llvm-project/llvm/test/CodeGen/X86/urem-seteq-optsize.ll

Ignoring revisions in .git-blame-ignore-revs. Click here to bypass and see the normal blame view.

72 lines
2.1 KiB
LLVM
Raw Normal View History

; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
; RUN: llc -mtriple=i686-unknown-linux-gnu < %s | FileCheck %s --check-prefixes=CHECK,X86
; RUN: llc -mtriple=x86_64-unknown-linux-gnu < %s | FileCheck %s --check-prefixes=CHECK,X64
; On X86, division in expensive. BuildRemEqFold should therefore run even
; when optimizing for size. Only optimizing for minimum size retains a plain div.
define i32 @test_minsize(i32 %X) optsize minsize nounwind readnone {
; X86-LABEL: test_minsize:
; X86: # %bb.0:
; X86-NEXT: movl {{[0-9]+}}(%esp), %eax
; X86-NEXT: pushl $5
; X86-NEXT: popl %ecx
; X86-NEXT: xorl %edx, %edx
; X86-NEXT: divl %ecx
; X86-NEXT: testl %edx, %edx
; X86-NEXT: je .LBB0_1
; X86-NEXT: # %bb.2:
; X86-NEXT: pushl $-10
; X86-NEXT: popl %eax
; X86-NEXT: retl
; X86-NEXT: .LBB0_1:
; X86-NEXT: pushl $42
; X86-NEXT: popl %eax
; X86-NEXT: retl
;
; X64-LABEL: test_minsize:
; X64: # %bb.0:
; X64-NEXT: movl %edi, %eax
; X64-NEXT: pushq $5
; X64-NEXT: popq %rcx
; X64-NEXT: xorl %edx, %edx
; X64-NEXT: divl %ecx
; X64-NEXT: testl %edx, %edx
; X64-NEXT: pushq $42
; X64-NEXT: popq %rcx
; X64-NEXT: pushq $-10
; X64-NEXT: popq %rax
; X64-NEXT: cmovel %ecx, %eax
; X64-NEXT: retq
%rem = urem i32 %X, 5
%cmp = icmp eq i32 %rem, 0
%ret = select i1 %cmp, i32 42, i32 -10
ret i32 %ret
}
define i32 @test_optsize(i32 %X) optsize nounwind readnone {
; X86-LABEL: test_optsize:
; X86: # %bb.0:
[CodeGen] [SelectionDAG] More efficient code for X % C == 0 (UREM case) (try 3) Summary: I'm submitting a new revision since i don't understand how to reclaim/reopen/take over the existing one, D50222. There is no such action in "Add Action" menu... This implements an optimization described in Hacker's Delight 10-17: when `C` is constant, the result of `X % C == 0` can be computed more cheaply without actually calculating the remainder. The motivation is discussed here: https://bugs.llvm.org/show_bug.cgi?id=35479. This is a recommit, the original commit rL364563 was reverted in rL364568 because test-suite detected miscompile - the new comparison constant 'Q' was being computed incorrectly (we divided by `D0` instead of `D`). Original patch D50222 by @hermord (Dmytro Shynkevych) Notes: - In principle, it's possible to also handle the `X % C1 == C2` case, as discussed on bugzilla. This seems to require an extra branch on overflow, so I refrained from implementing this for now. - An explicit check for when the `REM` can be reduced to just its LHS is included: the `X % C` == 0 optimization breaks `test1` in `test/CodeGen/X86/jump_sign.ll` otherwise. I hadn't managed to find a better way to not generate worse output in this case. - The `test/CodeGen/X86/jump_sign.ll` regresses, and is being fixed by a followup patch D63390. Reviewers: RKSimon, craig.topper, spatel, hermord, xbolva00 Reviewed By: RKSimon, xbolva00 Subscribers: dexonsmith, kristina, xbolva00, javed.absar, llvm-commits, hermord Tags: #llvm Differential Revision: https://reviews.llvm.org/D63391 llvm-svn: 364600
2019-06-28 05:52:10 +08:00
; X86-NEXT: imull $-858993459, {{[0-9]+}}(%esp), %eax # imm = 0xCCCCCCCD
; X86-NEXT: cmpl $858993460, %eax # imm = 0x33333334
; X86-NEXT: movl $42, %eax
[CodeGen] [SelectionDAG] More efficient code for X % C == 0 (UREM case) (try 3) Summary: I'm submitting a new revision since i don't understand how to reclaim/reopen/take over the existing one, D50222. There is no such action in "Add Action" menu... This implements an optimization described in Hacker's Delight 10-17: when `C` is constant, the result of `X % C == 0` can be computed more cheaply without actually calculating the remainder. The motivation is discussed here: https://bugs.llvm.org/show_bug.cgi?id=35479. This is a recommit, the original commit rL364563 was reverted in rL364568 because test-suite detected miscompile - the new comparison constant 'Q' was being computed incorrectly (we divided by `D0` instead of `D`). Original patch D50222 by @hermord (Dmytro Shynkevych) Notes: - In principle, it's possible to also handle the `X % C1 == C2` case, as discussed on bugzilla. This seems to require an extra branch on overflow, so I refrained from implementing this for now. - An explicit check for when the `REM` can be reduced to just its LHS is included: the `X % C` == 0 optimization breaks `test1` in `test/CodeGen/X86/jump_sign.ll` otherwise. I hadn't managed to find a better way to not generate worse output in this case. - The `test/CodeGen/X86/jump_sign.ll` regresses, and is being fixed by a followup patch D63390. Reviewers: RKSimon, craig.topper, spatel, hermord, xbolva00 Reviewed By: RKSimon, xbolva00 Subscribers: dexonsmith, kristina, xbolva00, javed.absar, llvm-commits, hermord Tags: #llvm Differential Revision: https://reviews.llvm.org/D63391 llvm-svn: 364600
2019-06-28 05:52:10 +08:00
; X86-NEXT: jb .LBB1_2
; X86-NEXT: # %bb.1:
; X86-NEXT: movl $-10, %eax
; X86-NEXT: .LBB1_2:
; X86-NEXT: retl
;
; X64-LABEL: test_optsize:
; X64: # %bb.0:
[CodeGen] [SelectionDAG] More efficient code for X % C == 0 (UREM case) (try 3) Summary: I'm submitting a new revision since i don't understand how to reclaim/reopen/take over the existing one, D50222. There is no such action in "Add Action" menu... This implements an optimization described in Hacker's Delight 10-17: when `C` is constant, the result of `X % C == 0` can be computed more cheaply without actually calculating the remainder. The motivation is discussed here: https://bugs.llvm.org/show_bug.cgi?id=35479. This is a recommit, the original commit rL364563 was reverted in rL364568 because test-suite detected miscompile - the new comparison constant 'Q' was being computed incorrectly (we divided by `D0` instead of `D`). Original patch D50222 by @hermord (Dmytro Shynkevych) Notes: - In principle, it's possible to also handle the `X % C1 == C2` case, as discussed on bugzilla. This seems to require an extra branch on overflow, so I refrained from implementing this for now. - An explicit check for when the `REM` can be reduced to just its LHS is included: the `X % C` == 0 optimization breaks `test1` in `test/CodeGen/X86/jump_sign.ll` otherwise. I hadn't managed to find a better way to not generate worse output in this case. - The `test/CodeGen/X86/jump_sign.ll` regresses, and is being fixed by a followup patch D63390. Reviewers: RKSimon, craig.topper, spatel, hermord, xbolva00 Reviewed By: RKSimon, xbolva00 Subscribers: dexonsmith, kristina, xbolva00, javed.absar, llvm-commits, hermord Tags: #llvm Differential Revision: https://reviews.llvm.org/D63391 llvm-svn: 364600
2019-06-28 05:52:10 +08:00
; X64-NEXT: imull $-858993459, %edi, %eax # imm = 0xCCCCCCCD
; X64-NEXT: cmpl $858993460, %eax # imm = 0x33333334
; X64-NEXT: movl $42, %ecx
; X64-NEXT: movl $-10, %eax
[CodeGen] [SelectionDAG] More efficient code for X % C == 0 (UREM case) (try 3) Summary: I'm submitting a new revision since i don't understand how to reclaim/reopen/take over the existing one, D50222. There is no such action in "Add Action" menu... This implements an optimization described in Hacker's Delight 10-17: when `C` is constant, the result of `X % C == 0` can be computed more cheaply without actually calculating the remainder. The motivation is discussed here: https://bugs.llvm.org/show_bug.cgi?id=35479. This is a recommit, the original commit rL364563 was reverted in rL364568 because test-suite detected miscompile - the new comparison constant 'Q' was being computed incorrectly (we divided by `D0` instead of `D`). Original patch D50222 by @hermord (Dmytro Shynkevych) Notes: - In principle, it's possible to also handle the `X % C1 == C2` case, as discussed on bugzilla. This seems to require an extra branch on overflow, so I refrained from implementing this for now. - An explicit check for when the `REM` can be reduced to just its LHS is included: the `X % C` == 0 optimization breaks `test1` in `test/CodeGen/X86/jump_sign.ll` otherwise. I hadn't managed to find a better way to not generate worse output in this case. - The `test/CodeGen/X86/jump_sign.ll` regresses, and is being fixed by a followup patch D63390. Reviewers: RKSimon, craig.topper, spatel, hermord, xbolva00 Reviewed By: RKSimon, xbolva00 Subscribers: dexonsmith, kristina, xbolva00, javed.absar, llvm-commits, hermord Tags: #llvm Differential Revision: https://reviews.llvm.org/D63391 llvm-svn: 364600
2019-06-28 05:52:10 +08:00
; X64-NEXT: cmovbl %ecx, %eax
; X64-NEXT: retq
%rem = urem i32 %X, 5
%cmp = icmp eq i32 %rem, 0
%ret = select i1 %cmp, i32 42, i32 -10
ret i32 %ret
}