Maciej Borzecki | 22 Dec 15:28 2014
Picon

[PATCH] wic: account for mmcblk device partition naming

MMC block device partitions are named differently than other block
devices and use the scheme: mmcblk<devnum>p<partnum>, ex: mmcblk0p1,
mmcblk0p2. The current code generates incorrect parition names missing
'p' infix for fstab entries. The patch resolves this problem.

Signed-off-by: Maciej Borzecki <maciej.borzecki@...>
Signed-off-by: Maciek Borzecki <maciek.borzecki@...>
---
 scripts/lib/wic/imager/direct.py | 16 ++++++++++++----
 1 file changed, 12 insertions(+), 4 deletions(-)

diff --git a/scripts/lib/wic/imager/direct.py b/scripts/lib/wic/imager/direct.py
index 6b2ab33..b1dc3e9 100644
--- a/scripts/lib/wic/imager/direct.py
+++ b/scripts/lib/wic/imager/direct.py
 <at>  <at>  -100,10 +100,18  <at>  <at>  class DirectImageCreator(BaseImageCreator):
         for num, p in enumerate(parts, 1):
             if not p.mountpoint or p.mountpoint == "/" or p.mountpoint == "/boot":
                 continue
-            if self._ptable_format == 'msdos' and num > 3:
-                device_name = "/dev/" + p.disk + str(num + 1)
-            else:
-                device_name = "/dev/" + p.disk + str(num)
+
+            part = ''
+            # mmc device partitions are named mmcblk0p1, mmcblk0p2..
+            if p.disk.startswith('mmcblk'):
+                part = 'p'
+
+            pnum = num
(Continue reading)

Awais Belal | 22 Dec 14:53 2014

[PATCH] gstreamer1.0-* use gitsm style SRC_URI for git recipes

All gstreamer components when fetched through git have a git
submodule type repo rather than a main repo style. This
will make builds fail when there's no network connection
while the source is already downloaded.
We fix this by explicitly using the gitsm styled SRC_URI for
these submodules.

Signed-off-by: Awais Belal <awais_belal@...>
---
 .../gstreamer/gstreamer1.0-libav_git.bb            |    2 +-
 .../gstreamer/gstreamer1.0-omx_git.bb              |    2 +-
 .../gstreamer/gstreamer1.0-plugins.inc             |    2 +-
 .../gstreamer/gstreamer1.0_git.bb                  |    2 +-
 4 files changed, 4 insertions(+), 4 deletions(-)

diff --git a/meta/recipes-multimedia/gstreamer/gstreamer1.0-libav_git.bb b/meta/recipes-multimedia/gstreamer/gstreamer1.0-libav_git.bb
index bf1b87e..376a7c4 100644
--- a/meta/recipes-multimedia/gstreamer/gstreamer1.0-libav_git.bb
+++ b/meta/recipes-multimedia/gstreamer/gstreamer1.0-libav_git.bb
 <at>  <at>  -11,7 +11,7  <at>  <at>  LIC_FILES_CHKSUM = "file://COPYING;md5=b234ee4d69f5fce4486a80fdaf4a4263 \
                     file://gst-libs/ext/libav/COPYING.LGPLv2.1;md5=bd7a443320af8c812e4c18d1b79df004 \
                     file://gst-libs/ext/libav/COPYING.LGPLv3;md5=e6a600fd5e1d9cbde2d983680233ad02"

-SRC_URI = "git://anongit.freedesktop.org/gstreamer/gst-libav;branch=master"
+SRC_URI = "gitsm://anongit.freedesktop.org/gstreamer/gst-libav;branch=master"
 S = "${WORKDIR}/git"

 SRCREV = "127202d6f65584891dabf92be031f0d170b0e7f1"
diff --git a/meta/recipes-multimedia/gstreamer/gstreamer1.0-omx_git.bb b/meta/recipes-multimedia/gstreamer/gstreamer1.0-omx_git.bb
index 931a7fc..df7c58c 100644
(Continue reading)

Enrico Scholz | 22 Dec 12:24 2014
Picon

How to enforce ${B} = ${S} builds globally?

Hi,

what is the recommended way to enable in-source builds globally?
Building in a separate directory might be interesting for qa, but it
lowers end-user experience significantly because __FILE__ expands now
to absolute names.

E.g.

| #include <assert.h>
| int main() { assert(0); }

gives

| test:
/srv/oe/cache/bld/f95aceac-8952-bd61-430a-dfcdf48db33c/work/cortexa9t2hf-vfp-neon-linux-gnueabi/test/1.0-r0/test-1.0/test.c:2:
main: Assertion `0' failed.

instead of

| test: test.c:2: main: Assertion `0' failed.

Enrico
--

-- 
Richard Purdie | 22 Dec 11:13 2014

[PATCH] runqemu-internal: Fix qemu networking for qemuarm64

This brings qemuarm64 into feature parity with the other qemu machines
and enables the automated testing on the autobuilder.

Signed-off-by: Richard Purdie <richard.purdie@...>

diff --git a/scripts/runqemu-internal b/scripts/runqemu-internal
index c219419..f6274e2 100755
--- a/scripts/runqemu-internal
+++ b/scripts/runqemu-internal
 <at>  <at>  -264,7 +264,7  <at>  <at>  else
             ROOTFS_OPTIONS="-hda $ROOTFS"
         fi
         if [ "$MACHINE" = "qemuarm64" ]; then
-            QEMU_NETWORK_CMD="-netdev user,id=net0 -device virtio-net-device,netdev=net0"
+            QEMU_NETWORK_CMD="-netdev tap,id=net0,ifname=$TAP,script=no,downscript=no -device
virtio-net-device,netdev=net0 "
             DROOT="/dev/vda"
             ROOTFS_OPTIONS="-drive id=disk0,file=$ROOTFS -device virtio-blk-device,drive=disk0"
         fi
 <at>  <at>  -381,7 +381,7  <at>  <at>  if [ "$MACHINE" = "qemuarm64" ]; then
     export QEMU_AUDIO_DRV="none"
     QEMU_UI_OPTIONS="$QEMU_UI_OPTIONS"
     if [ "${FSTYPE:0:3}" = "ext" -o "$FSTYPE" = "btrfs" ]; then
-        KERNCMDLINE="root=/dev/vda rw console=ttyAMA0,38400 mem=$QEMU_MEMORY highres=off"
+        KERNCMDLINE="root=/dev/vda rw console=ttyAMA0,38400 mem=$QEMU_MEMORY highres=off $KERNEL_NETWORK_CMD"
         # qemu-system-aarch64 only support '-machine virt -cpu cortex-a57' for now
         QEMUOPTIONS="$QEMU_NETWORK_CMD -machine virt -cpu cortex-a57 $ROOTFS_OPTIONS $QEMU_UI_OPTIONS"
     fi

--

-- 
(Continue reading)

Robert Yang | 22 Dec 06:28 2014

[PATCH 0/7] Packages upgrade

The following changes since commit c3598cf720e04ab27ab5d2817c09f2496b677560:

  kernel-devsrc: Ensure we have a dependency on the actual source (2014-12-21 13:51:29 +0000)

are available in the git repository at:

  git://git.openembedded.org/openembedded-core-contrib rbt/pu
  http://cgit.openembedded.org/cgit.cgi/openembedded-core-contrib/log/?h=rbt/pu

Robert Yang (7):
  file: upgrade to 5.21
  wget: upgrade to 1.16.1
  wget: add PACKAGECONFIG for util-linux
  rsync: upgrade to 3.1.1
  logrotate: upgrade to 3.8.8
  git: upgrade to 2.2.0
  make: upgrade to 4.1

 .../recipes-devtools/file/file/debian-742262.patch |    4 +--
 .../file/{file_5.18.bb => file_5.21.bb}            |    4 +--
 .../git/{git_2.0.1.bb => git_2.2.0.bb}             |    4 +--
 .../make/{make_4.0.bb => make_4.1.bb}              |    4 +--
 .../rsync/{rsync_3.1.0.bb => rsync_3.1.1.bb}       |    4 +--
 .../disable-check-different-filesystems.patch      |   34 --------------------
 .../{logrotate_3.8.7.bb => logrotate_3.8.8.bb}     |    5 ++-
 meta/recipes-extended/wget/wget.inc                |    5 ++-
 .../wget/{wget_1.16.bb => wget_1.16.1.bb}          |    4 +--
 9 files changed, 18 insertions(+), 50 deletions(-)
 rename meta/recipes-devtools/file/{file_5.18.bb => file_5.21.bb} (86%)
 rename meta/recipes-devtools/git/{git_2.0.1.bb => git_2.2.0.bb} (61%)
(Continue reading)

Armin Kuster | 22 Dec 05:54 2014
Picon

[PATCH 0/5] [RFC] Move packages from meta-oe to core for X11

Please consider moving xterm, xclock, twm and libxaw from 
meta-openembedded/recipes-graphics to OE-core.

This fixes 'core-image-x11' image dependancy on meta-oe packages
since xinit requires xterm, xclock and twm to 'startx'.

if approved and commited, I will post patches to remove those packages
from meta-oe.

Armin Kuster (5):
  xclock: move from meta-oe to core.
  twm: move from meta-oe to core.
  libxaw: move from meta-oe to core.
  xterm: move from meta-oe to core.
  xinit: fix depends issue

 meta/recipes-graphics/xorg-app/twm_1.0.8.bb     | 20 +++++++++++++++
 meta/recipes-graphics/xorg-app/xclock_1.0.5.bb  | 15 +++++++++++
 meta/recipes-graphics/xorg-app/xinit_1.3.3.bb   |  8 +++++-
 meta/recipes-graphics/xorg-app/xterm_313.bb     |  1 +
 meta/recipes-graphics/xorg-lib/libxaw_1.0.10.bb | 33 +++++++++++++++++++++++++
 5 files changed, 76 insertions(+), 1 deletion(-)
 create mode 100644 meta/recipes-graphics/xorg-app/twm_1.0.8.bb
 create mode 100644 meta/recipes-graphics/xorg-app/xclock_1.0.5.bb
 create mode 120000 meta/recipes-graphics/xorg-app/xterm_313.bb
 create mode 100644 meta/recipes-graphics/xorg-lib/libxaw_1.0.10.bb

--

-- 
1.9.1

(Continue reading)

Paul Barker | 21 Dec 23:11 2014
Picon

[PATCH 0/8] opkg recipe improvements

This patch series implements the changes I suggested on the mailing list a few
days ago.

Summary of changes:
  - The opkg recipe is cleanup up and opkg.inc is merged into opkg_0.2.4.bb.

  - opkg-collateral is updated and then merged into the main opkg recipe.

  - The cryptically named opkg-config-base is given the much more obvious name
    opkg-arch-config.

  - poky-feed-config-opkg is dropped, it is recommended that people use
    distro-feed-configs from meta-oe.

  - A new opkg-keyrings recipe is created to package up keys used to verify
    packages and package feeds. The functionality within opkg which this depends
    upon is backported.

After these changes:
  - All generic files are in 'opkg', including the main opkg config which should
    be upgraded in step with opkg itself.

  - Arch-specific config is only in 'opkg-arch-config' which has MACHINE_ARCH.
    Additional arch-specific config can easily be added to this package with a
    bbappend.

  - Distro-specific config is in 'distro-feed-config' and 'opkg-keyrings'.
    Additional distro-specific config is probably best added as extra recipes.

I think this is a much cleaner layout but let me know if you disagree, I'm happy
(Continue reading)

Richard Purdie | 21 Dec 17:52 2014

linux-imx issues with make 3.82

Hi Otavio,

https://autobuilder.yoctoproject.org/main/builders/nightly-fsl-arm/builds/133/steps/BuildImages/logs/stdio
https://autobuilder.yoctoproject.org/main/builders/nightly-fsl-arm-lsb/builds/133/steps/BuildImages/logs/stdio

It seems linux-imx is continuing to fail on some autobuilders but not
others (hence I thought we'd fixed it). It seems the problem is old
kernels with make 3.82. Would you be able to apply something like the
patch in:

http://www.gossamer-threads.com/lists/linux/kernel/1263151

to linux-imx so it works with make 3.82?

Cheers,

Richard

--

-- 
Richard Purdie | 21 Dec 13:30 2014

[PATCH] kernel-devsrc: Ensure we don't race against do_make_scripts from module-base.bbclass

do_install for kernel-devsrc can race against do_make_scripts from 
module-base.bbclass. Since there is a lock there to guard against concurrency
already, we can just use it here to avoid a race.

Ultimately, this can all likely be much more streamlined but this resolves
the immediate build failures.

Signed-off-by: Richard Purdie <richard.purdie@...>

diff --git a/meta/recipes-kernel/linux/kernel-devsrc.bb b/meta/recipes-kernel/linux/kernel-devsrc.bb
index 1557c27..9fb761b 100644
--- a/meta/recipes-kernel/linux/kernel-devsrc.bb
+++ b/meta/recipes-kernel/linux/kernel-devsrc.bb
 <at>  <at>  -55,6 +55,8  <at>  <at>  do_install() {
                 cp ${S}/arch/powerpc/lib/crtsavres.o $kerneldir/arch/powerpc/lib/crtsavres.o
         fi
 }
+# Ensure we don't race against "make scripts" during cpio
+do_install[lockfiles] = "${TMPDIR}/kernel-scripts.lock"

 PACKAGES = "kernel-devsrc"
 FILES_${PN} = "${KERNEL_SRC_PATH}"

--

-- 
Richard Purdie | 21 Dec 13:29 2014

[PATCH] kernel-devsrc: Handle ppc crtsaves.o explictly for now

Resolve kernel module build failures for qemuppc by including crtsaves.o.

I'm not particularly happy to be doing this, it should perhaps be contained 
in the kernel-dev package. Until the overlap between kernel-devsrc and 
kernel-dev is resolved, this at least removed the regressions.

Signed-off-by: Richard Purdie <richard.purdie@...>

diff --git a/meta/recipes-kernel/linux/kernel-devsrc.bb b/meta/recipes-kernel/linux/kernel-devsrc.bb
index 2872f17..1557c27 100644
--- a/meta/recipes-kernel/linux/kernel-devsrc.bb
+++ b/meta/recipes-kernel/linux/kernel-devsrc.bb
 <at>  <at>  -46,6 +46,14  <at>  <at>  do_install() {
         cd ${S}
         find . -type d -name '.git*' -prune -o -type f -print0 | cpio --null -pdlu $kerneldir
         oe_runmake -C $kerneldir CC="${KERNEL_CC}" LD="${KERNEL_LD}" clean _mrproper_scripts
+
+        # As of Linux kernel version 3.0.1, the clean target removes
+        # arch/powerpc/lib/crtsavres.o which is present in
+        # KBUILD_LDFLAGS_MODULE, making it required to build external modules.
+        if [ ${ARCH} = "powerpc" ]; then
+                mkdir -p $kerneldir/arch/powerpc/lib/
+                cp ${S}/arch/powerpc/lib/crtsavres.o $kerneldir/arch/powerpc/lib/crtsavres.o
+        fi
 }

 PACKAGES = "kernel-devsrc"

--

-- 
(Continue reading)

Richard Purdie | 20 Dec 14:57 2014

[PATCH] linux-libc-headers: Upgrade 3.17 -> 3.17.7

Whilst we don't usually take linux-libc-headers point release, 3.17.3
fixes a userspace header issue which caused gdb to fail to build on
mips. We therefore may as well upgrade to the latest point release.

Signed-off-by: Richard Purdie <richard.purdie@...>

diff --git a/meta/recipes-kernel/linux-libc-headers/linux-libc-headers_3.17.7.bb b/meta/recipes-kernel/linux-libc-headers/linux-libc-headers_3.17.7.bb
new file mode 100644
index 0000000..7332597
--- /dev/null
+++ b/meta/recipes-kernel/linux-libc-headers/linux-libc-headers_3.17.7.bb
 <at>  <at>  -0,0 +1,7  <at>  <at> 
+KORG_ARCHIVE_COMPRESSION = "xz"
+
+require linux-libc-headers.inc
+
+SRC_URI[md5sum] = "59cc7d0a236b08de743d6d9421f0f2c6"
+SRC_URI[sha256sum] = "a3c8e6aee912bfd2e69abf9df5a06058434a8771bcaa1784b5980bf1b5298e1c"
+
diff --git a/meta/recipes-kernel/linux-libc-headers/linux-libc-headers_3.17.bb b/meta/recipes-kernel/linux-libc-headers/linux-libc-headers_3.17.bb
deleted file mode 100644
index 3417697..0000000
--- a/meta/recipes-kernel/linux-libc-headers/linux-libc-headers_3.17.bb
+++ /dev/null
 <at>  <at>  -1,7 +0,0  <at>  <at> 
-KORG_ARCHIVE_COMPRESSION = "xz"
-
-require linux-libc-headers.inc
-
-SRC_URI[md5sum] = "fb30d0f29214d75cddd2faa94f73d5cf"
(Continue reading)


Gmane