Buildroot Archive on lore.kernel.org
 help / color / mirror / Atom feed
* [Buildroot] [git commit] dependencies.sh: only javac and jar are needed by classpath
From: Peter Korsgaard @ 2012-12-04 20:09 UTC (permalink / raw)
  To: buildroot

commit: http://git.buildroot.net/buildroot/commit/?id=694050ffa2f3cbbfe7234b434f16f02cb93fa940
branch: http://git.buildroot.net/buildroot/commit/?id=refs/heads/master

Signed-off-by: Peter Korsgaard <jacmet@sunsite.dk>
---
 support/dependencies/dependencies.sh |    2 +-
 1 files changed, 1 insertions(+), 1 deletions(-)

diff --git a/support/dependencies/dependencies.sh b/support/dependencies/dependencies.sh
index 1b79b23..7a02512 100755
--- a/support/dependencies/dependencies.sh
+++ b/support/dependencies/dependencies.sh
@@ -159,7 +159,7 @@ if grep ^BR2_TOOLCHAIN_BUILDROOT=y $CONFIG_FILE > /dev/null && \
    fi
 fi
 if grep -q ^BR2_PACKAGE_CLASSPATH=y $CONFIG_FILE ; then
-    for prog in java javac jar; do
+    for prog in javac jar; do
 	if ! which $prog > /dev/null ; then
 	    /bin/echo -e "\nYou must install '$prog' on your build machine\n" >&2
 	    exit 1

^ permalink raw reply related
* [Buildroot] [git commit] package: add classpath
From: Thomas Petazzoni @ 2012-12-04 19:43 UTC (permalink / raw)
  To: buildroot
In-Reply-To: <20121204190928.0DF9899716@busybox.osuosl.org>

Dear Peter Korsgaard,

On Tue, 4 Dec 2012 10:21:05 -0800, Peter Korsgaard wrote:
> commit:
> http://git.buildroot.net/buildroot/commit/?id=5062f530081913bd0487f0f879be180f30836ab8
> branch:
> http://git.buildroot.net/buildroot/commit/?id=refs/heads/master
> 
> Requires java, javac and jar on the build machine, similar to how we
> require gcc/g++.

So, it needs additional checks in support/dependencies/dependencies.sh
I guess.

> classpath: fixup

Doh ? :-)

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] elfutils and the perf userspace tool
From: Thomas Petazzoni @ 2012-12-04 19:42 UTC (permalink / raw)
  To: buildroot
In-Reply-To: <50BE4D1C.3020702@petroprogram.com>

Dear Stefan Fr?berg,

On Tue, 04 Dec 2012 21:21:00 +0200, Stefan Fr?berg wrote:
> And there was some other (uClibc related) stuff that needed patching
> too that I can't now
> remember
> 
> All in all, elfutils belongs sadly to the software category:
> 
>  "Im doing things my own way, Im using non-portable glibc stuff and
> I don't give a crab about portability. Nah-nah-naaa"
> 
> :-(

Don't reinvent the wheel:

  http://git.yoctoproject.org/cgit/cgit.cgi/poky/tree/meta/recipes-devtools/elfutils/elfutils/uclibc-support.patch

And possibly other patches in:

 http://git.yoctoproject.org/cgit/cgit.cgi/poky/tree/meta/recipes-devtools/elfutils/elfutils

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] xapp_init: bad substitution of mcookie path in startx
From: Peter Korsgaard @ 2012-12-04 19:37 UTC (permalink / raw)
  To: buildroot
In-Reply-To: <50BDEE24.6010601@wanadoo.fr>

>>>>> "Thierry" == Thierry Bultel <thierry.bultel@wanadoo.fr> writes:

 Thierry> Hi,
 Thierry> The generated "startx" has the host path for mcookie:



 Thierry> ----------
 Thierry>    esac

 Thierry>     authdisplay=${display:-:0}


 Thierry> mcookie=`/nous/BASYSTEMES/Buildroot/output.imx6neon/host/usr/bin/mcookie`

What buildroot version are you using? It seems to be correct here with
current git:

grep bin/mcookie output/target/usr/bin/startx
    mcookie=`/usr/bin/mcookie`

-- 
Bye, Peter Korsgaard

^ permalink raw reply
* [Buildroot] [PATCH] package/crosstool-ng: update to 1.17.0
From: Thomas Petazzoni @ 2012-12-04 19:35 UTC (permalink / raw)
  To: buildroot
In-Reply-To: <201212041909.25267.yann.morin.1998@free.fr>

Dear Yann E. MORIN,

On Tue, 4 Dec 2012 19:09:24 +0100, Yann E. MORIN wrote:

> > Thanks for doing this work. I think ct-ng now supports minimal
> > defconfig files. Would it make sense to use that instead of
> > full .config files in order to reduce the amount of churn when
> > updating between ct-ng versions? Yann?
> 
> Indeed, but the defconfig support in crosstool-NG is not perfect yet,
> so I think we should keep complete .config files for this release.
> 
> The problem with defconfigs in crosstool-NG is that there is no
> guarantee about new symbols coming and going between versions, so
> this can result in strange behaviors when a new version comes out,
> and a defconfig from a previous version is used. Here's what may
> happen:
> 
> In version X, we use the latest stable 3.6 linux kernel, which is
> 3.6.2. This is also the absolute latest version of the kernel. So we
> have this symbol: LINUX_VER_3_6_2
> 
> The user did select this version. As 3.6.3 is the latest version, it
> is the first item in the choice, so it does not appear in the
> defconfig, being the default of the choice.
> 
> Now, version X+1 is out, and the stable 3.6 kernel is updated to
> 3.6.7, so we get a new symbol, but the previous one no longer exists.
> Also, kernel 3.7 is out, so the 3.6.7 is not the latest version
> available.
> 
> Now, the default symbol in the choice is not 3.6.x, but 3.7.x, which
> is probably not what the user expected.
> 
> Thus, using the defconfig from the previous version will yield a
> completely different .config.
> 
> This can be even more anoying with the gcc version, as we only have
> one linaro release per gcc /generation/ (eg. 4.7). If the user did
> select the latest linaro version (eg. 4.7) which was the default of
> the choice, qnd the new release has a more recent mainstream version
> (eg. 4.8), but no corresponding linaro version, then the gcc version
> will be bumnped to 4.8, when the user did in fact want the linaro 4.7.
> 
> I have a pending change to fix this issue for all components, but it
> will go in only for the next release, and that shall make using
> defconfigs from a previous version relatively safe to use with a
> newer version.
> 
> Then, we can switch buildroot to using the ct-ng defconfigs, but until
> then I think it is safer to use full .config files.

To me, it is also the point of using defconfig from the first place: to
benefit from the new default values for configuration options that we
leave undefined.

Note that the crosstool-ng.config-* don't necessarily need to be pure
defconfig files: you can add a few explicit options, whose value is
currently the default one, but since you don't want them to change to
newer default values, you explicit their value in the defconfig. We
have this for a few options in Buildroot defconfig files as well.

defconfig files do not necessarily need to be the direct output of
"savedefconfig", they can be hand-edited to enforce the value of
certain options.

Best regards,

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] elfutils and the perf userspace tool
From: Stefan Fröberg @ 2012-12-04 19:21 UTC (permalink / raw)
  To: buildroot
In-Reply-To: <1354608794.90144.YahooMailNeo@web120304.mail.ne1.yahoo.com>

Hi Bogdan

4.12.2012 10:13, Bogdan Radulescu kirjoitti:
>
>
>
> ----- Original Message -----
> From: Bryan Hundven <bryanhundven@gmail.com>
> To: Bogdan Radulescu <bogdan_radulescu99@yahoo.com>
> Cc: "buildroot at busybox.net" <buildroot@busybox.net>
> Sent: Tuesday, December 4, 2012 5:33 AM
> Subject: Re: [Buildroot] elfutils and the perf userspace tool
>
> On Mon, Dec 3, 2012 at 2:20 PM, Bogdan Radulescu
> <bogdan_radulescu99@yahoo.com> wrote:
>>
>> Hello everybody,
>>
>> I am using buildroot for generating a distro for ARM. I am able to add additional packages to buildroot or to use arm-linux-gcc directly to compile simple C programs.
>>
>> Still I haven't yet understood how to cross compile the perf userspace tool and I think it would be a very nice addition to buildroot.
>>
>> The perf tool requires elfutils and the kernel sources it comes with.
>>
>> My first problem is that elfutils will not successfully compile, as you can see here:
>> http://pastie.org/pastes/5474687/text
>>
>> My elfutils.mk file looks like this http://pastie.org/pastes/5474697/text
>>
>> Please tell me how do I move further from this.
>>
>> Thanks,
>> Bogdan
>> _______________________________________________
>> buildroot mailing list
>> buildroot at busybox.net
>> http://lists.busybox.net/mailman/listinfo/buildroot
> I forgot to cc buildroot at ... eh.
>
> Original response:
>
> I think you need the argp-standalone, as I don't think uclibc provides libargp.
>
> -Bryan
>
>
>
> Hi Bryan,
>
> As you can see in the elfutils.mk file I already added argp-standalone as a dependency. It compiles OK.
> I am also linking against it, still something goes wrong at some point.
>
> Bogdan
>
> _______________________________________________
> buildroot mailing list
> buildroot at busybox.net
> http://lists.busybox.net/mailman/listinfo/buildroot

It's not enough to give -largp in LDFLAGS.

elfutils doesn't give a damn about LDFLAGS and goes it own way.

What you need is to add -largp explicitly into a line that links libdw

You could try saving the following and also remember to add
ELFUTILS_AUTORECONF = YES
line to your elfutils.mk

diff -Naur elfutils-0.155.org/libdw/Makefile.am
elfutils-0.155/libdw/Makefile.am
--- elfutils-0.155.org/libdw/Makefile.am    2012-12-04
21:02:32.309083801 +0200
+++ elfutils-0.155/libdw/Makefile.am    2012-12-04 21:06:41.205098676 +0200
@@ -111,7 +111,7 @@
         -Wl,--enable-new-dtags,-rpath,$(pkglibdir) \
         -Wl,--version-script,$<,--no-undefined \
         -Wl,--whole-archive $(filter-out $<,$^) -Wl,--no-whole-archive\
-        -ldl $(zip_LIBS)
+        -ldl $(zip_LIBS) -largp
     if readelf -d $@ | fgrep -q TEXTREL; then exit 1; fi
     ln -fs $@ $@.$(VERSION)

There might other places too where you have to add that -largp

You probably(if you are using uClibc) need to explicitly add also -lintl
to the same line with -largp.
At least that was what I had to do to make it compile even tought I had
gived configure script the option "--disable-nls".

Also you will need to probably (again if using uClibc) have fts.h with
your uClibc-installation
and sometime later you will probably hit into a obstack missing problem.
That can only be solved by including gnulib stuff into your source.

And there was some other (uClibc related) stuff that needed patching too
that I can't now
remember

All in all, elfutils belongs sadly to the software category:

 "Im doing things my own way, Im using non-portable glibc stuff and
I don't give a crab about portability. Nah-nah-naaa"

:-(

It's hard but doeable. And the end result might be that only the libelf
part of elfutils will work
(I only needed that so it did not matter in my case) and not the programs.

Regards
Stefan

^ permalink raw reply
* [Buildroot] Buildroot and Jamvm
From: Peter Korsgaard @ 2012-12-04 19:12 UTC (permalink / raw)
  To: buildroot
In-Reply-To: <CAH8HgbzkfN9RJWRC-32u+qJxyanNgt6b=Us64QkzzTJaWVQgyA@mail.gmail.com>

>>>>> "alain" == alain mouflet <alain.mouflet@gmail.com> writes:

Hi,

 alain> I have downloaded Buildroot to use with a calao board. The
 alain> 2012.08 version is without jamvm and I need this soft.  I tought
 alain> that is simple to add it and I did it : I took java path in a
 alain> last version (2010.08), I have copied this directory on the new
 alain> buildroot pakages. I have modified the config.in file to get the
 alain> choice with "make menuconfig" and it's ok but never classpath
 alain> and jamvm are downloaded. No message about classpath or jamvm
 alain> appears in the build.log This way is good ? I made a mistake
 alain> somewhere or may be it's impossible to get jamvm with the new
 alain> buildroot ?

Some of the buildroot infrastructure has changed a bit since 2010.08, so
that's probably why. I have just re-added jamvm + classpath to buildroot
git, so it would be good if you could give it a test.

I've done several test builds and tried to run a simple hello world
program, but nothing major.

For git access, see:

http://buildroot.net/git.html

-- 
Bye, Peter Korsgaard

^ permalink raw reply
* [Buildroot] [PATCH 1/3] jquery: bump to version 1.8.3
From: Peter Korsgaard @ 2012-12-04 19:08 UTC (permalink / raw)
  To: buildroot
In-Reply-To: <1354641847-5083-1-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 all 3, thanks.

-- 
Bye, Peter Korsgaard

^ permalink raw reply
* [Buildroot] [PATCH 1/2] kernel-headers: bump 3.{0, 4, 6}.x stable versions
From: Peter Korsgaard @ 2012-12-04 18:56 UTC (permalink / raw)
  To: buildroot
In-Reply-To: <1354631815-18202-1-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 both, thanks.

-- 
Bye, Peter Korsgaard

^ permalink raw reply
* [Buildroot] [git commit] jquery-sparkline: bump to version 2.1
From: Peter Korsgaard @ 2012-12-04 18:55 UTC (permalink / raw)
  To: buildroot

commit: http://git.buildroot.net/buildroot/commit/?id=f290aa705db2084c0864fbf6750e4414a82e48a6
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/jquery-sparkline/jquery-sparkline.mk |    2 +-
 1 files changed, 1 insertions(+), 1 deletions(-)

diff --git a/package/jquery-sparkline/jquery-sparkline.mk b/package/jquery-sparkline/jquery-sparkline.mk
index d79e762..76adad6 100644
--- a/package/jquery-sparkline/jquery-sparkline.mk
+++ b/package/jquery-sparkline/jquery-sparkline.mk
@@ -1,4 +1,4 @@
-JQUERY_SPARKLINE_VERSION = 1.6
+JQUERY_SPARKLINE_VERSION = 2.1
 JQUERY_SPARKLINE_SITE = http://www.omnipotent.net/jquery.sparkline/$(JQUERY_SPARKLINE_VERSION)
 JQUERY_SPARKLINE_SOURCE = jquery.sparkline.min.js
 

^ permalink raw reply related
* [Buildroot] [git commit] jquery-validation: bump to version 1.10.0
From: Peter Korsgaard @ 2012-12-04 18:54 UTC (permalink / raw)
  To: buildroot

commit: http://git.buildroot.net/buildroot/commit/?id=48e0da1d745435f1c64c17556fc8758c6e37ce42
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/jquery-validation/jquery-validation.mk |    6 +++---
 1 files changed, 3 insertions(+), 3 deletions(-)

diff --git a/package/jquery-validation/jquery-validation.mk b/package/jquery-validation/jquery-validation.mk
index 7d30270..746b534 100644
--- a/package/jquery-validation/jquery-validation.mk
+++ b/package/jquery-validation/jquery-validation.mk
@@ -1,13 +1,13 @@
-JQUERY_VALIDATION_VERSION = 1.9.0
+JQUERY_VALIDATION_VERSION = 1.10.0
 JQUERY_VALIDATION_SITE = http://jquery.bassistance.de/validate
 JQUERY_VALIDATION_SOURCE = jquery-validation-$(JQUERY_VALIDATION_VERSION).zip
 
 define JQUERY_VALIDATION_EXTRACT_CMDS
-	unzip -d $(BUILD_DIR) $(DL_DIR)/$(JQUERY_VALIDATION_SOURCE)
+	unzip -d $(@D) $(DL_DIR)/$(JQUERY_VALIDATION_SOURCE)
 endef
 
 define JQUERY_VALIDATION_INSTALL_TARGET_CMDS
-	$(INSTALL) -m 0644 -D $(@D)/jquery.validate.min.js \
+	$(INSTALL) -m 0644 -D $(@D)/dist/jquery.validate.min.js \
 		$(TARGET_DIR)/var/www/jquery.validate.js
 endef
 

^ permalink raw reply related
* [Buildroot] [git commit] jquery: bump to version 1.8.3
From: Peter Korsgaard @ 2012-12-04 18:54 UTC (permalink / raw)
  To: buildroot

commit: http://git.buildroot.net/buildroot/commit/?id=6dbd0cd4e1337e0f68d46414174813be382a9678
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/jquery/jquery.mk |    2 +-
 1 files changed, 1 insertions(+), 1 deletions(-)

diff --git a/package/jquery/jquery.mk b/package/jquery/jquery.mk
index fb3a1b4..871f7a0 100644
--- a/package/jquery/jquery.mk
+++ b/package/jquery/jquery.mk
@@ -1,4 +1,4 @@
-JQUERY_VERSION = 1.7.1
+JQUERY_VERSION = 1.8.3
 JQUERY_SITE = http://code.jquery.com
 JQUERY_SOURCE = jquery-$(JQUERY_VERSION).min.js
 

^ permalink raw reply related
* [Buildroot] [git commit] linux: bump 3.6.x stable version
From: Peter Korsgaard @ 2012-12-04 18:52 UTC (permalink / raw)
  To: buildroot

commit: http://git.buildroot.net/buildroot/commit/?id=2422f13aaf6909d22414dba2fd8722135b1bb02b
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>
---
 linux/Config.in |    4 ++--
 1 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/linux/Config.in b/linux/Config.in
index f408ad5..ef3b2a1 100644
--- a/linux/Config.in
+++ b/linux/Config.in
@@ -21,7 +21,7 @@ choice
 	prompt "Kernel version"
 
 config BR2_LINUX_KERNEL_3_6
-	bool "3.6.8"
+	bool "3.6.9"
 
 config BR2_LINUX_KERNEL_SAME_AS_HEADERS
 	bool "Same as toolchain kernel headers"
@@ -76,7 +76,7 @@ config BR2_LINUX_KERNEL_CUSTOM_GIT_VERSION
 
 config BR2_LINUX_KERNEL_VERSION
 	string
-	default "3.6.8" if BR2_LINUX_KERNEL_3_6
+	default "3.6.9" if BR2_LINUX_KERNEL_3_6
 	default BR2_DEFAULT_KERNEL_HEADERS if BR2_LINUX_KERNEL_SAME_AS_HEADERS
 	default BR2_LINUX_KERNEL_CUSTOM_VERSION_VALUE if BR2_LINUX_KERNEL_CUSTOM_VERSION
 	default "custom" if BR2_LINUX_KERNEL_CUSTOM_TARBALL

^ permalink raw reply related
* [Buildroot] [git commit] kernel-headers: bump 3.{0, 4, 6}.x stable versions
From: Peter Korsgaard @ 2012-12-04 18:52 UTC (permalink / raw)
  To: buildroot

commit: http://git.buildroot.net/buildroot/commit/?id=261f894bb8633646162280f5a9ac1a9abb12a2e5
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>
---
 toolchain/kernel-headers/Config.in                 |    6 +++---
 ...all-fix-__packed-in-exported-kernel-head.patch} |    0
 2 files changed, 3 insertions(+), 3 deletions(-)

diff --git a/toolchain/kernel-headers/Config.in b/toolchain/kernel-headers/Config.in
index 12b3956..6552c62 100644
--- a/toolchain/kernel-headers/Config.in
+++ b/toolchain/kernel-headers/Config.in
@@ -67,12 +67,12 @@ config BR2_DEFAULT_KERNEL_HEADERS
 	default "2.6.37.6"	if BR2_KERNEL_HEADERS_2_6_37
 	default "2.6.38.8"	if BR2_KERNEL_HEADERS_2_6_38
 	default "2.6.39.4"	if BR2_KERNEL_HEADERS_2_6_39
-	default "3.0.53"	if BR2_KERNEL_HEADERS_3_0
+	default "3.0.54"	if BR2_KERNEL_HEADERS_3_0
 	default "3.1.10"	if BR2_KERNEL_HEADERS_3_1
 	default "3.2.34"	if BR2_KERNEL_HEADERS_3_2
 	default "3.3.8"		if BR2_KERNEL_HEADERS_3_3
-	default "3.4.20"	if BR2_KERNEL_HEADERS_3_4
+	default "3.4.21"	if BR2_KERNEL_HEADERS_3_4
 	default "3.5.7"		if BR2_KERNEL_HEADERS_3_5
-	default "3.6.8"		if BR2_KERNEL_HEADERS_3_6
+	default "3.6.9"		if BR2_KERNEL_HEADERS_3_6
 	default "2.6"		if BR2_KERNEL_HEADERS_SNAP
 	default $BR2_DEFAULT_KERNEL_VERSION if BR2_KERNEL_HEADERS_VERSION
diff --git a/toolchain/kernel-headers/linux-3.0.53-headers_install-fix-__packed-in-exported-kernel-head.patch b/toolchain/kernel-headers/linux-3.0.54-headers_install-fix-__packed-in-exported-kernel-head.patch
similarity index 100%
rename from toolchain/kernel-headers/linux-3.0.53-headers_install-fix-__packed-in-exported-kernel-head.patch
rename to toolchain/kernel-headers/linux-3.0.54-headers_install-fix-__packed-in-exported-kernel-head.patch

^ permalink raw reply related
* [Buildroot] [git commit] dependencies.sh: check for java tools if classpath is enabled
From: Peter Korsgaard @ 2012-12-04 18:28 UTC (permalink / raw)
  To: buildroot

commit: http://git.buildroot.net/buildroot/commit/?id=4be86184e032652863ea826edb9018f5f9b64b7d
branch: http://git.buildroot.net/buildroot/commit/?id=refs/heads/master

Signed-off-by: Peter Korsgaard <jacmet@sunsite.dk>
---
 support/dependencies/dependencies.sh |    8 ++++++++
 1 files changed, 8 insertions(+), 0 deletions(-)

diff --git a/support/dependencies/dependencies.sh b/support/dependencies/dependencies.sh
index 9f0f6a9..1b79b23 100755
--- a/support/dependencies/dependencies.sh
+++ b/support/dependencies/dependencies.sh
@@ -158,3 +158,11 @@ 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 java javac jar; do
+	if ! which $prog > /dev/null ; then
+	    /bin/echo -e "\nYou must install '$prog' on your build machine\n" >&2
+	    exit 1
+	fi
+    done
+fi

^ permalink raw reply related
* [Buildroot] [git commit] package: add jamvm jvm
From: Peter Korsgaard @ 2012-12-04 18:21 UTC (permalink / raw)
  To: buildroot

commit: http://git.buildroot.net/buildroot/commit/?id=ae10e33a504770b0141dd7582c6ca26fc43b4642
branch: http://git.buildroot.net/buildroot/commit/?id=refs/heads/master

Signed-off-by: Peter Korsgaard <jacmet@sunsite.dk>
---
 package/Config.in       |    1 +
 package/jamvm/Config.in |    9 +++++++++
 package/jamvm/jamvm.mk  |   11 +++++++++++
 3 files changed, 21 insertions(+), 0 deletions(-)

diff --git a/package/Config.in b/package/Config.in
index 66e8352..f95e29d 100644
--- a/package/Config.in
+++ b/package/Config.in
@@ -268,6 +268,7 @@ endmenu
 menu "Interpreter languages and scripting"
 source "package/erlang/Config.in"
 source "package/haserl/Config.in"
+source "package/jamvm/Config.in"
 source "package/lua/Config.in"
 source "package/luajit/Config.in"
 if BR2_PACKAGE_LUA || BR2_PACKAGE_LUAJIT
diff --git a/package/jamvm/Config.in b/package/jamvm/Config.in
new file mode 100644
index 0000000..852c6e8
--- /dev/null
+++ b/package/jamvm/Config.in
@@ -0,0 +1,9 @@
+config BR2_PACKAGE_JAMVM
+	bool "jamvm"
+	select BR2_PACKAGE_ZLIB
+	select BR2_PACKAGE_CLASSPATH
+	help
+	  JamVM is a new Java Virtual Machine which conforms to the
+	  JVM specification version 2 (blue book).
+
+	  http://jamvm.sf.net
diff --git a/package/jamvm/jamvm.mk b/package/jamvm/jamvm.mk
new file mode 100644
index 0000000..f0765d7
--- /dev/null
+++ b/package/jamvm/jamvm.mk
@@ -0,0 +1,11 @@
+JAMVM_VERSION = 1.5.4
+JAMVM_SITE = http://downloads.sourceforge.net/project/jamvm/jamvm/JamVM%20$(JAMVM_VERSION)
+JAMVM_DEPENDENCIES = zlib classpath
+# int inlining seems to crash jamvm, don't build shared version of internal lib
+JAMVM_CONF_OPT = \
+	--with-classpath-install-dir=/usr \
+	--disable-int-inlining \
+	--disable-shared \
+	--without-pic
+
+$(eval $(autotools-package))

^ permalink raw reply related
* [Buildroot] [git commit] package: add classpath
From: Peter Korsgaard @ 2012-12-04 18:21 UTC (permalink / raw)
  To: buildroot

commit: http://git.buildroot.net/buildroot/commit/?id=5062f530081913bd0487f0f879be180f30836ab8
branch: http://git.buildroot.net/buildroot/commit/?id=refs/heads/master

Requires java, javac and jar on the build machine, similar to how we
require gcc/g++.

Signed-off-by: Peter Korsgaard <jacmet@sunsite.dk>

classpath: fixup
---
 package/Config.in              |    1 +
 package/classpath/Config.in    |   14 ++++++++++
 package/classpath/classpath.mk |   55 ++++++++++++++++++++++++++++++++++++++++
 3 files changed, 70 insertions(+), 0 deletions(-)

diff --git a/package/Config.in b/package/Config.in
index 5ba1f05..66e8352 100644
--- a/package/Config.in
+++ b/package/Config.in
@@ -489,6 +489,7 @@ endmenu
 menu "Other"
 source "package/apr/Config.in"
 source "package/apr-util/Config.in"
+source "package/classpath/Config.in"
 source "package/fftw/Config.in"
 source "package/libargtable2/Config.in"
 source "package/argp-standalone/Config.in"
diff --git a/package/classpath/Config.in b/package/classpath/Config.in
new file mode 100644
index 0000000..45d5984
--- /dev/null
+++ b/package/classpath/Config.in
@@ -0,0 +1,14 @@
+config BR2_PACKAGE_CLASSPATH
+	bool "classpath"
+	depends on BR2_PACKAGE_JAMVM
+	depends on BR2_INET_IPV6
+	help
+	  GNU Classpath, Essential Libraries for Java, is a GNU
+	  project to create free core class libraries for use with
+	  virtual machines and compilers for the java programming
+	  language.
+
+	  http://classpath.org
+
+comment "classpath requires a toolchain with IPV6 support"
+	depends on !BR2_INET_IPV6
diff --git a/package/classpath/classpath.mk b/package/classpath/classpath.mk
new file mode 100644
index 0000000..9142f69
--- /dev/null
+++ b/package/classpath/classpath.mk
@@ -0,0 +1,55 @@
+CLASSPATH_VERSION = 0.98
+CLASSPATH_SITE = $(BR2_GNU_MIRROR)/classpath
+CLASSPATH_CONF_OPT = \
+	--disable-examples \
+	--disable-plugin \
+	--disable-rpath \
+	--disable-Werror \
+	--disable-gconf-peer \
+	--disable-tools
+
+CLASSPATH_DEPENDENCIES = host-pkgconf
+
+ifeq ($(BR2_PACKAGE_ALSA_LIB),y)
+CLASSPATH_CONF_OPT += --enable-alsa
+CLASSPATH_DEPENDENCIES += alsa-lib
+else
+CLASSPATH_CONF_OPT += --disable-alsa
+endif
+
+ifeq ($(BR2_PACKAGE_GMP),y)
+CLASSPATH_CONF_OPT += --enable-gmp
+CLASSPATH_DEPENDENCIES += gmp
+else
+CLASSPATH_CONF_OPT += --disable-gmp
+endif
+
+ifeq ($(BR2_PACKAGE_GST_PLUGINS_BASE),y)
+CLASSPATH_CONF_OPT += --enable-gstreamer-peer
+CLASSPATH_DEPENDENCIES += gst-plugins-base
+else
+CLASSPATH_CONF_OPT += --disable-gstreamer-peer
+endif
+
+ifeq ($(BR2_PACKAGE_LIBGTK2)$(BR2_PACKAGE_XORG7),yy)
+CLASSPATH_CONF_OPT += --enable-gtk-peer
+CLASSPATH_DEPENDENCIES += libgtk2
+else
+CLASSPATH_CONF_OPT += --disable-gtk-peer
+endif
+
+ifeq ($(BR2_PACKAGE_QT),y)
+CLASSPATH_CONF_OPT += --enable-qt-peer
+CLASSPATH_DEPENDENCIES += qt
+else
+CLASSPATH_CONF_OPT += --disable-qt-peer
+endif
+
+ifeq ($(BR2_PACKAGE_LIBXML2)$(BR2_PACKAGE_LIBXSLT),yy)
+CLASSPATH_CONF_OPT += --enable-xmlj
+CLASSPATH_DEPENDENCIES += libxml2 libxslt
+else
+CLASSPATH_CONF_OPT += --disable-xmlj
+endif
+
+$(eval $(autotools-package))

^ permalink raw reply related
* [Buildroot] How to handle modularity in buildroot?
From: Charles Manning @ 2012-12-04 18:18 UTC (permalink / raw)
  To: buildroot
In-Reply-To: <20121204101042.3ae057bb@skate>

On Tue, Dec 4, 2012 at 10:10 PM, Thomas Petazzoni
<thomas.petazzoni@free-electrons.com> wrote:
> Dear Charles Manning,
>
> On Tue, 4 Dec 2012 12:49:14 +1300, Charles Manning wrote:
>
>> I am aware of this mechanism but I don't think it really gets me what I am
>> after.
>>
>> I think that with defconfig, I would be able to make defconfig, then tweak the
>> config to get a variant. But now if I change the base config, I then have to
>> go through the process again for the variant.
>>
>> What I would like is the ability to have:
>> * A base package
>> * Production rootfs = base + production files.
>> * Development rootfs = base  + development files.
>>
>> Then a change to the base package automatically flows through the production +
>> development rootfs.
>>
>> Is there a way to accomplish that?
>
> First of all, thanks for your nice words about Buildroot, definitely
> appreciated.
>
> That said, if Buildroot is certainly simpler than OE, it also means
> that Buildroot has a smaller feature set, and to some extent a bit less
> flexibility.

Thanks Thomas

I certainly would never want buildroot to bloat out to be like OE. The
most direct way to make something useless is to try to be all o be all
things to all people.

buildroot is very valuable and I intend to contribute where I can. I
live in New Zealand which makes it a bit hard to attend FOSDEM :-).

>
> What you can do to achieve what you want is use defconfig fragments.
> For example:
>
>  * mysystem_base_defconfig.frag would contain:
>
> BR2_arm=y
> BR2_TOOLCHAIN_EXTERNAL=y
> BR2_TOOLCHAIN_EXTERNAL_CUSTOM=y
> BR2_TOOLCHAIN_EXTERNAL_PATH="/home/thomas/x-tools/arm-br-arm926/usr/"
> BR2_TOOLCHAIN_EXTERNAL_LARGEFILE=y
> BR2_TOOLCHAIN_EXTERNAL_INET_IPV6=y
> BR2_TOOLCHAIN_EXTERNAL_LOCALE=y
> # BR2_TOOLCHAIN_EXTERNAL_HAS_THREADS_DEBUG is not set
> BR2_TOOLCHAIN_EXTERNAL_INET_RPC=y
> BR2_TOOLCHAIN_EXTERNAL_CXX=y
> BR2_PACKAGE_BUSYBOX=y
>
>  * mysystem_prod_defconfig.frag would contain:
>
> BR2_PACKAGE_APPLICATION1=y
> BR2_PACKAGE_APPLICATION2=y
>
>  * mysystem_dev_defconfig.frag would contain:
>
> BR2_PACKAGE_STRACE=y
> BR2_PACKAGE_LTRACE=y
>
> And then, when you want to do a production build:
>
>  make clean
>  cat mysystem_base_defconfig.frag mysystem_prod_defconfig.frag > configs/mysystem_prod_defconfig
>  make mysystem_prod_defconfig
>  make
>
> And when you want to do a development build:
>
>  make clean
>  cat mysystem_base_defconfig.frag mysystem_dev_defconfig.frag > configs/mysystem_dev_defconfig
>  make mysystem_dev_defconfig
>  make
>
> Basically, instead of having Buildroot providing this feature, we rely
> on the simple manipulation of the configuration file.

This sounds like the right idea.

I will experiment with this a bit and see where it takes me. If there
is something worth sharing I'll do that.

Thanks

Charles

^ permalink raw reply
* [Buildroot] [PATCH 2/2] usb_modeswitch_data: bump to version 20121109
From: Gustavo Zacarias @ 2012-12-04 18:13 UTC (permalink / raw)
  To: buildroot
In-Reply-To: <1354644802-28286-1-git-send-email-gustavo@zacarias.com.ar>

Signed-off-by: Gustavo Zacarias <gustavo@zacarias.com.ar>
---
 package/usb_modeswitch_data/usb_modeswitch_data.mk |    2 +-
 1 files changed, 1 insertions(+), 1 deletions(-)

diff --git a/package/usb_modeswitch_data/usb_modeswitch_data.mk b/package/usb_modeswitch_data/usb_modeswitch_data.mk
index a7b7072..479137c 100644
--- a/package/usb_modeswitch_data/usb_modeswitch_data.mk
+++ b/package/usb_modeswitch_data/usb_modeswitch_data.mk
@@ -4,7 +4,7 @@
 #
 #############################################################
 
-USB_MODESWITCH_DATA_VERSION = 20120815
+USB_MODESWITCH_DATA_VERSION = 20121109
 USB_MODESWITCH_DATA_SOURCE = usb-modeswitch-data-$(USB_MODESWITCH_DATA_VERSION).tar.bz2
 USB_MODESWITCH_DATA_SITE = http://www.draisberghof.de/usb_modeswitch
 USB_MODESWITCH_DATA_DEPENDENCIES = usb_modeswitch
-- 
1.7.8.6

^ permalink raw reply related
* [Buildroot] [PATCH 1/2] usb_modeswitch: bump to version 1.2.5
From: Gustavo Zacarias @ 2012-12-04 18:13 UTC (permalink / raw)
  To: buildroot

Signed-off-by: Gustavo Zacarias <gustavo@zacarias.com.ar>
---
 package/usb_modeswitch/usb_modeswitch.mk |    2 +-
 1 files changed, 1 insertions(+), 1 deletions(-)

diff --git a/package/usb_modeswitch/usb_modeswitch.mk b/package/usb_modeswitch/usb_modeswitch.mk
index c517991..653109d 100644
--- a/package/usb_modeswitch/usb_modeswitch.mk
+++ b/package/usb_modeswitch/usb_modeswitch.mk
@@ -4,7 +4,7 @@
 #
 #############################################################
 
-USB_MODESWITCH_VERSION = 1.2.4
+USB_MODESWITCH_VERSION = 1.2.5
 USB_MODESWITCH_SOURCE = usb-modeswitch-$(USB_MODESWITCH_VERSION).tar.bz2
 USB_MODESWITCH_SITE = http://www.draisberghof.de/usb_modeswitch
 USB_MODESWITCH_DEPENDENCIES = libusb-compat
-- 
1.7.8.6

^ permalink raw reply related
* [Buildroot] [PATCH] package/crosstool-ng: update to 1.17.0
From: Yann E. MORIN @ 2012-12-04 18:09 UTC (permalink / raw)
  To: buildroot
In-Reply-To: <20121204101228.2e78f9a1@skate>

Thomas, Carsten, All,

On Tuesday 04 December 2012 Thomas Petazzoni wrote:
> On Mon,  3 Dec 2012 20:47:45 +0100, Carsten Schoenert wrote:
> > From: Carsten Schoenert <c.schoenert@t-online.de>
> > 
> > Updating current crosstool-ng config files to 1.17.0
> >  crosstool-ng.config-eglibc
> >  crosstool-ng.config-glibc
> >  crosstool-ng.config-uClibc
> > 
> > Signed-off-by: Carsten Schoenert <c.schoenert@t-online.de>
> > ---
> >  package/crosstool-ng/crosstool-ng.mk               |    2 +-
> >  .../crosstool-ng.config-eglibc                     |   73 ++++++++++++++-----
> >  .../crosstool-ng.config-glibc                      |   74 +++++++++++++++-----
> >  .../crosstool-ng.config-uClibc                     |   71 ++++++++++++++-----
> >  4 files changed, 165 insertions(+), 55 deletions(-)
> 
> Thanks for doing this work. I think ct-ng now supports minimal
> defconfig files. Would it make sense to use that instead of
> full .config files in order to reduce the amount of churn when updating
> between ct-ng versions? Yann?

Indeed, but the defconfig support in crosstool-NG is not perfect yet, so
I think we should keep complete .config files for this release.

The problem with defconfigs in crosstool-NG is that there is no guarantee
about new symbols coming and going between versions, so this can result in
strange behaviors when a new version comes out, and a defconfig from a
previous version is used. Here's what may happen:

In version X, we use the latest stable 3.6 linux kernel, which is 3.6.2.
This is also the absolute latest version of the kernel. So we have this
symbol: LINUX_VER_3_6_2

The user did select this version. As 3.6.3 is the latest version, it is
the first item in the choice, so it does not appear in the defconfig,
being the default of the choice.

Now, version X+1 is out, and the stable 3.6 kernel is updated to 3.6.7,
so we get a new symbol, but the previous one no longer exists. Also,
kernel 3.7 is out, so the 3.6.7 is not the latest version available.

Now, the default symbol in the choice is not 3.6.x, but 3.7.x, which
is probably not what the user expected.

Thus, using the defconfig from the previous version will yield a
completely different .config.

This can be even more anoying with the gcc version, as we only have one
linaro release per gcc /generation/ (eg. 4.7). If the user did select
the latest linaro version (eg. 4.7) which was the default of the choice,
qnd the new release has a more recent mainstream version (eg. 4.8), but
no corresponding linaro version, then the gcc version will be bumnped
to 4.8, when the user did in fact want the linaro 4.7.

I have a pending change to fix this issue for all components, but it will
go in only for the next release, and that shall make using defconfigs from
a previous version relatively safe to use with a newer version.

Then, we can switch buildroot to using the ct-ng defconfigs, but until
then I think it is safer to use full .config files.

Regards,
Yann E. MORIN.

-- 
.-----------------.--------------------.------------------.--------------------.
|  Yann E. MORIN  | Real-Time Embedded | /"\ ASCII RIBBON | Erics' conspiracy: |
| +33 662 376 056 | Software  Designer | \ / CAMPAIGN     |  ___               |
| +33 223 225 172 `------------.-------:  X  AGAINST      |  \e/  There is no  |
| http://ymorin.is-a-geek.org/ | _/*\_ | / \ HTML MAIL    |   v   conspiracy.  |
'------------------------------^-------^------------------^--------------------'

^ permalink raw reply
* [Buildroot] [PATCH 3/3] jquery-sparkline: bump to version 2.1
From: Gustavo Zacarias @ 2012-12-04 17:24 UTC (permalink / raw)
  To: buildroot
In-Reply-To: <1354641847-5083-1-git-send-email-gustavo@zacarias.com.ar>

Signed-off-by: Gustavo Zacarias <gustavo@zacarias.com.ar>
---
 package/jquery-sparkline/jquery-sparkline.mk |    2 +-
 1 files changed, 1 insertions(+), 1 deletions(-)

diff --git a/package/jquery-sparkline/jquery-sparkline.mk b/package/jquery-sparkline/jquery-sparkline.mk
index d79e762..76adad6 100644
--- a/package/jquery-sparkline/jquery-sparkline.mk
+++ b/package/jquery-sparkline/jquery-sparkline.mk
@@ -1,4 +1,4 @@
-JQUERY_SPARKLINE_VERSION = 1.6
+JQUERY_SPARKLINE_VERSION = 2.1
 JQUERY_SPARKLINE_SITE = http://www.omnipotent.net/jquery.sparkline/$(JQUERY_SPARKLINE_VERSION)
 JQUERY_SPARKLINE_SOURCE = jquery.sparkline.min.js
 
-- 
1.7.8.6

^ permalink raw reply related
* [Buildroot] [PATCH 2/3] jquery-validation: bump to version 1.10.0
From: Gustavo Zacarias @ 2012-12-04 17:24 UTC (permalink / raw)
  To: buildroot
In-Reply-To: <1354641847-5083-1-git-send-email-gustavo@zacarias.com.ar>

Signed-off-by: Gustavo Zacarias <gustavo@zacarias.com.ar>
---
 package/jquery-validation/jquery-validation.mk |    6 +++---
 1 files changed, 3 insertions(+), 3 deletions(-)

diff --git a/package/jquery-validation/jquery-validation.mk b/package/jquery-validation/jquery-validation.mk
index 7d30270..746b534 100644
--- a/package/jquery-validation/jquery-validation.mk
+++ b/package/jquery-validation/jquery-validation.mk
@@ -1,13 +1,13 @@
-JQUERY_VALIDATION_VERSION = 1.9.0
+JQUERY_VALIDATION_VERSION = 1.10.0
 JQUERY_VALIDATION_SITE = http://jquery.bassistance.de/validate
 JQUERY_VALIDATION_SOURCE = jquery-validation-$(JQUERY_VALIDATION_VERSION).zip
 
 define JQUERY_VALIDATION_EXTRACT_CMDS
-	unzip -d $(BUILD_DIR) $(DL_DIR)/$(JQUERY_VALIDATION_SOURCE)
+	unzip -d $(@D) $(DL_DIR)/$(JQUERY_VALIDATION_SOURCE)
 endef
 
 define JQUERY_VALIDATION_INSTALL_TARGET_CMDS
-	$(INSTALL) -m 0644 -D $(@D)/jquery.validate.min.js \
+	$(INSTALL) -m 0644 -D $(@D)/dist/jquery.validate.min.js \
 		$(TARGET_DIR)/var/www/jquery.validate.js
 endef
 
-- 
1.7.8.6

^ permalink raw reply related
* [Buildroot] [PATCH 1/3] jquery: bump to version 1.8.3
From: Gustavo Zacarias @ 2012-12-04 17:24 UTC (permalink / raw)
  To: buildroot

Signed-off-by: Gustavo Zacarias <gustavo@zacarias.com.ar>
---
 package/jquery/jquery.mk |    2 +-
 1 files changed, 1 insertions(+), 1 deletions(-)

diff --git a/package/jquery/jquery.mk b/package/jquery/jquery.mk
index fb3a1b4..871f7a0 100644
--- a/package/jquery/jquery.mk
+++ b/package/jquery/jquery.mk
@@ -1,4 +1,4 @@
-JQUERY_VERSION = 1.7.1
+JQUERY_VERSION = 1.8.3
 JQUERY_SITE = http://code.jquery.com
 JQUERY_SOURCE = jquery-$(JQUERY_VERSION).min.js
 
-- 
1.7.8.6

^ permalink raw reply related
* [Buildroot] Qt library difference between E.so and just .so
From: Eric Bénard @ 2012-12-04 17:19 UTC (permalink / raw)
  To: buildroot
In-Reply-To: <CADKZYYJFAKFs5KmSbcU7o=wkWi1_wOVc6MhgZrOdALgr_iva+A@mail.gmail.com>

Hi Santhosh,

Le Sat, 1 Dec 2012 16:27:54 -0600,
Santhosh Ramani <srcoolz@gmail.com> a ?crit :
> I'm using buildroot 2012.08 to create rootfs for AM335x device and would
> like to compile in Qt libraries. I went through make menuconfig and I get
> libraries for Qt 4.8.2. However, when I try and run some of the same
> examples from the TI SDK it complains that it can't find *libQtGuiE.so.4.7.2
> *.
> 
TI'SDK is generated using Yocto/OpenEmbedded which allows the creation
of both Qt Embedded _and_ Qt for Xorg librairies. In order to make
the difference between both libraries which may both be integrated in
the same SDK, the Qt Embedded libraries have the 'E' suffix and the
SDK provides the right config files to make qmake happy.
So if you generate a rootfs using buildroot, you need to compile the
examples with buildroot's toolchain to have a consistent library naming.

Best regards,
Eric

^ permalink raw reply
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox