #0 {
>> ret void
>> }
>>
>> -; CHECK-LABEL: @store_vgpr_ptr
>> +; CHECK-LABEL: store_vgpr_ptr
>> ; CHECK: BUFFER_STORE_DWORD v{{[0-9]+}}, v{{\[[0-9]+:[0-9]+\]}}, s{{\[[0-9]+:[0-9]+\]}}, 0 addr64
>> define void @store_vgpr_ptr(i32 addrspace(1)* %out) #0 {
>> %tid = call i32 @llvm.r600.read.tidig.x() readnone
>> diff --git a/test/CodeGen/R600/mul.ll b/test/CodeGen/R600/mul.ll
>> index 11de3e3..79acd88 100644
>> --- a/test/CodeGen/R600/mul.ll
>> +++ b/test/CodeGen/R600/mul.ll
>> @@ -3,7 +3,7 @@
>>
>> ; mul24 and mad24 are affected
>>
>> -; FUNC-LABEL: @test_mul_v2i32
>> +; FUNC-LABEL: test_mul_v2i32
>> ; EG: MULLO_INT {{\*? *}}T{{[0-9]+\.[XYZW], T[0-9]+\.[XYZW], T[0-9]+\.[XYZW]}}
>> ; EG: MULLO_INT {{\*? *}}T{{[0-9]+\.[XYZW], T[0-9]+\.[XYZW], T[0-9]+\.[XYZW]}}
>>
>> @@ -19,7 +19,7 @@ define void @test_mul_v2i32(<2 x i32> addrspace(1)* %out, <2 x i32> addrspace(1)
>> ret void
>> }
>>
>> -; FUNC-LABEL: @v_mul_v4i32
>> +; FUNC-LABEL: v_mul_v4i32
>> ; EG: MULLO_INT {{\*? *}}T{{[0-9]+\.[XYZW], T[0-9]+\.[XYZW], T[0-9]+\.[XYZW]}}
>> ; EG: MULLO_INT {{\*? *}}T{{[0-9]+\.[XYZW], T[0-9]+\.[XYZW], T[0-9]+\.[XYZW]}}
>> ; EG: MULLO_INT {{\*? *}}T{{[0-9]+\.[XYZW], T[0-9]+\.[XYZW], T[0-9]+\.[XYZW]}}
>> @@ -39,7 +39,7 @@ define void @v_mul_v4i32(<4 x i32> addrspace(1)* %out, <4 x i32> addrspace(1)* %
>> ret void
>> }
>>
>> -; FUNC-LABEL: @s_trunc_i64_mul_to_i32
>> +; FUNC-LABEL: s_trunc_i64_mul_to_i32
>> ; SI: S_LOAD_DWORD
>> ; SI: S_LOAD_DWORD
>> ; SI: S_MUL_I32
>> @@ -51,7 +51,7 @@ define void @s_trunc_i64_mul_to_i32(i32 addrspace(1)* %out, i64 %a, i64 %b) {
>> ret void
>> }
>>
>> -; FUNC-LABEL: @v_trunc_i64_mul_to_i32
>> +; FUNC-LABEL: v_trunc_i64_mul_to_i32
>> ; SI: S_LOAD_DWORD
>> ; SI: S_LOAD_DWORD
>> ; SI: V_MUL_LO_I32
>> @@ -67,7 +67,7 @@ define void @v_trunc_i64_mul_to_i32(i32 addrspace(1)* %out, i64 addrspace(1)* %a
>>
>> ; This 64-bit multiply should just use MUL_HI and MUL_LO, since the top
>> ; 32-bits of both arguments are sign bits.
>> -; FUNC-LABEL: @mul64_sext_c
>> +; FUNC-LABEL: mul64_sext_c
>> ; EG-DAG: MULLO_INT
>> ; EG-DAG: MULHI_INT
>> ; SI-DAG: S_MUL_I32
>> @@ -80,7 +80,7 @@ entry:
>> ret void
>> }
>>
>> -; FUNC-LABEL: @v_mul64_sext_c:
>> +; FUNC-LABEL: v_mul64_sext_c:
>> ; EG-DAG: MULLO_INT
>> ; EG-DAG: MULHI_INT
>> ; SI-DAG: V_MUL_LO_I32
>> @@ -94,7 +94,7 @@ define void @v_mul64_sext_c(i64 addrspace(1)* %out, i32 addrspace(1)* %in) {
>> ret void
>> }
>>
>> -; FUNC-LABEL: @v_mul64_sext_inline_imm:
>> +; FUNC-LABEL: v_mul64_sext_inline_imm:
>> ; SI-DAG: V_MUL_LO_I32 v{{[0-9]+}}, 9, v{{[0-9]+}}
>> ; SI-DAG: V_MUL_HI_I32 v{{[0-9]+}}, 9, v{{[0-9]+}}
>> ; SI: S_ENDPGM
>> @@ -106,7 +106,7 @@ define void @v_mul64_sext_inline_imm(i64 addrspace(1)* %out, i32 addrspace(1)* %
>> ret void
>> }
>>
>> -; FUNC-LABEL: @s_mul_i32:
>> +; FUNC-LABEL: s_mul_i32:
>> ; SI: S_LOAD_DWORD [[SRC0:s[0-9]+]],
>> ; SI: S_LOAD_DWORD [[SRC1:s[0-9]+]],
>> ; SI: S_MUL_I32 [[SRESULT:s[0-9]+]], [[SRC0]], [[SRC1]]
>> @@ -119,7 +119,7 @@ define void @s_mul_i32(i32 addrspace(1)* %out, i32 %a, i32 %b) nounwind {
>> ret void
>> }
>>
>> -; FUNC-LABEL: @v_mul_i32
>> +; FUNC-LABEL: v_mul_i32
>> ; SI: V_MUL_LO_I32 v{{[0-9]+}}, v{{[0-9]+}}, v{{[0-9]+}}
>> define void @v_mul_i32(i32 addrspace(1)* %out, i32 addrspace(1)* %in) {
>> %b_ptr = getelementptr i32 addrspace(1)* %in, i32 1
>> @@ -137,14 +137,14 @@ define void @v_mul_i32(i32 addrspace(1)* %out, i32 addrspace(1)* %in) {
>> ; so this test just uses FUNC-LABEL to make sure the compiler does not
>> ; crash with a 'failed to select' error.
>>
>> -; FUNC-LABEL: @s_mul_i64:
>> +; FUNC-LABEL: s_mul_i64:
>> define void @s_mul_i64(i64 addrspace(1)* %out, i64 %a, i64 %b) nounwind {
>> %mul = mul i64 %a, %b
>> store i64 %mul, i64 addrspace(1)* %out, align 8
>> ret void
>> }
>>
>> -; FUNC-LABEL: @v_mul_i64
>> +; FUNC-LABEL: v_mul_i64
>> ; SI: V_MUL_LO_I32
>> define void @v_mul_i64(i64 addrspace(1)* %out, i64 addrspace(1)* %aptr, i64 addrspace(1)* %bptr) {
>> %a = load i64 addrspace(1)* %aptr, align 8
>> @@ -154,7 +154,7 @@ define void @v_mul_i64(i64 addrspace(1)* %out, i64 addrspace(1)* %aptr, i64 addr
>> ret void
>> }
>>
>> -; FUNC-LABEL: @mul32_in_branch
>> +; FUNC-LABEL: mul32_in_branch
>> ; SI: S_MUL_I32
>> define void @mul32_in_branch(i32 addrspace(1)* %out, i32 addrspace(1)* %in, i32 %a, i32 %b, i32 %c) {
>> entry:
>> @@ -175,7 +175,7 @@ endif:
>> ret void
>> }
>>
>> -; FUNC-LABEL: @mul64_in_branch
>> +; FUNC-LABEL: mul64_in_branch
>> ; SI-DAG: S_MUL_I32
>> ; SI-DAG: V_MUL_HI_U32
>> ; SI: S_ENDPGM
>> diff --git a/test/CodeGen/R600/mul_int24.ll b/test/CodeGen/R600/mul_int24.ll
>> index 046911b..0ad9a59 100644
>> --- a/test/CodeGen/R600/mul_int24.ll
>> +++ b/test/CodeGen/R600/mul_int24.ll
>> @@ -2,7 +2,7 @@
>> ; RUN: llc < %s -march=r600 -mcpu=cayman | FileCheck %s --check-prefix=CM --check-prefix=FUNC
>> ; RUN: llc < %s -march=r600 -mcpu=SI -verify-machineinstrs | FileCheck %s --check-prefix=SI --check-prefix=FUNC
>>
>> -; FUNC-LABEL: @i32_mul24
>> +; FUNC-LABEL: i32_mul24
>> ; Signed 24-bit multiply is not supported on pre-Cayman GPUs.
>> ; EG: MULLO_INT
>> ; Make sure we are not masking the inputs
>> diff --git a/test/CodeGen/R600/mul_uint24.ll b/test/CodeGen/R600/mul_uint24.ll
>> index 72bbe0f..7d5a41e 100644
>> --- a/test/CodeGen/R600/mul_uint24.ll
>> +++ b/test/CodeGen/R600/mul_uint24.ll
>> @@ -2,7 +2,7 @@
>> ; RUN: llc < %s -march=r600 -mcpu=cayman | FileCheck %s --check-prefix=EG --check-prefix=FUNC
>> ; RUN: llc < %s -march=r600 -mcpu=SI -verify-machineinstrs | FileCheck %s --check-prefix=SI --check-prefix=FUNC
>>
>> -; FUNC-LABEL: @u32_mul24
>> +; FUNC-LABEL: u32_mul24
>> ; EG: MUL_UINT24 {{[* ]*}}T{{[0-9]\.[XYZW]}}, KC0[2].Z, KC0[2].W
>> ; SI: V_MUL_U32_U24
>>
>> @@ -17,7 +17,7 @@ entry:
>> ret void
>> }
>>
>> -; FUNC-LABEL: @i16_mul24
>> +; FUNC-LABEL: i16_mul24
>> ; EG: MUL_UINT24 {{[* ]*}}T{{[0-9]}}.[[MUL_CHAN:[XYZW]]]
>> ; The result must be sign-extended
>> ; EG: BFE_INT {{[* ]*}}T{{[0-9]}}.{{[XYZW]}}, PV.[[MUL_CHAN]], 0.0, literal.x
>> @@ -32,7 +32,7 @@ entry:
>> ret void
>> }
>>
>> -; FUNC-LABEL: @i8_mul24
>> +; FUNC-LABEL: i8_mul24
>> ; EG: MUL_UINT24 {{[* ]*}}T{{[0-9]}}.[[MUL_CHAN:[XYZW]]]
>> ; The result must be sign-extended
>> ; EG: BFE_INT {{[* ]*}}T{{[0-9]}}.{{[XYZW]}}, PV.[[MUL_CHAN]], 0.0, literal.x
>> @@ -48,7 +48,7 @@ entry:
>> }
>>
>> ; Multiply with 24-bit inputs and 64-bit output
>> -; FUNC_LABEL: @mul24_i64
>> +; FUNC_LABEL: mul24_i64
>> ; EG; MUL_UINT24
>> ; EG: MULHI
>> ; SI: V_MUL_U32_U24
>> diff --git a/test/CodeGen/R600/no-initializer-constant-addrspace.ll b/test/CodeGen/R600/no-initializer-constant-addrspace.ll
>> index ab82e7e..6493469 100644
>> --- a/test/CodeGen/R600/no-initializer-constant-addrspace.ll
>> +++ b/test/CodeGen/R600/no-initializer-constant-addrspace.ll
>> @@ -3,7 +3,7 @@
>>
>> @extern_const_addrspace = external unnamed_addr addrspace(2) constant [5 x i32], align 4
>>
>> -; FUNC-LABEL: @load_extern_const_init
>> +; FUNC-LABEL: load_extern_const_init
>> define void @load_extern_const_init(i32 addrspace(1)* %out) nounwind {
>> %val = load i32 addrspace(2)* getelementptr ([5 x i32] addrspace(2)* @extern_const_addrspace, i64 0, i64 3), align 4
>> store i32 %val, i32 addrspace(1)* %out, align 4
>> @@ -12,7 +12,7 @@ define void @load_extern_const_init(i32 addrspace(1)* %out) nounwind {
>>
>> @undef_const_addrspace = unnamed_addr addrspace(2) constant [5 x i32] undef, align 4
>>
>> -; FUNC-LABEL: @load_undef_const_init
>> +; FUNC-LABEL: load_undef_const_init
>> define void @load_undef_const_init(i32 addrspace(1)* %out) nounwind {
>> %val = load i32 addrspace(2)* getelementptr ([5 x i32] addrspace(2)* @undef_const_addrspace, i64 0, i64 3), align 4
>> store i32 %val, i32 addrspace(1)* %out, align 4
>> diff --git a/test/CodeGen/R600/operand-spacing.ll b/test/CodeGen/R600/operand-spacing.ll
>> index d9e153d..b6b062a 100644
>> --- a/test/CodeGen/R600/operand-spacing.ll
>> +++ b/test/CodeGen/R600/operand-spacing.ll
>> @@ -2,7 +2,7 @@
>>
>> ; Make sure there isn't an extra space between the instruction name and first operands.
>>
>> -; SI-LABEL: @add_f32
>> +; SI-LABEL: add_f32
>> ; SI-DAG: S_LOAD_DWORD [[SREGA:s[0-9]+]], s{{\[[0-9]+:[0-9]+\]}}, 0xb
>> ; SI-DAG: S_LOAD_DWORD [[SREGB:s[0-9]+]], s{{\[[0-9]+:[0-9]+\]}}, 0xc
>> ; SI: V_MOV_B32_e32 [[VREGB:v[0-9]+]], [[SREGB]]
>> diff --git a/test/CodeGen/R600/or.ll b/test/CodeGen/R600/or.ll
>> index 913fda1..8fcbbbb 100644
>> --- a/test/CodeGen/R600/or.ll
>> +++ b/test/CodeGen/R600/or.ll
>> @@ -1,11 +1,11 @@
>> ;RUN: llc < %s -march=r600 -mcpu=redwood | FileCheck --check-prefix=EG %s
>> ;RUN: llc < %s -march=r600 -mcpu=verde -verify-machineinstrs | FileCheck --check-prefix=SI %s
>>
>> -; EG-LABEL: @or_v2i32
>> +; EG-LABEL: or_v2i32
>> ; EG: OR_INT {{\*? *}}T{{[0-9]+\.[XYZW], T[0-9]+\.[XYZW], T[0-9]+\.[XYZW]}}
>> ; EG: OR_INT {{\*? *}}T{{[0-9]+\.[XYZW], T[0-9]+\.[XYZW], T[0-9]+\.[XYZW]}}
>>
>> -; SI-LABEL: @or_v2i32
>> +; SI-LABEL: or_v2i32
>> ; SI: V_OR_B32_e32 v{{[0-9]+, v[0-9]+, v[0-9]+}}
>> ; SI: V_OR_B32_e32 v{{[0-9]+, v[0-9]+, v[0-9]+}}
>>
>> @@ -18,13 +18,13 @@ define void @or_v2i32(<2 x i32> addrspace(1)* %out, <2 x i32> addrspace(1)* %in)
>> ret void
>> }
>>
>> -; EG-LABEL: @or_v4i32
>> +; EG-LABEL: or_v4i32
>> ; EG: OR_INT {{\*? *}}T{{[0-9]+\.[XYZW], T[0-9]+\.[XYZW], T[0-9]+\.[XYZW]}}
>> ; EG: OR_INT {{\*? *}}T{{[0-9]+\.[XYZW], T[0-9]+\.[XYZW], T[0-9]+\.[XYZW]}}
>> ; EG: OR_INT {{\*? *}}T{{[0-9]+\.[XYZW], T[0-9]+\.[XYZW], T[0-9]+\.[XYZW]}}
>> ; EG: OR_INT {{\*? *}}T{{[0-9]+\.[XYZW], T[0-9]+\.[XYZW], T[0-9]+\.[XYZW]}}
>>
>> -; SI-LABEL: @or_v4i32
>> +; SI-LABEL: or_v4i32
>> ; SI: V_OR_B32_e32 v{{[0-9]+, v[0-9]+, v[0-9]+}}
>> ; SI: V_OR_B32_e32 v{{[0-9]+, v[0-9]+, v[0-9]+}}
>> ; SI: V_OR_B32_e32 v{{[0-9]+, v[0-9]+, v[0-9]+}}
>> @@ -39,7 +39,7 @@ define void @or_v4i32(<4 x i32> addrspace(1)* %out, <4 x i32> addrspace(1)* %in)
>> ret void
>> }
>>
>> -; SI-LABEL: @scalar_or_i32
>> +; SI-LABEL: scalar_or_i32
>> ; SI: S_OR_B32
>> define void @scalar_or_i32(i32 addrspace(1)* %out, i32 %a, i32 %b) {
>> %or = or i32 %a, %b
>> @@ -47,7 +47,7 @@ define void @scalar_or_i32(i32 addrspace(1)* %out, i32 %a, i32 %b) {
>> ret void
>> }
>>
>> -; SI-LABEL: @vector_or_i32
>> +; SI-LABEL: vector_or_i32
>> ; SI: V_OR_B32_e32 v{{[0-9]}}
>> define void @vector_or_i32(i32 addrspace(1)* %out, i32 addrspace(1)* %a, i32 %b) {
>> %loada = load i32 addrspace(1)* %a
>> @@ -56,7 +56,7 @@ define void @vector_or_i32(i32 addrspace(1)* %out, i32 addrspace(1)* %a, i32 %b)
>> ret void
>> }
>>
>> -; SI-LABEL: @scalar_or_literal_i32
>> +; SI-LABEL: scalar_or_literal_i32
>> ; SI: S_OR_B32 s{{[0-9]+}}, s{{[0-9]+}}, 0x1869f
>> define void @scalar_or_literal_i32(i32 addrspace(1)* %out, i32 %a) {
>> %or = or i32 %a, 99999
>> @@ -64,7 +64,7 @@ define void @scalar_or_literal_i32(i32 addrspace(1)* %out, i32 %a) {
>> ret void
>> }
>>
>> -; SI-LABEL: @vector_or_literal_i32
>> +; SI-LABEL: vector_or_literal_i32
>> ; SI: V_OR_B32_e32 v{{[0-9]+}}, 0xffff, v{{[0-9]+}}
>> define void @vector_or_literal_i32(i32 addrspace(1)* %out, i32 addrspace(1)* %a, i32 addrspace(1)* %b) {
>> %loada = load i32 addrspace(1)* %a, align 4
>> @@ -73,7 +73,7 @@ define void @vector_or_literal_i32(i32 addrspace(1)* %out, i32 addrspace(1)* %a,
>> ret void
>> }
>>
>> -; SI-LABEL: @vector_or_inline_immediate_i32
>> +; SI-LABEL: vector_or_inline_immediate_i32
>> ; SI: V_OR_B32_e32 v{{[0-9]+}}, 4, v{{[0-9]+}}
>> define void @vector_or_inline_immediate_i32(i32 addrspace(1)* %out, i32 addrspace(1)* %a, i32 addrspace(1)* %b) {
>> %loada = load i32 addrspace(1)* %a, align 4
>> @@ -82,10 +82,10 @@ define void @vector_or_inline_immediate_i32(i32 addrspace(1)* %out, i32 addrspac
>> ret void
>> }
>>
>> -; EG-LABEL: @scalar_or_i64
>> +; EG-LABEL: scalar_or_i64
>> ; EG-DAG: OR_INT * T{{[0-9]\.[XYZW]}}, KC0[2].W, KC0[3].Y
>> ; EG-DAG: OR_INT * T{{[0-9]\.[XYZW]}}, KC0[3].X, KC0[3].Z
>> -; SI-LABEL: @scalar_or_i64
>> +; SI-LABEL: scalar_or_i64
>> ; SI: S_OR_B64
>> define void @scalar_or_i64(i64 addrspace(1)* %out, i64 %a, i64 %b) {
>> %or = or i64 %a, %b
>> @@ -93,7 +93,7 @@ define void @scalar_or_i64(i64 addrspace(1)* %out, i64 %a, i64 %b) {
>> ret void
>> }
>>
>> -; SI-LABEL: @vector_or_i64
>> +; SI-LABEL: vector_or_i64
>> ; SI: V_OR_B32_e32 v{{[0-9]}}
>> ; SI: V_OR_B32_e32 v{{[0-9]}}
>> define void @vector_or_i64(i64 addrspace(1)* %out, i64 addrspace(1)* %a, i64 addrspace(1)* %b) {
>> @@ -104,7 +104,7 @@ define void @vector_or_i64(i64 addrspace(1)* %out, i64 addrspace(1)* %a, i64 add
>> ret void
>> }
>>
>> -; SI-LABEL: @scalar_vector_or_i64
>> +; SI-LABEL: scalar_vector_or_i64
>> ; SI: V_OR_B32_e32 v{{[0-9]}}
>> ; SI: V_OR_B32_e32 v{{[0-9]}}
>> define void @scalar_vector_or_i64(i64 addrspace(1)* %out, i64 addrspace(1)* %a, i64 %b) {
>> @@ -114,7 +114,7 @@ define void @scalar_vector_or_i64(i64 addrspace(1)* %out, i64 addrspace(1)* %a,
>> ret void
>> }
>>
>> -; SI-LABEL: @vector_or_i64_loadimm
>> +; SI-LABEL: vector_or_i64_loadimm
>> ; SI-DAG: S_MOV_B32 [[LO_S_IMM:s[0-9]+]], 0xdf77987f
>> ; SI-DAG: S_MOV_B32 [[HI_S_IMM:s[0-9]+]], 0x146f
>> ; SI-DAG: BUFFER_LOAD_DWORDX2 v{{\[}}[[LO_VREG:[0-9]+]]:[[HI_VREG:[0-9]+]]{{\]}},
>> @@ -129,7 +129,7 @@ define void @vector_or_i64_loadimm(i64 addrspace(1)* %out, i64 addrspace(1)* %a,
>> }
>>
>> ; FIXME: The or 0 should really be removed.
>> -; SI-LABEL: @vector_or_i64_imm
>> +; SI-LABEL: vector_or_i64_imm
>> ; SI: BUFFER_LOAD_DWORDX2 v{{\[}}[[LO_VREG:[0-9]+]]:[[HI_VREG:[0-9]+]]{{\]}},
>> ; SI: V_OR_B32_e32 {{v[0-9]+}}, 8, v[[LO_VREG]]
>> ; SI: V_OR_B32_e32 {{v[0-9]+}}, 0, {{.*}}
>> @@ -141,7 +141,7 @@ define void @vector_or_i64_imm(i64 addrspace(1)* %out, i64 addrspace(1)* %a, i64
>> ret void
>> }
>>
>> -; SI-LABEL: @trunc_i64_or_to_i32
>> +; SI-LABEL: trunc_i64_or_to_i32
>> ; SI: S_LOAD_DWORD s[[SREG0:[0-9]+]]
>> ; SI: S_LOAD_DWORD s[[SREG1:[0-9]+]]
>> ; SI: S_OR_B32 s[[SRESULT:[0-9]+]], s[[SREG1]], s[[SREG0]]
>> @@ -154,10 +154,10 @@ define void @trunc_i64_or_to_i32(i32 addrspace(1)* %out, i64 %a, i64 %b) {
>> ret void
>> }
>>
>> -; EG-CHECK: @or_i1
>> +; EG-CHECK: or_i1
>> ; EG-CHECK: OR_INT {{\** *}}T{{[0-9]+\.[XYZW], PV\.[XYZW], PS}}
>>
>> -; SI-CHECK: @or_i1
>> +; SI-CHECK: or_i1
>> ; SI-CHECK: S_OR_B64 s[{{[0-9]+:[0-9]+}}], s[{{[0-9]+:[0-9]+}}], s[{{[0-9]+:[0-9]+}}]
>> define void @or_i1(float addrspace(1)* %out, float addrspace(1)* %in0, float addrspace(1)* %in1) {
>> %a = load float addrspace(1) * %in0
>> diff --git a/test/CodeGen/R600/packetizer.ll b/test/CodeGen/R600/packetizer.ll
>> index 0a405c5..1f70c2f 100644
>> --- a/test/CodeGen/R600/packetizer.ll
>> +++ b/test/CodeGen/R600/packetizer.ll
>> @@ -1,7 +1,7 @@
>> ; RUN: llc < %s -march=r600 -mcpu=redwood | FileCheck %s
>> ; RUN: llc < %s -march=r600 -mcpu=cayman | FileCheck %s
>>
>> -; CHECK: @test
>> +; CHECK: test
>> ; CHECK: BIT_ALIGN_INT T{{[0-9]}}.X
>> ; CHECK: BIT_ALIGN_INT T{{[0-9]}}.Y
>> ; CHECK: BIT_ALIGN_INT T{{[0-9]}}.Z
>> diff --git a/test/CodeGen/R600/predicate-dp4.ll b/test/CodeGen/R600/predicate-dp4.ll
>> index e48d6a7..32d2711 100644
>> --- a/test/CodeGen/R600/predicate-dp4.ll
>> +++ b/test/CodeGen/R600/predicate-dp4.ll
>> @@ -1,6 +1,6 @@
>> ;RUN: llc < %s -march=r600 -mcpu=cayman
>>
>> -; CHECK-LABEL: @main
>> +; CHECK-LABEL: main
>> ; CHECK: PRED_SETE_INT * Pred,
>> ; CHECK: DOT4 T{{[0-9]+}}.X, T0.X, T0.X, Pred_sel_one
>> define void @main(<4 x float> inreg) #0 {
>> diff --git a/test/CodeGen/R600/predicates.ll b/test/CodeGen/R600/predicates.ll
>> index 902508f..a744600 100644
>> --- a/test/CodeGen/R600/predicates.ll
>> +++ b/test/CodeGen/R600/predicates.ll
>> @@ -3,7 +3,7 @@
>> ; These tests make sure the compiler is optimizing branches using predicates
>> ; when it is legal to do so.
>>
>> -; CHECK: @simple_if
>> +; CHECK: simple_if
>> ; CHECK: PRED_SET{{[EGN][ET]*}}_INT * Pred,
>> ; CHECK: LSHL * T{{[0-9]+\.[XYZW], T[0-9]+\.[XYZW]}}, 1, Pred_sel
>> define void @simple_if(i32 addrspace(1)* %out, i32 %in) {
>> @@ -21,7 +21,7 @@ ENDIF:
>> ret void
>> }
>>
>> -; CHECK: @simple_if_else
>> +; CHECK: simple_if_else
>> ; CHECK: PRED_SET{{[EGN][ET]*}}_INT * Pred,
>> ; CHECK: LSH{{[LR] \* T[0-9]+\.[XYZW], T[0-9]+\.[XYZW]}}, 1, Pred_sel
>> ; CHECK: LSH{{[LR] \* T[0-9]+\.[XYZW], T[0-9]+\.[XYZW]}}, 1, Pred_sel
>> @@ -44,7 +44,7 @@ ENDIF:
>> ret void
>> }
>>
>> -; CHECK: @nested_if
>> +; CHECK: nested_if
>> ; CHECK: ALU_PUSH_BEFORE
>> ; CHECK: JUMP
>> ; CHECK: POP
>> @@ -71,7 +71,7 @@ ENDIF:
>> ret void
>> }
>>
>> -; CHECK: @nested_if_else
>> +; CHECK: nested_if_else
>> ; CHECK: ALU_PUSH_BEFORE
>> ; CHECK: JUMP
>> ; CHECK: POP
>> diff --git a/test/CodeGen/R600/private-memory.ll b/test/CodeGen/R600/private-memory.ll
>> index 505ef6c..d7f7f5c 100644
>> --- a/test/CodeGen/R600/private-memory.ll
>> +++ b/test/CodeGen/R600/private-memory.ll
>> @@ -4,7 +4,7 @@
>>
>> declare i32 @llvm.r600.read.tidig.x() nounwind readnone
>>
>> -; FUNC-LABEL: @mova_same_clause
>> +; FUNC-LABEL: mova_same_clause
>>
>> ; R600: LDS_WRITE
>> ; R600: LDS_WRITE
>> @@ -45,7 +45,7 @@ entry:
>> ; XXX: This generated code has unnecessary MOVs, we should be able to optimize
>> ; this.
>>
>> -; FUNC-LABEL: @multiple_structs
>> +; FUNC-LABEL: multiple_structs
>> ; R600-NOT: MOVA_INT
>> ; SI-NOT: V_MOVREL
>> ; SI-NOT: V_MOVREL
>> @@ -76,7 +76,7 @@ entry:
>> ; loads and stores should be lowered to copies, so there shouldn't be any
>> ; MOVA instructions.
>>
>> -; FUNC-LABEL: @direct_loop
>> +; FUNC-LABEL: direct_loop
>> ; R600-NOT: MOVA_INT
>> ; SI-NOT: V_MOVREL
>>
>> @@ -112,7 +112,7 @@ for.end:
>> ret void
>> }
>>
>> -; FUNC-LABEL: @short_array
>> +; FUNC-LABEL: short_array
>>
>> ; R600: MOVA_INT
>>
>> @@ -133,7 +133,7 @@ entry:
>> ret void
>> }
>>
>> -; FUNC-LABEL: @char_array
>> +; FUNC-LABEL: char_array
>>
>> ; R600: MOVA_INT
>>
>> @@ -156,7 +156,7 @@ entry:
>>
>> ; Make sure we don't overwrite workitem information with private memory
>>
>> -; FUNC-LABEL: @work_item_info
>> +; FUNC-LABEL: work_item_info
>> ; R600-NOT: MOV T0.X
>> ; Additional check in case the move ends up in the last slot
>> ; R600-NOT: MOV * TO.X
>> @@ -179,7 +179,7 @@ entry:
>>
>> ; Test that two stack objects are not stored in the same register
>> ; The second stack object should be in T3.X
>> -; FUNC-LABEL: @no_overlap
>> +; FUNC-LABEL: no_overlap
>> ; R600_CHECK: MOV
>> ; R600_CHECK: [[CHAN:[XYZW]]]+
>> ; R600-NOT: [[CHAN]]+
>> diff --git a/test/CodeGen/R600/r600-encoding.ll b/test/CodeGen/R600/r600-encoding.ll
>> index b760c88..fe643f0 100644
>> --- a/test/CodeGen/R600/r600-encoding.ll
>> +++ b/test/CodeGen/R600/r600-encoding.ll
>> @@ -4,10 +4,10 @@
>> ; The earliest R600 GPUs have a slightly different encoding than the rest of
>> ; the VLIW4/5 GPUs.
>>
>> -; EG-CHECK: @test
>> +; EG-CHECK: test
>> ; EG-CHECK: MUL_IEEE {{[ *TXYZWPVxyzw.,0-9]+}} ; encoding: [{{0x[0-9a-f]+,0x[0-9a-f]+,0x[0-9a-f]+,0x[0-9a-f]+,0x10,0x01,0x[0-9a-f]+,0x[0-9a-f]+}}]
>>
>> -; R600-CHECK: @test
>> +; R600-CHECK: test
>> ; R600-CHECK: MUL_IEEE {{[ *TXYZWPVxyzw.,0-9]+}} ; encoding: [{{0x[0-9a-f]+,0x[0-9a-f]+,0x[0-9a-f]+,0x[0-9a-f]+,0x10,0x02,0x[0-9a-f]+,0x[0-9a-f]+}}]
>>
>> define void @test(<4 x float> inreg %reg0) #0 {
>> diff --git a/test/CodeGen/R600/register-count-comments.ll b/test/CodeGen/R600/register-count-comments.ll
>> index 6179013..30a7a9e 100644
>> --- a/test/CodeGen/R600/register-count-comments.ll
>> +++ b/test/CodeGen/R600/register-count-comments.ll
>> @@ -2,7 +2,7 @@
>>
>> declare i32 @llvm.SI.tid() nounwind readnone
>>
>> -; SI-LABEL: @foo:
>> +; SI-LABEL: foo:
>> ; SI: .section .AMDGPU.csdata
>> ; SI: ; Kernel info:
>> ; SI: ; NumSgprs: {{[0-9]+}}
>> @@ -19,7 +19,7 @@ define void @foo(i32 addrspace(1)* noalias %out, i32 addrspace(1)* %abase, i32 a
>> ret void
>> }
>>
>> -; SI-LABEL: @one_vgpr_used
>> +; SI-LABEL: one_vgpr_used
>> ; SI: NumVgprs: 1
>> define void @one_vgpr_used(i32 addrspace(1)* %out, i32 %x) nounwind {
>> store i32 %x, i32 addrspace(1)* %out, align 4
>> diff --git a/test/CodeGen/R600/reorder-stores.ll b/test/CodeGen/R600/reorder-stores.ll
>> index be2fcc6..171fda3 100644
>> --- a/test/CodeGen/R600/reorder-stores.ll
>> +++ b/test/CodeGen/R600/reorder-stores.ll
>> @@ -1,6 +1,6 @@
>> ; RUN: llc -march=r600 -mcpu=SI < %s | FileCheck -check-prefix=SI %s
>>
>> -; SI-LABEL: @no_reorder_v2f64_global_load_store
>> +; SI-LABEL: no_reorder_v2f64_global_load_store
>> ; SI: BUFFER_LOAD_DWORDX2
>> ; SI: BUFFER_LOAD_DWORDX2
>> ; SI: BUFFER_LOAD_DWORDX2
>> @@ -18,7 +18,7 @@ define void @no_reorder_v2f64_global_load_store(<2 x double> addrspace(1)* nocap
>> ret void
>> }
>>
>> -; SI-LABEL: @no_reorder_scalarized_v2f64_local_load_store
>> +; SI-LABEL: no_reorder_scalarized_v2f64_local_load_store
>> ; SI: DS_READ_B64
>> ; SI: DS_READ_B64
>> ; SI: DS_WRITE_B64
>> @@ -32,7 +32,7 @@ define void @no_reorder_scalarized_v2f64_local_load_store(<2 x double> addrspace
>> ret void
>> }
>>
>> -; SI-LABEL: @no_reorder_split_v8i32_global_load_store
>> +; SI-LABEL: no_reorder_split_v8i32_global_load_store
>> ; SI: BUFFER_LOAD_DWORD
>> ; SI: BUFFER_LOAD_DWORD
>> ; SI: BUFFER_LOAD_DWORD
>> @@ -82,7 +82,7 @@ define void @no_reorder_split_v8i32_global_load_store(<8 x i32> addrspace(1)* no
>> ret void
>> }
>>
>> -; SI-LABEL: @no_reorder_extload_64
>> +; SI-LABEL: no_reorder_extload_64
>> ; SI: DS_READ_B64
>> ; SI: DS_READ_B64
>> ; SI: DS_WRITE_B64
>> diff --git a/test/CodeGen/R600/rotl.i64.ll b/test/CodeGen/R600/rotl.i64.ll
>> index a221ce3..d2b6e375 100644
>> --- a/test/CodeGen/R600/rotl.i64.ll
>> +++ b/test/CodeGen/R600/rotl.i64.ll
>> @@ -1,6 +1,6 @@
>> ; RUN: llc -march=r600 -mcpu=SI -verify-machineinstrs < %s | FileCheck -check-prefix=SI -check-prefix=FUNC %s
>>
>> -; FUNC-LABEL: @s_rotl_i64:
>> +; FUNC-LABEL: s_rotl_i64:
>> ; SI-DAG: S_LSHL_B64
>> ; SI-DAG: S_SUB_I32
>> ; SI-DAG: S_LSHR_B64
>> @@ -16,7 +16,7 @@ entry:
>> ret void
>> }
>>
>> -; FUNC-LABEL: @v_rotl_i64:
>> +; FUNC-LABEL: v_rotl_i64:
>> ; SI-DAG: V_LSHL_B64
>> ; SI-DAG: V_SUB_I32
>> ; SI: V_LSHR_B64
>> diff --git a/test/CodeGen/R600/rotl.ll b/test/CodeGen/R600/rotl.ll
>> index a9dee8c..0f47191 100644
>> --- a/test/CodeGen/R600/rotl.ll
>> +++ b/test/CodeGen/R600/rotl.ll
>> @@ -1,7 +1,7 @@
>> ; RUN: llc -march=r600 -mcpu=redwood < %s | FileCheck --check-prefix=R600 -check-prefix=FUNC %s
>> ; RUN: llc -march=r600 -mcpu=SI -verify-machineinstrs < %s | FileCheck -check-prefix=SI -check-prefix=FUNC %s
>>
>> -; FUNC-LABEL: @rotl_i32:
>> +; FUNC-LABEL: rotl_i32:
>> ; R600: SUB_INT {{\** T[0-9]+\.[XYZW]}}, literal.x
>> ; R600-NEXT: 32
>> ; R600: BIT_ALIGN_INT {{T[0-9]+\.[XYZW]}}, KC0[2].Z, KC0[2].Z, PV.{{[XYZW]}}
>> @@ -19,7 +19,7 @@ entry:
>> ret void
>> }
>>
>> -; FUNC-LABEL: @rotl_v2i32
>> +; FUNC-LABEL: rotl_v2i32
>> ; SI-DAG: S_SUB_I32
>> ; SI-DAG: S_SUB_I32
>> ; SI-DAG: V_ALIGNBIT_B32
>> @@ -35,7 +35,7 @@ entry:
>> ret void
>> }
>>
>> -; FUNC-LABEL: @rotl_v4i32
>> +; FUNC-LABEL: rotl_v4i32
>> ; SI-DAG: S_SUB_I32
>> ; SI-DAG: V_ALIGNBIT_B32
>> ; SI-DAG: S_SUB_I32
>> diff --git a/test/CodeGen/R600/rotr.i64.ll b/test/CodeGen/R600/rotr.i64.ll
>> index d15fbc3..3cc4d9c 100644
>> --- a/test/CodeGen/R600/rotr.i64.ll
>> +++ b/test/CodeGen/R600/rotr.i64.ll
>> @@ -1,6 +1,6 @@
>> ; RUN: llc -march=r600 -mcpu=SI -verify-machineinstrs < %s | FileCheck -check-prefix=SI -check-prefix=FUNC %s
>>
>> -; FUNC-LABEL: @s_rotr_i64
>> +; FUNC-LABEL: s_rotr_i64
>> ; SI-DAG: S_SUB_I32
>> ; SI-DAG: S_LSHR_B64
>> ; SI-DAG: S_LSHL_B64
>> @@ -15,7 +15,7 @@ entry:
>> ret void
>> }
>>
>> -; FUNC-LABEL: @v_rotr_i64
>> +; FUNC-LABEL: v_rotr_i64
>> ; SI-DAG: V_SUB_I32
>> ; SI-DAG: V_LSHR_B64
>> ; SI-DAG: V_LSHL_B64
>> @@ -33,7 +33,7 @@ entry:
>> ret void
>> }
>>
>> -; FUNC-LABEL: @s_rotr_v2i64
>> +; FUNC-LABEL: s_rotr_v2i64
>> define void @s_rotr_v2i64(<2 x i64> addrspace(1)* %in, <2 x i64> %x, <2 x i64> %y) {
>> entry:
>> %tmp0 = sub <2 x i64> <i64 64, i64 64>, %y
>> @@ -44,7 +44,7 @@ entry:
>> ret void
>> }
>>
>> -; FUNC-LABEL: @v_rotr_v2i64
>> +; FUNC-LABEL: v_rotr_v2i64
>> define void @v_rotr_v2i64(<2 x i64> addrspace(1)* %in, <2 x i64> addrspace(1)* %xptr, <2 x i64> addrspace(1)* %yptr) {
>> entry:
>> %x = load <2 x i64> addrspace(1)* %xptr, align 8
>> diff --git a/test/CodeGen/R600/rotr.ll b/test/CodeGen/R600/rotr.ll
>> index a5a4da4..9fda9a8 100644
>> --- a/test/CodeGen/R600/rotr.ll
>> +++ b/test/CodeGen/R600/rotr.ll
>> @@ -1,7 +1,7 @@
>> ; RUN: llc -march=r600 -mcpu=redwood < %s | FileCheck --check-prefix=R600 -check-prefix=FUNC %s
>> ; RUN: llc -march=r600 -mcpu=SI -verify-machineinstrs < %s | FileCheck -check-prefix=SI -check-prefix=FUNC %s
>>
>> -; FUNC-LABEL: @rotr_i32:
>> +; FUNC-LABEL: rotr_i32:
>> ; R600: BIT_ALIGN_INT
>>
>> ; SI: V_ALIGNBIT_B32
>> @@ -15,7 +15,7 @@ entry:
>> ret void
>> }
>>
>> -; FUNC-LABEL: @rotr_v2i32:
>> +; FUNC-LABEL: rotr_v2i32:
>> ; R600: BIT_ALIGN_INT
>> ; R600: BIT_ALIGN_INT
>>
>> @@ -31,7 +31,7 @@ entry:
>> ret void
>> }
>>
>> -; FUNC-LABEL: @rotr_v4i32:
>> +; FUNC-LABEL: rotr_v4i32:
>> ; R600: BIT_ALIGN_INT
>> ; R600: BIT_ALIGN_INT
>> ; R600: BIT_ALIGN_INT
>> diff --git a/test/CodeGen/R600/rsq.ll b/test/CodeGen/R600/rsq.ll
>> index a9f3013..8fd50ee 100644
>> --- a/test/CodeGen/R600/rsq.ll
>> +++ b/test/CodeGen/R600/rsq.ll
>> @@ -4,7 +4,7 @@
>> declare float @llvm.sqrt.f32(float) nounwind readnone
>> declare double @llvm.sqrt.f64(double) nounwind readnone
>>
>> -; SI-LABEL: @rsq_f32
>> +; SI-LABEL: rsq_f32
>> ; SI: V_RSQ_F32_e32
>> ; SI: S_ENDPGM
>> define void @rsq_f32(float addrspace(1)* noalias %out, float addrspace(1)* noalias %in) nounwind {
>> @@ -15,7 +15,7 @@ define void @rsq_f32(float addrspace(1)* noalias %out, float addrspace(1)* noali
>> ret void
>> }
>>
>> -; SI-LABEL: @rsq_f64
>> +; SI-LABEL: rsq_f64
>> ; SI-UNSAFE: V_RSQ_F64_e32
>> ; SI-SAFE: V_SQRT_F64_e32
>> ; SI: S_ENDPGM
>> @@ -27,7 +27,7 @@ define void @rsq_f64(double addrspace(1)* noalias %out, double addrspace(1)* noa
>> ret void
>> }
>>
>> -; SI-LABEL: @rsq_f32_sgpr
>> +; SI-LABEL: rsq_f32_sgpr
>> ; SI: V_RSQ_F32_e32 {{v[0-9]+}}, {{s[0-9]+}}
>> ; SI: S_ENDPGM
>> define void @rsq_f32_sgpr(float addrspace(1)* noalias %out, float %val) nounwind {
>> diff --git a/test/CodeGen/R600/saddo.ll b/test/CodeGen/R600/saddo.ll
>> index c80480e..39b7793 100644
>> --- a/test/CodeGen/R600/saddo.ll
>> +++ b/test/CodeGen/R600/saddo.ll
>> @@ -4,7 +4,7 @@
>> declare { i32, i1 } @llvm.sadd.with.overflow.i32(i32, i32) nounwind readnone
>> declare { i64, i1 } @llvm.sadd.with.overflow.i64(i64, i64) nounwind readnone
>>
>> -; FUNC-LABEL: @saddo_i64_zext
>> +; FUNC-LABEL: saddo_i64_zext
>> define void @saddo_i64_zext(i64 addrspace(1)* %out, i64 %a, i64 %b) nounwind {
>> %sadd = call { i64, i1 } @llvm.sadd.with.overflow.i64(i64 %a, i64 %b) nounwind
>> %val = extractvalue { i64, i1 } %sadd, 0
>> @@ -15,7 +15,7 @@ define void @saddo_i64_zext(i64 addrspace(1)* %out, i64 %a, i64 %b) nounwind {
>> ret void
>> }
>>
>> -; FUNC-LABEL: @s_saddo_i32
>> +; FUNC-LABEL: s_saddo_i32
>> define void @s_saddo_i32(i32 addrspace(1)* %out, i1 addrspace(1)* %carryout, i32 %a, i32 %b) nounwind {
>> %sadd = call { i32, i1 } @llvm.sadd.with.overflow.i32(i32 %a, i32 %b) nounwind
>> %val = extractvalue { i32, i1 } %sadd, 0
>> @@ -25,7 +25,7 @@ define void @s_saddo_i32(i32 addrspace(1)* %out, i1 addrspace(1)* %carryout, i32
>> ret void
>> }
>>
>> -; FUNC-LABEL: @v_saddo_i32
>> +; FUNC-LABEL: v_saddo_i32
>> define void @v_saddo_i32(i32 addrspace(1)* %out, i1 addrspace(1)* %carryout, i32 addrspace(1)* %aptr, i32 addrspace(1)* %bptr) nounwind {
>> %a = load i32 addrspace(1)* %aptr, align 4
>> %b = load i32 addrspace(1)* %bptr, align 4
>> @@ -37,7 +37,7 @@ define void @v_saddo_i32(i32 addrspace(1)* %out, i1 addrspace(1)* %carryout, i32
>> ret void
>> }
>>
>> -; FUNC-LABEL: @s_saddo_i64
>> +; FUNC-LABEL: s_saddo_i64
>> define void @s_saddo_i64(i64 addrspace(1)* %out, i1 addrspace(1)* %carryout, i64 %a, i64 %b) nounwind {
>> %sadd = call { i64, i1 } @llvm.sadd.with.overflow.i64(i64 %a, i64 %b) nounwind
>> %val = extractvalue { i64, i1 } %sadd, 0
>> @@ -47,7 +47,7 @@ define void @s_saddo_i64(i64 addrspace(1)* %out, i1 addrspace(1)* %carryout, i64
>> ret void
>> }
>>
>> -; FUNC-LABEL: @v_saddo_i64
>> +; FUNC-LABEL: v_saddo_i64
>> ; SI: V_ADD_I32
>> ; SI: V_ADDC_U32
>> define void @v_saddo_i64(i64 addrspace(1)* %out, i1 addrspace(1)* %carryout, i64 addrspace(1)* %aptr, i64 addrspace(1)* %bptr) nounwind {
>> diff --git a/test/CodeGen/R600/salu-to-valu.ll b/test/CodeGen/R600/salu-to-valu.ll
>> index 98ccff6..2ddffe8 100644
>> --- a/test/CodeGen/R600/salu-to-valu.ll
>> +++ b/test/CodeGen/R600/salu-to-valu.ll
>> @@ -7,7 +7,7 @@
>> ; sgpr register pair and use that for the pointer operand
>> ; (low 64-bits of srsrc).
>>
>> -; CHECK-LABEL: @mubuf
>> +; CHECK-LABEL: mubuf
>>
>> ; Make sure we aren't using VGPRs for the source operand of S_MOV_B64
>> ; CHECK-NOT: S_MOV_B64 s[{{[0-9]+:[0-9]+}}], v
>> @@ -49,7 +49,7 @@ attributes #1 = { nounwind readnone }
>>
>> ; Test moving an SMRD instruction to the VALU
>>
>> -; CHECK-LABEL: @smrd_valu
>> +; CHECK-LABEL: smrd_valu
>> ; CHECK: BUFFER_LOAD_DWORD [[OUT:v[0-9]+]]
>> ; CHECK: BUFFER_STORE_DWORD [[OUT]]
>>
>> @@ -77,7 +77,7 @@ endif:
>>
>> ; Test moving ann SMRD with an immediate offset to the VALU
>>
>> -; CHECK-LABEL: @smrd_valu2
>> +; CHECK-LABEL: smrd_valu2
>> ; CHECK: BUFFER_LOAD_DWORD
>> define void @smrd_valu2(i32 addrspace(1)* %out, [8 x i32] addrspace(2)* %in) {
>> entry:
>> @@ -89,7 +89,7 @@ entry:
>> ret void
>> }
>>
>> -; CHECK-LABEL: @s_load_imm_v8i32
>> +; CHECK-LABEL: s_load_imm_v8i32
>> ; CHECK: BUFFER_LOAD_DWORDX4
>> ; CHECK: BUFFER_LOAD_DWORDX4
>> define void @s_load_imm_v8i32(<8 x i32> addrspace(1)* %out, i32 addrspace(2)* nocapture readonly %in) {
>> @@ -102,7 +102,7 @@ entry:
>> ret void
>> }
>>
>> -; CHECK-LABEL: @s_load_imm_v16i32
>> +; CHECK-LABEL: s_load_imm_v16i32
>> ; CHECK: BUFFER_LOAD_DWORDX4
>> ; CHECK: BUFFER_LOAD_DWORDX4
>> ; CHECK: BUFFER_LOAD_DWORDX4
>> diff --git a/test/CodeGen/R600/scalar_to_vector.ll b/test/CodeGen/R600/scalar_to_vector.ll
>> index bcccb06..afd26c7 100644
>> --- a/test/CodeGen/R600/scalar_to_vector.ll
>> +++ b/test/CodeGen/R600/scalar_to_vector.ll
>> @@ -1,7 +1,7 @@
>> ; RUN: llc -march=r600 -mcpu=SI -verify-machineinstrs < %s | FileCheck -check-prefix=SI -check-prefix=FUNC %s
>>
>>
>> -; FUNC-LABEL: @scalar_to_vector_v2i32
>> +; FUNC-LABEL: scalar_to_vector_v2i32
>> ; SI: BUFFER_LOAD_DWORD [[VAL:v[0-9]+]],
>> ; SI: V_LSHRREV_B32_e32 [[RESULT:v[0-9]+]], 16, [[VAL]]
>> ; SI: BUFFER_STORE_SHORT [[RESULT]]
>> @@ -17,7 +17,7 @@ define void @scalar_to_vector_v2i32(<4 x i16> addrspace(1)* %out, i32 addrspace(
>> ret void
>> }
>>
>> -; FUNC-LABEL: @scalar_to_vector_v2f32
>> +; FUNC-LABEL: scalar_to_vector_v2f32
>> ; SI: BUFFER_LOAD_DWORD [[VAL:v[0-9]+]],
>> ; SI: V_LSHRREV_B32_e32 [[RESULT:v[0-9]+]], 16, [[VAL]]
>> ; SI: BUFFER_STORE_SHORT [[RESULT]]
>> diff --git a/test/CodeGen/R600/schedule-global-loads.ll b/test/CodeGen/R600/schedule-global-loads.ll
>> index fcff65f..27ba96e 100644
>> --- a/test/CodeGen/R600/schedule-global-loads.ll
>> +++ b/test/CodeGen/R600/schedule-global-loads.ll
>> @@ -8,7 +8,7 @@ declare i32 @llvm.r600.read.tidig.x() #1
>> ; seems the only things areLoadsFromSameBasePtr is accomplishing is
>> ; ordering the loads so that the lower address loads come first.
>>
>> -; FUNC-LABEL: @cluster_global_arg_loads
>> +; FUNC-LABEL: cluster_global_arg_loads
>> ; SI-DAG: BUFFER_LOAD_DWORD [[REG0:v[0-9]+]], s{{\[[0-9]+:[0-9]+\]}}, 0{{$}}
>> ; SI-DAG: BUFFER_LOAD_DWORD [[REG1:v[0-9]+]], s{{\[[0-9]+:[0-9]+\]}}, 0 offset:0x4
>> ; SI: BUFFER_STORE_DWORD [[REG0]]
>> @@ -24,7 +24,7 @@ define void @cluster_global_arg_loads(i32 addrspace(1)* %out0, i32 addrspace(1)*
>>
>> ; Test for a crach in SIInstrInfo::areLoadsFromSameBasePtr() when checking
>> ; an MUBUF load which does not have a vaddr operand.
>> -; FUNC-LABEL: @same_base_ptr_crash
>> +; FUNC-LABEL: same_base_ptr_crash
>> ; SI: BUFFER_LOAD_DWORD
>> ; SI: BUFFER_LOAD_DWORD
>> define void @same_base_ptr_crash(i32 addrspace(1)* %out, i32 addrspace(1)* %in, i32 %offset) {
>> diff --git a/test/CodeGen/R600/schedule-kernel-arg-loads.ll b/test/CodeGen/R600/schedule-kernel-arg-loads.ll
>> index 34b7098..957f697 100644
>> --- a/test/CodeGen/R600/schedule-kernel-arg-loads.ll
>> +++ b/test/CodeGen/R600/schedule-kernel-arg-loads.ll
>> @@ -1,6 +1,6 @@
>> ; RUN: llc -march=r600 -mcpu=SI -verify-machineinstrs < %s | FileCheck -check-prefix=FUNC -check-prefix=SI %s
>>
>> -; FUNC-LABEL: @cluster_arg_loads
>> +; FUNC-LABEL: cluster_arg_loads
>> ; SI: S_LOAD_DWORDX2 s{{\[[0-9]+:[0-9]+\]}}, s{{\[[0-9]+:[0-9]+\]}}, 0x9
>> ; SI-NEXT: S_LOAD_DWORDX2 s{{\[[0-9]+:[0-9]+\]}}, s{{\[[0-9]+:[0-9]+\]}}, 0xb
>> ; SI-NEXT: S_LOAD_DWORD s{{[0-9]+}}, s{{\[[0-9]+:[0-9]+\]}}, 0xd
>> diff --git a/test/CodeGen/R600/schedule-vs-if-nested-loop-failure.ll b/test/CodeGen/R600/schedule-vs-if-nested-loop-failure.ll
>> index 3d2142d..070a850 100644
>> --- a/test/CodeGen/R600/schedule-vs-if-nested-loop-failure.ll
>> +++ b/test/CodeGen/R600/schedule-vs-if-nested-loop-failure.ll
>> @@ -5,7 +5,7 @@
>> declare void @llvm.AMDGPU.barrier.local() nounwind noduplicate
>>
>>
>> -; SI-LABEL: @main(
>> +; SI-LABEL: main(
>> define void @main(<4 x float> inreg %reg0, <4 x float> inreg %reg1) #0 {
>> main_body:
>> %0 = extractelement <4 x float> %reg1, i32 0
>> diff --git a/test/CodeGen/R600/sdiv.ll b/test/CodeGen/R600/sdiv.ll
>> index e922d5c..50d824b 100644
>> --- a/test/CodeGen/R600/sdiv.ll
>> +++ b/test/CodeGen/R600/sdiv.ll
>> @@ -10,7 +10,7 @@
>> ; This was fixed by adding an additional pattern in R600Instructions.td to
>> ; match this pattern with a CNDGE_INT.
>>
>> -; FUNC-LABEL: @sdiv_i32
>> +; FUNC-LABEL: sdiv_i32
>> ; EG: CF_END
>> define void @sdiv_i32(i32 addrspace(1)* %out, i32 addrspace(1)* %in) {
>> %den_ptr = getelementptr i32 addrspace(1)* %in, i32 1
>> @@ -21,7 +21,7 @@ define void @sdiv_i32(i32 addrspace(1)* %out, i32 addrspace(1)* %in) {
>> ret void
>> }
>>
>> -; FUNC-LABEL: @sdiv_i32_4
>> +; FUNC-LABEL: sdiv_i32_4
>> define void @sdiv_i32_4(i32 addrspace(1)* %out, i32 addrspace(1)* %in) {
>> %num = load i32 addrspace(1) * %in
>> %result = sdiv i32 %num, 4
>> @@ -32,7 +32,7 @@ define void @sdiv_i32_4(i32 addrspace(1)* %out, i32 addrspace(1)* %in) {
>> ; Multiply by a weird constant to make sure setIntDivIsCheap is
>> ; working.
>>
>> -; FUNC-LABEL: @slow_sdiv_i32_3435
>> +; FUNC-LABEL: slow_sdiv_i32_3435
>> ; SI: BUFFER_LOAD_DWORD [[VAL:v[0-9]+]],
>> ; SI: V_MOV_B32_e32 [[MAGIC:v[0-9]+]], 0x98a1930b
>> ; SI: V_MUL_HI_I32 [[TMP:v[0-9]+]], [[VAL]], [[MAGIC]]
>> diff --git a/test/CodeGen/R600/sdivrem24.ll b/test/CodeGen/R600/sdivrem24.ll
>> index 55f6d2c..da66662 100644
>> --- a/test/CodeGen/R600/sdivrem24.ll
>> +++ b/test/CodeGen/R600/sdivrem24.ll
>> @@ -1,7 +1,7 @@
>> ; RUN: llc -march=r600 -mcpu=SI < %s | FileCheck -check-prefix=SI -check-prefix=FUNC %s
>> ; RUN: llc -march=r600 -mcpu=redwood < %s | FileCheck -check-prefix=EG -check-prefix=FUNC %s
>>
>> -; FUNC-LABEL: @sdiv24_i8
>> +; FUNC-LABEL: sdiv24_i8
>> ; SI: V_CVT_F32_I32
>> ; SI: V_CVT_F32_I32
>> ; SI: V_RCP_F32
>> @@ -20,7 +20,7 @@ define void @sdiv24_i8(i8 addrspace(1)* %out, i8 addrspace(1)* %in) {
>> ret void
>> }
>>
>> -; FUNC-LABEL: @sdiv24_i16
>> +; FUNC-LABEL: sdiv24_i16
>> ; SI: V_CVT_F32_I32
>> ; SI: V_CVT_F32_I32
>> ; SI: V_RCP_F32
>> @@ -39,7 +39,7 @@ define void @sdiv24_i16(i16 addrspace(1)* %out, i16 addrspace(1)* %in) {
>> ret void
>> }
>>
>> -; FUNC-LABEL: @sdiv24_i32
>> +; FUNC-LABEL: sdiv24_i32
>> ; SI: V_CVT_F32_I32
>> ; SI: V_CVT_F32_I32
>> ; SI: V_RCP_F32
>> @@ -62,7 +62,7 @@ define void @sdiv24_i32(i32 addrspace(1)* %out, i32 addrspace(1)* %in) {
>> ret void
>> }
>>
>> -; FUNC-LABEL: @sdiv25_i32
>> +; FUNC-LABEL: sdiv25_i32
>> ; SI-NOT: V_CVT_F32_I32
>> ; SI-NOT: V_RCP_F32
>>
>> @@ -81,7 +81,7 @@ define void @sdiv25_i32(i32 addrspace(1)* %out, i32 addrspace(1)* %in) {
>> ret void
>> }
>>
>> -; FUNC-LABEL: @test_no_sdiv24_i32_1
>> +; FUNC-LABEL: test_no_sdiv24_i32_1
>> ; SI-NOT: V_CVT_F32_I32
>> ; SI-NOT: V_RCP_F32
>>
>> @@ -100,7 +100,7 @@ define void @test_no_sdiv24_i32_1(i32 addrspace(1)* %out, i32 addrspace(1)* %in)
>> ret void
>> }
>>
>> -; FUNC-LABEL: @test_no_sdiv24_i32_2
>> +; FUNC-LABEL: test_no_sdiv24_i32_2
>> ; SI-NOT: V_CVT_F32_I32
>> ; SI-NOT: V_RCP_F32
>>
>> @@ -119,7 +119,7 @@ define void @test_no_sdiv24_i32_2(i32 addrspace(1)* %out, i32 addrspace(1)* %in)
>> ret void
>> }
>>
>> -; FUNC-LABEL: @srem24_i8
>> +; FUNC-LABEL: srem24_i8
>> ; SI: V_CVT_F32_I32
>> ; SI: V_CVT_F32_I32
>> ; SI: V_RCP_F32
>> @@ -138,7 +138,7 @@ define void @srem24_i8(i8 addrspace(1)* %out, i8 addrspace(1)* %in) {
>> ret void
>> }
>>
>> -; FUNC-LABEL: @srem24_i16
>> +; FUNC-LABEL: srem24_i16
>> ; SI: V_CVT_F32_I32
>> ; SI: V_CVT_F32_I32
>> ; SI: V_RCP_F32
>> @@ -157,7 +157,7 @@ define void @srem24_i16(i16 addrspace(1)* %out, i16 addrspace(1)* %in) {
>> ret void
>> }
>>
>> -; FUNC-LABEL: @srem24_i32
>> +; FUNC-LABEL: srem24_i32
>> ; SI: V_CVT_F32_I32
>> ; SI: V_CVT_F32_I32
>> ; SI: V_RCP_F32
>> @@ -180,7 +180,7 @@ define void @srem24_i32(i32 addrspace(1)* %out, i32 addrspace(1)* %in) {
>> ret void
>> }
>>
>> -; FUNC-LABEL: @srem25_i32
>> +; FUNC-LABEL: srem25_i32
>> ; SI-NOT: V_CVT_F32_I32
>> ; SI-NOT: V_RCP_F32
>>
>> @@ -199,7 +199,7 @@ define void @srem25_i32(i32 addrspace(1)* %out, i32 addrspace(1)* %in) {
>> ret void
>> }
>>
>> -; FUNC-LABEL: @test_no_srem24_i32_1
>> +; FUNC-LABEL: test_no_srem24_i32_1
>> ; SI-NOT: V_CVT_F32_I32
>> ; SI-NOT: V_RCP_F32
>>
>> @@ -218,7 +218,7 @@ define void @test_no_srem24_i32_1(i32 addrspace(1)* %out, i32 addrspace(1)* %in)
>> ret void
>> }
>>
>> -; FUNC-LABEL: @test_no_srem24_i32_2
>> +; FUNC-LABEL: test_no_srem24_i32_2
>> ; SI-NOT: V_CVT_F32_I32
>> ; SI-NOT: V_RCP_F32
>>
>> diff --git a/test/CodeGen/R600/select-i1.ll b/test/CodeGen/R600/select-i1.ll
>> index 009dd7f..5419ffb 100644
>> --- a/test/CodeGen/R600/select-i1.ll
>> +++ b/test/CodeGen/R600/select-i1.ll
>> @@ -2,7 +2,7 @@
>>
>> ; FIXME: This should go in existing select.ll test, except the current testcase there is broken on SI
>>
>> -; FUNC-LABEL: @select_i1
>> +; FUNC-LABEL: select_i1
>> ; SI: V_CNDMASK_B32
>> ; SI-NOT: V_CNDMASK_B32
>> define void @select_i1(i1 addrspace(1)* %out, i32 %cond, i1 %a, i1 %b) nounwind {
>> diff --git a/test/CodeGen/R600/select-vectors.ll b/test/CodeGen/R600/select-vectors.ll
>> index 94605fe..8553f24 100644
>> --- a/test/CodeGen/R600/select-vectors.ll
>> +++ b/test/CodeGen/R600/select-vectors.ll
>> @@ -4,7 +4,7 @@
>> ; Evergreen not enabled since it seems to be having problems with doubles.
>>
>>
>> -; FUNC-LABEL: @select_v4i8
>> +; FUNC-LABEL: select_v4i8
>> ; SI: V_CNDMASK_B32_e64
>> ; SI: V_CNDMASK_B32_e64
>> ; SI: V_CNDMASK_B32_e64
>> @@ -16,7 +16,7 @@ define void @select_v4i8(<4 x i8> addrspace(1)* %out, <4 x i8> %a, <4 x i8> %b,
>> ret void
>> }
>>
>> -; FUNC-LABEL: @select_v4i16
>> +; FUNC-LABEL: select_v4i16
>> ; SI: V_CNDMASK_B32_e64
>> ; SI: V_CNDMASK_B32_e64
>> ; SI: V_CNDMASK_B32_e64
>> @@ -28,7 +28,7 @@ define void @select_v4i16(<4 x i16> addrspace(1)* %out, <4 x i16> %a, <4 x i16>
>> ret void
>> }
>>
>> -; FUNC-LABEL: @select_v2i32
>> +; FUNC-LABEL: select_v2i32
>> ; SI: V_CNDMASK_B32_e64
>> ; SI: V_CNDMASK_B32_e64
>> ; SI: BUFFER_STORE_DWORDX2
>> @@ -39,7 +39,7 @@ define void @select_v2i32(<2 x i32> addrspace(1)* %out, <2 x i32> %a, <2 x i32>
>> ret void
>> }
>>
>> -; FUNC-LABEL: @select_v4i32
>> +; FUNC-LABEL: select_v4i32
>> ; SI: V_CNDMASK_B32_e64
>> ; SI: V_CNDMASK_B32_e64
>> ; SI: V_CNDMASK_B32_e64
>> @@ -52,7 +52,7 @@ define void @select_v4i32(<4 x i32> addrspace(1)* %out, <4 x i32> %a, <4 x i32>
>> ret void
>> }
>>
>> -; FUNC-LABEL: @select_v8i32
>> +; FUNC-LABEL: select_v8i32
>> ; SI: V_CNDMASK_B32_e64
>> ; SI: V_CNDMASK_B32_e64
>> ; SI: V_CNDMASK_B32_e64
>> @@ -68,7 +68,7 @@ define void @select_v8i32(<8 x i32> addrspace(1)* %out, <8 x i32> %a, <8 x i32>
>> ret void
>> }
>>
>> -; FUNC-LABEL: @select_v2f32
>> +; FUNC-LABEL: select_v2f32
>> ; SI: BUFFER_STORE_DWORDX2
>> define void @select_v2f32(<2 x float> addrspace(1)* %out, <2 x float> %a, <2 x float> %b, i32 %c) nounwind {
>> %cmp = icmp eq i32 %c, 0
>> @@ -77,7 +77,7 @@ define void @select_v2f32(<2 x float> addrspace(1)* %out, <2 x float> %a, <2 x f
>> ret void
>> }
>>
>> -; FUNC-LABEL: @select_v4f32
>> +; FUNC-LABEL: select_v4f32
>> ; SI: BUFFER_STORE_DWORDX4
>> define void @select_v4f32(<4 x float> addrspace(1)* %out, <4 x float> %a, <4 x float> %b, i32 %c) nounwind {
>> %cmp = icmp eq i32 %c, 0
>> @@ -86,7 +86,7 @@ define void @select_v4f32(<4 x float> addrspace(1)* %out, <4 x float> %a, <4 x f
>> ret void
>> }
>>
>> -; FUNC-LABEL: @select_v8f32
>> +; FUNC-LABEL: select_v8f32
>> ; SI: V_CNDMASK_B32_e64
>> ; SI: V_CNDMASK_B32_e64
>> ; SI: V_CNDMASK_B32_e64
>> @@ -102,7 +102,7 @@ define void @select_v8f32(<8 x float> addrspace(1)* %out, <8 x float> %a, <8 x f
>> ret void
>> }
>>
>> -; FUNC-LABEL: @select_v2f64
>> +; FUNC-LABEL: select_v2f64
>> ; SI: V_CNDMASK_B32_e64
>> ; SI: V_CNDMASK_B32_e64
>> ; SI: V_CNDMASK_B32_e64
>> @@ -114,7 +114,7 @@ define void @select_v2f64(<2 x double> addrspace(1)* %out, <2 x double> %a, <2 x
>> ret void
>> }
>>
>> -; FUNC-LABEL: @select_v4f64
>> +; FUNC-LABEL: select_v4f64
>> ; SI: V_CNDMASK_B32_e64
>> ; SI: V_CNDMASK_B32_e64
>> ; SI: V_CNDMASK_B32_e64
>> @@ -130,7 +130,7 @@ define void @select_v4f64(<4 x double> addrspace(1)* %out, <4 x double> %a, <4 x
>> ret void
>> }
>>
>> -; FUNC-LABEL: @select_v8f64
>> +; FUNC-LABEL: select_v8f64
>> ; SI: V_CNDMASK_B32_e64
>> ; SI: V_CNDMASK_B32_e64
>> ; SI: V_CNDMASK_B32_e64
>> diff --git a/test/CodeGen/R600/select.ll b/test/CodeGen/R600/select.ll
>> index 7d51568..9dc1b4b 100644
>> --- a/test/CodeGen/R600/select.ll
>> +++ b/test/CodeGen/R600/select.ll
>> @@ -7,7 +7,7 @@
>> ; In order to avoid the select_cc optimization, this test case calculates the
>> ; condition for the select in a separate basic block.
>>
>> -; FUNC-LABEL: @select
>> +; FUNC-LABEL: select
>> ; EG-DAG: MEM_RAT_CACHELESS STORE_RAW T{{[0-9]+}}.X
>> ; EG-DAG: MEM_RAT_CACHELESS STORE_RAW T{{[0-9]+}}.X
>> ; EG-DAG: MEM_RAT_CACHELESS STORE_RAW T{{[0-9]+}}.XY
>> diff --git a/test/CodeGen/R600/select64.ll b/test/CodeGen/R600/select64.ll
>> index dba25e3..c6427bf 100644
>> --- a/test/CodeGen/R600/select64.ll
>> +++ b/test/CodeGen/R600/select64.ll
>> @@ -1,6 +1,6 @@
>> ; RUN: llc < %s -march=r600 -mcpu=SI -verify-machineinstrs | FileCheck %s
>>
>> -; CHECK-LABEL: @select0
>> +; CHECK-LABEL: select0
>> ; i64 select should be split into two i32 selects, and we shouldn't need
>> ; to use a shfit to extract the hi dword of the input.
>> ; CHECK-NOT: S_LSHR_B64
>> @@ -14,7 +14,7 @@ entry:
>> ret void
>> }
>>
>> -; CHECK-LABEL: @select_trunc_i64
>> +; CHECK-LABEL: select_trunc_i64
>> ; CHECK: V_CNDMASK_B32
>> ; CHECK-NOT: V_CNDMASK_B32
>> define void @select_trunc_i64(i32 addrspace(1)* %out, i32 %cond, i64 %in) nounwind {
>> @@ -25,7 +25,7 @@ define void @select_trunc_i64(i32 addrspace(1)* %out, i32 %cond, i64 %in) nounwi
>> ret void
>> }
>>
>> -; CHECK-LABEL: @select_trunc_i64_2
>> +; CHECK-LABEL: select_trunc_i64_2
>> ; CHECK: V_CNDMASK_B32
>> ; CHECK-NOT: V_CNDMASK_B32
>> define void @select_trunc_i64_2(i32 addrspace(1)* %out, i32 %cond, i64 %a, i64 %b) nounwind {
>> @@ -36,7 +36,7 @@ define void @select_trunc_i64_2(i32 addrspace(1)* %out, i32 %cond, i64 %a, i64 %
>> ret void
>> }
>>
>> -; CHECK-LABEL: @v_select_trunc_i64_2
>> +; CHECK-LABEL: v_select_trunc_i64_2
>> ; CHECK: V_CNDMASK_B32
>> ; CHECK-NOT: V_CNDMASK_B32
>> define void @v_select_trunc_i64_2(i32 addrspace(1)* %out, i32 %cond, i64 addrspace(1)* %aptr, i64 addrspace(1)* %bptr) nounwind {
>> diff --git a/test/CodeGen/R600/selectcc-opt.ll b/test/CodeGen/R600/selectcc-opt.ll
>> index bdb6867..94d5516 100644
>> --- a/test/CodeGen/R600/selectcc-opt.ll
>> +++ b/test/CodeGen/R600/selectcc-opt.ll
>> @@ -2,7 +2,7 @@
>> ; RUN: llc -march=r600 -mcpu=redwood < %s | FileCheck -check-prefix=EG -check-prefix=FUNC %s
>>
>>
>> -; FUNC-LABEL: @test_a
>> +; FUNC-LABEL: test_a
>> ; EG-NOT: CND
>> ; EG: SET{{[NEQGTL]+}}_DX10
>>
>> @@ -30,7 +30,7 @@ ENDIF:
>> ; Same as test_a, but the branch labels are swapped to produce the inverse cc
>> ; for the icmp instruction
>>
>> -; EG-LABEL: @test_b
>> +; EG-LABEL: test_b
>> ; EG: SET{{[GTEQN]+}}_DX10
>> ; EG-NEXT: PRED_
>> ; EG-NEXT: ALU clause starting
>> @@ -56,7 +56,7 @@ ENDIF:
>> }
>>
>> ; Test a CND*_INT instruction with float true/false values
>> -; EG-LABEL: @test_c
>> +; EG-LABEL: test_c
>> ; EG: CND{{[GTE]+}}_INT
>> define void @test_c(float addrspace(1)* %out, i32 %in) {
>> entry:
>> @@ -66,7 +66,7 @@ entry:
>> ret void
>> }
>>
>> -; FUNC-LABEL: @selectcc_bool
>> +; FUNC-LABEL: selectcc_bool
>> ; SI: V_CMP_NE_I32
>> ; SI-NEXT: V_CNDMASK_B32_e64
>> ; SI-NOT: CMP
>> diff --git a/test/CodeGen/R600/selectcc.ll b/test/CodeGen/R600/selectcc.ll
>> index a8f57cf..55ed945e 100644
>> --- a/test/CodeGen/R600/selectcc.ll
>> +++ b/test/CodeGen/R600/selectcc.ll
>> @@ -1,7 +1,7 @@
>> ; RUN: llc -verify-machineinstrs -march=r600 -mcpu=redwood < %s | FileCheck -check-prefix=EG -check-prefix=FUNC %s
>> ; RUN: llc -verify-machineinstrs -march=r600 -mcpu=SI < %s | FileCheck -check-prefix=SI -check-prefix=FUNC %s
>>
>> -; FUNC-LABEL: @selectcc_i64
>> +; FUNC-LABEL: selectcc_i64
>> ; EG: XOR_INT
>> ; EG: XOR_INT
>> ; EG: OR_INT
>> diff --git a/test/CodeGen/R600/set-dx10.ll b/test/CodeGen/R600/set-dx10.ll
>> index 5c7d499..fc28e2d 100644
>> --- a/test/CodeGen/R600/set-dx10.ll
>> +++ b/test/CodeGen/R600/set-dx10.ll
>> @@ -4,7 +4,7 @@
>> ; to store integer true (-1) and false (0) values are lowered to one of the
>> ; SET*DX10 instructions.
>>
>> -; CHECK: @fcmp_une_select_fptosi
>> +; CHECK: fcmp_une_select_fptosi
>> ; CHECK: SETNE_DX10 {{\** *}}T{{[0-9]+\.[XYZW]}}, KC0[2].Z, literal.x,
>> ; CHECK-NEXT: LSHR
>> ; CHECK-NEXT: 1084227584(5.000000e+00)
>> @@ -18,7 +18,7 @@ entry:
>> ret void
>> }
>>
>> -; CHECK: @fcmp_une_select_i32
>> +; CHECK: fcmp_une_select_i32
>> ; CHECK: SETNE_DX10 {{\** *}}T{{[0-9]+\.[XYZW]}}, KC0[2].Z, literal.x,
>> ; CHECK-NEXT: LSHR
>> ; CHECK-NEXT: 1084227584(5.000000e+00)
>> @@ -30,7 +30,7 @@ entry:
>> ret void
>> }
>>
>> -; CHECK: @fcmp_oeq_select_fptosi
>> +; CHECK: fcmp_oeq_select_fptosi
>> ; CHECK: SETE_DX10 {{\** *}}T{{[0-9]+\.[XYZW]}}, KC0[2].Z, literal.x,
>> ; CHECK-NEXT: LSHR
>> ; CHECK-NEXT: 1084227584(5.000000e+00)
>> @@ -44,7 +44,7 @@ entry:
>> ret void
>> }
>>
>> -; CHECK: @fcmp_oeq_select_i32
>> +; CHECK: fcmp_oeq_select_i32
>> ; CHECK: SETE_DX10 {{\** *}}T{{[0-9]+\.[XYZW]}}, KC0[2].Z, literal.x,
>> ; CHECK-NEXT: LSHR
>> ; CHECK-NEXT: 1084227584(5.000000e+00)
>> @@ -56,7 +56,7 @@ entry:
>> ret void
>> }
>>
>> -; CHECK: @fcmp_ogt_select_fptosi
>> +; CHECK: fcmp_ogt_select_fptosi
>> ; CHECK: SETGT_DX10 {{\** *}}T{{[0-9]+\.[XYZW]}}, KC0[2].Z, literal.x,
>> ; CHECK-NEXT: LSHR
>> ; CHECK-NEXT: 1084227584(5.000000e+00)
>> @@ -70,7 +70,7 @@ entry:
>> ret void
>> }
>>
>> -; CHECK: @fcmp_ogt_select_i32
>> +; CHECK: fcmp_ogt_select_i32
>> ; CHECK: SETGT_DX10 {{\** *}}T{{[0-9]+\.[XYZW]}}, KC0[2].Z, literal.x,
>> ; CHECK-NEXT: LSHR
>> ; CHECK-NEXT: 1084227584(5.000000e+00)
>> @@ -82,7 +82,7 @@ entry:
>> ret void
>> }
>>
>> -; CHECK: @fcmp_oge_select_fptosi
>> +; CHECK: fcmp_oge_select_fptosi
>> ; CHECK: SETGE_DX10 {{\** *}}T{{[0-9]+\.[XYZW]}}, KC0[2].Z, literal.x,
>> ; CHECK-NEXT: LSHR
>> ; CHECK-NEXT: 1084227584(5.000000e+00)
>> @@ -96,7 +96,7 @@ entry:
>> ret void
>> }
>>
>> -; CHECK: @fcmp_oge_select_i32
>> +; CHECK: fcmp_oge_select_i32
>> ; CHECK: SETGE_DX10 {{\** *}}T{{[0-9]+\.[XYZW]}}, KC0[2].Z, literal.x,
>> ; CHECK-NEXT: LSHR
>> ; CHECK-NEXT: 1084227584(5.000000e+00)
>> @@ -108,7 +108,7 @@ entry:
>> ret void
>> }
>>
>> -; CHECK: @fcmp_ole_select_fptosi
>> +; CHECK: fcmp_ole_select_fptosi
>> ; CHECK: SETGE_DX10 {{\** *}}T{{[0-9]+\.[XYZW]}}, literal.x, KC0[2].Z,
>> ; CHECK-NEXT: LSHR
>> ; CHECK-NEXT: 1084227584(5.000000e+00)
>> @@ -122,7 +122,7 @@ entry:
>> ret void
>> }
>>
>> -; CHECK: @fcmp_ole_select_i32
>> +; CHECK: fcmp_ole_select_i32
>> ; CHECK: SETGE_DX10 {{\** *}}T{{[0-9]+\.[XYZW]}}, literal.x, KC0[2].Z,
>> ; CHECK-NEXT: LSHR
>> ; CHECK-NEXT: 1084227584(5.000000e+00)
>> @@ -134,7 +134,7 @@ entry:
>> ret void
>> }
>>
>> -; CHECK: @fcmp_olt_select_fptosi
>> +; CHECK: fcmp_olt_select_fptosi
>> ; CHECK: SETGT_DX10 {{\** *}}T{{[0-9]+\.[XYZW]}}, literal.x, KC0[2].Z,
>> ; CHECK-NEXT: LSHR
>> ; CHECK-NEXT: 1084227584(5.000000e+00)
>> @@ -148,7 +148,7 @@ entry:
>> ret void
>> }
>>
>> -; CHECK: @fcmp_olt_select_i32
>> +; CHECK: fcmp_olt_select_i32
>> ; CHECK: SETGT_DX10 {{\** *}}T{{[0-9]+\.[XYZW]}}, literal.x, KC0[2].Z,
>> ; CHECK-NEXT: LSHR
>> ; CHECK-NEXT: 1084227584(5.000000e+00)
>> diff --git a/test/CodeGen/R600/setcc-equivalent.ll b/test/CodeGen/R600/setcc-equivalent.ll
>> index e21cecf..4dc1b73 100644
>> --- a/test/CodeGen/R600/setcc-equivalent.ll
>> +++ b/test/CodeGen/R600/setcc-equivalent.ll
>> @@ -1,6 +1,6 @@
>> ; RUN: llc -march=r600 -mcpu=cypress < %s | FileCheck -check-prefix=EG %s
>>
>> -; EG-LABEL: @and_setcc_setcc_i32
>> +; EG-LABEL: and_setcc_setcc_i32
>> ; EG: AND_INT
>> ; EG-NEXT: SETE_INT
>> define void @and_setcc_setcc_i32(i32 addrspace(1)* %out, i32 %a, i32 %b) {
>> @@ -12,7 +12,7 @@ define void @and_setcc_setcc_i32(i32 addrspace(1)* %out, i32 %a, i32 %b) {
>> ret void
>> }
>>
>> -; EG-LABEL: @and_setcc_setcc_v4i32
>> +; EG-LABEL: and_setcc_setcc_v4i32
>> ; EG: AND_INT
>> ; EG: AND_INT
>> ; EG: SETE_INT
>> diff --git a/test/CodeGen/R600/setcc-opt.ll b/test/CodeGen/R600/setcc-opt.ll
>> index 8e831e4..d784895 100644
>> --- a/test/CodeGen/R600/setcc-opt.ll
>> +++ b/test/CodeGen/R600/setcc-opt.ll
>> @@ -1,6 +1,6 @@
>> ; RUN: llc -march=r600 -mcpu=SI -verify-machineinstrs < %s | FileCheck -check-prefix=SI -check-prefix=FUNC %s
>>
>> -; SI-LABEL: @sext_bool_icmp_ne
>> +; SI-LABEL: sext_bool_icmp_ne
>> ; SI: V_CMP_NE_I32
>> ; SI-NEXT: V_CNDMASK_B32
>> ; SI-NOT: V_CMP_NE_I32
>> diff --git a/test/CodeGen/R600/setcc.ll b/test/CodeGen/R600/setcc.ll
>> index 5bd95b7..df0b3bd 100644
>> --- a/test/CodeGen/R600/setcc.ll
>> +++ b/test/CodeGen/R600/setcc.ll
>> @@ -1,7 +1,7 @@
>> ;RUN: llc < %s -march=r600 -mcpu=redwood | FileCheck --check-prefix=R600 --check-prefix=FUNC %s
>> ;RUN: llc < %s -march=r600 -mcpu=SI -verify-machineinstrs| FileCheck --check-prefix=SI --check-prefix=FUNC %s
>>
>> -; FUNC-LABEL: @setcc_v2i32
>> +; FUNC-LABEL: setcc_v2i32
>> ; R600-DAG: SETE_INT * T{{[0-9]+\.[XYZW]}}, KC0[3].X, KC0[3].Z
>> ; R600-DAG: SETE_INT * T{{[0-9]+\.[XYZW]}}, KC0[2].W, KC0[3].Y
>>
>> @@ -12,7 +12,7 @@ define void @setcc_v2i32(<2 x i32> addrspace(1)* %out, <2 x i32> %a, <2 x i32> %
>> ret void
>> }
>>
>> -; FUNC-LABEL: @setcc_v4i32
>> +; FUNC-LABEL: setcc_v4i32
>> ; R600-DAG: SETE_INT * T{{[0-9]+\.[XYZW], T[0-9]+\.[XYZW], T[0-9]+\.[XYZW]}}
>> ; R600-DAG: SETE_INT * T{{[0-9]+\.[XYZW], T[0-9]+\.[XYZW], T[0-9]+\.[XYZW]}}
>> ; R600-DAG: SETE_INT * T{{[0-9]+\.[XYZW], T[0-9]+\.[XYZW], T[0-9]+\.[XYZW]}}
>> @@ -32,7 +32,7 @@ define void @setcc_v4i32(<4 x i32> addrspace(1)* %out, <4 x i32> addrspace(1)* %
>> ;; Float comparisons
>> ;;;==========================================================================;;;
>>
>> -; FUNC-LABEL: @f32_oeq
>> +; FUNC-LABEL: f32_oeq
>> ; R600: SETE_DX10
>> ; SI: V_CMP_EQ_F32
>> define void @f32_oeq(i32 addrspace(1)* %out, float %a, float %b) {
>> @@ -43,7 +43,7 @@ entry:
>> ret void
>> }
>>
>> -; FUNC-LABEL: @f32_ogt
>> +; FUNC-LABEL: f32_ogt
>> ; R600: SETGT_DX10
>> ; SI: V_CMP_GT_F32
>> define void @f32_ogt(i32 addrspace(1)* %out, float %a, float %b) {
>> @@ -54,7 +54,7 @@ entry:
>> ret void
>> }
>>
>> -; FUNC-LABEL: @f32_oge
>> +; FUNC-LABEL: f32_oge
>> ; R600: SETGE_DX10
>> ; SI: V_CMP_GE_F32
>> define void @f32_oge(i32 addrspace(1)* %out, float %a, float %b) {
>> @@ -65,7 +65,7 @@ entry:
>> ret void
>> }
>>
>> -; FUNC-LABEL: @f32_olt
>> +; FUNC-LABEL: f32_olt
>> ; R600: SETGT_DX10
>> ; SI: V_CMP_LT_F32
>> define void @f32_olt(i32 addrspace(1)* %out, float %a, float %b) {
>> @@ -76,7 +76,7 @@ entry:
>> ret void
>> }
>>
>> -; FUNC-LABEL: @f32_ole
>> +; FUNC-LABEL: f32_ole
>> ; R600: SETGE_DX10
>> ; SI: V_CMP_LE_F32
>> define void @f32_ole(i32 addrspace(1)* %out, float %a, float %b) {
>> @@ -87,7 +87,7 @@ entry:
>> ret void
>> }
>>
>> -; FUNC-LABEL: @f32_one
>> +; FUNC-LABEL: f32_one
>> ; R600-DAG: SETE_DX10
>> ; R600-DAG: SETE_DX10
>> ; R600-DAG: AND_INT
>> @@ -107,7 +107,7 @@ entry:
>> ret void
>> }
>>
>> -; FUNC-LABEL: @f32_ord
>> +; FUNC-LABEL: f32_ord
>> ; R600-DAG: SETE_DX10
>> ; R600-DAG: SETE_DX10
>> ; R600-DAG: AND_INT
>> @@ -121,7 +121,7 @@ entry:
>> ret void
>> }
>>
>> -; FUNC-LABEL: @f32_ueq
>> +; FUNC-LABEL: f32_ueq
>> ; R600-DAG: SETNE_DX10
>> ; R600-DAG: SETNE_DX10
>> ; R600-DAG: OR_INT
>> @@ -141,7 +141,7 @@ entry:
>> ret void
>> }
>>
>> -; FUNC-LABEL: @f32_ugt
>> +; FUNC-LABEL: f32_ugt
>> ; R600: SETGE
>> ; R600: SETE_DX10
>> ; SI: V_CMP_U_F32
>> @@ -157,7 +157,7 @@ entry:
>> ret void
>> }
>>
>> -; FUNC-LABEL: @f32_uge
>> +; FUNC-LABEL: f32_uge
>> ; R600: SETGT
>> ; R600: SETE_DX10
>> ; SI: V_CMP_U_F32
>> @@ -173,7 +173,7 @@ entry:
>> ret void
>> }
>>
>> -; FUNC-LABEL: @f32_ult
>> +; FUNC-LABEL: f32_ult
>> ; R600: SETGE
>> ; R600: SETE_DX10
>> ; SI: V_CMP_U_F32
>> @@ -189,7 +189,7 @@ entry:
>> ret void
>> }
>>
>> -; FUNC-LABEL: @f32_ule
>> +; FUNC-LABEL: f32_ule
>> ; R600: SETGT
>> ; R600: SETE_DX10
>> ; SI: V_CMP_U_F32
>> @@ -205,7 +205,7 @@ entry:
>> ret void
>> }
>>
>> -; FUNC-LABEL: @f32_une
>> +; FUNC-LABEL: f32_une
>> ; R600: SETNE_DX10
>> ; SI: V_CMP_NEQ_F32
>> define void @f32_une(i32 addrspace(1)* %out, float %a, float %b) {
>> @@ -216,7 +216,7 @@ entry:
>> ret void
>> }
>>
>> -; FUNC-LABEL: @f32_uno
>> +; FUNC-LABEL: f32_uno
>> ; R600: SETNE_DX10
>> ; R600: SETNE_DX10
>> ; R600: OR_INT
>> @@ -234,7 +234,7 @@ entry:
>> ;; 32-bit integer comparisons
>> ;;;==========================================================================;;;
>>
>> -; FUNC-LABEL: @i32_eq
>> +; FUNC-LABEL: i32_eq
>> ; R600: SETE_INT
>> ; SI: V_CMP_EQ_I32
>> define void @i32_eq(i32 addrspace(1)* %out, i32 %a, i32 %b) {
>> @@ -245,7 +245,7 @@ entry:
>> ret void
>> }
>>
>> -; FUNC-LABEL: @i32_ne
>> +; FUNC-LABEL: i32_ne
>> ; R600: SETNE_INT
>> ; SI: V_CMP_NE_I32
>> define void @i32_ne(i32 addrspace(1)* %out, i32 %a, i32 %b) {
>> @@ -256,7 +256,7 @@ entry:
>> ret void
>> }
>>
>> -; FUNC-LABEL: @i32_ugt
>> +; FUNC-LABEL: i32_ugt
>> ; R600: SETGT_UINT
>> ; SI: V_CMP_GT_U32
>> define void @i32_ugt(i32 addrspace(1)* %out, i32 %a, i32 %b) {
>> @@ -267,7 +267,7 @@ entry:
>> ret void
>> }
>>
>> -; FUNC-LABEL: @i32_uge
>> +; FUNC-LABEL: i32_uge
>> ; R600: SETGE_UINT
>> ; SI: V_CMP_GE_U32
>> define void @i32_uge(i32 addrspace(1)* %out, i32 %a, i32 %b) {
>> @@ -278,7 +278,7 @@ entry:
>> ret void
>> }
>>
>> -; FUNC-LABEL: @i32_ult
>> +; FUNC-LABEL: i32_ult
>> ; R600: SETGT_UINT
>> ; SI: V_CMP_LT_U32
>> define void @i32_ult(i32 addrspace(1)* %out, i32 %a, i32 %b) {
>> @@ -289,7 +289,7 @@ entry:
>> ret void
>> }
>>
>> -; FUNC-LABEL: @i32_ule
>> +; FUNC-LABEL: i32_ule
>> ; R600: SETGE_UINT
>> ; SI: V_CMP_LE_U32
>> define void @i32_ule(i32 addrspace(1)* %out, i32 %a, i32 %b) {
>> @@ -300,7 +300,7 @@ entry:
>> ret void
>> }
>>
>> -; FUNC-LABEL: @i32_sgt
>> +; FUNC-LABEL: i32_sgt
>> ; R600: SETGT_INT
>> ; SI: V_CMP_GT_I32
>> define void @i32_sgt(i32 addrspace(1)* %out, i32 %a, i32 %b) {
>> @@ -311,7 +311,7 @@ entry:
>> ret void
>> }
>>
>> -; FUNC-LABEL: @i32_sge
>> +; FUNC-LABEL: i32_sge
>> ; R600: SETGE_INT
>> ; SI: V_CMP_GE_I32
>> define void @i32_sge(i32 addrspace(1)* %out, i32 %a, i32 %b) {
>> @@ -322,7 +322,7 @@ entry:
>> ret void
>> }
>>
>> -; FUNC-LABEL: @i32_slt
>> +; FUNC-LABEL: i32_slt
>> ; R600: SETGT_INT
>> ; SI: V_CMP_LT_I32
>> define void @i32_slt(i32 addrspace(1)* %out, i32 %a, i32 %b) {
>> @@ -333,7 +333,7 @@ entry:
>> ret void
>> }
>>
>> -; FUNC-LABEL: @i32_sle
>> +; FUNC-LABEL: i32_sle
>> ; R600: SETGE_INT
>> ; SI: V_CMP_LE_I32
>> define void @i32_sle(i32 addrspace(1)* %out, i32 %a, i32 %b) {
>> diff --git a/test/CodeGen/R600/setcc64.ll b/test/CodeGen/R600/setcc64.ll
>> index 54a33b3..e4b43d8 100644
>> --- a/test/CodeGen/R600/setcc64.ll
>> +++ b/test/CodeGen/R600/setcc64.ll
>> @@ -6,7 +6,7 @@
>> ;; Double comparisons
>> ;;;==========================================================================;;;
>>
>> -; FUNC-LABEL: @f64_oeq
>> +; FUNC-LABEL: f64_oeq
>> ; SI: V_CMP_EQ_F64
>> define void @f64_oeq(i32 addrspace(1)* %out, double %a, double %b) {
>> entry:
>> @@ -16,7 +16,7 @@ entry:
>> ret void
>> }
>>
>> -; FUNC-LABEL: @f64_ogt
>> +; FUNC-LABEL: f64_ogt
>> ; SI: V_CMP_GT_F64
>> define void @f64_ogt(i32 addrspace(1)* %out, double %a, double %b) {
>> entry:
>> @@ -26,7 +26,7 @@ entry:
>> ret void
>> }
>>
>> -; FUNC-LABEL: @f64_oge
>> +; FUNC-LABEL: f64_oge
>> ; SI: V_CMP_GE_F64
>> define void @f64_oge(i32 addrspace(1)* %out, double %a, double %b) {
>> entry:
>> @@ -36,7 +36,7 @@ entry:
>> ret void
>> }
>>
>> -; FUNC-LABEL: @f64_olt
>> +; FUNC-LABEL: f64_olt
>> ; SI: V_CMP_LT_F64
>> define void @f64_olt(i32 addrspace(1)* %out, double %a, double %b) {
>> entry:
>> @@ -46,7 +46,7 @@ entry:
>> ret void
>> }
>>
>> -; FUNC-LABEL: @f64_ole
>> +; FUNC-LABEL: f64_ole
>> ; SI: V_CMP_LE_F64
>> define void @f64_ole(i32 addrspace(1)* %out, double %a, double %b) {
>> entry:
>> @@ -56,7 +56,7 @@ entry:
>> ret void
>> }
>>
>> -; FUNC-LABEL: @f64_one
>> +; FUNC-LABEL: f64_one
>> ; SI: V_CMP_O_F64
>> ; SI: V_CMP_NEQ_F64
>> ; SI: V_CNDMASK_B32_e64
>> @@ -70,7 +70,7 @@ entry:
>> ret void
>> }
>>
>> -; FUNC-LABEL: @f64_ord
>> +; FUNC-LABEL: f64_ord
>> ; SI: V_CMP_O_F64
>> define void @f64_ord(i32 addrspace(1)* %out, double %a, double %b) {
>> entry:
>> @@ -80,7 +80,7 @@ entry:
>> ret void
>> }
>>
>> -; FUNC-LABEL: @f64_ueq
>> +; FUNC-LABEL: f64_ueq
>> ; SI: V_CMP_U_F64
>> ; SI: V_CMP_EQ_F64
>> ; SI: V_CNDMASK_B32_e64
>> @@ -94,7 +94,7 @@ entry:
>> ret void
>> }
>>
>> -; FUNC-LABEL: @f64_ugt
>> +; FUNC-LABEL: f64_ugt
>> ; SI: V_CMP_U_F64
>> ; SI: V_CMP_GT_F64
>> ; SI: V_CNDMASK_B32_e64
>> @@ -108,7 +108,7 @@ entry:
>> ret void
>> }
>>
>> -; FUNC-LABEL: @f64_uge
>> +; FUNC-LABEL: f64_uge
>> ; SI: V_CMP_U_F64
>> ; SI: V_CMP_GE_F64
>> ; SI: V_CNDMASK_B32_e64
>> @@ -122,7 +122,7 @@ entry:
>> ret void
>> }
>>
>> -; FUNC-LABEL: @f64_ult
>> +; FUNC-LABEL: f64_ult
>> ; SI: V_CMP_U_F64
>> ; SI: V_CMP_LT_F64
>> ; SI: V_CNDMASK_B32_e64
>> @@ -136,7 +136,7 @@ entry:
>> ret void
>> }
>>
>> -; FUNC-LABEL: @f64_ule
>> +; FUNC-LABEL: f64_ule
>> ; SI: V_CMP_U_F64
>> ; SI: V_CMP_LE_F64
>> ; SI: V_CNDMASK_B32_e64
>> @@ -150,7 +150,7 @@ entry:
>> ret void
>> }
>>
>> -; FUNC-LABEL: @f64_une
>> +; FUNC-LABEL: f64_une
>> ; SI: V_CMP_NEQ_F64
>> define void @f64_une(i32 addrspace(1)* %out, double %a, double %b) {
>> entry:
>> @@ -160,7 +160,7 @@ entry:
>> ret void
>> }
>>
>> -; FUNC-LABEL: @f64_uno
>> +; FUNC-LABEL: f64_uno
>> ; SI: V_CMP_U_F64
>> define void @f64_uno(i32 addrspace(1)* %out, double %a, double %b) {
>> entry:
>> @@ -174,7 +174,7 @@ entry:
>> ;; 64-bit integer comparisons
>> ;;;==========================================================================;;;
>>
>> -; FUNC-LABEL: @i64_eq
>> +; FUNC-LABEL: i64_eq
>> ; SI: V_CMP_EQ_I64
>> define void @i64_eq(i32 addrspace(1)* %out, i64 %a, i64 %b) {
>> entry:
>> @@ -184,7 +184,7 @@ entry:
>> ret void
>> }
>>
>> -; FUNC-LABEL: @i64_ne
>> +; FUNC-LABEL: i64_ne
>> ; SI: V_CMP_NE_I64
>> define void @i64_ne(i32 addrspace(1)* %out, i64 %a, i64 %b) {
>> entry:
>> @@ -194,7 +194,7 @@ entry:
>> ret void
>> }
>>
>> -; FUNC-LABEL: @i64_ugt
>> +; FUNC-LABEL: i64_ugt
>> ; SI: V_CMP_GT_U64
>> define void @i64_ugt(i32 addrspace(1)* %out, i64 %a, i64 %b) {
>> entry:
>> @@ -204,7 +204,7 @@ entry:
>> ret void
>> }
>>
>> -; FUNC-LABEL: @i64_uge
>> +; FUNC-LABEL: i64_uge
>> ; SI: V_CMP_GE_U64
>> define void @i64_uge(i32 addrspace(1)* %out, i64 %a, i64 %b) {
>> entry:
>> @@ -214,7 +214,7 @@ entry:
>> ret void
>> }
>>
>> -; FUNC-LABEL: @i64_ult
>> +; FUNC-LABEL: i64_ult
>> ; SI: V_CMP_LT_U64
>> define void @i64_ult(i32 addrspace(1)* %out, i64 %a, i64 %b) {
>> entry:
>> @@ -224,7 +224,7 @@ entry:
>> ret void
>> }
>>
>> -; FUNC-LABEL: @i64_ule
>> +; FUNC-LABEL: i64_ule
>> ; SI: V_CMP_LE_U64
>> define void @i64_ule(i32 addrspace(1)* %out, i64 %a, i64 %b) {
>> entry:
>> @@ -234,7 +234,7 @@ entry:
>> ret void
>> }
>>
>> -; FUNC-LABEL: @i64_sgt
>> +; FUNC-LABEL: i64_sgt
>> ; SI: V_CMP_GT_I64
>> define void @i64_sgt(i32 addrspace(1)* %out, i64 %a, i64 %b) {
>> entry:
>> @@ -244,7 +244,7 @@ entry:
>> ret void
>> }
>>
>> -; FUNC-LABEL: @i64_sge
>> +; FUNC-LABEL: i64_sge
>> ; SI: V_CMP_GE_I64
>> define void @i64_sge(i32 addrspace(1)* %out, i64 %a, i64 %b) {
>> entry:
>> @@ -254,7 +254,7 @@ entry:
>> ret void
>> }
>>
>> -; FUNC-LABEL: @i64_slt
>> +; FUNC-LABEL: i64_slt
>> ; SI: V_CMP_LT_I64
>> define void @i64_slt(i32 addrspace(1)* %out, i64 %a, i64 %b) {
>> entry:
>> @@ -264,7 +264,7 @@ entry:
>> ret void
>> }
>>
>> -; FUNC-LABEL: @i64_sle
>> +; FUNC-LABEL: i64_sle
>> ; SI: V_CMP_LE_I64
>> define void @i64_sle(i32 addrspace(1)* %out, i64 %a, i64 %b) {
>> entry:
>> diff --git a/test/CodeGen/R600/seto.ll b/test/CodeGen/R600/seto.ll
>> index eb1176f..7293f37 100644
>> --- a/test/CodeGen/R600/seto.ll
>> +++ b/test/CodeGen/R600/seto.ll
>> @@ -1,6 +1,6 @@
>> ; RUN: llc -march=r600 -mcpu=verde -verify-machineinstrs < %s | FileCheck %s
>>
>> -; CHECK-LABEL: @main
>> +; CHECK-LABEL: main
>> ; CHECK: V_CMP_O_F32_e64 [[CMP:s\[[0-9]+:[0-9]+\]]], [[SREG:s[0-9]+]], [[SREG]]
>> ; CHECK-NEXT: V_CNDMASK_B32_e64 {{v[0-9]+}}, 0, 1.0, [[CMP]]
>> define void @main(float %p) {
>> diff --git a/test/CodeGen/R600/setuo.ll b/test/CodeGen/R600/setuo.ll
>> index a78e8e6..1fe6600 100644
>> --- a/test/CodeGen/R600/setuo.ll
>> +++ b/test/CodeGen/R600/setuo.ll
>> @@ -1,6 +1,6 @@
>> ; RUN: llc -march=r600 -mcpu=verde -verify-machineinstrs < %s | FileCheck %s
>>
>> -; CHECK-LABEL: @main
>> +; CHECK-LABEL: main
>> ; CHECK: V_CMP_U_F32_e64 [[CMP:s\[[0-9]+:[0-9]+\]]], [[SREG:s[0-9]+]], [[SREG]]
>> ; CHECK-NEXT: V_CNDMASK_B32_e64 {{v[0-9]+}}, 0, 1.0, [[CMP]]
>> define void @main(float %p) {
>> diff --git a/test/CodeGen/R600/sext-in-reg.ll b/test/CodeGen/R600/sext-in-reg.ll
>> index ba0dbcb..3e216bb 100644
>> --- a/test/CodeGen/R600/sext-in-reg.ll
>> +++ b/test/CodeGen/R600/sext-in-reg.ll
>> @@ -4,7 +4,7 @@
>> declare i32 @llvm.AMDGPU.imax(i32, i32) nounwind readnone
>>
>>
>> -; FUNC-LABEL: @sext_in_reg_i1_i32
>> +; FUNC-LABEL: sext_in_reg_i1_i32
>> ; SI: S_LOAD_DWORD [[ARG:s[0-9]+]],
>> ; SI: S_BFE_I32 [[SEXTRACT:s[0-9]+]], [[ARG]], 0x10000
>> ; SI: V_MOV_B32_e32 [[EXTRACT:v[0-9]+]], [[SEXTRACT]]
>> @@ -20,7 +20,7 @@ define void @sext_in_reg_i1_i32(i32 addrspace(1)* %out, i32 %in) {
>> ret void
>> }
>>
>> -; FUNC-LABEL: @sext_in_reg_i8_to_i32
>> +; FUNC-LABEL: sext_in_reg_i8_to_i32
>> ; SI: S_ADD_I32 [[VAL:s[0-9]+]],
>> ; SI: S_SEXT_I32_I8 [[EXTRACT:s[0-9]+]], [[VAL]]
>> ; SI: V_MOV_B32_e32 [[VEXTRACT:v[0-9]+]], [[EXTRACT]]
>> @@ -38,7 +38,7 @@ define void @sext_in_reg_i8_to_i32(i32 addrspace(1)* %out, i32 %a, i32 %b) nounw
>> ret void
>> }
>>
>> -; FUNC-LABEL: @sext_in_reg_i16_to_i32
>> +; FUNC-LABEL: sext_in_reg_i16_to_i32
>> ; SI: S_ADD_I32 [[VAL:s[0-9]+]],
>> ; SI: S_SEXT_I32_I16 [[EXTRACT:s[0-9]+]], [[VAL]]
>> ; SI: V_MOV_B32_e32 [[VEXTRACT:v[0-9]+]], [[EXTRACT]]
>> @@ -56,7 +56,7 @@ define void @sext_in_reg_i16_to_i32(i32 addrspace(1)* %out, i32 %a, i32 %b) noun
>> ret void
>> }
>>
>> -; FUNC-LABEL: @sext_in_reg_i8_to_v1i32
>> +; FUNC-LABEL: sext_in_reg_i8_to_v1i32
>> ; SI: S_ADD_I32 [[VAL:s[0-9]+]],
>> ; SI: S_SEXT_I32_I8 [[EXTRACT:s[0-9]+]], [[VAL]]
>> ; SI: V_MOV_B32_e32 [[VEXTRACT:v[0-9]+]], [[EXTRACT]]
>> @@ -74,7 +74,7 @@ define void @sext_in_reg_i8_to_v1i32(<1 x i32> addrspace(1)* %out, <1 x i32> %a,
>> ret void
>> }
>>
>> -; FUNC-LABEL: @sext_in_reg_i1_to_i64
>> +; FUNC-LABEL: sext_in_reg_i1_to_i64
>> ; SI: S_MOV_B32 {{s[0-9]+}}, -1
>> ; SI: S_ADD_I32 [[VAL:s[0-9]+]],
>> ; SI: S_BFE_I32 s{{[0-9]+}}, s{{[0-9]+}}, 0x10000
>> @@ -87,7 +87,7 @@ define void @sext_in_reg_i1_to_i64(i64 addrspace(1)* %out, i64 %a, i64 %b) nounw
>> ret void
>> }
>>
>> -; FUNC-LABEL: @sext_in_reg_i8_to_i64
>> +; FUNC-LABEL: sext_in_reg_i8_to_i64
>> ; SI: S_MOV_B32 {{s[0-9]+}}, -1
>> ; SI: S_ADD_I32 [[VAL:s[0-9]+]],
>> ; SI: S_SEXT_I32_I8 [[EXTRACT:s[0-9]+]], [[VAL]]
>> @@ -111,7 +111,7 @@ define void @sext_in_reg_i8_to_i64(i64 addrspace(1)* %out, i64 %a, i64 %b) nounw
>> ret void
>> }
>>
>> -; FUNC-LABEL: @sext_in_reg_i16_to_i64
>> +; FUNC-LABEL: sext_in_reg_i16_to_i64
>> ; SI: S_MOV_B32 {{s[0-9]+}}, -1
>> ; SI: S_ADD_I32 [[VAL:s[0-9]+]],
>> ; SI: S_SEXT_I32_I16 [[EXTRACT:s[0-9]+]], [[VAL]]
>> @@ -135,7 +135,7 @@ define void @sext_in_reg_i16_to_i64(i64 addrspace(1)* %out, i64 %a, i64 %b) noun
>> ret void
>> }
>>
>> -; FUNC-LABEL: @sext_in_reg_i32_to_i64
>> +; FUNC-LABEL: sext_in_reg_i32_to_i64
>> ; SI: S_LOAD_DWORD
>> ; SI: S_LOAD_DWORD
>> ; SI: S_ADD_I32 [[ADD:s[0-9]+]],
>> @@ -161,7 +161,7 @@ define void @sext_in_reg_i32_to_i64(i64 addrspace(1)* %out, i64 %a, i64 %b) noun
>> }
>>
>> ; This is broken on Evergreen for some reason related to the <1 x i64> kernel arguments.
>> -; XFUNC-LABEL: @sext_in_reg_i8_to_v1i64
>> +; XFUNC-LABEL: sext_in_reg_i8_to_v1i64
>> ; XSI: S_BFE_I32 [[EXTRACT:s[0-9]+]], {{s[0-9]+}}, 524288
>> ; XSI: S_ASHR_I32 {{v[0-9]+}}, [[EXTRACT]], 31
>> ; XSI: BUFFER_STORE_DWORD
>> @@ -175,7 +175,7 @@ define void @sext_in_reg_i32_to_i64(i64 addrspace(1)* %out, i64 %a, i64 %b) noun
>> ; ret void
>> ; }
>>
>> -; FUNC-LABEL: @sext_in_reg_i1_in_i32_other_amount
>> +; FUNC-LABEL: sext_in_reg_i1_in_i32_other_amount
>> ; SI-NOT: BFE
>> ; SI: S_LSHL_B32 [[REG:s[0-9]+]], {{s[0-9]+}}, 6
>> ; SI: S_ASHR_I32 {{s[0-9]+}}, [[REG]], 7
>> @@ -194,7 +194,7 @@ define void @sext_in_reg_i1_in_i32_other_amount(i32 addrspace(1)* %out, i32 %a,
>> ret void
>> }
>>
>> -; FUNC-LABEL: @sext_in_reg_v2i1_in_v2i32_other_amount
>> +; FUNC-LABEL: sext_in_reg_v2i1_in_v2i32_other_amount
>> ; SI-DAG: S_LSHL_B32 [[REG0:s[0-9]+]], {{s[0-9]}}, 6
>> ; SI-DAG: S_ASHR_I32 {{s[0-9]+}}, [[REG0]], 7
>> ; SI-DAG: S_LSHL_B32 [[REG1:s[0-9]+]], {{s[0-9]}}, 6
>> @@ -218,7 +218,7 @@ define void @sext_in_reg_v2i1_in_v2i32_other_amount(<2 x i32> addrspace(1)* %out
>> }
>>
>>
>> -; FUNC-LABEL: @sext_in_reg_v2i1_to_v2i32
>> +; FUNC-LABEL: sext_in_reg_v2i1_to_v2i32
>> ; SI: S_BFE_I32 {{s[0-9]+}}, {{s[0-9]+}}, 0x10000
>> ; SI: S_BFE_I32 {{s[0-9]+}}, {{s[0-9]+}}, 0x10000
>> ; SI: BUFFER_STORE_DWORDX2
>> @@ -235,7 +235,7 @@ define void @sext_in_reg_v2i1_to_v2i32(<2 x i32> addrspace(1)* %out, <2 x i32> %
>> ret void
>> }
>>
>> -; FUNC-LABEL: @sext_in_reg_v4i1_to_v4i32
>> +; FUNC-LABEL: sext_in_reg_v4i1_to_v4i32
>> ; SI: S_BFE_I32 {{s[0-9]+}}, {{s[0-9]+}}, 0x10000
>> ; SI: S_BFE_I32 {{s[0-9]+}}, {{s[0-9]+}}, 0x10000
>> ; SI: S_BFE_I32 {{s[0-9]+}}, {{s[0-9]+}}, 0x10000
>> @@ -256,7 +256,7 @@ define void @sext_in_reg_v4i1_to_v4i32(<4 x i32> addrspace(1)* %out, <4 x i32> %
>> ret void
>> }
>>
>> -; FUNC-LABEL: @sext_in_reg_v2i8_to_v2i32
>> +; FUNC-LABEL: sext_in_reg_v2i8_to_v2i32
>> ; SI: S_SEXT_I32_I8 {{s[0-9]+}}, {{s[0-9]+}}
>> ; SI: S_SEXT_I32_I8 {{s[0-9]+}}, {{s[0-9]+}}
>> ; SI: BUFFER_STORE_DWORDX2
>> @@ -273,7 +273,7 @@ define void @sext_in_reg_v2i8_to_v2i32(<2 x i32> addrspace(1)* %out, <2 x i32> %
>> ret void
>> }
>>
>> -; FUNC-LABEL: @sext_in_reg_v4i8_to_v4i32
>> +; FUNC-LABEL: sext_in_reg_v4i8_to_v4i32
>> ; SI: S_SEXT_I32_I8 {{s[0-9]+}}, {{s[0-9]+}}
>> ; SI: S_SEXT_I32_I8 {{s[0-9]+}}, {{s[0-9]+}}
>> ; SI: S_SEXT_I32_I8 {{s[0-9]+}}, {{s[0-9]+}}
>> @@ -294,7 +294,7 @@ define void @sext_in_reg_v4i8_to_v4i32(<4 x i32> addrspace(1)* %out, <4 x i32> %
>> ret void
>> }
>>
>> -; FUNC-LABEL: @sext_in_reg_v2i16_to_v2i32
>> +; FUNC-LABEL: sext_in_reg_v2i16_to_v2i32
>> ; SI: S_SEXT_I32_I16 {{s[0-9]+}}, {{s[0-9]+}}
>> ; SI: S_SEXT_I32_I16 {{s[0-9]+}}, {{s[0-9]+}}
>> ; SI: BUFFER_STORE_DWORDX2
>> @@ -311,7 +311,7 @@ define void @sext_in_reg_v2i16_to_v2i32(<2 x i32> addrspace(1)* %out, <2 x i32>
>> ret void
>> }
>>
>> -; FUNC-LABEL: @testcase
>> +; FUNC-LABEL: testcase
>> define void @testcase(i8 addrspace(1)* %out, i8 %a) nounwind {
>> %and_a_1 = and i8 %a, 1
>> %cmp_eq = icmp eq i8 %and_a_1, 0
>> @@ -323,7 +323,7 @@ define void @testcase(i8 addrspace(1)* %out, i8 %a) nounwind {
>> ret void
>> }
>>
>> -; FUNC-LABEL: @testcase_3
>> +; FUNC-LABEL: testcase_3
>> define void @testcase_3(i8 addrspace(1)* %out, i8 %a) nounwind {
>> %and_a_1 = and i8 %a, 1
>> %cmp_eq = icmp eq i8 %and_a_1, 0
>> @@ -335,7 +335,7 @@ define void @testcase_3(i8 addrspace(1)* %out, i8 %a) nounwind {
>> ret void
>> }
>>
>> -; FUNC-LABEL: @vgpr_sext_in_reg_v4i8_to_v4i32
>> +; FUNC-LABEL: vgpr_sext_in_reg_v4i8_to_v4i32
>> ; SI: V_BFE_I32 [[EXTRACT:v[0-9]+]], {{v[0-9]+}}, 0, 8
>> ; SI: V_BFE_I32 [[EXTRACT:v[0-9]+]], {{v[0-9]+}}, 0, 8
>> ; SI: V_BFE_I32 [[EXTRACT:v[0-9]+]], {{v[0-9]+}}, 0, 8
>> @@ -350,7 +350,7 @@ define void @vgpr_sext_in_reg_v4i8_to_v4i32(<4 x i32> addrspace(1)* %out, <4 x i
>> ret void
>> }
>>
>> -; FUNC-LABEL: @vgpr_sext_in_reg_v4i16_to_v4i32
>> +; FUNC-LABEL: vgpr_sext_in_reg_v4i16_to_v4i32
>> ; SI: V_BFE_I32 [[EXTRACT:v[0-9]+]], {{v[0-9]+}}, 0, 16
>> ; SI: V_BFE_I32 [[EXTRACT:v[0-9]+]], {{v[0-9]+}}, 0, 16
>> define void @vgpr_sext_in_reg_v4i16_to_v4i32(<4 x i32> addrspace(1)* %out, <4 x i32> addrspace(1)* %a, <4 x i32> addrspace(1)* %b) nounwind {
>> @@ -366,7 +366,7 @@ define void @vgpr_sext_in_reg_v4i16_to_v4i32(<4 x i32> addrspace(1)* %out, <4 x
>> ; FIXME: The BFE should really be eliminated. I think it should happen
>> ; when computeKnownBitsForTargetNode is implemented for imax.
>>
>> -; FUNC-LABEL: @sext_in_reg_to_illegal_type
>> +; FUNC-LABEL: sext_in_reg_to_illegal_type
>> ; SI: BUFFER_LOAD_SBYTE
>> ; SI: V_MAX_I32
>> ; SI: V_BFE_I32
>> @@ -383,7 +383,7 @@ define void @sext_in_reg_to_illegal_type(i16 addrspace(1)* nocapture %out, i8 ad
>>
>> declare i32 @llvm.AMDGPU.bfe.i32(i32, i32, i32) nounwind readnone
>>
>> -; FUNC-LABEL: @bfe_0_width
>> +; FUNC-LABEL: bfe_0_width
>> ; SI-NOT: BFE
>> ; SI: S_ENDPGM
>> define void @bfe_0_width(i32 addrspace(1)* %out, i32 addrspace(1)* %ptr) nounwind {
>> @@ -393,7 +393,7 @@ define void @bfe_0_width(i32 addrspace(1)* %out, i32 addrspace(1)* %ptr) nounwin
>> ret void
>> }
>>
>> -; FUNC-LABEL: @bfe_8_bfe_8
>> +; FUNC-LABEL: bfe_8_bfe_8
>> ; SI: V_BFE_I32
>> ; SI-NOT: BFE
>> ; SI: S_ENDPGM
>> @@ -405,7 +405,7 @@ define void @bfe_8_bfe_8(i32 addrspace(1)* %out, i32 addrspace(1)* %ptr) nounwin
>> ret void
>> }
>>
>> -; FUNC-LABEL: @bfe_8_bfe_16
>> +; FUNC-LABEL: bfe_8_bfe_16
>> ; SI: V_BFE_I32 v{{[0-9]+}}, v{{[0-9]+}}, 0, 8
>> ; SI: S_ENDPGM
>> define void @bfe_8_bfe_16(i32 addrspace(1)* %out, i32 addrspace(1)* %ptr) nounwind {
>> @@ -417,7 +417,7 @@ define void @bfe_8_bfe_16(i32 addrspace(1)* %out, i32 addrspace(1)* %ptr) nounwi
>> }
>>
>> ; This really should be folded into 1
>> -; FUNC-LABEL: @bfe_16_bfe_8
>> +; FUNC-LABEL: bfe_16_bfe_8
>> ; SI: V_BFE_I32 v{{[0-9]+}}, v{{[0-9]+}}, 0, 8
>> ; SI-NOT: BFE
>> ; SI: S_ENDPGM
>> @@ -430,7 +430,7 @@ define void @bfe_16_bfe_8(i32 addrspace(1)* %out, i32 addrspace(1)* %ptr) nounwi
>> }
>>
>> ; Make sure there isn't a redundant BFE
>> -; FUNC-LABEL: @sext_in_reg_i8_to_i32_bfe
>> +; FUNC-LABEL: sext_in_reg_i8_to_i32_bfe
>> ; SI: S_SEXT_I32_I8 s{{[0-9]+}}, s{{[0-9]+}}
>> ; SI-NOT: BFE
>> ; SI: S_ENDPGM
>> @@ -443,7 +443,7 @@ define void @sext_in_reg_i8_to_i32_bfe(i32 addrspace(1)* %out, i32 %a, i32 %b) n
>> ret void
>> }
>>
>> -; FUNC-LABEL: @sext_in_reg_i8_to_i32_bfe_wrong
>> +; FUNC-LABEL: sext_in_reg_i8_to_i32_bfe_wrong
>> define void @sext_in_reg_i8_to_i32_bfe_wrong(i32 addrspace(1)* %out, i32 %a, i32 %b) nounwind {
>> %c = add i32 %a, %b ; add to prevent folding into extload
>> %bfe = call i32 @llvm.AMDGPU.bfe.i32(i32 %c, i32 8, i32 0) nounwind readnone
>> @@ -453,7 +453,7 @@ define void @sext_in_reg_i8_to_i32_bfe_wrong(i32 addrspace(1)* %out, i32 %a, i32
>> ret void
>> }
>>
>> -; FUNC-LABEL: @sextload_i8_to_i32_bfe
>> +; FUNC-LABEL: sextload_i8_to_i32_bfe
>> ; SI: BUFFER_LOAD_SBYTE
>> ; SI-NOT: BFE
>> ; SI: S_ENDPGM
>> @@ -467,7 +467,7 @@ define void @sextload_i8_to_i32_bfe(i32 addrspace(1)* %out, i8 addrspace(1)* %pt
>> ret void
>> }
>>
>> -; FUNC-LABEL: @sextload_i8_to_i32_bfe_0:
>> +; FUNC-LABEL: sextload_i8_to_i32_bfe_0:
>> ; SI-NOT: BFE
>> ; SI: S_ENDPGM
>> define void @sextload_i8_to_i32_bfe_0(i32 addrspace(1)* %out, i8 addrspace(1)* %ptr) nounwind {
>> @@ -480,7 +480,7 @@ define void @sextload_i8_to_i32_bfe_0(i32 addrspace(1)* %out, i8 addrspace(1)* %
>> ret void
>> }
>>
>> -; FUNC-LABEL: @sext_in_reg_i1_bfe_offset_0:
>> +; FUNC-LABEL: sext_in_reg_i1_bfe_offset_0:
>> ; SI-NOT: SHR
>> ; SI-NOT: SHL
>> ; SI: V_BFE_I32 v{{[0-9]+}}, v{{[0-9]+}}, 0, 1
>> @@ -494,7 +494,7 @@ define void @sext_in_reg_i1_bfe_offset_0(i32 addrspace(1)* %out, i32 addrspace(1
>> ret void
>> }
>>
>> -; FUNC-LABEL: @sext_in_reg_i1_bfe_offset_1
>> +; FUNC-LABEL: sext_in_reg_i1_bfe_offset_1
>> ; SI: BUFFER_LOAD_DWORD
>> ; SI-NOT: SHL
>> ; SI-NOT: SHR
>> @@ -509,7 +509,7 @@ define void @sext_in_reg_i1_bfe_offset_1(i32 addrspace(1)* %out, i32 addrspace(1
>> ret void
>> }
>>
>> -; FUNC-LABEL: @sext_in_reg_i2_bfe_offset_1:
>> +; FUNC-LABEL: sext_in_reg_i2_bfe_offset_1:
>> ; SI: BUFFER_LOAD_DWORD
>> ; SI: V_LSHLREV_B32_e32 v{{[0-9]+}}, 30, v{{[0-9]+}}
>> ; SI: V_ASHRREV_I32_e32 v{{[0-9]+}}, 30, v{{[0-9]+}}
>> diff --git a/test/CodeGen/R600/sgpr-control-flow.ll b/test/CodeGen/R600/sgpr-control-flow.ll
>> index 326b37a..76298ff 100644
>> --- a/test/CodeGen/R600/sgpr-control-flow.ll
>> +++ b/test/CodeGen/R600/sgpr-control-flow.ll
>> @@ -7,7 +7,7 @@
>> ; If the branch decision is made based on a value in an SGPR then all
>> ; threads will execute the same code paths, so we don't need to worry
>> ; about instructions in different blocks overwriting each other.
>> -; SI-LABEL: @sgpr_if_else_salu_br
>> +; SI-LABEL: sgpr_if_else_salu_br
>> ; SI: S_ADD
>> ; SI: S_ADD
>>
>> @@ -34,7 +34,7 @@ endif:
>> ; The two S_ADD instructions should write to different registers, since
>> ; different threads will take different control flow paths.
>>
>> -; SI-LABEL: @sgpr_if_else_valu_br
>> +; SI-LABEL: sgpr_if_else_valu_br
>> ; SI: S_ADD_I32 [[SGPR:s[0-9]+]]
>> ; SI-NOT: S_ADD_I32 [[SGPR]]
>>
>> diff --git a/test/CodeGen/R600/sgpr-copy-duplicate-operand.ll b/test/CodeGen/R600/sgpr-copy-duplicate-operand.ll
>> index 9d8a623..bc21910 100644
>> --- a/test/CodeGen/R600/sgpr-copy-duplicate-operand.ll
>> +++ b/test/CodeGen/R600/sgpr-copy-duplicate-operand.ll
>> @@ -3,7 +3,7 @@
>> ; Copy VGPR -> SGPR used twice as an instruction operand, which is then
>> ; used in an REG_SEQUENCE that also needs to be handled.
>>
>> -; SI-LABEL: @test_dup_operands:
>> +; SI-LABEL: test_dup_operands:
>> ; SI: V_ADD_I32_e32
>> define void @test_dup_operands(<2 x i32> addrspace(1)* noalias %out, <2 x i32> addrspace(1)* noalias %in) {
>> %a = load <2 x i32> addrspace(1)* %in
>> diff --git a/test/CodeGen/R600/sgpr-copy.ll b/test/CodeGen/R600/sgpr-copy.ll
>> index c7d5bf9..d3a1709 100644
>> --- a/test/CodeGen/R600/sgpr-copy.ll
>> +++ b/test/CodeGen/R600/sgpr-copy.ll
>> @@ -2,7 +2,7 @@
>>
>> ; This test checks that no VGPR to SGPR copies are created by the register
>> ; allocator.
>> -; CHECK-LABEL: @phi1
>> +; CHECK-LABEL: phi1
>> ; CHECK: S_BUFFER_LOAD_DWORD [[DST:s[0-9]]], {{s\[[0-9]+:[0-9]+\]}}, 0
>> ; CHECK: V_MOV_B32_e32 v{{[0-9]}}, [[DST]]
>>
>> @@ -29,7 +29,7 @@ ENDIF: ; preds = %main_body, %ELSE
>> }
>>
>> ; Make sure this program doesn't crash
>> -; CHECK-LABEL: @phi2
>> +; CHECK-LABEL: phi2
>> define void @phi2(<16 x i8> addrspace(2)* inreg, <16 x i8> addrspace(2)* inreg, <32 x i8> addrspace(2)* inreg, i32 inreg, <2 x i32>, <2 x i32>, <2 x i32>, <3 x i32>, <2 x i32>, <2 x i32>, <2 x i32>, float, float, float, float, float, float, float, float, float) #0 {
>> main_body:
>> %20 = getelementptr <16 x i8> addrspace(2)* %0, i32 0
>> @@ -149,7 +149,7 @@ ENDIF24: ; preds = %ENDIF, %IF25
>> }
>>
>> ; We just want ot make sure the program doesn't crash
>> -; CHECK-LABEL: @loop
>> +; CHECK-LABEL: loop
>>
>> define void @loop(<16 x i8> addrspace(2)* inreg, <16 x i8> addrspace(2)* inreg, <32 x i8> addrspace(2)* inreg, i32 inreg, <2 x i32>, <2 x i32>, <2 x i32>, <3 x i32>, <2 x i32>, <2 x i32>, <2 x i32>, float, float, float, float, float, float, float, float, float) #0 {
>> main_body:
>> @@ -227,7 +227,7 @@ declare i32 @llvm.SI.packf16(float, float) #1
>> ; registers were being identified as an SGPR regclass which was causing
>> ; an assertion failure.
>>
>> -; CHECK-LABEL: @sample_v3
>> +; CHECK-LABEL: sample_v3
>> ; CHECK: IMAGE_SAMPLE
>> ; CHECK: IMAGE_SAMPLE
>> ; CHECK: EXP
>> @@ -269,7 +269,7 @@ endif:
>>
>> !2 = metadata !{metadata !"const", null, i32 1}
>>
>> -; CHECK-LABEL: @copy1
>> +; CHECK-LABEL: copy1
>> ; CHECK: BUFFER_LOAD_DWORD
>> ; CHECK: V_ADD
>> ; CHECK: S_ENDPGM
>> @@ -296,7 +296,7 @@ endif:
>> }
>>
>> ; This test is just checking that we don't crash / assertion fail.
>> -; CHECK-LABEL: @copy2
>> +; CHECK-LABEL: copy2
>> ; CHECK: S_ENDPGM
>>
>> define void @copy2([17 x <16 x i8>] addrspace(2)* byval, [32 x <16 x i8>] addrspace(2)* byval, [16 x <32 x i8>] addrspace(2)* byval, float inreg, i32 inreg, <2 x i32>, <2 x i32>, <2 x i32>, <3 x i32>, <2 x i32>, <2 x i32>, <2 x i32>, float, float, float, float, float, float, float, float, float) #0 {
>> diff --git a/test/CodeGen/R600/shared-op-cycle.ll b/test/CodeGen/R600/shared-op-cycle.ll
>> index 0484fc9..b909f93 100644
>> --- a/test/CodeGen/R600/shared-op-cycle.ll
>> +++ b/test/CodeGen/R600/shared-op-cycle.ll
>> @@ -1,6 +1,6 @@
>> ; RUN: llc < %s -march=r600 -mcpu=redwood | FileCheck %s
>>
>> -; CHECK: @main
>> +; CHECK: main
>> ; CHECK: MULADD_IEEE *
>> ; CHECK-NOT: MULADD_IEEE *
>>
>> diff --git a/test/CodeGen/R600/shl.ll b/test/CodeGen/R600/shl.ll
>> index 43fab2a..240a4cf 100644
>> --- a/test/CodeGen/R600/shl.ll
>> +++ b/test/CodeGen/R600/shl.ll
>> @@ -1,11 +1,11 @@
>> ;RUN: llc < %s -march=r600 -mcpu=redwood | FileCheck --check-prefix=EG-CHECK %s
>> ;RUN: llc < %s -march=r600 -mcpu=verde -verify-machineinstrs | FileCheck --check-prefix=SI-CHECK %s
>>
>> -;EG-CHECK: @shl_v2i32
>> +;EG-CHECK: shl_v2i32
>> ;EG-CHECK: LSHL {{\*? *}}T{{[0-9]+\.[XYZW], T[0-9]+\.[XYZW], T[0-9]+\.[XYZW]}}
>> ;EG-CHECK: LSHL {{\*? *}}T{{[0-9]+\.[XYZW], T[0-9]+\.[XYZW], T[0-9]+\.[XYZW]}}
>>
>> -;SI-CHECK: @shl_v2i32
>> +;SI-CHECK: shl_v2i32
>> ;SI-CHECK: V_LSHL_B32_e32 v{{[0-9]+, v[0-9]+, v[0-9]+}}
>> ;SI-CHECK: V_LSHL_B32_e32 v{{[0-9]+, v[0-9]+, v[0-9]+}}
>>
>> @@ -18,13 +18,13 @@ define void @shl_v2i32(<2 x i32> addrspace(1)* %out, <2 x i32> addrspace(1)* %in
>> ret void
>> }
>>
>> -;EG-CHECK: @shl_v4i32
>> +;EG-CHECK: shl_v4i32
>> ;EG-CHECK: LSHL {{\*? *}}T{{[0-9]+\.[XYZW], T[0-9]+\.[XYZW], T[0-9]+\.[XYZW]}}
>> ;EG-CHECK: LSHL {{\*? *}}T{{[0-9]+\.[XYZW], T[0-9]+\.[XYZW], T[0-9]+\.[XYZW]}}
>> ;EG-CHECK: LSHL {{\*? *}}T{{[0-9]+\.[XYZW], T[0-9]+\.[XYZW], T[0-9]+\.[XYZW]}}
>> ;EG-CHECK: LSHL {{\*? *}}T{{[0-9]+\.[XYZW], T[0-9]+\.[XYZW], T[0-9]+\.[XYZW]}}
>>
>> -;SI-CHECK: @shl_v4i32
>> +;SI-CHECK: shl_v4i32
>> ;SI-CHECK: V_LSHL_B32_e32 v{{[0-9]+, v[0-9]+, v[0-9]+}}
>> ;SI-CHECK: V_LSHL_B32_e32 v{{[0-9]+, v[0-9]+, v[0-9]+}}
>> ;SI-CHECK: V_LSHL_B32_e32 v{{[0-9]+, v[0-9]+, v[0-9]+}}
>> @@ -39,7 +39,7 @@ define void @shl_v4i32(<4 x i32> addrspace(1)* %out, <4 x i32> addrspace(1)* %in
>> ret void
>> }
>>
>> -;EG-CHECK: @shl_i64
>> +;EG-CHECK: shl_i64
>> ;EG-CHECK: SUB_INT {{\*? *}}[[COMPSH:T[0-9]+\.[XYZW]]], {{literal.[xy]}}, [[SHIFT:T[0-9]+\.[XYZW]]]
>> ;EG-CHECK: LSHR {{\* *}}[[TEMP:T[0-9]+\.[XYZW]]], [[OPLO:T[0-9]+\.[XYZW]]], {{[[COMPSH]]|PV.[XYZW]}}
>> ;EG-CHECK: LSHR {{\*? *}}[[OVERF:T[0-9]+\.[XYZW]]], {{[[TEMP]]|PV.[XYZW]}}, 1
>> @@ -51,7 +51,7 @@ define void @shl_v4i32(<4 x i32> addrspace(1)* %out, <4 x i32> addrspace(1)* %in
>> ;EG-CHECK-DAG: CNDE_INT {{\*? *}}[[RESLO:T[0-9]+\.[XYZW]]], {{T[0-9]+\.[XYZW]}}
>> ;EG-CHECK-DAG: CNDE_INT {{\*? *}}[[RESHI:T[0-9]+\.[XYZW]]], {{T[0-9]+\.[XYZW], .*}}, 0.0
>>
>> -;SI-CHECK: @shl_i64
>> +;SI-CHECK: shl_i64
>> ;SI-CHECK: V_LSHL_B64 {{v\[[0-9]+:[0-9]+\], v\[[0-9]+:[0-9]+\], v[0-9]+}}
>>
>> define void @shl_i64(i64 addrspace(1)* %out, i64 addrspace(1)* %in) {
>> @@ -63,7 +63,7 @@ define void @shl_i64(i64 addrspace(1)* %out, i64 addrspace(1)* %in) {
>> ret void
>> }
>>
>> -;EG-CHECK: @shl_v2i64
>> +;EG-CHECK: shl_v2i64
>> ;EG-CHECK-DAG: SUB_INT {{\*? *}}[[COMPSHA:T[0-9]+\.[XYZW]]], {{literal.[xy]}}, [[SHA:T[0-9]+\.[XYZW]]]
>> ;EG-CHECK-DAG: SUB_INT {{\*? *}}[[COMPSHB:T[0-9]+\.[XYZW]]], {{literal.[xy]}}, [[SHB:T[0-9]+\.[XYZW]]]
>> ;EG-CHECK-DAG: LSHR {{\*? *}}[[COMPSHA]]
>> @@ -85,7 +85,7 @@ define void @shl_i64(i64 addrspace(1)* %out, i64 addrspace(1)* %in) {
>> ;EG-CHECK-DAG: CNDE_INT
>> ;EG-CHECK-DAG: CNDE_INT
>>
>> -;SI-CHECK: @shl_v2i64
>> +;SI-CHECK: shl_v2i64
>> ;SI-CHECK: V_LSHL_B64 {{v\[[0-9]+:[0-9]+\], v\[[0-9]+:[0-9]+\], v[0-9]+}}
>> ;SI-CHECK: V_LSHL_B64 {{v\[[0-9]+:[0-9]+\], v\[[0-9]+:[0-9]+\], v[0-9]+}}
>>
>> @@ -98,7 +98,7 @@ define void @shl_v2i64(<2 x i64> addrspace(1)* %out, <2 x i64> addrspace(1)* %in
>> ret void
>> }
>>
>> -;EG-CHECK: @shl_v4i64
>> +;EG-CHECK: shl_v4i64
>> ;EG-CHECK-DAG: SUB_INT {{\*? *}}[[COMPSHA:T[0-9]+\.[XYZW]]], {{literal.[xy]}}, [[SHA:T[0-9]+\.[XYZW]]]
>> ;EG-CHECK-DAG: SUB_INT {{\*? *}}[[COMPSHB:T[0-9]+\.[XYZW]]], {{literal.[xy]}}, [[SHB:T[0-9]+\.[XYZW]]]
>> ;EG-CHECK-DAG: SUB_INT {{\*? *}}[[COMPSHC:T[0-9]+\.[XYZW]]], {{literal.[xy]}}, [[SHC:T[0-9]+\.[XYZW]]]
>> @@ -140,7 +140,7 @@ define void @shl_v2i64(<2 x i64> addrspace(1)* %out, <2 x i64> addrspace(1)* %in
>> ;EG-CHECK-DAG: CNDE_INT
>> ;EG-CHECK-DAG: CNDE_INT
>>
>> -;SI-CHECK: @shl_v4i64
>> +;SI-CHECK: shl_v4i64
>> ;SI-CHECK: V_LSHL_B64 {{v\[[0-9]+:[0-9]+\], v\[[0-9]+:[0-9]+\], v[0-9]+}}
>> ;SI-CHECK: V_LSHL_B64 {{v\[[0-9]+:[0-9]+\], v\[[0-9]+:[0-9]+\], v[0-9]+}}
>> ;SI-CHECK: V_LSHL_B64 {{v\[[0-9]+:[0-9]+\], v\[[0-9]+:[0-9]+\], v[0-9]+}}
>> diff --git a/test/CodeGen/R600/shl_add_constant.ll b/test/CodeGen/R600/shl_add_constant.ll
>> index 60f35d7..8e2aba1 100644
>> --- a/test/CodeGen/R600/shl_add_constant.ll
>> +++ b/test/CodeGen/R600/shl_add_constant.ll
>> @@ -4,7 +4,7 @@ declare i32 @llvm.r600.read.tidig.x() #1
>>
>> ; Test with inline immediate
>>
>> -; FUNC-LABEL: @shl_2_add_9_i32
>> +; FUNC-LABEL: shl_2_add_9_i32
>> ; SI: V_LSHLREV_B32_e32 [[REG:v[0-9]+]], 2, {{v[0-9]+}}
>> ; SI: V_ADD_I32_e32 [[RESULT:v[0-9]+]], 36, [[REG]]
>> ; SI: BUFFER_STORE_DWORD [[RESULT]]
>> @@ -19,7 +19,7 @@ define void @shl_2_add_9_i32(i32 addrspace(1)* %out, i32 addrspace(1)* %in) #0 {
>> ret void
>> }
>>
>> -; FUNC-LABEL: @shl_2_add_9_i32_2_add_uses
>> +; FUNC-LABEL: shl_2_add_9_i32_2_add_uses
>> ; SI-DAG: V_ADD_I32_e32 [[ADDREG:v[0-9]+]], 9, {{v[0-9]+}}
>> ; SI-DAG: V_LSHLREV_B32_e32 [[SHLREG:v[0-9]+]], 2, {{v[0-9]+}}
>> ; SI-DAG: BUFFER_STORE_DWORD [[ADDREG]]
>> @@ -38,7 +38,7 @@ define void @shl_2_add_9_i32_2_add_uses(i32 addrspace(1)* %out0, i32 addrspace(1
>>
>> ; Test with add literal constant
>>
>> -; FUNC-LABEL: @shl_2_add_999_i32
>> +; FUNC-LABEL: shl_2_add_999_i32
>> ; SI: V_LSHLREV_B32_e32 [[REG:v[0-9]+]], 2, {{v[0-9]+}}
>> ; SI: V_ADD_I32_e32 [[RESULT:v[0-9]+]], 0xf9c, [[REG]]
>> ; SI: BUFFER_STORE_DWORD [[RESULT]]
>> @@ -53,7 +53,7 @@ define void @shl_2_add_999_i32(i32 addrspace(1)* %out, i32 addrspace(1)* %in) #0
>> ret void
>> }
>>
>> -; FUNC-LABEL: @test_add_shl_add_constant
>> +; FUNC-LABEL: test_add_shl_add_constant
>> ; SI-DAG: S_LOAD_DWORD [[X:s[0-9]+]], s{{\[[0-9]+:[0-9]+\]}}, 0xb
>> ; SI-DAG: S_LOAD_DWORD [[Y:s[0-9]+]], s{{\[[0-9]+:[0-9]+\]}}, 0xc
>> ; SI: S_LSHL_B32 [[SHL3:s[0-9]+]], [[X]], 3
>> @@ -69,7 +69,7 @@ define void @test_add_shl_add_constant(i32 addrspace(1)* %out, i32 %x, i32 %y) #
>> ret void
>> }
>>
>> -; FUNC-LABEL: @test_add_shl_add_constant_inv
>> +; FUNC-LABEL: test_add_shl_add_constant_inv
>> ; SI-DAG: S_LOAD_DWORD [[X:s[0-9]+]], s{{\[[0-9]+:[0-9]+\]}}, 0xb
>> ; SI-DAG: S_LOAD_DWORD [[Y:s[0-9]+]], s{{\[[0-9]+:[0-9]+\]}}, 0xc
>> ; SI: S_LSHL_B32 [[SHL3:s[0-9]+]], [[X]], 3
>> diff --git a/test/CodeGen/R600/shl_add_ptr.ll b/test/CodeGen/R600/shl_add_ptr.ll
>> index ecab1c8..fec2dc4 100644
>> --- a/test/CodeGen/R600/shl_add_ptr.ll
>> +++ b/test/CodeGen/R600/shl_add_ptr.ll
>> @@ -14,7 +14,7 @@ declare i32 @llvm.r600.read.tidig.x() #1
>>
>> ; Make sure the (add tid, 2) << 2 gets folded into the ds's offset as (tid << 2) + 8
>>
>> -; SI-LABEL: @load_shl_base_lds_0
>> +; SI-LABEL: load_shl_base_lds_0
>> ; SI: V_LSHLREV_B32_e32 [[PTR:v[0-9]+]], 2, {{v[0-9]+}}
>> ; SI: DS_READ_B32 {{v[0-9]+}}, [[PTR]], 0x8, [M0]
>> ; SI: S_ENDPGM
>> @@ -31,7 +31,7 @@ define void @load_shl_base_lds_0(float addrspace(1)* %out, i32 addrspace(1)* %ad
>> ; Make sure once the first use is folded into the addressing mode, the
>> ; remaining add use goes through the normal shl + add constant fold.
>>
>> -; SI-LABEL: @load_shl_base_lds_1
>> +; SI-LABEL: load_shl_base_lds_1
>> ; SI: V_LSHLREV_B32_e32 [[PTR:v[0-9]+]], 2, {{v[0-9]+}}
>> ; SI: DS_READ_B32 [[RESULT:v[0-9]+]], [[PTR]], 0x8, [M0]
>> ; SI: V_ADD_I32_e32 [[ADDUSE:v[0-9]+]], 8, v{{[0-9]+}}
>> @@ -51,7 +51,7 @@ define void @load_shl_base_lds_1(float addrspace(1)* %out, i32 addrspace(1)* %ad
>>
>> @maxlds = addrspace(3) global [65536 x i8] zeroinitializer, align 4
>>
>> -; SI-LABEL: @load_shl_base_lds_max_offset
>> +; SI-LABEL: load_shl_base_lds_max_offset
>> ; SI: DS_READ_U8 v{{[0-9]+}}, v{{[0-9]+}}, 0xffff
>> ; SI: S_ENDPGM
>> define void @load_shl_base_lds_max_offset(i8 addrspace(1)* %out, i8 addrspace(3)* %lds, i32 addrspace(1)* %add_use) #0 {
>> @@ -67,7 +67,7 @@ define void @load_shl_base_lds_max_offset(i8 addrspace(1)* %out, i8 addrspace(3)
>> ; The two globals are placed adjacent in memory, so the same base
>> ; pointer can be used with an offset into the second one.
>>
>> -; SI-LABEL: @load_shl_base_lds_2
>> +; SI-LABEL: load_shl_base_lds_2
>> ; SI: V_LSHLREV_B32_e32 [[PTR:v[0-9]+]], 2, {{v[0-9]+}}
>> ; SI-NEXT: DS_READ_B32 {{v[0-9]+}}, [[PTR]], 0x100, [M0]
>> ; SI-NEXT: DS_READ_B32 {{v[0-9]+}}, [[PTR]], 0x900, [M0]
>> @@ -84,7 +84,7 @@ define void @load_shl_base_lds_2(float addrspace(1)* %out) #0 {
>> ret void
>> }
>>
>> -; SI-LABEL: @store_shl_base_lds_0
>> +; SI-LABEL: store_shl_base_lds_0
>> ; SI: V_LSHLREV_B32_e32 [[PTR:v[0-9]+]], 2, {{v[0-9]+}}
>> ; SI: DS_WRITE_B32 [[PTR]], {{v[0-9]+}}, 0x8 [M0]
>> ; SI: S_ENDPGM
>> @@ -114,7 +114,7 @@ define void @store_shl_base_lds_0(float addrspace(1)* %out, i32 addrspace(1)* %a
>> ; }
>>
>>
>> -; SI-LABEL: @atomic_cmpxchg_shl_base_lds_0
>> +; SI-LABEL: atomic_cmpxchg_shl_base_lds_0
>> ; SI: V_LSHLREV_B32_e32 [[PTR:v[0-9]+]], 2, {{v[0-9]+}}
>> ; SI: DS_CMPST_RTN_B32 {{v[0-9]+}}, [[PTR]], {{v[0-9]+}}, {{v[0-9]+}}, 0x8
>> ; SI: S_ENDPGM
>> @@ -129,7 +129,7 @@ define void @atomic_cmpxchg_shl_base_lds_0(i32 addrspace(1)* %out, i32 addrspace
>> ret void
>> }
>>
>> -; SI-LABEL: @atomic_swap_shl_base_lds_0
>> +; SI-LABEL: atomic_swap_shl_base_lds_0
>> ; SI: V_LSHLREV_B32_e32 [[PTR:v[0-9]+]], 2, {{v[0-9]+}}
>> ; SI: DS_WRXCHG_RTN_B32 {{v[0-9]+}}, [[PTR]], {{v[0-9]+}}, 0x8
>> ; SI: S_ENDPGM
>> @@ -143,7 +143,7 @@ define void @atomic_swap_shl_base_lds_0(i32 addrspace(1)* %out, i32 addrspace(1)
>> ret void
>> }
>>
>> -; SI-LABEL: @atomic_add_shl_base_lds_0
>> +; SI-LABEL: atomic_add_shl_base_lds_0
>> ; SI: V_LSHLREV_B32_e32 [[PTR:v[0-9]+]], 2, {{v[0-9]+}}
>> ; SI: DS_ADD_RTN_U32 {{v[0-9]+}}, [[PTR]], {{v[0-9]+}}, 0x8
>> ; SI: S_ENDPGM
>> @@ -157,7 +157,7 @@ define void @atomic_add_shl_base_lds_0(i32 addrspace(1)* %out, i32 addrspace(1)*
>> ret void
>> }
>>
>> -; SI-LABEL: @atomic_sub_shl_base_lds_0
>> +; SI-LABEL: atomic_sub_shl_base_lds_0
>> ; SI: V_LSHLREV_B32_e32 [[PTR:v[0-9]+]], 2, {{v[0-9]+}}
>> ; SI: DS_SUB_RTN_U32 {{v[0-9]+}}, [[PTR]], {{v[0-9]+}}, 0x8
>> ; SI: S_ENDPGM
>> @@ -171,7 +171,7 @@ define void @atomic_sub_shl_base_lds_0(i32 addrspace(1)* %out, i32 addrspace(1)*
>> ret void
>> }
>>
>> -; SI-LABEL: @atomic_and_shl_base_lds_0
>> +; SI-LABEL: atomic_and_shl_base_lds_0
>> ; SI: V_LSHLREV_B32_e32 [[PTR:v[0-9]+]], 2, {{v[0-9]+}}
>> ; SI: DS_AND_RTN_B32 {{v[0-9]+}}, [[PTR]], {{v[0-9]+}}, 0x8
>> ; SI: S_ENDPGM
>> @@ -185,7 +185,7 @@ define void @atomic_and_shl_base_lds_0(i32 addrspace(1)* %out, i32 addrspace(1)*
>> ret void
>> }
>>
>> -; SI-LABEL: @atomic_or_shl_base_lds_0
>> +; SI-LABEL: atomic_or_shl_base_lds_0
>> ; SI: V_LSHLREV_B32_e32 [[PTR:v[0-9]+]], 2, {{v[0-9]+}}
>> ; SI: DS_OR_RTN_B32 {{v[0-9]+}}, [[PTR]], {{v[0-9]+}}, 0x8
>> ; SI: S_ENDPGM
>> @@ -199,7 +199,7 @@ define void @atomic_or_shl_base_lds_0(i32 addrspace(1)* %out, i32 addrspace(1)*
>> ret void
>> }
>>
>> -; SI-LABEL: @atomic_xor_shl_base_lds_0
>> +; SI-LABEL: atomic_xor_shl_base_lds_0
>> ; SI: V_LSHLREV_B32_e32 [[PTR:v[0-9]+]], 2, {{v[0-9]+}}
>> ; SI: DS_XOR_RTN_B32 {{v[0-9]+}}, [[PTR]], {{v[0-9]+}}, 0x8
>> ; SI: S_ENDPGM
>> @@ -223,7 +223,7 @@ define void @atomic_xor_shl_base_lds_0(i32 addrspace(1)* %out, i32 addrspace(1)*
>> ; ret void
>> ; }
>>
>> -; SI-LABEL: @atomic_min_shl_base_lds_0
>> +; SI-LABEL: atomic_min_shl_base_lds_0
>> ; SI: V_LSHLREV_B32_e32 [[PTR:v[0-9]+]], 2, {{v[0-9]+}}
>> ; SI: DS_MIN_RTN_I32 {{v[0-9]+}}, [[PTR]], {{v[0-9]+}}, 0x8
>> ; SI: S_ENDPGM
>> @@ -237,7 +237,7 @@ define void @atomic_min_shl_base_lds_0(i32 addrspace(1)* %out, i32 addrspace(1)*
>> ret void
>> }
>>
>> -; SI-LABEL: @atomic_max_shl_base_lds_0
>> +; SI-LABEL: atomic_max_shl_base_lds_0
>> ; SI: V_LSHLREV_B32_e32 [[PTR:v[0-9]+]], 2, {{v[0-9]+}}
>> ; SI: DS_MAX_RTN_I32 {{v[0-9]+}}, [[PTR]], {{v[0-9]+}}, 0x8
>> ; SI: S_ENDPGM
>> @@ -251,7 +251,7 @@ define void @atomic_max_shl_base_lds_0(i32 addrspace(1)* %out, i32 addrspace(1)*
>> ret void
>> }
>>
>> -; SI-LABEL: @atomic_umin_shl_base_lds_0
>> +; SI-LABEL: atomic_umin_shl_base_lds_0
>> ; SI: V_LSHLREV_B32_e32 [[PTR:v[0-9]+]], 2, {{v[0-9]+}}
>> ; SI: DS_MIN_RTN_U32 {{v[0-9]+}}, [[PTR]], {{v[0-9]+}}, 0x8
>> ; SI: S_ENDPGM
>> @@ -265,7 +265,7 @@ define void @atomic_umin_shl_base_lds_0(i32 addrspace(1)* %out, i32 addrspace(1)
>> ret void
>> }
>>
>> -; SI-LABEL: @atomic_umax_shl_base_lds_0
>> +; SI-LABEL: atomic_umax_shl_base_lds_0
>> ; SI: V_LSHLREV_B32_e32 [[PTR:v[0-9]+]], 2, {{v[0-9]+}}
>> ; SI: DS_MAX_RTN_U32 {{v[0-9]+}}, [[PTR]], {{v[0-9]+}}, 0x8
>> ; SI: S_ENDPGM
>> diff --git a/test/CodeGen/R600/si-annotate-cf-assertion.ll b/test/CodeGen/R600/si-annotate-cf-assertion.ll
>> index daa4667..7027bf4 100644
>> --- a/test/CodeGen/R600/si-annotate-cf-assertion.ll
>> +++ b/test/CodeGen/R600/si-annotate-cf-assertion.ll
>> @@ -4,7 +4,7 @@
>>
>>
>> define void @test(i32 addrspace(1)* %g, i8 addrspace(3)* %l, i32 %x) nounwind {
>> -; CHECK-LABEL: @test:
>> +; CHECK-LABEL: test:
>>
>> entry:
>> switch i32 %x, label %sw.default [
>> diff --git a/test/CodeGen/R600/si-lod-bias.ll b/test/CodeGen/R600/si-lod-bias.ll
>> index 8d7a79c..eb58301 100644
>> --- a/test/CodeGen/R600/si-lod-bias.ll
>> +++ b/test/CodeGen/R600/si-lod-bias.ll
>> @@ -3,7 +3,7 @@
>> ; This shader has the potential to generated illegal VGPR to SGPR copies if
>> ; the wrong register class is used for the REG_SEQUENCE instructions.
>>
>> -; CHECK: @main
>> +; CHECK: main
>> ; CHECK: IMAGE_SAMPLE_B v{{\[[0-9]:[0-9]\]}}, 15, 0, 0, 0, 0, 0, 0, 0, v{{\[[0-9]:[0-9]\]}}
>>
>> define void @main(<16 x i8> addrspace(2)* inreg, <16 x i8> addrspace(2)* inreg, <32 x i8> addrspace(2)* inreg, i32 inreg, <2 x i32>, <2 x i32>, <2 x i32>, <3 x i32>, <2 x i32>, <2 x i32>, <2 x i32>, float, float, float, float, float, float, float, float, float) #0 {
>> diff --git a/test/CodeGen/R600/si-sgpr-spill.ll b/test/CodeGen/R600/si-sgpr-spill.ll
>> index 53a0965..b28b64c 100644
>> --- a/test/CodeGen/R600/si-sgpr-spill.ll
>> +++ b/test/CodeGen/R600/si-sgpr-spill.ll
>> @@ -3,7 +3,7 @@
>> ; These tests check that the compiler won't crash when it needs to spill
>> ; SGPRs.
>>
>> -; CHECK-LABEL: @main
>> +; CHECK-LABEL: main
>> ; Writing to M0 from an SMRD instruction will hang the GPU.
>> ; CHECK-NOT: S_BUFFER_LOAD_DWORD m0
>> ; CHECK: S_ENDPGM
>> @@ -688,7 +688,7 @@ attributes #4 = { nounwind readonly }
>>
>> !0 = metadata !{metadata !"const", null, i32 1}
>>
>> -; CHECK-LABEL: @main1
>> +; CHECK-LABEL: main1
>> ; CHECK: S_ENDPGM
>> define void @main1([17 x <16 x i8>] addrspace(2)* byval, [32 x <16 x i8>] addrspace(2)* byval, [16 x <32 x i8>] addrspace(2)* byval, float inreg, i32 inreg, <2 x i32>, <2 x i32>, <2 x i32>, <3 x i32>, <2 x i32>, <2 x i32>, <2 x i32>, float, float, float, float, float, float, float, float, float) #0 {
>> main_body:
>> diff --git a/test/CodeGen/R600/si-vector-hang.ll b/test/CodeGen/R600/si-vector-hang.ll
>> index 093234f..7a76338 100644
>> --- a/test/CodeGen/R600/si-vector-hang.ll
>> +++ b/test/CodeGen/R600/si-vector-hang.ll
>> @@ -1,6 +1,6 @@
>> ; RUN: llc < %s -march=r600 -mcpu=verde -verify-machineinstrs | FileCheck %s
>>
>> -; CHECK: @test_8_min_char
>> +; CHECK: test_8_min_char
>> ; CHECK: BUFFER_STORE_BYTE
>> ; CHECK: BUFFER_STORE_BYTE
>> ; CHECK: BUFFER_STORE_BYTE
>> diff --git a/test/CodeGen/R600/sign_extend.ll b/test/CodeGen/R600/sign_extend.ll
>> index dbc2320..25f8abe 100644
>> --- a/test/CodeGen/R600/sign_extend.ll
>> +++ b/test/CodeGen/R600/sign_extend.ll
>> @@ -1,6 +1,6 @@
>> ; RUN: llc -march=r600 -mcpu=SI -verify-machineinstrs < %s | FileCheck -check-prefix=SI %s
>>
>> -; SI-LABEL: @s_sext_i1_to_i32:
>> +; SI-LABEL: s_sext_i1_to_i32:
>> ; SI: V_CNDMASK_B32_e64
>> ; SI: S_ENDPGM
>> define void @s_sext_i1_to_i32(i32 addrspace(1)* %out, i32 %a, i32 %b) nounwind {
>> @@ -10,7 +10,7 @@ define void @s_sext_i1_to_i32(i32 addrspace(1)* %out, i32 %a, i32 %b) nounwind {
>> ret void
>> }
>>
>> -; SI-LABEL: @test_s_sext_i32_to_i64:
>> +; SI-LABEL: test_s_sext_i32_to_i64:
>> ; SI: S_ASHR_I32
>> ; SI: S_ENDPG
>> define void @test_s_sext_i32_to_i64(i64 addrspace(1)* %out, i32 %a, i32 %b, i32 %c) nounwind {
>> @@ -22,7 +22,7 @@ entry:
>> ret void
>> }
>>
>> -; SI-LABEL: @s_sext_i1_to_i64:
>> +; SI-LABEL: s_sext_i1_to_i64:
>> ; SI: V_CNDMASK_B32_e64
>> ; SI: V_CNDMASK_B32_e64
>> ; SI: S_ENDPGM
>> @@ -33,7 +33,7 @@ define void @s_sext_i1_to_i64(i64 addrspace(1)* %out, i32 %a, i32 %b) nounwind {
>> ret void
>> }
>>
>> -; SI-LABEL: @s_sext_i32_to_i64:
>> +; SI-LABEL: s_sext_i32_to_i64:
>> ; SI: S_ASHR_I32
>> ; SI: S_ENDPGM
>> define void @s_sext_i32_to_i64(i64 addrspace(1)* %out, i32 %a) nounwind {
>> @@ -42,7 +42,7 @@ define void @s_sext_i32_to_i64(i64 addrspace(1)* %out, i32 %a) nounwind {
>> ret void
>> }
>>
>> -; SI-LABEL: @v_sext_i32_to_i64:
>> +; SI-LABEL: v_sext_i32_to_i64:
>> ; SI: V_ASHR
>> ; SI: S_ENDPGM
>> define void @v_sext_i32_to_i64(i64 addrspace(1)* %out, i32 addrspace(1)* %in) nounwind {
>> @@ -52,7 +52,7 @@ define void @v_sext_i32_to_i64(i64 addrspace(1)* %out, i32 addrspace(1)* %in) no
>> ret void
>> }
>>
>> -; SI-LABEL: @s_sext_i16_to_i64:
>> +; SI-LABEL: s_sext_i16_to_i64:
>> ; SI: S_ENDPGM
>> define void @s_sext_i16_to_i64(i64 addrspace(1)* %out, i16 %a) nounwind {
>> %sext = sext i16 %a to i64
>> diff --git a/test/CodeGen/R600/simplify-demanded-bits-build-pair.ll b/test/CodeGen/R600/simplify-demanded-bits-build-pair.ll
>> index e6f8ce8..b61fbf5 100644
>> --- a/test/CodeGen/R600/simplify-demanded-bits-build-pair.ll
>> +++ b/test/CodeGen/R600/simplify-demanded-bits-build-pair.ll
>> @@ -14,7 +14,7 @@ define void @trunc_select_i64(i32 addrspace(1)* %out, i64 %a, i64 %b, i32 %c) {
>> }
>>
>> ; FIXME: Fix truncating store for local memory
>> -; SI-LABEL: @trunc_load_alloca_i64:
>> +; SI-LABEL: trunc_load_alloca_i64:
>> ; SI: V_MOVRELS_B32
>> ; SI-NOT: V_MOVRELS_B32
>> ; SI: S_ENDPGM
>> diff --git a/test/CodeGen/R600/sint_to_fp.ll b/test/CodeGen/R600/sint_to_fp.ll
>> index b27dfda..db74c6c 100644
>> --- a/test/CodeGen/R600/sint_to_fp.ll
>> +++ b/test/CodeGen/R600/sint_to_fp.ll
>> @@ -1,10 +1,10 @@
>> ; RUN: llc < %s -march=r600 -mcpu=redwood | FileCheck %s --check-prefix=R600-CHECK
>> ; RUN: llc < %s -march=r600 -mcpu=SI -verify-machineinstrs | FileCheck %s --check-prefix=SI-CHECK
>>
>> -; R600-CHECK: @sint_to_fp_v2i32
>> +; R600-CHECK: sint_to_fp_v2i32
>> ; R600-CHECK-DAG: INT_TO_FLT * T{{[0-9]+\.[XYZW]}}, KC0[2].W
>> ; R600-CHECK-DAG: INT_TO_FLT * T{{[0-9]+\.[XYZW]}}, KC0[3].X
>> -; SI-CHECK: @sint_to_fp_v2i32
>> +; SI-CHECK: sint_to_fp_v2i32
>> ; SI-CHECK: V_CVT_F32_I32_e32
>> ; SI-CHECK: V_CVT_F32_I32_e32
>> define void @sint_to_fp_v2i32(<2 x float> addrspace(1)* %out, <2 x i32> %in) {
>> @@ -13,12 +13,12 @@ define void @sint_to_fp_v2i32(<2 x float> addrspace(1)* %out, <2 x i32> %in) {
>> ret void
>> }
>>
>> -; R600-CHECK: @sint_to_fp_v4i32
>> +; R600-CHECK: sint_to_fp_v4i32
>> ; R600-CHECK: INT_TO_FLT * T{{[0-9]+\.[XYZW], T[0-9]+\.[XYZW]}}
>> ; R600-CHECK: INT_TO_FLT * T{{[0-9]+\.[XYZW], T[0-9]+\.[XYZW]}}
>> ; R600-CHECK: INT_TO_FLT * T{{[0-9]+\.[XYZW], T[0-9]+\.[XYZW]}}
>> ; R600-CHECK: INT_TO_FLT * T{{[0-9]+\.[XYZW], T[0-9]+\.[XYZW]}}
>> -; SI-CHECK: @sint_to_fp_v4i32
>> +; SI-CHECK: sint_to_fp_v4i32
>> ; SI-CHECK: V_CVT_F32_I32_e32
>> ; SI-CHECK: V_CVT_F32_I32_e32
>> ; SI-CHECK: V_CVT_F32_I32_e32
>> @@ -30,7 +30,7 @@ define void @sint_to_fp_v4i32(<4 x float> addrspace(1)* %out, <4 x i32> addrspac
>> ret void
>> }
>>
>> -; FUNC-LABEL: @sint_to_fp_i1_f32:
>> +; FUNC-LABEL: sint_to_fp_i1_f32:
>> ; SI: V_CMP_EQ_I32_e64 [[CMP:s\[[0-9]+:[0-9]\]]],
>> ; SI-NEXT: V_CNDMASK_B32_e64 [[RESULT:v[0-9]+]], 0, -1.000000e+00, [[CMP]]
>> ; SI: BUFFER_STORE_DWORD [[RESULT]],
>> @@ -42,7 +42,7 @@ define void @sint_to_fp_i1_f32(float addrspace(1)* %out, i32 %in) {
>> ret void
>> }
>>
>> -; FUNC-LABEL: @sint_to_fp_i1_f32_load:
>> +; FUNC-LABEL: sint_to_fp_i1_f32_load:
>> ; SI: V_CNDMASK_B32_e64 [[RESULT:v[0-9]+]], 0, -1.000000e+00
>> ; SI: BUFFER_STORE_DWORD [[RESULT]],
>> ; SI: S_ENDPGM
>> diff --git a/test/CodeGen/R600/sint_to_fp64.ll b/test/CodeGen/R600/sint_to_fp64.ll
>> index 12b8cf5..9622dfc 100644
>> --- a/test/CodeGen/R600/sint_to_fp64.ll
>> +++ b/test/CodeGen/R600/sint_to_fp64.ll
>> @@ -1,6 +1,6 @@
>> ; RUN: llc -march=r600 -mcpu=SI -verify-machineinstrs < %s | FileCheck -check-prefix=SI %s
>>
>> -; SI: @sint_to_fp64
>> +; SI: sint_to_fp64
>> ; SI: V_CVT_F64_I32_e32
>> define void @sint_to_fp64(double addrspace(1)* %out, i32 %in) {
>> %result = sitofp i32 %in to double
>> @@ -8,7 +8,7 @@ define void @sint_to_fp64(double addrspace(1)* %out, i32 %in) {
>> ret void
>> }
>>
>> -; SI-LABEL: @sint_to_fp_i1_f64:
>> +; SI-LABEL: sint_to_fp_i1_f64:
>> ; SI: V_CMP_EQ_I32_e64 [[CMP:s\[[0-9]+:[0-9]\]]],
>> ; FIXME: We should the VGPR sources for V_CNDMASK are copied from SGPRs,
>> ; we should be able to fold the SGPRs into the V_CNDMASK instructions.
>> @@ -23,7 +23,7 @@ define void @sint_to_fp_i1_f64(double addrspace(1)* %out, i32 %in) {
>> ret void
>> }
>>
>> -; SI-LABEL: @sint_to_fp_i1_f64_load:
>> +; SI-LABEL: sint_to_fp_i1_f64_load:
>> ; SI: V_CNDMASK_B32_e64 [[IRESULT:v[0-9]]], 0, -1
>> ; SI-NEXT: V_CVT_F64_I32_e32 [[RESULT:v\[[0-9]+:[0-9]\]]], [[IRESULT]]
>> ; SI: BUFFER_STORE_DWORDX2 [[RESULT]]
>> diff --git a/test/CodeGen/R600/smrd.ll b/test/CodeGen/R600/smrd.ll
>> index b63b01e..b9b81d5 100644
>> --- a/test/CodeGen/R600/smrd.ll
>> +++ b/test/CodeGen/R600/smrd.ll
>> @@ -1,7 +1,7 @@
>> ; RUN: llc < %s -march=r600 -mcpu=SI -show-mc-encoding -verify-machineinstrs | FileCheck %s
>>
>> ; SMRD load with an immediate offset.
>> -; CHECK-LABEL: @smrd0
>> +; CHECK-LABEL: smrd0
>> ; CHECK: S_LOAD_DWORD s{{[0-9]}}, s[{{[0-9]:[0-9]}}], 0x1 ; encoding: [0x01
>> define void @smrd0(i32 addrspace(1)* %out, i32 addrspace(2)* %ptr) {
>> entry:
>> @@ -12,7 +12,7 @@ entry:
>> }
>>
>> ; SMRD load with the largest possible immediate offset.
>> -; CHECK-LABEL: @smrd1
>> +; CHECK-LABEL: smrd1
>> ; CHECK: S_LOAD_DWORD s{{[0-9]}}, s[{{[0-9]:[0-9]}}], 0xff ; encoding: [0xff
>> define void @smrd1(i32 addrspace(1)* %out, i32 addrspace(2)* %ptr) {
>> entry:
>> @@ -23,7 +23,7 @@ entry:
>> }
>>
>> ; SMRD load with an offset greater than the largest possible immediate.
>> -; CHECK-LABEL: @smrd2
>> +; CHECK-LABEL: smrd2
>> ; CHECK: S_MOV_B32 s[[OFFSET:[0-9]]], 0x400
>> ; CHECK: S_LOAD_DWORD s{{[0-9]}}, s[{{[0-9]:[0-9]}}], s[[OFFSET]] ; encoding: [0x0[[OFFSET]]
>> ; CHECK: S_ENDPGM
>> @@ -36,7 +36,7 @@ entry:
>> }
>>
>> ; SMRD load with a 64-bit offset
>> -; CHECK-LABEL: @smrd3
>> +; CHECK-LABEL: smrd3
>> ; CHECK-DAG: S_MOV_B32 s[[SLO:[0-9]+]], 0
>> ; CHECK-DAG: S_MOV_B32 s[[SHI:[0-9]+]], 4
>> ; FIXME: We don't need to copy these values to VGPRs
>> @@ -54,7 +54,7 @@ entry:
>> }
>>
>> ; SMRD load using the load.const intrinsic with an immediate offset
>> -; CHECK-LABEL: @smrd_load_const0
>> +; CHECK-LABEL: smrd_load_const0
>> ; CHECK: S_BUFFER_LOAD_DWORD s{{[0-9]}}, s[{{[0-9]:[0-9]}}], 0x4 ; encoding: [0x04
>> define void @smrd_load_const0(<16 x i8> addrspace(2)* inreg, <16 x i8> addrspace(2)* inreg, <32 x i8> addrspace(2)* inreg, i32 inreg, <2 x i32>, <2 x i32>, <2 x i32>, <3 x i32>, <2 x i32>, <2 x i32>, <2 x i32>, float, float, float, float, float, float, float, float, float) #0 {
>> main_body:
>> @@ -67,7 +67,7 @@ main_body:
>>
>> ; SMRD load using the load.const intrinsic with the largest possible immediate
>> ; offset.
>> -; CHECK-LABEL: @smrd_load_const1
>> +; CHECK-LABEL: smrd_load_const1
>> ; CHECK: S_BUFFER_LOAD_DWORD s{{[0-9]}}, s[{{[0-9]:[0-9]}}], 0xff ; encoding: [0xff
>> define void @smrd_load_const1(<16 x i8> addrspace(2)* inreg, <16 x i8> addrspace(2)* inreg, <32 x i8> addrspace(2)* inreg, i32 inreg, <2 x i32>, <2 x i32>, <2 x i32>, <3 x i32>, <2 x i32>, <2 x i32>, <2 x i32>, float, float, float, float, float, float, float, float, float) #0 {
>> main_body:
>> @@ -80,7 +80,7 @@ main_body:
>> ; SMRD load using the load.const intrinsic with an offset greater than the
>> ; largets possible immediate.
>> ; immediate offset.
>> -; CHECK-LABEL: @smrd_load_const2
>> +; CHECK-LABEL: smrd_load_const2
>> ; CHECK: S_MOV_B32 s[[OFFSET:[0-9]]], 0x400
>> ; CHECK: S_BUFFER_LOAD_DWORD s{{[0-9]}}, s[{{[0-9]:[0-9]}}], s[[OFFSET]] ; encoding: [0x0[[OFFSET]]
>> define void @smrd_load_const2(<16 x i8> addrspace(2)* inreg, <16 x i8> addrspace(2)* inreg, <32 x i8> addrspace(2)* inreg, i32 inreg, <2 x i32>, <2 x i32>, <2 x i32>, <3 x i32>, <2 x i32>, <2 x i32>, <2 x i32>, float, float, float, float, float, float, float, float, float) #0 {
>> diff --git a/test/CodeGen/R600/split-scalar-i64-add.ll b/test/CodeGen/R600/split-scalar-i64-add.ll
>> index cc50f89..71ef36a 100644
>> --- a/test/CodeGen/R600/split-scalar-i64-add.ll
>> +++ b/test/CodeGen/R600/split-scalar-i64-add.ll
>> @@ -8,7 +8,7 @@ declare i32 @llvm.r600.read.tidig.x() readnone
>> ; set in vcc, which is undefined since the low scalar half add sets
>> ; scc instead.
>>
>> -; FUNC-LABEL: @imp_def_vcc_split_i64_add_0
>> +; FUNC-LABEL: imp_def_vcc_split_i64_add_0
>> define void @imp_def_vcc_split_i64_add_0(i64 addrspace(1)* %out, i32 %val) {
>> %vec.0 = insertelement <2 x i32> undef, i32 %val, i32 0
>> %vec.1 = insertelement <2 x i32> %vec.0, i32 999999, i32 1
>> @@ -18,7 +18,7 @@ define void @imp_def_vcc_split_i64_add_0(i64 addrspace(1)* %out, i32 %val) {
>> ret void
>> }
>>
>> -; FUNC-LABEL: @imp_def_vcc_split_i64_add_1
>> +; FUNC-LABEL: imp_def_vcc_split_i64_add_1
>> define void @imp_def_vcc_split_i64_add_1(i64 addrspace(1)* %out, i32 %val0, i64 %val1) {
>> %vec.0 = insertelement <2 x i32> undef, i32 %val0, i32 0
>> %vec.1 = insertelement <2 x i32> %vec.0, i32 99999, i32 1
>> diff --git a/test/CodeGen/R600/sra.ll b/test/CodeGen/R600/sra.ll
>> index 9eb3dc5..643b761 100644
>> --- a/test/CodeGen/R600/sra.ll
>> +++ b/test/CodeGen/R600/sra.ll
>> @@ -1,11 +1,11 @@
>> ;RUN: llc < %s -march=r600 -mcpu=redwood | FileCheck --check-prefix=EG-CHECK %s
>> ;RUN: llc < %s -march=r600 -mcpu=verde -verify-machineinstrs | FileCheck --check-prefix=SI-CHECK %s
>>
>> -;EG-CHECK-LABEL: @ashr_v2i32
>> +;EG-CHECK-LABEL: ashr_v2i32
>> ;EG-CHECK: ASHR {{\*? *}}T{{[0-9]+\.[XYZW], T[0-9]+\.[XYZW], T[0-9]+\.[XYZW]}}
>> ;EG-CHECK: ASHR {{\*? *}}T{{[0-9]+\.[XYZW], T[0-9]+\.[XYZW], T[0-9]+\.[XYZW]}}
>>
>> -;SI-CHECK-LABEL: @ashr_v2i32
>> +;SI-CHECK-LABEL: ashr_v2i32
>> ;SI-CHECK: V_ASHR_I32_e32 v{{[0-9]+, v[0-9]+, v[0-9]+}}
>> ;SI-CHECK: V_ASHR_I32_e32 v{{[0-9]+, v[0-9]+, v[0-9]+}}
>>
>> @@ -18,13 +18,13 @@ define void @ashr_v2i32(<2 x i32> addrspace(1)* %out, <2 x i32> addrspace(1)* %i
>> ret void
>> }
>>
>> -;EG-CHECK-LABEL: @ashr_v4i32
>> +;EG-CHECK-LABEL: ashr_v4i32
>> ;EG-CHECK: ASHR {{\*? *}}T{{[0-9]+\.[XYZW], T[0-9]+\.[XYZW], T[0-9]+\.[XYZW]}}
>> ;EG-CHECK: ASHR {{\*? *}}T{{[0-9]+\.[XYZW], T[0-9]+\.[XYZW], T[0-9]+\.[XYZW]}}
>> ;EG-CHECK: ASHR {{\*? *}}T{{[0-9]+\.[XYZW], T[0-9]+\.[XYZW], T[0-9]+\.[XYZW]}}
>> ;EG-CHECK: ASHR {{\*? *}}T{{[0-9]+\.[XYZW], T[0-9]+\.[XYZW], T[0-9]+\.[XYZW]}}
>>
>> -;SI-CHECK-LABEL: @ashr_v4i32
>> +;SI-CHECK-LABEL: ashr_v4i32
>> ;SI-CHECK: V_ASHR_I32_e32 v{{[0-9]+, v[0-9]+, v[0-9]+}}
>> ;SI-CHECK: V_ASHR_I32_e32 v{{[0-9]+, v[0-9]+, v[0-9]+}}
>> ;SI-CHECK: V_ASHR_I32_e32 v{{[0-9]+, v[0-9]+, v[0-9]+}}
>> @@ -39,10 +39,10 @@ define void @ashr_v4i32(<4 x i32> addrspace(1)* %out, <4 x i32> addrspace(1)* %i
>> ret void
>> }
>>
>> -;EG-CHECK-LABEL: @ashr_i64
>> +;EG-CHECK-LABEL: ashr_i64
>> ;EG-CHECK: ASHR
>>
>> -;SI-CHECK-LABEL: @ashr_i64
>> +;SI-CHECK-LABEL: ashr_i64
>> ;SI-CHECK: S_ASHR_I64 s[{{[0-9]}}:{{[0-9]}}], s[{{[0-9]}}:{{[0-9]}}], 8
>> define void @ashr_i64(i64 addrspace(1)* %out, i32 %in) {
>> entry:
>> @@ -52,7 +52,7 @@ entry:
>> ret void
>> }
>>
>> -;EG-CHECK-LABEL: @ashr_i64_2
>> +;EG-CHECK-LABEL: ashr_i64_2
>> ;EG-CHECK: SUB_INT {{\*? *}}[[COMPSH:T[0-9]+\.[XYZW]]], {{literal.[xy]}}, [[SHIFT:T[0-9]+\.[XYZW]]]
>> ;EG-CHECK: LSHL {{\* *}}[[TEMP:T[0-9]+\.[XYZW]]], [[OPHI:T[0-9]+\.[XYZW]]], {{[[COMPSH]]|PV.[XYZW]}}
>> ;EG-CHECK: LSHL {{\*? *}}[[OVERF:T[0-9]+\.[XYZW]]], {{[[TEMP]]|PV.[XYZW]}}, 1
>> @@ -66,7 +66,7 @@ entry:
>> ;EG-CHECK-DAG: CNDE_INT {{\*? *}}[[RESLO:T[0-9]+\.[XYZW]]], {{T[0-9]+\.[XYZW]}}
>> ;EG-CHECK-DAG: CNDE_INT {{\*? *}}[[RESHI:T[0-9]+\.[XYZW]]], {{T[0-9]+\.[XYZW]}}
>>
>> -;SI-CHECK-LABEL: @ashr_i64_2
>> +;SI-CHECK-LABEL: ashr_i64_2
>> ;SI-CHECK: V_ASHR_I64 {{v\[[0-9]+:[0-9]+\], v\[[0-9]+:[0-9]+\], v[0-9]+}}
>> define void @ashr_i64_2(i64 addrspace(1)* %out, i64 addrspace(1)* %in) {
>> entry:
>> @@ -78,7 +78,7 @@ entry:
>> ret void
>> }
>>
>> -;EG-CHECK-LABEL: @ashr_v2i64
>> +;EG-CHECK-LABEL: ashr_v2i64
>> ;EG-CHECK-DAG: SUB_INT {{\*? *}}[[COMPSHA:T[0-9]+\.[XYZW]]], {{literal.[xy]}}, [[SHA:T[0-9]+\.[XYZW]]]
>> ;EG-CHECK-DAG: SUB_INT {{\*? *}}[[COMPSHB:T[0-9]+\.[XYZW]]], {{literal.[xy]}}, [[SHB:T[0-9]+\.[XYZW]]]
>> ;EG-CHECK-DAG: LSHL {{\*? *}}[[COMPSHA]]
>> @@ -104,7 +104,7 @@ entry:
>> ;EG-CHECK-DAG: CNDE_INT
>> ;EG-CHECK-DAG: CNDE_INT
>>
>> -;SI-CHECK-LABEL: @ashr_v2i64
>> +;SI-CHECK-LABEL: ashr_v2i64
>> ;SI-CHECK: V_ASHR_I64 {{v\[[0-9]+:[0-9]+\], v\[[0-9]+:[0-9]+\], v[0-9]+}}
>> ;SI-CHECK: V_ASHR_I64 {{v\[[0-9]+:[0-9]+\], v\[[0-9]+:[0-9]+\], v[0-9]+}}
>>
>> @@ -117,7 +117,7 @@ define void @ashr_v2i64(<2 x i64> addrspace(1)* %out, <2 x i64> addrspace(1)* %i
>> ret void
>> }
>>
>> -;EG-CHECK-LABEL: @ashr_v4i64
>> +;EG-CHECK-LABEL: ashr_v4i64
>> ;EG-CHECK-DAG: SUB_INT {{\*? *}}[[COMPSHA:T[0-9]+\.[XYZW]]], {{literal.[xy]}}, [[SHA:T[0-9]+\.[XYZW]]]
>> ;EG-CHECK-DAG: SUB_INT {{\*? *}}[[COMPSHB:T[0-9]+\.[XYZW]]], {{literal.[xy]}}, [[SHB:T[0-9]+\.[XYZW]]]
>> ;EG-CHECK-DAG: SUB_INT {{\*? *}}[[COMPSHC:T[0-9]+\.[XYZW]]], {{literal.[xy]}}, [[SHC:T[0-9]+\.[XYZW]]]
>> @@ -167,7 +167,7 @@ define void @ashr_v2i64(<2 x i64> addrspace(1)* %out, <2 x i64> addrspace(1)* %i
>> ;EG-CHECK-DAG: CNDE_INT
>> ;EG-CHECK-DAG: CNDE_INT
>>
>> -;SI-CHECK-LABEL: @ashr_v4i64
>> +;SI-CHECK-LABEL: ashr_v4i64
>> ;SI-CHECK: V_ASHR_I64 {{v\[[0-9]+:[0-9]+\], v\[[0-9]+:[0-9]+\], v[0-9]+}}
>> ;SI-CHECK: V_ASHR_I64 {{v\[[0-9]+:[0-9]+\], v\[[0-9]+:[0-9]+\], v[0-9]+}}
>> ;SI-CHECK: V_ASHR_I64 {{v\[[0-9]+:[0-9]+\], v\[[0-9]+:[0-9]+\], v[0-9]+}}
>> diff --git a/test/CodeGen/R600/srl.ll b/test/CodeGen/R600/srl.ll
>> index 44ad73f..c782e8e 100644
>> --- a/test/CodeGen/R600/srl.ll
>> +++ b/test/CodeGen/R600/srl.ll
>> @@ -1,11 +1,11 @@
>> ;RUN: llc < %s -march=r600 -mcpu=redwood | FileCheck --check-prefix=EG-CHECK %s
>> ;RUN: llc < %s -march=r600 -mcpu=verde -verify-machineinstrs | FileCheck --check-prefix=SI-CHECK %s
>>
>> -;EG-CHECK: @lshr_v2i32
>> +;EG-CHECK: lshr_v2i32
>> ;EG-CHECK: LSHR {{\*? *}}T{{[0-9]+\.[XYZW], T[0-9]+\.[XYZW], T[0-9]+\.[XYZW]}}
>> ;EG-CHECK: LSHR {{\*? *}}T{{[0-9]+\.[XYZW], T[0-9]+\.[XYZW], T[0-9]+\.[XYZW]}}
>>
>> -;SI-CHECK: @lshr_v2i32
>> +;SI-CHECK: lshr_v2i32
>> ;SI-CHECK: V_LSHR_B32_e32 v{{[0-9]+, v[0-9]+, v[0-9]+}}
>> ;SI-CHECK: V_LSHR_B32_e32 v{{[0-9]+, v[0-9]+, v[0-9]+}}
>>
>> @@ -19,13 +19,13 @@ define void @lshr_v2i32(<2 x i32> addrspace(1)* %out, <2 x i32> addrspace(1)* %i
>> }
>>
>>
>> -;EG-CHECK: @lshr_v4i32
>> +;EG-CHECK: lshr_v4i32
>> ;EG-CHECK: LSHR {{\*? *}}T{{[0-9]+\.[XYZW], T[0-9]+\.[XYZW], T[0-9]+\.[XYZW]}}
>> ;EG-CHECK: LSHR {{\*? *}}T{{[0-9]+\.[XYZW], T[0-9]+\.[XYZW], T[0-9]+\.[XYZW]}}
>> ;EG-CHECK: LSHR {{\*? *}}T{{[0-9]+\.[XYZW], T[0-9]+\.[XYZW], T[0-9]+\.[XYZW]}}
>> ;EG-CHECK: LSHR {{\*? *}}T{{[0-9]+\.[XYZW], T[0-9]+\.[XYZW], T[0-9]+\.[XYZW]}}
>>
>> -;SI-CHECK: @lshr_v4i32
>> +;SI-CHECK: lshr_v4i32
>> ;SI-CHECK: V_LSHR_B32_e32 v{{[0-9]+, v[0-9]+, v[0-9]+}}
>> ;SI-CHECK: V_LSHR_B32_e32 v{{[0-9]+, v[0-9]+, v[0-9]+}}
>> ;SI-CHECK: V_LSHR_B32_e32 v{{[0-9]+, v[0-9]+, v[0-9]+}}
>> @@ -40,7 +40,7 @@ define void @lshr_v4i32(<4 x i32> addrspace(1)* %out, <4 x i32> addrspace(1)* %i
>> ret void
>> }
>>
>> -;EG-CHECK: @lshr_i64
>> +;EG-CHECK: lshr_i64
>> ;EG-CHECK: SUB_INT {{\*? *}}[[COMPSH:T[0-9]+\.[XYZW]]], {{literal.[xy]}}, [[SHIFT:T[0-9]+\.[XYZW]]]
>> ;EG-CHECK: LSHL {{\* *}}[[TEMP:T[0-9]+\.[XYZW]]], [[OPHI:T[0-9]+\.[XYZW]]], {{[[COMPSH]]|PV.[XYZW]}}
>> ;EG-CHECK: LSHL {{\*? *}}[[OVERF:T[0-9]+\.[XYZW]]], {{[[TEMP]]|PV.[XYZW]}}, 1
>> @@ -53,7 +53,7 @@ define void @lshr_v4i32(<4 x i32> addrspace(1)* %out, <4 x i32> addrspace(1)* %i
>> ;EG-CHECK-DAG: CNDE_INT {{\*? *}}[[RESLO:T[0-9]+\.[XYZW]]], {{T[0-9]+\.[XYZW]}}
>> ;EG-CHECK-DAG: CNDE_INT {{\*? *}}[[RESHI:T[0-9]+\.[XYZW]]], {{T[0-9]+\.[XYZW], .*}}, 0.0
>>
>> -;SI-CHECK: @lshr_i64
>> +;SI-CHECK: lshr_i64
>> ;SI-CHECK: V_LSHR_B64 {{v\[[0-9]+:[0-9]+\], v\[[0-9]+:[0-9]+\], v[0-9]+}}
>>
>> define void @lshr_i64(i64 addrspace(1)* %out, i64 addrspace(1)* %in) {
>> @@ -65,7 +65,7 @@ define void @lshr_i64(i64 addrspace(1)* %out, i64 addrspace(1)* %in) {
>> ret void
>> }
>>
>> -;EG-CHECK: @lshr_v2i64
>> +;EG-CHECK: lshr_v2i64
>> ;EG-CHECK-DAG: SUB_INT {{\*? *}}[[COMPSHA:T[0-9]+\.[XYZW]]], {{literal.[xy]}}, [[SHA:T[0-9]+\.[XYZW]]]
>> ;EG-CHECK-DAG: SUB_INT {{\*? *}}[[COMPSHB:T[0-9]+\.[XYZW]]], {{literal.[xy]}}, [[SHB:T[0-9]+\.[XYZW]]]
>> ;EG-CHECK-DAG: LSHL {{\*? *}}[[COMPSHA]]
>> @@ -89,7 +89,7 @@ define void @lshr_i64(i64 addrspace(1)* %out, i64 addrspace(1)* %in) {
>> ;EG-CHECK-DAG: CNDE_INT
>> ;EG-CHECK-DAG: CNDE_INT
>>
>> -;SI-CHECK: @lshr_v2i64
>> +;SI-CHECK: lshr_v2i64
>> ;SI-CHECK: V_LSHR_B64 {{v\[[0-9]+:[0-9]+\], v\[[0-9]+:[0-9]+\], v[0-9]+}}
>> ;SI-CHECK: V_LSHR_B64 {{v\[[0-9]+:[0-9]+\], v\[[0-9]+:[0-9]+\], v[0-9]+}}
>>
>> @@ -103,7 +103,7 @@ define void @lshr_v2i64(<2 x i64> addrspace(1)* %out, <2 x i64> addrspace(1)* %i
>> }
>>
>>
>> -;EG-CHECK: @lshr_v4i64
>> +;EG-CHECK: lshr_v4i64
>> ;EG-CHECK-DAG: SUB_INT {{\*? *}}[[COMPSHA:T[0-9]+\.[XYZW]]], {{literal.[xy]}}, [[SHA:T[0-9]+\.[XYZW]]]
>> ;EG-CHECK-DAG: SUB_INT {{\*? *}}[[COMPSHB:T[0-9]+\.[XYZW]]], {{literal.[xy]}}, [[SHB:T[0-9]+\.[XYZW]]]
>> ;EG-CHECK-DAG: SUB_INT {{\*? *}}[[COMPSHC:T[0-9]+\.[XYZW]]], {{literal.[xy]}}, [[SHC:T[0-9]+\.[XYZW]]]
>> @@ -151,7 +151,7 @@ define void @lshr_v2i64(<2 x i64> addrspace(1)* %out, <2 x i64> addrspace(1)* %i
>> ;EG-CHECK-DAG: CNDE_INT
>> ;EG-CHECK-DAG: CNDE_INT
>>
>> -;SI-CHECK: @lshr_v4i64
>> +;SI-CHECK: lshr_v4i64
>> ;SI-CHECK: V_LSHR_B64 {{v\[[0-9]+:[0-9]+\], v\[[0-9]+:[0-9]+\], v[0-9]+}}
>> ;SI-CHECK: V_LSHR_B64 {{v\[[0-9]+:[0-9]+\], v\[[0-9]+:[0-9]+\], v[0-9]+}}
>> ;SI-CHECK: V_LSHR_B64 {{v\[[0-9]+:[0-9]+\], v\[[0-9]+:[0-9]+\], v[0-9]+}}
>> diff --git a/test/CodeGen/R600/ssubo.ll b/test/CodeGen/R600/ssubo.ll
>> index 066cdf5..307c2c3 100644
>> --- a/test/CodeGen/R600/ssubo.ll
>> +++ b/test/CodeGen/R600/ssubo.ll
>> @@ -4,7 +4,7 @@
>> declare { i32, i1 } @llvm.ssub.with.overflow.i32(i32, i32) nounwind readnone
>> declare { i64, i1 } @llvm.ssub.with.overflow.i64(i64, i64) nounwind readnone
>>
>> -; FUNC-LABEL: @ssubo_i64_zext
>> +; FUNC-LABEL: ssubo_i64_zext
>> define void @ssubo_i64_zext(i64 addrspace(1)* %out, i64 %a, i64 %b) nounwind {
>> %ssub = call { i64, i1 } @llvm.ssub.with.overflow.i64(i64 %a, i64 %b) nounwind
>> %val = extractvalue { i64, i1 } %ssub, 0
>> @@ -15,7 +15,7 @@ define void @ssubo_i64_zext(i64 addrspace(1)* %out, i64 %a, i64 %b) nounwind {
>> ret void
>> }
>>
>> -; FUNC-LABEL: @s_ssubo_i32
>> +; FUNC-LABEL: s_ssubo_i32
>> define void @s_ssubo_i32(i32 addrspace(1)* %out, i1 addrspace(1)* %carryout, i32 %a, i32 %b) nounwind {
>> %ssub = call { i32, i1 } @llvm.ssub.with.overflow.i32(i32 %a, i32 %b) nounwind
>> %val = extractvalue { i32, i1 } %ssub, 0
>> @@ -25,7 +25,7 @@ define void @s_ssubo_i32(i32 addrspace(1)* %out, i1 addrspace(1)* %carryout, i32
>> ret void
>> }
>>
>> -; FUNC-LABEL: @v_ssubo_i32
>> +; FUNC-LABEL: v_ssubo_i32
>> define void @v_ssubo_i32(i32 addrspace(1)* %out, i1 addrspace(1)* %carryout, i32 addrspace(1)* %aptr, i32 addrspace(1)* %bptr) nounwind {
>> %a = load i32 addrspace(1)* %aptr, align 4
>> %b = load i32 addrspace(1)* %bptr, align 4
>> @@ -37,7 +37,7 @@ define void @v_ssubo_i32(i32 addrspace(1)* %out, i1 addrspace(1)* %carryout, i32
>> ret void
>> }
>>
>> -; FUNC-LABEL: @s_ssubo_i64
>> +; FUNC-LABEL: s_ssubo_i64
>> ; SI: S_SUB_U32
>> ; SI: S_SUBB_U32
>> define void @s_ssubo_i64(i64 addrspace(1)* %out, i1 addrspace(1)* %carryout, i64 %a, i64 %b) nounwind {
>> @@ -49,7 +49,7 @@ define void @s_ssubo_i64(i64 addrspace(1)* %out, i1 addrspace(1)* %carryout, i64
>> ret void
>> }
>>
>> -; FUNC-LABEL: @v_ssubo_i64
>> +; FUNC-LABEL: v_ssubo_i64
>> ; SI: V_SUB_I32_e32
>> ; SI: V_SUBB_U32_e32
>> define void @v_ssubo_i64(i64 addrspace(1)* %out, i1 addrspace(1)* %carryout, i64 addrspace(1)* %aptr, i64 addrspace(1)* %bptr) nounwind {
>> diff --git a/test/CodeGen/R600/store-v3i32.ll b/test/CodeGen/R600/store-v3i32.ll
>> index 3357803..7a1d8a7 100644
>> --- a/test/CodeGen/R600/store-v3i32.ll
>> +++ b/test/CodeGen/R600/store-v3i32.ll
>> @@ -4,7 +4,7 @@
>> ; 3 vectors have the same size and alignment as 4 vectors, so this
>> ; should be done in a single store.
>>
>> -; SI-LABEL: @store_v3i32:
>> +; SI-LABEL: store_v3i32:
>> ; SI: BUFFER_STORE_DWORDX4
>> define void @store_v3i32(<3 x i32> addrspace(1)* %out, <3 x i32> %a) nounwind {
>> store <3 x i32> %a, <3 x i32> addrspace(1)* %out, align 16
>> diff --git a/test/CodeGen/R600/store-v3i64.ll b/test/CodeGen/R600/store-v3i64.ll
>> index 58d28b5..c58262a 100644
>> --- a/test/CodeGen/R600/store-v3i64.ll
>> +++ b/test/CodeGen/R600/store-v3i64.ll
>> @@ -1,7 +1,7 @@
>> ; XFAIL: *
>> ; RUN: llc -march=r600 -mcpu=SI -verify-machineinstrs< %s | FileCheck -check-prefix=SI
>>
>> -; SI-LABEL: @global_store_v3i64:
>> +; SI-LABEL: global_store_v3i64:
>> ; SI: BUFFER_STORE_DWORDX4
>> ; SI: BUFFER_STORE_DWORDX4
>> define void @global_store_v3i64(<3 x i64> addrspace(1)* %out, <3 x i64> %x) {
>> @@ -9,19 +9,19 @@ define void @global_store_v3i64(<3 x i64> addrspace(1)* %out, <3 x i64> %x) {
>> ret void
>> }
>>
>> -; SI-LABEL: @global_store_v3i64_unaligned:
>> +; SI-LABEL: global_store_v3i64_unaligned:
>> define void @global_store_v3i64_unaligned(<3 x i64> addrspace(1)* %out, <3 x i64> %x) {
>> store <3 x i64> %x, <3 x i64> addrspace(1)* %out, align 1
>> ret void
>> }
>>
>> -; SI-LABEL: @local_store_v3i64:
>> +; SI-LABEL: local_store_v3i64:
>> define void @local_store_v3i64(<3 x i64> addrspace(3)* %out, <3 x i64> %x) {
>> store <3 x i64> %x, <3 x i64> addrspace(3)* %out, align 32
>> ret void
>> }
>>
>> -; SI-LABEL: @local_store_v3i64_unaligned:
>> +; SI-LABEL: local_store_v3i64_unaligned:
>> define void @local_store_v3i64_unaligned(<3 x i64> addrspace(1)* %out, <3 x i64> %x) {
>> store <3 x i64> %x, <3 x i64> addrspace(1)* %out, align 1
>> ret void
>> diff --git a/test/CodeGen/R600/store-vector-ptrs.ll b/test/CodeGen/R600/store-vector-ptrs.ll
>> index b2ea846..2e3c64e 100644
>> --- a/test/CodeGen/R600/store-vector-ptrs.ll
>> +++ b/test/CodeGen/R600/store-vector-ptrs.ll
>> @@ -3,7 +3,7 @@
>> ; This tests for a bug that caused a crash in
>> ; AMDGPUDAGToDAGISel::SelectMUBUFScratch() which is used for selecting
>> ; scratch loads and stores.
>> -; CHECK-LABEL: @store_vector_ptrs
>> +; CHECK-LABEL: store_vector_ptrs
>> define void @store_vector_ptrs(<4 x i32*>* %out, <4 x [1024 x i32]*> %array) nounwind {
>> %p = getelementptr <4 x [1024 x i32]*> %array, <4 x i16> zeroinitializer, <4 x i16> <i16 16, i16 16, i16 16, i16 16>
>> store <4 x i32*> %p, <4 x i32*>* %out
>> diff --git a/test/CodeGen/R600/store.ll b/test/CodeGen/R600/store.ll
>> index dd27533..d20552f 100644
>> --- a/test/CodeGen/R600/store.ll
>> +++ b/test/CodeGen/R600/store.ll
>> @@ -5,7 +5,7 @@
>> ;===------------------------------------------------------------------------===;
>> ; Global Address Space
>> ;===------------------------------------------------------------------------===;
>> -; FUNC-LABEL: @store_i1
>> +; FUNC-LABEL: store_i1
>> ; EG-CHECK: MEM_RAT MSKOR
>> ; SI-CHECK: BUFFER_STORE_BYTE
>> define void @store_i1(i1 addrspace(1)* %out) {
>> @@ -15,7 +15,7 @@ entry:
>> }
>>
>> ; i8 store
>> -; EG-CHECK-LABEL: @store_i8
>> +; EG-CHECK-LABEL: store_i8
>> ; EG-CHECK: MEM_RAT MSKOR T[[RW_GPR:[0-9]]].XW, T{{[0-9]}}.X
>> ; EG-CHECK: VTX_READ_8 [[VAL:T[0-9]\.X]], [[VAL]]
>> ; IG 0: Get the byte index and truncate the value
>> @@ -34,7 +34,7 @@ entry:
>> ; EG-CHECK: MOV T[[RW_GPR]].Y, 0.0
>> ; EG-CHECK: MOV * T[[RW_GPR]].Z, 0.0
>>
>> -; SI-CHECK-LABEL: @store_i8
>> +; SI-CHECK-LABEL: store_i8
>> ; SI-CHECK: BUFFER_STORE_BYTE
>>
>> define void @store_i8(i8 addrspace(1)* %out, i8 %in) {
>> @@ -44,7 +44,7 @@ entry:
>> }
>>
>> ; i16 store
>> -; EG-CHECK-LABEL: @store_i16
>> +; EG-CHECK-LABEL: store_i16
>> ; EG-CHECK: MEM_RAT MSKOR T[[RW_GPR:[0-9]]].XW, T{{[0-9]}}.X
>> ; EG-CHECK: VTX_READ_16 [[VAL:T[0-9]\.X]], [[VAL]]
>> ; IG 0: Get the byte index and truncate the value
>> @@ -63,7 +63,7 @@ entry:
>> ; EG-CHECK: MOV T[[RW_GPR]].Y, 0.0
>> ; EG-CHECK: MOV * T[[RW_GPR]].Z, 0.0
>>
>> -; SI-CHECK-LABEL: @store_i16
>> +; SI-CHECK-LABEL: store_i16
>> ; SI-CHECK: BUFFER_STORE_SHORT
>> define void @store_i16(i16 addrspace(1)* %out, i16 %in) {
>> entry:
>> @@ -71,10 +71,10 @@ entry:
>> ret void
>> }
>>
>> -; EG-CHECK-LABEL: @store_v2i8
>> +; EG-CHECK-LABEL: store_v2i8
>> ; EG-CHECK: MEM_RAT MSKOR
>> ; EG-CHECK-NOT: MEM_RAT MSKOR
>> -; SI-CHECK-LABEL: @store_v2i8
>> +; SI-CHECK-LABEL: store_v2i8
>> ; SI-CHECK: BUFFER_STORE_BYTE
>> ; SI-CHECK: BUFFER_STORE_BYTE
>> define void @store_v2i8(<2 x i8> addrspace(1)* %out, <2 x i32> %in) {
>> @@ -85,11 +85,11 @@ entry:
>> }
>>
>>
>> -; EG-CHECK-LABEL: @store_v2i16
>> +; EG-CHECK-LABEL: store_v2i16
>> ; EG-CHECK: MEM_RAT_CACHELESS STORE_RAW
>> -; CM-CHECK-LABEL: @store_v2i16
>> +; CM-CHECK-LABEL: store_v2i16
>> ; CM-CHECK: MEM_RAT_CACHELESS STORE_DWORD
>> -; SI-CHECK-LABEL: @store_v2i16
>> +; SI-CHECK-LABEL: store_v2i16
>> ; SI-CHECK: BUFFER_STORE_SHORT
>> ; SI-CHECK: BUFFER_STORE_SHORT
>> define void @store_v2i16(<2 x i16> addrspace(1)* %out, <2 x i32> %in) {
>> @@ -99,11 +99,11 @@ entry:
>> ret void
>> }
>>
>> -; EG-CHECK-LABEL: @store_v4i8
>> +; EG-CHECK-LABEL: store_v4i8
>> ; EG-CHECK: MEM_RAT_CACHELESS STORE_RAW
>> -; CM-CHECK-LABEL: @store_v4i8
>> +; CM-CHECK-LABEL: store_v4i8
>> ; CM-CHECK: MEM_RAT_CACHELESS STORE_DWORD
>> -; SI-CHECK-LABEL: @store_v4i8
>> +; SI-CHECK-LABEL: store_v4i8
>> ; SI-CHECK: BUFFER_STORE_BYTE
>> ; SI-CHECK: BUFFER_STORE_BYTE
>> ; SI-CHECK: BUFFER_STORE_BYTE
>> @@ -116,11 +116,11 @@ entry:
>> }
>>
>> ; floating-point store
>> -; EG-CHECK-LABEL: @store_f32
>> +; EG-CHECK-LABEL: store_f32
>> ; EG-CHECK: MEM_RAT_CACHELESS STORE_RAW T{{[0-9]+\.X, T[0-9]+\.X}}, 1
>> -; CM-CHECK-LABEL: @store_f32
>> +; CM-CHECK-LABEL: store_f32
>> ; CM-CHECK: MEM_RAT_CACHELESS STORE_DWORD T{{[0-9]+\.X, T[0-9]+\.X}}
>> -; SI-CHECK-LABEL: @store_f32
>> +; SI-CHECK-LABEL: store_f32
>> ; SI-CHECK: BUFFER_STORE_DWORD
>>
>> define void @store_f32(float addrspace(1)* %out, float %in) {
>> @@ -128,13 +128,13 @@ define void @store_f32(float addrspace(1)* %out, float %in) {
>> ret void
>> }
>>
>> -; EG-CHECK-LABEL: @store_v4i16
>> +; EG-CHECK-LABEL: store_v4i16
>> ; EG-CHECK: MEM_RAT MSKOR
>> ; EG-CHECK: MEM_RAT MSKOR
>> ; EG-CHECK: MEM_RAT MSKOR
>> ; EG-CHECK: MEM_RAT MSKOR
>> ; EG-CHECK-NOT: MEM_RAT MSKOR
>> -; SI-CHECK-LABEL: @store_v4i16
>> +; SI-CHECK-LABEL: store_v4i16
>> ; SI-CHECK: BUFFER_STORE_SHORT
>> ; SI-CHECK: BUFFER_STORE_SHORT
>> ; SI-CHECK: BUFFER_STORE_SHORT
>> @@ -148,11 +148,11 @@ entry:
>> }
>>
>> ; vec2 floating-point stores
>> -; EG-CHECK-LABEL: @store_v2f32
>> +; EG-CHECK-LABEL: store_v2f32
>> ; EG-CHECK: MEM_RAT_CACHELESS STORE_RAW
>> -; CM-CHECK-LABEL: @store_v2f32
>> +; CM-CHECK-LABEL: store_v2f32
>> ; CM-CHECK: MEM_RAT_CACHELESS STORE_DWORD
>> -; SI-CHECK-LABEL: @store_v2f32
>> +; SI-CHECK-LABEL: store_v2f32
>> ; SI-CHECK: BUFFER_STORE_DWORDX2
>>
>> define void @store_v2f32(<2 x float> addrspace(1)* %out, float %a, float %b) {
>> @@ -163,13 +163,13 @@ entry:
>> ret void
>> }
>>
>> -; EG-CHECK-LABEL: @store_v4i32
>> +; EG-CHECK-LABEL: store_v4i32
>> ; EG-CHECK: MEM_RAT_CACHELESS STORE_RAW
>> ; EG-CHECK-NOT: MEM_RAT_CACHELESS STORE_RAW
>> -; CM-CHECK-LABEL: @store_v4i32
>> +; CM-CHECK-LABEL: store_v4i32
>> ; CM-CHECK: MEM_RAT_CACHELESS STORE_DWORD
>> ; CM-CHECK-NOT: MEM_RAT_CACHELESS STORE_DWORD
>> -; SI-CHECK-LABEL: @store_v4i32
>> +; SI-CHECK-LABEL: store_v4i32
>> ; SI-CHECK: BUFFER_STORE_DWORDX4
>> define void @store_v4i32(<4 x i32> addrspace(1)* %out, <4 x i32> %in) {
>> entry:
>> @@ -177,7 +177,7 @@ entry:
>> ret void
>> }
>>
>> -; FUNC-LABEL: @store_i64_i8
>> +; FUNC-LABEL: store_i64_i8
>> ; EG-CHECK: MEM_RAT MSKOR
>> ; SI-CHECK: BUFFER_STORE_BYTE
>> define void @store_i64_i8(i8 addrspace(1)* %out, i64 %in) {
>> @@ -187,7 +187,7 @@ entry:
>> ret void
>> }
>>
>> -; FUNC-LABEL: @store_i64_i16
>> +; FUNC-LABEL: store_i64_i16
>> ; EG-CHECK: MEM_RAT MSKOR
>> ; SI-CHECK: BUFFER_STORE_SHORT
>> define void @store_i64_i16(i16 addrspace(1)* %out, i64 %in) {
>> @@ -201,7 +201,7 @@ entry:
>> ; Local Address Space
>> ;===------------------------------------------------------------------------===;
>>
>> -; FUNC-LABEL: @store_local_i1
>> +; FUNC-LABEL: store_local_i1
>> ; EG-CHECK: LDS_BYTE_WRITE
>> ; SI-CHECK: DS_WRITE_B8
>> define void @store_local_i1(i1 addrspace(3)* %out) {
>> @@ -210,29 +210,29 @@ entry:
>> ret void
>> }
>>
>> -; EG-CHECK-LABEL: @store_local_i8
>> +; EG-CHECK-LABEL: store_local_i8
>> ; EG-CHECK: LDS_BYTE_WRITE
>> -; SI-CHECK-LABEL: @store_local_i8
>> +; SI-CHECK-LABEL: store_local_i8
>> ; SI-CHECK: DS_WRITE_B8
>> define void @store_local_i8(i8 addrspace(3)* %out, i8 %in) {
>> store i8 %in, i8 addrspace(3)* %out
>> ret void
>> }
>>
>> -; EG-CHECK-LABEL: @store_local_i16
>> +; EG-CHECK-LABEL: store_local_i16
>> ; EG-CHECK: LDS_SHORT_WRITE
>> -; SI-CHECK-LABEL: @store_local_i16
>> +; SI-CHECK-LABEL: store_local_i16
>> ; SI-CHECK: DS_WRITE_B16
>> define void @store_local_i16(i16 addrspace(3)* %out, i16 %in) {
>> store i16 %in, i16 addrspace(3)* %out
>> ret void
>> }
>>
>> -; EG-CHECK-LABEL: @store_local_v2i16
>> +; EG-CHECK-LABEL: store_local_v2i16
>> ; EG-CHECK: LDS_WRITE
>> -; CM-CHECK-LABEL: @store_local_v2i16
>> +; CM-CHECK-LABEL: store_local_v2i16
>> ; CM-CHECK: LDS_WRITE
>> -; SI-CHECK-LABEL: @store_local_v2i16
>> +; SI-CHECK-LABEL: store_local_v2i16
>> ; SI-CHECK: DS_WRITE_B16
>> ; SI-CHECK: DS_WRITE_B16
>> define void @store_local_v2i16(<2 x i16> addrspace(3)* %out, <2 x i16> %in) {
>> @@ -241,11 +241,11 @@ entry:
>> ret void
>> }
>>
>> -; EG-CHECK-LABEL: @store_local_v4i8
>> +; EG-CHECK-LABEL: store_local_v4i8
>> ; EG-CHECK: LDS_WRITE
>> -; CM-CHECK-LABEL: @store_local_v4i8
>> +; CM-CHECK-LABEL: store_local_v4i8
>> ; CM-CHECK: LDS_WRITE
>> -; SI-CHECK-LABEL: @store_local_v4i8
>> +; SI-CHECK-LABEL: store_local_v4i8
>> ; SI-CHECK: DS_WRITE_B8
>> ; SI-CHECK: DS_WRITE_B8
>> ; SI-CHECK: DS_WRITE_B8
>> @@ -256,13 +256,13 @@ entry:
>> ret void
>> }
>>
>> -; EG-CHECK-LABEL: @store_local_v2i32
>> +; EG-CHECK-LABEL: store_local_v2i32
>> ; EG-CHECK: LDS_WRITE
>> ; EG-CHECK: LDS_WRITE
>> -; CM-CHECK-LABEL: @store_local_v2i32
>> +; CM-CHECK-LABEL: store_local_v2i32
>> ; CM-CHECK: LDS_WRITE
>> ; CM-CHECK: LDS_WRITE
>> -; SI-CHECK-LABEL: @store_local_v2i32
>> +; SI-CHECK-LABEL: store_local_v2i32
>> ; SI-CHECK: DS_WRITE_B64
>> define void @store_local_v2i32(<2 x i32> addrspace(3)* %out, <2 x i32> %in) {
>> entry:
>> @@ -270,17 +270,17 @@ entry:
>> ret void
>> }
>>
>> -; EG-CHECK-LABEL: @store_local_v4i32
>> +; EG-CHECK-LABEL: store_local_v4i32
>> ; EG-CHECK: LDS_WRITE
>> ; EG-CHECK: LDS_WRITE
>> ; EG-CHECK: LDS_WRITE
>> ; EG-CHECK: LDS_WRITE
>> -; CM-CHECK-LABEL: @store_local_v4i32
>> +; CM-CHECK-LABEL: store_local_v4i32
>> ; CM-CHECK: LDS_WRITE
>> ; CM-CHECK: LDS_WRITE
>> ; CM-CHECK: LDS_WRITE
>> ; CM-CHECK: LDS_WRITE
>> -; SI-CHECK-LABEL: @store_local_v4i32
>> +; SI-CHECK-LABEL: store_local_v4i32
>> ; SI-CHECK: DS_WRITE_B32
>> ; SI-CHECK: DS_WRITE_B32
>> ; SI-CHECK: DS_WRITE_B32
>> @@ -291,7 +291,7 @@ entry:
>> ret void
>> }
>>
>> -; FUNC-LABEL: @store_local_i64_i8
>> +; FUNC-LABEL: store_local_i64_i8
>> ; EG-CHECK: LDS_BYTE_WRITE
>> ; SI-CHECK: DS_WRITE_B8
>> define void @store_local_i64_i8(i8 addrspace(3)* %out, i64 %in) {
>> @@ -301,7 +301,7 @@ entry:
>> ret void
>> }
>>
>> -; FUNC-LABEL: @store_local_i64_i16
>> +; FUNC-LABEL: store_local_i64_i16
>> ; EG-CHECK: LDS_SHORT_WRITE
>> ; SI-CHECK: DS_WRITE_B16
>> define void @store_local_i64_i16(i16 addrspace(3)* %out, i64 %in) {
>> @@ -318,11 +318,11 @@ entry:
>> ; Evergreen / Northern Islands don't support 64-bit stores yet, so there should
>> ; be two 32-bit stores.
>>
>> -; EG-CHECK-LABEL: @vecload2
>> +; EG-CHECK-LABEL: vecload2
>> ; EG-CHECK: MEM_RAT_CACHELESS STORE_RAW
>> -; CM-CHECK-LABEL: @vecload2
>> +; CM-CHECK-LABEL: vecload2
>> ; CM-CHECK: MEM_RAT_CACHELESS STORE_DWORD
>> -; SI-CHECK-LABEL: @vecload2
>> +; SI-CHECK-LABEL: vecload2
>> ; SI-CHECK: BUFFER_STORE_DWORDX2
>> define void @vecload2(i32 addrspace(1)* nocapture %out, i32 addrspace(2)* nocapture %mem) #0 {
>> entry:
>> @@ -339,7 +339,7 @@ attributes #0 = { nounwind "less-precise-fpmad"="false" "no-frame-pointer-elim"=
>>
>> ; When i128 was a legal type this program generated cannot select errors:
>>
>> -; FUNC-LABEL: @i128-const-store
>> +; FUNC-LABEL: i128-const-store
>> ; FIXME: We should be able to to this with one store instruction
>> ; EG-CHECK: STORE_RAW
>> ; EG-CHECK: STORE_RAW
>> diff --git a/test/CodeGen/R600/store.r600.ll b/test/CodeGen/R600/store.r600.ll
>> index 00589a0..6a0f36a 100644
>> --- a/test/CodeGen/R600/store.r600.ll
>> +++ b/test/CodeGen/R600/store.r600.ll
>> @@ -3,7 +3,7 @@
>> ; XXX: Merge this test into store.ll once it is supported on SI
>>
>> ; v4i32 store
>> -; EG-CHECK: @store_v4i32
>> +; EG-CHECK: store_v4i32
>> ; EG-CHECK: MEM_RAT_CACHELESS STORE_RAW T{{[0-9]+\.XYZW, T[0-9]+\.X}}, 1
>>
>> define void @store_v4i32(<4 x i32> addrspace(1)* %out, <4 x i32> addrspace(1)* %in) {
>> @@ -13,7 +13,7 @@ define void @store_v4i32(<4 x i32> addrspace(1)* %out, <4 x i32> addrspace(1)* %
>> }
>>
>> ; v4f32 store
>> -; EG-CHECK: @store_v4f32
>> +; EG-CHECK: store_v4f32
>> ; EG-CHECK: MEM_RAT_CACHELESS STORE_RAW T{{[0-9]+\.XYZW, T[0-9]+\.X}}, 1
>> define void @store_v4f32(<4 x float> addrspace(1)* %out, <4 x float> addrspace(1)* %in) {
>> %1 = load <4 x float> addrspace(1) * %in
>> diff --git a/test/CodeGen/R600/structurize.ll b/test/CodeGen/R600/structurize.ll
>> index c2acd93..78dea8f 100644
>> --- a/test/CodeGen/R600/structurize.ll
>> +++ b/test/CodeGen/R600/structurize.ll
>> @@ -13,7 +13,7 @@
>> ;
>> ;
>>
>> -; CHECK-LABEL: @branch_into_diamond
>> +; CHECK-LABEL: branch_into_diamond
>> ; === entry block:
>> ; CHECK: ALU_PUSH_BEFORE
>> ; === Branch instruction (IF):
>> diff --git a/test/CodeGen/R600/structurize1.ll b/test/CodeGen/R600/structurize1.ll
>> index 8c10301..79d95d0 100644
>> --- a/test/CodeGen/R600/structurize1.ll
>> +++ b/test/CodeGen/R600/structurize1.ll
>> @@ -16,7 +16,7 @@
>> ; }
>> ; }
>>
>> -; CHECK-LABEL: @if_inside_loop
>> +; CHECK-LABEL: if_inside_loop
>> ; CHECK: LOOP_START_DX10
>> ; CHECK: END_LOOP
>> define void @if_inside_loop(i32 addrspace(1)* %out, i32 %a, i32 %b, i32 %c, i32 %d) {
>> diff --git a/test/CodeGen/R600/sub.ll b/test/CodeGen/R600/sub.ll
>> index a8196a0..3cf285c 100644
>> --- a/test/CodeGen/R600/sub.ll
>> +++ b/test/CodeGen/R600/sub.ll
>> @@ -3,7 +3,7 @@
>>
>> declare i32 @llvm.r600.read.tidig.x() readnone
>>
>> -;FUNC-LABEL: @test2
>> +;FUNC-LABEL: test2
>> ;EG: SUB_INT {{\** *}}T{{[0-9]+\.[XYZW], T[0-9]+\.[XYZW], T[0-9]+\.[XYZW]}}
>> ;EG: SUB_INT {{\** *}}T{{[0-9]+\.[XYZW], T[0-9]+\.[XYZW], T[0-9]+\.[XYZW]}}
>>
>> @@ -19,7 +19,7 @@ define void @test2(<2 x i32> addrspace(1)* %out, <2 x i32> addrspace(1)* %in) {
>> ret void
>> }
>>
>> -;FUNC-LABEL: @test4
>> +;FUNC-LABEL: test4
>> ;EG: SUB_INT {{\** *}}T{{[0-9]+\.[XYZW], T[0-9]+\.[XYZW], T[0-9]+\.[XYZW]}}
>> ;EG: SUB_INT {{\** *}}T{{[0-9]+\.[XYZW], T[0-9]+\.[XYZW], T[0-9]+\.[XYZW]}}
>> ;EG: SUB_INT {{\** *}}T{{[0-9]+\.[XYZW], T[0-9]+\.[XYZW], T[0-9]+\.[XYZW]}}
>> @@ -39,7 +39,7 @@ define void @test4(<4 x i32> addrspace(1)* %out, <4 x i32> addrspace(1)* %in) {
>> ret void
>> }
>>
>> -; FUNC-LABEL: @s_sub_i64:
>> +; FUNC-LABEL: s_sub_i64:
>> ; SI: S_SUB_U32
>> ; SI: S_SUBB_U32
>>
>> @@ -54,7 +54,7 @@ define void @s_sub_i64(i64 addrspace(1)* noalias %out, i64 %a, i64 %b) nounwind
>> ret void
>> }
>>
>> -; FUNC-LABEL: @v_sub_i64:
>> +; FUNC-LABEL: v_sub_i64:
>> ; SI: V_SUB_I32_e32
>> ; SI: V_SUBB_U32_e32
>>
>> diff --git a/test/CodeGen/R600/swizzle-export.ll b/test/CodeGen/R600/swizzle-export.ll
>> index 0a68f76..4cd6bf2 100644
>> --- a/test/CodeGen/R600/swizzle-export.ll
>> +++ b/test/CodeGen/R600/swizzle-export.ll
>> @@ -1,6 +1,6 @@
>> ; RUN: llc < %s -march=r600 -mcpu=redwood | FileCheck --check-prefix=EG-CHECK %s
>>
>> -;EG-CHECK: @main
>> +;EG-CHECK: main
>> ;EG-CHECK: EXPORT T{{[0-9]+}}.XYXX
>> ;EG-CHECK: EXPORT T{{[0-9]+}}.ZXXX
>> ;EG-CHECK: EXPORT T{{[0-9]+}}.XXWX
>> @@ -92,7 +92,7 @@ main_body:
>> ret void
>> }
>>
>> -; EG-CHECK: @main2
>> +; EG-CHECK: main2
>> ; EG-CHECK: T{{[0-9]+}}.XY__
>> ; EG-CHECK: T{{[0-9]+}}.ZXY0
>>
>> diff --git a/test/CodeGen/R600/trunc-store-i1.ll b/test/CodeGen/R600/trunc-store-i1.ll
>> index a3975c8..c85152f 100644
>> --- a/test/CodeGen/R600/trunc-store-i1.ll
>> +++ b/test/CodeGen/R600/trunc-store-i1.ll
>> @@ -1,7 +1,7 @@
>> ; RUN: llc -march=r600 -mcpu=SI -verify-machineinstrs< %s | FileCheck -check-prefix=SI %s
>>
>>
>> -; SI-LABEL: @global_truncstore_i32_to_i1
>> +; SI-LABEL: global_truncstore_i32_to_i1
>> ; SI: S_LOAD_DWORD [[LOAD:s[0-9]+]],
>> ; SI: S_AND_B32 [[SREG:s[0-9]+]], [[LOAD]], 1
>> ; SI: V_MOV_B32_e32 [[VREG:v[0-9]+]], [[SREG]]
>> @@ -12,7 +12,7 @@ define void @global_truncstore_i32_to_i1(i1 addrspace(1)* %out, i32 %val) nounwi
>> ret void
>> }
>>
>> -; SI-LABEL: @global_truncstore_i64_to_i1
>> +; SI-LABEL: global_truncstore_i64_to_i1
>> ; SI: BUFFER_STORE_BYTE
>> define void @global_truncstore_i64_to_i1(i1 addrspace(1)* %out, i64 %val) nounwind {
>> %trunc = trunc i64 %val to i1
>> @@ -20,7 +20,7 @@ define void @global_truncstore_i64_to_i1(i1 addrspace(1)* %out, i64 %val) nounwi
>> ret void
>> }
>>
>> -; SI-LABEL: @global_truncstore_i16_to_i1
>> +; SI-LABEL: global_truncstore_i16_to_i1
>> ; SI: S_LOAD_DWORD [[LOAD:s[0-9]+]],
>> ; SI: S_AND_B32 [[SREG:s[0-9]+]], [[LOAD]], 1
>> ; SI: V_MOV_B32_e32 [[VREG:v[0-9]+]], [[SREG]]
>> diff --git a/test/CodeGen/R600/trunc-vector-store-assertion-failure.ll b/test/CodeGen/R600/trunc-vector-store-assertion-failure.ll
>> index ec959c2..82b64bf 100644
>> --- a/test/CodeGen/R600/trunc-vector-store-assertion-failure.ll
>> +++ b/test/CodeGen/R600/trunc-vector-store-assertion-failure.ll
>> @@ -4,7 +4,7 @@
>> ; vector stores at the end of a basic block were not being added to the
>> ; LegalizedNodes list, which triggered an assertion failure.
>>
>> -; CHECK-LABEL: @test
>> +; CHECK-LABEL: test
>> ; CHECK: MEM_RAT_CACHELESS STORE_RAW
>> define void @test(<4 x i8> addrspace(1)* %out, i32 %cond, <4 x i8> %in) {
>> entry:
>> diff --git a/test/CodeGen/R600/trunc.ll b/test/CodeGen/R600/trunc.ll
>> index d3b191d..a84d42c 100644
>> --- a/test/CodeGen/R600/trunc.ll
>> +++ b/test/CodeGen/R600/trunc.ll
>> @@ -2,12 +2,12 @@
>> ; RUN: llc -march=r600 -mcpu=cypress < %s | FileCheck -check-prefix=EG %s
>>
>> define void @trunc_i64_to_i32_store(i32 addrspace(1)* %out, i64 %in) {
>> -; SI-LABEL: @trunc_i64_to_i32_store
>> +; SI-LABEL: trunc_i64_to_i32_store
>> ; SI: S_LOAD_DWORD [[SLOAD:s[0-9]+]], s[0:1], 0xb
>> ; SI: V_MOV_B32_e32 [[VLOAD:v[0-9]+]], [[SLOAD]]
>> ; SI: BUFFER_STORE_DWORD [[VLOAD]]
>>
>> -; EG-LABEL: @trunc_i64_to_i32_store
>> +; EG-LABEL: trunc_i64_to_i32_store
>> ; EG: MEM_RAT_CACHELESS STORE_RAW T0.X, T1.X, 1
>> ; EG: LSHR
>> ; EG-NEXT: 2(
>> @@ -16,7 +16,7 @@ define void @trunc_i64_to_i32_store(i32 addrspace(1)* %out, i64 %in) {
>> ret void
>> }
>>
>> -; SI-LABEL: @trunc_load_shl_i64:
>> +; SI-LABEL: trunc_load_shl_i64:
>> ; SI-DAG: S_LOAD_DWORDX2
>> ; SI-DAG: S_LOAD_DWORD [[SREG:s[0-9]+]],
>> ; SI: S_LSHL_B32 [[SHL:s[0-9]+]], [[SREG]], 2
>> @@ -29,7 +29,7 @@ define void @trunc_load_shl_i64(i32 addrspace(1)* %out, i64 %a) {
>> ret void
>> }
>>
>> -; SI-LABEL: @trunc_shl_i64:
>> +; SI-LABEL: trunc_shl_i64:
>> ; SI: S_LOAD_DWORDX2 s{{\[}}[[LO_SREG:[0-9]+]]:{{[0-9]+\]}}, s{{\[[0-9]+:[0-9]+\]}}, 0xd
>> ; SI: S_LSHL_B64 s{{\[}}[[LO_SHL:[0-9]+]]:{{[0-9]+\]}}, s{{\[}}[[LO_SREG]]:{{[0-9]+\]}}, 2
>> ; SI: S_ADD_U32 s[[LO_SREG2:[0-9]+]], s[[LO_SHL]],
>> @@ -45,7 +45,7 @@ define void @trunc_shl_i64(i64 addrspace(1)* %out2, i32 addrspace(1)* %out, i64
>> ret void
>> }
>>
>> -; SI-LABEL: @trunc_i32_to_i1:
>> +; SI-LABEL: trunc_i32_to_i1:
>> ; SI: V_AND_B32_e32 v{{[0-9]+}}, 1, v{{[0-9]+}}
>> ; SI: V_CMP_EQ_I32
>> define void @trunc_i32_to_i1(i32 addrspace(1)* %out, i32 addrspace(1)* %ptr) {
>> @@ -56,7 +56,7 @@ define void @trunc_i32_to_i1(i32 addrspace(1)* %out, i32 addrspace(1)* %ptr) {
>> ret void
>> }
>>
>> -; SI-LABEL: @sgpr_trunc_i32_to_i1:
>> +; SI-LABEL: sgpr_trunc_i32_to_i1:
>> ; SI: V_AND_B32_e64 v{{[0-9]+}}, 1, s{{[0-9]+}}
>> ; SI: V_CMP_EQ_I32
>> define void @sgpr_trunc_i32_to_i1(i32 addrspace(1)* %out, i32 %a) {
>> diff --git a/test/CodeGen/R600/uaddo.ll b/test/CodeGen/R600/uaddo.ll
>> index 0b854b5..24c9000 100644
>> --- a/test/CodeGen/R600/uaddo.ll
>> +++ b/test/CodeGen/R600/uaddo.ll
>> @@ -4,7 +4,7 @@
>> declare { i32, i1 } @llvm.uadd.with.overflow.i32(i32, i32) nounwind readnone
>> declare { i64, i1 } @llvm.uadd.with.overflow.i64(i64, i64) nounwind readnone
>>
>> -; FUNC-LABEL: @uaddo_i64_zext
>> +; FUNC-LABEL: uaddo_i64_zext
>> ; SI: ADD
>> ; SI: ADDC
>> ; SI: ADDC
>> @@ -18,7 +18,7 @@ define void @uaddo_i64_zext(i64 addrspace(1)* %out, i64 %a, i64 %b) nounwind {
>> ret void
>> }
>>
>> -; FUNC-LABEL: @s_uaddo_i32
>> +; FUNC-LABEL: s_uaddo_i32
>> ; SI: S_ADD_I32
>> define void @s_uaddo_i32(i32 addrspace(1)* %out, i1 addrspace(1)* %carryout, i32 %a, i32 %b) nounwind {
>> %uadd = call { i32, i1 } @llvm.uadd.with.overflow.i32(i32 %a, i32 %b) nounwind
>> @@ -29,7 +29,7 @@ define void @s_uaddo_i32(i32 addrspace(1)* %out, i1 addrspace(1)* %carryout, i32
>> ret void
>> }
>>
>> -; FUNC-LABEL: @v_uaddo_i32
>> +; FUNC-LABEL: v_uaddo_i32
>> ; SI: V_ADD_I32
>> define void @v_uaddo_i32(i32 addrspace(1)* %out, i1 addrspace(1)* %carryout, i32 addrspace(1)* %aptr, i32 addrspace(1)* %bptr) nounwind {
>> %a = load i32 addrspace(1)* %aptr, align 4
>> @@ -42,7 +42,7 @@ define void @v_uaddo_i32(i32 addrspace(1)* %out, i1 addrspace(1)* %carryout, i32
>> ret void
>> }
>>
>> -; FUNC-LABEL: @s_uaddo_i64
>> +; FUNC-LABEL: s_uaddo_i64
>> ; SI: S_ADD_U32
>> ; SI: S_ADDC_U32
>> define void @s_uaddo_i64(i64 addrspace(1)* %out, i1 addrspace(1)* %carryout, i64 %a, i64 %b) nounwind {
>> @@ -54,7 +54,7 @@ define void @s_uaddo_i64(i64 addrspace(1)* %out, i1 addrspace(1)* %carryout, i64
>> ret void
>> }
>>
>> -; FUNC-LABEL: @v_uaddo_i64
>> +; FUNC-LABEL: v_uaddo_i64
>> ; SI: V_ADD_I32
>> ; SI: V_ADDC_U32
>> define void @v_uaddo_i64(i64 addrspace(1)* %out, i1 addrspace(1)* %carryout, i64 addrspace(1)* %aptr, i64 addrspace(1)* %bptr) nounwind {
>> diff --git a/test/CodeGen/R600/udiv.ll b/test/CodeGen/R600/udiv.ll
>> index 5371321..9285e69 100644
>> --- a/test/CodeGen/R600/udiv.ll
>> +++ b/test/CodeGen/R600/udiv.ll
>> @@ -1,7 +1,7 @@
>> ;RUN: llc < %s -march=r600 -mcpu=redwood | FileCheck --check-prefix=EG-CHECK %s
>> ;RUN: llc < %s -march=r600 -mcpu=verde -verify-machineinstrs | FileCheck --check-prefix=SI-CHECK %s
>>
>> -;EG-CHECK-LABEL: @test
>> +;EG-CHECK-LABEL: test
>> ;EG-CHECK-NOT: SETGE_INT
>> ;EG-CHECK: CF_END
>>
>> @@ -18,9 +18,9 @@ define void @test(i32 addrspace(1)* %out, i32 addrspace(1)* %in) {
>> ;The goal of this test is to make sure the ISel doesn't fail when it gets
>> ;a v4i32 udiv
>>
>> -;EG-CHECK-LABEL: @test2
>> +;EG-CHECK-LABEL: test2
>> ;EG-CHECK: CF_END
>> -;SI-CHECK-LABEL: @test2
>> +;SI-CHECK-LABEL: test2
>> ;SI-CHECK: S_ENDPGM
>>
>> define void @test2(<2 x i32> addrspace(1)* %out, <2 x i32> addrspace(1)* %in) {
>> @@ -32,9 +32,9 @@ define void @test2(<2 x i32> addrspace(1)* %out, <2 x i32> addrspace(1)* %in) {
>> ret void
>> }
>>
>> -;EG-CHECK-LABEL: @test4
>> +;EG-CHECK-LABEL: test4
>> ;EG-CHECK: CF_END
>> -;SI-CHECK-LABEL: @test4
>> +;SI-CHECK-LABEL: test4
>> ;SI-CHECK: S_ENDPGM
>>
>> define void @test4(<4 x i32> addrspace(1)* %out, <4 x i32> addrspace(1)* %in) {
>> diff --git a/test/CodeGen/R600/udivrem.ll b/test/CodeGen/R600/udivrem.ll
>> index 5f5753a..fcc6b88 100644
>> --- a/test/CodeGen/R600/udivrem.ll
>> +++ b/test/CodeGen/R600/udivrem.ll
>> @@ -1,7 +1,7 @@
>> ; RUN: llc -march=r600 -mcpu=SI -verify-machineinstrs < %s | FileCheck --check-prefix=SI --check-prefix=FUNC %s
>> ; RUN: llc -march=r600 -mcpu=redwood < %s | FileCheck --check-prefix=EG --check-prefix=FUNC %s
>>
>> -; FUNC-LABEL: @test_udivrem
>> +; FUNC-LABEL: test_udivrem
>> ; EG: RECIP_UINT
>> ; EG-DAG: MULHI
>> ; EG-DAG: MULLO_INT
>> @@ -58,7 +58,7 @@ define void @test_udivrem(i32 addrspace(1)* %out, i32 %x, i32 %y) {
>> ret void
>> }
>>
>> -; FUNC-LABEL: @test_udivrem_v2
>> +; FUNC-LABEL: test_udivrem_v2
>> ; EG-DAG: RECIP_UINT
>> ; EG-DAG: MULHI
>> ; EG-DAG: MULLO_INT
>> @@ -162,7 +162,7 @@ define void @test_udivrem_v2(<2 x i32> addrspace(1)* %out, <2 x i32> %x, <2 x i3
>> }
>>
>>
>> -; FUNC-LABEL: @test_udivrem_v4
>> +; FUNC-LABEL: test_udivrem_v4
>> ; EG-DAG: RECIP_UINT
>> ; EG-DAG: MULHI
>> ; EG-DAG: MULLO_INT
>> diff --git a/test/CodeGen/R600/udivrem24.ll b/test/CodeGen/R600/udivrem24.ll
>> index 219c662..2a4e62e 100644
>> --- a/test/CodeGen/R600/udivrem24.ll
>> +++ b/test/CodeGen/R600/udivrem24.ll
>> @@ -1,7 +1,7 @@
>> ; RUN: llc -march=r600 -mcpu=SI < %s | FileCheck -check-prefix=SI -check-prefix=FUNC %s
>> ; RUN: llc -march=r600 -mcpu=redwood < %s | FileCheck -check-prefix=EG -check-prefix=FUNC %s
>>
>> -; FUNC-LABEL: @udiv24_i8
>> +; FUNC-LABEL: udiv24_i8
>> ; SI: V_CVT_F32_UBYTE
>> ; SI: V_CVT_F32_UBYTE
>> ; SI: V_RCP_F32
>> @@ -20,7 +20,7 @@ define void @udiv24_i8(i8 addrspace(1)* %out, i8 addrspace(1)* %in) {
>> ret void
>> }
>>
>> -; FUNC-LABEL: @udiv24_i16
>> +; FUNC-LABEL: udiv24_i16
>> ; SI: V_CVT_F32_U32
>> ; SI: V_CVT_F32_U32
>> ; SI: V_RCP_F32
>> @@ -39,7 +39,7 @@ define void @udiv24_i16(i16 addrspace(1)* %out, i16 addrspace(1)* %in) {
>> ret void
>> }
>>
>> -; FUNC-LABEL: @udiv24_i32
>> +; FUNC-LABEL: udiv24_i32
>> ; SI: V_CVT_F32_U32
>> ; SI-DAG: V_CVT_F32_U32
>> ; SI-DAG: V_RCP_F32
>> @@ -62,7 +62,7 @@ define void @udiv24_i32(i32 addrspace(1)* %out, i32 addrspace(1)* %in) {
>> ret void
>> }
>>
>> -; FUNC-LABEL: @udiv25_i32
>> +; FUNC-LABEL: udiv25_i32
>> ; RCP_IFLAG is for URECIP in the full 32b alg
>> ; SI: V_RCP_IFLAG
>> ; SI-NOT: V_RCP_F32
>> @@ -82,7 +82,7 @@ define void @udiv25_i32(i32 addrspace(1)* %out, i32 addrspace(1)* %in) {
>> ret void
>> }
>>
>> -; FUNC-LABEL: @test_no_udiv24_i32_1
>> +; FUNC-LABEL: test_no_udiv24_i32_1
>> ; RCP_IFLAG is for URECIP in the full 32b alg
>> ; SI: V_RCP_IFLAG
>> ; SI-NOT: V_RCP_F32
>> @@ -102,7 +102,7 @@ define void @test_no_udiv24_i32_1(i32 addrspace(1)* %out, i32 addrspace(1)* %in)
>> ret void
>> }
>>
>> -; FUNC-LABEL: @test_no_udiv24_i32_2
>> +; FUNC-LABEL: test_no_udiv24_i32_2
>> ; RCP_IFLAG is for URECIP in the full 32b alg
>> ; SI: V_RCP_IFLAG
>> ; SI-NOT: V_RCP_F32
>> @@ -122,7 +122,7 @@ define void @test_no_udiv24_i32_2(i32 addrspace(1)* %out, i32 addrspace(1)* %in)
>> ret void
>> }
>>
>> -; FUNC-LABEL: @urem24_i8
>> +; FUNC-LABEL: urem24_i8
>> ; SI: V_CVT_F32_UBYTE
>> ; SI: V_CVT_F32_UBYTE
>> ; SI: V_RCP_F32
>> @@ -141,7 +141,7 @@ define void @urem24_i8(i8 addrspace(1)* %out, i8 addrspace(1)* %in) {
>> ret void
>> }
>>
>> -; FUNC-LABEL: @urem24_i16
>> +; FUNC-LABEL: urem24_i16
>> ; SI: V_CVT_F32_U32
>> ; SI: V_CVT_F32_U32
>> ; SI: V_RCP_F32
>> @@ -160,7 +160,7 @@ define void @urem24_i16(i16 addrspace(1)* %out, i16 addrspace(1)* %in) {
>> ret void
>> }
>>
>> -; FUNC-LABEL: @urem24_i32
>> +; FUNC-LABEL: urem24_i32
>> ; SI: V_CVT_F32_U32
>> ; SI: V_CVT_F32_U32
>> ; SI: V_RCP_F32
>> @@ -183,7 +183,7 @@ define void @urem24_i32(i32 addrspace(1)* %out, i32 addrspace(1)* %in) {
>> ret void
>> }
>>
>> -; FUNC-LABEL: @urem25_i32
>> +; FUNC-LABEL: urem25_i32
>> ; RCP_IFLAG is for URECIP in the full 32b alg
>> ; SI: V_RCP_IFLAG
>> ; SI-NOT: V_RCP_F32
>> @@ -203,7 +203,7 @@ define void @urem25_i32(i32 addrspace(1)* %out, i32 addrspace(1)* %in) {
>> ret void
>> }
>>
>> -; FUNC-LABEL: @test_no_urem24_i32_1
>> +; FUNC-LABEL: test_no_urem24_i32_1
>> ; RCP_IFLAG is for URECIP in the full 32b alg
>> ; SI: V_RCP_IFLAG
>> ; SI-NOT: V_RCP_F32
>> @@ -223,7 +223,7 @@ define void @test_no_urem24_i32_1(i32 addrspace(1)* %out, i32 addrspace(1)* %in)
>> ret void
>> }
>>
>> -; FUNC-LABEL: @test_no_urem24_i32_2
>> +; FUNC-LABEL: test_no_urem24_i32_2
>> ; RCP_IFLAG is for URECIP in the full 32b alg
>> ; SI: V_RCP_IFLAG
>> ; SI-NOT: V_RCP_F32
>> diff --git a/test/CodeGen/R600/udivrem64.ll b/test/CodeGen/R600/udivrem64.ll
>> index a71315a..e2aac46 100644
>> --- a/test/CodeGen/R600/udivrem64.ll
>> +++ b/test/CodeGen/R600/udivrem64.ll
>> @@ -1,7 +1,7 @@
>> ;XUN: llc < %s -march=r600 -mcpu=SI -verify-machineinstrs| FileCheck --check-prefix=SI --check-prefix=FUNC %s
>> ;RUN: llc < %s -march=r600 -mcpu=redwood | FileCheck --check-prefix=EG --check-prefix=FUNC %s
>>
>> -;FUNC-LABEL: @test_udiv
>> +;FUNC-LABEL: test_udiv
>> ;EG: RECIP_UINT
>> ;EG: LSHL {{.*}}, 1,
>> ;EG: BFE_UINT
>> @@ -41,7 +41,7 @@ define void @test_udiv(i64 addrspace(1)* %out, i64 %x, i64 %y) {
>> ret void
>> }
>>
>> -;FUNC-LABEL: @test_urem
>> +;FUNC-LABEL: test_urem
>> ;EG: RECIP_UINT
>> ;EG: BFE_UINT
>> ;EG: BFE_UINT
>> diff --git a/test/CodeGen/R600/uint_to_fp.f64.ll b/test/CodeGen/R600/uint_to_fp.f64.ll
>> index 9a41796..cf01395 100644
>> --- a/test/CodeGen/R600/uint_to_fp.f64.ll
>> +++ b/test/CodeGen/R600/uint_to_fp.f64.ll
>> @@ -1,6 +1,6 @@
>> ; RUN: llc -march=r600 -mcpu=SI -verify-machineinstrs < %s | FileCheck -check-prefix=SI %s
>>
>> -; SI-LABEL: @uint_to_fp_f64_i32
>> +; SI-LABEL: uint_to_fp_f64_i32
>> ; SI: V_CVT_F64_U32_e32
>> ; SI: S_ENDPGM
>> define void @uint_to_fp_f64_i32(double addrspace(1)* %out, i32 %in) {
>> @@ -9,7 +9,7 @@ define void @uint_to_fp_f64_i32(double addrspace(1)* %out, i32 %in) {
>> ret void
>> }
>>
>> -; SI-LABEL: @uint_to_fp_i1_f64:
>> +; SI-LABEL: uint_to_fp_i1_f64:
>> ; SI: V_CMP_EQ_I32_e64 [[CMP:s\[[0-9]+:[0-9]\]]],
>> ; FIXME: We should the VGPR sources for V_CNDMASK are copied from SGPRs,
>> ; we should be able to fold the SGPRs into the V_CNDMASK instructions.
>> @@ -24,7 +24,7 @@ define void @uint_to_fp_i1_f64(double addrspace(1)* %out, i32 %in) {
>> ret void
>> }
>>
>> -; SI-LABEL: @uint_to_fp_i1_f64_load:
>> +; SI-LABEL: uint_to_fp_i1_f64_load:
>> ; SI: V_CNDMASK_B32_e64 [[IRESULT:v[0-9]]], 0, 1
>> ; SI-NEXT: V_CVT_F64_U32_e32 [[RESULT:v\[[0-9]+:[0-9]\]]], [[IRESULT]]
>> ; SI: BUFFER_STORE_DWORDX2 [[RESULT]]
>> diff --git a/test/CodeGen/R600/uint_to_fp.ll b/test/CodeGen/R600/uint_to_fp.ll
>> index 9d4dacf..6665bdd 100644
>> --- a/test/CodeGen/R600/uint_to_fp.ll
>> +++ b/test/CodeGen/R600/uint_to_fp.ll
>> @@ -1,7 +1,7 @@
>> ; RUN: llc -march=r600 -mcpu=SI -verify-machineinstrs < %s | FileCheck -check-prefix=SI -check-prefix=FUNC %s
>> ; RUN: llc -march=r600 -mcpu=redwood < %s | FileCheck -check-prefix=R600 -check-prefix=FUNC %s
>>
>> -; FUNC-LABEL: @uint_to_fp_v2i32
>> +; FUNC-LABEL: uint_to_fp_v2i32
>> ; R600-DAG: UINT_TO_FLT * T{{[0-9]+\.[XYZW]}}, KC0[2].W
>> ; R600-DAG: UINT_TO_FLT * T{{[0-9]+\.[XYZW]}}, KC0[3].X
>>
>> @@ -14,7 +14,7 @@ define void @uint_to_fp_v2i32(<2 x float> addrspace(1)* %out, <2 x i32> %in) {
>> ret void
>> }
>>
>> -; FUNC-LABEL: @uint_to_fp_v4i32
>> +; FUNC-LABEL: uint_to_fp_v4i32
>> ; R600: UINT_TO_FLT * T{{[0-9]+\.[XYZW], T[0-9]+\.[XYZW]}}
>> ; R600: UINT_TO_FLT * T{{[0-9]+\.[XYZW], T[0-9]+\.[XYZW]}}
>> ; R600: UINT_TO_FLT * T{{[0-9]+\.[XYZW], T[0-9]+\.[XYZW]}}
>> @@ -32,7 +32,7 @@ define void @uint_to_fp_v4i32(<4 x float> addrspace(1)* %out, <4 x i32> addrspac
>> ret void
>> }
>>
>> -; FUNC-LABEL: @uint_to_fp_i64_f32
>> +; FUNC-LABEL: uint_to_fp_i64_f32
>> ; R600: UINT_TO_FLT
>> ; R600: UINT_TO_FLT
>> ; R600: MULADD_IEEE
>> @@ -47,7 +47,7 @@ entry:
>> ret void
>> }
>>
>> -; FUNC-LABEL: @uint_to_fp_i1_f32:
>> +; FUNC-LABEL: uint_to_fp_i1_f32:
>> ; SI: V_CMP_EQ_I32_e64 [[CMP:s\[[0-9]+:[0-9]\]]],
>> ; SI-NEXT: V_CNDMASK_B32_e64 [[RESULT:v[0-9]+]], 0, 1.0, [[CMP]]
>> ; SI: BUFFER_STORE_DWORD [[RESULT]],
>> @@ -59,7 +59,7 @@ define void @uint_to_fp_i1_f32(float addrspace(1)* %out, i32 %in) {
>> ret void
>> }
>>
>> -; FUNC-LABEL: @uint_to_fp_i1_f32_load:
>> +; FUNC-LABEL: uint_to_fp_i1_f32_load:
>> ; SI: V_CNDMASK_B32_e64 [[RESULT:v[0-9]+]], 0, 1.0
>> ; SI: BUFFER_STORE_DWORD [[RESULT]],
>> ; SI: S_ENDPGM
>> diff --git a/test/CodeGen/R600/unaligned-load-store.ll b/test/CodeGen/R600/unaligned-load-store.ll
>> index cca7df5..104f139 100644
>> --- a/test/CodeGen/R600/unaligned-load-store.ll
>> +++ b/test/CodeGen/R600/unaligned-load-store.ll
>> @@ -1,7 +1,7 @@
>> ; RUN: llc -march=r600 -mcpu=SI -verify-machineinstrs< %s | FileCheck -check-prefix=SI %s
>>
>> ; FIXME: This is probably wrong. This probably needs to expand to 8-bit reads and writes.
>> -; SI-LABEL: @unaligned_load_store_i32:
>> +; SI-LABEL: unaligned_load_store_i32:
>> ; SI: DS_READ_U16
>> ; SI: DS_READ_U16
>> ; SI: DS_WRITE_B32
>> @@ -12,7 +12,7 @@ define void @unaligned_load_store_i32(i32 addrspace(3)* %p, i32 addrspace(3)* %r
>> ret void
>> }
>>
>> -; SI-LABEL: @unaligned_load_store_v4i32:
>> +; SI-LABEL: unaligned_load_store_v4i32:
>> ; SI: DS_READ_U16
>> ; SI: DS_READ_U16
>> ; SI: DS_READ_U16
>> @@ -32,7 +32,7 @@ define void @unaligned_load_store_v4i32(<4 x i32> addrspace(3)* %p, <4 x i32> ad
>> ret void
>> }
>>
>> -; SI-LABEL: @load_lds_i64_align_4
>> +; SI-LABEL: load_lds_i64_align_4
>> ; SI: DS_READ2_B32
>> ; SI: S_ENDPGM
>> define void @load_lds_i64_align_4(i64 addrspace(1)* nocapture %out, i64 addrspace(3)* %in) #0 {
>> @@ -41,7 +41,7 @@ define void @load_lds_i64_align_4(i64 addrspace(1)* nocapture %out, i64 addrspac
>> ret void
>> }
>>
>> -; SI-LABEL: @load_lds_i64_align_4_with_offset
>> +; SI-LABEL: load_lds_i64_align_4_with_offset
>> ; SI: DS_READ2_B32 v[{{[0-9]+}}:{{[0-9]+}}], v{{[0-9]}}, 0x8, 0x9
>> ; SI: S_ENDPGM
>> define void @load_lds_i64_align_4_with_offset(i64 addrspace(1)* nocapture %out, i64 addrspace(3)* %in) #0 {
>> @@ -51,7 +51,7 @@ define void @load_lds_i64_align_4_with_offset(i64 addrspace(1)* nocapture %out,
>> ret void
>> }
>>
>> -; SI-LABEL: @load_lds_i64_align_4_with_split_offset
>> +; SI-LABEL: load_lds_i64_align_4_with_split_offset
>> ; The tests for the case where the lo offset is 8-bits, but the hi offset is 9-bits
>> ; SI: DS_READ2_B32 v[{{[0-9]+}}:{{[0-9]+}}], v{{[0-9]}}, 0x0, 0x1
>> ; SI: S_ENDPGM
>> @@ -71,7 +71,7 @@ define void @load_lds_i64_align_4_with_split_offset(i64 addrspace(1)* nocapture
>> ; ret void
>> ; }
>>
>> -; SI-LABEL: @store_lds_i64_align_4
>> +; SI-LABEL: store_lds_i64_align_4
>> ; SI: DS_WRITE2_B32
>> ; SI: S_ENDPGM
>> define void @store_lds_i64_align_4(i64 addrspace(3)* %out, i64 %val) #0 {
>> @@ -79,7 +79,7 @@ define void @store_lds_i64_align_4(i64 addrspace(3)* %out, i64 %val) #0 {
>> ret void
>> }
>>
>> -; SI-LABEL: @store_lds_i64_align_4_with_offset
>> +; SI-LABEL: store_lds_i64_align_4_with_offset
>> ; SI: DS_WRITE2_B32 v{{[0-9]+}}, v{{[0-9]+}}, v{{[0-9]+}}, 0x8, 0x9
>> ; SI: S_ENDPGM
>> define void @store_lds_i64_align_4_with_offset(i64 addrspace(3)* %out) #0 {
>> @@ -88,7 +88,7 @@ define void @store_lds_i64_align_4_with_offset(i64 addrspace(3)* %out) #0 {
>> ret void
>> }
>>
>> -; SI-LABEL: @store_lds_i64_align_4_with_split_offset
>> +; SI-LABEL: store_lds_i64_align_4_with_split_offset
>> ; The tests for the case where the lo offset is 8-bits, but the hi offset is 9-bits
>> ; SI: DS_WRITE2_B32 v{{[0-9]+}}, v{{[0-9]+}}, v{{[0-9]}}, 0x0, 0x1
>> ; SI: S_ENDPGM
>> diff --git a/test/CodeGen/R600/unhandled-loop-condition-assertion.ll b/test/CodeGen/R600/unhandled-loop-condition-assertion.ll
>> index e4129c5..f9be782 100644
>> --- a/test/CodeGen/R600/unhandled-loop-condition-assertion.ll
>> +++ b/test/CodeGen/R600/unhandled-loop-condition-assertion.ll
>> @@ -5,7 +5,7 @@
>>
>> ; SI hits an assertion at -O0, evergreen hits a not implemented unreachable.
>>
>> -; COMMON-LABEL: @branch_true:
>> +; COMMON-LABEL: branch_true:
>> define void @branch_true(i8 addrspace(1)* nocapture %main, i32 %main_stride) #0 {
>> entry:
>> br i1 true, label %for.end, label %for.body.lr.ph
>> @@ -39,7 +39,7 @@ for.end: ; preds = %for.body, %entry
>> ret void
>> }
>>
>> -; COMMON-LABEL: @branch_false:
>> +; COMMON-LABEL: branch_false:
>> ; SI: .text
>> ; SI-NEXT: S_ENDPGM
>> define void @branch_false(i8 addrspace(1)* nocapture %main, i32 %main_stride) #0 {
>> @@ -75,7 +75,7 @@ for.end: ; preds = %for.body, %entry
>> ret void
>> }
>>
>> -; COMMON-LABEL: @branch_undef:
>> +; COMMON-LABEL: branch_undef:
>> ; SI: .text
>> ; SI-NEXT: S_ENDPGM
>> define void @branch_undef(i8 addrspace(1)* nocapture %main, i32 %main_stride) #0 {
>> diff --git a/test/CodeGen/R600/unroll.ll b/test/CodeGen/R600/unroll.ll
>> index e0035ea..39ecc42 100644
>> --- a/test/CodeGen/R600/unroll.ll
>> +++ b/test/CodeGen/R600/unroll.ll
>> @@ -7,7 +7,7 @@ target triple = "r600--"
>> ; private memory. We want to make sure these kinds of loops are always
>> ; unrolled, because private memory is slow.
>>
>> -; CHECK-LABEL: @test
>> +; CHECK-LABEL: test
>> ; CHECK-NOT: alloca
>> ; CHECK: store i32 5, i32 addrspace(1)* %out
>> define void @test(i32 addrspace(1)* %out) {
>> diff --git a/test/CodeGen/R600/unsupported-cc.ll b/test/CodeGen/R600/unsupported-cc.ll
>> index f986a02..dd015a3 100644
>> --- a/test/CodeGen/R600/unsupported-cc.ll
>> +++ b/test/CodeGen/R600/unsupported-cc.ll
>> @@ -2,7 +2,7 @@
>>
>> ; These tests are for condition codes that are not supported by the hardware
>>
>> -; CHECK-LABEL: @slt
>> +; CHECK-LABEL: slt
>> ; CHECK: SETGT_INT {{\** *}}T{{[0-9]+\.[XYZW]}}, literal.x, KC0[2].Z
>> ; CHECK-NEXT: LSHR
>> ; CHECK-NEXT: 5(7.006492e-45)
>> @@ -14,7 +14,7 @@ entry:
>> ret void
>> }
>>
>> -; CHECK-LABEL: @ult_i32
>> +; CHECK-LABEL: ult_i32
>> ; CHECK: SETGT_UINT {{\** *}}T{{[0-9]+\.[XYZW]}}, literal.x, KC0[2].Z
>> ; CHECK-NEXT: LSHR
>> ; CHECK-NEXT: 5(7.006492e-45)
>> @@ -26,7 +26,7 @@ entry:
>> ret void
>> }
>>
>> -; CHECK-LABEL: @ult_float
>> +; CHECK-LABEL: ult_float
>> ; CHECK: SETGE * T{{[0-9]}}.[[CHAN:[XYZW]]], KC0[2].Z, literal.x
>> ; CHECK-NEXT: 1084227584(5.000000e+00)
>> ; CHECK-NEXT: SETE T{{[0-9]\.[XYZW]}}, PV.[[CHAN]], 0.0
>> @@ -39,7 +39,7 @@ entry:
>> ret void
>> }
>>
>> -; CHECK-LABEL: @ult_float_native
>> +; CHECK-LABEL: ult_float_native
>> ; CHECK: SETGE T{{[0-9]\.[XYZW]}}, KC0[2].Z, literal.x
>> ; CHECK-NEXT: LSHR *
>> ; CHECK-NEXT: 1084227584(5.000000e+00)
>> @@ -51,7 +51,7 @@ entry:
>> ret void
>> }
>>
>> -; CHECK-LABEL: @olt
>> +; CHECK-LABEL: olt
>> ; CHECK: SETGT T{{[0-9]+\.[XYZW]}}, literal.x, KC0[2].Z
>> ; CHECK-NEXT: LSHR *
>> ; CHECK-NEXT: 1084227584(5.000000e+00)
>> @@ -63,7 +63,7 @@ entry:
>> ret void
>> }
>>
>> -; CHECK-LABEL: @sle
>> +; CHECK-LABEL: sle
>> ; CHECK: SETGT_INT {{\** *}}T{{[0-9]+\.[XYZW]}}, literal.x, KC0[2].Z
>> ; CHECK-NEXT: LSHR
>> ; CHECK-NEXT: 6(8.407791e-45)
>> @@ -75,7 +75,7 @@ entry:
>> ret void
>> }
>>
>> -; CHECK-LABEL: @ule_i32
>> +; CHECK-LABEL: ule_i32
>> ; CHECK: SETGT_UINT {{\** *}}T{{[0-9]+\.[XYZW]}}, literal.x, KC0[2].Z
>> ; CHECK-NEXT: LSHR
>> ; CHECK-NEXT: 6(8.407791e-45)
>> @@ -87,7 +87,7 @@ entry:
>> ret void
>> }
>>
>> -; CHECK-LABEL: @ule_float
>> +; CHECK-LABEL: ule_float
>> ; CHECK: SETGT * T{{[0-9]}}.[[CHAN:[XYZW]]], KC0[2].Z, literal.x
>> ; CHECK-NEXT: 1084227584(5.000000e+00)
>> ; CHECK-NEXT: SETE T{{[0-9]\.[XYZW]}}, PV.[[CHAN]], 0.0
>> @@ -100,7 +100,7 @@ entry:
>> ret void
>> }
>>
>> -; CHECK-LABEL: @ule_float_native
>> +; CHECK-LABEL: ule_float_native
>> ; CHECK: SETGT T{{[0-9]\.[XYZW]}}, KC0[2].Z, literal.x
>> ; CHECK-NEXT: LSHR *
>> ; CHECK-NEXT: 1084227584(5.000000e+00)
>> @@ -112,7 +112,7 @@ entry:
>> ret void
>> }
>>
>> -; CHECK-LABEL: @ole
>> +; CHECK-LABEL: ole
>> ; CHECK: SETGE T{{[0-9]\.[XYZW]}}, literal.x, KC0[2].Z
>> ; CHECK-NEXT: LSHR *
>> ; CHECK-NEXT:1084227584(5.000000e+00)
>> diff --git a/test/CodeGen/R600/urem.ll b/test/CodeGen/R600/urem.ll
>> index 8045145..a0aee5f 100644
>> --- a/test/CodeGen/R600/urem.ll
>> +++ b/test/CodeGen/R600/urem.ll
>> @@ -5,9 +5,9 @@
>> ;The goal of this test is to make sure the ISel doesn't fail when it gets
>> ;a v2i32/v4i32 urem
>>
>> -;EG-CHECK: @test2
>> +;EG-CHECK: test2
>> ;EG-CHECK: CF_END
>> -;SI-CHECK: @test2
>> +;SI-CHECK: test2
>> ;SI-CHECK: S_ENDPGM
>>
>> define void @test2(<2 x i32> addrspace(1)* %out, <2 x i32> addrspace(1)* %in) {
>> @@ -19,9 +19,9 @@ define void @test2(<2 x i32> addrspace(1)* %out, <2 x i32> addrspace(1)* %in) {
>> ret void
>> }
>>
>> -;EG-CHECK: @test4
>> +;EG-CHECK: test4
>> ;EG-CHECK: CF_END
>> -;SI-CHECK: @test4
>> +;SI-CHECK: test4
>> ;SI-CHECK: S_ENDPGM
>>
>> define void @test4(<4 x i32> addrspace(1)* %out, <4 x i32> addrspace(1)* %in) {
>> diff --git a/test/CodeGen/R600/use-sgpr-multiple-times.ll b/test/CodeGen/R600/use-sgpr-multiple-times.ll
>> index 5c00718..5e83467 100644
>> --- a/test/CodeGen/R600/use-sgpr-multiple-times.ll
>> +++ b/test/CodeGen/R600/use-sgpr-multiple-times.ll
>> @@ -5,7 +5,7 @@ declare float @llvm.fmuladd.f32(float, float, float) #1
>> declare i32 @llvm.AMDGPU.imad24(i32, i32, i32) #1
>>
>>
>> -; SI-LABEL: @test_sgpr_use_twice_binop:
>> +; SI-LABEL: test_sgpr_use_twice_binop:
>> ; SI: S_LOAD_DWORD [[SGPR:s[0-9]+]],
>> ; SI: V_ADD_F32_e64 [[RESULT:v[0-9]+]], [[SGPR]], [[SGPR]]
>> ; SI: BUFFER_STORE_DWORD [[RESULT]]
>> @@ -15,7 +15,7 @@ define void @test_sgpr_use_twice_binop(float addrspace(1)* %out, float %a) #0 {
>> ret void
>> }
>>
>> -; SI-LABEL: @test_sgpr_use_three_ternary_op:
>> +; SI-LABEL: test_sgpr_use_three_ternary_op:
>> ; SI: S_LOAD_DWORD [[SGPR:s[0-9]+]],
>> ; SI: V_FMA_F32 [[RESULT:v[0-9]+]], [[SGPR]], [[SGPR]], [[SGPR]]
>> ; SI: BUFFER_STORE_DWORD [[RESULT]]
>> @@ -25,7 +25,7 @@ define void @test_sgpr_use_three_ternary_op(float addrspace(1)* %out, float %a)
>> ret void
>> }
>>
>> -; SI-LABEL: @test_sgpr_use_twice_ternary_op_a_a_b:
>> +; SI-LABEL: test_sgpr_use_twice_ternary_op_a_a_b:
>> ; SI: S_LOAD_DWORD [[SGPR0:s[0-9]+]], s{{\[[0-9]+:[0-9]+\]}}, 0xb
>> ; SI: S_LOAD_DWORD [[SGPR1:s[0-9]+]], s{{\[[0-9]+:[0-9]+\]}}, 0xc
>> ; SI: V_MOV_B32_e32 [[VGPR1:v[0-9]+]], [[SGPR1]]
>> @@ -37,7 +37,7 @@ define void @test_sgpr_use_twice_ternary_op_a_a_b(float addrspace(1)* %out, floa
>> ret void
>> }
>>
>> -; SI-LABEL: @test_sgpr_use_twice_ternary_op_a_b_a:
>> +; SI-LABEL: test_sgpr_use_twice_ternary_op_a_b_a:
>> ; SI: S_LOAD_DWORD [[SGPR0:s[0-9]+]], s{{\[[0-9]+:[0-9]+\]}}, 0xb
>> ; SI: S_LOAD_DWORD [[SGPR1:s[0-9]+]], s{{\[[0-9]+:[0-9]+\]}}, 0xc
>> ; SI: V_MOV_B32_e32 [[VGPR1:v[0-9]+]], [[SGPR1]]
>> @@ -49,7 +49,7 @@ define void @test_sgpr_use_twice_ternary_op_a_b_a(float addrspace(1)* %out, floa
>> ret void
>> }
>>
>> -; SI-LABEL: @test_sgpr_use_twice_ternary_op_b_a_a:
>> +; SI-LABEL: test_sgpr_use_twice_ternary_op_b_a_a:
>> ; SI: S_LOAD_DWORD [[SGPR0:s[0-9]+]], s{{\[[0-9]+:[0-9]+\]}}, 0xb
>> ; SI: S_LOAD_DWORD [[SGPR1:s[0-9]+]], s{{\[[0-9]+:[0-9]+\]}}, 0xc
>> ; SI: V_MOV_B32_e32 [[VGPR1:v[0-9]+]], [[SGPR1]]
>> @@ -61,7 +61,7 @@ define void @test_sgpr_use_twice_ternary_op_b_a_a(float addrspace(1)* %out, floa
>> ret void
>> }
>>
>> -; SI-LABEL: @test_sgpr_use_twice_ternary_op_a_a_imm:
>> +; SI-LABEL: test_sgpr_use_twice_ternary_op_a_a_imm:
>> ; SI: S_LOAD_DWORD [[SGPR:s[0-9]+]]
>> ; SI: V_FMA_F32 [[RESULT:v[0-9]+]], [[SGPR]], [[SGPR]], 2.0
>> ; SI: BUFFER_STORE_DWORD [[RESULT]]
>> @@ -71,7 +71,7 @@ define void @test_sgpr_use_twice_ternary_op_a_a_imm(float addrspace(1)* %out, fl
>> ret void
>> }
>>
>> -; SI-LABEL: @test_sgpr_use_twice_ternary_op_a_imm_a:
>> +; SI-LABEL: test_sgpr_use_twice_ternary_op_a_imm_a:
>> ; SI: S_LOAD_DWORD [[SGPR:s[0-9]+]]
>> ; SI: V_FMA_F32 [[RESULT:v[0-9]+]], [[SGPR]], 2.0, [[SGPR]]
>> ; SI: BUFFER_STORE_DWORD [[RESULT]]
>> @@ -82,7 +82,7 @@ define void @test_sgpr_use_twice_ternary_op_a_imm_a(float addrspace(1)* %out, fl
>> }
>>
>> ; Don't use fma since fma c, x, y is canonicalized to fma x, c, y
>> -; SI-LABEL: @test_sgpr_use_twice_ternary_op_imm_a_a:
>> +; SI-LABEL: test_sgpr_use_twice_ternary_op_imm_a_a:
>> ; SI: S_LOAD_DWORD [[SGPR:s[0-9]+]]
>> ; SI: V_MAD_I32_I24 [[RESULT:v[0-9]+]], 2, [[SGPR]], [[SGPR]]
>> ; SI: BUFFER_STORE_DWORD [[RESULT]]
>> diff --git a/test/CodeGen/R600/usubo.ll b/test/CodeGen/R600/usubo.ll
>> index c293ad7..1601f66 100644
>> --- a/test/CodeGen/R600/usubo.ll
>> +++ b/test/CodeGen/R600/usubo.ll
>> @@ -4,7 +4,7 @@
>> declare { i32, i1 } @llvm.usub.with.overflow.i32(i32, i32) nounwind readnone
>> declare { i64, i1 } @llvm.usub.with.overflow.i64(i64, i64) nounwind readnone
>>
>> -; FUNC-LABEL: @usubo_i64_zext
>> +; FUNC-LABEL: usubo_i64_zext
>> define void @usubo_i64_zext(i64 addrspace(1)* %out, i64 %a, i64 %b) nounwind {
>> %usub = call { i64, i1 } @llvm.usub.with.overflow.i64(i64 %a, i64 %b) nounwind
>> %val = extractvalue { i64, i1 } %usub, 0
>> @@ -15,7 +15,7 @@ define void @usubo_i64_zext(i64 addrspace(1)* %out, i64 %a, i64 %b) nounwind {
>> ret void
>> }
>>
>> -; FUNC-LABEL: @s_usubo_i32
>> +; FUNC-LABEL: s_usubo_i32
>> ; SI: S_SUB_I32
>> define void @s_usubo_i32(i32 addrspace(1)* %out, i1 addrspace(1)* %carryout, i32 %a, i32 %b) nounwind {
>> %usub = call { i32, i1 } @llvm.usub.with.overflow.i32(i32 %a, i32 %b) nounwind
>> @@ -26,7 +26,7 @@ define void @s_usubo_i32(i32 addrspace(1)* %out, i1 addrspace(1)* %carryout, i32
>> ret void
>> }
>>
>> -; FUNC-LABEL: @v_usubo_i32
>> +; FUNC-LABEL: v_usubo_i32
>> ; SI: V_SUBREV_I32_e32
>> define void @v_usubo_i32(i32 addrspace(1)* %out, i1 addrspace(1)* %carryout, i32 addrspace(1)* %aptr, i32 addrspace(1)* %bptr) nounwind {
>> %a = load i32 addrspace(1)* %aptr, align 4
>> @@ -39,7 +39,7 @@ define void @v_usubo_i32(i32 addrspace(1)* %out, i1 addrspace(1)* %carryout, i32
>> ret void
>> }
>>
>> -; FUNC-LABEL: @s_usubo_i64
>> +; FUNC-LABEL: s_usubo_i64
>> ; SI: S_SUB_U32
>> ; SI: S_SUBB_U32
>> define void @s_usubo_i64(i64 addrspace(1)* %out, i1 addrspace(1)* %carryout, i64 %a, i64 %b) nounwind {
>> @@ -51,7 +51,7 @@ define void @s_usubo_i64(i64 addrspace(1)* %out, i1 addrspace(1)* %carryout, i64
>> ret void
>> }
>>
>> -; FUNC-LABEL: @v_usubo_i64
>> +; FUNC-LABEL: v_usubo_i64
>> ; SI: V_SUB_I32
>> ; SI: V_SUBB_U32
>> define void @v_usubo_i64(i64 addrspace(1)* %out, i1 addrspace(1)* %carryout, i64 addrspace(1)* %aptr, i64 addrspace(1)* %bptr) nounwind {
>> diff --git a/test/CodeGen/R600/v1i64-kernel-arg.ll b/test/CodeGen/R600/v1i64-kernel-arg.ll
>> index 2aa1221..0b3a04a 100644
>> --- a/test/CodeGen/R600/v1i64-kernel-arg.ll
>> +++ b/test/CodeGen/R600/v1i64-kernel-arg.ll
>> @@ -2,14 +2,14 @@
>> ; XFAIL: *
>> ; RUN: llc -march=r600 -mcpu=cypress < %s | FileCheck %s
>>
>> -; CHECK-LABEL: @kernel_arg_i64
>> +; CHECK-LABEL: kernel_arg_i64
>> define void @kernel_arg_i64(i64 addrspace(1)* %out, i64 %a) nounwind {
>> store i64 %a, i64 addrspace(1)* %out, align 8
>> ret void
>> }
>>
>> ; i64 arg works, v1i64 arg does not.
>> -; CHECK-LABEL: @kernel_arg_v1i64
>> +; CHECK-LABEL: kernel_arg_v1i64
>> define void @kernel_arg_v1i64(<1 x i64> addrspace(1)* %out, <1 x i64> %a) nounwind {
>> store <1 x i64> %a, <1 x i64> addrspace(1)* %out, align 8
>> ret void
>> diff --git a/test/CodeGen/R600/v_cndmask.ll b/test/CodeGen/R600/v_cndmask.ll
>> index 51e3d8b..04abda1 100644
>> --- a/test/CodeGen/R600/v_cndmask.ll
>> +++ b/test/CodeGen/R600/v_cndmask.ll
>> @@ -2,7 +2,7 @@
>>
>> declare i32 @llvm.r600.read.tidig.x() #1
>>
>> -; SI-LABEL: @v_cnd_nan_nosgpr
>> +; SI-LABEL: v_cnd_nan_nosgpr:
>> ; SI: V_CNDMASK_B32_e64 v{{[0-9]}}, v{{[0-9]}}, -1, s{{\[[0-9]+:[0-9]+\]}}
>> ; SI-DAG: v{{[0-9]}}
>> ; All nan values are converted to 0xffffffff
>> @@ -22,7 +22,7 @@ define void @v_cnd_nan_nosgpr(float addrspace(1)* %out, i32 %c, float addrspace(
>> ; single constant bus SGPR usage is the last operand, and it should
>> ; never be moved.
>>
>> -; SI-LABEL: @v_cnd_nan
>> +; SI-LABEL: v_cnd_nan:
>> ; SI: V_CNDMASK_B32_e64 v{{[0-9]}}, v{{[0-9]}}, -1, s{{\[[0-9]+:[0-9]+\]}}
>> ; SI-DAG: v{{[0-9]}}
>> ; All nan values are converted to 0xffffffff
>> diff --git a/test/CodeGen/R600/vector-alloca.ll b/test/CodeGen/R600/vector-alloca.ll
>> index ec1995f..8e08a1c 100644
>> --- a/test/CodeGen/R600/vector-alloca.ll
>> +++ b/test/CodeGen/R600/vector-alloca.ll
>> @@ -2,7 +2,7 @@
>> ; RUN: llc -march=r600 -mcpu=verde -mattr=-promote-alloca -verify-machineinstrs < %s | FileCheck -check-prefix=SI-ALLOCA -check-prefix=SI -check-prefix=FUNC %s
>> ; RUN: llc -march=r600 -mcpu=verde -mattr=+promote-alloca -verify-machineinstrs < %s | FileCheck -check-prefix=SI-PROMOTE -check-prefix=SI -check-prefix=FUNC %s
>>
>> -; FUNC-LABEL: @vector_read
>> +; FUNC-LABEL: vector_read
>> ; EG: MOV
>> ; EG: MOV
>> ; EG: MOV
>> @@ -25,7 +25,7 @@ entry:
>> ret void
>> }
>>
>> -; FUNC-LABEL: @vector_write
>> +; FUNC-LABEL: vector_write
>> ; EG: MOV
>> ; EG: MOV
>> ; EG: MOV
>> @@ -53,7 +53,7 @@ entry:
>>
>> ; This test should be optimize to:
>> ; store i32 0, i32 addrspace(1)* %out
>> -; FUNC-LABEL: @bitcast_gep
>> +; FUNC-LABEL: bitcast_gep
>> ; EG: STORE_RAW
>> define void @bitcast_gep(i32 addrspace(1)* %out, i32 %w_index, i32 %r_index) {
>> entry:
>> diff --git a/test/CodeGen/R600/vertex-fetch-encoding.ll b/test/CodeGen/R600/vertex-fetch-encoding.ll
>> index 7ea7a5c..a21bcfe 100644
>> --- a/test/CodeGen/R600/vertex-fetch-encoding.ll
>> +++ b/test/CodeGen/R600/vertex-fetch-encoding.ll
>> @@ -1,9 +1,9 @@
>> ; RUN: llc < %s -march=r600 -show-mc-encoding -mcpu=barts | FileCheck --check-prefix=NI-CHECK %s
>> ; RUN: llc < %s -march=r600 -show-mc-encoding -mcpu=cayman | FileCheck --check-prefix=CM-CHECK %s
>>
>> -; NI-CHECK: @vtx_fetch32
>> +; NI-CHECK: vtx_fetch32
>> ; NI-CHECK: VTX_READ_32 T[[GPR:[0-9]]].X, T[[GPR]].X, 0 ; encoding: [0x40,0x01,0x0[[GPR]],0x10,0x0[[GPR]],0xf0,0x5f,0x13,0x00,0x00,0x08,0x00
>> -; CM-CHECK: @vtx_fetch32
>> +; CM-CHECK: vtx_fetch32
>> ; CM-CHECK: VTX_READ_32 T[[GPR:[0-9]]].X, T[[GPR]].X, 0 ; encoding: [0x40,0x01,0x0[[GPR]],0x00,0x0[[GPR]],0xf0,0x5f,0x13,0x00,0x00,0x00,0x00
>>
>> define void @vtx_fetch32(i32 addrspace(1)* %out, i32 addrspace(1)* %in) {
>> @@ -13,7 +13,7 @@ entry:
>> ret void
>> }
>>
>> -; NI-CHECK: @vtx_fetch128
>> +; NI-CHECK: vtx_fetch128
>> ; NI-CHECK: VTX_READ_128 T[[DST:[0-9]]].XYZW, T[[SRC:[0-9]]].X, 0 ; encoding: [0x40,0x01,0x0[[SRC]],0x40,0x0[[DST]],0x10,0x8d,0x18,0x00,0x00,0x08,0x00
>> ; XXX: Add a case for Cayman when v4i32 stores are supported.
>>
>> diff --git a/test/CodeGen/R600/vop-shrink.ll b/test/CodeGen/R600/vop-shrink.ll
>> index bf1aae4..ee7197b 100644
>> --- a/test/CodeGen/R600/vop-shrink.ll
>> +++ b/test/CodeGen/R600/vop-shrink.ll
>> @@ -1,7 +1,7 @@
>> ; RUN: llc -march=r600 -mcpu=SI -verify-machineinstrs< %s | FileCheck -check-prefix=SI -check-prefix=FUNC %s
>>
>> ; Test that we correctly commute a sub instruction
>> -; FUNC-LABEL: @sub_rev
>> +; FUNC-LABEL: sub_rev
>> ; SI-NOT: V_SUB_I32_e32 v{{[0-9]+}}, s
>> ; SI: V_SUBREV_I32_e32 v{{[0-9]+}}, s
>>
>> @@ -32,7 +32,7 @@ endif: ; preds = %else, %if
>> ; Test that we fold an immediate that was illegal for a 64-bit op into the
>> ; 32-bit op when we shrink it.
>>
>> -; FUNC-LABEL: @add_fold
>> +; FUNC-LABEL: add_fold
>> ; SI: V_ADD_F32_e32 v{{[0-9]+}}, 0x44800000
>> define void @add_fold(float addrspace(1)* %out) {
>> entry:
>> diff --git a/test/CodeGen/R600/vselect.ll b/test/CodeGen/R600/vselect.ll
>> index dca7b06..90ebdc4 100644
>> --- a/test/CodeGen/R600/vselect.ll
>> +++ b/test/CodeGen/R600/vselect.ll
>> @@ -1,11 +1,11 @@
>> ;RUN: llc < %s -march=r600 -mcpu=redwood | FileCheck --check-prefix=EG-CHECK %s
>> ;RUN: llc < %s -march=r600 -mcpu=SI -verify-machineinstrs | FileCheck --check-prefix=SI-CHECK %s
>>
>> -;EG-CHECK: @test_select_v2i32
>> +;EG-CHECK: test_select_v2i32
>> ;EG-CHECK: CNDE_INT {{\** *}}T{{[0-9]+\.[XYZW], PV\.[XYZW], T[0-9]+\.[XYZW], T[0-9]+\.[XYZW]}}
>> ;EG-CHECK: CNDE_INT {{\** *}}T{{[0-9]+\.[XYZW], PV\.[XYZW], T[0-9]+\.[XYZW], T[0-9]+\.[XYZW]}}
>>
>> -;SI-CHECK: @test_select_v2i32
>> +;SI-CHECK: test_select_v2i32
>> ;SI-CHECK: V_CNDMASK_B32_e64
>> ;SI-CHECK: V_CNDMASK_B32_e64
>>
>> @@ -19,11 +19,11 @@ entry:
>> ret void
>> }
>>
>> -;EG-CHECK: @test_select_v2f32
>> +;EG-CHECK: test_select_v2f32
>> ;EG-CHECK: CNDE_INT {{\** *}}T{{[0-9]+\.[XYZW], PV\.[XYZW], T[0-9]+\.[XYZW], T[0-9]+\.[XYZW]}}
>> ;EG-CHECK: CNDE_INT {{\** *}}T{{[0-9]+\.[XYZW], PV\.[XYZW], T[0-9]+\.[XYZW], T[0-9]+\.[XYZW]}}
>>
>> -;SI-CHECK: @test_select_v2f32
>> +;SI-CHECK: test_select_v2f32
>> ;SI-CHECK: V_CNDMASK_B32_e64
>> ;SI-CHECK: V_CNDMASK_B32_e64
>>
>> @@ -37,13 +37,13 @@ entry:
>> ret void
>> }
>>
>> -;EG-CHECK: @test_select_v4i32
>> +;EG-CHECK: test_select_v4i32
>> ;EG-CHECK: CNDE_INT {{\** *}}T{{[0-9]+\.[XYZW], PV\.[XYZW], T[0-9]+\.[XYZW], T[0-9]+\.[XYZW]}}
>> ;EG-CHECK: CNDE_INT {{\** *}}T{{[0-9]+\.[XYZW], T[0-9]+\.[XYZW], T[0-9]+\.[XYZW], T[0-9]+\.[XYZW]}}
>> ;EG-CHECK: CNDE_INT {{\** *}}T{{[0-9]+\.[XYZW], PV\.[XYZW], T[0-9]+\.[XYZW], T[0-9]+\.[XYZW]}}
>> ;EG-CHECK: CNDE_INT {{\** *}}T{{[0-9]+\.[XYZW], PV\.[XYZW], T[0-9]+\.[XYZW], T[0-9]+\.[XYZW]}}
>>
>> -;SI-CHECK: @test_select_v4i32
>> +;SI-CHECK: test_select_v4i32
>> ;SI-CHECK: V_CNDMASK_B32_e64
>> ;SI-CHECK: V_CNDMASK_B32_e64
>> ;SI-CHECK: V_CNDMASK_B32_e64
>> @@ -59,7 +59,7 @@ entry:
>> ret void
>> }
>>
>> -;EG-CHECK: @test_select_v4f32
>> +;EG-CHECK: test_select_v4f32
>> ;EG-CHECK: CNDE_INT {{\** *}}T{{[0-9]+\.[XYZW], PV\.[XYZW], T[0-9]+\.[XYZW], T[0-9]+\.[XYZW]}}
>> ;EG-CHECK: CNDE_INT {{\** *}}T{{[0-9]+\.[XYZW], T[0-9]+\.[XYZW], T[0-9]+\.[XYZW], T[0-9]+\.[XYZW]}}
>> ;EG-CHECK: CNDE_INT {{\** *}}T{{[0-9]+\.[XYZW], PV\.[XYZW], T[0-9]+\.[XYZW], T[0-9]+\.[XYZW]}}
>> diff --git a/test/CodeGen/R600/vselect64.ll b/test/CodeGen/R600/vselect64.ll
>> index 604695b..4b532be 100644
>> --- a/test/CodeGen/R600/vselect64.ll
>> +++ b/test/CodeGen/R600/vselect64.ll
>> @@ -1,7 +1,7 @@
>> ; RUN: llc < %s -march=r600 -mcpu=redwood | FileCheck %s
>> ; XXX: Merge this test into vselect.ll once SI supports 64-bit select.
>>
>> -; CHECK-LABEL: @test_select_v4i64
>> +; CHECK-LABEL: test_select_v4i64
>> ; Make sure the vectors aren't being stored on the stack. We know they are
>> ; being stored on the stack if the shaders uses at leat 10 registers.
>> ; CHECK-NOT: {{\**}} MOV T{{[0-9][0-9]}}.X
>> diff --git a/test/CodeGen/R600/vtx-fetch-branch.ll b/test/CodeGen/R600/vtx-fetch-branch.ll
>> index 0fc99de..4805bb2 100644
>> --- a/test/CodeGen/R600/vtx-fetch-branch.ll
>> +++ b/test/CodeGen/R600/vtx-fetch-branch.ll
>> @@ -6,7 +6,7 @@
>> ; after the fetch clause.
>>
>>
>> -; CHECK-LABEL: @test
>> +; CHECK-LABEL: test
>> ; CHECK-NOT: ALU_POP_AFTER
>> ; CHECK: TEX
>> ; CHECK-NEXT: POP
>> diff --git a/test/CodeGen/R600/vtx-schedule.ll b/test/CodeGen/R600/vtx-schedule.ll
>> index ce852c5..0d727ed 100644
>> --- a/test/CodeGen/R600/vtx-schedule.ll
>> +++ b/test/CodeGen/R600/vtx-schedule.ll
>> @@ -4,7 +4,7 @@
>> ; the result of another VTX_READ instruction were being grouped in the
>> ; same fetch clasue.
>>
>> -; CHECK: @test
>> +; CHECK: test
>> ; CHECK: Fetch clause
>> ; CHECK: VTX_READ_32 [[IN0:T[0-9]+\.X]], [[IN0]], 0
>> ; CHECK: Fetch clause
>> diff --git a/test/CodeGen/R600/wait.ll b/test/CodeGen/R600/wait.ll
>> index b0b7e91..f349a6b 100644
>> --- a/test/CodeGen/R600/wait.ll
>> +++ b/test/CodeGen/R600/wait.ll
>> @@ -1,6 +1,6 @@
>> ; RUN: llc -march=r600 -mcpu=SI -verify-machineinstrs < %s | FileCheck -strict-whitespace %s
>>
>> -; CHECK-LABEL: @main
>> +; CHECK-LABEL: main
>> ; CHECK: S_LOAD_DWORDX4
>> ; CHECK: S_LOAD_DWORDX4
>> ; CHECK: S_WAITCNT lgkmcnt(0){{$}}
>> diff --git a/test/CodeGen/R600/work-item-intrinsics.ll b/test/CodeGen/R600/work-item-intrinsics.ll
>> index 1dbd9b8..022a99b 100644
>> --- a/test/CodeGen/R600/work-item-intrinsics.ll
>> +++ b/test/CodeGen/R600/work-item-intrinsics.ll
>> @@ -2,7 +2,7 @@
>> ; RUN: llc -march=r600 -mcpu=SI -verify-machineinstrs < %s | FileCheck -check-prefix=SI -check-prefix=FUNC %s
>>
>>
>> -; FUNC-LABEL: @ngroups_x
>> +; FUNC-LABEL: ngroups_x
>> ; EG: MEM_RAT_CACHELESS STORE_RAW [[VAL:T[0-9]+\.X]]
>> ; EG: MOV [[VAL]], KC0[0].X
>>
>> @@ -16,7 +16,7 @@ entry:
>> ret void
>> }
>>
>> -; FUNC-LABEL: @ngroups_y
>> +; FUNC-LABEL: ngroups_y
>> ; EG: MEM_RAT_CACHELESS STORE_RAW [[VAL:T[0-9]+\.X]]
>> ; EG: MOV [[VAL]], KC0[0].Y
>>
>> @@ -30,7 +30,7 @@ entry:
>> ret void
>> }
>>
>> -; FUNC-LABEL: @ngroups_z
>> +; FUNC-LABEL: ngroups_z
>> ; EG: MEM_RAT_CACHELESS STORE_RAW [[VAL:T[0-9]+\.X]]
>> ; EG: MOV [[VAL]], KC0[0].Z
>>
>> @@ -44,7 +44,7 @@ entry:
>> ret void
>> }
>>
>> -; FUNC-LABEL: @global_size_x
>> +; FUNC-LABEL: global_size_x
>> ; EG: MEM_RAT_CACHELESS STORE_RAW [[VAL:T[0-9]+\.X]]
>> ; EG: MOV [[VAL]], KC0[0].W
>>
>> @@ -58,7 +58,7 @@ entry:
>> ret void
>> }
>>
>> -; FUNC-LABEL: @global_size_y
>> +; FUNC-LABEL: global_size_y
>> ; EG: MEM_RAT_CACHELESS STORE_RAW [[VAL:T[0-9]+\.X]]
>> ; EG: MOV [[VAL]], KC0[1].X
>>
>> @@ -72,7 +72,7 @@ entry:
>> ret void
>> }
>>
>> -; FUNC-LABEL: @global_size_z
>> +; FUNC-LABEL: global_size_z
>> ; EG: MEM_RAT_CACHELESS STORE_RAW [[VAL:T[0-9]+\.X]]
>> ; EG: MOV [[VAL]], KC0[1].Y
>>
>> @@ -86,7 +86,7 @@ entry:
>> ret void
>> }
>>
>> -; FUNC-LABEL: @local_size_x
>> +; FUNC-LABEL: local_size_x
>> ; EG: MEM_RAT_CACHELESS STORE_RAW [[VAL:T[0-9]+\.X]]
>> ; EG: MOV [[VAL]], KC0[1].Z
>>
>> @@ -100,7 +100,7 @@ entry:
>> ret void
>> }
>>
>> -; FUNC-LABEL: @local_size_y
>> +; FUNC-LABEL: local_size_y
>> ; EG: MEM_RAT_CACHELESS STORE_RAW [[VAL:T[0-9]+\.X]]
>> ; EG: MOV [[VAL]], KC0[1].W
>>
>> @@ -114,7 +114,7 @@ entry:
>> ret void
>> }
>>
>> -; FUNC-LABEL: @local_size_z
>> +; FUNC-LABEL: local_size_z
>> ; EG: MEM_RAT_CACHELESS STORE_RAW [[VAL:T[0-9]+\.X]]
>> ; EG: MOV [[VAL]], KC0[2].X
>>
>> @@ -132,7 +132,7 @@ entry:
>> ; Currently we always use exactly 2 user sgprs for the pointer to the
>> ; kernel arguments, but this may change in the future.
>>
>> -; FUNC-LABEL: @tgid_x
>> +; FUNC-LABEL: tgid_x
>> ; SI: V_MOV_B32_e32 [[VVAL:v[0-9]+]], s4
>> ; SI: BUFFER_STORE_DWORD [[VVAL]]
>> define void @tgid_x (i32 addrspace(1)* %out) {
>> @@ -142,7 +142,7 @@ entry:
>> ret void
>> }
>>
>> -; FUNC-LABEL: @tgid_y
>> +; FUNC-LABEL: tgid_y
>> ; SI: V_MOV_B32_e32 [[VVAL:v[0-9]+]], s5
>> ; SI: BUFFER_STORE_DWORD [[VVAL]]
>> define void @tgid_y (i32 addrspace(1)* %out) {
>> @@ -152,7 +152,7 @@ entry:
>> ret void
>> }
>>
>> -; FUNC-LABEL: @tgid_z
>> +; FUNC-LABEL: tgid_z
>> ; SI: V_MOV_B32_e32 [[VVAL:v[0-9]+]], s6
>> ; SI: BUFFER_STORE_DWORD [[VVAL]]
>> define void @tgid_z (i32 addrspace(1)* %out) {
>> @@ -162,7 +162,7 @@ entry:
>> ret void
>> }
>>
>> -; FUNC-LABEL: @tidig_x
>> +; FUNC-LABEL: tidig_x
>> ; SI: BUFFER_STORE_DWORD v0
>> define void @tidig_x (i32 addrspace(1)* %out) {
>> entry:
>> @@ -171,7 +171,7 @@ entry:
>> ret void
>> }
>>
>> -; FUNC-LABEL: @tidig_y
>> +; FUNC-LABEL: tidig_y
>> ; SI: BUFFER_STORE_DWORD v1
>> define void @tidig_y (i32 addrspace(1)* %out) {
>> entry:
>> @@ -180,7 +180,7 @@ entry:
>> ret void
>> }
>>
>> -; FUNC-LABEL: @tidig_z
>> +; FUNC-LABEL: tidig_z
>> ; SI: BUFFER_STORE_DWORD v2
>> define void @tidig_z (i32 addrspace(1)* %out) {
>> entry:
>> diff --git a/test/CodeGen/R600/wrong-transalu-pos-fix.ll b/test/CodeGen/R600/wrong-transalu-pos-fix.ll
>> index b1cbe3f..9332fd5 100644
>> --- a/test/CodeGen/R600/wrong-transalu-pos-fix.ll
>> +++ b/test/CodeGen/R600/wrong-transalu-pos-fix.ll
>> @@ -1,7 +1,7 @@
>> ; RUN: llc < %s -march=r600 -mcpu=redwood | FileCheck %s
>>
>> ; We want all MULLO_INT inst to be last in their instruction group
>> -;CHECK: @fill3d
>> +;CHECK: fill3d
>> ;CHECK-NOT: MULLO_INT T[0-9]+
>>
>> ; ModuleID = 'radeon'
>> diff --git a/test/CodeGen/R600/xor.ll b/test/CodeGen/R600/xor.ll
>> index 8c2c80e..1e4803e 100644
>> --- a/test/CodeGen/R600/xor.ll
>> +++ b/test/CodeGen/R600/xor.ll
>> @@ -1,11 +1,11 @@
>> ;RUN: llc < %s -march=r600 -mcpu=redwood | FileCheck --check-prefix=EG-CHECK %s
>> ;RUN: llc < %s -march=r600 -mcpu=verde -verify-machineinstrs | FileCheck --check-prefix=SI-CHECK %s
>>
>> -;EG-CHECK: @xor_v2i32
>> +;EG-CHECK: xor_v2i32
>> ;EG-CHECK: XOR_INT {{\** *}}T{{[0-9]+\.[XYZW], T[0-9]+\.[XYZW], T[0-9]+\.[XYZW]}}
>> ;EG-CHECK: XOR_INT {{\** *}}T{{[0-9]+\.[XYZW], T[0-9]+\.[XYZW], T[0-9]+\.[XYZW]}}
>>
>> -;SI-CHECK: @xor_v2i32
>> +;SI-CHECK: xor_v2i32
>> ;SI-CHECK: V_XOR_B32_e32 v{{[0-9]+, v[0-9]+, v[0-9]+}}
>> ;SI-CHECK: V_XOR_B32_e32 v{{[0-9]+, v[0-9]+, v[0-9]+}}
>>
>> @@ -18,13 +18,13 @@ define void @xor_v2i32(<2 x i32> addrspace(1)* %out, <2 x i32> addrspace(1)* %in
>> ret void
>> }
>>
>> -;EG-CHECK: @xor_v4i32
>> +;EG-CHECK: xor_v4i32
>> ;EG-CHECK: XOR_INT {{\** *}}T{{[0-9]+\.[XYZW], T[0-9]+\.[XYZW], T[0-9]+\.[XYZW]}}
>> ;EG-CHECK: XOR_INT {{\** *}}T{{[0-9]+\.[XYZW], T[0-9]+\.[XYZW], T[0-9]+\.[XYZW]}}
>> ;EG-CHECK: XOR_INT {{\** *}}T{{[0-9]+\.[XYZW], T[0-9]+\.[XYZW], T[0-9]+\.[XYZW]}}
>> ;EG-CHECK: XOR_INT {{\** *}}T{{[0-9]+\.[XYZW], T[0-9]+\.[XYZW], T[0-9]+\.[XYZW]}}
>>
>> -;SI-CHECK: @xor_v4i32
>> +;SI-CHECK: xor_v4i32
>> ;SI-CHECK: V_XOR_B32_e32 {{v[0-9]+, v[0-9]+, v[0-9]+}}
>> ;SI-CHECK: V_XOR_B32_e32 {{v[0-9]+, v[0-9]+, v[0-9]+}}
>> ;SI-CHECK: V_XOR_B32_e32 {{v[0-9]+, v[0-9]+, v[0-9]+}}
>> @@ -38,10 +38,10 @@ define void @xor_v4i32(<4 x i32> addrspace(1)* %out, <4 x i32> addrspace(1)* %in
>> ret void
>> }
>>
>> -;EG-CHECK: @xor_i1
>> +;EG-CHECK: xor_i1
>> ;EG-CHECK: XOR_INT {{\** *}}T{{[0-9]+\.[XYZW], PV\.[XYZW], PS}}
>>
>> -;SI-CHECK: @xor_i1
>> +;SI-CHECK: xor_i1
>> ;SI-CHECK: V_XOR_B32_e32 v{{[0-9]+}}, v{{[0-9]+}}, v{{[0-9]+}}
>>
>> define void @xor_i1(float addrspace(1)* %out, float addrspace(1)* %in0, float addrspace(1)* %in1) {
>> @@ -55,7 +55,7 @@ define void @xor_i1(float addrspace(1)* %out, float addrspace(1)* %in0, float ad
>> ret void
>> }
>>
>> -; SI-CHECK-LABEL: @vector_xor_i32
>> +; SI-CHECK-LABEL: vector_xor_i32
>> ; SI-CHECK: V_XOR_B32_e32
>> define void @vector_xor_i32(i32 addrspace(1)* %out, i32 addrspace(1)* %in0, i32 addrspace(1)* %in1) {
>> %a = load i32 addrspace(1)* %in0
>> @@ -65,7 +65,7 @@ define void @vector_xor_i32(i32 addrspace(1)* %out, i32 addrspace(1)* %in0, i32
>> ret void
>> }
>>
>> -; SI-CHECK-LABEL: @scalar_xor_i32
>> +; SI-CHECK-LABEL: scalar_xor_i32
>> ; SI-CHECK: S_XOR_B32
>> define void @scalar_xor_i32(i32 addrspace(1)* %out, i32 %a, i32 %b) {
>> %result = xor i32 %a, %b
>> @@ -73,7 +73,7 @@ define void @scalar_xor_i32(i32 addrspace(1)* %out, i32 %a, i32 %b) {
>> ret void
>> }
>>
>> -; SI-CHECK-LABEL: @scalar_not_i32
>> +; SI-CHECK-LABEL: scalar_not_i32
>> ; SI-CHECK: S_NOT_B32
>> define void @scalar_not_i32(i32 addrspace(1)* %out, i32 %a) {
>> %result = xor i32 %a, -1
>> @@ -81,7 +81,7 @@ define void @scalar_not_i32(i32 addrspace(1)* %out, i32 %a) {
>> ret void
>> }
>>
>> -; SI-CHECK-LABEL: @vector_not_i32
>> +; SI-CHECK-LABEL: vector_not_i32
>> ; SI-CHECK: V_NOT_B32
>> define void @vector_not_i32(i32 addrspace(1)* %out, i32 addrspace(1)* %in0, i32 addrspace(1)* %in1) {
>> %a = load i32 addrspace(1)* %in0
>> @@ -91,7 +91,7 @@ define void @vector_not_i32(i32 addrspace(1)* %out, i32 addrspace(1)* %in0, i32
>> ret void
>> }
>>
>> -; SI-CHECK-LABEL: @vector_xor_i64
>> +; SI-CHECK-LABEL: vector_xor_i64
>> ; SI-CHECK: V_XOR_B32_e32
>> ; SI-CHECK: V_XOR_B32_e32
>> ; SI-CHECK: S_ENDPGM
>> @@ -103,7 +103,7 @@ define void @vector_xor_i64(i64 addrspace(1)* %out, i64 addrspace(1)* %in0, i64
>> ret void
>> }
>>
>> -; SI-CHECK-LABEL: @scalar_xor_i64
>> +; SI-CHECK-LABEL: scalar_xor_i64
>> ; SI-CHECK: S_XOR_B64
>> ; SI-CHECK: S_ENDPGM
>> define void @scalar_xor_i64(i64 addrspace(1)* %out, i64 %a, i64 %b) {
>> @@ -112,7 +112,7 @@ define void @scalar_xor_i64(i64 addrspace(1)* %out, i64 %a, i64 %b) {
>> ret void
>> }
>>
>> -; SI-CHECK-LABEL: @scalar_not_i64
>> +; SI-CHECK-LABEL: scalar_not_i64
>> ; SI-CHECK: S_NOT_B64
>> define void @scalar_not_i64(i64 addrspace(1)* %out, i64 %a) {
>> %result = xor i64 %a, -1
>> @@ -120,7 +120,7 @@ define void @scalar_not_i64(i64 addrspace(1)* %out, i64 %a) {
>> ret void
>> }
>>
>> -; SI-CHECK-LABEL: @vector_not_i64
>> +; SI-CHECK-LABEL: vector_not_i64
>> ; SI-CHECK: V_NOT_B32
>> ; SI-CHECK: V_NOT_B32
>> define void @vector_not_i64(i64 addrspace(1)* %out, i64 addrspace(1)* %in0, i64 addrspace(1)* %in1) {
>> @@ -135,7 +135,7 @@ define void @vector_not_i64(i64 addrspace(1)* %out, i64 addrspace(1)* %in0, i64
>> ; Note that in the future the backend may be smart enough to
>> ; use an SALU instruction for this.
>>
>> -; SI-CHECK-LABEL: @xor_cf
>> +; SI-CHECK-LABEL: xor_cf
>> ; SI-CHECK: S_XOR_B64
>> define void @xor_cf(i64 addrspace(1)* %out, i64 addrspace(1)* %in, i64 %a, i64 %b) {
>> entry:
>> diff --git a/test/CodeGen/R600/zero_extend.ll b/test/CodeGen/R600/zero_extend.ll
>> index 1a0fd73..3b997a7 100644
>> --- a/test/CodeGen/R600/zero_extend.ll
>> +++ b/test/CodeGen/R600/zero_extend.ll
>> @@ -1,11 +1,11 @@
>> ; RUN: llc < %s -march=r600 -mcpu=redwood | FileCheck %s --check-prefix=R600-CHECK
>> ; RUN: llc < %s -march=r600 -mcpu=SI -verify-machineinstrs | FileCheck %s --check-prefix=SI-CHECK
>>
>> -; R600-CHECK: @test
>> +; R600-CHECK: test
>> ; R600-CHECK: MEM_RAT_CACHELESS STORE_RAW
>> ; R600-CHECK: MEM_RAT_CACHELESS STORE_RAW
>>
>> -; SI-CHECK: @test
>> +; SI-CHECK: test
>> ; SI-CHECK: S_MOV_B32 [[ZERO:s[0-9]]], 0{{$}}
>> ; SI-CHECK: V_MOV_B32_e32 v[[V_ZERO:[0-9]]], [[ZERO]]
>> ; SI-CHECK: BUFFER_STORE_DWORDX2 v[0:[[V_ZERO]]{{\]}}
>> @@ -18,7 +18,7 @@ entry:
>> ret void
>> }
>>
>> -; SI-CHECK-LABEL: @testi1toi32
>> +; SI-CHECK-LABEL: testi1toi32
>> ; SI-CHECK: V_CNDMASK_B32
>> define void @testi1toi32(i32 addrspace(1)* %out, i32 %a, i32 %b) {
>> entry:
>> @@ -28,7 +28,7 @@ entry:
>> ret void
>> }
>>
>> -; SI-CHECK-LABEL: @zext_i1_to_i64
>> +; SI-CHECK-LABEL: zext_i1_to_i64
>> ; SI-CHECK: V_CMP_EQ_I32
>> ; SI-CHECK: V_CNDMASK_B32
>> ; SI-CHECK: S_MOV_B32 s{{[0-9]+}}, 0
>> --
>> 1.8.3.1
>>
>> _______________________________________________
>> llvm-commits mailing list
>> llvm-commits at cs.uiuc.edu
>> http://lists.cs.uiuc.edu/mailman/listinfo/llvm-commits
>
More information about the llvm-commits
mailing list