Buffer overruns

Talk about anything here.
Post Reply
guest

Post by guest »

anyone report buffer overruns since 1.5?
User avatar
Doobie
Site Admin
Posts: 1121
Joined: Wed Jan 28, 2004 5:34 pm
Location: Calgary Alberta
Contact:

Post by Doobie »

I've never encountered a buffer overrun vulnerability in myth due to the fact that input boxes always control the length of user input. But now that you mention it, I have a couple ideas of how to get around that... mb i'll look into this. I suppose map making has some potential in this regard as well.
Want to play Myth? New and old Myth players can playmyth online at Mariusnet.com.
KyleMaclean.net
Myrd
Site Admin
Posts: 4029
Joined: Fri Mar 19, 2004 10:17 pm

Post by Myrd »

A few existing ones were fixed in 1.5. Usually, buffer overruns are fairly hard (but not impossible) to discover without looking at the code in question. And no, we haven't had any reports (although feel free to make one if you have something!).



Edited By Myrd on 1085012518
lerabu
Posts: 57
Joined: Wed Apr 14, 2004 12:54 pm

Post by lerabu »

It occurs often in m3 - Sprite Buffer Overload
User avatar
haravikk
Site Admin
Posts: 987
Joined: Mon Mar 22, 2004 12:56 pm
Location: Scotland

Post by haravikk »

iirc that was because of the sprite buffer being increased by MythDev to get better graphical performance, but if you have a slow or low memory graphics card then this might dominate it and cause problems. Wouldn't hurt to put a note on the bug-tracker though :)
Being Haravikk gets you girls like these:
Image
CIK
Posts: 1127
Joined: Mon Mar 01, 2004 9:08 pm

Post by CIK »

Actually the Sprite buffer in m3 isn't a buffer overrun, it just runs out of space that's all and drops Sprites.

The problem has been with m3 since 1.0, in the early days it caused sprite freakout which fixed in 1.1.2 or something like that.

There was a time when I tried increasing the buffer size, but it caused issues for a # of folks and even the enlarged one could still not be big enough in certain instances.
lerabu
Posts: 57
Joined: Wed Apr 14, 2004 12:54 pm

Post by lerabu »

So we're gonna have to wait 5 more uears untill there is a quantum computer powerful enough to run at at max specs without a glitch? ;)
User avatar
GHOST®
Posts: 673
Joined: Sat Mar 20, 2004 12:19 am
Location: MiChIgAn
Contact:

Post by GHOST® »

[color=DFC99B]
lerabu wrote:So we're gonna have to wait 5 more uears untill there is a quantum computer powerful enough to run at at max specs without a glitch? ;)
MB not THAT long. ;op

Last I talked to Myrd, he was doing some prelim investigation into MIII and is processor usage (iirc), he discovered that where as MII`s Overhead map uses only 3% of the proceesing time/power, in MIII that same feature is using a whopping 36%. :oO

A few more finds/fixes like this, and you could almost run MIII on lesser machines than stated in the System Requirements mb. ;op

Anyways, just saying that after watching our programmers work on MII, I have much higher hopes for MIII now than Ive ever had. :oD[/color]
"Do you want to be healed now? Or would you prefer to bleed to death so I can try my hand at resurrection?"
Post Reply