Skip to content

[LoongArch][NFC] Pre-commit for lowering vector mask generation to [X]VMSK{LT,GE,NE}Z #142108

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 2 commits into from
Jun 5, 2025
Merged

[LoongArch][NFC] Pre-commit for lowering vector mask generation to [X]VMSK{LT,GE,NE}Z #142108

merged 2 commits into from
Jun 5, 2025

Conversation

heiher
Copy link
Member

@heiher heiher commented May 30, 2025

No description provided.

@llvmbot
Copy link
Member

llvmbot commented May 30, 2025

@llvm/pr-subscribers-backend-loongarch

Author: hev (heiher)

Changes

Patch is 76.46 KiB, truncated to 20.00 KiB below, full version: https://github.com/llvm/llvm-project/pull/142108.diff

2 Files Affected:

  • (added) llvm/test/CodeGen/LoongArch/lasx/xvmskcond.ll (+1406)
  • (added) llvm/test/CodeGen/LoongArch/lsx/vmskcond.ll (+688)
diff --git a/llvm/test/CodeGen/LoongArch/lasx/xvmskcond.ll b/llvm/test/CodeGen/LoongArch/lasx/xvmskcond.ll
new file mode 100644
index 0000000000000..a934e5a549c4b
--- /dev/null
+++ b/llvm/test/CodeGen/LoongArch/lasx/xvmskcond.ll
@@ -0,0 +1,1406 @@
+; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py UTC_ARGS: --version 5
+; RUN: llc --mtriple=loongarch64 --mattr=+lasx < %s | FileCheck %s
+
+define i32 @xmsk_eq_allzeros_i8(<32 x i8 > %a) {
+; CHECK-LABEL: xmsk_eq_allzeros_i8:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    addi.d $sp, $sp, -64
+; CHECK-NEXT:    .cfi_def_cfa_offset 64
+; CHECK-NEXT:    st.d $ra, $sp, 56 # 8-byte Folded Spill
+; CHECK-NEXT:    st.d $fp, $sp, 48 # 8-byte Folded Spill
+; CHECK-NEXT:    .cfi_offset 1, -8
+; CHECK-NEXT:    .cfi_offset 22, -16
+; CHECK-NEXT:    addi.d $fp, $sp, 64
+; CHECK-NEXT:    .cfi_def_cfa 22, 0
+; CHECK-NEXT:    bstrins.d $sp, $zero, 4, 0
+; CHECK-NEXT:    xvseqi.b $xr0, $xr0, 0
+; CHECK-NEXT:    xvst $xr0, $sp, 0
+; CHECK-NEXT:    vpickve2gr.b $a0, $vr0, 0
+; CHECK-NEXT:    vpickve2gr.b $a1, $vr0, 1
+; CHECK-NEXT:    andi $a1, $a1, 1
+; CHECK-NEXT:    bstrins.d $a0, $a1, 63, 1
+; CHECK-NEXT:    vpickve2gr.b $a1, $vr0, 2
+; CHECK-NEXT:    bstrins.d $a0, $a1, 2, 2
+; CHECK-NEXT:    vpickve2gr.b $a1, $vr0, 3
+; CHECK-NEXT:    bstrins.d $a0, $a1, 3, 3
+; CHECK-NEXT:    vpickve2gr.b $a1, $vr0, 4
+; CHECK-NEXT:    bstrins.d $a0, $a1, 4, 4
+; CHECK-NEXT:    vpickve2gr.b $a1, $vr0, 5
+; CHECK-NEXT:    bstrins.d $a0, $a1, 5, 5
+; CHECK-NEXT:    vpickve2gr.b $a1, $vr0, 6
+; CHECK-NEXT:    andi $a1, $a1, 1
+; CHECK-NEXT:    slli.d $a1, $a1, 6
+; CHECK-NEXT:    or $a0, $a0, $a1
+; CHECK-NEXT:    vpickve2gr.b $a1, $vr0, 7
+; CHECK-NEXT:    andi $a1, $a1, 1
+; CHECK-NEXT:    slli.d $a1, $a1, 7
+; CHECK-NEXT:    or $a0, $a0, $a1
+; CHECK-NEXT:    vpickve2gr.b $a1, $vr0, 8
+; CHECK-NEXT:    andi $a1, $a1, 1
+; CHECK-NEXT:    slli.d $a1, $a1, 8
+; CHECK-NEXT:    or $a0, $a0, $a1
+; CHECK-NEXT:    vpickve2gr.b $a1, $vr0, 9
+; CHECK-NEXT:    andi $a1, $a1, 1
+; CHECK-NEXT:    slli.d $a1, $a1, 9
+; CHECK-NEXT:    or $a0, $a0, $a1
+; CHECK-NEXT:    vpickve2gr.b $a1, $vr0, 10
+; CHECK-NEXT:    andi $a1, $a1, 1
+; CHECK-NEXT:    slli.d $a1, $a1, 10
+; CHECK-NEXT:    or $a0, $a0, $a1
+; CHECK-NEXT:    vpickve2gr.b $a1, $vr0, 11
+; CHECK-NEXT:    andi $a1, $a1, 1
+; CHECK-NEXT:    slli.d $a1, $a1, 11
+; CHECK-NEXT:    or $a0, $a0, $a1
+; CHECK-NEXT:    vpickve2gr.b $a1, $vr0, 12
+; CHECK-NEXT:    andi $a1, $a1, 1
+; CHECK-NEXT:    slli.d $a1, $a1, 12
+; CHECK-NEXT:    or $a0, $a0, $a1
+; CHECK-NEXT:    vpickve2gr.b $a1, $vr0, 13
+; CHECK-NEXT:    andi $a1, $a1, 1
+; CHECK-NEXT:    slli.d $a1, $a1, 13
+; CHECK-NEXT:    or $a0, $a0, $a1
+; CHECK-NEXT:    vpickve2gr.b $a1, $vr0, 14
+; CHECK-NEXT:    andi $a1, $a1, 1
+; CHECK-NEXT:    slli.d $a1, $a1, 14
+; CHECK-NEXT:    or $a0, $a0, $a1
+; CHECK-NEXT:    vpickve2gr.b $a1, $vr0, 15
+; CHECK-NEXT:    andi $a1, $a1, 1
+; CHECK-NEXT:    ld.bu $a2, $sp, 16
+; CHECK-NEXT:    slli.d $a1, $a1, 15
+; CHECK-NEXT:    or $a0, $a0, $a1
+; CHECK-NEXT:    ld.bu $a1, $sp, 17
+; CHECK-NEXT:    andi $a2, $a2, 1
+; CHECK-NEXT:    slli.d $a2, $a2, 16
+; CHECK-NEXT:    or $a0, $a0, $a2
+; CHECK-NEXT:    andi $a1, $a1, 1
+; CHECK-NEXT:    ld.bu $a2, $sp, 18
+; CHECK-NEXT:    slli.d $a1, $a1, 17
+; CHECK-NEXT:    or $a0, $a0, $a1
+; CHECK-NEXT:    ld.bu $a1, $sp, 19
+; CHECK-NEXT:    andi $a2, $a2, 1
+; CHECK-NEXT:    slli.d $a2, $a2, 18
+; CHECK-NEXT:    or $a0, $a0, $a2
+; CHECK-NEXT:    andi $a1, $a1, 1
+; CHECK-NEXT:    ld.bu $a2, $sp, 20
+; CHECK-NEXT:    slli.d $a1, $a1, 19
+; CHECK-NEXT:    or $a0, $a0, $a1
+; CHECK-NEXT:    ld.bu $a1, $sp, 21
+; CHECK-NEXT:    andi $a2, $a2, 1
+; CHECK-NEXT:    slli.d $a2, $a2, 20
+; CHECK-NEXT:    or $a0, $a0, $a2
+; CHECK-NEXT:    andi $a1, $a1, 1
+; CHECK-NEXT:    ld.bu $a2, $sp, 22
+; CHECK-NEXT:    slli.d $a1, $a1, 21
+; CHECK-NEXT:    or $a0, $a0, $a1
+; CHECK-NEXT:    ld.bu $a1, $sp, 23
+; CHECK-NEXT:    andi $a2, $a2, 1
+; CHECK-NEXT:    slli.d $a2, $a2, 22
+; CHECK-NEXT:    or $a0, $a0, $a2
+; CHECK-NEXT:    andi $a1, $a1, 1
+; CHECK-NEXT:    ld.bu $a2, $sp, 24
+; CHECK-NEXT:    slli.d $a1, $a1, 23
+; CHECK-NEXT:    or $a0, $a0, $a1
+; CHECK-NEXT:    ld.bu $a1, $sp, 25
+; CHECK-NEXT:    andi $a2, $a2, 1
+; CHECK-NEXT:    slli.d $a2, $a2, 24
+; CHECK-NEXT:    or $a0, $a0, $a2
+; CHECK-NEXT:    andi $a1, $a1, 1
+; CHECK-NEXT:    ld.bu $a2, $sp, 26
+; CHECK-NEXT:    slli.d $a1, $a1, 25
+; CHECK-NEXT:    or $a0, $a0, $a1
+; CHECK-NEXT:    ld.bu $a1, $sp, 27
+; CHECK-NEXT:    andi $a2, $a2, 1
+; CHECK-NEXT:    slli.d $a2, $a2, 26
+; CHECK-NEXT:    or $a0, $a0, $a2
+; CHECK-NEXT:    andi $a1, $a1, 1
+; CHECK-NEXT:    ld.bu $a2, $sp, 28
+; CHECK-NEXT:    slli.d $a1, $a1, 27
+; CHECK-NEXT:    or $a0, $a0, $a1
+; CHECK-NEXT:    ld.bu $a1, $sp, 29
+; CHECK-NEXT:    andi $a2, $a2, 1
+; CHECK-NEXT:    slli.d $a2, $a2, 28
+; CHECK-NEXT:    or $a0, $a0, $a2
+; CHECK-NEXT:    andi $a1, $a1, 1
+; CHECK-NEXT:    ld.bu $a2, $sp, 30
+; CHECK-NEXT:    slli.d $a1, $a1, 29
+; CHECK-NEXT:    or $a0, $a0, $a1
+; CHECK-NEXT:    ld.b $a1, $sp, 31
+; CHECK-NEXT:    andi $a2, $a2, 1
+; CHECK-NEXT:    slli.d $a2, $a2, 30
+; CHECK-NEXT:    or $a0, $a0, $a2
+; CHECK-NEXT:    slli.d $a1, $a1, 31
+; CHECK-NEXT:    or $a0, $a0, $a1
+; CHECK-NEXT:    bstrpick.d $a0, $a0, 31, 0
+; CHECK-NEXT:    addi.d $sp, $fp, -64
+; CHECK-NEXT:    ld.d $fp, $sp, 48 # 8-byte Folded Reload
+; CHECK-NEXT:    ld.d $ra, $sp, 56 # 8-byte Folded Reload
+; CHECK-NEXT:    addi.d $sp, $sp, 64
+; CHECK-NEXT:    ret
+entry:
+  %1 = icmp eq <32 x i8> %a, splat (i8 0)
+  %2 = bitcast <32 x i1> %1 to i32
+  ret i32 %2
+}
+
+define i32 @xmsk_eq_allones_i8(<32 x i8 > %a) {
+; CHECK-LABEL: xmsk_eq_allones_i8:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    addi.d $sp, $sp, -64
+; CHECK-NEXT:    .cfi_def_cfa_offset 64
+; CHECK-NEXT:    st.d $ra, $sp, 56 # 8-byte Folded Spill
+; CHECK-NEXT:    st.d $fp, $sp, 48 # 8-byte Folded Spill
+; CHECK-NEXT:    .cfi_offset 1, -8
+; CHECK-NEXT:    .cfi_offset 22, -16
+; CHECK-NEXT:    addi.d $fp, $sp, 64
+; CHECK-NEXT:    .cfi_def_cfa 22, 0
+; CHECK-NEXT:    bstrins.d $sp, $zero, 4, 0
+; CHECK-NEXT:    xvseqi.b $xr0, $xr0, -1
+; CHECK-NEXT:    xvst $xr0, $sp, 0
+; CHECK-NEXT:    vpickve2gr.b $a0, $vr0, 0
+; CHECK-NEXT:    vpickve2gr.b $a1, $vr0, 1
+; CHECK-NEXT:    andi $a1, $a1, 1
+; CHECK-NEXT:    bstrins.d $a0, $a1, 63, 1
+; CHECK-NEXT:    vpickve2gr.b $a1, $vr0, 2
+; CHECK-NEXT:    bstrins.d $a0, $a1, 2, 2
+; CHECK-NEXT:    vpickve2gr.b $a1, $vr0, 3
+; CHECK-NEXT:    bstrins.d $a0, $a1, 3, 3
+; CHECK-NEXT:    vpickve2gr.b $a1, $vr0, 4
+; CHECK-NEXT:    bstrins.d $a0, $a1, 4, 4
+; CHECK-NEXT:    vpickve2gr.b $a1, $vr0, 5
+; CHECK-NEXT:    bstrins.d $a0, $a1, 5, 5
+; CHECK-NEXT:    vpickve2gr.b $a1, $vr0, 6
+; CHECK-NEXT:    andi $a1, $a1, 1
+; CHECK-NEXT:    slli.d $a1, $a1, 6
+; CHECK-NEXT:    or $a0, $a0, $a1
+; CHECK-NEXT:    vpickve2gr.b $a1, $vr0, 7
+; CHECK-NEXT:    andi $a1, $a1, 1
+; CHECK-NEXT:    slli.d $a1, $a1, 7
+; CHECK-NEXT:    or $a0, $a0, $a1
+; CHECK-NEXT:    vpickve2gr.b $a1, $vr0, 8
+; CHECK-NEXT:    andi $a1, $a1, 1
+; CHECK-NEXT:    slli.d $a1, $a1, 8
+; CHECK-NEXT:    or $a0, $a0, $a1
+; CHECK-NEXT:    vpickve2gr.b $a1, $vr0, 9
+; CHECK-NEXT:    andi $a1, $a1, 1
+; CHECK-NEXT:    slli.d $a1, $a1, 9
+; CHECK-NEXT:    or $a0, $a0, $a1
+; CHECK-NEXT:    vpickve2gr.b $a1, $vr0, 10
+; CHECK-NEXT:    andi $a1, $a1, 1
+; CHECK-NEXT:    slli.d $a1, $a1, 10
+; CHECK-NEXT:    or $a0, $a0, $a1
+; CHECK-NEXT:    vpickve2gr.b $a1, $vr0, 11
+; CHECK-NEXT:    andi $a1, $a1, 1
+; CHECK-NEXT:    slli.d $a1, $a1, 11
+; CHECK-NEXT:    or $a0, $a0, $a1
+; CHECK-NEXT:    vpickve2gr.b $a1, $vr0, 12
+; CHECK-NEXT:    andi $a1, $a1, 1
+; CHECK-NEXT:    slli.d $a1, $a1, 12
+; CHECK-NEXT:    or $a0, $a0, $a1
+; CHECK-NEXT:    vpickve2gr.b $a1, $vr0, 13
+; CHECK-NEXT:    andi $a1, $a1, 1
+; CHECK-NEXT:    slli.d $a1, $a1, 13
+; CHECK-NEXT:    or $a0, $a0, $a1
+; CHECK-NEXT:    vpickve2gr.b $a1, $vr0, 14
+; CHECK-NEXT:    andi $a1, $a1, 1
+; CHECK-NEXT:    slli.d $a1, $a1, 14
+; CHECK-NEXT:    or $a0, $a0, $a1
+; CHECK-NEXT:    vpickve2gr.b $a1, $vr0, 15
+; CHECK-NEXT:    andi $a1, $a1, 1
+; CHECK-NEXT:    ld.bu $a2, $sp, 16
+; CHECK-NEXT:    slli.d $a1, $a1, 15
+; CHECK-NEXT:    or $a0, $a0, $a1
+; CHECK-NEXT:    ld.bu $a1, $sp, 17
+; CHECK-NEXT:    andi $a2, $a2, 1
+; CHECK-NEXT:    slli.d $a2, $a2, 16
+; CHECK-NEXT:    or $a0, $a0, $a2
+; CHECK-NEXT:    andi $a1, $a1, 1
+; CHECK-NEXT:    ld.bu $a2, $sp, 18
+; CHECK-NEXT:    slli.d $a1, $a1, 17
+; CHECK-NEXT:    or $a0, $a0, $a1
+; CHECK-NEXT:    ld.bu $a1, $sp, 19
+; CHECK-NEXT:    andi $a2, $a2, 1
+; CHECK-NEXT:    slli.d $a2, $a2, 18
+; CHECK-NEXT:    or $a0, $a0, $a2
+; CHECK-NEXT:    andi $a1, $a1, 1
+; CHECK-NEXT:    ld.bu $a2, $sp, 20
+; CHECK-NEXT:    slli.d $a1, $a1, 19
+; CHECK-NEXT:    or $a0, $a0, $a1
+; CHECK-NEXT:    ld.bu $a1, $sp, 21
+; CHECK-NEXT:    andi $a2, $a2, 1
+; CHECK-NEXT:    slli.d $a2, $a2, 20
+; CHECK-NEXT:    or $a0, $a0, $a2
+; CHECK-NEXT:    andi $a1, $a1, 1
+; CHECK-NEXT:    ld.bu $a2, $sp, 22
+; CHECK-NEXT:    slli.d $a1, $a1, 21
+; CHECK-NEXT:    or $a0, $a0, $a1
+; CHECK-NEXT:    ld.bu $a1, $sp, 23
+; CHECK-NEXT:    andi $a2, $a2, 1
+; CHECK-NEXT:    slli.d $a2, $a2, 22
+; CHECK-NEXT:    or $a0, $a0, $a2
+; CHECK-NEXT:    andi $a1, $a1, 1
+; CHECK-NEXT:    ld.bu $a2, $sp, 24
+; CHECK-NEXT:    slli.d $a1, $a1, 23
+; CHECK-NEXT:    or $a0, $a0, $a1
+; CHECK-NEXT:    ld.bu $a1, $sp, 25
+; CHECK-NEXT:    andi $a2, $a2, 1
+; CHECK-NEXT:    slli.d $a2, $a2, 24
+; CHECK-NEXT:    or $a0, $a0, $a2
+; CHECK-NEXT:    andi $a1, $a1, 1
+; CHECK-NEXT:    ld.bu $a2, $sp, 26
+; CHECK-NEXT:    slli.d $a1, $a1, 25
+; CHECK-NEXT:    or $a0, $a0, $a1
+; CHECK-NEXT:    ld.bu $a1, $sp, 27
+; CHECK-NEXT:    andi $a2, $a2, 1
+; CHECK-NEXT:    slli.d $a2, $a2, 26
+; CHECK-NEXT:    or $a0, $a0, $a2
+; CHECK-NEXT:    andi $a1, $a1, 1
+; CHECK-NEXT:    ld.bu $a2, $sp, 28
+; CHECK-NEXT:    slli.d $a1, $a1, 27
+; CHECK-NEXT:    or $a0, $a0, $a1
+; CHECK-NEXT:    ld.bu $a1, $sp, 29
+; CHECK-NEXT:    andi $a2, $a2, 1
+; CHECK-NEXT:    slli.d $a2, $a2, 28
+; CHECK-NEXT:    or $a0, $a0, $a2
+; CHECK-NEXT:    andi $a1, $a1, 1
+; CHECK-NEXT:    ld.bu $a2, $sp, 30
+; CHECK-NEXT:    slli.d $a1, $a1, 29
+; CHECK-NEXT:    or $a0, $a0, $a1
+; CHECK-NEXT:    ld.b $a1, $sp, 31
+; CHECK-NEXT:    andi $a2, $a2, 1
+; CHECK-NEXT:    slli.d $a2, $a2, 30
+; CHECK-NEXT:    or $a0, $a0, $a2
+; CHECK-NEXT:    slli.d $a1, $a1, 31
+; CHECK-NEXT:    or $a0, $a0, $a1
+; CHECK-NEXT:    bstrpick.d $a0, $a0, 31, 0
+; CHECK-NEXT:    addi.d $sp, $fp, -64
+; CHECK-NEXT:    ld.d $fp, $sp, 48 # 8-byte Folded Reload
+; CHECK-NEXT:    ld.d $ra, $sp, 56 # 8-byte Folded Reload
+; CHECK-NEXT:    addi.d $sp, $sp, 64
+; CHECK-NEXT:    ret
+entry:
+  %1 = icmp eq <32 x i8> %a, splat (i8 -1)
+  %2 = bitcast <32 x i1> %1 to i32
+  ret i32 %2
+}
+
+define i32 @xmsk_sgt_allones_i8(<32 x i8 > %a) {
+; CHECK-LABEL: xmsk_sgt_allones_i8:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    addi.d $sp, $sp, -64
+; CHECK-NEXT:    .cfi_def_cfa_offset 64
+; CHECK-NEXT:    st.d $ra, $sp, 56 # 8-byte Folded Spill
+; CHECK-NEXT:    st.d $fp, $sp, 48 # 8-byte Folded Spill
+; CHECK-NEXT:    .cfi_offset 1, -8
+; CHECK-NEXT:    .cfi_offset 22, -16
+; CHECK-NEXT:    addi.d $fp, $sp, 64
+; CHECK-NEXT:    .cfi_def_cfa 22, 0
+; CHECK-NEXT:    bstrins.d $sp, $zero, 4, 0
+; CHECK-NEXT:    xvrepli.b $xr1, -1
+; CHECK-NEXT:    xvslt.b $xr0, $xr1, $xr0
+; CHECK-NEXT:    xvst $xr0, $sp, 0
+; CHECK-NEXT:    vpickve2gr.b $a0, $vr0, 0
+; CHECK-NEXT:    vpickve2gr.b $a1, $vr0, 1
+; CHECK-NEXT:    andi $a1, $a1, 1
+; CHECK-NEXT:    bstrins.d $a0, $a1, 63, 1
+; CHECK-NEXT:    vpickve2gr.b $a1, $vr0, 2
+; CHECK-NEXT:    bstrins.d $a0, $a1, 2, 2
+; CHECK-NEXT:    vpickve2gr.b $a1, $vr0, 3
+; CHECK-NEXT:    bstrins.d $a0, $a1, 3, 3
+; CHECK-NEXT:    vpickve2gr.b $a1, $vr0, 4
+; CHECK-NEXT:    bstrins.d $a0, $a1, 4, 4
+; CHECK-NEXT:    vpickve2gr.b $a1, $vr0, 5
+; CHECK-NEXT:    bstrins.d $a0, $a1, 5, 5
+; CHECK-NEXT:    vpickve2gr.b $a1, $vr0, 6
+; CHECK-NEXT:    andi $a1, $a1, 1
+; CHECK-NEXT:    slli.d $a1, $a1, 6
+; CHECK-NEXT:    or $a0, $a0, $a1
+; CHECK-NEXT:    vpickve2gr.b $a1, $vr0, 7
+; CHECK-NEXT:    andi $a1, $a1, 1
+; CHECK-NEXT:    slli.d $a1, $a1, 7
+; CHECK-NEXT:    or $a0, $a0, $a1
+; CHECK-NEXT:    vpickve2gr.b $a1, $vr0, 8
+; CHECK-NEXT:    andi $a1, $a1, 1
+; CHECK-NEXT:    slli.d $a1, $a1, 8
+; CHECK-NEXT:    or $a0, $a0, $a1
+; CHECK-NEXT:    vpickve2gr.b $a1, $vr0, 9
+; CHECK-NEXT:    andi $a1, $a1, 1
+; CHECK-NEXT:    slli.d $a1, $a1, 9
+; CHECK-NEXT:    or $a0, $a0, $a1
+; CHECK-NEXT:    vpickve2gr.b $a1, $vr0, 10
+; CHECK-NEXT:    andi $a1, $a1, 1
+; CHECK-NEXT:    slli.d $a1, $a1, 10
+; CHECK-NEXT:    or $a0, $a0, $a1
+; CHECK-NEXT:    vpickve2gr.b $a1, $vr0, 11
+; CHECK-NEXT:    andi $a1, $a1, 1
+; CHECK-NEXT:    slli.d $a1, $a1, 11
+; CHECK-NEXT:    or $a0, $a0, $a1
+; CHECK-NEXT:    vpickve2gr.b $a1, $vr0, 12
+; CHECK-NEXT:    andi $a1, $a1, 1
+; CHECK-NEXT:    slli.d $a1, $a1, 12
+; CHECK-NEXT:    or $a0, $a0, $a1
+; CHECK-NEXT:    vpickve2gr.b $a1, $vr0, 13
+; CHECK-NEXT:    andi $a1, $a1, 1
+; CHECK-NEXT:    slli.d $a1, $a1, 13
+; CHECK-NEXT:    or $a0, $a0, $a1
+; CHECK-NEXT:    vpickve2gr.b $a1, $vr0, 14
+; CHECK-NEXT:    andi $a1, $a1, 1
+; CHECK-NEXT:    slli.d $a1, $a1, 14
+; CHECK-NEXT:    or $a0, $a0, $a1
+; CHECK-NEXT:    vpickve2gr.b $a1, $vr0, 15
+; CHECK-NEXT:    andi $a1, $a1, 1
+; CHECK-NEXT:    ld.bu $a2, $sp, 16
+; CHECK-NEXT:    slli.d $a1, $a1, 15
+; CHECK-NEXT:    or $a0, $a0, $a1
+; CHECK-NEXT:    ld.bu $a1, $sp, 17
+; CHECK-NEXT:    andi $a2, $a2, 1
+; CHECK-NEXT:    slli.d $a2, $a2, 16
+; CHECK-NEXT:    or $a0, $a0, $a2
+; CHECK-NEXT:    andi $a1, $a1, 1
+; CHECK-NEXT:    ld.bu $a2, $sp, 18
+; CHECK-NEXT:    slli.d $a1, $a1, 17
+; CHECK-NEXT:    or $a0, $a0, $a1
+; CHECK-NEXT:    ld.bu $a1, $sp, 19
+; CHECK-NEXT:    andi $a2, $a2, 1
+; CHECK-NEXT:    slli.d $a2, $a2, 18
+; CHECK-NEXT:    or $a0, $a0, $a2
+; CHECK-NEXT:    andi $a1, $a1, 1
+; CHECK-NEXT:    ld.bu $a2, $sp, 20
+; CHECK-NEXT:    slli.d $a1, $a1, 19
+; CHECK-NEXT:    or $a0, $a0, $a1
+; CHECK-NEXT:    ld.bu $a1, $sp, 21
+; CHECK-NEXT:    andi $a2, $a2, 1
+; CHECK-NEXT:    slli.d $a2, $a2, 20
+; CHECK-NEXT:    or $a0, $a0, $a2
+; CHECK-NEXT:    andi $a1, $a1, 1
+; CHECK-NEXT:    ld.bu $a2, $sp, 22
+; CHECK-NEXT:    slli.d $a1, $a1, 21
+; CHECK-NEXT:    or $a0, $a0, $a1
+; CHECK-NEXT:    ld.bu $a1, $sp, 23
+; CHECK-NEXT:    andi $a2, $a2, 1
+; CHECK-NEXT:    slli.d $a2, $a2, 22
+; CHECK-NEXT:    or $a0, $a0, $a2
+; CHECK-NEXT:    andi $a1, $a1, 1
+; CHECK-NEXT:    ld.bu $a2, $sp, 24
+; CHECK-NEXT:    slli.d $a1, $a1, 23
+; CHECK-NEXT:    or $a0, $a0, $a1
+; CHECK-NEXT:    ld.bu $a1, $sp, 25
+; CHECK-NEXT:    andi $a2, $a2, 1
+; CHECK-NEXT:    slli.d $a2, $a2, 24
+; CHECK-NEXT:    or $a0, $a0, $a2
+; CHECK-NEXT:    andi $a1, $a1, 1
+; CHECK-NEXT:    ld.bu $a2, $sp, 26
+; CHECK-NEXT:    slli.d $a1, $a1, 25
+; CHECK-NEXT:    or $a0, $a0, $a1
+; CHECK-NEXT:    ld.bu $a1, $sp, 27
+; CHECK-NEXT:    andi $a2, $a2, 1
+; CHECK-NEXT:    slli.d $a2, $a2, 26
+; CHECK-NEXT:    or $a0, $a0, $a2
+; CHECK-NEXT:    andi $a1, $a1, 1
+; CHECK-NEXT:    ld.bu $a2, $sp, 28
+; CHECK-NEXT:    slli.d $a1, $a1, 27
+; CHECK-NEXT:    or $a0, $a0, $a1
+; CHECK-NEXT:    ld.bu $a1, $sp, 29
+; CHECK-NEXT:    andi $a2, $a2, 1
+; CHECK-NEXT:    slli.d $a2, $a2, 28
+; CHECK-NEXT:    or $a0, $a0, $a2
+; CHECK-NEXT:    andi $a1, $a1, 1
+; CHECK-NEXT:    ld.bu $a2, $sp, 30
+; CHECK-NEXT:    slli.d $a1, $a1, 29
+; CHECK-NEXT:    or $a0, $a0, $a1
+; CHECK-NEXT:    ld.b $a1, $sp, 31
+; CHECK-NEXT:    andi $a2, $a2, 1
+; CHECK-NEXT:    slli.d $a2, $a2, 30
+; CHECK-NEXT:    or $a0, $a0, $a2
+; CHECK-NEXT:    slli.d $a1, $a1, 31
+; CHECK-NEXT:    or $a0, $a0, $a1
+; CHECK-NEXT:    bstrpick.d $a0, $a0, 31, 0
+; CHECK-NEXT:    addi.d $sp, $fp, -64
+; CHECK-NEXT:    ld.d $fp, $sp, 48 # 8-byte Folded Reload
+; CHECK-NEXT:    ld.d $ra, $sp, 56 # 8-byte Folded Reload
+; CHECK-NEXT:    addi.d $sp, $sp, 64
+; CHECK-NEXT:    ret
+entry:
+  %1 = icmp sgt <32 x i8> %a, splat (i8 -1)
+  %2 = bitcast <32 x i1> %1 to i32
+  ret i32 %2
+}
+
+define i32 @xmsk_sge_allzeros_i8(<32 x i8 > %a) {
+; CHECK-LABEL: xmsk_sge_allzeros_i8:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    addi.d $sp, $sp, -64
+; CHECK-NEXT:    .cfi_def_cfa_offset 64
+; CHECK-NEXT:    st.d $ra, $sp, 56 # 8-byte Folded Spill
+; CHECK-NEXT:    st.d $fp, $sp, 48 # 8-byte Folded Spill
+; CHECK-NEXT:    .cfi_offset 1, -8
+; CHECK-NEXT:    .cfi_offset 22, -16
+; CHECK-NEXT:    addi.d $fp, $sp, 64
+; CHECK-NEXT:    .cfi_def_cfa 22, 0
+; CHECK-NEXT:    bstrins.d $sp, $zero, 4, 0
+; CHECK-NEXT:    xvrepli.b $xr1, 0
+; CHECK-NEXT:    xvsle.b $xr0, $xr1, $xr0
+; CHECK-NEXT:    xvst $xr0, $sp, 0
+; CHECK-NEXT:    vpickve2gr.b $a0, $vr0, 0
+; CHECK-NEXT:    vpickve2gr.b $a1, $vr0, 1
+; CHECK-NEXT:    andi $a1, $a1, 1
+; CHECK-NEXT:    bstrins.d $a0, $a1, 63, 1
+; CHECK-NEXT:    vpickve2gr.b $a1, $vr0, 2
+; CHECK-NEXT:    bstrins.d $a0, $a1, 2, 2
+; CHECK-NEXT:    vpickve2gr.b $a1, $vr0, 3
+; CHECK-NEXT:    bstrins.d $a0, $a1, 3, 3
+; CHECK-NEXT:    vpickve2gr.b $a1, $vr0, 4
+; CHECK-NEXT:    bstrins.d $a0, $a1, 4, 4
+; CHECK-NEXT:    vpickve2gr.b $a1, $vr0, 5
+; CHECK-NEXT:    bstrins.d $a0, $a1, 5, 5
+; CHECK-NEXT:    vpickve2gr.b $a1, $vr0, 6
+; CHECK-NEXT:    andi $a1, $a1, 1
+; CHECK-NEXT:    slli.d $a1, $a1, 6
+; CHECK-NEXT:    or $a0, $a0, $a1
+; CHECK-NEXT:    vpickve2gr.b $a1, $vr0, 7
+; CHECK-NEXT:    andi $a1, $a1, 1
+; CHECK-NEXT:    slli.d $a1, $a1, 7
+; CHECK-NEXT:    or $a0, $a0, $a1
+; CHECK-NEXT:    vpickve2gr.b $a1, $vr0, 8
+; CHECK-NEXT:    andi $a1, $a1, 1
+; CHECK-NEXT:    slli.d $a1, $a1, 8
+; CHECK-NEXT:    or $a0, $a0, $a1
+; CHECK-NEXT:    vpickve2gr.b $a1, $vr0, 9
+; CHECK-NEXT:    andi $a1, $a1, 1
+; CHECK-NEXT:    slli.d $a1, $a1, 9
+; CHECK-NEXT:    or $a0, $a0, $a1
+; CHECK-NEXT:    vpickve2gr.b $a1, $vr0, 10
+; CHECK-NEXT:    andi $a1, $a1, 1
+; CHECK-NEXT:    slli.d $a1, $a1, 10
+; CHECK-NEXT:    or $a0, $a0, $a1
+; CHECK-NEXT:    vpickve2gr.b $a1, $vr0, 11
+; CHECK-NEXT:    andi $a1, $a1, 1
+; CHECK-NEXT:    slli.d $a1, $a1, 11
+; CHECK-NEXT:    or $a0, $a0, $a1
+; CHECK-NEXT:    vpickve2gr.b $a1, $vr0, 12
+; CHECK-NEXT:    andi $a1, $a1, 1
+; CHECK-NEXT:    slli.d $a1, $a1, 12
+; CHECK-NEXT:    or $a0, $a0, $a1
+; CHECK-NEXT:    vpickve2gr.b $a1, $vr0, 13
+; CHECK-NEXT:    andi $a1, $a1, 1
+; CHECK-NEXT:    slli.d $a1, $a1, 13
+; CHECK-NEXT:    or $a0, $a0, $a1
+; CHECK-NEXT:    vpickve2gr.b $a1, $vr0, 14
+; CHECK-NEXT:    andi $a1, $a1, 1
+; CHECK-NEXT:    slli.d $a1, $a1, 14
+; CHECK-NEXT:    or $a0, $a0, $a1
+; CHECK-NEXT:    vpickve2gr.b $a1, $vr0, 15
+; CHECK-NEXT:    andi $a1, $a1, 1
+; CHECK-NEXT:    ld.bu $a2, $sp, 16
+; CHECK-NEXT:    slli.d $a1, $a1, 15
+; CHECK-NEXT:    or $a0, $a0, $a1
+; CHECK-NEXT:    ld.bu $a1, $sp, 17
+; CHECK-NEXT:    andi $a2, $a2, 1
+; CHECK-NEXT:    slli.d $a2, $a2, 16
+; CHECK-NEXT:    or $a0, $a0, $a2
+; CHECK-NEXT:    andi $a1, $a1, 1
+; CHECK-NEXT:    ld.bu $a2, $sp, 18
+; CHECK-NEXT:    slli.d $a1, $a1, 17
+; CHECK-NEXT:    or $a0, $a0, $a1
+; CHECK-NEXT:    ld.bu $a1, $sp, 19
+; CHECK-NEXT:    andi $a2, $a2, 1
+; CHECK-NEXT:    slli.d $a2, $a2, 18
+; CHECK-NEXT:    or $a0, $a0, $a2
+; CHECK-NEXT:    andi $a1, $a1, 1
+; CHECK-NEXT:    ld.bu $a2, $sp, 20
+; CHECK-NEXT:    slli.d $a1, $a1, 19
+; CHECK-NEXT:    or $a0, $a0, $a1
+; CHECK-NEXT:    ld.bu $a1, $sp, 21
+; CHECK-NEXT:    andi $a2, $a2, 1
+; CHECK-NEXT:    slli.d $a2, $a2, 20
+; CHECK-NEXT:    or $a0, $a0, $a2
+; CHECK-NEXT:    andi $a1, $a1, 1
+; CHECK-NEXT:    ld.bu $a2, $sp, 22
+; CHECK-NEXT:    slli.d $a1, $a1, 21
+; CHECK-NEXT:    or $a0, $a0, $a1
+; CHECK-NEXT:    ld.bu $a1, $sp, 23
+; CHECK-NEXT:    andi $a2, $a2, 1
+; CHECK-NEXT:    slli.d $a2, $a2, 22
+; CHECK-NEXT:    or $a0, $a0, $a2
+; CHECK-NEXT:    andi $a1, $a1, 1
+; CHECK-NEXT:    ld.bu $a2, $sp, 24
+; CHECK-NEXT:    slli.d $a1, $a1, 23
+; CHECK-NEXT:    or $a0, $a0, $a1
+; CHECK-NEXT:    ld.bu $a1, $sp, 25
+; CHECK-NEXT:    andi $a2, $a2, 1
+; CHECK-NEXT:    slli.d $a2, $a2, 24
+; CHECK-NEXT:    or $a0, $a0, $a2
+; CHECK-NEXT:    andi $a1, $a1, 1
+; CHECK-NEXT:    ld.bu $a2, $sp, 26
+; CHECK-NEXT:    slli.d $a1, $a1, 25
+; CHECK-NEXT:    or $a0, $a0, $a1
+; CHECK-NEXT:    ld.bu $a1, $sp, 27
...
[truncated]

@heiher heiher requested a review from tangaac May 30, 2025 09:24
@heiher heiher merged commit d979423 into llvm:main Jun 5, 2025
11 checks passed
@heiher heiher deleted the vmsk-1 branch June 5, 2025 12:26
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Reviewers

@tangaac tangaac tangaac approved these changes

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
@heiher @llvmbot @tangaac