llvm/test/CodeGen/NVPTX/blocksareclusters-kernel-attr.ll

; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py UTC_ARGS: --version 5
; RUN: llc < %s -mcpu=sm_90 -mattr=+ptx90 | FileCheck %s

target triple = "nvptx64-nvidia-cuda"

; Test "blocksareclusters" attribute with full "reqntid" and "cluster_dim"
; attributes.
define ptx_kernel void @kernel1(ptr %input, ptr %output) #0 #1 #2 {
; CHECK-LABEL: kernel1(
; CHECK:       .reqntid 1024, 1, 1
; CHECK-NEXT:  .reqnctapercluster 2, 2, 2
; CHECK-NEXT:  .blocksareclusters
; CHECK-NEXT:  {
; CHECK-EMPTY:
; CHECK-EMPTY:
; CHECK-NEXT:  // %bb.0:
; CHECK-NEXT:    ret;
  ret void
}

; Test "blocksareclusters" attribute with single dimension "reqntid" and
; "cluster_dim" attributes.
define ptx_kernel void @kernel2(ptr %input, ptr %output) #0 #3 #4 {
; CHECK-LABEL: kernel2(
; CHECK:       .reqntid 1024
; CHECK-NEXT:  .reqnctapercluster 2
; CHECK-NEXT:  .blocksareclusters // @kernel2
; CHECK-NEXT:  {
; CHECK-EMPTY:
; CHECK-EMPTY:
; CHECK-NEXT:  // %bb.0:
; CHECK-NEXT:    ret;
  ret void
}

; Test "blocksareclusters" attribute with two dimensions(not z dimension)
; "reqntid" and "cluster_dim" attributes.
define ptx_kernel void @kernel3(ptr %input, ptr %output) #0 #5 #6 {
; CHECK-LABEL: kernel3(
; CHECK:       .reqntid 512, 2
; CHECK-NEXT:  .reqnctapercluster 2, 2
; CHECK-NEXT:  .blocksareclusters // @kernel3
; CHECK-NEXT:  {
; CHECK-EMPTY:
; CHECK-EMPTY:
; CHECK-NEXT:  // %bb.0:
; CHECK-NEXT:    ret;
  ret void
}

attributes #0 = { "nvvm.blocksareclusters" }

attributes #1 = { "nvvm.reqntid"="1024,1,1" }
attributes #2 = { "nvvm.cluster_dim"="2,2,2" }

attributes #3 = { "nvvm.reqntid"="1024" }
attributes #4 = { "nvvm.cluster_dim"="2" }

attributes #5 = { "nvvm.reqntid"="512,2" }
attributes #6 = { "nvvm.cluster_dim"="2,2" }