Pull request for branch 2010-05-15-fixes
The following changes since commit 5e25bc482134bee62e3cc9d55af694aaa977d5b5:
Thomas Petazzoni (1):
external toolchain: check BR2_INSTALL_LIBSTDCPP
are available in the git repository at:
git://git.busybox.net/~tpetazzoni/git/buildroot 2010-05-15-fixes
Thomas Petazzoni (12):
iconv: disallow selection when BR2_ENABLE_LOCALE is enabled
lmbench: needs BR2_INET_RPC
startup-notification: add dependency on xlib_libX11
ntp: fix build failure
samba: only require libiconv if !BR2_ENABLE_LOCALE
pango: run pango-querymodules on the target instead of on the host
cairo: remove host variant
mutt: fix build and convert to autotools
lvm2: do not override CC and similar variables
uemacs: don't forget CFLAGS at link time
vpnc: pass CFLAGS at link time
netsnmp: disable build with external toolchain
package/cairo/cairo.mk | 3 - package/libiconv/Config.in | 1 + package/lmbench/Config.in | 2 + package/lvm2/lvm2.mk | 2 +- package/mutt/mutt-1.5.16-makedoc-hostcc.patch | 32 --------- package/mutt/mutt-1.5.17+20080114-nodoc.patch | 21 ++++++ package/mutt/mutt.mk | 73 +++----------------- package/netsnmp/Config.in | 3 + package/ntp/ntp-4.2.4_p5-nano.patch | 36 ++++++++++ package/pango/S25pango | 20 ++++++ package/pango/pango.mk | 18 +---- package/samba/Config.in | 2 +- package/samba/samba.mk | 2 +- package/startup-notification/Config.in | 2 + .../startup-notification/startup-notification.mk | 1 + package/uemacs/uemacs.mk | 2 +- package/vpnc/vpnc.mk | 4 +- 17 files changed, 105 insertions(+), 119 deletions(-) delete mode 100644 package/mutt/mutt-1.5.16-makedoc-hostcc.patch create mode 100644 package/mutt/mutt-1.5.17+20080114-nodoc.patch create mode 100644 package/ntp/ntp-4.2.4_p5-nano.patch create mode 100644 package/pango/S25pango
Thanks,
Signature · 1 line
Thomas Petazzoni
PATCH 1/12 iconv: disallow selection when BR2_ENABLE_LOCALE is enabled
A C library compiled with locale support already has an iconv() implementation. In that case, libiconv is useless. Therefore, disallow the selection of libiconv when the configuration says that the toolchain has locale support.
This fixes build issues similar to the one we had with gettext:
libiconv headers are installed, but -liconv is not added because the ./configure script detects that the C library supports iconv.
--- package/libiconv/Config.in | 1 + 1 files changed, 1 insertions(+), 0 deletions(-)
package/libiconv/Config.in +1 -0
@@ -1,5 +1,6 @@ config BR2_PACKAGE_LIBICONV bool "libiconv" + depends on !BR2_ENABLE_LOCALE help unicode conversion library
Signature · 1 line
1.6.3.3
PATCH 2/12 lmbench: needs BR2_INET_RPC
In reply to Thomas PetazzoniThis package uses several pmap_*() functions in the C library, that are only available if the C library has RPC support.
--- package/lmbench/Config.in | 2 ++ 1 files changed, 2 insertions(+), 0 deletions(-)
package/lmbench/Config.in +2 -0
@@ -1,5 +1,7 @@ config BR2_PACKAGE_LMBENCH bool "lmbench" + # Uses pmap_set, pmap__unset, pmap_getport, etc. + depends on BR2_INET_RPC help LMbench is a suite of simple, portable, ANSI/C microbenchmarks for UNIX/POSIX.
Signature · 1 line
1.6.3.3
PATCH 2/12 lmbench: needs BR2_INET_RPC
Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line
"Thomas" == Thomas Petazzoni <[email protected]> writes:
Thomas> This package uses several pmap_*() functions in the C library, that Thomas> are only available if the C library has RPC support.
A comment "lmbench needs a toolchain with RPC support" would be good, but I can add that myself.
Signature · 1 line
Bye, Peter Korsgaard
PATCH 3/12 startup-notification: add dependency on xlib_libX11
In reply to Thomas Petazzoni--- package/startup-notification/Config.in | 2 ++ .../startup-notification/startup-notification.mk | 1 + 2 files changed, 3 insertions(+), 0 deletions(-)
package/startup-notification/Config.in +2 -0
@@ -1,5 +1,7 @@ config BR2_PACKAGE_STARTUP_NOTIFICATION bool "startup-notification" + select BR2_PACKAGE_XLIB_LIBX11 + depends on BR2_PACKAGE_XORG7 help Startup-notification is a library used to monitor application startup.
package/startup-notification/startup-notification.mk +1 -0
@@ -9,6 +9,7 @@ STARTUP_NOTIFICATION_SITE = http://freedesktop.org/software/startup-notification STARTUP_NOTIFICATION_AUTORECONF = NO STARTUP_NOTIFICATION_INSTALL_STAGING = YES STARTUP_NOTIFICATION_INSTALL_TARGET = YES +STARTUP_NOTIFICATION = xlib_libX11 STARTUP_NOTIFICATION_CONF_ENV = lf_cv_sane_realloc=yes
Signature · 1 line
1.6.3.3
PATCH 4/12 ntp: fix build failure
In reply to Thomas PetazzoniThis patch adds a patch to NTP to fix the following build failure, using a workaround patch found on the Fedora bugzilla:
ntp_loopfilter.c: In function 'local_clock':
ntp_loopfilter.c:571: error: 'MOD_NANO' undeclared (first use in this function) ntp_loopfilter.c:571: error: (Each undeclared identifier is reported only once ntp_loopfilter.c:571: error: for each function it appears in.) ntp_loopfilter.c: In function 'loop_config':
ntp_loopfilter.c:896: error: 'MOD_NANO' undeclared (first use in this function)
--- package/ntp/ntp-4.2.4_p5-nano.patch | 36 +++++++++++++++++++++++++++++++++++ 1 files changed, 36 insertions(+), 0 deletions(-) create mode 100644 package/ntp/ntp-4.2.4_p5-nano.patch
package/ntp/ntp-4.2.4_p5-nano.patch new file +36 -0
@@ -0,0 +1,36 @@ +Fixes the MOD_NANO build failure + +ntp_loopfilter.c: In function 'local_clock': +ntp_loopfilter.c:571: error: 'MOD_NANO' undeclared (first use in this function) +ntp_loopfilter.c:571: error: (Each undeclared identifier is reported only once +ntp_loopfilter.c:571: error: for each function it appears in.) +ntp_loopfilter.c: In function 'loop_config': +ntp_loopfilter.c:896: error: 'MOD_NANO' undeclared (first use in this function) + +Taken from +http://cvs.fedoraproject.org/viewvc/devel/ntp/ntp-4.2.4p7-nano.patch?revision=1.1&content-type=text/plain&view=co + +See the discussion at +https://bugs.launchpad.net/ubuntu/+source/ntp/+bug/412242 for more +details and links. The below fix is not a correct fix, but only a +workaround. The real fix required changes to kernel/libc headers. + +Index: ntp-4.2.4p5/include/ntp_syscall.h +=================================================================== +--- ntp-4.2.4p5.orig/include/ntp_syscall.h 2010-05-14 17:34:48.000000000 +0200 ++++ ntp-4.2.4p5/include/ntp_syscall.h 2010-05-14 17:35:13.000000000 +0200 +@@ -14,6 +14,14 @@ + # include <sys/timex.h> + #endif + ++#if defined(ADJ_NANO) && !defined(MOD_NANO) ++#define MOD_NANO ADJ_NANO ++#endif ++ ++#if defined(ADJ_TAI) && !defined(MOD_TAI) ++#define MOD_TAI ADJ_TAI ++#endif ++ + #ifndef NTP_SYSCALLS_LIBC + #ifdef NTP_SYSCALLS_STD + # define ntp_adjtime(t) syscall(SYS_ntp_adjtime, (t))
Signature · 1 line
1.6.3.3
PATCH 5/12 samba: only require libiconv if !BR2_ENABLE_LOCALE
In reply to Thomas Petazzoni--- package/samba/Config.in | 2 +- package/samba/samba.mk | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-)
package/samba/Config.in +1 -1
@@ -1,6 +1,6 @@ config BR2_PACKAGE_SAMBA bool "samba" - select BR2_PACKAGE_LIBICONV + select BR2_PACKAGE_LIBICONV if !BR2_ENABLE_LOCALE help Provides print services to all manner of SMB/CIFS clients, including the numerous versions of Microsoft Windows
package/samba/samba.mk +1 -1
@@ -16,7 +16,7 @@ SAMBA_INSTALL_TARGET = YES SAMBA_DEPENDENCIES = \ - libiconv \ + $(if $(BR2_ENABLE_LOCALE),,libiconv) \ $(if $(BR2_PACKAGE_SAMBA_RPCCLIENT),readline) \ $(if $(BR2_PACKAGE_SAMBA_SMBCLIENT),readline) \ $(if $(BR2_PACKAGE_SAMBA_AVAHI),avahi) \
Signature · 1 line
1.6.3.3
PATCH 6/12 pango: run pango-querymodules on the target instead of on the host
In reply to Thomas PetazzoniWe needed to build pango for the host in order to compile pango-querymodules for the host, which was used to produce /etc/pango/pango.modules. Unfortunately:
* This produces an incorrect /etc/pango/pango.modules (no modules detected in my case), probably because the host pango-querymodules was looking at host pango modules
* This requires to build pango for the host, which requires to build cairo for the host, which requires to build X11 for the host.
To make things work and remove the dependency between pango and host-pango, we introduce a S25pango script that creates /etc/pango/pango.modules on startup if it doesn't exist, just as we do with libgtk2 for /etc/gtk-2.0/gdk-pixbuf.loaders.
Since host-pango is no longer needed, we remove all definitions related to it.
--- package/pango/S25pango | 20 ++++++++++++++++++++ package/pango/pango.mk | 18 ++---------------- 2 files changed, 22 insertions(+), 16 deletions(-) create mode 100644 package/pango/S25pango
package/pango/S25pango new file +20 -0
@@ -0,0 +1,20 @@ +#!/bin/sh +# +# run pango-querymodules if needed + +FILE=/etc/pango/pango.modules + +case "$1" in + start|"") + if [ ! -f $FILE ] ; then + mkdir -p /etc/pango + /usr/bin/pango-querymodules > "$FILE" + fi + ;; + stop) + ;; + *) + echo "Usage: $0 {start|stop}" >&2 + exit 1 + ;; +esac
package/pango/pango.mk +2 -16
@@ -40,16 +40,7 @@ PANGO_CONF_ENV = ac_cv_func_posix_getpwuid_r=yes glib_cv_stack_grows=no \ PANGO_CONF_OPT = --enable-shared --enable-static \ --enable-explicit-deps=no --disable-debug -HOST_PANGO_CONF_OPT = \ - --disable-static \ - $(if $(BR2_PACKAGE_XORG7),--with-x,--without-x) \ - --disable-debug \ - -PANGO_DEPENDENCIES = $(if $(BR2_NEEDS_GETTEXT),gettext libintl) host-pkg-config host-pango libglib2 cairo - -HOST_PANGO_DEPENDENCIES = host-pkg-config host-cairo host-libglib2 host-autoconf host-automake - -HOST_PANGO_AUTORECONF = YES +PANGO_DEPENDENCIES = $(if $(BR2_NEEDS_GETTEXT),gettext libintl) host-pkg-config libglib2 cairo ifeq ($(BR2_PACKAGE_XORG7),y) PANGO_CONF_OPT += --with-x \
@@ -61,12 +52,7 @@ else endif $(eval $(call AUTOTARGETS,package,pango)) -$(eval $(call AUTOTARGETS,package,pango,host)) $(PANGO_HOOK_POST_INSTALL): - mkdir -p $(TARGET_DIR)/etc/pango - $(PANGO_HOST_BINARY) > $(TARGET_DIR)/etc/pango/pango.modules - $(SED) 's~$(HOST_DIR)~~g' $(TARGET_DIR)/etc/pango/pango.modules + $(INSTALL) -m 755 package/pango/S25pango $(TARGET_DIR)/etc/init.d/ touch $@ - -PANGO_HOST_BINARY:=$(HOST_DIR)/usr/bin/pango-querymodules
Signature · 1 line
1.6.3.3
PATCH 7/12 cairo: remove host variant
In reply to Thomas PetazzoniSince we no longer need to build pango on the host, cairo on the host is no longer needed.
--- package/cairo/cairo.mk | 3 --- 1 files changed, 0 insertions(+), 3 deletions(-)
package/cairo/cairo.mk +0 -3
@@ -79,8 +79,6 @@ else CAIRO_CONF_OPT += --disable-svg endif -HOST_CAIRO_DEPENDENCIES = host-pkg-config host-pixman host-fontconfig - HOST_CAIRO_CONF_OPT = \ --enable-ps \ --enable-pdf \
@@ -90,4 +88,3 @@ HOST_CAIRO_CONF_OPT = \ --disable-svg $(eval $(call AUTOTARGETS,package,cairo)) -$(eval $(call AUTOTARGETS,package,cairo,host))
Signature · 1 line
1.6.3.3
PATCH 8/12 mutt: fix build and convert to autotools
In reply to Thomas Petazzonimutt has been upgraded, but the mutt-1.5.16-makedoc-hostcc.patch wasn't upgraded accordingly, causing build failures in the documentation. Therefore, we add a new patch mutt-1.5.17+20080114-nodoc.patch that disables the construction of the documentation.
As this patch modifies Makefile.am, we need to autoreconf the package. The simplest and cleanest way to do this is to convert the package to the autotools infrastructure.
--- package/mutt/mutt-1.5.16-makedoc-hostcc.patch | 32 ----------- package/mutt/mutt-1.5.17+20080114-nodoc.patch | 21 +++++++ package/mutt/mutt.mk | 73 ++++--------------------- 3 files changed, 32 insertions(+), 94 deletions(-) delete mode 100644 package/mutt/mutt-1.5.16-makedoc-hostcc.patch create mode 100644 package/mutt/mutt-1.5.17+20080114-nodoc.patch
package/mutt/mutt-1.5.16-makedoc-hostcc.patch deleted +0 -32
@@ -1,32 +0,0 @@ ---- mutt-1.5.16.oorig/Makefile.in 2007-09-02 22:16:23.000000000 +0200 -+++ mutt-1.5.16/Makefile.in 2007-09-02 22:19:49.000000000 +0200 -@@ -164,12 +164,6 @@ - DBX = @DBX@ - DEBUGGER = @DEBUGGER@ - --# $(makedoc_OBJECTS): $(makedoc_SOURCES) --# $(HOST_CC) $(DEFS) $(AM_CPPFLAGS) $(AM_CPPFLAGS) $(CPPFLAGS) $(AM_CFLAGS) -c $< -- --# makedoc: $(makedoc_OBJECTS) $(makedoc_DEPENDENCIES) --# @rm -rf makedoc --# $(HOST_CC) $(AM_CFLAGS) $(LDFLAGS) $(makedoc_LDFLAGS) $(makedoc_OBJECTS) $(makedoc_LDADD) -o makedoc - DEFS = -DPKGDATADIR=\"$(pkgdatadir)\" -DSYSCONFDIR=\"$(sysconfdir)\" \ - -DBINDIR=\"$(bindir)\" -DMUTTLOCALEDIR=\"$(datadir)/locale\" \ - -DHAVE_CONFIG_H=1 -@@ -447,9 +441,15 @@ - - clean-binPROGRAMS: - -test -z "$(bin_PROGRAMS)" || rm -f $(bin_PROGRAMS) -+ -+$(makedoc_OBJECTS): $(makedoc_SOURCES) -+ $(CC_FOR_BUILD) -DHAVE_UNISTD_H -DHAVE_GETOPT_H -DHAVE_STRERROR -c $< ;#$(DEFS) $(AM_CPPFLAGS) $(AM_CPPFLAGS) $(CPPFLAGS) $(AM_CFLAGS) -c $< -+#makedoc: $(makedoc_OBJECTS) $(makedoc_DEPENDENCIES) -+# @rm -rf makedoc -+# $(CC_FOR_BUILD) $(AM_CFLAGS) $(LDFLAGS) $(makedoc_LDFLAGS) $(makedoc_OBJECTS) $(makedoc_LDADD) -o makedoc - makedoc$(EXEEXT): $(makedoc_OBJECTS) $(makedoc_DEPENDENCIES) - @rm -f makedoc$(EXEEXT) -- $(LINK) $(makedoc_LDFLAGS) $(makedoc_OBJECTS) $(makedoc_LDADD) $(LIBS) -+ $(CC_FOR_BUILD) $(makedoc_LDFLAGS) $(makedoc_OBJECTS) $(makedoc_LDADD) $(LIBS) -o $@ - mutt$(EXEEXT): $(mutt_OBJECTS) $(mutt_DEPENDENCIES) - @rm -f mutt$(EXEEXT) - $(LINK) $(mutt_LDFLAGS) $(mutt_OBJECTS) $(mutt_LDADD) $(LIBS)
package/mutt/mutt-1.5.17+20080114-nodoc.patch new file +21 -0
@@ -0,0 +1,21 @@ +Disable the documentation build + +The documentation tool is compiled for the target, but executed on the +host. We don't bother fixing this, as we don't care of the +documentation in Buildroot. + +Signed-off-by: Thomas Petazzoni <[email protected]> + +Index: mutt-1.5.17+20080114/Makefile.am +=================================================================== +--- mutt-1.5.17+20080114.orig/Makefile.am 2010-05-15 18:15:08.000000000 +0200 ++++ mutt-1.5.17+20080114/Makefile.am 2010-05-15 18:15:17.000000000 +0200 +@@ -9,7 +9,7 @@ + IMAP_INCLUDES = -I$(top_srcdir)/imap + endif + +-SUBDIRS = m4 po intl doc contrib $(IMAP_SUBDIR) ++SUBDIRS = m4 po intl contrib $(IMAP_SUBDIR) + + bin_SCRIPTS = muttbug flea @SMIMEAUX_TARGET@ +
package/mutt/mutt.mk +11 -62
82 lines not on this page · raw message
Signature · 1 line
1.6.3.3
PATCH 9/12 lvm2: do not override CC and similar variables
In reply to Thomas PetazzoniSince lvm2 uses autoconf, there's no point in overriding CC and other variables at build time. This causes build breakage as CFLAGS aren't used at link time, therefore --sysroot lacks, which breaks the build with external toolchains.
--- package/lvm2/lvm2.mk | 2 +- 1 files changed, 1 insertions(+), 1 deletions(-)
package/lvm2/lvm2.mk +1 -1
@@ -74,7 +74,7 @@ $(LVM2_DIR)/.configured: $(LVM2_DIR)/.unpacked $(LVM2_DIR)/.built: $(LVM2_DIR)/.configured - $(MAKE1) CC=$(TARGET_CC) RANLIB=$(TARGET_RANLIB) AR=$(TARGET_AR) -C $(LVM2_DIR) DESTDIR=$(STAGING_DIR) + $(MAKE1) -C $(LVM2_DIR) DESTDIR=$(STAGING_DIR) $(MAKE1) -C $(LVM2_DIR) DESTDIR=$(STAGING_DIR) install # Fixup write permissions so that the files can be overwritten # several times in the $(TARGET_DIR)
Signature · 1 line
1.6.3.3
PATCH 9/12 lvm2: do not override CC and similar variables
Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line
"Thomas" == Thomas Petazzoni <[email protected]> writes:
Thomas> Since lvm2 uses autoconf, there's no point in overriding CC and Thomas> other variables at build time. This causes build breakage as Thomas> CFLAGS aren't used at link time, therefore --sysroot lacks, Thomas> which breaks the build with external toolchains.
Thanks. Is there anything stopping us from converting it to Makefile.autotools.in format?
Signature · 1 line
Bye, Peter Korsgaard
PATCH 9/12 lvm2: do not override CC and similar variables
Quoted 2 lines · On Tue, 18 May 2010 13:43:14 +0200 Peter Korsgaard <[email protected]> wrote:
Thanks. Is there anything stopping us from converting it to Makefile.autotools.in format?
Probably not. In this particular case, I didn't do it because we are in the stabilization phase and I didn't want to introduce too much unneeded changes. For mutt, I had to do it because it actually fixed some problems.
Thomas
Signature · 4 lines
Thomas Petazzoni, Free Electrons Kernel, drivers, real-time and embedded Linux development, consulting, training and support. http://free-electrons.com
PATCH 9/12 lvm2: do not override CC and similar variables
Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line
"Thomas" == Thomas Petazzoni <[email protected]> writes:
Thomas> On Tue, 18 May 2010 13:43:14 +0200
Quoted 2 lines · Thomas> Peter Korsgaard <[email protected]> wrote:
Quoted 2 lines
Thanks. Is there anything stopping us from converting it to Makefile.autotools.in format?
Thomas> Probably not. In this particular case, I didn't do it because Thomas> we are in the stabilization phase and I didn't want to Thomas> introduce too much unneeded changes. For mutt, I had to do it Thomas> because it actually fixed some problems.
OK.
Signature · 1 line
Bye, Peter Korsgaard
PATCH 10/12 uemacs: don't forget CFLAGS at link time
In reply to Thomas PetazzoniWithout CFLAGS, the build breaks because --sysroot is missing at link time, and this option is absolutely mandatory for external toolchain builds.
--- package/uemacs/uemacs.mk | 2 +- 1 files changed, 1 insertions(+), 1 deletions(-)
package/uemacs/uemacs.mk +1 -1
@@ -23,7 +23,7 @@ $(UEMACS_DIR)/.unpacked: $(DL_DIR)/$(UEMACS_SOURCE) $(UEMACS_DIR)/$(UEMACS_BINARY): $(UEMACS_DIR)/.unpacked $(MAKE) -C $(UEMACS_DIR) \ - CC="$(TARGET_CC)" DEFINES="-DAUTOCONF -DPOSIX -DUSG" CFLAGS+="$(TARGET_CFLAGS) " LIBS=-lncurses + CC="$(TARGET_CC)" DEFINES="-DAUTOCONF -DPOSIX -DUSG" CFLAGS+="$(TARGET_CFLAGS) " LIBS="$(TARGET_CFLAGS) -lncurses" $(STRIPCMD) $(UEMACS_DIR)/$(UEMACS_BINARY) $(TARGET_DIR)/$(UEMACS_TARGET_BINARY): $(UEMACS_DIR)/$(UEMACS_BINARY)
Signature · 1 line
1.6.3.3
PATCH 11/12 vpnc: pass CFLAGS at link time
In reply to Thomas PetazzoniExternal toolchain builds absolutely require the --sysroot option to be passed at link time.
--- package/vpnc/vpnc.mk | 4 ++-- 1 files changed, 2 insertions(+), 2 deletions(-)
package/vpnc/vpnc.mk +2 -2
@@ -26,7 +26,7 @@ $(VPNC_BINARY): $(VPNC_DIR)/.unpacked rm -f $@ $(MAKE) $(TARGET_CONFIGURE_OPTS) INCLUDE=$(STAGING_DIR)/usr/include \ CFLAGS="$(TARGET_CFLAGS)" \ - LDFLAGS+=-lgcrypt LDFLAGS+=-lgpg-error \ + LDFLAGS+=-lgcrypt LDFLAGS+=-lgpg-error LDFLAGS+="$(TARGET_CFLAGS)" \ CC="$(TARGET_CC)" -C $(VPNC_DIR) $(VPNC_TARGET_BINARY): $(VPNC_BINARY)
@@ -38,7 +38,7 @@ $(VPNC_TARGET_BINARY): $(VPNC_BINARY) MANDIR=/usr/share/man \ VERSION=$(VPNC_VERSION) \ INCLUDE=$(STAGING_DIR)/usr/include \ - LDFLAGS="-lgcrypt -lgpg-error" \ + LDFLAGS="-lgcrypt -lgpg-error $(TARGET_CFLAGS)" \ -C $(VPNC_DIR) install $(STRIPCMD) $(STRIP_STRIP_UNNEEDED) $(VPNC_TARGET_BINARY)
Signature · 1 line
1.6.3.3
PATCH 12/12 netsnmp: disable build with external toolchain
In reply to Thomas PetazzoniThere is a build problem with netsnmp on external toolchain that isn't easy to fix: libtool strips the --sysroot that we are passing it, which make the link step fail.
In preparation for the release, just make sure this package isn't visible to external toolchain users. Of course, on the long run, we should fix this.
--- package/netsnmp/Config.in | 3 +++ 1 files changed, 3 insertions(+), 0 deletions(-)
package/netsnmp/Config.in +3 -0
@@ -1,5 +1,8 @@ config BR2_PACKAGE_NETSNMP bool "netsnmp" + # Build with external toolchains is broken, libtool strips + # --sysroot at link time + depends on !BR2_TOOLCHAIN_EXTERNAL help Suite of applications used to implement SNMP v1, SNMP v2c, and SNMP v3 using both IPv4 and IPv6.
Signature · 1 line
1.6.3.3
PATCH 12/12 netsnmp: disable build with external toolchain
Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line
"Thomas" == Thomas Petazzoni <[email protected]> writes:
Thomas> There is a build problem with netsnmp on external toolchain that isn't Thomas> easy to fix: libtool strips the --sysroot that we are passing it, Thomas> which make the link step fail.
Thomas> In preparation for the release, just make sure this package isn't Thomas> visible to external toolchain users. Of course, on the long run, we Thomas> should fix this.
Hmm, yeah. We'll need to do the same for BR2_PACKAGE_QUAGGA_SNMP then.
Maybe a comment "netsnmp is broken with external toolchain" would be good.
Signature · 1 line
Bye, Peter Korsgaard
Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line
"Thomas" == Thomas Petazzoni <[email protected]> writes:
Thomas> The following changes since commit 5e25bc482134bee62e3cc9d55af694aaa977d5b5:
Thomas> Thomas Petazzoni (1):
Thomas> external toolchain: check BR2_INSTALL_LIBSTDCPP
Thomas> are available in the git repository at:
Thomas> git://git.busybox.net/~tpetazzoni/git/buildroot 2010-05-15-fixes
Thomas> Thomas Petazzoni (12):
Thomas> iconv: disallow selection when BR2_ENABLE_LOCALE is enabled Thomas> lmbench: needs BR2_INET_RPC Thomas> startup-notification: add dependency on xlib_libX11 Thomas> ntp: fix build failure Thomas> samba: only require libiconv if !BR2_ENABLE_LOCALE Thomas> pango: run pango-querymodules on the target instead of on the host Thomas> cairo: remove host variant Thomas> mutt: fix build and convert to autotools Thomas> lvm2: do not override CC and similar variables Thomas> uemacs: don't forget CFLAGS at link time Thomas> vpnc: pass CFLAGS at link time Thomas> netsnmp: disable build with external toolchain
I wanted to pull this, but that's not what I got:
From git://git.busybox.net/~tpetazzoni/git/buildroot
* branch 2010-05-15-fixes -> FETCH_HEAD
Removing package/mutt/mutt-1.5.16-makedoc-hostcc.patch Merge made by recursive.
package/avahi/Config.in | 5 +- package/avahi/avahi.mk | 2 +- package/cairo/cairo.mk | 3 - package/gettext/Config.in | 2 + package/gmpc/Config.in | 4 +- package/gmpc/gmpc.mk | 3 +- package/grep/Config.in | 4 +- package/grep/grep.mk | 2 +- package/hal/Config.in | 4 +- package/hal/hal.mk | 2 +- package/libcap/libcap.mk | 4 +- ...cgicc-3.2.9-disable-documentation-option.patch} | 0 package/libcgicc/libcgicc.mk | 2 + package/libglib2/Config.in | 4 +- package/libglib2/libglib2.mk | 2 +- package/libiconv/Config.in | 1 + package/libidn/libidn.mk | 2 +- package/libsoup/libsoup.mk | 2 +- package/lmbench/Config.in | 2 + package/ltrace/ltrace.mk | 3 +- package/lvm2/lvm2.mk | 2 +- package/make/Config.in | 4 +- package/make/make.mk | 2 +- package/multimedia/libmpd/Config.in | 2 - package/mutt/mutt-1.5.16-makedoc-hostcc.patch | 32 --------- package/mutt/mutt-1.5.17+20080114-nodoc.patch | 21 ++++++ package/mutt/mutt.mk | 73 +++----------------- package/netsnmp/Config.in | 3 + package/ntp/ntp-4.2.4_p5-nano.patch | 36 ++++++++++ package/pango/S25pango | 20 ++++++ package/pango/pango.mk | 18 +---- package/php/Config.ext | 2 +- package/php/php.mk | 2 +- package/psmisc/Config.in | 4 +- package/psmisc/psmisc.mk | 8 +-- package/samba/Config.in | 2 +- package/samba/samba.mk | 2 +- package/sshfs/Config.in | 4 +- package/sshfs/sshfs.mk | 2 +- package/startup-notification/Config.in | 2 + .../startup-notification/startup-notification.mk | 1 + package/uemacs/uemacs.mk | 2 +- package/util-linux/Config.in | 4 +- package/util-linux/util-linux.mk | 4 +- package/vpnc/vpnc.mk | 4 +- toolchain/Config.in.2 | 6 ++ toolchain/external-toolchain/ext-tool.mk | 12 +++ 47 files changed, 164 insertions(+), 163 deletions(-)
E.G. the branch also contained your older pull request, including the avahi ipv6 dependency that I NACKed.
Could you fix that and send a new pull request, please?
Signature · 1 line
Bye, Peter Korsgaard
Quoted 6 lines · On Tue, 18 May 2010 13:56:51 +0200 Peter Korsgaard <[email protected]> wrote:
I wanted to pull this, but that's not what I got:
E.G. the branch also contained your older pull request, including the avahi ipv6 dependency that I NACKed.
Could you fix that and send a new pull request, please?
Argh, yes, of course. In fact, it's because I started my second branch on top of the first, because some of the fixes of the second branch depend on fixes on the first branch.
The easiest is probably that I merge both branches, get rid of the unwanted Avahi IPv6 fix, and send a new pull request for the whole thing. Would that be ok for you ?
Thomas
Signature · 4 lines
Thomas Petazzoni, Free Electrons Kernel, drivers, real-time and embedded Linux development, consulting, training and support. http://free-electrons.com
Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line
"Thomas" == Thomas Petazzoni <[email protected]> writes:
Hi,
Quoted 1 line
Quoted 1 line
Could you fix that and send a new pull request, please?
Thomas> Argh, yes, of course. In fact, it's because I started my second branch Thomas> on top of the first, because some of the fixes of the second branch Thomas> depend on fixes on the first branch.
Thomas> The easiest is probably that I merge both branches, get rid of the Thomas> unwanted Avahi IPv6 fix, and send a new pull request for the whole Thomas> thing. Would that be ok for you ?
Yes, that would be perfect - Thanks.
It would be even better if you fix the minor comments I had while you're at it, but I can do that as well otherwise.
Signature · 1 line
Bye, Peter Korsgaard