bind: bump to version 9.9.5-p1

Fixes runtime issues when built with gcc 4.9

Signed-off-by: Gustavo Zacarias <gustavo@zacarias.com.ar>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

atftp: new package

[Peter: note that readline is optional, drop trailing Config.in line]
Signed-off-by: Ryan Barnett <ryan.barnett@rockwellcollins.com>
CC: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

radvd: bump to version 1.13

Signed-off-by: Gustavo Zacarias <gustavo@zacarias.com.ar>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

packages: fix the header package name

According to the documentation:

"Header: The file starts with a header. It contains the module name,
preferably in lowercase, enclosed between separators made of 80 hashes."

This patch makes the appropriate changes.

Signed-off-by: Jerzy Grzegorek <jerzy.grzegorek@trzebnica.net>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

network-manager: bump version

Patch is now upstream.

Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

network-manager: needs kernel headers v3.7

NetworkManager uses struct ifla_vxlan_port_range that was introduced in kernel
version 3.7.

Fixes:
http://autobuild.buildroot.net/results/6f2/6f214e3206ad0b12c633a7632c1e32c369e447ff/

Signed-off-by: Baruch Siach <baruch@tkos.co.il>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

package/libeXosip2: fix build for static linking

Add a new patch to use pkg-config to detect openssl.

[Peter: fix minor typos in description]
Signed-off-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

perl-mojolicious: bump to version 5.08

- switch to BackPan in order to prevent build breakage
(like http://autobuild.buildroot.net/results/358/358f531f2db90b9bc3b1e4e2158c68d2bf6587fc/)
- add license file

Signed-off-by: Francois Perrad <francois.perrad@gadz.org>
Tested-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Acked-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

samba: security bump to version 3.6.24

Fixes:
CVE-2014-0244 (Denial of service - CPU loop)
CVE-2014-3493 (Denial of service - Server crash/memory corruption)

Signed-off-by: Gustavo Zacarias <gustavo@zacarias.com.ar>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

samba4: security bump to version 4.1.9

Fixes:
CVE-2014-0244 (Denial of service - CPU loop)
CVE-2014-3493 (Denial of service - Server crash/memory corruption)

Signed-off-by: Gustavo Zacarias <gustavo@zacarias.com.ar>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

orc: bump version

From the release notes:

  - Add libtool versioning to the linker flags again. This was accidentially
    removed in 0.4.20 but should not cause any problems on platforms other
    than OS X (Sebastian Dröge)

Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

avahi: fixup init scripts after default user removal

Commit d3ccfa362b (avahi: run as avahi user/group instead of default)
changed avahi-autoipd to run as the avahi user, but forgot to update the
init script/systemd config to match.

Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

orc: bump version

Tarball no longer available in .gz format. From the release notes:

  - Fix list corruption when splitting code memory chunks, causing crashes
    when allocating a lot of code memory and trying to free it later
    (Tim-Philipp Müller)
  - Add some extra checks for the number of variables used in ORC code to
    prevent overflows and crashes in the compiler (Vincent Penquerc'h)
  - Various compiler warnings, coverity warnings and static code analysis
    fixes (Sebastian Dröge)

Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

wpa_supplicant: disable for avr32

Fixes:
http://autobuild.buildroot.net/results/209/20908f479b33c1e2952622f5e8ad6b60d58af693/

Signed-off-by: Baruch Siach <baruch@tkos.co.il>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

board/avnet/zedboard/: improve readme.txt explanations

[Thomas: fix small grammar issue noticed by Spenser Gilliland, adjust
commit title.]

Signed-off-by: Javier Domingo Cansino <javierdo1@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

legal-info: rename legal-warning-pkg-savednothing helper

This helper was called when none of the sources or license
files were saved.

Now we handle license files separately from the sources,
this is no longer the case: they are only called when the
sources are not saved.

Rename the handler and change the warning message accordingly.

Signed-off-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Cc: Luca Ceresoli <luca@lucaceresoli.net>
Cc: Thomas De Schampheleire <patrickdepinguin@gmail.com>
Cc: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Cc: Fabio Porcedda <fabio.porcedda@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

legal-info: add a comment about what packages we save the tarballs of

As the legal-info infra only (rightfully) saves the tarballs of packages
that:
  - we want to redistribute,
  - and are not local,
  - and are not overriden,

add a comment stating so.

This should clarify the code-block, which although trivial to read,
was not easy to interpret without thinking thouroughly about it.

Signed-off-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Cc: Luca Ceresoli <luca@lucaceresoli.net>
Cc: Thomas De Schampheleire <patrickdepinguin@gmail.com>
Cc: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Cc: Fabio Porcedda <fabio.porcedda@gmail.com>
Reviewed-by: Luca Ceresoli <luca@lucaceresoli.net>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

legal-info: save license files even for local or overridden packages

Even if we do not save the sources for local or overridden packages because
it is too complex, we can still quite easily save the license files.

Also, having the license files is a very important part of complying with
the licenses.

Move the copy of license files out of the non-local, non-overridden package
case, but still in the case where packages have a _SOURCE defined, to
avoid catching packages bundled in Buildroot (eg. mkpasswd et al.)

Reported-by: Luca Ceresoli <luca@lucaceresoli.net>
Signed-off-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Cc: Luca Ceresoli <luca@lucaceresoli.net>
Cc: Thomas De Schampheleire <patrickdepinguin@gmail.com>
Cc: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Cc: Fabio Porcedda <fabio.porcedda@gmail.com>
Reviewed-by: Luca Ceresoli <luca@lucaceresoli.net>
Tested-by: Luca Ceresoli <luca@lucaceresoli.net>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

legal-info: extract even no-redistribute packages

Currently, if a package is marked _REDISTRIBUTE = NO, then legal-info
will not try to extract it first.

If that package also declares some _LICENSE_FILES, legal-info fails
if it is the only action we're trying to run:

    $ cat defconfig
    BR2_arm=y
    BR2_TOOLCHAIN_BUILDROOT_EGLIBC=y
    BR2_PACKAGE_LIBFSLCODEC=y
    $ make BR2_DEFCONFIG=$(pwd)/defconfig defconfig
    $ make legal-info
    [--SNIP--]
    cat: /home/ymorin/dev/buildroot/O/build/libfslcodec-3.5.7-1.0.0/EULA: No such file or directory

Fix this by always having legal-info extract the archives if one or
more _LICENSE_FILES are specified.

We do this for all types of packages: overridden, local or 'normal'
remote packages. Even though we do not save the sources for the
overridden or local packages, we need to save their licensing info,
so we need to extract them.

This implies that we now need only PKG-source, not PKG-extract anymore,
as a dependency of legal-info for packages we want to save (ie.
redistributable, non-local and non-overriden packages.)

Signed-off-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Cc: Luca Ceresoli <luca@lucaceresoli.net>
Cc: Thomas De Schampheleire <patrickdepinguin@gmail.com>
Cc: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Cc: Fabio Porcedda <fabio.porcedda@gmail.com>
Reviewed-by: Luca Ceresoli <luca@lucaceresoli.net>
Tested-by: Luca Ceresoli <luca@lucaceresoli.net>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

php-geoip: new package

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

knock: fix static build

Use pcap-config to determine link dependencies of libpcap.

Fixes:
http://autobuild.buildroot.net/results/1eb/1eb877ebd9469c9261bf114fc7733819ae3fd562/

Signed-off-by: Baruch Siach <baruch@tkos.co.il>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

geoip: new package

[Thomas: slightly reformat the Config.in help text. Add the 'LICENSE'
file to GEOIP_LICENSE_FILES.]

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

perl-mojolicious: new package

[Thomas: as noted by Arnout, remove trailing whitespace, and fix the
license to Artistic-2.0. Also, adjust the indentation in
package/Config.in to the new standard.]

Signed-off-by: Francois Perrad <francois.perrad@gadz.org>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

support/script/scancpan: populates Config.in with homepage

Signed-off-by: Francois Perrad <francois.perrad@gadz.org>
Reviewed-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

Makefile: test if "dot" exists in "<pkg>-graph-depends"

To be able to check the "dot" command availability in
"<pkg>-graph-depends" move the check to the "graph-depends-requirements" rule.
Also don't use a subshell for the exit command to be sure that the error
will be returned by the shell.

Signed-off-by: Fabio Porcedda <fabio.porcedda@gmail.com>
Cc: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Cc: Yann E. MORIN <yann.morin.1998@free.fr>
Tested-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Acked-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

manual: (faq-boot-hang-after-starting) fix config entries

Signed-off-by: Vivien Didelot <vivien.didelot@savoirfairelinux.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

gst-ffmpeg: fix build on armv6 without thumb2 (e.g. RaspberryPi)

[Thomas: rename existing patch and the one added by Arnout to follow
the patch naming convention.]

Cc: Marco Trapanese <marcotrapanese@gmail.com>
Signed-off-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>
Acked-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

manual: add requirements for graph generation

Add "graphviz" and "python-matplotlib" as requirements for graph
generation.

Signed-off-by: Fabio Porcedda <fabio.porcedda@gmail.com>
Reported-by: Dallas Clement <dallas.a.clement@gmail.com>
Cc: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Cc: Yann E. MORIN <yann.morin.1998@free.fr>
Acked-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

graph-depends.py: support python3

This patch is the result of 2to3.

In addition, universal_newlines=True is added to the Popen calls. In
python3, this makes sure that the output is decoded so that we get a
string instead of a buffer object.

Cc: Vivien Didelot <vivien.didelot@savoirfairelinux.com>
Signed-off-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>
Reviewed-by: Vivien Didelot <vivien.didelot@savoirfairelinux.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

faifa: needs threads

Fixes:
http://autobuild.buildroot.net/results/a93/a93395c656169e0f08f471f12bc05f29c9ea2976/

Signed-off-by: Gustavo Zacarias <gustavo@zacarias.com.ar>
Acked-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

libhid: fix for nommu

Make it accept uclinux in the tuple as linux. Fixes:
http://autobuild.buildroot.net/results/07f/07f2a560d9915ff7bad830be11f95aa856ce0e73/

Upstream seems dead with the last commit in svn being 5+ years ago and
with some recent patches in the mailing list just sitting there.

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

dhcpcd: fix build with ccache

After the bump to dhcpcd 6.4.0, building dhcpcd with ccache fails at the
configure step:

Using compiler .. <buildroot>/output/host/usr/bin/ccache <buildroot>/output/host/usr/bin/i686-pc-linux-gnu-gcc
<buildroot>/output/host/usr/bin/ccache <buildroot>/output/host/usr/bin/i686-pc-linux-gnu-gcc is not an executable
make: *** [<buildroot>/output/build/dhcpcd-6.4.0/.stamp_configured] Error 1

This patch backports an upstream patch to fix this issue.

Signed-off-by: Thomas De Schampheleire <thomas.de.schampheleire@gmail.com>
Acked-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

imagemagick: bump to version 6.8.9-3

Signed-off-by: Jerzy Grzegorek <jerzy.grzegorek@trzebnica.net>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

torsmo: needs mmu

Fixes:
http://autobuild.buildroot.net/results/28f/28ffabf47643dfdb646ec98a56978db75fc84c05/

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

strace: fix ptrace {PEEK, POKE}USER arguments decoding on xtensa

Backported from: 57fac759ca0945f5608ed6a2228c697f8fb67cd6
Signed-off-by: Max Filippov <jcmvbkbc@gmail.com>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

strace: add serial number to existing patch file names

Signed-off-by: Max Filippov <jcmvbkbc@gmail.com>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

valgrind: fix build for glibc 2.19

Fix build for glibc 2.19-based toolchains, fixes:
http://autobuild.buildroot.net/results/be4/be4c033e2088c324f93fc70230f05d761108cfba/

Signed-off-by: Gustavo Zacarias <gustavo@zacarias.com.ar>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

kmod: bump version

Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

tcpreplay: use pcap-config

Since other libs can be pulled in by libpcap switch to using
pcap-config, it's the correct way and simpler. Fixes:
http://autobuild.buildroot.net/results/c02/c023bacd35e9ef1addeb881cd3d572f97f9b4b56/

Signed-off-by: Gustavo Zacarias <gustavo@zacarias.com.ar>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

pax-utils: needs largefile support

Fixes:
http://autobuild.buildroot.net/results/2f8/2f833cc285eac99daba96949b9601f4e64b6f6be/

Signed-off-by: Gustavo Zacarias <gustavo@zacarias.com.ar>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

lftp: bump to version 4.5.2

Signed-off-by: Gustavo Zacarias <gustavo@zacarias.com.ar>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

libnss: security bump to version 3.16.1

Fixes:

CVE-2014-1492 - The cert_TestHostName function in lib/certdb/certdb.c in
the certificate-checking implementation in Mozilla Network Security
Services (NSS) before 3.16 accepts a wildcard character that is embedded
in an internationalized domain name's U-label, which might allow
man-in-the-middle attackers to spoof SSL servers via a crafted
certificate.

CVE-2014-1491 - Mozilla Network Security Services (NSS) before 3.15.4,
as used in Mozilla Firefox before 27.0, Firefox ESR 24.x before 24.3,
Thunderbird before 24.3, SeaMonkey before 2.24, and other products, does
not properly restrict public values in Diffie-Hellman key exchanges,
which makes it easier for remote attackers to bypass cryptographic
protection mechanisms in ticket handling by leveraging use of a certain
value.

CVE-2014-1490 - Race condition in libssl in Mozilla Network Security
Services (NSS) before 3.15.4, as used in Mozilla Firefox before 27.0,
Firefox ESR 24.x before 24.3, Thunderbird before 24.3, SeaMonkey before
2.24, and other products, allows remote attackers to cause a denial of
service (use-after-free) or possibly have unspecified other impact via
vectors involving a resumption handshake that triggers incorrect
replacement of a session ticket.

CVE-2013-1740 - The ssl_Do1stHandshake function in sslsecur.c in libssl
in Mozilla Network Security Services (NSS) before 3.15.4, when the TLS
False Start feature is enabled, allows man-in-the-middle attackers to
spoof SSL servers by using an arbitrary X.509 certificate during certain
handshake traffic.

Signed-off-by: Gustavo Zacarias <gustavo@zacarias.com.ar>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

libnspr: security bump to version 4.10.6

Fixes CVE-2014-1545 - Mozilla Netscape Portable Runtime (NSPR) before
4.10.6 allows remote attackers to execute arbitrary code or cause a
denial of service (out-of-bounds write) via vectors involving the
sprintf and console functions.

Signed-off-by: Gustavo Zacarias <gustavo@zacarias.com.ar>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

tcpreplay: bump to version 4.0.4

Signed-off-by: Gustavo Zacarias <gustavo@zacarias.com.ar>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

libpcap: make dbus support deterministic

Signed-off-by: Gustavo Zacarias <gustavo@zacarias.com.ar>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

dbus: don't forcibly disable static builds

It's there for some historical reason and breaks libpcap with dbus
support for static linkage scenarios (like the one used by tcpreplay).

Signed-off-by: Gustavo Zacarias <gustavo@zacarias.com.ar>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

genimage: bump to version 6

Signed-off-by: Vivien Didelot <vivien.didelot@savoirfairelinux.com>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

configs: add lego_ev3_defconfig

Add initial support for the Lego Mindstorms EV3 programmable brick.

The Lego Mindstorms EV3 brick comprises a Texas Instruments AM1808 SoC, with
an ARM 926EJ-S main processor running at 300 MHz.
https://en.wikipedia.org/wiki/Lego_Mindstorms_EV3

This configuration uses the Linux kernel of the ev3dev project.
https://github.com/mindboards/ev3sources

More info is available in the board/lego/ev3/readme.txt file, shamelessly
documented in the same way as the SoCkit folks did.

[Peter: lock kernel headers to match]
Signed-off-by: Vincent Stehlé <vincent.stehle@laposte.net>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

argus: bump to vanilla version 3.6.0.1

All of the new toolchain requirements (ipv6, threads, rpc) aren't
actually new - this package failed to build for ages, it just wasn't
picked up by the autobuilders because the main Makefile just doesn't
care about bailing out properly.

Signed-off-by: Gustavo Zacarias <gustavo@zacarias.com.ar>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

stunnel: bump to version 5.02

Signed-off-by: Gustavo Zacarias <gustavo@zacarias.com.ar>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

libsoc: bump to 0.6.2

Signed-off-by: Yegor Yefremov <yegorslists@googlemail.com>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

linux: bump default to version 3.15.1

Signed-off-by: Gustavo Zacarias <gustavo@zacarias.com.ar>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

linux-headers: bump 3.{4, 10, 14, 15}.x series

Signed-off-by: Gustavo Zacarias <gustavo@zacarias.com.ar>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

freerdp: bump to version 1.0.2

And switch to github helper.

Signed-off-by: Gustavo Zacarias <gustavo@zacarias.com.ar>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

msgpack: disable for ARC

msgpack uses the __sync_fetch_and_add() atomic primitive that is not
available for ARC. Fixes:
http://autobuild.buildroot.net/results/829/8292e67354602738cae12d8ebea83b513ba35482/

Signed-off-by: Gustavo Zacarias <gustavo@zacarias.com.ar>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

netatalk: add support for ACLs

Signed-off-by: Gustavo Zacarias <gustavo@zacarias.com.ar>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

netatalk: add support for zeroconf

[Peter: check for avahi-daemon, not just the base avahi package]
Signed-off-by: Gustavo Zacarias <gustavo@zacarias.com.ar>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

netatalk: bump to version 3.1.2

Signed-off-by: Gustavo Zacarias <gustavo@zacarias.com.ar>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

knock: bump to version 0.7

Bump knock to version 0.7 and switch away from a git snapshot, it's
nicer and avoids the need to autoreconf.

Signed-off-by: Gustavo Zacarias <gustavo@zacarias.com.ar>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

dbus: bump to version 1.8.4

Upstream removed support for libxml2 as xml backend, select expat
unconditionally.

Signed-off-by: Baruch Siach <baruch@tkos.co.il>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

can-utils: switch homepage now that berlios is gone

Unfortunately the gitorious page/wiki says nothing useful so use the
elinux.org page for now.

Signed-off-by: Gustavo Zacarias <gustavo@zacarias.com.ar>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

ngircd: switch site now that berlios is gone

Signed-off-by: Gustavo Zacarias <gustavo@zacarias.com.ar>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

openocd: switch homepage now that berlios is gone

Signed-off-by: Gustavo Zacarias <gustavo@zacarias.com.ar>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

tstools: switch homepage now that berlios is gone

Signed-off-by: Gustavo Zacarias <gustavo@zacarias.com.ar>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

tslib: switch homepage now that berlios is gone

Signed-off-by: Gustavo Zacarias <gustavo@zacarias.com.ar>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

gpsd: switch homepage now that berlios is gone

Signed-off-by: Gustavo Zacarias <gustavo@zacarias.com.ar>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

wireless-regdb: bump to version 2014.06.13

Signed-off-by: Gustavo Zacarias <gustavo@zacarias.com.ar>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

libusb: bump to version 1.0.19

Signed-off-by: Gustavo Zacarias <gustavo@zacarias.com.ar>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

kbd: bump to version 1.15.5

Signed-off-by: Gustavo Zacarias <gustavo@zacarias.com.ar>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

sysstat: bump to version 11.0.0

Signed-off-by: Gustavo Zacarias <gustavo@zacarias.com.ar>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

psmisc: bump to version 22.21

Signed-off-by: Gustavo Zacarias <gustavo@zacarias.com.ar>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

vim: bump to version 7.4-333

Signed-off-by: Gustavo Zacarias <gustavo@zacarias.com.ar>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

blackfin: 2012R1 toolchain is gone, remove kludges

Remove ADI toolchain 2012R1 package kludges since that version is gone
since the last bump.

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

dhcpcd: bump to version 6.4.0

Now with support for static builds.

Signed-off-by: Gustavo Zacarias <gustavo@zacarias.com.ar>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

libpng: bump to version 1.6.12

Signed-off-by: Gustavo Zacarias <gustavo@zacarias.com.ar>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

file: bump to version 5.19

Signed-off-by: Gustavo Zacarias <gustavo@zacarias.com.ar>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

btrfs-progs: bump to version 3.14.2

Signed-off-by: Gustavo Zacarias <gustavo@zacarias.com.ar>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

opus-tools: bump to version 0.1.9

Signed-off-by: Gustavo Zacarias <gustavo@zacarias.com.ar>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

mpg123: bump to version 1.20.1

Signed-off-by: Gustavo Zacarias <gustavo@zacarias.com.ar>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

pax-utils: bump to version 0.8.1

Signed-off-by: Gustavo Zacarias <gustavo@zacarias.com.ar>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

vlc: security bump to version 2.1.4

Fixes CVE-2014-1684: The ASF_ReadObject_file_properties function in
modules/demux/asf/libasf.c in the ASF Demuxer in VideoLAN VLC Media
Player before 2.1.3 allows remote attackers to cause a denial of service
(divide-by-zero error and crash) via a zero minimum and maximum data
packet size in an ASF file.

Signed-off-by: Gustavo Zacarias <gustavo@zacarias.com.ar>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

olsr: bump to version 0.6.6.2

Also make the menu entry less melodramatic.

Signed-off-by: Gustavo Zacarias <gustavo@zacarias.com.ar>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

links: add security fix for CVE-2013-6050

Also fix LICENSE typo.

Signed-off-by: Gustavo Zacarias <gustavo@zacarias.com.ar>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

svgtiny: create BUILDDIR before $(BUILDDIR)/src_colors.c

Fixes:
  http://autobuild.buildroot.net/results/5fe/5feca450d70a8de48fa2463e55c9e84db5d755b0/

[Peter: renumber existing patch and this]
Signed-off-by: Max Filippov <jcmvbkbc@gmail.com>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

uclibc: prevent rebuilding after installation to staging

Currently we configure uClibc to use kernel headers from "staging" folder with
KERNEL_HEADERS="$(STAGING_DIR)/usr/include". This path is added to include
search path of uClibc build system in Rules.mak "CFLAGS += -I$(KERNEL_HEADERS)".

At the same time on uClibc installation to "staging" we point to the same
location "$(STAGING_DIR)/usr" (headers effectively go in "usr/include").

So after every installation to "staging" dependences get touched (even though we
copy the same headers every time) and so we may see lots of sources in uClibc
get rebuilt.

This has 2 consequences:
1. Longer build time - becase even on ordinary buildroot build uClibc is built
twice. On "uclibc building" and on "uclibc installation to target".

2. Symbols in libuClibc built initially (that is later installed in
"staging/sysroot") are situated with different offset compared to second build
(later copied in "target"). This happens because as described above only part
of sources get rebuilt and then on final linkage object files are linked in
different order.

And (2) leads to problems on remote rebugging: gdbserver reports offsets that
correspond to pointless assembly in libuClibc on host.

Here's how it looks like.

Before this patch:
$ cd ~/br2_output/i586/target/lib
$ i586-buildroot-linux-uclibc-readelf -s libuClibc-0.9.33.2.so | grep kill
   423: 0000c42c    54 FUNC    GLOBAL DEFAULT    7 kill

$ cd ~/br2_output/i586/staging/lib
$ i586-buildroot-linux-uclibc-readelf -s libuClibc-0.9.33.2.so | grep kill
   423: 0000b518    54 FUNC    GLOBAL DEFAULT    7 kill

After this patch:
$ cd ~/br2_output/i586/target/lib
$ i586-buildroot-linux-uclibc-readelf -s libuClibc-0.9.33.2.so | grep kill
   423: 0000b518    54 FUNC    GLOBAL DEFAULT    7 kill

$ cd ~/br2_output/i586/staging/lib
$ i586-buildroot-linux-uclibc-readelf -s libuClibc-0.9.33.2.so | grep kill
   423: 0000b518    54 FUNC    GLOBAL DEFAULT    7 kill

Signed-off-by: Alexey Brodkin <abrodkin@synopsys.com>
Cc: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Cc: Peter Korsgaard <peter@korsgaard.com>
Cc: Noam Camus <noamc@ezchip.com>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

fltk: drop custom install options

The fltk buildsystem no longer tries to strip binaries during installation,
so these can be dropped.

Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

fmlib: only show comment if we're on a suitable arch

Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

fmc: fmlib needs a linux kernel, so propagate dependency

At the same time fixup the comments to only be visible if we are
on a suitable arch.

Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

ccache: provide capability to do initial ccache setup

For example, if your project is known to require more space than the
default max cache size, then you might want to increase the cache size
to a suitable amount using the -M (--max-size) option.

The string you specify here is passed verbatim to ccache.  Refer to
ccache documentation for more details.

These initial settings are applied after ccache has been compiled.

Signed-off-by: Danomi Manchego <danomimanchego123@gmail.com>
Reviewed-by: Thomas De Schampheleire <thomas.de.schampheleire@gmail.com>
Tested-by: Thomas De Schampheleire <thomas.de.schampheleire@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

ccache: change default cache directory path to match config setting

Commit 433290761fceb476b095548eec10adf72405e050 changed the hard-coded
ccache directory location to use BR_CACHE_DIR (then BUILDROOT_CACHE_DIR),
which is exported by Makefile based on the BR2_CCACHE_DIR config option.
This allowed the cache location to be changed on-the-fly by setting a
"make" command line variable, but left the default location of ccache's
normal default at "$HOME/.ccache".  Since this location does not match the
default for BR2_CCACHE_DIR, it is basically almost never correct, so
direct invocation of ccache outside of the buildroot Makefile, such as for
increasing the cache size, becomes cumbersome.

This patch changes the last-ditch cache location from "$HOME/.ccache" to
the BR_CCACHE_DIR value defined when host-ccache is configured.  Note that
the ability to later override the cache location by using a BR_CACHE_DIR
command line variable is left intact.

Signed-off-by: Danomi Manchego <danomimanchego123@gmail.com>
Reviewed-by: Thomas De Schampheleire <thomas.de.schampheleire@gmail.com>
Tested-by: Thomas De Schampheleire <thomas.de.schampheleire@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

classpath: Use generic check for host program

Remove the specific check that was done in dependencies.sh to use the
generic one that were introduced by the previous patch.

Also, introduce, BR2_NEEDS_HOST_JAVAC and BR2_NEEDS_HOST_JAR as it is
needed by classpath.

Signed-off-by: Maxime Hadjinlian <maxime.hadjinlian@gmail.com>
Reviewed-by: Thomas De Schampheleire <thomas.de.schampheleire@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

infra: Add generic check_prog_host function

Avoid copy/pasting the same block of code to check if a program is
available on the host machine.

Signed-off-by: Maxime Hadjinlian <maxime.hadjinlian@gmail.com>
Reviewed-by: Thomas De Schampheleire <thomas.de.schampheleire@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

xserver_xorg-server: Fix race condition when installing man pages

Reproduced the bug using BR2_JLEVEL=16

Fixes race condition by disabling parallel make in man/

http://autobuild.buildroot.net/results/11e/11ece737078ca8da2fee4249432148675b4e630e/
http://autobuild.buildroot.net/results/a95/a952932b832ee3296d6c225c7a9df408ca3eef9f/

Signed-off-by: Bernd Kuhls <bernd.kuhls@t-online.de>
Reviewed-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

mkpasswd: change version from 'undefined' to 'buildroot-$(BR2_VERSION)'

The sources of the mkpasswd package are shipped with Buildroot, rather than
downloaded from an external location. As a result, no explicit version is
defined, causing build messages and build directory to show 'undefined' as
version.

This patch sets the version for mkpasswd to 'buildroot-$(BR2_VERSION), which
would for example expand to 'buildroot-2014.05'.

Signed-off-by: Thomas De Schampheleire <thomas.de.schampheleire@gmail.com>
Acked-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

makedevs: change version from 'undefined' to 'buildroot-$(BR2_VERSION)'

The sources of the makedevs package are shipped with Buildroot, rather than
downloaded from an external location. As a result, no explicit version is
defined, causing build messages and build directory to show 'undefined' as
version.

This patch sets the version for makedevs to 'buildroot-$(BR2_VERSION), which
would for example expand to 'buildroot-2014.05'.

Signed-off-by: Thomas De Schampheleire <thomas.de.schampheleire@gmail.com>
Acked-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

toolchain/toolchain-buildroot: migrate to virtual package infrastructure

This patch migrates the toolchain and toolchain-buildroot packages to the
virtual package infrastructure, causing the log messages to change from:

>>> toolchain undefined Downloading
>>> toolchain undefined Extracting
...

to

>>> toolchain virtual Downloading
>>> toolchain virtual Extracting
...

and similar for 'toolchain-buildroot', simply because it looks nicer.

At the same time, the directory names also become toolchain-virtual,
toolchain-buildroot-virtual instead of the corresponding 'undefined'
variants.

Signed-off-by: Thomas De Schampheleire <thomas.de.schampheleire@gmail.com>
Acked-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Acked-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

pkg-virtual: simplify definition of FOO_VERSION to 'virtual'

As mentioned in the e-mail accompanying the introduction of the pkg-virtual
infrastructure [1], the definition of FOO_VERSION is 'strange'.

After the cleanup of single/double dollar signs in inner-generic-package,
the special construction in pkg-virtual is no longer needed and can be
simplified.

Signed-off-by: Thomas De Schampheleire <thomas.de.schampheleire@gmail.com>
Acked-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
[1] http://lists.busybox.net/pipermail/buildroot/2014-April/093670.html
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

infra: add comment describing single/double dollar-sign rules

As the rules with respect to variable and function references and the need
for single or double dollar signs are not trivial, add a comment in
pkg-generic.mk describing them.

Signed-off-by: Thomas De Schampheleire <thomas.de.schampheleire@gmail.com>
Acked-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Reviewed-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

infra: consistently use double dollar signs inside inner-xxx-targets

The inner-xxx-targets in the buildroot package infrastructures are
evaluated using $(eval) which causes variable references to be a bit
different than in regular make code. As we want most references to be
expanded only at the time of the $(eval) we should not use standard
references $(VAR) but rather use double dollar signs $$(VAR). This includes
function references like $(call), $(subst), etc. The only exception is the
reference to pkgdir/pkgname and numbered variables, which are parameters to
the inner block: $(1), $(2), etc.

This patch introduces consistent usage of double-dollar signs throughout the
different inner-xxx-targets blocks.

In some cases, this would potentially cause circular references, in
particular when the value of HOST_FOO_VAR would be obtained from the
corresponding FOO_VAR if HOST_FOO_VAR is not defined. In these cases, a test
is added to check for a host package (the only case where such constructions
are relevant; these are not circular).

Benefits of these changes are:
- behavior of variables is now again as expected. For example, setting
  $(2)_VERSION = virtual in pkg-virtual.mk will effectively work, while
  originally it would cause very odd results.

- The output of 'make printvars' is now much more useful. This target shows
  the value of all variables, and the expression that led to that value.
  However, if the expression was coming from an inner-xxx-targets block, and
  was using single dollar signs, it would show in printvars as
    VAR = value (value)
  while if double dollar signs are used, it would effectively look like
    VAR = value (actual expression)
  as is intended.
  This improvement is for example effective for FOO_DL_VERSION, FOO_RAWNAME,
  FOO_SITE_METHOD and FOO_MAKE.

The correctness of this patch has been verified using 'make printvars',
'make manual' and 'make legal-info' before and after applying this patch,
and comparing the output.

Insight-provided-by: Arnout Vandecappelle <arnout@mind.be>
Signed-off-by: Thomas De Schampheleire <thomas.de.schampheleire@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

fmc: fix package/Config.in

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