]> code.ossystems Code Review - openembedded-core.git/log
openembedded-core.git
8 years agonative.bbclass: populate native recipe with it's files
Saul Wold [Mon, 20 Feb 2017 05:40:43 +0000 (21:40 -0800)]
native.bbclass: populate native recipe with it's files

This allows a native package's recipe-sysroot-native to be populated with
that packages native image files.  This in turns allows it to be used by
scripts or other tools without creating un-necessary DEPENDS.

An example of this is systemtap-native and the crosstap script.

Signed-off-by: Saul Wold <sgw@linux.intel.com>
Signed-off-by: Ross Burton <ross.burton@intel.com>
8 years agokernel.bbclass: do_deploy requires depmodwrapper
Robert Yang [Thu, 16 Feb 2017 08:18:59 +0000 (00:18 -0800)]
kernel.bbclass: do_deploy requires depmodwrapper

Fixed:
run.do_deploy.27809: line 155: depmodwrapper: command not found

Signed-off-by: Robert Yang <liezhi.yang@windriver.com>
Signed-off-by: Ross Burton <ross.burton@intel.com>
8 years agodevtool: upgrade: fix error of import recipeutils
Luck Hoang [Mon, 20 Feb 2017 09:37:58 +0000 (16:37 +0700)]
devtool: upgrade: fix error of import recipeutils

upgrade.py imports oe.recipeutils in meta/lib/ but path to oe.recipeutils
is not provided. This fails populate_sdk_ext.

Signed-off-by: Luck Hoang <huyht1205@gmail.com>
Signed-off-by: Ross Burton <ross.burton@intel.com>
8 years agoterminal.bbclass: export configure vars
Robert Yang [Mon, 20 Feb 2017 08:45:27 +0000 (00:45 -0800)]
terminal.bbclass: export configure vars

So that we can easily reproduce configure by:

$ $CACHED_CONFIGUREVARS ./configure $CONFIGUREOPTS $EXTRA_OECONF

[YOCTO #9423]

Signed-off-by: Robert Yang <liezhi.yang@windriver.com>
Signed-off-by: Ross Burton <ross.burton@intel.com>
8 years agouseradd-staticids: fix typo for '--expiredate'
Peter Bergin [Sun, 19 Feb 2017 20:30:13 +0000 (21:30 +0100)]
useradd-staticids: fix typo for '--expiredate'

There is a typo in the useradd-staticids.bbclass that
causes error during do_install.

If build was configured to use usedadd-staticids and
the recipe included (for example):

USERADD_PARAM_${PN} = "--expiredate '' testUser"

Due to the typo this added the wrong option '--expiredata '
that caused an error.

Signed-off-by: Peter Bergin <peter@berginkonsult.se>
Signed-off-by: Ross Burton <ross.burton@intel.com>
8 years agoarch-arm*.inc: squash whitespace within TUNE_FEATURES strings
Andre McCurdy [Fri, 17 Feb 2017 18:47:01 +0000 (10:47 -0800)]
arch-arm*.inc: squash whitespace within TUNE_FEATURES strings

TUNE_FEATURES is include in BUILDCFG_VARS, so any whitespace is
visible to the user during the build process. Remove the extra
whitespace added during the 2.1 development cycle:

  http://git.openembedded.org/openembedded-core/commit/?id=f774b44fa007a2a756ada892ede832b1251d940c

For consistency, squash whitespace within PACKAGE_EXTRA_ARCHS strings
too.

Whitespace within TUNE_FEATURES strings in the tune-cortexa*.inc
files has been fixed in a separate commit:

  http://git.openembedded.org/openembedded-core/commit/?id=5610c6397ee098dd998b7417b343494de77179f9

(From OE-Core rev: 3cd0c5ef748ad072f1bd9a8d42157e9643bf97eb)

Signed-off-by: Andre McCurdy <armccurdy@gmail.com>
Signed-off-by: Ross Burton <ross.burton@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
8 years agoscripts/buildstats-diff: fix epoch numbers in --ver-diff
Markus Lehtonen [Wed, 15 Feb 2017 10:45:27 +0000 (12:45 +0200)]
scripts/buildstats-diff: fix epoch numbers in --ver-diff

Incorrect data was printed (recipe name instead of epoch number) when
displaying changes in epoch.

(From OE-Core rev: 5e2b1bfb684dc76963f692172f7457c2249c3266)

Signed-off-by: Markus Lehtonen <markus.lehtonen@linux.intel.com>
Signed-off-by: Ross Burton <ross.burton@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
8 years agolib/oe/patch: Support xz-compressed patches
Dominic Sacré [Wed, 15 Feb 2017 10:44:49 +0000 (11:44 +0100)]
lib/oe/patch: Support xz-compressed patches

Add .xz to the list of extensions recognized by patch_path(), so that
compressed patches ending in .patch.xz or .diff.xz are automatically
applied.

(From OE-Core rev: f1a2c45765d14d3ca09657ad1f6b526554af2bb6)

Signed-off-by: Dominic Sacré <dominic.sacre@gmx.de>
Signed-off-by: Ross Burton <ross.burton@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
8 years agobitbake.conf: Point KERNELORG_MIRROR to cdn.kernel.org
Dominic Sacré [Thu, 16 Feb 2017 09:06:24 +0000 (10:06 +0100)]
bitbake.conf: Point KERNELORG_MIRROR to cdn.kernel.org

Use cdn.kernel.org to distribute the server load and improve download speeds.
Leave www.kernel.org in MIRRORS as a fallback.

See https://www.kernel.org/introducing-fastly-cdn.html

(From OE-Core rev: 7c81b680a3cc4602c9c153398103d5477d7fd894)

Signed-off-by: Dominic Sacré <dominic.sacre@gmx.de>
Signed-off-by: Ross Burton <ross.burton@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
8 years agooeqa/runtime/context.py: Fix use of getTarget() with testexport
Mariano Lopez [Tue, 21 Feb 2017 10:11:00 +0000 (10:11 +0000)]
oeqa/runtime/context.py: Fix use of getTarget() with testexport

The idea on getTarget is to use kwargs to send custom variables
to different targets, instead of this, a new variable was added
(just used for custom targets) and this broke testexport. So
in order to fix it, just add the custom variable to kwargs.

This fixes the use of getTarget() in testexport class that was
introduced in 1dc8010afd71fe46fb28bb86fb7c07a5fbd3d7cf.

Signed-off-by: Mariano Lopez <mariano.lopez@linux.intel.com>
Signed-off-by: Ross Burton <ross.burton@intel.com>
8 years agooeqa: fix typo
Humberto Ibarra [Mon, 13 Feb 2017 17:25:37 +0000 (11:25 -0600)]
oeqa: fix typo

The word 'uninstall' was spelled incorrectly in various parts of
oeqa files.

(From OE-Core rev: af365e025030436f83b233fa51ecc9c58c58ce3c)

Signed-off-by: Humberto Ibarra <humberto.ibarra.lopez@intel.com>
Signed-off-by: Ross Burton <ross.burton@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
8 years agoimage_types_wic: fix expansion error
Ed Bartosh [Wed, 22 Feb 2017 14:39:35 +0000 (16:39 +0200)]
image_types_wic: fix expansion error

Variable WKS_SEARCH_PATH was not fully expanded when used directly
in this expression:
    WKS_FULL_PATH = "${@wks_search('${WKS_FILES}'.split(), '${WKS_SEARCH_PATH}') or ''}"

This caused expansion error: bb.data_smart.ExpansionError:
Failure expanding variable WKS_FULL_PATH, .... which triggered exception
SyntaxError: invalid syntax (WKS_FULL_PATH, line 1)

Calling d.getVar instead of directly referencing variables
should fix the issue.

Signed-off-by: Ed Bartosh <ed.bartosh@linux.intel.com>
Signed-off-by: Ross Burton <ross.burton@intel.com>
8 years agolayer.conf: bump version for change in eSDK selftest behaviour
Joshua Lock [Fri, 17 Feb 2017 15:49:16 +0000 (15:49 +0000)]
layer.conf: bump version for change in eSDK selftest behaviour

Since 59a99cd8 "oeqa/sdk: Updates sanity tests for minimal eSDK"
we can now run oe-selftests against a minimal eSDK.

We need to increase the layer version so that we can detect this
change in behaviour in our automated testing.

Signed-off-by: Joshua Lock <joshua.g.lock@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
8 years agoselftest/containerimage.py: Add container IMAGE_FSTYPES test
Randy Witt [Mon, 30 Jan 2017 21:16:28 +0000 (13:16 -0800)]
selftest/containerimage.py: Add container IMAGE_FSTYPES test

This test checks to make sure only the files expected exist in a
container image. Currently only ROOTFS_BOOTSTRAP_INSTALL, gets added to
all images without the user specifying it.

But this test should help if a developer in the future ever silently
adds more than just ROOTFS_BOOTSTRAP_INSTALL, and that the developer can
make sure it also gets removed from a container image.

[YOCTO #9502]

Signed-off-by: Randy Witt <randy.e.witt@linux.intel.com>
Signed-off-by: Ross Burton <ross.burton@intel.com>
8 years agoimage-container.bbclass: Error if not using linux-dummy
Randy Witt [Wed, 18 Jan 2017 18:31:51 +0000 (10:31 -0800)]
image-container.bbclass: Error if not using linux-dummy

In the typical case of a container, the user doesn't particularly care
about the kernel version used. Since using linux-dummy prevents building
a kernel in the most common case, alert the user if they choose a
container image, but don't also specify linux-dummy.

In the cases the user actually does want to build against particular
kernel headers, etc, they can override by setting
IMAGE_CONTAINER_NO_DUMMY = "1".

[YOCTO #9502]

Signed-off-by: Randy Witt <randy.e.witt@linux.intel.com>
Signed-off-by: Ross Burton <ross.burton@intel.com>
8 years agoimage-container.bbclass: Add the "container" IMAGE_FSTYPES
Randy Witt [Wed, 18 Jan 2017 18:31:50 +0000 (10:31 -0800)]
image-container.bbclass: Add the "container" IMAGE_FSTYPES

The "container" fstype does very little other than pick tar.bz2 as the
actual image type and disable installation of ROOTFS_BOOTSTRAP_INSTALL.

[YOCTO #9502]

Signed-off-by: Randy Witt <randy.e.witt@linux.intel.com>
Signed-off-by: Ross Burton <ross.burton@intel.com>
8 years agooeqa/selftest/bbtests: Update after changes to dependency graph file generation
Richard Purdie [Wed, 25 Jan 2017 11:47:06 +0000 (11:47 +0000)]
oeqa/selftest/bbtests: Update after changes to dependency graph file generation

Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
8 years agowget: Upgrade 1.19 -> 1.19.1
Richard Purdie [Sun, 19 Feb 2017 14:13:11 +0000 (06:13 -0800)]
wget: Upgrade 1.19 -> 1.19.1

This fixes musl build issues and avoids autobuilder failures.

Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
8 years agorootfspostcommands: remove shadow backup files instead of trying to sort
Patrick Ohly [Fri, 10 Feb 2017 11:29:14 +0000 (12:29 +0100)]
rootfspostcommands: remove shadow backup files instead of trying to sort

Backup are files sometimes are inconsistent and then cannot be
sorted (YOCTO #11043), and more importantly, are not needed in
the initial rootfs, so they get deleted.

Fixes: [YOCTO #11007]
Signed-off-by: Patrick Ohly <patrick.ohly@intel.com>
8 years agolinux-yocto-rt/4.9: update to 4.9-rt6
Bruce Ashfield [Tue, 14 Feb 2017 21:14:25 +0000 (16:14 -0500)]
linux-yocto-rt/4.9: update to 4.9-rt6

Updating the SRCREVs to make Paul Gortmaker's port of 4.9-rt6 active.

Signed-off-by: Bruce Ashfield <bruce.ashfield@windriver.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
8 years agokernel-yocto: trivial: remove extra } from bsp_definition
Bruce Ashfield [Tue, 14 Feb 2017 21:14:24 +0000 (16:14 -0500)]
kernel-yocto: trivial: remove extra } from bsp_definition

A typo introduced an extra } in the logging of the bsp_definition.
While this won't cause problems .. it needs to be fixed!

Signed-off-by: Bruce Ashfield <bruce.ashfield@windriver.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
8 years agolinux-yocto/4.4: meta/configuration updates
Bruce Ashfield [Tue, 14 Feb 2017 21:14:23 +0000 (16:14 -0500)]
linux-yocto/4.4: meta/configuration updates

Integrating the following commit already in 4.9 and 4.8:

  intel-quark: Add intel-quark-preempt-rt bsp configuration

  While there are intel-quark configurations for the KTYPEs standard
  and tiny in bsp/intel-common, there's none for the preempt-rt KTYPE.
  Trying to build preempt-rt enabled kernels such as linux-yocto-rt for
  intel-quark yields a .config having a potentially misconfigured
  architecture. More importantly, however, preempt-rt related CONFIG
  options are not enabled. Hence, a build of, e.g., linux-yocto-rt, does
  not result in a preempt-rt enabled kernel.

  This patch qualifies to be (back)ported to other branches than
  master.

Signed-off-by: Christian Storm <christian.storm@siemens.com>
Signed-off-by: Jan Kiszka <jan.kiszka@siemens.com>
Signed-off-by: Bruce Ashfield <bruce.ashfield@windriver.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
8 years agolinux-yocto/4.9: update to v4.9.9
Bruce Ashfield [Tue, 14 Feb 2017 21:14:21 +0000 (16:14 -0500)]
linux-yocto/4.9: update to v4.9.9

Updating to the korg -stable release. Changelog is as follows:

   d2e4b66b4ef2 Linux 4.9.9
   f2a0409a0850 drm/i915/execlists: Reset RING registers upon resume
   72cd604cfd86 fs: break out of iomap_file_buffered_write on fatal signals
   bf9ab22c350d iw_cxgb4: set correct FetchBurstMax for QPs
   13363b6988f6 x86/irq: Make irq activate operations symmetric
   e02136282296 irqdomain: Avoid activating interrupts more than once
   c6b0333baea0 iio: health: max30100: fixed parenthesis around FIFO count check
   12c2fcf6bcf7 iio: dht11: Use usleep_range instead of msleep for start signal
   7baa8491d47d iio: health: afe4403: retrieve a valid iio_dev in suspend/resume
   5f0ee562605b iio: health: afe4404: retrieve a valid iio_dev in suspend/resume
   54d2ccc4003b iio: adc: palmas_gpadc: retrieve a valid iio_dev in suspend/resume
   db0e02ef6b82 staging: greybus: timesync: validate platform state callback
   89cc65c1117f USB: serial: option: add device ID for HP lt2523 (Novatel E371)
   12a9c11c2e19 usb: gadget: f_fs: Assorted buffer overflow checks.
   e40d15fce929 usb: musb: Fix host mode error -71 regression
   cbd819e7db3a USB: Add quirk for WORLDE easykey.25 MIDI keyboard
   4807725aab0b USB: serial: pl2303: add ATEN device ID
   8bc382a9652b USB: serial: qcserial: add Dell DW5570 QDL
   0dcbd0aa44e6 KVM: x86: do not save guest-unsupported XSAVE state
   bc05a2e940fe dmaengine: cppi41: Fix oops in cppi41_runtime_resume
   2c2e7fe7c9c5 dmaengine: cppi41: Fix runtime PM timeouts with USB mass storage
   bebb9d75e84d perf/x86/intel/uncore: Clean up hotplug conversion fallout
   e6bd712154aa HID: wacom: Fix poor prox handling in 'wacom_pl_irq'
   f24bc920612c HID: hid-lg: Fix immediate disconnection of Logitech Rumblepad 2
   80246551c7d7 HID: usbhid: Quirk a AMI virtual mouse and keyboard with ALWAYS_POLL
   40add19d3897 iwlwifi: mvm: avoid crash on restart w/o reserved queues
   97663735a96f iwlwifi: fix double hyphen in MODULE_FIRMWARE for 8000
   3d8ec7d2d5f9 pinctrl: intel: merrifield: Add missed check in mrfld_config_set()
   2cf6c49264e4 pinctrl: baytrail: Debounce register is one per community
   449d3ecfbd76 Revert "vring: Force use of DMA API for ARM-based systems with legacy devices"
   16f61dee7e7c Revert "bcma: init serial console directly from ChipCommon code"
   12f822d23dee percpu-refcount: fix reference leak during percpu-atomic transition
   8ee8ff9e2652 regulator: axp20x: AXP806: Fix dcdcb being set instead of dcdce
   1594edd9ea0d vhost: fix initialization for vq->is_le
   04eb7db25bb1 mmc: sdhci: Ignore unexpected CARD_INT interrupts
   1d88791d5ed5 cgroup: don't online subsystems before cgroup_name/path() are operational
   a150e08704b2 can: bcm: fix hrtimer/tasklet termination in bcm op removal
   a93ae8dccc3c tracing: Fix hwlat kthread migration
   b67c7d39bc28 mm, fs: check for fatal signals in do_generic_file_read()
   6cb0497aec81 base/memory, hotplug: fix a kernel oops in show_valid_zones()
   72f7419610c8 mm/memory_hotplug.c: check start_pfn in test_pages_in_a_zone()
   9e255997c2e1 cifs: initialize file_info_lock
   f0c3a0ac3349 zswap: disable changing params if init fails
   a3d729526f2f svcrpc: fix oops in absence of krb5 module
   743146d347f3 NFSD: Fix a null reference case in find_or_create_lock_stateid()
   4c953848c95b powerpc/mm: Use the correct pointer when setting a 2MB pte
   8f415333be3a powerpc: Fix build failure with clang due to BUILD_BUG_ON()
   bbf69e5197da powerpc: Add missing error check to prom_find_boot_cpu()
   73d45909780e powerpc/eeh: Fix wrong flag passed to eeh_unfreeze_pe()
   4b70d598c2ee libata: Fix ATA request sense
   6d08607ef35b libata: apply MAX_SEC_1024 to all CX1-JB*-HP devices
   fc794153c407 ata: sata_mv:- Handle return value of devm_ioremap.
   b41615aa7056 perf/core: Fix PERF_RECORD_MMAP2 prot/flags for anonymous memory
   3996a91e3bdc perf/core: Fix use-after-free bug
   53bed1f64025 crypto: arm64/aes-blk - honour iv_out requirement in CBC and CTR modes
   b04a39f88ced crypto: api - Clear CRYPTO_ALG_DEAD bit before registering an alg
   2eb8f7c4225a drm/nouveau/nv1a,nv1f/disp: fix memory clock rate retrieval
   bd5cefed1ccc drm/nouveau/disp/gt215: Fix HDA ELD handling (thus, HDMI audio) on gt215
   c9fb422fd938 drm/amdgpu/si: fix crash on headless asics
   20658b3df8b9 pinctrl: baytrail: Add missing spinlock usage in byt_gpio_irq_handler
   7396685a1bca HID: cp2112: fix gpio-callback error handling
   a18c4584a489 HID: cp2112: fix sleep-while-atomic
   dfd713307adb xtensa: fix noMMU build on cores with MMU
   f2e24dd91818 efi/fdt: Avoid FDT manipulation after ExitBootServices()
   f0c7412edfc1 x86/efi: Always map the first physical page into the EFI pagetables
   13e6ef99d23b ext4: validate s_first_meta_bg at mount time
   610c2b7ff8f6 PCI/ASPM: Handle PCI-to-PCIe bridges as roots of PCIe hierarchies

Signed-off-by: Bruce Ashfield <bruce.ashfield@windriver.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
8 years agolinux-yocto-dev: bump to v4.10+
Bruce Ashfield [Tue, 14 Feb 2017 21:14:22 +0000 (16:14 -0500)]
linux-yocto-dev: bump to v4.10+

Signed-off-by: Bruce Ashfield <bruce.ashfield@windriver.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
8 years agolinux-yocto/4.8/4.9: fix tpm.cfg and add dm-verity
Bruce Ashfield [Tue, 14 Feb 2017 21:14:20 +0000 (16:14 -0500)]
linux-yocto/4.8/4.9: fix tpm.cfg and add dm-verity

Integrating the following commits to fix a configuration warning
and to add a tpm configuration feature:

  8f3bc608ae61 features: add dm-verity
  0ad64750fef1 tpm: avoid warning due to double CONFIG_TCG_TPM

Signed-off-by: Bruce Ashfield <bruce.ashfield@windriver.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
8 years agolinux-yocto/4.9: Intel-quark: Add intel-quark-preempt-rt bsp configuration
Bruce Ashfield [Tue, 14 Feb 2017 21:14:19 +0000 (16:14 -0500)]
linux-yocto/4.9: Intel-quark: Add intel-quark-preempt-rt bsp configuration

While there are intel-quark configurations for the KTYPEs standard and
tiny in bsp/intel-common, there's none for the preempt-rt KTYPE.
Trying to build preempt-rt enabled kernels such as linux-yocto-rt for
intel-quark yields a .config having a potentially misconfigured
architecture. More importantly, however, preempt-rt related CONFIG
options are not enabled. Hence, a build of, e.g., linux-yocto-rt, does
not result in a preempt-rt enabled kernel.

Signed-off-by: Christian Storm <christian.storm@siemens.com>
Signed-off-by: Jan Kiszka <jan.kiszka@siemens.com>
Signed-off-by: Bruce Ashfield <bruce.ashfield@windriver.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
8 years agolinux-yocto-rt/4.9: integrate 4.9-rt4
Bruce Ashfield [Tue, 14 Feb 2017 21:14:18 +0000 (16:14 -0500)]
linux-yocto-rt/4.9: integrate 4.9-rt4

Integrating the 4.9-rt4 new changes:

  56a3dd6fb4e9 x86/mm/cpa: avoid wbinvd() for PREEMPT
  221db3244e71 pinctrl: qcom: Use raw spinlock variants
  43158f23e188 radix-tree: use local locks
  109a8fea2c4a cpuset: Convert callback_lock to raw_spinlock_t
  3741fb19d20f softirq: wake the timer softirq if needed

Signed-off-by: Bruce Ashfield <bruce.ashfield@windriver.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
8 years agolinux-yocto/4.8: update to v4.8.18
Bruce Ashfield [Tue, 14 Feb 2017 21:14:17 +0000 (16:14 -0500)]
linux-yocto/4.8: update to v4.8.18

Paul Gortmaker has started doing 4.8 -stable updates. This integrates
his 4.8.18 queue from:

   git://git.kernel.org/pub/scm/linux/kernel/git/paulg/linux-4.8.y.git

Change log is as follows:

   28dda89acd7c Linux 4.8.18
   936e79682048 tick/broadcast: Prevent NULL pointer dereference
   f22b03e15427 clocksource/dummy_timer: Move hotplug callback after the real timers
   8f78db7e5df0 xfs: use GPF_NOFS when allocating btree cursors
   1d818a81bac6 xfs: ignore leaf attr ichdr.count in verifier during log replay
   62ef2c9c1d94 xfs: don't allow di_size with high bit set
   db28776559c7 xfs: error out if trying to add attrs and anextents > 0
   fb70f2a64910 xfs: don't crash if reading a directory results in an unexpected hole
   ebff27d76e7e xfs: complain if we don't get nextents bmap records
   14c3f0b270aa xfs: check for bogus values in btree block headers
   31f0f90742a3 xfs: pass state not whichfork to trace_xfs_extlist
   430a362685e9 xfs: Move AGI buffer type setting to xfs_read_agi
   20aa52b53586 xfs: new inode extent list lookup helpers
   c7b75fcea4e6 xfs: fix unbalanced inode reclaim flush locking
   e89df591f833 xfs: check minimum block size for CRC filesystems
   b04a2c02f4d7 xfs: provide helper for counting extents from if_bytes
   f4e7f4cd52b1 xfs: don't BUG() on mixed direct and mapped I/O
   49f159ae6e80 xfs: check return value of _trans_reserve_quota_nblks
   23bb39fdbb96 xfs: don't call xfs_sb_quota_from_disk twice
   d3c7d2aa00d4 tpm_tis: Check return values from get_burstcount.
   9fc58e555d78 drm/i915/gen9: fix the WM memory bandwidth WA for Y tiling cases
   4f075608d2c4 drm/i915/gen9: unconditionally apply the memory bandwidth WA
   55e9c9d00d70 drm/i915: disable PSR by default on HSW/BDW
   2c266a9a3631 drm/radeon: Always store CRTC relative radeon_crtc->cursor_x/y values
   e4d1900e15db s390/topology: always use s390 specific sched_domain_topology_level
   b04e5322d8a4 powerpc/pci/rpadlpar: Fix device reference leaks
   458e63f135d6 PCI: Enable access to non-standard VPD for Chelsio devices (cxgb3)
   05a725c46f56 PCI: Support INTx masking on ConnectX-4 with firmware x.14.1100+
   8d12795b1eac PCI: Convert Mellanox broken INTx quirks to be for listed devices only
   24af87eb2ffa PCI: Convert broken INTx masking quirks from HEADER to FINAL
   ec9f277f52e9 PCI: Add Mellanox device IDs
   9e7c8ebc1316 staging: media: davinci_vpfe: unlock on error in vpfe_reqbufs()
   8c6c73a692c9 f2fs: hide a maybe-uninitialized warning
   a91f2c1d4f45 f2fs: remove percpu_count due to performance regression
   93634e1cc928 md: MD_RECOVERY_NEEDED is set for mddev->recovery
   239b31c50bde crypto: arm64/aes-ce - fix for big endian
   a3141467e092 crypto: arm64/aes-xts-ce: fix for big endian
   7c26ef18f05a crypto: arm64/sha1-ce - fix for big endian
   cb70151453a1 crypto: arm64/aes-neon - fix for big endian
   c1c451bea4af crypto: arm64/aes-ccm-ce: fix for big endian
   c4957563b255 crypto: arm/aes-ce - fix for big endian
   dcb0c4ebc7ad crypto: arm64/ghash-ce - fix for big endian
   d60ecc7f02a4 crypto: arm64/sha2-ce - fix for big endian
   f9315853c7c1 s390/crypto: unlock on error in prng_tdes_read()
   a76c795f53e7 mm, compaction: fix NR_ISOLATED_* stats for pfn based migration
   d3b82f9a9716 mm: khugepaged: fix radix tree node leak in shmem collapse error path
   ab533014582f mm: khugepaged: close use-after-free race during shmem collapsing
   c9777f749c4c mm/hugetlb.c: use the right pte val for compare in hugetlb_cow
   0ff00251359e mmc: mmc_test: Uninitialized return value
   abe558647a0f PM / wakeirq: Fix dedicated wakeirq for drivers not using autosuspend
   b90b6b8cce53 irqchip/bcm7038-l1: Implement irq_cpu_offline() callback
   d6229a05dd06 ima: fix memory leak in ima_release_policy
   8cbe9b689467 sbp-target: Fix second argument of percpu_ida_alloc()
   fed8068fd1de target/iscsi: Fix double free in lio_target_tiqn_addtpg()
   ec92af68808e scsi: mvsas: fix command_active typo
   efe08cc6939b ASoC: samsung: i2s: Fixup last IRQ unsafe spin lock call
   a5a550bd0768 ASoC: cht_bsw_rt5645: Fix leftover kmalloc
   3fcdecb0f9f6 iommu/vt-d: Flush old iommu caches for kdump when the device gets context mapped
   22604a045edb iommu/vt-d: Fix pasid table size encoding
   f7dd8813279b iommu/amd: Fix the left value check of cmd buffer
   9a134d29af48 iommu/amd: Missing error code in amd_iommu_init_device()
   35a294dd5723 clk: renesas: mstp: Support 8-bit registers for r7s72100
   8a7022b2ffad clk: imx31: fix rewritten input argument of mx31_clocks_init()
   8b4940dd9224 clk: sunxi-ng: sun8i-h3: Set CLK_SET_RATE_PARENT for audio module clocks
   a13af6a13e68 clk: ti: dra7: fix "failed to lookup clock node gmac_gmii_ref_clk_div" boot message
   c54f34533601 clk: clk-wm831x: fix a logic error
   36c3481984b5 clk: qcom: ipq806x: Fix board clk rates
   fdd91f12bebc hwmon: (g762) Fix overflows and crash seen when writing limit attributes
   18d749bb756f hwmon: (nct7802) Fix overflows seen when writing into limit attributes
   1e6b9473280a hwmon: (ds620) Fix overflows seen when writing temperature limits
   0bc18f1d87d9 hwmon: (amc6821) sign extension temperature
   4c8493cb90b9 hwmon: (scpi) Fix module autoload
   d618b1647e27 platform/x86: fujitsu-laptop: use brightness_set_blocking for LED-setting callbacks
   b803e5984015 x86/cpu: Probe CPUID leaf 6 even when cpuid_level == 6
   1121b5f49eb7 debugfs: improve DEFINE_DEBUGFS_ATTRIBUTE for !CONFIG_DEBUG_FS
   f4b9eaba036c clk: renesas: cpg-mssr: Fix inverted debug check
   a25cb9b5744b efi/efivar_ssdt_load: Don't return success on allocation failure
   b36f676475bd cris: Only build flash rescue image if CONFIG_ETRAX_AXISFLASHMAP is selected
   40b0bbe612bb ath10k: use the right length of "background"
   3e0f61361a82 nl80211: Use different attrs for BSSID and random MAC addr in scan req
   3d04e6a06abc mac80211: fix tid_agg_rx NULL dereference
   f4ded840a335 drm/i915: tune down the fast link training vs boot fail
   2ed2b3cc6ae3 drm/i915/dp: add lane_count check in intel_dp_check_link_status
   ac2370c93a40 usb: dwc3: gadget: always unmap EP0 requests
   6039272a760e usb: dwc3: ep0: explicitly call dwc3_ep0_prepare_one_trb()
   4f76ead23734 usb: dwc3: ep0: add dwc3_ep0_prepare_one_trb()
   7acc87fae524 iio: accel: st_accel: fix LIS3LV02 reading and scaling
   d59378efa7ac staging: iio: ad7606: fix improper setting of oversampling pins
   401316ef7253 mei: move write cb to completion on credentials failures
   3ab87b8fd267 mei: bus: fix mei_cldev_enable KDoc
   c103d55bb7d7 USB: serial: io_ti: bind to interface after fw download
   d302588a910f ARM: dts: sun7i: bananapi-m1-plus: Enable USB PHY for USB host support
   56dce131ec69 arm64: dts: hip06: Correct hardware pin number of usb node
   0c03658a3e7b USB: phy: am335x-control: fix device and of_node leaks
   518a4d8d533c ARM: dts: r8a7794: Correct hsusb parent clock
   7780eb2fd54b usb: gadget: Fix second argument of percpu_ida_alloc()
   a3ff51875d83 USB: serial: kl5kusb105: abort on open exception path
   5f8a0a2fd33b ALSA: usb-audio: Fix bogus error return in snd_usb_create_stream()
   fb106dc8b847 usb: musb: blackfin: add bfin_fifo_offset in bfin_ops
   1bff22506598 usb: gadget: udc: core: fix return code of usb_gadget_probe_driver()
   c7c4ad7fce6d usb: hub: Move hub_port_disable() to fix warning if PM is disabled
   b7fb0479ecd6 usb: musb: Fix trying to free already-free IRQ 4
   217c03eaf10c usb: dwc3: gadget: Fix full speed mode
   7de3c659811a usb: dwc3: pci: Fix dr_mode misspelling
   6fdac0047799 usb: dwc3: pci: add Intel Gemini Lake PCI ID
   220d6521aac2 xhci: Fix race related to abort operation
   d3f3b0bd0756 xhci: Use delayed_work instead of timer for command timeout
   0da62a5b5242 usb: xhci-mem: use passed in GFP flags instead of GFP_KERNEL
   997d598c694f USB: serial: mos7720: fix parallel probe
   eb9e22ac9265 USB: serial: mos7720: fix parport use-after-free on probe errors
   c91243377c8d USB: serial: mos7720: fix use-after-free on probe errors
   b756fba5e563 USB: serial: mos7720: fix NULL-deref at open
   8cc2ae8e130e USB: serial: mos7840: fix NULL-deref at open
   68ab1ccfc085 USB: serial: kobil_sct: fix NULL-deref in write
   37ca8378e77c USB: serial: cyberjack: fix NULL-deref at open
   fd435e99f5ed USB: serial: oti6858: fix NULL-deref at open
   1653f66ba3fb USB: serial: io_edgeport: fix NULL-deref at open
   73ec9d58b1c7 USB: serial: ti_usb_3410_5052: fix NULL-deref at open
   d1706b42d164 USB: serial: garmin_gps: fix memory leak on failed URB submit
   b653000d9b0c USB: serial: iuu_phoenix: fix NULL-deref at open
   427bac51ce67 USB: serial: io_ti: fix I/O after disconnect
   59f740cd834c USB: serial: io_ti: fix another NULL-deref at open
   2ef82cf063c8 USB: serial: io_ti: fix NULL-deref at open
   8d5f29b2d578 USB: serial: spcp8x5: fix NULL-deref at open
   e1fe8df45b74 USB: serial: keyspan_pda: verify endpoints at probe
   10da95b6a2eb USB: serial: pl2303: fix NULL-deref at open
   e29498ba6e1f USB: serial: quatech2: fix sleep-while-atomic in close
   05859d754d6f USB: serial: omninet: fix NULL-derefs at open and disconnect
   be17e1c916ce usb: return error code when platform_get_irq fails
   0e9fc72194ce usb: xhci: hold lock over xhci_abort_cmd_ring()
   527f810d8a86 xhci: Handle command completion and timeout race
   c18ca85fd6de usb: host: xhci: Fix possible wild pointer when handling abort command
   67a7de5a2f35 usb: xhci: fix return value of xhci_setup_device()
   775a2bec9cdf xhci: free xhci virtual devices with leaf nodes first
   80c502e0ce6f usb: xhci: apply XHCI_PME_STUCK_QUIRK to Intel Apollo Lake
   d3a9f7c7e8c4 usb: xhci: fix possible wild pointer
   5ded28d83128 usb: gadget: composite: Test get_alt() presence instead of set_alt()
   16718d267e09 USB: dummy-hcd: fix bug in stop_activity (handle ep0)
   47acb8434892 USB: fix problems with duplicate endpoint addresses
   a45f34e820c6 USB: gadgetfs: fix checks of wTotalLength in config descriptors
   27b4d04d1b87 USB: gadgetfs: fix use-after-free bug
   cd4da96b1bac USB: gadgetfs: fix unbounded memory allocation bug
   1fa95887b663 usb: gadgetfs: restrict upper bound on device configuration size
   e4efa88511eb usb: storage: unusual_uas: Add JMicron JMS56x to unusual device
   05ace125078d usb: musb: dsps: implement clear_ep_rxintr() callback
   38caa4d8986e usb: musb: core: add clear_ep_rxintr() to musb_platform_ops
   483c56831853 KVM: MIPS: Flush KVM entry code from icache globally
   08aa895b61e1 KVM: MIPS: Don't clobber CP0_Status.UX
   d33b04df932e KVM: x86: reset MMU on KVM_SET_VCPU_EVENTS
   e3843cdbf81a drm/i915: Force VDD off on the new power seqeuencer before starting to use it
   dbce182845b4 gcc-plugins: update gcc-common.h for gcc-7
   970150316c04 mac80211: initialize fast-xmit 'info' later
   cf3b94f8fe70 ARM: davinci: da850: don't add emac clock to lookup table twice
   391c21a9ab0a parisc: Mark cr16 clocksource unstable on SMP systems
   684f6df20889 fscrypt: fix renaming and linking special files
   6c55d3ad7fb3 ALSA: usb-audio: Fix irq/process data synchronization
   3074297922e2 ALSA: hda - Apply asus-mode8 fixup to ASUS X71SL
   81a35981225c ALSA: hda - Fix up GPIO for ASUS ROG Ranger
   576b9d7789d8 iio: bmi160: Fix time needed to sleep after command execution
   ad2a1ae984c6 iio: max44000: correct value in illuminance_integration_time_available
   ee8e4a73dbf6 iio: common: st_sensors: fix channel data parsing

Signed-off-by: Bruce Ashfield <bruce.ashfield@windriver.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
8 years agolinux-yocto/4.9: bump to v4.9.8
Bruce Ashfield [Tue, 14 Feb 2017 21:14:16 +0000 (16:14 -0500)]
linux-yocto/4.9: bump to v4.9.8

Integrating the -stable releases. Changelog is as follows:

   c8ea2f3b8247 Linux 4.9.8
   b5b4d4a9141e xfs: fix bmv_count confusion w/ shared extents
   5d44dd54bd57 xfs: clear _XBF_PAGES from buffers when readahead page
   29f96b7e9023 xfs: extsize hints are not unlikely in xfs_bmap_btalloc
   aab858dabd5e xfs: remove racy hasattr check from attr ops
   29094164ea99 xfs: verify dirblocklog correctly
   214d55efa255 xfs: fix COW writeback race
   29f319275e76 xfs: fix xfs_mode_to_ftype() prototype
   d062d90c35f2 xfs: don't wrap ID in xfs_dq_get_next_id
   d3201a14b36a xfs: sanity check inode di_mode
   43ce5921773b xfs: sanity check inode mode when creating new dentry
   b5f68e24cc7b xfs: replace xfs_mode_to_ftype table with switch statement
   4fac84ba1da7 xfs: add missing include dependencies to xfs_dir2.h
   e5325fcf70b1 xfs: sanity check directory inode di_size
   624e54b5aff1 xfs: make the ASSERT() condition likely
   4f4d5082ec73 xfs: don't print warnings when xfs_log_force fails
   e9b776519107 xfs: don't rely on ->total in xfs_alloc_space_available
   6b81365b1e57 xfs: adjust allocation length in xfs_alloc_space_available
   c63f4d3aa09d xfs: fix bogus minleft manipulations
   d20e4ad06c4b xfs: bump up reserved blocks in xfs_alloc_set_aside
   9f42bc4f9c1c net: dsa: Bring back device detaching in dsa_slave_suspend()
   e972cce0c833 lwtunnel: Fix oops on state free after encap module unload
   89c2588627c2 net: Specify the owning module for lwtunnel ops
   087c2ecb21af qmi_wwan/cdc_ether: add device ID for HP lt2523 (Novatel E371) WWAN card
   93ff5e03bcba af_unix: move unix_mknod() out of bindlock
   37b27b20a8a9 r8152: don't execute runtime suspend if the tx is not empty
   ad864d9fce0e net: mpls: Fix multipath selection for LSR use case
   74423145d931 bridge: netlink: call br_changelink() during br_dev_newlink()
   087dced61adc net/mlx5e: Do not recycle pages from emergency reserve
   0c687a735cdb tcp: initialize max window for a new fastopen socket
   79453ab88164 ipv6: addrconf: Avoid addrconf_disable_change() using RCU read-side lock
   e9db042dca20 lwtunnel: fix autoload of lwt modules
   b335e6566661 net: phy: bcm63xx: Utilize correct config_intr function
   948e137ad9ba net: fix harmonize_features() vs NETIF_F_HIGHDMA
   d1c95f9ce102 vxlan: fix byte order of vxlan-gpe port number
   1e7cbb413f63 virtio-net: restore VIRTIO_HDR_F_DATA_VALID on receiving
   3eab5dd0eb19 virtio: don't set VIRTIO_NET_HDR_F_DATA_VALID on xmit
   b260a714a638 net sched actions: fix refcnt when GETing of action after bind
   2d6b61ec9d3a ax25: Fix segfault after sock connection timeout
   c7a5df92d1e8 ip6_tunnel: Account for tunnel header in tunnel MTU
   08e650708dd0 ravb: do not use zero-length alignment DMA descriptor
   77ce30dc4df4 mlx4: do not call napi_schedule() without care
   18767acb7b67 openvswitch: maintain correct checksum state in conntrack actions
   3524f64224bd tcp: fix tcp_fastopen unaligned access complaints on sparc
   b66b1f5ac33d net: systemport: Decouple flow control from __bcm_sysport_tx_reclaim
   958bb1bdc2c2 net: ipv4: fix table id in getroute response
   6980c52c4efb net: lwtunnel: Handle lwtunnel_fill_encap failure
   ec1aa8d49532 mlxsw: pci: Fix EQE structure definition
   4ec59d1fe470 mlxsw: switchx2: Fix memory leak at skb reallocation
   7c249f3306b6 mlxsw: spectrum: Fix memory leak at skb reallocation
   5b3df4401064 netvsc: add rcu_read locking to netvsc callback
   a37f2311e63c r8152: fix the sw rx checksum is unavailable
   fd2ffe57dda0 Linux 4.9.7
   b59dd202f231 drm/i915: Remove WaDisableLSQCROPERFforOCL KBL workaround.
   922813f4d66f perf/core: Fix concurrent sys_perf_event_open() vs. 'move_group' race
   f5f415c13209 mm, memcg: do not retry precharge charges
   98185d4b18a1 platform/x86: intel_mid_powerbtn: Set IRQ_ONESHOT
   adb260d15134 platform/x86: mlx-platform: free first dev on error
   776050a9b55e virtio_mmio: Set DMA masks appropriately
   143a9ad4e68c memory_hotplug: make zone_can_shift() return a boolean value
   da1fdb8456ac pinctrl: baytrail: Rectify debounce support
   583eded5860b pinctrl: uniphier: fix Ethernet (RMII) pin-mux setting for LD20
   35948ae694c0 pinctrl: broxton: Use correct PADCFGLOCK offset
   668a827a6c9d s5k4ecgx: select CRC32 helper
   9a335996bfcf IB/rxe: Prevent from completer to operate on non valid QP
   fed1e89a16e3 IB/rxe: Fix rxe dev insertion to rxe_dev_list
   c81ee4ed9c0d IB/umem: Release pid in error and ODP flow
   d7c3d7e453b7 drm/i915: Check for NULL atomic state in intel_crtc_disable_noatomic()
   181478cdb75e drm/i915: Fix calculation of rotated x and y offsets for planar formats
   ff4956555513 drm/i915: Don't init hpd polling for vlv and chv from runtime_suspend()
   8d7c76ae613e drm/i915: Don't leak edid in intel_crt_detect_ddc()
   2cff678d0bb5 drm/i915: prevent crash with .disable_display parameter
   9913aca2d587 drm/i915: Clear ret before unbinding in i915_gem_evict_something()
   1b8ff91ae545 v4l: tvp5150: Don't override output pinmuxing at stream on/off time
   11e5015ae1d1 v4l: tvp5150: Fix comment regarding output pin muxing
   d25f9bfeb62f v4l: tvp5150: Reset device at probe time, not in get/set format handlers
   fe6531075e1d pctv452e: move buffer to heap, no mutex
   0bd3cb8d470b iw_cxgb4: free EQ queue memory on last deref
   cb1d48f55a6d SUNRPC: cleanup ida information when removing sunrpc module
   5637949edb50 NFSv4.0: always send mode in SETATTR after EXCLUSIVE4
   0a7023506112 NFSv4.1: Fix a deadlock in layoutget
   73fdda3b01cd nfs: Don't increment lock sequence ID after NFS4ERR_MOVED
   2b95f1210e50 parisc: Don't use BITS_PER_LONG in userspace-exported swab.h header
   ca332b96ba62 ARC: [arcompact] handle unaligned access delay slot corner case
   9d5f2c151ec0 ARC: udelay: fix inline assembler by adding LP_COUNT to clobber list
   50f5972cc231 can: ti_hecc: add missing prepare and unprepare of the clock
   9f56548b0077 can: c_can_pci: fix null-pointer-deref in c_can_start() - set device pointer
   a1af471b4151 IB/srp: fix invalid indirect_sg_entries parameter value
   c2293e76babe IB/srp: fix mr allocation when the device supports sg gaps
   24be606cd3b4 IB/iser: Fix sg_tablesize calculation
   95600605ffaa IB/cxgb3: fix misspelling in header guard
   d7f56ee1198a s390/ptrace: Preserve previous registers for short regset write
   62d7f2123f19 s390/mm: Fix cmma unused transfer from pgste into pte
   97a2e39b7ab9 RDMA/cma: Fix unknown symbol when CONFIG_IPV6 is not enabled
   ffb97c11d05f Btrfs: remove ->{get, set}_acl() from btrfs_dir_ro_inode_operations
   ad80fada9d6d Btrfs: disable xattr operations on subvolume directories
   79babd4a6ce2 Btrfs: remove old tree_root case in btrfs_read_locked_inode()
   959f9709c025 ISDN: eicon: silence misleading array-bounds warning
   485952414360 xfs: prevent quotacheck from overloading inode lru
   03707d6c36f9 sysctl: fix proc_doulongvec_ms_jiffies_minmax()
   c7556867782e userns: Make ucounts lock irq-safe
   13e39d5930ec vring: Force use of DMA API for ARM-based systems with legacy devices
   96e5cec10e7a mm, page_alloc: fix premature OOM when racing with cpuset mems update
   b678e4ff7ce0 mm, page_alloc: move cpuset seqcount checking to slowpath
   d1656c5aef4d mm, page_alloc: fix fast-path race with cpuset update or removal
   ade7afe9dca6 mm, page_alloc: fix check for NULL preferred_zone
   9b1a1ae9b528 mm/mempolicy.c: do not put mempolicy before using its nodemask
   6676aa65464f mm/huge_memory.c: respect FOLL_FORCE/FOLL_COW for thp
   a2104c7cd3b2 drm/atomic: clear out fence when duplicating state
   bbae3c452596 Revert "drm/radeon: always apply pci shutdown callbacks"
   5270c017f19f drm/vc4: fix a bounds check
   cfba2a001d0e drm/vc4: Return -EINVAL on the overflow checks failing.
   b9edac54cb85 drm/vc4: Fix an integer overflow in temporary allocation layout.
   32600835ebe1 drm/vc4: Fix memory leak of the CRTC state.
   4c741e2adb35 drm/i915: Ignore bogus plane coordinates on SKL when the plane is not visible
   f1dc9aaee029 drm: Fix broken VT switch with video=1366x768 option
   2abb7f408f7c drm: Schedule the output_poll_work with 1s delay if we have delayed event
   e4be4d4942b9 tile/ptrace: Preserve previous registers for short regset write
   544160b6ea18 fbdev: color map copying bounds checking

Signed-off-by: Bruce Ashfield <bruce.ashfield@windriver.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
8 years agokernel-yocto.bbclass: ensure repatching when HEAD is checked out
Awais Belal [Tue, 14 Feb 2017 21:14:15 +0000 (16:14 -0500)]
kernel-yocto.bbclass: ensure repatching when HEAD is checked out

In some cases it is seen that kernel_checkout and validate_branches are
run again in simultaneous builds. During do_patch the kgit-s2q mechanism
looks for a sentinel file inside the .git directory, finds a fence post
and starts picking up patches after that.

This can create trouble as validate_branches checks out the HEAD of the
branch and so the patches should be reapplied rather than skipped due to
finding of the fence post.

We can call kgit-s2q --clean to remove the sentinel file when the
branches are checked out.

Signed-off-by: Awais Belal <awais_belal@mentor.com>
Signed-off-by: Bruce Ashfield <bruce.ashfield@windriver.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
8 years agowic: direct: move creation of PartitionedImage to __init__
Ed Bartosh [Fri, 10 Feb 2017 14:44:03 +0000 (16:44 +0200)]
wic: direct: move creation of PartitionedImage to __init__

Moved creation of PartitionedImage object from DirectPlugin.create
method to init. It makes the code a bit more readable and logical.

Signed-off-by: Ed Bartosh <ed.bartosh@linux.intel.com>
8 years agowic: direct: move generation of part.realnum to PartitionedImage
Ed Bartosh [Fri, 10 Feb 2017 14:37:22 +0000 (16:37 +0200)]
wic: direct: move generation of part.realnum to PartitionedImage

Moved the code that generates real partition numbers from DirectPlugin
to PartitionedImage.

The idea is to have all code that works with partitions
in PartitionedImage class.

Signed-off-by: Ed Bartosh <ed.bartosh@linux.intel.com>
8 years agowic: direct: add PartitionedImage.prepare method
Ed Bartosh [Fri, 10 Feb 2017 14:23:59 +0000 (16:23 +0200)]
wic: direct: add PartitionedImage.prepare method

Moved code that calls prepare method of Partition objects
from DirectPlugin to PartitionedImage.prepare.

The idea is to have all code that works with partitions
in PartitionedImage class.

Signed-off-by: Ed Bartosh <ed.bartosh@linux.intel.com>
8 years agowic: direct: set bootloader.source in the __init__
Ed Bartosh [Fri, 10 Feb 2017 10:57:13 +0000 (12:57 +0200)]
wic: direct: set bootloader.source in the __init__

Moved setting of bootloader source from do_create method
to __init__ as it doesn't have anything to do with image
creation.

Signed-off-by: Ed Bartosh <ed.bartosh@linux.intel.com>
8 years agowic: direct: move UUID generation to PartitionedImage
Ed Bartosh [Fri, 10 Feb 2017 10:41:15 +0000 (12:41 +0200)]
wic: direct: move UUID generation to PartitionedImage

Moved code that generates partition UUIDs from DirectPlugin to
PartitionedImage class as it's more logical to have it there.
This allows the code to be reused by other imager plugins.

Got rid of having yet another list of partitions in PartitionedImage.
Reused the list passed from DirectPlugin.

Signed-off-by: Ed Bartosh <ed.bartosh@linux.intel.com>
8 years agowic: direct: add 'realnum' attribute to partition
Ed Bartosh [Fri, 10 Feb 2017 10:10:22 +0000 (12:10 +0200)]
wic: direct: add 'realnum' attribute to partition

Replaced call of _get_part_num method with an attribute.
This eliminates the need to call the method and loop over
partitions every time we need to know realnum for partition.

Signed-off-by: Ed Bartosh <ed.bartosh@linux.intel.com>
8 years agowic: direct: remove useless code
Ed Bartosh [Thu, 9 Feb 2017 18:45:20 +0000 (20:45 +0200)]
wic: direct: remove useless code

Removed catching CreatorError and raising it again.

Signed-off-by: Ed Bartosh <ed.bartosh@linux.intel.com>
8 years agowic: direct: don't catch ImagerError
Ed Bartosh [Thu, 9 Feb 2017 18:16:45 +0000 (20:16 +0200)]
wic: direct: don't catch ImagerError

Don't transform ImagerError exception into warning.
Let wic to catch it on the upper level.

Signed-off-by: Ed Bartosh <ed.bartosh@linux.intel.com>
8 years agowic: direct: remove set_bootimg_dir setter
Ed Bartosh [Thu, 9 Feb 2017 14:48:55 +0000 (16:48 +0200)]
wic: direct: remove set_bootimg_dir setter

Removed java-like setter set_bootimg_dir. It's more pythonic
to access public attributes directly.

Signed-off-by: Ed Bartosh <ed.bartosh@linux.intel.com>
8 years agowic: ksparser: set default disk to 'sda'
Ed Bartosh [Thu, 9 Feb 2017 14:37:57 +0000 (16:37 +0200)]
wic: ksparser: set default disk to 'sda'

Set default value of --ondisk to 'sda' to ensure
we always have disk name for the partition.

This is a first step of replacing --ondisk with
disk <name> attribute of .wks. This is better as
all partitions share the same disk.

Signed-off-by: Ed Bartosh <ed.bartosh@linux.intel.com>
8 years agowic: remove utils/oe/__init__.py
Ed Bartosh [Thu, 9 Feb 2017 14:32:40 +0000 (16:32 +0200)]
wic: remove utils/oe/__init__.py

This file and utils/oe folder are not needed anymore as
all modules were removed or moved out of this directory.

Signed-off-by: Ed Bartosh <ed.bartosh@linux.intel.com>
8 years agowic: move PartitionedImage class to direct.py
Ed Bartosh [Thu, 9 Feb 2017 14:28:54 +0000 (16:28 +0200)]
wic: move PartitionedImage class to direct.py

As PartitionedImage is only used in direct.py it makes sense
to move it there. It's easier to maintain (and refactor) it
this way.

Signed-off-by: Ed Bartosh <ed.bartosh@linux.intel.com>
8 years agowic: move disk operations to PartitionImage class
Ed Bartosh [Thu, 9 Feb 2017 13:52:55 +0000 (15:52 +0200)]
wic: move disk operations to PartitionImage class

Disk operations were spread over DirectPlugin, DiskImage and Image
code making the code hard to understand.

Renamed Image class to PartitionedImage.
Removed DiskImage class.
Moved disk operations to PartitionedImage.

There was an implicit support for multiple disks: if different devices
were specified in .wks file (e.g. --ondisk sda and --ondisk sdb), wic
would theoretically generate multiple images. This is quite confusing
option and the code supporting it was broken for a long time. The same
effect (multiple output images) can be achieved in obvious and clear
way - by using multiple .wks files.

This functionality was removed. PartitionedImage works only with
one image. This makes the code less complex and easier to maintain.

Signed-off-by: Ed Bartosh <ed.bartosh@linux.intel.com>
8 years agoselftest/wic: Add tests for --exclude-dir option.
Kristian Amlie [Mon, 6 Feb 2017 16:16:47 +0000 (17:16 +0100)]
selftest/wic: Add tests for --exclude-dir option.

Based partially on an earlier patch by Maciej Borzecki.

Note that because tools now reside in recipe specific sysroots, we
need to import the path from bitbake and apply it during the test.

Signed-off-by: Kristian Amlie <kristian.amlie@mender.io>
Signed-off-by: Ross Burton <ross.burton@intel.com>
8 years agowic: Add --exclude-path option to rootfs source plugin.
Kristian Amlie [Mon, 6 Feb 2017 16:16:46 +0000 (17:16 +0100)]
wic: Add --exclude-path option to rootfs source plugin.

It will omit the given path from the resulting partition, and if the
given path ends in a slash, it will only delete the content, and keep
the directory.

Since mkfs only accepts whole directories as input, we need to copy
the rootfs directory to the workdir so that we can selectively delete
files from it.

Since we want to use the copyhardlinktree() function, we need to put
the generic oe lib in the module search path.

Signed-off-by: Kristian Amlie <kristian.amlie@mender.io>
Signed-off-by: Ross Burton <ross.burton@intel.com>
8 years agowic: remove unused argument scripts_path
Ed Bartosh [Wed, 8 Feb 2017 18:51:33 +0000 (20:51 +0200)]
wic: remove unused argument scripts_path

There is no need to pass scripts_path from main wic module
down the stack as it's not used there.

Removed scripts_path argument from DirectPlugin class
and wic_create function.

Signed-off-by: Ed Bartosh <ed.bartosh@linux.intel.com>
Signed-off-by: Ross Burton <ross.burton@intel.com>
8 years agowic: direct: remove unused import
Ed Bartosh [Wed, 8 Feb 2017 18:51:32 +0000 (20:51 +0200)]
wic: direct: remove unused import

Removed unused import of wic.errors module.

Signed-off-by: Ed Bartosh <ed.bartosh@linux.intel.com>
Signed-off-by: Ross Burton <ross.burton@intel.com>
8 years agowic: direct: remove unused plugin attributes
Ed Bartosh [Wed, 8 Feb 2017 18:51:31 +0000 (20:51 +0200)]
wic: direct: remove unused plugin attributes

Removed unused _disks, _disk_format and _disk_names
attributes from DirectPlugin class.

Signed-off-by: Ed Bartosh <ed.bartosh@linux.intel.com>
Signed-off-by: Ross Burton <ross.burton@intel.com>
8 years agowic: make sure layout_partitions is called once
Ed Bartosh [Wed, 8 Feb 2017 18:51:30 +0000 (20:51 +0200)]
wic: make sure layout_partitions is called once

Removed artificial _partitions_layed_out attribute and
unneeded call of layout_partitions method.

Signed-off-by: Ed Bartosh <ed.bartosh@linux.intel.com>
Signed-off-by: Ross Burton <ross.burton@intel.com>
8 years agowic: use the same partition object in direct and partitionedfs
Ed Bartosh [Wed, 8 Feb 2017 18:51:29 +0000 (20:51 +0200)]
wic: use the same partition object in direct and partitionedfs

Partition attributes were copied to the dictionary in partitionedfs
code, which makes the code hard to follow.

Used partition object passed from direct.py module as is in
partitionedfs.

Signed-off-by: Ed Bartosh <ed.bartosh@linux.intel.com>
Signed-off-by: Ross Burton <ross.burton@intel.com>
8 years agowic: direct: get rid of _get_parts getter
Ed Bartosh [Wed, 8 Feb 2017 18:51:28 +0000 (20:51 +0200)]
wic: direct: get rid of _get_parts getter

Replaced _get_parts getter with direct attribute
access to self.parts

Removed code that implicitly created partition
if there are no partitions mentioned in .wks file

Signed-off-by: Ed Bartosh <ed.bartosh@linux.intel.com>
Signed-off-by: Ross Burton <ross.burton@intel.com>
8 years agowic: partitionedfs: rename __create_partition and __add_disk
Ed Bartosh [Wed, 8 Feb 2017 18:51:27 +0000 (20:51 +0200)]
wic: partitionedfs: rename __create_partition and __add_disk

Renamed private methods with leading double underscores:
__create_partition -> _create_partition
__add_disk -> _add_disk

There is no point to have those names mangled, one underscore
is enough.

Signed-off-by: Ed Bartosh <ed.bartosh@linux.intel.com>
Signed-off-by: Ross Burton <ross.burton@intel.com>
8 years agowic: partitionedfs: get rid of __add_partition
Ed Bartosh [Wed, 8 Feb 2017 18:51:26 +0000 (20:51 +0200)]
wic: partitionedfs: get rid of __add_partition

3 lines long private method __add_partition is called only
from add_partition method. Merged them together to increase
readability.

Signed-off-by: Ed Bartosh <ed.bartosh@linux.intel.com>
Signed-off-by: Ross Burton <ross.burton@intel.com>
8 years agowic: partitionedfs: merged __format_disks and create
Ed Bartosh [Wed, 8 Feb 2017 18:51:25 +0000 (20:51 +0200)]
wic: partitionedfs: merged __format_disks and create

Private method __format_disks is called only from create
method making the code less readable. Merged the code
into one method.

Signed-off-by: Ed Bartosh <ed.bartosh@linux.intel.com>
Signed-off-by: Ross Burton <ross.burton@intel.com>
8 years agom4: use wildcard for selftest bbappend
Ross Burton [Fri, 10 Feb 2017 14:43:03 +0000 (14:43 +0000)]
m4: use wildcard for selftest bbappend

Signed-off-by: Ross Burton <ross.burton@intel.com>
8 years agobash: fix CVE-2016-9401
Li Zhou [Mon, 13 Feb 2017 02:53:24 +0000 (10:53 +0800)]
bash: fix CVE-2016-9401

popd in bash might allow local users to bypass the restricted shell
and cause a use-after-free via a crafted address.

Porting patch from <https://ftp.gnu.org/pub/gnu/bash/bash-4.4-patches/
bash44-006> to solve CVE-2016-9401.

Signed-off-by: Li Zhou <li.zhou@windriver.com>
Signed-off-by: Ross Burton <ross.burton@intel.com>
8 years agolibxslt: Add build fix (with ld-is-gold)
Jussi Kukkonen [Fri, 10 Feb 2017 14:04:38 +0000 (16:04 +0200)]
libxslt: Add build fix (with ld-is-gold)

Libraries must be linked with "-lm", otherwise gold fails to link
binaries with those libraries.

Signed-off-by: Jussi Kukkonen <jussi.kukkonen@intel.com>
Signed-off-by: Ross Burton <ross.burton@intel.com>
8 years agolighttpd: Upgrade 1.4.43 -> 1.4.45
Andrej Valek [Fri, 10 Feb 2017 12:42:29 +0000 (13:42 +0100)]
lighttpd: Upgrade 1.4.43 -> 1.4.45

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 agom4: 1.4.17 -> 1.4.18
Robert Yang [Fri, 10 Feb 2017 03:23:15 +0000 (19:23 -0800)]
m4: 1.4.17 -> 1.4.18

Rebased remove-gets.patch.

Signed-off-by: Robert Yang <liezhi.yang@windriver.com>
Signed-off-by: Ross Burton <ross.burton@intel.com>
8 years agomklibs-native: 0.1.41 -> 0.1.43
Robert Yang [Fri, 10 Feb 2017 03:23:14 +0000 (19:23 -0800)]
mklibs-native: 0.1.41 -> 0.1.43

Signed-off-by: Robert Yang <liezhi.yang@windriver.com>
Signed-off-by: Ross Burton <ross.burton@intel.com>
8 years agostrace: fix gawk's path
Robert Yang [Fri, 10 Feb 2017 03:23:13 +0000 (19:23 -0800)]
strace: fix gawk's path

Fixed:
strace-4.15: /usr/lib/strace/ptest/tests/caps-abbrev.awk contained in package strace-ptest requires /bin/gawk, but no providers found in RDEPENDS_strace-ptest? [file-rdeps]

The path should be /usr/bin/gawk as other scripts use in this package.

Signed-off-by: Robert Yang <liezhi.yang@windriver.com>
Signed-off-by: Ross Burton <ross.burton@intel.com>
8 years agostrace: 4.14 -> 4.15
Robert Yang [Fri, 10 Feb 2017 03:23:12 +0000 (19:23 -0800)]
strace: 4.14 -> 4.15

Signed-off-by: Robert Yang <liezhi.yang@windriver.com>
Signed-off-by: Ross Burton <ross.burton@intel.com>
8 years agowget: 1.18 -> 1.19
Robert Yang [Fri, 10 Feb 2017 03:23:11 +0000 (19:23 -0800)]
wget: 1.18 -> 1.19

Signed-off-by: Robert Yang <liezhi.yang@windriver.com>
Signed-off-by: Ross Burton <ross.burton@intel.com>
8 years agolibpcre: 8.39 -> 8.40
Robert Yang [Fri, 10 Feb 2017 03:23:10 +0000 (19:23 -0800)]
libpcre: 8.39 -> 8.40

The LIC_FILES_CHKSUM is changed because the date is changed, here is
diff result:
< Copyright (c) 1997-2017 University of Cambridge

Signed-off-by: Ross Burton <ross.burton@intel.com>
8 years agohdparm: 9.50 -> 9.51
Robert Yang [Fri, 10 Feb 2017 03:23:09 +0000 (19:23 -0800)]
hdparm: 9.50 -> 9.51

Signed-off-by: Robert Yang <liezhi.yang@windriver.com>
Signed-off-by: Ross Burton <ross.burton@intel.com>
8 years agognu-efi: 3.0.4 -> 3.0.5
Robert Yang [Fri, 10 Feb 2017 03:23:08 +0000 (19:23 -0800)]
gnu-efi: 3.0.4 -> 3.0.5

* Remove aarch64-initplat.c-fix-const-qualifier.patch which is already
  in the source.

Signed-off-by: Robert Yang <liezhi.yang@windriver.com>
Signed-off-by: Ross Burton <ross.burton@intel.com>
8 years agogit: 2.11.0 -> 2.11.1
Robert Yang [Fri, 10 Feb 2017 03:23:07 +0000 (19:23 -0800)]
git: 2.11.0 -> 2.11.1

Signed-off-by: Robert Yang <liezhi.yang@windriver.com>
Signed-off-by: Ross Burton <ross.burton@intel.com>
8 years agoe2fsprogs: 1.43.3 -> 1.43.4
Robert Yang [Fri, 10 Feb 2017 03:23:06 +0000 (19:23 -0800)]
e2fsprogs: 1.43.3 -> 1.43.4

* Rebased Revert-mke2fs-enable-the-metadata_csum-and-64bit-fea.patch.
* The mkfs.ext4dev is gone.

Signed-off-by: Robert Yang <liezhi.yang@windriver.com>
Signed-off-by: Ross Burton <ross.burton@intel.com>
8 years agousbutils: allow udev-hwdb to be optional
Gary Thomas [Fri, 10 Feb 2017 06:05:03 +0000 (07:05 +0100)]
usbutils: allow udev-hwdb to be optional

Use RRECOMMENDS for the udev hardware data base, to allow for this
to be suppressed if desired (saves many MB - useful for smaller systems)

Signed-off-by: Gary Thomas <gary@mlbassoc.com>
Signed-off-by: Ross Burton <ross.burton@intel.com>
8 years agoperf: add PACKAGECONFIG for jvmti
Stefan Müller-Klieser [Thu, 9 Feb 2017 11:49:39 +0000 (12:49 +0100)]
perf: add PACKAGECONFIG for jvmti

Recent kernel build jvmti, java profiling agent, if a host jdk is
detected. The library will be packaged in -dev and throws a qa error.
Disable the jvmti build per default.

Signed-off-by: Stefan Müller-Klieser <s.mueller-klieser@phytec.de>
Signed-off-by: Ross Burton <ross.burton@intel.com>
8 years agodbus: remove pointless cd
Ross Burton [Thu, 9 Feb 2017 16:47:21 +0000 (16:47 +0000)]
dbus: remove pointless cd

There's no need to cd to the target directory when creating symlinks, so don't.

[ YOCTO #10985 ]

Signed-off-by: Ross Burton <ross.burton@intel.com>
8 years agoelfutils: update homepage and upstream souce
Hongxu Jia [Wed, 8 Feb 2017 03:01:35 +0000 (22:01 -0500)]
elfutils: update homepage and upstream souce

In git://sourceware.org/git/elfutils.git:
----------------
commit 1700fd25e6caf26663af2bd994d1d99fab9df59f
Author: Mark Wielaard <mark@klomp.org>
Date:   Sat Dec 24 22:31:41 2016 +0100

    http://elfutils.org/ is now hosted at http://sourceware.org/elfutils/

    fedorahosted used to be our home, but we are now hosted at sourceware.

    Change the elfutils project home to http://elfutils.org/
    Point hosted services (email, release, git, bug tracker and web pages)
    to https://sourceware.org/elfutils/

    Move design notes from README to NOTES.
    Add URLs for home, releases, bugs, git and mailinglist to README.

    Make the --version output of all tools the same by using a common
    print_version function and update the publicly shown copyright holder
    to the elfutils developers.

Signed-off-by: Mark Wielaard <mark@klomp.org>
------------------

Signed-off-by: Hongxu Jia <hongxu.jia@windriver.com>
8 years agoelfutils: 0.166 -> 0.168
Hongxu Jia [Mon, 6 Feb 2017 09:34:46 +0000 (04:34 -0500)]
elfutils: 0.166 -> 0.168

- Backport patches from debian to 0.168 and add US tags.

- Rebase 0001-build-Provide-alternatives-for-glibc-assumptions-hel.patch to support libc musl

Signed-off-by: Hongxu Jia <hongxu.jia@windriver.com>
8 years agolibgpg-error: 1.25 -> 1.26
Hongxu Jia [Tue, 7 Feb 2017 04:30:46 +0000 (23:30 -0500)]
libgpg-error: 1.25 -> 1.26

- errorref.txt is part of the documentation so should be in PN-doc.

Signed-off-by: Hongxu Jia <hongxu.jia@windriver.com>
8 years agogpgme: allow building -native variant
Alexander Kanavin [Mon, 13 Feb 2017 15:05:31 +0000 (17:05 +0200)]
gpgme: allow building -native variant

This is required by various pieces in the dnf stack

Signed-off-by: Alexander Kanavin <alexander.kanavin@linux.intel.com>
Signed-off-by: Ross Burton <ross.burton@intel.com>
8 years agogpgme: 1.6.0 -> 1.8.0
Hongxu Jia [Tue, 7 Feb 2017 03:29:07 +0000 (22:29 -0500)]
gpgme: 1.6.0 -> 1.8.0

- Drop gpgme.pc, it is duplicated with pkgconfig.patch

- Rebase pkgconfig.patch to 1.8.0

- Split cmake config to ${PN}-dev

- Update LIC_FILES_CHKSUM, because copytright year changed.

- The libgpgme-pthread is obsolete by upstream;

- Add package libgpgcpp;

TODO: Do not remove gpgme-pthread.pc from pkgconfig.patch,
      the upstream explains:
      git://git.gnupg.org/gpgme.git
      commit 09b64554328445e99a8cc78fc34ea49c2ea2e7f9
      ------------------------------
      libgpgme-pthread is removed but gpgme-config still supports
      --thread=pthread for compatibility with find scripts.
      ------------------------------

Signed-off-by: Hongxu Jia <hongxu.jia@windriver.com>
8 years agosecurity_flags: disable -pie for gpgme
Ross Burton [Mon, 13 Feb 2017 14:23:02 +0000 (14:23 +0000)]
security_flags: disable -pie for gpgme

Otherwise this fails to build the libraries:

relocation R_X86_64_PC32 against undefined hidden symbol `__init_array_start'
can not be used when making a shared object

Signed-off-by: Ross Burton <ross.burton@intel.com>
8 years agognupg: 2.1.14 -> 2.1.18
Hongxu Jia [Tue, 7 Feb 2017 05:51:48 +0000 (00:51 -0500)]
gnupg: 2.1.14 -> 2.1.18

Update LIC_FILES_CHKSUM md5 checksum, because "http" has been
changed to "https".

Signed-off-by: Hongxu Jia <hongxu.jia@windriver.com>
8 years agorpcbind: 0.2.3 -> 0.2.4
Hongxu Jia [Tue, 7 Feb 2017 05:37:22 +0000 (00:37 -0500)]
rpcbind: 0.2.3 -> 0.2.4

Drop obsolete patch
- 0001-Avoid-use-of-glibc-sys-cdefs.h-header.patch

Drop backport patches:
- 0001-handle_reply-Don-t-use-the-xp_auth-pointer-directly.patch
- cve-2015-7236.patch

Signed-off-by: Hongxu Jia <hongxu.jia@windriver.com>
8 years agoman-pages: 4.08 -> 4.09
Hongxu Jia [Tue, 7 Feb 2017 05:05:55 +0000 (00:05 -0500)]
man-pages: 4.08 -> 4.09

Signed-off-by: Hongxu Jia <hongxu.jia@windriver.com>
8 years agolibgcrypt: 1.7.3 -> 1.7.6
Hongxu Jia [Tue, 7 Feb 2017 04:23:02 +0000 (23:23 -0500)]
libgcrypt: 1.7.3 -> 1.7.6

Signed-off-by: Hongxu Jia <hongxu.jia@windriver.com>
8 years agosocat: 1.7.3.1 -> 1.7.3.2
Hongxu Jia [Mon, 6 Feb 2017 08:57:03 +0000 (03:57 -0500)]
socat: 1.7.3.1 -> 1.7.3.2

Signed-off-by: Hongxu Jia <hongxu.jia@windriver.com>
8 years agopax-utils: 1.1.6 -> 1.2.2
Hongxu Jia [Mon, 6 Feb 2017 08:48:09 +0000 (03:48 -0500)]
pax-utils: 1.1.6 -> 1.2.2

Export GNULIB_OVERRIDES_WINT_T to fix compile failure
---------
1099 make[4]: *** [printf-args.o] Error 1
1100 make[4]: *** Waiting for unfinished jobs....
1101 In file included from ../../../pax-utils-1.2.2/autotools/gnulib/printf-args.h:41:0,
1102                  from ../../../pax-utils-1.2.2/autotools/gnulib/printf-parse.h:29,
1103                  from ../../../pax-utils-1.2.2/autotools/gnulib/printf-parse.c:36:
1104 ./wchar.h:476:6: error: #if with no expression
1105  # if
1106       ^
1107 make[4]: *** [printf-parse.o] Error 1
---------

Signed-off-by: Hongxu Jia <hongxu.jia@windriver.com>
8 years agowatchdog: Add RCONFLICTS to watchdog-keepalive and split RDEPENDS
Fabio Berton [Wed, 25 Jan 2017 11:22:05 +0000 (09:22 -0200)]
watchdog: Add RCONFLICTS to watchdog-keepalive and split RDEPENDS

watchdog-keepalive is a simplified version of the watchdog daemon
and can't be installed along with watchdog daemon, we need to
choose which daemon install.

Signed-off-by: Fabio Berton <fabio.berton@ossystems.com.br>
Signed-off-by: Ross Burton <ross.burton@intel.com>
8 years agolibcap: Add gperf-native to depends
Khem Raj [Wed, 8 Feb 2017 15:57:59 +0000 (07:57 -0800)]
libcap: Add gperf-native to depends

Fixes
|  gperf_case_strncmp (register const char *s1, register const char *s2, register size_t n)
|                                                                                 ^~~~~~
| ./_caps_output.gperf:96:53: error: unknown type name 'size_t'
|  __cap_hash_name (register const char *str, register size_t len)
|                                                      ^~~~~~
| ./_caps_output.gperf:200:55: error: unknown type name 'size_t'
|  __cap_lookup_name (register const char *str, register size_t len)
|                                                        ^~~~~~
| ./_caps_output.gperf:200:1: error: conflicting types for '__cap_lookup_name'
|  __cap_lookup_name (register const char *str, register size_t len)
|  ^~~~~~~~~~~~~~~~~
| ./_caps_output.gperf:33:29: note: previous declaration of '__cap_lookup_name' was here
|  const struct __cap_token_s *__cap_lookup_name(const char *, unsigned int);
|                              ^~~~~~~~~~~~~~~~~
| cap_text.c: In function 'cap_to_name':
| cap_text.c:291:2: warning: ignoring return value of 'asprintf', declared with attribute warn_unused_result [-Wunused-result]
|   asprintf(&tmp, "%u", cap);
|   ^~~~~~~~~~~~~~~~~~~~~~~~~
| make[1]: *** [Makefile:63: cap_text.o] Error 1

Signed-off-by: Khem Raj <raj.khem@gmail.com>
Signed-off-by: Ross Burton <ross.burton@intel.com>
8 years agolttng-tools: no need to delete libxml.m4
Ross Burton [Wed, 8 Feb 2017 14:44:29 +0000 (14:44 +0000)]
lttng-tools: no need to delete libxml.m4

The configure.ac uses pkg-config now and this macro doesn't exist in the tarball
anymore, so remove the deletion.

Signed-off-by: Ross Burton <ross.burton@intel.com>
8 years agosecurity_flags: enable string formatting check for console tools
Chen Qi [Wed, 8 Feb 2017 09:25:37 +0000 (17:25 +0800)]
security_flags: enable string formatting check for console tools

As we've fixed the string formatting issue in console-tools, we don't
need to override SECURITY_STRINGFORMAT for console-tools any more.

[YOCTO #9540]

Signed-off-by: Chen Qi <Qi.Chen@windriver.com>
Signed-off-by: Ross Burton <ross.burton@intel.com>
8 years agoconsole-tools: fix string formatting issue
Chen Qi [Wed, 8 Feb 2017 09:25:36 +0000 (17:25 +0800)]
console-tools: fix string formatting issue

Fix string formatting issue in console-tools. Otherwise, when compiling with
'-Wformat -Wformat-security -Werror=format-security', we would have compilation
error.

[YOCTO #9540]

Signed-off-by: Chen Qi <Qi.Chen@windriver.com>
Signed-off-by: Ross Burton <ross.burton@intel.com>
8 years agoltp: Fix containers/userns05 failure for lib32
He Zhe [Wed, 8 Feb 2017 02:58:46 +0000 (10:58 +0800)]
ltp: Fix containers/userns05 failure for lib32

Backport a patch to fix userns05 case:
<<<test_output>>>
user_namespace5 1 TFAIL : userns05.c:95: userns:parent should be not equal to cpid2
user_namespace5 1 TFAIL : userns05.c:95: userns:parent should be not equal to cpid2
user_namespace5 0 TINFO : Child process returned TPASS
user_namespace5 0 TINFO : Child process returned TPASS
user_namespace5 0 TINFO : Child process returned TFAIL
incrementing stop

Rename the previous patch to add a number prefix

Signed-off-by: He Zhe <zhe.he@windriver.com>
Signed-off-by: Ross Burton <ross.burton@intel.com>
8 years agopython: Don't use getentropy on Linux
Andreas Oberritter [Tue, 7 Feb 2017 23:48:54 +0000 (00:48 +0100)]
python: Don't use getentropy on Linux

Backport a patch from 2.7 branch to fix a regression with glibc
2.24 causing "OSError: [Errno 38] Function not implemented" when
calling urandom() with older kernels.

Signed-off-by: Andreas Oberritter <obi@opendreambox.org>
Signed-off-by: Ross Burton <ross.burton@intel.com>
8 years agopseudo: update to 1.8.2
Joshua Lock [Tue, 7 Feb 2017 15:53:00 +0000 (15:53 +0000)]
pseudo: update to 1.8.2

Update to the newly minted 1.8.2, dropping several patches we'd
backported since the last release.

Signed-off-by: Joshua Lock <joshua.g.lock@intel.com>
Signed-off-by: Ross Burton <ross.burton@intel.com>
8 years agosmartpm: if RPM transaction fails with an empty list of problems in attempt mode...
Herve Jourdain [Thu, 9 Jun 2016 07:41:31 +0000 (15:41 +0800)]
smartpm: if RPM transaction fails with an empty list of problems in attempt mode, ignore it!

Since commit be0cabf816a84c59e0e951b92a570121ed843822 on 2016/05/17, smartpm will retry the transaction if ts.run() returns with an empty list of problems, and we're in attempt mode.
But then, since all the packages are already installed, the transaction will fail with "package already installed" error.

In attempt mode, it's fine to not retry nor generate an error when the transaction fails for no reason linked to files/packages conflicts, so this patch fixes this behaviour.

Signed-off-by: Herve Jourdain <herve.jourdain@neuf.fr>
Signed-off-by: Ross Burton <ross.burton@intel.com>
8 years agoglibc: Update to 2.25 release
Khem Raj [Mon, 6 Feb 2017 19:38:04 +0000 (11:38 -0800)]
glibc: Update to 2.25 release

Signed-off-by: Khem Raj <raj.khem@gmail.com>
Signed-off-by: Ross Burton <ross.burton@intel.com>
8 years agopulseaudio: 9.0 -> 10.0
Tanu Kaskinen [Fri, 3 Feb 2017 07:06:35 +0000 (09:06 +0200)]
pulseaudio: 9.0 -> 10.0

Relase notes:
https://www.freedesktop.org/wiki/Software/PulseAudio/Notes/10.0/

The checksum of the LICENSE file changed due to some clarifications.
There were no changes to the actual licensing terms.

The LICENSE variable was not accurate, so I made changes to it.
Specifically:
 * there's no GPL code in PulseAudio so I dropped GPL from the list
 * the LGPL code allows using later versions of the license rather than
   limiting to just 2.1
 * there are some MIT and BSD licensed bits

I added more files to LIC_FILES_CHKSUM to have better coverage of all
the differently licensed code.

Dropped json-c and gdbm from DEPENDS. The new release doesn't use json-c
any more. gdbm isn't used when --with-database=simple is passed to
configure, so it should have been removed from DEPENDS a long time ago.

The new release dropped the Xen module, so the --without-xen configure
option isn't needed any more.

Added a comment for why --without-fftw is used.

Disabled the adrian echo canceller, because it has an unusual license,
and disabling the code was simpler than adding a new license to OE-Core.

Dropped upstreamed patches.

Signed-off-by: Ross Burton <ross.burton@intel.com>
8 years agoxz: upgrade to 5.2.3
Andreas Horsthemke [Sat, 4 Feb 2017 13:34:55 +0000 (14:34 +0100)]
xz: upgrade to 5.2.3

Signed-off-by: Andreas Horsthemke <andhorsthemke@gmail.com>
Signed-off-by: Ross Burton <ross.burton@intel.com>
8 years agolibepoxy: upgrade to 1.4.0
Ross Burton [Mon, 6 Feb 2017 23:26:34 +0000 (23:26 +0000)]
libepoxy: upgrade to 1.4.0

Drop the two patches that add the ability to disable GLX, this has been added
upstream.

Drop the Python3 patch as the code generator works with Py3 now.

(From OE-Core rev: 8ba197b3ae10a90ca988f0021597dfd20463c2d2)

Signed-off-by: Ross Burton <ross.burton@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
8 years agoselftest: remove result file log awkward checking
Benjamin Esquivel [Sun, 12 Feb 2017 19:00:38 +0000 (19:00 +0000)]
selftest: remove result file log awkward checking

because of the way that the test cases are looked in the results,
the file logger is incompatible with extended implementations of the
unittest runner.

as the xml runner extends the unittest runner, it shares the id()
method which returns the full name of the test, not only the test
function name. With that, a single check of the full name reviews at
the same time the class name as well as the function name.

[YOCTO#11012]

Signed-off-by: Benjamin Esquivel <benjamin.esquivel@linux.intel.com>
Signed-off-by: Ross Burton <ross.burton@intel.com>
8 years agobuild-perf-test-wrapper.sh: fix issues with non-existing path arguments
Markus Lehtonen [Fri, 27 Jan 2017 13:24:25 +0000 (15:24 +0200)]
build-perf-test-wrapper.sh: fix issues with non-existing path arguments

Without the '-s' option realpath will error out if the given path does
not exist.

Signed-off-by: Markus Lehtonen <markus.lehtonen@linux.intel.com>
Signed-off-by: Ross Burton <ross.burton@intel.com>
8 years agooe-build-perf-test: drop --commit-results
Markus Lehtonen [Fri, 3 Feb 2017 17:58:54 +0000 (19:58 +0200)]
oe-build-perf-test: drop --commit-results

Drop support for committing results into Git repository. The
functionality was not directly related to testing and feels unnecessary
complication of the script. The functionality has been moved into a
separate oe-git-archive script.

[YOCTO #10582]

Signed-off-by: Markus Lehtonen <markus.lehtonen@linux.intel.com>
Signed-off-by: Ross Burton <ross.burton@intel.com>
8 years agobuild-perf-test-wrapper.sh: use oe-git-archive
Markus Lehtonen [Fri, 3 Feb 2017 17:53:55 +0000 (19:53 +0200)]
build-perf-test-wrapper.sh: use oe-git-archive

Start to use the new helper script for archiving results data in Git.

[YOCTO #10582]

Signed-off-by: Markus Lehtonen <markus.lehtonen@linux.intel.com>
Signed-off-by: Ross Burton <ross.burton@intel.com>