John Hawkinson wrote:

> | Module Name:	sharesrc
> | Committed By:	gmcgarry
> | Date:		Sun Jun  4 22:38:58 UTC 2000
> | 
> | Modified Files:
> | 	  sharesrc/share/man/man4/man4.i386: apm.4
> | 
> | Log Message:
> | Document APM_ALLOW_BOGUS_SEGMENTS.  Update description for APM_POWER_CHANGE,
> | addresses PR10066 from Takahiro Kambe <taca@sky.yamashina.kyoto.jp>.
> 
> Err, what does APM_ALLOW_BOGUS_SEGMENTS really do?

Allow the use of data segments which are in unexpected locations.

> It seems to me that in the spirit of "be liberal in what you accept...",
> we should always accept bogus segments unless there is a very good
> reason not to do so.

I raised these issues on port-i386 some months ago.  I received no
comments.  I committed the code, preserving the existing behaviour.

My notebook uses a data segment at 3MB.  I suspect a kernel with full
symbol table would overlay this segment and I'm going to be see silent
kernel corruption using apm.  Therefore, the default setting should
prohibit the use of apm unless the user is willing to accept this risk.

	-- Gregory McGarry <g.mcgarry@ieee.org>