[prev in list] [next in list] [prev in thread] [next in thread] 

List:       eros-arch
Subject:    Re: [eros-arch] Problem with Mapped Registers
From:       Norman Hardy <norm () cap-lore ! com>
Date:       2003-12-31 17:13:34
Message-ID: AA9199A2-3BB4-11D8-85D5-000393121E30 () cap-lore ! com
[Download RAW message or body]


On Dec 31, 2003, at 3:48 AM, Jonathan S. Shapiro wrote:

> On Sun, 2003-12-28 at 00:24, Charles Landau wrote:
>> If you decree that every process must have a mapped page for
>> communication with the EROS kernel, an emulator no longer has the
>> complete control over the process's address space that it needs for
>> an accurate emulation.
>
> 1. On most architectures this is already a lost cause, because portions
> of the address space are visibly occupied by the EROS kernel itself.

For the X86 see "Kludge" in this 
<http://www.cap-lore.com/CapTheory/PrivMap.html> for a way for the 
kernel to hide itself. This is a pain but will work on most machines.

> 2. There is almost always a page that can be stolen, subject to the
> proviso that the emulated code is not actively looking to guard against
> emulation.

.......

Norm Hardy <http://cap-lore.com>

_______________________________________________
eros-arch mailing list
eros-arch@mail.eros-os.org
http://www.eros-os.org/mailman/listinfo/eros-arch
[prev in list] [next in list] [prev in thread] [next in thread]