Skip to contentSkip to conversation

random misc trivia

PATCH In review Bernhard Reutner-Fischer 7 participants 32 messages Feb 8, 2010 – Feb 15, 2010

lore
Share & tools
lore
#0 lore raw

cover random misc trivia

Hi,

A small series of unrelated patches that i could easily isolate. Please apply.

thanks,
Bernhard

#1 lore raw

PATCH 1/12 trim superfluous trailing whitespace

---
 recipes/busybox/busybox.inc |    6 +++---
 1 files changed, 3 insertions(+), 3 deletions(-)
recipes/busybox/busybox.inc +3 -3
index 519fcd6..7f67f47 100644
@@ -55,7 +55,7 @@ FILES_${PN} += "${datadir}/udhcpc"
 INITSCRIPT_PACKAGES = "${PN}-httpd ${PN}-syslog ${PN}-udhcpd"
 INITSCRIPT_NAME_${PN}-httpd = "busybox-httpd"
 INITSCRIPT_NAME_${PN}-syslog = "syslog"
-INITSCRIPT_NAME_${PN}-udhcpd = "busybox-udhcpd" 
+INITSCRIPT_NAME_${PN}-udhcpd = "busybox-udhcpd"
 CONFFILES_${PN}-syslog = "${sysconfdir}/syslog.conf"
 
 # This disables the syslog startup links in slugos (see slugos-init)
@@ -137,7 +137,7 @@ pkg_postinst_${PN} () {
 	# (update-alternatives have no problem replacing links later anyway)
 	test -n 2> /dev/null || alias test='busybox test'
 	if test "x$D" = "x"; then while read link; do if test ! -h "$link"; then case "$link" in /*/*/*) to="../../bin/busybox";; /bin/*) to="busybox";; /*/*) to="../bin/busybox";; /*) to="/bin/busybox";; esac; busybox ln -s $to $link; fi; done </etc/busybox.links; fi
-	
+
 	# This adds the links, remember that this has to work when building an image too, hence the $D
 	while read link; do case "$link" in /*/*/*) to="../../bin/busybox";; /bin/*) to="busybox";; /*/*) to="../bin/busybox";; /*) to="/bin/busybox";; esac; bn=`basename $link`; update-alternatives --install $link $bn $to 50; done <$D/etc/busybox.links
 }
@@ -167,7 +167,7 @@ pkg_prerm_${PN} () {
 			/*/*/*) to="../../bin/busybox";;
 			/bin/*) to="busybox";;
 			/*/*) to="../bin/busybox";;
-			/*) to="/bin/busybox";;		
+			/*) to="/bin/busybox";;
 		esac
 		bn=`basename $link`
 		sh /usr/bin/update-alternatives --remove $bn $to
Signature · 1 line
1.6.5
#2 lore raw

PATCH 2/12 some more !getopt_long fixes

---
 recipes/busybox/files/hwclock.sh |   24 ++++++++++++------------
 1 files changed, 12 insertions(+), 12 deletions(-)
recipes/busybox/files/hwclock.sh +12 -12
index 6c0af34..57c9fc7 100644
@@ -14,27 +14,27 @@
 
 . /etc/default/rcS
 
-[ "$UTC" = yes ] && UTC=--utc || UTC=--localtime
+[ "$UTC" = "yes" ] && UTC=-u || UTC=-l
 
 case "$1" in
         start)
-                if [ "$VERBOSE" != no ]
+                if [ "$VERBOSE" != "no" ]
                 then
                         echo "System time was `date`."
                         echo "Setting the System Clock using the Hardware Clock as reference..."
                 fi
 
-		if [ "$HWCLOCKACCESS" != no ]
+		if [ "$HWCLOCKACCESS" != "no" ]
 		then
 			if [ -z "$TZ" ]
 			then
-	                   hwclock -s $UTC;# --hctosys
+	                   hwclock -s $UTC
 			else
-			   TZ="$TZ" hwclock -s $UTC;# --hctosys
+			   TZ="$TZ" hwclock -s $UTC
 			fi
 		fi
 
-                if [ "$VERBOSE" != no ]
+                if [ "$VERBOSE" != "no" ]
                 then
                         echo "System Clock set. System local time is now `date`."
                 fi
@@ -47,24 +47,24 @@ case "$1" in
 		# WARNING: If you disable this, any changes to the system
 		#          clock will not be carried across reboots.
 		#
-		if [ "$VERBOSE" != no ]
+		if [ "$VERBOSE" != "no" ]
 		then
 			echo "Saving the System Clock time to the Hardware Clock..."
 		fi
-		if [ "$HWCLOCKACCESS" != no ]
+		if [ "$HWCLOCKACCESS" != "no" ]
 		then
-			hwclock -w $UTC;# --systohc
+			hwclock -w $UTC
 		fi
-		if [ "$VERBOSE" != no ]
+		if [ "$VERBOSE" != "no" ]
 		then
 			echo "Hardware Clock updated to `date`."
 		fi
                 exit 0
                 ;;
 	show)
-		if [ "$HWCLOCKACCESS" != no ]
+		if [ "$HWCLOCKACCESS" != "no" ]
 		then
-			hwclock -r $UTC;# --show
+			hwclock -r $UTC
 		fi
 		;;
         *)
Signature · 1 line
1.6.5
#3 lore raw

PATCH 3/12 osso-application-installer: nonworking

osso-application-installer depends on nonworking hildon-lgpl

---
 .../fix-buildsystem.patch                          |   65 --------------------
 .../maemo/osso-application-installer_1.0.03-1.bb   |   19 ------
 .../fix-buildsystem.patch                          |   65 ++++++++++++++++++++
 .../osso-application-installer_1.0.03-1.bb         |   19 ++++++
 4 files changed, 84 insertions(+), 84 deletions(-)
 delete mode 100644 recipes/maemo/osso-application-installer/fix-buildsystem.patch
 delete mode 100644 recipes/maemo/osso-application-installer_1.0.03-1.bb
 create mode 100644 recipes/nonworking/osso-application-installer/fix-buildsystem.patch
 create mode 100644 recipes/nonworking/osso-application-installer/osso-application-installer_1.0.03-1.bb
recipes/maemo/osso-application-installer/fix-buildsystem.patch deleted +0 -65
deleted file mode 100644
index d89522b..0000000
@@ -1,65 +0,0 @@
-diff -urbB osso-application-installer-1.0.03_old/configure.ac osso-application-installer-1.0.03/configure.ac
---- osso-application-installer-1.0.03_old/configure.ac	2005-05-22 22:06:52.000000000 +0200
-+++ osso-application-installer-1.0.03/configure.ac	2005-06-30 12:19:24.000000000 +0200
-@@ -38,9 +38,9 @@
- AC_SUBST(CP_CFLAGS)
- AC_SUBST(CP_LIBS)
- 
--cp_pluginlibdir=`pkg-config hildon-control-panel --variable=pluginlibdir`
-+cp_pluginlibdir=${libdir}/hildon-control-panel
- AC_SUBST(cp_pluginlibdir)
--cp_plugindesktopdir=`pkg-config hildon-control-panel --variable=plugindesktopentrydir`
-+cp_plugindesktopdir=${datadir}/applications/hildon-control-panel
- AC_SUBST(cp_plugindesktopdir)
- 
- # Localisation-related
-@@ -52,12 +52,12 @@
- AM_GLIB_GNU_GETTEXT
- 
- # Mime crap
--dnl MIME-type registration check (2.6 / 2.8 version)
--AC_PATH_PROG(UPDATE_MIME_DATABASE, update-mime-database, no)
--AM_CONDITIONAL(HAVE_26_MIME, test x$UPDATE_MIME_DATABASE != xno)
-+#dnl MIME-type registration check (2.6 / 2.8 version)
-+#AC_PATH_PROG(UPDATE_MIME_DATABASE, update-mime-database, no)
-+#AM_CONDITIONAL(HAVE_26_MIME, test x$UPDATE_MIME_DATABASE != xno)
- 
--AC_PATH_PROG(UPDATE_DESKTOP_DATABASE, update-desktop-database, no)
--AM_CONDITIONAL(HAVE_28_MIME, test x$UPDATE_DESKTOP_DATABASE != xno)
-+#AC_PATH_PROG(UPDATE_DESKTOP_DATABASE, update-desktop-database, no)
-+#AM_CONDITIONAL(HAVE_28_MIME, test x$UPDATE_DESKTOP_DATABASE != xno)
- 
- 
- applicationinstallerlocaledir=${datadir}/locale
-@@ -76,7 +76,7 @@
- AC_SUBST(hildonpixmapdir)
- AC_SUBST(hildonstatusbarlibdir)
- 
--dbuslibpathdir=`$PKG_CONFIG --variable=libdir dbus-1`
-+dbuslibpathdir=${libdir}
- dbusconfdir=/etc/dbus-1/system.d
- dbusservdir=${dbuslibpathdir}/dbus-1.0/services
- AC_SUBST(dbuslibpathdir)
-diff -urbB osso-application-installer-1.0.03_old/mime/Makefile.am osso-application-installer-1.0.03/mime/Makefile.am
---- osso-application-installer-1.0.03_old/mime/Makefile.am	2005-05-22 22:06:52.000000000 +0200
-+++ osso-application-installer-1.0.03/mime/Makefile.am	2005-06-30 12:19:24.000000000 +0200
-@@ -2,19 +2,3 @@
- mime_DATA = osso-application-installer.xml
- 
- EXTRA_DIST = $(mime_DATA)
--
--install-data-hook:
--if HAVE_26_MIME
--	$(UPDATE_MIME_DATABASE) "$(DESTDIR)$(datadir)/mime"
--endif
--if HAVE_28_MIME
--	$(UPDATE_DESKTOP_DATABASE)
--endif
--
--uninstall-hook:
--if HAVE_26_MIME
--	$(UPDATE_MIME_DATABASE) "$(DESTDIR)$(datadir)/mime"
--endif
--if HAVE_28_MIME
--	$(UPDATE_DESKTOP_DATABASE) 
--endif
recipes/maemo/osso-application-installer_1.0.03-1.bb deleted +0 -19
deleted file mode 100644
index 7380255..0000000
@@ -1,19 +0,0 @@
-PR         = "r0"
-LICENSE    = "GPL"
-
-DEPENDS = "hildon-lgpl hildon-base-lib libosso hildon-libs hildon-control-panel shared-mime-info"
-RDEPENDS = "shared-mime-info"
-
-SRC_URI = "http://repository.maemo.org/pool/maemo/ossw/source/o/${PN}/${PN}_${PV}.tar.gz \
-           file://fix-buildsystem.patch;patch=1"
-
-S = "${WORKDIR}/osso-application-installer-1.0.03"
-
-inherit autotools pkgconfig
-
-FILES_${PN} += "${libdir}/dbus-1.0 ${libdir}/hildon-control-panel/*.so ${datadir}/applications ${datadir}/mime"
-
-pkg_postinst () {
-  echo "Updating MIME database... this may take a while."
-  ${bindir}/update-mime-database ${datadir}/mime
-}
recipes/nonworking/osso-application-installer/fix-buildsystem.patch new file +65 -0

65 lines not on this page · raw message

recipes/nonworking/osso-application-installer/osso-application-installer_1.0.03-1.bb new file +19 -0
new file mode 100644
index 0000000..7380255
@@ -0,0 +1,19 @@
+PR         = "r0"
+LICENSE    = "GPL"
+
+DEPENDS = "hildon-lgpl hildon-base-lib libosso hildon-libs hildon-control-panel shared-mime-info"
+RDEPENDS = "shared-mime-info"
+
+SRC_URI = "http://repository.maemo.org/pool/maemo/ossw/source/o/${PN}/${PN}_${PV}.tar.gz \
+           file://fix-buildsystem.patch;patch=1"
+
+S = "${WORKDIR}/osso-application-installer-1.0.03"
+
+inherit autotools pkgconfig
+
+FILES_${PN} += "${libdir}/dbus-1.0 ${libdir}/hildon-control-panel/*.so ${datadir}/applications ${datadir}/mime"
+
+pkg_postinst () {
+  echo "Updating MIME database... this may take a while."
+  ${bindir}/update-mime-database ${datadir}/mime
+}
Signature · 1 line
1.6.5
#4 lore raw

PATCH 4/12 matchbox-panel-2: fix DEPENDS

libacpi has no recipe as opposed to acpid

---
 recipes/matchbox2/matchbox-panel-2_svn.bb |    2 +-
 1 files changed, 1 insertions(+), 1 deletions(-)
recipes/matchbox2/matchbox-panel-2_svn.bb +1 -1
index ae3b122..1a61370 100644
@@ -2,7 +2,7 @@ DESCRIPTION = "matchbox-panel-2 is a lightweight dock (system tray) application
 LICENSE = "GPLv2"
 SECTION = "x11/panels"
 DEPENDS = "gtk+ startup-notification dbus dbus-glib matchbox-panel-2-icon-themes"
-DEPENDS += " ${@base_contains("MACHINE_FEATURES", "acpi", "libacpi", "",d)}"
+DEPENDS += " ${@base_contains("MACHINE_FEATURES", "acpi", "acpid", "",d)}"
 DEPENDS += " ${@base_contains("MACHINE_FEATURES", "apm", "apmd", "",d)}"
 RDEPENDS_${PN} = "matchbox-panel-2-icon-theme"
 PACKAGE_ARCH = "${MACHINE_ARCH}"
Signature · 1 line
1.6.5
#5 lore raw

PATCH 5/12 no need to set LD nowadays

---
 recipes/busybox/busybox.inc |    3 ++-
 1 files changed, 2 insertions(+), 1 deletions(-)
recipes/busybox/busybox.inc +2 -1
index 7f67f47..c80b3bd 100644
@@ -66,7 +66,7 @@ RDEPENDS_${PN}-syslog += "${PN}"
 RDEPENDS_${PN}-udhcpd += "${PN}"
 
 # Use gcc for linking so LDFLAGS actually makes sense
-LD = "${CC} -nostdlib"
+#LD = "${CC} -nostdlib"
 
 inherit cml1 update-rc.d
 
@@ -78,6 +78,7 @@ do_configure () {
 
 do_compile() {
 	unset CFLAGS CPPFLAGS CXXFLAGS
+	export EXTRA_CFLAGS="$CFLAGS"
 	base_do_compile
 }
 
Signature · 1 line
1.6.5
#6 lore raw

PATCH 6/12 quilt: use builtin date routines

quilt tries to use an RFC822 date format which is obsoleted by rfc2822 since 8 years now.
It has a builtin date handler, so use that and not the date binary (which may only support rfc2822 by now, as busybox does)

---
 recipes/quilt/quilt-native.inc |    2 +-
 1 files changed, 1 insertions(+), 1 deletions(-)
recipes/quilt/quilt-native.inc +1 -1
index f815969..cad1b74 100644
@@ -9,7 +9,7 @@ INHIBIT_AUTOTOOLS_DEPS = "1"
 inherit autotools native
 
 PATCHTOOL = "patch"
-EXTRA_OECONF = "--disable-nls"
+EXTRA_OECONF = "--disable-nls --without-date"
 
 do_configure () {
 	oe_runconf
Signature · 1 line
1.6.5
#7 lore raw

PATCH 7/12 Document proxy setup in local.conf.sample

---
 conf/local.conf.sample |    3 +++
 1 files changed, 3 insertions(+), 0 deletions(-)
conf/local.conf.sample +3 -0
index 643ae0b..e7c4559 100644
@@ -30,6 +30,9 @@ BBFILES := "${@bb.fatal('Edit your conf/local.conf: BBFILES')}"
 # This is a regulary expression, so be sure to get your parenthesis balanced.
 BBMASK = ""
 
+# Proxy settings:
+# BB_ENV_EXTRAWHITE="http_proxy ftp_proxy no_proxy"
+
 # Uncomment this if you want to use a prebuilt toolchain. You will need to
 # provide packages for toolchain and additional libraries yourself. You also
 # have to set PATH in your environment to make sure BitBake finds additional binaries.
Signature · 1 line
1.6.5
#8 lore raw

PATCH 8/12 commentary typo fix

---
 recipes/uclibc/uclibc_git.bb |    2 +-
 1 files changed, 1 insertions(+), 1 deletions(-)
recipes/uclibc/uclibc_git.bb +1 -1
index 5f44466..757a538 100644
@@ -1,4 +1,4 @@
-# UCLIBC_BASE should be the latest released revision of uclibc (that way
+# UCLIBC_BASE should be the latest released version of uclibc (that way
 # the config files will typically be correct!)  uclibc-svn takes precedence
 # over uclibc-${UCLIBC_BASE}, if a config file in uclibc-snv is out of date
 # try removing it
Signature · 1 line
1.6.5
#9 lore raw

PATCH 9/12 flex: remove version specifics from generic inc

---
 recipes/flex/flex-native_2.5.31.bb |    4 ++--
 recipes/flex/flex.inc              |    3 ---
 recipes/flex/flex_2.5.31.bb        |    5 ++++-
 3 files changed, 6 insertions(+), 6 deletions(-)
recipes/flex/flex-native_2.5.31.bb +2 -2
index 1e9354a..de3f215 100644
@@ -1,6 +1,6 @@
-require flex.inc
+require flex_${PV}.bb
 inherit native
-PR = "r2"
+PR = "r3"
 
 do_stage () {
 	install -m 0755 flex ${STAGING_BINDIR}
recipes/flex/flex.inc +0 -3
index 4ebd44a..37e34fe 100644
@@ -2,9 +2,6 @@ DESCRIPTION = "Flex is a tool for generating programs that recognize lexical pat
 SECTION = "devel"
 LICENSE = "BSD"
 
-SRC_URI = "${SOURCEFORGE_MIRROR}/flex/flex-2.5.31.tar.bz2 \
-           file://flex-lvalue.diff;patch=1 \
-           file://fix-gen.patch;patch=1"
 S = "${WORKDIR}/flex-${PV}"
 
 inherit autotools
recipes/flex/flex_2.5.31.bb +4 -1
index e247a03..ae9e6f9 100644
@@ -1,7 +1,10 @@
 require flex.inc
 PR = "r5"
 
-SRC_URI += "file://include.patch;patch=1"
+SRC_URI = "${SOURCEFORGE_MIRROR}/flex/flex-${PV}.tar.bz2 \
+           file://flex-lvalue.diff;patch=1 \
+           file://fix-gen.patch;patch=1 \
+           file://include.patch;patch=1"
 
 do_stage() {
 	oe_libinstall -a libfl ${STAGING_LIBDIR}
Signature · 1 line
1.6.5
#10 lore raw

PATCH 10/12 git: remove version specifics from generic inc

---
 recipes/git/git-native_1.6.0.4.bb |    1 +
 recipes/git/git.inc               |    5 ++---
 recipes/git/git_1.6.0.4.bb        |    3 ++-
 3 files changed, 5 insertions(+), 4 deletions(-)
recipes/git/git-native_1.6.0.4.bb +1 -0
index 71b111b..c17ce6e 100644
@@ -2,3 +2,4 @@ require git.inc
 inherit native
 DEPENDS = "openssl-native curl-native zlib-native expat-native"
 PR = "r4"
+SRC_URI += "file://autotools.patch;patch=1"
recipes/git/git.inc +2 -3
index fb88fd4..644e159 100644
@@ -2,9 +2,8 @@ DESCRIPTION = "The git revision control system used by the Linux kernel develope
 SECTION = "console/utils"
 LICENSE = "GPL"
 
-SRC_URI = "http://www.kernel.org/pub/software/scm/git/git-${PV}.tar.bz2 \
-           file://autotools.patch;patch=1 \
-	   "
+SRC_URI = "http://www.kernel.org/pub/software/scm/git/git-${PV}.tar.bz2"
+
 S = "${WORKDIR}/git-${PV}"
 
 inherit autotools
recipes/git/git_1.6.0.4.bb +2 -1
index e839ab9..c37e3d0 100644
@@ -1,6 +1,7 @@
 require git.inc
 
-SRC_URI += "file://snprintf-returns-bogus.patch;patch=1"
+SRC_URI += "file://autotools.patch;patch=1 \
+            file://snprintf-returns-bogus.patch;patch=1"
 
 DEPENDS = "openssl curl zlib expat"
 RDEPENDS = "perl perl-module-file-path cpio findutils sed"
Signature · 1 line
1.6.5
#11 lore raw

PATCH 11/12 fix binutils patch to apply to 2.20

---
 .../binutils-x86_64_i386_biarch.patch              |    2 +-
 1 files changed, 1 insertions(+), 1 deletions(-)
recipes/binutils/binutils-2.20/binutils-x86_64_i386_biarch.patch +1 -1
index 6054434..4cd9e22 100644
@@ -11,9 +11,9 @@
 --- binutils/ld/emulparams/elf_i386.sh
 +++ binutils/ld/emulparams/elf_i386.sh
 @@ -12,3 +12,13 @@
- NO_SMALL_DATA=yes
  SEPARATE_GOTPLT=12
  SHARABLE_SECTIONS=yes
+ IREL_IN_PLT=
 +
 +# Linux modify the default library search path to first include
 +# a 32-bit specific directory.
Signature · 1 line
1.6.5
#12 lore raw

PATCH 12/12 task-base.bb: use uniform test for set

---
 recipes/tasks/task-base.bb |    4 +++-
 1 files changed, 3 insertions(+), 1 deletions(-)
recipes/tasks/task-base.bb +3 -1
index d59acba..fc04b46 100644
@@ -136,7 +136,9 @@ python __anonymous () {
 
     import bb
 
-    if not hasattr(__builtins__, 'set'):
+    try:
+	set
+    except NameError:
 	from sets import Set as set
 
     distro_features = set(bb.data.getVar("DISTRO_FEATURES", d, 1).split())
Signature · 1 line
1.6.5
#13 lore raw

Re: [PATCH 12/12] task-base.bb: use uniform test for set

recipes/tasks/task-base.bb python __anonymous () { · line 142

Quoted 5 lines · Bernhard Reutner-Fischer
recipes/tasks/task-base.bb +3 -1
-    if not hasattr(__builtins__, 'set'):
+    try:
+	set
+    except NameError:
 	from sets import Set as set

It's not completely obvious to me that this patch is an improvement: it replaces a single line of code with three new ones and doesn't seem to change the outcome. Maybe one of our Python überhackers can comment on whether or not this is really better.

p.

#14 lore raw

Re: [PATCH 12/12] task-base.bb: use uniform test for set

Quoted 11 lines · Phil Blundell
Quoted 5 lines · Bernhard Reutner-Fischer

-    if not hasattr(__builtins__, 'set'):
+    try:
+     set
+    except NameError:
      from sets import Set as set

It's not completely obvious to me that this patch is an improvement: it replaces a single line of code with three new ones and doesn't seem to change the outcome.  Maybe one of our Python überhackers can comment on whether or not this is really better.

hasatts uses getattr and catches the exception from getattr. There was a problem in hasattr where it returned False on
any exception generated from getaddr so the result could be unreliable. But this may not be applicable in this particular case

I still see

NOTE: Handling BitBake files: \ (7015/8051) [87 %]__builtin__:10:
DeprecationWarning: the sets module is deprecated
NOTE: Handling BitBake files: | (8051/8051) [100 %] Parsing of 8051 .bb files complete (0 cached, 8051 parsed). 7800 targets, 333 skipped, 157 masked, 0 errors.

Quoted 10 lines

p.

_______________________________________________
Openembedded-devel mailing list
[email protected]
http://lists.linuxtogo.org/cgi-bin/mailman/listinfo/openembedded-devel

#16 lore raw

Re: [PATCH 07/12] Document proxy setup in local.conf.sample

In reply to Bernhard Reutner-Fischer

conf/local.conf.sample BBFILES := "${@bb.fatal('Edit your conf/local.conf: BBFILES')}" · line 34

Quoted 10 lines · Bernhard Reutner-Fischer
conf/local.conf.sample +2 -0
index 643ae0b..e7c4559 100644
@@ -30,6 +30,9 @@ BBFILES := "${@bb.fatal('Edit your conf/local.conf: BBFILES')}"
 # This is a regulary expression, so be sure to get your parenthesis balanced.
 BBMASK = ""
 
+# Proxy settings:
+# BB_ENV_EXTRAWHITE="http_proxy ftp_proxy no_proxy"

I'm sorry to inform you, but that won't work with the standard bitbake! Which uses own HTTP_PROXY, FTP_PROXY, HTTP_PROXY_IGNORE and FTP_PROXY_IGNORE vars.

I tried to persuade RP to change that back to standard http_proxy, ftp_proxy and no_proxy to no avail[1].

Because of that I had to keep my own version of bitbake[2] with the necessary patches for my users - check the top 4 patches in that tree. Although, GIT_PROXY_COMMAND one is already in the bitbake master...

And looks like Khem has already pushed the above change though...

[1] http://thread.gmane.org/gmane.comp.handhelds.openembedded/19716
[2] http://arago-project.org/git/?p=arago-bitbake.git;a=shortlog;h=refs/heads/next

Signature · 1 line
Denys
#17 lore raw

Re: [PATCH 07/12] Document proxy setup in local.conf.sample

Quoted 15 lines · Denys Dmytriyenko

conf/local.conf.sample BBFILES := "${@bb.fatal('Edit your conf/local.conf: BBFILES')}" · line 34

Quoted 10 lines · Bernhard Reutner-Fischer
conf/local.conf.sample +2 -0
index 643ae0b..e7c4559 100644
@@ -30,6 +30,9 @@ BBFILES := "${@bb.fatal('Edit your conf/local.conf: BBFILES')}"
 # This is a regulary expression, so be sure to get your parenthesis balanced.
 BBMASK = ""
 
+# Proxy settings:
+# BB_ENV_EXTRAWHITE="http_proxy ftp_proxy no_proxy"

I'm sorry to inform you, but that won't work with the standard bitbake! Which uses own HTTP_PROXY, FTP_PROXY, HTTP_PROXY_IGNORE and FTP_PROXY_IGNORE vars.

Quoted 8 lines

I tried to persuade RP to change that back to standard http_proxy, ftp_proxy and no_proxy to no avail[1].

Because of that I had to keep my own version of bitbake[2] with the necessary patches for my users - check the top 4 patches in that tree. Although, GIT_PROXY_COMMAND one is already in the bitbake master...

And looks like Khem has already pushed the above change though...

drats, you're right. I was erroneously checking against a patched bitbake.

khem, i'm very sorry for my sloppiness! Please
revert 3f602f72f0936ef773aa4767636d41bb82ecf3b0 :
---8<---
Unfortunately bitbake master allegedly uses the non-standard {HT,F}TP_PROXY{,_IGNORE} config variables instead of the established http_proxy, ftp_proxy, no_proxy environment variables. ---8<---
lynx has a good description for these, and especially a sane rule for no_proxy (with domain.tld:port csv entries) which works very well for me in practise, FYI.
http://lynx.isc.org/lynx2.8.5/lynx2-8-5/lynx_help/keystrokes/environments.html#proxy

Sorry for the wrong patch and thanks, Denys, for mentioning this!

Quoted 3 lines

[1] http://thread.gmane.org/gmane.comp.handhelds.openembedded/19716
[2] http://arago-project.org/git/?p=arago-bitbake.git;a=shortlog;h=refs/heads/next

#18 lore raw

Re: [PATCH 07/12] Document proxy setup in local.conf.sample

Quoted 23 lines · Bernhard Reutner-Fischer
Quoted 22 lines · Denys Dmytriyenko

conf/local.conf.sample BBFILES := "${@bb.fatal('Edit your conf/local.conf: BBFILES')}" · line 34

Quoted 10 lines · Bernhard Reutner-Fischer
conf/local.conf.sample +2 -0
index 643ae0b..e7c4559 100644
@@ -30,6 +30,9 @@ BBFILES := "${@bb.fatal('Edit your conf/local.conf: BBFILES')}"
 # This is a regulary expression, so be sure to get your parenthesis balanced.
 BBMASK = ""
 
+# Proxy settings:
+# BB_ENV_EXTRAWHITE="http_proxy ftp_proxy no_proxy"

I'm sorry to inform you, but that won't work with the standard bitbake! Which uses own HTTP_PROXY, FTP_PROXY, HTTP_PROXY_IGNORE and FTP_PROXY_IGNORE vars.

I tried to persuade RP to change that back to standard http_proxy, ftp_proxy and no_proxy to no avail[1].

Because of that I had to keep my own version of bitbake[2] with the necessary patches for my users - check the top 4 patches in that tree. Although, GIT_PROXY_COMMAND one is already in the bitbake master...

Quoted 6 lines
Quoted 1 line

And looks like Khem has already pushed the above change though...

drats, you're right. I was erroneously checking against a patched bitbake.

khem, i'm very sorry for my sloppiness! Please
revert 3f602f72f0936ef773aa4767636d41bb82ecf3b0 :

Reverted.

Quoted 14 lines

---8<---
Unfortunately bitbake master allegedly uses the non-standard {HT,F}TP_PROXY{,_IGNORE} config variables instead of the established http_proxy, ftp_proxy, no_proxy environment variables. ---8<---
lynx has a good description for these, and especially a sane rule for no_proxy (with domain.tld:port csv entries) which works very well for me in practise, FYI.
http://lynx.isc.org/lynx2.8.5/lynx2-8-5/lynx_help/keystrokes/environments.html#proxy

Sorry for the wrong patch and thanks, Denys, for mentioning this!

Quoted 3 lines

[1] http://thread.gmane.org/gmane.comp.handhelds.openembedded/19716
[2] http://arago-project.org/git/?p=arago-bitbake.git;a=shortlog;h=refs/heads/next

Signature · 1 line
Denys
#19 lore raw

Re: [PATCH 06/12] quilt: use builtin date routines

In reply to Bernhard Reutner-Fischer
Quoted 4 lines · Bernhard Reutner-Fischer

quilt tries to use an RFC822 date format which is obsoleted by rfc2822 since 8 years now.
It has a builtin date handler, so use that and not the date binary (which may only support rfc2822 by now, as busybox does)

Thanks for the patch. I checked this in.

p.

#20 lore raw

Re: [PATCH 05/12] no need to set LD nowadays

In reply to Bernhard Reutner-Fischer

recipes/busybox/busybox.inc RDEPENDS_${PN}-syslog += "${PN}" · line 69

Quoted 4 lines · Bernhard Reutner-Fischer

# Use gcc for linking so LDFLAGS actually makes sense

recipes/busybox/busybox.inc +1 -1
-LD = "${CC} -nostdlib"
+#LD = "${CC} -nostdlib"

Thanks for the patch. When you say "nowadays", do you know which actual versions of busybox this is true for? Is this definitely correct for all the versions that we have in the .dev tree?

Also, if that line is genuinely no longer required, you might as well just delete it. There is no point in leaving commented-out cruft in these recipes indefinitely.

p.

#21 lore raw

Re: [PATCH 05/12] no need to set LD nowadays

Quoted 8 lines · Phil Blundell

recipes/busybox/busybox.inc RDEPENDS_${PN}-syslog += "${PN}" · line 69

Quoted 4 lines · Bernhard Reutner-Fischer

# Use gcc for linking so LDFLAGS actually makes sense

recipes/busybox/busybox.inc +1 -1
-LD = "${CC} -nostdlib"
+#LD = "${CC} -nostdlib"

Thanks for the patch. When you say "nowadays", do you know which actual versions of busybox this is true for? Is this definitely correct for

busybox >= 1.4.0 switched to kbuild, i.e.
make CROSS_COMPILE=alpha-linux- EXTRA_CFLAGS=-fwhatever \
EXTRA_LDFLAGS=-Wl,--warn-multiple-gp

For the usual case you'd either set CONFIG_CROSS_COMPILER_PREFIX in the .config or pass CC down to make.

If you happen to have an odd host compiler then you want those too:
HOSTCC=ccc HOSTCFLAGS="-ieee" make ....

Quoted 1 line

all the versions that we have in the .dev tree?

It seems that the .dev tree has 1.2.1, 1.2.2, 1.3.1 still in there but i, personally would strongly advise against using those (same for the standalone tinylogin, but that's a different thing and not relevant right now).

Quoted 4 lines

Also, if that line is genuinely no longer required, you might as well just delete it. There is no point in leaving commented-out cruft in these recipes indefinitely.

The ancient cruft versions made me leave in that comment.

#22 lore raw

Re: [PATCH 05/12] no need to set LD nowadays

Quoted 4 lines · Bernhard Reutner-Fischer

It seems that the .dev tree has 1.2.1, 1.2.2, 1.3.1 still in there but i, personally would strongly advise against using those (same for the standalone tinylogin, but that's a different thing and not relevant right now).

As long as those versions are there, I don't think it would be acceptable to check in a patch to busybox.inc which would delete them. Leaving the ${LD} setting as a comment doesn't really help with that.

If you wanted to generate a new patch which added the ${LD} setting to the individual recipes that still require it, and simultaneously delete it from busybox.inc, then I think that would be fine.

p.

#23 lore raw

Re: [PATCH 05/12] no need to set LD nowadays

Quoted 6 lines · Phil Blundell
Quoted 4 lines · Bernhard Reutner-Fischer

It seems that the .dev tree has 1.2.1, 1.2.2, 1.3.1 still in there but i, personally would strongly advise against using those (same for the standalone tinylogin, but that's a different thing and not relevant right now).

Quoted 7 lines

As long as those versions are there, I don't think it would be acceptable to check in a patch to busybox.inc which would delete them. Leaving the ${LD} setting as a comment doesn't really help with that.

If you wanted to generate a new patch which added the ${LD} setting to the individual recipes that still require it, and simultaneously delete it from busybox.inc, then I think that would be fine.

Ack, i retract this patch for now.
Thanks,

#24 lore raw

Re: [PATCH 05/12] no need to set LD nowadays

In reply to Bernhard Reutner-Fischer

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Quoted 10 lines · Bernhard Reutner-Fischer
---
 recipes/busybox/busybox.inc |    3 ++-
 1 files changed, 2 insertions(+), 1 deletions(-)
recipes/busybox/busybox.inc +0 -0
index 7f67f47..c80b3bd 100644

recipes/busybox/busybox.inc RDEPENDS_${PN}-syslog += "${PN}" · line 69

Quoted 5 lines
recipes/busybox/busybox.inc +1 -1
@@ -66,7 +66,7 @@ RDEPENDS_${PN}-syslog += "${PN}"
 RDEPENDS_${PN}-udhcpd += "${PN}"
 
 # Use gcc for linking so LDFLAGS actually makes sense
-LD = "${CC} -nostdlib"
+#LD = "${CC} -nostdlib"

We have an scm, so you can just delete both the statement and comment. -----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.5 (Darwin)

iD8DBQFLcRh2MkyGM64RGpERApVjAJ0X74dRU/v8yAyQvqhcqrCz9v1LwgCfZ52X tRQfra5kwqse4cUBbK04+60=
=q9ca
-----END PGP SIGNATURE-----

#25 lore raw

Re: [PATCH 04/12] matchbox-panel-2: fix DEPENDS

In reply to Bernhard Reutner-Fischer
Quoted 6 lines · Bernhard Reutner-Fischer

libacpi has no recipe as opposed to acpid

---
 recipes/matchbox2/matchbox-panel-2_svn.bb |    2 +-
 1 files changed, 1 insertions(+), 1 deletions(-)

Um, really? I do seem to have a recipe for libacpi in my tree.

-rw-r--r-- 1 pb pb 463 Nov 23 14:53 recipes/libacpi/libacpi_0.2.bb

Also, from a quick look at the acpid package, it doesn't seem to stage any libraries or headers so it is not clear to me that it would actually satisfy matchbox-panel-2's requirements. Did you verify that matchbox-panel-2's ACPI code really is buildable with your patch applied?

p.

#26 lore raw

Re: [PATCH 04/12] matchbox-panel-2: fix DEPENDS

Quoted 9 lines · Phil Blundell
Quoted 6 lines · Bernhard Reutner-Fischer

libacpi has no recipe as opposed to acpid

---
 recipes/matchbox2/matchbox-panel-2_svn.bb |    2 +-
 1 files changed, 1 insertions(+), 1 deletions(-)

Um, really? I do seem to have a recipe for libacpi in my tree.

Oh, there was none in september. I withdraw this patch (and sorry for not re-checking. Will do from now on..)

Thanks,

#27 lore raw

Re: [PATCH 03/12] osso-application-installer: nonworking

In reply to Bernhard Reutner-Fischer
Quoted 1 line · Bernhard Reutner-Fischer

osso-application-installer depends on nonworking hildon-lgpl

I hereby withdraw this 'bitbake help' hinting on 'world' prompted but severely misguided attempt from september to get that world straight. Sorry and cheers,
Bernhard

PS: +1 for teaching bitbake to emit sensible targets within reach.. :/

#28 lore raw

Re: [PATCH 02/12] some more !getopt_long fixes

In reply to Bernhard Reutner-Fischer

Review

Quoted 2 lines · Bernhard Reutner-Fischer
(quoted diff) +1 -1
-                if [ "$VERBOSE" != no ]
+                if [ "$VERBOSE" != "no" ]

I don't see any need for this change. Why doesn't the commit message does not explain why it was done?

I assume it is a stylistic change only and should that's why committed separately (with a different commit log), right?

Jens

#29 lore raw

Re: [PATCH 02/12] some more !getopt_long fixes

Quoted 9 lines · Jens Seidel

Review

Quoted 2 lines · Bernhard Reutner-Fischer
(quoted diff) +1 -1
-                if [ "$VERBOSE" != no ]
+                if [ "$VERBOSE" != "no" ]

I don't see any need for this change. Why doesn't the commit message does not explain why it was done?

I assume it is a stylistic change only and should that's why committed separately (with a different commit log), right?

Yes, it is a stylistic change only and could be committed separately. I'll factor them out and resend.

#31 lore raw

Re: [PATCH 00/12] misc trivia

In reply to Bernhard Reutner-Fischer

On Mon, Feb 08, 2010 at 10:14:22PM +0100, Bernhard Reutner-Fischer wrote:

All were applied or retracted except these 2 patches that asked for refactoring and which were resent accordingly today:

Jens Seidel:
resent as [PATCH 1/8] hwclock.sh: some more !getopt_long fixes [was: [PATCH 02/12] some more !getopt_long fixes] Phil Blundell:
dropped, now [PATCH 5/8] busybox: use .tar.bz2 [was: [PATCH 05/12] no need to set LD nowadays]

Thanks!