Re: Strange bug with ilisp 5.12.0 on NT

Alain Picard <[email protected]> writes:

> On Windows NT, with ilisp-5.12.0, and Xanalys Lispworks 4.2 enterprise,

Always remember to mention the version number, the type (XEmacs, Emacs) and
the compilation method (cygwin, native) of your emacs as well.

> emacs gets into an infinite loop, consuming 100% of the

Always?  After some specific action?

> CPU.  Even multiple control-g's don't interrupt it; I have
> to go to the task manager and gun it down.
> The version I was using previously, ilisp version 5.10.1,
> does not suffer from this problem.

Such "cannot be interrupted" problems are sometimes (often?) Emacs problems
(perhaps in addition to a problem in a package which triggers it), but in
this case the many bindings of inhibit-quit to `t' in ILISP might cause this
(that you cannot interrupt, that is; the reason for it being stuck is another
question).  You could try modifying them all from inhibit-quit <- t to
inhibit-quit <- nil and then, when Emacs/ILISP gets stuck, try to interrupt
it with C-g.  If you are lucky, you'll get a backtrace (load sources instead
of compiled files so that you can make more sense of it) and see where ILISP
is stuck (use edebug for more precise debugging, then).

> How can I diagnose fix this or diagnose this further?

If that inhibit-quit trick does not help, you probably have an emacs problem.
In this case, at least Emacs (perhaps XEmacs as well, I don't know) comes
with a file DEBUG (in the etc directory) which explains this.  The sections
"** If the symptom of the bug is that Emacs fails to respond" and "** Some
suggestions for debugging on MS Windows:" might be of help.

-- 
Hannu
Please don't send copies of list mail

_______________________________________________________________

Don't miss the 2002 Sprint PCS Application Developer's Conference
August 25-28 in Las Vegas -- http://devcon.sprintpcs.com/adp/index.cfm