caio/clang-p2996
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity
Files
6b5bde697bb6f2ec8564c9b35bbf9f2d5b703f13
clang-p2996/llvm/test/CodeGen/AArch64/fptrunc.ll
David Green 5a81a559d6 [GISel] Explicitly disable BF16 tablegen patterns. (#124113) 
We currently have an issue where bf16 patters can be used to match fp16
types, as GISel does not know about the difference between the two. This
patch explicitly disables them to make sure that they are never used.

The opposite can also happen too, where fp16 patterns are used for
operators that should be bf16. So this also changes any operations with
bf16 types to now cause a fallback to SDAG.

The pass setup for GISel has been slightly adjusted to make sure that a
verify pass does not get added between AMD-SDAG and SIFixSGPRCopiesPass,
which otherwise can cause verifier issues when falling back.
2025-01-27 22:21:12 +00:00

406 lines
14 KiB
LLVM
Raw Blame History

; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py UTC_ARGS: --version 3
; RUN: llc -mtriple=aarch64 -global-isel=0 -verify-machineinstrs %s -o - | FileCheck %s --check-prefixes=CHECK,CHECK-SD
; RUN: llc -mtriple=aarch64 -global-isel=1 -verify-machineinstrs %s -o - | FileCheck %s --check-prefixes=CHECK,CHECK-GI
; RUN: llc -mtriple=aarch64 -global-isel=1 -mattr=+fullfp16,+bf16 -verify-machineinstrs %s -o - | FileCheck %s --check-prefixes=CHECK,CHECK-GI
define float @fptrunc_f64_f32(double %a) {
; CHECK-LABEL: fptrunc_f64_f32:
; CHECK: // %bb.0: // %entry
; CHECK-NEXT: fcvt s0, d0
; CHECK-NEXT: ret
entry:
%c = fptrunc double %a to float
ret float %c
}
define half @fptrunc_f64_f16(double %a) {
; CHECK-LABEL: fptrunc_f64_f16:
; CHECK: // %bb.0: // %entry
; CHECK-NEXT: fcvt h0, d0
; CHECK-NEXT: ret
entry:
%c = fptrunc double %a to half
ret half %c
}
define half @fptrunc_f32_f16(float %a) {
; CHECK-LABEL: fptrunc_f32_f16:
; CHECK: // %bb.0: // %entry
; CHECK-NEXT: fcvt h0, s0
; CHECK-NEXT: ret
entry:
%c = fptrunc float %a to half
ret half %c
}
define half @fptrunc_f128_f16(fp128 %a) {
; CHECK-SD-LABEL: fptrunc_f128_f16:
; 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 __trunctfhf2
; CHECK-SD-NEXT: ldr x30, [sp], #16 // 8-byte Folded Reload
; CHECK-SD-NEXT: ret
;
; CHECK-GI-LABEL: fptrunc_f128_f16:
; CHECK-GI: // %bb.0: // %entry
; CHECK-GI-NEXT: b __trunctfhf2
entry:
%c = fptrunc fp128 %a to half
ret half %c
}
define float @fptrunc_f128_f32(fp128 %a) {
; CHECK-SD-LABEL: fptrunc_f128_f32:
; 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 __trunctfsf2
; CHECK-SD-NEXT: ldr x30, [sp], #16 // 8-byte Folded Reload
; CHECK-SD-NEXT: ret
;
; CHECK-GI-LABEL: fptrunc_f128_f32:
; CHECK-GI: // %bb.0: // %entry
; CHECK-GI-NEXT: b __trunctfsf2
entry:
%c = fptrunc fp128 %a to float
ret float %c
}
define double @fptrunc_f128_f64(fp128 %a) {
; CHECK-SD-LABEL: fptrunc_f128_f64:
; 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 __trunctfdf2
; CHECK-SD-NEXT: ldr x30, [sp], #16 // 8-byte Folded Reload
; CHECK-SD-NEXT: ret
;
; CHECK-GI-LABEL: fptrunc_f128_f64:
; CHECK-GI: // %bb.0: // %entry
; CHECK-GI-NEXT: b __trunctfdf2
entry:
%c = fptrunc fp128 %a to double
ret double %c
}
define <2 x half> @fptrunc_v2f128_v2f16(<2 x fp128> %a) {
; CHECK-SD-LABEL: fptrunc_v2f128_v2f16:
; CHECK-SD: // %bb.0: // %entry
; CHECK-SD-NEXT: sub sp, sp, #48
; CHECK-SD-NEXT: str x30, [sp, #32] // 8-byte Folded Spill
; CHECK-SD-NEXT: .cfi_def_cfa_offset 48
; CHECK-SD-NEXT: .cfi_offset w30, -16
; CHECK-SD-NEXT: str q0, [sp] // 16-byte Folded Spill
; CHECK-SD-NEXT: mov v0.16b, v1.16b
; CHECK-SD-NEXT: bl __trunctfhf2
; CHECK-SD-NEXT: // kill: def $h0 killed $h0 def $q0
; CHECK-SD-NEXT: str q0, [sp, #16] // 16-byte Folded Spill
; CHECK-SD-NEXT: ldr q0, [sp] // 16-byte Folded Reload
; CHECK-SD-NEXT: bl __trunctfhf2
; CHECK-SD-NEXT: ldr q1, [sp, #16] // 16-byte Folded Reload
; CHECK-SD-NEXT: // kill: def $h0 killed $h0 def $q0
; CHECK-SD-NEXT: ldr x30, [sp, #32] // 8-byte Folded Reload
; CHECK-SD-NEXT: mov v0.h[1], v1.h[0]
; CHECK-SD-NEXT: // kill: def $d0 killed $d0 killed $q0
; CHECK-SD-NEXT: add sp, sp, #48
; CHECK-SD-NEXT: ret
;
; CHECK-GI-LABEL: fptrunc_v2f128_v2f16:
; CHECK-GI: // %bb.0: // %entry
; CHECK-GI-NEXT: sub sp, sp, #64
; CHECK-GI-NEXT: str x30, [sp, #48] // 8-byte Folded Spill
; CHECK-GI-NEXT: .cfi_def_cfa_offset 64
; CHECK-GI-NEXT: .cfi_offset w30, -16
; CHECK-GI-NEXT: mov v2.d[0], x8
; CHECK-GI-NEXT: str q1, [sp] // 16-byte Folded Spill
; CHECK-GI-NEXT: mov v2.d[1], x8
; CHECK-GI-NEXT: str q2, [sp, #32] // 16-byte Folded Spill
; CHECK-GI-NEXT: bl __trunctfhf2
; CHECK-GI-NEXT: // kill: def $h0 killed $h0 def $q0
; CHECK-GI-NEXT: str q0, [sp, #16] // 16-byte Folded Spill
; CHECK-GI-NEXT: ldr q0, [sp] // 16-byte Folded Reload
; CHECK-GI-NEXT: bl __trunctfhf2
; CHECK-GI-NEXT: // kill: def $h0 killed $h0 def $q0
; CHECK-GI-NEXT: str q0, [sp] // 16-byte Folded Spill
; CHECK-GI-NEXT: ldr q0, [sp, #32] // 16-byte Folded Reload
; CHECK-GI-NEXT: bl __trunctfhf2
; CHECK-GI-NEXT: ldr q0, [sp, #32] // 16-byte Folded Reload
; CHECK-GI-NEXT: bl __trunctfhf2
; CHECK-GI-NEXT: ldp q1, q0, [sp] // 32-byte Folded Reload
; CHECK-GI-NEXT: ldr x30, [sp, #48] // 8-byte Folded Reload
; CHECK-GI-NEXT: mov v0.h[1], v1.h[0]
; CHECK-GI-NEXT: // kill: def $d0 killed $d0 killed $q0
; CHECK-GI-NEXT: add sp, sp, #64
; CHECK-GI-NEXT: ret
entry:
%c = fptrunc <2 x fp128> %a to <2 x half>
ret <2 x half> %c
}
define <2 x float> @fptrunc_v2f128_v2f32(<2 x fp128> %a) {
; CHECK-SD-LABEL: fptrunc_v2f128_v2f32:
; CHECK-SD: // %bb.0: // %entry
; CHECK-SD-NEXT: sub sp, sp, #48
; CHECK-SD-NEXT: str x30, [sp, #32] // 8-byte Folded Spill
; CHECK-SD-NEXT: .cfi_def_cfa_offset 48
; CHECK-SD-NEXT: .cfi_offset w30, -16
; CHECK-SD-NEXT: str q0, [sp] // 16-byte Folded Spill
; CHECK-SD-NEXT: mov v0.16b, v1.16b
; CHECK-SD-NEXT: bl __trunctfsf2
; CHECK-SD-NEXT: // kill: def $s0 killed $s0 def $q0
; CHECK-SD-NEXT: str q0, [sp, #16] // 16-byte Folded Spill
; CHECK-SD-NEXT: ldr q0, [sp] // 16-byte Folded Reload
; CHECK-SD-NEXT: bl __trunctfsf2
; CHECK-SD-NEXT: ldr q1, [sp, #16] // 16-byte Folded Reload
; CHECK-SD-NEXT: // kill: def $s0 killed $s0 def $q0
; CHECK-SD-NEXT: ldr x30, [sp, #32] // 8-byte Folded Reload
; CHECK-SD-NEXT: mov v0.s[1], v1.s[0]
; CHECK-SD-NEXT: // kill: def $d0 killed $d0 killed $q0
; CHECK-SD-NEXT: add sp, sp, #48
; CHECK-SD-NEXT: ret
;
; CHECK-GI-LABEL: fptrunc_v2f128_v2f32:
; CHECK-GI: // %bb.0: // %entry
; CHECK-GI-NEXT: sub sp, sp, #48
; CHECK-GI-NEXT: str x30, [sp, #32] // 8-byte Folded Spill
; CHECK-GI-NEXT: .cfi_def_cfa_offset 48
; CHECK-GI-NEXT: .cfi_offset w30, -16
; CHECK-GI-NEXT: str q1, [sp] // 16-byte Folded Spill
; CHECK-GI-NEXT: bl __trunctfsf2
; CHECK-GI-NEXT: // kill: def $s0 killed $s0 def $q0
; CHECK-GI-NEXT: str q0, [sp, #16] // 16-byte Folded Spill
; CHECK-GI-NEXT: ldr q0, [sp] // 16-byte Folded Reload
; CHECK-GI-NEXT: bl __trunctfsf2
; CHECK-GI-NEXT: ldr q1, [sp, #16] // 16-byte Folded Reload
; CHECK-GI-NEXT: // kill: def $s0 killed $s0 def $q0
; CHECK-GI-NEXT: ldr x30, [sp, #32] // 8-byte Folded Reload
; CHECK-GI-NEXT: mov v1.s[1], v0.s[0]
; CHECK-GI-NEXT: fmov d0, d1
; CHECK-GI-NEXT: add sp, sp, #48
; CHECK-GI-NEXT: ret
entry:
%c = fptrunc <2 x fp128> %a to <2 x float>
ret <2 x float> %c
}
define <2 x double> @fptrunc_v2f128_v2f64(<2 x fp128> %a) {
; CHECK-SD-LABEL: fptrunc_v2f128_v2f64:
; CHECK-SD: // %bb.0: // %entry
; CHECK-SD-NEXT: sub sp, sp, #48
; CHECK-SD-NEXT: str x30, [sp, #32] // 8-byte Folded Spill
; CHECK-SD-NEXT: .cfi_def_cfa_offset 48
; CHECK-SD-NEXT: .cfi_offset w30, -16
; CHECK-SD-NEXT: str q0, [sp] // 16-byte Folded Spill
; CHECK-SD-NEXT: mov v0.16b, v1.16b
; CHECK-SD-NEXT: bl __trunctfdf2
; CHECK-SD-NEXT: // kill: def $d0 killed $d0 def $q0
; CHECK-SD-NEXT: str q0, [sp, #16] // 16-byte Folded Spill
; CHECK-SD-NEXT: ldr q0, [sp] // 16-byte Folded Reload
; CHECK-SD-NEXT: bl __trunctfdf2
; CHECK-SD-NEXT: ldr q1, [sp, #16] // 16-byte Folded Reload
; CHECK-SD-NEXT: // kill: def $d0 killed $d0 def $q0
; CHECK-SD-NEXT: ldr x30, [sp, #32] // 8-byte Folded Reload
; CHECK-SD-NEXT: mov v0.d[1], v1.d[0]
; CHECK-SD-NEXT: add sp, sp, #48
; CHECK-SD-NEXT: ret
;
; CHECK-GI-LABEL: fptrunc_v2f128_v2f64:
; CHECK-GI: // %bb.0: // %entry
; CHECK-GI-NEXT: sub sp, sp, #48
; CHECK-GI-NEXT: str x30, [sp, #32] // 8-byte Folded Spill
; CHECK-GI-NEXT: .cfi_def_cfa_offset 48
; CHECK-GI-NEXT: .cfi_offset w30, -16
; CHECK-GI-NEXT: str q1, [sp] // 16-byte Folded Spill
; CHECK-GI-NEXT: bl __trunctfdf2
; CHECK-GI-NEXT: // kill: def $d0 killed $d0 def $q0
; CHECK-GI-NEXT: str q0, [sp, #16] // 16-byte Folded Spill
; CHECK-GI-NEXT: ldr q0, [sp] // 16-byte Folded Reload
; CHECK-GI-NEXT: bl __trunctfdf2
; CHECK-GI-NEXT: ldr q1, [sp, #16] // 16-byte Folded Reload
; CHECK-GI-NEXT: // kill: def $d0 killed $d0 def $q0
; CHECK-GI-NEXT: ldr x30, [sp, #32] // 8-byte Folded Reload
; CHECK-GI-NEXT: mov v1.d[1], v0.d[0]
; CHECK-GI-NEXT: mov v0.16b, v1.16b
; CHECK-GI-NEXT: add sp, sp, #48
; CHECK-GI-NEXT: ret
entry:
%c = fptrunc <2 x fp128> %a to <2 x double>
ret <2 x double> %c
}
define <2 x float> @fptrunc_v2f64_v2f32(<2 x double> %a) {
; CHECK-LABEL: fptrunc_v2f64_v2f32:
; CHECK: // %bb.0: // %entry
; CHECK-NEXT: fcvtn v0.2s, v0.2d
; CHECK-NEXT: ret
entry:
%c = fptrunc <2 x double> %a to <2 x float>
ret <2 x float> %c
}
define <3 x float> @fptrunc_v3f64_v3f32(<3 x double> %a) {
; CHECK-SD-LABEL: fptrunc_v3f64_v3f32:
; CHECK-SD: // %bb.0: // %entry
; CHECK-SD-NEXT: // kill: def $d0 killed $d0 def $q0
; CHECK-SD-NEXT: // kill: def $d1 killed $d1 def $q1
; CHECK-SD-NEXT: // kill: def $d2 killed $d2 def $q2
; CHECK-SD-NEXT: mov v0.d[1], v1.d[0]
; CHECK-SD-NEXT: fcvtn v0.2s, v0.2d
; CHECK-SD-NEXT: fcvtn2 v0.4s, v2.2d
; CHECK-SD-NEXT: ret
;
; CHECK-GI-LABEL: fptrunc_v3f64_v3f32:
; CHECK-GI: // %bb.0: // %entry
; CHECK-GI-NEXT: // kill: def $d0 killed $d0 def $q0
; CHECK-GI-NEXT: // kill: def $d1 killed $d1 def $q1
; CHECK-GI-NEXT: fcvt s2, d2
; CHECK-GI-NEXT: mov v0.d[1], v1.d[0]
; CHECK-GI-NEXT: fcvtn v1.2s, v0.2d
; CHECK-GI-NEXT: mov v0.s[0], v1.s[0]
; CHECK-GI-NEXT: mov v0.s[1], v1.s[1]
; CHECK-GI-NEXT: mov v0.s[2], v2.s[0]
; CHECK-GI-NEXT: ret
entry:
%c = fptrunc <3 x double> %a to <3 x float>
ret <3 x float> %c
}
define <4 x float> @fptrunc_v4f64_v4f32(<4 x double> %a) {
; CHECK-LABEL: fptrunc_v4f64_v4f32:
; CHECK: // %bb.0: // %entry
; CHECK-NEXT: fcvtn v0.2s, v0.2d
; CHECK-NEXT: fcvtn2 v0.4s, v1.2d
; CHECK-NEXT: ret
entry:
%c = fptrunc <4 x double> %a to <4 x float>
ret <4 x float> %c
}
define <2 x half> @fptrunc_v2f64_v2f16(<2 x double> %a) {
; CHECK-SD-LABEL: fptrunc_v2f64_v2f16:
; CHECK-SD: // %bb.0: // %entry
; CHECK-SD-NEXT: fcvtxn v0.2s, v0.2d
; CHECK-SD-NEXT: fcvtn v0.4h, v0.4s
; CHECK-SD-NEXT: ret
;
; CHECK-GI-LABEL: fptrunc_v2f64_v2f16:
; CHECK-GI: // %bb.0: // %entry
; CHECK-GI-NEXT: mov d1, v0.d[1]
; CHECK-GI-NEXT: fcvt h0, d0
; CHECK-GI-NEXT: fcvt h1, d1
; CHECK-GI-NEXT: mov v0.h[1], v1.h[0]
; CHECK-GI-NEXT: // kill: def $d0 killed $d0 killed $q0
; CHECK-GI-NEXT: ret
entry:
%c = fptrunc <2 x double> %a to <2 x half>
ret <2 x half> %c
}
define <3 x half> @fptrunc_v3f64_v3f16(<3 x double> %a) {
; CHECK-SD-LABEL: fptrunc_v3f64_v3f16:
; CHECK-SD: // %bb.0: // %entry
; CHECK-SD-NEXT: fcvt h1, d1
; CHECK-SD-NEXT: fcvt h0, d0
; CHECK-SD-NEXT: mov v0.h[1], v1.h[0]
; CHECK-SD-NEXT: fcvt h1, d2
; CHECK-SD-NEXT: mov v0.h[2], v1.h[0]
; CHECK-SD-NEXT: // kill: def $d0 killed $d0 killed $q0
; CHECK-SD-NEXT: ret
;
; CHECK-GI-LABEL: fptrunc_v3f64_v3f16:
; CHECK-GI: // %bb.0: // %entry
; CHECK-GI-NEXT: fcvt h0, d0
; CHECK-GI-NEXT: fcvt h1, d1
; CHECK-GI-NEXT: fcvt h2, d2
; CHECK-GI-NEXT: mov v0.h[1], v1.h[0]
; CHECK-GI-NEXT: mov v0.h[2], v2.h[0]
; CHECK-GI-NEXT: // kill: def $d0 killed $d0 killed $q0
; CHECK-GI-NEXT: ret
entry:
%c = fptrunc <3 x double> %a to <3 x half>
ret <3 x half> %c
}
define <4 x half> @fptrunc_v4f64_v4f16(<4 x double> %a) {
; CHECK-SD-LABEL: fptrunc_v4f64_v4f16:
; CHECK-SD: // %bb.0: // %entry
; CHECK-SD-NEXT: fcvtxn v0.2s, v0.2d
; CHECK-SD-NEXT: fcvtxn2 v0.4s, v1.2d
; CHECK-SD-NEXT: fcvtn v0.4h, v0.4s
; CHECK-SD-NEXT: ret
;
; CHECK-GI-LABEL: fptrunc_v4f64_v4f16:
; CHECK-GI: // %bb.0: // %entry
; CHECK-GI-NEXT: mov d2, v0.d[1]
; CHECK-GI-NEXT: fcvt h0, d0
; CHECK-GI-NEXT: mov d3, v1.d[1]
; CHECK-GI-NEXT: fcvt h1, d1
; CHECK-GI-NEXT: fcvt h2, d2
; CHECK-GI-NEXT: mov v0.h[1], v2.h[0]
; CHECK-GI-NEXT: fcvt h2, d3
; CHECK-GI-NEXT: mov v0.h[2], v1.h[0]
; CHECK-GI-NEXT: mov v0.h[3], v2.h[0]
; CHECK-GI-NEXT: // kill: def $d0 killed $d0 killed $q0
; CHECK-GI-NEXT: ret
entry:
%c = fptrunc <4 x double> %a to <4 x half>
ret <4 x half> %c
}
define <2 x half> @fptrunc_v2f32_v2f16(<2 x float> %a) {
; CHECK-SD-LABEL: fptrunc_v2f32_v2f16:
; CHECK-SD: // %bb.0: // %entry
; CHECK-SD-NEXT: // kill: def $d0 killed $d0 def $q0
; CHECK-SD-NEXT: fcvtn v0.4h, v0.4s
; CHECK-SD-NEXT: ret
;
; CHECK-GI-LABEL: fptrunc_v2f32_v2f16:
; CHECK-GI: // %bb.0: // %entry
; CHECK-GI-NEXT: // kill: def $d0 killed $d0 def $q0
; CHECK-GI-NEXT: mov v1.s[0], v0.s[0]
; CHECK-GI-NEXT: mov v1.s[1], v0.s[1]
; CHECK-GI-NEXT: fcvtn v0.4h, v1.4s
; CHECK-GI-NEXT: ret
entry:
%c = fptrunc <2 x float> %a to <2 x half>
ret <2 x half> %c
}
define <3 x half> @fptrunc_v3f32_v3f16(<3 x float> %a) {
; CHECK-LABEL: fptrunc_v3f32_v3f16:
; CHECK: // %bb.0: // %entry
; CHECK-NEXT: fcvtn v0.4h, v0.4s
; CHECK-NEXT: ret
entry:
%c = fptrunc <3 x float> %a to <3 x half>
ret <3 x half> %c
}
define <4 x half> @fptrunc_v4f32_v4f16(<4 x float> %a) {
; CHECK-LABEL: fptrunc_v4f32_v4f16:
; CHECK: // %bb.0: // %entry
; CHECK-NEXT: fcvtn v0.4h, v0.4s
; CHECK-NEXT: ret
entry:
%c = fptrunc <4 x float> %a to <4 x half>
ret <4 x half> %c
}
define <8 x half> @fptrunc_v8f32_v8f16(<8 x float> %a) {
; CHECK-LABEL: fptrunc_v8f32_v8f16:
; CHECK: // %bb.0: // %entry
; CHECK-NEXT: fcvtn v0.4h, v0.4s
; CHECK-NEXT: fcvtn2 v0.8h, v1.4s
; CHECK-NEXT: ret
entry:
%c = fptrunc <8 x float> %a to <8 x half>
ret <8 x half> %c
}
Reference in New Issue View Git Blame Copy Permalink