|
|
Expecting good explanation
|
|
|
|
|
|
The size of object of a class that contains int variable is 4bytes then why the size of class 1byte is not considered along with 4bytes?.
|
|
|
|
|
I don't quite understand what you are asking? But I will make a guess. You are saying that a class containing since int will have objects of size 4 bytes. Then why does empty class takes 1 byte instead of 4 bytes? Because 1 byte is enough for the purpose of differentiating two objects of an empty class. So why use 4 bytes and waste 3 bytes per object?
-Saurabh
|
|
|
|
|
My previous question is not so clear, let me clear my doubt here
class A
{
};
class B
{
int k;
};
int main()
{
A a;
int i = sizeof(a); -----> the value of i is 1byte
B b;
int j = sizeof(b); -----> the value of j is 4bytes
return 0;
}
why the value of j is 4bytes why not 5 bytes[size of int variable(4bytes) + size of class(1byte))
|
|
|
|
|
class objects have to have non zero memory size.
In case of class A it is empty since compiler cant determinne the size it will allocate minimum allocation unit i.e. 1 byte for its object.
In case of class B it has int variable so compiler knows its size and hence will allocate appropriate size i.e 4 bytes for int for its object.
Regards,
Sandip.
|
|
|
|
|
Because class B now has data, and doesn't need to insert a byte to maintain memory integrity.
|
|
|
|
|
|
Kwanalouie wrote: The problem I have is that when the cursor is at the top of the client window and I click the right mouse button, the value of point.y is around 72 instead of zero. When I use this same code in the OnLeftMouseDown routine, the value of point.y at the top of the Client window is zero.
Because OnContextMenu point parameter is in screen co-ordinates and your right click handler point is in client co-ordinates so call ClientToScreen/MapWindowPoints to convert to screen co-ordinates. That's why it's zero in right click handler and 72 in context menu handler.
From MSDN for OnContextMenu point parameter
Position of the cursor, in screen coordinates, at the time of the mouse click.
From MSDN for OnLButtonDown point parameter
Specifies the x- and y-coordinate of the cursor. These coordinates are always relative to the upper-left corner of the window.
Nibu thomas
Microsoft MVP for VC++
Code must be written to be read, not by the compiler, but by another human being.
Programming Blog: http://nibuthomas.wordpress.com
modified on Thursday, June 5, 2008 11:52 PM
|
|
|
|
|
Thanks a lot. That did the trick!
|
|
|
|
|
Which one and can it be done taking VS 6.0 C++ code and bringing it into one of the new VS 2xxx versions? We would like not to recode at this time and just import/rebuild under the newer environments.
Any problems with doing this?
thanks
|
|
|
|
|
Kyle P wrote: ...can it be done taking VS 6.0 C++ code and bringing it into one of the new VS 2xxx versions?
Yes.
Kyle P wrote: We would like not to recode at this time and just import/rebuild under the newer environments.
Highly unlikely. Unless you already had some real clean (i.e., compliant) C++ code, there'll be code changes.
"Love people and use things, not love things and use people." - Unknown
"To have a respect for ourselves guides our morals; to have deference for others governs our manners." - Laurence Sterne
|
|
|
|
|
Is there a guide of what to look at first and common things to change? Any documentation form MS on this or other guides?
thanks again
|
|
|
|
|
The most important issues deals with template and partial specialization.
Pay special attention to the use of the typename keyword, and to the way parameter lists are declared while reusing template declarations, since some scope rule are different.
2 bugs found.
> recompile ...
65534 bugs found.
|
|
|
|
|
I suspect what you'll likely run into will be Unicode and deprecation issues.
"Love people and use things, not love things and use people." - Unknown
"To have a respect for ourselves guides our morals; to have deference for others governs our manners." - Laurence Sterne
|
|
|
|
|
Hi, I'm using C++ with MFC and VC++ 2005.
My problem is this:
I have two different projects, one dependent on the other. The first project compiles to produce a static lib. It also includes a namespace, let's call it "NAM" for the purpose of this question. The second project then includes the headers from the first project, and the static lib in its settings.
This all seems fine, and the first project (the lib) compiles with no problems. However when I compile the second project, I receive linker errors for all of the namespace functions called in the first project's single .cpp file. So for instance, in project 1, I might have the function:
NAM::Find_String(void)
Which compiles just fine when I compile that project alone. But in the second project, I get this weird linker error that says:
error LNK2001: unresolved external symbol _Find_String
What am I doing wrong? It's like somehow the second project doesn't see the 'namespace' declared in the first project even though all the headers (there are only 3) are included and the lib is linked correctly.
KR
|
|
|
|
|
Please check whether you have properly included the staticlib in its settings. Second thing, you have to specify the path in which the static liv is located in Project->Settings->Link->Input. Also please check whether you have included the namespace of the first project in the second one.
akt
|
|
|
|
|
Hi there.
I've been messing around with programming Windows threads in Visual C++ 2008 and I just recently noticed how much CPU my computer uses when it runs only ONE thread.
For a dual core processor running at 1.86Ghz each, on task manager it shows roughly 50% CPU usage when ONE EMPTY thread is running in a loop. 
Can anyone tell me WHY on earth it's using so much CPU and how to solve this problem?
Is it due to how I've created and called the thread? the current code I have got is:
hThread = (void *)_beginthreadex(0, 0, &TestThread, 0, 0, &uThread);
where "TestThread" is the thread function, as shown below:
unsigned _stdcall TestThread(void *)
{
while(true);
_endthreadex(0);
return 0;
}
thanks in advance,
OCrowley.
|
|
|
|
|
OCrowley wrote: For a dual core processor running at 1.86Ghz each, on task manager it shows roughly 50% CPU usage when ONE EMPTY thread is running in a loop. [WTF]
Your system contains a lot of other system processes and services running in background which ofcourse will grab the cpu.
If you want to know the time used by your process, then take perfmon.exe and add a watch on "Processor Time" on your test application. Look here for how to use it. - http://www.windowsnetworking.com/articles_tutorials/Windows_2003_Performance_Monitor.html[^]
I thought you were asking about the processor time utilized by your process. Just ignore my reply.
Regards,
Jijo.
_____________________________________________________
http://weseetips.com[ ^] Visual C++ tips and tricks. Updated daily.
modified on Friday, June 6, 2008 1:18 AM
|
|
|
|
|
Misunderstood his question.
Regards,
Jijo.
_____________________________________________________
http://weseetips.com[ ^] Visual C++ tips and tricks. Updated daily.
|
|
|
|
|
Your thread is an infinite loop, so it's probably eating up all the time there. I don't see where you end your thread, as it can't end itself.
You could put a
Sleep(500); in your thread, and you'll see the usage drop.
In reality a useful thread would probably be performing some function and then suspending itself until it's time to do that function again. While it's working, it'll use whatever CPU time it needs to complete the task.
Hope that helps.
Karl - WK5M
PP-ASEL-IA (N43CS)
PGP Key: 0xDB02E193
PGP Key Fingerprint: 8F06 5A2E 2735 892B 821C 871A 0411 94EA DB02 E193
|
|
|
|
|
Your thread is using its entire quanta/time-slice. As the previous poster said, put in a sleep or, even better, create a global event handle and wait on it, exiting when it becomes signaled.
Anyone who thinks he has a better idea of what's good for people than people do is a swine.
- P.J. O'Rourke
|
|
|
|
|
Just to add to the previous two answers: the only reason it is not showing 100% CPU time, is that it is dual core, and only one core is assigned per thread.
|
|
|
|
|
Hi again.
I've included the sleep functions in all threads I'm using and that's done the job! Back to 0-1% again! 
Thanks all for your great help!
"create a global event handle and wait on it, exiting when it becomes signaled."
I'll look into that as well. Thanks very much.
OCrowley.
|
|
|
|
General 
News 
Suggestion 
Question 
Bug 
Answer 
Joke 
Praise 
Rant 
Admin 
Submit an article or tip
