From: Roman Lebedev <lebedev.ri@gmail.com>
Date: Thu, 23 May 2019 10:55:13 +0000 (+0000)
Subject: [NFC][X86] Fix check prefixes and autogenerate fold-pcmpeqd-2.ll test
X-Git-Url: https://granicus.if.org/sourcecode?a=commitdiff_plain;h=685cac1a6b03f4a9858a69469236bcccacc4a9ca;p=llvm

[NFC][X86] Fix check prefixes and autogenerate fold-pcmpeqd-2.ll test

Being affected by (sub %x, c) -> (add %x, (sub 0, c))
patch in an uncertain way.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@361483 91177308-0d34-0410-b5e6-96231b3b80d8
---

diff --git a/test/CodeGen/X86/fold-pcmpeqd-2.ll b/test/CodeGen/X86/fold-pcmpeqd-2.ll
index d95c6323de4..55c3287028c 100644
--- a/test/CodeGen/X86/fold-pcmpeqd-2.ll
+++ b/test/CodeGen/X86/fold-pcmpeqd-2.ll
@@ -1,5 +1,6 @@
-; RUN: llc < %s -mtriple=i386-apple-darwin -mcpu=yonah -regalloc=basic | FileCheck %s
-; RUN: llc < %s -mtriple=x86_64-apple-darwin -regalloc=basic | FileCheck %s
+; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
+; RUN: llc < %s -mtriple=i386-apple-darwin -mcpu=yonah -regalloc=basic | FileCheck %s --check-prefixes=ALL,X32
+; RUN: llc < %s -mtriple=x86_64-apple-darwin -regalloc=basic | FileCheck %s --check-prefixes=ALL,X64
 
 ; This testcase should need to spill the -1 value on both x86-32 and x86-64,
 ; so it shouldn't use pcmpeqd to materialize an all-ones vector; it
@@ -7,18 +8,177 @@
 ;
 ; RAGreedy defeats the test by splitting live ranges.
 
-; Constant pool all-ones vector:
-; CHECK: .space 16,255
-
-; No pcmpeqd instructions, everybody uses the constant pool.
-; CHECK-LABEL: program_1:
-; CHECK-NOT: pcmpeqd
+; There should be no pcmpeqd instructions, everybody should the constant pool.
 
 	%struct.__ImageExecInfo = type <{ <4 x i32>, <4 x float>, <2 x i64>, i8*, i8*, i8*, i32, i32, i32, i32, i32 }>
 	%struct._cl_image_format_t = type <{ i32, i32, i32 }>
 	%struct._image2d_t = type <{ i8*, %struct._cl_image_format_t, i32, i32, i32, i32, i32, i32 }>
 
 define void @program_1(%struct._image2d_t* %dest, %struct._image2d_t* %t0, <4 x float> %p0, <4 x float> %p1, <4 x float> %p4, <4 x float> %p5, <4 x float> %p6) nounwind {
+; X32-LABEL: program_1:
+; X32:       ## %bb.0: ## %entry
+; X32-NEXT:    cmpl $0, 0
+; X32-NEXT:    jle LBB0_2
+; X32-NEXT:  ## %bb.1: ## %forcond
+; X32-NEXT:    cmpl $0, 0
+; X32-NEXT:    jg LBB0_3
+; X32-NEXT:  LBB0_2: ## %ifthen
+; X32-NEXT:    retl
+; X32-NEXT:  LBB0_3: ## %forbody
+; X32-NEXT:    pushl %esi
+; X32-NEXT:    subl $88, %esp
+; X32-NEXT:    movaps {{.*#+}} xmm1 = [1.28E+2,1.28E+2,1.28E+2,1.28E+2]
+; X32-NEXT:    minps LCPI0_3, %xmm1
+; X32-NEXT:    cvttps2dq %xmm1, %xmm0
+; X32-NEXT:    cvtdq2ps %xmm0, %xmm0
+; X32-NEXT:    subps %xmm0, %xmm1
+; X32-NEXT:    movaps %xmm1, {{[-0-9]+}}(%e{{[sb]}}p) ## 16-byte Spill
+; X32-NEXT:    movaps {{[-0-9]+}}(%e{{[sb]}}p), %xmm0 ## 16-byte Reload
+; X32-NEXT:    mulps LCPI0_3, %xmm0
+; X32-NEXT:    movaps %xmm0, {{[-0-9]+}}(%e{{[sb]}}p) ## 16-byte Spill
+; X32-NEXT:    movaps {{[-0-9]+}}(%e{{[sb]}}p), %xmm0 ## 16-byte Reload
+; X32-NEXT:    addps LCPI0_1, %xmm0
+; X32-NEXT:    movaps %xmm0, {{[-0-9]+}}(%e{{[sb]}}p) ## 16-byte Spill
+; X32-NEXT:    movaps {{[-0-9]+}}(%e{{[sb]}}p), %xmm0 ## 16-byte Reload
+; X32-NEXT:    mulps %xmm1, %xmm0
+; X32-NEXT:    movaps %xmm0, {{[-0-9]+}}(%e{{[sb]}}p) ## 16-byte Spill
+; X32-NEXT:    movaps {{[-0-9]+}}(%e{{[sb]}}p), %xmm0 ## 16-byte Reload
+; X32-NEXT:    addps LCPI0_2, %xmm0
+; X32-NEXT:    movaps %xmm0, {{[-0-9]+}}(%e{{[sb]}}p) ## 16-byte Spill
+; X32-NEXT:    movdqa {{[-0-9]+}}(%e{{[sb]}}p), %xmm0 ## 16-byte Reload
+; X32-NEXT:    psubd LCPI0_4, %xmm0
+; X32-NEXT:    movdqa %xmm0, {{[-0-9]+}}(%e{{[sb]}}p) ## 16-byte Spill
+; X32-NEXT:    movaps {{[-0-9]+}}(%e{{[sb]}}p), %xmm0 ## 16-byte Reload
+; X32-NEXT:    mulps LCPI0_3, %xmm0
+; X32-NEXT:    movaps %xmm0, {{[-0-9]+}}(%e{{[sb]}}p) ## 16-byte Spill
+; X32-NEXT:    xorps %xmm0, %xmm0
+; X32-NEXT:    movaps %xmm0, {{[-0-9]+}}(%e{{[sb]}}p) ## 16-byte Spill
+; X32-NEXT:    movaps {{[-0-9]+}}(%e{{[sb]}}p), %xmm0 ## 16-byte Reload
+; X32-NEXT:    mulps %xmm0, %xmm0
+; X32-NEXT:    movaps %xmm0, {{[-0-9]+}}(%e{{[sb]}}p) ## 16-byte Spill
+; X32-NEXT:    movaps {{[-0-9]+}}(%e{{[sb]}}p), %xmm0 ## 16-byte Reload
+; X32-NEXT:    mulps LCPI0_3, %xmm0
+; X32-NEXT:    movaps %xmm0, {{[-0-9]+}}(%e{{[sb]}}p) ## 16-byte Spill
+; X32-NEXT:    xorps %xmm0, %xmm0
+; X32-NEXT:    movaps %xmm0, {{[-0-9]+}}(%e{{[sb]}}p) ## 16-byte Spill
+; X32-NEXT:    movaps {{[-0-9]+}}(%e{{[sb]}}p), %xmm0 ## 16-byte Reload
+; X32-NEXT:    cmpunordps %xmm0, %xmm0
+; X32-NEXT:    movaps %xmm0, {{[-0-9]+}}(%e{{[sb]}}p) ## 16-byte Spill
+; X32-NEXT:    movaps {{[-0-9]+}}(%e{{[sb]}}p), %xmm0 ## 16-byte Reload
+; X32-NEXT:    minps LCPI0_3, %xmm0
+; X32-NEXT:    movaps %xmm0, {{[-0-9]+}}(%e{{[sb]}}p) ## 16-byte Spill
+; X32-NEXT:    xorps %xmm0, %xmm0
+; X32-NEXT:    movaps %xmm0, {{[0-9]+}}(%esp)
+; X32-NEXT:    movl $0, (%esp)
+; X32-NEXT:    xorl %esi, %esi
+; X32-NEXT:    xorps %xmm3, %xmm3
+; X32-NEXT:    movaps {{[-0-9]+}}(%e{{[sb]}}p), %xmm0 ## 16-byte Reload
+; X32-NEXT:    movdqa {{[-0-9]+}}(%e{{[sb]}}p), %xmm1 ## 16-byte Reload
+; X32-NEXT:    movaps {{[-0-9]+}}(%e{{[sb]}}p), %xmm2 ## 16-byte Reload
+; X32-NEXT:    calll *%esi
+; X32-NEXT:    movaps {{[-0-9]+}}(%e{{[sb]}}p), %xmm0 ## 16-byte Reload
+; X32-NEXT:    minps LCPI0_3, %xmm0
+; X32-NEXT:    movaps %xmm0, {{[-0-9]+}}(%e{{[sb]}}p) ## 16-byte Spill
+; X32-NEXT:    pxor %xmm1, %xmm1
+; X32-NEXT:    psubd {{[-0-9]+}}(%e{{[sb]}}p), %xmm1 ## 16-byte Folded Reload
+; X32-NEXT:    movdqa {{[-0-9]+}}(%e{{[sb]}}p), %xmm0 ## 16-byte Reload
+; X32-NEXT:    psubd LCPI0_4, %xmm0
+; X32-NEXT:    movdqa %xmm0, {{[-0-9]+}}(%e{{[sb]}}p) ## 16-byte Spill
+; X32-NEXT:    movdqa {{[-0-9]+}}(%e{{[sb]}}p), %xmm0 ## 16-byte Reload
+; X32-NEXT:    por %xmm1, %xmm0
+; X32-NEXT:    movdqa %xmm0, {{[-0-9]+}}(%e{{[sb]}}p) ## 16-byte Spill
+; X32-NEXT:    pxor %xmm0, %xmm0
+; X32-NEXT:    movdqa %xmm0, {{[0-9]+}}(%esp)
+; X32-NEXT:    movl $0, (%esp)
+; X32-NEXT:    xorps %xmm3, %xmm3
+; X32-NEXT:    movaps {{[-0-9]+}}(%e{{[sb]}}p), %xmm0 ## 16-byte Reload
+; X32-NEXT:    movaps {{[-0-9]+}}(%e{{[sb]}}p), %xmm1 ## 16-byte Reload
+; X32-NEXT:    movaps {{[-0-9]+}}(%e{{[sb]}}p), %xmm2 ## 16-byte Reload
+; X32-NEXT:    calll *%esi
+; X32-NEXT:    ud2
+;
+; X64-LABEL: program_1:
+; X64:       ## %bb.0: ## %entry
+; X64-NEXT:    cmpl $0, 0
+; X64-NEXT:    jle LBB0_2
+; X64-NEXT:  ## %bb.1: ## %forcond
+; X64-NEXT:    cmpl $0, 0
+; X64-NEXT:    jg LBB0_3
+; X64-NEXT:  LBB0_2: ## %ifthen
+; X64-NEXT:    retq
+; X64-NEXT:  LBB0_3: ## %forbody
+; X64-NEXT:    pushq %rbx
+; X64-NEXT:    subq $64, %rsp
+; X64-NEXT:    xorps %xmm0, %xmm0
+; X64-NEXT:    movaps %xmm0, {{[-0-9]+}}(%r{{[sb]}}p) ## 16-byte Spill
+; X64-NEXT:    movaps {{.*#+}} xmm1 = [1.28E+2,1.28E+2,1.28E+2,1.28E+2]
+; X64-NEXT:    minps {{.*}}(%rip), %xmm1
+; X64-NEXT:    cvttps2dq %xmm1, %xmm0
+; X64-NEXT:    cvtdq2ps %xmm0, %xmm0
+; X64-NEXT:    subps %xmm0, %xmm1
+; X64-NEXT:    movaps %xmm1, (%rsp) ## 16-byte Spill
+; X64-NEXT:    movaps (%rsp), %xmm0 ## 16-byte Reload
+; X64-NEXT:    mulps {{.*}}(%rip), %xmm0
+; X64-NEXT:    movaps %xmm0, (%rsp) ## 16-byte Spill
+; X64-NEXT:    movaps (%rsp), %xmm0 ## 16-byte Reload
+; X64-NEXT:    addps {{.*}}(%rip), %xmm0
+; X64-NEXT:    movaps %xmm0, (%rsp) ## 16-byte Spill
+; X64-NEXT:    movaps (%rsp), %xmm0 ## 16-byte Reload
+; X64-NEXT:    mulps %xmm1, %xmm0
+; X64-NEXT:    movaps %xmm0, (%rsp) ## 16-byte Spill
+; X64-NEXT:    movaps (%rsp), %xmm0 ## 16-byte Reload
+; X64-NEXT:    addps {{.*}}(%rip), %xmm0
+; X64-NEXT:    movaps %xmm0, (%rsp) ## 16-byte Spill
+; X64-NEXT:    movdqa (%rsp), %xmm0 ## 16-byte Reload
+; X64-NEXT:    psubd {{.*}}(%rip), %xmm0
+; X64-NEXT:    movdqa %xmm0, (%rsp) ## 16-byte Spill
+; X64-NEXT:    movaps (%rsp), %xmm0 ## 16-byte Reload
+; X64-NEXT:    mulps {{.*}}(%rip), %xmm0
+; X64-NEXT:    movaps %xmm0, (%rsp) ## 16-byte Spill
+; X64-NEXT:    xorps %xmm0, %xmm0
+; X64-NEXT:    movaps %xmm0, {{[-0-9]+}}(%r{{[sb]}}p) ## 16-byte Spill
+; X64-NEXT:    movaps {{[-0-9]+}}(%r{{[sb]}}p), %xmm0 ## 16-byte Reload
+; X64-NEXT:    mulps %xmm0, %xmm0
+; X64-NEXT:    movaps %xmm0, {{[-0-9]+}}(%r{{[sb]}}p) ## 16-byte Spill
+; X64-NEXT:    movaps (%rsp), %xmm0 ## 16-byte Reload
+; X64-NEXT:    mulps {{.*}}(%rip), %xmm0
+; X64-NEXT:    movaps %xmm0, (%rsp) ## 16-byte Spill
+; X64-NEXT:    xorps %xmm0, %xmm0
+; X64-NEXT:    movaps %xmm0, {{[-0-9]+}}(%r{{[sb]}}p) ## 16-byte Spill
+; X64-NEXT:    movaps {{[-0-9]+}}(%r{{[sb]}}p), %xmm0 ## 16-byte Reload
+; X64-NEXT:    cmpunordps %xmm0, %xmm0
+; X64-NEXT:    movaps %xmm0, {{[-0-9]+}}(%r{{[sb]}}p) ## 16-byte Spill
+; X64-NEXT:    movaps {{[-0-9]+}}(%r{{[sb]}}p), %xmm0 ## 16-byte Reload
+; X64-NEXT:    minps {{.*}}(%rip), %xmm0
+; X64-NEXT:    movaps %xmm0, {{[-0-9]+}}(%r{{[sb]}}p) ## 16-byte Spill
+; X64-NEXT:    xorl %ebx, %ebx
+; X64-NEXT:    xorps %xmm3, %xmm3
+; X64-NEXT:    xorps %xmm4, %xmm4
+; X64-NEXT:    movaps (%rsp), %xmm0 ## 16-byte Reload
+; X64-NEXT:    movaps {{[-0-9]+}}(%r{{[sb]}}p), %xmm1 ## 16-byte Reload
+; X64-NEXT:    movaps {{[-0-9]+}}(%r{{[sb]}}p), %xmm2 ## 16-byte Reload
+; X64-NEXT:    xorl %edi, %edi
+; X64-NEXT:    callq *%rbx
+; X64-NEXT:    movaps (%rsp), %xmm0 ## 16-byte Reload
+; X64-NEXT:    minps {{.*}}(%rip), %xmm0
+; X64-NEXT:    movaps %xmm0, (%rsp) ## 16-byte Spill
+; X64-NEXT:    movdqa {{[-0-9]+}}(%r{{[sb]}}p), %xmm0 ## 16-byte Reload
+; X64-NEXT:    psubd {{[-0-9]+}}(%r{{[sb]}}p), %xmm0 ## 16-byte Folded Reload
+; X64-NEXT:    movdqa %xmm0, {{[-0-9]+}}(%r{{[sb]}}p) ## 16-byte Spill
+; X64-NEXT:    movdqa {{[-0-9]+}}(%r{{[sb]}}p), %xmm0 ## 16-byte Reload
+; X64-NEXT:    psubd {{.*}}(%rip), %xmm0
+; X64-NEXT:    movdqa %xmm0, {{[-0-9]+}}(%r{{[sb]}}p) ## 16-byte Spill
+; X64-NEXT:    movaps {{[-0-9]+}}(%r{{[sb]}}p), %xmm0 ## 16-byte Reload
+; X64-NEXT:    orps {{[-0-9]+}}(%r{{[sb]}}p), %xmm0 ## 16-byte Folded Reload
+; X64-NEXT:    movaps %xmm0, {{[-0-9]+}}(%r{{[sb]}}p) ## 16-byte Spill
+; X64-NEXT:    xorps %xmm3, %xmm3
+; X64-NEXT:    xorps %xmm4, %xmm4
+; X64-NEXT:    movaps {{[-0-9]+}}(%r{{[sb]}}p), %xmm0 ## 16-byte Reload
+; X64-NEXT:    movaps {{[-0-9]+}}(%r{{[sb]}}p), %xmm1 ## 16-byte Reload
+; X64-NEXT:    movaps (%rsp), %xmm2 ## 16-byte Reload
+; X64-NEXT:    xorl %edi, %edi
+; X64-NEXT:    callq *%rbx
+; X64-NEXT:    ud2
 entry:
 	%tmp3.i = load i32, i32* null		; <i32> [#uses=1]
 	%cmp = icmp slt i32 0, %tmp3.i		; <i1> [#uses=1]