|
You should post your note at the very tutorial page, or, at least, post here a link of such tutorial.
THESE PEOPLE REALLY BOTHER ME!! How can they know what you should do without knowing what you want done?!?!
-- C++ FQA Lite
|
|
|
|
|
Hi,
i am confused about the return value of the clock_getres functions. I am running on custom hw with armv4 cpu and kernel version 2.6.26.
I get values 7812500 (nanoseconds) for CLOCK_REALTIME and CLOCK_MONOTONIC.
Can i trust that the resolution of gettimeofday (uses CLOCK_REALTIME?) calls is ~7.8ms? Does the CPU clock speed affect this value?
Regards
ele lont
|
|
|
|
|
Most time functions have a resolution that is about on this order of magnitude. I've read somewhere that on most current systems the resolution is about 10-20ms, but 7.8ms sounds entirely reasonable too. And no, CPU speed doesn't matter.
Note that there are also functions for high-res time measurement. There is plenty of info about it on the web
GOTOs are a bit like wire coat hangers: they tend to breed in the darkness, such that where there once were few, eventually there are many, and the program's architecture collapses beneath them. (Fran Poretto)
|
|
|
|
|
You should. However you could also make a test.
THESE PEOPLE REALLY BOTHER ME!! How can they know what you should do without knowing what you want done?!?!
-- C++ FQA Lite
|
|
|
|
|
|
|
|
<pre lang="text">
Could someone please help me to find my bug.
This is C++ code "running" on Arduino.
It compiles and runs ,but...
I am using a struct to define / collect data which belong together.
The bug is when I enter a character array (using pointer) into the structure variable, also a character array.
The actual variable I am having problem with is ctextPrompts. I am sure there will be more, but this is the first bug.
It is entered properly - verified by test code immediately after - see TOK here note.
However, when I initialize next member of the struture cTextEntry - it wipes out the ctextPrompts.
Initializing next structure member ctextEmulate puts ctextEmulate characters into ctextPrompts.
This is my first attempt to utilize pointers and structure and I MUST have done something really stupid to cause this bug.
I am using LCD to debug the code.
Since the rest of the code works as expected I am pretty sure it is not a "memory" issue, I do verify the size of available RAM on initialization.
I doubt it is a Arduino compiler issue, just plain wrong usage of pointers which I just do not see.
I do apologize for the long code, but I feel it would help to see the debbuging attempts I have made so far.
Thanks for your help.
Cheers
Vaclav
PS While testing I am using both pointer (*) and indexing [0] with same results
</pre>char
*ctextEntry[] ={
" ", " "," " };
// emulating data
char *ctextEmulate[] = { // emulate entries
"14260123 ", // start frequency
"14360000 ", // end frequency
"10000 ", // step frequency
"1 ", // step speed
"1 ", // 0 single shot 1 repeat
"Option" //
}; // temporary not completed
char *ctextPrompts[] ={
"Start frequency ","End frequency ","Step frequency ","Step speed ","Single / Repeat","Optional "};
typedef struct DataRecordTAG
{
char *cPrompts[]; // prompt text
char *cEmulate[]; // emulate input
char *cEntry[]; // response
char cTEST;
int iIndex;
}
DataRecord;
DataRecord Record[DATA_STRUCT]; // variable
int igRecordIndex = 0; // global access to records
.......
void InitializeRecords(void)
{
// changed to global igRecordIndex igRecordIndex
// igRecordIndex
// initialize prompts and some test entries
int iPromptIndex = 0;
for(igRecordIndex = 0; igRecordIndex != DATA_STRUCT ; igRecordIndex++) // fence post error Optional missing
{
Record[igRecordIndex].iIndex = igRecordIndex; //test index not really needed
/* TOK
#ifdef FLOW_LCD
lcd.clear();
lcd.print("Record[igRecordIndex].iIndex"); //too long
lcd.setCursor(0,1);
lcd.print(Record[igRecordIndex].iIndex); // too long
delay(DELAY_LCD);
#endif
*/
#ifdef FLOW_LCD
TOK verified
lcd.clear();
lcd.print("cPrompts empty ");
lcd.setCursor(0,1);
lcd.print(*Record[0].cPrompts);
delay(5*DELAY_LCD);
#endif
// prompt bug
*Record[igRecordIndex].cPrompts = ctextPrompts[igRecordIndex];
//ReadRecords();
#ifdef FLOW_LCD
TOK here
lcd.clear();
lcd.print("cPrompts entered");
lcd.setCursor(0,1);
lcd.print(*Record[0].cPrompts);
delay(5*DELAY_LCD);
#endif
/*
// BUG writes over previous record how ??
for(int i = 0; i < DATA_STRUCT; i++)
{
#ifdef FLOW_LCD
lcd.clear();
lcd.print("igRecordIndex");
lcd.setCursor(0,1);
lcd.print(igRecordIndex);
delay(DELAY_LCD);
lcd.clear();
lcd.print("Record[i].cPrompts");
lcd.setCursor(0,1);
lcd.print(*Record[i].cPrompts);
delay(5*DELAY_LCD);
#endif
}
// BUG writes over previous record how ??
*/
/* TOK
#ifdef FLOW_LCD
lcd.clear();
lcd.print("Record[igRecordIndex].cPrompts"); //too long
lcd.setCursor(0,1);
lcd.print(*Record[igRecordIndex].cPrompts); // too long
delay(DELAY_LCD);
#endif
*/
Record[igRecordIndex].cEntry[0] = ctextEntry[igRecordIndex]; // [DIGITS]
#ifdef FLOW_LCD
BUG now the cPrompts is empty
lcd.clear();
lcd.print("cText entered");
lcd.setCursor(0,1);
lcd.print(*Record[0].cPrompts);
delay(5*DELAY_LCD);
#endif
*Record[igRecordIndex].cEmulate = ctextEmulate[igRecordIndex]; // emulate input
#ifdef FLOW_LCD
BUG the cPrompts contains ctextEmulate - index 0
lcd.clear();
lcd.print("ctextEmulate entered");
lcd.setCursor(0,1);
lcd.print(*Record[0].cPrompts);
delay(5*DELAY_LCD);
#endif
/* TOK
#ifdef FLOW_LCD
lcd.clear();
lcd.print("Emulate "); //too long
lcd.setCursor(0,1);
lcd.print(Record[igRecordIndex].cEmulate); // too long
delay(DELAY_LCD);
#endif
*/
}
igRecordIndex = 0;
ReadRecords();
for(;;);
/* TOK
#ifdef FLOW_SERIAL
Serial.println("Prompts initialized DONE ");
#endif
#ifdef FLOW_LCD
lcd.clear();
lcd.print("Prompts initialized DONE ");
delay(DELAY_LCD);
#endif
*/
}
|
|
|
|
|
Quote: *Record[igRecordIndex].cPrompts = ctextPrompts[igRecordIndex];
What are you doing with this line?
THESE PEOPLE REALLY BOTHER ME!! How can they know what you should do without knowing what you want done?!?!
-- C++ FQA Lite
|
|
|
|
|
Initializing the struct member cPrompts (char array) from ctextPrompts (char array).
According to next verification code it works.
But than things go crazy as described.
|
|
|
|
|
There maybe be several bugs, but I believe you should remove the []:
ypedef struct DataRecordTAG
{
char *cPrompts;
char *cEmulate;
char *cEntry;
char cTEST;
int iIndex;
}
|
|
|
|
|
I think what I am doing is assigning pointers in the structure to point to an array.
Is that legal / right?
Or do I need to physically copy the array to the structure?
That will probably wipe out the memory for good.
I am sure lost in this .
Maybe I need to not use the structure at all.
|
|
|
|
|
Now that does not "copy" the entire array into the structure.
But I think I need to work on making the structure pointer equal the array pointer.
|
|
|
|
|
I did remove the array and so far it is working. But I need to check the rest of the code .
Thanks
Vaclav
So basically if read my code in English = assigning a char pointer to point to pointer to array of characters.
What I did was wrongly assign a char array pointer to point to array of characters.
Thanks to both of you guys to nudge me the correct this.
Vaclav
-- modified 11-Sep-14 19:03pm.
|
|
|
|
|
Hi
I want to call a ASP.Net (.asmx) from a MFC 2008 project.
My PC has Windows 7, but I want to run my program from Windows XP and 7 (I need to support Winsows XP)
(Some years ago I did this in VC++2003 but add web refrence is not included in VC++2008)
Any suggetstion?
Regards
www.logicsims.ir
|
|
|
|
|
Hello. How do I execute certain batch files and shell scripts along with my project? Are these settings specified in Pre Build and Post Build events in visual studio? Thanks for any input.
This world is going to explode due to international politics, SOON.
|
|
|
|
|
AmbiguousName wrote: Are these settings specified in Pre Build and Post Build events in visual studio? Thanks for any input.
Yes, depending on which part of the build process you want them.
|
|
|
|
|
All right let's suppose I want Pre Build. Now how do I tell Visual Studio to use xxxx.sh and xxxx.bat files?
This world is going to explode due to international politics, SOON.
|
|
|
|
|
Go to the Project Properties, click on Build Events, Pre-build, and add the details as required.
|
|
|
|
|
Since I have never done this, so bear with me. Do I have to add some kind of path? Suppose a xxxx.bat file is at desktop, would the entry be like this: call ..\..\Desktop\xxx.bat ?
This world is going to explode due to international politics, SOON.
|
|
|
|
|
Yes, the same as you would do in any application or command shell. However it probably makes more sense to add the batch files into your project directory so they stay with the project. That also allows you to manage them within your source control system as well.
|
|
|
|
|
<blockquote class="quote"><div class="op">Quote:</div>float determinant1(float **a,int n)
{
int i,j,j1,j2 ;
float det = 0 ;
float **m = NULL ;
if (n < 1) { }
else if (n == 1) {
det = a[0][0] ;
}
else if (n == 2) {
det = a[0][0] * a[1][1] - a[1][0] * a[0][1] ;
}
else {
det = 0 ;
for (j1 = 0 ; j1 < n ; j1++) {
m = (float **) malloc((n-1)* sizeof(float *)) ;
for (i = 0 ; i < n-1 ; i++)
m[i] = (float *) malloc((n-1)* sizeof(float)) ;
for (i = 1 ; i < n ; i++) {
j2 = 0 ;
for (j = 0 ; j < n ; j++) {
if (j == j1) continue ;
m[i-1][j2] = a[i][j] ;
j2++ ;
}
}
det += pow(-1.0,1.0 + j1 + 1.0) * a[0][j1] * determinant1(m,n-1) ;
printf("\n%f",det);
for (i = 0;i<n-1;i++) free(m[i]) ;
free(m) ;
}
}
return(det) ;
}
</blockquote>
|
|
|
|
|
Check your compiler documentation on how to increase the stacksize. Very likely your program dies due to stack overflow. This tends to happen when you have very deep levels of function calls, and 100 levels of recursion qualifies as very deep!
Alternately, test your program with smaller matrices, say 10x10 rather than 100x100. If it works, it's a stack overflow issue.
To prevent stack overflow, the only safe method is to avoid recursion, i. e. you should transform your program into one that doesn't require recursion. Otherwise you need to set a maximum recursion level and a set a stacksize that is sufficient to run the program with the maximum defined recursion level; you should exit the program with an error message, if the input would lead to a recursion level exceeding the predefined maximum.
GOTOs are a bit like wire coat hangers: they tend to breed in the darkness, such that where there once were few, eventually there are many, and the program's architecture collapses beneath them. (Fran Poretto)
|
|
|
|
|
thank you for ur suggestion but can u kindly tell me how to set the maximum recrusion size or else the without recrusion process..
|
|
|
|
|
What I meant is, check programatically the depth of your recursion. E. g. define a static counter within your recursive function, and always compare it to the maximum allowed value, like this:
const int max_recursion_level = 20;
int foo(int n) {
int result = 0;
static int recursion_level = 0;
++recursion_level;
if (recursion_level > max_recursion_level)
throw("recursion level exceeded!");
if (n<=0)
result = 0;
else if (n==1)
result = 1;
else
result = foo(n-1) + foo(n-2);
--recursion_level;
return result;
}
As for calculating the determinant without recursion, I am sure there are plenty of algorithms on the web if you search for it. I?ve found one in the first response to this question: http://cboard.cprogramming.com/cplusplus-programming/30001-determinant-calculation.html[^]
GOTOs are a bit like wire coat hangers: they tend to breed in the darkness, such that where there once were few, eventually there are many, and the program's architecture collapses beneath them. (Fran Poretto)
|
|
|
|
General 
News 
Suggestion 
Question 
Bug 
Answer 
Joke 
Praise 
Rant 
Admin 
Submit an article or tip
