[Mesa-dev] [PATCH] R600/SI: Use i32 vectors for resources and samplers
Matt Arsenault
arsenm2 at gmail.com
Mon Jul 7 11:16:17 PDT 2014
On Jul 7, 2014, at 8:28 AM, Marek Olšák <maraeo at gmail.com> wrote:
> From: Marek Olšák <marek.olsak at amd.com>
>
> This affects new intrinsics only.
>
> What surprises me is that v32i8 still works.
> ---
> lib/Target/R600/SIInstructions.td | 4 +-
> lib/Target/R600/SIIntrinsics.td | 6 +--
> test/CodeGen/R600/llvm.SI.image.ll | 12 ++---
> test/CodeGen/R600/llvm.SI.image.sample.ll | 80 ++++++++++++++---------------
> test/CodeGen/R600/llvm.SI.image.sample.o.ll | 80 ++++++++++++++---------------
> 5 files changed, 91 insertions(+), 91 deletions(-)
LGTM
>
> diff --git a/lib/Target/R600/SIInstructions.td b/lib/Target/R600/SIInstructions.td
> index 25a9362..8aecbb9 100644
> --- a/lib/Target/R600/SIInstructions.td
> +++ b/lib/Target/R600/SIInstructions.td
> @@ -1851,7 +1851,7 @@ def : Pat <
>
> // Image + sampler
> class SampleRawPattern<SDPatternOperator name, MIMG opcode, ValueType vt> : Pat <
> - (name vt:$addr, v32i8:$rsrc, v16i8:$sampler, i32:$dmask, i32:$unorm,
> + (name vt:$addr, v8i32:$rsrc, v4i32:$sampler, i32:$dmask, i32:$unorm,
> i32:$r128, i32:$da, i32:$glc, i32:$slc, i32:$tfe, i32:$lwe),
> (opcode (as_i32imm $dmask), (as_i1imm $unorm), (as_i1imm $glc), (as_i1imm $da),
> (as_i1imm $r128), (as_i1imm $tfe), (as_i1imm $lwe), (as_i1imm $slc),
> @@ -1868,7 +1868,7 @@ multiclass SampleRawPatterns<SDPatternOperator name, string opcode> {
>
> // Image only
> class ImagePattern<SDPatternOperator name, MIMG opcode, ValueType vt> : Pat <
> - (name vt:$addr, v32i8:$rsrc, i32:$dmask, i32:$unorm,
> + (name vt:$addr, v8i32:$rsrc, i32:$dmask, i32:$unorm,
> i32:$r128, i32:$da, i32:$glc, i32:$slc, i32:$tfe, i32:$lwe),
> (opcode (as_i32imm $dmask), (as_i1imm $unorm), (as_i1imm $glc), (as_i1imm $da),
> (as_i1imm $r128), (as_i1imm $tfe), (as_i1imm $lwe), (as_i1imm $slc),
> diff --git a/lib/Target/R600/SIIntrinsics.td b/lib/Target/R600/SIIntrinsics.td
> index fc0c850..027a0a2 100644
> --- a/lib/Target/R600/SIIntrinsics.td
> +++ b/lib/Target/R600/SIIntrinsics.td
> @@ -58,8 +58,8 @@ let TargetPrefix = "SI", isTarget = 1 in {
> class SampleRaw : Intrinsic <
> [llvm_v4f32_ty], // vdata(VGPR)
> [llvm_anyint_ty, // vaddr(VGPR)
> - llvm_v32i8_ty, // rsrc(SGPR)
> - llvm_v16i8_ty, // sampler(SGPR)
> + llvm_v8i32_ty, // rsrc(SGPR)
> + llvm_v4i32_ty, // sampler(SGPR)
> llvm_i32_ty, // dmask(imm)
> llvm_i32_ty, // unorm(imm)
> llvm_i32_ty, // r128(imm)
> @@ -74,7 +74,7 @@ let TargetPrefix = "SI", isTarget = 1 in {
> class Image : Intrinsic <
> [llvm_v4f32_ty], // vdata(VGPR)
> [llvm_anyint_ty, // vaddr(VGPR)
> - llvm_v32i8_ty, // rsrc(SGPR)
> + llvm_v8i32_ty, // rsrc(SGPR)
> llvm_i32_ty, // dmask(imm)
> llvm_i32_ty, // unorm(imm)
> llvm_i32_ty, // r128(imm)
> diff --git a/test/CodeGen/R600/llvm.SI.image.ll b/test/CodeGen/R600/llvm.SI.image.ll
> index debe7c7..eac0b8e 100644
> --- a/test/CodeGen/R600/llvm.SI.image.ll
> +++ b/test/CodeGen/R600/llvm.SI.image.ll
> @@ -4,7 +4,7 @@
> ;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:
> - %r = call <4 x float> @llvm.SI.image.load.v4i32(<4 x i32> undef, <32 x i8> undef, i32 15, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0)
> + %r = call <4 x float> @llvm.SI.image.load.v4i32(<4 x i32> undef, <8 x i32> undef, i32 15, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0)
> %r0 = extractelement <4 x float> %r, i32 0
> %r1 = extractelement <4 x float> %r, i32 1
> %r2 = extractelement <4 x float> %r, i32 2
> @@ -17,7 +17,7 @@ main_body:
> ;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:
> - %r = call <4 x float> @llvm.SI.image.load.mip.v4i32(<4 x i32> undef, <32 x i8> undef, i32 15, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0)
> + %r = call <4 x float> @llvm.SI.image.load.mip.v4i32(<4 x i32> undef, <8 x i32> undef, i32 15, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0)
> %r0 = extractelement <4 x float> %r, i32 0
> %r1 = extractelement <4 x float> %r, i32 1
> %r2 = extractelement <4 x float> %r, i32 2
> @@ -30,7 +30,7 @@ main_body:
> ;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:
> - %r = call <4 x float> @llvm.SI.getresinfo.i32(i32 undef, <32 x i8> undef, i32 15, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0)
> + %r = call <4 x float> @llvm.SI.getresinfo.i32(i32 undef, <8 x i32> undef, i32 15, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0)
> %r0 = extractelement <4 x float> %r, i32 0
> %r1 = extractelement <4 x float> %r, i32 1
> %r2 = extractelement <4 x float> %r, i32 2
> @@ -39,9 +39,9 @@ main_body:
> ret void
> }
>
> -declare <4 x float> @llvm.SI.image.load.v4i32(<4 x i32>, <32 x i8>, i32, i32, i32, i32, i32, i32, i32, i32) #1
> -declare <4 x float> @llvm.SI.image.load.mip.v4i32(<4 x i32>, <32 x i8>, i32, i32, i32, i32, i32, i32, i32, i32) #1
> -declare <4 x float> @llvm.SI.getresinfo.i32(i32, <32 x i8>, i32, i32, i32, i32, i32, i32, i32, i32) #1
> +declare <4 x float> @llvm.SI.image.load.v4i32(<4 x i32>, <8 x i32>, i32, i32, i32, i32, i32, i32, i32, i32) #1
> +declare <4 x float> @llvm.SI.image.load.mip.v4i32(<4 x i32>, <8 x i32>, i32, i32, i32, i32, i32, i32, i32, i32) #1
> +declare <4 x float> @llvm.SI.getresinfo.i32(i32, <8 x i32>, i32, i32, i32, i32, i32, i32, i32, i32) #1
>
> declare void @llvm.SI.export(i32, i32, i32, i32, i32, float, float, float, float)
>
> diff --git a/test/CodeGen/R600/llvm.SI.image.sample.ll b/test/CodeGen/R600/llvm.SI.image.sample.ll
> index 20358b2..14dff7e 100644
> --- a/test/CodeGen/R600/llvm.SI.image.sample.ll
> +++ b/test/CodeGen/R600/llvm.SI.image.sample.ll
> @@ -4,7 +4,7 @@
> ;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:
> - %r = call <4 x float> @llvm.SI.image.sample.v4i32(<4 x i32> undef, <32 x i8> undef, <16 x i8> undef, i32 15, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0)
> + %r = call <4 x float> @llvm.SI.image.sample.v4i32(<4 x i32> undef, <8 x i32> undef, <4 x i32> undef, i32 15, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0)
> %r0 = extractelement <4 x float> %r, i32 0
> %r1 = extractelement <4 x float> %r, i32 1
> %r2 = extractelement <4 x float> %r, i32 2
> @@ -17,7 +17,7 @@ main_body:
> ;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:
> - %r = call <4 x float> @llvm.SI.image.sample.cl.v4i32(<4 x i32> undef, <32 x i8> undef, <16 x i8> undef, i32 15, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0)
> + %r = call <4 x float> @llvm.SI.image.sample.cl.v4i32(<4 x i32> undef, <8 x i32> undef, <4 x i32> undef, i32 15, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0)
> %r0 = extractelement <4 x float> %r, i32 0
> %r1 = extractelement <4 x float> %r, i32 1
> %r2 = extractelement <4 x float> %r, i32 2
> @@ -30,7 +30,7 @@ main_body:
> ;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:
> - %r = call <4 x float> @llvm.SI.image.sample.d.v4i32(<4 x i32> undef, <32 x i8> undef, <16 x i8> undef, i32 15, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0)
> + %r = call <4 x float> @llvm.SI.image.sample.d.v4i32(<4 x i32> undef, <8 x i32> undef, <4 x i32> undef, i32 15, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0)
> %r0 = extractelement <4 x float> %r, i32 0
> %r1 = extractelement <4 x float> %r, i32 1
> %r2 = extractelement <4 x float> %r, i32 2
> @@ -43,7 +43,7 @@ main_body:
> ;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:
> - %r = call <4 x float> @llvm.SI.image.sample.d.cl.v4i32(<4 x i32> undef, <32 x i8> undef, <16 x i8> undef, i32 15, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0)
> + %r = call <4 x float> @llvm.SI.image.sample.d.cl.v4i32(<4 x i32> undef, <8 x i32> undef, <4 x i32> undef, i32 15, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0)
> %r0 = extractelement <4 x float> %r, i32 0
> %r1 = extractelement <4 x float> %r, i32 1
> %r2 = extractelement <4 x float> %r, i32 2
> @@ -56,7 +56,7 @@ main_body:
> ;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:
> - %r = call <4 x float> @llvm.SI.image.sample.l.v4i32(<4 x i32> undef, <32 x i8> undef, <16 x i8> undef, i32 15, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0)
> + %r = call <4 x float> @llvm.SI.image.sample.l.v4i32(<4 x i32> undef, <8 x i32> undef, <4 x i32> undef, i32 15, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0)
> %r0 = extractelement <4 x float> %r, i32 0
> %r1 = extractelement <4 x float> %r, i32 1
> %r2 = extractelement <4 x float> %r, i32 2
> @@ -69,7 +69,7 @@ main_body:
> ;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:
> - %r = call <4 x float> @llvm.SI.image.sample.b.v4i32(<4 x i32> undef, <32 x i8> undef, <16 x i8> undef, i32 15, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0)
> + %r = call <4 x float> @llvm.SI.image.sample.b.v4i32(<4 x i32> undef, <8 x i32> undef, <4 x i32> undef, i32 15, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0)
> %r0 = extractelement <4 x float> %r, i32 0
> %r1 = extractelement <4 x float> %r, i32 1
> %r2 = extractelement <4 x float> %r, i32 2
> @@ -82,7 +82,7 @@ main_body:
> ;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:
> - %r = call <4 x float> @llvm.SI.image.sample.b.cl.v4i32(<4 x i32> undef, <32 x i8> undef, <16 x i8> undef, i32 15, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0)
> + %r = call <4 x float> @llvm.SI.image.sample.b.cl.v4i32(<4 x i32> undef, <8 x i32> undef, <4 x i32> undef, i32 15, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0)
> %r0 = extractelement <4 x float> %r, i32 0
> %r1 = extractelement <4 x float> %r, i32 1
> %r2 = extractelement <4 x float> %r, i32 2
> @@ -95,7 +95,7 @@ main_body:
> ;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:
> - %r = call <4 x float> @llvm.SI.image.sample.lz.v4i32(<4 x i32> undef, <32 x i8> undef, <16 x i8> undef, i32 15, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0)
> + %r = call <4 x float> @llvm.SI.image.sample.lz.v4i32(<4 x i32> undef, <8 x i32> undef, <4 x i32> undef, i32 15, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0)
> %r0 = extractelement <4 x float> %r, i32 0
> %r1 = extractelement <4 x float> %r, i32 1
> %r2 = extractelement <4 x float> %r, i32 2
> @@ -108,7 +108,7 @@ main_body:
> ;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:
> - %r = call <4 x float> @llvm.SI.image.sample.cd.v4i32(<4 x i32> undef, <32 x i8> undef, <16 x i8> undef, i32 15, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0)
> + %r = call <4 x float> @llvm.SI.image.sample.cd.v4i32(<4 x i32> undef, <8 x i32> undef, <4 x i32> undef, i32 15, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0)
> %r0 = extractelement <4 x float> %r, i32 0
> %r1 = extractelement <4 x float> %r, i32 1
> %r2 = extractelement <4 x float> %r, i32 2
> @@ -121,7 +121,7 @@ main_body:
> ;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:
> - %r = call <4 x float> @llvm.SI.image.sample.cd.cl.v4i32(<4 x i32> undef, <32 x i8> undef, <16 x i8> undef, i32 15, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0)
> + %r = call <4 x float> @llvm.SI.image.sample.cd.cl.v4i32(<4 x i32> undef, <8 x i32> undef, <4 x i32> undef, i32 15, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0)
> %r0 = extractelement <4 x float> %r, i32 0
> %r1 = extractelement <4 x float> %r, i32 1
> %r2 = extractelement <4 x float> %r, i32 2
> @@ -134,7 +134,7 @@ main_body:
> ;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:
> - %r = call <4 x float> @llvm.SI.image.sample.c.v4i32(<4 x i32> undef, <32 x i8> undef, <16 x i8> undef, i32 15, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0)
> + %r = call <4 x float> @llvm.SI.image.sample.c.v4i32(<4 x i32> undef, <8 x i32> undef, <4 x i32> undef, i32 15, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0)
> %r0 = extractelement <4 x float> %r, i32 0
> %r1 = extractelement <4 x float> %r, i32 1
> %r2 = extractelement <4 x float> %r, i32 2
> @@ -147,7 +147,7 @@ main_body:
> ;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:
> - %r = call <4 x float> @llvm.SI.image.sample.c.cl.v4i32(<4 x i32> undef, <32 x i8> undef, <16 x i8> undef, i32 15, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0)
> + %r = call <4 x float> @llvm.SI.image.sample.c.cl.v4i32(<4 x i32> undef, <8 x i32> undef, <4 x i32> undef, i32 15, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0)
> %r0 = extractelement <4 x float> %r, i32 0
> %r1 = extractelement <4 x float> %r, i32 1
> %r2 = extractelement <4 x float> %r, i32 2
> @@ -160,7 +160,7 @@ main_body:
> ;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:
> - %r = call <4 x float> @llvm.SI.image.sample.c.d.v4i32(<4 x i32> undef, <32 x i8> undef, <16 x i8> undef, i32 15, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0)
> + %r = call <4 x float> @llvm.SI.image.sample.c.d.v4i32(<4 x i32> undef, <8 x i32> undef, <4 x i32> undef, i32 15, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0)
> %r0 = extractelement <4 x float> %r, i32 0
> %r1 = extractelement <4 x float> %r, i32 1
> %r2 = extractelement <4 x float> %r, i32 2
> @@ -173,7 +173,7 @@ main_body:
> ;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:
> - %r = call <4 x float> @llvm.SI.image.sample.c.d.cl.v4i32(<4 x i32> undef, <32 x i8> undef, <16 x i8> undef, i32 15, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0)
> + %r = call <4 x float> @llvm.SI.image.sample.c.d.cl.v4i32(<4 x i32> undef, <8 x i32> undef, <4 x i32> undef, i32 15, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0)
> %r0 = extractelement <4 x float> %r, i32 0
> %r1 = extractelement <4 x float> %r, i32 1
> %r2 = extractelement <4 x float> %r, i32 2
> @@ -186,7 +186,7 @@ main_body:
> ;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:
> - %r = call <4 x float> @llvm.SI.image.sample.c.l.v4i32(<4 x i32> undef, <32 x i8> undef, <16 x i8> undef, i32 15, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0)
> + %r = call <4 x float> @llvm.SI.image.sample.c.l.v4i32(<4 x i32> undef, <8 x i32> undef, <4 x i32> undef, i32 15, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0)
> %r0 = extractelement <4 x float> %r, i32 0
> %r1 = extractelement <4 x float> %r, i32 1
> %r2 = extractelement <4 x float> %r, i32 2
> @@ -199,7 +199,7 @@ main_body:
> ;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:
> - %r = call <4 x float> @llvm.SI.image.sample.c.b.v4i32(<4 x i32> undef, <32 x i8> undef, <16 x i8> undef, i32 15, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0)
> + %r = call <4 x float> @llvm.SI.image.sample.c.b.v4i32(<4 x i32> undef, <8 x i32> undef, <4 x i32> undef, i32 15, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0)
> %r0 = extractelement <4 x float> %r, i32 0
> %r1 = extractelement <4 x float> %r, i32 1
> %r2 = extractelement <4 x float> %r, i32 2
> @@ -212,7 +212,7 @@ main_body:
> ;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:
> - %r = call <4 x float> @llvm.SI.image.sample.c.b.cl.v4i32(<4 x i32> undef, <32 x i8> undef, <16 x i8> undef, i32 15, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0)
> + %r = call <4 x float> @llvm.SI.image.sample.c.b.cl.v4i32(<4 x i32> undef, <8 x i32> undef, <4 x i32> undef, i32 15, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0)
> %r0 = extractelement <4 x float> %r, i32 0
> %r1 = extractelement <4 x float> %r, i32 1
> %r2 = extractelement <4 x float> %r, i32 2
> @@ -225,7 +225,7 @@ main_body:
> ;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:
> - %r = call <4 x float> @llvm.SI.image.sample.c.lz.v4i32(<4 x i32> undef, <32 x i8> undef, <16 x i8> undef, i32 15, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0)
> + %r = call <4 x float> @llvm.SI.image.sample.c.lz.v4i32(<4 x i32> undef, <8 x i32> undef, <4 x i32> undef, i32 15, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0)
> %r0 = extractelement <4 x float> %r, i32 0
> %r1 = extractelement <4 x float> %r, i32 1
> %r2 = extractelement <4 x float> %r, i32 2
> @@ -238,7 +238,7 @@ main_body:
> ;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:
> - %r = call <4 x float> @llvm.SI.image.sample.c.cd.v4i32(<4 x i32> undef, <32 x i8> undef, <16 x i8> undef, i32 15, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0)
> + %r = call <4 x float> @llvm.SI.image.sample.c.cd.v4i32(<4 x i32> undef, <8 x i32> undef, <4 x i32> undef, i32 15, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0)
> %r0 = extractelement <4 x float> %r, i32 0
> %r1 = extractelement <4 x float> %r, i32 1
> %r2 = extractelement <4 x float> %r, i32 2
> @@ -251,7 +251,7 @@ main_body:
> ;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:
> - %r = call <4 x float> @llvm.SI.image.sample.c.cd.cl.v4i32(<4 x i32> undef, <32 x i8> undef, <16 x i8> undef, i32 15, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0)
> + %r = call <4 x float> @llvm.SI.image.sample.c.cd.cl.v4i32(<4 x i32> undef, <8 x i32> undef, <4 x i32> undef, i32 15, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0)
> %r0 = extractelement <4 x float> %r, i32 0
> %r1 = extractelement <4 x float> %r, i32 1
> %r2 = extractelement <4 x float> %r, i32 2
> @@ -261,27 +261,27 @@ main_body:
> }
>
>
> -declare <4 x float> @llvm.SI.image.sample.v4i32(<4 x i32>, <32 x i8>, <16 x i8>, i32, i32, i32, i32, i32, i32, i32, i32) #1
> -declare <4 x float> @llvm.SI.image.sample.cl.v4i32(<4 x i32>, <32 x i8>, <16 x i8>, i32, i32, i32, i32, i32, i32, i32, i32) #1
> -declare <4 x float> @llvm.SI.image.sample.d.v4i32(<4 x i32>, <32 x i8>, <16 x i8>, i32, i32, i32, i32, i32, i32, i32, i32) #1
> -declare <4 x float> @llvm.SI.image.sample.d.cl.v4i32(<4 x i32>, <32 x i8>, <16 x i8>, i32, i32, i32, i32, i32, i32, i32, i32) #1
> -declare <4 x float> @llvm.SI.image.sample.l.v4i32(<4 x i32>, <32 x i8>, <16 x i8>, i32, i32, i32, i32, i32, i32, i32, i32) #1
> -declare <4 x float> @llvm.SI.image.sample.b.v4i32(<4 x i32>, <32 x i8>, <16 x i8>, i32, i32, i32, i32, i32, i32, i32, i32) #1
> -declare <4 x float> @llvm.SI.image.sample.b.cl.v4i32(<4 x i32>, <32 x i8>, <16 x i8>, i32, i32, i32, i32, i32, i32, i32, i32) #1
> -declare <4 x float> @llvm.SI.image.sample.lz.v4i32(<4 x i32>, <32 x i8>, <16 x i8>, i32, i32, i32, i32, i32, i32, i32, i32) #1
> -declare <4 x float> @llvm.SI.image.sample.cd.v4i32(<4 x i32>, <32 x i8>, <16 x i8>, i32, i32, i32, i32, i32, i32, i32, i32) #1
> -declare <4 x float> @llvm.SI.image.sample.cd.cl.v4i32(<4 x i32>, <32 x i8>, <16 x i8>, i32, i32, i32, i32, i32, i32, i32, i32) #1
> +declare <4 x float> @llvm.SI.image.sample.v4i32(<4 x i32>, <8 x i32>, <4 x i32>, i32, i32, i32, i32, i32, i32, i32, i32) #1
> +declare <4 x float> @llvm.SI.image.sample.cl.v4i32(<4 x i32>, <8 x i32>, <4 x i32>, i32, i32, i32, i32, i32, i32, i32, i32) #1
> +declare <4 x float> @llvm.SI.image.sample.d.v4i32(<4 x i32>, <8 x i32>, <4 x i32>, i32, i32, i32, i32, i32, i32, i32, i32) #1
> +declare <4 x float> @llvm.SI.image.sample.d.cl.v4i32(<4 x i32>, <8 x i32>, <4 x i32>, i32, i32, i32, i32, i32, i32, i32, i32) #1
> +declare <4 x float> @llvm.SI.image.sample.l.v4i32(<4 x i32>, <8 x i32>, <4 x i32>, i32, i32, i32, i32, i32, i32, i32, i32) #1
> +declare <4 x float> @llvm.SI.image.sample.b.v4i32(<4 x i32>, <8 x i32>, <4 x i32>, i32, i32, i32, i32, i32, i32, i32, i32) #1
> +declare <4 x float> @llvm.SI.image.sample.b.cl.v4i32(<4 x i32>, <8 x i32>, <4 x i32>, i32, i32, i32, i32, i32, i32, i32, i32) #1
> +declare <4 x float> @llvm.SI.image.sample.lz.v4i32(<4 x i32>, <8 x i32>, <4 x i32>, i32, i32, i32, i32, i32, i32, i32, i32) #1
> +declare <4 x float> @llvm.SI.image.sample.cd.v4i32(<4 x i32>, <8 x i32>, <4 x i32>, i32, i32, i32, i32, i32, i32, i32, i32) #1
> +declare <4 x float> @llvm.SI.image.sample.cd.cl.v4i32(<4 x i32>, <8 x i32>, <4 x i32>, i32, i32, i32, i32, i32, i32, i32, i32) #1
>
> -declare <4 x float> @llvm.SI.image.sample.c.v4i32(<4 x i32>, <32 x i8>, <16 x i8>, i32, i32, i32, i32, i32, i32, i32, i32) #1
> -declare <4 x float> @llvm.SI.image.sample.c.cl.v4i32(<4 x i32>, <32 x i8>, <16 x i8>, i32, i32, i32, i32, i32, i32, i32, i32) #1
> -declare <4 x float> @llvm.SI.image.sample.c.d.v4i32(<4 x i32>, <32 x i8>, <16 x i8>, i32, i32, i32, i32, i32, i32, i32, i32) #1
> -declare <4 x float> @llvm.SI.image.sample.c.d.cl.v4i32(<4 x i32>, <32 x i8>, <16 x i8>, i32, i32, i32, i32, i32, i32, i32, i32) #1
> -declare <4 x float> @llvm.SI.image.sample.c.l.v4i32(<4 x i32>, <32 x i8>, <16 x i8>, i32, i32, i32, i32, i32, i32, i32, i32) #1
> -declare <4 x float> @llvm.SI.image.sample.c.b.v4i32(<4 x i32>, <32 x i8>, <16 x i8>, i32, i32, i32, i32, i32, i32, i32, i32) #1
> -declare <4 x float> @llvm.SI.image.sample.c.b.cl.v4i32(<4 x i32>, <32 x i8>, <16 x i8>, i32, i32, i32, i32, i32, i32, i32, i32) #1
> -declare <4 x float> @llvm.SI.image.sample.c.lz.v4i32(<4 x i32>, <32 x i8>, <16 x i8>, i32, i32, i32, i32, i32, i32, i32, i32) #1
> -declare <4 x float> @llvm.SI.image.sample.c.cd.v4i32(<4 x i32>, <32 x i8>, <16 x i8>, i32, i32, i32, i32, i32, i32, i32, i32) #1
> -declare <4 x float> @llvm.SI.image.sample.c.cd.cl.v4i32(<4 x i32>, <32 x i8>, <16 x i8>, i32, i32, i32, i32, i32, i32, i32, i32) #1
> +declare <4 x float> @llvm.SI.image.sample.c.v4i32(<4 x i32>, <8 x i32>, <4 x i32>, i32, i32, i32, i32, i32, i32, i32, i32) #1
> +declare <4 x float> @llvm.SI.image.sample.c.cl.v4i32(<4 x i32>, <8 x i32>, <4 x i32>, i32, i32, i32, i32, i32, i32, i32, i32) #1
> +declare <4 x float> @llvm.SI.image.sample.c.d.v4i32(<4 x i32>, <8 x i32>, <4 x i32>, i32, i32, i32, i32, i32, i32, i32, i32) #1
> +declare <4 x float> @llvm.SI.image.sample.c.d.cl.v4i32(<4 x i32>, <8 x i32>, <4 x i32>, i32, i32, i32, i32, i32, i32, i32, i32) #1
> +declare <4 x float> @llvm.SI.image.sample.c.l.v4i32(<4 x i32>, <8 x i32>, <4 x i32>, i32, i32, i32, i32, i32, i32, i32, i32) #1
> +declare <4 x float> @llvm.SI.image.sample.c.b.v4i32(<4 x i32>, <8 x i32>, <4 x i32>, i32, i32, i32, i32, i32, i32, i32, i32) #1
> +declare <4 x float> @llvm.SI.image.sample.c.b.cl.v4i32(<4 x i32>, <8 x i32>, <4 x i32>, i32, i32, i32, i32, i32, i32, i32, i32) #1
> +declare <4 x float> @llvm.SI.image.sample.c.lz.v4i32(<4 x i32>, <8 x i32>, <4 x i32>, i32, i32, i32, i32, i32, i32, i32, i32) #1
> +declare <4 x float> @llvm.SI.image.sample.c.cd.v4i32(<4 x i32>, <8 x i32>, <4 x i32>, i32, i32, i32, i32, i32, i32, i32, i32) #1
> +declare <4 x float> @llvm.SI.image.sample.c.cd.cl.v4i32(<4 x i32>, <8 x i32>, <4 x i32>, i32, i32, i32, i32, i32, i32, i32, i32) #1
>
> declare void @llvm.SI.export(i32, i32, i32, i32, i32, float, float, float, float)
>
> diff --git a/test/CodeGen/R600/llvm.SI.image.sample.o.ll b/test/CodeGen/R600/llvm.SI.image.sample.o.ll
> index 6a79466..ed3ef91 100644
> --- a/test/CodeGen/R600/llvm.SI.image.sample.o.ll
> +++ b/test/CodeGen/R600/llvm.SI.image.sample.o.ll
> @@ -4,7 +4,7 @@
> ;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:
> - %r = call <4 x float> @llvm.SI.image.sample.o.v4i32(<4 x i32> undef, <32 x i8> undef, <16 x i8> undef, i32 15, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0)
> + %r = call <4 x float> @llvm.SI.image.sample.o.v4i32(<4 x i32> undef, <8 x i32> undef, <4 x i32> undef, i32 15, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0)
> %r0 = extractelement <4 x float> %r, i32 0
> %r1 = extractelement <4 x float> %r, i32 1
> %r2 = extractelement <4 x float> %r, i32 2
> @@ -17,7 +17,7 @@ main_body:
> ;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:
> - %r = call <4 x float> @llvm.SI.image.sample.cl.o.v4i32(<4 x i32> undef, <32 x i8> undef, <16 x i8> undef, i32 15, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0)
> + %r = call <4 x float> @llvm.SI.image.sample.cl.o.v4i32(<4 x i32> undef, <8 x i32> undef, <4 x i32> undef, i32 15, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0)
> %r0 = extractelement <4 x float> %r, i32 0
> %r1 = extractelement <4 x float> %r, i32 1
> %r2 = extractelement <4 x float> %r, i32 2
> @@ -30,7 +30,7 @@ main_body:
> ;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:
> - %r = call <4 x float> @llvm.SI.image.sample.d.o.v4i32(<4 x i32> undef, <32 x i8> undef, <16 x i8> undef, i32 15, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0)
> + %r = call <4 x float> @llvm.SI.image.sample.d.o.v4i32(<4 x i32> undef, <8 x i32> undef, <4 x i32> undef, i32 15, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0)
> %r0 = extractelement <4 x float> %r, i32 0
> %r1 = extractelement <4 x float> %r, i32 1
> %r2 = extractelement <4 x float> %r, i32 2
> @@ -43,7 +43,7 @@ main_body:
> ;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:
> - %r = call <4 x float> @llvm.SI.image.sample.d.cl.o.v4i32(<4 x i32> undef, <32 x i8> undef, <16 x i8> undef, i32 15, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0)
> + %r = call <4 x float> @llvm.SI.image.sample.d.cl.o.v4i32(<4 x i32> undef, <8 x i32> undef, <4 x i32> undef, i32 15, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0)
> %r0 = extractelement <4 x float> %r, i32 0
> %r1 = extractelement <4 x float> %r, i32 1
> %r2 = extractelement <4 x float> %r, i32 2
> @@ -56,7 +56,7 @@ main_body:
> ;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:
> - %r = call <4 x float> @llvm.SI.image.sample.l.o.v4i32(<4 x i32> undef, <32 x i8> undef, <16 x i8> undef, i32 15, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0)
> + %r = call <4 x float> @llvm.SI.image.sample.l.o.v4i32(<4 x i32> undef, <8 x i32> undef, <4 x i32> undef, i32 15, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0)
> %r0 = extractelement <4 x float> %r, i32 0
> %r1 = extractelement <4 x float> %r, i32 1
> %r2 = extractelement <4 x float> %r, i32 2
> @@ -69,7 +69,7 @@ main_body:
> ;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:
> - %r = call <4 x float> @llvm.SI.image.sample.b.o.v4i32(<4 x i32> undef, <32 x i8> undef, <16 x i8> undef, i32 15, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0)
> + %r = call <4 x float> @llvm.SI.image.sample.b.o.v4i32(<4 x i32> undef, <8 x i32> undef, <4 x i32> undef, i32 15, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0)
> %r0 = extractelement <4 x float> %r, i32 0
> %r1 = extractelement <4 x float> %r, i32 1
> %r2 = extractelement <4 x float> %r, i32 2
> @@ -82,7 +82,7 @@ main_body:
> ;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:
> - %r = call <4 x float> @llvm.SI.image.sample.b.cl.o.v4i32(<4 x i32> undef, <32 x i8> undef, <16 x i8> undef, i32 15, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0)
> + %r = call <4 x float> @llvm.SI.image.sample.b.cl.o.v4i32(<4 x i32> undef, <8 x i32> undef, <4 x i32> undef, i32 15, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0)
> %r0 = extractelement <4 x float> %r, i32 0
> %r1 = extractelement <4 x float> %r, i32 1
> %r2 = extractelement <4 x float> %r, i32 2
> @@ -95,7 +95,7 @@ main_body:
> ;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:
> - %r = call <4 x float> @llvm.SI.image.sample.lz.o.v4i32(<4 x i32> undef, <32 x i8> undef, <16 x i8> undef, i32 15, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0)
> + %r = call <4 x float> @llvm.SI.image.sample.lz.o.v4i32(<4 x i32> undef, <8 x i32> undef, <4 x i32> undef, i32 15, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0)
> %r0 = extractelement <4 x float> %r, i32 0
> %r1 = extractelement <4 x float> %r, i32 1
> %r2 = extractelement <4 x float> %r, i32 2
> @@ -108,7 +108,7 @@ main_body:
> ;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:
> - %r = call <4 x float> @llvm.SI.image.sample.cd.o.v4i32(<4 x i32> undef, <32 x i8> undef, <16 x i8> undef, i32 15, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0)
> + %r = call <4 x float> @llvm.SI.image.sample.cd.o.v4i32(<4 x i32> undef, <8 x i32> undef, <4 x i32> undef, i32 15, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0)
> %r0 = extractelement <4 x float> %r, i32 0
> %r1 = extractelement <4 x float> %r, i32 1
> %r2 = extractelement <4 x float> %r, i32 2
> @@ -121,7 +121,7 @@ main_body:
> ;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:
> - %r = call <4 x float> @llvm.SI.image.sample.cd.cl.o.v4i32(<4 x i32> undef, <32 x i8> undef, <16 x i8> undef, i32 15, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0)
> + %r = call <4 x float> @llvm.SI.image.sample.cd.cl.o.v4i32(<4 x i32> undef, <8 x i32> undef, <4 x i32> undef, i32 15, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0)
> %r0 = extractelement <4 x float> %r, i32 0
> %r1 = extractelement <4 x float> %r, i32 1
> %r2 = extractelement <4 x float> %r, i32 2
> @@ -134,7 +134,7 @@ main_body:
> ;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:
> - %r = call <4 x float> @llvm.SI.image.sample.c.o.v4i32(<4 x i32> undef, <32 x i8> undef, <16 x i8> undef, i32 15, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0)
> + %r = call <4 x float> @llvm.SI.image.sample.c.o.v4i32(<4 x i32> undef, <8 x i32> undef, <4 x i32> undef, i32 15, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0)
> %r0 = extractelement <4 x float> %r, i32 0
> %r1 = extractelement <4 x float> %r, i32 1
> %r2 = extractelement <4 x float> %r, i32 2
> @@ -147,7 +147,7 @@ main_body:
> ;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:
> - %r = call <4 x float> @llvm.SI.image.sample.c.cl.o.v4i32(<4 x i32> undef, <32 x i8> undef, <16 x i8> undef, i32 15, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0)
> + %r = call <4 x float> @llvm.SI.image.sample.c.cl.o.v4i32(<4 x i32> undef, <8 x i32> undef, <4 x i32> undef, i32 15, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0)
> %r0 = extractelement <4 x float> %r, i32 0
> %r1 = extractelement <4 x float> %r, i32 1
> %r2 = extractelement <4 x float> %r, i32 2
> @@ -160,7 +160,7 @@ main_body:
> ;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:
> - %r = call <4 x float> @llvm.SI.image.sample.c.d.o.v4i32(<4 x i32> undef, <32 x i8> undef, <16 x i8> undef, i32 15, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0)
> + %r = call <4 x float> @llvm.SI.image.sample.c.d.o.v4i32(<4 x i32> undef, <8 x i32> undef, <4 x i32> undef, i32 15, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0)
> %r0 = extractelement <4 x float> %r, i32 0
> %r1 = extractelement <4 x float> %r, i32 1
> %r2 = extractelement <4 x float> %r, i32 2
> @@ -173,7 +173,7 @@ main_body:
> ;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:
> - %r = call <4 x float> @llvm.SI.image.sample.c.d.cl.o.v4i32(<4 x i32> undef, <32 x i8> undef, <16 x i8> undef, i32 15, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0)
> + %r = call <4 x float> @llvm.SI.image.sample.c.d.cl.o.v4i32(<4 x i32> undef, <8 x i32> undef, <4 x i32> undef, i32 15, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0)
> %r0 = extractelement <4 x float> %r, i32 0
> %r1 = extractelement <4 x float> %r, i32 1
> %r2 = extractelement <4 x float> %r, i32 2
> @@ -186,7 +186,7 @@ main_body:
> ;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:
> - %r = call <4 x float> @llvm.SI.image.sample.c.l.o.v4i32(<4 x i32> undef, <32 x i8> undef, <16 x i8> undef, i32 15, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0)
> + %r = call <4 x float> @llvm.SI.image.sample.c.l.o.v4i32(<4 x i32> undef, <8 x i32> undef, <4 x i32> undef, i32 15, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0)
> %r0 = extractelement <4 x float> %r, i32 0
> %r1 = extractelement <4 x float> %r, i32 1
> %r2 = extractelement <4 x float> %r, i32 2
> @@ -199,7 +199,7 @@ main_body:
> ;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:
> - %r = call <4 x float> @llvm.SI.image.sample.c.b.o.v4i32(<4 x i32> undef, <32 x i8> undef, <16 x i8> undef, i32 15, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0)
> + %r = call <4 x float> @llvm.SI.image.sample.c.b.o.v4i32(<4 x i32> undef, <8 x i32> undef, <4 x i32> undef, i32 15, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0)
> %r0 = extractelement <4 x float> %r, i32 0
> %r1 = extractelement <4 x float> %r, i32 1
> %r2 = extractelement <4 x float> %r, i32 2
> @@ -212,7 +212,7 @@ main_body:
> ;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:
> - %r = call <4 x float> @llvm.SI.image.sample.c.b.cl.o.v4i32(<4 x i32> undef, <32 x i8> undef, <16 x i8> undef, i32 15, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0)
> + %r = call <4 x float> @llvm.SI.image.sample.c.b.cl.o.v4i32(<4 x i32> undef, <8 x i32> undef, <4 x i32> undef, i32 15, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0)
> %r0 = extractelement <4 x float> %r, i32 0
> %r1 = extractelement <4 x float> %r, i32 1
> %r2 = extractelement <4 x float> %r, i32 2
> @@ -225,7 +225,7 @@ main_body:
> ;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:
> - %r = call <4 x float> @llvm.SI.image.sample.c.lz.o.v4i32(<4 x i32> undef, <32 x i8> undef, <16 x i8> undef, i32 15, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0)
> + %r = call <4 x float> @llvm.SI.image.sample.c.lz.o.v4i32(<4 x i32> undef, <8 x i32> undef, <4 x i32> undef, i32 15, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0)
> %r0 = extractelement <4 x float> %r, i32 0
> %r1 = extractelement <4 x float> %r, i32 1
> %r2 = extractelement <4 x float> %r, i32 2
> @@ -238,7 +238,7 @@ main_body:
> ;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:
> - %r = call <4 x float> @llvm.SI.image.sample.c.cd.o.v4i32(<4 x i32> undef, <32 x i8> undef, <16 x i8> undef, i32 15, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0)
> + %r = call <4 x float> @llvm.SI.image.sample.c.cd.o.v4i32(<4 x i32> undef, <8 x i32> undef, <4 x i32> undef, i32 15, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0)
> %r0 = extractelement <4 x float> %r, i32 0
> %r1 = extractelement <4 x float> %r, i32 1
> %r2 = extractelement <4 x float> %r, i32 2
> @@ -251,7 +251,7 @@ main_body:
> ;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:
> - %r = call <4 x float> @llvm.SI.image.sample.c.cd.cl.o.v4i32(<4 x i32> undef, <32 x i8> undef, <16 x i8> undef, i32 15, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0)
> + %r = call <4 x float> @llvm.SI.image.sample.c.cd.cl.o.v4i32(<4 x i32> undef, <8 x i32> undef, <4 x i32> undef, i32 15, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0)
> %r0 = extractelement <4 x float> %r, i32 0
> %r1 = extractelement <4 x float> %r, i32 1
> %r2 = extractelement <4 x float> %r, i32 2
> @@ -261,27 +261,27 @@ main_body:
> }
>
>
> -declare <4 x float> @llvm.SI.image.sample.o.v4i32(<4 x i32>, <32 x i8>, <16 x i8>, i32, i32, i32, i32, i32, i32, i32, i32) #1
> -declare <4 x float> @llvm.SI.image.sample.cl.o.v4i32(<4 x i32>, <32 x i8>, <16 x i8>, i32, i32, i32, i32, i32, i32, i32, i32) #1
> -declare <4 x float> @llvm.SI.image.sample.d.o.v4i32(<4 x i32>, <32 x i8>, <16 x i8>, i32, i32, i32, i32, i32, i32, i32, i32) #1
> -declare <4 x float> @llvm.SI.image.sample.d.cl.o.v4i32(<4 x i32>, <32 x i8>, <16 x i8>, i32, i32, i32, i32, i32, i32, i32, i32) #1
> -declare <4 x float> @llvm.SI.image.sample.l.o.v4i32(<4 x i32>, <32 x i8>, <16 x i8>, i32, i32, i32, i32, i32, i32, i32, i32) #1
> -declare <4 x float> @llvm.SI.image.sample.b.o.v4i32(<4 x i32>, <32 x i8>, <16 x i8>, i32, i32, i32, i32, i32, i32, i32, i32) #1
> -declare <4 x float> @llvm.SI.image.sample.b.cl.o.v4i32(<4 x i32>, <32 x i8>, <16 x i8>, i32, i32, i32, i32, i32, i32, i32, i32) #1
> -declare <4 x float> @llvm.SI.image.sample.lz.o.v4i32(<4 x i32>, <32 x i8>, <16 x i8>, i32, i32, i32, i32, i32, i32, i32, i32) #1
> -declare <4 x float> @llvm.SI.image.sample.cd.o.v4i32(<4 x i32>, <32 x i8>, <16 x i8>, i32, i32, i32, i32, i32, i32, i32, i32) #1
> -declare <4 x float> @llvm.SI.image.sample.cd.cl.o.v4i32(<4 x i32>, <32 x i8>, <16 x i8>, i32, i32, i32, i32, i32, i32, i32, i32) #1
> +declare <4 x float> @llvm.SI.image.sample.o.v4i32(<4 x i32>, <8 x i32>, <4 x i32>, i32, i32, i32, i32, i32, i32, i32, i32) #1
> +declare <4 x float> @llvm.SI.image.sample.cl.o.v4i32(<4 x i32>, <8 x i32>, <4 x i32>, i32, i32, i32, i32, i32, i32, i32, i32) #1
> +declare <4 x float> @llvm.SI.image.sample.d.o.v4i32(<4 x i32>, <8 x i32>, <4 x i32>, i32, i32, i32, i32, i32, i32, i32, i32) #1
> +declare <4 x float> @llvm.SI.image.sample.d.cl.o.v4i32(<4 x i32>, <8 x i32>, <4 x i32>, i32, i32, i32, i32, i32, i32, i32, i32) #1
> +declare <4 x float> @llvm.SI.image.sample.l.o.v4i32(<4 x i32>, <8 x i32>, <4 x i32>, i32, i32, i32, i32, i32, i32, i32, i32) #1
> +declare <4 x float> @llvm.SI.image.sample.b.o.v4i32(<4 x i32>, <8 x i32>, <4 x i32>, i32, i32, i32, i32, i32, i32, i32, i32) #1
> +declare <4 x float> @llvm.SI.image.sample.b.cl.o.v4i32(<4 x i32>, <8 x i32>, <4 x i32>, i32, i32, i32, i32, i32, i32, i32, i32) #1
> +declare <4 x float> @llvm.SI.image.sample.lz.o.v4i32(<4 x i32>, <8 x i32>, <4 x i32>, i32, i32, i32, i32, i32, i32, i32, i32) #1
> +declare <4 x float> @llvm.SI.image.sample.cd.o.v4i32(<4 x i32>, <8 x i32>, <4 x i32>, i32, i32, i32, i32, i32, i32, i32, i32) #1
> +declare <4 x float> @llvm.SI.image.sample.cd.cl.o.v4i32(<4 x i32>, <8 x i32>, <4 x i32>, i32, i32, i32, i32, i32, i32, i32, i32) #1
>
> -declare <4 x float> @llvm.SI.image.sample.c.o.v4i32(<4 x i32>, <32 x i8>, <16 x i8>, i32, i32, i32, i32, i32, i32, i32, i32) #1
> -declare <4 x float> @llvm.SI.image.sample.c.cl.o.v4i32(<4 x i32>, <32 x i8>, <16 x i8>, i32, i32, i32, i32, i32, i32, i32, i32) #1
> -declare <4 x float> @llvm.SI.image.sample.c.d.o.v4i32(<4 x i32>, <32 x i8>, <16 x i8>, i32, i32, i32, i32, i32, i32, i32, i32) #1
> -declare <4 x float> @llvm.SI.image.sample.c.d.cl.o.v4i32(<4 x i32>, <32 x i8>, <16 x i8>, i32, i32, i32, i32, i32, i32, i32, i32) #1
> -declare <4 x float> @llvm.SI.image.sample.c.l.o.v4i32(<4 x i32>, <32 x i8>, <16 x i8>, i32, i32, i32, i32, i32, i32, i32, i32) #1
> -declare <4 x float> @llvm.SI.image.sample.c.b.o.v4i32(<4 x i32>, <32 x i8>, <16 x i8>, i32, i32, i32, i32, i32, i32, i32, i32) #1
> -declare <4 x float> @llvm.SI.image.sample.c.b.cl.o.v4i32(<4 x i32>, <32 x i8>, <16 x i8>, i32, i32, i32, i32, i32, i32, i32, i32) #1
> -declare <4 x float> @llvm.SI.image.sample.c.lz.o.v4i32(<4 x i32>, <32 x i8>, <16 x i8>, i32, i32, i32, i32, i32, i32, i32, i32) #1
> -declare <4 x float> @llvm.SI.image.sample.c.cd.o.v4i32(<4 x i32>, <32 x i8>, <16 x i8>, i32, i32, i32, i32, i32, i32, i32, i32) #1
> -declare <4 x float> @llvm.SI.image.sample.c.cd.cl.o.v4i32(<4 x i32>, <32 x i8>, <16 x i8>, i32, i32, i32, i32, i32, i32, i32, i32) #1
> +declare <4 x float> @llvm.SI.image.sample.c.o.v4i32(<4 x i32>, <8 x i32>, <4 x i32>, i32, i32, i32, i32, i32, i32, i32, i32) #1
> +declare <4 x float> @llvm.SI.image.sample.c.cl.o.v4i32(<4 x i32>, <8 x i32>, <4 x i32>, i32, i32, i32, i32, i32, i32, i32, i32) #1
> +declare <4 x float> @llvm.SI.image.sample.c.d.o.v4i32(<4 x i32>, <8 x i32>, <4 x i32>, i32, i32, i32, i32, i32, i32, i32, i32) #1
> +declare <4 x float> @llvm.SI.image.sample.c.d.cl.o.v4i32(<4 x i32>, <8 x i32>, <4 x i32>, i32, i32, i32, i32, i32, i32, i32, i32) #1
> +declare <4 x float> @llvm.SI.image.sample.c.l.o.v4i32(<4 x i32>, <8 x i32>, <4 x i32>, i32, i32, i32, i32, i32, i32, i32, i32) #1
> +declare <4 x float> @llvm.SI.image.sample.c.b.o.v4i32(<4 x i32>, <8 x i32>, <4 x i32>, i32, i32, i32, i32, i32, i32, i32, i32) #1
> +declare <4 x float> @llvm.SI.image.sample.c.b.cl.o.v4i32(<4 x i32>, <8 x i32>, <4 x i32>, i32, i32, i32, i32, i32, i32, i32, i32) #1
> +declare <4 x float> @llvm.SI.image.sample.c.lz.o.v4i32(<4 x i32>, <8 x i32>, <4 x i32>, i32, i32, i32, i32, i32, i32, i32, i32) #1
> +declare <4 x float> @llvm.SI.image.sample.c.cd.o.v4i32(<4 x i32>, <8 x i32>, <4 x i32>, i32, i32, i32, i32, i32, i32, i32, i32) #1
> +declare <4 x float> @llvm.SI.image.sample.c.cd.cl.o.v4i32(<4 x i32>, <8 x i32>, <4 x i32>, i32, i32, i32, i32, i32, i32, i32, i32) #1
>
> declare void @llvm.SI.export(i32, i32, i32, i32, i32, float, float, float, float)
>
> --
> 1.9.1
>
> _______________________________________________
> mesa-dev mailing list
> mesa-dev at lists.freedesktop.org
> http://lists.freedesktop.org/mailman/listinfo/mesa-dev
More information about the llvm-commits
mailing list