sambuc/codezero
1
0
Fork 0
mirror of https://github.com/drasko/codezero.git synced 2026-01-10 17:56:38 +01:00
Code Issues Actions Packages Projects Releases Wiki Activity

Updated LICENSE/README files, removed unused code from tasks.

Browse Source
This commit is contained in:
Bahadir Balban
2009-06-15 14:58:41 +03:00
parent 0dd8918ae5
commit ba1cc0c6bc
17 changed files with 44 additions and 810 deletions
Download Patch File Download Diff File Expand all files Collapse all files

5
LICENSE
View File

@@ -2,14 +2,13 @@
The source code under the codezero directory tree is governed by the
license below and this version only, unless it is stated otherwise as
part of the file. To be more precise, for every source file where it
says: "Copyright (C) 2007, 2008 Bahadir Balban" or a similar wording
says: "Copyright (C) 2007-2009 Bahadir Balban" or a similar wording
(capitalisation, years or the format of the name may change), the below
license has effect. Any other source file may or may not be inclusive.
For questions please contact me on bbalban@b-labs.co.uk
For questions please contact me on bahadir@l4dev.org
Bahadir Balban
GNU GENERAL PUBLIC LICENSE
Version 3, 29 June 2007

197
README
View File

@@ -1,100 +1,44 @@
Codezero Microkernel Pre-release
Codezero Microkernel v0.1 Release
Copyright (C) 2007, 2008 Bahadir Balban
Copyright (C) 2007-2009 Bahadir Bilgehan Balban
1.) What is Codezero?
Codezero is a new L4 microkernel that has been written from scratch. It is a
modern microkernel implementation that provides capabilities for virtualization
and implementation of native OS services. Codezero can act as a virtualisation
platform, a hardware abstraction layer, and as a basis for developing operating
system services. Codezero's primary focus is on embedded systems.
What is Codezero?
Codezero is a new microkernel that has been developed for embedded systems. It
implements a simple API that is based on the L4 Microkernel, and it can be used
as a base to develop or run new operating systems. As part of the project,
server tasks are provided that implement memory management and a virtual
filesystem layer. These servers are built upon the base Codezero API, and they
currently support a small but essential subset of the POSIX API.
Codezero targets high-end embedded systems that support virtual memory and it
has an emphasis on the ARM architecture. Open source development practices are
used in its development. Users can use Codezero's POSIX-like server tasks or
another operating system built upon it, or both at the same time. Codezero aims
to differ from other systems by implementing a modern embedded operating system
and yet provide more flexibility than a single operating system API.
Why the name Codezero?
2.) Why the name Codezero?
The project focuses on simplicity and clarity in software design. Everything is
kept simple, from the user-level tasks down to the build system that does not
take part in the actual microkernel.
kept simple, from the user-level services down to the build system that does
not take part in the actual microkernel. Codezero is a microkernel that is
easy-to-use and modify, providing a rapid kernel software development
environment.
Who could use Codezero?
3.) Why use Codezero?
Currently, Codezero can be used on any embedded system that requires
multitasking and virtual memory support but not all the detailed features of a
complex operating system. Codezero's initial advantage in that respect would be
its clarity and simplicity that makes it easier for users to grasp and use. In
the near future, the real-time features will be optimised and it will also be a
good candidate for real-time applications. Finally it will be used on systems
that require high dependability. Tightly integrated embedded systems are one
such example where multiple isolated application domains exist side-by-side,
sharing the same cpu and memory system.
Codezero aims to provide easy hands-on experience; you may simply download the
source tree, install the tools using this guide, and get up and running with
development on your virtual embedded platform from the comfort of your host PC.
Codezero strives to incorporate the most modern design features and presents
them in a well-written source code base. In other words, Codezero strives to be
technically cutting edge. Updates are rapid and open community is encouraged to
get involved in core development.
What are the design features?
4.) What is the license?
Codezero attempts to incorporate many modern features that are present in
today's operating systems. Some of them are presented below.
Based on the L4 microkernel design principles, there are only a few system
calls in Codezero. These system calls provide purely mechanism; threads and
address spaces, and the methods of inter-process communication between them.
Anything beyond these are policy and they are implemented in the userspace. Due
to this rigorously simple design the same microkernel can be used to design
completely different operating systems.
In terms of other features, the microkernel is preemptive, and smp-ready.
Currently only synchronous communication is implemented, but this will change
in the near future. The microkernel also incorporates a simple priority-based
scheduler, and all blocking operations (locking, ipc, waiting) are
interruptible. Even though the microkernel needs to be optimised in its
real-time capabilities, it does incorporate the necessary architecture to
support real-time performance.
There are two system tasks built upon the base microkernel that manage memory
and file-based I/O, called MM0 and FS0. MM0 is the system task that implements
memory management. It contains allocators and manages the page cache. It
implements demand paging by managing page faults, physical pages and their
file/task associations. MM0 provides the default paging mechanism on Codezero.
FS0 is the system task that implements a simple, modern virtual filesystem
layer. It is designed to serve file requests from MM0. Since it abstracts the
low-level filesystem details, it is a relatively easy job to port a new
filesystem to be used under FS0.
MM0 and FS0 both reside in the userspace, and they are not mandatory services.
For example the virtual and physical memory resources can be partitioned by
Codezero among pagers, and a third-party pager can override Codezero's pager on
the same run-time, and implement an independent paging behaviour for its own
memory partition. This feature provides the option of having an adjustable
mixture of generalisation and specialisation of system services at the same
run-time. For instance, Codezero's abstract posix-like page/file management
services can be used in combination with an application-specific pager that
depends on its own paging abilities. A critical task could both use mm0 and
fs0's posix-like files benefiting from the abstraction and simplification that
it brings, but at the same time rely on its own specialised page-fault handling
mechanism for its critical data. Similarly, a complete operating system can be
virtualised and both native and virtualised applications can run on the same
run-time.
What will the license be?
The current release is distributed under GNU General Public License Version 3
and this version only. Any next version will be released in the same license,
but there are intentions to keep the project in a dual-licensed manner. In any
case, one version of the source code will always be released as open source as
in the OSI definition.
The current release is distributed under GNU General Public License Version 3.
For contributions we ask for a copyright share agreement and you may freely
contribute to the project this way. This is our current model, but if you
object to this, feel free to mention your ideas in our mailing list.
The third party source code under the directories loader/ tools/ libs/c
libs/elf have their own copyright and licenses, separate from this project. All
@@ -102,79 +46,22 @@ third party source code is open source in the OSI definition. Please check
these directories for their respective licenses.
Why yet another POSIX microkernel?
5.) Why not improve on a popular open source kernel instead of Codezero?
There are many open source POSIX operating systems with advanced features such
as BSD versions and Linux. However, neither of these were originally designed
for embedded systems. Multiple problems arise due to this fact.
Open source POSIX kernels are not designed for running on embedded platforms.
* Their source code is encumbered and cluttered with legacy or unrelated
components.
* Their user bases are saturated, their core developers are engaged and
focused on non-embedded platforms.
* Their source code base is too complex to grasp and most components enforce
unnecessary policy on embedded applications.
Unix itself and all the tools built upon weren't meant for using on small
devices. Accordingly, existing Unix operating systems contain a lot of
historical code. Their code base is so big, that it gets more and more
difficult to understand how their internals work. On these systems usually much
of the existing code base is irrelevant to newly developed software, and
embedded systems need new software often. Codezero is written from scratch to
solely target embedded systems and as such the source code is %100 relevant.
It is small and free from legacy code.
Codezero is simple, clean, and focused on embedded systems. Still, existing
open source kernels are valuable with some of their features and drivers. As a
result, we attempt to virtualize them on top of Codezero.
From a design perspective, these kernels have a monolithic design, and as such
they may have issues with dependability due to much of the code sharing the
same address space. This is an important issue on embedded systems since their
operation is more sensitive to disruptions. Being a microkernel design,
Codezero aims to defeat this problem and increase dependability.
From a support perspective, most unix operating systems like BSD and Linux have
a highly saturated user base. The developers focus on these existing users and
often the systems they support are not embedded computers. Codezero will focus
completely on embedded systems, aiming to meet the support need for this type
of systems.
Other than modern Unix kernels, there are established operating systems
targeting embedded devices. Codezero will contrast and compete with some of
them by its simplicity, some by its openness and some by its feature set, but
mostly by providing a more flexible development model.
Finally, POSIX compliance is only a step, or a partial aim for the Codezero
microkernel. It is not limited to the goal of just complying with POSIX, which
has been done many times by other operating systems. Codezero microkernel will
provide a dependable software environment where isolated application domains
can run side-by-side in the same run-time. In addition, user-level servers MM0
and FS0 will implement native system services and provide a POSIX-like API for
these application domains.  By supplying a variety of system software options,
the applications will be able to choose among different speed, safety,
determinism policies at the same run-time. This is expected to prove useful in
embedded systems.
Furthermore there are new ideas in literature that would improve systems
software but aren't implemented either because they have no existing users or
may break compatibility (e.g. some are presented in Plan 9). For example file
abstractions could be used more liberally to cover data exchange and control of
devices, services and network communication. Existing kernels already have
established methods of doing such operations and they would oppose major design
overhauls, which limits their innovation capability for this kind of
experimentation. In contrast, Codezero's partitioned nature provides the
opportunity to implement innovative feature services in small and isolated
parts, without cluttering the rest of the system. As well as natively
supporting existing APIs such as POSIX, Codezero project aims to keep up with
the latest OS literature and provide the opportunity to incorporate the latest
ideas in OS technology.
Can you summarise? Why should I use Codezero, again?
Codezero is an operating system that targets embedded systems with virtual
memory support. It implements modern features such as demand-paging and a
virtual filesystem layer under a POSIX-like API. Different from most other
POSIX-like systems, it is based on a microkernel design. The microkernel has
been carefully designed so that it is small and well-focused. It has a cleanly
separated set of system services that can be used as a base for implementing or
running other operating systems. It can also be used as a barebones system that
provides multitasking and thread communication. Its source code is also freely
available (See LICENSE heading for details). Codezero aims to differ from other
systems by implementing an open and modern embedded operating system that
provides more flexibility than a single operating system API. Since currently
there's a very small user base, it can be easily adopted for any custom
embedded system project that needs focused developer attention.
Other embedded open source kernels are also out there. Codezero aims to differ
by clarity, simplicity and its cutting edge features.

130
tasks/blkdev0/SConstruct
View File

@@ -1,130 +0,0 @@
#
# User space application build script
#
# Copyright (C) 2007 Bahadir Balban
#
import os
import sys
import shutil
from string import split
from os.path import join
from glob import glob
task_name = "blkdev0"
# The root directory of the repository where this file resides:
project_root = "../.."
tools_root = join(project_root, "tools")
prev_image = join(project_root, "tasks/mm0/mm0.axf")
libs_path = join(project_root, "libs")
ld_script = "include/linker.lds"
physical_base_ld_script = "include/physical_base.lds"
# libc paths:
libc_variant = "userspace"
libc_libpath = join(libs_path, "c/build/%s" % libc_variant)
libc_incpath = join(libc_libpath, "include")
libc_crt0 = join(libs_path, "c/build/crt/sys-userspace/arch-arm/crt0.o")
libc_name = "c-%s" % libc_variant
# libl4 paths:
libl4_path = "../libl4"
libl4_incpath = join(libl4_path, "include")
#libmem paths:
libmem_path = "../libmem"
libmem_incpath = "../libmem"
# kernel paths:
kernel_incpath = join(project_root, "include")
# Kernel config header.
config_h = join(project_root, "include/l4/config.h")
# If crt0 is in its library path, it becomes hard to link with it.
# For instance the linker script must use an absolute path for it.
def copy_crt0(source, target, env):
os.system("cp " + str(source[0]) + " " + str(target[0]))
def get_physical_base(source, target, env):
os.system(join(tools_root, "pyelf/readelf.py --first-free-page " + \
prev_image +" >> " + physical_base_ld_script))
# The kernel build environment:
env = Environment(CC = 'arm-none-linux-gnueabi-gcc',
# We don't use -nostdinc because sometimes we need standard headers,
# such as stdarg.h e.g. for variable args, as in printk().
CCFLAGS = ['-g', '-nostdlib', '-ffreestanding', '-std=gnu99', '-Wall', '-Werror'],
LINKFLAGS = ['-nostdlib', '-T' + ld_script, "-L" + libc_libpath, "-L" + libl4_path, "-L" + libmem_path],
ASFLAGS = ['-D__ASSEMBLY__'],
PROGSUFFIX = '.axf', # The suffix to use for final executable
ENV = {'PATH' : os.environ['PATH']}, # Inherit shell path
LIBS = [libc_name, 'libl4', 'libmm', 'libmc', 'libkm', \
'gcc', libc_name], # libgcc.a - This is required for division routines.
CPPFLAGS = "-D__USERSPACE__",
CPPPATH = ['#include', libl4_incpath, libc_incpath, kernel_incpath, libmem_incpath])
def extract_arch_subarch_plat(config_header):
'''
From the autogenerated kernel config.h, extracts platform, archictecture,
subarchitecture information. This is used to include the relevant headers
from the kernel directories.
'''
arch = None
subarch = None
plat = None
if not os.path.exists(config_header):
print "\n\nconfig.h does not exist. "\
"Please run: `scons configure' first\n\n"
sys.exit()
f = open(config_h, "r")
while True:
line = f.readline()
if line == "":
break
parts = split(line)
if len(parts) > 0:
if parts[0] == "#define":
if parts[1] == "__ARCH__":
arch = parts[2]
elif parts[1] == "__PLATFORM__":
plat = parts[2]
elif parts[1] == "__SUBARCH__":
subarch = parts[2]
f.close()
if arch == None:
print "Error: No config symbol found for architecture"
sys.exit()
if subarch == None:
print "Error: No config symbol found for subarchitecture"
sys.exit()
if plat == None:
print "Error: No config symbol found for platform"
sys.exit()
return arch, subarch, plat
def create_symlinks(arch):
arch_path = "include/arch"
arch_path2 ="src/arch"
if os.path.exists(arch_path):
os.system("rm %s" % (arch_path))
os.system("ln -s %s %s" % ("arch-" + arch, arch_path))
if os.path.exists(arch_path2):
os.system("rm %s" % (arch_path2))
os.system("ln -s %s %s" % ("arch-" + arch, arch_path2))
arch, subarch, plat = extract_arch_subarch_plat(config_h)
create_symlinks(arch) # Creates symlinks to architecture specific directories.
src = [glob("src/*.c"), glob("fsbin/*.S"), glob("*.c"), glob("src/arch/*.c")]
objs = env.Object(src)
physical_base = env.Command(physical_base_ld_script, prev_image, get_physical_base)
crt0_copied = env.Command("crt0.o", libc_crt0, copy_crt0)
task = env.Program(task_name, objs + [crt0_copied])
env.Alias(task_name, task)
env.Depends(task, physical_base)

43
tasks/blkdev0/fsbin/SConstruct
View File

@@ -1,43 +0,0 @@
#
# Binary to elf build script
# Works by including a binary in an assembler file and linking it.
#
# Copyright (C) 2007 Bahadir Balban
#
import os
import sys
import shutil
from os.path import join
from glob import glob
elfimg_name = "romfs"
# The root directory of the repository where this file resides:
project_root = "../.."
tools_root = join(project_root, "tools")
prev_image = join(project_root, "tasks/test0/test0.axf")
ld_script = "include/linker.lds"
physical_base_ld_script = "include/physical_base.lds"
def get_physical_base(source, target, env):
os.system(join(tools_root, "pyelf/readelf.py --first-free-page " + \
prev_image + " >> " + physical_base_ld_script))
# The kernel build environment:
env = Environment(CC = 'arm-none-linux-gnueabi-gcc',
# We don't use -nostdinc because sometimes we need standard headers,
# such as stdarg.h e.g. for variable args, as in printk().
CCFLAGS = ['-g', '-nostdlib', '-ffreestanding', '-std=gnu99', '-Wall', '-Werror'],
LINKFLAGS = ['-nostdlib', '-T' + ld_script],
ASFLAGS = ['-D__ASSEMBLY__'],
PROGSUFFIX = '.axf', # The suffix to use for final executable
ENV = {'PATH' : os.environ['PATH']}, # Inherit shell path
CPPFLAGS = "-D__USERSPACE__",
CPPPATH = ['#include'])
src = [glob("*.S")]
objs = env.Object(src)
physical_base = env.Command(physical_base_ld_script, prev_image, get_physical_base)
task = env.Program(elfimg_name, objs)
env.Alias(elfimg_name, task)
env.Depends(task, physical_base)

7
tasks/blkdev0/fsbin/incbin.S
View File

@@ -1,7 +0,0 @@
/* This is used to place a binary file into an elf section */
.section .data.fs
.incbin "fsbin/romfs.bin"
.align 4

BIN
tasks/blkdev0/fsbin/romfs.bin
View File

Binary file not shown.

1
tasks/blkdev0/include/arch
View File

@@ -1 +0,0 @@
arch-arm

31
tasks/blkdev0/include/blkdev.h
View File

@@ -1,31 +0,0 @@
#ifndef __BLKDEV_H__
#define __BLKDEV_H__
#include <l4lib/types.h>
struct block_device;
struct block_device_ops {
int (*init)(struct block_device *bdev);
int (*open)(struct block_device *bdev);
int (*read)(struct block_device *bdev, unsigned long offset,
int size, void *buf);
int (*write)(struct block_device *bdev, unsigned long offset,
int size, void *buf);
int (*read_page)(struct block_device *bdev,
unsigned long pfn, void *buf);
int (*write_page)(struct block_device *bdev,
unsigned long pfn, void *buf);
};
struct block_device {
char *name;
void *private; /* Low-level device specific data */
u64 size;
struct block_device_ops ops;
};
void init_blkdev(void);
#endif /* __BLKDEV_H__ */

47
tasks/blkdev0/include/linker.lds
View File

@@ -1,47 +0,0 @@
/*
* Simple linker script for userspace or svc tasks.
*
* Copyright (C) 2007 Bahadir Balban
*/
/*
* The only catch with this linker script is that everything
* is linked starting at virtual_base, and loaded starting
* at physical_base. virtual_base is the predefined region
* of virtual memory for userland applications. physical_base
* is determined at build-time, it is one of the subsequent pages
* that come after the kernel image's load area.
*/
/* USER_AREA_START, see memlayout.h */
virtual_base = 0x10000000;
__stack = 0x20000000;
INCLUDE "include/physical_base.lds"
/* physical_base = 0x228000; */
offset = virtual_base - physical_base;
ENTRY(_start)
SECTIONS
{
. = virtual_base;
_start_text = .;
.text : AT (ADDR(.text) - offset) { crt0.o(.text) *(.text) }
/* rodata is needed else your strings will link at physical! */
.rodata : AT (ADDR(.rodata) - offset) { *(.rodata) }
.rodata1 : AT (ADDR(.rodata1) - offset) { *(.rodata1) }
.data : AT (ADDR(.data) - offset)
{
. = ALIGN(4K);
_start_ramdisk0 = .;
*(.data.romfs)
_end_ramdisk0 = .;
. = ALIGN(4K);
_start_ramdisk1 = .;
*(.data.sfs)
_end_ramdisk1 = .;
*(.data)
}
.bss : AT (ADDR(.bss) - offset) { *(.bss) }
_end = .;
}

7
tasks/blkdev0/include/ramdisk.h
View File

@@ -1,7 +0,0 @@
#ifndef __RAMDISK_H__
#define __RAMDISK_H__
extern struct block_device ramdisk[];
extern struct block_device ramdisk[];
#endif

59
tasks/blkdev0/main.c
View File

@@ -1,59 +0,0 @@
/*
* High-level block device handling.
*
* Copyright (C) 2007, 2008 Bahadir Balban
*/
#include <stdio.h>
#include <blkdev.h>
/*
* Handles block device requests from fs0 using a combination of
* server-specific and posix shm semantics
*/
void handle_block_device_request()
{
u32 mr[MR_UNUSED_TOTAL];
l4id_t sender;
int err;
u32 tag;
printf("%s: Listening requests.\n", __TASKNAME__);
if ((err = l4_receive(L4_ANYTHREAD)) < 0) {
printf("%s: %s: IPC Error: %d. Quitting...\n", __TASKNAME__,
__FUNCTION__, err);
BUG();
}
/* Read conventional ipc data */
tag = l4_get_tag();
sender = l4_get_sender();
/* Read mrs not used by syslib */
for (int i = 0; i < MR_UNUSED_TOTAL; i++)
mr[i] = read_mr(i);
switch(tag) {
case L4_IPC_TAG_WAIT:
printf("%s: Synced with waiting thread.\n", __TASKNAME__);
break;
case L4_IPC_TAG_BLOCK_OPEN:
sys_open(sender, (void *)mr[0], (int)mr[1], (u32)mr[2]);
break;
default:
printf("%s: Unrecognised ipc tag (%d)"
"received. Ignoring.\n", __TASKNAME__, mr[MR_TAG]);
}
}
int main(void)
{
/* Initialise the block devices */
init_blkdev();
while (1) {
handle_block_device_request();
}
return 0;
}

11
tasks/blkdev0/src/init.c
View File

@@ -1,11 +0,0 @@
#include <blkdev.h>
#include <ramdisk.h>
void init_blkdev(void)
{
ramdisk[0].ops.init(&ramdisk[0]);
ramdisk[1].ops.init(&ramdisk[1]);
}

116
tasks/blkdev0/src/ramdisk.c
View File

@@ -1,116 +0,0 @@
/*
* A basic ramdisk implementation.
*
* Copyright (C) 2007, 2008 Bahadir Balban
*
* The ramdisk binary is embedded in the data section of the ramdisk device
* executable. Read/writes simply occur to this area. The disk area could
* have any filesystem layout e.g. romfs, which is irrelevant for this code.
*/
#include <blkdev.h>
#include <ramdisk.h>
#include <l4/macros.h>
#include <l4/config.h>
#include <l4/types.h>
#include <string.h>
#include INC_SUBARCH(mm.h)
#include INC_GLUE(memory.h)
/* Ramdisk section markers for ramdisk inside this executable image */
extern char _start_ramdisk0[];
extern char _end_ramdisk0[];
extern char _start_ramdisk1[];
extern char _end_ramdisk1[];
struct ramdisk_data {
u64 base;
u64 end;
};
struct ramdisk_data rddata[2];
__attribute__((section(".data.sfs"))) char sfsdisk[SZ_16MB];
int ramdisk_init(struct block_device *ramdisk)
{
struct ramdisk_data *rddata = ramdisk->private;
if (!strcmp("ramdisk0", ramdisk->name)) {
rddata->base = (u64)(unsigned long)_start_ramdisk0;
rddata->end = (u64)(unsigned long)_end_ramdisk0;
ramdisk->size = (u64)((unsigned long)_end_ramdisk0 -
(unsigned long)_start_ramdisk0);
} else if (!strcmp("ramdisk1", ramdisk->name)) {
rddata->base = (u64)(unsigned long)_start_ramdisk1;
rddata->end = (u64)(unsigned long)_end_ramdisk1;
ramdisk->size = (u64)((unsigned long)_end_ramdisk1 -
(unsigned long)_start_ramdisk1);
} else
return -1;
return 0;
}
int ramdisk_open(struct block_device *ramdisk)
{
return 0;
}
int ramdisk_read(struct block_device *b, unsigned long offset, int size, void *buf)
{
struct ramdisk_data *data = b->private;
void *src = (void *)(unsigned long)(data->base + offset);
memcpy(buf, src, size);
return 0;
}
int ramdisk_write(struct block_device *b, unsigned long offset,
int size, void *buf)
{
struct ramdisk_data *data = b->private;
void *dst = (void *)(unsigned long)(data->base + offset);
memcpy(dst, buf, size);
return 0;
}
int ramdisk_readpage(struct block_device *b, unsigned long pfn, void *buf)
{
ramdisk_read(b, __pfn_to_addr(pfn), PAGE_SIZE, buf);
return 0;
}
int ramdisk_writepage(struct block_device *b, unsigned long pfn, void *buf)
{
ramdisk_write(b, __pfn_to_addr(pfn), PAGE_SIZE, buf);
return 0;
}
struct block_device ramdisk[2] = {
[0] = {
.name = "ramdisk0",
.private = &rddata[0],
.ops = {
.init = ramdisk_init,
.open = ramdisk_open,
.read = ramdisk_read,
.write = ramdisk_write,
.read_page = ramdisk_readpage,
.write_page = ramdisk_writepage,
},
},
[1] = {
.name = "ramdisk1",
.private = &rddata[1],
.ops = {
.init = ramdisk_init,
.open = ramdisk_open,
.read = ramdisk_read,
.write = ramdisk_write,
.read_page = ramdisk_readpage,
.write_page = ramdisk_writepage,
},
}
};

28
tasks/blkdev0/tools/generate_bootdesc.py
View File

@@ -1,28 +0,0 @@
#!/usr/bin/python
import os
import sys
compiler_prefix = "arm-none-linux-gnueabi-"
objdump = "objdump"
command = "-t"
image_name = "inittask.axf"
linkoutput_file_suffix = "-linkinfo.txt"
linkoutput_file = image_name + linkoutput_file_suffix
def generate_bootdesc():
command = compiler_prefix + objdump + " -t " + image_name + " > " + linkoutput_file
print command
os.system(command)
f = open(linkoutput_file, "r")
while True:
line = f.readline()
if len(line) is 0:
break
if "_start" in line or "_end" in line:
print line
f.close()
if __name__ == "__main__":
generate_bootdesc()

18
tasks/sigma0/channel.c
View File

@@ -1,18 +0,0 @@
struct channel {
int dir; /* Direction */
char *name; /* Name */
int cd; /* Channel descriptor */
};
struct interface {
struct channel chan[];
};
int main(int argc, char *argv[])
{
void *buf = malloc(sizeof(struct channel)*10);
struct interface *intf = buf;
}

154
tasks/sigma0/main.c
View File

@@ -1,154 +0,0 @@
/*
* Lightweight and simple RPC primitives.
*
* Copyright (c) 2007 Bahadir Balban
*
* Some methodology is needed for efficient and functional message passing
* among processes, particulary when shared memory or same address space
* messaging is available. The rpc primitives here attempt to fill this gap.
*
* The idea is to generate as little bloat as possible. To that end we don't
* need encryption, marshalling, type tracking and discovery. Also Very minor
* boilerplated code is produced from C macros rather than a notorious rpc tool.
*/
#include <stdio.h>
#include <stdlib.h>
#include <string.h>
/* What default value arg0 is given for rpc calls */
#define ARG0 8
#define PAGE_SIZE 0x1000
char pagebuf[PAGE_SIZE];
struct ipc_shmem {
unsigned long vaddr;
};
struct tlmsg {
struct ipc_shmem shmem;
unsigned long arg[3];
unsigned long ret;
unsigned long method_num;
};
typedef int (*remote_call_t)(struct tlmsg *m);
int square(int i)
{
printf("%s: Entering...\n", __FUNCTION__);
return i*i;
}
/* Hand coded wrapper for non-complex argument types. */
int rpc_square(struct tlmsg *m)
{
printf("%s: Entering...\n", __FUNCTION__);
return square(m->arg[0]);
}
struct complex_struct {
int item;
int item2;
};
/* Use this to declare an RPC wrapper for your function. Your function
* must return a primitive type (e.g, int, float, long etc.) and can
* have a ref to a single argument of complex type (e.g. a struct). Apart
* from these limitations it is essentially a regular C function. Note
* that one can pass a lot of data back and forth using a single struct.
* The wrappers are very lightweight, and thanks to many possibilities of
* shared memory (e.g. same address-space, shared pages) data need not be
* copied when passed back and forth. */
#define DECLARE_RPC_BYREF(ret, func, type0) \
static inline ret rpc_byref_##func(struct tlmsg *m) \
{ /* Find struct address */ \
unsigned long multiword_struct = m->arg[0] + \
m->shmem.vaddr; /* Data passed by a shared entity */ \
return func((type0 *)multiword_struct); /* Call real function */\
}
/* Same as above, but passing a structure by value instead of reference.
* This is much slower due to lots of copying involved. It is not
* recommended but included for completion. */
#define DECLARE_RPC_BYVAL(ret, func, type0) \
static inline ret rpc_byval_##func(struct tlmsg *m) \
{ /* Find struct address */ \
unsigned long multiword_struct = m->arg[0] + \
m->shmem.vaddr; /* Data passed by a shared entity */ \
/* Call real function, by value */ \
return func((type0)(*(type0 *)multiword_struct)); \
}
/* Use these directly to declare the function, *and* its RPC wrapper. */
#define RPC_FUNC_BYREF(ret, func, type0) \
ret func(type0 *); \
DECLARE_RPC_BYREF(ret, func, type0) \
ret func(type0 *arg0)
#define RPC_FUNC_BYVAL(ret, func, type0) \
ret func(type0); \
DECLARE_RPC_BYVAL(ret, func, type0) \
ret func(type0 arg0)
RPC_FUNC_BYVAL(int, complex_byval, struct complex_struct)
{
printf("%s: Entering...\n", __FUNCTION__);
arg0.item++;
return 0;
}
RPC_FUNC_BYREF(int, complex_byref, struct complex_struct)
{
printf("%s: Entering...\n", __FUNCTION__);
arg0->item++;
return 0;
}
#define RPC_NAME(func, by) rpc_##by##_##func
remote_call_t remote_call_array[] = {
rpc_square,
RPC_NAME(complex_byval, byval),
RPC_NAME(complex_byref, byref),
};
struct tlmsg *getmsg(int x)
{
struct tlmsg *m = (struct tlmsg *)malloc(sizeof(struct tlmsg));
m->method_num = x;
m->arg[0] = ARG0;
m->shmem.vaddr = (unsigned long)&pagebuf;
return m;
}
void putmsg(struct tlmsg *m)
{
free((void *)m);
}
void check_rpc_success()
{
struct complex_struct *cs = (struct complex_struct *)(pagebuf + ARG0);
printf("complex struct at offset 0x%x. cs->item: %d, expected: 1.\n", ARG0, cs->item);
}
int main(void)
{
struct tlmsg *m[3];
unsigned int ret;
remote_call_t call[3];
memset((void *)pagebuf, 0, PAGE_SIZE);
for (int i = 0; i < 3; i++) {
m[i] = getmsg(i);
printf("Calling remote function %d according to incoming msg.\n", i);
call[i] = remote_call_array[m[i]->method_num];
ret = call[i](m[i]); /* i.e. call rpc function number i, with
* message number i as argument */
printf("Call returned %d\n", ret);
m[i]->ret = ret;
putmsg(m[i]);
}
check_rpc_success();
}

BIN
tasks/sigma0/rpc
View File

Binary file not shown.