hak5
/
openwrt
mirror of https://github.com/hak5/openwrt.git
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity
openwrt/target/linux/brcm63xx/image/Makefile

84 lines
2.5 KiB
Makefile
Raw Normal View History

credit where credit is due SVN-Revision: 4091
2006-06-27 00:35:46 +00:00
#
# Copyright (C) 2006 OpenWrt.org
#
# This is free software, licensed under the GNU General Public License v2.
# See /LICENSE for more information.
#
Merge premilinary Broadcom 63xx support (Livebox, Siemens SE515...) Kernel compiles fine, does not include linux-2.6.8.1 linux-mips specific fixes and neither includes broadcom changes to netfiler Needs testing of course :) SVN-Revision: 4084
2006-06-26 11:45:23 +00:00
include $(TOPDIR)/rules.mk
include $(INCLUDE_DIR)/image.mk
bcm963xx: fix load address / kernel entry Current kernels have the same kernel entry as load address, so just set them like this. Also, the BCM963xx uses 0x80010000 as its load address, so use this too. Signed-off-by: Axel Gembe <ago@bastart.eu.org> SVN-Revision: 11169
2008-05-18 17:29:39 +00:00
LOADADDR = 0x80010000 # RAM start + 16M
KERNEL_ENTRY = $(LOADADDR) # Newer kernels add a jmp to the kernel_entry at the start of the binary
Merge premilinary Broadcom 63xx support (Livebox, Siemens SE515...) Kernel compiles fine, does not include linux-2.6.8.1 linux-mips specific fixes and neither includes broadcom changes to netfiler Needs testing of course :) SVN-Revision: 4084
2006-06-26 11:45:23 +00:00
RAMSIZE = 0x01000000 # 64MB
LOADER_MAKEOPTS= \
KDIR=$(KDIR) \
LOADADDR=$(LOADADDR) \
KERNEL_ENTRY=$(KERNEL_ENTRY) \
RAMSIZE=$(RAMSIZE)
Major cleanup : - no more Broadcom kerSys* code - working runtime boot loader detection - updated flash map driver SVN-Revision: 6896
2007-04-08 17:20:01 +00:00
define trxalign/jffs2-128k
-a 0x20000
endef
define trxalign/jffs2-64k
-a 0x10000
endef
define trxalign/squashfs
-a 1024
endef
bcm963xx: image generation using image tagger This adds CFE tagged image generation in the BCM963xx target image Makefile. It has to generate its own LZMA compressed kernel because CFE is a LZMA nazi. It also seems that the Broadcom image tagger as well as CFE use AdvanceCOMP, which is based on a slightly older LZMA SDK. Anyways, some of the code is GPL, some of it is LGPL, so it might be that Broadcom owes some sources. Also, LZMA has a bug which causes it to generate different output when you are using stdin / stdout compared to plain files. I've just worked around the issue by using plain files. Signed-off-by: Axel Gembe <ago@bastart.eu.org> SVN-Revision: 11171
2008-05-18 17:30:04 +00:00
define Image/Build/CFE
# Generate the tagged image
$(STAGING_DIR_HOST)/bin/imagetag -i $(KDIR)/vmlinux.lzma.cfe -f $(KDIR)/root.$(1) \
-o $(BIN_DIR)/openwrt-$(2)-$(1)-cfe.bin \
-b $(2) -c $(3) -e $(LOADADDR) -l $(LOADADDR)
# -b $(2) -c $(3) -e $(KERNEL_ENTRY) -l $(LOADADDR)
$(call prepare_generic_squashfs,$(BIN_DIR)/openwrt-$(2)-$(1)-cfe.bin)
endef
Merge premilinary Broadcom 63xx support (Livebox, Siemens SE515...) Kernel compiles fine, does not include linux-2.6.8.1 linux-mips specific fixes and neither includes broadcom changes to netfiler Needs testing of course :) SVN-Revision: 4084
2006-06-26 11:45:23 +00:00
define Build/Clean
Major cleanup : - no more Broadcom kerSys* code - working runtime boot loader detection - updated flash map driver SVN-Revision: 6896
2007-04-08 17:20:01 +00:00
$(MAKE) -C lzma-loader clean
Merge premilinary Broadcom 63xx support (Livebox, Siemens SE515...) Kernel compiles fine, does not include linux-2.6.8.1 linux-mips specific fixes and neither includes broadcom changes to netfiler Needs testing of course :) SVN-Revision: 4084
2006-06-26 11:45:23 +00:00
endef
define Image/Prepare
bcm963xx: image generation using image tagger This adds CFE tagged image generation in the BCM963xx target image Makefile. It has to generate its own LZMA compressed kernel because CFE is a LZMA nazi. It also seems that the Broadcom image tagger as well as CFE use AdvanceCOMP, which is based on a slightly older LZMA SDK. Anyways, some of the code is GPL, some of it is LGPL, so it might be that Broadcom owes some sources. Also, LZMA has a bug which causes it to generate different output when you are using stdin / stdout compared to plain files. I've just worked around the issue by using plain files. Signed-off-by: Axel Gembe <ago@bastart.eu.org> SVN-Revision: 11171
2008-05-18 17:30:04 +00:00
# Standard LZMA kernel
build system cleanup/restructuring as described in http://lists.openwrt.org/pipermail/openwrt-devel/2007-August/001159.html SVN-Revision: 8362
2007-08-07 00:04:25 +00:00
cat $(KDIR)/vmlinux | $(STAGING_DIR_HOST)/bin/lzma e -si -so -eos -lc1 -lp2 -pb2 > $(KDIR)/vmlinux.lzma
bcm963xx: image generation using image tagger This adds CFE tagged image generation in the BCM963xx target image Makefile. It has to generate its own LZMA compressed kernel because CFE is a LZMA nazi. It also seems that the Broadcom image tagger as well as CFE use AdvanceCOMP, which is based on a slightly older LZMA SDK. Anyways, some of the code is GPL, some of it is LGPL, so it might be that Broadcom owes some sources. Also, LZMA has a bug which causes it to generate different output when you are using stdin / stdout compared to plain files. I've just worked around the issue by using plain files. Signed-off-by: Axel Gembe <ago@bastart.eu.org> SVN-Revision: 11171
2008-05-18 17:30:04 +00:00
# CFE is a LZMA nazi! It took me hours to find out the parameters!
# Also I think lzma has a bug cause it generates different output depending on
# if you use stdin / stdout or not. Use files instead of stdio here, cause
# otherwise CFE will complain and not boot the image.
$(STAGING_DIR_HOST)/bin/lzma e -d22 -fb64 -a1 $(KDIR)/vmlinux $(KDIR)/vmlinux.lzma.tmp
# Strip out the length, CFE doesn't like this
dd if=$(KDIR)/vmlinux.lzma.tmp of=$(KDIR)/vmlinux.lzma.cfe bs=5 count=1
dd if=$(KDIR)/vmlinux.lzma.tmp of=$(KDIR)/vmlinux.lzma.cfe ibs=13 obs=5 skip=1 seek=1 conv=notrunc
rm -f $(KDIR)/vmlinux.lzma.tmp
# Build the LZMA loader
Major cleanup : - no more Broadcom kerSys* code - working runtime boot loader detection - updated flash map driver SVN-Revision: 6896
2007-04-08 17:20:01 +00:00
rm -f $(KDIR)/loader.gz
$(MAKE) -C lzma-loader \
BUILD_DIR="$(KDIR)" \
TARGET="$(KDIR)" \
clean install
bcm963xx: image generation using image tagger This adds CFE tagged image generation in the BCM963xx target image Makefile. It has to generate its own LZMA compressed kernel because CFE is a LZMA nazi. It also seems that the Broadcom image tagger as well as CFE use AdvanceCOMP, which is based on a slightly older LZMA SDK. Anyways, some of the code is GPL, some of it is LGPL, so it might be that Broadcom owes some sources. Also, LZMA has a bug which causes it to generate different output when you are using stdin / stdout compared to plain files. I've just worked around the issue by using plain files. Signed-off-by: Axel Gembe <ago@bastart.eu.org> SVN-Revision: 11171
2008-05-18 17:30:04 +00:00
Major cleanup : - no more Broadcom kerSys* code - working runtime boot loader detection - updated flash map driver SVN-Revision: 6896
2007-04-08 17:20:01 +00:00
echo -ne "\\x00" >> $(KDIR)/loader.gz
brcm63xx image generation cleanups (#3169) SVN-Revision: 10703
2008-04-01 21:10:47 +00:00
rm -f $(KDIR)/fs_mark
touch $(KDIR)/fs_mark
$(call prepare_generic_squashfs,$(KDIR)/fs_mark)
bcm963xx: image generation using image tagger This adds CFE tagged image generation in the BCM963xx target image Makefile. It has to generate its own LZMA compressed kernel because CFE is a LZMA nazi. It also seems that the Broadcom image tagger as well as CFE use AdvanceCOMP, which is based on a slightly older LZMA SDK. Anyways, some of the code is GPL, some of it is LGPL, so it might be that Broadcom owes some sources. Also, LZMA has a bug which causes it to generate different output when you are using stdin / stdout compared to plain files. I've just worked around the issue by using plain files. Signed-off-by: Axel Gembe <ago@bastart.eu.org> SVN-Revision: 11171
2008-05-18 17:30:04 +00:00
Merge premilinary Broadcom 63xx support (Livebox, Siemens SE515...) Kernel compiles fine, does not include linux-2.6.8.1 linux-mips specific fixes and neither includes broadcom changes to netfiler Needs testing of course :) SVN-Revision: 4084
2006-06-26 11:45:23 +00:00
endef
define Image/Build
make the image filenames a bit shorter SVN-Revision: 9533
2007-11-10 23:08:11 +00:00
$(STAGING_DIR_HOST)/bin/trx -o $(BIN_DIR)/openwrt-$(BOARD)-$(1).trx -f $(KDIR)/loader.gz -f $(KDIR)/vmlinux.lzma $(call trxalign/$(1)) -f $(KDIR)/root.$(1)
bcm963xx: image generation using image tagger This adds CFE tagged image generation in the BCM963xx target image Makefile. It has to generate its own LZMA compressed kernel because CFE is a LZMA nazi. It also seems that the Broadcom image tagger as well as CFE use AdvanceCOMP, which is based on a slightly older LZMA SDK. Anyways, some of the code is GPL, some of it is LGPL, so it might be that Broadcom owes some sources. Also, LZMA has a bug which causes it to generate different output when you are using stdin / stdout compared to plain files. I've just worked around the issue by using plain files. Signed-off-by: Axel Gembe <ago@bastart.eu.org> SVN-Revision: 11171
2008-05-18 17:30:04 +00:00
$(call Image/Build/CFE,$(1),96345GW2,6345)
Generate valid images for the bcm96348gw board SVN-Revision: 13298
2008-11-20 21:22:45 +00:00
$(call Image/Build/CFE,$(1),96348GW,6348)
Add definitions for bcm96350vw, add back bcm63xx mtd driver, be careful when testing with 8MB devices SVN-Revision: 13015
2008-10-20 12:55:28 +00:00
# Neufbox4
$(call Image/Build/CFE,$(1),96358VW,6358)
This patch is for broadcom 96348GW-11 boards, to get USB and Ethernet working. Also it patches mtd map driver for bcm96xx boards to be sure that CFE1CFE1 code is detected when booting from CFE. (#4201) SVN-Revision: 13213
2008-11-15 11:30:22 +00:00
# Comtrend 536
$(call Image/Build/CFE,$(1),96348GW-11,6348)
Merge premilinary Broadcom 63xx support (Livebox, Siemens SE515...) Kernel compiles fine, does not include linux-2.6.8.1 linux-mips specific fixes and neither includes broadcom changes to netfiler Needs testing of course :) SVN-Revision: 4084
2006-06-26 11:45:23 +00:00
endef
$(eval $(call BuildImage))