diff options
Diffstat (limited to 'llvm/test/CodeGen/X86')
25 files changed, 545 insertions, 151 deletions
diff --git a/llvm/test/CodeGen/X86/2007-08-09-IllegalX86-64Asm.ll b/llvm/test/CodeGen/X86/2007-08-09-IllegalX86-64Asm.ll index 28b4541..7bdc4e1 100644 --- a/llvm/test/CodeGen/X86/2007-08-09-IllegalX86-64Asm.ll +++ b/llvm/test/CodeGen/X86/2007-08-09-IllegalX86-64Asm.ll @@ -44,7 +44,7 @@ define ptr @ubyte_divmod(ptr %a, ptr %b) { ; CHECK-NEXT: leaq {{[0-9]+}}(%rsp), %rsi ; CHECK-NEXT: callq __ubyte_convert_to_ctype ; CHECK-NEXT: testl %eax, %eax -; CHECK-NEXT: js LBB0_6 +; CHECK-NEXT: js LBB0_4 ; CHECK-NEXT: ## %bb.1: ## %cond_next.i ; CHECK-NEXT: leaq {{[0-9]+}}(%rsp), %rsi ; CHECK-NEXT: movq %rbx, %rdi @@ -53,84 +53,81 @@ define ptr @ubyte_divmod(ptr %a, ptr %b) { ; CHECK-NEXT: sarl $31, %ecx ; CHECK-NEXT: andl %eax, %ecx ; CHECK-NEXT: cmpl $-2, %ecx -; CHECK-NEXT: je LBB0_10 +; CHECK-NEXT: je LBB0_8 ; CHECK-NEXT: ## %bb.2: ## %cond_next.i ; CHECK-NEXT: cmpl $-1, %ecx -; CHECK-NEXT: jne LBB0_3 -; CHECK-NEXT: LBB0_8: ## %bb4 +; CHECK-NEXT: jne LBB0_6 +; CHECK-NEXT: LBB0_3: ## %bb4 ; CHECK-NEXT: movq _PyArray_API@GOTPCREL(%rip), %rax ; CHECK-NEXT: movq (%rax), %rax ; CHECK-NEXT: movq 16(%rax), %rax -; CHECK-NEXT: jmp LBB0_9 -; CHECK-NEXT: LBB0_6: ## %_ubyte_convert2_to_ctypes.exit +; CHECK-NEXT: jmp LBB0_10 +; CHECK-NEXT: LBB0_4: ## %_ubyte_convert2_to_ctypes.exit ; CHECK-NEXT: cmpl $-2, %eax -; CHECK-NEXT: je LBB0_10 -; CHECK-NEXT: ## %bb.7: ## %_ubyte_convert2_to_ctypes.exit -; CHECK-NEXT: cmpl $-1, %eax ; CHECK-NEXT: je LBB0_8 -; CHECK-NEXT: LBB0_3: ## %bb35 +; CHECK-NEXT: ## %bb.5: ## %_ubyte_convert2_to_ctypes.exit +; CHECK-NEXT: cmpl $-1, %eax +; CHECK-NEXT: je LBB0_3 +; CHECK-NEXT: LBB0_6: ## %bb35 ; CHECK-NEXT: movq _PyUFunc_API@GOTPCREL(%rip), %r14 ; CHECK-NEXT: movq (%r14), %rax ; CHECK-NEXT: callq *216(%rax) ; CHECK-NEXT: movzbl {{[0-9]+}}(%rsp), %edx ; CHECK-NEXT: testb %dl, %dl -; CHECK-NEXT: je LBB0_4 -; CHECK-NEXT: ## %bb.12: ## %cond_false.i -; CHECK-NEXT: setne %dil +; CHECK-NEXT: je LBB0_11 +; CHECK-NEXT: ## %bb.7: ## %cond_false.i ; CHECK-NEXT: movzbl {{[0-9]+}}(%rsp), %esi ; CHECK-NEXT: movzbl %sil, %ecx ; CHECK-NEXT: movl %ecx, %eax ; CHECK-NEXT: divb %dl ; CHECK-NEXT: movl %eax, %r15d ; CHECK-NEXT: testb %cl, %cl -; CHECK-NEXT: setne %al -; CHECK-NEXT: testb %dil, %al -; CHECK-NEXT: jne LBB0_5 -; CHECK-NEXT: LBB0_13: ## %cond_true.i200 -; CHECK-NEXT: testb %dl, %dl -; CHECK-NEXT: jne LBB0_15 -; CHECK-NEXT: ## %bb.14: ## %cond_true14.i -; CHECK-NEXT: movl $4, %edi -; CHECK-NEXT: callq _feraiseexcept -; CHECK-NEXT: LBB0_15: ## %ubyte_ctype_remainder.exit -; CHECK-NEXT: xorl %ebx, %ebx -; CHECK-NEXT: jmp LBB0_16 -; CHECK-NEXT: LBB0_10: ## %bb17 +; CHECK-NEXT: jne LBB0_12 +; CHECK-NEXT: jmp LBB0_14 +; CHECK-NEXT: LBB0_8: ## %bb17 ; CHECK-NEXT: callq _PyErr_Occurred ; CHECK-NEXT: testq %rax, %rax -; CHECK-NEXT: jne LBB0_23 -; CHECK-NEXT: ## %bb.11: ## %cond_next +; CHECK-NEXT: jne LBB0_27 +; CHECK-NEXT: ## %bb.9: ## %cond_next ; CHECK-NEXT: movq _PyArray_API@GOTPCREL(%rip), %rax ; CHECK-NEXT: movq (%rax), %rax ; CHECK-NEXT: movq 80(%rax), %rax -; CHECK-NEXT: LBB0_9: ## %bb4 +; CHECK-NEXT: LBB0_10: ## %bb4 ; CHECK-NEXT: movq 96(%rax), %rax ; CHECK-NEXT: movq %r14, %rdi ; CHECK-NEXT: movq %rbx, %rsi ; CHECK-NEXT: callq *40(%rax) -; CHECK-NEXT: jmp LBB0_24 -; CHECK-NEXT: LBB0_4: ## %cond_true.i +; CHECK-NEXT: jmp LBB0_28 +; CHECK-NEXT: LBB0_11: ## %cond_true.i ; CHECK-NEXT: movl $4, %edi ; CHECK-NEXT: callq _feraiseexcept ; CHECK-NEXT: movzbl {{[0-9]+}}(%rsp), %edx ; CHECK-NEXT: movzbl {{[0-9]+}}(%rsp), %esi +; CHECK-NEXT: xorl %r15d, %r15d ; CHECK-NEXT: testb %sil, %sil -; CHECK-NEXT: sete %al +; CHECK-NEXT: je LBB0_14 +; CHECK-NEXT: LBB0_12: ## %cond_false.i ; CHECK-NEXT: testb %dl, %dl -; CHECK-NEXT: sete %cl -; CHECK-NEXT: xorl %r15d, %r15d -; CHECK-NEXT: orb %al, %cl -; CHECK-NEXT: jne LBB0_13 -; CHECK-NEXT: LBB0_5: ## %cond_next17.i +; CHECK-NEXT: je LBB0_14 +; CHECK-NEXT: ## %bb.13: ## %cond_next17.i ; CHECK-NEXT: movzbl %sil, %eax ; CHECK-NEXT: divb %dl ; CHECK-NEXT: movzbl %ah, %ebx -; CHECK-NEXT: LBB0_16: ## %ubyte_ctype_remainder.exit +; CHECK-NEXT: jmp LBB0_18 +; CHECK-NEXT: LBB0_14: ## %cond_true.i200 +; CHECK-NEXT: testb %dl, %dl +; CHECK-NEXT: jne LBB0_17 +; CHECK-NEXT: ## %bb.16: ## %cond_true14.i +; CHECK-NEXT: movl $4, %edi +; CHECK-NEXT: callq _feraiseexcept +; CHECK-NEXT: LBB0_17: ## %ubyte_ctype_remainder.exit +; CHECK-NEXT: xorl %ebx, %ebx +; CHECK-NEXT: LBB0_18: ## %ubyte_ctype_remainder.exit ; CHECK-NEXT: movq (%r14), %rax ; CHECK-NEXT: callq *224(%rax) ; CHECK-NEXT: testl %eax, %eax -; CHECK-NEXT: je LBB0_19 -; CHECK-NEXT: ## %bb.17: ## %cond_true61 +; CHECK-NEXT: je LBB0_21 +; CHECK-NEXT: ## %bb.19: ## %cond_true61 ; CHECK-NEXT: movl %eax, %ebp ; CHECK-NEXT: movq (%r14), %rax ; CHECK-NEXT: movq _.str5@GOTPCREL(%rip), %rdi @@ -139,8 +136,8 @@ define ptr @ubyte_divmod(ptr %a, ptr %b) { ; CHECK-NEXT: leaq {{[0-9]+}}(%rsp), %rcx ; CHECK-NEXT: callq *200(%rax) ; CHECK-NEXT: testl %eax, %eax -; CHECK-NEXT: js LBB0_23 -; CHECK-NEXT: ## %bb.18: ## %cond_next73 +; CHECK-NEXT: js LBB0_27 +; CHECK-NEXT: ## %bb.20: ## %cond_next73 ; CHECK-NEXT: movl $1, {{[0-9]+}}(%rsp) ; CHECK-NEXT: movq (%r14), %rax ; CHECK-NEXT: movq {{[0-9]+}}(%rsp), %rsi @@ -149,13 +146,13 @@ define ptr @ubyte_divmod(ptr %a, ptr %b) { ; CHECK-NEXT: movl %ebp, %edx ; CHECK-NEXT: callq *232(%rax) ; CHECK-NEXT: testl %eax, %eax -; CHECK-NEXT: jne LBB0_23 -; CHECK-NEXT: LBB0_19: ## %cond_next89 +; CHECK-NEXT: jne LBB0_27 +; CHECK-NEXT: LBB0_21: ## %cond_next89 ; CHECK-NEXT: movl $2, %edi ; CHECK-NEXT: callq _PyTuple_New ; CHECK-NEXT: testq %rax, %rax -; CHECK-NEXT: je LBB0_23 -; CHECK-NEXT: ## %bb.20: ## %cond_next97 +; CHECK-NEXT: je LBB0_27 +; CHECK-NEXT: ## %bb.22: ## %cond_next97 ; CHECK-NEXT: movq %rax, %r14 ; CHECK-NEXT: movq _PyArray_API@GOTPCREL(%rip), %r12 ; CHECK-NEXT: movq (%r12), %rax @@ -163,8 +160,8 @@ define ptr @ubyte_divmod(ptr %a, ptr %b) { ; CHECK-NEXT: xorl %esi, %esi ; CHECK-NEXT: callq *304(%rdi) ; CHECK-NEXT: testq %rax, %rax -; CHECK-NEXT: je LBB0_21 -; CHECK-NEXT: ## %bb.25: ## %cond_next135 +; CHECK-NEXT: je LBB0_25 +; CHECK-NEXT: ## %bb.23: ## %cond_next135 ; CHECK-NEXT: movb %r15b, 16(%rax) ; CHECK-NEXT: movq %rax, 24(%r14) ; CHECK-NEXT: movq (%r12), %rax @@ -172,22 +169,22 @@ define ptr @ubyte_divmod(ptr %a, ptr %b) { ; CHECK-NEXT: xorl %esi, %esi ; CHECK-NEXT: callq *304(%rdi) ; CHECK-NEXT: testq %rax, %rax -; CHECK-NEXT: je LBB0_21 -; CHECK-NEXT: ## %bb.26: ## %cond_next182 +; CHECK-NEXT: je LBB0_25 +; CHECK-NEXT: ## %bb.24: ## %cond_next182 ; CHECK-NEXT: movb %bl, 16(%rax) ; CHECK-NEXT: movq %rax, 32(%r14) ; CHECK-NEXT: movq %r14, %rax -; CHECK-NEXT: jmp LBB0_24 -; CHECK-NEXT: LBB0_21: ## %cond_true113 +; CHECK-NEXT: jmp LBB0_28 +; CHECK-NEXT: LBB0_25: ## %cond_true113 ; CHECK-NEXT: decq (%r14) -; CHECK-NEXT: jne LBB0_23 -; CHECK-NEXT: ## %bb.22: ## %cond_true126 +; CHECK-NEXT: jne LBB0_27 +; CHECK-NEXT: ## %bb.26: ## %cond_true126 ; CHECK-NEXT: movq 8(%r14), %rax ; CHECK-NEXT: movq %r14, %rdi ; CHECK-NEXT: callq *48(%rax) -; CHECK-NEXT: LBB0_23: ## %UnifiedReturnBlock +; CHECK-NEXT: LBB0_27: ## %UnifiedReturnBlock ; CHECK-NEXT: xorl %eax, %eax -; CHECK-NEXT: LBB0_24: ## %UnifiedReturnBlock +; CHECK-NEXT: LBB0_28: ## %UnifiedReturnBlock ; CHECK-NEXT: addq $32, %rsp ; CHECK-NEXT: popq %rbx ; CHECK-NEXT: popq %r12 diff --git a/llvm/test/CodeGen/X86/absolute-symbol-kernel-code-model.ll b/llvm/test/CodeGen/X86/absolute-symbol-kernel-code-model.ll new file mode 100644 index 0000000..ce7024d --- /dev/null +++ b/llvm/test/CodeGen/X86/absolute-symbol-kernel-code-model.ll @@ -0,0 +1,34 @@ +; RUN: llc --code-model=kernel < %s -asm-verbose=0 | FileCheck %s + +target triple = "x86_64-unknown-linux-gnu" + +; CHECK-LABEL: func_no_abs_sym +define i64 @func_no_abs_sym() nounwind { + ; CHECK: movq $no_abs_sym, %rax + %1 = ptrtoint ptr @no_abs_sym to i64 + ret i64 %1 +} + +; CHECK-LABEL: func_abs_sym +define i64 @func_abs_sym() nounwind { + ; CHECK: movabsq $abs_sym, %rax + %1 = ptrtoint ptr @abs_sym to i64 + ret i64 %1 +} + +; CHECK-LABEL: func_abs_sym_in_range +define i64 @func_abs_sym_in_range() nounwind { + ;; The absolute_symbol range fits in 32 bits but we still use movabs + ;; since there's no benefit to using the sign extending instruction + ;; with absolute symbols. + ; CHECK: movabsq $abs_sym_in_range, %rax + %1 = ptrtoint ptr @abs_sym_in_range to i64 + ret i64 %1 +} + +@no_abs_sym = external hidden global [0 x i8] +@abs_sym = external hidden global [0 x i8], !absolute_symbol !0 +@abs_sym_in_range = external hidden global [0 x i8], !absolute_symbol !1 + +!0 = !{i64 -1, i64 -1} ;; Full range +!1 = !{i64 -2147483648, i64 2147483648} ;; In range diff --git a/llvm/test/CodeGen/X86/apx/cf.ll b/llvm/test/CodeGen/X86/apx/cf.ll index b2651e9..de9caa5 100644 --- a/llvm/test/CodeGen/X86/apx/cf.ll +++ b/llvm/test/CodeGen/X86/apx/cf.ll @@ -230,6 +230,24 @@ entry: ret void } +define void @and_cond(i32 %a, i1 %b) { +; CHECK-LABEL: and_cond: +; CHECK: # %bb.0: +; CHECK-NEXT: testl %edi, %edi +; CHECK-NEXT: setg %al +; CHECK-NEXT: notb %sil +; CHECK-NEXT: xorl %ecx, %ecx +; CHECK-NEXT: testb %al, %sil +; CHECK-NEXT: cfcmovnel %ecx, 0 +; CHECK-NEXT: retq + %is_pos = icmp sgt i32 %a, 0 + %not_b = xor i1 %b, true + %cond = and i1 %not_b, %is_pos + %mask = insertelement <1 x i1> zeroinitializer, i1 %cond, i64 0 + call void @llvm.masked.store.v1i32.p0(<1 x i32> zeroinitializer, ptr null, i32 1, <1 x i1> %mask) + ret void +} + define i64 @redundant_test(i64 %num, ptr %p1, i64 %in) { ; CHECK-LABEL: redundant_test: ; CHECK: # %bb.0: diff --git a/llvm/test/CodeGen/X86/call-graph-section-addrtaken.ll b/llvm/test/CodeGen/X86/call-graph-section-addrtaken.ll index 2aea9c1..632d90d 100644 --- a/llvm/test/CodeGen/X86/call-graph-section-addrtaken.ll +++ b/llvm/test/CodeGen/X86/call-graph-section-addrtaken.ll @@ -27,7 +27,7 @@ entry: !1 = !{i64 0, !"_ZTSFivE.generalized"} !2 = !{i64 0, !"_ZTSFviE.generalized"} -; CHECK: .section .callgraph,"o",@progbits,.text +; CHECK: .section .llvm.callgraph,"o",@progbits,.text ;; Version ; CHECK-NEXT: .byte 0 ;; Flags -- Potential indirect target so LSB is set to 1. Other bits are 0. diff --git a/llvm/test/CodeGen/X86/call-graph-section-assembly.ll b/llvm/test/CodeGen/X86/call-graph-section-assembly.ll index 1aabf66..ed6849a 100644 --- a/llvm/test/CodeGen/X86/call-graph-section-assembly.ll +++ b/llvm/test/CodeGen/X86/call-graph-section-assembly.ll @@ -1,8 +1,8 @@ ;; Test if temporary labels are generated for each indirect callsite. -;; Test if the .callgraph section contains the MD5 hash of callees' type (type id) +;; Test if the .llvm.callgraph section contains the MD5 hash of callees' type (type id) ;; is correctly paired with its corresponding temporary label generated for indirect ;; call sites annotated with !callee_type metadata. -;; Test if the .callgraph section contains unique direct callees. +;; Test if the .llvm.callgraph section contains unique direct callees. ; RUN: llc -mtriple=x86_64-unknown-linux --call-graph-section -o - < %s | FileCheck %s @@ -36,7 +36,7 @@ entry: !4 = !{!5} !5 = !{i64 0, !"_ZTSFPvS_E.generalized"} -; CHECK: .section .callgraph,"o",@progbits,.text +; CHECK: .section .llvm.callgraph,"o",@progbits,.text ;; Version ; CHECK-NEXT: .byte 0 ;; Flags diff --git a/llvm/test/CodeGen/X86/call-graph-section-tailcall.ll b/llvm/test/CodeGen/X86/call-graph-section-tailcall.ll index 34dc5b8..49cc335 100644 --- a/llvm/test/CodeGen/X86/call-graph-section-tailcall.ll +++ b/llvm/test/CodeGen/X86/call-graph-section-tailcall.ll @@ -1,7 +1,10 @@ -;; Tests that we store the type identifiers in .callgraph section of the object file for tailcalls. +;; Tests that we store the type identifiers in .llvm.callgraph section of the object file for tailcalls. + +; REQUIRES: x86-registered-target +; REQUIRES: arm-registered-target ; RUN: llc -mtriple=x86_64-unknown-linux --call-graph-section -filetype=obj -o - < %s | \ -; RUN: llvm-readelf -x .callgraph - | FileCheck %s +; RUN: llvm-readelf -x .llvm.callgraph - | FileCheck %s define i32 @check_tailcall(ptr %func, i8 %x) !type !0 { entry: @@ -27,7 +30,7 @@ declare !type !2 i32 @bar(i8 signext) !2 = !{i64 0, !"_ZTSFicE.generalized"} !3 = !{i64 0, !"_ZTSFiiE.generalized"} -; CHECK: Hex dump of section '.callgraph': +; CHECK: Hex dump of section '.llvm.callgraph': ; CHECK-NEXT: 0x00000000 00050000 00000000 00008e19 0b7f3326 ; CHECK-NEXT: 0x00000010 e3000154 86bc5981 4b8e3000 05000000 ;; Verify that the type id 0x308e4b8159bc8654 is in section. diff --git a/llvm/test/CodeGen/X86/call-graph-section.ll b/llvm/test/CodeGen/X86/call-graph-section.ll index c144a24..8a1c6ca 100644 --- a/llvm/test/CodeGen/X86/call-graph-section.ll +++ b/llvm/test/CodeGen/X86/call-graph-section.ll @@ -1,7 +1,10 @@ -;; Tests that we store the type identifiers in .callgraph section of the object file. +;; Tests that we store the type identifiers in .llvm.callgraph section of the object file. + +; REQUIRES: x86-registered-target +; REQUIRES: arm-registered-target ; RUN: llc -mtriple=x86_64-unknown-linux --call-graph-section -filetype=obj -o - < %s | \ -; RUN: llvm-readelf -x .callgraph - | FileCheck %s +; RUN: llvm-readelf -x .llvm.callgraph - | FileCheck %s declare !type !0 void @foo() @@ -31,7 +34,7 @@ entry: ;; Make sure following type IDs are in call graph section ;; 0x5eecb3e2444f731f, 0x814b8e305486bc59, 0xf897fd777ade6814 -; CHECK: Hex dump of section '.callgraph': +; CHECK: Hex dump of section '.llvm.callgraph': ; CHECK-NEXT: 0x00000000 00050000 00000000 00000000 00000000 ; CHECK-NEXT: 0x00000010 00000324 44f731f5 eecb3e54 86bc5981 ; CHECK-NEXT: 0x00000020 4b8e307a de6814f8 97fd77 diff --git a/llvm/test/CodeGen/X86/combine-udiv.ll b/llvm/test/CodeGen/X86/combine-udiv.ll index 5571519..c90344b8 100644 --- a/llvm/test/CodeGen/X86/combine-udiv.ll +++ b/llvm/test/CodeGen/X86/combine-udiv.ll @@ -502,11 +502,11 @@ define <8 x i16> @combine_vec_udiv_nonuniform(<8 x i16> %x) { ; SSE2-NEXT: por %xmm2, %xmm1 ; SSE2-NEXT: pmulhuw {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm1 # [25645,61681,8195,9363,512,32769,32897,2] ; SSE2-NEXT: psubw %xmm1, %xmm0 -; SSE2-NEXT: pmulhuw {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm0 +; SSE2-NEXT: pmulhuw {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm0 # [32768,0,0,0] ; SSE2-NEXT: paddw %xmm1, %xmm0 ; SSE2-NEXT: movdqa {{.*#+}} xmm1 = [65535,65535,65535,0,0,65535,65535,0] ; SSE2-NEXT: pandn %xmm0, %xmm1 -; SSE2-NEXT: pmulhuw {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm0 +; SSE2-NEXT: pmulhuw {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm0 # [0,16,0,8,8,0,0,0,0,0,2,0,2,0,0,0] ; SSE2-NEXT: por %xmm1, %xmm0 ; SSE2-NEXT: retq ; @@ -517,7 +517,7 @@ define <8 x i16> @combine_vec_udiv_nonuniform(<8 x i16> %x) { ; SSE41-NEXT: pblendw {{.*#+}} xmm1 = xmm0[0,1,2],xmm1[3],xmm0[4,5,6,7] ; SSE41-NEXT: pmulhuw {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm1 # [25645,61681,8195,9363,512,32769,32897,2] ; SSE41-NEXT: psubw %xmm1, %xmm0 -; SSE41-NEXT: pmulhuw {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm0 +; SSE41-NEXT: pmulhuw {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm0 # [32768,0,0,0] ; SSE41-NEXT: paddw %xmm1, %xmm0 ; SSE41-NEXT: movdqa {{.*#+}} xmm1 = [4096,2048,8,u,u,2,2,u] ; SSE41-NEXT: pmulhuw %xmm0, %xmm1 @@ -530,7 +530,7 @@ define <8 x i16> @combine_vec_udiv_nonuniform(<8 x i16> %x) { ; AVX-NEXT: vpblendw {{.*#+}} xmm1 = xmm0[0,1,2],xmm1[3],xmm0[4,5,6,7] ; AVX-NEXT: vpmulhuw {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm1, %xmm1 # [25645,61681,8195,9363,512,32769,32897,2] ; AVX-NEXT: vpsubw %xmm1, %xmm0, %xmm0 -; AVX-NEXT: vpmulhuw {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm0, %xmm0 +; AVX-NEXT: vpmulhuw {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm0, %xmm0 # [32768,0,0,0] ; AVX-NEXT: vpaddw %xmm1, %xmm0, %xmm0 ; AVX-NEXT: vpmulhuw {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm0, %xmm1 # [4096,2048,8,u,u,2,2,u] ; AVX-NEXT: vpblendw {{.*#+}} xmm0 = xmm1[0,1,2],xmm0[3,4],xmm1[5,6],xmm0[7] @@ -541,7 +541,7 @@ define <8 x i16> @combine_vec_udiv_nonuniform(<8 x i16> %x) { ; XOP-NEXT: vpshlw {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm0, %xmm1 ; XOP-NEXT: vpmulhuw {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm1, %xmm1 # [25645,61681,8195,9363,512,32769,32897,2] ; XOP-NEXT: vpsubw %xmm1, %xmm0, %xmm0 -; XOP-NEXT: vpmulhuw {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm0, %xmm0 +; XOP-NEXT: vpmulhuw {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm0, %xmm0 # [32768,0,0,0] ; XOP-NEXT: vpaddw %xmm1, %xmm0, %xmm0 ; XOP-NEXT: vpshlw {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm0, %xmm0 ; XOP-NEXT: retq @@ -630,7 +630,7 @@ define <16 x i8> @combine_vec_udiv_nonuniform4(<16 x i8> %x) { ; SSE2-NEXT: pand %xmm1, %xmm2 ; SSE2-NEXT: pxor %xmm3, %xmm3 ; SSE2-NEXT: punpcklbw {{.*#+}} xmm0 = xmm0[0],xmm3[0],xmm0[1],xmm3[1],xmm0[2],xmm3[2],xmm0[3],xmm3[3],xmm0[4],xmm3[4],xmm0[5],xmm3[5],xmm0[6],xmm3[6],xmm0[7],xmm3[7] -; SSE2-NEXT: pmullw {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm0 +; SSE2-NEXT: pmullw {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm0 # [171,0,0,0] ; SSE2-NEXT: psrlw $15, %xmm0 ; SSE2-NEXT: pandn %xmm0, %xmm1 ; SSE2-NEXT: por %xmm2, %xmm1 @@ -641,7 +641,7 @@ define <16 x i8> @combine_vec_udiv_nonuniform4(<16 x i8> %x) { ; SSE41: # %bb.0: ; SSE41-NEXT: movdqa %xmm0, %xmm1 ; SSE41-NEXT: pmovzxbw {{.*#+}} xmm2 = xmm0[0],zero,xmm0[1],zero,xmm0[2],zero,xmm0[3],zero,xmm0[4],zero,xmm0[5],zero,xmm0[6],zero,xmm0[7],zero -; SSE41-NEXT: pmullw {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm2 +; SSE41-NEXT: pmullw {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm2 # [171,0,0,0] ; SSE41-NEXT: psrlw $8, %xmm2 ; SSE41-NEXT: packuswb %xmm2, %xmm2 ; SSE41-NEXT: psrlw $7, %xmm2 @@ -654,7 +654,7 @@ define <16 x i8> @combine_vec_udiv_nonuniform4(<16 x i8> %x) { ; AVX-LABEL: combine_vec_udiv_nonuniform4: ; AVX: # %bb.0: ; AVX-NEXT: vpmovzxbw {{.*#+}} xmm1 = xmm0[0],zero,xmm0[1],zero,xmm0[2],zero,xmm0[3],zero,xmm0[4],zero,xmm0[5],zero,xmm0[6],zero,xmm0[7],zero -; AVX-NEXT: vpmullw {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm1, %xmm1 +; AVX-NEXT: vpmullw {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm1, %xmm1 # [171,0,0,0] ; AVX-NEXT: vpsrlw $8, %xmm1, %xmm1 ; AVX-NEXT: vpackuswb %xmm1, %xmm1, %xmm1 ; AVX-NEXT: vpsrlw $7, %xmm1, %xmm1 @@ -691,7 +691,7 @@ define <8 x i16> @pr38477(<8 x i16> %a0) { ; SSE2-NEXT: psubw %xmm3, %xmm0 ; SSE2-NEXT: pmulhuw {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm0 # [u,32768,0,0,0,0,0,32768] ; SSE2-NEXT: paddw %xmm3, %xmm0 -; SSE2-NEXT: pmulhuw {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm0 +; SSE2-NEXT: pmulhuw {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm0 # [u,u,0,4,0,4,16,0,4,0,0,4,0,0,0,16] ; SSE2-NEXT: pand {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm3 ; SSE2-NEXT: por %xmm3, %xmm0 ; SSE2-NEXT: pand %xmm1, %xmm0 diff --git a/llvm/test/CodeGen/X86/cpus-intel.ll b/llvm/test/CodeGen/X86/cpus-intel.ll index 40c38c2..646629d 100644 --- a/llvm/test/CodeGen/X86/cpus-intel.ll +++ b/llvm/test/CodeGen/X86/cpus-intel.ll @@ -38,6 +38,8 @@ ; RUN: llc < %s -o /dev/null -mtriple=i686-unknown-unknown -mcpu=lunarlake 2>&1 | FileCheck %s --check-prefix=CHECK-NO-ERROR --allow-empty ; RUN: llc < %s -o /dev/null -mtriple=i686-unknown-unknown -mcpu=gracemont 2>&1 | FileCheck %s --check-prefix=CHECK-NO-ERROR --allow-empty ; RUN: llc < %s -o /dev/null -mtriple=i686-unknown-unknown -mcpu=pantherlake 2>&1 | FileCheck %s --check-prefix=CHECK-NO-ERROR --allow-empty +; RUN: llc < %s -o /dev/null -mtriple=i686-unknown-unknown -mcpu=wildcatlake 2>&1 | FileCheck %s --check-prefix=CHECK-NO-ERROR --allow-empty +; RUN: llc < %s -o /dev/null -mtriple=i686-unknown-unknown -mcpu=novalake 2>&1 | FileCheck %s --check-prefix=CHECK-NO-ERROR --allow-empty ; RUN: llc < %s -o /dev/null -mtriple=i686-unknown-unknown -mcpu=clearwaterforest 2>&1 | FileCheck %s --check-prefix=CHECK-NO-ERROR --allow-empty ; RUN: llc < %s -o /dev/null -mtriple=i686-unknown-unknown -mcpu=diamondrapids 2>&1 | FileCheck %s --check-prefix=CHECK-NO-ERROR --allow-empty @@ -104,6 +106,8 @@ ; RUN: llc < %s -o /dev/null -mtriple=x86_64-unknown-unknown -mcpu=lunarlake 2>&1 | FileCheck %s --check-prefix=CHECK-NO-ERROR --allow-empty ; RUN: llc < %s -o /dev/null -mtriple=x86_64-unknown-unknown -mcpu=gracemont 2>&1 | FileCheck %s --check-prefix=CHECK-NO-ERROR --allow-empty ; RUN: llc < %s -o /dev/null -mtriple=x86_64-unknown-unknown -mcpu=pantherlake 2>&1 | FileCheck %s --check-prefix=CHECK-NO-ERROR --allow-empty +; RUN: llc < %s -o /dev/null -mtriple=x86_64-unknown-unknown -mcpu=wildcatlake 2>&1 | FileCheck %s --check-prefix=CHECK-NO-ERROR --allow-empty +; RUN: llc < %s -o /dev/null -mtriple=x86_64-unknown-unknown -mcpu=novalake 2>&1 | FileCheck %s --check-prefix=CHECK-NO-ERROR --allow-empty ; RUN: llc < %s -o /dev/null -mtriple=x86_64-unknown-unknown -mcpu=clearwaterforest 2>&1 | FileCheck %s --check-prefix=CHECK-NO-ERROR --allow-empty ; RUN: llc < %s -o /dev/null -mtriple=x86_64-unknown-unknown -mcpu=diamondrapids 2>&1 | FileCheck %s --check-prefix=CHECK-NO-ERROR --allow-empty diff --git a/llvm/test/CodeGen/X86/global-variable-partition-with-dap.ll b/llvm/test/CodeGen/X86/global-variable-partition-with-dap.ll index f3950b7..b2b0a6d 100644 --- a/llvm/test/CodeGen/X86/global-variable-partition-with-dap.ll +++ b/llvm/test/CodeGen/X86/global-variable-partition-with-dap.ll @@ -1,17 +1,101 @@ target datalayout = "e-m:e-p270:32:32-p271:32:32-p272:64:64-i64:64-i128:128-f80:128-n8:16:32:64-S128" target triple = "x86_64-unknown-linux-gnu" -;; A minimal test case. Subsequent PRs will expand on this test case -;; (e.g., with more functions, variables and profiles) and test the hotness -;; reconcillation implementation. +;; Requires asserts for -debug-only. +; REQUIRES: asserts + +; RUN: rm -rf %t && split-file %s %t && cd %t + +; RUN: llc -mtriple=x86_64-unknown-linux-gnu -relocation-model=pic \ +; RUN: -partition-static-data-sections=true \ +; RUN: -debug-only=static-data-profile-info \ +; RUN: -data-sections=true -unique-section-names=false \ +; RUN: input-with-data-access-prof-on.ll -o - 2>&1 | FileCheck %s --check-prefixes=LOG,IR + ; RUN: llc -mtriple=x86_64-unknown-linux-gnu -relocation-model=pic \ ; RUN: -partition-static-data-sections=true \ +; RUN: -debug-only=static-data-profile-info \ ; RUN: -data-sections=true -unique-section-names=false \ -; RUN: %s -o - 2>&1 | FileCheck %s --check-prefix=IR +; RUN: input-with-data-access-prof-off.ll -o - 2>&1 | FileCheck %s --check-prefixes=OFF + +; LOG: hot_bss has section prefix hot, the max from data access profiles as hot and PGO counters as hot +; LOG: data_unknown_hotness has section prefix <empty>, the max from data access profiles as <empty> and PGO counters as unlikely +; LOG: external_relro_array has section prefix unlikely, solely from data access profiles + +; IR: .type hot_bss,@object +; IR-NEXT: .section .bss.hot.,"aw" +; IR: .type data_unknown_hotness,@object +; IR-NEXT: .section .data,"aw" +; IR: .type external_relro_array,@object +; IR-NEXT: .section .data.rel.ro.unlikely.,"aw" + + +; OFF: .type hot_bss,@object +; OFF-NEXT: .section .bss.hot.,"aw" +; OFF: .type data_unknown_hotness,@object +; OFF-NEXT: .section .data.unlikely.,"aw" +;; Global variable section prefix metadata is not used when +;; module flag `EnableDataAccessProf` is 0, and @external_relro_array has +;; external linkage, so analysis based on PGO counters doesn't apply. +; OFF: .type external_relro_array,@object # @external_relro_array +; OFF-NEXT: .section .data.rel.ro,"aw" + +;--- input-with-data-access-prof-on.ll +; Internal vars +@hot_bss = internal global i32 0, !section_prefix !17 +@data_unknown_hotness = internal global i32 1 +; External vars +@external_relro_array = constant [2 x ptr] [ptr @hot_bss, ptr @data_unknown_hotness], !section_prefix !18 + +define void @cold_func() !prof !15 { + %9 = load i32, ptr @data_unknown_hotness + %11 = call i32 (...) @func_taking_arbitrary_param(i32 %9) + ret void +} + +define void @hot_func() !prof !14 { + %9 = load i32, ptr @hot_bss + %11 = call i32 (...) @func_taking_arbitrary_param(i32 %9) + ret void +} + +declare i32 @func_taking_arbitrary_param(...) -; IR: .section .bss.hot.,"aw" +!llvm.module.flags = !{!0, !1} +!0 = !{i32 2, !"EnableDataAccessProf", i32 1} +!1 = !{i32 1, !"ProfileSummary", !2} +!2 = !{!3, !4, !5, !6, !7, !8, !9, !10} +!3 = !{!"ProfileFormat", !"InstrProf"} +!4 = !{!"TotalCount", i64 1460183} +!5 = !{!"MaxCount", i64 849024} +!6 = !{!"MaxInternalCount", i64 32769} +!7 = !{!"MaxFunctionCount", i64 849024} +!8 = !{!"NumCounts", i64 23627} +!9 = !{!"NumFunctions", i64 3271} +!10 = !{!"DetailedSummary", !11} +!11 = !{!12, !13} +!12 = !{i32 990000, i64 166, i32 73} +!13 = !{i32 999999, i64 3, i32 1443} +!14 = !{!"function_entry_count", i64 100000} +!15 = !{!"function_entry_count", i64 1} +!16 = !{!"branch_weights", i32 1, i32 99999} +!17 = !{!"section_prefix", !"hot"} +!18 = !{!"section_prefix", !"unlikely"} + +;--- input-with-data-access-prof-off.ll +; Same as file above except that module flag `EnableDataAccessProf` has value 0. +; Internal vars @hot_bss = internal global i32 0, !section_prefix !17 +@data_unknown_hotness = internal global i32 1 +; External vars +@external_relro_array = constant [2 x ptr] [ptr @hot_bss, ptr @data_unknown_hotness], !section_prefix !18 + +define void @cold_func() !prof !15 { + %9 = load i32, ptr @data_unknown_hotness + %11 = call i32 (...) @func_taking_arbitrary_param(i32 %9) + ret void +} define void @hot_func() !prof !14 { %9 = load i32, ptr @hot_bss @@ -21,8 +105,9 @@ define void @hot_func() !prof !14 { declare i32 @func_taking_arbitrary_param(...) -!llvm.module.flags = !{!1} +!llvm.module.flags = !{!0, !1} +!0 = !{i32 2, !"EnableDataAccessProf", i32 0} !1 = !{i32 1, !"ProfileSummary", !2} !2 = !{!3, !4, !5, !6, !7, !8, !9, !10} !3 = !{!"ProfileFormat", !"InstrProf"} @@ -40,3 +125,4 @@ declare i32 @func_taking_arbitrary_param(...) !15 = !{!"function_entry_count", i64 1} !16 = !{!"branch_weights", i32 1, i32 99999} !17 = !{!"section_prefix", !"hot"} +!18 = !{!"section_prefix", !"unlikely"} diff --git a/llvm/test/CodeGen/X86/madd.ll b/llvm/test/CodeGen/X86/madd.ll index bdb7c30..4ec54d8 100644 --- a/llvm/test/CodeGen/X86/madd.ll +++ b/llvm/test/CodeGen/X86/madd.ll @@ -2071,7 +2071,7 @@ define <4 x i32> @pmaddwd_negative2(<8 x i16> %A) { ; AVX1: # %bb.0: ; AVX1-NEXT: vpmovsxwd %xmm0, %xmm1 ; AVX1-NEXT: vpunpckhwd {{.*#+}} xmm0 = xmm0[4,4,5,5,6,6,7,7] -; AVX1-NEXT: vpmaddwd {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm0, %xmm0 +; AVX1-NEXT: vpmaddwd {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm0, %xmm0 # [1,7,42,32] ; AVX1-NEXT: vpmulld {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm1, %xmm1 ; AVX1-NEXT: vphaddd %xmm0, %xmm1, %xmm0 ; AVX1-NEXT: retq diff --git a/llvm/test/CodeGen/X86/masked_gather_scatter.ll b/llvm/test/CodeGen/X86/masked_gather_scatter.ll index 4cde581..caec02e 100644 --- a/llvm/test/CodeGen/X86/masked_gather_scatter.ll +++ b/llvm/test/CodeGen/X86/masked_gather_scatter.ll @@ -4765,6 +4765,66 @@ define void @scaleidx_scatter_outofrange(<8 x float> %value, ptr %base, <8 x i32 } declare void @llvm.masked.scatter.v8f32.v8p0(<8 x float>, <8 x ptr>, i32 immarg, <8 x i1>) +define <16 x i32> @pr163023_sext(ptr %a0, <16 x i32> %a1) { +; X64-LABEL: pr163023_sext: +; X64: # %bb.0: +; X64-NEXT: kxnorw %k0, %k0, %k1 +; X64-NEXT: vpxor %xmm1, %xmm1, %xmm1 +; X64-NEXT: vpgatherdd (%rdi,%zmm0), %zmm1 {%k1} +; X64-NEXT: vmovdqa64 %zmm1, %zmm0 +; X64-NEXT: retq +; +; X86-LABEL: pr163023_sext: +; X86: # %bb.0: +; X86-NEXT: movl {{[0-9]+}}(%esp), %eax +; X86-NEXT: kxnorw %k0, %k0, %k1 +; X86-NEXT: vpxor %xmm1, %xmm1, %xmm1 +; X86-NEXT: vpgatherdd (%eax,%zmm0), %zmm1 {%k1} +; X86-NEXT: vmovdqa64 %zmm1, %zmm0 +; X86-NEXT: retl + %addr.p = ptrtoint ptr %a0 to i64 + %addr.v = insertelement <1 x i64> poison, i64 %addr.p, i64 0 + %addr.splat = shufflevector <1 x i64> %addr.v, <1 x i64> poison, <16 x i32> zeroinitializer + %ofs = sext <16 x i32> %a1 to <16 x i64> + %addr = add nuw <16 x i64> %addr.splat, %ofs + %ptr = inttoptr <16 x i64> %addr to <16 x ptr> + %gather = call <16 x i32> @llvm.masked.gather.v16i32.v16p0(<16 x ptr> %ptr, i32 4, <16 x i1> splat (i1 true), <16 x i32> poison) + ret <16 x i32> %gather +} + +define <16 x i32> @pr163023_zext(ptr %a0, <16 x i32> %a1) { +; X64-LABEL: pr163023_zext: +; X64: # %bb.0: +; X64-NEXT: vpmovzxdq {{.*#+}} zmm1 = ymm0[0],zero,ymm0[1],zero,ymm0[2],zero,ymm0[3],zero,ymm0[4],zero,ymm0[5],zero,ymm0[6],zero,ymm0[7],zero +; X64-NEXT: vextracti64x4 $1, %zmm0, %ymm0 +; X64-NEXT: vpmovzxdq {{.*#+}} zmm0 = ymm0[0],zero,ymm0[1],zero,ymm0[2],zero,ymm0[3],zero,ymm0[4],zero,ymm0[5],zero,ymm0[6],zero,ymm0[7],zero +; X64-NEXT: kxnorw %k0, %k0, %k1 +; X64-NEXT: vpxor %xmm2, %xmm2, %xmm2 +; X64-NEXT: vpxor %xmm3, %xmm3, %xmm3 +; X64-NEXT: kxnorw %k0, %k0, %k2 +; X64-NEXT: vpgatherqd (%rdi,%zmm0), %ymm3 {%k2} +; X64-NEXT: vpgatherqd (%rdi,%zmm1), %ymm2 {%k1} +; X64-NEXT: vinserti64x4 $1, %ymm3, %zmm2, %zmm0 +; X64-NEXT: retq +; +; X86-LABEL: pr163023_zext: +; X86: # %bb.0: +; X86-NEXT: movl {{[0-9]+}}(%esp), %eax +; X86-NEXT: kxnorw %k0, %k0, %k1 +; X86-NEXT: vpxor %xmm1, %xmm1, %xmm1 +; X86-NEXT: vpgatherdd (%eax,%zmm0), %zmm1 {%k1} +; X86-NEXT: vmovdqa64 %zmm1, %zmm0 +; X86-NEXT: retl + %addr.p = ptrtoint ptr %a0 to i64 + %addr.v = insertelement <1 x i64> poison, i64 %addr.p, i64 0 + %addr.splat = shufflevector <1 x i64> %addr.v, <1 x i64> poison, <16 x i32> zeroinitializer + %ofs = zext <16 x i32> %a1 to <16 x i64> + %addr = add nuw <16 x i64> %addr.splat, %ofs + %ptr = inttoptr <16 x i64> %addr to <16 x ptr> + %gather = call <16 x i32> @llvm.masked.gather.v16i32.v16p0(<16 x ptr> %ptr, i32 4, <16 x i1> splat (i1 true), <16 x i32> poison) + ret <16 x i32> %gather +} + ; ; PR45906 ; This used to cause fast-isel to generate bad copy instructions that would diff --git a/llvm/test/CodeGen/X86/min-legal-vector-width.ll b/llvm/test/CodeGen/X86/min-legal-vector-width.ll index d752659..04f0a65 100644 --- a/llvm/test/CodeGen/X86/min-legal-vector-width.ll +++ b/llvm/test/CodeGen/X86/min-legal-vector-width.ll @@ -1,5 +1,5 @@ ; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py -; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mcpu=skylake-avx512 -mattr=prefer-256-bit | FileCheck %s --check-prefixes=CHECK,CHECK-SKX +; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mcpu=skylake-avx512 -mattr=prefer-256-bit | FileCheck %s --check-prefixes=CHECK,CHECK-SKX,CHECK-SKX-NOVBMI ; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mcpu=skylake-avx512 -mattr=prefer-256-bit,avx512vbmi | FileCheck %s --check-prefixes=CHECK,CHECK-SKX,CHECK-SKX-VBMI ; Make sure CPUs default to prefer-256-bit. avx512vnni isn't interesting as it just adds an isel peephole for vpmaddwd+vpaddd ; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mcpu=skylake-avx512 | FileCheck %s --check-prefixes=CHECK,CHECK-AVX512 @@ -883,6 +883,30 @@ define <16 x i16> @test_16f32tosb_512(ptr %ptr, <16 x i16> %passthru) "min-legal } define dso_local void @mul256(ptr %a, ptr %b, ptr %c) "min-legal-vector-width"="256" { +; CHECK-SKX-NOVBMI-LABEL: mul256: +; CHECK-SKX-NOVBMI: # %bb.0: +; CHECK-SKX-NOVBMI-NEXT: vmovdqa (%rdi), %ymm0 +; CHECK-SKX-NOVBMI-NEXT: vmovdqa 32(%rdi), %ymm1 +; CHECK-SKX-NOVBMI-NEXT: vmovdqa (%rsi), %ymm2 +; CHECK-SKX-NOVBMI-NEXT: vmovdqa 32(%rsi), %ymm3 +; CHECK-SKX-NOVBMI-NEXT: vpbroadcastd {{.*#+}} ymm4 = [255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255] +; CHECK-SKX-NOVBMI-NEXT: vpand %ymm4, %ymm3, %ymm5 +; CHECK-SKX-NOVBMI-NEXT: vpmaddubsw %ymm5, %ymm1, %ymm5 +; CHECK-SKX-NOVBMI-NEXT: vpandn %ymm3, %ymm4, %ymm3 +; CHECK-SKX-NOVBMI-NEXT: vpmaddubsw %ymm3, %ymm1, %ymm1 +; CHECK-SKX-NOVBMI-NEXT: vpsllw $8, %ymm1, %ymm1 +; CHECK-SKX-NOVBMI-NEXT: vpternlogq {{.*#+}} ymm1 = ymm1 | (ymm5 & ymm4) +; CHECK-SKX-NOVBMI-NEXT: vpand %ymm4, %ymm2, %ymm3 +; CHECK-SKX-NOVBMI-NEXT: vpmaddubsw %ymm3, %ymm0, %ymm3 +; CHECK-SKX-NOVBMI-NEXT: vpandn %ymm2, %ymm4, %ymm2 +; CHECK-SKX-NOVBMI-NEXT: vpmaddubsw %ymm2, %ymm0, %ymm0 +; CHECK-SKX-NOVBMI-NEXT: vpsllw $8, %ymm0, %ymm0 +; CHECK-SKX-NOVBMI-NEXT: vpternlogq {{.*#+}} ymm0 = ymm0 | (ymm3 & ymm4) +; CHECK-SKX-NOVBMI-NEXT: vmovdqa %ymm0, (%rdx) +; CHECK-SKX-NOVBMI-NEXT: vmovdqa %ymm1, 32(%rdx) +; CHECK-SKX-NOVBMI-NEXT: vzeroupper +; CHECK-SKX-NOVBMI-NEXT: retq +; ; CHECK-SKX-VBMI-LABEL: mul256: ; CHECK-SKX-VBMI: # %bb.0: ; CHECK-SKX-VBMI-NEXT: vmovdqa (%rdi), %ymm0 @@ -960,6 +984,21 @@ define dso_local void @mul256(ptr %a, ptr %b, ptr %c) "min-legal-vector-width"=" } define dso_local void @mul512(ptr %a, ptr %b, ptr %c) "min-legal-vector-width"="512" { +; CHECK-SKX-NOVBMI-LABEL: mul512: +; CHECK-SKX-NOVBMI: # %bb.0: +; CHECK-SKX-NOVBMI-NEXT: vmovdqa64 (%rdi), %zmm0 +; CHECK-SKX-NOVBMI-NEXT: vmovdqa64 (%rsi), %zmm1 +; CHECK-SKX-NOVBMI-NEXT: vpbroadcastd {{.*#+}} zmm2 = [255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255] +; CHECK-SKX-NOVBMI-NEXT: vpandq %zmm2, %zmm1, %zmm3 +; CHECK-SKX-NOVBMI-NEXT: vpmaddubsw %zmm3, %zmm0, %zmm3 +; CHECK-SKX-NOVBMI-NEXT: vpandnq %zmm1, %zmm2, %zmm1 +; CHECK-SKX-NOVBMI-NEXT: vpmaddubsw %zmm1, %zmm0, %zmm0 +; CHECK-SKX-NOVBMI-NEXT: vpsllw $8, %zmm0, %zmm0 +; CHECK-SKX-NOVBMI-NEXT: vpternlogq {{.*#+}} zmm0 = zmm0 | (zmm3 & zmm2) +; CHECK-SKX-NOVBMI-NEXT: vmovdqa64 %zmm0, (%rdx) +; CHECK-SKX-NOVBMI-NEXT: vzeroupper +; CHECK-SKX-NOVBMI-NEXT: retq +; ; CHECK-SKX-VBMI-LABEL: mul512: ; CHECK-SKX-VBMI: # %bb.0: ; CHECK-SKX-VBMI-NEXT: vmovdqa64 (%rdi), %zmm0 @@ -1137,6 +1176,14 @@ define <16 x i16> @trunc_v16i32_v16i16_zeroes(ptr %x) nounwind "min-legal-vector } define <32 x i8> @trunc_v32i16_v32i8_zeroes(ptr %x) nounwind "min-legal-vector-width"="256" { +; CHECK-SKX-NOVBMI-LABEL: trunc_v32i16_v32i8_zeroes: +; CHECK-SKX-NOVBMI: # %bb.0: +; CHECK-SKX-NOVBMI-NEXT: vpsrlw $8, 32(%rdi), %ymm0 +; CHECK-SKX-NOVBMI-NEXT: vpsrlw $8, (%rdi), %ymm1 +; CHECK-SKX-NOVBMI-NEXT: vpackuswb %ymm0, %ymm1, %ymm0 +; CHECK-SKX-NOVBMI-NEXT: vpermq {{.*#+}} ymm0 = ymm0[0,2,1,3] +; CHECK-SKX-NOVBMI-NEXT: retq +; ; CHECK-SKX-VBMI-LABEL: trunc_v32i16_v32i8_zeroes: ; CHECK-SKX-VBMI: # %bb.0: ; CHECK-SKX-VBMI-NEXT: vmovdqa (%rdi), %ymm1 @@ -1192,6 +1239,14 @@ define <16 x i16> @trunc_v16i32_v16i16_sign(ptr %x) nounwind "min-legal-vector-w } define <32 x i8> @trunc_v32i16_v32i8_sign(ptr %x) nounwind "min-legal-vector-width"="256" { +; CHECK-SKX-NOVBMI-LABEL: trunc_v32i16_v32i8_sign: +; CHECK-SKX-NOVBMI: # %bb.0: +; CHECK-SKX-NOVBMI-NEXT: vpsrlw $8, 32(%rdi), %ymm0 +; CHECK-SKX-NOVBMI-NEXT: vpsrlw $8, (%rdi), %ymm1 +; CHECK-SKX-NOVBMI-NEXT: vpackuswb %ymm0, %ymm1, %ymm0 +; CHECK-SKX-NOVBMI-NEXT: vpermq {{.*#+}} ymm0 = ymm0[0,2,1,3] +; CHECK-SKX-NOVBMI-NEXT: retq +; ; CHECK-SKX-VBMI-LABEL: trunc_v32i16_v32i8_sign: ; CHECK-SKX-VBMI: # %bb.0: ; CHECK-SKX-VBMI-NEXT: vmovdqa (%rdi), %ymm1 diff --git a/llvm/test/CodeGen/X86/pr160612.ll b/llvm/test/CodeGen/X86/pr160612.ll new file mode 100644 index 0000000..6572c42 --- /dev/null +++ b/llvm/test/CodeGen/X86/pr160612.ll @@ -0,0 +1,74 @@ +; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py +; RUN: llc < %s -mtriple=x86_64-unknown-linux-gnu -O2 | FileCheck %s + +; Test for issue #160612: OR conditions in branches should use multiple branches +; instead of materializing booleans with SETCC when no special optimizations apply. + +declare void @subroutine_foo() +declare void @subroutine_bar() + +; Original issue: (x == 0 || y == 0) was generating SETCC + TEST + BRANCH +; instead of using two conditional branches directly. +define void @func_a(i32 noundef %x, i32 noundef %y) { +; CHECK-LABEL: func_a: +; CHECK: # %bb.0: # %entry +; CHECK-NEXT: testl %edi, %edi +; CHECK-NEXT: je subroutine_foo@PLT # TAILCALL +; CHECK-NEXT: # %bb.1: # %entry +; CHECK-NEXT: testl %esi, %esi +; CHECK-NEXT: jne subroutine_bar@PLT # TAILCALL +; CHECK-NEXT: # %bb.2: # %if.then +; CHECK-NEXT: jmp subroutine_foo@PLT # TAILCALL +entry: + %cmp = icmp eq i32 %x, 0 + %cmp1 = icmp eq i32 %y, 0 + %or.cond = or i1 %cmp, %cmp1 + br i1 %or.cond, label %if.then, label %if.else + +if.then: + tail call void @subroutine_foo() + br label %if.end + +if.else: + tail call void @subroutine_bar() + br label %if.end + +if.end: + ret void +} + +; Reference implementation that already generated optimal code. +; This should continue to generate the same optimal code. +define void @func_b(i32 noundef %x, i32 noundef %y) { +; CHECK-LABEL: func_b: +; CHECK: # %bb.0: # %entry +; CHECK-NEXT: testl %edi, %edi +; CHECK-NEXT: je subroutine_foo@PLT # TAILCALL +; CHECK-NEXT: # %bb.1: # %if.else +; CHECK-NEXT: testl %esi, %esi +; CHECK-NEXT: je subroutine_foo@PLT # TAILCALL +; CHECK-NEXT: # %bb.2: # %if.else3 +; CHECK-NEXT: jmp subroutine_bar@PLT # TAILCALL +entry: + %cmp = icmp eq i32 %x, 0 + br i1 %cmp, label %if.then, label %if.else + +if.then: + tail call void @subroutine_foo() + br label %if.end4 + +if.else: + %cmp1 = icmp eq i32 %y, 0 + br i1 %cmp1, label %if.then2, label %if.else3 + +if.then2: + tail call void @subroutine_foo() + br label %if.end4 + +if.else3: + tail call void @subroutine_bar() + br label %if.end4 + +if.end4: + ret void +} diff --git a/llvm/test/CodeGen/X86/setcc-wide-types.ll b/llvm/test/CodeGen/X86/setcc-wide-types.ll index 69abf6e..d018c53 100644 --- a/llvm/test/CodeGen/X86/setcc-wide-types.ll +++ b/llvm/test/CodeGen/X86/setcc-wide-types.ll @@ -1493,15 +1493,23 @@ define i1 @allbits_i128_load_arg(ptr %w) { } define i1 @anybits_i256_load_arg(ptr %w) { -; ANY-LABEL: anybits_i256_load_arg: -; ANY: # %bb.0: -; ANY-NEXT: movq (%rdi), %rax -; ANY-NEXT: movq 8(%rdi), %rcx -; ANY-NEXT: orq 24(%rdi), %rcx -; ANY-NEXT: orq 16(%rdi), %rax -; ANY-NEXT: orq %rcx, %rax -; ANY-NEXT: setne %al -; ANY-NEXT: retq +; SSE-LABEL: anybits_i256_load_arg: +; SSE: # %bb.0: +; SSE-NEXT: movq (%rdi), %rax +; SSE-NEXT: movq 8(%rdi), %rcx +; SSE-NEXT: orq 24(%rdi), %rcx +; SSE-NEXT: orq 16(%rdi), %rax +; SSE-NEXT: orq %rcx, %rax +; SSE-NEXT: setne %al +; SSE-NEXT: retq +; +; AVXANY-LABEL: anybits_i256_load_arg: +; AVXANY: # %bb.0: +; AVXANY-NEXT: vmovdqu (%rdi), %ymm0 +; AVXANY-NEXT: vptest %ymm0, %ymm0 +; AVXANY-NEXT: setne %al +; AVXANY-NEXT: vzeroupper +; AVXANY-NEXT: retq %ld = load i256, ptr %w %cmp = icmp ne i256 %ld, 0 ret i1 %cmp @@ -1552,21 +1560,30 @@ define i1 @allbits_i256_load_arg(ptr %w) { } define i1 @anybits_i512_load_arg(ptr %w) { -; ANY-LABEL: anybits_i512_load_arg: -; ANY: # %bb.0: -; ANY-NEXT: movq 16(%rdi), %rax -; ANY-NEXT: movq (%rdi), %rcx -; ANY-NEXT: movq 8(%rdi), %rdx -; ANY-NEXT: movq 24(%rdi), %rsi -; ANY-NEXT: orq 56(%rdi), %rsi -; ANY-NEXT: orq 40(%rdi), %rdx -; ANY-NEXT: orq %rsi, %rdx -; ANY-NEXT: orq 48(%rdi), %rax -; ANY-NEXT: orq 32(%rdi), %rcx -; ANY-NEXT: orq %rax, %rcx -; ANY-NEXT: orq %rdx, %rcx -; ANY-NEXT: setne %al -; ANY-NEXT: retq +; NO512-LABEL: anybits_i512_load_arg: +; NO512: # %bb.0: +; NO512-NEXT: movq 16(%rdi), %rax +; NO512-NEXT: movq (%rdi), %rcx +; NO512-NEXT: movq 8(%rdi), %rdx +; NO512-NEXT: movq 24(%rdi), %rsi +; NO512-NEXT: orq 56(%rdi), %rsi +; NO512-NEXT: orq 40(%rdi), %rdx +; NO512-NEXT: orq %rsi, %rdx +; NO512-NEXT: orq 48(%rdi), %rax +; NO512-NEXT: orq 32(%rdi), %rcx +; NO512-NEXT: orq %rax, %rcx +; NO512-NEXT: orq %rdx, %rcx +; NO512-NEXT: setne %al +; NO512-NEXT: retq +; +; AVX512-LABEL: anybits_i512_load_arg: +; AVX512: # %bb.0: +; AVX512-NEXT: vmovdqu64 (%rdi), %zmm0 +; AVX512-NEXT: vptestmd %zmm0, %zmm0, %k0 +; AVX512-NEXT: kortestw %k0, %k0 +; AVX512-NEXT: setne %al +; AVX512-NEXT: vzeroupper +; AVX512-NEXT: retq %ld = load i512, ptr %w %cmp = icmp ne i512 %ld, 0 ret i1 %cmp diff --git a/llvm/test/CodeGen/X86/sqrt-fastmath-mir.ll b/llvm/test/CodeGen/X86/sqrt-fastmath-mir.ll index 42617c1..18588aa 100644 --- a/llvm/test/CodeGen/X86/sqrt-fastmath-mir.ll +++ b/llvm/test/CodeGen/X86/sqrt-fastmath-mir.ll @@ -24,7 +24,7 @@ define float @sqrt_ieee_ninf(float %f) #0 { ; CHECK-NEXT: {{ $}} ; CHECK-NEXT: [[COPY:%[0-9]+]]:fr32 = COPY $xmm0 ; CHECK-NEXT: [[DEF:%[0-9]+]]:fr32 = IMPLICIT_DEF - ; CHECK-NEXT: [[VRSQRTSSr:%[0-9]+]]:fr32 = VRSQRTSSr killed [[DEF]], [[COPY]] + ; CHECK-NEXT: [[VRSQRTSSr:%[0-9]+]]:fr32 = ninf afn VRSQRTSSr killed [[DEF]], [[COPY]] ; CHECK-NEXT: [[VMULSSrr:%[0-9]+]]:fr32 = ninf afn nofpexcept VMULSSrr [[COPY]], [[VRSQRTSSr]], implicit $mxcsr ; CHECK-NEXT: [[VMOVSSrm_alt:%[0-9]+]]:fr32 = VMOVSSrm_alt $rip, 1, $noreg, %const.0, $noreg :: (load (s32) from constant-pool) ; CHECK-NEXT: [[VFMADD213SSr:%[0-9]+]]:fr32 = ninf afn nofpexcept VFMADD213SSr [[VRSQRTSSr]], killed [[VMULSSrr]], [[VMOVSSrm_alt]], implicit $mxcsr @@ -71,7 +71,7 @@ define float @sqrt_daz_ninf(float %f) #1 { ; CHECK-NEXT: {{ $}} ; CHECK-NEXT: [[COPY:%[0-9]+]]:fr32 = COPY $xmm0 ; CHECK-NEXT: [[DEF:%[0-9]+]]:fr32 = IMPLICIT_DEF - ; CHECK-NEXT: [[VRSQRTSSr:%[0-9]+]]:fr32 = VRSQRTSSr killed [[DEF]], [[COPY]] + ; CHECK-NEXT: [[VRSQRTSSr:%[0-9]+]]:fr32 = ninf afn VRSQRTSSr killed [[DEF]], [[COPY]] ; CHECK-NEXT: [[VMULSSrr:%[0-9]+]]:fr32 = ninf afn nofpexcept VMULSSrr [[COPY]], [[VRSQRTSSr]], implicit $mxcsr ; CHECK-NEXT: [[VMOVSSrm_alt:%[0-9]+]]:fr32 = VMOVSSrm_alt $rip, 1, $noreg, %const.0, $noreg :: (load (s32) from constant-pool) ; CHECK-NEXT: [[VFMADD213SSr:%[0-9]+]]:fr32 = ninf afn nofpexcept VFMADD213SSr [[VRSQRTSSr]], killed [[VMULSSrr]], [[VMOVSSrm_alt]], implicit $mxcsr diff --git a/llvm/test/CodeGen/X86/srem-seteq-illegal-types.ll b/llvm/test/CodeGen/X86/srem-seteq-illegal-types.ll index cc4bda8..650b562 100644 --- a/llvm/test/CodeGen/X86/srem-seteq-illegal-types.ll +++ b/llvm/test/CodeGen/X86/srem-seteq-illegal-types.ll @@ -1,6 +1,5 @@ ; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py ; RUN: llc -mtriple=i686-unknown-linux-gnu < %s | FileCheck %s --check-prefixes=X86 -; RUN: llc -mtriple=x86_64-unknown-linux-gnu < %s | FileCheck %s --check-prefixes=X64 ; RUN: llc -mtriple=x86_64-unknown-linux-gnu -mattr=+sse2 < %s | FileCheck %s --check-prefixes=X64,SSE2 ; RUN: llc -mtriple=x86_64-unknown-linux-gnu -mattr=+sse4.1 < %s | FileCheck %s --check-prefixes=X64,SSE41 ; RUN: llc -mtriple=x86_64-unknown-linux-gnu -mattr=+avx < %s | FileCheck %s --check-prefixes=X64,AVX1 diff --git a/llvm/test/CodeGen/X86/urem-seteq-illegal-types.ll b/llvm/test/CodeGen/X86/urem-seteq-illegal-types.ll index 7c1a1e2..874d885 100644 --- a/llvm/test/CodeGen/X86/urem-seteq-illegal-types.ll +++ b/llvm/test/CodeGen/X86/urem-seteq-illegal-types.ll @@ -1,6 +1,5 @@ ; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py ; RUN: llc -mtriple=i686-unknown-linux-gnu < %s | FileCheck %s --check-prefixes=X86 -; RUN: llc -mtriple=x86_64-unknown-linux-gnu < %s | FileCheck %s --check-prefixes=X64 ; RUN: llc -mtriple=x86_64-unknown-linux-gnu -mattr=+sse2 < %s | FileCheck %s --check-prefixes=X64,SSE2 ; RUN: llc -mtriple=x86_64-unknown-linux-gnu -mattr=+sse4.1 < %s | FileCheck %s --check-prefixes=X64,SSE41 ; RUN: llc -mtriple=x86_64-unknown-linux-gnu -mattr=+avx < %s | FileCheck %s --check-prefixes=X64,AVX1 diff --git a/llvm/test/CodeGen/X86/var-permute-128.ll b/llvm/test/CodeGen/X86/var-permute-128.ll index 6174011..83a0ddb 100644 --- a/llvm/test/CodeGen/X86/var-permute-128.ll +++ b/llvm/test/CodeGen/X86/var-permute-128.ll @@ -5,9 +5,9 @@ ; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=+xop | FileCheck %s --check-prefixes=AVX,AVXNOVLBW,XOP ; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=+avx | FileCheck %s --check-prefixes=AVX,AVXNOVLBW,AVX1 ; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=+avx2 | FileCheck %s --check-prefixes=AVX,AVXNOVLBW,AVX2 -; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=+avx512f | FileCheck %s --check-prefixes=AVX,AVXNOVLBW,AVX512 -; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=+avx512bw | FileCheck %s --check-prefixes=AVX,AVXNOVLBW,AVX512 -; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=+avx512vbmi | FileCheck %s --check-prefixes=AVX,AVXNOVLBW,AVX512 +; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=+avx512f | FileCheck %s --check-prefixes=AVX,AVXNOVLBW,AVX512,AVX512F +; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=+avx512bw | FileCheck %s --check-prefixes=AVX,AVXNOVLBW,AVX512,AVX512BW +; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=+avx512vbmi | FileCheck %s --check-prefixes=AVX,AVXNOVLBW,AVX512,AVX512BW ; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=+avx512bw,+avx512vl | FileCheck %s --check-prefixes=AVX,AVX512VL,AVX512VLBW ; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=+avx512bw,+avx512vl,+avx512vbmi | FileCheck %s --check-prefixes=AVX,AVX512VL,VLVBMI @@ -598,6 +598,33 @@ define <8 x i16> @var_shuffle_zero_v8i16(<8 x i16> %v, <8 x i16> %indices) nounw ; AVX2-NEXT: vpshufb %xmm1, %xmm0, %xmm0 ; AVX2-NEXT: retq ; +; AVX512F-LABEL: var_shuffle_zero_v8i16: +; AVX512F: # %bb.0: +; AVX512F-NEXT: vpmaxuw {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm1, %xmm2 +; AVX512F-NEXT: vpcmpeqw %xmm2, %xmm1, %xmm2 +; AVX512F-NEXT: vpor %xmm1, %xmm2, %xmm1 +; AVX512F-NEXT: vpmullw {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm1, %xmm1 # [514,514,514,514,514,514,514,514] +; AVX512F-NEXT: vpaddw {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm1, %xmm1 +; AVX512F-NEXT: vpor %xmm2, %xmm1, %xmm1 +; AVX512F-NEXT: vpshufb %xmm1, %xmm0, %xmm0 +; AVX512F-NEXT: retq +; +; AVX512BW-LABEL: var_shuffle_zero_v8i16: +; AVX512BW: # %bb.0: +; AVX512BW-NEXT: # kill: def $xmm1 killed $xmm1 def $zmm1 +; AVX512BW-NEXT: vpbroadcastw {{.*#+}} xmm2 = [7,7,7,7,7,7,7,7] +; AVX512BW-NEXT: vpcmpnleuw %zmm2, %zmm1, %k1 +; AVX512BW-NEXT: vpcmpeqd %xmm2, %xmm2, %xmm2 +; AVX512BW-NEXT: vmovdqu16 %zmm2, %zmm1 {%k1} +; AVX512BW-NEXT: vpmullw {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm1, %xmm1 # [514,514,514,514,514,514,514,514] +; AVX512BW-NEXT: vpaddw {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm1, %xmm1 +; AVX512BW-NEXT: vpshufb %xmm1, %xmm0, %xmm0 +; AVX512BW-NEXT: vpxor %xmm1, %xmm1, %xmm1 +; AVX512BW-NEXT: vmovdqu16 %zmm1, %zmm0 {%k1} +; AVX512BW-NEXT: # kill: def $xmm0 killed $xmm0 killed $zmm0 +; AVX512BW-NEXT: vzeroupper +; AVX512BW-NEXT: retq +; ; AVX512VL-LABEL: var_shuffle_zero_v8i16: ; AVX512VL: # %bb.0: ; AVX512VL-NEXT: vpcmpnleuw {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm1, %k1 @@ -921,6 +948,28 @@ define <16 x i8> @var_shuffle_zero_v16i8(<16 x i8> %v, <16 x i8> %indices) nounw ; AVX2-NEXT: vpshufb %xmm1, %xmm0, %xmm0 ; AVX2-NEXT: retq ; +; AVX512F-LABEL: var_shuffle_zero_v16i8: +; AVX512F: # %bb.0: +; AVX512F-NEXT: vpmaxub {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm1, %xmm2 +; AVX512F-NEXT: vpcmpeqb %xmm2, %xmm1, %xmm2 +; AVX512F-NEXT: vpor %xmm1, %xmm2, %xmm1 +; AVX512F-NEXT: vpshufb %xmm1, %xmm0, %xmm0 +; AVX512F-NEXT: retq +; +; AVX512BW-LABEL: var_shuffle_zero_v16i8: +; AVX512BW: # %bb.0: +; AVX512BW-NEXT: # kill: def $xmm1 killed $xmm1 def $zmm1 +; AVX512BW-NEXT: vpbroadcastb {{.*#+}} xmm2 = [15,15,15,15,15,15,15,15,15,15,15,15,15,15,15,15] +; AVX512BW-NEXT: vpcmpnleub %zmm2, %zmm1, %k1 +; AVX512BW-NEXT: vpcmpeqd %xmm2, %xmm2, %xmm2 +; AVX512BW-NEXT: vmovdqu8 %zmm2, %zmm1 {%k1} +; AVX512BW-NEXT: vpshufb %xmm1, %xmm0, %xmm0 +; AVX512BW-NEXT: vpxor %xmm1, %xmm1, %xmm1 +; AVX512BW-NEXT: vmovdqu8 %zmm1, %zmm0 {%k1} +; AVX512BW-NEXT: # kill: def $xmm0 killed $xmm0 killed $zmm0 +; AVX512BW-NEXT: vzeroupper +; AVX512BW-NEXT: retq +; ; AVX512VL-LABEL: var_shuffle_zero_v16i8: ; AVX512VL: # %bb.0: ; AVX512VL-NEXT: vpcmpnleub {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm1, %k1 diff --git a/llvm/test/CodeGen/X86/vector-fshr-128.ll b/llvm/test/CodeGen/X86/vector-fshr-128.ll index 9b52857..d16b28a 100644 --- a/llvm/test/CodeGen/X86/vector-fshr-128.ll +++ b/llvm/test/CodeGen/X86/vector-fshr-128.ll @@ -1872,7 +1872,7 @@ define <8 x i16> @constant_funnnel_v8i16(<8 x i16> %x, <8 x i16> %y) nounwind { ; SSE2: # %bb.0: ; SSE2-NEXT: movdqa {{.*#+}} xmm2 = [0,65535,65535,65535,65535,65535,65535,65535] ; SSE2-NEXT: pandn %xmm1, %xmm2 -; SSE2-NEXT: pmulhuw {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm1 +; SSE2-NEXT: pmulhuw {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm1 # [0,0,0,128,0,64,0,32,0,16,0,8,0,4,0,2] ; SSE2-NEXT: por %xmm1, %xmm2 ; SSE2-NEXT: paddw %xmm0, %xmm0 ; SSE2-NEXT: pmullw {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm0 # [32768,16384,8192,4096,2048,1024,512,256] @@ -1964,7 +1964,7 @@ define <8 x i16> @constant_funnnel_v8i16(<8 x i16> %x, <8 x i16> %y) nounwind { ; X86-SSE2: # %bb.0: ; X86-SSE2-NEXT: movdqa {{.*#+}} xmm2 = [0,65535,65535,65535,65535,65535,65535,65535] ; X86-SSE2-NEXT: pandn %xmm1, %xmm2 -; X86-SSE2-NEXT: pmulhuw {{\.?LCPI[0-9]+_[0-9]+}}, %xmm1 +; X86-SSE2-NEXT: pmulhuw {{\.?LCPI[0-9]+_[0-9]+}}, %xmm1 # [0,0,0,128,0,64,0,32,0,16,0,8,0,4,0,2] ; X86-SSE2-NEXT: por %xmm1, %xmm2 ; X86-SSE2-NEXT: paddw %xmm0, %xmm0 ; X86-SSE2-NEXT: pmullw {{\.?LCPI[0-9]+_[0-9]+}}, %xmm0 # [32768,16384,8192,4096,2048,1024,512,256] diff --git a/llvm/test/CodeGen/X86/vector-reduce-add-mask.ll b/llvm/test/CodeGen/X86/vector-reduce-add-mask.ll index 983ae59..3d85d55 100644 --- a/llvm/test/CodeGen/X86/vector-reduce-add-mask.ll +++ b/llvm/test/CodeGen/X86/vector-reduce-add-mask.ll @@ -851,7 +851,7 @@ define i16 @test_v4i16_v4i8(<4 x i16> %a0) { ; SSE2: # %bb.0: ; SSE2-NEXT: movdqa {{.*#+}} xmm1 = [0,65535,65535,65535,65535,65535,65535,65535] ; SSE2-NEXT: pandn %xmm0, %xmm1 -; SSE2-NEXT: pmulhuw {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm0 +; SSE2-NEXT: pmulhuw {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm0 # [0,0,0,128,0,64,0,32,u,u,u,u,u,u,u,u] ; SSE2-NEXT: por %xmm1, %xmm0 ; SSE2-NEXT: pshufd {{.*#+}} xmm1 = xmm0[1,1,1,1] ; SSE2-NEXT: paddw %xmm0, %xmm1 diff --git a/llvm/test/CodeGen/X86/vector-shift-ashr-128.ll b/llvm/test/CodeGen/X86/vector-shift-ashr-128.ll index d565ef0..1602cde 100644 --- a/llvm/test/CodeGen/X86/vector-shift-ashr-128.ll +++ b/llvm/test/CodeGen/X86/vector-shift-ashr-128.ll @@ -1673,7 +1673,7 @@ define <16 x i8> @constant_shift_v16i8_pairs(<16 x i8> %a) nounwind { ; SSE2: # %bb.0: ; SSE2-NEXT: movdqa {{.*#+}} xmm1 = [65535,65535,65535,65535,65535,0,65535,65535] ; SSE2-NEXT: pandn %xmm0, %xmm1 -; SSE2-NEXT: pmulhuw {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm0 +; SSE2-NEXT: pmulhuw {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm0 # [0,128,0,16,0,2,0,32,0,64,0,0,0,8,0,4] ; SSE2-NEXT: por %xmm1, %xmm0 ; SSE2-NEXT: pand {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm0 ; SSE2-NEXT: movdqa {{.*#+}} xmm1 = [64,64,8,8,1,1,16,16,32,32,128,128,4,4,2,2] @@ -1750,7 +1750,7 @@ define <16 x i8> @constant_shift_v16i8_pairs(<16 x i8> %a) nounwind { ; X86-SSE: # %bb.0: ; X86-SSE-NEXT: movdqa {{.*#+}} xmm1 = [65535,65535,65535,65535,65535,0,65535,65535] ; X86-SSE-NEXT: pandn %xmm0, %xmm1 -; X86-SSE-NEXT: pmulhuw {{\.?LCPI[0-9]+_[0-9]+}}, %xmm0 +; X86-SSE-NEXT: pmulhuw {{\.?LCPI[0-9]+_[0-9]+}}, %xmm0 # [0,128,0,16,0,2,0,32,0,64,0,0,0,8,0,4] ; X86-SSE-NEXT: por %xmm1, %xmm0 ; X86-SSE-NEXT: pand {{\.?LCPI[0-9]+_[0-9]+}}, %xmm0 ; X86-SSE-NEXT: movdqa {{.*#+}} xmm1 = [64,64,8,8,1,1,16,16,32,32,128,128,4,4,2,2] diff --git a/llvm/test/CodeGen/X86/vector-shift-lshr-128.ll b/llvm/test/CodeGen/X86/vector-shift-lshr-128.ll index 8cb2c7b..a847da6 100644 --- a/llvm/test/CodeGen/X86/vector-shift-lshr-128.ll +++ b/llvm/test/CodeGen/X86/vector-shift-lshr-128.ll @@ -1223,7 +1223,7 @@ define <8 x i16> @constant_shift_v8i16(<8 x i16> %a) nounwind { ; SSE2: # %bb.0: ; SSE2-NEXT: movdqa {{.*#+}} xmm1 = [0,65535,65535,65535,65535,65535,65535,65535] ; SSE2-NEXT: pandn %xmm0, %xmm1 -; SSE2-NEXT: pmulhuw {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm0 +; SSE2-NEXT: pmulhuw {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm0 # [0,0,0,128,0,64,0,32,0,16,0,8,0,4,0,2] ; SSE2-NEXT: por %xmm1, %xmm0 ; SSE2-NEXT: retq ; @@ -1275,7 +1275,7 @@ define <8 x i16> @constant_shift_v8i16(<8 x i16> %a) nounwind { ; X86-SSE: # %bb.0: ; X86-SSE-NEXT: movdqa {{.*#+}} xmm1 = [0,65535,65535,65535,65535,65535,65535,65535] ; X86-SSE-NEXT: pandn %xmm0, %xmm1 -; X86-SSE-NEXT: pmulhuw {{\.?LCPI[0-9]+_[0-9]+}}, %xmm0 +; X86-SSE-NEXT: pmulhuw {{\.?LCPI[0-9]+_[0-9]+}}, %xmm0 # [0,0,0,128,0,64,0,32,0,16,0,8,0,4,0,2] ; X86-SSE-NEXT: por %xmm1, %xmm0 ; X86-SSE-NEXT: retl %shift = lshr <8 x i16> %a, <i16 0, i16 1, i16 2, i16 3, i16 4, i16 5, i16 6, i16 7> diff --git a/llvm/test/CodeGen/X86/vector-shift-lshr-sub128.ll b/llvm/test/CodeGen/X86/vector-shift-lshr-sub128.ll index 57874c4..eb39b6a 100644 --- a/llvm/test/CodeGen/X86/vector-shift-lshr-sub128.ll +++ b/llvm/test/CodeGen/X86/vector-shift-lshr-sub128.ll @@ -1480,7 +1480,7 @@ define <4 x i16> @constant_shift_v4i16(<4 x i16> %a) nounwind { ; SSE2: # %bb.0: ; SSE2-NEXT: movdqa {{.*#+}} xmm1 = [0,65535,65535,65535,65535,65535,65535,65535] ; SSE2-NEXT: pandn %xmm0, %xmm1 -; SSE2-NEXT: pmulhuw {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm0 +; SSE2-NEXT: pmulhuw {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm0 # [0,0,0,128,0,64,0,32,u,u,u,u,u,u,u,u] ; SSE2-NEXT: por %xmm1, %xmm0 ; SSE2-NEXT: retq ; @@ -1532,7 +1532,7 @@ define <4 x i16> @constant_shift_v4i16(<4 x i16> %a) nounwind { ; X86-SSE: # %bb.0: ; X86-SSE-NEXT: movdqa {{.*#+}} xmm1 = [0,65535,65535,65535,65535,65535,65535,65535] ; X86-SSE-NEXT: pandn %xmm0, %xmm1 -; X86-SSE-NEXT: pmulhuw {{\.?LCPI[0-9]+_[0-9]+}}, %xmm0 +; X86-SSE-NEXT: pmulhuw {{\.?LCPI[0-9]+_[0-9]+}}, %xmm0 # [0,0,0,128,0,64,0,32,u,u,u,u,u,u,u,u] ; X86-SSE-NEXT: por %xmm1, %xmm0 ; X86-SSE-NEXT: retl %shift = lshr <4 x i16> %a, <i16 0, i16 1, i16 2, i16 3> diff --git a/llvm/test/CodeGen/X86/x86-shrink-wrap-unwind.ll b/llvm/test/CodeGen/X86/x86-shrink-wrap-unwind.ll index b2064b1..02d4d88 100644 --- a/llvm/test/CodeGen/X86/x86-shrink-wrap-unwind.ll +++ b/llvm/test/CodeGen/X86/x86-shrink-wrap-unwind.ll @@ -181,40 +181,38 @@ define zeroext i1 @segmentedStack(ptr readonly %vk1, ptr readonly %vk2, i64 %key ; CHECK-LABEL: segmentedStack: ; CHECK: ## %bb.0: ; CHECK-NEXT: cmpq %gs:816, %rsp -; CHECK-NEXT: jbe LBB3_6 +; CHECK-NEXT: jbe LBB3_7 ; CHECK-NEXT: LBB3_1: ## %entry ; CHECK-NEXT: pushq %rax ; CHECK-NEXT: .cfi_def_cfa_offset 16 -; CHECK-NEXT: testq %rdi, %rdi -; CHECK-NEXT: sete %al -; CHECK-NEXT: testq %rsi, %rsi -; CHECK-NEXT: sete %cl -; CHECK-NEXT: orb %al, %cl ; CHECK-NEXT: movq %rdi, %rax ; CHECK-NEXT: orq %rsi, %rax ; CHECK-NEXT: sete %al -; CHECK-NEXT: testb %cl, %cl -; CHECK-NEXT: jne LBB3_4 -; CHECK-NEXT: ## %bb.2: ## %if.end4.i +; CHECK-NEXT: testq %rdi, %rdi +; CHECK-NEXT: je LBB3_5 +; CHECK-NEXT: ## %bb.2: ## %entry +; CHECK-NEXT: testq %rsi, %rsi +; CHECK-NEXT: je LBB3_5 +; CHECK-NEXT: ## %bb.3: ## %if.end4.i ; CHECK-NEXT: movq 8(%rdi), %rdx ; CHECK-NEXT: cmpq 8(%rsi), %rdx -; CHECK-NEXT: jne LBB3_5 -; CHECK-NEXT: ## %bb.3: ## %land.rhs.i.i +; CHECK-NEXT: jne LBB3_6 +; CHECK-NEXT: ## %bb.4: ## %land.rhs.i.i ; CHECK-NEXT: movq (%rsi), %rsi ; CHECK-NEXT: movq (%rdi), %rdi ; CHECK-NEXT: callq _memcmp ; CHECK-NEXT: testl %eax, %eax ; CHECK-NEXT: sete %al -; CHECK-NEXT: LBB3_4: ## %__go_ptr_strings_equal.exit +; CHECK-NEXT: LBB3_5: ## %__go_ptr_strings_equal.exit ; CHECK-NEXT: ## kill: def $al killed $al killed $eax ; CHECK-NEXT: popq %rcx ; CHECK-NEXT: retq -; CHECK-NEXT: LBB3_5: +; CHECK-NEXT: LBB3_6: ; CHECK-NEXT: xorl %eax, %eax ; CHECK-NEXT: ## kill: def $al killed $al killed $eax ; CHECK-NEXT: popq %rcx ; CHECK-NEXT: retq -; CHECK-NEXT: LBB3_6: +; CHECK-NEXT: LBB3_7: ; CHECK-NEXT: movl $8, %r10d ; CHECK-NEXT: movl $0, %r11d ; CHECK-NEXT: callq ___morestack @@ -224,43 +222,41 @@ define zeroext i1 @segmentedStack(ptr readonly %vk1, ptr readonly %vk2, i64 %key ; NOCOMPACTUNWIND-LABEL: segmentedStack: ; NOCOMPACTUNWIND: # %bb.0: ; NOCOMPACTUNWIND-NEXT: cmpq %fs:112, %rsp -; NOCOMPACTUNWIND-NEXT: jbe .LBB3_6 +; NOCOMPACTUNWIND-NEXT: jbe .LBB3_7 ; NOCOMPACTUNWIND-NEXT: .LBB3_1: # %entry ; NOCOMPACTUNWIND-NEXT: pushq %rax ; NOCOMPACTUNWIND-NEXT: .cfi_def_cfa_offset 16 -; NOCOMPACTUNWIND-NEXT: testq %rdi, %rdi -; NOCOMPACTUNWIND-NEXT: sete %al -; NOCOMPACTUNWIND-NEXT: testq %rsi, %rsi -; NOCOMPACTUNWIND-NEXT: sete %cl -; NOCOMPACTUNWIND-NEXT: orb %al, %cl ; NOCOMPACTUNWIND-NEXT: movq %rdi, %rax ; NOCOMPACTUNWIND-NEXT: orq %rsi, %rax ; NOCOMPACTUNWIND-NEXT: sete %al -; NOCOMPACTUNWIND-NEXT: testb %cl, %cl -; NOCOMPACTUNWIND-NEXT: jne .LBB3_4 -; NOCOMPACTUNWIND-NEXT: # %bb.2: # %if.end4.i +; NOCOMPACTUNWIND-NEXT: testq %rdi, %rdi +; NOCOMPACTUNWIND-NEXT: je .LBB3_5 +; NOCOMPACTUNWIND-NEXT: # %bb.2: # %entry +; NOCOMPACTUNWIND-NEXT: testq %rsi, %rsi +; NOCOMPACTUNWIND-NEXT: je .LBB3_5 +; NOCOMPACTUNWIND-NEXT: # %bb.3: # %if.end4.i ; NOCOMPACTUNWIND-NEXT: movq 8(%rdi), %rdx ; NOCOMPACTUNWIND-NEXT: cmpq 8(%rsi), %rdx -; NOCOMPACTUNWIND-NEXT: jne .LBB3_5 -; NOCOMPACTUNWIND-NEXT: # %bb.3: # %land.rhs.i.i +; NOCOMPACTUNWIND-NEXT: jne .LBB3_6 +; NOCOMPACTUNWIND-NEXT: # %bb.4: # %land.rhs.i.i ; NOCOMPACTUNWIND-NEXT: movq (%rsi), %rsi ; NOCOMPACTUNWIND-NEXT: movq (%rdi), %rdi ; NOCOMPACTUNWIND-NEXT: callq memcmp@PLT ; NOCOMPACTUNWIND-NEXT: testl %eax, %eax ; NOCOMPACTUNWIND-NEXT: sete %al -; NOCOMPACTUNWIND-NEXT: .LBB3_4: # %__go_ptr_strings_equal.exit +; NOCOMPACTUNWIND-NEXT: .LBB3_5: # %__go_ptr_strings_equal.exit ; NOCOMPACTUNWIND-NEXT: # kill: def $al killed $al killed $eax ; NOCOMPACTUNWIND-NEXT: popq %rcx ; NOCOMPACTUNWIND-NEXT: .cfi_def_cfa_offset 8 ; NOCOMPACTUNWIND-NEXT: retq -; NOCOMPACTUNWIND-NEXT: .LBB3_5: +; NOCOMPACTUNWIND-NEXT: .LBB3_6: ; NOCOMPACTUNWIND-NEXT: .cfi_def_cfa_offset 16 ; NOCOMPACTUNWIND-NEXT: xorl %eax, %eax ; NOCOMPACTUNWIND-NEXT: # kill: def $al killed $al killed $eax ; NOCOMPACTUNWIND-NEXT: popq %rcx ; NOCOMPACTUNWIND-NEXT: .cfi_def_cfa_offset 8 ; NOCOMPACTUNWIND-NEXT: retq -; NOCOMPACTUNWIND-NEXT: .LBB3_6: +; NOCOMPACTUNWIND-NEXT: .LBB3_7: ; NOCOMPACTUNWIND-NEXT: movl $8, %r10d ; NOCOMPACTUNWIND-NEXT: movl $0, %r11d ; NOCOMPACTUNWIND-NEXT: callq __morestack |