[llvm] [RISCV] Mark sincos libcalls as available for RISC-V (PR #168708)

[via llvm-commits]

llvmbot wrote:


<!--LLVM PR SUMMARY COMMENT-->

@llvm/pr-subscribers-llvm-ir

Author: Alex Bradbury (asb)

<details>
<summary>Changes</summary>

A standard glibc build for RISC-V does contain these libcalls, and GCC will generate calls to them. Found by comparing benchmark data for SPEC, spotting that the GCC build of 526.blender_r spends 3.02% of dynamically executed instructions in libm functions, while the Clang build spends 3.41%. Calling sincos rather than separate sin and cos functions is the reason for the difference.

---
Full diff: https://github.com/llvm/llvm-project/pull/168708.diff


2 Files Affected:

- (modified) llvm/include/llvm/IR/RuntimeLibcalls.td (+1) 
- (modified) llvm/test/CodeGen/RISCV/sincos-expansion.ll (+16-24) 


``````````diff
diff --git a/llvm/include/llvm/IR/RuntimeLibcalls.td b/llvm/include/llvm/IR/RuntimeLibcalls.td
index ce7e836f66446..9230b39e6f45e 100644
--- a/llvm/include/llvm/IR/RuntimeLibcalls.td
+++ b/llvm/include/llvm/IR/RuntimeLibcalls.td
@@ -2452,6 +2452,7 @@ def RISCVSystemLibrary
       (add DefaultRuntimeLibcallImpls,
            exp10f, exp10, exp10l_f128,
            __riscv_flush_icache,
+           LibmHasSinCosF32, LibmHasSinCosF64, LibmHasSinCosF128,
            LibcallImpls<(add Int128RTLibcalls), isRISCV64>,
            DefaultStackProtector)>;
 
diff --git a/llvm/test/CodeGen/RISCV/sincos-expansion.ll b/llvm/test/CodeGen/RISCV/sincos-expansion.ll
index 6f37d67134cb3..606768a8e2f53 100644
--- a/llvm/test/CodeGen/RISCV/sincos-expansion.ll
+++ b/llvm/test/CodeGen/RISCV/sincos-expansion.ll
@@ -7,20 +7,16 @@
 define float @test_sincos_f32(float %f) nounwind {
 ; CHECK-LABEL: test_sincos_f32:
 ; CHECK:       # %bb.0:
-; CHECK-NEXT:    addi sp, sp, -32
-; CHECK-NEXT:    sd ra, 24(sp) # 8-byte Folded Spill
-; CHECK-NEXT:    fsd fs0, 16(sp) # 8-byte Folded Spill
-; CHECK-NEXT:    fsd fs1, 8(sp) # 8-byte Folded Spill
-; CHECK-NEXT:    fmv.s fs0, fa0
-; CHECK-NEXT:    call sinf
-; CHECK-NEXT:    fmv.s fs1, fa0
-; CHECK-NEXT:    fmv.s fa0, fs0
-; CHECK-NEXT:    call cosf
-; CHECK-NEXT:    fadd.s fa0, fs1, fa0
-; CHECK-NEXT:    ld ra, 24(sp) # 8-byte Folded Reload
-; CHECK-NEXT:    fld fs0, 16(sp) # 8-byte Folded Reload
-; CHECK-NEXT:    fld fs1, 8(sp) # 8-byte Folded Reload
-; CHECK-NEXT:    addi sp, sp, 32
+; CHECK-NEXT:    addi sp, sp, -16
+; CHECK-NEXT:    sd ra, 8(sp) # 8-byte Folded Spill
+; CHECK-NEXT:    addi a0, sp, 4
+; CHECK-NEXT:    mv a1, sp
+; CHECK-NEXT:    call sincosf
+; CHECK-NEXT:    flw fa5, 0(sp)
+; CHECK-NEXT:    flw fa4, 4(sp)
+; CHECK-NEXT:    fadd.s fa0, fa4, fa5
+; CHECK-NEXT:    ld ra, 8(sp) # 8-byte Folded Reload
+; CHECK-NEXT:    addi sp, sp, 16
 ; CHECK-NEXT:    ret
   %sin = call float @sinf(float %f) readnone
   %cos = call float @cosf(float %f) readnone
@@ -57,17 +53,13 @@ define double @test_sincos_f64(double %f) nounwind {
 ; CHECK:       # %bb.0:
 ; CHECK-NEXT:    addi sp, sp, -32
 ; CHECK-NEXT:    sd ra, 24(sp) # 8-byte Folded Spill
-; CHECK-NEXT:    fsd fs0, 16(sp) # 8-byte Folded Spill
-; CHECK-NEXT:    fsd fs1, 8(sp) # 8-byte Folded Spill
-; CHECK-NEXT:    fmv.d fs0, fa0
-; CHECK-NEXT:    call sin
-; CHECK-NEXT:    fmv.d fs1, fa0
-; CHECK-NEXT:    fmv.d fa0, fs0
-; CHECK-NEXT:    call cos
-; CHECK-NEXT:    fadd.d fa0, fs1, fa0
+; CHECK-NEXT:    addi a0, sp, 16
+; CHECK-NEXT:    addi a1, sp, 8
+; CHECK-NEXT:    call sincos
+; CHECK-NEXT:    fld fa5, 8(sp)
+; CHECK-NEXT:    fld fa4, 16(sp)
+; CHECK-NEXT:    fadd.d fa0, fa4, fa5
 ; CHECK-NEXT:    ld ra, 24(sp) # 8-byte Folded Reload
-; CHECK-NEXT:    fld fs0, 16(sp) # 8-byte Folded Reload
-; CHECK-NEXT:    fld fs1, 8(sp) # 8-byte Folded Reload
 ; CHECK-NEXT:    addi sp, sp, 32
 ; CHECK-NEXT:    ret
   %sin = call double @sin(double %f) readnone

``````````

</details>


https://github.com/llvm/llvm-project/pull/168708