* [Buildroot] [PATCH 1/7] docs/manual: update 'adding packages' with the new _AVAILABLE symbol
From: Arnout Vandecappelle @ 2012-11-01 22:40 UTC (permalink / raw)
To: buildroot
In-Reply-To: <201211011721.23581.yann.morin.1998@free.fr>
On 11/01/12 17:21, Yann E. MORIN wrote:
> Arnout, All,
>
> On Thursday 01 November 2012 Arnout Vandecappelle wrote:
>> On 09/10/12 01:40, Yann E. MORIN wrote:
[snip]
>>> +21: ifeq ($(BR2_PACKAGE_LIBFOO_GOO),y)
>>> +22: LIBFOO_DEPENDENCIES += goo
>>> +23: LIBFOO_CONF_OPT += --enable-goo
>>> +24: else
>>> +25: LIBFOO_CONF_OPT += --disable-goo
>>> +26: endif
>>
>> While you're at it, an automatic enable/disable example would
>> be nice too.
>
> What do you mean by "automatic"?
ifeq ($(BR2_PACKAGE_OPENSSL),y)
LIBFOO_DEPENDENCIES += openssl
LIBFOO_CONF_OPT += --enable-openssl
else
LIBFOO_CONF_OPT += --disable-openssl
endif
>
>>> --- a/docs/manual/adding-packages-directory.txt
>>> +++ b/docs/manual/adding-packages-directory.txt
>>> +* The syntax of the +Config.in+ file is the same as the one for the kernel
>>> + Kconfig file. The documentation for this syntax is available at:
>>> + https://git.kernel.org/?p=linux/kernel/git/torvalds/linux.git;a=blob;f=Documentation/kbuild/kconfig-language.txt[]
>>
>> http://kernel.org/doc/Documentation/kbuild/kconfig-language.txt
>
> Does this URL always match the latest version in the git tree?
I'm not sure, but anyway we don't sync with the upstream kconfig very
often either.
>
>>> +1. The +BR2_PACKAGE_LIBFOO_AVAILABLE+ symbol shall +depends on+ any
>>> +other package's +_AVAILABLE+ symbol. It may also depend on any other
>>> +symbol, such as toolchain features, but should not directly depend on
>>> +any package's main symbol.
>>
>> ... except for _XORG7, _PYTHON, etc.
>
> Well, my opinion (FWIW) is those packages should not be treated
> differently just because they are /big/.
>
> (One of) the goal(s) of _AVAILABLE is to allow the user to say either:
> - I want this package, enable whatever dependencies are required.
> or:
> - I need this package, but I have to provide a toolchain that has
> such and such feature
>
> _AVAILABLE makes that easy.
>
> Then, it's up to the user to understand what pulling-in a package implies.
I actually agree, but that's not the current reality.
OTOH, it makes sense to promote the wanted reality in the documentation.
Regards,
Arnout
[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
^ permalink raw reply
* [Buildroot] [PATCH 1/1] Create a symbolic link to the target kernel directory. example
From: Thomas Petazzoni @ 2012-11-01 22:24 UTC (permalink / raw)
To: buildroot
In-Reply-To: <D4127A04D6B846E881F6B42F61985647@JohanW7>
Dear Sagaert Johan,
On Thu, 1 Nov 2012 20:10:26 +0100, Sagaert Johan wrote:
> KERNELDIR=/home/buildroot12git/output/build/linux-76e10d158efb6d4516018846f60c2ab5501900bc
Turn this into:
KERNELDIR=$(shell ls -1 /home/buildroot12git/output/build/linux-[0-9a-f]*)
or, if you want something cleaner, add a <pkg>-show-builddir target for
all packages in pkg-generic.mk (we could probably get this patch
upstream), and do:
KERNELDIR=($shell make -C /home/buildroot12git/ linux-show-builddir)
And voil?.
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 v2] fs/common: Create initial console device
From: Arnout Vandecappelle @ 2012-11-01 22:21 UTC (permalink / raw)
To: buildroot
In-Reply-To: <1351765685-8128-1-git-send-email-markos.chandras@gmail.com>
On 11/01/12 11:28, Markos Chandras wrote:
> From: Markos Chandras<markos.chandras@imgtec.com>
>
> A /dev/console node must be present in rootfs when the Linux kernel
> boots otherwise the kernel will print the following warning:
> "Warning: unable to open an initial console"
>
> This is because when we use an initramfs the /dev directory is not
> populated at this point. This can cause problems when a program
> (e.g ldso with early debugging enabled) opens a standard file
> descriptor for read/write before these descriptors are actually
> created by the init process later on.
>
> Signed-off-by: Markos Chandras<markos.chandras@imgtec.com>
> ---
> fs/cpio/cpio.mk | 2 ++
> 1 files changed, 2 insertions(+), 0 deletions(-)
>
> diff --git a/fs/cpio/cpio.mk b/fs/cpio/cpio.mk
> index aa20b41..3293bc9 100644
> --- a/fs/cpio/cpio.mk
> +++ b/fs/cpio/cpio.mk
> @@ -21,6 +21,8 @@ define ROOTFS_CPIO_ADD_INIT
> fi
> endef
>
> +PACKAGES_PERMISSIONS_TABLE += /dev/console c 622 0 0 5 1
The last three entries are missing (start, inc, count). Also, there should
be a $(sep) at the end of the line; otherwise, any subsequent entries are
appended at the end of the line. (This will probably be the last line in
the table anyway, because the fs/ tree is included at the end of Makefile,
but we shouldn't rely on that).
Regards,
Arnout
> +
> endif # BR2_ROOTFS_DEVICE_CREATION_STATIC
>
> ROOTFS_CPIO_PRE_GEN_HOOKS += ROOTFS_CPIO_ADD_INIT
--
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] Build gcc debugger for the Target
From: Arnout Vandecappelle @ 2012-11-01 22:11 UTC (permalink / raw)
To: buildroot
In-Reply-To: <CABtOAfxVfcmYSAa+X6ksTdbVVCq4_piWCyO-L0-JAdAnrq_TSA@mail.gmail.com>
On 11/01/12 11:41, Alexander Khryukin wrote:
> Yes, i enabled Devel files on target, but gcc not available into "Development Tools".
>
> Also i using external toolchain, not buildroot.
gcc on the target is only available with a buildroot toolchain.
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 1/1] bump latest stable kernel headers
From: Sagaert Johan @ 2012-11-01 19:32 UTC (permalink / raw)
To: buildroot
bump latest stable kernel headers.
Signed-off-by: Sagaert Johan <sagaert.johan@skynet.be>
---
| 8 ++++----
1 file changed, 4 insertions(+), 4 deletions(-)
--git a/toolchain/kernel-headers/Config.in b/toolchain/kernel-headers/Config.in
index 2270e0c..d596870 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 "3.0.49" if BR2_KERNEL_HEADERS_3_0
+ default "3.0.50" if BR2_KERNEL_HEADERS_3_0
default "3.1.10" if BR2_KERNEL_HEADERS_3_1
- default "3.2.32" if BR2_KERNEL_HEADERS_3_2
+ default "3.2.33" if BR2_KERNEL_HEADERS_3_2
default "3.3.8" if BR2_KERNEL_HEADERS_3_3
- default "3.4.16" if BR2_KERNEL_HEADERS_3_4
+ default "3.4.17" if BR2_KERNEL_HEADERS_3_4
default "3.5.7" if BR2_KERNEL_HEADERS_3_5
- default "3.6.4" if BR2_KERNEL_HEADERS_3_6
+ default "3.6.5" if BR2_KERNEL_HEADERS_3_6
default "2.6" if BR2_KERNEL_HEADERS_SNAP
default $BR2_DEFAULT_KERNEL_VERSION if BR2_KERNEL_HEADERS_VERSION
--
1.8.0
^ permalink raw reply related
* [Buildroot] [PATCH 1/1] Create a symbolic link to the target kernel directory. example
From: Sagaert Johan @ 2012-11-01 19:10 UTC (permalink / raw)
To: buildroot
In-Reply-To: <20121101101912.310579d7@skate>
Hi
An example makefile explains it all :
With my patch I can write a makefile for my kernel modules like this:
MODULENAME := digitalio
obj-m += $(MODULENAME).o
KERNELDIR=/home/buildroot12git/targetkernel
all:$(MODULENAME).ko
$(MODULENAME).ko:$(MODULENAME).c
make -C $(KERNELDIR) SUBDIRS=$(PWD) modules CC=arm-linux-gcc LD=arm-linux-ld ARCH=arm
.PHONY clean:
make -C $(KERNELDIR) M=$(PWD) clean
Without the symbolic link I had to use this makefile, and change it everytime I used another kernel version.
MODULENAME := digitalio
obj-m += $(MODULENAME).o
KERNELDIR=/home/buildroot12git/output/build/linux-76e10d158efb6d4516018846f60c2ab5501900bc
all:$(MODULENAME).ko
$(MODULENAME).ko:$(MODULENAME).c
make -C $(KERNELDIR) SUBDIRS=$(PWD) modules CC=arm-linux-gcc LD=arm-linux-ld ARCH=arm
.PHONY clean:
make -C $(KERNELDIR) M=$(PWD) clean
As I onderstood from Arnout, I could build the module from within buildroot (making use of the buildroot vars) but this not an
option while still developing a module. Every buildroot make would also force a regeneration of the rootfs tar, and this takes a
serious ammount of time with a large rootfs.
So I develop it outside the build system and use the symlink to the kernel I have selected.
Regards, Johan
-----Oorspronkelijk bericht-----
Van: Thomas Petazzoni [mailto:thomas.petazzoni at free-electrons.com]
Verzonden: donderdag 1 november 2012 10:19
Aan: Sagaert Johan
CC: 'Arnout Vandecappelle'; buildroot at busybox.net
Onderwerp: Re: [Buildroot] [PATCH 1/1] Create a symbolic link to the target kernel directory.
On Thu, 1 Nov 2012 02:14:35 +0100, Sagaert Johan wrote:
> I always build modules outside of the buildroot system.
> Is there another (fast ) way ? I don't like to wait 20 seconds for
> buildroot tarring the rootfs.
I don't see the relation between waiting 20 seconds for Buildroot to tar the rootfs and the patch you're posting. Care to explain?
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 2/7] support/scripts: add a script to add a new package
From: Yann E. MORIN @ 2012-11-01 17:25 UTC (permalink / raw)
To: buildroot
In-Reply-To: <5091D7B6.8070703@mind.be>
Arnout, All,
On Thursday 01 November 2012 Arnout Vandecappelle wrote:
> On 09/10/12 01:40, Yann E. MORIN wrote:
> > This script asks a few questions to the user, and creates the skeleton
> > files (Config.in and package.mk).
[--SNIP--]
> > +# --------------------------------------
> > +# Can't use 'cat<<-_EOF_', as Config.in uses leading tabs.
>
> I don't think the indented Config.in block is very readable; I'd use a
> plain <<_EOF_ with no extra indentation. Then you can use cat after all.
Hmm. Maybe that's because my default tabstop is 4, not 8.
I just tried with 8, and it is indeed less readable.
Yet, I prefer that here-documents be indented rather than not.
I'll see to make it readable for tabstop=8.
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 2/7] support/scripts: add a script to add a new package
From: Yann E. MORIN @ 2012-11-01 17:00 UTC (permalink / raw)
To: buildroot
In-Reply-To: <20121101100910.0cfe5b84@skate>
Thomas, All,
On Thursday 01 November 2012 Thomas Petazzoni wrote:
> However, I also wonder if it is a good idea to make this script
> ask questions, as compared to a more conventional script that takes
> command line arguments. But it's true that a script asking questions is
> more like a wizard, probably easier to use.
Indeed, the script is designed as a wizard for newbies, so they do not
find it too complex to add a new package, especially since the
_AVAILABLE stuff can be a bit misleading.
It's also meant for experts to quickly add a new package. For them, it
might be better to get the values from the arguments, rather than asking
questions, right. This script can be refined later for this use-case.
> > > + define ${PKG_NAME}_INSTALL_TARGET_CMDS
> >
> > Putting a sample
> > install -D -m 0644 $(@D)/... $(TARGET_DIR)/...
> > isn't a bad idea.
>
> Not sure I agree here. People too often do manual installation in
> generic packages, while they should use 'make install', possibly after
> tunning/patching the Makefile.
Agreed.
Regards,
Yann E. MORIN.
--
.-----------------.--------------------.------------------.--------------------.
| Yann E. MORIN | Real-Time Embedded | /"\ ASCII RIBBON | Erics' conspiracy: |
| +33 662 376 056 | Software Designer | \ / CAMPAIGN | ___ |
| +33 223 225 172 `------------.-------: X AGAINST | \e/ There is no |
| http://ymorin.is-a-geek.org/ | _/*\_ | / \ HTML MAIL | v conspiracy. |
'------------------------------^-------^------------------^--------------------'
^ permalink raw reply
* [Buildroot] [PATCH 2/7] support/scripts: add a script to add a new package
From: Yann E. MORIN @ 2012-11-01 16:56 UTC (permalink / raw)
To: buildroot
In-Reply-To: <5091D7B6.8070703@mind.be>
Arnout, All,
On Thursday 01 November 2012 Arnout Vandecappelle wrote:
> On 09/10/12 01:40, Yann E. MORIN wrote:
> > --- /dev/null
> > +++ b/docs/manual/adding-packages-script.txt
> > @@ -0,0 +1,41 @@
> > +$ ./support/script/pkg-new
> > +Name of the package: libfoo
> > +
> > +1) none
> > +2) multimedia
> > +3) java
> > +4) x11r7
> > +5) games
> > +Category of your package: 1
>
> I would skip this question and always create it in packages/libfoo. The
> subdirectories are rare, we want to get rid of them, and anyway you can
> easily move the generated directory to a different place after the fact.
OK
> > --- a/docs/manual/adding-packages.txt
> > +++ b/docs/manual/adding-packages.txt
> > @@ -19,4 +19,6 @@ include::adding-packages-handwritten.txt[]
> >
> > include::adding-packages-gettext.txt[]
> >
> > +include::adding-packages-script.txt[]
> > +
>
> I would put this before the rest of adding-packages.
I've poundered this, too, but the script just an /implementation/ of the
documentation, so I think it should go last.
> > --- /dev/null
> > +++ b/support/scripts/pkg-new
> > @@ -0,0 +1,151 @@
> > +#!/bin/bash
>
> Does it have to be bash? Hm, yes, for the arrays... It would be better
> if we can avoid relying on bash for new functionality.
I'll check if I can make it a POSIX-compliant script.
> > +# --------------------------------------
> > +# Can't use 'cat<<-_EOF_', as Config.in uses leading tabs.
>
> I don't think the indented Config.in block is very readable; I'd use a
> plain <<_EOF_ with no extra indentation. Then you can use cat after all.
Indenting here-documents allows one to easily see the end of it. I
personally have difficulties reading scripts where this is not the case.
> > +# --------------------------------------
> > +# Create the package.mk file
> > +cat>"${pkg_dir}/${pkg_name}.mk"<<-_EOF_
>
> Same here, indentation doesn't look natural to me.
Hmmm. Let's ask a neutral party, then! ;-)
> > + #############################################################
> > + #
> > + # ${pkg_name}
> > + #
> > + #############################################################
> > +
> > + ${PKG_NAME}_VERSION =
> > + ${PKG_NAME}_SOURCE =
> > + ${PKG_NAME}_SITE =
> > + ${PKG_NAME}_DEPENDENCIES =
> > + ${PKG_NAME}_LICENSE =
> > + ${PKG_NAME}_LICENSE_FILES =
> > +
> > +_EOF_
>
> For autotools-package and cmake-package, _CONF_OPT is also a very useful one.
> Maybe also add _INSTALL_STAGING.
I think we should fill-in only strictly-required variables.
> > +if [ "${pkg_bs}" = "generic" ]; then
> > + cat>>"${pkg_dir}/${pkg_name}.mk"<<-_EOF_
> > + # See docs/manual/ for the complete list of actions that can
> > + # be defined; only the most common ones are listed below:
> > +
> > + define ${PKG_NAME}_CONFIGURE_CMDS
> > + endef
> > +
> > + define ${PKG_NAME}_BUILD_CMDS
> > + endef
> > +
> > + define ${PKG_NAME}_INSTALL_TARGET_CMDS
>
> Putting a sample
> install -D -m 0644 $(@D)/... $(TARGET_DIR)/...
> isn't a bad idea.
I don't like it. This is just a skeleton file. No default value has been
provided for any variable.
Regards,
Yann E. MORIN.
--
.-----------------.--------------------.------------------.--------------------.
| Yann E. MORIN | Real-Time Embedded | /"\ ASCII RIBBON | Erics' conspiracy: |
| +33 662 376 056 | Software Designer | \ / CAMPAIGN | ___ |
| +33 223 225 172 `------------.-------: X AGAINST | \e/ There is no |
| http://ymorin.is-a-geek.org/ | _/*\_ | / \ HTML MAIL | v conspiracy. |
'------------------------------^-------^------------------^--------------------'
^ permalink raw reply
* [Buildroot] [PATCH 1/7] docs/manual: update 'adding packages' with the new _AVAILABLE symbol
From: Yann E. MORIN @ 2012-11-01 16:21 UTC (permalink / raw)
To: buildroot
In-Reply-To: <5091D0B8.6050404@mind.be>
Arnout, All,
On Thursday 01 November 2012 Arnout Vandecappelle wrote:
> On 09/10/12 01:40, Yann E. MORIN wrote:
> > --- a/docs/manual/adding-packages-autotools.txt
> > +++ b/docs/manual/adding-packages-autotools.txt
> > +12:
> > +13: ifeq ($(BR2_PACKAGE_LIBFOO_FROBBLE),y)
> > +14: LIBFOO_CONF_OPT += --enable-frobble
> > +15: else
> > +16: LIBFOO_CONF_OPT += --disable-frobble
> > +17: endif
> > +18:
> > +19: LIBFOO_CONF_OPT += --with-gazzle-level=$(BR2_PACKAGE_LIBFOO_GAZZLE_LEVEL)
>
> This one is going a bit too far... And if you keep it, it should
> probably have a qstrip.
It's an 'int', so is not quoted in the .config.
(but I'll use a 'string' to address your other comment, so
I'll qstrip it.)
> > +20:
> > +21: ifeq ($(BR2_PACKAGE_LIBFOO_GOO),y)
> > +22: LIBFOO_DEPENDENCIES += goo
> > +23: LIBFOO_CONF_OPT += --enable-goo
> > +24: else
> > +25: LIBFOO_CONF_OPT += --disable-goo
> > +26: endif
>
> While you're at it, an automatic enable/disable example would
> be nice too.
What do you mean by "automatic"?
> > --- a/docs/manual/adding-packages-directory.txt
> > +++ b/docs/manual/adding-packages-directory.txt
> > +* The syntax of the +Config.in+ file is the same as the one for the kernel
> > + Kconfig file. The documentation for this syntax is available at:
> > + https://git.kernel.org/?p=linux/kernel/git/torvalds/linux.git;a=blob;f=Documentation/kbuild/kconfig-language.txt[]
>
> http://kernel.org/doc/Documentation/kbuild/kconfig-language.txt
Does this URL always match the latest version in the git tree?
> > +1. The +BR2_PACKAGE_LIBFOO_AVAILABLE+ symbol shall +depends on+ any
> > +other package's +_AVAILABLE+ symbol. It may also depend on any other
> > +symbol, such as toolchain features, but should not directly depend on
> > +any package's main symbol.
>
> ... except for _XORG7, _PYTHON, etc.
Well, my opinion (FWIW) is those packages should not be treated
differently just because they are /big/.
(One of) the goal(s) of _AVAILABLE is to allow the user to say either:
- I want this package, enable whatever dependencies are required.
or:
- I need this package, but I have to provide a toolchain that has
such and such feature
_AVAILABLE makes that easy.
Then, it's up to the user to understand what pulling-in a package implies.
> > +config BR2_PACKAGE_LIBFOO_GAZZLE_LEVEL
> > + int "Gazzle level"
> > + range 0 10
>
> We currently don't have a single range config option, and only one int.
> So the example is a bit contrived... If you add an extra example, a string
> option that is qstripped in the .mk file is more appropriate.
OK.
> > --- a/docs/manual/adding-packages-generic.txt
> > +++ b/docs/manual/adding-packages-generic.txt
> > @@ -79,11 +102,11 @@ steps should be performed to install the package in the staging space.
> > +LIBFOO_INSTALL_TARGET_CMDS+ tells what steps should be
> > performed to install the package in the target space.
>
> Shouldn't _DEVICES and _PERMISSIONS be explained?
Da, tovarich!
Thank you!
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 9/9] libsoup: Remove redundant dependency on gettext
From: Arnout Vandecappelle @ 2012-11-01 14:36 UTC (permalink / raw)
To: buildroot
In-Reply-To: <20121101143539.8550.70150.stgit@localhost>
From: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>
libsoup doesn't use intl stuff directly, but uses gi18n. libglib2
already has all required dependencies on gettext.
Signed-off-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>
---
package/libsoup/Config.in | 1 -
package/libsoup/libsoup.mk | 2 +-
2 files changed, 1 insertion(+), 2 deletions(-)
diff --git a/package/libsoup/Config.in b/package/libsoup/Config.in
index 57b425f..b79108e 100644
--- a/package/libsoup/Config.in
+++ b/package/libsoup/Config.in
@@ -3,7 +3,6 @@ config BR2_PACKAGE_LIBSOUP
depends on BR2_USE_WCHAR # glib2 and gnutls
select BR2_PACKAGE_LIBXML2
select BR2_PACKAGE_LIBGLIB2
- select BR2_PACKAGE_GETTEXT if BR2_NEEDS_GETTEXT
help
libsoup is an HTTP client/server library. It uses GObject
and the GLib main loop, to integrate well with GNOME
diff --git a/package/libsoup/libsoup.mk b/package/libsoup/libsoup.mk
index 2f173df..2a2d8bf 100644
--- a/package/libsoup/libsoup.mk
+++ b/package/libsoup/libsoup.mk
@@ -19,7 +19,7 @@ endif
LIBSOUP_CONF_OPT = --disable-glibtest --without-gnome
-LIBSOUP_DEPENDENCIES = $(if $(BR2_NEEDS_GETTEXT_IF_LOCALE),gettext) host-pkgconf host-libglib2 libglib2 libxml2
+LIBSOUP_DEPENDENCIES = host-pkgconf host-libglib2 libglib2 libxml2
ifeq ($(BR2_PACKAGE_LIBSOUP_SSL),y)
LIBSOUP_DEPENDENCIES += glib-networking
^ permalink raw reply related
* [Buildroot] [PATCH 3 8/9] gdk-pixbuf: remove dependency on gettext
From: Arnout Vandecappelle @ 2012-11-01 14:36 UTC (permalink / raw)
To: buildroot
In-Reply-To: <20121101143539.8550.70150.stgit@localhost>
From: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>
The dependency on gettext comes from libglib2, not from gdk-pixbuf
itself.
Signed-off-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>
---
package/gdk-pixbuf/Config.in | 1 -
package/gdk-pixbuf/gdk-pixbuf.mk | 1 -
2 files changed, 2 deletions(-)
diff --git a/package/gdk-pixbuf/Config.in b/package/gdk-pixbuf/Config.in
index 33b5f7c..f205793 100644
--- a/package/gdk-pixbuf/Config.in
+++ b/package/gdk-pixbuf/Config.in
@@ -2,7 +2,6 @@ config BR2_PACKAGE_GDK_PIXBUF
bool "gdk-pixbuf"
depends on BR2_USE_WCHAR # glib2
select BR2_PACKAGE_LIBGLIB2
- select BR2_PACKAGE_GETTEXT if BR2_NEEDS_GETTEXT
select BR2_PACKAGE_LIBICONV if !BR2_ENABLE_LOCALE
help
Gdk-Pixbuf is an image loader and scaler. It uses GObject
diff --git a/package/gdk-pixbuf/gdk-pixbuf.mk b/package/gdk-pixbuf/gdk-pixbuf.mk
index 4e00c40..8c785cd 100644
--- a/package/gdk-pixbuf/gdk-pixbuf.mk
+++ b/package/gdk-pixbuf/gdk-pixbuf.mk
@@ -40,7 +40,6 @@ GDK_PIXBUF_DEPENDENCIES += xlib_libX11
endif
GDK_PIXBUF_DEPENDENCIES += \
- $(if $(BR2_NEEDS_GETTEXT_IF_LOCALE),gettext) \
$(if $(BR2_ENABLE_LOCALE),,libiconv) \
host-pkgconf libglib2
^ permalink raw reply related
* [Buildroot] [PATCH 3 7/9] pulseaudio: remove dependency on gettext
From: Arnout Vandecappelle @ 2012-11-01 14:36 UTC (permalink / raw)
To: buildroot
In-Reply-To: <20121101143539.8550.70150.stgit@localhost>
From: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>
The dependency on gettext is only due to libglib2, not due to
pulseaudio itself. It works fine without gettext if libglib2 is
not selected.
Signed-off-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>
---
v2: actually remove the gettext dependency this time; thanks Thomas
---
package/multimedia/pulseaudio/Config.in | 1 -
package/multimedia/pulseaudio/pulseaudio.mk | 1 -
2 files changed, 2 deletions(-)
diff --git a/package/multimedia/pulseaudio/Config.in b/package/multimedia/pulseaudio/Config.in
index 6c767bb..0844192 100644
--- a/package/multimedia/pulseaudio/Config.in
+++ b/package/multimedia/pulseaudio/Config.in
@@ -5,7 +5,6 @@ config BR2_PACKAGE_PULSEAUDIO
select BR2_PACKAGE_JSON_C
select BR2_PACKAGE_LIBSNDFILE
select BR2_PACKAGE_SPEEX
- select BR2_PACKAGE_GETTEXT if BR2_NEEDS_GETTEXT
help
PulseAudio is a sound system for POSIX OSes, meaning that it
is a proxy for your sound applications. It allows you to do
diff --git a/package/multimedia/pulseaudio/pulseaudio.mk b/package/multimedia/pulseaudio/pulseaudio.mk
index e78eb42..a842180 100644
--- a/package/multimedia/pulseaudio/pulseaudio.mk
+++ b/package/multimedia/pulseaudio/pulseaudio.mk
@@ -16,7 +16,6 @@ PULSEAUDIO_CONF_OPT = \
PULSEAUDIO_DEPENDENCIES = \
host-pkgconf libtool json-c libsndfile speex host-intltool \
- $(if $(BR2_NEEDS_GETTEXT),gettext) \
$(if $(BR2_PACKAGE_LIBATOMIC_OPS),libatomic_ops) \
$(if $(BR2_PACKAGE_LIBSAMPLERATE),libsamplerate) \
$(if $(BR2_PACKAGE_ALSA_LIB),alsa-lib) \
^ permalink raw reply related
* [Buildroot] [PATCH 3 6/9] avahi: remove double dependency on gettext
From: Arnout Vandecappelle @ 2012-11-01 14:36 UTC (permalink / raw)
To: buildroot
In-Reply-To: <20121101143539.8550.70150.stgit@localhost>
From: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>
Signed-off-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>
Acked-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
---
package/avahi/avahi.mk | 5 +----
1 file changed, 1 insertion(+), 4 deletions(-)
diff --git a/package/avahi/avahi.mk b/package/avahi/avahi.mk
index 5f07a2f..d96f24b 100644
--- a/package/avahi/avahi.mk
+++ b/package/avahi/avahi.mk
@@ -133,10 +133,7 @@ else
AVAHI_CONF_OPT += --disable-python
endif
-ifeq ($(BR2_PACKAGE_GETTEXT),y)
-AVAHI_DEPENDENCIES += gettext
-AVAHI_MAKE_OPT = LIBS=-lintl
-endif
+AVAHI_MAKE_OPT += $(if $(BR2_NEEDS_GETTEXT_IF_LOCALE),LIBS=-lintl)
define AVAHI_REMOVE_INITSCRIPT
rm -rf $(TARGET_DIR)/etc/init.d/avahi-*
^ permalink raw reply related
* [Buildroot] [PATCH 3 5/9] lshw: fix build with no LOCALE support
From: Arnout Vandecappelle @ 2012-11-01 14:36 UTC (permalink / raw)
To: buildroot
In-Reply-To: <20121101143539.8550.70150.stgit@localhost>
From: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>
lshw doesn't need libintl if -DNONLS is defined.
Signed-off-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>
---
package/lshw/Config.in | 2 +-
package/lshw/lshw.mk | 12 +++++++++---
2 files changed, 10 insertions(+), 4 deletions(-)
diff --git a/package/lshw/Config.in b/package/lshw/Config.in
index 98beacd..afefa6c 100644
--- a/package/lshw/Config.in
+++ b/package/lshw/Config.in
@@ -3,7 +3,7 @@ config BR2_PACKAGE_LSHW
depends on BR2_INSTALL_LIBSTDCPP
depends on BR2_LARGEFILE
depends on BR2_USE_WCHAR
- select BR2_PACKAGE_GETTEXT if BR2_NEEDS_GETTEXT
+ select BR2_PACKAGE_GETTEXT if BR2_NEEDS_GETTEXT_IF_LOCALE
help
lshw (Hardware Lister) is a small tool to provide
detailed information on the hardware configuration of the machine.
diff --git a/package/lshw/lshw.mk b/package/lshw/lshw.mk
index 68daa0d..b3c2214 100644
--- a/package/lshw/lshw.mk
+++ b/package/lshw/lshw.mk
@@ -6,10 +6,16 @@
LSHW_VERSION = B.02.16
LSHW_SITE = http://ezix.org/software/files
+
+LSHW_CFLAGS = $(TARGET_CFLAGS)
+ifeq ($(BR2_ENABLE_LOCALE),)
+LSHW_CFLAGS += -DNONLS
+endif
+
LSHW_MAKE_OPT = CC="$(TARGET_CC)" CXX="$(TARGET_CXX)" AR="$(TARGET_AR)" \
- RPM_OPT_FLAGS="$(TARGET_CFLAGS)" all
-LSHW_MAKE_ENV = LIBS="$(if $(BR2_NEEDS_GETTEXT),-lintl)"
-LSHW_DEPENDENCIES = $(if $(BR2_NEEDS_GETTEXT),gettext)
+ RPM_OPT_FLAGS="$(LSHW_CFLAGS)" all
+LSHW_MAKE_ENV = LIBS="$(if $(BR2_NEEDS_GETTEXT_IF_LOCALE),-lintl)"
+LSHW_DEPENDENCIES = $(if $(BR2_NEEDS_GETTEXT_IF_LOCALE),gettext)
define LSHW_BUILD_CMDS
$(LSHW_MAKE_ENV) $(MAKE) -C $(@D)/src $(LSHW_MAKE_OPT)
^ permalink raw reply related
* [Buildroot] [PATCH 3 4/9] util-linux: only needs gettext if locale is selected
From: Arnout Vandecappelle @ 2012-11-01 14:35 UTC (permalink / raw)
To: buildroot
In-Reply-To: <20121101143539.8550.70150.stgit@localhost>
From: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>
Also add missing select in Config.in.
Signed-off-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>
Acked-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
---
package/util-linux/Config.in | 1 +
package/util-linux/util-linux.mk | 2 +-
2 files changed, 2 insertions(+), 1 deletion(-)
diff --git a/package/util-linux/Config.in b/package/util-linux/Config.in
index 43a8444..88d08e7 100644
--- a/package/util-linux/Config.in
+++ b/package/util-linux/Config.in
@@ -2,6 +2,7 @@ config BR2_PACKAGE_UTIL_LINUX
bool "util-linux"
depends on BR2_LARGEFILE
depends on BR2_USE_WCHAR
+ select BR2_PACKAGE_GETTEXT if BR2_NEEDS_GETTEXT_IF_LOCALE
help
Various useful/essential Linux utilities.
diff --git a/package/util-linux/util-linux.mk b/package/util-linux/util-linux.mk
index f6028f4..f15baa7 100644
--- a/package/util-linux/util-linux.mk
+++ b/package/util-linux/util-linux.mk
@@ -29,7 +29,7 @@ else
UTIL_LINUX_CONF_OPT += --without-ncurses
endif
-ifeq ($(BR2_PACKAGE_GETTEXT),y)
+ifeq ($(BR2_NEEDS_GETTEXT_IF_LOCALE),y)
UTIL_LINUX_DEPENDENCIES += gettext
UTIL_LINUX_MAKE_OPT += LIBS=-lintl
endif
^ permalink raw reply related
* [Buildroot] [PATCH 3 3/9] ndisc6: only needs gettext if locale is selected
From: Arnout Vandecappelle @ 2012-11-01 14:35 UTC (permalink / raw)
To: buildroot
In-Reply-To: <20121101143539.8550.70150.stgit@localhost>
From: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>
Also add missing select in Config.in.
Signed-off-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>
Acked-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
---
package/ndisc6/Config.in | 1 +
package/ndisc6/ndisc6.mk | 2 +-
2 files changed, 2 insertions(+), 1 deletion(-)
diff --git a/package/ndisc6/Config.in b/package/ndisc6/Config.in
index 126bfa2..6abe4ce 100644
--- a/package/ndisc6/Config.in
+++ b/package/ndisc6/Config.in
@@ -4,6 +4,7 @@ comment "ndisc6 requires a toolchain with IPv6 support"
config BR2_PACKAGE_NDISC6
bool "ndisc6 tools"
depends on BR2_INET_IPV6
+ select BR2_PACKAGE_GETTEXT if BR2_NEEDS_GETTEXT_IF_LOCALE
help
NDisc6 is a small collection of useful tools for IPv6 networking
diff --git a/package/ndisc6/ndisc6.mk b/package/ndisc6/ndisc6.mk
index f58cc13..1d3c0e7 100644
--- a/package/ndisc6/ndisc6.mk
+++ b/package/ndisc6/ndisc6.mk
@@ -10,7 +10,7 @@ NDISC6_SITE = http://www.remlab.net/files/ndisc6/
NDISC6_CONF_ENV = CC="$(TARGET_CC) -std=gnu99"
NDISC6_CONF_OPT = --localstatedir=/var --disable-rpath --disable-suid-install
-ifeq ($(BR2_PACKAGE_GETTEXT),y)
+ifeq ($(BR2_NEEDS_GETTEXT_IF_LOCALE),y)
NDISC6_DEPENDENCIES += gettext
NDISC6_CONF_ENV += LDFLAGS="$(TARGET_LDFLAGS) -lintl"
endif
^ permalink raw reply related
* [Buildroot] [PATCH 3 2/9] flex: only needs gettext if locale is selected
From: Arnout Vandecappelle @ 2012-11-01 14:35 UTC (permalink / raw)
To: buildroot
In-Reply-To: <20121101143539.8550.70150.stgit@localhost>
From: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>
Also add missing select in Config.in.
Signed-off-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>
Acked-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
---
package/flex/Config.in | 1 +
package/flex/flex.mk | 3 +--
2 files changed, 2 insertions(+), 2 deletions(-)
diff --git a/package/flex/Config.in b/package/flex/Config.in
index b10eb00..8ce4ccd 100644
--- a/package/flex/Config.in
+++ b/package/flex/Config.in
@@ -1,5 +1,6 @@
config BR2_PACKAGE_FLEX
bool "flex"
+ select BR2_PACKAGE_GETTEXT if BR2_NEEDS_GETTEXT_IF_LOCALE
help
A fast lexical analyser generator. A tool for generating
programs that perform pattern-matching on text.
diff --git a/package/flex/flex.mk b/package/flex/flex.mk
index 2599cdc..fe21432 100644
--- a/package/flex/flex.mk
+++ b/package/flex/flex.mk
@@ -11,8 +11,7 @@ FLEX_PATCH = flex_$(FLEX_VERSION)-$(FLEX_PATCH_VERSION).diff.gz
FLEX_SITE = $(BR2_DEBIAN_MIRROR)/debian/pool/main/f/flex
FLEX_DIR = $(BUILD_DIR)/flex-$(FLEX_VERSION)
FLEX_INSTALL_STAGING = YES
-FLEX_DEPENDENCIES = \
- $(if $(BR2_PACKAGE_GETTEXT),gettext)
+FLEX_DEPENDENCIES = $(if $(BR2_NEEDS_GETTEXT_IF_LOCALE),gettext)
# we don't have a host-gettext/libintl
HOST_FLEX_DEPENDENCIES =
^ permalink raw reply related
* [Buildroot] [PATCH 3 1/9] diffutils: only needs gettext if locale is selected
From: Arnout Vandecappelle @ 2012-11-01 14:35 UTC (permalink / raw)
To: buildroot
From: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>
Also add missing select in Config.in.
Signed-off-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>
Acked-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
---
package/diffutils/Config.in | 1 +
package/diffutils/diffutils.mk | 3 +--
2 files changed, 2 insertions(+), 2 deletions(-)
diff --git a/package/diffutils/Config.in b/package/diffutils/Config.in
index 86da5d3..4cc16e2 100644
--- a/package/diffutils/Config.in
+++ b/package/diffutils/Config.in
@@ -1,6 +1,7 @@
config BR2_PACKAGE_DIFFUTILS
bool"diffutils"
depends on BR2_USE_WCHAR
+ select BR2_PACKAGE_GETTEXT if BR2_NEEDS_GETTEXT_IF_LOCALE
help
GNU diff. Compare files per line.
diff --git a/package/diffutils/diffutils.mk b/package/diffutils/diffutils.mk
index 168d040..2530763 100644
--- a/package/diffutils/diffutils.mk
+++ b/package/diffutils/diffutils.mk
@@ -6,8 +6,7 @@
DIFFUTILS_VERSION = 3.2
DIFFUTILS_SITE = $(BR2_GNU_MIRROR)/diffutils
-DIFFUTILS_DEPENDENCIES = \
- $(if $(BR2_PACKAGE_GETTEXT),gettext)
+DIFFUTILS_DEPENDENCIES = $(if $(BR2_NEEDS_GETTEXT_IF_LOCALE),gettext)
ifeq ($(BR2_PACKAGE_BUSYBOX),y)
DIFFUTILS_DEPENDENCIES += busybox
^ permalink raw reply related
* [Buildroot] [PATCH 10/10] libsoup: remove redundant dependency on gettext
From: Arnout Vandecappelle @ 2012-11-01 14:26 UTC (permalink / raw)
To: buildroot
In-Reply-To: <201210142324.20324.yann.morin.1998@free.fr>
On 10/14/12 23:24, Yann E. MORIN wrote:
> Or maybe you meant that gettext is needed by libglib2, not by libsoup
> itself? In which case you want to remove the select altogether.
In fact, libsoup doesn't use libintl directly, but does use gi18n. So
I guess the dependency can indeed be removed.
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 09/10] glib-networking: remove redundant dependency on gettext
From: Arnout Vandecappelle @ 2012-11-01 14:02 UTC (permalink / raw)
To: buildroot
In-Reply-To: <50927D91.6060804@mind.be>
On 11/01/12 14:48, Arnout Vandecappelle wrote:
> On 10/14/12 23:21, Yann E. MORIN wrote:
>> Arnout, All,
>>
>> On Sunday 14 October 2012 Arnout Vandecappelle (Essensium/Mind) wrote:
>>> Signed-off-by: Arnout Vandecappelle (Essensium/Mind)<arnout@mind.be>
>>> ---
>>> package/glib-networking/Config.in | 2 +-
>>> 1 file changed, 1 insertion(+), 1 deletion(-)
>>>
>>> diff --git a/package/glib-networking/Config.in b/package/glib-networking/Config.in
>>> index dab35ea..f1aa7ee 100644
>>> --- a/package/glib-networking/Config.in
>>> +++ b/package/glib-networking/Config.in
>>> @@ -2,7 +2,7 @@ config BR2_PACKAGE_GLIB_NETWORKING
>>> bool "glib-networking"
>>> depends on BR2_USE_WCHAR # glib2
>>> select BR2_PACKAGE_LIBGLIB2
>>> - select BR2_PACKAGE_GETTEXT if BR2_NEEDS_GETTEXT
>>> + select BR2_PACKAGE_GETTEXT if BR2_NEEDS_GETTEXT_IF_LOCALE
>>> help
>>> Network-related GIO modules for glib.
>>
>> The commit message says "remove redundant dependency on gettext", but this
>> change changes the dependency on plain gettext to gettext-if-locales.
>>
>> Thus, I think the subject is wrong, and should be changed to something like:
>> glib-networking: only depends on gettext if locales are used
>
> Actually, on second revision: glib-networking uses gettext directly so we should
> therefore select it here as well. It worked fine in testing because libglib2
> anyway depends on locale.
>
> Therefore, I'll replace this with unconditionally depending on gettext in
> the .mk file.
... but of course, the whole point of this series was to remove BR2_NEEDS_GETTEXT...
I'm just going to drop this patch for the time being, which leaves us with 3
packages using BR2_NEEDS_GETTEXT: php (gettext module), libglib2, glib-networking.
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 09/10] glib-networking: remove redundant dependency on gettext
From: Arnout Vandecappelle @ 2012-11-01 13:48 UTC (permalink / raw)
To: buildroot
In-Reply-To: <201210142321.11131.yann.morin.1998@free.fr>
On 10/14/12 23:21, Yann E. MORIN wrote:
> Arnout, All,
>
> On Sunday 14 October 2012 Arnout Vandecappelle (Essensium/Mind) wrote:
>> Signed-off-by: Arnout Vandecappelle (Essensium/Mind)<arnout@mind.be>
>> ---
>> package/glib-networking/Config.in | 2 +-
>> 1 file changed, 1 insertion(+), 1 deletion(-)
>>
>> diff --git a/package/glib-networking/Config.in b/package/glib-networking/Config.in
>> index dab35ea..f1aa7ee 100644
>> --- a/package/glib-networking/Config.in
>> +++ b/package/glib-networking/Config.in
>> @@ -2,7 +2,7 @@ config BR2_PACKAGE_GLIB_NETWORKING
>> bool "glib-networking"
>> depends on BR2_USE_WCHAR # glib2
>> select BR2_PACKAGE_LIBGLIB2
>> - select BR2_PACKAGE_GETTEXT if BR2_NEEDS_GETTEXT
>> + select BR2_PACKAGE_GETTEXT if BR2_NEEDS_GETTEXT_IF_LOCALE
>> help
>> Network-related GIO modules for glib.
>
> The commit message says "remove redundant dependency on gettext", but this
> change changes the dependency on plain gettext to gettext-if-locales.
>
> Thus, I think the subject is wrong, and should be changed to something like:
> glib-networking: only depends on gettext if locales are used
Actually, on second revision: glib-networking uses gettext directly so we should
therefore select it here as well. It worked fine in testing because libglib2
anyway depends on locale.
Therefore, I'll replace this with unconditionally depending on gettext in
the .mk file.
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] uClibc: install libc.so even if BR2_PREFER_STATIC_LIB is enabled
From: Eial Czerwacki @ 2012-11-01 11:28 UTC (permalink / raw)
To: buildroot
In-Reply-To: <1351764953-24251-1-git-send-email-arnout@mind.be>
On 11/01/2012 12:15 PM, Arnout Vandecappelle (Essensium/Mind) wrote:
> BR2_PREFER_STATIC_LIB _prefers_ static linking, but doesn't force it for
> all packages. So some binaries may still be built without -static, and
> they will need libc.so on the target.
>
> Therefore, put libc.so.0 in the target unconditionally.
>
> Signed-off-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>
> ---
> Untested!
>
> toolchain/uClibc/uclibc.mk | 2 --
> 1 file changed, 2 deletions(-)
>
> diff --git a/toolchain/uClibc/uclibc.mk b/toolchain/uClibc/uclibc.mk
> index 397cdd4..65ff90b 100644
> --- a/toolchain/uClibc/uclibc.mk
> +++ b/toolchain/uClibc/uclibc.mk
> @@ -486,9 +486,7 @@ $(TARGET_DIR)/usr/bin/ldd: $(cross_compiler)
> PREFIX=$(TARGET_DIR) utils install_utils
> touch -c $@
>
> -ifneq ($(BR2_PREFER_STATIC_LIB),y)
> UCLIBC_TARGETS=$(TARGET_DIR)/lib/libc.so.0
> -endif
>
> ifeq ($(BR2_UCLIBC_INSTALL_TEST_SUITE),y)
> UCLIBC_TARGETS+=uclibc-test
I can confirm it works:
# find / -name "libc*"
/lib/libcrypt.so.0
/lib/libc.so.0
/lib/libcrypt-0.9.32.1.so
/usr/include/bits/libc-lock.h
/usr/lib/libc.a
/usr/lib/libcrypt_pic.a
/usr/lib/libc_pic.a
/usr/lib/libcurses.a
/usr/lib/libcrypt.a
^ permalink raw reply
* [Buildroot] [PATCH 2/2] linux: bump 3.6.x stable version
From: Gustavo Zacarias @ 2012-11-01 11:07 UTC (permalink / raw)
To: buildroot
In-Reply-To: <1351768068-2241-1-git-send-email-gustavo@zacarias.com.ar>
Signed-off-by: Gustavo Zacarias <gustavo@zacarias.com.ar>
---
linux/Config.in | 4 ++--
1 files changed, 2 insertions(+), 2 deletions(-)
diff --git a/linux/Config.in b/linux/Config.in
index a849849..60d7794 100644
--- a/linux/Config.in
+++ b/linux/Config.in
@@ -21,7 +21,7 @@ choice
prompt "Kernel version"
config BR2_LINUX_KERNEL_3_6
- bool "3.6.4"
+ bool "3.6.5"
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 "3.6.4" if BR2_LINUX_KERNEL_3_6
+ default "3.6.5" 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
--
1.7.8.6
^ permalink raw reply related
* [Buildroot] [PATCH 1/2] kernel-headers: bump 3.{0, 2, 4, 6}.x stable versions
From: Gustavo Zacarias @ 2012-11-01 11:07 UTC (permalink / raw)
To: buildroot
Signed-off-by: Gustavo Zacarias <gustavo@zacarias.com.ar>
---
| 8 ++++----
...all-fix-__packed-in-exported-kernel-head.patch} | 0
2 files changed, 4 insertions(+), 4 deletions(-)
rename toolchain/kernel-headers/{linux-3.0.49-headers_install-fix-__packed-in-exported-kernel-head.patch => linux-3.0.50-headers_install-fix-__packed-in-exported-kernel-head.patch} (100%)
--git a/toolchain/kernel-headers/Config.in b/toolchain/kernel-headers/Config.in
index 2270e0c..d596870 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 "3.0.49" if BR2_KERNEL_HEADERS_3_0
+ default "3.0.50" if BR2_KERNEL_HEADERS_3_0
default "3.1.10" if BR2_KERNEL_HEADERS_3_1
- default "3.2.32" if BR2_KERNEL_HEADERS_3_2
+ default "3.2.33" if BR2_KERNEL_HEADERS_3_2
default "3.3.8" if BR2_KERNEL_HEADERS_3_3
- default "3.4.16" if BR2_KERNEL_HEADERS_3_4
+ default "3.4.17" if BR2_KERNEL_HEADERS_3_4
default "3.5.7" if BR2_KERNEL_HEADERS_3_5
- default "3.6.4" if BR2_KERNEL_HEADERS_3_6
+ default "3.6.5" 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-3.0.49-headers_install-fix-__packed-in-exported-kernel-head.patch b/toolchain/kernel-headers/linux-3.0.50-headers_install-fix-__packed-in-exported-kernel-head.patch
similarity index 100%
rename from toolchain/kernel-headers/linux-3.0.49-headers_install-fix-__packed-in-exported-kernel-head.patch
rename to toolchain/kernel-headers/linux-3.0.50-headers_install-fix-__packed-in-exported-kernel-head.patch
--
1.7.8.6
^ 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