[llvm] Revert "[NVPTX] Support lowering of `(l)lround`" (PR #202500)
via llvm-commits
llvm-commits at lists.llvm.org
Mon Jun 8 20:54:00 PDT 2026
llvmorg-github-actions[bot] wrote:
<!--LLVM PR SUMMARY COMMENT-->
@llvm/pr-subscribers-backend-nvptx
Author: Aiden Grossman (boomanaiden154)
<details>
<summary>Changes</summary>
Reverts llvm/llvm-project#<!-- -->183901
Looks like using removeFromUseLists from Transforms doesn’t work in certain configurations.
---
Full diff: https://github.com/llvm/llvm-project/pull/202500.diff
2 Files Affected:
- (modified) llvm/lib/Target/NVPTX/NVPTXISelLowering.cpp (-2)
- (modified) llvm/test/CodeGen/NVPTX/math-intrins.ll (-151)
``````````diff
diff --git a/llvm/lib/Target/NVPTX/NVPTXISelLowering.cpp b/llvm/lib/Target/NVPTX/NVPTXISelLowering.cpp
index 924384c0298f9..e13911f87eed5 100644
--- a/llvm/lib/Target/NVPTX/NVPTXISelLowering.cpp
+++ b/llvm/lib/Target/NVPTX/NVPTXISelLowering.cpp
@@ -1013,8 +1013,6 @@ NVPTXTargetLowering::NVPTXTargetLowering(const NVPTXTargetMachine &TM,
setOperationAction(ISD::FROUND, MVT::bf16, Promote);
AddPromotedToType(ISD::FROUND, MVT::bf16, MVT::f32);
- setOperationAction({ISD::LROUND, ISD::LLROUND}, {MVT::f32, MVT::f64}, Expand);
-
// 'Expand' implements FCOPYSIGN without calling an external library.
setOperationAction(ISD::FCOPYSIGN, MVT::f16, Expand);
setOperationAction(ISD::FCOPYSIGN, MVT::v2f16, Expand);
diff --git a/llvm/test/CodeGen/NVPTX/math-intrins.ll b/llvm/test/CodeGen/NVPTX/math-intrins.ll
index 3bae69d76ef82..1ed296269c521 100644
--- a/llvm/test/CodeGen/NVPTX/math-intrins.ll
+++ b/llvm/test/CodeGen/NVPTX/math-intrins.ll
@@ -216,157 +216,6 @@ define double @round_double(double %a) {
ret double %b
}
-; ---- lround ----
-
-define i32 @lround_i32_float(float %a) {
-; CHECK-LABEL: lround_i32_float(
-; CHECK: {
-; CHECK-NEXT: .reg .pred %p<3>;
-; CHECK-NEXT: .reg .b32 %r<11>;
-; CHECK-EMPTY:
-; CHECK-NEXT: // %bb.0:
-; CHECK-NEXT: ld.param.b32 %r1, [lround_i32_float_param_0];
-; CHECK-NEXT: and.b32 %r2, %r1, -2147483648;
-; CHECK-NEXT: or.b32 %r3, %r2, 1056964608;
-; CHECK-NEXT: add.rn.f32 %r4, %r1, %r3;
-; CHECK-NEXT: cvt.rzi.f32.f32 %r5, %r4;
-; CHECK-NEXT: abs.f32 %r6, %r1;
-; CHECK-NEXT: setp.gt.f32 %p1, %r6, 0f4B000000;
-; CHECK-NEXT: selp.f32 %r7, %r1, %r5, %p1;
-; CHECK-NEXT: cvt.rzi.f32.f32 %r8, %r1;
-; CHECK-NEXT: setp.lt.f32 %p2, %r6, 0f3F000000;
-; CHECK-NEXT: selp.f32 %r9, %r8, %r7, %p2;
-; CHECK-NEXT: cvt.rzi.s32.f32 %r10, %r9;
-; CHECK-NEXT: st.param.b32 [func_retval0], %r10;
-; CHECK-NEXT: ret;
- %b = call i32 @llvm.lround.i32.f32(float %a)
- ret i32 %b
-}
-
-define i64 @lround_i64_float(float %a) {
-; CHECK-LABEL: lround_i64_float(
-; CHECK: {
-; CHECK-NEXT: .reg .pred %p<3>;
-; CHECK-NEXT: .reg .b32 %r<10>;
-; CHECK-NEXT: .reg .b64 %rd<2>;
-; CHECK-EMPTY:
-; CHECK-NEXT: // %bb.0:
-; CHECK-NEXT: ld.param.b32 %r1, [lround_i64_float_param_0];
-; CHECK-NEXT: and.b32 %r2, %r1, -2147483648;
-; CHECK-NEXT: or.b32 %r3, %r2, 1056964608;
-; CHECK-NEXT: add.rn.f32 %r4, %r1, %r3;
-; CHECK-NEXT: cvt.rzi.f32.f32 %r5, %r4;
-; CHECK-NEXT: abs.f32 %r6, %r1;
-; CHECK-NEXT: setp.gt.f32 %p1, %r6, 0f4B000000;
-; CHECK-NEXT: selp.f32 %r7, %r1, %r5, %p1;
-; CHECK-NEXT: cvt.rzi.f32.f32 %r8, %r1;
-; CHECK-NEXT: setp.lt.f32 %p2, %r6, 0f3F000000;
-; CHECK-NEXT: selp.f32 %r9, %r8, %r7, %p2;
-; CHECK-NEXT: cvt.rzi.s64.f32 %rd1, %r9;
-; CHECK-NEXT: st.param.b64 [func_retval0], %rd1;
-; CHECK-NEXT: ret;
- %b = call i64 @llvm.lround.i64.f32(float %a)
- ret i64 %b
-}
-
-define i32 @lround_i32_double(double %a) {
-; CHECK-LABEL: lround_i32_double(
-; CHECK: {
-; CHECK-NEXT: .reg .pred %p<3>;
-; CHECK-NEXT: .reg .b32 %r<2>;
-; CHECK-NEXT: .reg .b64 %rd<8>;
-; CHECK-EMPTY:
-; CHECK-NEXT: // %bb.0:
-; CHECK-NEXT: ld.param.b64 %rd1, [lround_i32_double_param_0];
-; CHECK-NEXT: abs.f64 %rd2, %rd1;
-; CHECK-NEXT: setp.lt.f64 %p1, %rd2, 0d3FE0000000000000;
-; CHECK-NEXT: add.rn.f64 %rd3, %rd2, 0d3FE0000000000000;
-; CHECK-NEXT: cvt.rzi.f64.f64 %rd4, %rd3;
-; CHECK-NEXT: selp.f64 %rd5, 0d0000000000000000, %rd4, %p1;
-; CHECK-NEXT: copysign.f64 %rd6, %rd1, %rd5;
-; CHECK-NEXT: setp.gt.f64 %p2, %rd2, 0d4330000000000000;
-; CHECK-NEXT: selp.f64 %rd7, %rd1, %rd6, %p2;
-; CHECK-NEXT: cvt.rzi.s32.f64 %r1, %rd7;
-; CHECK-NEXT: st.param.b32 [func_retval0], %r1;
-; CHECK-NEXT: ret;
- %b = call i32 @llvm.lround.i32.f64(double %a)
- ret i32 %b
-}
-
-define i64 @lround_i64_double(double %a) {
-; CHECK-LABEL: lround_i64_double(
-; CHECK: {
-; CHECK-NEXT: .reg .pred %p<3>;
-; CHECK-NEXT: .reg .b64 %rd<9>;
-; CHECK-EMPTY:
-; CHECK-NEXT: // %bb.0:
-; CHECK-NEXT: ld.param.b64 %rd1, [lround_i64_double_param_0];
-; CHECK-NEXT: abs.f64 %rd2, %rd1;
-; CHECK-NEXT: setp.lt.f64 %p1, %rd2, 0d3FE0000000000000;
-; CHECK-NEXT: add.rn.f64 %rd3, %rd2, 0d3FE0000000000000;
-; CHECK-NEXT: cvt.rzi.f64.f64 %rd4, %rd3;
-; CHECK-NEXT: selp.f64 %rd5, 0d0000000000000000, %rd4, %p1;
-; CHECK-NEXT: copysign.f64 %rd6, %rd1, %rd5;
-; CHECK-NEXT: setp.gt.f64 %p2, %rd2, 0d4330000000000000;
-; CHECK-NEXT: selp.f64 %rd7, %rd1, %rd6, %p2;
-; CHECK-NEXT: cvt.rzi.s64.f64 %rd8, %rd7;
-; CHECK-NEXT: st.param.b64 [func_retval0], %rd8;
-; CHECK-NEXT: ret;
- %b = call i64 @llvm.lround.i64.f64(double %a)
- ret i64 %b
-}
-
-; ---- llround ----
-
-define i64 @cpu_llround_i64_float(float %a) {
-; CHECK-LABEL: cpu_llround_i64_float(
-; CHECK: {
-; CHECK-NEXT: .reg .pred %p<3>;
-; CHECK-NEXT: .reg .b32 %r<10>;
-; CHECK-NEXT: .reg .b64 %rd<2>;
-; CHECK-EMPTY:
-; CHECK-NEXT: // %bb.0:
-; CHECK-NEXT: ld.param.b32 %r1, [cpu_llround_i64_float_param_0];
-; CHECK-NEXT: and.b32 %r2, %r1, -2147483648;
-; CHECK-NEXT: or.b32 %r3, %r2, 1056964608;
-; CHECK-NEXT: add.rn.f32 %r4, %r1, %r3;
-; CHECK-NEXT: cvt.rzi.f32.f32 %r5, %r4;
-; CHECK-NEXT: abs.f32 %r6, %r1;
-; CHECK-NEXT: setp.gt.f32 %p1, %r6, 0f4B000000;
-; CHECK-NEXT: selp.f32 %r7, %r1, %r5, %p1;
-; CHECK-NEXT: cvt.rzi.f32.f32 %r8, %r1;
-; CHECK-NEXT: setp.lt.f32 %p2, %r6, 0f3F000000;
-; CHECK-NEXT: selp.f32 %r9, %r8, %r7, %p2;
-; CHECK-NEXT: cvt.rzi.s64.f32 %rd1, %r9;
-; CHECK-NEXT: st.param.b64 [func_retval0], %rd1;
-; CHECK-NEXT: ret;
- %b = call i64 @llvm.llround.i64.f32(float %a)
- ret i64 %b
-}
-
-define i64 @cpu_lround_i64_double(double %a) {
-; CHECK-LABEL: cpu_lround_i64_double(
-; CHECK: {
-; CHECK-NEXT: .reg .pred %p<3>;
-; CHECK-NEXT: .reg .b64 %rd<9>;
-; CHECK-EMPTY:
-; CHECK-NEXT: // %bb.0:
-; CHECK-NEXT: ld.param.b64 %rd1, [cpu_lround_i64_double_param_0];
-; CHECK-NEXT: abs.f64 %rd2, %rd1;
-; CHECK-NEXT: setp.lt.f64 %p1, %rd2, 0d3FE0000000000000;
-; CHECK-NEXT: add.rn.f64 %rd3, %rd2, 0d3FE0000000000000;
-; CHECK-NEXT: cvt.rzi.f64.f64 %rd4, %rd3;
-; CHECK-NEXT: selp.f64 %rd5, 0d0000000000000000, %rd4, %p1;
-; CHECK-NEXT: copysign.f64 %rd6, %rd1, %rd5;
-; CHECK-NEXT: setp.gt.f64 %p2, %rd2, 0d4330000000000000;
-; CHECK-NEXT: selp.f64 %rd7, %rd1, %rd6, %p2;
-; CHECK-NEXT: cvt.rzi.s64.f64 %rd8, %rd7;
-; CHECK-NEXT: st.param.b64 [func_retval0], %rd8;
-; CHECK-NEXT: ret;
- %b = call i64 @llvm.lround.i64.f64(double %a)
- ret i64 %b
-}
-
; ---- nearbyint ----
define float @nearbyint_float(float %a) {
``````````
</details>
https://github.com/llvm/llvm-project/pull/202500
More information about the llvm-commits
mailing list