SelectionDAG: Lower some range metadata to AssertZext

If a range has a lower bound of 0, add an AssertZext from the
nearest floor power of two.

This allows operations with some workitem intrinsics with known
maximum ranges to use fast 24-bit multiplies.

llvm-svn: 260109

2 files changed, 90 insertions, 0 deletions

diff --git a/llvm/test/CodeGen/AArch64/lower-range-metadata-func-call.ll b/llvm/test/CodeGen/AArch64/lower-range-metadata-func-call.ll
new file mode 100644
index 00000000000..fd4b2f5ba30
--- /dev/null
+++ b/llvm/test/CodeGen/AArch64/lower-range-metadata-func-call.ll
@@ -0,0 +1,44 @@
+; RUN: llc -march=aarch64 -mtriple=aarch64-none-linux-gnu < %s | FileCheck %s
+
+; and can be eliminated
+; CHECK-LABEL: {{^}}test_call_known_max_range:
+; CHECK: bl foo
+; CHECK-NOT: and
+; CHECK: ret
+define i32 @test_call_known_max_range() #0 {
+entry:
+  %id = tail call i32 @foo(), !range !0
+  %and = and i32 %id, 1023
+  ret i32 %and
+}
+
+; CHECK-LABEL: {{^}}test_call_known_trunc_1_bit_range:
+; CHECK: bl foo
+; CHECK: and w{{[0-9]+}}, w0, #0x1ff
+; CHECK: ret
+define i32 @test_call_known_trunc_1_bit_range() #0 {
+entry:
+  %id = tail call i32 @foo(), !range !0
+  %and = and i32 %id, 511
+  ret i32 %and
+}
+
+; CHECK-LABEL: {{^}}test_call_known_max_range_m1:
+; CHECK: bl foo
+; CHECK: and w{{[0-9]+}}, w0, #0xff
+; CHECK: ret
+define i32 @test_call_known_max_range_m1() #0 {
+entry:
+  %id = tail call i32 @foo(), !range !1
+  %and = and i32 %id, 255
+  ret i32 %and
+}
+
+
+declare i32 @foo()
+
+attributes #0 = { norecurse nounwind }
+attributes #1 = { nounwind readnone }
+
+!0 = !{i32 0, i32 1024}
+!1 = !{i32 0, i32 1023}
diff --git a/llvm/test/CodeGen/AMDGPU/lower-range-metadata-intrinsic-call.ll b/llvm/test/CodeGen/AMDGPU/lower-range-metadata-intrinsic-call.ll
new file mode 100644
index 00000000000..b6a9179a212
--- /dev/null
+++ b/llvm/test/CodeGen/AMDGPU/lower-range-metadata-intrinsic-call.ll
@@ -0,0 +1,46 @@
+; RUN: llc -march=amdgcn -mtriple=amdgcn-unknown-amdhsa < %s | FileCheck %s
+; RUN: llc -march=amdgcn -mtriple=amdgcn-unknown-unknown < %s | FileCheck %s
+
+; and can be eliminated
+; CHECK-LABEL: {{^}}test_workitem_id_x_known_max_range:
+; CHECK-NOT: v0
+; CHECK: {{flat|buffer}}_store_dword v0
+define void @test_workitem_id_x_known_max_range(i32 addrspace(1)* nocapture %out) #0 {
+entry:
+  %id = tail call i32 @llvm.amdgcn.workitem.id.x(), !range !0
+  %and = and i32 %id, 1023
+  store i32 %and, i32 addrspace(1)* %out, align 4
+  ret void
+}
+
+; CHECK-LABEL: {{^}}test_workitem_id_x_known_trunc_1_bit_range:
+; CHECK: v_and_b32_e32 [[MASKED:v[0-9]+]], 0x1ff, v0
+; CHECK: {{flat|buffer}}_store_dword [[MASKED]]
+define void @test_workitem_id_x_known_trunc_1_bit_range(i32 addrspace(1)* nocapture %out) #0 {
+entry:
+  %id = tail call i32 @llvm.amdgcn.workitem.id.x(), !range !0
+  %and = and i32 %id, 511
+  store i32 %and, i32 addrspace(1)* %out, align 4
+  ret void
+}
+
+; CHECK-LABEL: {{^}}test_workitem_id_x_known_max_range_m1:
+; CHECK-NOT: v0
+; CHECK: v_and_b32_e32 [[MASKED:v[0-9]+]], 0xff, v0
+; CHECK: {{flat|buffer}}_store_dword [[MASKED]]
+define void @test_workitem_id_x_known_max_range_m1(i32 addrspace(1)* nocapture %out) #0 {
+entry:
+  %id = tail call i32 @llvm.amdgcn.workitem.id.x(), !range !1
+  %and = and i32 %id, 255
+  store i32 %and, i32 addrspace(1)* %out, align 4
+  ret void
+}
+
+
+declare i32 @llvm.amdgcn.workitem.id.x() #1
+
+attributes #0 = { norecurse nounwind }
+attributes #1 = { nounwind readnone }
+
+!0 = !{i32 0, i32 1024}
+!1 = !{i32 0, i32 1023}