[Openmcl-devel] Problems with asdf-install

[Gary Byers]

When you get thrown into the kernel debugger, what does the backtrace
look like ? (As a wild guess, I wonder if anything's trying to do
I/O to a defunct stream early in the startup process.  That would
error, try to print an error message on a defunct stream, error,
try to print a message on a defunct stream ... and getting thrown
into the kernel debugger after this goes on for a while is arguably
slightly better than stack-overflowing in relative silence.)  Whether
it's because of what I'm calling a defunct stream or for some other
reason, the "error reporting error" message generally means that
the error system has concluded that something is very, very wrong.

I don't use Slime, and may be misrembering and/or misunderstanding
what I remember about it, but I wonder what it does when sets up the
standard streams (*TERMINAL-IO*, etc.) for the thread that it creates
and how pervasive that is.  I don't know whether even the vague
suspicion/nervousness I'm expressing about that is at all justified;
it's probably a good idea to first see if this is a "defunct stream"
issue or something else entirely.

On Mon, 2 Jan 2006, Ralph Richard Cook wrote:

> I'm using OpenMCL 1.0 with Slime. If I load asdf-install and, for
> example portable aserve, and save with ccl:save-application, then try
> to load that image via Slime, I'll get
>
>
> Welcome to OpenMCL Version 1.0 (DarwinPPC32)!
> ? ;Loading #P"/Users/rrc/slime/swank-loader.lisp"...
> ;Loading #P"/Users/rrc/.slime/fasl/openmcl-1.0-darwin-powerpc/swank-
> backend.dfsl"...
> ;Loading #P"/Users/rrc/.slime/fasl/openmcl-1.0-darwin-powerpc/
> nregex.dfsl"...
> ;Loading #P"/Users/rrc/.slime/fasl/openmcl-1.0-darwin-powerpc/
> metering.dfsl"...
> ;Loading #P"/Users/rrc/.slime/fasl/openmcl-1.0-darwin-powerpc/swank-
> openmcl.dfsl"...
> ;Loading #P"/Users/rrc/.slime/fasl/openmcl-1.0-darwin-powerpc/swank-
> gray.dfsl"...
> ;Loading #P"/Users/rrc/.slime/fasl/openmcl-1.0-darwin-powerpc/
> swank.dfsl"...Bug in OpenMCL system code:
> Error reporting error
> ? for help
> [829] OpenMCL kernel debugger:
>
> Anyone else seen anything like this?
>
> Thanks,
> Richard
>
> _______________________________________________
> Openmcl-devel mailing list
> Openmcl-devel at clozure.com
> http://clozure.com/mailman/listinfo/openmcl-devel
>
>