Microsoft MVP성태의 닷넷 이야기
.NET Framework: 789. C# 컴파일 옵션 - Check for arithmetic overflow/underflow [링크 복사], [링크+제목 복사]
조회: 14775
글쓴 사람
정성태 (techsharer at outlook.com)
홈페이지
첨부 파일
 
(연관된 글이 3개 있습니다.)

C# 컴파일 옵션 - Check for arithmetic overflow/underflow

아래의 글 덕분에 처음으로 overflow/underflow에 대해 심각하게 생각해 보았습니다. ^^

오버플로우와 언더플로우
; https://www.sysnet.pe.kr/3/0/5019

위의 의견에 따르면, (예를 들어) MaxValue + 1일 때의 상황이 overflow라고 불리기도 하지만 (예를 들어) MinValue - 1일 때의 상황 역시 overflow입니다. 실제로 wiki 정의에서도 이것을 찾아볼 수 있습니다.

Integer overflow
; https://en.wikipedia.org/wiki/Integer_overflow

In computer programming, an integer overflow occurs when an arithmetic operation attempts to create a numeric value that is outside of the range that can be represented with a given number of bits - either larger than the maximum or lower than the minimum representable value.


그리고 옵션 명에도 나오지만 특히나 "arithmetic underflow"는,
Arithmetic underflow
; https://en.wikipedia.org/wiki/Arithmetic_underflow

The term arithmetic underflow (or "floating point underflow", or just "underflow") is a condition in a computer program where the result of a calculation is a number of smaller absolute value than the computer can actually represent in memory on its CPU.

Arithmetic underflow can occur when the true result of a floating point operation is smaller in magnitude (that is, closer to zero) than the smallest value representable as a normal floating point number in the target datatype.[1] Underflow can in part be regarded as negative overflow of the exponent of the floating point value. For example, if the exponent part can represent values from ?128 to 127, then a result with a value less than ?128 may cause underflow.


이렇게 부동 소수점에서 (0인 듯 0이 아닌 0 같을 정도로) 0에 가까운 값이지만 정밀도로 표현할 수 없어 그냥 0으로 설정해 버리는 경우를 의미합니다.




이건 그냥 제 생각이지만, 해당 옵션의 문자열을 지정한 마이크로소프트 측의 개발자는 여기서의 underflow를 "integer underflow"로 여겼을 가능성이 크고, 아래의 링크에서와 같이 많은 글들이 (암묵적으로) MinValue보다 값이 낮아졌을 때의 상황을 의미하는 단어로 쓴 것 같습니다.

CWE-191: Integer Underflow (Wrap or Wraparound)
; https://cwe.mitre.org/data/definitions/191.html

Java SE 8's New Compact Profiles and Integer APIs
; http://www.informit.com/articles/article.aspx?p=2216988

Overflow And Underflow of Data Types in Java 
; https://dzone.com/articles/overflow-and-underflow-dat

Integer Overflow/Underflow and Floating Point Imprecision.
; https://medium.com/@taabishm2/integer-overflow-underflow-and-floating-point-imprecision-6ba869a99033

Integer underflow
; https://scc.ustc.edu.cn/zlsc/sugon/intel/ssadiag_docs/pt_reference/references/sc_int_underflow.htm

재미있는 것은, MSIL 코드를 보면 빼기(Subtract) 연산에서 overflow라는 의미의 "Ovf" 접두사를 넣었다는 것입니다.

OpCodes.Sub_Ovf 필드
; https://docs.microsoft.com/en-us/dotnet/api/system.reflection.emit.opcodes.sub_ovf

IL 코드 설계자가 Add_Ovf와 함께 빼기에서도 (Sub_Udf가 아닌) Sub_Ovf를 사용했다는 것은 overflow의 엄밀한 의미를 알고 지킨 것 같습니다.

또한, "OpCodes.Sub_Ovf 필드" 도움말 내용을 보면,

This operation is performed on signed integers; for floating-point values, use Sub.

부동 소수점에 대해서는 ovf 성격의 연산을 지원하지 않고 정수형에 대해서만 지원합니다. 그리고 Sub 명령어의 도움말을 보면,

Floating-point overflow returns +inf (PositiveInfinity) or -inf (NegativeInfinity).

부동 소수점의 경우 overflow 예외를 발생시키지 않는다고 나오며 underflow는 언급조차 안 하고 있습니다.

즉, .NET에서는 부동 소수점에서의 overflow/underflow에 대한 예외는 발생하지 않습니다. 실제로 "Check for arithmetic overflow/underflow" 옵션을 켜고 부동 소수점 빼기/나누기 등을 해보면, 단순하게 sub, div IL 명령어로만 대체될 뿐 ovf 의미의 IL 코드가 사용되지 않습니다.




종합해 봤을 때, 프로젝트 속성 창의 "Check for arithmetic overflow/underflow"에서의 underflow 의미는 (MinValue - 1 상황에서 발생하는) "integer underflow"를 의미하는 것 같습니다. 일단, 이에 대해서는 다음의 이슈를 통해 마이크로소프트에 문의를 해두었으니,

Is "Check for arithmetic overflow/underflow" option name valid?  #3792 
; https://github.com/dotnet/project-system/issues/3792

추이를 지켜보고 향후 진도를 나가는 것이 좋겠다는 판단입니다.




[이 글에 대해서 여러분들과 의견을 공유하고 싶습니다. 틀리거나 미흡한 부분 또는 의문 사항이 있으시면 언제든 댓글 남겨주십시오.]

[연관 글]






[최초 등록일: ]
[최종 수정일: 7/13/2021]

Creative Commons License
이 저작물은 크리에이티브 커먼즈 코리아 저작자표시-비영리-변경금지 2.0 대한민국 라이센스에 따라 이용하실 수 있습니다.
by SeongTae Jeong, mailto:techsharer at outlook.com

비밀번호

댓글 작성자
 



2020-02-06 08시43분
비주얼 스튜디오 16.6에 "underflow" 문구가 삭제될 거라고 합니다. ^^

Remove "underflow" from /checked description
; https://github.com/dotnet/project-system/pull/5855
정성태
2021-03-18 12시53분
오늘 문득 확인을 하게 되었는데, 옵션 명이 "Check for arithmetic overflow"로 바뀌었습니다.
정성태

... 16  17  18  [19]  20  21  22  23  24  25  26  27  28  29  30  ...
NoWriterDateCnt.TitleFile(s)
13152정성태11/1/20225616.NET Framework: 2061. ASP.NET Core - DI로 추가한 클래스의 초기화 방법 [1]
13151정성태10/31/20225718C/C++: 161. Windows 11 환경에서 raw socket 테스트하는 방법파일 다운로드1
13150정성태10/30/20225769C/C++: 160. Visual Studio 2022로 빌드한 C++ 프로그램을 위한 다른 PC에서 실행하는 방법
13149정성태10/27/20225697오류 유형: 825. C# - CLR ETW 이벤트 수신이 GCHeapStats_V1/V2에 대해 안 되는 문제파일 다운로드1
13148정성태10/26/20225702오류 유형: 824. msbuild 에러 - error NETSDK1005: Assets file '...\project.assets.json' doesn't have a target for 'net5.0'. Ensure that restore has run and that you have included 'net5.0' in the TargetFramew
13147정성태10/25/20224810오류 유형: 823. Visual Studio 2022 - Unable to attach to CoreCLR. The debugger's protocol is incompatible with the debuggee.
13146정성태10/24/20225647.NET Framework: 2060. C# - Java의 Xmx와 유사한 힙 메모리 최댓값 제어 옵션 HeapHardLimit
13145정성태10/21/20225924오류 유형: 822. db2 - Password validation for user db2inst1 failed with rc = -2146500508
13144정성태10/20/20225760.NET Framework: 2059. ClrMD를 이용해 윈도우 환경의 메모리 덤프로부터 닷넷 모듈을 추출하는 방법파일 다운로드1
13143정성태10/19/20226279오류 유형: 821. windbg/sos - Error code - 0x000021BE
13142정성태10/18/20225061도서: 시작하세요! C# 12 프로그래밍
13141정성태10/17/20226768.NET Framework: 2058. [in,out] 배열을 C#에서 C/C++로 넘기는 방법 - 세 번째 이야기파일 다운로드1
13140정성태10/11/20226129C/C++: 159. C/C++ - 리눅스 환경에서 u16string 문자열을 출력하는 방법 [2]
13139정성태10/9/20225965.NET Framework: 2057. 리눅스 환경의 .NET Core 3/5+ 메모리 덤프로부터 모든 닷넷 모듈을 추출하는 방법파일 다운로드1
13138정성태10/8/20227266.NET Framework: 2056. C# - await 비동기 호출을 기대한 메서드가 동기로 호출되었을 때의 부작용 [1]
13137정성태10/8/20225636.NET Framework: 2055. 리눅스 환경의 .NET Core 3/5+ 메모리 덤프로부터 닷넷 모듈을 추출하는 방법
13136정성태10/7/20226205.NET Framework: 2054. .NET Core/5+ SDK 설치 없이 dotnet-dump 사용하는 방법
13135정성태10/5/20226448.NET Framework: 2053. 리눅스 환경의 .NET Core 3/5+ 메모리 덤프를 분석하는 방법 - 두 번째 이야기
13134정성태10/4/20225184오류 유형: 820. There is a problem with AMD Radeon RX 5600 XT device. For more information, search for 'graphics device driver error code 31'
13133정성태10/4/20225498Windows: 211. Windows - (commit이 아닌) reserved 메모리 사용량 확인 방법 [1]
13132정성태10/3/20225369스크립트: 42. 파이썬 - latexify-py 패키지 소개 - 함수를 mathjax 식으로 표현
13131정성태10/3/20228064.NET Framework: 2052. C# - Windows Forms의 데이터 바인딩 지원(DataBinding, DataSource) [2]파일 다운로드1
13130정성태9/28/20225146.NET Framework: 2051. .NET Core/5+ - 에러 로깅을 위한 Middleware가 동작하지 않는 경우파일 다운로드1
13129정성태9/27/20225444.NET Framework: 2050. .NET Core를 IIS에서 호스팅하는 경우 .NET Framework CLR이 함께 로드되는 환경
13128정성태9/23/20228014C/C++: 158. Visual C++ - IDL 구문 중 "unsigned long"을 인식하지 못하는 #import파일 다운로드1
13127정성태9/22/20226451Windows: 210. WSL에 systemd 도입
... 16  17  18  [19]  20  21  22  23  24  25  26  27  28  29  30  ...