There was talk about this some months back on this list. An article
about mem leaks was mentioned too.
http://www.fifthgate.org/articles/palm_mem_leaks.html
[EMAIL PROTECTED] wrote:
Hi,
i am allocating a large ammount of memory in my application.
will it be de-allocated automatically when i
after line MemHandleFree(fldHnd); try adding
FldSetTextHandle((FieldPtr)FrmGetObjectPtr(frmPtr,
objectIndex),NULL);
On 1/10/07, Ron [EMAIL PROTECTED] wrote:
LionScribe thank you for your reply. I have tried to do as you say but I get a
SysFatalAlert as follows:
---
TestGen
Thank you for your reply Dmitry. It got rid of the non-word-align error, but I
am getting a different error:
--
TestGen (unknown version) called SysFatalAlert with the message: MemoryMgr.c,
Line:4384, Free handle.
--
Thanks
--
For information on using the PalmSource
Didn't test, but that may be from freeing the locked handle from a non-editable
field.
By the way, why do you check for objectIndex, you already have the index.
LionScribe
--
For information on using the PalmSource Developer Forums, or to unsubscribe,
please see
Palm OS Reference states: FrmRemoveObject does not free memory allocated to a
string in an editable field.
Therefore you will first have to call FldGetTextHandle if a valid handle is
returned call MemHandleFree on all fields.
In addition I suggest removing all objects in reverse, thus lomiting
Don't know what's happening, but Yanoff is adding all crazy symbols to my
previous post. So please forgive.
LionScribe
--
For information on using the PalmSource Developer Forums, or to unsubscribe,
please see http://www.palmos.com/dev/support/forums/
LionScribe thank you for your reply. I have tried to do as you say but I get a
SysFatalAlert as follows:
---
TestGen (unknown version) called SysFatalAlert with the message: MemoryMgr.c,
Line:4372, Non-word-aligned handle.
---
Here is the code:
Thanks a lot cbruner
It is done and working very well
Regards,
Micky
-Original Message-
From: [EMAIL PROTECTED]
[mailto:[EMAIL PROTECTED] On Behalf Of cbruner
Sent: Thursday, July 27, 2006 11:45 PM
To: Palm Developer Forum
Subject: Re: Memory Leak
Helmut A. Bender wrote:
Am Donnerstag
At 04:16 AM 7/28/2006, you wrote:
Subject: Re: Memory Leak
From: cbruner [EMAIL PROTECTED]
Date: Thu, 27 Jul 2006 14:15:22 -0400
X-Message-Number: 11
Helmut A. Bender wrote:
Am Donnerstag, 27. Juli 2006 15:40 schrieb Micky MeNeZeS:
I have got following code.
[...]
It seems you don't
Am Donnerstag, 27. Juli 2006 15:40 schrieb Micky MeNeZeS:
I have got following code.
[...]
It seems you don't initialize the buffer you create. So the check in
StringCopy
if (StrLen(aBuffer) == 0)
won't work, since StrLen will get a random value. Set the first byte of the
buffer to
Helmut A. Bender wrote:
Am Donnerstag, 27. Juli 2006 15:40 schrieb Micky MeNeZeS:
I have got following code.
[...]
It seems you don't initialize the buffer you create. So the check in
StringCopy
if (StrLen(aBuffer) == 0)
won't work, since StrLen will get a random
Is it possible to develop an application which will exactly work as a simulator
also catch memory leaks (for 68K and ARM) and other bugs.
Thanks
John
--
For information on using the PalmSource Developer Forums, or to unsubscribe,
please see http://www.palmos.com/dev/support/forums/
the current public version will not. but if there is interest, i can
make a version to do that.
On 6/13/06, [EMAIL PROTECTED] [EMAIL PROTECTED] wrote:
Thanks for the reply, but now I want to know whether UDMH will pin point where
the leaks are, just like in IBM's Rational Purify for Windows.
Dmitry Grinberg schrieb:
the current public version will not. but if there is interest, i can
make a version to do that.
Yes, of course. Especially for exisiting apps it would be a good way to
check if and where memory leaks are. Since the Simulator does not report
that as good as the
but! this will NOT run on the simulator. only a real device with an
arm processor.
On 6/14/06, Thomas Damme [EMAIL PROTECTED] wrote:
Dmitry Grinberg schrieb:
the current public version will not. but if there is interest, i can
make a version to do that.
Yes, of course. Especially for
not too hard to make.just trap all allocations and deallocations. UDMH
does this, [debug version also reports on any leaks, release just
frees them for the app that forgot]
On 6/13/06, [EMAIL PROTECTED] [EMAIL PROTECTED] wrote:
Hi All,
I would like to know whether there is any commercial
Thanks for the reply, but now I want to know whether UDMH will pin point where
the leaks are, just like in IBM's Rational Purify for Windows. This is because
I know that my application is leaking lots of memory and so now I want to know
from where.
Thanks
John
--
For information on using
At 04:15 AM 5/28/2006, you wrote:
Subject: Re: Memory Leak
From: Carlos Gonzalez [EMAIL PROTECTED]
Date: Sat, 27 May 2006 18:31:44 -
X-Message-Number: 10
Thanks everyone for all your inputs. Here is the final code that
works fine with no more memory leak. Thanks againn everyone.
textH
Carlos, Try using with FldSetTextPtr/FldGetTextPtr as per your requirement instead of usingFldSetTextHandle. Then there won't be any memory leaks.Regards, HarshaCarlos Gonzalez [EMAIL PROTECTED] wrote: Eduardo,Here is part of my code and it is the only place where I use MemHandleLock and
At 04:16 AM 5/27/2006, you wrote:
Subject: RE: Memory Leak
From: Stefan Mohr [EMAIL PROTECTED]
Date: Fri, 26 May 2006 11:33:23 -0700
X-Message-Number: 22
You can try debugging your code using the Simulator and keep an eye on
the dynamically allocated variables, but you may find there's actually
At 04:16 AM 5/27/2006, you wrote:
Subject: Re: Memory Leak
From: Carlos Gonzalez [EMAIL PROTECTED]
Date: Fri, 26 May 2006 20:05:45 -
Here is part of my code and it is the only place where I use
MemHandleLock and unlock.. that is it.
Boolean CalculateRisk()
{
ControlType* ctl;
Boolean
Thanks everyone for all your inputs. Here is the final code that works fine
with no more memory leak. Thanks againn everyone.
textH = FldGetTextHandle(GetObjectPtr(MainFieldRisk));
if (!textH)
{
textH = MemHandleNew(8);
}
str = MemHandleLock(textH);
doubleToStr(str, PI, 2);
You can try debugging your code using the Simulator and keep an eye on
the dynamically allocated variables, but you may find there's actually
no problem. Our emulator quite frequently says we've got 40-odd memory
leaks when we shut it down but it's been thoroughly investigated and no
memory leaks
thanks, the problem is that my debuger fails when try to launch, this started
happening just today... I have no idea why.
Also when I run my application in my palm, it gives me this error:
MemoryMgr.C, Line 4384, Free Handle .
Carlos
--
For information on using the PalmSource Developer Forums,
Carlos.
Look at the POSE logs, theres a plenty info about the memory leaks.
Stefan.
Theres always a reason why POSE complains about a memory leaks, there are
harmless on the behalf of the app, but means that some data is left in
memory.
When I build my first app, I had the same idea that
Carlos.
Could you try to debug this a little more, so can help you better. This
could be generated by a lot of different things, i.e. for overFreeing a
memory chunk, or an iexisten one, or you could be trying to free a MemHandle
used to store the result of DmNewRecord() after calling
Eduardo,
Here is part of my code and it is the only place where I use MemHandleLock and
unlock.. that is it.
Boolean CalculateRisk()
{
ControlType* ctl;
Boolean handled = true;
MemHandle textH;
char*str;
double PI;
...
PI += -1.236215274 - 2.598564382 - 2.119863737 +
Carlos.
You should first call MemHandleNew(size) where size should be big enougth to
store the result of doubleToStr(), this may be also the problem of the
memory leak you talked before, because you had a handle that point to a
random address of memory that is not allocated for your use, then
]
Newsgroups: palm-dev-forum
To: Palm Developer Forum palm-dev-forum@news.palmos.com
Sent: Friday, May 26, 2006 2:44 PM
Subject: Re: Memory Leak
Carlos.
Could you try to debug this a little more, so can help you better. This
could be generated by a lot of different things, i.e. for overFreeing
Eduardo thanks, this is my original code but, still giving me the same error..
It alway gives the erro when I run it twice the function... the second time
gives me this error:.. MemoryMgr.c, Line 4384, Free Handle...
...
PI += -1.236215274 - 2.598564382 - 2.119863737 ;
PI = exp(PI);
PI = (1.0
Maybe, if you are using C++, this helps:
I've written some classes that encapsulate the MemHandle locking and
unlocking and work as smart pointers. I've attached them to this mail.
Regards,
Juergen
Del Ventruella schrieb:
In my first (and only) experience, the memory leaks were issues that
Subject: Re: Memory Leak
Eduardo thanks, this is my original code but, still giving me the same
error.. It alway gives the erro when I run it twice the function... the
second time gives me this error:.. MemoryMgr.c, Line 4384, Free Handle...
...
PI += -1.236215274 - 2.598564382 - 2.119863737 ;
PI = exp
Carlos.
Here's the big deal, textH should exist as long as your text field so textH
has to be a global or an arg to pass to your functions, so I'm guessing that
you run this funtion on frmOpenEvent or on ctlSelectEvent (i.e. button tap),
so you have to deal with it when you get an appStopEvent
yep, I have seen this before. Are you doing FrmEraseForm and FrmDeleteForm
before switching to the new form?
This is how I implement form transitions in general:
--
fP = FrmGetActiveForm();
if ( fP ) {
FrmEraseForm( fP );
FrmDeleteForm( fP );
}
FrmGotoForm( formID
Hi,
i m new to PODS can u tell me any way how to Find the Memory Leak problem
With PODs
--
For information on using the PalmSource Developer Forums, or to unsubscribe,
please see http://www.palmos.com/dev/support/forums/
Jeetu schrieb:
Hi,
i m new to PODS can u tell me any way how to Find the Memory Leak problem
With PODs
PalmOS Emulator will tell you if your program leaks memory. Run it in
Emulator and leave your program, then Emulator will tell you if your
program didn't free any memory chunks.
Jürgen
Jeetu wrote:
i m new to PODS can u tell me any way how to Find the Memory Leak problem
With PODs
If you can run your program under the Emulator (NOT the simulator!!!),
the Emulator will detect memory leaks when the application exits.
If you look in the Emulator's log file, it will show the
Thanx for ur Kind attention,
Here is what i m exactly doing,
I want to fill my ListBox with some values from DB,
so i have created a dynamic array that will contain the values as per the no.
of records in DB.Then i m inserting each of those values from array inside the
ListBox.
Initially my
Thanx for ur Kind attention,
Here is what i m exactly doing,
I want to fill my ListBox with some values from DB,
so i have created a dynamic array that will contain the values as per the no.
of records in DB.Then i m inserting each of those values from array inside the
ListBox.
Initially my
vamshi wrote:
hi,
I get 'Relocatable chunk leaked at' after my application exit from PilotMain,
The log file shows that the leak happened at a location which I never allocated directly.
(My application uses tables)
I have attached the log file, please find it.
please help me at solving out
Free all heap memories before exiting
-Original Message-
From: [EMAIL PROTECTED]
[mailto:[EMAIL PROTECTED] On Behalf Of vamshi
Sent: Monday, July 25, 2005 7:13 PM
To: Palm Developer Forum
Subject: Memory leak problem
I get this problem each time I exit my application 'Relocatable chunk
If you allocate memory, you have to deallocate it before you leave your app.
--
Scott Erickson
Software Engineer, FB-4, Inc.
[EMAIL PROTECTED]
vamshi [EMAIL PROTECTED] wrote in message news:[EMAIL PROTECTED]
I get this problem each time I exit my application 'Relocatable chunk
leaked at'
Looks like you aren't deallocating the current pointer in the field
when you assign a new handle to it.
Adam
On Jul 25, 2005, at 10:39 PM, vamshi wrote:
hi,
I get 'Relocatable chunk leaked at' after my application exit from
PilotMain,
The log file shows that the leak happened at a
Jeff Ondich wrote:
Many OS 4 devices only had 256K for the dynamic heap -- the larger
dynamic heap sizes didn't really start appearing until Palm OS 5. If
you need that much storage, it might be worth only supporting OS 5 and
later.
Is there a way to test for the available heap size?
Henk Jonas wrote:
Jeff Ondich wrote:
Yes, POSE gives me the memory leak error during exectuion. While
waiting for a computation that usually takes a second or two, I get this:
Found 1855 memory leaks for [my application]. Information
concerning the leaks can be found in the log file.
Maybe you app is exting at this moment. Question is just why...
because when you use new and there is no free memory it just exits your
application. i fought a lot with this...
to fix it you must override new operator so it won't cause exception but return
NULL instead (or catch the
Michal--that's a good idea, and I have now played quite a bit with the
new operator today. But I'm having trouble with operator new[] and
delete[]. I have found lots of advice on-line on how to override new
and delete, but I haven't found any samples of how to override new[] and
delete[]
At 06:22 PM 7/19/2005, you wrote:
Michal--that's a good idea, and I have now played quite a bit with the new
operator today. But I'm having trouble with operator new[] and
delete[]. I have found lots of advice on-line on how to override new and
delete, but I haven't found any samples of how
Ben Combee wrote:
Many OS 4 devices only had 256K for the dynamic heap -- the larger
dynamic heap sizes didn't really start appearing until Palm OS 5. If
you need that much storage, it might be worth only supporting OS 5 and
later.
Thanks, Ben. That's what I need to know.
--
For
Many OS 4 devices only had 256K for the dynamic heap -- the larger
dynamic heap sizes didn't really start appearing until Palm OS 5. If
you need that much storage, it might be worth only supporting OS 5 and
later.
Is there a way to test for the available heap size?
--
For information on
Hello Jeff,
I don't know how many small blocks of memory you are talking about, but
I've worked on applications with tons of small new/delete, arrays of
objects containing dynamically allocated stuff, etc... and never got
into any kind of limits. Each and every time I got a memory leaked
Thanks, Jerome. I always assume these things are my own fault, but this
one has me confused. It's claiming I have a memory leak before the
objects in question even go out of scope, right in the middle of
allocating some other object. It may be that I'm clobbering a pointer
and POSE is smart
One thing I don't understant... POSE reports memory leaks after your
application has terminated (before going back to the launcher). Are you
saying you get the memory leak message -during- your application execution?
Jerome
Jeff Ondich wrote:
Thanks, Jerome. I always assume these things
Yes, POSE gives me the memory leak error during exectuion. While
waiting for a computation that usually takes a second or two, I get this:
Found 1855 memory leaks for [my application]. Information
concerning the leaks can be found in the log file.
And indeed, the log file shows lots and
Jeff Ondich wrote:
Yes, POSE gives me the memory leak error during exectuion. While
waiting for a computation that usually takes a second or two, I get this:
Found 1855 memory leaks for [my application]. Information concerning
the leaks can be found in the log file.
And indeed, the log
Roel Wijmans [EMAIL PROTECTED] wrote in message
news:[EMAIL PROTECTED]
I seem to be having a memory leak when running a database of a
memory card. The emulator doesn't pick it up though. All is running
fine if the database runs of the device, but as soon as i run it of
the memory card, it i
Drew Haninger [EMAIL PROTECTED] a écrit dans le message de news:
[EMAIL PROTECTED]
here's the whole function, just notice the code that is commented...
ot_FntDefineFontreplaces FntDefineFont
(ot stands for OliveTree)
just the T5 actual device, notice we are applying the code to all OS 5.0
devices.
- Original Message -
From: Regis St-Gelais [EMAIL PROTECTED]
Newsgroups: palm-dev-forum
To: Palm Developer Forum palm-dev-forum@palm.lyris.net
Sent: Tuesday, February 01, 2005 4:27 AM
Subject: Re: Memory
Drew Haninger [EMAIL PROTECTED] a écrit dans le message de news:
[EMAIL PROTECTED]
it can only be done on OS 5.0, discoved through via experimentation and a
tip on this newsgroup, could not find any documentation.
Drew
Drew,
How do you test/mesure that memory leak?
Do you have a small test
: Monday, January 31, 2005 4:15 PM
Subject: Re: Memory leak in T5?
Drew Haninger [EMAIL PROTECTED] a crit dans le message de news:
[EMAIL PROTECTED]
it can only be done on OS 5.0, discoved through via experimentation and
a
tip on this newsgroup, could not find any documentation.
Drew
Drew Haninger [EMAIL PROTECTED] a écrit dans le message de news:
[EMAIL PROTECTED]
it can only be done on OS 5.0, discoved through via experimentation and a
tip on this newsgroup, could not find any documentation.
Drew
Is that leak problem showing on all OS5 devices or only on the T5?
--
there is a leak in FntDefineFont(), one needs to set the second parameter to
zero to get the memory back, here is a code snippet:
//*
// causes CRASHES in some sumulators, but not the actual devices.
// BUT necessary for the Palm T5, else it
Drew Haninger [EMAIL PROTECTED] a écrit dans le message de news:
[EMAIL PROTECTED]
there is a leak in FntDefineFont(), one needs to set the second parameter
to
zero to get the memory back, here is a code snippet:
Is this documented somewhere because the docs says:
Palm OS 4.0 and later do
: Wednesday, January 26, 2005 9:01 AM
Subject: Re: Memory leak in T5?
Drew Haninger [EMAIL PROTECTED] a crit dans le message de news:
[EMAIL PROTECTED]
there is a leak in FntDefineFont(), one needs to set the second
parameter
to
zero to get the memory back, here is a code snippet
Okay, I have switched to OS 4.12 and now have two more questions having to do
with the debug version.
1) It appears whenever I load the app from the debugger, I get the following
error:
UIAppShell (unknown version) called SysFatalAlert with the message: DataMgr.c,
Line: 8638, Records left
In case I was not clear with my questions,
1) Why is FrmCloseAllForms() needed? It seems like the OS should close the
forms automatically and if not the call should be in the samples.
2) One way it seems to me to get around these two memory leaks would be to
detect if the menu is up, such as
At 09:33 AM 11/20/2004, you wrote:
In case I was not clear with my questions,
1) Why is FrmCloseAllForms() needed? It seems like the OS should close the
forms automatically and if not the call should be in the samples.
This is a bug in the samples. I know a version of the Palm OS app
stationery
Thank you for the information on the missing function call in the earlier
versions of CW. I started with an earlier version and guess I just noticed it.
I actually had not touched it for a while partly because of being out of the
country on business.
Concerning the menus . . . the ONLY thing I
I just had a thought and tried it out . . .
1) Start Emulator.exe (I have the newest one)
2) Create a new session with palmos33-en-v.rom, Palm Vx, Standard-English, and
8192K.
3) Hit the Home button
4) Start the Address app.
5) Hit the Menu button to bring up the menu.
6) Hit the Home button
7)
I wish I could edit my previous posting rather than create another. Oh well, I
can create.
If this is not my problem, which it does not seem to be (but I could be wrong),
what suggestions would you have to get around this problem so that I can do
Gremlin testing and still have it check for
At 04:24 PM 11/20/2004, you wrote:
I wish I could edit my previous posting rather than create another. Oh
well, I can create.
If this is not my problem, which it does not seem to be (but I could be
wrong), what suggestions would you have to get around this problem so that
I can do Gremlin
On Nov 20, 2004, at 2:48 PM, Ben Combee wrote:
At 04:24 PM 11/20/2004, you wrote:
I wish I could edit my previous posting rather than create another.
Oh well, I can create.
If this is not my problem, which it does not seem to be (but I could
be wrong), what suggestions would you have to get
Thank you very much for your helps (the information). I believe in creating
very clean code, but will have to ignore this since I have not proven it is not
me. Thank you again.
--
For information on using the Palm Developer Forums, or to unsubscribe, please
see
Hays, Jonathan (GE Infrastructure) wrote / skrev:
I believe that when you call FrmInitForm, if the cell is a
textItemType, the system will allocate an empty string. In
that case, when you point the cell to your custom text,
you need to make sure that the old text handle is freed.
You can
Hays, Jonathan (GE Infrastructure) wrote / skrev:
Not sure if this is it, but when you call FldSetText, the previous
handle that was allocated doesn't get freed. You can call
FldGetTextHandle to the the old handle and then free it after
you have set the new one with FldSetText.
Thanks for
]
[mailto:[EMAIL PROTECTED] Behalf Of Veronica
Loell
Sent: Tuesday, October 05, 2004 12:42 AM
To: Palm Developer Forum
Subject: Re: Memory leak, table - SelectTextItem, AllocateLines
Hays, Jonathan (GE Infrastructure) wrote / skrev:
Not sure if this is it, but when you call FldSetText, the previous
Not sure if this is it, but when you call FldSetText, the previous handle that was
allocated doesn't get freed. You can call FldGetTextHandle to the the old handle and
then free it after you have set the new one with FldSetText.
-Jon
---
Jonathan Hays
you have to free this memory on the frmCloseEvent and all goes well
for(i=0;ikNumTextColumns;i++)
for(j=0;jkNumRows;j++)
{
if(gHandles[i][j])
MemHandleFree(gHandles[i][j]);
}
Best Regards
Manu
El 07/04/2004, a las 02:35, Pinaki
Pinaki Roy wrote:
more thing -- if i do lot of scrolling -- and then
press the exit button -it's giving lot of mem leaks
(in hundreds) -- so i added
That kind of suggests that your memory leaks are coming coming from a
custom draw function or from your scroll function maybe. Why don't you
Hi Matt,
Thanx for your response -- i tried adding
gHandles[i][j] = 0 but it's still not working -- one
more thing -- if i do lot of scrolling -- and then
press the exit button -it's giving lot of mem leaks
(in hundreds) -- so i added
this code again in sclRepeatEvent just before calling
Pinaki Roy wrote:
- but still it's giving memory leak whenever i press
the exit button and i 'm sure it's bocz the memory
allocated for ghandle -- how i can free the mem for
ghandle prpperly??
some more observations -
1) i'm calling FrmCloseAllForms() in my AppStop()
function -- and after
From: laclists [EMAIL PROTECTED]
I'm learning to develop for Palm using prc-tools and
gcc. I've created a normal program, but when I load
the first form I get three mem leaks from the emulator.
(snip)
But I'm using neither MemPtrNew() nor MemHandleNew().
I'm just loading the form
The emulator should tell you whether it's in LogLoadRecord() or
SetFieldTextFromStr().
If it's LogLoadRecord() try posting the entire function and make sure it's
not returning prematurely.
If it's leaking from the field, it would be harder to see w/o the whole
file.
Matt
David Beers [EMAIL
Uh, forget that part about the system disposing of the field memory in a
FrmReturnToForm(). I know better, and yes, I am calling
FldSetTextHandle(field, 0) on the ctlSelectEvents that close my modal form.
Still getting the leak, though.
- Original Message -
From: David Beers [EMAIL
David Beers wrote:
Uh, forget that part about the system disposing of the field memory in a
FrmReturnToForm(). I know better, and yes, I am calling
FldSetTextHandle(field, 0) on the ctlSelectEvents that close my modal form.
Still getting the leak, though.
Did you explicitly free the Field's
Hi, folks!
meg wrote:
eliminated. However, I can't figure out how to fix this. The problem is
definitely the MemPtrNew allocation in BuildAnswerString, but when I try and
Yup. Righto.
free the pointer with MemPtrFree at the very end of the function, the
database doesn't end up with the
I'm betting that the problem is in the definition of questions_record_type.
It appears (though I can't tell for sure) that you are allocating
dynamic strings when you build your questions, and then storing
pointers to those strings in your database. When you exit your
program, those strings
Thanks Hal, for the quick response.
Here is my definition for the question record type structure:
typedef struct {
char* Question1Answer;
char* Question2Answer;
char* Question3Answer;
char* Question4Answer;
char* Question5Answer;
char* Question6Answer;
char* Question7Answer;
char*
Hmmm,
also, when I replace the call to BuildAnswerString with just a StrCopy
everything works, even without the DmWrite calls changed. So I was figuring
the problem was in BuildAnswerString?
--
For information on using the Palm Developer Forums, or to unsubscribe, please see
As defined, there is no space in your database for the question text.
One (bloated) way to do it would be
Char Question1Answer[32];
Char Question2Answer[32];
Char Question3Answer[32];
etc.
UInt8 sessionFinished;
Another (nicer) way to do it would be
UInt8 sessionFinished;
Char
Thanks a lot!!
Dave Carrigan [EMAIL PROTECTED] ??:[EMAIL PROTECTED]
On Thu, Aug 14, 2003 at 12:53:10AM +0100, gamepower111 wrote:
In the FormHandleEvent function, i used the following codes, when the
form
can be closed correctly and return to the main form of the project. but
if i
Hi, use Char nameP[8]=['\0','\0','\0','\0','\0','\0','\0','\0'];
and close all forms in the stop application routine with FrmCloseAllForms(),
generally make the leaks.
--
Open WebMail Project (http://openwebmail.org)
-- Original Message ---
From: David Martin [EMAIL PROTECTED]
To:
On Thu, Aug 14, 2003 at 12:53:10AM +0100, gamepower111 wrote:
In the FormHandleEvent function, i used the following codes, when the form
can be closed correctly and return to the main form of the project. but if i
exit the application, memory leak occured. If I delete the frmCloseEvent
codes
Somewhere your linked list is not completely cleaning up after itself. If
you allocated the memory, you need to deallocate it.
- Original Message -
From: gamepower111 [EMAIL PROTECTED]
Newsgroups: palm-dev-forum
To: Palm Developer Forum [EMAIL PROTECTED]
Sent: Thursday, August 14, 2003
You are unlocking, but not freeing your memory handle. Lookup
MemHandleFree in the reference.
At 06:32 05/08/2003 -0700, you wrote:
Greetings
Here is a subroutine that does a frequently-wanted thing: it retrieves
the hotsync name. It also leaks maxsize bytes every time you call it.
I
Hi padiyar,
Do not assign NULL value to first for loop. Then check your code.
Rangababu TNV.
-Original Message-
From: veena padiyar [mailto:[EMAIL PROTECTED]
Sent: Friday, May 30, 2003 11:00 AM
To: Palm Developer Forum
Subject: Memory leak problem
hello,
I am using a 2
I didn't look so well on the code but I have two suggestions:
1. I would use: ( I repeat, maybe your code is good, but this is more
simple )
Int16 **arrayColOne=new Int16 *[6];
for(int i=0;i6;i++) arrayColOne[i]=new Int16[size];
//later
for(int i=0;i6;i++) delete[]
Developer Forum [EMAIL PROTECTED]
Sent: Monday, January 20, 2003 11:44 PM
Subject: Re: Memory Leak tracking?
At 11:35 PM +0100 1/20/03, Jan Slodicka wrote:
I repeat what was said in this forum within a few last days:
Look at the threads POSE reports memory leak even when I don't allocate
At 12:28 PM +0100 1/21/03, Jan Slodicka wrote:
But allow me one question: Is there some mechanism checking for writes past
allocated buffer?
Yes, as long as the write (or read, for that matter) accesses
something illegal (like the block header of the following block, or
the contents of an
At 14:21 2003-1-21 -0800, you wrote:
At 12:28 PM +0100 1/21/03, Jan Slodicka wrote:
But allow me one question: Is there some mechanism checking for writes past
allocated buffer?
Yes, as long as the write (or read, for that matter) accesses something
illegal (like the block header of the
1 - 100 of 127 matches
Mail list logo
