]> code.ossystems Code Review - openembedded-core.git/log
openembedded-core.git
8 years agostaging: Add missing parameter to debug message
Richard Purdie [Mon, 6 Feb 2017 15:47:28 +0000 (15:47 +0000)]
staging: Add missing parameter to debug message

Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
8 years agostaging: Improve file creation resiliance
Richard Purdie [Mon, 6 Feb 2017 14:47:55 +0000 (14:47 +0000)]
staging: Improve file creation resiliance

If you abort a build mid execution of the staging extend_sysroot function
there are currently races and cleanup of that function may fail.

This change splits the code into separate phases so that the manifests are
manipulated before files are installed, meaning we should be able to reverse
actions if builds fail, crash or are interrupted.

Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
8 years agostaging: Add shared manifest support
Richard Purdie [Sat, 4 Feb 2017 17:33:27 +0000 (17:33 +0000)]
staging: Add shared manifest support

Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
8 years agoperl-native: Remove .packlist file
Richard Purdie [Sat, 4 Feb 2017 17:31:17 +0000 (17:31 +0000)]
perl-native: Remove .packlist file

This is large and needs fixups to relocate it in each case. We can drop it, save
the work and the ~150MB disk space its various copies take up. Its not needed
for anything that I can see.

Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
8 years agostaging: Don't put ptest files into the sysroot
Richard Purdie [Sat, 4 Feb 2017 17:30:15 +0000 (17:30 +0000)]
staging: Don't put ptest files into the sysroot

This wastes space/time as the ptest files aren't going to be used in the
sysroot so add them to the blacklist.

Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
8 years agoglibc: Separate locale files to their own sstate task
Richard Purdie [Sat, 4 Feb 2017 17:28:18 +0000 (17:28 +0000)]
glibc: Separate locale files to their own sstate task

Putting the locale and script files into the sysroot for use by their
specific recipes used to be a simple way to share the files. With RSS,
we don't want to copy these into many different recipes so put these
files in their own sstate task.

Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
8 years agoRevert "yocto-bsps: add 4.9 bbappend"
Richard Purdie [Mon, 6 Feb 2017 14:45:31 +0000 (14:45 +0000)]
Revert "yocto-bsps: add 4.9 bbappend"

This reverts commit e024cb13eaf7d5bd484995390ba75f3ab85ec0ef as applied
to the wrong repo.

8 years agogcc-6: Sync gcc stddef.h with musl stddef.h
Khem Raj [Fri, 3 Feb 2017 22:45:26 +0000 (14:45 -0800)]
gcc-6: Sync gcc stddef.h with musl stddef.h

GCC provides some of std* headers including stddef.h
and it syncs with glibc definitions via __needed* defines
to find which datatypes are expected to be defined on top
of glibc. we need same for musl.

Drop unused 0048-ARM-PR-target-71056-Don-t-use-vectorized-builtins-wh.patch

Signed-off-by: Khem Raj <raj.khem@gmail.com>
Signed-off-by: Ross Burton <ross.burton@intel.com>
8 years agooeqa.buildperf: reword test descriptions
Markus Lehtonen [Fri, 3 Feb 2017 08:27:07 +0000 (10:27 +0200)]
oeqa.buildperf: reword test descriptions

In an attempt to make them more compact and coherent.

Signed-off-by: Markus Lehtonen <markus.lehtonen@linux.intel.com>
Signed-off-by: Ross Burton <ross.burton@intel.com>
8 years agosdkext/cases oeqa/selftest: Updates decorators
Francisco Pedraza [Mon, 30 Jan 2017 20:14:36 +0000 (14:14 -0600)]
sdkext/cases oeqa/selftest: Updates decorators

Updates decorators for eSDK in test cases at:
sdkext/cases/devtool.py and oeqa/selftest/eSDK.py

Signed-off-by: Francisco Pedraza <francisco.j.pedraza.gonzalez@intel.com>
Signed-off-by: Ross Burton <ross.burton@intel.com>
8 years agobash: use update-alternatives for the bash binary
Andreas Oberritter [Wed, 30 Nov 2016 18:50:26 +0000 (19:50 +0100)]
bash: use update-alternatives for the bash binary

Busybox may offer a bash applet. If enabled, the alternatives mechanism
avoids breaking logins if bash gets deinstalled while /bin/bash is configured
as a login shell.

Signed-off-by: Andreas Oberritter <obi@opendreambox.org>
Signed-off-by: Ross Burton <ross.burton@intel.com>
8 years agowic: fix call of serial_console_form_kargs
Ed Bartosh [Fri, 3 Feb 2017 15:54:40 +0000 (17:54 +0200)]
wic: fix call of serial_console_form_kargs

As syslinux module has been recently removed and
serial_consloe_form_kargs became local API in rootfs_pcbios_ext plugin
it should be called without syslinux. prefix.

Signed-off-by: Ed Bartosh <ed.bartosh@linux.intel.com>
Signed-off-by: Ross Burton <ross.burton@intel.com>
8 years agoselftest: wic: stop using iso image type
Ed Bartosh [Fri, 3 Feb 2017 15:22:08 +0000 (17:22 +0200)]
selftest: wic: stop using iso image type

Removed 'IMAGE_FSTYPES = "iso" as this functionality depends on
do_bootimg, which is going to be obsoleted soon.

As wic doesn't depend on bootimg it's safe to remove this.

Signed-off-by: Ed Bartosh <ed.bartosh@linux.intel.com>
Signed-off-by: Ross Burton <ross.burton@intel.com>
8 years agoselftest: wic: use wic-tools recipe to get STAGING_DIR
Ed Bartosh [Fri, 3 Feb 2017 14:58:15 +0000 (16:58 +0200)]
selftest: wic: use wic-tools recipe to get STAGING_DIR

STAGING_DIR variable is used to get path to a boot dir.
It's better to use wic-tools recipe to it as it contains
all bootloader artifacts.

Modified test_build_artifacts and test_rootfs_artifacts to
use wic-tools target to get STAGING_DIR.

Signed-off-by: Ed Bartosh <ed.bartosh@linux.intel.com>
Signed-off-by: Ross Burton <ross.burton@intel.com>
8 years agowic: isoimage-isohybrid: use wic-tools to get syslinux path
Ed Bartosh [Fri, 3 Feb 2017 13:49:56 +0000 (15:49 +0200)]
wic: isoimage-isohybrid: use wic-tools to get syslinux path

wic-tools recipe specific sysroot contains syslinux as wic-tools
depends on it. Used wic-tools target to get syslinux path should
guarantee that syslinux is installed there and can be used.

Signed-off-by: Ed Bartosh <ed.bartosh@linux.intel.com>
Signed-off-by: Ross Burton <ross.burton@intel.com>
8 years agoimage_types_wic: remove dependency to do_bootimg
Ed Bartosh [Thu, 2 Feb 2017 21:46:55 +0000 (23:46 +0200)]
image_types_wic: remove dependency to do_bootimg

Removing task dependency do_wic -> do_bootimg as wic
doesn't depend on hddimg/booimg anymore.

Signed-off-by: Ed Bartosh <ed.bartosh@linux.intel.com>
Signed-off-by: Ross Burton <ross.burton@intel.com>
8 years agoimage: rename image-wic -> image_types_wic
Ed Bartosh [Wed, 1 Feb 2017 15:27:49 +0000 (17:27 +0200)]
image: rename image-wic -> image_types_wic

Make name of the wic image type class consistent with
existing naming scheme for image types.

Signed-off-by: Ed Bartosh <ed.bartosh@linux.intel.com>
Signed-off-by: Ross Burton <ross.burton@intel.com>
8 years agolibgpg-error_1.25.bb: Enable nativesdk
Nathan Rossi [Mon, 30 Jan 2017 07:47:45 +0000 (17:47 +1000)]
libgpg-error_1.25.bb: Enable nativesdk

Enable nativesdk on this recipe.

Signed-off-by: Nathan Rossi <nathan@nathanrossi.com>
Signed-off-by: Ross Burton <ross.burton@intel.com>
8 years agolibgcrypt.inc: Add PACKAGECONFIG for 'capabilities'
Nathan Rossi [Mon, 30 Jan 2017 07:47:45 +0000 (17:47 +1000)]
libgcrypt.inc: Add PACKAGECONFIG for 'capabilities'

Add PACKAGECONFIG options for 'capabilities' which enables the libcap
dependency.

Signed-off-by: Nathan Rossi <nathan@nathanrossi.com>
Signed-off-by: Ross Burton <ross.burton@intel.com>
8 years agowic: flatten imager class hierarchy
Ed Bartosh [Fri, 3 Feb 2017 20:26:06 +0000 (22:26 +0200)]
wic: flatten imager class hierarchy

wic code is hard to follow due to deep and twiggy class
inheritance tree.

Flatten imager tree:
 wic -> wic_create -> Creator -> DirectPlugin -> DirectImageCreator
to
 wic -> wic_create -> DirectPlugin
by
 removing Creator class and creator module
 merging DirectImageCreator into DirectPlugin

Changed APIs to use the same parameters names.

Passed parsed command line options as an object down the stack.

Signed-off-by: Ed Bartosh <ed.bartosh@linux.intel.com>
Signed-off-by: Ross Burton <ross.burton@intel.com>
8 years agolibgcrypt.inc: Enable nativesdk
Nathan Rossi [Mon, 30 Jan 2017 07:47:45 +0000 (17:47 +1000)]
libgcrypt.inc: Enable nativesdk

Enable nativesdk on this recipe.

Signed-off-by: Nathan Rossi <nathan@nathanrossi.com>
Signed-off-by: Ross Burton <ross.burton@intel.com>
8 years agooeqa/buildproject: Run gnu-configize by default
Richard Purdie [Fri, 3 Feb 2017 10:32:40 +0000 (10:32 +0000)]
oeqa/buildproject: Run gnu-configize by default

The runtime oeqa tests were failing in the on-target case, only for
aarch64. This was because an old config.sub was being used.

Similar to the SDK testing, call gnu-configize in the on-target
case too to resolve the failing tests.

Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
8 years agooeqa/runtime: Improve failure log output
Richard Purdie [Fri, 3 Feb 2017 10:30:59 +0000 (10:30 +0000)]
oeqa/runtime: Improve failure log output

Printing a message which says "configure failed" without the log output
is effectively useless. If a command fails, print the output by default
and simplify the calling code which makes debugging any of these failures
much easier.

Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
8 years agogstreamer: update to 1.10.3
Nicolas Dechesne [Tue, 31 Jan 2017 21:03:03 +0000 (22:03 +0100)]
gstreamer: update to 1.10.3

New bugfix release:
* updated version/checksums
* removed 1 patch in -good which was merged upstream

Signed-off-by: Nicolas Dechesne <nicolas.dechesne@linaro.org>
Signed-off-by: Ross Burton <ross.burton@intel.com>
8 years agoopenssl: Updgrade 1.0.2j -> 1.0.2k
Andrej Valek [Thu, 2 Feb 2017 08:35:00 +0000 (09:35 +0100)]
openssl: Updgrade 1.0.2j -> 1.0.2k

Signed-off-by: Andrej Valek <andrej.valek@siemens.com>
Signed-off-by: Pascal Bach <pascal.bach@siemens.com>
Signed-off-by: Ross Burton <ross.burton@intel.com>
8 years agogdb 7.12: fix armv8b build
Koen Kooi [Thu, 2 Feb 2017 10:28:28 +0000 (11:28 +0100)]
gdb 7.12: fix armv8b build

Backport fix from GDB upstream to fix big-endian aarch64 build.

Signed-off-by: Koen Kooi <koen.kooi@linaro.org>
Signed-off-by: Ross Burton <ross.burton@intel.com>
8 years agoglib-2.0: native package should not depend on DISTRO_FEATURES
Gary Thomas [Sat, 21 Jan 2017 05:02:18 +0000 (06:02 +0100)]
glib-2.0: native package should not depend on DISTRO_FEATURES

xxx-native packages should not depend on ${DISTRO} settings.  Doing
so feels inherently wrong and limits the usefulness of sstate-cache.

This patch changes how this package is installed, in particular
removing the dependency on the ${DISTRO_FEATURES} variable in
glib-2.0-native.  This will further improve the ability to share
native packages between builds with differences in ${DISTRO_FEATURES}

Signed-off-by: Gary Thomas <gary@mlbassoc.com>
Signed-off-by: Ross Burton <ross.burton@intel.com>
8 years agoxserver-xorg: Fix X server 1.19 crash
Tom Hochstein [Wed, 1 Feb 2017 15:14:10 +0000 (09:14 -0600)]
xserver-xorg: Fix X server 1.19 crash

Commit 6a5a4e60373c1386b311b2a8bb666c32d68a9d99 removes the configure of
useSIGIO option:

xfree86 SIGIO support is reworked to use internal versions of
OsBlockSIGIO and OsReleaseSIGIO

The check for useSIGIO is no longer needed.

Signed-off-by: Tom Hochstein <tom.hochstein@nxp.com>
Signed-off-by: Ross Burton <ross.burton@intel.com>
8 years agoyocto-bsps: add 4.9 bbappend
Bruce Ashfield [Wed, 1 Feb 2017 15:43:19 +0000 (10:43 -0500)]
yocto-bsps: add 4.9 bbappend

Introduce a 4.9 bbappend for the h/w reference BSPs. Since 4.8 will
be removed, we need to have this file in place.

Signed-off-by: Bruce Ashfield <bruce.ashfield@windriver.com>
Signed-off-by: Ross Burton <ross.burton@intel.com>
8 years agolinux-yocto/4.9: yaffs2 fixes
Bruce Ashfield [Wed, 1 Feb 2017 15:30:36 +0000 (10:30 -0500)]
linux-yocto/4.9: yaffs2 fixes

Merging three fixes to yaffs2, which adjust to mainline changes in the
vfs subsystem:

  4700f2f8b9db fs: yaffs2: fix the prototype of function yaffs_rename()
  56e654cab1db fs: yaffs2: switch to the generic xattr handler
  102082f3c245 fs/yaffs2: adjust to the change of inode_change_ok()

Signed-off-by: Kevin Hao <kexin.hao@windriver.com>
Signed-off-by: Bruce Ashfield <bruce.ashfield@windriver.com>
Signed-off-by: Ross Burton <ross.burton@intel.com>
8 years agokernel-yocto: log the BSP definition file
Bruce Ashfield [Wed, 1 Feb 2017 15:30:35 +0000 (10:30 -0500)]
kernel-yocto: log the BSP definition file

When debugging a kernel configuration issue, one of the first questions
is "what BSP was used". To answer this qusetion, we log the BSP .scc
file that was used to generate the kernel configuration in the kernel
source meta directory.

Signed-off-by: Bruce Ashfield <bruce.ashfield@windriver.com>
Signed-off-by: Ross Burton <ross.burton@intel.com>
8 years agolinux-yocto/4.9: update to -rt3
Bruce Ashfield [Wed, 1 Feb 2017 15:30:34 +0000 (10:30 -0500)]
linux-yocto/4.9: update to -rt3

Paul Gortmaker has refreshed the 4.9 -rt support to -rt3.

Signed-off-by: Bruce Ashfield <bruce.ashfield@windriver.com>
Signed-off-by: Ross Burton <ross.burton@intel.com>
8 years agokernel-yocto/meta: common-pc: add pci-siov to feature fragments
Bruce Ashfield [Wed, 1 Feb 2017 15:30:33 +0000 (10:30 -0500)]
kernel-yocto/meta: common-pc: add pci-siov to feature fragments

The common-pc ethernet selection has drivers that depend on pci_iov
being defined. As such, we should include that feature fragment
or we get build warnings:

    ---------- CONFIG_BNX2X_SRIOV -----------------
    Config: CONFIG_BNX2X_SRIOV
    From:   work-shared/qemux86-64/kernel-source/.kernel-meta/configs/standard/bsp/common-pc/common-pc-eth.cfg
    Requested value:  CONFIG_BNX2X_SRIOV=y
    Actual value:

    Config 'BNX2X_SRIOV' has the following conditionals:
    BNX2X && PCI_IOV (value: "n")
    BNX2X && PCI_IOV (value: "n")
    Dependency values are:
      BNX2X [m] PCI_IOV [n] y [y]

    ---------- CONFIG_BNXT_SRIOV -----------------
    Config: CONFIG_BNXT_SRIOV
    From: work-shared/qemux86-64/kernel-source/.kernel-meta/configs/standard/bsp/common-pc/common-pc-eth.cfg
    Requested value:  CONFIG_BNXT_SRIOV=y
    Actual value:

    Config 'BNXT_SRIOV' has the following conditionals:
    BNXT && PCI_IOV (value: "n")
    BNXT && PCI_IOV (value: "n")
    Dependency values are:
      y [y] BNXT [m] PCI_IOV [n]

Signed-off-by: Bruce Ashfield <bruce.ashfield@windriver.com>
Signed-off-by: Ross Burton <ross.burton@intel.com>
8 years agolinux-yocto/4.9: update to v4.9.6
Bruce Ashfield [Wed, 1 Feb 2017 15:30:32 +0000 (10:30 -0500)]
linux-yocto/4.9: update to v4.9.6

Integrating the 4.9.5 and 4.9.6 -stable updates. The commit logs
are as follows:

   09f886dc5a69 Linux 4.9.6
   f77ef5348d4b libceph: stop allocating a new cipher on every crypto request
   5b482bf58868 libceph: uninline ceph_crypto_key_destroy()
   12274f2c17f2 tools/virtio/ringtest: fix run-on-all.sh for offline cpus
   fa555d021d2b selftest/powerpc: Wrong PMC initialized in pmc56_overflow test
   f37b7a3004bb soc: ti: wkup_m3_ipc: Fix error return code in wkup_m3_ipc_probe()
   97d5e2057564 spi: pxa2xx: add missed break
   d21814a8068a dmaengine: pl330: Fix runtime PM support for terminated transfers
   172270c74348 dmaengine: rcar-dmac: unmap slave resource when channel is freed
   3bef7578e05f s5p-mfc: Fix clock management in s5p_mfc_release() function
   d47e1e7c46fe s5p-cec: mark PM functions as __maybe_unused again
   dfe8e5730fa1 st-hva: fix some error handling in hva_hw_probe()
   b9dc16170dbf ite-cir: initialize use_demodulator before using it
   278997a8e002 gs1662: drop kfree for memory allocated with devm_kzalloc
   cec5ef6ac520 platform: pxa_camera: add VIDEO_V4L2 dependency
   2a3060531768 blackfin: check devm_pinctrl_get() for errors
   fcdab6ca9c31 rpmsg: virtio_rpmsg_bus: fix channel creation
   0f3418442d7b mtd: spi-nor: Fix some error codes in cqspi_setup_flash()
   e55e6c026b7c mtd: spi-nor: Off by one in cqspi_setup_flash()
   ebdfcaa14eef PM / devfreq: Fix the bug of devfreq_add_device when governor is NULL
   795983547317 PM / devfreq: exynos-bus: Fix the wrong return value
   16236802bfec scsi: mpt3sas: fix hang on ata passthrough commands
   a07a122ad2a2 scsi: ses: Fix SAS device detection in enclosure
   41c6b3e8989e swiotlb: Add swiotlb=noforce debug option
   1fd1e6cd6314 swiotlb: Convert swiotlb_force from int to enum
   776c2b2d165d arm64: Fix swiotlb fallback allocation
   962957889d74 arm64: mm: avoid name clash in __page_to_voff()
   d34b6684e60f xprtrdma: Squelch "max send, max recv" messages at connect time
   8ade1c2b4530 xprtrdma: Make FRWR send queue entry accounting more accurate
   a193c7247596 libceph: make sure ceph_aes_crypt() IV is aligned
   6e9fa67c58cc ceph: fix endianness bug in frag_tree_split_cmp
   2e4f2131b66f ceph: fix endianness of getattr mask in ceph_d_revalidate
   8934e069674a ceph: fix ceph_get_caps() interruption
   48baa924108e ceph: fix scheduler warning due to nested blocking
   04c9fe63166f ARM: 8613/1: Fix the uaccess crash on PB11MPCore
   dd8334a5e17e ARM: ux500: fix prcmu_is_cpu_in_wfi() calculation
   cd9601caa2fa ARM: dts: omap3: Fix Card Detect and Write Protect on Logic PD SOM-LV
   a075ac9c0a40 ARM: dts: imx6qdl-nitrogen6_max: fix sgtl5000 pinctrl init
   cfcb94b3a498 ARM: dts: omap2: Add an empty chosen node to top level DTSI
   5921b26bf744 ARM: dts: omap3: Add an empty chosen node to top level DTSI
   bec062cd47bd ARM: dts: am4372: Add an empty chosen node to top level DTSI
   c3f7ca43b2d4 ARM: dts: omap5: Add an empty chosen node to top level DTSI
   835bf872d924 ARM: dts: omap4: Add an empty chosen node to top level DTSI
   355a8fced2bf ARM: dts: am33xx: Add an empty chosen node to top level DTSI
   3e1c70972204 ARM: dts: dm814x: Add an empty chosen node to top level DTSI
   ab6dc01db1f7 ARM: dts: dm816x: Add an empty chosen node to top level DTSI
   d4f12aa133db ARM: dts: dra7: Add an empty chosen node to top level DTSI
   b8add6715c9a libceph: remove now unused ceph_*{en,de}crypt*() functions
   2982b9c92a66 libceph: switch ceph_x_decrypt() to ceph_crypt()
   717a145bd5a9 libceph: switch ceph_x_encrypt() to ceph_crypt()
   6e371f9a4144 libceph: tweak calcu_signature() a little
   788a0bbc7011 libceph: rename and align ceph_x_authorizer::reply_buf
   ecf7ced85628 libceph: introduce ceph_crypt() for in-place en/decryption
   0548b8298938 libceph: introduce ceph_x_encrypt_offset()
   be60457612a2 libceph: old_key in process_one_ticket() is redundant
   2e62bf3c6fe9 libceph: ceph_x_encrypt_buflen() takes in_len
   6d9b544d88a4 Input: ALPS - fix TrackStick support for SS5 hardware
   6e53a62a0d52 arm64/ptrace: Reject attempts to set incomplete hardware breakpoint fields
   f9081dd0c8be arm64/ptrace: Avoid uninitialised struct padding in fpr_set()
   5c5839be0842 arm64/ptrace: Preserve previous registers for short regset write - 3
   a4aafb8c4204 arm64/ptrace: Preserve previous registers for short regset write - 2
   357cfd6c83ee arm64/ptrace: Preserve previous registers for short regset write
   de327948c009 arm64: avoid returning from bad_mode
   71c496495514 ARM: dts: da850-evm: fix read access to SPI flash
   5b6618615215 ARM: dts: OMAP5 / DRA7: indicate that SATA port 0 is available.
   1f75575aca7b ceph: fix bad endianness handling in parse_reply_info_extra
   a14aeccb65e5 ibmvscsis: Fix max transfer length
   51cff2c64d20 ibmvscsis: Fix sleeping in interrupt context
   df35a8f51fcb ARM: 8634/1: hw_breakpoint: blacklist Scorpion CPUs
   73a2e2405d30 svcrdma: avoid duplicate dma unmapping during error recovery
   c49b31b585a1 clocksource/exynos_mct: Clear interrupt when cpu is shut down
   ce5c52f039cf ubifs: Fix journal replay wrt. xattr nodes
   bab10a549fd6 mac80211: implement multicast forwarding on fast-RX path
   beecb1e72cc5 qla2xxx: Fix crash due to null pointer access
   33038189bcab x86/ioapic: Restore IO-APIC irq_chip retrigger callback
   bf15e45eee7a powerpc: Ignore reserved field in DCSR and PVR reads and writes
   e09778546713 powerpc/ptrace: Preserve previous TM fprs/vsrs on short regset write
   9e10a0fc7312 powerpc/ptrace: Preserve previous fprs/vsrs on short regset write
   98e60a8345ae powerpc/perf: Fix PM_BRU_CMPL event code for power9
   09187cdf50c1 powerpc/icp-opal: Fix missing KVM case and harden replay
   26c4d513b6af KVM: arm/arm64: vgic: Fix deadlock on error handling
   e0641f201114 KVM: s390: do not expose random data via facility bitmap
   001ed27d088a mtd: nand: xway: fix build because of module functions
   c8dce6e0ef2c mtd: nand: xway: disable module support
   d8d82ac42e27 mtd: nand: lpc32xx: fix invalid error handling of a requested irq
   86249aef233e ieee802154: atusb: do not use the stack for buffers to make them DMA able
   ad17175732ca mmc: mxs-mmc: Fix additional cycles after transmission stop
   c1274eeb2fe3 mmc: sdhci-acpi: Only powered up enabled acpi child devices
   35028599bc88 HID: corsair: fix control-transfer error handling
   63df1ccb937f HID: corsair: fix DMA buffers on stack
   4922a6a5cfa7 PCI: Enumerate switches below PCI-to-PCIe bridges
   c6bdc450fe93 PCI: designware: Check for iATU unroll only on platforms that use ATU
   07f0267561c6 fuse: fix time_to_jiffies nsec sanity check
   0181b3603681 fuse: clear FR_PENDING flag when moving requests out of pending queue
   bb82fb48df8c ARC: module: Fix !CONFIG_ARC_DW2_UNWIND builds
   3c4d83a1a41e libnvdimm, namespace: fix pmem namespace leak, delete when size set to zero
   f29f3616b932 svcrpc: don't leak contexts on PROC_DESTROY
   a297ed84b928 sunrpc: don't call sleeping functions from the notifier block callbacks
   90687fc3c8c3 rcu: Narrow early boot window of illegal synchronous grace periods
   bdeaa468e221 rcu: Remove cond_resched() from Tiny synchronize_sched()
   f779e00d2e86 x86/PCI: Ignore _CRS on Supermicro X8DTH-i/6/iF/6F
   782b361c9306 tmpfs: clear S_ISGID when setting posix ACLs
   9170948a24b7 ARM: dts: omap3: Add DTS for Logic PD SOM-LV 37xx Dev Kit
   efba7a37efe2 ARM: dts: imx31: fix AVIC base address
   173470b1b10a ARM: dts: imx31: move CCM device node to AIPS2 bus devices
   979f63227c98 ARM: dts: imx31: fix clock control module interrupts description
   e95bd2ca11a9 ARM: dts: imx6q-cm-fx6: fix fec pinctrl
   387812143cbe ARM: dts: r8a7794: remove Z clock
   57a10f29e0b3 ARM: dts: r8a7794: Use SYSC "always-on" PM Domain for sound
   97136d1ef47a ARM: dts: bcm283x: fix typo in mailbox address
   03f5be20ec9b perf jit: Enable jitdump support without dwarf
   3ef6a3c0136b perf scripting: Avoid leaking the scripting_context variable
   ea738532666a perf callchain: Fixup help/config for no-unwinding
   a1d308914090 perf diff: Do not overwrite valid build id
   0f242c0ae0a6 perf trace: Check if MAP_32BIT is defined (again)
   8b5484fab145 perf mem: Fix --all-user/--all-kernel options
   567eea578016 perf trace: Use the syscall raw_syscalls:sys_enter timestamp
   d622b626270a IB/IPoIB: Remove can't use GFP_NOIO warning
   656a7d657533 IB/mlx4: Check if GRH is available before using it
   661600ff77f1 IB/mlx4: When no DMFS for IPoIB, don't allow NET_IF QPs
   8f95735412ad IB/mlx4: Fix port query for 56Gb Ethernet links
   1039b7f5c7ef IB/mlx4: Handle well-known-gid in mad_demux processing
   618d14b01a45 IB/mlx4: Fix out-of-range array index in destroy qp flow
   970f6806c2b1 IB/mlx4: Set traffic class in AH
   0d64860aa638 IB/mlx5: Wait for all async command completions to complete
   7c8441c89043 IB/mlx5: Assign SRQ type earlier
   dae9f4f14411 IB/mlx5: Fix reported max SGE calculation
   4db097fe49ea IB/mlx5: Avoid system crash when enabling many VFs
   45f3a7eb2ca0 IB/rxe: avoid putting a large struct rxe_qp on stack
   25c72b518734 IB/rxe: Increase max number of completions to 32k
   b7c551559834 IB/core: Release allocated memory in cache setup failure
   40bf0662fe3f Linux 4.9.5
   4e764538512e pinctrl: sh-pfc: Do not unconditionally support PIN_CONFIG_BIAS_DISABLE
   b90a617fef04 arm64: hugetlb: fix the wrong return value for huge_ptep_set_access_flags
   2021e55d719a arm64: hugetlb: remove the wrong pmd check in find_num_contig()
   736aec92c4e9 arm64: hugetlb: fix the wrong address for several functions
   3083593700dd powerpc/powernv: Don't warn on PE init if unfreeze is unsupported
   dcd15c20c208 powerpc/ibmebus: Fix device reference leaks in sysfs interface
   25bba7152f2a powerpc/ibmebus: Fix further device reference leaks
   ae4c24b2d584 powerpc/mm: Correct process and partition table max size
   42eded92c42b bus: vexpress-config: fix device reference leak
   d9c19f90f3a5 blk-mq: Always schedule hctx->next_cpu
   cb50d45c3d42 power: supply: bq27xxx_battery: Fix register map for BQ27510 and BQ27520
   df386a0c2804 bq24190_charger: Fix PM runtime use for bq24190_battery_set_property
   4b07d85374f5 iw_cxgb4: Fix error return code in c4iw_rdev_open()
   aa3ee06a2ae2 powercap/intel_rapl: fix and tidy up error handling
   f53cfb081f00 ACPI / APEI: Fix NMI notification handling
   f57d8710096d block: cfq_cpd_alloc() should use @gfp
   f99694cdaf76 block: Change extern inline to static inline
   a4e73cc229d8 ACPI / CPPC: set an error code on probe error path
   26991b28b47c regulators: helpers: Fix handling of bypass_val_on in get_bypass_regmap
   89c728ed9237 cpufreq: powernv: Disable preemption while checking CPU throttling state
   e5f33ef0a601 powerpc/64: Simplify adaptation to new ISA v3.00 HPTE format
   f607fd883654 remoteproc: st: Fix error return code in st_rproc_probe()
   0e38a0b7acbb remoteproc: qcom_wcnss: Fix circular module dependency
   b4a69e3877ca drm: Initialise drm_mm.head_node.allocated
   5bdb57418845 drm/i915: Move the min_pixclk[] handling to the end of readout
   699fbc4cb7d4 drm/panel: simple: Check against num_timings when setting preferred for timing
   214417a0419b drm: avoid uninitialized timestamp use in wait_vblank
   3b376640445a drm/i915/gen9: Fix PCODE polling during SAGV disabling
   26eae206b786 i2c: mux: pca954x: fix i2c mux selection caching
   396b25173370 NFSv4.1: nfs4_fl_prepare_ds must be careful about reporting success.
   e331f2f2b1fb NFS: Fix a performance regression in readdir
   4c4d4bec6c90 pNFS: Fix race in pnfs_wait_on_layoutreturn
   633b57037699 NFS: fix typo in parameter description
   a5209aa99042 pinctrl: meson: fix gpio request disabling other modes
   7a1dcd92fd39 btrfs: fix error handling when run_delayed_extent_op fails
   003e3163fcc7 btrfs: fix locking when we put back a delayed ref that's too new
   0aefd99f37f3 nvme: apply DELAY_BEFORE_CHK_RDY quirk at probe time too
   9bae3c370a9e x86/cpu: Fix bootup crashes by sanitizing the argument of the 'clearcpuid=' command-line option
   64e236812197 i2c: piix4: Avoid race conditions with IMC
   ce31072b4342 net/mlx5: Only cancel recovery work when cleaning up device
   2f946414a080 USB: serial: ch341: fix modem-control and B0 handling
   12a26bd342b5 drm/amdgpu: drop verde dpm quirks
   d52e714b39ce drm/amdgpu: update si kicker smc firmware
   0c37b07002d5 drm/radeon: drop verde dpm quirks
   7cd7aea3dc78 drm/radeon: update smc firmware selection for SI
   5375b71f3b17 drm: Clean up planes in atomic commit helper failure path
   6d15a83af982 drm/i915/gen9: Fix PCODE polling timeout in stable backport
   9297e0c189f5 net/af_iucv: don't use paged skbs for TX on HiperSockets
   00cf64fbaa1e sysctl: Drop reference added by grab_header in proc_sys_readdir
   eca02f01be96 Clearing FIFOs in RS485 emulation mode causes subsequent transmits to break
   86820a103f51 extcon: return error code on failure
   ef8ee4495171 sysrq: attach sysrq handler correctly for 32-bit kernel
   4a940d6503e1 orinoco: Use shash instead of ahash for MIC calculations
   0edcc47cdb7f ibmvscsis: Fix srp_transfer_data fail return code
   ba04d869974e tty/serial: atmel_serial: BUG: stop DMA from transmitting in stop_tx
   2d789bd1fc24 tty/serial: atmel: RS485 half duplex w/DMA: enable RX after TX is done
   199c89fd32b2 virtio_blk: avoid DMA to stack for the sense buffer
   6c6ae8c43df9 dmaengine: omap-dma: Fix dynamic lch_map allocation
   3fbaff3adc76 drivers: char: mem: Fix thinkos in kmem address checks
   1a62a0f76556 mnt: Protect the mountpoint hashtable with mount_lock
   52fd0ab07676 pid: fix lockdep deadlock warning due to ucount_lock
   57bfd5a3710b vme: Fix wrong pointer utilization in ca91cx42_slave_get
   1f363639eb30 Revert "tty: serial: 8250: add CON_CONSDEV to flags"
   f9cf776b0555 ASoC: hdmi-codec: use unsigned type to structure members with bit-field
   28dad9aa9b36 btrfs: fix crash when tracepoint arguments are freed by wq callbacks
   4d0f302bf56a xhci: fix deadlock at host remove by running watchdog correctly
   d06367ac1730 fix a fencepost error in pipe_advance()
   ab8957396a69 i2c: fix kernel memory disclosure in dev interface
   93c94ec23fae i2c: print correct device invalid address
   61a8c3372adb Input: elants_i2c - avoid divide by 0 errors on bad touchscreen data
   0556a65e8b7d USB: serial: ch341: fix open and resume after B0
   3ed1f6da3a17 USB: serial: ch341: fix control-message error handling
   139556a98511 USB: serial: ch341: fix open error handling
   1685daad0b0c USB: serial: ch341: fix resume after reset
   4aeab97a0515 USB: serial: ch341: fix initial modem-control state
   58ede4beda66 USB: serial: kl5kusb105: fix line-state error handling
   dfd48efcde84 usb: musb: fix runtime PM in debugfs
   88d3670a1de4 wusbcore: Fix one more crypto-on-the-stack bug
   99ff99b830c1 x86/CPU/AMD: Fix Bulldozer topology
   bd7e769457f9 x86/bugs: Separate AMD E400 erratum and C1E bug
   e2d9ad2c540b x86/cpu/AMD: Clean up cpu_llc_id assignment per topology feature
   259495a0440f bridge: netfilter: Fix dropping packets that moving through bridge interface
   6ba35da690f3 xfs: Timely free truncated dirty pages
   86673e9331c9 gpio: Move freeing of GPIO hogs before numbing of the device
   0a28f5393689 nl80211: fix sched scan netlink socket owner destruction
   14d6c966744d x86/efi: Don't allocate memmap through memblock after mm_init()
   99b17ac0014b efi/x86: Prune invalid memory map entries and fix boot regression
   74ce3fd64bc4 efi/libstub/arm*: Pass latest memory map to the kernel
   736e77c07fba KVM: x86: Introduce segmented_write_std
   83fedbb76051 KVM: x86: emulate FXSAVE and FXRSTOR
   aae8f3464b1f KVM: x86: add asm_safe wrapper
   bc5e1316efd6 KVM: x86: add Align16 instruction flag
   90f70fcd6f72 KVM: x86: fix NULL deref in vcpu_scan_ioapic
   5ed21cc0cf26 KVM: x86: flush pending lapic jump label updates on module unload
   483ecebb22c1 jump_labels: API for flushing deferred jump label updates
   7caf473f99b8 KVM: eventfd: fix NULL deref irqbypass consumer
   7718ffcf9a64 KVM: x86: fix emulation of "MOV SS, null selector"
   1e26cec60668 mm/hugetlb.c: fix reservation race when freeing surplus pages
   8315c22ea879 mm/slab.c: fix SLAB freelist randomization duplicate entries
   6ca29ee3ca0d mm: support anonymous stable page
   07fc9575e88a mm, memcg: fix the active list aging for lowmem requests when memcg is enabled
   6c9bd81cb9ea ocfs2: fix crash caused by stale lvb with fsdlm plugin
   692755b1006d mm: fix devm_memremap_pages crash, use mem_hotplug_{begin, done}
   8edd365ee94c mm: pmd dirty emulation in page fault handler
   87fa6f37fa29 dax: fix deadlock with DAX 4k holes
   2e264fb546fa zram: support BDI_CAP_STABLE_WRITES
   ad4764b4c8eb zram: revalidate disk under init_lock
   057ac4429aef selftests: do not require bash for the generated test
   91ee732cb563 selftests: do not require bash to run netsocktests testcase
   d65146c7fb91 drm/savage: dereferencing an error pointer
   c730a84aff6f drm/vc4: Fix a couple error codes in vc4_cl_lookup_bos()
   a63bb198dc14 drm/tegra: dpaux: Fix error handling
   6b94626c9edf regulator: axp20x: Fix axp809 ldo_io registration error on cold boot
   8ac055af47ae regulator: tps65086: Fix 25mV ranges for BUCK regulators
   9229336861cb pinctrl: sh-pfc: Add helper to handle bias lookup table
   5e1595223a2c pinctrl: sh-pfc: r8a7795: Use lookup function for bias data
   b01bbf221fcc pinctrl: imx: fix imx_pinctrl_desc initialization
   f34fbb92fe81 Input: i8042 - add Pegatron touchpad to noloop table
   5975358bf2ad Input: xpad - use correct product id for x360w controllers

Signed-off-by: Bruce Ashfield <bruce.ashfield@windriver.com>
Signed-off-by: Ross Burton <ross.burton@intel.com>
8 years agover_linux: Use /usr/bin/awk instead of /bin/awk
Bruce Ashfield [Wed, 1 Feb 2017 15:30:31 +0000 (10:30 -0500)]
ver_linux: Use /usr/bin/awk instead of /bin/awk

To avoid kernel-devsrc failing with missing a dependency on "/bin/awk".
Due to the way this script is invoked, using #!/usr/bin/env can run into
issue when invoked.

Since most distros have awk in /usr/bin and not /bin, we change the
script while this is sorted out upstream.

Signed-off-by: Saul Wold <sgw@linux.intel.com>
Signed-off-by: Bruce Ashfield <bruce.ashfield@windriver.com>
Signed-off-by: Ross Burton <ross.burton@intel.com>
8 years agokernel-yocto: ecryptfs, NFC and CAN bus config updates
Bruce Ashfield [Wed, 1 Feb 2017 15:30:30 +0000 (10:30 -0500)]
kernel-yocto: ecryptfs, NFC and CAN bus config updates

Integrating the following kernel config updates:

  f7f388ec4d89 Add support and drivers for CAN bus as feature
  2b20935eb14b Filesystem encryption support
  8520e18f2956 Update NFC support
  a079d66845cd Add eCryptFS filesystem feature

Signed-off-by: Jussi Laako <jussi.laako@linux.intel.com>
Signed-off-by: Bruce Ashfield <bruce.ashfield@windriver.com>
Signed-off-by: Ross Burton <ross.burton@intel.com>
8 years agokern-tools: re-enable scc merge command
Bruce Ashfield [Wed, 1 Feb 2017 15:30:29 +0000 (10:30 -0500)]
kern-tools: re-enable scc merge command

The ability to merge two branches directly from a .scc file was
dropped during the streamlining of the tools.

As was pointed out by David Vincent <freesilicon@gmail.com>, there is
once again a valid use case for this functionality, so we restore the
capability.

Signed-off-by: Bruce Ashfield <bruce.ashfield@windriver.com>
Signed-off-by: Ross Burton <ross.burton@intel.com>
8 years agokernel-yocto/features: enable TPM
Bruce Ashfield [Wed, 1 Feb 2017 15:30:28 +0000 (10:30 -0500)]
kernel-yocto/features: enable TPM

Adding a feature fragment that enables a broad range of TPM
drivers. They service as a baseline for production kernels.

Signed-off-by: Patrick Ohly <patrick.ohly@intel.com>
Signed-off-by: Bruce Ashfield <bruce.ashfield@windriver.com>
Signed-off-by: Ross Burton <ross.burton@intel.com>
8 years agolinux-yocto/4.9: update to 4.9.4
Bruce Ashfield [Wed, 1 Feb 2017 15:30:27 +0000 (10:30 -0500)]
linux-yocto/4.9: update to 4.9.4

Integrating the korg -stable release with the following changes:

   75353ac8ff43 Linux 4.9.4
   6fea974494af rtlwifi: rtl_usb: Fix missing entry in USB driver's private data
   7dae85b5c355 rtlwifi: Fix enter/exit power_save
   3f41ee3a45cb drm/i915/gen9: Fix PCODE polling during CDCLK change notification
   93f2976eb027 ALSA: usb-audio: Add a quirk for Plantronics BT600
   f52e670a5b29 spi: mvebu: fix baudrate calculation for armada variant
   05b7bdf1c3d8 ARM: omap2+: am437x: rollback to use omap3_gptimer_timer_init()
   b8ba5faa7a6b ARM: 8631/1: clkdev: Detect errors in clk_hw_register_clkdev() for mass registration
   87dbf3dc1652 ARM: OMAP4+: Fix bad fallthrough for cpuidle
   b336dc57bc92 ARM: OMAP5: Fix build for PM code
   0f665deba9bc ARM: OMAP5: Fix mpuss_early_init
   aa1c7b01c9c7 bus: arm-ccn: Prevent hotplug callback leak
   bd99e7a6036e svcrdma: Clear xpt_bc_xps in xprt_setup_rdma_bc() error exit arm
   c2ce1c4133b3 ARM: qcom_defconfig: Fix MDM9515 LCC and GCC config
   e925eb342659 ARM: zynq: Reserve correct amount of non-DMA RAM
   78e2d9405e2d ARM: pxa: fix pxa25x interrupt init
   596ff0afbe8e ARM64: dts: bcm2835: Fix bcm2837 compatible string
   e3937bc1cc0b ARM64: dts: bcm2837-rpi-3-b: remove incorrect pwr LED
   d40152d5ac67 arm64: dts: mt8173: Fix auxadc node
   08aed6e8883d tools/virtio: fix READ_ONCE()
   e7d05ec1923e powerpc: Fix build warning on 32-bit PPC
   2fc33ff4ba81 ALSA: firewire-tascam: Fix to handle error from initialization of stream data
   2c867216c555 HID: hid-cypress: validate length of report
   e425ed1d3c75 net: vrf: do not allow table id 0
   7b7a5a85b1d9 net: ipv4: Fix multipath selection with vrf
   7cc73483a4c7 net/mlx5e: Remove WARN_ONCE from adaptive moderation code
   17a561b19a27 gro: Disable frag0 optimization on IPv6 ext headers
   934ca017c850 gro: use min_t() in skb_gro_reset_offset()
   ec0fdcb88c6f gro: Enter slow-path if there is no tailroom
   33364eee1fe4 net: add the AF_QIPCRTR entries to family name tables
   2ff4a0243c9e net: dsa: Ensure validity of dst->ds[0]
   66f24d624baa r8152: fix rx issue for runtime suspend
   c8a89b4f5248 r8152: split rtl8152_suspend function
   294f2c889637 net: dsa: bcm_sf2: Utilize nested MDIO read/write
   ac77aab46168 net: dsa: bcm_sf2: Do not clobber b53_switch_ops
   b55f6ca7380d bpf: change back to orig prog on too many passes
   a4d205a59521 net: vrf: Add missing Rx counters
   efc455f08ea8 ipv4: Do not allow MAIN to be alias for new LOCAL w/ custom rules
   fe1e13cfe2c4 igmp: Make igmp group member RFC 3376 compliant
   7826d11cf44c flow_dissector: Update pptp handling to avoid null pointer deref.
   9f65f5d4746b drop_monitor: consider inserted data in genlmsg_end
   9f7551e05b0f drop_monitor: add missing call to genlmsg_end
   a8a213f296ae net: ipv4: dst for local input routes should use l3mdev if relevant
   e7422080e35d net: fix incorrect original ingress device index in PKTINFO
   2ffc694b5727 rtnl: stats - add missing netlink message size checks
   8cb7d6277f01 net/mlx5e: Disable netdev after close
   ee9f2fd3f6b6 net/mlx5e: Don't sync netdev state when not registered
   33c782dd1514 net/mlx5: Prevent setting multicast macs for VFs
   b22c86ff8e78 net/mlx5: Mask destination mac value in ethtool steering rules
   efbbc75c00fc net/mlx5: Avoid shadowing numa_node
   ca8a64467f2a net/mlx5: Cancel recovery work in remove flow
   7bf1de7f2749 net/mlx5: Check FW limitations on log_max_qp before setting it
   9b4a34ff8987 net/sched: cls_flower: Fix missing addr_type in classify
   99f40c6bf565 net: stmmac: Fix race between stmmac_drv_probe and stmmac_open
   09babe4ce12e net, sched: fix soft lockup in tc_classify
   ee99e2bc5e8a ipv6: handle -EFAULT from skb_copy_bits
   d36a1cb1e328 inet: fix IP(V6)_RECVORIGDSTADDR for udp sockets
   ed3cc329c7bc sctp: sctp_transport_lookup_process should rcu_read_unlock when transport is null
   8b8fbe5c25ab net: vrf: Drop conntrack data after pass through VRF device on Tx
   d4a0b2e40c46 net: vrf: Fix NAT within a VRF

Signed-off-by: Bruce Ashfield <bruce.ashfield@windriver.com>
Signed-off-by: Ross Burton <ross.burton@intel.com>
8 years agomesa: update to 13.0.4
Nicolas Dechesne [Wed, 1 Feb 2017 13:56:53 +0000 (14:56 +0100)]
mesa: update to 13.0.4

Bug fixes release.

This is really needed for Qualcomm based h/w since freedreno
driver had a regression in 13.0.x branch which is fixed in 13.0.4.

Signed-off-by: Nicolas Dechesne <nicolas.dechesne@linaro.org>
Signed-off-by: Ross Burton <ross.burton@intel.com>
8 years agoqemu: Upgrade to 2.8.0
Aníbal Limón [Tue, 31 Jan 2017 22:10:24 +0000 (16:10 -0600)]
qemu: Upgrade to 2.8.0

Added patches:

- target-ppc-fix-user-mode.patch

Rebased patches:

- exclude-some-arm-EABI-obsolete-syscalls.patc

Removed patches (already in upstream):

- 0003-fix-CVE-2016-7908.patch
- 0004-fix-CVE-2016-7909.patch
- 0001-target-mips-add-24KEc-CPU-definition.patch

Changelog,

http://wiki.qemu.org/ChangeLog/2.8

Signed-off-by: Aníbal Limón <anibal.limon@linux.intel.com>
Signed-off-by: Ross Burton <ross.burton@intel.com>
8 years agolibtasn1: Upgrade 4.9 -> 4.10
Maxin B. John [Mon, 30 Jan 2017 11:55:58 +0000 (13:55 +0200)]
libtasn1: Upgrade 4.9 -> 4.10

Removed the following Backported patches:
 1. 0001-configure-don-t-add-Werror-to-build-flags.patch
 2. 0002-ASN.y-corrected-compiler-warning.patch
 3. 0003-parser_aux-corrected-potential-null-pointer-derefere.patch
 4. 0004-tools-eliminated-compiler-warnings.patch

fixed the following build error with musl
...
|                  from ../../libtasn1-4.10/gl/getopt.c:28:
| ./stdint.h:89:5: error: #if with no expression
|  #if
|      ^

Signed-off-by: Maxin B. John <maxin.john@intel.com>
Signed-off-by: Ross Burton <ross.burton@intel.com>
8 years agoffmpeg: disable cpu autodetection
Andreas Oberritter [Mon, 30 Jan 2017 21:59:50 +0000 (22:59 +0100)]
ffmpeg: disable cpu autodetection

Just pass the correct -mcpu= configure switch (fallback to generic).

Avoids warnings like:
| warning: switch -mcpu=cortex-a15 conflicts with -march=armv7-a switch

Signed-off-by: Andreas Oberritter <obi@opendreambox.org>
Signed-off-by: Ross Burton <ross.burton@intel.com>
8 years agopython-3-manifest: split out typing
Anders Darander [Fri, 27 Jan 2017 08:18:38 +0000 (09:18 +0100)]
python-3-manifest: split out typing

This allows us to use typing.py without having to add the whole
python3-misc package.

Signed-off-by: Anders Darander <anders@chargestorm.se>
Signed-off-by: Ross Burton <ross.burton@intel.com>
8 years agopython-3-manifest: add ipaddress to python3-io
Anders Darander [Fri, 27 Jan 2017 08:09:43 +0000 (09:09 +0100)]
python-3-manifest: add ipaddress to python3-io

This allows us to use ipaddress without requiring the add the whole
python3-misc.

Signed-off-by: Anders Darander <anders@chargestorm.se>
Signed-off-by: Ross Burton <ross.burton@intel.com>
8 years agopython-3-manifest: add _compat_pickle to python3-pickle
Anders Darander [Fri, 27 Jan 2017 08:07:17 +0000 (09:07 +0100)]
python-3-manifest: add _compat_pickle to python3-pickle

This allows us to depend on _compat_pickle.* wihtout having to add the whole
python3-misc.

Signed-off-by: Anders Darander <anders@chargestorm.se>
Signed-off-by: Ross Burton <ross.burton@intel.com>
8 years agopython-3-manifest: add socketserver to netserver
Anders Darander [Fri, 27 Jan 2017 08:04:29 +0000 (09:04 +0100)]
python-3-manifest: add socketserver to netserver

socketserver.* should be part of python3-netserver.

Signed-off-by: Anders Darander <anders@chargestorm.se>
Signed-off-by: Ross Burton <ross.burton@intel.com>
8 years agopython-3-manifest: add argparse to RDEPENDS for netclient
Anders Darander [Fri, 27 Jan 2017 08:02:38 +0000 (09:02 +0100)]
python-3-manifest: add argparse to RDEPENDS for netclient

http/server.py requires argparse.

Signed-off-by: Anders Darander <anders@chargestorm.se>
Signed-off-by: Ross Burton <ross.burton@intel.com>
8 years agopython3-manifest: move htlm.py to python3-html
Anders Darander [Fri, 27 Jan 2017 07:55:59 +0000 (08:55 +0100)]
python3-manifest: move htlm.py to python3-html

This allows us to use html.py without importing misc.

Signed-off-by: Anders Darander <anders@chargestorm.se>
Signed-off-by: Ross Burton <ross.burton@intel.com>
8 years agopython-3-manifest: fix adding imp to importlib
Anders Darander [Fri, 27 Jan 2017 07:50:59 +0000 (08:50 +0100)]
python-3-manifest: fix adding imp to importlib

Commit: 512334f102a33833d39af53467894315f0715d07
"python-3.5-manifest: Add imp to importlib"

added imp to importlib in the generated manifest, but not in the generator script.

Signed-off-by: Anders Darander <anders@chargestorm.se>
Signed-off-by: Ross Burton <ross.burton@intel.com>
8 years agolibbsd: libbsd-overlay.pc: Set Cflags to use -I instead of -isystem
Fabio Berton [Mon, 30 Jan 2017 16:27:00 +0000 (14:27 -0200)]
libbsd: libbsd-overlay.pc: Set Cflags to use -I instead of -isystem

pkg-config currently only handles -I and -L correctly, but misses
-isystem, so we need to do this workaround to fix this issue.

There's a open bug in:
https://bugzilla.freedesktop.org/show_bug.cgi?id=99516

This appears when building lldpd and shows error:

ERROR: lldpd-0.9.6-r0 do_configure: This autoconf log indicates
errors, it looked at host include and/or library paths while
determining system capabilities.
Rerun configure task after fixing this.
ERROR: lldpd-0.9.6-r0 do_configure: Function failed: do_qa_configure

Change-Id: I83255b5952e19bb2f516b1186276cd03eb67eed3
Signed-off-by: Fabio Berton <fabio.berton@ossystems.com.br>
Signed-off-by: Ross Burton <ross.burton@intel.com>
8 years agolibsolv: split libsolvext into it's own pkg-config
Alejandro del Castillo [Fri, 27 Jan 2017 22:38:57 +0000 (16:38 -0600)]
libsolv: split libsolvext into it's own pkg-config

Opkg only depends on libsolv proper, splitting libsolv-ext pkg-config
reduces opkg dependencies.

Signed-off-by: Alejandro del Castillo <alejandro.delcastillo@ni.com>
Signed-off-by: Ross Burton <ross.burton@intel.com>
8 years agoxserver-xorg: fix 16bit modes
Andreas Müller [Thu, 26 Jan 2017 15:33:56 +0000 (16:33 +0100)]
xserver-xorg: fix 16bit modes

Discovered on beaglebone black

Signed-off-by: Andreas Müller <schnitzeltony@googlemail.com>
Signed-off-by: Ross Burton <ross.burton@intel.com>
8 years agopkgconfig: fix typo introduced during recent conversion to PACKAGECONFIG
Andre McCurdy [Thu, 2 Feb 2017 20:22:26 +0000 (12:22 -0800)]
pkgconfig: fix typo introduced during recent conversion to PACKAGECONFIG

Signed-off-by: Andre McCurdy <armccurdy@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
8 years agoeudev: add RPROVIDES so eudev-hwdb provides udev-hwdb
Ross Burton [Thu, 2 Feb 2017 18:33:19 +0000 (18:33 +0000)]
eudev: add RPROVIDES so eudev-hwdb provides udev-hwdb

Otherwise the common name udev-hwdb is only provided by systemd, meaning that
other recipes can't depend on a single name.

Signed-off-by: Ross Burton <ross.burton@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
8 years agouninative: Make patchelf modified files sparse
Richard Purdie [Thu, 2 Feb 2017 14:38:13 +0000 (14:38 +0000)]
uninative: Make patchelf modified files sparse

When we switched to recipe specific sysroots (rss), performance took a nose dive. Its
easy to blame rss but it turns out not to be entirely at fault.

Three configurations are compared here:

a) Pre-RSS (revision 45df694a9f472ac2f684aadac4d864c3dfdc48a7)
b) Post-RSS (revision 226a508da955439b881b2f0a544a3aee76e59919)
c) as b) with this change

Overall build times:

a) 22794.25user 2687.88system 30:32.84elapsed 1390%CPU (0avgtext+0avgdata 919056maxresident)k
b) 22677.25user 3238.79system 36:16.68elapsed 1190%CPU (0avgtext+0avgdata 918896maxresident)k
c) 23571.84user 3383.65system 31:36.83elapsed 1421%CPU (0avgtext+0avgdata 919068maxresident)k

For the overall build and sstate directories, du -s shows:
a)
3992588   build-pre-rss/sstate-cache
30804484  build-pre-rss/tmp
b)
4013272   build-with-rss/sstate-cache
36519084  build-with-rss/tmp
c)
4014744   build-with-rss2/sstate-cache
35336960  build-with-rss2/tmp

However more worryingly:

$ du -s build-pre-rss/tmp/sysroots/
2506092 build-pre-rss/tmp/sysroots/
$ du -s build-with-rss/tmp/sysroots-components/
3790712 build-with-rss/tmp/sysroots-components/
$ du -s build-with-rss2/tmp/sysroots-components/
2467544 build-with-rss2/tmp/sysroots-components/

These numbers *should* be equivalent but as you can see, b) is ~1.2GB larger. The reason turned out
to be patchelf. Taking a specific binary from a specific recipe, bc from bc-native, in a) its 82kb
(stripped) yet in b) its 2.17MB.

$ ./patchelf --set-interpreter /bin/rp bc
warning: working around a Linux kernel bug by creating a hole of 2084864 bytes in ‘bc’

https://github.com/NixOS/patchelf/blob/master/src/patchelf.cc#L710 shows that this "hole" is just
padded zeros using memset, its not a proper sparse hole.

This patch copies files with cp --sparse=always after modifying them with patchelf, then replacing
the original file. The better fix will be to fix this in patchself itself and seek() there
when writing the new file but that means new uninative tarballs and will take a bit of work
so I'm proposing this workaround in the meantime.

Also, this patch drops error handling since subprocess check_output() tracebacks will print this
information if the command fails so we can simplify the code.

8 years agowic: get rid of image_output_dir variable
Ed Bartosh [Thu, 2 Feb 2017 13:32:59 +0000 (15:32 +0200)]
wic: get rid of image_output_dir variable

Used options.outdir instead of image_output_dir.
There is no sense to use extra variable for this.

Signed-off-by: Ed Bartosh <ed.bartosh@linux.intel.com>
Signed-off-by: Ross Burton <ross.burton@intel.com>
8 years agowic: direct: fix creation of work directory
Ed Bartosh [Thu, 2 Feb 2017 13:32:58 +0000 (15:32 +0200)]
wic: direct: fix creation of work directory

It was a typo in current code: mktemp was used instead of
mkdtemp to create work directory. This is fixed by using
mkdtemp.

Create work directory as a subdirectory of output directory
to make sure both are on the same partition to make moving
of result image faster.

This also fixes possible disk space issues as mkdtemp uses
TMPDIR, TEMP or TMP environment variables to get default value
of its 'dir' parameter. Those variables are usually pointing
to /tmp, which is not the best location to create huge images.

Signed-off-by: Ed Bartosh <ed.bartosh@linux.intel.com>
Signed-off-by: Ross Burton <ross.burton@intel.com>
8 years agowic: engine: create output dir
Ed Bartosh [Thu, 2 Feb 2017 13:32:57 +0000 (15:32 +0200)]
wic: engine: create output dir

Make sure output directory exists before creating an image.
Create it if it doesn't exist.

Signed-off-by: Ed Bartosh <ed.bartosh@linux.intel.com>
Signed-off-by: Ross Burton <ross.burton@intel.com>
8 years agousbutils: add dependency on udev-hwdb, not libudev
Ross Burton [Thu, 2 Feb 2017 13:58:50 +0000 (13:58 +0000)]
usbutils: add dependency on udev-hwdb, not libudev

libudev will be autodetected by the linkage, the intention here was to depend on
udev-hwdb to ensure that the USB ID lists are installed.

Signed-off-by: Ross Burton <ross.burton@intel.com>
8 years agoscripts/runqemu: fix a typo
Ming Liu [Wed, 1 Feb 2017 12:04:19 +0000 (13:04 +0100)]
scripts/runqemu: fix a typo

Signed-off-by: Ming Liu <peter.x.liu@external.atlascopco.com>
Signed-off-by: Ross Burton <ross.burton@intel.com>
8 years agoselftest/runtime-tests: Modify positins test to reduce execution time
Jose Perez Carranza [Mon, 30 Jan 2017 13:48:09 +0000 (07:48 -0600)]
selftest/runtime-tests: Modify positins test to reduce execution time

Currently test_postinst_roofs_and_boot is building a full-cmdline image
this is taking a lot of time to execute the test, so a minimal image
will be build instead.

Signed-off-by: Jose Perez Carranza <jose.perez.carranza@linux.intel.com>
Signed-off-by: Ross Burton <ross.burton@intel.com>
8 years agobuild-perf-test-wrapper.sh: implement locking
Markus Lehtonen [Thu, 2 Feb 2017 10:42:12 +0000 (12:42 +0200)]
build-perf-test-wrapper.sh: implement locking

In order to prevent multiple instances of the script running at the same
time.

Signed-off-by: Markus Lehtonen <markus.lehtonen@linux.intel.com>
Signed-off-by: Ross Burton <ross.burton@intel.com>
8 years agowic: remove syslinux.py
Ed Bartosh [Wed, 1 Feb 2017 10:29:32 +0000 (12:29 +0200)]
wic: remove syslinux.py

This module contains singe function serial_console_form_kargs, which
is used only by rootfs_pcbios_ext plugin. Moved it there and removed
syslinux module to make it easy to find and mainain plugin code.

[YOCTO #10619]

8 years agowic: code cleanup
Ed Bartosh [Tue, 31 Jan 2017 17:31:55 +0000 (19:31 +0200)]
wic: code cleanup

Fixed indentation, unused imports, trailing lines etc.

[YOCTO #10619]

Signed-off-by: Ed Bartosh <ed.bartosh@linux.intel.com>
8 years agowic: msger.py: remove unused APIs
Ed Bartosh [Tue, 31 Jan 2017 16:29:40 +0000 (18:29 +0200)]
wic: msger.py: remove unused APIs

Removed unused enable_logstderr and disable_logstderr APIs.

[YOCTO #10619]

Signed-off-by: Ed Bartosh <ed.bartosh@linux.intel.com>
8 years agowic: removed code from __init__.py
Ed Bartosh [Tue, 31 Jan 2017 16:27:31 +0000 (18:27 +0200)]
wic: removed code from __init__.py

The code deals with non-existing directory
and can be removed.

[YOCTO #10619]

Signed-off-by: Ed Bartosh <ed.bartosh@linux.intel.com>
8 years agowic: move oe/misc.py one level up
Ed Bartosh [Tue, 31 Jan 2017 15:56:56 +0000 (17:56 +0200)]
wic: move oe/misc.py one level up

Flattened directory structure:
   moved wic/utils/oe/misc.py -> wic/utils/misc.py

[YOCTO #10619]

Signed-off-by: Ed Bartosh <ed.bartosh@linux.intel.com>
8 years agowic: move 2 APIs to wic.engine
Ed Bartosh [Tue, 31 Jan 2017 11:32:03 +0000 (13:32 +0200)]
wic: move 2 APIs to wic.engine

Moved find_canned and get_custom_config APIs to engine module.
Removed empty wic.utils.misc module.

[YOCTO #10619]

Signed-off-by: Ed Bartosh <ed.bartosh@linux.intel.com>
8 years agowci: misc: removed build_name API
Ed Bartosh [Tue, 31 Jan 2017 10:45:01 +0000 (12:45 +0200)]
wci: misc: removed build_name API

This API is not used in wic code.

[YOCTO #10619]

Signed-off-by: Ed Bartosh <ed.bartosh@linux.intel.com>
8 years agowic: partition: simlify calling plugin methods
Ed Bartosh [Tue, 31 Jan 2017 10:37:45 +0000 (12:37 +0200)]
wic: partition: simlify calling plugin methods

Replaced parse_sourceparams function with list comprehension.
Used local variables instead of attributes.
Moved global variable to the local scope.

[YOCTO #10619]

Signed-off-by: Ed Bartosh <ed.bartosh@linux.intel.com>
8 years agowic: removed test file
Ed Bartosh [Tue, 31 Jan 2017 09:51:52 +0000 (11:51 +0200)]
wic: removed test file

This file is not used anywhere in the wic code.

[YOCTO #10619]

Signed-off-by: Ed Bartosh <ed.bartosh@linux.intel.com>
8 years agowic: renamd direct_plugin.py -> direct.py
Ed Bartosh [Tue, 31 Jan 2017 09:46:11 +0000 (11:46 +0200)]
wic: renamd direct_plugin.py -> direct.py

As this files is located in plugins/imager subdirectory it's
obvious that it's an imager plugin. Renamed to direct.py to
be consistent with plugin naming scheme.

[YOCTO #10619]

Signed-off-by: Ed Bartosh <ed.bartosh@linux.intel.com>
8 years agowic: simplified code of direct_plugin
Ed Bartosh [Tue, 31 Jan 2017 02:05:49 +0000 (04:05 +0200)]
wic: simplified code of direct_plugin

Removed unused methods.
Got rid of get_default_source_plugin and _full_name methods.

[YOCTO #10619]

Signed-off-by: Ed Bartosh <ed.bartosh@linux.intel.com>
8 years agowic: pylinted direct_plugin
Ed Bartosh [Tue, 31 Jan 2017 01:54:41 +0000 (03:54 +0200)]
wic: pylinted direct_plugin

Fixed wrong continued indentation, unused import and
trailing new line pyling warnings.

[YOCTO #10619]

Signed-off-by: Ed Bartosh <ed.bartosh@linux.intel.com>
8 years agowic: improve naming in direct_plugin classes
Ed Bartosh [Tue, 31 Jan 2017 01:43:56 +0000 (03:43 +0200)]
wic: improve naming in direct_plugin classes

Synchronized attribure names in DirectImageCreator and
DirectPlugin for better readability. Simplified code,
removed unneeded global variable disk_methods.

Signed-off-by: Ed Bartosh <ed.bartosh@linux.intel.com>
8 years agowic: get rid of __rootfs_dir_to_dict method
Ed Bartosh [Tue, 31 Jan 2017 01:25:10 +0000 (03:25 +0200)]
wic: get rid of __rootfs_dir_to_dict method

Replaced class method __rootfs_dir_to_dict with a list
comprehension.

[YOCTO #10619]

Signed-off-by: Ed Bartosh <ed.bartosh@linux.intel.com>
8 years agowic: moved content of direct.py to direct_plugin
Ed Bartosh [Mon, 30 Jan 2017 21:54:29 +0000 (23:54 +0200)]
wic: moved content of direct.py to direct_plugin

This move simplifies directory structure and makes
further refactoring easier. The code from direct.py was used
only in direct_plugin, so it's safe to move it there.

[YOCTO #10619]

Signed-off-by: Ed Bartosh <ed.bartosh@linux.intel.com>
8 years agowic: removed conf.py and empty config file.
Ed Bartosh [Mon, 30 Jan 2017 21:37:18 +0000 (23:37 +0200)]
wic: removed conf.py and empty config file.

Removed as they're not used anymore in wic code.

Signed-off-by: Ed Bartosh <ed.bartosh@linux.intel.com>
8 years agowic: direct_plugin: stop using config manager
Ed Bartosh [Mon, 30 Jan 2017 21:31:53 +0000 (23:31 +0200)]
wic: direct_plugin: stop using config manager

This is a preparation to removing conf.py and config/wic.conf
from the codebase.

Got rid of using configmgr global object in direct_plugin and direct
modules. It was used to implicitly parse kickstart file and set
couple of variables.

Replaced usage of configmgr by passing parameters directly to the
DirectImageCreator.

[YOCTO #10619]

Signed-off-by: Ed Bartosh <ed.bartosh@linux.intel.com>
8 years agowic: creator: stop using config manager
Ed Bartosh [Mon, 30 Jan 2017 21:23:00 +0000 (23:23 +0200)]
wic: creator: stop using config manager

This is a preparation to removing conf.py and config/wic.conf
from the codebase.

confmgr object is complicated for no reason and almost
useless as all configuration info comes from command line and
bitbake variables. It's used it creator.py to store information
about output directory, logs and some never used functionality
like tmpfs for future use, which doesn't actually happen.

[YOCTO #10619]

Signed-off-by: Ed Bartosh <ed.bartosh@linux.intel.com>
8 years agoselftest/bbtests: use write_config instead of local.conf file
Leonardo Sandoval [Wed, 1 Feb 2017 18:04:55 +0000 (12:04 -0600)]
selftest/bbtests: use write_config instead of local.conf file

Extra configuration data should be write using the write_config
method instead of manually appending to the local.conf file

Signed-off-by: Leonardo Sandoval <leonardo.sandoval.gonzalez@linux.intel.com>
Signed-off-by: Ross Burton <ross.burton@intel.com>
8 years agoselftest: wic: stop using hddimg in FSTYPES
Ed Bartosh [Wed, 1 Feb 2017 16:00:05 +0000 (18:00 +0200)]
selftest: wic: stop using hddimg in FSTYPES

Removed hddimg from FSTYPES in wic test suite as
wic doesn't depend on hddimg anymore.

[YOCTO #10835]

Signed-off-by: Ed Bartosh <ed.bartosh@linux.intel.com>
Signed-off-by: Ross Burton <ross.burton@intel.com>
8 years agowic-tools: Do not remove sysroot with rm_work, because wic needs it.
Kristian Amlie [Mon, 30 Jan 2017 15:43:38 +0000 (16:43 +0100)]
wic-tools: Do not remove sysroot with rm_work, because wic needs it.

The sysroot of wic-tools is needed for wic, but if rm_work is enabled,
it will be removed before wic has a chance to use it, hence this fix.

Signed-off-by: Kristian Amlie <kristian.amlie@mender.io>
Signed-off-by: Ross Burton <ross.burton@intel.com>
8 years agopkgconfig: use packageconfig
Ross Burton [Tue, 31 Jan 2017 20:27:26 +0000 (20:27 +0000)]
pkgconfig: use packageconfig

8 years agopkg-config: depend on pkgconfig-native for target builds
Ross Burton [Tue, 31 Jan 2017 20:24:32 +0000 (20:24 +0000)]
pkg-config: depend on pkgconfig-native for target builds

When building for the target, pkg-config uses the target glib-2.0 instead of
it's own minimal fork. To find this it needs to use pkg-config so ensure this
dependency exists in case it doesn't exist on the host already.

Signed-off-by: Ross Burton <ross.burton@intel.com>
8 years agocups: add missing pkgconfig inherit
Ross Burton [Tue, 31 Jan 2017 20:24:28 +0000 (20:24 +0000)]
cups: add missing pkgconfig inherit

Signed-off-by: Ross Burton <ross.burton@intel.com>
8 years agobuild-appliance-image: Update to master head revision
Richard Purdie [Wed, 1 Feb 2017 23:10:58 +0000 (23:10 +0000)]
build-appliance-image: Update to master head revision

Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
8 years agometa-environment: Clean up the task structure to reduce manifest warnings
Richard Purdie [Wed, 1 Feb 2017 11:31:47 +0000 (11:31 +0000)]
meta-environment: Clean up the task structure to reduce manifest warnings

This puts the dependencies on the correct task and removes pointless
noexec tasks allowing for a slightly cleaner task structure.

Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
8 years agostaging: Reduce the number of mkdirs calls
Richard Purdie [Wed, 1 Feb 2017 11:29:06 +0000 (11:29 +0000)]
staging: Reduce the number of mkdirs calls

The number of mkdir calls was showing up high on the profile charts since
it was getting called once per file which is excessive. Each call results
in one or more syscalls which is bad for performance. Cache which
directories we've seen to reduce the calls to a more reasonable number
and speed up recipe specific sysroots.

Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
8 years agosstate/staging: Batch log messages for performance
Richard Purdie [Wed, 1 Feb 2017 11:25:32 +0000 (11:25 +0000)]
sstate/staging: Batch log messages for performance

According to profile data, repeated calls to bb.debug and bb.note in
the extend_recipe_sysroot() codepath were accounting for 75% of the time
(1.5s) in calls from tasks like do_image_complete.

This batches up the log messages into one call into the logging system
which gives similar behaviour to disabling the logging but retains the
debug information.

Since setscene_depvalid is also called from bitbake's setscene code,
we have to be a little creative with the function parameters and leave
the other debug output mechanism in place. This should hopefully
speed up recipe specific sysroots.

Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
8 years agoxmlto: Don't hardcode the path to tail
Richard Purdie [Mon, 30 Jan 2017 23:18:28 +0000 (23:18 +0000)]
xmlto: Don't hardcode the path to tail

We don't need to hardcode a path to tail, follow the other tools examples and
don't specify a path since PATH is good enough for us.

Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
8 years agobinutils-cross: Remove exec_prefix from the linker search path
Richard Purdie [Mon, 30 Jan 2017 23:18:09 +0000 (23:18 +0000)]
binutils-cross: Remove exec_prefix from the linker search path

We don't put target libs into a native/cross ${exec_prefix} but having
this in the default search path means all linker scripts have to be relocated.
This is a considerable chunk of files to create multiple copies of for no good
reason.

Instead, patch out the paths we don't need.

Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
8 years agodistro/defaultsetup: Enable removal of libtool .la files by default
Richard Purdie [Mon, 30 Jan 2017 23:11:32 +0000 (23:11 +0000)]
distro/defaultsetup: Enable removal of libtool .la files by default

Relocation of native .la files during recipe specific sysroot relocation
is probably the final straw in just killing these files off.

Change things so this class is inherited by default. If distros don't want to
do this, they can opt out but it seems like the best thing to do now since
.la files aren't needed on Linux.

Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
8 years agorelocatable: Make native .pc files relocatable
Richard Purdie [Mon, 30 Jan 2017 23:08:38 +0000 (23:08 +0000)]
relocatable: Make native .pc files relocatable

The native .pc files currently have hardcoded paths in them meaning each has
to be relocated at final install time. pkg-config has built in functionality
to avoid this, namely the pcfiledir variable.

This function translates .pc files to use the variable meaning further
relocation later is unnecessary.

Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
8 years agoutils: Avoid need to relocate wrapper scripts
Richard Purdie [Mon, 30 Jan 2017 23:08:11 +0000 (23:08 +0000)]
utils: Avoid need to relocate wrapper scripts

Whilst the path to the executable is dynamically determined, the passed in
environment variables or parameters are not relocatable and rely on the sstate
remapping code. In the recipe specific sysroot case this has become more costly.

This patch translates such paths into relocatable entries which means that
a sed replacement at final installation isn't needed.

Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
8 years agooeqa/core/context.py: Add validation for run-tests option
Aníbal Limón [Tue, 31 Jan 2017 21:05:53 +0000 (15:05 -0600)]
oeqa/core/context.py: Add validation for run-tests option

The run-tests option is optional so if isn't specified set
to None instead of crash on split().

Signed-off-by: Aníbal Limón <anibal.limon@linux.intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
8 years agooeqa/runtime/cases: Rename syslog module to oe_syslog
Aníbal Limón [Tue, 31 Jan 2017 21:05:54 +0000 (15:05 -0600)]
oeqa/runtime/cases: Rename syslog module to oe_syslog

Debian based distros has a builtin syslog module so when
try to load tests using unittest it references the builtin
module instead of runtime/cases.

[YOCTO #10964]

Signed-off-by: Aníbal Limón <anibal.limon@linux.intel.com>
Signed-off-by: Mariano Lopez <mariano.lopez@linux.intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
8 years agotestimage.bbclass: Fixes about suites and dependencies
Mariano Lopez [Tue, 31 Jan 2017 21:05:52 +0000 (15:05 -0600)]
testimage.bbclass: Fixes about suites and dependencies

There are some missing suites on sato-sdk and lsb-sdk images so add it.

The createrepo-native needs to be built before run testimage in
order to create a repository from install packages.

The DL_DIR data also needs to be updated from the current bitbake
environment instead of use the value from testdata file.

[YOCTO #10964]

Signed-off-by: Mariano Lopez <mariano.lopez@linux.intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
8 years agoruntime/cases: Fix case numbers, missing cases and unused classes
Mariano Lopez [Tue, 31 Jan 2017 21:05:51 +0000 (15:05 -0600)]
runtime/cases: Fix case numbers, missing cases and unused classes

gcc: Removed unneded lines.

multilib: fixed case number.

syslog: added a missing test.

[YOCTO #10964]

Signed-off-by: Mariano Lopez <mariano.lopez@linux.intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>