uboot: improve help text of BR2_TARGET_UBOOT_BOARDNAME

Signed-off-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>
Signed-off-by: Peter Korsgaard <jacmet@sunsite.dk>

u-boot: add version 2012.07

Signed-off-by: Gustavo Zacarias <gustavo@zacarias.com.ar>
Signed-off-by: Peter Korsgaard <jacmet@sunsite.dk>

Move BR2_SPARC_TYPE to an uClibc specific place

BR2_SPARC_TYPE is a hidden configuration option that is only used for
the configuration of uClibc, therefore, we move it from
target/Config.arch.in to toolchain/uClibc/Config.in.

Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Signed-off-by: Peter Korsgaard <jacmet@sunsite.dk>

Move BR2_ARM_TYPE to an uClibc specific place

BR2_ARM_TYPE is a hidden configuration option that is only used for
the configuration of uClibc, therefore, we move it from
target/Config.arch.in to toolchain/uClibc/Config.in.

We also add a comment that explains that this stuff is only useful for
uClibc <= 0.9.32. Starting from 0.9.33, uClibc build process simply
uses the compiler flags to find the ARM processor that should be
used. So, someday, we'll be able to remove this.

Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Signed-off-by: Peter Korsgaard <jacmet@sunsite.dk>

dnsmasq: bump to version 2.63

Signed-off-by: Gustavo Zacarias <gustavo@zacarias.com.ar>
Signed-off-by: Peter Korsgaard <jacmet@sunsite.dk>

alsamixergui: use a normal 'select' statement for the alsa-lib dependency

Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Signed-off-by: Peter Korsgaard <jacmet@sunsite.dk>

alsa-lib: require thread support

Some parts of alsa-lib correctly use threads only when available, but
some other parts, especially certain PCM plugins, unconditionally
require threads. While it would certainly be possible to fix alsa-lib
to only use threads when available, it probably doesn't make much
sense, since on an embedded system that has audio, we are probably
powerful enough to enable thread support in the C library.

Fixes:

  http://autobuild.buildroot.org/results/e14469be7f6171f4c8c0c09c8e32943819f7938b/build-end.log

Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Signed-off-by: Peter Korsgaard <jacmet@sunsite.dk>

alsa-lib: put BR2_PACKAGE_ALSA_LIB_PYTHON under the existing if condition

Instead of duplicating the "depends on BR2_PACKAGE_ALSA_LIB", just put
BR2_PACKAGE_ALSA_LIB_PYTHON under the existing condition "if
BR2_PACKAGE_ALSA_LIB".

Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Acked-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>
Signed-off-by: Peter Korsgaard <jacmet@sunsite.dk>

package/pulseaudio: fix build-dependency on gettext

In its Config.in, pulseaudio declares a dependency on gettext
using the documented construct:
    select BR2_PACKAGE_GETTEXT if BR2_NEEDS_GETTEXT

but in its .mk file, declares a dependency on the package:
    PULSEAUDIO_DEPENDENCIES = [...] $(if $(BR2_PACKAGE_GETTEXT),gettext) [...]

instead of the documented:
    $(if $(BR2_NEEDS_GETTEXT),gettext)

Fix that.

Signed-off-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Tested-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>
Acked-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>
CC: Samuel Martin <s.martin49@gmail.com>
Signed-off-by: Peter Korsgaard <jacmet@sunsite.dk>

package/lshw: add missing dependency on gettext

lshw.mk defines a build-dependency on gettext,
but not in its Config.in.

Signed-off-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Tested-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>
Acked-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>
CC: Samuel Martin <s.martin49@gmail.com>
Signed-off-by: Peter Korsgaard <jacmet@sunsite.dk>

doc: update after gettext changes

Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Acked-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Tested-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>
Acked-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>
CC: Samuel Martin <s.martin49@gmail.com>
Signed-off-by: Peter Korsgaard <jacmet@sunsite.dk>

gettext: convert to autotools-package

Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Tested-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
[yann.morin.1998@free.fr: add license info, build libintl if locales disabled]
Signed-off-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
CC: Samuel Martin <s.martin49@gmail.com>
Signed-off-by: Peter Korsgaard <jacmet@sunsite.dk>

Remove all references to libintl

From now on, packages only need to select the BR2_PACKAGE_GETTEXT
option and depend on the 'gettext' package to get the necessary i18n
libraries installed on the target.

Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Tested-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
[yann.morin.1998@free.fr: remove BR2_PACKAGE_LIBINTL]
Signed-off-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Tested-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>
Acked-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>
CC: Samuel Martin <s.martin49@gmail.com>
Signed-off-by: Peter Korsgaard <jacmet@sunsite.dk>

gettext: remove libintl handling

Now, the gettext package always:

 * install everything to the staging directory
 * install the libraries to the target directory

Note that this commit removes BR2_PACKAGE_LIBINTL without updating the
packages, it is done in a separate commit to ease the review process.

Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Tested-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
[yann.morin.1998@free.fr: keep BR2_PACKAGE_LIBINTL as blind option]
Signed-off-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Tested-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>
Acked-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>
CC: Samuel Martin <s.martin49@gmail.com>
Signed-off-by: Peter Korsgaard <jacmet@sunsite.dk>

gettext: remove the gettext-target make target, unused

Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Tested-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Acked-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Tested-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>
Acked-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>
CC: Samuel Martin <s.martin49@gmail.com>
Signed-off-by: Peter Korsgaard <jacmet@sunsite.dk>

gettext: remove option to build statically

Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Tested-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Acked-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Tested-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>
Acked-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>
CC: Samuel Martin <s.martin49@gmail.com>
Signed-off-by: Peter Korsgaard <jacmet@sunsite.dk>

fbdump: only build VGA16 support on x86/x86-64

The VGA16 dumping feature probably doesn't make much sense on anything
else than x86 and x86-64, as it is accessing I/O ports that seem
PC-specific.

Moreover, some architectures such as SH4, do not have an
implementation of outw() in uClibc, but we quite certainly don't care
about VGA16 support for such architectures.

Fixes:

  http://autobuild.buildroot.org/results/e1c860830dd585830cf5ca5f935b81856e992334/build-end.log

Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Signed-off-by: Peter Korsgaard <jacmet@sunsite.dk>

leafpad: needs host-intltool

Fixes:

  http://autobuild.buildroot.org/results/d313a67b3deca262844d1043eafa5f6c5b6b7cad/build-end.log

Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Signed-off-by: Peter Korsgaard <jacmet@sunsite.dk>

dhcp: security bump to version 4.1-ESV-R7

Bump dhcp to version 4.1-ESV-R7 to fix CVE-2012-3955

Signed-off-by: Gustavo Zacarias <gustavo@zacarias.com.ar>
Signed-off-by: Peter Korsgaard <jacmet@sunsite.dk>

mplayer: not supported on sh4a/sh4aeb

The mplayer configure script says that sh4a is not supported,
according to:

 http://autobuild.buildroot.org/results/c6457c2dea7c1df4d8e0f23995ebfb3285a85aed/build-end.log

Apparently, sh4 is supported, but not sh4a, even though it seems that
sh4a is a super-set of sh4. So it would probably not be too difficult
to get mplayer to build/work for sh4a, but we don't have hardware to
test this easily, and we can leave this work to the first developer
who will need mplayer on sh4a: let's only support what the upstream
package supports.

Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Signed-off-by: Peter Korsgaard <jacmet@sunsite.dk>

vala: bump version

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

libnspr: requires thread support in toolchain

Fixes

 http://autobuild.buildroot.org/results/58e0784e2289a60fa51589d2ce0a7a8ffe40f07e/build-end.log

Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Signed-off-by: Peter Korsgaard <jacmet@sunsite.dk>

vala: fix documentation build problem

On gcc10, the build of the host-vala documentation causes a
segmentation fault of xsltproc, for some unknown reason. See:

  http://autobuild.buildroot.org/results/42fea25ac4fe33a56c02d90c361da1c494f79ba3/build-end.log

Therefore, this patch fixes the problem by:

 * Ensuring that host-libxslt is amongst the dependency of vala when
   documentation is enabled so that we build a well-known version of
   libxslt/xsltproc to build the documentation. If documentation is
   not enabled, then we force vala to *not* find xsltproc so that it
   doesn't try to use the xsltproc of the build machine distribution
   if available.

 * Disabling the build of the documentation for host-vala. Note that
   we need to pass ':' as the xsltproc binary name to teach the
   configure script that xsltproc is not available.

Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Signed-off-by: Peter Korsgaard <jacmet@sunsite.dk>

Revert "poco: fix parallel build failure"

This reverts commit 32f4957b153bdabe7af60d529942aca7d1a4783d.

Adding the phony objdir target as a prerequisite of all the .o targets
doesn't work, as they are then always considered out of date, leading
to poco getting rebuilt (without the proper MYSQL_{INC,LIB}DIR settings)
at staging-install / target-install time.

Instead simply use MAKE1 for the build step.

Fixes http://autobuild.buildroot.net/results/1cca8b4115674f12884bcbc0c680efed0fb939ae

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

erlang: don't build the Java stuff on the host

We don't need the Java stuff, and it causes build failures on certain
machines:

 http://autobuild.buildroot.org/results/99035e89e101c41d5e8b5ef6a025856f4af88ae2/build-end.log

Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Signed-off-by: Peter Korsgaard <jacmet@sunsite.dk>

libmad: add Thumb2 patches from Debian

This fixes recurrent libmad build failures reported by the
autobuilder. See:

  http://autobuild.buildroot.net/results/a10c943e74b6e67e69183a35b5762482e357a75f/build-end.log

Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Signed-off-by: Peter Korsgaard <jacmet@sunsite.dk>

linux: bump default to kernel version 3.5.4

Signed-off-by: Gustavo Zacarias <gustavo@zacarias.com.ar>
Signed-off-by: Peter Korsgaard <jacmet@sunsite.dk>

kernel-headers: bump 3.{0, 4, 5}.x stable versions

Signed-off-by: Gustavo Zacarias <gustavo@zacarias.com.ar>
Signed-off-by: Peter Korsgaard <jacmet@sunsite.dk>

package: add new filesystem utilities category

Add the new filesystem utilities category and move a bunch of packages
into it.

Signed-off-by: Gustavo Zacarias <gustavo@zacarias.com.ar>
Acked-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Signed-off-by: Peter Korsgaard <jacmet@sunsite.dk>

poco: fix parallel build failure

This added patch should fix
http://autobuild.buildroot.net/results/1bf7e51ef30af9bbf5e423e80ef07212b83bdaf8/.

CROSSENV is now required also in the install stages to prevent a rebuild using
the host native toolchain.

Signed-off-by: Baruch Siach <baruch@tkos.co.il>
Signed-off-by: Peter Korsgaard <jacmet@sunsite.dk>

kmod: bump version

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

qt: bump version

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

vsftpd: add linux-pam support

Signed-off-by: Danomi Manchego <danomimanchego123@gmail.com>
Signed-off-by: Peter Korsgaard <jacmet@sunsite.dk>

wpa_supplicant: add license information

v2 - license type is GPLv2/BSD-3c as pointed out by Baruch Siach

Signed-off-by: Mark Jackson <mpfj@newflow.co.uk>
Signed-off-by: Peter Korsgaard <jacmet@sunsite.dk>

patch: bump to version 2.7

Signed-off-by: Gustavo Zacarias <gustavo@zacarias.com.ar>
Signed-off-by: Peter Korsgaard <jacmet@sunsite.dk>

unionfs: bump to version 0.26

Signed-off-by: Gustavo Zacarias <gustavo@zacarias.com.ar>
Signed-off-by: Peter Korsgaard <jacmet@sunsite.dk>

hiawatha: bump to version 8.5

[Peter: license is explictly GPLv2, not GPLv2+]
Signed-off-by: Gustavo Zacarias <gustavo@zacarias.com.ar>
Signed-off-by: Peter Korsgaard <jacmet@sunsite.dk>

kernel-headers: bump 3.2.x stable version

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

gst-plugin-bad: add musepack option

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

gst-plugins-bad: opus + jpegformat: unbreak non-debug build

opus + jpegformat plugin builds fail when gstreamer is configured with
--disable-gst-debug as they are checking the GST_DISABLE_DEBUG symbol
instead of GST_DISABLE_GST_DEBUG.

Submitted upstream as https://bugzilla.gnome.org/show_bug.cgi?id=683850

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

gst-plugin-bad: add opus option

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

package: add opus-tools

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

package: add opus audio codec library

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

lcdproc: fix build with high BR2_JLEVEL setting

Fixes http://autobuild.buildroot.net/results/646db4b6b284beb4392759a57efffc8d10ee9f49

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

pppd: fix build with 3.5+ kernel headers

Fixes http://autobuild.buildroot.net/results/8a9369fece696e012249b865708ffee533c5318a

Pppd bundles some, but not all needed kernel headers. In Linux 3.5 (and
before that 2.6.35) if_pppol2tp.h was changed, so the embedded copy was no
longer compatible.

if_pppol2tp.h has been part of the exported kernel headers since 2.6.23, so
simply drop the bundled copy and use the kernel headers instead, similar
to how it is done in Gentoo:

https://bugs.gentoo.org/show_bug.cgi?id=427684

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

linux-firmware: update to a more recent git snapshot

This update pulls in new ath6k and realtek drivers and updates for
libertas and iwlwifi:

a707c25 Merge branch 'master' of git://dev.laptop.org/users/dsd/linux-firmware
adb6547 rtlwifi/rtl8723e: Add firmware for new driver
4e17394 linux-firmware: ath6k: add firmware for AR6004 hw1.3
bac3567 Marvell SD8787: update to v14.66.9.p96
6a08d14 Marvell SD8686: update to v9.70.20.p0
bbede96 iwlwifi: update firmware for 6205

Signed-off-by: Sven Neumann <s.neumann@raumfeld.com>
Signed-off-by: Peter Korsgaard <jacmet@sunsite.dk>

pkg-download: allow relative BR2_DL_DIR

And move the BUILDROOT_DL_DIR override here as well for consistency.

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

schifra: new package

[Peter: minor fixes/cleanups]
Signed-off-by: Simon Dawson <spdawson@gmail.com>
Signed-off-by: Peter Korsgaard <jacmet@sunsite.dk>

sudo: bump version

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

openssh: bump to version 6.1p1

Signed-off-by: Gustavo Zacarias <gustavo@zacarias.com.ar>
Signed-off-by: Peter Korsgaard <jacmet@sunsite.dk>

iproute2: bump to version 3.5.1

Signed-off-by: Gustavo Zacarias <gustavo@zacarias.com.ar>
Signed-off-by: Peter Korsgaard <jacmet@sunsite.dk>

bison: bump to version 2.6.2

[Peter: License is GPLv3+]
Signed-off-by: Gustavo Zacarias <gustavo@zacarias.com.ar>
Signed-off-by: Peter Korsgaard <jacmet@sunsite.dk>

grep: bump to version 2.14

[Peter: License is GPLv3+]
Signed-off-by: Gustavo Zacarias <gustavo@zacarias.com.ar>
Signed-off-by: Peter Korsgaard <jacmet@sunsite.dk>

gnutls: bump to 2.12.20

[Peter: License is GPLv3+ LGPLv2.1+]
Signed-off-by: Gustavo Zacarias <gustavo@zacarias.com.ar>
Signed-off-by: Peter Korsgaard <jacmet@sunsite.dk>

mpc: bump to version 1.0.1 and license change

Signed-off-by: Gustavo Zacarias <gustavo@zacarias.com.ar>
Signed-off-by: Peter Korsgaard <jacmet@sunsite.dk>

automake: bump to version 1.11.6

Signed-off-by: Gustavo Zacarias <gustavo@zacarias.com.ar>
Signed-off-by: Peter Korsgaard <jacmet@sunsite.dk>

less: bump to version 451

[Peter: License is GPLv3+]
Signed-off-by: Gustavo Zacarias <gustavo@zacarias.com.ar>
Signed-off-by: Peter Korsgaard <jacmet@sunsite.dk>

libnl: bump to version 3.2.13

Signed-off-by: Gustavo Zacarias <gustavo@zacarias.com.ar>
Signed-off-by: Peter Korsgaard <jacmet@sunsite.dk>

libfuse: fix build failures with iconv

Fix build failures when iconv is present for libfuse users that use
pkgconfig.
For example:
http://autobuild.buildroot.net/results/4d455831a7b467b9ea1ad54bea6b57607f5560a8/build-end.log

Signed-off-by: Gustavo Zacarias <gustavo@zacarias.com.ar>
Signed-off-by: Peter Korsgaard <jacmet@sunsite.dk>

json-c: fix build when no thread support is available

Fixes
http://autobuild.buildroot.org/results/1d9a2e44da3aa627265b03763ca324c609c0c61c/build-end.log.

Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

cjson: bug fix - use -fPIC instead of -fpic

This change is to fix the autobuild failure at:
http://autobuild.buildroot.net/results/45e2c08d5e3c868b2d7fdf26a7c5f88de5ff8f61/build-end.log
I can't actually reproduce this bug, because my machine cannot
run the 64-bit microblaze toolchain.  However, the log makes
it clear that -fPIC is needed, and grepping for fPIC and fpic
under packages makes it clear that cjson should have used -fPIC
anyway.  So even if the bug isn't fixed, it must surely be improved.

Signed-off-by: Danomi Manchego <danomimanchego123@gmail.com>
Signed-off-by: Peter Korsgaard <jacmet@sunsite.dk>

vala: move to 'Development tools' submenu

vala isn't an X application.

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

vala: bump version

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

zeromq: add license information

Note that the COPYING.LESSER file states that "Parts of the
software are licensed under the MIT (X11) license".
On the 0MQ mailing list, Pieter Hintjens has confirmed that
the code no longer contains any parts that are under the
MIT/X11 license. (The COPYING.LESSER file will be fixed in
a future release.)

Signed-off-by: Simon Dawson <simond@trainfx.com>
Signed-off-by: Peter Korsgaard <jacmet@sunsite.dk>

rp-pppoe: bump to version 3.11

Signed-off-by: Gustavo Zacarias <gustavo@zacarias.com.ar>
Signed-off-by: Peter Korsgaard <jacmet@sunsite.dk>

linux-firmware: add firmware for SDIO based Marvell Libertas modules

Signed-off-by: Sven Neumann <s.neumann@raumfeld.com>
Signed-off-by: Peter Korsgaard <jacmet@sunsite.dk>

pkg-infra: warn about use of deprecated GENTARGETS

With the replacement of GENTARGETS by package-generic, there is a risk
that local packages don't work anymore without any indication of what
is wrong.  Therefore, generate an error message if the GENTARGETS,
AUTOTARGETS or CMAKETARGETS macro is still used.

Signed-off-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>
Signed-off-by: Peter Korsgaard <jacmet@sunsite.dk>

libpcap: fix build on microblaze

On microblaze, libpcap needs to be built with -fPIC rather than -fpic as
reported by mrueg on IRC.

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

erlang: fix host version dependencies

Fixes http://autobuild.buildroot.net/results/8816fe16534062cdf0d3a682a7408f9adbea2985

The host version of erlang does not need to enable support for & depend on
host versions of ncurses/openssl/zlib, even if the target version does.

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

linux-firmware: add Realtek firmware

Signed-off-by: Yegor Yefremov <yegorslists@googlemail.com>
Signed-off-by: Peter Korsgaard <jacmet@sunsite.dk>

linux-firmware: add ti-connectivity fw v5

Signed-off-by: Yegor Yefremov <yegorslists@googlemail.com>
Signed-off-by: Peter Korsgaard <jacmet@sunsite.dk>

qemu/mips64-malta: add new sample config

Signed-off-by: Gustavo Zacarias <gustavo@zacarias.com.ar>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Signed-off-by: Peter Korsgaard <jacmet@sunsite.dk>

mips64: remove toolchain kludge

-mno-abicalls is an old kludge for some (probably) old issue.
Remove it since it's actually harmful, static busybox doesn't build with
it for a modern-ish toolchain (defaults as of this commit, uClibc
0.9.33.2 + gcc 4.5.4).

Signed-off-by: Gustavo Zacarias <gustavo@zacarias.com.ar>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Signed-off-by: Peter Korsgaard <jacmet@sunsite.dk>

Clarify MIPS ABIs support

Practically speaking, MIPS has three useful ABIs:

 * o32 is for 32-bits CPUs, or 64-bit CPUs running only a 32-bit subset
   of the instruction set.
 * n32 is for 64-bits CPUs only. It has 32-bits pointers and long
   integers.
 * n64 is for 64-bits CPUs only. It has 64-bits pointers and long
   integers.

See http://www.linux-mips.org/wiki/MIPS_ABI_History and
http://www.linux-mips.org/wiki/WhatsWrongWithO32N32N64 for more
details.

So, this commit reworks the Buildroot MIPS support by:

 * Add separate mips64/mips64el top-level architectures.

 * Renaming the n32 ABI option to BR2_MIPS_NABI32, for consistency
   with BR2_MIPS_OABI32.

 * Renaming the n64 ABI option to BR2_MIPS_NABI64, for consistency
   with BR2_MIPS_OABI32.

 * Make the n32 and n64 ABI selections select the BR2_ARCH_IS_64,
   since those ABIs are valid on 64-bits CPUs only.

 * Removing the o64 ABI, which is practicaly never used.

 * Removing the "none" ABI, which really doesn't make sense.

 * Introduce the mips64 and mips64el architecture names when a 64-bits
   MIPS ABI is choosen. This will fix build issue like
   http://autobuild.buildroot.org/results/9b8c5ea86c953a89e85e7b67e9221de41773f652/build-end.log
   where gmp was confused by the fact of having a 32 bits architecture
   (detected by the mips- architecture part of the tuple) but 64 bits
   integer size when compiling.

 * Adjust the uclibc.mk logic to support the new mips64/mips64el
   architecture names, and take into account the renaming of the ABI
   options.

This has been build tested by generating Buildroot toolchains and
compiling a few packages for MIPS o32, MIPS n32 and MIPS n64.

This work is originally based on prior work done by Gustavo Zacarias.

Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Signed-off-by: Peter Korsgaard <jacmet@sunsite.dk>

uClibc: use BR2_ENDIAN to simplify endianess selection

Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Signed-off-by: Peter Korsgaard <jacmet@sunsite.dk>

php: bump to version 5.3.16

Signed-off-by: Gustavo Zacarias <gustavo@zacarias.com.ar>
Signed-off-by: Peter Korsgaard <jacmet@sunsite.dk>

linux: bump default to kernel version 3.5.3

Signed-off-by: Gustavo Zacarias <gustavo@zacarias.com.ar>
Signed-off-by: Peter Korsgaard <jacmet@sunsite.dk>

gcc: default to 4.6.x

4.5.x is no longer maintained upstream.

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

kernel-headers: add 3.5.x, remove 2.6.36.x

And deprecate 2.6.38.x.

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

kernel-headers: bump 3.{0,2,4}.x stable versions

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

busybox: add 1.19.4 / 1.20.2 fixes

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

collectd: fix rrdtool suboption dependency

Fixes warning: (BR2_PACKAGE_COLLECTD_RRDTOOL) selects BR2_PACKAGE_RRDTOOL
which has unmet direct dependencies (BR2_USE_WCHAR).

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

e2fsprogs: fix download URL

It is /v$VERSION, not just /$VERSION.

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

usb_modeswitch_data: bump version

Previous version is no longer available upstream.

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

Merge branch 'next'

Conflicts:
package/e2fsprogs/e2fsprogs.mk
package/libfuse/libfuse.mk
package/multimedia/mpd/mpd.mk
package/smartmontools/smartmontools.mk

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

Kickoff 2012.11 cycle

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

toolchain/helpers.mk: fix check_glibc sysroot typo

Introduced by 68973cca2 (adjust logic to support Linaro 2012.05)

Reported-by: R Zhong <rzhong@hotmail.com>
Signed-off-by: Peter Korsgaard <jacmet@sunsite.dk>

Prepare for 2012.08

Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

Update CHANGES with resolved bugs in the 2012.08 cycle

Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

rrdtool: fix build with high BR2_JLEVEL settings

Fixes http://autobuild.buildroot.net/results/3dc64c2d74c54c665f16649992052a39b1665d4d

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

mtdev2tuio: new package

mtdev2tuio is a simple application for converting touch events
to the standart TUIO 1.1 protocol.

https://github.com/olivopaolo/mtdev2tuio

[thomas.petazzoni@free-electrons.com: use TARGET_CONFIGURE_OPTS, fix
indentation, add dependency on thread support in toolchain]

Signed-off-by: Stephan Hoffmann <sho@relinux.de>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

liblo: new pachage

liblo is an implementation of the Open Sound Control protocol
for POSIX systems.

http://liblo.sourceforge.net/

[thomas.petazzoni@free-electrons.com: add dependency on thread support
in toolchain]

Signed-off-by: Stephan Hoffmann <sho@relinux.de>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

mtdev: new package

The mtdev is a stand-alone library which transforms all variants
of kernel multitouch events to the slotted type B protocol.

http://bitmath.org/code/mtdev/

Signed-off-by: Stephan Hoffmann <sho@relinux.de>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

Remove BR2_SOURCEFORGE_MIRROR variable

Now that all packages have been converted to use the
downloads.sourceforge.net URLs that automatically selects an available
Sourceforge mirror, we can get rid of the BR2_SOURCEFORGE_MIRROR
configuration variable.

Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

prboom, oprofile: Add explicit <foo>_SITE statements

The default <foo>_SITE value that pkg-generic sets when the .mk file
doesn't declare it is using BR2_SOURCEFORGE_MIRROR and therefore isn't
consistent with the usage of downloads.sourceforge.net we have
generalized for all drivers.

Since the downloads.sourceforge.net URLs are inherently specific to
each package, we can no longer provide a default in pkg-generic.mk,
and therefore each package *must* declare its own <foo>_SITE value.

This patch fixes the only two packages that were lacking a <foo>_SITE
value.

Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

remove rest of the BR2_SOURCEFORGE_MIRROR references

Signed-off-by: Stefan Fröberg <stefan.froberg@petroprogram.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

sqlite: add license info

Signed-off-by: Danomi Manchego <danomimanchego123@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

luajit: bug fix - prevent target CFLAGS from being used in host tool compile.

The luajit amalgamation compile starts with building a host tool, and then
uses it to build itself.  However, when CFLAGS is specified, as opposed to
TARGET_CFLAGS, then it is used for both HOST and TARGET builds.  So if you
add something target specific into 'Target Optimizations' (for example,
'-mfpu=neon -mfloat-abi=softfp' for ARM Cortex-A8), then it gets into the
host tool compile, which then fails (because my build machine is not an
ARM Cortext-A8).  This can be fixed by using TARGET_CFLAGS and
TARGET_LDFLAGS, instead of CFLAGS and LDFLAGS, respectively.

Signed-off-by: Danomi Manchego <danomimanchego123@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

cups: uses fork(), only available on MMU platforms

Fixes
http://autobuild.buildroot.net/results/6db1eb27387696bfa6b4d8ffb30eff00a2f842d0/build-end.log.

Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

microperl: uses fork(), only available on MMU platforms

Fixes
http://autobuild.buildroot.net/results/0d5e272b594bc6cc84df291134e2b23d33b97d17/build-end.log.

Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>