* [Buildroot] [PATCH 01/10] manual: add section about storing the configuration.
From: Arnout Vandecappelle @ 2012-10-20 23:45 UTC (permalink / raw)
To: buildroot
In-Reply-To: <1350776731-8467-1-git-send-email-arnout@mind.be>
From: "Arnout Vandecappelle (Essensium/Mind)" <arnout@mind.be>
Reuse part of board-support.txt, and remove that one because it
was unused.
Signed-off-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>
---
docs/manual/adding-packages-directory.txt | 1 +
docs/manual/board-support.txt | 35 -----
docs/manual/customize-store.txt | 241 +++++++++++++++++++++++++++++
docs/manual/customize.txt | 2 +
4 files changed, 244 insertions(+), 35 deletions(-)
delete mode 100644 docs/manual/board-support.txt
create mode 100644 docs/manual/customize-store.txt
diff --git a/docs/manual/adding-packages-directory.txt b/docs/manual/adding-packages-directory.txt
index 4a96415..ce9c5ce 100644
--- a/docs/manual/adding-packages-directory.txt
+++ b/docs/manual/adding-packages-directory.txt
@@ -147,6 +147,7 @@ package.
The +.mk+ file
~~~~~~~~~~~~~~
+[[adding-packages-mk]]
Finally, here's the hardest part. Create a file named +libfoo.mk+. It
describes how the package should be downloaded, configured, built,
diff --git a/docs/manual/board-support.txt b/docs/manual/board-support.txt
deleted file mode 100644
index d1d9d63..0000000
--- a/docs/manual/board-support.txt
+++ /dev/null
@@ -1,35 +0,0 @@
-Creating your own board support
-===============================
-
-Creating your own board support in Buildroot allows users of a
-particular hardware platform to easily build a system that is known to
-work.
-
-To do so, you need to create a normal Buildroot configuration that
-builds a basic system for the hardware: toolchain, kernel, bootloader,
-filesystem and a simple Busybox-only userspace. No specific package
-should be selected: the configuration should be as minimal as
-possible, and should only build a working basic Busybox system for the
-target platform. You can of course use more complicated configurations
-for your internal projects, but the Buildroot project will only
-integrate basic board configurations. This is because package
-selections are highly application-specific.
-
-Once you have a known working configuration, run +make
-savedefconfig+. This will generate a minimal +defconfig+ file at the
-root of the Buildroot source tree. Move this file into the +configs/+
-directory, and rename it +MYBOARD_defconfig+.
-
-It is recommended to use as much as possible upstream versions of the
-Linux kernel and bootloaders, and to use as much as possible default
-kernel and bootloader configurations. If they are incorrect for your
-platform, we encourage you to send fixes to the corresponding upstream
-projects.
-
-However, in the mean time, you may want to store kernel or bootloader
-configuration or patches specific to your target platform. To do so,
-create a directory +board/MANUFACTURER+ and a subdirectory
-+board/MANUFACTURER/BOARDNAME+ (after replacing, of course,
-MANUFACTURER and BOARDNAME with the appropriate values, in lower case
-letters). You can then store your patches and configurations in these
-directories, and reference them from the main Buildroot configuration.
diff --git a/docs/manual/customize-store.txt b/docs/manual/customize-store.txt
new file mode 100644
index 0000000..cad1700
--- /dev/null
+++ b/docs/manual/customize-store.txt
@@ -0,0 +1,241 @@
+Storing the configuration
+-------------------------
+[[customize-store]]
+
+When you have a buildroot configuration that you are satisfied with
+and you want to move to share it with others, put it under revision
+control or move on to a different buildroot project, you need to store
+the configuration so it can be rebuilt later. The configuration that
+needs to be stored consists of the buildroot configuration, the
+configuration files for packages that you use (kernel, busybox,
+uClibc, ...), and your rootfs modifications.
+
+Basics for storing the configuration
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+[[customize-store-basics]]
+
+Buildroot configuration
+^^^^^^^^^^^^^^^^^^^^^^^
+
+For storing the buildroot configuration itself, buildroot offers the
+following command: +make savedefconfig+
+
+This strips the buildroot configuration down by removing configuration
+options that are at their default value. The result is stored in a file
+called +defconfig+. Copy this file to +foo_defconfig+ in the +configs+
+directory. The configuration can then be rebuilt by running
++make foo_defconfig+
+
+Alternatively, you can copy the file to any other place and rebuild with
++make BR2_DEFCONFIG=<path-to-defconfig> defconfig+
+
+
+Other package configuration
+^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+The configuration files for busybox, the linux kernel, uClibc and
+crosstool-NG should be stored as well. For each of these, a
+buildroot configuration option exists to point to an input configuration
+file, e.g. +BR2_LINUX_KERNEL_CUSTOM_CONFIG_FILE+. To save their
+configuration, set those configuration options to a path outside
+your output directory. Then, copy the configuration files
+to that path.
+
+Make sure that you create a configuration file 'before' changing
+the +BR2_LINUX_KERNEL_CUSTOM_CONFIG_FILE+ etc. options. Otherwise,
+buildroot will try to access this config file, which doesn't exist
+yet, and will fail.
+
+Buildroot provides a few helper targets to make the saving of
+configuration files easier.
+
+* +make linux-update-defconfig+ saves the linux configuration to the
+ path specified by +BR2_LINUX_KERNEL_CUSTOM_CONFIG_FILE+. It
+ simplifies the config file by removing default values. However,
+ this only works with kernels starting from 2.6.33. For earlier
+ kernels, use +make linux-update-config+.
+* +make busybox-update-config+ saves the busybox configuration to the
+ path specified by +BR2_PACKAGE_BUSYBOX_CONFIG+.
+* +make uclibc-update-config+ saves the uClibc configuration to the
+ path specified by +BR2_UCLIBC_CONFIG+.
+* For crosstool-NG, no helper exists so you have to copy the config
+ file manually to +BR2_TOOLCHAIN_CTNG_CONFIG+.
+
+
+Creating your own board support
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+Creating your own board support in Buildroot allows users of a
+particular hardware platform to easily build a system that is known to
+work.
+
+To do so, you need to create a normal Buildroot configuration that
+builds a basic system for the hardware: toolchain, kernel, bootloader,
+filesystem and a simple Busybox-only userspace. No specific package
+should be selected: the configuration should be as minimal as
+possible, and should only build a working basic Busybox system for the
+target platform. You can of course use more complicated configurations
+for your internal projects, but the Buildroot project will only
+integrate basic board configurations. This is because package
+selections are highly application-specific.
+
+Once you have a known working configuration, run +make
+savedefconfig+. This will generate a minimal +defconfig+ file at the
+root of the Buildroot source tree. Move this file into the +configs/+
+directory, and rename it +<boardname>_defconfig+.
+
+It is recommended to use as much as possible upstream versions of the
+Linux kernel and bootloaders, and to use as much as possible default
+kernel and bootloader configurations. If they are incorrect for your
+platform, we encourage you to send fixes to the corresponding upstream
+projects.
+
+However, in the mean time, you may want to store kernel or bootloader
+configuration or patches specific to your target platform. To do so,
+create a directory +board/<manufacturer>+ and a subdirectory
++board/<manufacturer>/<boardname>+. You can then store your patches
+and configurations in these directories, and reference them from the main
+Buildroot configuration.
+
+
+Step-by-step instructions for storing configuration inside the buildroot tree
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+To store the configuration for a specific product, device or
+application, it is advisable to use the same conventions as for the
+board support: put the buildroot defconfig in the +configs+ directory,
+and any other files in a subdirectory of the +boards+ directory. This
+section gives step-by-step instructions about how to do that. Of course,
+you can skip the steps that are not relevant for your use case.
+
+1. +make menuconfig+ to configure toolchain, packages and kernel.
+1. +make linux-menuconfig+ to update the kernel config, similar for
+ other configuration.
+1. +mkdir -p board/<manufacturer>/<boardname>+
+1. Set the following options to +board/<manufacturer>/<boardname>/<package>.config+
+ (as far as they are relevant):
+ * +BR2_LINUX_KERNEL_CUSTOM_CONFIG_FILE+
+ * +BR2_PACKAGE_BUSYBOX_CONFIG+
+ * +BR2_TOOLCHAIN_CTNG_CONFIG+
+ * +BR2_UCLIBC_CONFIG+
+ * +BR2_TARGET_AT91BOOTSTRAP3_CUSTOM_CONFIG_FILE+
+1. Write the configuration files:
+ * +make linux-update-defconfig+
+ * +make busybox-update-config+
+ * +cp <output>/build/build-toolchain/.config board/<manufacturer>/<boardname>/ctng.config+
+ * +make uclibc-update-config+
+ * +cp <output>/build/at91bootstrap3-*/.config board/<manufacturer>/<boardname>/at91bootstrap3.config+
+1. Create +board/<manufacturer>/<boardname>/fs-overlay+ and fill it
+ with additional files you need on your rootfs, e.g.
+ +board/<manufacturer>/<boardname>/etc/inittab+.
+1. Create a post-build script
+ +board/<manufacturer>/<boardname>/post-build.sh+. It should contain
+ the following command:
++
+------------
+rsync -a --exclude .empty --exclude '*~' ${0%/*}/fs-overlay $1
+------------
++
+1. Set +BR2_ROOTFS_POST_BUILD_SCRIPT+ to +board/<manufacturer>/<boardname>/post-build.sh+
+1. If additional setuid permissions have to be set or device nodes have
+ to be created, create +board/<manufacturer>/<boardname>/device_table.txt+
+ and add that path to +BR2_ROOTFS_DEVICE_TABLE+.
+1. +make savedefconfig+ to save the buildroot configuration.
+1. +cp defconfig configs/<boardname>_defconfig+
+
+
+Step-by-step instructions for storing configuration outside the buildroot tree
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+When you use buildroot in different projects, you may want to keep
+each project separate from buildroot itself. This requires an extra
+script or Makefile, but is otherwise easy to do with buildroot.
+Take the following steps (very similar to storing configuration inside
+the buildroot tree).
+
+1. +make menuconfig+ to configure toolchain, packages and kernel.
+1. +make linux-menuconfig+ to update the kernel config, similar for
+ other configuration.
+1. +mkdir -p <path-to-board-directory>+
+1. Set the following options to +$(BOARDDIR)/<package>.config+
+ (as far as they are relevant):
+ * +BR2_LINUX_KERNEL_CUSTOM_CONFIG_FILE+
+ * +BR2_PACKAGE_BUSYBOX_CONFIG+
+ * +BR2_TOOLCHAIN_CTNG_CONFIG+
+ * +BR2_UCLIBC_CONFIG+
+ * +BR2_TARGET_AT91BOOTSTRAP3_CUSTOM_CONFIG_FILE+
+1. Write the configuration files:
+ * +make linux-update-defconfig+
+ * +make busybox-update-config+
+ * +cp <output>/build/build-toolchain/.config <path-to-board-directory>/ctng.config+
+ * +make uclibc-update-config+
+ * +cp <output>/build/at91bootstrap3-*/.config <path-to-board-directory>/at91bootstrap3.config+
+1. Create +<path-to-board-directory>/fs-overlay+ and fill it
+ with additional files you need on your rootfs, e.g.
+ +<path-to-board-directory>/etc/inittab+.
+1. Create a post-build script
+ +<path-to-board-directory>/post-build.sh+. It should contain
+ the following command:
++
+------------
+rsync -a --exclude .empty --exclude '*~' ${0%/*}/fs-overlay $1
+------------
++
+1. Set +BR2_ROOTFS_POST_BUILD_SCRIPT+ to +$(BOARDDIR)/post-build.sh+
+1. If additional setuid permissions have to be set or device nodes have
+ to be created, create +<path-to-board-directory>/device_table.txt+
+ and add that path to +BR2_ROOTFS_DEVICE_TABLE+.
+1. +make savedefconfig+ to save the buildroot configuration.
+1. +cp defconfig <path-to-board-directory>/buildroot.config+
+1. Create a script or Makefile in the board directory that calls
+ buildroot:
++
+------------
+make -C <path-to-buildroot> O=<path-to-board-directory>/output BR2_DEFCONFIG=<path-to-board-directory>/buildroot.config defconfig
+make -C <path-to-buildroot> O=<path-to-board-directory>/output BOARDDIR=<path-to-board-directory>
+------------
++
+1. If you have additional proprietary binaries that are compiled
+ outside of buildroot, you can copy them into the rootfs as
+ part of +post-build.sh+
+1. If you need additional packages that are not available in buildroot
+ and that you don't want to make available (e.g. proprietary
+ applications), set +BR2_PACKAGE_OVERRIDE_FILE+ to
+ +$(BOARDDIR)/local.mk+. Add the following to +local.mk+ to
+ add package 'foo': +include foo/foo.mk+. Add the following in
+ +foo.mk+ (see xref:adding-packages-mk[Adding packages to buildroot]):
++
+------------
+BR2_PACKAGE_FOO = y
+FOO_OVERRIDE_SRCDIR = $(BOARDDIR)/foo
+
+define FOO_BUILD_CMDS
+ $(TARGET_MAKE_ENV) $(MAKE) -C $(@D)
+endef
+
+define FOO_INSTALL_TARGET_CMDS
+ $(INSTALL) -D -m 0755 $(@D)/foo $(TARGET_DIR)/usr/bin/foo
+endef
+
+$(eval $(generic-package))
+------------
++
+The +BR2_PACKAGE_FOO=y+ makes sure the package is always selected,
+so there is no need for a +Config.in+ file. The +FOO_OVERRIDE_SRCDIR+
+tells buildroot to fetch the sources from the +foo+ directory, and
+also makes sure that they are re-synchronized when you call +make
+foo-rebuild+
++
+1. To add patches to the linux build, set +BR2_LINUX_KERNEL_PATCH+ to
+ +$(BOARDDIR)/patches/linux+ and add your patches in that directory.
+ Similar for U-Boot, barebox, at91bootstrap and at91bootstrap3.
+1. To add patches for some other package 'foo', put them in the
+ +patches/foo+ directory and add the following to +local.mk+:
++
+------------
+define FOO_LOCAL_PATCHES
+support/scripts/apply-patches.sh $(@D) $(BOARDDIR)/patches/foo foo-\*.patch
+endef
+
+FOO_POST_PATCH_HOOKS += FOO_LOCAL_PATCHES
+------------
diff --git a/docs/manual/customize.txt b/docs/manual/customize.txt
index e8235de..6bd5811 100644
--- a/docs/manual/customize.txt
+++ b/docs/manual/customize.txt
@@ -10,3 +10,5 @@ include::customize-uclibc-config.txt[]
include::customize-kernel-config.txt[]
include::customize-toolchain.txt[]
+
+include::customize-store.txt[]
--
1.7.10.4
^ permalink raw reply related
* [Buildroot] [PATCH 00/10] Various simplifications of the buildroot configuration
From: Arnout Vandecappelle @ 2012-10-20 23:45 UTC (permalink / raw)
To: buildroot
Second attempt at my "project directory" series - this time leaving out
the project directory.
First patch adds a section to the manual to explain how to store the
configuration. It's basically Thomas's ELC-E presentation of last
year, but I've added a section for when it's out-of-tree. I also added
a very brief explanation about how to add packages and patches
out-of-tree.
The second patch adds the fs-overlay as a config option. Skeleton
is not deprecated this time round.
The next seven patches add the update-all-config target. At
Thomas S.'s suggestion, it's split up more. The first three
add missing -update-config targets and fix the busybox one;
the last three ones make sure that the buildroot default
config files are not overwritten by update-all-config.
The final patch adds the BR2_DEFCONFIG to .config, so
update-all-config can save directly to the right config file.
Regards,
Arnout
----------------------------------------------------------------
Arnout Vandecappelle (Essensium/Mind) (10):
manual: add section about storing the configuration.
target/generic: add filesystem overlay option
ctng: add ctng-update-config target
busybox: busybox-update-config should depend on busybox-configure
at91bootstrap3: add -update-config target
Add update-all-config target
busybox: update-all-config shouldn't update default busybox config
crosstool-ng: update-all-config shouldn't update default busybox config
uClibc: update-all-config shouldn't update default uClibc config
Make savedefconfig save to a configured file.
Config.in | 12 ++
Makefile | 32 +++-
boot/at91bootstrap3/at91bootstrap3.mk | 8 +
docs/manual/adding-packages-directory.txt | 1 +
docs/manual/board-support.txt | 35 ----
docs/manual/customize-rootfs.txt | 21 +-
docs/manual/customize-store.txt | 223 ++++++++++++++++++++++
docs/manual/customize.txt | 2 +
linux/linux.mk | 2 +
package/busybox/Config.in | 7 +-
package/busybox/busybox.mk | 13 +-
target/generic/Config.in | 11 ++
toolchain/toolchain-crosstool-ng/Config.in | 6 +-
toolchain/toolchain-crosstool-ng/crosstool-ng.mk | 21 +-
toolchain/uClibc/Config.in | 7 +-
toolchain/uClibc/uclibc.mk | 25 +--
16 files changed, 340 insertions(+), 86 deletions(-)
delete mode 100644 docs/manual/board-support.txt
create mode 100644 docs/manual/customize-store.txt
^ permalink raw reply
* [Buildroot] [PATCH] pkg-infra: produce legal info for proprietary packages
From: Richard Braun @ 2012-10-20 21:55 UTC (permalink / raw)
To: buildroot
In-Reply-To: <50830EB4.9090200@lucaceresoli.net>
On Sat, Oct 20, 2012 at 10:51:00PM +0200, Luca Ceresoli wrote:
> This in confusing. In your previous e-mail dated Sep 20you quoted the
> "Intel Software License Agreement", that one is supposed to accept before
> downloading the package from the web page:
Confusing indeed.
> What hurts me a little is that, if intel-microcode turns out to be
> redistributable, we would have no use case in mainline Buildroot that
> makes use of such a feature to the legal-info code.
That's what troubles me as well. I don't want to introduce dead code.
But if you do it as you seem to have planned, then I won't worry about
it and just wait for the feature.
I'll try to get some clarification from Intel in the meanwhile. Thanks
for your answer.
--
Richard Braun
^ permalink raw reply
* [Buildroot] [PATCH v2] Add package raspberrypi-firmware.
From: Belisko Marek @ 2012-10-20 21:27 UTC (permalink / raw)
To: buildroot
In-Reply-To: <87vce4u96t.fsf@macbook.be.48ers.dk>
On Sat, Oct 20, 2012 at 11:26 PM, Peter Korsgaard <jacmet@uclibc.org> wrote:
>>>>>> "Alex" == Alex Bradbury <asb@asbradbury.org> writes:
>
> Alex> On 20 October 2012 20:49, Belisko Marek <marek.belisko@gmail.com> wrote:
> >> Hi,
> >>
> >> just want to ask it this patch is in queue. It's month it was submitted.
>
> Yes, sorry about that. I just started looking at the patch now.
>
> Alex> I'd suggest it's probably not worth committing as-is, because this
> Alex> weekend we finally got rid of the multiple start_*.elf files.
>
> Ahh, interesting - So how is the memory layout then defined?
Described here:
https://github.com/raspberrypi/firmware/commit/c57ea9dd367f12bf4fb41b7b86806a2dc6281176
>
> --
> Bye, Peter Korsgaard
Marek
--
as simple and primitive as possible
-------------------------------------------------
Marek Belisko - OPEN-NANDRA
Freelance Developer
Ruska Nova Ves 219 | Presov, 08005 Slovak Republic
Tel: +421 915 052 184
skype: marekwhite
twitter: #opennandra
web: http://open-nandra.com
^ permalink raw reply
* [Buildroot] [PATCH v2] Add package raspberrypi-firmware.
From: Peter Korsgaard @ 2012-10-20 21:26 UTC (permalink / raw)
To: buildroot
In-Reply-To: <CA+wH294BUL_+SP55er=j9CgEf6N3VAQsrfbtjfvzywRVn2jb-Q@mail.gmail.com>
>>>>> "Alex" == Alex Bradbury <asb@asbradbury.org> writes:
Alex> On 20 October 2012 20:49, Belisko Marek <marek.belisko@gmail.com> wrote:
>> Hi,
>>
>> just want to ask it this patch is in queue. It's month it was submitted.
Yes, sorry about that. I just started looking at the patch now.
Alex> I'd suggest it's probably not worth committing as-is, because this
Alex> weekend we finally got rid of the multiple start_*.elf files.
Ahh, interesting - So how is the memory layout then defined?
--
Bye, Peter Korsgaard
^ permalink raw reply
* [Buildroot] [PATCH v2] Add package raspberrypi-firmware.
From: Belisko Marek @ 2012-10-20 21:24 UTC (permalink / raw)
To: buildroot
In-Reply-To: <CA+wH294BUL_+SP55er=j9CgEf6N3VAQsrfbtjfvzywRVn2jb-Q@mail.gmail.com>
Hi,
On Sat, Oct 20, 2012 at 11:10 PM, Alex Bradbury <asb@asbradbury.org> wrote:
> On 20 October 2012 20:49, Belisko Marek <marek.belisko@gmail.com> wrote:
>> Hi,
>>
>> just want to ask it this patch is in queue. It's month it was submitted.
>
> I'd suggest it's probably not worth committing as-is, because this
> weekend we finally got rid of the multiple start_*.elf files.
OK. I'll resend new version of patch. Thanks for info.
>
> Alex
Cheers,
marek
--
as simple and primitive as possible
-------------------------------------------------
Marek Belisko - OPEN-NANDRA
Freelance Developer
Ruska Nova Ves 219 | Presov, 08005 Slovak Republic
Tel: +421 915 052 184
skype: marekwhite
twitter: #opennandra
web: http://open-nandra.com
^ permalink raw reply
* [Buildroot] [PATCH v2] Add package raspberrypi-firmware.
From: Alex Bradbury @ 2012-10-20 21:10 UTC (permalink / raw)
To: buildroot
In-Reply-To: <CAAfyv37jOjsuDLX4jLendyTJq-Yv=eGMh8r+g4_Kvn_b74bHLg@mail.gmail.com>
On 20 October 2012 20:49, Belisko Marek <marek.belisko@gmail.com> wrote:
> Hi,
>
> just want to ask it this patch is in queue. It's month it was submitted.
I'd suggest it's probably not worth committing as-is, because this
weekend we finally got rid of the multiple start_*.elf files.
Alex
^ permalink raw reply
* [Buildroot] [PATCH] pkg-infra: produce legal info for proprietary packages
From: Luca Ceresoli @ 2012-10-20 20:51 UTC (permalink / raw)
To: buildroot
In-Reply-To: <20121016154229.GA15417@mail.sceen.net>
Richard Braun wrote:
> On Tue, Oct 02, 2012 at 03:41:38PM +0200, Luca Ceresoli wrote:
>> intel-microcode is clearly not fitting any of the two categories: we want to
>> describe its license, but we are not allowed to redistribute it freely, as
>> the license text reported from Richard seems to signify.
> Actually, there is a license text embedded in the microcode file. It
> reads :
>
> Redistribution. Redistribution and use in binary form, without modification, are
> permitted provided that the following conditions are met:
> .Redistributions must reproduce the above copyright notice and the following
> disclaimer in the documentation and/or other materials provided with the
> distribution.
> .Neither the name of Intel Corporation nor the names of its suppliers may be used
> to endorse or promote products derived from this software without specific prior
> written permission.
> .No reverse engineering, decompilation, or disassembly of this software is
> permitted.
> ."Binary form" includes any format commonly used for electronic conveyance
> which is a reversible, bit-exact translation of binary representation to ASCII or
> ISO text, for example, "uuencode."
This in confusing. In your previous e-mail dated Sep 20you quoted the
"Intel Software License Agreement", that one is supposed to accept before
downloading the package from the web page:
> "OEM LICENSE: You may reproduce and distribute the Software only as an
> integral part of or incorporated in Your product or as a standalone
> Software maintenance update for existing end users of Your products,
> excluding any other standalone products,
(http://downloadcenter.intel.com/confirm.aspx?httpDown=http://downloadmirror.intel.com/21385/eng/microcode-20120606.tgz&lang=eng&Dwnldid=21385&keyword=microcode)
This means we cannot redistribute the code, except as part of a product.
OTOH in the downloaded file, that one may simply download at
http://downloadmirror.intel.com/21385/eng/microcode-20120606.tgz without
accepting any agreement, contains the (different) license that you're
quoting right now, which would permit redistribution.
I used to think big companies can afford good lawyers, but this does not
seem to be happening at Intel.
> The disclaimer is a common 'this software is distributed "as is"'
> notice.
>
>
> I'm not exactly sure what "redistribute it freely" means here, since I'm
> much more used to free licenses, but it seems to me that redistribution
> is actually allowed as buildroot isn't in any way violating any of these
> conditions, as long as this text appears in the list of licenses, which
> my patch takes care of.
>
> Do you agree with that, and if yes, how would that change the rework
> proposal ?
I agree with your interpretation of the license in the .dat file, although
this does not clarify whether the applicable text is the Agreement on
the web page or this one license.
In both cases I don't think my proposal needs to be changed. The principle
is still true: one package may be redistributable or not. If it is not,
one may still want to describe its license in the _LICENSE variable with a
more descriptive text than "PROPRIETARY".
What hurts me a little is that, if intel-microcode turns out to be
redistributable, we would have no use case in mainline Buildroot that
makes use of such a feature to the legal-info code.
But I would go on and implement it anyway, since it could be useful to
other Buildroot users and not hurt the other ones.
Luca
^ permalink raw reply
* [Buildroot] [PATCH v2] Add package raspberrypi-firmware.
From: Belisko Marek @ 2012-10-20 19:49 UTC (permalink / raw)
To: buildroot
In-Reply-To: <506B1924.2090208@mind.be>
Hi,
just want to ask it this patch is in queue. It's month it was submitted.
---------- Forwarded message ----------
From: Arnout Vandecappelle <arnout@mind.be>
Date: Tue, Oct 2, 2012 at 6:41 PM
Subject: Re: [Buildroot] [PATCH v2] Add package raspberrypi-firmware.
To: Marek Belisko <marek.belisko@open-nandra.com>
Cc: buildroot at busybox.net
On 20/09/12 22:37, Marek Belisko wrote:
>
> Add support for raspberrypi bootloader and VideoCore.
>
> Signed-off-by: Marek Belisko<marek.belisko@open-nandra.com>
>
Still some comments, but these can be fixed while committing, so:
Acked-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>
(untested).
There should be a
---
(three dashes on a line by itself) between the SOB and the changelog,
otherwise the changelog becomes part of the commit message, which is
not what we want.
> Changes from V1:
> - fix names in Config.in (BR2_PACKAGE_BOOTLOADER_ARM192 to
> BR2_PACKAGE_RASPBERRYPI_BOOTLOADER_ARM192)
> - remove unnecessary SOURCE in mk file
> - fix installing VideoCore libraries and include
> - copy bootloader files directly to images directory
> - rework handling selection of GPU& ARM memory split
>
> (inspired by Yann patch)
> - bump to latest "turbo mode" firmware version
> (http://www.raspberrypi.org/archives/2008)
>
[snip]
> +if BR2_PACKAGE_RASPBERRYPI_BOOTLOADER
> +choice
> +
> +prompt "Memory& GPU split"
prompt should come immediately below choice (no empty line) and should be
indented with 1 tab.
> + default BR2_PACKAGE_RASPBERRYPI_BOOTLOADER_ARM192
> + help
> + Select how much memory use for system and
> + how much for GPU.
[snip]
> +ifeq ($(BR2_PACKAGE_RASPBERRYPI_VIDEOCORE),y)
> +RASPBERRYPI_FIRMWARE_INSTALL_STAGING = YES
> +
> +define RASPBERRYPI_FIRMWARE_INSTALL_STAGING_CMDS
> + cp -r -T $(@D)/hardfp/opt/vc/include $(STAGING_DIR)/usr/include
> + cp -r -T $(@D)/hardfp/opt/vc/lib $(STAGING_DIR)/usr/lib
I would prefer to use
rsync -a $(@D)/hardfp/opt/vc/include/ $(STAGING_DIR)/usr/include
but that's probably just personal preference.
[snip]
--
Arnout Vandecappelle arnout at mind be
Senior Embedded Software Architect +32-16-286540
Essensium/Mind http://www.mind.be
G.Geenslaan 9, 3001 Leuven, Belgium BE 872 984 063 RPR Leuven
LinkedIn profile: http://www.linkedin.com/in/arnoutvandecappelle
GPG fingerprint: 7CB5 E4CC 6C2E EFD4 6E3D A754 F963 ECAB 2450 2F1F
Thanks.
Marek
--
as simple and primitive as possible
-------------------------------------------------
Marek Belisko - OPEN-NANDRA
Freelance Developer
Ruska Nova Ves 219 | Presov, 08005 Slovak Republic
Tel: +421 915 052 184
skype: marekwhite
twitter: #opennandra
web: http://open-nandra.com
^ permalink raw reply
* [Buildroot] [PATCH 1/3] package/liburcu: Bump version to 0.7.5.
From: Peter Korsgaard @ 2012-10-20 19:11 UTC (permalink / raw)
To: buildroot
In-Reply-To: <CAFbHwiRP=LPt_yd3mYsZwGqMZfOU-ihB00qZP6VnNMKEix7N3g@mail.gmail.com>
>>>>> "Will" == Will Newton <will.newton@gmail.com> writes:
Will> Signed-off-by: Will Newton <will.newton@imgtec.com>
Committed all 3, thanks.
--
Bye, Peter Korsgaard
^ permalink raw reply
* [Buildroot] [git commit] package/lttng-babeltrace: Bump version to 1.0.0-rc6.
From: Peter Korsgaard @ 2012-10-20 19:11 UTC (permalink / raw)
To: buildroot
commit: http://git.buildroot.net/buildroot/commit/?id=3d48c41284d0daee0b70312022c7c6d31f730eb7
branch: http://git.buildroot.net/buildroot/commit/?id=refs/heads/master
Signed-off-by: Will Newton <will.newton@imgtec.com>
Signed-off-by: Peter Korsgaard <jacmet@sunsite.dk>
---
...-babeltrace-no-posix-fallocate-in-uclibc.patch} | 0
package/lttng-babeltrace/lttng-babeltrace.mk | 11 ++---------
2 files changed, 2 insertions(+), 9 deletions(-)
diff --git a/package/lttng-babeltrace/lttng-babeltrace-0.8-no-posix-fallocate-in-uclibc.patch b/package/lttng-babeltrace/lttng-babeltrace-no-posix-fallocate-in-uclibc.patch
similarity index 100%
rename from package/lttng-babeltrace/lttng-babeltrace-0.8-no-posix-fallocate-in-uclibc.patch
rename to package/lttng-babeltrace/lttng-babeltrace-no-posix-fallocate-in-uclibc.patch
diff --git a/package/lttng-babeltrace/lttng-babeltrace.mk b/package/lttng-babeltrace/lttng-babeltrace.mk
index 0f3200f..dd701df 100644
--- a/package/lttng-babeltrace/lttng-babeltrace.mk
+++ b/package/lttng-babeltrace/lttng-babeltrace.mk
@@ -1,14 +1,7 @@
-LTTNG_BABELTRACE_SITE = http://lttng.org/files/bundles/20111214/
-LTTNG_BABELTRACE_VERSION = 0.8
+LTTNG_BABELTRACE_SITE = http://lttng.org/files/babeltrace/
+LTTNG_BABELTRACE_VERSION = 1.0.0-rc6
LTTNG_BABELTRACE_SOURCE = babeltrace-$(LTTNG_BABELTRACE_VERSION).tar.bz2
-# Needed to fix libtool handling, otherwise the build fails when
-# building the ctf-parser-test program, which depends on libctf-ast.so
-# which itself depends on libbabeltrace_types.so.0 (and libtool gets
-# lost in the middle of this).
-LTTNG_BABELTRACE_AUTORECONF = YES
-HOST_LTTNG_BABELTRACE_AUTORECONF = YES
-
LTTNG_BABELTRACE_DEPENDENCIES = popt util-linux libglib2
$(eval $(autotools-package))
^ permalink raw reply related
* [Buildroot] [git commit] package/lttng-tools: Bump to version 2.0.4.
From: Peter Korsgaard @ 2012-10-20 19:10 UTC (permalink / raw)
To: buildroot
commit: http://git.buildroot.net/buildroot/commit/?id=aa866e877b98aabd54f25cfff14a31ee01716c3a
branch: http://git.buildroot.net/buildroot/commit/?id=refs/heads/master
Signed-off-by: Will Newton <will.newton@imgtec.com>
Signed-off-by: Peter Korsgaard <jacmet@sunsite.dk>
---
package/lttng-tools/lttng-tools.mk | 2 +-
1 files changed, 1 insertions(+), 1 deletions(-)
diff --git a/package/lttng-tools/lttng-tools.mk b/package/lttng-tools/lttng-tools.mk
index 7a0a7b4..fe48ac6 100644
--- a/package/lttng-tools/lttng-tools.mk
+++ b/package/lttng-tools/lttng-tools.mk
@@ -3,7 +3,7 @@
# LTTng-Tools: the trace control client
#
#############################################################
-LTTNG_TOOLS_VERSION = 2.0.1
+LTTNG_TOOLS_VERSION = 2.0.4
LTTNG_TOOLS_SITE = http://lttng.org/files/lttng-tools/
LTTNG_TOOLS_SOURCE = lttng-tools-$(LTTNG_TOOLS_VERSION).tar.bz2
^ permalink raw reply related
* [Buildroot] [git commit] package/liburcu: Bump version to 0.7.5.
From: Peter Korsgaard @ 2012-10-20 19:10 UTC (permalink / raw)
To: buildroot
commit: http://git.buildroot.net/buildroot/commit/?id=fdb79a1b57c0b33b517e5799befd9ca4840ead8a
branch: http://git.buildroot.net/buildroot/commit/?id=refs/heads/master
Signed-off-by: Will Newton <will.newton@imgtec.com>
Signed-off-by: Peter Korsgaard <jacmet@sunsite.dk>
---
package/liburcu/liburcu.mk | 2 +-
1 files changed, 1 insertions(+), 1 deletions(-)
diff --git a/package/liburcu/liburcu.mk b/package/liburcu/liburcu.mk
index bd69137..0bfba37 100644
--- a/package/liburcu/liburcu.mk
+++ b/package/liburcu/liburcu.mk
@@ -3,7 +3,7 @@
# liburc: userspace RCU (read-copy-update) library
#
#############################################################
-LIBURCU_VERSION = 0.7.3
+LIBURCU_VERSION = 0.7.5
LIBURCU_SITE = http://lttng.org/files/urcu/
LIBURCU_SOURCE = userspace-rcu-$(LIBURCU_VERSION).tar.bz2
^ permalink raw reply related
* [Buildroot] [PATCH] package/time: New package for GNU time.
From: Peter Korsgaard @ 2012-10-20 19:10 UTC (permalink / raw)
To: buildroot
In-Reply-To: <CAFbHwiRDGH0W+eRMGGui4gA+_iRhSYtHc2dhHwPDYf9uvqF5-w@mail.gmail.com>
>>>>> "Will" == Will Newton <will.newton@gmail.com> writes:
Will> Signed-off-by: Will Newton <will.newton@imgtec.com>
Committed, thanks.
--
Bye, Peter Korsgaard
^ permalink raw reply
* [Buildroot] [git commit] package/time: New package for GNU time.
From: Peter Korsgaard @ 2012-10-20 19:09 UTC (permalink / raw)
To: buildroot
commit: http://git.buildroot.net/buildroot/commit/?id=2073c21ec041b6404b19e17e0f139ddbc639a35f
branch: http://git.buildroot.net/buildroot/commit/?id=refs/heads/master
Signed-off-by: Will Newton <will.newton@imgtec.com>
Acked-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>
Tested-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>
Signed-off-by: Peter Korsgaard <jacmet@sunsite.dk>
---
package/Config.in | 1 +
package/time/Config.in | 6 ++++++
package/time/time.mk | 17 +++++++++++++++++
3 files changed, 24 insertions(+), 0 deletions(-)
diff --git a/package/Config.in b/package/Config.in
index d7b3db6..dd4c97e 100644
--- a/package/Config.in
+++ b/package/Config.in
@@ -672,6 +672,7 @@ source "package/logsurfer/Config.in"
source "package/screen/Config.in"
source "package/sudo/Config.in"
if BR2_PACKAGE_BUSYBOX_SHOW_OTHERS
+source "package/time/Config.in"
source "package/which/Config.in"
endif
source "package/xmlstarlet/Config.in"
diff --git a/package/time/Config.in b/package/time/Config.in
new file mode 100644
index 0000000..634d41a
--- /dev/null
+++ b/package/time/Config.in
@@ -0,0 +1,6 @@
+config BR2_PACKAGE_TIME
+ bool "time"
+ help
+ The GNU time utility.
+
+ http://savannah.gnu.org/projects/time/
diff --git a/package/time/time.mk b/package/time/time.mk
new file mode 100644
index 0000000..7b53286
--- /dev/null
+++ b/package/time/time.mk
@@ -0,0 +1,17 @@
+#############################################################
+#
+# time
+#
+#############################################################
+
+TIME_VERSION = 1.7
+TIME_SITE = $(BR2_GNU_MIRROR)/time
+TIME_CONF_ENV = ac_cv_func_wait3=yes
+
+# time uses an old version of automake that does not support
+# installing in DESTDIR.
+define TIME_INSTALL_TARGET_CMDS
+ install -D -m 755 $(@D)/time $(TARGET_DIR)/usr/bin/time
+endef
+
+$(eval $(autotools-package))
^ permalink raw reply related
* [Buildroot] [PATCH] toolchain: have check_glibc to search deeper for ld-linux
From: Peter Korsgaard @ 2012-10-20 19:06 UTC (permalink / raw)
To: buildroot
In-Reply-To: <1350073550-779-1-git-send-email-jean-mickael.guerin@6wind.com>
>>>>> "Jean-Mickael" == Jean-Mickael Guerin <jean-mickael.guerin@6wind.com> writes:
Jean-Mickael> ld-linux*.so may not be present in lib/ directory, it could be
Jean-Mickael> in lib32 and/or lib64 only. But check_glibc reports
Jean-Mickael> "Incorrect selection of the C library" in this case, which is
Jean-Mickael> not true.
Jean-Mickael> Fixed by extending the search to SYSROOT/*/*.
Committed, thanks.
--
Bye, Peter Korsgaard
^ permalink raw reply
* [Buildroot] [git commit] toolchain: have check_glibc to search deeper for ld-linux
From: Peter Korsgaard @ 2012-10-20 19:05 UTC (permalink / raw)
To: buildroot
commit: http://git.buildroot.net/buildroot/commit/?id=c8fd94218ee6284061552ca5de2256417a21bcb9
branch: http://git.buildroot.net/buildroot/commit/?id=refs/heads/master
ld-linux*.so may not be present in lib/ directory, it could be
in lib32 and/or lib64 only. But check_glibc reports
"Incorrect selection of the C library" in this case, which is
not true.
Fixed by extending the search to SYSROOT/*/*.
Signed-off-by: Jean-Mickael Guerin <jean-mickael.guerin@6wind.com>
Acked-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>
Signed-off-by: Peter Korsgaard <jacmet@sunsite.dk>
---
toolchain/helpers.mk | 2 +-
1 files changed, 1 insertions(+), 1 deletions(-)
diff --git a/toolchain/helpers.mk b/toolchain/helpers.mk
index d5ab91b..497cfff 100644
--- a/toolchain/helpers.mk
+++ b/toolchain/helpers.mk
@@ -196,7 +196,7 @@ check_glibc_feature = \
#
check_glibc = \
SYSROOT_DIR="$(strip $1)"; \
- if test `find $${SYSROOT_DIR}/lib/ -maxdepth 1 -name 'ld-linux*.so.*' -o -name 'ld.so.*' | wc -l` -eq 0 ; then \
+ if test `find $${SYSROOT_DIR}/ -maxdepth 2 -name 'ld-linux*.so.*' -o -name 'ld.so.*' | wc -l` -eq 0 ; then \
echo "Incorrect selection of the C library"; \
exit -1; \
fi; \
^ permalink raw reply related
* [Buildroot] [PATCH] hostapd: add fix for CVE-2012-4445
From: Peter Korsgaard @ 2012-10-20 19:05 UTC (permalink / raw)
To: buildroot
In-Reply-To: <1350044109-9853-1-git-send-email-gustavo@zacarias.com.ar>
>>>>> "Gustavo" == Gustavo Zacarias <gustavo@zacarias.com.ar> writes:
Gustavo> Signed-off-by: Gustavo Zacarias <gustavo@zacarias.com.ar>
Committed, thanks.
--
Bye, Peter Korsgaard
^ permalink raw reply
* [Buildroot] [git commit] hostapd: add fix for CVE-2012-4445
From: Peter Korsgaard @ 2012-10-20 19:04 UTC (permalink / raw)
To: buildroot
commit: http://git.buildroot.net/buildroot/commit/?id=7aed4faa9b0e6b21dba82ad5d3be5db796517580
branch: http://git.buildroot.net/buildroot/commit/?id=refs/heads/master
Signed-off-by: Gustavo Zacarias <gustavo@zacarias.com.ar>
Signed-off-by: Peter Korsgaard <jacmet@sunsite.dk>
---
package/hostapd/hostapd-cve-2012-4445.patch | 49 +++++++++++++++++++++++++++
1 files changed, 49 insertions(+), 0 deletions(-)
diff --git a/package/hostapd/hostapd-cve-2012-4445.patch b/package/hostapd/hostapd-cve-2012-4445.patch
new file mode 100644
index 0000000..034a458
--- /dev/null
+++ b/package/hostapd/hostapd-cve-2012-4445.patch
@@ -0,0 +1,49 @@
+From 567bacefd73782508bfe72d3624df495f0df4cd1 Mon Sep 17 00:00:00 2001
+From: Jouni Malinen <j@w1.fi>
+Date: Sun, 7 Oct 2012 20:06:29 +0300
+Subject: [PATCH] EAP-TLS server: Fix TLS Message Length validation
+
+EAP-TLS/PEAP/TTLS/FAST server implementation did not validate TLS
+Message Length value properly and could end up trying to store more
+information into the message buffer than the allocated size if the first
+fragment is longer than the indicated size. This could result in hostapd
+process terminating in wpabuf length validation. Fix this by rejecting
+messages that have invalid TLS Message Length value.
+
+This would affect cases that use the internal EAP authentication server
+in hostapd either directly with IEEE 802.1X or when using hostapd as a
+RADIUS authentication server and when receiving an incorrectly
+constructed EAP-TLS message. Cases where hostapd uses an external
+authentication are not affected.
+
+Thanks to Timo Warns for finding and reporting this issue.
+
+Signed-hostap: Jouni Malinen <j@w1.fi>
+intended-for: hostap-1
+(cherry picked from commit 586c446e0ff42ae00315b014924ec669023bd8de)
+---
+ src/eap_server/eap_server_tls_common.c | 8 ++++++++
+ 1 files changed, 8 insertions(+), 0 deletions(-)
+
+diff --git a/src/eap_server/eap_server_tls_common.c b/src/eap_server/eap_server_tls_common.c
+index e149ee3..2cbe700 100644
+--- a/src/eap_server/eap_server_tls_common.c
++++ b/src/eap_server/eap_server_tls_common.c
+@@ -224,6 +224,14 @@ static int eap_server_tls_process_fragment(struct eap_ssl_data *data,
+ return -1;
+ }
+
++ if (len > message_length) {
++ wpa_printf(MSG_INFO, "SSL: Too much data (%d bytes) in "
++ "first fragment of frame (TLS Message "
++ "Length %d bytes)",
++ (int) len, (int) message_length);
++ return -1;
++ }
++
+ data->tls_in = wpabuf_alloc(message_length);
+ if (data->tls_in == NULL) {
+ wpa_printf(MSG_DEBUG, "SSL: No memory for message");
+--
+1.7.4-rc1
+
^ permalink raw reply related
* [Buildroot] [PATCH] version update pcre 8.31
From: Peter Korsgaard @ 2012-10-20 19:04 UTC (permalink / raw)
To: buildroot
In-Reply-To: <1350035032-10549-1-git-send-email-alexander@mezon.ru>
>>>>> "Alexander" == Alexander Khryukin <alexander@mezon.ru> writes:
Alexander> Signed-off-by: Alexander Khryukin <alexander@mezon.ru>
Committed, thanks.
--
Bye, Peter Korsgaard
^ permalink raw reply
* [Buildroot] [git commit] pcre: bump version
From: Peter Korsgaard @ 2012-10-20 19:04 UTC (permalink / raw)
To: buildroot
commit: http://git.buildroot.net/buildroot/commit/?id=b1570168d86db59a0c9101ca757fc77b909ce3da
branch: http://git.buildroot.net/buildroot/commit/?id=refs/heads/master
Signed-off-by: Alexander Khryukin <alexander@mezon.ru>
Signed-off-by: Peter Korsgaard <jacmet@sunsite.dk>
---
package/pcre/pcre.mk | 2 +-
1 files changed, 1 insertions(+), 1 deletions(-)
diff --git a/package/pcre/pcre.mk b/package/pcre/pcre.mk
index 8fd6b21..9947c4b 100644
--- a/package/pcre/pcre.mk
+++ b/package/pcre/pcre.mk
@@ -4,7 +4,7 @@
#
#############################################################
-PCRE_VERSION = 8.30
+PCRE_VERSION = 8.31
PCRE_SITE = ftp://ftp.csx.cam.ac.uk/pub/software/programming/pcre
PCRE_LICENSE = BSD-3c
PCRE_LICENSE_FILES = LICENCE
^ permalink raw reply related
* [Buildroot] [PATCH] netsnmp: bump to version 5.7.2
From: Peter Korsgaard @ 2012-10-20 19:03 UTC (permalink / raw)
To: buildroot
In-Reply-To: <1350688324-25979-1-git-send-email-gustavo@zacarias.com.ar>
>>>>> "Gustavo" == Gustavo Zacarias <gustavo@zacarias.com.ar> writes:
Gustavo> Signed-off-by: Gustavo Zacarias <gustavo@zacarias.com.ar>
Committed, thanks.
--
Bye, Peter Korsgaard
^ permalink raw reply
* [Buildroot] [git commit] netsnmp: bump to version 5.7.2
From: Peter Korsgaard @ 2012-10-20 19:02 UTC (permalink / raw)
To: buildroot
commit: http://git.buildroot.net/buildroot/commit/?id=9546bc4576215df39183e583c6e6daa874e64f44
branch: http://git.buildroot.net/buildroot/commit/?id=refs/heads/master
Signed-off-by: Gustavo Zacarias <gustavo@zacarias.com.ar>
Signed-off-by: Peter Korsgaard <jacmet@sunsite.dk>
---
package/netsnmp/netsnmp-CVE-2012-2141.patch | 36 ---------------------------
package/netsnmp/netsnmp-swinst-crash.patch | 17 ------------
package/netsnmp/netsnmp.mk | 2 +-
3 files changed, 1 insertions(+), 54 deletions(-)
diff --git a/package/netsnmp/netsnmp-CVE-2012-2141.patch b/package/netsnmp/netsnmp-CVE-2012-2141.patch
deleted file mode 100644
index 1b34b9c..0000000
--- a/package/netsnmp/netsnmp-CVE-2012-2141.patch
+++ /dev/null
@@ -1,36 +0,0 @@
-From 4c5633f1603e4bd03ed05c37d782ec8911759c47 Mon Sep 17 00:00:00 2001
-From: Robert Story <rstory@freesnmp.com>
-Date: Mon, 14 May 2012 11:40:06 -0400
-Subject: [PATCH] NEWS: snmp: BUG: 3526549: CVE-2012-2141 Array index error leading to crash
-
----
- agent/mibgroup/agent/extend.c | 6 +++++-
- 1 files changed, 5 insertions(+), 1 deletions(-)
-
-diff --git a/agent/mibgroup/agent/extend.c b/agent/mibgroup/agent/extend.c
-index d00475f..1f8586a 100644
---- a/agent/mibgroup/agent/extend.c
-+++ b/agent/mibgroup/agent/extend.c
-@@ -1126,7 +1126,7 @@ _extend_find_entry( netsnmp_request_info *request,
- * ...and check the line requested is valid
- */
- line_idx = *table_info->indexes->next_variable->val.integer;
-- if (eptr->numlines < line_idx)
-+ if (line_idx < 1 || line_idx > eptr->numlines)
- return NULL;
- }
- return eptr;
-@@ -1299,6 +1299,10 @@ handle_nsExtendOutput2Table(netsnmp_mib_handler *handler,
- * Determine which line we've been asked for....
- */
- line_idx = *table_info->indexes->next_variable->val.integer;
-+ if (line_idx < 1 || line_idx > extension->numlines) {
-+ netsnmp_set_request_error(reqinfo, request, SNMP_NOSUCHINSTANCE);
-+ continue;
-+ }
- cp = extension->lines[line_idx-1];
-
- /*
---
-1.7.4.1
-
diff --git a/package/netsnmp/netsnmp-swinst-crash.patch b/package/netsnmp/netsnmp-swinst-crash.patch
deleted file mode 100644
index f7298a6..0000000
--- a/package/netsnmp/netsnmp-swinst-crash.patch
+++ /dev/null
@@ -1,17 +0,0 @@
-Fixes runtime error with uClibc (and possibly others).
-
-Original patch and bug description:
-http://sourceforge.net/tracker/?func=detail&aid=3436528&group_id=12694&atid=312694
-
---- net-snmp-5.7.1/agent/mibgroup/host/data_access/swinst_pkginfo.c
-+++ /home/fabled//net-snmp-5.7.1.patched/agent/mibgroup/host/data_access/swinst_pkginfo.c
-@@ -140,7 +140,8 @@
- memcpy( entry->swDate, cp, date_len );
- entry->swDate_len = date_len;
- }
-- closedir( d );
-+ if (d != NULL)
-+ closedir( d );
-
- DEBUGMSGTL(("swinst:load:arch"," loaded %d entries\n",
- (int)CONTAINER_SIZE(container)));
diff --git a/package/netsnmp/netsnmp.mk b/package/netsnmp/netsnmp.mk
index bbe0dbb..2670cf3 100644
--- a/package/netsnmp/netsnmp.mk
+++ b/package/netsnmp/netsnmp.mk
@@ -4,7 +4,7 @@
#
#############################################################
-NETSNMP_VERSION = 5.7.1
+NETSNMP_VERSION = 5.7.2
NETSNMP_SITE = http://downloads.sourceforge.net/project/net-snmp/net-snmp/$(NETSNMP_VERSION)
NETSNMP_SOURCE = net-snmp-$(NETSNMP_VERSION).tar.gz
NETSNMP_LICENSE = Various BSD-like
^ permalink raw reply related
* [Buildroot] [PATCH 10/13] Add update-all-config target
From: Arnout Vandecappelle @ 2012-10-20 16:52 UTC (permalink / raw)
To: buildroot
In-Reply-To: <5082D5B1.1020609@mind.be>
On 20/10/12 18:47, Arnout Vandecappelle wrote:
>> - forced copy
>
> Ah, except that cp -f doesn't mean forced copy; it means: remove an
> already existing file before the copy. That means that permissions are
> lost and if it's a symlink, it's not the file it points to that gets
> updated. So I think cp -f is _not_ what we want here...
Wrong again, Arnout :-( The -f only removes the symlink it the file
couldn't be written to in the first place, so it's OK to have it.
Regards,
Arnout
--
Arnout Vandecappelle arnout at mind be
Senior Embedded Software Architect +32-16-286540
Essensium/Mind http://www.mind.be
G.Geenslaan 9, 3001 Leuven, Belgium BE 872 984 063 RPR Leuven
LinkedIn profile: http://www.linkedin.com/in/arnoutvandecappelle
GPG fingerprint: 7CB5 E4CC 6C2E EFD4 6E3D A754 F963 ECAB 2450 2F1F
^ permalink raw reply
* [Buildroot] [PATCH 10/13] Add update-all-config target
From: Arnout Vandecappelle @ 2012-10-20 16:47 UTC (permalink / raw)
To: buildroot
In-Reply-To: <CAAXf6LWwO6OkBhz03vsBekTWv4=9Bb7eO-sskvzhdTCD58K7oA@mail.gmail.com>
On 14/10/12 20:45, Thomas De Schampheleire wrote:
>> +
>> > +ctng-update-config: $(CTNG_DIR)/.config
>> > + $(Q)cp $< $(CTNG_CONFIG_FILE)
>> > +
>> > +ifneq ($(CTNG_CONFIG_FILE),)
>> > +UPDATE_ALL_CONFIG_TARGETS += ctng-update-config
>> > +endif
> I would personally have split the addition of an update-config feature
> for crosstool-ng from this patch that wraps the different
> update-config targets in one update-all-config.
Good point.
> I wonder why ctng-update-config uses a different type of recipe than
> their counterparts in busybox, uclibc, ...
>
> For busybox it is:
>> > busybox-update-config:
>> > cp -f $(BUSYBOX_BUILD_CONFIG) $(BUSYBOX_CONFIG_FILE)
> while for ctng, as proposed above:
>> > +ctng-update-config: $(CTNG_DIR)/.config
>> > + $(Q)cp $< $(CTNG_CONFIG_FILE)
> Differences:
>
> - target dependency
linux and uclibc have it, only busybox is missing it.
> - $(Q)
My bad.
> - forced copy
Ah, except that cp -f doesn't mean forced copy; it means: remove an
already existing file before the copy. That means that permissions are
lost and if it's a symlink, it's not the file it points to that gets
updated. So I think cp -f is _not_ what we want here...
> I would say that since all these packages are using the same kind of
> config system, the way to save their config files should also line up.
> It's possible that the new ctng-update-config is better in some ways,
> but in this case I think the other ones should be updated as well
True, so I'll fix the use of -f in a separate patch. And add the
dependency for busybox as well.
Regards,
Arnout
--
Arnout Vandecappelle arnout at mind be
Senior Embedded Software Architect +32-16-286540
Essensium/Mind http://www.mind.be
G.Geenslaan 9, 3001 Leuven, Belgium BE 872 984 063 RPR Leuven
LinkedIn profile: http://www.linkedin.com/in/arnoutvandecappelle
GPG fingerprint: 7CB5 E4CC 6C2E EFD4 6E3D A754 F963 ECAB 2450 2F1F
^ permalink raw reply
page: next (older) | prev (newer) | latest
- recent:[subjects (threaded)|topics (new)|topics (active)]
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox