Re: Broken kernel compile on 5.1-RELEASE / 5-CURRENT (SMP, PAE &scsi)

From: Terry Lambert <tlambert2_at_mindspring.com>
Date: Wed, 20 Aug 2003 00:52:56 -0700
Robert Watson wrote:
> Can't speak to the specifics of this, but you want to be very careful not
> to use kernel modules with PAE: modules are currently without the context
> of the kernel configuration file, and PAE introduces possible binary
> incompatibility with modules that dig into VM (which many drivers do).

This is a good argument for defining it as it's own port, like
IA64 or SPARC or whatever.

> Various
> conversations have happened regarding how to address this problem, and I'm
> not sure we've come up with the right answer yet.  There seem to be two
> conflicting directions: build modules in the context of a kernel module
> (and get conditionally compiled type/structure/code/... pieces), and try
> to make the module build entirely independent of a kernel configuration.

The approach I prefer is: PAE gives you access to up to 64G of
real memory, and you really only use it when you have more than
4G of physical memory (my understanding of the memory architecture
on the hardware where this would be used is that the minimum ends
up being 6G).  So you should have no objection to compiling all
the drivers in statically, since it's not like you need to save
the space.  8-) 8-).

-- Terry
Received on Tue Aug 19 2003 - 22:53:59 UTC

This archive was generated by hypermail 2.4.0 : Wed May 19 2021 - 11:37:19 UTC