[llvm] f1a29ec - [AArch64] Add i128 and fp128 tests to fptoi. NFC
David Green via llvm-commits
llvm-commits at lists.llvm.org
Fri Jun 14 02:52:06 PDT 2024
Author: David Green
Date: 2024-06-14T10:52:01+01:00
New Revision: f1a29ec082ead82c6a4d61e515222d6bcf046a5b
URL: https://github.com/llvm/llvm-project/commit/f1a29ec082ead82c6a4d61e515222d6bcf046a5b
DIFF: https://github.com/llvm/llvm-project/commit/f1a29ec082ead82c6a4d61e515222d6bcf046a5b.diff
LOG: [AArch64] Add i128 and fp128 tests to fptoi. NFC
Added:
Modified:
llvm/test/CodeGen/AArch64/fptoi.ll
Removed:
################################################################################
diff --git a/llvm/test/CodeGen/AArch64/fptoi.ll b/llvm/test/CodeGen/AArch64/fptoi.ll
index 01585d02adcb1..3b8054a635bcd 100644
--- a/llvm/test/CodeGen/AArch64/fptoi.ll
+++ b/llvm/test/CodeGen/AArch64/fptoi.ll
@@ -1,8 +1,55 @@
; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py UTC_ARGS: --version 2
; RUN: llc -mtriple=aarch64 -verify-machineinstrs %s -o - | FileCheck %s --check-prefixes=CHECK,CHECK-SD,CHECK-SD-NOFP16
; RUN: llc -mtriple=aarch64 -mattr=+fullfp16 -verify-machineinstrs %s -o - | FileCheck %s --check-prefixes=CHECK,CHECK-SD,CHECK-SD-FP16
-; RUN: llc -mtriple=aarch64 -global-isel -verify-machineinstrs %s -o - | FileCheck %s --check-prefixes=CHECK,CHECK-GI,CHECK-GI-NOFP16
-; RUN: llc -mtriple=aarch64 -mattr=+fullfp16 -global-isel -verify-machineinstrs %s -o - | FileCheck %s --check-prefixes=CHECK,CHECK-GI,CHECK-GI-FP16
+; RUN: llc -mtriple=aarch64 -global-isel -verify-machineinstrs -global-isel-abort=2 %s -o - 2>&1 | FileCheck %s --check-prefixes=CHECK,CHECK-GI,CHECK-GI-NOFP16
+; RUN: llc -mtriple=aarch64 -mattr=+fullfp16 -global-isel -verify-machineinstrs -global-isel-abort=2 %s -o - 2>&1 | FileCheck %s --check-prefixes=CHECK,CHECK-GI,CHECK-GI-FP16
+
+; CHECK-GI: warning: Instruction selection used fallback path for fptos_f64_i128
+; CHECK-GI-NEXT: warning: Instruction selection used fallback path for fptou_f64_i128
+; CHECK-GI-NEXT: warning: Instruction selection used fallback path for fptos_f32_i128
+; CHECK-GI-NEXT: warning: Instruction selection used fallback path for fptou_f32_i128
+; CHECK-GI-NEXT: warning: Instruction selection used fallback path for fptos_f128_i64
+; CHECK-GI-NEXT: warning: Instruction selection used fallback path for fptou_f128_i64
+; CHECK-GI-NEXT: warning: Instruction selection used fallback path for fptos_f128_i32
+; CHECK-GI-NEXT: warning: Instruction selection used fallback path for fptou_f128_i32
+; CHECK-GI-NEXT: warning: Instruction selection used fallback path for fptos_f128_i16
+; CHECK-GI-NEXT: warning: Instruction selection used fallback path for fptou_f128_i16
+; CHECK-GI-NEXT: warning: Instruction selection used fallback path for fptos_f128_i8
+; CHECK-GI-NEXT: warning: Instruction selection used fallback path for fptou_f128_i8
+; CHECK-GI-NEXT: warning: Instruction selection used fallback path for fptos_f128_i128
+; CHECK-GI-NEXT: warning: Instruction selection used fallback path for fptou_f128_i128
+; CHECK-GI-NEXT: warning: Instruction selection used fallback path for fptos_v2f64_v2i128
+; CHECK-GI-NEXT: warning: Instruction selection used fallback path for fptou_v2f64_v2i128
+; CHECK-GI-NEXT: warning: Instruction selection used fallback path for fptos_v3f64_v3i128
+; CHECK-GI-NEXT: warning: Instruction selection used fallback path for fptou_v3f64_v3i128
+; CHECK-GI-NEXT: warning: Instruction selection used fallback path for fptos_v2f32_v2i128
+; CHECK-GI-NEXT: warning: Instruction selection used fallback path for fptou_v2f32_v2i128
+; CHECK-GI-NEXT: warning: Instruction selection used fallback path for fptos_v3f32_v3i128
+; CHECK-GI-NEXT: warning: Instruction selection used fallback path for fptou_v3f32_v3i128
+; CHECK-GI-NEXT: warning: Instruction selection used fallback path for fptos_v2f16_v2i128
+; CHECK-GI-NEXT: warning: Instruction selection used fallback path for fptou_v2f16_v2i128
+; CHECK-GI-NEXT: warning: Instruction selection used fallback path for fptos_v3f16_v3i128
+; CHECK-GI-NEXT: warning: Instruction selection used fallback path for fptou_v3f16_v3i128
+; CHECK-GI-NEXT: warning: Instruction selection used fallback path for fptos_v2f128_v2i64
+; CHECK-GI-NEXT: warning: Instruction selection used fallback path for fptou_v2f128_v2i64
+; CHECK-GI-NEXT: warning: Instruction selection used fallback path for fptos_v3f128_v3i64
+; CHECK-GI-NEXT: warning: Instruction selection used fallback path for fptou_v3f128_v3i64
+; CHECK-GI-NEXT: warning: Instruction selection used fallback path for fptos_v2f128_v2i32
+; CHECK-GI-NEXT: warning: Instruction selection used fallback path for fptou_v2f128_v2i32
+; CHECK-GI-NEXT: warning: Instruction selection used fallback path for fptos_v3f128_v3i32
+; CHECK-GI-NEXT: warning: Instruction selection used fallback path for fptou_v3f128_v3i32
+; CHECK-GI-NEXT: warning: Instruction selection used fallback path for fptos_v2f128_v2i16
+; CHECK-GI-NEXT: warning: Instruction selection used fallback path for fptou_v2f128_v2i16
+; CHECK-GI-NEXT: warning: Instruction selection used fallback path for fptos_v3f128_v3i16
+; CHECK-GI-NEXT: warning: Instruction selection used fallback path for fptou_v3f128_v3i16
+; CHECK-GI-NEXT: warning: Instruction selection used fallback path for fptos_v2f128_v2i8
+; CHECK-GI-NEXT: warning: Instruction selection used fallback path for fptou_v2f128_v2i8
+; CHECK-GI-NEXT: warning: Instruction selection used fallback path for fptos_v3f128_v3i8
+; CHECK-GI-NEXT: warning: Instruction selection used fallback path for fptou_v3f128_v3i8
+; CHECK-GI-NEXT: warning: Instruction selection used fallback path for fptos_v2f128_v2i128
+; CHECK-GI-NEXT: warning: Instruction selection used fallback path for fptou_v2f128_v2i128
+; CHECK-GI-NEXT: warning: Instruction selection used fallback path for fptos_v3f128_v3i128
+; CHECK-GI-NEXT: warning: Instruction selection used fallback path for fptou_v3f128_v3i128
define i64 @fptos_f64_i64(double %a) {
; CHECK-LABEL: fptos_f64_i64:
@@ -94,6 +141,34 @@ entry:
ret i8 %c
}
+define i128 @fptos_f64_i128(double %a) {
+; CHECK-LABEL: fptos_f64_i128:
+; CHECK: // %bb.0: // %entry
+; CHECK-NEXT: str x30, [sp, #-16]! // 8-byte Folded Spill
+; CHECK-NEXT: .cfi_def_cfa_offset 16
+; CHECK-NEXT: .cfi_offset w30, -16
+; CHECK-NEXT: bl __fixdfti
+; CHECK-NEXT: ldr x30, [sp], #16 // 8-byte Folded Reload
+; CHECK-NEXT: ret
+entry:
+ %c = fptosi double %a to i128
+ ret i128 %c
+}
+
+define i128 @fptou_f64_i128(double %a) {
+; CHECK-LABEL: fptou_f64_i128:
+; CHECK: // %bb.0: // %entry
+; CHECK-NEXT: str x30, [sp, #-16]! // 8-byte Folded Spill
+; CHECK-NEXT: .cfi_def_cfa_offset 16
+; CHECK-NEXT: .cfi_offset w30, -16
+; CHECK-NEXT: bl __fixunsdfti
+; CHECK-NEXT: ldr x30, [sp], #16 // 8-byte Folded Reload
+; CHECK-NEXT: ret
+entry:
+ %c = fptoui double %a to i128
+ ret i128 %c
+}
+
define i64 @fptos_f32_i64(float %a) {
; CHECK-LABEL: fptos_f32_i64:
; CHECK: // %bb.0: // %entry
@@ -184,6 +259,34 @@ entry:
ret i8 %c
}
+define i128 @fptos_f32_i128(float %a) {
+; CHECK-LABEL: fptos_f32_i128:
+; CHECK: // %bb.0: // %entry
+; CHECK-NEXT: str x30, [sp, #-16]! // 8-byte Folded Spill
+; CHECK-NEXT: .cfi_def_cfa_offset 16
+; CHECK-NEXT: .cfi_offset w30, -16
+; CHECK-NEXT: bl __fixsfti
+; CHECK-NEXT: ldr x30, [sp], #16 // 8-byte Folded Reload
+; CHECK-NEXT: ret
+entry:
+ %c = fptosi float %a to i128
+ ret i128 %c
+}
+
+define i128 @fptou_f32_i128(float %a) {
+; CHECK-LABEL: fptou_f32_i128:
+; CHECK: // %bb.0: // %entry
+; CHECK-NEXT: str x30, [sp, #-16]! // 8-byte Folded Spill
+; CHECK-NEXT: .cfi_def_cfa_offset 16
+; CHECK-NEXT: .cfi_offset w30, -16
+; CHECK-NEXT: bl __fixunssfti
+; CHECK-NEXT: ldr x30, [sp], #16 // 8-byte Folded Reload
+; CHECK-NEXT: ret
+entry:
+ %c = fptoui float %a to i128
+ ret i128 %c
+}
+
define i64 @fptos_f16_i64(half %a) {
; CHECK-SD-NOFP16-LABEL: fptos_f16_i64:
; CHECK-SD-NOFP16: // %bb.0: // %entry
@@ -400,6 +503,200 @@ entry:
ret i8 %c
}
+define i128 @fptos_f16_i128(half %a) {
+; CHECK-SD-LABEL: fptos_f16_i128:
+; CHECK-SD: // %bb.0: // %entry
+; CHECK-SD-NEXT: str x30, [sp, #-16]! // 8-byte Folded Spill
+; CHECK-SD-NEXT: .cfi_def_cfa_offset 16
+; CHECK-SD-NEXT: .cfi_offset w30, -16
+; CHECK-SD-NEXT: bl __fixhfti
+; CHECK-SD-NEXT: ldr x30, [sp], #16 // 8-byte Folded Reload
+; CHECK-SD-NEXT: ret
+;
+; CHECK-GI-NOFP16-LABEL: fptos_f16_i128:
+; CHECK-GI-NOFP16: // %bb.0: // %entry
+; CHECK-GI-NOFP16-NEXT: fcvt s0, h0
+; CHECK-GI-NOFP16-NEXT: fcvtzs x0, s0
+; CHECK-GI-NOFP16-NEXT: asr x1, x0, #63
+; CHECK-GI-NOFP16-NEXT: ret
+;
+; CHECK-GI-FP16-LABEL: fptos_f16_i128:
+; CHECK-GI-FP16: // %bb.0: // %entry
+; CHECK-GI-FP16-NEXT: fcvtzs x0, h0
+; CHECK-GI-FP16-NEXT: asr x1, x0, #63
+; CHECK-GI-FP16-NEXT: ret
+entry:
+ %c = fptosi half %a to i128
+ ret i128 %c
+}
+
+define i128 @fptou_f16_i128(half %a) {
+; CHECK-SD-LABEL: fptou_f16_i128:
+; CHECK-SD: // %bb.0: // %entry
+; CHECK-SD-NEXT: str x30, [sp, #-16]! // 8-byte Folded Spill
+; CHECK-SD-NEXT: .cfi_def_cfa_offset 16
+; CHECK-SD-NEXT: .cfi_offset w30, -16
+; CHECK-SD-NEXT: bl __fixunshfti
+; CHECK-SD-NEXT: ldr x30, [sp], #16 // 8-byte Folded Reload
+; CHECK-SD-NEXT: ret
+;
+; CHECK-GI-NOFP16-LABEL: fptou_f16_i128:
+; CHECK-GI-NOFP16: // %bb.0: // %entry
+; CHECK-GI-NOFP16-NEXT: fcvt s0, h0
+; CHECK-GI-NOFP16-NEXT: mov x1, xzr
+; CHECK-GI-NOFP16-NEXT: fcvtzu x0, s0
+; CHECK-GI-NOFP16-NEXT: ret
+;
+; CHECK-GI-FP16-LABEL: fptou_f16_i128:
+; CHECK-GI-FP16: // %bb.0: // %entry
+; CHECK-GI-FP16-NEXT: fcvtzu x0, h0
+; CHECK-GI-FP16-NEXT: mov x1, xzr
+; CHECK-GI-FP16-NEXT: ret
+entry:
+ %c = fptoui half %a to i128
+ ret i128 %c
+}
+
+define i64 @fptos_f128_i64(fp128 %a) {
+; CHECK-LABEL: fptos_f128_i64:
+; CHECK: // %bb.0: // %entry
+; CHECK-NEXT: str x30, [sp, #-16]! // 8-byte Folded Spill
+; CHECK-NEXT: .cfi_def_cfa_offset 16
+; CHECK-NEXT: .cfi_offset w30, -16
+; CHECK-NEXT: bl __fixtfdi
+; CHECK-NEXT: ldr x30, [sp], #16 // 8-byte Folded Reload
+; CHECK-NEXT: ret
+entry:
+ %c = fptosi fp128 %a to i64
+ ret i64 %c
+}
+
+define i64 @fptou_f128_i64(fp128 %a) {
+; CHECK-LABEL: fptou_f128_i64:
+; CHECK: // %bb.0: // %entry
+; CHECK-NEXT: str x30, [sp, #-16]! // 8-byte Folded Spill
+; CHECK-NEXT: .cfi_def_cfa_offset 16
+; CHECK-NEXT: .cfi_offset w30, -16
+; CHECK-NEXT: bl __fixunstfdi
+; CHECK-NEXT: ldr x30, [sp], #16 // 8-byte Folded Reload
+; CHECK-NEXT: ret
+entry:
+ %c = fptoui fp128 %a to i64
+ ret i64 %c
+}
+
+define i32 @fptos_f128_i32(fp128 %a) {
+; CHECK-LABEL: fptos_f128_i32:
+; CHECK: // %bb.0: // %entry
+; CHECK-NEXT: str x30, [sp, #-16]! // 8-byte Folded Spill
+; CHECK-NEXT: .cfi_def_cfa_offset 16
+; CHECK-NEXT: .cfi_offset w30, -16
+; CHECK-NEXT: bl __fixtfsi
+; CHECK-NEXT: ldr x30, [sp], #16 // 8-byte Folded Reload
+; CHECK-NEXT: ret
+entry:
+ %c = fptosi fp128 %a to i32
+ ret i32 %c
+}
+
+define i32 @fptou_f128_i32(fp128 %a) {
+; CHECK-LABEL: fptou_f128_i32:
+; CHECK: // %bb.0: // %entry
+; CHECK-NEXT: str x30, [sp, #-16]! // 8-byte Folded Spill
+; CHECK-NEXT: .cfi_def_cfa_offset 16
+; CHECK-NEXT: .cfi_offset w30, -16
+; CHECK-NEXT: bl __fixunstfsi
+; CHECK-NEXT: ldr x30, [sp], #16 // 8-byte Folded Reload
+; CHECK-NEXT: ret
+entry:
+ %c = fptoui fp128 %a to i32
+ ret i32 %c
+}
+
+define i16 @fptos_f128_i16(fp128 %a) {
+; CHECK-LABEL: fptos_f128_i16:
+; CHECK: // %bb.0: // %entry
+; CHECK-NEXT: str x30, [sp, #-16]! // 8-byte Folded Spill
+; CHECK-NEXT: .cfi_def_cfa_offset 16
+; CHECK-NEXT: .cfi_offset w30, -16
+; CHECK-NEXT: bl __fixtfsi
+; CHECK-NEXT: ldr x30, [sp], #16 // 8-byte Folded Reload
+; CHECK-NEXT: ret
+entry:
+ %c = fptosi fp128 %a to i16
+ ret i16 %c
+}
+
+define i16 @fptou_f128_i16(fp128 %a) {
+; CHECK-LABEL: fptou_f128_i16:
+; CHECK: // %bb.0: // %entry
+; CHECK-NEXT: str x30, [sp, #-16]! // 8-byte Folded Spill
+; CHECK-NEXT: .cfi_def_cfa_offset 16
+; CHECK-NEXT: .cfi_offset w30, -16
+; CHECK-NEXT: bl __fixtfsi
+; CHECK-NEXT: ldr x30, [sp], #16 // 8-byte Folded Reload
+; CHECK-NEXT: ret
+entry:
+ %c = fptoui fp128 %a to i16
+ ret i16 %c
+}
+
+define i8 @fptos_f128_i8(fp128 %a) {
+; CHECK-LABEL: fptos_f128_i8:
+; CHECK: // %bb.0: // %entry
+; CHECK-NEXT: str x30, [sp, #-16]! // 8-byte Folded Spill
+; CHECK-NEXT: .cfi_def_cfa_offset 16
+; CHECK-NEXT: .cfi_offset w30, -16
+; CHECK-NEXT: bl __fixtfsi
+; CHECK-NEXT: ldr x30, [sp], #16 // 8-byte Folded Reload
+; CHECK-NEXT: ret
+entry:
+ %c = fptosi fp128 %a to i8
+ ret i8 %c
+}
+
+define i8 @fptou_f128_i8(fp128 %a) {
+; CHECK-LABEL: fptou_f128_i8:
+; CHECK: // %bb.0: // %entry
+; CHECK-NEXT: str x30, [sp, #-16]! // 8-byte Folded Spill
+; CHECK-NEXT: .cfi_def_cfa_offset 16
+; CHECK-NEXT: .cfi_offset w30, -16
+; CHECK-NEXT: bl __fixtfsi
+; CHECK-NEXT: ldr x30, [sp], #16 // 8-byte Folded Reload
+; CHECK-NEXT: ret
+entry:
+ %c = fptoui fp128 %a to i8
+ ret i8 %c
+}
+
+define i128 @fptos_f128_i128(fp128 %a) {
+; CHECK-LABEL: fptos_f128_i128:
+; CHECK: // %bb.0: // %entry
+; CHECK-NEXT: str x30, [sp, #-16]! // 8-byte Folded Spill
+; CHECK-NEXT: .cfi_def_cfa_offset 16
+; CHECK-NEXT: .cfi_offset w30, -16
+; CHECK-NEXT: bl __fixtfti
+; CHECK-NEXT: ldr x30, [sp], #16 // 8-byte Folded Reload
+; CHECK-NEXT: ret
+entry:
+ %c = fptosi fp128 %a to i128
+ ret i128 %c
+}
+
+define i128 @fptou_f128_i128(fp128 %a) {
+; CHECK-LABEL: fptou_f128_i128:
+; CHECK: // %bb.0: // %entry
+; CHECK-NEXT: str x30, [sp, #-16]! // 8-byte Folded Spill
+; CHECK-NEXT: .cfi_def_cfa_offset 16
+; CHECK-NEXT: .cfi_offset w30, -16
+; CHECK-NEXT: bl __fixunstfti
+; CHECK-NEXT: ldr x30, [sp], #16 // 8-byte Folded Reload
+; CHECK-NEXT: ret
+entry:
+ %c = fptoui fp128 %a to i128
+ ret i128 %c
+}
+
define <2 x i64> @fptos_v2f64_v2i64(<2 x double> %a) {
; CHECK-LABEL: fptos_v2f64_v2i64:
; CHECK: // %bb.0: // %entry
@@ -1183,14 +1480,14 @@ define <8 x i16> @fptos_v8f64_v8i16(<8 x double> %a) {
; CHECK-SD: // %bb.0: // %entry
; CHECK-SD-NEXT: fcvtzs v3.2d, v3.2d
; CHECK-SD-NEXT: fcvtzs v2.2d, v2.2d
-; CHECK-SD-NEXT: adrp x8, .LCPI54_0
+; CHECK-SD-NEXT: adrp x8, .LCPI70_0
; CHECK-SD-NEXT: fcvtzs v1.2d, v1.2d
; CHECK-SD-NEXT: fcvtzs v0.2d, v0.2d
; CHECK-SD-NEXT: xtn v6.2s, v3.2d
; CHECK-SD-NEXT: xtn v5.2s, v2.2d
; CHECK-SD-NEXT: xtn v4.2s, v1.2d
; CHECK-SD-NEXT: xtn v3.2s, v0.2d
-; CHECK-SD-NEXT: ldr q0, [x8, :lo12:.LCPI54_0]
+; CHECK-SD-NEXT: ldr q0, [x8, :lo12:.LCPI70_0]
; CHECK-SD-NEXT: tbl v0.16b, { v3.16b, v4.16b, v5.16b, v6.16b }, v0.16b
; CHECK-SD-NEXT: ret
;
@@ -1214,14 +1511,14 @@ define <8 x i16> @fptou_v8f64_v8i16(<8 x double> %a) {
; CHECK-SD: // %bb.0: // %entry
; CHECK-SD-NEXT: fcvtzs v3.2d, v3.2d
; CHECK-SD-NEXT: fcvtzs v2.2d, v2.2d
-; CHECK-SD-NEXT: adrp x8, .LCPI55_0
+; CHECK-SD-NEXT: adrp x8, .LCPI71_0
; CHECK-SD-NEXT: fcvtzs v1.2d, v1.2d
; CHECK-SD-NEXT: fcvtzs v0.2d, v0.2d
; CHECK-SD-NEXT: xtn v6.2s, v3.2d
; CHECK-SD-NEXT: xtn v5.2s, v2.2d
; CHECK-SD-NEXT: xtn v4.2s, v1.2d
; CHECK-SD-NEXT: xtn v3.2s, v0.2d
-; CHECK-SD-NEXT: ldr q0, [x8, :lo12:.LCPI55_0]
+; CHECK-SD-NEXT: ldr q0, [x8, :lo12:.LCPI71_0]
; CHECK-SD-NEXT: tbl v0.16b, { v3.16b, v4.16b, v5.16b, v6.16b }, v0.16b
; CHECK-SD-NEXT: ret
;
@@ -1245,7 +1542,7 @@ define <16 x i16> @fptos_v16f64_v16i16(<16 x double> %a) {
; CHECK-SD: // %bb.0: // %entry
; CHECK-SD-NEXT: fcvtzs v3.2d, v3.2d
; CHECK-SD-NEXT: fcvtzs v7.2d, v7.2d
-; CHECK-SD-NEXT: adrp x8, .LCPI56_0
+; CHECK-SD-NEXT: adrp x8, .LCPI72_0
; CHECK-SD-NEXT: fcvtzs v2.2d, v2.2d
; CHECK-SD-NEXT: fcvtzs v6.2d, v6.2d
; CHECK-SD-NEXT: fcvtzs v1.2d, v1.2d
@@ -1258,7 +1555,7 @@ define <16 x i16> @fptos_v16f64_v16i16(<16 x double> %a) {
; CHECK-SD-NEXT: xtn v22.2s, v6.2d
; CHECK-SD-NEXT: xtn v17.2s, v1.2d
; CHECK-SD-NEXT: xtn v21.2s, v5.2d
-; CHECK-SD-NEXT: ldr q1, [x8, :lo12:.LCPI56_0]
+; CHECK-SD-NEXT: ldr q1, [x8, :lo12:.LCPI72_0]
; CHECK-SD-NEXT: xtn v16.2s, v0.2d
; CHECK-SD-NEXT: xtn v20.2s, v4.2d
; CHECK-SD-NEXT: tbl v0.16b, { v16.16b, v17.16b, v18.16b, v19.16b }, v1.16b
@@ -1292,7 +1589,7 @@ define <16 x i16> @fptou_v16f64_v16i16(<16 x double> %a) {
; CHECK-SD: // %bb.0: // %entry
; CHECK-SD-NEXT: fcvtzs v3.2d, v3.2d
; CHECK-SD-NEXT: fcvtzs v7.2d, v7.2d
-; CHECK-SD-NEXT: adrp x8, .LCPI57_0
+; CHECK-SD-NEXT: adrp x8, .LCPI73_0
; CHECK-SD-NEXT: fcvtzs v2.2d, v2.2d
; CHECK-SD-NEXT: fcvtzs v6.2d, v6.2d
; CHECK-SD-NEXT: fcvtzs v1.2d, v1.2d
@@ -1305,7 +1602,7 @@ define <16 x i16> @fptou_v16f64_v16i16(<16 x double> %a) {
; CHECK-SD-NEXT: xtn v22.2s, v6.2d
; CHECK-SD-NEXT: xtn v17.2s, v1.2d
; CHECK-SD-NEXT: xtn v21.2s, v5.2d
-; CHECK-SD-NEXT: ldr q1, [x8, :lo12:.LCPI57_0]
+; CHECK-SD-NEXT: ldr q1, [x8, :lo12:.LCPI73_0]
; CHECK-SD-NEXT: xtn v16.2s, v0.2d
; CHECK-SD-NEXT: xtn v20.2s, v4.2d
; CHECK-SD-NEXT: tbl v0.16b, { v16.16b, v17.16b, v18.16b, v19.16b }, v1.16b
@@ -1352,7 +1649,7 @@ define <32 x i16> @fptos_v32f64_v32i16(<32 x double> %a) {
; CHECK-SD-NEXT: .cfi_offset b15, -64
; CHECK-SD-NEXT: fcvtzs v3.2d, v3.2d
; CHECK-SD-NEXT: fcvtzs v18.2d, v2.2d
-; CHECK-SD-NEXT: adrp x8, .LCPI58_0
+; CHECK-SD-NEXT: adrp x8, .LCPI74_0
; CHECK-SD-NEXT: fcvtzs v19.2d, v1.2d
; CHECK-SD-NEXT: ldp q20, q21, [sp, #160]
; CHECK-SD-NEXT: fcvtzs v22.2d, v0.2d
@@ -1387,7 +1684,7 @@ define <32 x i16> @fptos_v32f64_v32i16(<32 x double> %a) {
; CHECK-SD-NEXT: xtn v12.2s, v16.2d
; CHECK-SD-NEXT: xtn v27.2s, v5.2d
; CHECK-SD-NEXT: xtn v26.2s, v4.2d
-; CHECK-SD-NEXT: ldr q4, [x8, :lo12:.LCPI58_0]
+; CHECK-SD-NEXT: ldr q4, [x8, :lo12:.LCPI74_0]
; CHECK-SD-NEXT: tbl v0.16b, { v0.16b, v1.16b, v2.16b, v3.16b }, v4.16b
; CHECK-SD-NEXT: tbl v2.16b, { v8.16b, v9.16b, v10.16b, v11.16b }, v4.16b
; CHECK-SD-NEXT: tbl v3.16b, { v12.16b, v13.16b, v14.16b, v15.16b }, v4.16b
@@ -1456,7 +1753,7 @@ define <32 x i16> @fptou_v32f64_v32i16(<32 x double> %a) {
; CHECK-SD-NEXT: .cfi_offset b15, -64
; CHECK-SD-NEXT: fcvtzs v3.2d, v3.2d
; CHECK-SD-NEXT: fcvtzs v18.2d, v2.2d
-; CHECK-SD-NEXT: adrp x8, .LCPI59_0
+; CHECK-SD-NEXT: adrp x8, .LCPI75_0
; CHECK-SD-NEXT: fcvtzs v19.2d, v1.2d
; CHECK-SD-NEXT: ldp q20, q21, [sp, #160]
; CHECK-SD-NEXT: fcvtzs v22.2d, v0.2d
@@ -1491,7 +1788,7 @@ define <32 x i16> @fptou_v32f64_v32i16(<32 x double> %a) {
; CHECK-SD-NEXT: xtn v12.2s, v16.2d
; CHECK-SD-NEXT: xtn v27.2s, v5.2d
; CHECK-SD-NEXT: xtn v26.2s, v4.2d
-; CHECK-SD-NEXT: ldr q4, [x8, :lo12:.LCPI59_0]
+; CHECK-SD-NEXT: ldr q4, [x8, :lo12:.LCPI75_0]
; CHECK-SD-NEXT: tbl v0.16b, { v0.16b, v1.16b, v2.16b, v3.16b }, v4.16b
; CHECK-SD-NEXT: tbl v2.16b, { v8.16b, v9.16b, v10.16b, v11.16b }, v4.16b
; CHECK-SD-NEXT: tbl v3.16b, { v12.16b, v13.16b, v14.16b, v15.16b }, v4.16b
@@ -1992,6 +2289,158 @@ entry:
ret <32 x i8> %c
}
+define <2 x i128> @fptos_v2f64_v2i128(<2 x double> %a) {
+; CHECK-LABEL: fptos_v2f64_v2i128:
+; CHECK: // %bb.0: // %entry
+; CHECK-NEXT: sub sp, sp, #48
+; CHECK-NEXT: str x30, [sp, #16] // 8-byte Folded Spill
+; CHECK-NEXT: stp x20, x19, [sp, #32] // 16-byte Folded Spill
+; CHECK-NEXT: .cfi_def_cfa_offset 48
+; CHECK-NEXT: .cfi_offset w19, -8
+; CHECK-NEXT: .cfi_offset w20, -16
+; CHECK-NEXT: .cfi_offset w30, -32
+; CHECK-NEXT: str q0, [sp] // 16-byte Folded Spill
+; CHECK-NEXT: mov d0, v0.d[1]
+; CHECK-NEXT: bl __fixdfti
+; CHECK-NEXT: ldr q0, [sp] // 16-byte Folded Reload
+; CHECK-NEXT: mov x19, x0
+; CHECK-NEXT: mov x20, x1
+; CHECK-NEXT: // kill: def $d0 killed $d0 killed $q0
+; CHECK-NEXT: bl __fixdfti
+; CHECK-NEXT: fmov d0, x0
+; CHECK-NEXT: mov x2, x19
+; CHECK-NEXT: mov x3, x20
+; CHECK-NEXT: ldp x20, x19, [sp, #32] // 16-byte Folded Reload
+; CHECK-NEXT: ldr x30, [sp, #16] // 8-byte Folded Reload
+; CHECK-NEXT: mov v0.d[1], x1
+; CHECK-NEXT: fmov x0, d0
+; CHECK-NEXT: add sp, sp, #48
+; CHECK-NEXT: ret
+entry:
+ %c = fptosi <2 x double> %a to <2 x i128>
+ ret <2 x i128> %c
+}
+
+define <2 x i128> @fptou_v2f64_v2i128(<2 x double> %a) {
+; CHECK-LABEL: fptou_v2f64_v2i128:
+; CHECK: // %bb.0: // %entry
+; CHECK-NEXT: sub sp, sp, #48
+; CHECK-NEXT: str x30, [sp, #16] // 8-byte Folded Spill
+; CHECK-NEXT: stp x20, x19, [sp, #32] // 16-byte Folded Spill
+; CHECK-NEXT: .cfi_def_cfa_offset 48
+; CHECK-NEXT: .cfi_offset w19, -8
+; CHECK-NEXT: .cfi_offset w20, -16
+; CHECK-NEXT: .cfi_offset w30, -32
+; CHECK-NEXT: str q0, [sp] // 16-byte Folded Spill
+; CHECK-NEXT: mov d0, v0.d[1]
+; CHECK-NEXT: bl __fixunsdfti
+; CHECK-NEXT: ldr q0, [sp] // 16-byte Folded Reload
+; CHECK-NEXT: mov x19, x0
+; CHECK-NEXT: mov x20, x1
+; CHECK-NEXT: // kill: def $d0 killed $d0 killed $q0
+; CHECK-NEXT: bl __fixunsdfti
+; CHECK-NEXT: fmov d0, x0
+; CHECK-NEXT: mov x2, x19
+; CHECK-NEXT: mov x3, x20
+; CHECK-NEXT: ldp x20, x19, [sp, #32] // 16-byte Folded Reload
+; CHECK-NEXT: ldr x30, [sp, #16] // 8-byte Folded Reload
+; CHECK-NEXT: mov v0.d[1], x1
+; CHECK-NEXT: fmov x0, d0
+; CHECK-NEXT: add sp, sp, #48
+; CHECK-NEXT: ret
+entry:
+ %c = fptoui <2 x double> %a to <2 x i128>
+ ret <2 x i128> %c
+}
+
+define <3 x i128> @fptos_v3f64_v3i128(<3 x double> %a) {
+; CHECK-LABEL: fptos_v3f64_v3i128:
+; CHECK: // %bb.0: // %entry
+; CHECK-NEXT: stp d9, d8, [sp, #-64]! // 16-byte Folded Spill
+; CHECK-NEXT: str x30, [sp, #16] // 8-byte Folded Spill
+; CHECK-NEXT: stp x22, x21, [sp, #32] // 16-byte Folded Spill
+; CHECK-NEXT: stp x20, x19, [sp, #48] // 16-byte Folded Spill
+; CHECK-NEXT: .cfi_def_cfa_offset 64
+; CHECK-NEXT: .cfi_offset w19, -8
+; CHECK-NEXT: .cfi_offset w20, -16
+; CHECK-NEXT: .cfi_offset w21, -24
+; CHECK-NEXT: .cfi_offset w22, -32
+; CHECK-NEXT: .cfi_offset w30, -48
+; CHECK-NEXT: .cfi_offset b8, -56
+; CHECK-NEXT: .cfi_offset b9, -64
+; CHECK-NEXT: fmov d9, d0
+; CHECK-NEXT: fmov d0, d1
+; CHECK-NEXT: fmov d8, d2
+; CHECK-NEXT: bl __fixdfti
+; CHECK-NEXT: fmov d0, d8
+; CHECK-NEXT: mov x19, x0
+; CHECK-NEXT: mov x20, x1
+; CHECK-NEXT: bl __fixdfti
+; CHECK-NEXT: fmov d0, d9
+; CHECK-NEXT: mov x21, x0
+; CHECK-NEXT: mov x22, x1
+; CHECK-NEXT: bl __fixdfti
+; CHECK-NEXT: fmov d0, x0
+; CHECK-NEXT: mov x2, x19
+; CHECK-NEXT: mov x3, x20
+; CHECK-NEXT: mov x4, x21
+; CHECK-NEXT: mov x5, x22
+; CHECK-NEXT: ldr x30, [sp, #16] // 8-byte Folded Reload
+; CHECK-NEXT: ldp x20, x19, [sp, #48] // 16-byte Folded Reload
+; CHECK-NEXT: mov v0.d[1], x1
+; CHECK-NEXT: ldp x22, x21, [sp, #32] // 16-byte Folded Reload
+; CHECK-NEXT: fmov x0, d0
+; CHECK-NEXT: ldp d9, d8, [sp], #64 // 16-byte Folded Reload
+; CHECK-NEXT: ret
+entry:
+ %c = fptosi <3 x double> %a to <3 x i128>
+ ret <3 x i128> %c
+}
+
+define <3 x i128> @fptou_v3f64_v3i128(<3 x double> %a) {
+; CHECK-LABEL: fptou_v3f64_v3i128:
+; CHECK: // %bb.0: // %entry
+; CHECK-NEXT: stp d9, d8, [sp, #-64]! // 16-byte Folded Spill
+; CHECK-NEXT: str x30, [sp, #16] // 8-byte Folded Spill
+; CHECK-NEXT: stp x22, x21, [sp, #32] // 16-byte Folded Spill
+; CHECK-NEXT: stp x20, x19, [sp, #48] // 16-byte Folded Spill
+; CHECK-NEXT: .cfi_def_cfa_offset 64
+; CHECK-NEXT: .cfi_offset w19, -8
+; CHECK-NEXT: .cfi_offset w20, -16
+; CHECK-NEXT: .cfi_offset w21, -24
+; CHECK-NEXT: .cfi_offset w22, -32
+; CHECK-NEXT: .cfi_offset w30, -48
+; CHECK-NEXT: .cfi_offset b8, -56
+; CHECK-NEXT: .cfi_offset b9, -64
+; CHECK-NEXT: fmov d9, d0
+; CHECK-NEXT: fmov d0, d1
+; CHECK-NEXT: fmov d8, d2
+; CHECK-NEXT: bl __fixunsdfti
+; CHECK-NEXT: fmov d0, d8
+; CHECK-NEXT: mov x19, x0
+; CHECK-NEXT: mov x20, x1
+; CHECK-NEXT: bl __fixunsdfti
+; CHECK-NEXT: fmov d0, d9
+; CHECK-NEXT: mov x21, x0
+; CHECK-NEXT: mov x22, x1
+; CHECK-NEXT: bl __fixunsdfti
+; CHECK-NEXT: fmov d0, x0
+; CHECK-NEXT: mov x2, x19
+; CHECK-NEXT: mov x3, x20
+; CHECK-NEXT: mov x4, x21
+; CHECK-NEXT: mov x5, x22
+; CHECK-NEXT: ldr x30, [sp, #16] // 8-byte Folded Reload
+; CHECK-NEXT: ldp x20, x19, [sp, #48] // 16-byte Folded Reload
+; CHECK-NEXT: mov v0.d[1], x1
+; CHECK-NEXT: ldp x22, x21, [sp, #32] // 16-byte Folded Reload
+; CHECK-NEXT: fmov x0, d0
+; CHECK-NEXT: ldp d9, d8, [sp], #64 // 16-byte Folded Reload
+; CHECK-NEXT: ret
+entry:
+ %c = fptoui <3 x double> %a to <3 x i128>
+ ret <3 x i128> %c
+}
+
define <2 x i64> @fptos_v2f32_v2i64(<2 x float> %a) {
; CHECK-LABEL: fptos_v2f32_v2i64:
; CHECK: // %bb.0: // %entry
@@ -3120,6 +3569,160 @@ entry:
ret <32 x i8> %c
}
+define <2 x i128> @fptos_v2f32_v2i128(<2 x float> %a) {
+; CHECK-LABEL: fptos_v2f32_v2i128:
+; CHECK: // %bb.0: // %entry
+; CHECK-NEXT: sub sp, sp, #48
+; CHECK-NEXT: str x30, [sp, #16] // 8-byte Folded Spill
+; CHECK-NEXT: stp x20, x19, [sp, #32] // 16-byte Folded Spill
+; CHECK-NEXT: .cfi_def_cfa_offset 48
+; CHECK-NEXT: .cfi_offset w19, -8
+; CHECK-NEXT: .cfi_offset w20, -16
+; CHECK-NEXT: .cfi_offset w30, -32
+; CHECK-NEXT: // kill: def $d0 killed $d0 def $q0
+; CHECK-NEXT: str q0, [sp] // 16-byte Folded Spill
+; CHECK-NEXT: mov s0, v0.s[1]
+; CHECK-NEXT: bl __fixsfti
+; CHECK-NEXT: ldr q0, [sp] // 16-byte Folded Reload
+; CHECK-NEXT: mov x19, x0
+; CHECK-NEXT: mov x20, x1
+; CHECK-NEXT: // kill: def $s0 killed $s0 killed $q0
+; CHECK-NEXT: bl __fixsfti
+; CHECK-NEXT: fmov d0, x0
+; CHECK-NEXT: mov x2, x19
+; CHECK-NEXT: mov x3, x20
+; CHECK-NEXT: ldp x20, x19, [sp, #32] // 16-byte Folded Reload
+; CHECK-NEXT: ldr x30, [sp, #16] // 8-byte Folded Reload
+; CHECK-NEXT: mov v0.d[1], x1
+; CHECK-NEXT: fmov x0, d0
+; CHECK-NEXT: add sp, sp, #48
+; CHECK-NEXT: ret
+entry:
+ %c = fptosi <2 x float> %a to <2 x i128>
+ ret <2 x i128> %c
+}
+
+define <2 x i128> @fptou_v2f32_v2i128(<2 x float> %a) {
+; CHECK-LABEL: fptou_v2f32_v2i128:
+; CHECK: // %bb.0: // %entry
+; CHECK-NEXT: sub sp, sp, #48
+; CHECK-NEXT: str x30, [sp, #16] // 8-byte Folded Spill
+; CHECK-NEXT: stp x20, x19, [sp, #32] // 16-byte Folded Spill
+; CHECK-NEXT: .cfi_def_cfa_offset 48
+; CHECK-NEXT: .cfi_offset w19, -8
+; CHECK-NEXT: .cfi_offset w20, -16
+; CHECK-NEXT: .cfi_offset w30, -32
+; CHECK-NEXT: // kill: def $d0 killed $d0 def $q0
+; CHECK-NEXT: str q0, [sp] // 16-byte Folded Spill
+; CHECK-NEXT: mov s0, v0.s[1]
+; CHECK-NEXT: bl __fixunssfti
+; CHECK-NEXT: ldr q0, [sp] // 16-byte Folded Reload
+; CHECK-NEXT: mov x19, x0
+; CHECK-NEXT: mov x20, x1
+; CHECK-NEXT: // kill: def $s0 killed $s0 killed $q0
+; CHECK-NEXT: bl __fixunssfti
+; CHECK-NEXT: fmov d0, x0
+; CHECK-NEXT: mov x2, x19
+; CHECK-NEXT: mov x3, x20
+; CHECK-NEXT: ldp x20, x19, [sp, #32] // 16-byte Folded Reload
+; CHECK-NEXT: ldr x30, [sp, #16] // 8-byte Folded Reload
+; CHECK-NEXT: mov v0.d[1], x1
+; CHECK-NEXT: fmov x0, d0
+; CHECK-NEXT: add sp, sp, #48
+; CHECK-NEXT: ret
+entry:
+ %c = fptoui <2 x float> %a to <2 x i128>
+ ret <2 x i128> %c
+}
+
+define <3 x i128> @fptos_v3f32_v3i128(<3 x float> %a) {
+; CHECK-LABEL: fptos_v3f32_v3i128:
+; CHECK: // %bb.0: // %entry
+; CHECK-NEXT: sub sp, sp, #64
+; CHECK-NEXT: str x30, [sp, #16] // 8-byte Folded Spill
+; CHECK-NEXT: stp x22, x21, [sp, #32] // 16-byte Folded Spill
+; CHECK-NEXT: stp x20, x19, [sp, #48] // 16-byte Folded Spill
+; CHECK-NEXT: .cfi_def_cfa_offset 64
+; CHECK-NEXT: .cfi_offset w19, -8
+; CHECK-NEXT: .cfi_offset w20, -16
+; CHECK-NEXT: .cfi_offset w21, -24
+; CHECK-NEXT: .cfi_offset w22, -32
+; CHECK-NEXT: .cfi_offset w30, -48
+; CHECK-NEXT: str q0, [sp] // 16-byte Folded Spill
+; CHECK-NEXT: ext v0.16b, v0.16b, v0.16b, #8
+; CHECK-NEXT: // kill: def $s0 killed $s0 killed $q0
+; CHECK-NEXT: bl __fixsfti
+; CHECK-NEXT: ldr q0, [sp] // 16-byte Folded Reload
+; CHECK-NEXT: mov x19, x0
+; CHECK-NEXT: mov x20, x1
+; CHECK-NEXT: mov s0, v0.s[1]
+; CHECK-NEXT: bl __fixsfti
+; CHECK-NEXT: ldr q0, [sp] // 16-byte Folded Reload
+; CHECK-NEXT: mov x21, x0
+; CHECK-NEXT: mov x22, x1
+; CHECK-NEXT: // kill: def $s0 killed $s0 killed $q0
+; CHECK-NEXT: bl __fixsfti
+; CHECK-NEXT: fmov d0, x0
+; CHECK-NEXT: mov x2, x21
+; CHECK-NEXT: mov x3, x22
+; CHECK-NEXT: mov x4, x19
+; CHECK-NEXT: mov x5, x20
+; CHECK-NEXT: ldr x30, [sp, #16] // 8-byte Folded Reload
+; CHECK-NEXT: ldp x20, x19, [sp, #48] // 16-byte Folded Reload
+; CHECK-NEXT: mov v0.d[1], x1
+; CHECK-NEXT: ldp x22, x21, [sp, #32] // 16-byte Folded Reload
+; CHECK-NEXT: fmov x0, d0
+; CHECK-NEXT: add sp, sp, #64
+; CHECK-NEXT: ret
+entry:
+ %c = fptosi <3 x float> %a to <3 x i128>
+ ret <3 x i128> %c
+}
+
+define <3 x i128> @fptou_v3f32_v3i128(<3 x float> %a) {
+; CHECK-LABEL: fptou_v3f32_v3i128:
+; CHECK: // %bb.0: // %entry
+; CHECK-NEXT: sub sp, sp, #64
+; CHECK-NEXT: str x30, [sp, #16] // 8-byte Folded Spill
+; CHECK-NEXT: stp x22, x21, [sp, #32] // 16-byte Folded Spill
+; CHECK-NEXT: stp x20, x19, [sp, #48] // 16-byte Folded Spill
+; CHECK-NEXT: .cfi_def_cfa_offset 64
+; CHECK-NEXT: .cfi_offset w19, -8
+; CHECK-NEXT: .cfi_offset w20, -16
+; CHECK-NEXT: .cfi_offset w21, -24
+; CHECK-NEXT: .cfi_offset w22, -32
+; CHECK-NEXT: .cfi_offset w30, -48
+; CHECK-NEXT: str q0, [sp] // 16-byte Folded Spill
+; CHECK-NEXT: ext v0.16b, v0.16b, v0.16b, #8
+; CHECK-NEXT: // kill: def $s0 killed $s0 killed $q0
+; CHECK-NEXT: bl __fixunssfti
+; CHECK-NEXT: ldr q0, [sp] // 16-byte Folded Reload
+; CHECK-NEXT: mov x19, x0
+; CHECK-NEXT: mov x20, x1
+; CHECK-NEXT: mov s0, v0.s[1]
+; CHECK-NEXT: bl __fixunssfti
+; CHECK-NEXT: ldr q0, [sp] // 16-byte Folded Reload
+; CHECK-NEXT: mov x21, x0
+; CHECK-NEXT: mov x22, x1
+; CHECK-NEXT: // kill: def $s0 killed $s0 killed $q0
+; CHECK-NEXT: bl __fixunssfti
+; CHECK-NEXT: fmov d0, x0
+; CHECK-NEXT: mov x2, x21
+; CHECK-NEXT: mov x3, x22
+; CHECK-NEXT: mov x4, x19
+; CHECK-NEXT: mov x5, x20
+; CHECK-NEXT: ldr x30, [sp, #16] // 8-byte Folded Reload
+; CHECK-NEXT: ldp x20, x19, [sp, #48] // 16-byte Folded Reload
+; CHECK-NEXT: mov v0.d[1], x1
+; CHECK-NEXT: ldp x22, x21, [sp, #32] // 16-byte Folded Reload
+; CHECK-NEXT: fmov x0, d0
+; CHECK-NEXT: add sp, sp, #64
+; CHECK-NEXT: ret
+entry:
+ %c = fptoui <3 x float> %a to <3 x i128>
+ ret <3 x i128> %c
+}
+
define <2 x i64> @fptos_v2f16_v2i64(<2 x half> %a) {
; CHECK-SD-NOFP16-LABEL: fptos_v2f16_v2i64:
; CHECK-SD-NOFP16: // %bb.0: // %entry
@@ -6177,3 +6780,745 @@ entry:
%c = fptoui <32 x half> %a to <32 x i8>
ret <32 x i8> %c
}
+
+define <2 x i128> @fptos_v2f16_v2i128(<2 x half> %a) {
+; CHECK-LABEL: fptos_v2f16_v2i128:
+; CHECK: // %bb.0: // %entry
+; CHECK-NEXT: sub sp, sp, #48
+; CHECK-NEXT: str x30, [sp, #16] // 8-byte Folded Spill
+; CHECK-NEXT: stp x20, x19, [sp, #32] // 16-byte Folded Spill
+; CHECK-NEXT: .cfi_def_cfa_offset 48
+; CHECK-NEXT: .cfi_offset w19, -8
+; CHECK-NEXT: .cfi_offset w20, -16
+; CHECK-NEXT: .cfi_offset w30, -32
+; CHECK-NEXT: // kill: def $d0 killed $d0 def $q0
+; CHECK-NEXT: str q0, [sp] // 16-byte Folded Spill
+; CHECK-NEXT: mov h0, v0.h[1]
+; CHECK-NEXT: bl __fixhfti
+; CHECK-NEXT: ldr q0, [sp] // 16-byte Folded Reload
+; CHECK-NEXT: mov x19, x0
+; CHECK-NEXT: mov x20, x1
+; CHECK-NEXT: // kill: def $h0 killed $h0 killed $q0
+; CHECK-NEXT: bl __fixhfti
+; CHECK-NEXT: fmov d0, x0
+; CHECK-NEXT: mov x2, x19
+; CHECK-NEXT: mov x3, x20
+; CHECK-NEXT: ldp x20, x19, [sp, #32] // 16-byte Folded Reload
+; CHECK-NEXT: ldr x30, [sp, #16] // 8-byte Folded Reload
+; CHECK-NEXT: mov v0.d[1], x1
+; CHECK-NEXT: fmov x0, d0
+; CHECK-NEXT: add sp, sp, #48
+; CHECK-NEXT: ret
+entry:
+ %c = fptosi <2 x half> %a to <2 x i128>
+ ret <2 x i128> %c
+}
+
+define <2 x i128> @fptou_v2f16_v2i128(<2 x half> %a) {
+; CHECK-LABEL: fptou_v2f16_v2i128:
+; CHECK: // %bb.0: // %entry
+; CHECK-NEXT: sub sp, sp, #48
+; CHECK-NEXT: str x30, [sp, #16] // 8-byte Folded Spill
+; CHECK-NEXT: stp x20, x19, [sp, #32] // 16-byte Folded Spill
+; CHECK-NEXT: .cfi_def_cfa_offset 48
+; CHECK-NEXT: .cfi_offset w19, -8
+; CHECK-NEXT: .cfi_offset w20, -16
+; CHECK-NEXT: .cfi_offset w30, -32
+; CHECK-NEXT: // kill: def $d0 killed $d0 def $q0
+; CHECK-NEXT: str q0, [sp] // 16-byte Folded Spill
+; CHECK-NEXT: mov h0, v0.h[1]
+; CHECK-NEXT: bl __fixunshfti
+; CHECK-NEXT: ldr q0, [sp] // 16-byte Folded Reload
+; CHECK-NEXT: mov x19, x0
+; CHECK-NEXT: mov x20, x1
+; CHECK-NEXT: // kill: def $h0 killed $h0 killed $q0
+; CHECK-NEXT: bl __fixunshfti
+; CHECK-NEXT: fmov d0, x0
+; CHECK-NEXT: mov x2, x19
+; CHECK-NEXT: mov x3, x20
+; CHECK-NEXT: ldp x20, x19, [sp, #32] // 16-byte Folded Reload
+; CHECK-NEXT: ldr x30, [sp, #16] // 8-byte Folded Reload
+; CHECK-NEXT: mov v0.d[1], x1
+; CHECK-NEXT: fmov x0, d0
+; CHECK-NEXT: add sp, sp, #48
+; CHECK-NEXT: ret
+entry:
+ %c = fptoui <2 x half> %a to <2 x i128>
+ ret <2 x i128> %c
+}
+
+define <3 x i128> @fptos_v3f16_v3i128(<3 x half> %a) {
+; CHECK-LABEL: fptos_v3f16_v3i128:
+; CHECK: // %bb.0: // %entry
+; CHECK-NEXT: sub sp, sp, #64
+; CHECK-NEXT: str x30, [sp, #16] // 8-byte Folded Spill
+; CHECK-NEXT: stp x22, x21, [sp, #32] // 16-byte Folded Spill
+; CHECK-NEXT: stp x20, x19, [sp, #48] // 16-byte Folded Spill
+; CHECK-NEXT: .cfi_def_cfa_offset 64
+; CHECK-NEXT: .cfi_offset w19, -8
+; CHECK-NEXT: .cfi_offset w20, -16
+; CHECK-NEXT: .cfi_offset w21, -24
+; CHECK-NEXT: .cfi_offset w22, -32
+; CHECK-NEXT: .cfi_offset w30, -48
+; CHECK-NEXT: // kill: def $d0 killed $d0 def $q0
+; CHECK-NEXT: str q0, [sp] // 16-byte Folded Spill
+; CHECK-NEXT: mov h0, v0.h[1]
+; CHECK-NEXT: bl __fixhfti
+; CHECK-NEXT: ldr q0, [sp] // 16-byte Folded Reload
+; CHECK-NEXT: mov x19, x0
+; CHECK-NEXT: mov x20, x1
+; CHECK-NEXT: mov h0, v0.h[2]
+; CHECK-NEXT: bl __fixhfti
+; CHECK-NEXT: ldr q0, [sp] // 16-byte Folded Reload
+; CHECK-NEXT: mov x21, x0
+; CHECK-NEXT: mov x22, x1
+; CHECK-NEXT: // kill: def $h0 killed $h0 killed $q0
+; CHECK-NEXT: bl __fixhfti
+; CHECK-NEXT: fmov d0, x0
+; CHECK-NEXT: mov x2, x19
+; CHECK-NEXT: mov x3, x20
+; CHECK-NEXT: mov x4, x21
+; CHECK-NEXT: mov x5, x22
+; CHECK-NEXT: ldr x30, [sp, #16] // 8-byte Folded Reload
+; CHECK-NEXT: ldp x20, x19, [sp, #48] // 16-byte Folded Reload
+; CHECK-NEXT: mov v0.d[1], x1
+; CHECK-NEXT: ldp x22, x21, [sp, #32] // 16-byte Folded Reload
+; CHECK-NEXT: fmov x0, d0
+; CHECK-NEXT: add sp, sp, #64
+; CHECK-NEXT: ret
+entry:
+ %c = fptosi <3 x half> %a to <3 x i128>
+ ret <3 x i128> %c
+}
+
+define <3 x i128> @fptou_v3f16_v3i128(<3 x half> %a) {
+; CHECK-LABEL: fptou_v3f16_v3i128:
+; CHECK: // %bb.0: // %entry
+; CHECK-NEXT: sub sp, sp, #64
+; CHECK-NEXT: str x30, [sp, #16] // 8-byte Folded Spill
+; CHECK-NEXT: stp x22, x21, [sp, #32] // 16-byte Folded Spill
+; CHECK-NEXT: stp x20, x19, [sp, #48] // 16-byte Folded Spill
+; CHECK-NEXT: .cfi_def_cfa_offset 64
+; CHECK-NEXT: .cfi_offset w19, -8
+; CHECK-NEXT: .cfi_offset w20, -16
+; CHECK-NEXT: .cfi_offset w21, -24
+; CHECK-NEXT: .cfi_offset w22, -32
+; CHECK-NEXT: .cfi_offset w30, -48
+; CHECK-NEXT: // kill: def $d0 killed $d0 def $q0
+; CHECK-NEXT: str q0, [sp] // 16-byte Folded Spill
+; CHECK-NEXT: mov h0, v0.h[1]
+; CHECK-NEXT: bl __fixunshfti
+; CHECK-NEXT: ldr q0, [sp] // 16-byte Folded Reload
+; CHECK-NEXT: mov x19, x0
+; CHECK-NEXT: mov x20, x1
+; CHECK-NEXT: mov h0, v0.h[2]
+; CHECK-NEXT: bl __fixunshfti
+; CHECK-NEXT: ldr q0, [sp] // 16-byte Folded Reload
+; CHECK-NEXT: mov x21, x0
+; CHECK-NEXT: mov x22, x1
+; CHECK-NEXT: // kill: def $h0 killed $h0 killed $q0
+; CHECK-NEXT: bl __fixunshfti
+; CHECK-NEXT: fmov d0, x0
+; CHECK-NEXT: mov x2, x19
+; CHECK-NEXT: mov x3, x20
+; CHECK-NEXT: mov x4, x21
+; CHECK-NEXT: mov x5, x22
+; CHECK-NEXT: ldr x30, [sp, #16] // 8-byte Folded Reload
+; CHECK-NEXT: ldp x20, x19, [sp, #48] // 16-byte Folded Reload
+; CHECK-NEXT: mov v0.d[1], x1
+; CHECK-NEXT: ldp x22, x21, [sp, #32] // 16-byte Folded Reload
+; CHECK-NEXT: fmov x0, d0
+; CHECK-NEXT: add sp, sp, #64
+; CHECK-NEXT: ret
+entry:
+ %c = fptoui <3 x half> %a to <3 x i128>
+ ret <3 x i128> %c
+}
+
+define <2 x i64> @fptos_v2f128_v2i64(<2 x fp128> %a) {
+; CHECK-LABEL: fptos_v2f128_v2i64:
+; CHECK: // %bb.0: // %entry
+; CHECK-NEXT: sub sp, sp, #48
+; CHECK-NEXT: str x30, [sp, #32] // 8-byte Folded Spill
+; CHECK-NEXT: .cfi_def_cfa_offset 48
+; CHECK-NEXT: .cfi_offset w30, -16
+; CHECK-NEXT: str q0, [sp] // 16-byte Folded Spill
+; CHECK-NEXT: mov v0.16b, v1.16b
+; CHECK-NEXT: bl __fixtfdi
+; CHECK-NEXT: fmov d0, x0
+; CHECK-NEXT: str q0, [sp, #16] // 16-byte Folded Spill
+; CHECK-NEXT: ldr q0, [sp] // 16-byte Folded Reload
+; CHECK-NEXT: bl __fixtfdi
+; CHECK-NEXT: fmov d0, x0
+; CHECK-NEXT: ldr q1, [sp, #16] // 16-byte Folded Reload
+; CHECK-NEXT: ldr x30, [sp, #32] // 8-byte Folded Reload
+; CHECK-NEXT: mov v0.d[1], v1.d[0]
+; CHECK-NEXT: add sp, sp, #48
+; CHECK-NEXT: ret
+entry:
+ %c = fptosi <2 x fp128> %a to <2 x i64>
+ ret <2 x i64> %c
+}
+
+define <2 x i64> @fptou_v2f128_v2i64(<2 x fp128> %a) {
+; CHECK-LABEL: fptou_v2f128_v2i64:
+; CHECK: // %bb.0: // %entry
+; CHECK-NEXT: sub sp, sp, #48
+; CHECK-NEXT: str x30, [sp, #32] // 8-byte Folded Spill
+; CHECK-NEXT: .cfi_def_cfa_offset 48
+; CHECK-NEXT: .cfi_offset w30, -16
+; CHECK-NEXT: str q0, [sp] // 16-byte Folded Spill
+; CHECK-NEXT: mov v0.16b, v1.16b
+; CHECK-NEXT: bl __fixunstfdi
+; CHECK-NEXT: fmov d0, x0
+; CHECK-NEXT: str q0, [sp, #16] // 16-byte Folded Spill
+; CHECK-NEXT: ldr q0, [sp] // 16-byte Folded Reload
+; CHECK-NEXT: bl __fixunstfdi
+; CHECK-NEXT: fmov d0, x0
+; CHECK-NEXT: ldr q1, [sp, #16] // 16-byte Folded Reload
+; CHECK-NEXT: ldr x30, [sp, #32] // 8-byte Folded Reload
+; CHECK-NEXT: mov v0.d[1], v1.d[0]
+; CHECK-NEXT: add sp, sp, #48
+; CHECK-NEXT: ret
+entry:
+ %c = fptoui <2 x fp128> %a to <2 x i64>
+ ret <2 x i64> %c
+}
+
+define <3 x i64> @fptos_v3f128_v3i64(<3 x fp128> %a) {
+; CHECK-LABEL: fptos_v3f128_v3i64:
+; CHECK: // %bb.0: // %entry
+; CHECK-NEXT: sub sp, sp, #64
+; CHECK-NEXT: str d8, [sp, #48] // 8-byte Folded Spill
+; CHECK-NEXT: str x30, [sp, #56] // 8-byte Folded Spill
+; CHECK-NEXT: .cfi_def_cfa_offset 64
+; CHECK-NEXT: .cfi_offset w30, -8
+; CHECK-NEXT: .cfi_offset b8, -16
+; CHECK-NEXT: stp q0, q1, [sp] // 32-byte Folded Spill
+; CHECK-NEXT: mov v0.16b, v2.16b
+; CHECK-NEXT: bl __fixtfdi
+; CHECK-NEXT: fmov d0, x0
+; CHECK-NEXT: str q0, [sp, #32] // 16-byte Folded Spill
+; CHECK-NEXT: ldr q0, [sp] // 16-byte Folded Reload
+; CHECK-NEXT: bl __fixtfdi
+; CHECK-NEXT: ldr q0, [sp, #16] // 16-byte Folded Reload
+; CHECK-NEXT: fmov d8, x0
+; CHECK-NEXT: bl __fixtfdi
+; CHECK-NEXT: fmov d0, d8
+; CHECK-NEXT: ldr q2, [sp, #32] // 16-byte Folded Reload
+; CHECK-NEXT: ldr x30, [sp, #56] // 8-byte Folded Reload
+; CHECK-NEXT: ldr d8, [sp, #48] // 8-byte Folded Reload
+; CHECK-NEXT: fmov d1, x0
+; CHECK-NEXT: // kill: def $d2 killed $d2 killed $q2
+; CHECK-NEXT: add sp, sp, #64
+; CHECK-NEXT: ret
+entry:
+ %c = fptosi <3 x fp128> %a to <3 x i64>
+ ret <3 x i64> %c
+}
+
+define <3 x i64> @fptou_v3f128_v3i64(<3 x fp128> %a) {
+; CHECK-LABEL: fptou_v3f128_v3i64:
+; CHECK: // %bb.0: // %entry
+; CHECK-NEXT: sub sp, sp, #64
+; CHECK-NEXT: str d8, [sp, #48] // 8-byte Folded Spill
+; CHECK-NEXT: str x30, [sp, #56] // 8-byte Folded Spill
+; CHECK-NEXT: .cfi_def_cfa_offset 64
+; CHECK-NEXT: .cfi_offset w30, -8
+; CHECK-NEXT: .cfi_offset b8, -16
+; CHECK-NEXT: stp q0, q1, [sp] // 32-byte Folded Spill
+; CHECK-NEXT: mov v0.16b, v2.16b
+; CHECK-NEXT: bl __fixunstfdi
+; CHECK-NEXT: fmov d0, x0
+; CHECK-NEXT: str q0, [sp, #32] // 16-byte Folded Spill
+; CHECK-NEXT: ldr q0, [sp] // 16-byte Folded Reload
+; CHECK-NEXT: bl __fixunstfdi
+; CHECK-NEXT: ldr q0, [sp, #16] // 16-byte Folded Reload
+; CHECK-NEXT: fmov d8, x0
+; CHECK-NEXT: bl __fixunstfdi
+; CHECK-NEXT: fmov d0, d8
+; CHECK-NEXT: ldr q2, [sp, #32] // 16-byte Folded Reload
+; CHECK-NEXT: ldr x30, [sp, #56] // 8-byte Folded Reload
+; CHECK-NEXT: ldr d8, [sp, #48] // 8-byte Folded Reload
+; CHECK-NEXT: fmov d1, x0
+; CHECK-NEXT: // kill: def $d2 killed $d2 killed $q2
+; CHECK-NEXT: add sp, sp, #64
+; CHECK-NEXT: ret
+entry:
+ %c = fptoui <3 x fp128> %a to <3 x i64>
+ ret <3 x i64> %c
+}
+
+define <2 x i32> @fptos_v2f128_v2i32(<2 x fp128> %a) {
+; CHECK-LABEL: fptos_v2f128_v2i32:
+; CHECK: // %bb.0: // %entry
+; CHECK-NEXT: sub sp, sp, #48
+; CHECK-NEXT: str x30, [sp, #32] // 8-byte Folded Spill
+; CHECK-NEXT: .cfi_def_cfa_offset 48
+; CHECK-NEXT: .cfi_offset w30, -16
+; CHECK-NEXT: str q1, [sp, #16] // 16-byte Folded Spill
+; CHECK-NEXT: bl __fixtfsi
+; CHECK-NEXT: fmov s0, w0
+; CHECK-NEXT: str q0, [sp] // 16-byte Folded Spill
+; CHECK-NEXT: ldr q0, [sp, #16] // 16-byte Folded Reload
+; CHECK-NEXT: bl __fixtfsi
+; CHECK-NEXT: ldr q0, [sp] // 16-byte Folded Reload
+; CHECK-NEXT: ldr x30, [sp, #32] // 8-byte Folded Reload
+; CHECK-NEXT: mov v0.s[1], w0
+; CHECK-NEXT: // kill: def $d0 killed $d0 killed $q0
+; CHECK-NEXT: add sp, sp, #48
+; CHECK-NEXT: ret
+entry:
+ %c = fptosi <2 x fp128> %a to <2 x i32>
+ ret <2 x i32> %c
+}
+
+define <2 x i32> @fptou_v2f128_v2i32(<2 x fp128> %a) {
+; CHECK-LABEL: fptou_v2f128_v2i32:
+; CHECK: // %bb.0: // %entry
+; CHECK-NEXT: sub sp, sp, #48
+; CHECK-NEXT: str x30, [sp, #32] // 8-byte Folded Spill
+; CHECK-NEXT: .cfi_def_cfa_offset 48
+; CHECK-NEXT: .cfi_offset w30, -16
+; CHECK-NEXT: str q1, [sp, #16] // 16-byte Folded Spill
+; CHECK-NEXT: bl __fixunstfsi
+; CHECK-NEXT: fmov s0, w0
+; CHECK-NEXT: str q0, [sp] // 16-byte Folded Spill
+; CHECK-NEXT: ldr q0, [sp, #16] // 16-byte Folded Reload
+; CHECK-NEXT: bl __fixunstfsi
+; CHECK-NEXT: ldr q0, [sp] // 16-byte Folded Reload
+; CHECK-NEXT: ldr x30, [sp, #32] // 8-byte Folded Reload
+; CHECK-NEXT: mov v0.s[1], w0
+; CHECK-NEXT: // kill: def $d0 killed $d0 killed $q0
+; CHECK-NEXT: add sp, sp, #48
+; CHECK-NEXT: ret
+entry:
+ %c = fptoui <2 x fp128> %a to <2 x i32>
+ ret <2 x i32> %c
+}
+
+define <3 x i32> @fptos_v3f128_v3i32(<3 x fp128> %a) {
+; CHECK-LABEL: fptos_v3f128_v3i32:
+; CHECK: // %bb.0: // %entry
+; CHECK-NEXT: sub sp, sp, #64
+; CHECK-NEXT: str x30, [sp, #48] // 8-byte Folded Spill
+; CHECK-NEXT: .cfi_def_cfa_offset 64
+; CHECK-NEXT: .cfi_offset w30, -16
+; CHECK-NEXT: stp q1, q2, [sp] // 32-byte Folded Spill
+; CHECK-NEXT: bl __fixtfsi
+; CHECK-NEXT: fmov s0, w0
+; CHECK-NEXT: str q0, [sp, #32] // 16-byte Folded Spill
+; CHECK-NEXT: ldr q0, [sp] // 16-byte Folded Reload
+; CHECK-NEXT: bl __fixtfsi
+; CHECK-NEXT: ldr q0, [sp, #32] // 16-byte Folded Reload
+; CHECK-NEXT: mov v0.s[1], w0
+; CHECK-NEXT: str q0, [sp, #32] // 16-byte Folded Spill
+; CHECK-NEXT: ldr q0, [sp, #16] // 16-byte Folded Reload
+; CHECK-NEXT: bl __fixtfsi
+; CHECK-NEXT: ldr q0, [sp, #32] // 16-byte Folded Reload
+; CHECK-NEXT: ldr x30, [sp, #48] // 8-byte Folded Reload
+; CHECK-NEXT: mov v0.s[2], w0
+; CHECK-NEXT: add sp, sp, #64
+; CHECK-NEXT: ret
+entry:
+ %c = fptosi <3 x fp128> %a to <3 x i32>
+ ret <3 x i32> %c
+}
+
+define <3 x i32> @fptou_v3f128_v3i32(<3 x fp128> %a) {
+; CHECK-LABEL: fptou_v3f128_v3i32:
+; CHECK: // %bb.0: // %entry
+; CHECK-NEXT: sub sp, sp, #64
+; CHECK-NEXT: str x30, [sp, #48] // 8-byte Folded Spill
+; CHECK-NEXT: .cfi_def_cfa_offset 64
+; CHECK-NEXT: .cfi_offset w30, -16
+; CHECK-NEXT: stp q1, q2, [sp] // 32-byte Folded Spill
+; CHECK-NEXT: bl __fixunstfsi
+; CHECK-NEXT: fmov s0, w0
+; CHECK-NEXT: str q0, [sp, #32] // 16-byte Folded Spill
+; CHECK-NEXT: ldr q0, [sp] // 16-byte Folded Reload
+; CHECK-NEXT: bl __fixunstfsi
+; CHECK-NEXT: ldr q0, [sp, #32] // 16-byte Folded Reload
+; CHECK-NEXT: mov v0.s[1], w0
+; CHECK-NEXT: str q0, [sp, #32] // 16-byte Folded Spill
+; CHECK-NEXT: ldr q0, [sp, #16] // 16-byte Folded Reload
+; CHECK-NEXT: bl __fixunstfsi
+; CHECK-NEXT: ldr q0, [sp, #32] // 16-byte Folded Reload
+; CHECK-NEXT: ldr x30, [sp, #48] // 8-byte Folded Reload
+; CHECK-NEXT: mov v0.s[2], w0
+; CHECK-NEXT: add sp, sp, #64
+; CHECK-NEXT: ret
+entry:
+ %c = fptoui <3 x fp128> %a to <3 x i32>
+ ret <3 x i32> %c
+}
+
+define <2 x i16> @fptos_v2f128_v2i16(<2 x fp128> %a) {
+; CHECK-LABEL: fptos_v2f128_v2i16:
+; CHECK: // %bb.0: // %entry
+; CHECK-NEXT: sub sp, sp, #48
+; CHECK-NEXT: str x30, [sp, #32] // 8-byte Folded Spill
+; CHECK-NEXT: .cfi_def_cfa_offset 48
+; CHECK-NEXT: .cfi_offset w30, -16
+; CHECK-NEXT: str q1, [sp, #16] // 16-byte Folded Spill
+; CHECK-NEXT: bl __fixtfsi
+; CHECK-NEXT: fmov s0, w0
+; CHECK-NEXT: str q0, [sp] // 16-byte Folded Spill
+; CHECK-NEXT: ldr q0, [sp, #16] // 16-byte Folded Reload
+; CHECK-NEXT: bl __fixtfsi
+; CHECK-NEXT: ldr q0, [sp] // 16-byte Folded Reload
+; CHECK-NEXT: ldr x30, [sp, #32] // 8-byte Folded Reload
+; CHECK-NEXT: mov v0.s[1], w0
+; CHECK-NEXT: // kill: def $d0 killed $d0 killed $q0
+; CHECK-NEXT: add sp, sp, #48
+; CHECK-NEXT: ret
+entry:
+ %c = fptosi <2 x fp128> %a to <2 x i16>
+ ret <2 x i16> %c
+}
+
+define <2 x i16> @fptou_v2f128_v2i16(<2 x fp128> %a) {
+; CHECK-LABEL: fptou_v2f128_v2i16:
+; CHECK: // %bb.0: // %entry
+; CHECK-NEXT: sub sp, sp, #48
+; CHECK-NEXT: str x30, [sp, #32] // 8-byte Folded Spill
+; CHECK-NEXT: .cfi_def_cfa_offset 48
+; CHECK-NEXT: .cfi_offset w30, -16
+; CHECK-NEXT: str q1, [sp, #16] // 16-byte Folded Spill
+; CHECK-NEXT: bl __fixtfsi
+; CHECK-NEXT: fmov s0, w0
+; CHECK-NEXT: str q0, [sp] // 16-byte Folded Spill
+; CHECK-NEXT: ldr q0, [sp, #16] // 16-byte Folded Reload
+; CHECK-NEXT: bl __fixtfsi
+; CHECK-NEXT: ldr q0, [sp] // 16-byte Folded Reload
+; CHECK-NEXT: ldr x30, [sp, #32] // 8-byte Folded Reload
+; CHECK-NEXT: mov v0.s[1], w0
+; CHECK-NEXT: // kill: def $d0 killed $d0 killed $q0
+; CHECK-NEXT: add sp, sp, #48
+; CHECK-NEXT: ret
+entry:
+ %c = fptoui <2 x fp128> %a to <2 x i16>
+ ret <2 x i16> %c
+}
+
+define <3 x i16> @fptos_v3f128_v3i16(<3 x fp128> %a) {
+; CHECK-LABEL: fptos_v3f128_v3i16:
+; CHECK: // %bb.0: // %entry
+; CHECK-NEXT: sub sp, sp, #48
+; CHECK-NEXT: str d8, [sp, #32] // 8-byte Folded Spill
+; CHECK-NEXT: str x30, [sp, #40] // 8-byte Folded Spill
+; CHECK-NEXT: .cfi_def_cfa_offset 48
+; CHECK-NEXT: .cfi_offset w30, -8
+; CHECK-NEXT: .cfi_offset b8, -16
+; CHECK-NEXT: stp q0, q1, [sp] // 32-byte Folded Spill
+; CHECK-NEXT: mov v0.16b, v2.16b
+; CHECK-NEXT: bl __fixtfsi
+; CHECK-NEXT: ldr q0, [sp] // 16-byte Folded Reload
+; CHECK-NEXT: fmov s8, w0
+; CHECK-NEXT: bl __fixtfsi
+; CHECK-NEXT: fmov s0, w0
+; CHECK-NEXT: str q0, [sp] // 16-byte Folded Spill
+; CHECK-NEXT: ldr q0, [sp, #16] // 16-byte Folded Reload
+; CHECK-NEXT: bl __fixtfsi
+; CHECK-NEXT: ldr q0, [sp] // 16-byte Folded Reload
+; CHECK-NEXT: ldr x30, [sp, #40] // 8-byte Folded Reload
+; CHECK-NEXT: mov v0.s[1], w0
+; CHECK-NEXT: uzp1 v0.4h, v0.4h, v8.4h
+; CHECK-NEXT: ldr d8, [sp, #32] // 8-byte Folded Reload
+; CHECK-NEXT: add sp, sp, #48
+; CHECK-NEXT: ret
+entry:
+ %c = fptosi <3 x fp128> %a to <3 x i16>
+ ret <3 x i16> %c
+}
+
+define <3 x i16> @fptou_v3f128_v3i16(<3 x fp128> %a) {
+; CHECK-LABEL: fptou_v3f128_v3i16:
+; CHECK: // %bb.0: // %entry
+; CHECK-NEXT: sub sp, sp, #48
+; CHECK-NEXT: str d8, [sp, #32] // 8-byte Folded Spill
+; CHECK-NEXT: str x30, [sp, #40] // 8-byte Folded Spill
+; CHECK-NEXT: .cfi_def_cfa_offset 48
+; CHECK-NEXT: .cfi_offset w30, -8
+; CHECK-NEXT: .cfi_offset b8, -16
+; CHECK-NEXT: stp q0, q1, [sp] // 32-byte Folded Spill
+; CHECK-NEXT: mov v0.16b, v2.16b
+; CHECK-NEXT: bl __fixtfsi
+; CHECK-NEXT: ldr q0, [sp] // 16-byte Folded Reload
+; CHECK-NEXT: fmov s8, w0
+; CHECK-NEXT: bl __fixtfsi
+; CHECK-NEXT: fmov s0, w0
+; CHECK-NEXT: str q0, [sp] // 16-byte Folded Spill
+; CHECK-NEXT: ldr q0, [sp, #16] // 16-byte Folded Reload
+; CHECK-NEXT: bl __fixtfsi
+; CHECK-NEXT: ldr q0, [sp] // 16-byte Folded Reload
+; CHECK-NEXT: ldr x30, [sp, #40] // 8-byte Folded Reload
+; CHECK-NEXT: mov v0.s[1], w0
+; CHECK-NEXT: uzp1 v0.4h, v0.4h, v8.4h
+; CHECK-NEXT: ldr d8, [sp, #32] // 8-byte Folded Reload
+; CHECK-NEXT: add sp, sp, #48
+; CHECK-NEXT: ret
+entry:
+ %c = fptoui <3 x fp128> %a to <3 x i16>
+ ret <3 x i16> %c
+}
+
+define <2 x i8> @fptos_v2f128_v2i8(<2 x fp128> %a) {
+; CHECK-LABEL: fptos_v2f128_v2i8:
+; CHECK: // %bb.0: // %entry
+; CHECK-NEXT: sub sp, sp, #48
+; CHECK-NEXT: str x30, [sp, #32] // 8-byte Folded Spill
+; CHECK-NEXT: .cfi_def_cfa_offset 48
+; CHECK-NEXT: .cfi_offset w30, -16
+; CHECK-NEXT: str q1, [sp, #16] // 16-byte Folded Spill
+; CHECK-NEXT: bl __fixtfsi
+; CHECK-NEXT: fmov s0, w0
+; CHECK-NEXT: str q0, [sp] // 16-byte Folded Spill
+; CHECK-NEXT: ldr q0, [sp, #16] // 16-byte Folded Reload
+; CHECK-NEXT: bl __fixtfsi
+; CHECK-NEXT: ldr q0, [sp] // 16-byte Folded Reload
+; CHECK-NEXT: ldr x30, [sp, #32] // 8-byte Folded Reload
+; CHECK-NEXT: mov v0.s[1], w0
+; CHECK-NEXT: // kill: def $d0 killed $d0 killed $q0
+; CHECK-NEXT: add sp, sp, #48
+; CHECK-NEXT: ret
+entry:
+ %c = fptosi <2 x fp128> %a to <2 x i8>
+ ret <2 x i8> %c
+}
+
+define <2 x i8> @fptou_v2f128_v2i8(<2 x fp128> %a) {
+; CHECK-LABEL: fptou_v2f128_v2i8:
+; CHECK: // %bb.0: // %entry
+; CHECK-NEXT: sub sp, sp, #48
+; CHECK-NEXT: str x30, [sp, #32] // 8-byte Folded Spill
+; CHECK-NEXT: .cfi_def_cfa_offset 48
+; CHECK-NEXT: .cfi_offset w30, -16
+; CHECK-NEXT: str q1, [sp, #16] // 16-byte Folded Spill
+; CHECK-NEXT: bl __fixtfsi
+; CHECK-NEXT: fmov s0, w0
+; CHECK-NEXT: str q0, [sp] // 16-byte Folded Spill
+; CHECK-NEXT: ldr q0, [sp, #16] // 16-byte Folded Reload
+; CHECK-NEXT: bl __fixtfsi
+; CHECK-NEXT: ldr q0, [sp] // 16-byte Folded Reload
+; CHECK-NEXT: ldr x30, [sp, #32] // 8-byte Folded Reload
+; CHECK-NEXT: mov v0.s[1], w0
+; CHECK-NEXT: // kill: def $d0 killed $d0 killed $q0
+; CHECK-NEXT: add sp, sp, #48
+; CHECK-NEXT: ret
+entry:
+ %c = fptoui <2 x fp128> %a to <2 x i8>
+ ret <2 x i8> %c
+}
+
+define <3 x i8> @fptos_v3f128_v3i8(<3 x fp128> %a) {
+; CHECK-LABEL: fptos_v3f128_v3i8:
+; CHECK: // %bb.0: // %entry
+; CHECK-NEXT: sub sp, sp, #48
+; CHECK-NEXT: str d8, [sp, #32] // 8-byte Folded Spill
+; CHECK-NEXT: str x30, [sp, #40] // 8-byte Folded Spill
+; CHECK-NEXT: .cfi_def_cfa_offset 48
+; CHECK-NEXT: .cfi_offset w30, -8
+; CHECK-NEXT: .cfi_offset b8, -16
+; CHECK-NEXT: stp q0, q1, [sp] // 32-byte Folded Spill
+; CHECK-NEXT: mov v0.16b, v2.16b
+; CHECK-NEXT: bl __fixtfsi
+; CHECK-NEXT: ldr q0, [sp] // 16-byte Folded Reload
+; CHECK-NEXT: fmov s8, w0
+; CHECK-NEXT: bl __fixtfsi
+; CHECK-NEXT: fmov s0, w0
+; CHECK-NEXT: str q0, [sp] // 16-byte Folded Spill
+; CHECK-NEXT: ldr q0, [sp, #16] // 16-byte Folded Reload
+; CHECK-NEXT: bl __fixtfsi
+; CHECK-NEXT: ldr q0, [sp] // 16-byte Folded Reload
+; CHECK-NEXT: ldr x30, [sp, #40] // 8-byte Folded Reload
+; CHECK-NEXT: mov v0.s[1], w0
+; CHECK-NEXT: uzp1 v0.4h, v0.4h, v8.4h
+; CHECK-NEXT: ldr d8, [sp, #32] // 8-byte Folded Reload
+; CHECK-NEXT: umov w0, v0.h[0]
+; CHECK-NEXT: umov w1, v0.h[1]
+; CHECK-NEXT: umov w2, v0.h[2]
+; CHECK-NEXT: add sp, sp, #48
+; CHECK-NEXT: ret
+entry:
+ %c = fptosi <3 x fp128> %a to <3 x i8>
+ ret <3 x i8> %c
+}
+
+define <3 x i8> @fptou_v3f128_v3i8(<3 x fp128> %a) {
+; CHECK-LABEL: fptou_v3f128_v3i8:
+; CHECK: // %bb.0: // %entry
+; CHECK-NEXT: sub sp, sp, #48
+; CHECK-NEXT: str d8, [sp, #32] // 8-byte Folded Spill
+; CHECK-NEXT: str x30, [sp, #40] // 8-byte Folded Spill
+; CHECK-NEXT: .cfi_def_cfa_offset 48
+; CHECK-NEXT: .cfi_offset w30, -8
+; CHECK-NEXT: .cfi_offset b8, -16
+; CHECK-NEXT: stp q0, q1, [sp] // 32-byte Folded Spill
+; CHECK-NEXT: mov v0.16b, v2.16b
+; CHECK-NEXT: bl __fixtfsi
+; CHECK-NEXT: ldr q0, [sp] // 16-byte Folded Reload
+; CHECK-NEXT: fmov s8, w0
+; CHECK-NEXT: bl __fixtfsi
+; CHECK-NEXT: fmov s0, w0
+; CHECK-NEXT: str q0, [sp] // 16-byte Folded Spill
+; CHECK-NEXT: ldr q0, [sp, #16] // 16-byte Folded Reload
+; CHECK-NEXT: bl __fixtfsi
+; CHECK-NEXT: ldr q0, [sp] // 16-byte Folded Reload
+; CHECK-NEXT: ldr x30, [sp, #40] // 8-byte Folded Reload
+; CHECK-NEXT: mov v0.s[1], w0
+; CHECK-NEXT: uzp1 v0.4h, v0.4h, v8.4h
+; CHECK-NEXT: ldr d8, [sp, #32] // 8-byte Folded Reload
+; CHECK-NEXT: umov w0, v0.h[0]
+; CHECK-NEXT: umov w1, v0.h[1]
+; CHECK-NEXT: umov w2, v0.h[2]
+; CHECK-NEXT: add sp, sp, #48
+; CHECK-NEXT: ret
+entry:
+ %c = fptoui <3 x fp128> %a to <3 x i8>
+ ret <3 x i8> %c
+}
+
+define <2 x i128> @fptos_v2f128_v2i128(<2 x fp128> %a) {
+; CHECK-LABEL: fptos_v2f128_v2i128:
+; CHECK: // %bb.0: // %entry
+; CHECK-NEXT: sub sp, sp, #48
+; CHECK-NEXT: str x30, [sp, #16] // 8-byte Folded Spill
+; CHECK-NEXT: stp x20, x19, [sp, #32] // 16-byte Folded Spill
+; CHECK-NEXT: .cfi_def_cfa_offset 48
+; CHECK-NEXT: .cfi_offset w19, -8
+; CHECK-NEXT: .cfi_offset w20, -16
+; CHECK-NEXT: .cfi_offset w30, -32
+; CHECK-NEXT: str q0, [sp] // 16-byte Folded Spill
+; CHECK-NEXT: mov v0.16b, v1.16b
+; CHECK-NEXT: bl __fixtfti
+; CHECK-NEXT: ldr q0, [sp] // 16-byte Folded Reload
+; CHECK-NEXT: mov x19, x0
+; CHECK-NEXT: mov x20, x1
+; CHECK-NEXT: bl __fixtfti
+; CHECK-NEXT: fmov d0, x0
+; CHECK-NEXT: mov x2, x19
+; CHECK-NEXT: mov x3, x20
+; CHECK-NEXT: ldp x20, x19, [sp, #32] // 16-byte Folded Reload
+; CHECK-NEXT: ldr x30, [sp, #16] // 8-byte Folded Reload
+; CHECK-NEXT: mov v0.d[1], x1
+; CHECK-NEXT: fmov x0, d0
+; CHECK-NEXT: add sp, sp, #48
+; CHECK-NEXT: ret
+entry:
+ %c = fptosi <2 x fp128> %a to <2 x i128>
+ ret <2 x i128> %c
+}
+
+define <2 x i128> @fptou_v2f128_v2i128(<2 x fp128> %a) {
+; CHECK-LABEL: fptou_v2f128_v2i128:
+; CHECK: // %bb.0: // %entry
+; CHECK-NEXT: sub sp, sp, #48
+; CHECK-NEXT: str x30, [sp, #16] // 8-byte Folded Spill
+; CHECK-NEXT: stp x20, x19, [sp, #32] // 16-byte Folded Spill
+; CHECK-NEXT: .cfi_def_cfa_offset 48
+; CHECK-NEXT: .cfi_offset w19, -8
+; CHECK-NEXT: .cfi_offset w20, -16
+; CHECK-NEXT: .cfi_offset w30, -32
+; CHECK-NEXT: str q0, [sp] // 16-byte Folded Spill
+; CHECK-NEXT: mov v0.16b, v1.16b
+; CHECK-NEXT: bl __fixunstfti
+; CHECK-NEXT: ldr q0, [sp] // 16-byte Folded Reload
+; CHECK-NEXT: mov x19, x0
+; CHECK-NEXT: mov x20, x1
+; CHECK-NEXT: bl __fixunstfti
+; CHECK-NEXT: fmov d0, x0
+; CHECK-NEXT: mov x2, x19
+; CHECK-NEXT: mov x3, x20
+; CHECK-NEXT: ldp x20, x19, [sp, #32] // 16-byte Folded Reload
+; CHECK-NEXT: ldr x30, [sp, #16] // 8-byte Folded Reload
+; CHECK-NEXT: mov v0.d[1], x1
+; CHECK-NEXT: fmov x0, d0
+; CHECK-NEXT: add sp, sp, #48
+; CHECK-NEXT: ret
+entry:
+ %c = fptoui <2 x fp128> %a to <2 x i128>
+ ret <2 x i128> %c
+}
+
+define <3 x i128> @fptos_v3f128_v3i128(<3 x fp128> %a) {
+; CHECK-LABEL: fptos_v3f128_v3i128:
+; CHECK: // %bb.0: // %entry
+; CHECK-NEXT: sub sp, sp, #80
+; CHECK-NEXT: str x30, [sp, #32] // 8-byte Folded Spill
+; CHECK-NEXT: stp x22, x21, [sp, #48] // 16-byte Folded Spill
+; CHECK-NEXT: stp x20, x19, [sp, #64] // 16-byte Folded Spill
+; CHECK-NEXT: .cfi_def_cfa_offset 80
+; CHECK-NEXT: .cfi_offset w19, -8
+; CHECK-NEXT: .cfi_offset w20, -16
+; CHECK-NEXT: .cfi_offset w21, -24
+; CHECK-NEXT: .cfi_offset w22, -32
+; CHECK-NEXT: .cfi_offset w30, -48
+; CHECK-NEXT: stp q2, q0, [sp] // 32-byte Folded Spill
+; CHECK-NEXT: mov v0.16b, v1.16b
+; CHECK-NEXT: bl __fixtfti
+; CHECK-NEXT: ldr q0, [sp] // 16-byte Folded Reload
+; CHECK-NEXT: mov x19, x0
+; CHECK-NEXT: mov x20, x1
+; CHECK-NEXT: bl __fixtfti
+; CHECK-NEXT: ldr q0, [sp, #16] // 16-byte Folded Reload
+; CHECK-NEXT: mov x21, x0
+; CHECK-NEXT: mov x22, x1
+; CHECK-NEXT: bl __fixtfti
+; CHECK-NEXT: fmov d0, x0
+; CHECK-NEXT: mov x2, x19
+; CHECK-NEXT: mov x3, x20
+; CHECK-NEXT: mov x4, x21
+; CHECK-NEXT: mov x5, x22
+; CHECK-NEXT: ldr x30, [sp, #32] // 8-byte Folded Reload
+; CHECK-NEXT: ldp x20, x19, [sp, #64] // 16-byte Folded Reload
+; CHECK-NEXT: mov v0.d[1], x1
+; CHECK-NEXT: ldp x22, x21, [sp, #48] // 16-byte Folded Reload
+; CHECK-NEXT: fmov x0, d0
+; CHECK-NEXT: add sp, sp, #80
+; CHECK-NEXT: ret
+entry:
+ %c = fptosi <3 x fp128> %a to <3 x i128>
+ ret <3 x i128> %c
+}
+
+define <3 x i128> @fptou_v3f128_v3i128(<3 x fp128> %a) {
+; CHECK-LABEL: fptou_v3f128_v3i128:
+; CHECK: // %bb.0: // %entry
+; CHECK-NEXT: sub sp, sp, #80
+; CHECK-NEXT: str x30, [sp, #32] // 8-byte Folded Spill
+; CHECK-NEXT: stp x22, x21, [sp, #48] // 16-byte Folded Spill
+; CHECK-NEXT: stp x20, x19, [sp, #64] // 16-byte Folded Spill
+; CHECK-NEXT: .cfi_def_cfa_offset 80
+; CHECK-NEXT: .cfi_offset w19, -8
+; CHECK-NEXT: .cfi_offset w20, -16
+; CHECK-NEXT: .cfi_offset w21, -24
+; CHECK-NEXT: .cfi_offset w22, -32
+; CHECK-NEXT: .cfi_offset w30, -48
+; CHECK-NEXT: stp q2, q0, [sp] // 32-byte Folded Spill
+; CHECK-NEXT: mov v0.16b, v1.16b
+; CHECK-NEXT: bl __fixunstfti
+; CHECK-NEXT: ldr q0, [sp] // 16-byte Folded Reload
+; CHECK-NEXT: mov x19, x0
+; CHECK-NEXT: mov x20, x1
+; CHECK-NEXT: bl __fixunstfti
+; CHECK-NEXT: ldr q0, [sp, #16] // 16-byte Folded Reload
+; CHECK-NEXT: mov x21, x0
+; CHECK-NEXT: mov x22, x1
+; CHECK-NEXT: bl __fixunstfti
+; CHECK-NEXT: fmov d0, x0
+; CHECK-NEXT: mov x2, x19
+; CHECK-NEXT: mov x3, x20
+; CHECK-NEXT: mov x4, x21
+; CHECK-NEXT: mov x5, x22
+; CHECK-NEXT: ldr x30, [sp, #32] // 8-byte Folded Reload
+; CHECK-NEXT: ldp x20, x19, [sp, #64] // 16-byte Folded Reload
+; CHECK-NEXT: mov v0.d[1], x1
+; CHECK-NEXT: ldp x22, x21, [sp, #48] // 16-byte Folded Reload
+; CHECK-NEXT: fmov x0, d0
+; CHECK-NEXT: add sp, sp, #80
+; CHECK-NEXT: ret
+entry:
+ %c = fptoui <3 x fp128> %a to <3 x i128>
+ ret <3 x i128> %c
+}
More information about the llvm-commits
mailing list