Author |
Message |
CypherJF I gargle nitroglycerin
Gender: Joined: Aug 14 2003 Posts: 2582 Location: USA Offline
|
Posted: Thu Nov 04, 2004 8:50 am Post subject: |
|
|
|
|
Interesting.. Thanks for bearing w/ me on this... I still think it's a user control creating an issue. :/
edit:
Do me a favor try running this special version I made for you; it'll run a message box for each step of loading it does (10 steps). Let me know of the results...
edit:
attachment removed. thanks for reporting back. _________________ Performance is often the art of cheating carefully. - James Gosling
Last edited by CypherJF on Fri Nov 05, 2004 12:27 am, edited 1 time in total |
|
Back to top |
|
|
D1st0rt Miss Directed Wannabe
Age:36 Gender: Joined: Aug 31 2003 Posts: 2247 Location: Blacksburg, VA Offline
|
Posted: Thu Nov 04, 2004 5:34 pm Post subject: |
|
|
|
|
Speaking of memory overloads, whats the fastest way to chew up ram of say, a school computer? Do I just put a big malloc inside a while loop? _________________
|
|
Back to top |
|
|
Mr Ekted Movie Geek
Gender: Joined: Feb 09 2004 Posts: 1379 Offline
|
Posted: Thu Nov 04, 2004 5:59 pm Post subject: |
|
|
|
|
calloc() is better. That will guarantee all allocated virtual RAM gets committed. _________________ 4,691 irradiated haggis! |
|
Back to top |
|
|
Grelminar Creator of Asss
Joined: Feb 26 2003 Posts: 378 Offline
|
Posted: Thu Nov 04, 2004 11:58 pm Post subject: |
|
|
|
|
This is really pedantic, but in theory, calloc could be implemented without actually writing to memory, if the C library knows that the mechanism it's using to get new pages from the OS are pre-zeroed. This is true for anonymously mmaped pages on unix systems. To really be sure you're forcing physical RAM and swap to be allocated, do the memset yourself. |
|
Back to top |
|
|
Chocko Guest
Offline
|
Posted: Fri Nov 05, 2004 12:22 am Post subject: |
|
|
|
|
finishes step 2 and then goes out of memory |
|
Back to top |
|
|
CypherJF I gargle nitroglycerin
Gender: Joined: Aug 14 2003 Posts: 2582 Location: USA Offline
|
Posted: Fri Nov 05, 2004 1:00 am Post subject: |
|
|
|
|
It's exiting on/near
call frmmain.show()
hmmm, it's got to be something on that form that's causing it. I really don't know. :-/ sorry. |
|
Back to top |
|
|
Mr Ekted Movie Geek
Gender: Joined: Feb 09 2004 Posts: 1379 Offline
|
Posted: Fri Nov 05, 2004 1:16 am Post subject: |
|
|
|
|
Grelminar wrote: | This is really pedantic, but in theory, calloc could be implemented without actually writing to memory, if the C library knows that the mechanism it's using to get new pages from the OS are pre-zeroed. This is true for anonymously mmaped pages on unix systems. To really be sure you're forcing physical RAM and swap to be allocated, do the memset yourself. |
Or to be even more efficient, malloc() the RAM and write just a single word every 4K (or whatever VM granularity is). |
|
Back to top |
|
|
CypherJF I gargle nitroglycerin
Gender: Joined: Aug 14 2003 Posts: 2582 Location: USA Offline
|
Posted: Fri Nov 05, 2004 5:02 pm Post subject: |
|
|
|
|
Out of Memory Issue:
Chocko and I finally got to sit down and figured it out. D With Windows XP you can follow steps such as this:
1.) Goto c:/windows/system32/, and backup your mswinsck.ocx
2.) Locate and install mswinsck.ocx into the /system32/ directory.
3.) Goto run type in: regsvr32 "c:/windows/system32/mswinsck.ocx"
The problem should be fixed. Note: I gave Chocko my copy of the Winsock control, so it is possible download on the link I have provided may not be a proper version.
Version 1.2 has been released; click here to go there. |
|
Back to top |
|
|
|