[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: Seg 11 in GCC



gvb@elentari.cs.wcu.edu (Glenn Bily)  wrote on 29.03.96 in <[🔎] 199603291856.NAA03775@elentari.cs.wcu.edu>:

> There are lots of explanations:
>
> The seg faults in question are more likely caused by improperly
> installing GCC or corrupted libraries. Or compiling new code on older

Umm ... did you READ what I wrote?!

> > Then how do _you_ explain that about 99% of all "gcc got signall 11"
> > problems can be solved by getting new RAM, changing waitstates, and
> > similar actions?

This won't solve improper installs or corrupted files.

> versions of GCC. I really want to know how many people you are talking

Well, judging from news and mailing lists, *nearly all* complaints about  
gcc sig 11 are eventually solved by tweaking the hardware.

That's why the answer has become automatic - it usually works.

> about. The fact is that RAM is solid state and the odds of it failure
> are almost non-existant. (A collegue of mine who has been in the
> industry for 35+ years says he has seen 1 RAM chip fail). You might want

Well, I've been "in the industry" for about 5 years, and I've seen about  
two dozen SIMMs fail _at least_. And lots more cases where changing wait  
states helped. And just about none of them was found by the on-board RAM  
test.

I've no idea what your collegue does, but I get to see end user  
problems.[1]

(I've also seen at least half a dozen 486's with broken on-chip-cache, and  
double that with broken external cache. Simply disabling the cache in  
question made any problems go away. It's so bad that nowadays, this is my  
first reaction to strange problems.)

The state of the chip industry is horrible. There's a simple reason -  
very, very slim margins.

> OS/2 is a whole different story. I'm not qualified to answer that.

It's the exact same story. And nowadays, Windows does it as well. Oh yes,  
I've also seen it happen with Netware.

[1] I tended to view viruses as something that gets talked about, but  
doesn't really happen, as well - until I got that job. I still haven't  
seen a single virus on my home system, but at work I've seen lots and lots  
of infections - mostly boot sector viruses.

MfG Kai



Reply to: