Subject: Re: kernel symbol tables (was Re: vmstat, iostat etc no longer work?)
To: None <current-users@NetBSD.ORG>
From: Mike Long <mike.long@analog.com>
List: current-users
Date: 11/14/1996 17:30:59
>Date: Thu, 14 Nov 1996 16:15:29 -0500
>From: Chris G Demetriou <Chris_G_Demetriou@ux2.sp.cs.cmu.edu>

[Greg Hudson wrote:]
>> You're Confused (tm).  The argument is that the in-memory symbol table
>> could include symbols from LKMs, in addition to the symbols from the
>> basic kernel.  I don't really know why this is important, but unless
>> I'm completely misunderstanding things, Mike was not arguing that the
>> LKM interface could use this symbol table.
>
>So, i figured that he wasn't arguing for just that, since i couldn't
>figure out any reason why it would be important/useful, except in the
>context of loading other LKMs...

Yup.  We need kernel symbols for three things:

1) DDB:  Symbols must be resident and nonpageable "forever".
2) KVM:  Symbols may be pageable, and may be discarded once kvm_mkdb
         has been run.
3) LKM:  Symbols may be pageable, and may be discarded when
         securelevel > 0.

In all cases it would be nice (but not strictly necessary)
if symbols from LKMs could be added to the symbols from the kernel
image.

>I'm not saying that it's necessarily bad,w just that the corner cases
>shouldn't be overlooked, and that logically incorrect justifications
>shouldn't be given as to why it's a "good solution."

Well, let me put it this way:  How many people run NetBSD on machines
where configuring DDB into the kernel would make it too big?
-- 
Mike Long <mike.long@analog.com>     <URL:http://www.shore.net/~mikel>
VLSI Design Engineer         finger mikel@shore.net for PGP public key
Analog Devices, CPD Division          CCBF225E7D3F7ECB2C8F7ABB15D9BE7B
Norwood, MA 02062 USA       (eq (opinion 'ADI) (opinion 'mike)) -> nil