Re: basic_string::append() acting weird at release mode

• Previous message: Mark Davis [MSFT]: "Re: Applying a member function to a range of elements"
• In reply to: Igor Tandetnik: "Re: basic_string::append() acting weird at release mode"
• Messages sorted by: [ date ] [ thread ]

Date: Sat, 28 Feb 2004 01:57:00 GMT

As Igor mentions, you should turn off optimizations if you want to debug
your release build.

Debugging a release build is often useful because the debug and release
compilers behave differently (e.g. variables are initialised by the debug
compiler, but not release - a common problem).

Mark Davis (MSFT)

--
This posting is provided "AS IS" with no warranties, and confers no rights.
Use of included information and code is subject to the terms specified at 
http://www.microsoft.com/info/cpyright.htm.
--------------------
>From: "Igor Tandetnik" <itandetnik@mvps.org>
>References: <5fc41c7e.0402251647.175eee12@posting.google.com>
>Subject: Re: basic_string::append() acting weird at release mode
>Date: Wed, 25 Feb 2004 19:57:43 -0500
>Lines: 35
>X-Priority: 3
>X-MSMail-Priority: Normal
>X-Newsreader: Microsoft Outlook Express 6.00.2800.1158
>X-MIMEOLE: Produced By Microsoft MimeOLE V6.00.2800.1165
>Message-ID: <#mtGMOA$DHA.2664@TK2MSFTNGP09.phx.gbl>
>Newsgroups: microsoft.public.vc.stl
>NNTP-Posting-Host: 142.winstar.net 207.98.207.142
>Path: cpmsftngxa06.phx.gbl!TK2MSFTNGP08.phx.gbl!TK2MSFTNGP09.phx.gbl
>Xref: cpmsftngxa06.phx.gbl microsoft.public.vc.stl:22102
>X-Tomcat-NG: microsoft.public.vc.stl
>
>"ph0ngwh0ng" <ph0ngwh0ng@hotmail.com> wrote in message
>news:5fc41c7e.0402251647.175eee12@posting.google.com...
>> Stepping into the code with F10, when I step over tstrKey.append(...),
>> the step cursor jumps to the end of the function(the } ).  Pressing
>> F10 again sends the step cursor back to tstrkey.append(...).  I have
>> to press F10 9 times after this to have the step cursor going to
>> RegOpenKeyEx().  Also, pressing F11 instead reveals the same
>> behaviour, in addition that stepping into basic_string::append(const
>> _Elem *_Ptr) leads me in basic_string::append(const _Elem *_Ptr,
>> size_type _Count) instead.
>
>That's to be expected when stepping through Release build (why do you
>want to, anyway?) The optimizer can rearrange code in odd ways, so that
>the resulting machine code bears just a sligh resemblance to the
>original source. As a result, there's no good way to associate line
>numbers with machine instructions - that's why the current execution
>point jumps back and forth when single-stepping.
>
>> When I'm finally at the end of the function, the step cursor goes into
>>
>> (seccook.c)
>> void __declspec(naked) __fastcall __security_check_cookie(DWORD_PTR
>> cookie)
>> {
>
>That's the buffer overrun check introduced by /GS option. You asked for
>it - you got it.
>-- 
>With best wishes,
>    Igor Tandetnik
>
>"For every complex problem, there is a solution that is simple, neat,
>and wrong." H.L. Mencken
>
>
>

• Previous message: Mark Davis [MSFT]: "Re: Applying a member function to a range of elements"
• In reply to: Igor Tandetnik: "Re: basic_string::append() acting weird at release mode"
• Messages sorted by: [ date ] [ thread ]