[Buildroot] [PATCH 1/8] connman: bump to version 1.25

[Buildroot] [PATCH 1/8] connman: bump to version 1.25

[Jérôme Pouiller]

Signed-off-by: J?r?me Pouiller <jezz@sysmic.org>
---
 package/connman/connman.mk | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/package/connman/connman.mk b/package/connman/connman.mk
index 9b5d15f..e9ce1e6 100644
--- a/package/connman/connman.mk
+++ b/package/connman/connman.mk
@@ -4,7 +4,7 @@
 #
 ################################################################################
 
-CONNMAN_VERSION = 1.23
+CONNMAN_VERSION = 1.25
 CONNMAN_SOURCE = connman-$(CONNMAN_VERSION).tar.xz
 CONNMAN_SITE = $(BR2_KERNEL_MIRROR)/linux/network/connman
 CONNMAN_DEPENDENCIES = libglib2 dbus iptables
-- 
1.9.1

[Buildroot] [PATCH 2/8] simicsfs: bump to version 1.18

[Jérôme Pouiller]

Fix support for kernel > 3.15

Signed-off-by: J?r?me Pouiller <jezz@sysmic.org>
---
 package/simicsfs/simicsfs.mk | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/package/simicsfs/simicsfs.mk b/package/simicsfs/simicsfs.mk
index 3acd265..99dda84 100644
--- a/package/simicsfs/simicsfs.mk
+++ b/package/simicsfs/simicsfs.mk
@@ -4,7 +4,7 @@
 #
 ###############################################################################
 
-SIMICSFS_VERSION = 1.17
+SIMICSFS_VERSION = 1.18
 SIMICSFS_SITE = http://download.simics.net/pub
 SIMICSFS_LICENSE = GPLv2+
 SIMICSFS_LICENSE_FILES = hostfs.h
-- 
1.9.1

[Buildroot] [PATCH 3/8] valgrind: bump to version 3.10

[Jérôme Pouiller]

Patches status:

  valgrind-0001-workaround-SIGSEGV-on-PPC: Don't know. In doubt, I
     prefer to keep it.

  valgrind-0002-don-t-enable-largefile-support-unconditionally-on-uC:
     Seems still necessary

  valgrind-0003-Add-replacement-for-a.out.h: Upstreamed

  valgrind-0004-remove-default-mips-flags: Upstreamed

  valgrind-0005-glibc-2.19: Upstream now support glibc up to 2.20

Signed-off-by: J?r?me Pouiller <jezz@sysmic.org>
---
 .../valgrind-0001-workaround-SIGSEGV-on-PPC.patch  |  8 +-
 ...e-largefile-support-unconditionally-on-uC.patch | 14 ++--
 ...valgrind-0003-Add-replacement-for-a.out.h.patch | 58 --------------
 .../valgrind-0004-remove-default-mips-flags.patch  | 93 ----------------------
 package/valgrind/valgrind-0005-glibc-2.19.patch    | 32 --------
 package/valgrind/valgrind.mk                       |  2 +-
 6 files changed, 12 insertions(+), 195 deletions(-)
 delete mode 100644 package/valgrind/valgrind-0003-Add-replacement-for-a.out.h.patch
 delete mode 100644 package/valgrind/valgrind-0004-remove-default-mips-flags.patch
 delete mode 100644 package/valgrind/valgrind-0005-glibc-2.19.patch

diff --git a/package/valgrind/valgrind-0001-workaround-SIGSEGV-on-PPC.patch b/package/valgrind/valgrind-0001-workaround-SIGSEGV-on-PPC.patch
index 9e891be..1c5966b 100644
--- a/package/valgrind/valgrind-0001-workaround-SIGSEGV-on-PPC.patch
+++ b/package/valgrind/valgrind-0001-workaround-SIGSEGV-on-PPC.patch
@@ -1,7 +1,7 @@
-From 872ade34d6af973ca88d5accd9e819216ecb94ba Mon Sep 17 00:00:00 2001
+From bd9d2af89e45081132c150e5e4bdf9a12dfca693 Mon Sep 17 00:00:00 2001
 From: Pierre Habouzit <madcoder@debian.org>
 Date: Tue, 10 May 2011 23:11:45 +0200
-Subject: [PATCH 1/3] workaround SIGSEGV on PPC.
+Subject: [PATCH 1/2] workaround SIGSEGV on PPC.
 MIME-Version: 1.0
 Content-Type: text/plain; charset=UTF-8
 Content-Transfer-Encoding: 8bit
@@ -13,7 +13,7 @@ Signed-off-by: J?r?me Pouiller <jezz@sysmic.org>
  1 file changed, 4 insertions(+)
 
 diff --git a/coregrind/m_machine.c b/coregrind/m_machine.c
-index 28a180e..8426551 100644
+index 42d8ce8..3c9900f 100644
 --- a/coregrind/m_machine.c
 +++ b/coregrind/m_machine.c
 @@ -27,6 +27,10 @@
@@ -28,5 +28,5 @@ index 28a180e..8426551 100644
  #include "pub_core_vki.h"
  #include "pub_core_libcsetjmp.h"   // setjmp facilities
 -- 
-1.8.3.2
+1.9.1
 
diff --git a/package/valgrind/valgrind-0002-don-t-enable-largefile-support-unconditionally-on-uC.patch b/package/valgrind/valgrind-0002-don-t-enable-largefile-support-unconditionally-on-uC.patch
index 1df318a..57ab2c3 100644
--- a/package/valgrind/valgrind-0002-don-t-enable-largefile-support-unconditionally-on-uC.patch
+++ b/package/valgrind/valgrind-0002-don-t-enable-largefile-support-unconditionally-on-uC.patch
@@ -1,7 +1,7 @@
-From 271d855c0ad94a2ec164e20173257a06cd83288d Mon Sep 17 00:00:00 2001
+From 6bb94d51e25b394bc9c52699f744a7cc06ffca15 Mon Sep 17 00:00:00 2001
 From: =?UTF-8?q?J=C3=A9r=C3=B4me=20Pouiller?= <jezz@sysmic.org>
 Date: Thu, 2 Jan 2014 16:00:06 +0100
-Subject: [PATCH 2/3] don't enable largefile support unconditionally on uClibc
+Subject: [PATCH 2/2] don't enable largefile support unconditionally on uClibc
 MIME-Version: 1.0
 Content-Type: text/plain; charset=UTF-8
 Content-Transfer-Encoding: 8bit
@@ -19,7 +19,7 @@ Signed-off-by: J?r?me Pouiller <jezz@sysmic.org>
  3 files changed, 12 insertions(+)
 
 diff --git a/coregrind/m_initimg/initimg-linux.c b/coregrind/m_initimg/initimg-linux.c
-index 9bcc05c..ffee30f 100644
+index a8e7d27..fa670cb 100644
 --- a/coregrind/m_initimg/initimg-linux.c
 +++ b/coregrind/m_initimg/initimg-linux.c
 @@ -55,7 +55,11 @@
@@ -35,10 +35,10 @@ index 9bcc05c..ffee30f 100644
  #include <elf.h>
  /* --- !!! --- EXTERNAL HEADERS end --- !!! --- */
 diff --git a/coregrind/m_main.c b/coregrind/m_main.c
-index 35c11e1..0372946 100644
+index e7159a8..4371a1f 100644
 --- a/coregrind/m_main.c
 +++ b/coregrind/m_main.c
-@@ -2997,7 +2997,11 @@ asm(
+@@ -3143,7 +3143,11 @@ asm(
  
  /* --- !!! --- EXTERNAL HEADERS start --- !!! --- */
  #define _GNU_SOURCE
@@ -51,7 +51,7 @@ index 35c11e1..0372946 100644
  #include <elf.h>
  /* --- !!! --- EXTERNAL HEADERS end --- !!! --- */
 diff --git a/coregrind/m_ume/elf.c b/coregrind/m_ume/elf.c
-index b5b30e5..0766c74 100644
+index 4615da3..06992a4 100644
 --- a/coregrind/m_ume/elf.c
 +++ b/coregrind/m_ume/elf.c
 @@ -48,7 +48,11 @@
@@ -67,5 +67,5 @@ index b5b30e5..0766c74 100644
  #include <elf.h>
  /* --- !!! --- EXTERNAL HEADERS end --- !!! --- */
 -- 
-1.8.3.2
+1.9.1
 
diff --git a/package/valgrind/valgrind-0003-Add-replacement-for-a.out.h.patch b/package/valgrind/valgrind-0003-Add-replacement-for-a.out.h.patch
deleted file mode 100644
index f538d3c..0000000
--- a/package/valgrind/valgrind-0003-Add-replacement-for-a.out.h.patch
+++ /dev/null
@@ -1,58 +0,0 @@
-From d560106f300dd2547c659511c8668d88a50eb577 Mon Sep 17 00:00:00 2001
-From: =?UTF-8?q?J=C3=A9r=C3=B4me=20Pouiller?= <jezz@sysmic.org>
-Date: Thu, 2 Jan 2014 16:00:49 +0100
-Subject: [PATCH 3/3] Add replacement for <a.out.h>
-MIME-Version: 1.0
-Content-Type: text/plain; charset=UTF-8
-Content-Transfer-Encoding: 8bit
-
-Valgrind includes <a.out.h> to get the definition of 'struct
-nlist'. However, while glibc directly defines 'struct nlist' in
-<a.out.h>, uClibc relies on it being defined by kernel headers (i.e
-<a.out.h> simply includes <linux/a.out.h>). This works for most
-architectures, but not for PowerPC, on which the a.out binary format
-has never been supported, and therefore the <linux/a.out.h> kernel
-header does not exist.
-
-One solution would have been to use the <nlist.h> header, but this one
-is only available in glibc, and it also has a slightly different
-definition than the one in <a.out.h>. So, for the time being, the
-easiest solution is to just replace the #include <a.out.h> in Valgrind
-code by a copy/paste of the 'struct nlist' definition.
-
-Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
-Signed-off-by: J?r?me Pouiller <jezz@sysmic.org>
----
- coregrind/m_debuginfo/readstabs.c | 16 +++++++++++++++-
- 1 file changed, 15 insertions(+), 1 deletion(-)
-
-diff --git a/coregrind/m_debuginfo/readstabs.c b/coregrind/m_debuginfo/readstabs.c
-index 70ad319..47f5bc7 100644
---- a/coregrind/m_debuginfo/readstabs.c
-+++ b/coregrind/m_debuginfo/readstabs.c
-@@ -54,7 +54,21 @@
- 
- /* --- !!! --- EXTERNAL HEADERS start --- !!! --- */
- #if defined(VGO_linux)
--#  include <a.out.h> /* stabs defns */
-+/* Copied from a.out.h, because it is otherwise not available on
-+   PowerPC/uClibc */
-+struct nlist
-+{
-+  union
-+    {
-+      char *n_name;
-+      struct nlist *n_next;
-+      long n_strx;
-+    } n_un;
-+  unsigned char n_type;
-+  char n_other;
-+  short n_desc;
-+  unsigned long n_value;
-+};
- #elif defined(VGO_darwin)
- #  include <mach-o/nlist.h>
- #  define n_other n_sect
--- 
-1.8.3.2
-
diff --git a/package/valgrind/valgrind-0004-remove-default-mips-flags.patch b/package/valgrind/valgrind-0004-remove-default-mips-flags.patch
deleted file mode 100644
index 14a493b..0000000
--- a/package/valgrind/valgrind-0004-remove-default-mips-flags.patch
+++ /dev/null
@@ -1,93 +0,0 @@
-From 374743e6faa9d3fc2fba1489e2ead8ea8ebd1f62 Mon Sep 17 00:00:00 2001
-From: Vicente Olivert Riera <Vincent.Riera@imgtec.com>
-Date: Mon, 31 Mar 2014 12:00:39 +0100
-Subject: [PATCH] mips32/64: Remove default flags -mips32 and -mips64 from Makefile.all.am
-
-During configuration of Valgrind we check does the compiler support
--march=mips32 and -march=mips64. If compiler supports these flags we are
-using them as default flags for mips32 and mips64.
-
-Original upstream patch:
-   https://github.com/svn2github/valgrind/commit/fdf6c5aea4671c3c43c90230510735d215dd1e1c
-
-Signed-off-by: Vicente Olivert Riera <Vincent.Riera@imgtec.com>
----
- Makefile.all.am |    8 ++++----
- configure.ac    |   40 ++++++++++++++++++++++++++++++++++++++++
- 2 files changed, 44 insertions(+), 4 deletions(-)
-
-diff --git a/Makefile.all.am b/Makefile.all.am
-index 1f69802..405f11f 100644
---- a/Makefile.all.am
-+++ b/Makefile.all.am
-@@ -173,12 +173,12 @@ AM_CFLAGS_S390X_LINUX     = @FLAG_M64@ $(AM_CFLAGS_BASE) -fomit-frame-pointer
- AM_CCASFLAGS_S390X_LINUX  = @FLAG_M64@ -g -mzarch -march=z900
- 
- AM_FLAG_M3264_MIPS32_LINUX = @FLAG_M32@
--AM_CFLAGS_MIPS32_LINUX     = @FLAG_M32@ $(AM_CFLAGS_BASE) -mips32
--AM_CCASFLAGS_MIPS32_LINUX  = @FLAG_M32@ -mips32 -g
-+AM_CFLAGS_MIPS32_LINUX     = @FLAG_M32@ $(AM_CFLAGS_BASE) @FLAG_MIPS32@
-+AM_CCASFLAGS_MIPS32_LINUX  = @FLAG_M32@ -g @FLAG_MIPS32@
- 
- AM_FLAG_M3264_MIPS64_LINUX = @FLAG_M64@
--AM_CFLAGS_MIPS64_LINUX     = @FLAG_M64@ $(AM_CFLAGS_BASE) -mips64
--AM_CCASFLAGS_MIPS64_LINUX  = @FLAG_M64@ -mips64 -g
-+AM_CFLAGS_MIPS64_LINUX     = @FLAG_M64@ $(AM_CFLAGS_BASE) @FLAG_MIPS64@
-+AM_CCASFLAGS_MIPS64_LINUX  = @FLAG_M64@ -g @FLAG_MIPS64@
- 
- # Flags for the primary target.  These must be used to build the
- # regtests and performance tests.  In fact, these must be used to
-diff --git a/configure.ac b/configure.ac
-index 229ab98..0e3884c 100644
---- a/configure.ac
-+++ b/configure.ac
-@@ -1450,6 +1450,46 @@ CFLAGS=$safe_CFLAGS
- AC_SUBST(FLAG_M64)
- 
- 
-+# does this compiler support -march=mips32 (mips32 default) ?
-+AC_MSG_CHECKING([if gcc accepts -march=mips32])
-+
-+safe_CFLAGS=$CFLAGS
-+CFLAGS="$CFLAGS -march=mips32"
-+
-+AC_COMPILE_IFELSE([AC_LANG_PROGRAM([[]], [[
-+  return 0;
-+]])], [
-+FLAG_MIPS32="-march=mips32"
-+AC_MSG_RESULT([yes])
-+], [
-+FLAG_MIPS32=""
-+AC_MSG_RESULT([no])
-+])
-+CFLAGS=$safe_CFLAGS
-+
-+AC_SUBST(FLAG_MIPS32)
-+
-+
-+# does this compiler support -march=mips64 (mips64 default) ?
-+AC_MSG_CHECKING([if gcc accepts -march=mips64])
-+
-+safe_CFLAGS=$CFLAGS
-+CFLAGS="$CFLAGS -march=mips64"
-+
-+AC_COMPILE_IFELSE([AC_LANG_PROGRAM([[]], [[
-+  return 0;
-+]])], [
-+FLAG_MIPS64="-march=mips64"
-+AC_MSG_RESULT([yes])
-+], [
-+FLAG_MIPS64=""
-+AC_MSG_RESULT([no])
-+])
-+CFLAGS=$safe_CFLAGS
-+
-+AC_SUBST(FLAG_MIPS64)
-+
-+
- # does this compiler support -mmmx ?
- AC_MSG_CHECKING([if gcc accepts -mmmx])
- 
--- 
-1.7.1
-
diff --git a/package/valgrind/valgrind-0005-glibc-2.19.patch b/package/valgrind/valgrind-0005-glibc-2.19.patch
deleted file mode 100644
index e386a79..0000000
--- a/package/valgrind/valgrind-0005-glibc-2.19.patch
+++ /dev/null
@@ -1,32 +0,0 @@
-Enable valgrind for glibc 2.19 targets, and update the error from 2.17
-to 2.19 (which in reality was 2.18).
-Status: upstream.
-
-Signed-off-by: Gustavo Zacarias <gustavo@zacarias.com.ar>
-
-diff -Nura valgrind-3.9.0.orig/configure.ac valgrind-3.9.0/configure.ac
---- valgrind-3.9.0.orig/configure.ac	2014-06-18 19:06:39.212571710 -0300
-+++ valgrind-3.9.0/configure.ac	2014-06-18 19:08:37.385557736 -0300
-@@ -918,6 +918,13 @@
- 	DEFAULT_SUPP="glibc-2.34567-NPTL-helgrind.supp ${DEFAULT_SUPP}"
- 	DEFAULT_SUPP="glibc-2.X-drd.supp ${DEFAULT_SUPP}"
- 	;;
-+     2.19)
-+        AC_MSG_RESULT(2.19 family)
-+	AC_DEFINE([GLIBC_2_19], 1, [Define to 1 if you're using glibc 2.19.x])
-+	DEFAULT_SUPP="glibc-2.X.supp ${DEFAULT_SUPP}"
-+	DEFAULT_SUPP="glibc-2.34567-NPTL-helgrind.supp ${DEFAULT_SUPP}"
-+	DEFAULT_SUPP="glibc-2.X-drd.supp ${DEFAULT_SUPP}"
-+	;;
-      darwin)
- 	AC_MSG_RESULT(Darwin)
- 	AC_DEFINE([DARWIN_LIBC], 1, [Define to 1 if you're using Darwin])
-@@ -931,7 +938,7 @@
- 
-      *)
- 	AC_MSG_RESULT([unsupported version ${GLIBC_VERSION}])
--	AC_MSG_ERROR([Valgrind requires glibc version 2.2 - 2.17])
-+	AC_MSG_ERROR([Valgrind requires glibc version 2.2 - 2.19])
- 	AC_MSG_ERROR([or Darwin libc])
- 	;;
- esac
diff --git a/package/valgrind/valgrind.mk b/package/valgrind/valgrind.mk
index aee7c9d..24d0290 100644
--- a/package/valgrind/valgrind.mk
+++ b/package/valgrind/valgrind.mk
@@ -4,7 +4,7 @@
 #
 ################################################################################
 
-VALGRIND_VERSION = 3.9.0
+VALGRIND_VERSION = 3.10.0
 VALGRIND_SITE    = http://valgrind.org/downloads
 VALGRIND_SOURCE  = valgrind-$(VALGRIND_VERSION).tar.bz2
 VALGRIND_LICENSE = GPLv2 GFDLv1.2
-- 
1.9.1

[Buildroot] [PATCH 4/8] lttng-modules: bump to version 2.5.0

[Jérôme Pouiller]

This version should support kernel up to 3.15.

Signed-off-by: J?r?me Pouiller <jezz@sysmic.org>
---
 package/lttng-modules/lttng-modules.mk | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/package/lttng-modules/lttng-modules.mk b/package/lttng-modules/lttng-modules.mk
index 346214a..72d2eb7 100644
--- a/package/lttng-modules/lttng-modules.mk
+++ b/package/lttng-modules/lttng-modules.mk
@@ -4,7 +4,7 @@
 #
 ################################################################################
 
-LTTNG_MODULES_VERSION = 2.4.1
+LTTNG_MODULES_VERSION = 2.5.0
 LTTNG_MODULES_SITE    = http://lttng.org/files/lttng-modules
 LTTNG_MODULES_SOURCE  = lttng-modules-$(LTTNG_MODULES_VERSION).tar.bz2
 LTTNG_MODULES_LICENSE = LGPLv2.1/GPLv2 for kernel modules; MIT for lib/bitfield.h and lib/prio_heap/*
-- 
1.9.1

[Buildroot] [PATCH 4/8] lttng-modules: bump to version 2.5.0

[Jérôme Pouiller]

On Tuesday 23 September 2014 11:45:43 J?r?me Pouiller wrote:
> This version should support kernel up to 3.15.
> 
> Signed-off-by: J?r?me Pouiller <jezz@sysmic.org>
> ---
>  package/lttng-modules/lttng-modules.mk | 2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)
> 
> diff --git a/package/lttng-modules/lttng-modules.mk
> b/package/lttng-modules/lttng-modules.mk index 346214a..72d2eb7 100644
> --- a/package/lttng-modules/lttng-modules.mk
> +++ b/package/lttng-modules/lttng-modules.mk
> @@ -4,7 +4,7 @@
>  #
>  ###########################################################################
> #####
> 
> -LTTNG_MODULES_VERSION = 2.4.1
> +LTTNG_MODULES_VERSION = 2.5.0
>  LTTNG_MODULES_SITE    = http://lttng.org/files/lttng-modules
>  LTTNG_MODULES_SOURCE  = lttng-modules-$(LTTNG_MODULES_VERSION).tar.bz2
>  LTTNG_MODULES_LICENSE = LGPLv2.1/GPLv2 for kernel modules; MIT for
> lib/bitfield.h and lib/prio_heap/*
Oops, this patch would not have been in this serie. I have also to bump 
versions of other lttng packages.


-- 
J?r?me Pouiller, Sysmic
Embedded Linux specialist
http://www.sysmic.fr

[Buildroot] [PATCH 5/8] libseccomp: bump to version 2.1.1

[Jérôme Pouiller]

Version 1.0.0 was not compatible with systemd. With 1.0.0, systemd
compilation produce:

src/shared/seccomp-util.c: In function 'seccomp_add_secondary_archs':
src/shared/seccomp-util.c:73:9: warning: implicit declaration of function 'seccomp_arch_add' [-Wimplicit-function-declaration]
         r = seccomp_arch_add(c, SCMP_ARCH_X86);
         ^
src/shared/seccomp-util.c:73:9: warning: nested extern declaration of 'seccomp_arch_add' [-Wnested-externs]
src/shared/seccomp-util.c:73:33: error: 'SCMP_ARCH_X86' undeclared (first use in this function)
         r = seccomp_arch_add(c, SCMP_ARCH_X86);
                                 ^
src/shared/seccomp-util.c:77:33: error: 'SCMP_ARCH_X86_64' undeclared (first use in this function)
         r = seccomp_arch_add(c, SCMP_ARCH_X86_64);
                                 ^
src/shared/seccomp-util.c:81:33: error: 'SCMP_ARCH_X32' undeclared (first use in this function)
         r = seccomp_arch_add(c, SCMP_ARCH_X32);

Signed-off-by: J?r?me Pouiller <jezz@sysmic.org>
---
 package/libseccomp/libseccomp.mk | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/package/libseccomp/libseccomp.mk b/package/libseccomp/libseccomp.mk
index da1dadd..3002388 100644
--- a/package/libseccomp/libseccomp.mk
+++ b/package/libseccomp/libseccomp.mk
@@ -4,8 +4,8 @@
 #
 ################################################################################
 
-LIBSECCOMP_VERSION         = 1.0.0
-LIBSECCOMP_SITE            = http://downloads.sourceforge.net/project/libseccomp
+LIBSECCOMP_VERSION         = 2.1.1
+LIBSECCOMP_SITE            = http://downloads.sourceforge.net/project/libseccomp/libseccomp-$(LIBSECCOMP_VERSION)/
 LIBSECCOMP_LICENSE         = LGPLv2.1
 LIBSECCOMP_LICENSE_FILES   = LICENSE
 LIBSECCOMP_INSTALL_STAGING = YES
-- 
1.9.1

[Buildroot] [PATCH 6/8] systemd: make libseccomp usage determinist

[Jérôme Pouiller]

Signed-off-by: J?r?me Pouiller <jezz@sysmic.org>
---
 package/systemd/systemd.mk | 7 +++++++
 1 file changed, 7 insertions(+)

diff --git a/package/systemd/systemd.mk b/package/systemd/systemd.mk
index 39b458e..c7cff0d 100644
--- a/package/systemd/systemd.mk
+++ b/package/systemd/systemd.mk
@@ -76,6 +76,13 @@ else
 SYSTEMD_CONF_OPT += --disable-gudev
 endif
 
+ifeq ($(BR2_PACKAGE_LIBSECCOMP),y)
+SYSTEMD_CONF_OPT += --enable-seccomp
+SYSTEMD_DEPENDENCIES += libseccomp
+else
+SYSTEMD_CONF_OPT += --disable-seccomp
+endif
+
 ifeq ($(BR2_PACKAGE_SYSTEMD_ALL_EXTRAS),y)
 SYSTEMD_DEPENDENCIES += \
 	xz 		\
-- 
1.9.1

[Buildroot] [PATCH 7/8] libssh2: fix static link with libgcrypt

[Jérôme Pouiller]

BR call configure script with environment variable
lt_cv_sys_lib_search_path_spec="" (see packages/Makefile.in). This prevent
libtool to detect compiler builtin paths. Thus, in case where
BR2_PREFER_STATIC_LIB is enabled, libtool does not find libgpg-error
dependency:

/home/jezz/br-continuous/cfgs/31-x86_64-core2-uclibc-static/host/usr/i686-buildroot-linux-uclibc/sysroot/usr/lib/libgcrypt.a(rsa.o):
In function `run_selftests':
rsa.c:(.text+0xd2): undefined reference to `gpg_strerror'
rsa.c:(.text+0xfc): undefined reference to `gpg_strerror'

For this case, we may remove $lt_cv_sys_lib_search_path_spec from environment.
However, it not easy to remove a variable from default environment. So, we
prefer to pass a correct value.

Signed-off-by: J?r?me Pouiller <jezz@sysmic.org>
---
 package/libssh2/libssh2.mk | 1 +
 1 file changed, 1 insertion(+)

diff --git a/package/libssh2/libssh2.mk b/package/libssh2/libssh2.mk
index 50bd265..193714e 100644
--- a/package/libssh2/libssh2.mk
+++ b/package/libssh2/libssh2.mk
@@ -10,6 +10,7 @@ LIBSSH2_LICENSE = BSD
 LIBSSH2_LICENSE_FILES = COPYING
 LIBSSH2_INSTALL_STAGING = YES
 LIBSSH2_CONF_OPT = --disable-examples-build
+LIBSSH2_CONF_ENV += lt_cv_sys_lib_search_path_spec="$(STAGING_DIR)/usr/lib"
 
 # libssh2 must use either libgcrypt or OpenSSL
 # Only select openssl if libgcrypt is not selected
-- 
1.9.1

[Buildroot] [PATCH 7/8] libssh2: fix static link with libgcrypt

[Thomas Petazzoni]

Dear J?r?me Pouiller,

On Tue, 23 Sep 2014 11:45:46 +0200, J?r?me Pouiller wrote:
> BR call configure script with environment variable
> lt_cv_sys_lib_search_path_spec="" (see packages/Makefile.in). This prevent
> libtool to detect compiler builtin paths. Thus, in case where
> BR2_PREFER_STATIC_LIB is enabled, libtool does not find libgpg-error
> dependency:
> 
> /home/jezz/br-continuous/cfgs/31-x86_64-core2-uclibc-static/host/usr/i686-buildroot-linux-uclibc/sysroot/usr/lib/libgcrypt.a(rsa.o):
> In function `run_selftests':
> rsa.c:(.text+0xd2): undefined reference to `gpg_strerror'
> rsa.c:(.text+0xfc): undefined reference to `gpg_strerror'
> 
> For this case, we may remove $lt_cv_sys_lib_search_path_spec from environment.
> However, it not easy to remove a variable from default environment. So, we
> prefer to pass a correct value.
> 
> Signed-off-by: J?r?me Pouiller <jezz@sysmic.org>

Since this patch, some other changes have been made to the libssh2
package, and it now links fine in a static configuration. I've marked
this patch as Rejected in patchwork, then.

Thanks,

Thomas
-- 
Thomas Petazzoni, CTO, Free Electrons
Embedded Linux, Kernel and Android engineering
http://free-electrons.com

[Buildroot] [PATCH 8/8] directfb: disable multi application if no MMU

[Jérôme Pouiller]

directfb-fusion use madvise which is not available on no-MMU targets.

It seems it would be possible to simply remove call to madvise (there is
only one call) if not available. However, since I am not sure, I prefer
to completly disable directfb-fusion.

Signed-off-by: J?r?me Pouiller <jezz@sysmic.org>
---
 package/directfb/Config.in | 1 +
 1 file changed, 1 insertion(+)

diff --git a/package/directfb/Config.in b/package/directfb/Config.in
index d53f756..3aa51cb 100644
--- a/package/directfb/Config.in
+++ b/package/directfb/Config.in
@@ -12,6 +12,7 @@ if BR2_PACKAGE_DIRECTFB
 config BR2_PACKAGE_DIRECTFB_MULTI
 	bool "directfb multi application"
 	select BR2_PACKAGE_LINUX_FUSION
+	depends on BR2_USE_MMU # madvise()
 	depends on BR2_LINUX_KERNEL
 	help
 	  Enable use of multiple concurrent DirectFB applications
-- 
1.9.1

[Buildroot] [PATCH 1/8] connman: bump to version 1.25

[Thomas Petazzoni]

Dear J?r?me Pouiller,

On Tue, 23 Sep 2014 11:45:40 +0200, J?r?me Pouiller wrote:
> Signed-off-by: J?r?me Pouiller <jezz@sysmic.org>
> ---
>  package/connman/connman.mk | 2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)

Applied patch 1, 2, 3, 5, 6 and 8 of your patch series.

Patch 4 you said yourself that it should be withdrawn, so I've marked
it as Rejected in patchwork.

Patch 7 is a bit more complicated, so I'd like to give it some time for
reviewers/testers to comment.

Thanks,

Thomas
-- 
Thomas Petazzoni, CTO, Free Electrons
Embedded Linux, Kernel and Android engineering
http://free-electrons.com