home *** CD-ROM | disk | FTP | other *** search
- Path: sparky!uunet!zaphod.mps.ohio-state.edu!wupost!darwin.sura.net!ra!tantalus!eric
- From: eric@tantalus.dell.com (Eric Youngdale)
- Newsgroups: comp.os.linux
- Subject: Re: GP faults and other trivia. . .
- Message-ID: <3544@ra.nrl.navy.mil>
- Date: 9 Sep 92 15:08:42 GMT
- References: <1992Sep8.104048.1@ualr.edu> <1992Sep9.094232.17759@klaava.Helsinki.FI>
- Sender: usenet@ra.nrl.navy.mil
- Organization: Naval Research Laboratory
- Lines: 54
-
- In article <1992Sep9.094232.17759@klaava.Helsinki.FI> torvalds@klaava.Helsinki.FI (Linus Benedict Torvalds) writes:
- >If you see intermittent system failure (core dumping in gcc etc) that
- >are not easily repeatable, and don't necessarily happen at the same
- >location every time, there are a couple of possible reasons:
- >
- > (a) kernel bugs
- > (b) weird/buggy hardware
- > (c) installation problems
- >
-
- There is one other possibility, (which is probably a subset of (a)),
- and that is a compiler bug which leads to incorrect code being generated.
- This could lead to a buggy kernel, but you could stare at the source code all
- day long and never have a clue as to what is going wrong. We do need to keep
- in mind that gcc 2.2.2 is still considered a beta release, and RMS does
- consider 1.40 (and 1.41) to be a more stable compiler. In fact, if there are
- weird unexplainable problems with the kernel, we might want to consider
- compiling the kernel with 1.40 (or 1.41), and seeing if the problems go away
- (although I do not know if it is still possible to compile the kernel with
- 1.40). RMS has been making noises about a gcc 2.3 release sometime in early
- October.
-
- I should point out that I have not yet seen any evidence of a compiler
- bug being responsible for any problems with the kernel.
-
- >So the first things to check when seeing problems like the above is if
- >it's hardware-related: one good way to do this is usually to slow down
- >the machine to 8MHz or whetever, and see if the errors go away. If they
- >do, it's probably not a software bug (although races etc can be
- >timing-dependent: not very frequent). Other things you can do is try
- >out some system testing software: but note that linux usually is a
- >better system tester than most of these especially if they run onder
- >16-bit DOS and don't check 32-bit accesses to high memory etc.
-
- I would not berate the system testing software out there as much as
- Linus does. It is probably true that 32 bit memory access is probably not
- tested, but if you have a bad SIMM (or a bad contact in the SIMM socket), it is
- a fairly painless way to find out about it.
-
- In general I have found Linux to be quite stable, and whenever
- I do see strange intermittent behavior, it is one of two things:
-
- a) the CMOS settings are corrupted.
- b) One of the SIMMs is not making good contact.
-
- I have found (a) to happen much more often that I would like, but whenever
- I get weird behavior, I will simply reload the CMOS from the defaults in the
- BIOS, and this usually gets things going. I have only seen (b) once.
-
- -Eric
-
- --
- Eric Youngdale
- eric@tantalus.nrl.navy.mil
-