I am working on adding my own module to the build of buildroot using $BR2_External. The make file of my package is as follows,
##############################################################
#
# GPIO
#
##############################################################
GPIO_VERSION = '2851a05c9b613c1736f79faa185a11118b229852'
GPIO_SITE = '<URL of git repo>'
GPIO_SITE_METHOD = git
GPIO_GIT_SUBMODULES = YES
GPIO_MODULE_SUBDIRS = GPIO_driver/
# GPIO_MODULE_SUBDIRS += GPIO_driver/
# define LDD_BUILD_CMDS
# $(MAKE) $(TARGET_CONFIGURE_OPTS) -C $(#D)/misc-modules
# $(MAKE) $(TARGET_CONFIGURE_OPTS) -C $(#D)/scull
# endef
#
# # TODO add your writer, finder and finder-test utilities/scripts to the installation steps below
define GPIO_INSTALL_TARGET_CMDS
#module
# $(INSTALL) -m 0755 $(#D)/01_simple_LKM/* $(TARGET_DIR)/usr/bin
$(INSTALL) -m 0755 $(#D)/GPIO_driver/* $(TARGET_DIR)/usr/bin
endef
$(eval $(kernel-module))
$(eval $(generic-package))
This make file always pulls only a specific commit (mentioned in GPIO_VERSION variable) from gitHub. This is getting a little frustrating as, everytime I push new code to git I have to update the make file with the new commit number as well. So, is there any way to write the make file such that the most recent commit is pulled.
Related
Sometimes, We meet a situation that remote source code fetched by a recipe need to be modified so that suit a specific machine.
How do we create a patch for remote source code locally? After that everytime we build the recipe (even clean it all) we can patch the remote source code automatically.
For example, I have a special machine with architecture A which is not common, so the remote source code need to be modified so that support architecture A.
Suppose there was a file called utils.h (which is code that we fetched by example.bb from remote git repository)
#if defined(__x86_64__) || \
defined(__mips__) || \
defined(__powerpc__) || defined(__ppc__) || defined(__ppc64__) \
#define SOME_FUNCTIONALITY 1
Apparently I need to add archtecture A support in the file.
#if defined(__x86_64__) || \
defined(__mips__) || \
defined(__powerpc__) || defined(__ppc__) || defined(__ppc64__) || \
defined(__A__) \
#define SOME_FUNCTIONALITY 1
But if we just modified like that, next time we execute
bitbake -c cleanall example
bitbake example
then we get a unchanged copies again(which means we have to modify it again).
How do we create a Add-architecture-A-support.patch locally so that we can patch the remote source code automatically?
This is a simple one from answers.
(Note: If there was no git in the source code directory, before modifying the source code, you need to create a git repository and commit all in the top directory of the source code.)
git init # create a git repository
git add .
git commit -m "First commit" # first commit
After change the utils.h as above, we can check the git status. It usually looks like that.
$ git status
HEAD detached from 87b933c420
Changes not staged for commit:
(use "git add <file>..." to update what will be comitted)
(use "git checkout -- <file>..." to discard changes in working directory)
modified: ../../utils.h
...
no changes added to commit (use "git add" and/or "git commit -a")
Then we add and commit the change locally (usually we don't have the permission to push to upper stream).
$ git add utils.h
$ git commit -m "Patch test"
After that we can use git to create a patch for the recent commit.
$ git show >Add-architecture-A-support.patch
It will creat a patch in the current directory with contents looks like that
commit a79e523...
Author: 杨...
Date: ...
Patch test
diff --git a/somedir/utils.h b/somedir/utils.h
index 20bfd36c84..
--- a/somedir/utils.h
+++ b/somedir/utils.h
...
+ defined(__A__) \
...
Then we can move the patch to the local layer where the recipe stayed.
recipe-example
|-- example
| |-- Add-architecture-A-support.patch
|-- example.bb
And add the patch in example.bb with this.
SRC_URI += "\
file://Add-architecture-A-support.patch \
"
Work finished. (Also, if want to undo the local commit after creating the patch, you can use git reset HEAD^ utils.h. emmm, I think so, maybe there are some faults, just google it)
looking to disable some SELinux modules (set to off) and create others in modules.conf. I don't see an obvious way of updating modules.conf as I tried adding my changes as a modules.conf patch but it failed given that the modules.conf file gets built and is not just downloaded by BR so it is not available for patching like other things under the refpolicy directory:
Build window output:
refpolicy 2.20190609 PatchingApplying 0001-refpolicy-update-modules-conf.patch using patch:
can't find file to patch at input line 3
I did see in the log that there is a support/sedoctool.py that autogenerates the policy/modules.conf file so that the file is NOT patchable like most other things in the ref policy.
The relevant section of the buildroot/output/build/refpolicy-2.20190609/Makefile:
# policy building support tools
support := support
genxml := $(PYTHON) $(support)/segenxml.py
gendoc := $(PYTHON) $(support)/sedoctool.py
<...snip...>
########################################
#
# Create config files
#
conf: $(mod_conf) $(booleans) generate$(booleans) $(mod_conf): conf.intermediate.INTERMEDIATE: conf.intermediate
conf.intermediate: $(polxml)
#echo "Updating $(booleans) and $(mod_conf)"
$(verbose) $(gendoc) -b $(booleans) -m $(mod_conf) -x $(polxml)
Part of the hsmlinux build.log showing the sedoctool.py (gendoc) being run:
Updating policy/booleans.conf and policy/modules.conf
.../build-buildroot-sawshark/buildroot/output/host/usr/bin/python3 support/sedoctool.py -b policy/booleans.conf -m policy/modules.conf -x doc/policy.xml
I'm sure there is a standard way of doing this, just doesn't seem to be documented anywhere I can find.
Thanks.
Turns out that the sedoctool.py script is reading the doc/policy.xml. Looking at sedoctool.py:
#modules enabled and disabled values
MOD_BASE = "base"
MOD_ENABLED = "module"
MOD_DISABLED = "off"
<...snip...>
def gen_module_conf(doc, file_name, namevalue_list):
"""
Generates the module configuration file using the XML provided and the
previous module configuration.
"""
# If file exists, preserve settings and modify if needed.
# Otherwise, create it.
<...snip...>
mod_name = node.getAttribute("name")
mod_layer = node.parentNode.getAttribute("name")
<...snip...>
if mod_name and mod_layer:
file_name.write("# Layer: %s\n# Module: %s\n" % (mod_layer,mod_name))
if required:
file_name.write("# Required in base\n")
file_name.write("#\n")
if [mod_name, MOD_DISABLED] in namevalue_list:
file_name.write("%s = %s\n\n" % (mod_name, MOD_DISABLED))
# If the module is set as enabled.
elif [mod_name, MOD_ENABLED] in namevalue_list:
file_name.write("%s = %s\n\n" % (mod_name, MOD_ENABLED))
# If the module is set as base.
elif [mod_name, MOD_BASE] in namevalue_list:
file_name.write("%s = %s\n\n" % (mod_name, MOD_BASE))
So sedoctool.py has the nice feature of: "# If file exists, preserve settings and modify if needed." and modules.conf can just be added whole here via a complete file patch and the modules that are not desired set as "off" : refpolicy-2.20190609/policy/modules.conf and the script will update as needed based on desired policy.
One more detail is that in the next stage of the refpolicy Makefile (Building) the modules.conf with the updates is deleted in the beginning which kind of clashes with the ability of sedoctool to preserve the patched version of modules.conf...so patched the removal in the Building stage of the Makefile.
[7m>>> refpolicy 2.20190609 Building^[
<...snip...>
rm -f policy/modules.conf
The Makefile in refpolicy-2.20190609 has this line that I patched out because we are patching in our own modules.conf:
bare: clean
<...snip...>
$(verbose) rm -f $(mod_conf)
That patch looks like:
--- BUILDROOT/Makefile 2020-08-17 13:25:06.963804709 -0400
+++ FIX/Makefile 2020-08-17 19:25:29.540607763 -0400
## -636,7 +636,6 ##
$(verbose) rm -f $(modxml)
$(verbose) rm -f $(tunxml)
$(verbose) rm -f $(boolxml)
- $(verbose) rm -f $(mod_conf)
$(verbose) rm -f $(booleans)
$(verbose) rm -fR $(htmldir)
$(verbose) rm -f $(tags)
BTW,
Creating a patch with a complete new file in pp1:q!:
diff -crB --new-file pp0 pp1 > pp0.patch
I followed the steps mentioned in the link
https://github.com/RTEMS/rtems-libbsd
for sparc and 4.12 version.
# cd /opt
# mkdir RTEMS
# cd RTEMS
# sandbox="$PWD/sandbox"
# mkdir sandbox
# cd "$sandbox"
# git clone git://git.rtems.org/rtems-source-builder.git
# git clone git://git.rtems.org/rtems.git
# git clone git://git.rtems.org/rtems-libbsd.git
Build and install the tools.
# cd rtems-source-builder/rtems
# ../source-builder/sb-set-builder --prefix="$sandbox/rtems-4.12" 4.12/rtems-sparc
Bootstrap the RTEMS sources:
-----------------------------
# cd "$sandbox"
# cd rtems
# PATH="$sandbox/rtems-4.12/bin:$PATH"
# ./bootstrap
# cd "$sandbox" or cd ..
# mkdir b-sis
# cd b-sis
# "$sandbox/rtems/configure" --target=sparc-rtems4.12 --prefix="$sandbox/rtems-4.12" --disable-networking --enable-tests=samples --enable-rtemsbsp=sis
# make
# make install
Build and install rtems-libbsd
================================
# cd "$sandbox"
# cd rtems-libbsd
# git submodule init
# git submodule update rtems_waf
# waf configure --prefix="$sandbox/rtems-4.12" --rtems-bsps=sparc/sis
In this step I got an error
Setting top to : /home/subhilash/RTEMS/sandbox/rtems-libbsd
Setting out to : /home/subhilash/RTEMS/sandbox/rtems-libbsd/build
No valid arch/bsps found
The error means waf configure was unable to find sparc/sis installed in your prefix. Probably, configure and make failed without an obvious error due to the removal of the sis BSP from RTEMS during the 4.12 development cycle. Try using erc32 instead of sis.
You may get timelier responses about RTEMS by inquiring on the users mailing list.
You should also be aware that the sis simulator for the erc32 does not have a simulated NIC. And as I right this, the greth driver that you probably want for LEON CPUs doesn't yet have a driver in the rtems-libbsd TCP/IP stack. It is supported by the legacy IPV4 stack.
We would welcome a contribution to port this driver to the new stack.
I don't know if the greth driver is supported by qemu but the basic leon3 is supported.
I have a working Yocto image for a RaspberryPi3. I want to add 3 script files /etc/ppp/peers/. I would have thought that adding non-compiled files to the root file-system was a fairly generic thing to do but the only examples I can find are using compiled files and inheriting the autotools recipe.
Is there an example of how to add text files or script files to a Yocto root filesystem this somewhere?
Either a How To write up or an existing recipe that takes a set of text files and places them onto the target's rootfs.
I must be missing something because I cannot get the file files onto the system.
I tried using do_deploy, but that puts files into my ../tmp/deploy/images/raspberrypi3/etc/ppp/ which would be helpful for scripts to aid in image deployment. It is not what I want though as the scripts need to be on the target.
Running a do_install() with or without a blank do_compile() has not resulted in things getting onto the target either. Unless there is something about using ${sysconfdir} or ${IMAGE_ROOTFS} or ${S} or ${D} or ${DEPLOYDIR} or ${WORKDIR} which is particular to the Pi. I'd provide an example of my script but having changed it so many times in the last two days there is not much worth of sharing just one iteration.
Anything that resembles the following with;
${IMAGE_ROOTFS} possibly substituted for ${D} or missing
do_install replaced with do_deploy.
There are probably other permutations that I have tried.
#
# Copy the ppp script files for <vendor> chips to the target filesystem
# These files are based on the details provided in
#
SUMMARY = "PPP Scripts for ..."
SECTION = "net"
LICENSE = "MIT"
LIC_FILES_CHKSUM = "file://${COMMON_LICENSE_DIR}/MIT;md5=0835ade698e0bcf8506ecda2f7b4f302"
DESCRIPTION = "A set of Linux PPP scripts blar blar"
RDEPENDS_${PN} = "ppp"
SRC_URI += "file://<provider>-ppp"
SRC_URI += "file://<provider>-chat-connect"
SRC_URI += "file://<provider>-chat-disconnect"
S = "${WORKDIR}"
#PACKAGES =+ "${PN} ${PN}-staticdev"
#DEPLOYDIR = "${WORKDIR}/deploy-${PN}"
#D = "${DEPLOYDIR}"
inherit allarch
# Install script on target's root file-system
do_install () {
# Install init script and default settings
install -d ${IMAGE_ROOTFS}${sysconfdir}
install -d ${IMAGE_ROOTFS}${sysconfdir}/ppp/
install -d ${IMAGE_ROOTFS}${sysconfdir}/ppp/peers/
install -m 0755 ${S}/<provider>-ppp ${IMAGE_ROOTFS}${sysconfdir}/ppp/peers/
install -m 0755 ${S}/<provider>-chat-connect ${IMAGE_ROOTFS}${sysconfdir}/ppp/peers/
install -m 0755 ${S}/<provider>-chat-disconnect ${IMAGE_ROOTFS}${sysconfdir}/ppp/peers/
}
# Mark the files which are part of this package
FILES_${PN} += "${sysconfdir}/ppp/"
FILES_${PN} += "${sysconfdir}/ppp/peers/"
FILES_${PN} += "${sysconfdir}/ppp/peers/<provider>-ppp"
FILES_${PN} += "${sysconfdir}/ppp/peers/<provider>-chat-connect"
FILES_${PN} += "${sysconfdir}/ppp/peers/<provider>-chat-disconnect"
I can find a lot of helloworld.c and automate examples. There must be some basic ones for adding scripts somewhere? My googlefu is very weak, I blame a lingering cold.
You should be using install -m 0755 ${WORKDIR}/<provider>-ppp ${D}${sysconfdir}/ppp/peer in your recipe. Have you added the resulting package to your image recipe? You could look at ${WORKDIR}/packages-split/${PN} to confirm that your files have been properly packaged.
When I run git status, *.m.swp files are showing up in the "untracked list" because I currently have these files open in MacVim (The originals are MATLAB files with *.m file extensions).
I have tried adding *.m.swp, and various permutations of this, to my .gitignore file so that the files are ignored, but nothing seems to work for me.
See an example of git status output below:
git status
# On branch mybranch1
# Changes to be committed:
# (use "git reset HEAD <file>..." to unstage)
#
# modified: ../dir1/file1.m
# new file: file2.m
# new file: file3.m
# modified: file4.m
#
# Untracked files:
# (use "git add <file>..." to include in what will be committed)
#
# ../../dir2/.file5.m.swp
# ../dir1/.file6.m.swp
# ../dir1/.file1.m.swp
# ../dir1/.file7.m.swp
# ../dir1/.file8.swp
# .file9.m.swp
# .file4.m.swp
How can I get git to ignore these? Thanks in advance for any help!
.*.m.swp
should work: I have tested it in my msysgit1.7.4 environment.
So: not "*..." but ".*...".
Don't forget to add your modified .gitignore to the index before doing a new git status.
.*\\.m\\.swp
do this instead, since unescaped .'s match anything