Buildroot Archive on lore.kernel.org
 help / color / mirror / Atom feed
* [Buildroot] [PATCH 1/2] toolchain/external: only copt the pthread lib if needed
From: Yann E. MORIN @ 2010-05-28 21:23 UTC (permalink / raw)
  To: buildroot

From: Yann E. MORIN <yann.morin.1998@anciens.enib.fr>

If threads are disabled, do not try to copy the libpthread.so from the
external library.

It is still expected that the BR configuration matches the external
toolchain setup, and no check is done to enforce that.
---
 toolchain/external-toolchain/ext-tool.mk |    6 +++++-
 1 files changed, 5 insertions(+), 1 deletions(-)

diff --git a/toolchain/external-toolchain/ext-tool.mk b/toolchain/external-toolchain/ext-tool.mk
index bb4809c..e858edb 100644
--- a/toolchain/external-toolchain/ext-tool.mk
+++ b/toolchain/external-toolchain/ext-tool.mk
@@ -251,7 +251,7 @@ check_cross_compiler_exists = \
 
 uclibc: dependencies $(STAMP_DIR)/ext-toolchain-installed
 
-EXTERNAL_LIBS=libc.so libcrypt.so libdl.so libgcc_s.so libm.so libnsl.so libpthread.so libresolv.so librt.so libutil.so
+EXTERNAL_LIBS=libc.so libcrypt.so libdl.so libgcc_s.so libm.so libnsl.so libresolv.so librt.so libutil.so
 ifeq ($(BR2_TOOLCHAIN_EXTERNAL_UCLIBC),y)
 EXTERNAL_LIBS+=ld-uClibc.so
 else
@@ -262,6 +262,10 @@ ifeq ($(BR2_INSTALL_LIBSTDCPP),y)
 EXTERNAL_LIBS+=libstdc++.so
 endif
 
+ifneq ($(BR2_PTHREADS_NONE),y)
+EXTERNAL_LIBS+=libpthread.so
+endif # ! no threads
+
 # SYSROOT_DIR selection. We first try the -print-sysroot option,
 # available in gcc 4.4.x and in some Codesourcery toolchains. If this
 # option is not available, we fallback to the value of --with-sysroot

^ permalink raw reply related

* [Buildroot] [PATCH 1/2] Remove locale dependency from xserver
From: Peter Korsgaard @ 2010-05-28 20:48 UTC (permalink / raw)
  To: buildroot
In-Reply-To: <20100528145834.11609.86446.stgit@localhost.localdomain>

>>>>> "Paulius" == Paulius Zaleckas <paulius.zaleckas@gmail.com> writes:

 Paulius> XServer compiles just fine without locale support
 Paulius> Signed-off-by: Paulius Zaleckas <paulius.zaleckas@gmail.com>

Committed both, thanks!

-- 
Bye, Peter Korsgaard

^ permalink raw reply

* [Buildroot] [git commit master] xserver_xorg-server: Use $(MAKE1) for make install
From: Peter Korsgaard @ 2010-05-28 20:47 UTC (permalink / raw)
  To: buildroot


commit: http://git.buildroot.net/buildroot/commit/?id=b7acacaf6e84ec05ed76dd419e645cfbd6683f7a
branch: http://git.buildroot.net/buildroot/commit/?id=refs/heads/master

Xserver installation fails with parallel make jobs.

Installation to staging is not needed. So just remove it
instead of converting to $(MAKE1).

Signed-off-by: Paulius Zaleckas <paulius.zaleckas@gmail.com>
Signed-off-by: Peter Korsgaard <jacmet@sunsite.dk>
---
 .../xserver_xorg-server/xserver_xorg-server.mk     |    4 +++-
 1 files changed, 3 insertions(+), 1 deletions(-)

diff --git a/package/x11r7/xserver_xorg-server/xserver_xorg-server.mk b/package/x11r7/xserver_xorg-server/xserver_xorg-server.mk
index 6c6017b..a49c180 100644
--- a/package/x11r7/xserver_xorg-server/xserver_xorg-server.mk
+++ b/package/x11r7/xserver_xorg-server/xserver_xorg-server.mk
@@ -9,7 +9,6 @@ XSERVER_XORG_SERVER_SOURCE = xorg-server-$(XSERVER_XORG_SERVER_VERSION).tar.bz2
 XSERVER_XORG_SERVER_SITE = http://xorg.freedesktop.org/releases/individual/xserver
 XSERVER_XORG_SERVER_AUTORECONF = NO
 XSERVER_XORG_SERVER_LIBTOOL_PATCH = NO
-XSERVER_XORG_SERVER_INSTALL_STAGING = YES
 XSERVER_XORG_SERVER_USE_CONFIG_CACHE = NO # overrides CFLAGS
 XSERVER_XORG_SERVER_INSTALL_STAGING_OPT = DESTDIR=$(STAGING_DIR) install install-data
 
@@ -137,5 +136,8 @@ ifneq ($(BR2_PACKAGE_XLIB_LIBDMX),y)
 XSERVER_XORG_SERVER_CONF_OPT += --disable-dmx
 endif
 
+define XSERVER_XORG_SERVER_INSTALL_TARGET_CMDS
+	$(MAKE1) $(XSERVER_XORG_SERVER_INSTALL_TARGET_OPT) -C $(@D)
+endef
 
 $(eval $(call AUTOTARGETS,package/x11r7,xserver_xorg-server))
-- 
1.6.3.3

^ permalink raw reply related

* [Buildroot] [git commit master] Remove locale dependency from xserver
From: Peter Korsgaard @ 2010-05-28 20:47 UTC (permalink / raw)
  To: buildroot


commit: http://git.buildroot.net/buildroot/commit/?id=b578f670d153a04cc410e0696013e9c606f38439
branch: http://git.buildroot.net/buildroot/commit/?id=refs/heads/master

XServer compiles just fine without locale support

Signed-off-by: Paulius Zaleckas <paulius.zaleckas@gmail.com>
Signed-off-by: Peter Korsgaard <jacmet@sunsite.dk>
---
 package/x11r7/Config.in |    7 +++----
 1 files changed, 3 insertions(+), 4 deletions(-)

diff --git a/package/x11r7/Config.in b/package/x11r7/Config.in
index 55524a0..b93cc90 100644
--- a/package/x11r7/Config.in
+++ b/package/x11r7/Config.in
@@ -19,11 +19,10 @@ choice
 
 config BR2_PACKAGE_XSERVER_xorg
 	bool "Modular X.org"
-	# depending on BR2_ENABLE_LOCALE gives a recursion error with dbus
-	depends on BR2_LARGEFILE && BR2_INSTALL_LIBSTDCPP # && BR2_ENABLE_LOCALE
+	depends on BR2_LARGEFILE && BR2_INSTALL_LIBSTDCPP
 
-comment 'X.org requires largefile, locale and C++ support'
-	depends on !(BR2_LARGEFILE && BR2_INSTALL_LIBSTDCPP && BR2_ENABLE_LOCALE)
+comment 'X.org requires largefile and C++ support'
+	depends on !(BR2_LARGEFILE && BR2_INSTALL_LIBSTDCPP)
 
 config BR2_PACKAGE_XSERVER_tinyx
 	bool "TinyX"
-- 
1.6.3.3

^ permalink raw reply related

* [Buildroot] Buildroot 2010.05-rc3 released
From: Yann E. MORIN @ 2010-05-28 20:20 UTC (permalink / raw)
  To: buildroot
In-Reply-To: <1274997359.436.32.camel@ubuntu>

Kris, All,

On Thursday 27 May 2010 23:55:59 Microbit_Ubuntu wrote:
> On Thu, 2010-05-27 at 23:15 +0200, Yann E. MORIN wrote:
> > libthread_db.so.1 is dlopen(3)ed by gdbserver, so you have to get it on
> > your target, in the standard libraries search path.
> When gdbserver was bleating about libthread_db.so.1, it was in the staging directory,
> but it wasn't placed on the target rootfs.
> When I manually added it on the target (libthread_db.so.1 I mean), gdbserver worked good as gold then...

I've just submitted a fix for this:
  http://lists.busybox.net/pipermail/buildroot/2010-May/034848.html

Care to test it and report, please?

Regards,
Yann E. MORIN.

-- 
.-----------------.--------------------.------------------.--------------------.
|  Yann E. MORIN  | Real-Time Embedded | /"\ ASCII RIBBON | Erics' conspiracy: |
| +33 662 376 056 | Software  Designer | \ / CAMPAIGN     |  ___               |
| +33 223 225 172 `------------.-------:  X  AGAINST      |  \e/  There is no  |
| http://ymorin.is-a-geek.org/ | _/*\_ | / \ HTML MAIL    |   v   conspiracy.  |
'------------------------------^-------^------------------^--------------------'

^ permalink raw reply

* [Buildroot] [PATCH] external toolchain: also copy the libthread_db.so for gdbserver
From: Yann E. MORIN @ 2010-05-28 20:05 UTC (permalink / raw)
  To: buildroot

From: Yann E. MORIN <yann.morin.1998@anciens.enib.fr>

gdbserver dlopen(3)s libthread_db.so at runtime, so there is no
dependency on it (does not appear as being (NEEDED)).

Copy libthread_db.so from external toolchain when gdbserver is enbled.

Signed-off-by: "Yann E. MORIN" <yann.morin.1998@anciens.enib.fr>
---
 toolchain/external-toolchain/ext-tool.mk |    4 ++++
 1 files changed, 4 insertions(+), 0 deletions(-)

diff --git a/toolchain/external-toolchain/ext-tool.mk b/toolchain/external-toolchain/ext-tool.mk
index bb4809c..ae9dfad 100644
--- a/toolchain/external-toolchain/ext-tool.mk
+++ b/toolchain/external-toolchain/ext-tool.mk
@@ -262,6 +262,10 @@ ifeq ($(BR2_INSTALL_LIBSTDCPP),y)
 EXTERNAL_LIBS+=libstdc++.so
 endif
 
+ifeq ($(BR2_PACKAGE_GDB_SERVER),y)
+EXTERNAL_LIBS+=libthread_db.so
+endif
+
 # SYSROOT_DIR selection. We first try the -print-sysroot option,
 # available in gcc 4.4.x and in some Codesourcery toolchains. If this
 # option is not available, we fallback to the value of --with-sysroot

^ permalink raw reply related

* [Buildroot] [Bug 1843] Fix libcap build failure
From: bugzilla at busybox.net @ 2010-05-28 19:48 UTC (permalink / raw)
  To: buildroot
In-Reply-To: <bug-1843-163@https.bugs.busybox.net/>

https://bugs.busybox.net/show_bug.cgi?id=1843

Peter Korsgaard <jacmet@uclibc.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |FIXED

--- Comment #1 from Peter Korsgaard <jacmet@uclibc.org>  ---
Committed, thanks

-- 
Configure bugmail: https://bugs.busybox.net/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

^ permalink raw reply

* [Buildroot] [git commit master] libcap: fix build issue when host pam is around
From: Peter Korsgaard @ 2010-05-28 19:47 UTC (permalink / raw)
  To: buildroot


commit: http://git.buildroot.net/buildroot/commit/?id=2816c33c3d802f002bdb53d41f249eb5571a7cb9
branch: http://git.buildroot.net/buildroot/commit/?id=refs/heads/master

Closes #1843

Signed-off-by: Gustavo Zacarias <gustavo@zacarias.com.ar>
Signed-off-by: Peter Korsgaard <jacmet@sunsite.dk>
---
 CHANGES                                       |    6 +++-
 package/libcap/libcap-2.19-build-system.patch |   43 +++++++++++--------------
 2 files changed, 24 insertions(+), 25 deletions(-)

diff --git a/CHANGES b/CHANGES
index 44ffa75..9eb6f81 100644
--- a/CHANGES
+++ b/CHANGES
@@ -2,7 +2,11 @@
 
 	Fixes all over the tree.
 
-	Updated/fixed packages: coreutils, hal
+	Updated/fixed packages: coreutils, hal, libcap
+
+	Issues resolved (http://bugs.uclibc.org):
+
+	#1843: Fix libcap build failure
 
 2010.05-rc3, Released May 27th, 2010:
 
diff --git a/package/libcap/libcap-2.19-build-system.patch b/package/libcap/libcap-2.19-build-system.patch
index 42a5518..51b73af 100644
--- a/package/libcap/libcap-2.19-build-system.patch
+++ b/package/libcap/libcap-2.19-build-system.patch
@@ -1,13 +1,19 @@
----
- Make.Rules      |   15 ++++++++-------
- libcap/Makefile |    2 +-
- 2 files changed, 9 insertions(+), 8 deletions(-)
-
-Index: libcap-2.19/Make.Rules
-===================================================================
---- libcap-2.19.orig/Make.Rules
-+++ libcap-2.19/Make.Rules
-@@ -45,24 +45,25 @@
+diff -Nura libcap-2.19.orig/libcap/Makefile libcap-2.19/libcap/Makefile
+--- libcap-2.19.orig/libcap/Makefile	2008-12-04 05:03:12.000000000 -0200
++++ libcap-2.19/libcap/Makefile	2010-05-28 14:32:06.290297694 -0300
+@@ -33,7 +33,7 @@
+ endif
+ 
+ _makenames: _makenames.c cap_names.list.h
+-	$(BUILD_CC) $(BUILD_CFLAGS) $< -o $@
++	$(CC_FOR_BUILD) $(CFLAGS_FOR_BUILD) $< -o $@
+ 
+ cap_names.h: _makenames
+ 	./_makenames > cap_names.h
+diff -Nura libcap-2.19.orig/Make.Rules libcap-2.19/Make.Rules
+--- libcap-2.19.orig/Make.Rules	2010-01-13 23:04:58.000000000 -0300
++++ libcap-2.19/Make.Rules	2010-05-28 14:32:31.802296186 -0300
+@@ -45,25 +45,26 @@
  KERNEL_HEADERS := $(topdir)/libcap/include
  IPATH += -I$(topdir)/libcap/include -I$(KERNEL_HEADERS)
  
@@ -36,20 +42,9 @@ Index: libcap-2.19/Make.Rules
  INCS=$(topdir)/libcap/include/sys/capability.h
  LDFLAGS += -L$(topdir)/libcap
  CFLAGS += -Dlinux $(WARNINGS) $(DEBUG) $(IPATH)
+-PAM_CAP := $(shell if [ -f /usr/include/security/pam_modules.h ]; then echo yes ; else echo no ; fi)
 +CFLAGS_FOR_BUILD += $(IPATH)
- PAM_CAP := $(shell if [ -f /usr/include/security/pam_modules.h ]; then echo yes ; else echo no ; fi)
++PAM_CAP := no
  INDENT := $(shell if [ -n "$(which indent 2>/dev/null)" ]; then echo "| indent -kr" ; fi)
  DYNAMIC := $(shell if [ ! -d "$(topdir)/.git" ]; then echo yes; fi)
-Index: libcap-2.19/libcap/Makefile
-===================================================================
---- libcap-2.19.orig/libcap/Makefile
-+++ libcap-2.19/libcap/Makefile
-@@ -33,7 +33,7 @@
- endif
- 
- _makenames: _makenames.c cap_names.list.h
--	$(BUILD_CC) $(BUILD_CFLAGS) $< -o $@
-+	$(CC_FOR_BUILD) $(CFLAGS_FOR_BUILD) $< -o $@
- 
- cap_names.h: _makenames
- 	./_makenames > cap_names.h
+ LIBATTR := yes
-- 
1.6.3.3

^ permalink raw reply related

* [Buildroot] git access via http
From: Peter Korsgaard @ 2010-05-28 18:54 UTC (permalink / raw)
  To: buildroot
In-Reply-To: <201005281824.09647.yann.morin.1998@anciens.enib.fr>

>>>>> "Yann" == Yann E MORIN <yann.morin.1998@anciens.enib.fr> writes:

 Yann> Peter, Phil, All,
 Yann> On Friday 28 May 2010 12:19:10 Peter Korsgaard wrote:
 >> Hmm, seems like git-over-http was never set up by the osuosl.org
 >> people. Is this important for you? The native git protocol is a lot
 >> better.

 Yann> Yes, but in some companies, the policy is to acces the "internet" only
 Yann> through http/https proxies. Some even disallow the 'CONNECT' command to
 Yann> anything else but http/https/ftp, and do not offer socks proxies. These
 Yann> rule out anything but plain http/ftp.

 Yann> So, having buildroot's git available through http would be nice, IMHO.

Ok, I'll ask the osuosl guys to set it up.

-- 
Bye, Peter Korsgaard

^ permalink raw reply

* [Buildroot] [Bug 1843] New: Fix libcap build failure
From: bugzilla at busybox.net @ 2010-05-28 17:43 UTC (permalink / raw)
  To: buildroot

https://bugs.busybox.net/show_bug.cgi?id=1843

              Host: i686-linux
            Target: arm-softfloat-linux-uclibcgnueabi
           Summary: Fix libcap build failure
           Product: buildroot
           Version: unspecified
          Platform: PC
        OS/Version: Linux
            Status: NEW
          Severity: normal
          Priority: P5
         Component: Other
        AssignedTo: unassigned at buildroot.uclibc.org
        ReportedBy: gustavo at zacarias.com.ar
                CC: buildroot at uclibc.org
   Estimated Hours: 0.0


Created attachment 1777
  --> https://bugs.busybox.net/attachment.cgi?id=1777
Forcibly disable PAM support in libcap

The libcap package fails to build because it tries to detect PAM presence via
shell file detection against the host includes.
Fix it by disabling PAM in Make.Rules since we don't have PAM support in
buildroot at the moment.

-- 
Configure bugmail: https://bugs.busybox.net/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

^ permalink raw reply

* [Buildroot] git access via http
From: Yann E. MORIN @ 2010-05-28 16:24 UTC (permalink / raw)
  To: buildroot
In-Reply-To: <87typstjkx.fsf@macbook.be.48ers.dk>

Peter, Phil, All,

On Friday 28 May 2010 12:19:10 Peter Korsgaard wrote:
> Hmm, seems like git-over-http was never set up by the osuosl.org
> people. Is this important for you? The native git protocol is a lot
> better.

Yes, but in some companies, the policy is to acces the "internet" only
through http/https proxies. Some even disallow the 'CONNECT' command to
anything else but http/https/ftp, and do not offer socks proxies. These
rule out anything but plain http/ftp.

So, having buildroot's git available through http would be nice, IMHO.

Regards,
Yann E. MORIN.

-- 
.-----------------.--------------------.------------------.--------------------.
|  Yann E. MORIN  | Real-Time Embedded | /"\ ASCII RIBBON | Erics' conspiracy: |
| +33 662 376 056 | Software  Designer | \ / CAMPAIGN     |  ___               |
| +33 223 225 172 `------------.-------:  X  AGAINST      |  \e/  There is no  |
| http://ymorin.is-a-geek.org/ | _/*\_ | / \ HTML MAIL    |   v   conspiracy.  |
'------------------------------^-------^------------------^--------------------'

^ permalink raw reply

* [Buildroot] [PATCH 2/2] Use only $(MAKE1) when installing Xserver
From: Paulius Zaleckas @ 2010-05-28 14:58 UTC (permalink / raw)
  To: buildroot
In-Reply-To: <20100528145834.11609.86446.stgit@localhost.localdomain>

Xserver installation fails with parallel make jobs.

Installation to staging is not needed. So jus remove it
instead of converting to $(MAKE1).

Signed-off-by: Paulius Zaleckas <paulius.zaleckas@gmail.com>
---

 .../xserver_xorg-server/xserver_xorg-server.mk     |    4 +++-
 1 files changed, 3 insertions(+), 1 deletions(-)

diff --git a/package/x11r7/xserver_xorg-server/xserver_xorg-server.mk b/package/x11r7/xserver_xorg-server/xserver_xorg-server.mk
index 6c6017b..a49c180 100644
--- a/package/x11r7/xserver_xorg-server/xserver_xorg-server.mk
+++ b/package/x11r7/xserver_xorg-server/xserver_xorg-server.mk
@@ -9,7 +9,6 @@ XSERVER_XORG_SERVER_SOURCE = xorg-server-$(XSERVER_XORG_SERVER_VERSION).tar.bz2
 XSERVER_XORG_SERVER_SITE = http://xorg.freedesktop.org/releases/individual/xserver
 XSERVER_XORG_SERVER_AUTORECONF = NO
 XSERVER_XORG_SERVER_LIBTOOL_PATCH = NO
-XSERVER_XORG_SERVER_INSTALL_STAGING = YES
 XSERVER_XORG_SERVER_USE_CONFIG_CACHE = NO # overrides CFLAGS
 XSERVER_XORG_SERVER_INSTALL_STAGING_OPT = DESTDIR=$(STAGING_DIR) install install-data
 
@@ -137,5 +136,8 @@ ifneq ($(BR2_PACKAGE_XLIB_LIBDMX),y)
 XSERVER_XORG_SERVER_CONF_OPT += --disable-dmx
 endif
 
+define XSERVER_XORG_SERVER_INSTALL_TARGET_CMDS
+	$(MAKE1) $(XSERVER_XORG_SERVER_INSTALL_TARGET_OPT) -C $(@D)
+endef
 
 $(eval $(call AUTOTARGETS,package/x11r7,xserver_xorg-server))

^ permalink raw reply related

* [Buildroot] [PATCH 1/2] Remove locale dependency from xserver
From: Paulius Zaleckas @ 2010-05-28 14:58 UTC (permalink / raw)
  To: buildroot

XServer compiles just fine without locale support

Signed-off-by: Paulius Zaleckas <paulius.zaleckas@gmail.com>
---

 package/x11r7/Config.in |    7 +++----
 1 files changed, 3 insertions(+), 4 deletions(-)

diff --git a/package/x11r7/Config.in b/package/x11r7/Config.in
index 55524a0..b93cc90 100644
--- a/package/x11r7/Config.in
+++ b/package/x11r7/Config.in
@@ -19,11 +19,10 @@ choice
 
 config BR2_PACKAGE_XSERVER_xorg
 	bool "Modular X.org"
-	# depending on BR2_ENABLE_LOCALE gives a recursion error with dbus
-	depends on BR2_LARGEFILE && BR2_INSTALL_LIBSTDCPP # && BR2_ENABLE_LOCALE
+	depends on BR2_LARGEFILE && BR2_INSTALL_LIBSTDCPP
 
-comment 'X.org requires largefile, locale and C++ support'
-	depends on !(BR2_LARGEFILE && BR2_INSTALL_LIBSTDCPP && BR2_ENABLE_LOCALE)
+comment 'X.org requires largefile and C++ support'
+	depends on !(BR2_LARGEFILE && BR2_INSTALL_LIBSTDCPP)
 
 config BR2_PACKAGE_XSERVER_tinyx
 	bool "TinyX"

^ permalink raw reply related

* [Buildroot] Buildroot 2010.05-rc3 released uClibc 0.9.30.1 patches missing ?
From: Peter Korsgaard @ 2010-05-28 12:44 UTC (permalink / raw)
  To: buildroot
In-Reply-To: <1274994677.436.2.camel@ubuntu>

>>>>> "Microbit" == Microbit Ubuntu <microbit@virginbroadband.com.au> writes:

 Microbit> Hallo peter,
 Microbit> Hoe gaat het ? :-)

Goed (I'm actually from Denmark rather than Belgium).

 Microbit> I don;t know if I'm off base here, but I notice that rc3 does
 Microbit> not have the patches for uClibc 0.9.30.1 at toolchain/uClibc
 Microbit> folder in the tarball ?????

Indeed, as we don't have support for the obsolete 0.9.30.1 version
anymore (the closest is 0.9.30.3).

-- 
Bye, Peter Korsgaard

^ permalink raw reply

* [Buildroot] git access via http
From: Peter Korsgaard @ 2010-05-28 10:19 UTC (permalink / raw)
  To: buildroot
In-Reply-To: <AB12B32E73474741A2C5361C433A44DE01CD4EEB@rte-ben-exch.RTE.ADWIN.RENESAS.COM>

>>>>> "Phil" == Phil Edworthy <Phil.Edworthy@renesas.com> writes:

 Phil> Hi,
 Phil> I just tried cloning the Buildroot repo via http but git just sits there
 Phil> after printing out "Initialized empty Git repository in
 Phil> /home/user/buildroot/.git/"

 Phil> I'm using git 1.6.6 and have no problems accessing github repos via http.
 Phil> Any ideas why this doesn't work?

Hmm, seems like git-over-http was never set up by the osuosl.org
people. Is this important for you? The native git protocol is a lot
better.

-- 
Bye, Peter Korsgaard

^ permalink raw reply

* [Buildroot] [PATCH] coreutils: add 'join'
From: Peter Korsgaard @ 2010-05-28 10:07 UTC (permalink / raw)
  To: buildroot
In-Reply-To: <1274991984-6652-1-git-send-email-bjorn.forsman@gmail.com>

>>>>> "Bj?rn" == Bj?rn Forsman <bjorn.forsman@gmail.com> writes:

 Bj?rn> coreutils: add 'join' to the list of files that are copied to target
 Bj?rn> rootfs.
 Bj?rn> ---
 Bj?rn>  package/coreutils/coreutils.mk |    3 ++-
 Bj?rn>  1 files changed, 2 insertions(+), 1 deletions(-)

Committed, thanks.

-- 
Bye, Peter Korsgaard

^ permalink raw reply

* [Buildroot] [git commit master] linux: zImage target no longer available on x86
From: Peter Korsgaard @ 2010-05-28 10:06 UTC (permalink / raw)
  To: buildroot


commit: http://git.buildroot.net/buildroot/commit/?id=98de259aee970b70f06f7e98253fbf20419057a0
branch: http://git.buildroot.net/buildroot/commit/?id=refs/heads/master

The legacy zImage target for x86 was removed from the kernel in 2.6.30,
and we state in Config.in that we'll use bzImage if BR2_PACKAGE_LINUX_FORMAT
isn't set, so ensure we do so for x86.

Signed-off-by: Peter Korsgaard <jacmet@sunsite.dk>
---
 target/linux/Makefile.in          |    7 ++++++-
 target/linux/Makefile.in.advanced |    5 +++++
 2 files changed, 11 insertions(+), 1 deletions(-)

diff --git a/target/linux/Makefile.in b/target/linux/Makefile.in
index 6983d25..f019824 100644
--- a/target/linux/Makefile.in
+++ b/target/linux/Makefile.in
@@ -10,7 +10,7 @@ ifeq ($(DOWNLOAD_LINUX26_VERSION),)
 # User did not define linux version, try using headers
 ifeq ($(LINUX_HEADERS_VERSION),)
 # We did not have headers (this is for real???)
-# Version of linuc before patches
+# Version of linux before patches
 DOWNLOAD_LINUX26_VERSION=2.6.22.1
 # Version of Linux after applying any patches
 LINUX26_VERSION=2.6.22.1
@@ -44,9 +44,14 @@ ifndef LINUX26_FORMAT
 ifneq ($(call qstrip,$(BR2_PACKAGE_LINUX_FORMAT)),)
 LINUX26_FORMAT=$(call qstrip,$(BR2_PACKAGE_LINUX_FORMAT))
 else
+ifneq ($(filter i386 x86-64,$(KERNEL_ARCH)),)
+# zImage target no longer available on x86 since 2.6.30
+LINUX26_FORMAT=bzImage
+else
 LINUX26_FORMAT=zImage
 endif
 endif
+endif
 
 # Has to be set by the target/device
 ifndef LINUX26_BINLOC
diff --git a/target/linux/Makefile.in.advanced b/target/linux/Makefile.in.advanced
index 9e495f3..86d94ca 100644
--- a/target/linux/Makefile.in.advanced
+++ b/target/linux/Makefile.in.advanced
@@ -111,9 +111,14 @@ ifndef LINUX26_FORMAT
 ifneq ($(call qstrip,$(BR2_PACKAGE_LINUX_FORMAT)),)
 LINUX26_FORMAT:=$(call qstrip,$(BR2_PACKAGE_LINUX_FORMAT))
 else
+ifneq ($(filter i386 x86-64,$(KERNEL_ARCH)),)
+# zImage target no longer available on x86 since 2.6.30
+LINUX26_FORMAT=bzImage
+else
 LINUX26_FORMAT:=zImage
 endif
 endif
+endif
 
 # -----------------------------------------------------------------------------
 # Has to be set by the target/device
-- 
1.6.3.3

^ permalink raw reply related

* [Buildroot] [git commit master] coreutils: add 'join'
From: Bjørn Forsman @ 2010-05-28 10:06 UTC (permalink / raw)
  To: buildroot


commit: http://git.buildroot.net/buildroot/commit/?id=c0b3dc20059a29bc36c699f5a71e038a44cf7544
branch: http://git.buildroot.net/buildroot/commit/?id=refs/heads/master

coreutils: add 'join' to the list of files that are copied to target
rootfs.

Signed-off-by: Bj?rn Forsman <bjorn.forsman@gmail.com>
Signed-off-by: Peter Korsgaard <jacmet@sunsite.dk>
---
 CHANGES                        |    2 +-
 package/coreutils/coreutils.mk |    3 ++-
 2 files changed, 3 insertions(+), 2 deletions(-)

diff --git a/CHANGES b/CHANGES
index f6a4ecd..44ffa75 100644
--- a/CHANGES
+++ b/CHANGES
@@ -2,7 +2,7 @@
 
 	Fixes all over the tree.
 
-	Updated/fixed packages: hal
+	Updated/fixed packages: coreutils, hal
 
 2010.05-rc3, Released May 27th, 2010:
 
diff --git a/package/coreutils/coreutils.mk b/package/coreutils/coreutils.mk
index 2081994..e2f7bc5 100644
--- a/package/coreutils/coreutils.mk
+++ b/package/coreutils/coreutils.mk
@@ -12,7 +12,8 @@ COREUTILS_DIR:=$(BUILD_DIR)/coreutils-$(COREUTILS_VERSION)
 COREUTILS_BINARY:=src/vdir
 COREUTILS_TARGET_BINARY:=bin/vdir
 BIN_PROGS:=cat chgrp chmod chown cp date dd df dir echo false hostname \
-	ln ls mkdir mknod mv pwd rm rmdir vdir sleep stty sync touch true uname
+	ln ls mkdir mknod mv pwd rm rmdir vdir sleep stty sync touch true \
+	uname join
 
 $(DL_DIR)/$(COREUTILS_SOURCE):
 	 $(call DOWNLOAD,$(COREUTILS_SITE),$(COREUTILS_SOURCE))
-- 
1.6.3.3

^ permalink raw reply related

* [Buildroot] Buildroot 2010.05-rc3 released
From: Microbit_Ubuntu @ 2010-05-28  0:49 UTC (permalink / raw)
  To: buildroot
In-Reply-To: <201005272315.11295.yann.morin.1998@anciens.enib.fr>

Hi Yann,

On Thu, 2010-05-27 at 23:15 +0200, Yann E. MORIN wrote:
> Kris, All,
> 
> On Thursday 27 May 2010 20:58:43 Microbit_Ubuntu wrote:
> [--SNIP--]
> > PS : ct-ng only offers "linuxthreads".
> 
> That's only true if you build a uClibc-based toolchain. If you build a
> glibc- or a eglibc-based toolchain, then crosstool-NG allows you to
> use NPTL.
> 
> So I'll take that you build a uClibc-based toolchain...
> 
> > So are we supposed to use 
> > "stable/old" in BR or "new" (latter, I guess) ????
> 
> That has to match the way you configured uClibc in your toolchain.
> Check the uClibc .config file for that.
> 
> > I seem to have had problems with that - invoking gdbserver prompts :
> > "can't load libthread_db.so.1".
> > This worked fine with internal toolchain... Anyone ?
> 
> libthread_db.so.1 is dlopen(3)ed by gdbserver, so you have to get it on
> your target, in the standard libraries search path.
> 
> Also, there is a known dangling issue wrt cross-gdb + gdbserver.
> See this thread:
>   http://lists.busybox.net/pipermail/buildroot/2010-May/034435.html
> 
> In short, you have to get the cross-gdb *and* the gdbserver from the
> same version of gdb, to be sure they know how to speak to each other.
> So: either get both from crosstool-NG, or get both from buildroot.
> 
> Regards,
> Yann E. MORIN.
> 

I hope the list will forgive me sending to you & list.
Your posts have not reached me at all yet through busybox.net :-(

A follow-up as promised.
I'm rather surprised that the new build still bleats about 'can't load
libthread_db.so.1' when starting gdbserver...
That lib is in ../staging/lib as a symlink to libthread_db-0.9.30.1.so.
There's also a symlink in /staging/usr/lib called libthread_db.so which
points to libthread_db.so.1 - hmm.

As mentioned, the previous rootfs runs fine, however there is no trace
of this lib on the target.

I'm a bit stumped on this one.
I've used linuxthreads (new) in uclibc 0.9.30.1 config.

TIA
Ici avec beaucoup de soucis !!!
(rather than "sans soucis" :-)
-- Kris

^ permalink raw reply

* [Buildroot] Buildroot 2010.05-rc3 released :: 2 build errors - lockfile-progs & gdb
From: Microbit_Ubuntu @ 2010-05-28  0:37 UTC (permalink / raw)
  To: buildroot
In-Reply-To: <877hmpv7km.fsf@macbook.be.48ers.dk>

Hi Peter,

On Thu, 2010-05-27 at 14:43 +0200, Peter Korsgaard wrote:
> Hi,
> 
> Buildroot 2010.05-rc3 is released - Go download it at:
> 
> http://buildroot.net/downloads/buildroot-2010.05-rc3.tar.gz
> 
> or
> 
> http://buildroot.net/downloads/buildroot-2010.05-rc3.tar.bz2
> 
> Or get it from Git:
> 
> git://git.buildroot.net/buildroot
> 
> Please give it a spin and report any problems to the mailing list or bug
> tracker. Many thanks to the people contributing to this release:
> 
> git shortlog -s -n 2010.05_rc2..
>     49  Peter Korsgaard
>     32  Thomas Petazzoni
>      1  Baruch Siach
>      1  Bj?rn Forsman
>      1  H Hartley Sweeten
>      1  Julien Boibessot
>      1  Nick Leverton
>      1  Yann E. MORIN
>      1  cmchao
> 
> Unless any big issues are found, expect the final 2010.05 release by the
> end of the week - So please give it a good test.
> 
> -- 
> Bye, Peter Korsgaard
> _______________________________________________
> buildroot mailing list
> buildroot at busybox.net
> http://lists.busybox.net/mailman/listinfo/buildroot

I still come across 2 build errors with rc3 :

1. lockfile-progs-0.1.11
========================
The generated Makefile causes a build error in :
CFLAGS    := -g -Wall -Werror -O2
I still need to remove the -Werror to resume the build ??

2. gdb symlink
==============
ct-ng generates arm-unknown-linux-uclibcgnueabi where I have
arm-elf-xxx as an alias.
The build stops when the symlink for gdb fails. I need to manually
insert a folder 'arm-unknown-linux-uclibcgnueabi' with 'bin' as a
subfolder under ../output/staging/usr

The error is some such :

install
-c /home/kris/buildroot-2010.05-rc3/output/toolchain/gdbhost-6.8/gdb/gdb /home/kris/x-tools/arm-unknown-linux-uclibcgnueabi/bin/arm-elf-gdb
ln -snf ../../bin/arm-unknown-linux-uclibcgnueabi-gdb \
		/home/kris/buildroot-2010.05-rc3/output/staging/usr/arm-unknown-linux-uclibcgnueabi/bin/gdb
ln: creating symbolic link
`/home/kris/buildroot-2010.05-rc3/output/staging/usr/arm-unknown-linux-uclibcgnueabi/bin/gdb': No such file or directory
make: ***
[/home/kris/x-tools/arm-unknown-linux-uclibcgnueabi/bin/arm-elf-gdb]
Error 1


AFAIK I'm using the tool turnkey manner.
Can you advise ?

-- Kris

^ permalink raw reply

* [Buildroot] Buildroot 2010.05-rc3 released
From: Microbit_Ubuntu @ 2010-05-27 21:55 UTC (permalink / raw)
  To: buildroot
In-Reply-To: <201005272315.11295.yann.morin.1998@anciens.enib.fr>

A propos,

On Thu, 2010-05-27 at 23:15 +0200, Yann E. MORIN wrote:
> Kris, All,
> 
> On Thursday 27 May 2010 20:58:43 Microbit_Ubuntu wrote:
> [--SNIP--]
> > PS : ct-ng only offers "linuxthreads".
> 
> That's only true if you build a uClibc-based toolchain. If you build a
> glibc- or a eglibc-based toolchain, then crosstool-NG allows you to
> use NPTL.
> 
> So I'll take that you build a uClibc-based toolchain...
> 
> > So are we supposed to use 
> > "stable/old" in BR or "new" (latter, I guess) ????
> 
> That has to match the way you configured uClibc in your toolchain.
> Check the uClibc .config file for that.
> 
> > I seem to have had problems with that - invoking gdbserver prompts :
> > "can't load libthread_db.so.1".
> > This worked fine with internal toolchain... Anyone ?
> 
> libthread_db.so.1 is dlopen(3)ed by gdbserver, so you have to get it on
> your target, in the standard libraries search path.
> 
> Also, there is a known dangling issue wrt cross-gdb + gdbserver.
> See this thread:
>   http://lists.busybox.net/pipermail/buildroot/2010-May/034435.html
> 
> In short, you have to get the cross-gdb *and* the gdbserver from the
> same version of gdb, to be sure they know how to speak to each other.
> So: either get both from crosstool-NG, or get both from buildroot.
> 
> Regards,
> Yann E. MORIN.
> 




> libthread_db.so.1 is dlopen(3)ed by gdbserver, so you have to get it on
> your target, in the standard libraries search path.

When gdbserver was bleating about libthread_db.so.1, it was in the staging directory,
but it wasn't placed on the target rootfs.
When I manually added it on the target (libthread_db.so.1 I mean), gdbserver worked good as gold then...

-- Kris

^ permalink raw reply

* [Buildroot] Buildroot 2010.05-rc3 released
From: Microbit_Ubuntu @ 2010-05-27 21:51 UTC (permalink / raw)
  To: buildroot
In-Reply-To: <201005272315.11295.yann.morin.1998@anciens.enib.fr>

Hi Yann, All,

On Thu, 2010-05-27 at 23:15 +0200, Yann E. MORIN wrote:
> Kris, All,
> 
> On Thursday 27 May 2010 20:58:43 Microbit_Ubuntu wrote:
> [--SNIP--]
> > PS : ct-ng only offers "linuxthreads".
> 
> That's only true if you build a uClibc-based toolchain. If you build a
> glibc- or a eglibc-based toolchain, then crosstool-NG allows you to
> use NPTL.
> 
> So I'll take that you build a uClibc-based toolchain...
> 
> > So are we supposed to use 
> > "stable/old" in BR or "new" (latter, I guess) ????
> 
> That has to match the way you configured uClibc in your toolchain.
> Check the uClibc .config file for that.
> 
> > I seem to have had problems with that - invoking gdbserver prompts :
> > "can't load libthread_db.so.1".
> > This worked fine with internal toolchain... Anyone ?
> 
> libthread_db.so.1 is dlopen(3)ed by gdbserver, so you have to get it on
> your target, in the standard libraries search path.
> 
> Also, there is a known dangling issue wrt cross-gdb + gdbserver.
> See this thread:
>   http://lists.busybox.net/pipermail/buildroot/2010-May/034435.html
> 
> In short, you have to get the cross-gdb *and* the gdbserver from the
> same version of gdb, to be sure they know how to speak to each other.
> So: either get both from crosstool-NG, or get both from buildroot.
> 
> Regards,
> Yann E. MORIN.
> 


I've been compiling with both from buildroot at all times, so that
should be OK.
While I was doing a sanity check vis-a-vis (sorry for the missing accent
grave) module inserting (solved), I used an older kernel uImage
(generated when I still used internal BR toolchain).

gdbserver of course is still happy there, but I can't find a trace of
libthread_db.so.1 on my rootfs target. (looked in /lib and /usr/lib).
I do have libpthread.so.0 etc.
I tried with an ltrace but that's not helping much.. (only see
__uClib_main)

I'll compile shortly and keep you posted.

Thanks for the help Yann, it's greatly appreciated !

-- Kris

^ permalink raw reply

* [Buildroot] Buildroot 2010.05-rc3 released uClibc 0.9.30.1 patches missing ?
From: Microbit_Ubuntu @ 2010-05-27 21:32 UTC (permalink / raw)
  To: buildroot
In-Reply-To: <201005272319.54338.yann.morin.1998@anciens.enib.fr>

Bonjour Yann, tout le monde,

On Thu, 2010-05-27 at 23:19 +0200, Yann E. MORIN wrote:
> Kris, All,
> 
> On Thursday 27 May 2010 23:11:17 Microbit_Ubuntu wrote:
> > I don;t know if I'm off base here, but I notice that rc3 does not have
> > the patches for uClibc 0.9.30.1 at toolchain/uClibc folder in the
> > tarball ?????
> 
> Indeed, that because the 0.9.30.x that is used is .3, and .1 is not present
> in the version choice, in the menuconfig. See:
>   toolchain/uClibc/Config.in at 40
> 
> Regards,
> Yann E. MORIN.
> 

Thanks for reviewing.
This might have sounded like a dumb question - as of course uClibc was
compiled by ct-ng...
I haven't yet compiled BR, am about to though. I saw uClibc 0.1.29
patches, so I thought "where's 0.9.30.1 ??".

I mean, if 0.9.30.3 is used, why bother with 0.9.29 patches ???
Oh well, false alarm in any case !!

The *biggest* problem so far I had was the fact that ct-ng-1.6.1 patches
uClibc 0.9.30.1 BUT does NOT apply the ARM sysdep patch.
I kept getting "invalid module format" on my target. A rebuild with
(indeed) "new" linuxthreads enabled in uClibc of the x-tools toolchain
fixed it after manually applying the sysdeps patch !!

I'll still reply to your other post as well.

Au revoir,
Kris

^ permalink raw reply

* [Buildroot] Buildroot 2010.05-rc3 released uClibc 0.9.30.1 patches missing ?
From: Yann E. MORIN @ 2010-05-27 21:19 UTC (permalink / raw)
  To: buildroot
In-Reply-To: <1274994677.436.2.camel@ubuntu>

Kris, All,

On Thursday 27 May 2010 23:11:17 Microbit_Ubuntu wrote:
> I don;t know if I'm off base here, but I notice that rc3 does not have
> the patches for uClibc 0.9.30.1 at toolchain/uClibc folder in the
> tarball ?????

Indeed, that because the 0.9.30.x that is used is .3, and .1 is not present
in the version choice, in the menuconfig. See:
  toolchain/uClibc/Config.in at 40

Regards,
Yann E. MORIN.

-- 
.-----------------.--------------------.------------------.--------------------.
|  Yann E. MORIN  | Real-Time Embedded | /"\ ASCII RIBBON | Erics' conspiracy: |
| +33 662 376 056 | Software  Designer | \ / CAMPAIGN     |  ___               |
| +33 223 225 172 `------------.-------:  X  AGAINST      |  \e/  There is no  |
| http://ymorin.is-a-geek.org/ | _/*\_ | / \ HTML MAIL    |   v   conspiracy.  |
'------------------------------^-------^------------------^--------------------'

^ permalink raw reply

* [Buildroot] Buildroot 2010.05-rc3 released
From: Yann E. MORIN @ 2010-05-27 21:15 UTC (permalink / raw)
  To: buildroot
In-Reply-To: <1274986723.7293.8.camel@ubuntu>

Kris, All,

On Thursday 27 May 2010 20:58:43 Microbit_Ubuntu wrote:
[--SNIP--]
> PS : ct-ng only offers "linuxthreads".

That's only true if you build a uClibc-based toolchain. If you build a
glibc- or a eglibc-based toolchain, then crosstool-NG allows you to
use NPTL.

So I'll take that you build a uClibc-based toolchain...

> So are we supposed to use 
> "stable/old" in BR or "new" (latter, I guess) ????

That has to match the way you configured uClibc in your toolchain.
Check the uClibc .config file for that.

> I seem to have had problems with that - invoking gdbserver prompts :
> "can't load libthread_db.so.1".
> This worked fine with internal toolchain... Anyone ?

libthread_db.so.1 is dlopen(3)ed by gdbserver, so you have to get it on
your target, in the standard libraries search path.

Also, there is a known dangling issue wrt cross-gdb + gdbserver.
See this thread:
  http://lists.busybox.net/pipermail/buildroot/2010-May/034435.html

In short, you have to get the cross-gdb *and* the gdbserver from the
same version of gdb, to be sure they know how to speak to each other.
So: either get both from crosstool-NG, or get both from buildroot.

Regards,
Yann E. MORIN.

-- 
.-----------------.--------------------.------------------.--------------------.
|  Yann E. MORIN  | Real-Time Embedded | /"\ ASCII RIBBON | Erics' conspiracy: |
| +33 662 376 056 | Software  Designer | \ / CAMPAIGN     |  ___               |
| +33 223 225 172 `------------.-------:  X  AGAINST      |  \e/  There is no  |
| http://ymorin.is-a-geek.org/ | _/*\_ | / \ HTML MAIL    |   v   conspiracy.  |
'------------------------------^-------^------------------^--------------------'

^ permalink raw reply


This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox