blob: 945ebe1789ea71dae39d81228a693da5e267e3dd (
plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
|
LADEN
=====
laden is the bootloader for the Hurd running on the L4 microkernel.
It contains the architecture specific part of booting the L4 kernel
and setting up sigma0 and the rootserver.
OPERATION
=========
laden is started by an architecture bootloader, and is provided with
ELF images of the kernel, sigma0 and the rootserver (and optionally
sigma1). It then loads these ELF images into the appropriate memory
location. Afterwards, the kernel interface page is located and
patched with the necessary information about the inital servers, the
memory configuration and the boot info. Finally it jumps to the
kernel's entry point.
PORTING
=======
If you want to port laden to a new architecture, you need to glue it
to the architecture's boot loader to enable laden to find the kernel,
sigma0 and the rootserver in memory. You also need to port the ELF
loader. Furthermore, you need to write output routines and some
functions like sleep(), reset() and halt().
OUTPUT
======
The available output drivers are architecture specific (see below for
details). You can suppress all output by using the output driver
"none", which is available on all architectures.
Architecture Specific Notes
===========================
ia32
----
On ia32, laden is a multiboot compliant kernel that can be booted by
GRUB. laden expects that the first multiboot module is the L4 kernel,
the second module is sigma0, and the third module is the rootserver.
All of these must be statically linked ELF binary images with load
addresses so that the programs do not overlap.
The boot info field in the KIP will be set to the multiboot
information structure.
On this architecture laden supports VGA output.
|
