* [Buildroot] [PATCH] pkg-download.mk: Fix shallow clone
From: Stephan Hoffmann @ 2012-12-06 10:59 UTC (permalink / raw)
To: buildroot
In-Reply-To: <50C0432E.2080005@mind.be>
Am 06.12.2012 08:03, schrieb Arnout Vandecappelle:
> On 06/12/12 03:14, Danomi Manchego wrote:
>> So I'm wondering why the recent changes in this area used grep at
>> all, rather than asking ls-remote directly. I.e.:
>>
>> ((test `git ls-remote $($(PKG)_SITE) $($(PKG)_DL_VERSION) | cut -f
>> 2-` && \
>>
>> If there is feedback that the above is desirable, then I can send in
>> the patch.
>
> The argument of test should be quoted to avoid getting the same
> problem again
> when the ref matches more than once for some reason.
>
> Or even better: use the exit code:
>
> git ls-remote --exit-code $($(PKG)_SITE) $($(PKG)_DL_VERSION) >
> /dev/null 2>&1 && \
>
> (hopefully the --exit-code option exists in older gits as well...)
Hi,
sorry to say, but my git 1.7.0 does not know anything about a
--exit-code option and ls-remote always gives a exit code of 0 as long
as the repository can be connected.
Maybe its worth to think about adding a host-git to buildroot to
overcome such compatibility issues?
Kind regards
Stephan
>
> Regards,
> Arnout
>
--
reLinux - Stephan Hoffmann
Am Schmidtgrund 124 50765 K?ln
Tel. +49.221.95595-19 Fax: -64
www.reLinux.de sho at reLinux.de
^ permalink raw reply
* [Buildroot] [PATCH] pkg-download.mk: Fix shallow clone
From: Danomi Manchego @ 2012-12-06 11:28 UTC (permalink / raw)
To: buildroot
In-Reply-To: <50C07A7A.2070204@relinux.de>
Arnout - thanks for the feedback.
Stephan - thanks for looking into the the --exit-status on older gits.
Since the --exit-status isn't universally available, maybe the simplest
thing to do is to add quotes, and drop the "cut" and "grep":
((test "`git ls-remote $($(PKG)_SITE) $($(PKG)_DL_VERSION)`" && \
Assume no objections or better ideas come in, I'll send in a patch tonight.
Thanks,
Danomi -
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.busybox.net/pipermail/buildroot/attachments/20121206/651c3a47/attachment.html>
^ permalink raw reply
* [Buildroot] Support of iMX6 hardware ?
From: Thomas Petazzoni @ 2012-12-06 11:48 UTC (permalink / raw)
To: buildroot
In-Reply-To: <50C06206.6060905@wanadoo.fr>
Dear Thierry Bultel,
On Thu, 06 Dec 2012 10:14:46 +0100, Thierry Bultel wrote:
> I have 2 questions around the Buildroot support for iMX6 from
> Freescale, especially around graphics.
>
> 1) Is there current work on that for a coming release ?
Everything is in the Buildroot Git repository, and besides that I
haven't heard of people working on i.MX6 "advanced" features. We have a
defconfig for the nitrogen6x board that was done by Peter, but I guess
Peter hasn't investigated the entire graphics stack of i.MX6 for now.
It would indeed be really, really great to improve i.MX6 support in
Buildroot (and also for other SoCs).
> I have noticed that openbricks seem to already have integrated the
> imx6-lib and xdriver-xf86-video-imx
>
> 2) I have trying to get both of them integrated in buildroot
> The imx6-lib, which will be a 'generic-package', needs includes of
> the target kernel; these includes are not in the sysroot.
>
> What is the most appropriate way to proceed ? The underlying Makefile
> has got a "INCLUDE" var to be set, shall I point it to my
> linux-custom build directory ? That does not seem very good.
To me, it's the right solution. If you have userspace code that depends
on very specific kernel headers that are not part of a traditional
toolchain, then pointing to the kernel sources is the only solution.
Of course, it means that the package must "depends on BR2_LINUX_KERNEL"
and have <foobar>_DEPENDENCIES = linux.
Thomas
--
Thomas Petazzoni, Free Electrons
Kernel, drivers, real-time and embedded Linux
development, consulting, training and support.
http://free-electrons.com
^ permalink raw reply
* [Buildroot] [PATCH v3] Add support for custom package patches
From: spdawson at gmail.com @ 2012-12-06 12:46 UTC (permalink / raw)
To: buildroot
From: Simon Dawson <spdawson@gmail.com>
Add a configuration item to allow a space-separated list of directories
to be specified, from which custom package patches will be applied.
Signed-off-by: Simon Dawson <spdawson@gmail.com>
Acked-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>
---
v3: Add documentation, as suggested by Samuel Martin. Incorporate further
improvements suggested by Arnout Vandecappelle
v2: Incorporate improvements suggested by Arnout Vandecappelle, and
fix handling of the case where the new configuration item is not set
Config.in | 7 +++++++
docs/manual/customize-packages.txt | 15 +++++++++++++++
docs/manual/customize.txt | 2 ++
package/pkg-generic.mk | 29 +++++++++++++++++------------
4 files changed, 41 insertions(+), 12 deletions(-)
create mode 100644 docs/manual/customize-packages.txt
diff --git a/Config.in b/Config.in
index b319ac7..25910e5 100644
--- a/Config.in
+++ b/Config.in
@@ -419,6 +419,13 @@ config BR2_PACKAGE_OVERRIDE_FILE
as the source directory for a particular package. See the
Buildroot documentation for more details on this feature.
+config BR2_EXTRA_PACKAGE_PATCH_DIRS
+ string "extra package patch directories"
+ help
+ You may specify a space-separated list of directories containing
+ extra package patches. Note that any extra patches for a package
+ will be applied AFTER the Buildroot patch set for the package.
+
endmenu
source "toolchain/Config.in"
diff --git a/docs/manual/customize-packages.txt b/docs/manual/customize-packages.txt
new file mode 100644
index 0000000..4dea34a
--- /dev/null
+++ b/docs/manual/customize-packages.txt
@@ -0,0 +1,15 @@
+// -*- mode:doc -*- ;
+
+[[packages-custom]]
+Customizing packages
+~~~~~~~~~~~~~~~~~~~~
+
+It is sometimes useful to apply 'extra' patches to packages - over and
+above those provided in Buildroot. This might be used to support custom
+features in a project, for example, or when working on a new architecture.
+
+The +BR2_EXTRA_PACKAGE_PATCH_DIRS+ configuration file option can be
+used to specify a set of extra package patch directories. The option
+should be set to a space-separated list of directories containing the
+extra package patches. Note that any extra patches for a package
+will be applied *after* the Buildroot patch set for the package.
diff --git a/docs/manual/customize.txt b/docs/manual/customize.txt
index 1e6f4e8..a2705e5 100644
--- a/docs/manual/customize.txt
+++ b/docs/manual/customize.txt
@@ -12,3 +12,5 @@ include::customize-uclibc-config.txt[]
include::customize-kernel-config.txt[]
include::customize-toolchain.txt[]
+
+include::customize-packages.txt[]
diff --git a/package/pkg-generic.mk b/package/pkg-generic.mk
index fc05dc6..31493b9 100644
--- a/package/pkg-generic.mk
+++ b/package/pkg-generic.mk
@@ -69,6 +69,22 @@ else
@true # Nothing to do to source a local package
endif
+# Apply package patches from the specified directory
+define pkg-apply-patch-directory
+ if test -d $(1); then \
+ if test "$(wildcard $(1)/$(NAMEVER)*.patch*)"; then \
+ support/scripts/apply-patches.sh $(@D) $(1) $(NAMEVER)\*.patch $(NAMEVER)\*.patch.$(ARCH) || exit 1; \
+ else \
+ support/scripts/apply-patches.sh $(@D) $(1) $(RAWNAME)\*.patch $(RAWNAME)\*.patch.$(ARCH) || exit 1; \
+ if test -d $(1)/$(NAMEVER); then \
+ support/scripts/apply-patches.sh $(@D) $(1)/$(NAMEVER) \*.patch \*.patch.$(ARCH) || exit 1; \
+ fi; \
+ fi; \
+ fi;
+endef
+
+PACKAGE_PATCH_DIRS = $($(PKG)_DIR_PREFIX)/$(RAWNAME) $(call qstrip,$(BR2_EXTRA_PACKAGE_PATCH_DIRS))
+
# Patch
#
# The RAWNAME variable is the lowercased package name, which allows to
@@ -79,18 +95,7 @@ $(BUILD_DIR)/%/.stamp_patched:
@$(call MESSAGE,"Patching $($(PKG)_DIR_PREFIX)/$(RAWNAME)")
$(foreach hook,$($(PKG)_PRE_PATCH_HOOKS),$(call $(hook))$(sep))
$(if $($(PKG)_PATCH),support/scripts/apply-patches.sh $(@D) $(DL_DIR) $($(PKG)_PATCH))
- $(Q)( \
- if test -d $($(PKG)_DIR_PREFIX)/$(RAWNAME); then \
- if test "$(wildcard $($(PKG)_DIR_PREFIX)/$(RAWNAME)/$(NAMEVER)*.patch*)"; then \
- support/scripts/apply-patches.sh $(@D) $($(PKG)_DIR_PREFIX)/$(RAWNAME) $(NAMEVER)\*.patch $(NAMEVER)\*.patch.$(ARCH) || exit 1; \
- else \
- support/scripts/apply-patches.sh $(@D) $($(PKG)_DIR_PREFIX)/$(RAWNAME) $(RAWNAME)\*.patch $(RAWNAME)\*.patch.$(ARCH) || exit 1; \
- if test -d $($(PKG)_DIR_PREFIX)/$(RAWNAME)/$(NAMEVER); then \
- support/scripts/apply-patches.sh $(@D) $($(PKG)_DIR_PREFIX)/$(RAWNAME)/$(NAMEVER) \*.patch \*.patch.$(ARCH) || exit 1; \
- fi; \
- fi; \
- fi; \
- )
+ $(Q)$(foreach patchdir,$(PACKAGE_PATCH_DIRS),$(call pkg-apply-patch-directory,$(patchdir)))
$(foreach hook,$($(PKG)_POST_PATCH_HOOKS),$(call $(hook))$(sep))
$(Q)touch $@
--
1.7.10.4
^ permalink raw reply related
* [Buildroot] [PATCH] [RFC] new target: live filesystem
From: Jeremy Rosen @ 2012-12-06 13:22 UTC (permalink / raw)
To: buildroot
In-Reply-To: <50BF84E0.9090502@mind.be>
> > unless you want me to make it relative again... I believe absolute
> > (whith the possibility to use env vars to find the buildroot
> > location) makes most sense but that might be worth discussing
>
> Like most other paths in buildroot, it's relative to $(TOP_DIR)
> (i.e. the buildroot directory).
>
my bad, you're right about that, i'll correct that
>
> >>> +endef
> >>> +
> >>> +define ROOTFS_LIVE_INIT
> >>> + if [ -z $(shell which sudo) ] ; then echo "sudo seems to not
> >>> be
> >>> installed on the host system" ; false ; fi ; \
> >>
> >> This really should be checked in
> >> support/dependencies/dependencies.sh.
> >>
> >
> > I didn't know about that file, but it seems to be more about
> > dependencies for buildroot core than dependencies for a particular
> > config option...
>
> I forgot to add: similar to the java dependencies for classpath.
>
>
I didn't find anything like that in buildroot... there are some config-dependant checks in dependencies.sh (checking is locale is installed) but nothing class path related
am I missing something or should I do it in a way similar to locale ?
Regards
J?r?my Rosen
^ permalink raw reply
* [Buildroot] [PATCH] [RFC] new target: live filesystem
From: Jérémy Rosen @ 2012-12-06 13:36 UTC (permalink / raw)
To: buildroot
In-Reply-To: <1354531204-17481-1-git-send-email-jeremy.rosen@openwide.fr>
add a new target to deploy a live filesystem to be used with NFS or as a chroot
Signed-off-by: J?r?my Rosen <jeremy.rosen@openwide.fr>
---
v2 : implement Arnoult's suggestion, update manual entry
v3 : improve documentation for the chroot case, more suggestions by Arnoult
---
docs/manual/beyond-buildroot.txt | 20 ++++++++------------
fs/Config.in | 1 +
fs/live/Config.in | 14 ++++++++++++++
fs/live/live.mk | 19 +++++++++++++++++++
support/dependencies/dependencies.sh | 8 ++++++++
5 files changed, 50 insertions(+), 12 deletions(-)
create mode 100644 fs/live/Config.in
create mode 100644 fs/live/live.mk
diff --git a/docs/manual/beyond-buildroot.txt b/docs/manual/beyond-buildroot.txt
index a87b584..17ccc1a 100644
--- a/docs/manual/beyond-buildroot.txt
+++ b/docs/manual/beyond-buildroot.txt
@@ -9,19 +9,15 @@ Boot the generated images
NFS boot
~~~~~~~~
-To achieve NFS-boot, enable _tar root filesystem_ in the _Filesystem
-images_ menu.
+To achieve NFS-boot, enable _live root filesystem_ in the _Filesystem
+images_ menu and select a _live image location_ to choose where the live
+filesystem will be deployed. you can use _$(BINARIES_DIR)_ to easily
+build in +/path/to/output_dir/images+
-After a complete build, just run the following commands to setup the
-NFS-root directory:
+You will be asked for a password during the build. This is needed to create
+device entries in the target filesystem
--------------------
-sudo tar -xavf /path/to/output_dir/rootfs.tar -C /path/to/nfs_root_dir
--------------------
-
-Remember to add this path to +/etc/exports+.
-
-Then, you can execute a NFS-boot from your target.
+You will need to add the _live image location_ to +/etc/exports+.
Chroot
------
@@ -29,7 +25,7 @@ Chroot
If you want to chroot in a generated image, then there are few thing
you should be aware of:
-* you should setup the new root from the _tar root filesystem_ image;
+* you should use the _live root filesystem_ image;
* either the selected target architecture is compatible with your host
machine, or you should use some +qemu-*+ binary and correctly set it
diff --git a/fs/Config.in b/fs/Config.in
index da4c5ff..664d2f6 100644
--- a/fs/Config.in
+++ b/fs/Config.in
@@ -11,5 +11,6 @@ source "fs/romfs/Config.in"
source "fs/squashfs/Config.in"
source "fs/tar/Config.in"
source "fs/ubifs/Config.in"
+source "fs/live/Config.in"
endmenu
diff --git a/fs/live/Config.in b/fs/live/Config.in
new file mode 100644
index 0000000..a79f1dc
--- /dev/null
+++ b/fs/live/Config.in
@@ -0,0 +1,14 @@
+config BR2_TARGET_ROOTFS_LIVE
+ bool "live root filesystem"
+ help
+ Create a live image of the root filesystem that is directly
+ usable as over NFS or chroot.
+
+config BR2_TARGET_ROOTFS_LIVE_DEST
+ string "live image location"
+ depends on BR2_TARGET_ROOTFS_LIVE
+ default "$(BINARIES_DIR)/live"
+ help
+ The directory where the image should be stored.
+ This directory will be emptied and recreated
+
diff --git a/fs/live/live.mk b/fs/live/live.mk
new file mode 100644
index 0000000..52f7444
--- /dev/null
+++ b/fs/live/live.mk
@@ -0,0 +1,19 @@
+#############################################################
+#
+# Build the live root filesystem directory
+#
+#############################################################
+
+
+define ROOTFS_LIVE_CMD
+ sudo rsync -a --no-o --no-g --delete-during $(TARGET_DIR)/ \
+ $(BR2_TARGET_ROOTFS_LIVE_DEST)/
+endef
+
+define ROOTFS_LIVE_INIT
+ if [ -z $(shell which sudo) ] ; then echo "sudo seems to not be installed on the host system" ; false ; fi
+endef
+
+ROOTFS_LIVE_PRE_GEN_HOOKS += ROOTFS_LIVE_INIT
+
+$(eval $(call ROOTFS_TARGET,live))
diff --git a/support/dependencies/dependencies.sh b/support/dependencies/dependencies.sh
index 7a02512..ebaabbb 100755
--- a/support/dependencies/dependencies.sh
+++ b/support/dependencies/dependencies.sh
@@ -158,6 +158,7 @@ if grep ^BR2_TOOLCHAIN_BUILDROOT=y $CONFIG_FILE > /dev/null && \
exit 1 ;
fi
fi
+
if grep -q ^BR2_PACKAGE_CLASSPATH=y $CONFIG_FILE ; then
for prog in javac jar; do
if ! which $prog > /dev/null ; then
@@ -166,3 +167,10 @@ if grep -q ^BR2_PACKAGE_CLASSPATH=y $CONFIG_FILE ; then
fi
done
fi
+
+if grep ^BR2_TARGET_ROOTFS_LIVE=y $CONFIG_FILE > /dev/null ; then
+ if ! which sudo > /dev/null ; then
+ /bin/echo -e "\nYou need sudo installed on your build machine to build a live filesystem\n"
+ exit 1 ;
+ fi
+fi
--
1.7.10.4
^ permalink raw reply related
* [Buildroot] [PATCH] [RFC] new target: live filesystem
From: Jeremy Rosen @ 2012-12-06 13:36 UTC (permalink / raw)
To: buildroot
In-Reply-To: <3e51b13a-6944-4355-935f-76b551203888@zimbra2.corp.accelance.fr>
> >
> > I forgot to add: similar to the java dependencies for classpath.
> >
> >
>
> I didn't find anything like that in buildroot... there are some
> config-dependant checks in dependencies.sh (checking is locale is
> installed) but nothing class path related
>
> am I missing something or should I do it in a way similar to locale ?
>
ok, again it's my fault, I hadn't rebased in the last couple of days, v3 should be ready now
^ permalink raw reply
* [Buildroot] buildroot-2012.11 openssh: Bad trap error
From: Gustavo Zacarias @ 2012-12-06 14:56 UTC (permalink / raw)
To: buildroot
In-Reply-To: <50C06C77.7020509@shf.de>
On 12/06/2012 06:59 AM, Jan Schunke wrote:
> I am using buildroot-2012.11 on Powerpc 440 (Xilinx Virtex5) without FPU
>
> when I boot I get:
>
> ...
> Starting network...
> Generating RSA Key...
> [ 5.527467] Bad trap at PC: 48063f20, SR: 2d000, vector=2010
> Trace/breakpoint trap
> Generating RSA Key...
> [ 5.572257] Bad trap at PC: 48063f20, SR: 2d000, vector=2010
> Trace/breakpoint trap
> Generating DSA Key...
> THIS CAN TAKE A MINUTE OR TWO DEPENDING ON YOUR PROCESSOR!
>
> [ 5.594576] Bad trap at PC: 48063f20, SR: 2d000, vector=2010
> Trace/breakpoint trap
> Generating ECDSA Key...
> THIS CAN TAKE A MINUTE OR TWO DEPENDING ON YOUR PROCESSOR!
>
> [ 5.615732] Bad trap at PC: 48063f20, SR: 2d000, vector=2010
> Trace/breakpoint trap
> Starting sshd: [ 5.940254] Bad trap at PC: 48063f20, SR: 2d000,
> vector=2010
> Trace/breakpoint trap
> OK
> ...
>
> I can achieve a similar behaviour if I compile a program with -mhard-float.
>
> So I guess for one of the following packages softfloat is not set correctly.
>
> Maybe someone more competent can take a look into this.
>
> When I roll back
>
> openssh from 6.1p1 to 6.0p1
> openssl from 1.0.1c to 1.0.0j
> zlib from 1.27 to 1.26
>
> everything is fine.
Hi.
Ok, i managed to build a similar scenario as yours under qemu
(virtex-ml507 emulation).
With similar toolchain/versions/options i'm unable to hit the issue.
It's most likely that qemu is being too optimistic in the CPU emulation
department (handling the invalid opcodes for a 440 core).
Did you try switching just a single version?
I'm betting on openssl since it's got powerpc-specific optimizations
whereas zlib and openssh don't.
You could also try disabling the powerpc optimizations in
package/openssl/openssl.mk, just remove the block from ifeq
($(ARCH),powerpc) to the first endif (3 lines).
And let us know :)
Regards.
^ permalink raw reply
* [Buildroot] [PATCH] sconeserver: fix imagemagick build failures
From: spdawson at gmail.com @ 2012-12-06 15:12 UTC (permalink / raw)
To: buildroot
From: Simon Dawson <spdawson@gmail.com>
The sconeserver configure script uses Magick++-config to obtain compiler and
linker arguments for the imagemagick libraries. This doesn't work in
Buildroot, and causes build failures such as the following
http://autobuild.buildroot.net/results/7833edd60bbb2c291aea1efb5ccf407da7f4a51f/build-end.log
The upstream sconeserver code has been changed to use pkg-config directly,
instead of Magick++-config; this patch adapts the sconeserver package to
the upstream change.
Both this patch and the associated upstream change are based on a Buildroot
patch submitted by Samuel Martin: http://patchwork.ozlabs.org/patch/200901/
Signed-off-by: Simon Dawson <spdawson@gmail.com>
---
package/sconeserver/sconeserver.mk | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/package/sconeserver/sconeserver.mk b/package/sconeserver/sconeserver.mk
index 15a1edc..ab3d1fc 100644
--- a/package/sconeserver/sconeserver.mk
+++ b/package/sconeserver/sconeserver.mk
@@ -5,7 +5,7 @@
#############################################################
# Release 0.6.0 doesn't build cleanly, so use a recent
# Subversion trunk snapshot.
-SCONESERVER_VERSION = 180
+SCONESERVER_VERSION = 183
SCONESERVER_SITE = \
https://sconeserver.svn.sourceforge.net/svnroot/sconeserver/trunk
SCONESERVER_SITE_METHOD = svn
@@ -51,7 +51,7 @@ else
endif
ifeq ($(BR2_PACKAGE_SCONESERVER_HTTP_SCONESITE_IMAGE),y)
- SCONESERVER_DEPENDENCIES += imagemagick
+ SCONESERVER_DEPENDENCIES += imagemagick host-pkgconf
SCONESERVER_CONF_OPT += --with-sconesite-image
else
SCONESERVER_CONF_OPT += --without-sconesite-image
--
1.7.10.4
^ permalink raw reply related
* [Buildroot] [PATCH] fix cjson package
From: Viallard Anthony @ 2012-12-06 15:18 UTC (permalink / raw)
To: buildroot
Hi buildroot guys,
I've try to use cjson package but there a download problem. So, I fix up
the package Makefile and here the change :
diff --git a/package/cjson/cjson.mk b/package/cjson/cjson.mk
index ccc6010..728639d 100644
--- a/package/cjson/cjson.mk
+++ b/package/cjson/cjson.mk
@@ -3,8 +3,8 @@
# cjson
#
#############################################################
-CJSON_VERSION = undefined
-CJSON_SOURCE = cJSONFiles.zip
+CJSON_VERSION = 2011-10-10
+CJSON_SOURCE = cJSONFiles-$(CJSON_VERSION).zip
CJSON_SITE = http://downloads.sourceforge.net/project/cjson/
CJSON_INSTALL_STAGING = YES
CJSON_LICENSE = MIT
Cheers,
Viallard Anthony.
^ permalink raw reply related
* [Buildroot] Support of iMX6 hardware ?
From: Peter Korsgaard @ 2012-12-06 15:35 UTC (permalink / raw)
To: buildroot
In-Reply-To: <20121206124855.796d4d38@skate>
>>>>> "Thomas" == Thomas Petazzoni <thomas.petazzoni@free-electrons.com> writes:
Hi,
>> 1) Is there current work on that for a coming release ?
Thomas> Everything is in the Buildroot Git repository, and besides that I
Thomas> haven't heard of people working on i.MX6 "advanced" features. We have a
Thomas> defconfig for the nitrogen6x board that was done by Peter, but I guess
Thomas> Peter hasn't investigated the entire graphics stack of i.MX6 for now.
Indeed. I've only touched the open source bits. From the looks of it, I
doubt I will be able to spend much time on further imx6 support during
the 2013.02 cycle - So help would be ver much appreciated.
--
Bye, Peter Korsgaard
^ permalink raw reply
* [Buildroot] elfutils and the perf userspace tool
From: Bogdan Radulescu @ 2012-12-06 16:10 UTC (permalink / raw)
To: buildroot
In-Reply-To: <20121205163111.025fed52@skate>
Hello Thomas,
----- Original Message -----
> From: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
> To: Bogdan Radulescu <bogdan_radulescu99@yahoo.com>
> Cc: Stefan Fr?berg <stefan.froberg@petroprogram.com>; "buildroot at busybox.net" <buildroot@busybox.net>
> Sent: Wednesday, December 5, 2012 4:31 PM
> Subject: Re: [Buildroot] elfutils and the perf userspace tool
>
> Dear Bogdan Radulescu,
>
> On Wed, 5 Dec 2012 07:07:53 -0800 (PST), Bogdan Radulescu wrote:
>> I had a look at the yoctoproject patches and they are a bit confusing
>> to me. There are way to many patches. Some of them fix problems I
>> don't have and are very old. I decided to go with Stefans suggestions
>> and I could go further, but now I am stuck again. Yes, I am using
>> uClibc. Do you guys have a suggestion for this:
>> http://pastie.org/pastes/5484094/text
>>
>> I would really like to have perf in buildroot and this thing is in
>> the way.
>
> Did you notice that you can build perf without libelf, by passing
> NO_LIBELF=1 to perf's Makefile?
>
> Thomas
> --
> Thomas Petazzoni, Free Electrons
> Kernel, drivers, real-time and embedded Linux
> development, consulting, training and support.
> http://free-electrons.com
>
I noticed now that this is in the newer versions. I tried to build it without libelf even though it will be pretty crippled but still I didn't have any luck.
http://pastie.org/pastes/5489305/text
I tried with using 3.7-rc3. In 3.4.22 isn't patched for building without libelf.
On which version did you manage to build it for ARM?
Bogdan
^ permalink raw reply
* [Buildroot] elfutils and the perf userspace tool
From: Thomas Petazzoni @ 2012-12-06 16:11 UTC (permalink / raw)
To: buildroot
In-Reply-To: <1354810212.82454.YahooMailNeo@web120304.mail.ne1.yahoo.com>
Dear Bogdan Radulescu,
On Thu, 6 Dec 2012 08:10:12 -0800 (PST), Bogdan Radulescu wrote:
> I noticed now that this is in the newer versions. I tried to build it
> without libelf even though it will be pretty crippled but still I
> didn't have any luck. http://pastie.org/pastes/5489305/text
>
> I tried with using 3.7-rc3. In 3.4.22 isn't patched for building
> without libelf. On which version did you manage to build it for ARM?
I never built perf for ARM so far, so I can't really help,
unfortunately.
Thomas
--
Thomas Petazzoni, Free Electrons
Kernel, drivers, real-time and embedded Linux
development, consulting, training and support.
http://free-electrons.com
^ permalink raw reply
* [Buildroot] [PATCH] make legal-info: fails with OVERRIDE_SRCDIR
From: Peter Korsgaard @ 2012-12-06 17:07 UTC (permalink / raw)
To: buildroot
In-Reply-To: <50C072C7.30901@relinux.de>
>>>>> "Stephan" == Stephan Hoffmann <sho@relinux.de> writes:
Hi,
Stephan> Committed, thanks.
Stephan> Hi,
Stephan> sorry, but I forgot that Arnout's patch
Stephan> http://lists.busybox.net/pipermail/
Stephan> buildroot/2012-October/059910.html is not yet applied. My
Stephan> patch does not work without this.
Ahh, ok. Arnout, you mentioned you would rework and resend the patch:
http://lists.busybox.net/pipermail/buildroot/2012-October/059941.html
--
Bye, Peter Korsgaard
^ permalink raw reply
* [Buildroot] [PATCH v5] neard: new package
From: Peter Korsgaard @ 2012-12-06 17:30 UTC (permalink / raw)
To: buildroot
In-Reply-To: <877gow9uqw.fsf@dell.be.48ers.dk>
>>>>> "Peter" == Peter Korsgaard <jacmet@uclibc.org> writes:
>>>>> "spdawson" == spdawson <spdawson@gmail.com> writes:
spdawson> From: Simon Dawson <spdawson@gmail.com>
spdawson> Signed-off-by: Simon Dawson <spdawson@gmail.com>
spdawson> ---
spdawson> v5: Resend
spdawson> v4: No changes; patch series version bumped
spdawson> v3: Incorporated improvements suggested by Baruch Siach,
spdawson> add Arnout Vandecappelle's acks
spdawson> v2: Incorporated improvements suggested by Thomas Petazzoni
spdawson> +config BR2_PACKAGE_NEARD
spdawson> + bool "neard"
spdawson> + select BR2_PACKAGE_DBUS
spdawson> + select BR2_PACKAGE_LIBGLIB2
spdawson> + select BR2_PACKAGE_LIBNL
Peter> You forgot to add the dependencies for the toolchain options needed by
Peter> dbus/libglib2/libnl.
Peter> Committed with that fixed, thanks.
It seems to cause build failures though - Care to look at it?
http://autobuild.buildroot.net/results/eb2e3e8c604e26738f73b55e788b969d4a3dee23/build-end.log
--
Bye, Peter Korsgaard
^ permalink raw reply
* [Buildroot] [git commit] libpcap: fix canusb related build issue
From: Peter Korsgaard @ 2012-12-06 17:41 UTC (permalink / raw)
To: buildroot
commit: http://git.buildroot.net/buildroot/commit/?id=e63dcf20e2e457534346b144b32a2c4abe469a29
branch: http://git.buildroot.net/buildroot/commit/?id=refs/heads/master
Fixes http://autobuild.buildroot.net/results/1b57e479a643df11ea3b20f848085af4ef40799b
The libusb auto detection oddly enough succeeds, even though libusb isn't
available. Fix it by explicitly enabling/disabling canusb support.
Signed-off-by: Peter Korsgaard <jacmet@sunsite.dk>
---
package/libpcap/libpcap.mk | 10 ++++++++--
1 files changed, 8 insertions(+), 2 deletions(-)
diff --git a/package/libpcap/libpcap.mk b/package/libpcap/libpcap.mk
index d35c228..7e68f51 100644
--- a/package/libpcap/libpcap.mk
+++ b/package/libpcap/libpcap.mk
@@ -20,12 +20,18 @@ LIBPCAP_INSTALL_STAGING = YES
# configure script correctly handles --enable-shared/--disable-shared.
LIBPCAP_AUTORECONF = YES
-LIBPCAP_DEPENDENCIES = zlib \
- $(if $(BR2_PACKAGE_LIBUSB),libusb)
+LIBPCAP_DEPENDENCIES = zlib
LIBPCAP_CONF_ENV = ac_cv_linux_vers=2 \
ac_cv_header_linux_wireless_h=yes # configure misdetects this
LIBPCAP_CONF_OPT = --disable-yydebug --with-pcap=linux
+ifeq ($(BR2_PACKAGE_LIBUSB),y)
+LIBPCAP_CONF_OPT += --enable-canusb
+LIBPCAP_DEPENDENCIES += libusb
+else
+LIBPCAP_CONF_OPT += --disable-canusb
+endif
+
# microblaze needs -fPIC instead of -fpic
ifeq ($(BR2_microblaze),y)
LIBPCAP_CONF_ENV += CFLAGS="$(TARGET_CFLAGS) -fPIC"
^ permalink raw reply related
* [Buildroot] [PATCH] qemu/ppc-virtex-ml507: add new sample config
From: Gustavo Zacarias @ 2012-12-06 18:45 UTC (permalink / raw)
To: buildroot
Add new powerpc-virtex-ml507 sample qemu config.
Useful for powerpc softfloat testing.
Signed-off-by: Gustavo Zacarias <gustavo@zacarias.com.ar>
---
board/qemu/powerpc-virtex-ml507/readme.txt | 8 ++++++++
configs/qemu_ppc_virtex_ml507_defconfig | 22 ++++++++++++++++++++++
2 files changed, 30 insertions(+), 0 deletions(-)
create mode 100644 board/qemu/powerpc-virtex-ml507/readme.txt
create mode 100644 configs/qemu_ppc_virtex_ml507_defconfig
diff --git a/board/qemu/powerpc-virtex-ml507/readme.txt b/board/qemu/powerpc-virtex-ml507/readme.txt
new file mode 100644
index 0000000..0500a95
--- /dev/null
+++ b/board/qemu/powerpc-virtex-ml507/readme.txt
@@ -0,0 +1,8 @@
+Run the emulation with:
+
+ cp output/images/virtex440-ml507.dtb ppc.dtb
+ qemu-system-ppc -M virtex-ml507 -kernel output/images/vmlinux -m 256 -nographic -append "console=ttyS0"
+
+The login prompt will appear in the terminal that started Qemu.
+
+Tested with QEMU 1.2.0
diff --git a/configs/qemu_ppc_virtex_ml507_defconfig b/configs/qemu_ppc_virtex_ml507_defconfig
new file mode 100644
index 0000000..541d84c
--- /dev/null
+++ b/configs/qemu_ppc_virtex_ml507_defconfig
@@ -0,0 +1,22 @@
+# Architecture
+BR2_powerpc=y
+BR2_powerpc_440=y
+
+# Filesystem
+# BR2_TARGET_ROOTFS_TAR is not set
+BR2_TARGET_ROOTFS_INITRAMFS=y
+
+# Lock to 3.6 headers to avoid breaking with newer kernels
+BR2_KERNEL_HEADERS_3_6=y
+
+# Use soft float
+BR2_SOFT_FLOAT=y
+
+# Kernel
+BR2_LINUX_KERNEL=y
+BR2_LINUX_KERNEL_CUSTOM_VERSION=y
+BR2_LINUX_KERNEL_CUSTOM_VERSION_VALUE="3.6.9"
+BR2_LINUX_KERNEL_DEFCONFIG="44x/virtex5"
+BR2_LINUX_KERNEL_VMLINUX=y
+BR2_LINUX_KERNEL_DTS_SUPPORT=y
+BR2_LINUX_KERNEL_INTREE_DTS_NAME="44x/virtex440-ml507"
--
1.7.8.6
^ permalink raw reply related
* [Buildroot] [PATCHv2] qemu/ppc-virtex-ml507: add new sample config
From: Gustavo Zacarias @ 2012-12-06 18:59 UTC (permalink / raw)
To: buildroot
Add new powerpc-virtex-ml507 sample qemu config.
Useful for powerpc softfloat testing.
Signed-off-by: Gustavo Zacarias <gustavo@zacarias.com.ar>
---
board/qemu/powerpc-virtex-ml507/readme.txt | 8 ++++++++
configs/qemu_ppc_virtex_ml507_defconfig | 22 ++++++++++++++++++++++
2 files changed, 30 insertions(+), 0 deletions(-)
create mode 100644 board/qemu/powerpc-virtex-ml507/readme.txt
create mode 100644 configs/qemu_ppc_virtex_ml507_defconfig
diff --git a/board/qemu/powerpc-virtex-ml507/readme.txt b/board/qemu/powerpc-virtex-ml507/readme.txt
new file mode 100644
index 0000000..0500a95
--- /dev/null
+++ b/board/qemu/powerpc-virtex-ml507/readme.txt
@@ -0,0 +1,8 @@
+Run the emulation with:
+
+ cp output/images/virtex440-ml507.dtb ppc.dtb
+ qemu-system-ppc -M virtex-ml507 -kernel output/images/vmlinux -m 256 -nographic -append "console=ttyS0"
+
+The login prompt will appear in the terminal that started Qemu.
+
+Tested with QEMU 1.2.0
diff --git a/configs/qemu_ppc_virtex_ml507_defconfig b/configs/qemu_ppc_virtex_ml507_defconfig
new file mode 100644
index 0000000..6b4ea74
--- /dev/null
+++ b/configs/qemu_ppc_virtex_ml507_defconfig
@@ -0,0 +1,22 @@
+# Architecture
+BR2_powerpc=y
+BR2_powerpc_440=y
+
+# Filesystem
+# BR2_TARGET_ROOTFS_TAR is not set
+BR2_TARGET_ROOTFS_INITRAMFS=y
+
+# Lock to 3.6 headers to avoid breaking with newer kernels
+BR2_KERNEL_HEADERS_3_6=y
+
+# Use soft float
+BR2_SOFT_FLOAT=y
+
+# Kernel
+BR2_LINUX_KERNEL=y
+BR2_LINUX_KERNEL_CUSTOM_VERSION=y
+BR2_LINUX_KERNEL_CUSTOM_VERSION_VALUE="3.6.9"
+BR2_LINUX_KERNEL_DEFCONFIG="44x/virtex5"
+BR2_LINUX_KERNEL_VMLINUX=y
+BR2_LINUX_KERNEL_DTS_SUPPORT=y
+BR2_LINUX_KERNEL_INTREE_DTS_NAME="virtex440-ml507"
--
1.7.8.6
^ permalink raw reply related
* [Buildroot] [PATCH v5] neard: new package
From: Simon Dawson @ 2012-12-06 19:41 UTC (permalink / raw)
To: buildroot
In-Reply-To: <87d2yn142e.fsf@dell.be.48ers.dk>
Hi Peter.
On 6 December 2012 17:30, Peter Korsgaard <jacmet@sunsite.dk> wrote:
> It seems to cause build failures though - Care to look at it?
>
> http://autobuild.buildroot.net/results/eb2e3e8c604e26738f73b55e788b969d4a3dee23/build-end.log
The linux/nfc.h kernel header does not exist for pre-3.1.x kernels.
The toolchain used for the autobuild linked above uses 2.6.38 kernel
headers.
To my knowledge, we do not have a mechanism for marking a package as
requiring a [toolchain using a] minimum kernel header version. Is
there a simple way of achieving this, or is it going to require some
new infrastructure?
Simon.
^ permalink raw reply
* [Buildroot] [git commit] toolchain-external: bfin 2012R1-BETA toolchain no longer available
From: Peter Korsgaard @ 2012-12-06 19:43 UTC (permalink / raw)
To: buildroot
commit: http://git.buildroot.net/buildroot/commit/?id=36c76398dcb5ee604ae24f384d7055ef37b23568
branch: http://git.buildroot.net/buildroot/commit/?id=refs/heads/master
Replace it by -RC2 instead.
Signed-off-by: Peter Korsgaard <jacmet@sunsite.dk>
---
toolchain/toolchain-external/ext-tool.mk | 8 ++++----
1 files changed, 4 insertions(+), 4 deletions(-)
diff --git a/toolchain/toolchain-external/ext-tool.mk b/toolchain/toolchain-external/ext-tool.mk
index beba49f..6236235 100644
--- a/toolchain/toolchain-external/ext-tool.mk
+++ b/toolchain/toolchain-external/ext-tool.mk
@@ -245,10 +245,10 @@ TOOLCHAIN_EXTERNAL_SITE_2 = http://blackfin.uclinux.org/gf/download/frsrelease
TOOLCHAIN_EXTERNAL_SOURCE_2 = blackfin-toolchain-uclibc-full-2011R1-RC4.i386.tar.bz2
TOOLCHAIN_EXTERNAL_SOURCE = $(TOOLCHAIN_EXTERNAL_SOURCE_1) $(TOOLCHAIN_EXTERNAL_SOURCE_2)
else ifeq ($(BR2_TOOLCHAIN_EXTERNAL_BLACKFIN_UCLINUX_2012R1),y)
-TOOLCHAIN_EXTERNAL_SITE_1 = http://blackfin.uclinux.org/gf/download/frsrelease/544/9749/
-TOOLCHAIN_EXTERNAL_SOURCE_1 = blackfin-toolchain-2012R1-BETA1.i386.tar.bz2
-TOOLCHAIN_EXTERNAL_SITE_2 = http://blackfin.uclinux.org/gf/download/frsrelease/544/9773/
-TOOLCHAIN_EXTERNAL_SOURCE_2 = blackfin-toolchain-uclibc-full-2012R1-BETA1.i386.tar.bz2
+TOOLCHAIN_EXTERNAL_SITE_1 = http://blackfin.uclinux.org/gf/download/frsrelease/559/9858/
+TOOLCHAIN_EXTERNAL_SOURCE_1 = blackfin-toolchain-2012R1-RC2.i386.tar.bz2
+TOOLCHAIN_EXTERNAL_SITE_2 = http://blackfin.uclinux.org/gf/download/frsrelease/559/9866/
+TOOLCHAIN_EXTERNAL_SOURCE_2 = blackfin-toolchain-uclibc-full-2012R1-RC2.i386.tar.bz2
TOOLCHAIN_EXTERNAL_SOURCE = $(TOOLCHAIN_EXTERNAL_SOURCE_1) $(TOOLCHAIN_EXTERNAL_SOURCE_2)
else ifeq ($(BR2_TOOLCHAIN_EXTERNAL_XILINX_MICROBLAZEEL_V2),y)
TOOLCHAIN_EXTERNAL_SITE=http://git.xilinx.com/?p=xldk/microblaze_v2.0_le.git;a=blob;h=d7b493c5dbcc24ba9cc3be2e4c14d6d9701e6805;hb=00163583b771bb4e937632765dd0c5516b3e31c4;f=
^ permalink raw reply related
* [Buildroot] [git commit] alsa-lib: fix build on bfin
From: Peter Korsgaard @ 2012-12-06 19:43 UTC (permalink / raw)
To: buildroot
commit: http://git.buildroot.net/buildroot/commit/?id=9a6d3456f3e7965f30648fd09ef54b13109d210d
branch: http://git.buildroot.net/buildroot/commit/?id=refs/heads/master
Bfin toolchains don't have versionsort (which alsa-lib needs), so fake it
using alphasort instead.
Fixes http://autobuild.buildroot.net/results/cf760651b2e238842813bf0882011b5fbb898c0d
Signed-off-by: Peter Korsgaard <jacmet@sunsite.dk>
---
package/alsa-lib/alsa-lib.mk | 6 ++++++
1 files changed, 6 insertions(+), 0 deletions(-)
diff --git a/package/alsa-lib/alsa-lib.mk b/package/alsa-lib/alsa-lib.mk
index eddd3d2..ba222db 100644
--- a/package/alsa-lib/alsa-lib.mk
+++ b/package/alsa-lib/alsa-lib.mk
@@ -70,6 +70,12 @@ ifeq ($(BR2_SOFT_FLOAT),y)
ALSA_LIB_CONF_OPT += --with-softfloat
endif
+ifeq ($(BR2_bfin),y)
+# blackfin external toolchains don't have versionsort. Fake it using alphasort
+# instead
+ALSA_LIB_CFLAGS += -Dversionsort=alphasort
+endif
+
ALSA_LIB_CONF_ENV = CFLAGS="$(ALSA_LIB_CFLAGS)" \
LDFLAGS="$(TARGET_LDFLAGS) -lm"
^ permalink raw reply related
* [Buildroot] [PATCH - v2] tcllib: new package
From: Davide Viti @ 2012-12-06 21:33 UTC (permalink / raw)
To: buildroot
I've updated the patch applying the fixes suggested during the review.
In particular:
- remove depends on BR2_PACKAGE_TCL from Config.in
- fix download URL and fetch smaller bz2 upstream tarball
- remove $(eval $(host-autotools-package))
- depend on host-tcl
- invoke "configure" instead of installer.tcl
- remove TCLLIB_LICENSE (is not BSD-2c). Just specify
TCLLIB_LICENSE_FILES
thanx in advance for your feedback
regards,
Davide
^ permalink raw reply
* [Buildroot] [PATCH] tcllib: new package
From: Davide Viti @ 2012-12-06 21:33 UTC (permalink / raw)
To: buildroot
In-Reply-To: <1354829598-16782-1-git-send-email-zinosat@tiscali.it>
CC: yann.morin.1998 at free.fr
Signed-off-by: Davide Viti <zinosat@tiscali.it>
---
package/Config.in | 5 +++++
package/tcllib/Config.in | 11 +++++++++++
package/tcllib/tcllib.mk | 12 ++++++++++++
3 files changed, 28 insertions(+), 0 deletions(-)
create mode 100644 package/tcllib/Config.in
create mode 100644 package/tcllib/tcllib.mk
diff --git a/package/Config.in b/package/Config.in
index 74e439e..cad1221 100644
--- a/package/Config.in
+++ b/package/Config.in
@@ -313,6 +313,11 @@ endmenu
endif
source "package/ruby/Config.in"
source "package/tcl/Config.in"
+if BR2_PACKAGE_TCL
+menu "tcl libraries/modules"
+source "package/tcllib/Config.in"
+endmenu
+endif
endmenu
menu "Libraries"
diff --git a/package/tcllib/Config.in b/package/tcllib/Config.in
new file mode 100644
index 0000000..b0a4ad0
--- /dev/null
+++ b/package/tcllib/Config.in
@@ -0,0 +1,11 @@
+config BR2_PACKAGE_TCLLIB
+ bool "tcllib"
+ help
+ Tcllib is a collection of utility modules for Tcl. These
+ modules provide a wide variety of functionality, from
+ implementations of standard data structures to
+ implementations of common networking protocols. The intent
+ is to collect commonly used function into a single library,
+ which users can rely on to be available and stable
+
+ http://tcl.activestate.com/software/tcllib/
diff --git a/package/tcllib/tcllib.mk b/package/tcllib/tcllib.mk
new file mode 100644
index 0000000..fa8968d
--- /dev/null
+++ b/package/tcllib/tcllib.mk
@@ -0,0 +1,12 @@
+#############################################################
+#
+# tcllib
+#
+#############################################################
+TCLLIB_VERSION = 1.13
+TCLLIB_SOURCE = tcllib-$(TCLLIB_VERSION).tar.bz2
+TCLLIB_SITE = http://downloads.sourceforge.net/project/tcllib/tcllib/$(TCLLIB_VERSION)
+TCLLIB_LICENSE_FILES = license.terms
+TCLLIB_DEPENDENCIES = host-tcl tcl
+
+$(eval $(autotools-package))
--
1.7.2.5
^ permalink raw reply related
* [Buildroot] [PATCH - v3] tcllib: new package
From: Davide Viti @ 2012-12-06 22:46 UTC (permalink / raw)
To: buildroot
As dicussed on IRC, here follows another update:
- Set TCLLIB_LICENSE to "tcllib license"
DESCRIPTION.txt says "Rights: BSD" without specifying the BSD clause
- remove "tcl" from the list of dependencies, since "tcllib" can be
selected only if tcl has been selected
^ permalink raw reply
* [Buildroot] [PATCH] tcllib: new package
From: Davide Viti @ 2012-12-06 22:46 UTC (permalink / raw)
To: buildroot
In-Reply-To: <1354833972-17700-1-git-send-email-zinosat@tiscali.it>
Signed-off-by: Davide Viti <zinosat@tiscali.it>
---
package/Config.in | 5 +++++
package/tcllib/Config.in | 11 +++++++++++
package/tcllib/tcllib.mk | 13 +++++++++++++
3 files changed, 29 insertions(+), 0 deletions(-)
create mode 100644 package/tcllib/Config.in
create mode 100644 package/tcllib/tcllib.mk
diff --git a/package/Config.in b/package/Config.in
index 74e439e..cad1221 100644
--- a/package/Config.in
+++ b/package/Config.in
@@ -313,6 +313,11 @@ endmenu
endif
source "package/ruby/Config.in"
source "package/tcl/Config.in"
+if BR2_PACKAGE_TCL
+menu "tcl libraries/modules"
+source "package/tcllib/Config.in"
+endmenu
+endif
endmenu
menu "Libraries"
diff --git a/package/tcllib/Config.in b/package/tcllib/Config.in
new file mode 100644
index 0000000..b0a4ad0
--- /dev/null
+++ b/package/tcllib/Config.in
@@ -0,0 +1,11 @@
+config BR2_PACKAGE_TCLLIB
+ bool "tcllib"
+ help
+ Tcllib is a collection of utility modules for Tcl. These
+ modules provide a wide variety of functionality, from
+ implementations of standard data structures to
+ implementations of common networking protocols. The intent
+ is to collect commonly used function into a single library,
+ which users can rely on to be available and stable
+
+ http://tcl.activestate.com/software/tcllib/
diff --git a/package/tcllib/tcllib.mk b/package/tcllib/tcllib.mk
new file mode 100644
index 0000000..471720d
--- /dev/null
+++ b/package/tcllib/tcllib.mk
@@ -0,0 +1,13 @@
+#############################################################
+#
+# tcllib
+#
+#############################################################
+TCLLIB_VERSION = 1.13
+TCLLIB_SOURCE = tcllib-$(TCLLIB_VERSION).tar.bz2
+TCLLIB_SITE = http://downloads.sourceforge.net/project/tcllib/tcllib/$(TCLLIB_VERSION)
+TCLLIB_LICENSE = tcllib license
+TCLLIB_LICENSE_FILES = license.terms
+TCLLIB_DEPENDENCIES = host-tcl
+
+$(eval $(autotools-package))
--
1.7.2.5
^ permalink raw reply related
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