Re: Idea: Move kernel to upper memory

[BVK Chaitanya <bayapuneni_chaitanya@symantec.com>]

Hi,


What is the conclusion of this thread?  is this idea still explored?


Robert Millan wrote:
> 
> The first concern that comes to mind is how would GRUB coexist with the
> payload area which precisely starts at 0x100000.  But I expect we'd face
> many unexpected issues.
> 

Does this mean, GRUB needs to fit within 1MB limit when all its modules 
are loaded?  More clearly, are there any limits on where (and how many) 
grub modules can be dynamically loaded currently?

Since payload(s) needs to be present at their precise positions only 
after the boot command, we can always deterministically /schedule/ 
memmove operation to be executed as part of boot command.

Since boot command is the last command executed by GRUB (correct me if 
not), we can overwrite any GRUB code/data at that point.  This also 
needs boot command to exist in low memory.

The main advantage i see here is that it removes any arbitrary limits on 
number of modules or their placement, if at all limits exists.


thanks,
--
bvk-chaitanya