Skip to content

LoongArch: Add test for sincos intrinsic #147471

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom
Merged
merged 1 commit into from
Jul 8, 2025
Merged

LoongArch: Add test for sincos intrinsic #147471

merged 1 commit into from
Jul 8, 2025

Conversation

arsenm
Copy link
Contributor

@arsenm arsenm commented Jul 8, 2025

No description provided.

@arsenm arsenm mentioned this pull request Jul 8, 2025
Merged
@arsenm Graphite App
Copy link
Contributor Author

arsenm commented Jul 8, 2025 •
edited
Loading

This stack of pull requests is managed by Graphite. Learn more about stacking.

@arsenm arsenm mentioned this pull request Jul 8, 2025
Merged
@llvmbot
Copy link
Member

llvmbot commented Jul 8, 2025

@llvm/pr-subscribers-backend-loongarch

Author: Matt Arsenault (arsenm)

Changes

Patch is 35.05 KiB, truncated to 20.00 KiB below, full version: https://github.com/llvm/llvm-project/pull/147471.diff

1 Files Affected:

  • (added) llvm/test/CodeGen/LoongArch/llvm.sincos.ll (+866)
diff --git a/llvm/test/CodeGen/LoongArch/llvm.sincos.ll b/llvm/test/CodeGen/LoongArch/llvm.sincos.ll
new file mode 100644
index 0000000000000..ffedd7f9e9438
--- /dev/null
+++ b/llvm/test/CodeGen/LoongArch/llvm.sincos.ll
@@ -0,0 +1,866 @@
+; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py UTC_ARGS: --version 5
+; RUN: llc --mtriple=loongarch32 --mattr=+d < %s | FileCheck %s --check-prefix=LA32
+; RUN: llc --mtriple=loongarch64 --mattr=+d < %s | FileCheck %s --check-prefix=LA64
+
+define { half, half } @test_sincos_f16(half %a) #0 {
+; LA32-LABEL: test_sincos_f16:
+; LA32:       # %bb.0:
+; LA32-NEXT:    addi.w $sp, $sp, -32
+; LA32-NEXT:    st.w $ra, $sp, 28 # 4-byte Folded Spill
+; LA32-NEXT:    st.w $fp, $sp, 24 # 4-byte Folded Spill
+; LA32-NEXT:    fst.d $fs0, $sp, 16 # 8-byte Folded Spill
+; LA32-NEXT:    fst.d $fs1, $sp, 8 # 8-byte Folded Spill
+; LA32-NEXT:    bl __extendhfsf2
+; LA32-NEXT:    fmov.s $fs0, $fa0
+; LA32-NEXT:    bl cosf
+; LA32-NEXT:    bl __truncsfhf2
+; LA32-NEXT:    fmov.s $fs1, $fa0
+; LA32-NEXT:    fmov.s $fa0, $fs0
+; LA32-NEXT:    bl sinf
+; LA32-NEXT:    movfr2gr.s $fp, $fs1
+; LA32-NEXT:    bl __truncsfhf2
+; LA32-NEXT:    movfr2gr.s $a0, $fa0
+; LA32-NEXT:    lu12i.w $a1, -16
+; LA32-NEXT:    or $a0, $a0, $a1
+; LA32-NEXT:    movgr2fr.w $fa0, $a0
+; LA32-NEXT:    or $a0, $fp, $a1
+; LA32-NEXT:    movgr2fr.w $fa1, $a0
+; LA32-NEXT:    fld.d $fs1, $sp, 8 # 8-byte Folded Reload
+; LA32-NEXT:    fld.d $fs0, $sp, 16 # 8-byte Folded Reload
+; LA32-NEXT:    ld.w $fp, $sp, 24 # 4-byte Folded Reload
+; LA32-NEXT:    ld.w $ra, $sp, 28 # 4-byte Folded Reload
+; LA32-NEXT:    addi.w $sp, $sp, 32
+; LA32-NEXT:    ret
+;
+; LA64-LABEL: test_sincos_f16:
+; LA64:       # %bb.0:
+; LA64-NEXT:    addi.d $sp, $sp, -32
+; LA64-NEXT:    st.d $ra, $sp, 24 # 8-byte Folded Spill
+; LA64-NEXT:    st.d $fp, $sp, 16 # 8-byte Folded Spill
+; LA64-NEXT:    fst.d $fs0, $sp, 8 # 8-byte Folded Spill
+; LA64-NEXT:    fst.d $fs1, $sp, 0 # 8-byte Folded Spill
+; LA64-NEXT:    pcaddu18i $ra, %call36(__extendhfsf2)
+; LA64-NEXT:    jirl $ra, $ra, 0
+; LA64-NEXT:    fmov.s $fs0, $fa0
+; LA64-NEXT:    pcaddu18i $ra, %call36(cosf)
+; LA64-NEXT:    jirl $ra, $ra, 0
+; LA64-NEXT:    pcaddu18i $ra, %call36(__truncsfhf2)
+; LA64-NEXT:    jirl $ra, $ra, 0
+; LA64-NEXT:    fmov.s $fs1, $fa0
+; LA64-NEXT:    fmov.s $fa0, $fs0
+; LA64-NEXT:    pcaddu18i $ra, %call36(sinf)
+; LA64-NEXT:    jirl $ra, $ra, 0
+; LA64-NEXT:    movfr2gr.s $fp, $fs1
+; LA64-NEXT:    pcaddu18i $ra, %call36(__truncsfhf2)
+; LA64-NEXT:    jirl $ra, $ra, 0
+; LA64-NEXT:    movfr2gr.s $a0, $fa0
+; LA64-NEXT:    lu12i.w $a1, -16
+; LA64-NEXT:    or $a0, $a0, $a1
+; LA64-NEXT:    movgr2fr.w $fa0, $a0
+; LA64-NEXT:    or $a0, $fp, $a1
+; LA64-NEXT:    movgr2fr.w $fa1, $a0
+; LA64-NEXT:    fld.d $fs1, $sp, 0 # 8-byte Folded Reload
+; LA64-NEXT:    fld.d $fs0, $sp, 8 # 8-byte Folded Reload
+; LA64-NEXT:    ld.d $fp, $sp, 16 # 8-byte Folded Reload
+; LA64-NEXT:    ld.d $ra, $sp, 24 # 8-byte Folded Reload
+; LA64-NEXT:    addi.d $sp, $sp, 32
+; LA64-NEXT:    ret
+  %result = call { half, half } @llvm.sincos.f16(half %a)
+  ret { half, half } %result
+}
+
+define half @test_sincos_f16_only_use_sin(half %a) #0 {
+; LA32-LABEL: test_sincos_f16_only_use_sin:
+; LA32:       # %bb.0:
+; LA32-NEXT:    addi.w $sp, $sp, -16
+; LA32-NEXT:    st.w $ra, $sp, 12 # 4-byte Folded Spill
+; LA32-NEXT:    bl __extendhfsf2
+; LA32-NEXT:    bl sinf
+; LA32-NEXT:    bl __truncsfhf2
+; LA32-NEXT:    movfr2gr.s $a0, $fa0
+; LA32-NEXT:    lu12i.w $a1, -16
+; LA32-NEXT:    or $a0, $a0, $a1
+; LA32-NEXT:    movgr2fr.w $fa0, $a0
+; LA32-NEXT:    ld.w $ra, $sp, 12 # 4-byte Folded Reload
+; LA32-NEXT:    addi.w $sp, $sp, 16
+; LA32-NEXT:    ret
+;
+; LA64-LABEL: test_sincos_f16_only_use_sin:
+; LA64:       # %bb.0:
+; LA64-NEXT:    addi.d $sp, $sp, -16
+; LA64-NEXT:    st.d $ra, $sp, 8 # 8-byte Folded Spill
+; LA64-NEXT:    pcaddu18i $ra, %call36(__extendhfsf2)
+; LA64-NEXT:    jirl $ra, $ra, 0
+; LA64-NEXT:    pcaddu18i $ra, %call36(sinf)
+; LA64-NEXT:    jirl $ra, $ra, 0
+; LA64-NEXT:    pcaddu18i $ra, %call36(__truncsfhf2)
+; LA64-NEXT:    jirl $ra, $ra, 0
+; LA64-NEXT:    movfr2gr.s $a0, $fa0
+; LA64-NEXT:    lu12i.w $a1, -16
+; LA64-NEXT:    or $a0, $a0, $a1
+; LA64-NEXT:    movgr2fr.w $fa0, $a0
+; LA64-NEXT:    ld.d $ra, $sp, 8 # 8-byte Folded Reload
+; LA64-NEXT:    addi.d $sp, $sp, 16
+; LA64-NEXT:    ret
+  %result = call { half, half } @llvm.sincos.f16(half %a)
+  %result.0 = extractvalue { half, half } %result, 0
+  ret half %result.0
+}
+
+define half @test_sincos_f16_only_use_cos(half %a) #0 {
+; LA32-LABEL: test_sincos_f16_only_use_cos:
+; LA32:       # %bb.0:
+; LA32-NEXT:    addi.w $sp, $sp, -16
+; LA32-NEXT:    st.w $ra, $sp, 12 # 4-byte Folded Spill
+; LA32-NEXT:    bl __extendhfsf2
+; LA32-NEXT:    bl cosf
+; LA32-NEXT:    bl __truncsfhf2
+; LA32-NEXT:    movfr2gr.s $a0, $fa0
+; LA32-NEXT:    lu12i.w $a1, -16
+; LA32-NEXT:    or $a0, $a0, $a1
+; LA32-NEXT:    movgr2fr.w $fa0, $a0
+; LA32-NEXT:    ld.w $ra, $sp, 12 # 4-byte Folded Reload
+; LA32-NEXT:    addi.w $sp, $sp, 16
+; LA32-NEXT:    ret
+;
+; LA64-LABEL: test_sincos_f16_only_use_cos:
+; LA64:       # %bb.0:
+; LA64-NEXT:    addi.d $sp, $sp, -16
+; LA64-NEXT:    st.d $ra, $sp, 8 # 8-byte Folded Spill
+; LA64-NEXT:    pcaddu18i $ra, %call36(__extendhfsf2)
+; LA64-NEXT:    jirl $ra, $ra, 0
+; LA64-NEXT:    pcaddu18i $ra, %call36(cosf)
+; LA64-NEXT:    jirl $ra, $ra, 0
+; LA64-NEXT:    pcaddu18i $ra, %call36(__truncsfhf2)
+; LA64-NEXT:    jirl $ra, $ra, 0
+; LA64-NEXT:    movfr2gr.s $a0, $fa0
+; LA64-NEXT:    lu12i.w $a1, -16
+; LA64-NEXT:    or $a0, $a0, $a1
+; LA64-NEXT:    movgr2fr.w $fa0, $a0
+; LA64-NEXT:    ld.d $ra, $sp, 8 # 8-byte Folded Reload
+; LA64-NEXT:    addi.d $sp, $sp, 16
+; LA64-NEXT:    ret
+  %result = call { half, half } @llvm.sincos.f16(half %a)
+  %result.1 = extractvalue { half, half } %result, 1
+  ret half %result.1
+}
+
+define { <2 x half>, <2 x half> } @test_sincos_v2f16(<2 x half> %a) #0 {
+; LA32-LABEL: test_sincos_v2f16:
+; LA32:       # %bb.0:
+; LA32-NEXT:    addi.w $sp, $sp, -64
+; LA32-NEXT:    st.w $ra, $sp, 60 # 4-byte Folded Spill
+; LA32-NEXT:    st.w $fp, $sp, 56 # 4-byte Folded Spill
+; LA32-NEXT:    st.w $s0, $sp, 52 # 4-byte Folded Spill
+; LA32-NEXT:    st.w $s1, $sp, 48 # 4-byte Folded Spill
+; LA32-NEXT:    st.w $s2, $sp, 44 # 4-byte Folded Spill
+; LA32-NEXT:    fst.d $fs0, $sp, 32 # 8-byte Folded Spill
+; LA32-NEXT:    fst.d $fs1, $sp, 24 # 8-byte Folded Spill
+; LA32-NEXT:    fst.d $fs2, $sp, 16 # 8-byte Folded Spill
+; LA32-NEXT:    fst.d $fs3, $sp, 8 # 8-byte Folded Spill
+; LA32-NEXT:    move $fp, $a0
+; LA32-NEXT:    movgr2fr.w $fs0, $a2
+; LA32-NEXT:    movgr2fr.w $fa0, $a1
+; LA32-NEXT:    bl __extendhfsf2
+; LA32-NEXT:    fmov.s $fs1, $fa0
+; LA32-NEXT:    bl cosf
+; LA32-NEXT:    bl __truncsfhf2
+; LA32-NEXT:    fmov.s $fs2, $fa0
+; LA32-NEXT:    fmov.s $fa0, $fs0
+; LA32-NEXT:    bl __extendhfsf2
+; LA32-NEXT:    fmov.s $fs0, $fa0
+; LA32-NEXT:    bl sinf
+; LA32-NEXT:    bl __truncsfhf2
+; LA32-NEXT:    fmov.s $fs3, $fa0
+; LA32-NEXT:    fmov.s $fa0, $fs1
+; LA32-NEXT:    bl sinf
+; LA32-NEXT:    bl __truncsfhf2
+; LA32-NEXT:    fmov.s $fs1, $fa0
+; LA32-NEXT:    fmov.s $fa0, $fs0
+; LA32-NEXT:    bl cosf
+; LA32-NEXT:    movfr2gr.s $s0, $fs1
+; LA32-NEXT:    movfr2gr.s $s1, $fs3
+; LA32-NEXT:    movfr2gr.s $s2, $fs2
+; LA32-NEXT:    bl __truncsfhf2
+; LA32-NEXT:    movfr2gr.s $a0, $fa0
+; LA32-NEXT:    st.h $a0, $fp, 6
+; LA32-NEXT:    st.h $s2, $fp, 4
+; LA32-NEXT:    st.h $s1, $fp, 2
+; LA32-NEXT:    st.h $s0, $fp, 0
+; LA32-NEXT:    fld.d $fs3, $sp, 8 # 8-byte Folded Reload
+; LA32-NEXT:    fld.d $fs2, $sp, 16 # 8-byte Folded Reload
+; LA32-NEXT:    fld.d $fs1, $sp, 24 # 8-byte Folded Reload
+; LA32-NEXT:    fld.d $fs0, $sp, 32 # 8-byte Folded Reload
+; LA32-NEXT:    ld.w $s2, $sp, 44 # 4-byte Folded Reload
+; LA32-NEXT:    ld.w $s1, $sp, 48 # 4-byte Folded Reload
+; LA32-NEXT:    ld.w $s0, $sp, 52 # 4-byte Folded Reload
+; LA32-NEXT:    ld.w $fp, $sp, 56 # 4-byte Folded Reload
+; LA32-NEXT:    ld.w $ra, $sp, 60 # 4-byte Folded Reload
+; LA32-NEXT:    addi.w $sp, $sp, 64
+; LA32-NEXT:    ret
+;
+; LA64-LABEL: test_sincos_v2f16:
+; LA64:       # %bb.0:
+; LA64-NEXT:    addi.d $sp, $sp, -80
+; LA64-NEXT:    st.d $ra, $sp, 72 # 8-byte Folded Spill
+; LA64-NEXT:    st.d $fp, $sp, 64 # 8-byte Folded Spill
+; LA64-NEXT:    st.d $s0, $sp, 56 # 8-byte Folded Spill
+; LA64-NEXT:    st.d $s1, $sp, 48 # 8-byte Folded Spill
+; LA64-NEXT:    st.d $s2, $sp, 40 # 8-byte Folded Spill
+; LA64-NEXT:    fst.d $fs0, $sp, 32 # 8-byte Folded Spill
+; LA64-NEXT:    fst.d $fs1, $sp, 24 # 8-byte Folded Spill
+; LA64-NEXT:    fst.d $fs2, $sp, 16 # 8-byte Folded Spill
+; LA64-NEXT:    fst.d $fs3, $sp, 8 # 8-byte Folded Spill
+; LA64-NEXT:    move $s0, $a2
+; LA64-NEXT:    move $fp, $a0
+; LA64-NEXT:    movgr2fr.w $fa0, $a1
+; LA64-NEXT:    pcaddu18i $ra, %call36(__extendhfsf2)
+; LA64-NEXT:    jirl $ra, $ra, 0
+; LA64-NEXT:    fmov.s $fs0, $fa0
+; LA64-NEXT:    pcaddu18i $ra, %call36(cosf)
+; LA64-NEXT:    jirl $ra, $ra, 0
+; LA64-NEXT:    pcaddu18i $ra, %call36(__truncsfhf2)
+; LA64-NEXT:    jirl $ra, $ra, 0
+; LA64-NEXT:    fmov.s $fs1, $fa0
+; LA64-NEXT:    movgr2fr.w $fa0, $s0
+; LA64-NEXT:    pcaddu18i $ra, %call36(__extendhfsf2)
+; LA64-NEXT:    jirl $ra, $ra, 0
+; LA64-NEXT:    fmov.s $fs2, $fa0
+; LA64-NEXT:    pcaddu18i $ra, %call36(sinf)
+; LA64-NEXT:    jirl $ra, $ra, 0
+; LA64-NEXT:    pcaddu18i $ra, %call36(__truncsfhf2)
+; LA64-NEXT:    jirl $ra, $ra, 0
+; LA64-NEXT:    fmov.s $fs3, $fa0
+; LA64-NEXT:    fmov.s $fa0, $fs0
+; LA64-NEXT:    pcaddu18i $ra, %call36(sinf)
+; LA64-NEXT:    jirl $ra, $ra, 0
+; LA64-NEXT:    pcaddu18i $ra, %call36(__truncsfhf2)
+; LA64-NEXT:    jirl $ra, $ra, 0
+; LA64-NEXT:    fmov.s $fs0, $fa0
+; LA64-NEXT:    fmov.s $fa0, $fs2
+; LA64-NEXT:    pcaddu18i $ra, %call36(cosf)
+; LA64-NEXT:    jirl $ra, $ra, 0
+; LA64-NEXT:    movfr2gr.s $s0, $fs0
+; LA64-NEXT:    movfr2gr.s $s1, $fs3
+; LA64-NEXT:    movfr2gr.s $s2, $fs1
+; LA64-NEXT:    pcaddu18i $ra, %call36(__truncsfhf2)
+; LA64-NEXT:    jirl $ra, $ra, 0
+; LA64-NEXT:    movfr2gr.s $a0, $fa0
+; LA64-NEXT:    st.h $a0, $fp, 6
+; LA64-NEXT:    st.h $s2, $fp, 4
+; LA64-NEXT:    st.h $s1, $fp, 2
+; LA64-NEXT:    st.h $s0, $fp, 0
+; LA64-NEXT:    fld.d $fs3, $sp, 8 # 8-byte Folded Reload
+; LA64-NEXT:    fld.d $fs2, $sp, 16 # 8-byte Folded Reload
+; LA64-NEXT:    fld.d $fs1, $sp, 24 # 8-byte Folded Reload
+; LA64-NEXT:    fld.d $fs0, $sp, 32 # 8-byte Folded Reload
+; LA64-NEXT:    ld.d $s2, $sp, 40 # 8-byte Folded Reload
+; LA64-NEXT:    ld.d $s1, $sp, 48 # 8-byte Folded Reload
+; LA64-NEXT:    ld.d $s0, $sp, 56 # 8-byte Folded Reload
+; LA64-NEXT:    ld.d $fp, $sp, 64 # 8-byte Folded Reload
+; LA64-NEXT:    ld.d $ra, $sp, 72 # 8-byte Folded Reload
+; LA64-NEXT:    addi.d $sp, $sp, 80
+; LA64-NEXT:    ret
+  %result = call { <2 x half>, <2 x half> } @llvm.sincos.v2f16(<2 x half> %a)
+  ret { <2 x half>, <2 x half> } %result
+}
+
+define { float, float } @test_sincos_f32(float %a) #0 {
+; LA32-LABEL: test_sincos_f32:
+; LA32:       # %bb.0:
+; LA32-NEXT:    addi.w $sp, $sp, -32
+; LA32-NEXT:    st.w $ra, $sp, 28 # 4-byte Folded Spill
+; LA32-NEXT:    fst.d $fs0, $sp, 16 # 8-byte Folded Spill
+; LA32-NEXT:    fst.d $fs1, $sp, 8 # 8-byte Folded Spill
+; LA32-NEXT:    fmov.s $fs0, $fa0
+; LA32-NEXT:    bl sinf
+; LA32-NEXT:    fmov.s $fs1, $fa0
+; LA32-NEXT:    fmov.s $fa0, $fs0
+; LA32-NEXT:    bl cosf
+; LA32-NEXT:    fmov.s $fa1, $fa0
+; LA32-NEXT:    fmov.s $fa0, $fs1
+; LA32-NEXT:    fld.d $fs1, $sp, 8 # 8-byte Folded Reload
+; LA32-NEXT:    fld.d $fs0, $sp, 16 # 8-byte Folded Reload
+; LA32-NEXT:    ld.w $ra, $sp, 28 # 4-byte Folded Reload
+; LA32-NEXT:    addi.w $sp, $sp, 32
+; LA32-NEXT:    ret
+;
+; LA64-LABEL: test_sincos_f32:
+; LA64:       # %bb.0:
+; LA64-NEXT:    addi.d $sp, $sp, -32
+; LA64-NEXT:    st.d $ra, $sp, 24 # 8-byte Folded Spill
+; LA64-NEXT:    fst.d $fs0, $sp, 16 # 8-byte Folded Spill
+; LA64-NEXT:    fst.d $fs1, $sp, 8 # 8-byte Folded Spill
+; LA64-NEXT:    fmov.s $fs0, $fa0
+; LA64-NEXT:    pcaddu18i $ra, %call36(sinf)
+; LA64-NEXT:    jirl $ra, $ra, 0
+; LA64-NEXT:    fmov.s $fs1, $fa0
+; LA64-NEXT:    fmov.s $fa0, $fs0
+; LA64-NEXT:    pcaddu18i $ra, %call36(cosf)
+; LA64-NEXT:    jirl $ra, $ra, 0
+; LA64-NEXT:    fmov.s $fa1, $fa0
+; LA64-NEXT:    fmov.s $fa0, $fs1
+; LA64-NEXT:    fld.d $fs1, $sp, 8 # 8-byte Folded Reload
+; LA64-NEXT:    fld.d $fs0, $sp, 16 # 8-byte Folded Reload
+; LA64-NEXT:    ld.d $ra, $sp, 24 # 8-byte Folded Reload
+; LA64-NEXT:    addi.d $sp, $sp, 32
+; LA64-NEXT:    ret
+  %result = call { float, float } @llvm.sincos.f32(float %a)
+  ret { float, float } %result
+}
+
+define { <2 x float>, <2 x float> } @test_sincos_v2f32(<2 x float> %a) #0 {
+; LA32-LABEL: test_sincos_v2f32:
+; LA32:       # %bb.0:
+; LA32-NEXT:    addi.w $sp, $sp, -48
+; LA32-NEXT:    st.w $ra, $sp, 44 # 4-byte Folded Spill
+; LA32-NEXT:    st.w $fp, $sp, 40 # 4-byte Folded Spill
+; LA32-NEXT:    fst.d $fs0, $sp, 32 # 8-byte Folded Spill
+; LA32-NEXT:    fst.d $fs1, $sp, 24 # 8-byte Folded Spill
+; LA32-NEXT:    fst.d $fs2, $sp, 16 # 8-byte Folded Spill
+; LA32-NEXT:    fst.d $fs3, $sp, 8 # 8-byte Folded Spill
+; LA32-NEXT:    fmov.s $fs0, $fa1
+; LA32-NEXT:    fmov.s $fs1, $fa0
+; LA32-NEXT:    move $fp, $a0
+; LA32-NEXT:    bl sinf
+; LA32-NEXT:    fmov.s $fs2, $fa0
+; LA32-NEXT:    fmov.s $fa0, $fs0
+; LA32-NEXT:    bl sinf
+; LA32-NEXT:    fmov.s $fs3, $fa0
+; LA32-NEXT:    fmov.s $fa0, $fs1
+; LA32-NEXT:    bl cosf
+; LA32-NEXT:    fmov.s $fs1, $fa0
+; LA32-NEXT:    fmov.s $fa0, $fs0
+; LA32-NEXT:    bl cosf
+; LA32-NEXT:    fst.s $fa0, $fp, 12
+; LA32-NEXT:    fst.s $fs1, $fp, 8
+; LA32-NEXT:    fst.s $fs3, $fp, 4
+; LA32-NEXT:    fst.s $fs2, $fp, 0
+; LA32-NEXT:    fld.d $fs3, $sp, 8 # 8-byte Folded Reload
+; LA32-NEXT:    fld.d $fs2, $sp, 16 # 8-byte Folded Reload
+; LA32-NEXT:    fld.d $fs1, $sp, 24 # 8-byte Folded Reload
+; LA32-NEXT:    fld.d $fs0, $sp, 32 # 8-byte Folded Reload
+; LA32-NEXT:    ld.w $fp, $sp, 40 # 4-byte Folded Reload
+; LA32-NEXT:    ld.w $ra, $sp, 44 # 4-byte Folded Reload
+; LA32-NEXT:    addi.w $sp, $sp, 48
+; LA32-NEXT:    ret
+;
+; LA64-LABEL: test_sincos_v2f32:
+; LA64:       # %bb.0:
+; LA64-NEXT:    addi.d $sp, $sp, -64
+; LA64-NEXT:    st.d $ra, $sp, 56 # 8-byte Folded Spill
+; LA64-NEXT:    vst $vr0, $sp, 0 # 16-byte Folded Spill
+; LA64-NEXT:    vreplvei.w $vr0, $vr0, 0
+; LA64-NEXT:    vst $vr0, $sp, 32 # 16-byte Folded Spill
+; LA64-NEXT:    # kill: def $f0 killed $f0 killed $vr0
+; LA64-NEXT:    pcaddu18i $ra, %call36(sinf)
+; LA64-NEXT:    jirl $ra, $ra, 0
+; LA64-NEXT:    # kill: def $f0 killed $f0 def $vr0
+; LA64-NEXT:    vst $vr0, $sp, 16 # 16-byte Folded Spill
+; LA64-NEXT:    vld $vr0, $sp, 0 # 16-byte Folded Reload
+; LA64-NEXT:    vreplvei.w $vr0, $vr0, 1
+; LA64-NEXT:    vst $vr0, $sp, 0 # 16-byte Folded Spill
+; LA64-NEXT:    # kill: def $f0 killed $f0 killed $vr0
+; LA64-NEXT:    pcaddu18i $ra, %call36(sinf)
+; LA64-NEXT:    jirl $ra, $ra, 0
+; LA64-NEXT:    # kill: def $f0 killed $f0 def $vr0
+; LA64-NEXT:    vld $vr1, $sp, 16 # 16-byte Folded Reload
+; LA64-NEXT:    vpackev.w $vr0, $vr0, $vr1
+; LA64-NEXT:    vst $vr0, $sp, 16 # 16-byte Folded Spill
+; LA64-NEXT:    vld $vr0, $sp, 32 # 16-byte Folded Reload
+; LA64-NEXT:    # kill: def $f0 killed $f0 killed $vr0
+; LA64-NEXT:    pcaddu18i $ra, %call36(cosf)
+; LA64-NEXT:    jirl $ra, $ra, 0
+; LA64-NEXT:    # kill: def $f0 killed $f0 def $vr0
+; LA64-NEXT:    vst $vr0, $sp, 32 # 16-byte Folded Spill
+; LA64-NEXT:    vld $vr0, $sp, 0 # 16-byte Folded Reload
+; LA64-NEXT:    # kill: def $f0 killed $f0 killed $vr0
+; LA64-NEXT:    pcaddu18i $ra, %call36(cosf)
+; LA64-NEXT:    jirl $ra, $ra, 0
+; LA64-NEXT:    # kill: def $f0 killed $f0 def $vr0
+; LA64-NEXT:    vld $vr1, $sp, 32 # 16-byte Folded Reload
+; LA64-NEXT:    vpackev.w $vr1, $vr0, $vr1
+; LA64-NEXT:    vld $vr0, $sp, 16 # 16-byte Folded Reload
+; LA64-NEXT:    ld.d $ra, $sp, 56 # 8-byte Folded Reload
+; LA64-NEXT:    addi.d $sp, $sp, 64
+; LA64-NEXT:    ret
+  %result = call { <2 x float>, <2 x float> } @llvm.sincos.v2f32(<2 x float> %a)
+  ret { <2 x float>, <2 x float> } %result
+}
+
+define { <3 x float>, <3 x float> } @test_sincos_v3f32(<3 x float> %a) #0 {
+; LA32-LABEL: test_sincos_v3f32:
+; LA32:       # %bb.0:
+; LA32-NEXT:    addi.w $sp, $sp, -64
+; LA32-NEXT:    st.w $ra, $sp, 60 # 4-byte Folded Spill
+; LA32-NEXT:    st.w $fp, $sp, 56 # 4-byte Folded Spill
+; LA32-NEXT:    fst.d $fs0, $sp, 48 # 8-byte Folded Spill
+; LA32-NEXT:    fst.d $fs1, $sp, 40 # 8-byte Folded Spill
+; LA32-NEXT:    fst.d $fs2, $sp, 32 # 8-byte Folded Spill
+; LA32-NEXT:    fst.d $fs3, $sp, 24 # 8-byte Folded Spill
+; LA32-NEXT:    fst.d $fs4, $sp, 16 # 8-byte Folded Spill
+; LA32-NEXT:    fst.d $fs5, $sp, 8 # 8-byte Folded Spill
+; LA32-NEXT:    fmov.s $fs0, $fa2
+; LA32-NEXT:    fmov.s $fs1, $fa1
+; LA32-NEXT:    fmov.s $fs2, $fa0
+; LA32-NEXT:    move $fp, $a0
+; LA32-NEXT:    bl sinf
+; LA32-NEXT:    fmov.s $fs3, $fa0
+; LA32-NEXT:    fmov.s $fa0, $fs1
+; LA32-NEXT:    bl sinf
+; LA32-NEXT:    fmov.s $fs4, $fa0
+; LA32-NEXT:    fmov.s $fa0, $fs0
+; LA32-NEXT:    bl sinf
+; LA32-NEXT:    fmov.s $fs5, $fa0
+; LA32-NEXT:    fmov.s $fa0, $fs2
+; LA32-NEXT:    bl cosf
+; LA32-NEXT:    fmov.s $fs2, $fa0
+; LA32-NEXT:    fmov.s $fa0, $fs1
+; LA32-NEXT:    bl cosf
+; LA32-NEXT:    fmov.s $fs1, $fa0
+; LA32-NEXT:    fmov.s $fa0, $fs0
+; LA32-NEXT:    bl cosf
+; LA32-NEXT:    fst.s $fa0, $fp, 24
+; LA32-NEXT:    fst.s $fs1, $fp, 20
+; LA32-NEXT:    fst.s $fs2, $fp, 16
+; LA32-NEXT:    fst.s $fs5, $fp, 8
+; LA32-NEXT:    fst.s $fs4, $fp, 4
+; LA32-NEXT:    fst.s $fs3, $fp, 0
+; LA32-NEXT:    fld.d $fs5, $sp, 8 # 8-byte Folded Reload
+; LA32-NEXT:    fld.d $fs4, $sp, 16 # 8-byte Folded Reload
+; LA32-NEXT:    fld.d $fs3, $sp, 24 # 8-byte Folded Reload
+; LA32-NEXT:    fld.d $fs2, $sp, 32 # 8-byte Folded Reload
+; LA32-NEXT:    fld.d $fs1, $sp, 40 # 8-byte Folded Reload
+; LA32-NEXT:    fld.d $fs0, $sp, 48 # 8-byte Folded Reload
+; LA32-NEXT:    ld.w $fp, $sp, 56 # 4-byte Folded Reload
+; LA32-NEXT:    ld.w $ra, $sp, 60 # 4-byte Folded Reload
+; LA32-NEXT:    addi.w $sp, $sp, 64
+; LA32-NEXT:    ret
+;
+; LA64-LABEL: test_sincos_v3f32:
+; LA64:       # %bb.0:
+; LA64-NEXT:    addi.d $sp, $sp, -96
+; LA64-NEXT:    st.d $ra, $sp, 88 # 8-byte Folded Spill
+; LA64-NEXT:    vst $vr0, $sp, 32 # 16-byte Folded Spill
+; LA64-NEXT:    vreplvei.w $vr0, $vr0, 2
+; LA64-NEXT:    vst $vr0, $sp, 16 # 16-byte Folded Spill
+; LA64-NEXT:    # kill: def $f0 killed $f0 killed $vr0
+; LA64-NEXT:    pcaddu18i $ra, %call36(sinf)
+; LA64-NEXT:    jirl $ra, $ra, 0
+; LA64-NEXT:    fst.s $fa0, $sp, 72
+; LA64-NEXT:    vld $vr0, $sp, 32 # 16-byte Folded Reload
+; LA64-NEXT:    vreplvei.w $vr0, $vr0, 1
+; LA64-NEXT:    vst $vr0, $sp, 0 # 16-byte Folded Spill
+; LA64-NEXT:    # kill: def $f0 killed $f0 killed $vr0
+; LA64-NEXT:    pcaddu18i $ra, %call36(sinf)
+; LA64-NEXT:    jirl $ra, $ra, 0
+; LA64-NEXT:    fst.s $fa0, $sp, 68
+; LA64-NEXT:    vld $vr0, $sp, 32 # 16-byte Folded Reload
+; LA64-NEXT:    vreplvei.w $vr0, $vr0, 0
+; LA64-NEXT:    vst $vr0, $sp, 32 # 16-byte Folded Spill
+; LA64-NEXT:    # kill: def $f0 killed $f0 killed $vr0
+; LA64-NEXT:    pcaddu18i $ra, %call36(sinf)
+; LA64-NEXT:    jirl $ra, $ra, 0
+; LA64-NEXT:    fst.s $fa0, $sp, 64
+; LA64-NEXT:    vld $vr0, $sp, 16 # 16-byte Folded Reload
+; LA64-NEXT:    # kill: def $f0 killed $f0 killed $vr0
+; LA64-NEXT:    pcaddu18i $ra, %call36(cosf)
+; LA64-NEXT:    jirl $ra, $ra, 0
+; LA64-NEXT:    fst.s $fa0, $sp, 56
+; LA64-NEXT:    vld $vr0, $sp, 0 # 16-byte Folded Reload
+; LA64-NEXT:    # kill: def $f0 killed $f0 killed $vr0
+; LA64-NEXT:    pcaddu18i $ra, %call36(cosf)
+; LA64-NEXT:    jirl $ra, $ra, 0
+; LA64-NEXT:    fst.s $fa0, $sp, 52
+; LA64-NEXT:    vld $vr0, $sp, 32 # 16-byte Folded Reload
+; LA64-NEXT:    # kill: def $f0 killed $f0 killed $vr0
+; LA64-NEXT:    pcaddu18i $ra, %call36(cosf)
+; LA64-NEXT:    jirl $ra, $ra, 0
+; LA64-NEXT:    fst.s $fa0, $sp, 48
+; LA64-NEXT:    vld $vr0, $sp, 64
+; LA64-NEXT:    vld $vr1, $sp, 48
+; LA64-NEXT:    ld.d $ra, $sp, 88 # 8-byte Folded Reload
+; LA64-NEXT:    addi.d $sp, $sp, 96
+; LA64-NEXT:    ret
+  %result = call { <3 x float>, <3 x float> } @llvm.sincos.v3f32(<3 x float> %a)
+  ret { <...
[truncated]

@arsenm arsenm marked this pull request as ready for review July 8, 2025 07:38
This was referenced Jul 8, 2025
Open
Open
heiher
heiher approved these changes Jul 8, 2025
View reviewed changes
Copy link
Member

@heiher heiher left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

LGTM.

@arsenm arsenm force-pushed the users/arsenm/loongarch/add-sincos-test branch from 5f21740 to 8ee5b85 Compare July 8, 2025 15:25
@arsenm arsenm force-pushed the users/arsenm/dag/fix-assert-soften-sincos-no-libcall branch from 65011f9 to 248eb92 Compare July 8, 2025 15:25
@arsenm Graphite App
Copy link
Contributor Author

arsenm commented Jul 8, 2025 •
edited
Loading

Merge activity

  • Jul 8, 4:42 PM UTC: A user started a stack merge that includes this pull request via Graphite.
  • Jul 8, 4:53 PM UTC: Graphite rebased this pull request as part of a merge.
  • Jul 8, 4:56 PM UTC: Graphite rebased this pull request as part of a merge.
  • Jul 8, 4:59 PM UTC: Graphite rebased this pull request as part of a merge.
  • Jul 8, 5:01 PM UTC: @arsenm merged this pull request with Graphite.

@arsenm arsenm force-pushed the users/arsenm/dag/fix-assert-soften-sincos-no-libcall branch from 248eb92 to aceabe9 Compare July 8, 2025 16:49
Base automatically changed from users/arsenm/dag/fix-assert-soften-sincos-no-libcall to main July 8, 2025 16:52
@arsenm arsenm force-pushed the users/arsenm/loongarch/add-sincos-test branch 2 times, most recently from 3baa828 to 5e1d92b Compare July 8, 2025 16:55
@arsenm arsenm force-pushed the users/arsenm/loongarch/add-sincos-test branch from 5e1d92b to 281f2ca Compare July 8, 2025 16:58
@arsenm arsenm merged commit 3614d49 into main Jul 8, 2025
7 of 9 checks passed
@arsenm arsenm deleted the users/arsenm/loongarch/add-sincos-test branch July 8, 2025 17:01
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Reviewers

@heiher heiher heiher approved these changes

@gonglingqin gonglingqin Awaiting requested review from gonglingqin

@SixWeining SixWeining Awaiting requested review from SixWeining

Assignees
No one assigned
Labels
backend:loongarch
Projects
None yet
Milestone
No milestone
Development

Successfully merging this pull request may close these issues.
3 participants
@arsenm @llvmbot @heiher