Can library files not in use cause crashing?

[redrumloa]

Just like the title asks. Can a something.library file actually cause a crash even if it isn't in use?

[Doobrey]

What do you mean by in use?
Opened by an app, but not actually using it?
or just sitting there on the HD,unopened?

 If you mean the latter, then no. If no app has opened the library, then it can`t cause a crash.

[CaptainHIT]

LOL :-D

Nice April joke! ;-)
I see only one guy believed that question :-P

[cecilia]

i just figured Red had made a mistake and was actually using Windows
:roflmao:

[Ilwrath]

Can a something.library file actually cause a crash even if it isn't in use?

Well, I hate to be like Bill Clinton... But that depends on your definition of "cause" and of "in use."  

The presence of a library in LIBS: may be queried by a certain program, even if it doesn't plan on using it.  IIRC, several CD32 games queried the existance of lowlevel.library, without actually using it.  If it found a lowlevel.library, the game would assume it was on a CD32, and might crash an A1200.  Thus, a library not in use COULD cause a crash of a badly written app.  I suggest digging around with SnoopDOS to figure out what is really happening.  :-)

[redrumloa]

Umm this was a real question and not an april fools!

@llwrath

I am trying to figure out a major instability. I am pretty sure it is
hardware related, but I want to eliminate everything else.

It's a A4k CSPPC w/OS3.9. The problem is it is even crashing on
startup before even launching any programs sometimes. Other times it
will wotk for a while.

[Ilwrath]

It's a A4k CSPPC w/OS3.9. The problem is it is even crashing on startup before even launching any programs sometimes. Other times it will wotk for a while.

Ick....  That does sound hardware-ish, doesn't it?

I might suggest cold-booting the system up into 3.1 from floppy/zip, or some other media, to make sure, though.  The 040/060 libs are called by Setpatch, and an improper version (or corruption) of one of these libs can trash a system fairly early in startup.  Also, the ROM update file could cause errors that look like hardware faults.  I actually had a partially corrupted Boing Bag 2 install (OS3.9) cause Yellow-Screen boot faults on my A4000!  

[NightShade737]

I copied a load of Libraries onto my UAE setup (not overwriting anything) and suddenly a load of things stopped working and others crashed (things like Execute Command and Icon Info etc). So I think the answer is yes.

[Piru]

@redrumloa

Can a something.library file actually cause a crash even if it isn't in use?

No.

Probably you just don't notice the library is in fact used by something. If the library is not used at all, it can't cause any stability problems.

@NightShade737

I copied a load of Libraries onto my UAE setup (not overwriting anything) and suddenly a load of things stopped working and others crashed (things like Execute Command and Icon Info etc). So I think the answer is yes.

Just because you don't overwrite a library when copying libs over doesn't mean the library would not be used on next boot.

So probably you copied some library that the system will open on next boot, and that will mess things up.

[Damion]

Ick.... That does sound hardware-ish, doesn't it?

I agree with Ilwrath...sounds definately like a hardware fault.







....or...I 'could' just be April foolsin' ya...you decide.:-P

[Framiga]

hi red,

crash or freeze?

I've had the same problem here with CSPPC233-060, and i've found that was a mechanical problem.

Pushing down (a little bit) on the 060 cpu, the system booted/worked perfectely.

So i've cleaned all the sockets (060CPU and the 200pins one) reseatted properly and now all works OK.

Now i'm start to think, to use the system used by Rigger Bolted down CSPPC

Ciao

[NightShade737]

Also, do a search on AmiNet for CSPPC and you will find this:

Short:    Workaround for CSPPC/233 hardware bug
Author:   lesueur@club-internet.fr, frank@phoenix.owl.de (Frank Wille & Emmanuel Lesueur)
Uploader: frank@phoenix.owl.de
Type:     hard/misc
Requires: powerpc.library or ppc.library

Some CyberStormPPC 604e/233MHz boards seem to suffer from annoying
random crashes under WarpOS as well as under PowerUp. They might be
caused by some kind of hardware bug, which sets bit 0 or bit 2 of the
PowerPC's LR register in certain situations. This will result into an
instruction access fault, when a program returns from a sub routine.

If you ever got an instruction access exception where the first digit
of LR is not '0', then this program might solve most of your troubles.

Whether this problem comes from a defective board, or a bad power supply,
or from too much heat is unknown. But surprisingly it only affects
CSPPCs with a 233 MHz CPU. 200 MHz is fine, I have one for comparison!

This archive contains two exception handlers, one for PowerUp
(CSPPC233Fix_ppc) and one for WarpOS (CSPPC233Fix_wos), which reset
those trashed bits in case of an exception which is recognized as the
"LR-Bug problem". The PPC program is able to continue after LR was fixed.


USAGE:

CSPPC233Fix_ppc (PowerUp version):
----------------------------------
  Can be started from CLI and from Workbench.
  CLI Usage:
      LRBug NOREQ/S,QUIET/S,NOWAIT/S

  NOREQ: don't ask before attempting to restart a task.
  QUIET: don't write messages with kprinf() when restarting a task.
  NOWAIT: don't install a handler, just restart crashed tasks and quit.

  WB Usage:
      Tooltypes NOREQ, QUIET, and NOWAIT, with the same meaning.

  To remove the handler, send a CTRL-C.

CSPPC233Fix_wos (WarpOS version):
---------------------------------
  Just start from CLI. No options. To remove the handler, press CTRL-C.
  Whenever an LR-Bug occurs, the hit will be printed to stdout. If
  you don't want that, redirect output to NIL:. All other exceptions
  will be invoke the WarpOS exception window, as usual.


REBUILD:

The full source text of both programs is included and a makefile is
provided to rebuild them with vbcc. The PowerUp version can also be
compiled with SAS/C and gcc. The WarpOS version with StormC (gcc-WarpOS
is not recommended for an exception handler, because of different ABIs).

[Framiga]

Hi NightShade737

yes i've already downloaded the LRBug fix but, my problem isn't software related.

It is a mere mechanical problem. (i'm quite sure that 70% of the PUP sent to DCE for repairs, had a similar problem)

BTW, do you suggest me to install it anyway?

Cheers

PS- should be useful, to start a thread or so, PUP related with all the tips-suggestions-fixes made by the users.

[redrumloa]

@all
Thanks for the responses, it helps me confirm it's a hardware problem which I pretty much knew already..

Out of curiosity, do lib files get loaded into RAM in startup or do the lay unused in LIBS until a program calls for it?

crash or freeze?

Well I am getting all kinds of random weirdness, but part of that may be some unrealsed stuff I am testing. But to focus on boot up, the system locks up occasionally. The picture is still there, but the mouse won't move. The keyboard does respond to a 3 finger salute. I really have a pretty plain jane OS3.9 setup, so it's not software hacks.

[PiR]

Hi

Have you ever considered simple lack of power?
It can show like you describe - everything freezes, including mouse pointer.

Another tip (my own :-D): CapsLock test:

Keyboard has its own simple CPU with small buffer.
When your system freezes try hitting CapsLock several times and observe its diode.
If after several hits (lets say no more then twenty) you notice that led is not changing (after the last hit it stayed on or off ignoring you're still hitting it), this means that keyboard buffer is full - waiting the main CPU to fetch the data. This means that CPU stopped and this may happen because the voltage dropped so much that CPU refuses to work any more.

3fingers works anyway, as it is forced via the back-door, omitting this keyboard buffer.

Invended this on my A1200.

Out of curiosity, do lib files get loaded into RAM in startup or do the lay unused in LIBS until a program calls for it?

Everything stays on disk and getts loaded only when needed and that means - at the first try to open it.
It is worth to mention that it is unloaded ... no, not, when the last one who used the lib is finished. It is unloaded only when the system is trying to organize any more memory and this lib not used at the moment, and this means, when system runs short of memory and somewhere someone tries to allocate some.


Good luck