Invalid NETSTAT command crashes SRB2

[Fawfulfan]

NETSTAT 1 can be used to determine the framerate of a map.

signal_handler(): segment violation

Clearly, something is wrong with either SRB2 or the wiki.

 

[MascaraSnake]

So why didn't you post this in the crash report topic, why didn't you give an EIP and why didn't tell us on what map and in which gametype it happened? NETSTAT works just fine for everybody else, you know.

 

[Fawfulfan]

Well, it happened uniformly for me, in EVERY map under EVERY circumstance. I wasn't aware that NETSTAT worked for other people.

 

[Mr. Mystery]

Well, it happened uniformly for me, in EVERY map under EVERY circumstance. I wasn't aware that NETSTAT worked for other people.

You thought wrong

 

[Fawfulfan]

Wait a minute...I just posted a bug report about that for the main SRB2 EXE. Now I think about it, I'm pretty sure I'm using Cinefast's modified SRB2 EXE. So could it be that one of Cinefast's fixes--some fix he made to both vanilla SRB2 and Cineblast--breaks NETSTAT?

Apparently people are encountering this problem in Cineblast as well. I seem to be the only one getting it in standard SRB2, though, and that may be why.

 

[jloy1]

I can use it fine in both and it hasn't crashed for me!
SO, I'm guessing, using it in certain places could cause it to crash?

EDIT: Cineblast is the only one that crashes on NETSTAT 1

 

[Prime 2.0]

Hey, can we please get an EIP here instead of flailing about wondering why different people get different results?

 

[Fawfulfan]

Hey, can we please get an EIP here instead of flailing about wondering why different people get different results?

EIP=00000000

I'm guessing that isn't very helpful, though.

 

[Ezer'Arch]

I'm guessing that isn't very helpful, though.

http://mb.srb2.org/showthread.php?p=698068#post698068

 

[Sonict]

Since it seems like this isn't actually an issue with a standard SRB2 build, I'll just mark this bug as invalid for now.

Now, If I am wrong. Feel free to talk about this more and I will change it back. I won't lock this just yet.

 

[Prime 2.0]

Just to make absolutely sure, I've uploaded the official binary here. Give it a whirl and see if it still happens, Fawfulfan.

...and remember, we only accept bug reports for the official binary packaged with 2.0.6.

 

[Fawfulfan]

Well, it looks like NETSTAT doesn't crash the game in the official EXE. I think it seems pretty clear that Cinefast's modifications are to blame.

And better still, the very problem Cinefast's binary fixed doesn't seem to be an issue for me anymore, so I'll stick to the official one from now on.

So...I guess you can mark this bug report as invalid.

 

[Cinefast]

Well, it looks like NETSTAT doesn't crash the game in the official EXE. I think it seems pretty clear that Cinefast's modifications are to blame.

And better still, the very problem Cinefast's binary fixed doesn't seem to be an issue for me anymore, so I'll stick to the official one from now on.

So...I guess you can mark this bug report as invalid.

This is interesting - the thing SRB2Cineblast and the OGL-Fix EXE have in common is that both were compiled with GCC 4.5.0. I'll have to see if the problem goes away with GCC 3.4.5 or some other old version.

OK, I ran it in gdb and the stacktrace clearly points to an error in the C runtime. This should affect SRB2-trunk if compiled with GCC 4.5.0 too.

 

[Prime 2.0]

This is interesting - the thing SRB2Cineblast and the OGL-Fix EXE have in common is that both were compiled with GCC 4.5.0. I'll have to see if the problem goes away with GCC 3.4.5 or some other old version.

OK, I ran it in gdb and the stacktrace clearly points to an error in the C runtime. This should affect SRB2-trunk if compiled with GCC 4.5.0 too.

Yeah, I recall hearing Alam say that GCC 4.5 had some issues with SRB2 a while back.