]> code.ossystems Code Review - openembedded-core.git/log
openembedded-core.git
4 years agoasciidoc: Switch to using the main branch
Peter Kjellerstedt [Wed, 24 Feb 2021 13:37:44 +0000 (14:37 +0100)]
asciidoc: Switch to using the main branch

Upstream has renamed the "master" branch to "main".

Signed-off-by: Peter Kjellerstedt <peter.kjellerstedt@axis.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
(cherry picked from commit 05637733cf89a81fd5d787402f5a917f279a1a0d)
Signed-off-by: Anuj Mittal <anuj.mittal@intel.com>
4 years agosstatesig.py: show an error instead of warning when sstate manifest isn't found
Martin Jansa [Tue, 23 Feb 2021 13:35:53 +0000 (14:35 +0100)]
sstatesig.py: show an error instead of warning when sstate manifest isn't found

* not sure if there are some valid use-cases for missing manifest, but
  recently I'm seeing increasing number of build failures where something
  from native recipe is missing (seen it with pseudo, autoconf, nodejs
  recently) and the only indication that something is wrong (before showing
  sometimes misleading error like:
  recipe-sysroot-native/usr/bin/node: No such file or directory
  is this warning:
  NOTE: Running task 7844 of 12431 (/meta-oe/meta-oe/recipes-devtools/nodejs/nodejs_12.20.2.bb:do_prepare_recipe_sysroot)
  NOTE: recipe nodejs-12.20.2-r0: task do_prepare_recipe_sysroot: Started
  WARNING: nodejs-12.20.2-r0 do_prepare_recipe_sysroot: Manifest /BUILD/sstate-control/manifest-x86_64_ubuntu-18.04-nodejs-native.populate_sysroot not found in x86_64 x86_64_ubuntu-18.04 (variant '')?
  NOTE: Running task 7845 of 12431 (/meta-oe/meta-oe/recipes-devtools/nodejs/nodejs_12.20.2.bb:do_unpack)
  NOTE: recipe nodejs-12.20.2-r0: task do_unpack: Started
  WARNING: nodejs-12.20.2-r0 do_prepare_recipe_sysroot: Manifest /BUILD/sstate-control/manifest-x86_64_ubuntu-18.04-nodejs-native.populate_sysroot not found in x86_64 x86_64_ubuntu-18.04 (variant '')?
  NOTE: recipe nodejs-12.20.2-r0: task do_prepare_recipe_sysroot: Succeeded

  if I rebuild that native dependency, then it gets fixed and I don't
  see these failures in clean builds (as without sstate and with empty
  TMPDIR), only in incremental builds

* but if there isn't valid reason for missing manifest file, then I think
  it would be better to error early (or even bb.fatal())

Signed-off-by: Martin Jansa <Martin.Jansa@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
(cherry picked from commit 95607a26854d873399d2b9d7e5fcbffc0cbdba4c)
Signed-off-by: Anuj Mittal <anuj.mittal@intel.com>
4 years agowpa-supplicant: fix CVE-2021-0326
Stefan Ghinea [Tue, 23 Feb 2021 19:20:28 +0000 (21:20 +0200)]
wpa-supplicant: fix CVE-2021-0326

In p2p_copy_client_info of p2p.c, there is a possible out of bounds write
due to a missing bounds check. This could lead to remote code execution
if the target device is performing a Wi-Fi Direct search, with no
additional execution privileges needed. User interaction is not needed
for exploitation.Product: AndroidVersions: Android-10 Android-11
Android-8.1 Android-9 Android ID: A-172937525

References:
https://nvd.nist.gov/vuln/detail/CVE-2021-0326

Upstream patches:
https://w1.fi/cgit/hostap/commit/?id=947272febe24a8f0ea828b5b2f35f13c3821901e<links_for_CVE_patches>

Signed-off-by: Stefan Ghinea <stefan.ghinea@windriver.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
(cherry picked from commit b7940edabe100512e8f558cc37f9da836feae74d)
Signed-off-by: Anuj Mittal <anuj.mittal@intel.com>
4 years agobitbake.conf: Introduce FAKEROOTLOGS variable used by bitbake to print pseudo.log
Tomasz Dziendzielski [Mon, 22 Feb 2021 22:44:01 +0000 (23:44 +0100)]
bitbake.conf: Introduce FAKEROOTLOGS variable used by bitbake to print pseudo.log

Bitbake depends on this variable while looking for pseudo.log file in
case of fakeroot task failure.

Signed-off-by: Tomasz Dziendzielski <tomasz.dziendzielski@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
(cherry picked from commit e8e9e5f62b144828387e20eec2e66fcd61edd0d3)
Signed-off-by: Anuj Mittal <anuj.mittal@intel.com>
4 years agobabeltrace2: Fix reproducibility
Richard Purdie [Tue, 23 Feb 2021 13:34:25 +0000 (13:34 +0000)]
babeltrace2: Fix reproducibility

babeltrace2-ptest was referencing the host's python version in it's
makefiles. Whilst this could be fixed, its easier to correct the
python version being found to the target one which works much
more effectively and solves the issue.

Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
(cherry picked from commit 559f16d2fef4f8b04a9db1104b4a5b63f7876c68)
Signed-off-by: Anuj Mittal <anuj.mittal@intel.com>
4 years agoreport-error.bbclass: Add layer and bitbake version info to error report
Milan Shah [Wed, 6 Jan 2021 13:38:37 +0000 (19:08 +0530)]
report-error.bbclass: Add layer and bitbake version info to error report

Instead of just providing local.conf info, add layer names and their
revisions with bitbake version information into error report
makes it easier to understand and reproduce failed build.

[YOCTO #9700]

Signed-off-by: Milan Shah <mshah@mvista.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
(cherry picked from commit a5afd7b1857c0012ab6e3d8bbafa67a96ff9e9e0)
Signed-off-by: Anuj Mittal <anuj.mittal@intel.com>
4 years agopython3: Fix python interpreter line length for nativesdk
Mingli Yu [Mon, 22 Feb 2021 01:21:31 +0000 (09:21 +0800)]
python3: Fix python interpreter line length for nativesdk

Make sure the python interpreter is "#!/usr/bin/env python3" for
nativesdk to avoid the shebang path exceeding the limit when install
the sdk under the directory with long path.

Before:
 $ cd ${target_sdk_dir}
 $ vi ./sysroots/x86_64-oesdk-linux/usr/bin/2to3
 #!${target_sdk_dir}/sysroots/${SDK_SYS}/usr/bin/python3.9
 [snip]

After:
 $ cd ${target_sdk_dir}
 $ vi ./sysroots/x86_64-oesdk-linux/usr/bin/2to3
 #!/usr/bin/env python3
 [snip]

Signed-off-by: Mingli Yu <mingli.yu@windriver.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
(cherry picked from commit a748e6098fa7d2ff594319937c7e0cf97700e83a)
Signed-off-by: Anuj Mittal <anuj.mittal@intel.com>
4 years agolibevdev: Update patch status to backport
Richard Purdie [Sat, 20 Feb 2021 15:08:23 +0000 (15:08 +0000)]
libevdev: Update patch status to backport

The patch was submitted and merged upstream.

Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
(cherry picked from commit 871bce0703ca9d14e5c44f6ee0b66fcb13cfb630)
Signed-off-by: Anuj Mittal <anuj.mittal@intel.com>
4 years agorsync: Fix group name determinism issue
Richard Purdie [Mon, 22 Feb 2021 12:35:03 +0000 (12:35 +0000)]
rsync: Fix group name determinism issue

configure falls back to inspecting the host#s /etc/group for the
nobody/nogroup naming. Pass the correct value to configure to
ensure host differences don't sneak in.

Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
(cherry picked from commit e5f809068b32df32e506f3067cd6086195e03f12)
Signed-off-by: Anuj Mittal <anuj.mittal@intel.com>
4 years agorsync: Fix a file sorting determinism issue
Richard Purdie [Sun, 21 Feb 2021 09:45:48 +0000 (09:45 +0000)]
rsync: Fix a file sorting determinism issue

Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
(cherry picked from commit f6b7f395ae2d98363e365d31f5e95ce41a12e5b4)
Signed-off-by: Anuj Mittal <anuj.mittal@intel.com>
4 years agoopenssl: upgrade 1.1.1i -> 1.1.1j
Wang Mingyu [Mon, 22 Feb 2021 06:11:30 +0000 (14:11 +0800)]
openssl: upgrade 1.1.1i -> 1.1.1j

Signed-off-by: Wang Mingyu <wangmy@cn.fujitsu.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
(cherry picked from commit a67635ca2c7a016efcf450e4011f2032883e995d)
Signed-off-by: Anuj Mittal <anuj.mittal@intel.com>
4 years agosystemd: Fix importd requirements comment
Vivien Didelot [Sat, 20 Feb 2021 22:59:53 +0000 (17:59 -0500)]
systemd: Fix importd requirements comment

The comment above the importd config states curl, which does not
exist in PACKAGECONFIG. Adding it results in the following error
(and then misleading compilation failures):

    WARNING: systemd-1_244.5-r0 do_configure: QA Issue: systemd: invalid PACKAGECONFIG: curl [invalid-packageconfig]

Support for curl is currently enabled through journal-upload:

    PACKAGECONFIG[journal-upload] = "-Dlibcurl=true,-Dlibcurl=false,curl"

While it might be more appropriate to deprecate "journal-upload"
in favor of a "curl" PACKAGECONFIG entry, simply fix the importd
comment for now by stating upload-journal instead of curl.

Signed-off-by: Vivien Didelot <vdidelot@pbsc.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
(cherry picked from commit e6a0bced605da58ff43ff06ea6a1d67e84c98d91)
Signed-off-by: Anuj Mittal <anuj.mittal@intel.com>
4 years agolinux-firmware: upgrade 20201218 -> 20210208
Richard Purdie [Tue, 16 Feb 2021 05:05:06 +0000 (05:05 +0000)]
linux-firmware: upgrade 20201218 -> 20210208

License file had a number of new additions for new firmware but no
new licenses that didn't already exist.

Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
(cherry picked from commit eab46e65feae24311f0ee0cd05b7ebb2e3b07363)
Signed-off-by: Anuj Mittal <anuj.mittal@intel.com>
4 years agowpebackend-fdo: Fix missing .so symlink when using dev package
Zbigniew Bodek [Fri, 19 Feb 2021 17:19:54 +0000 (02:19 +0900)]
wpebackend-fdo: Fix missing .so symlink when using dev package

Problem spotted on dunfell-23.0.4

Normally, unversioned .so is removed from -dev (default) package
however it is still being referred by other agents, such as webkitgtk.
That can be observed while executing i.e. Epiphany Browser.

For instance:
platformInitializeWebProcess()
calls
wpe_loader_init("libWPEBackend-fdo-1.0.so")

as a result one can see:
--------------------
wpe_loader_init could not load the library 'libWPEBackend-fdo-1.0.so':
libWPEBackend-fdo-1.0.so: cannot open shared object file:
No such file or directory
--------------------

"master" branch contains a bumped-up version of webkitgtk
to which this issue does not apply (the code has changed).

Forcefully keep this .so symlink to resolve the problem in dunfell.

Signed-off-by: Zbigniew Bodek <zbigniew.bodek@huawei.com>
Signed-off-by: Anuj Mittal <anuj.mittal@intel.com>
4 years agooeqa/commands: Fix compatibility with python 3.9
Richard Purdie [Sat, 21 Nov 2020 10:50:07 +0000 (10:50 +0000)]
oeqa/commands: Fix compatibility with python 3.9

Python 3.9 dropped isAlive() so use the preferred is_alive().

Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
(cherry picked from commit 9bb06428cbb2ac0f3d98a1696f050d3393385503)
Signed-off-by: Anuj Mittal <anuj.mittal@intel.com>
4 years agooe/recipeutils: Fix copying patches when BBLAYERS entries are not normalised
Andrei Gherzan [Thu, 18 Feb 2021 21:14:03 +0000 (21:14 +0000)]
oe/recipeutils: Fix copying patches when BBLAYERS entries are not normalised

`devtool` uses `copy_recipe_files` for the upgrade operation when
creating the new, workspace recipe. Before handling the copy operations,
the function checks the entry in `SRC_URI` against `FILE` while in turn
uses absolute paths. When BBLAYERS contains entries that are not
normalised, this check will fail resulting in having the recipe in the
workspace without the initial patches.

Signef-off-by: Robert Drab <robert.drab@huawei.com>
Signed-off-by: Andrei Gherzan <andrei.gherzan@huawei.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
(cherry picked from commit 777fc24ab2c332954b56dac28cd9b3032808828c)
Signed-off-by: Anuj Mittal <anuj.mittal@intel.com>
4 years agopackage_rpm: Enable use_source_date_epoch_as_buildtime in package_rpm class
Jan-Simon Möller [Thu, 18 Feb 2021 14:25:16 +0000 (15:25 +0100)]
package_rpm: Enable use_source_date_epoch_as_buildtime in package_rpm class

Adding --define 'use_source_date_epoch_as_buildtime 1' to rpmbuild
ensure that the rpm header does have a consistent BUILDTIME tag.

This allows the rpms to be built reproducibly.

Signed-off-by: Jan-Simon Möller <jsmoeller@linuxfoundation.org>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
(cherry picked from commit f16721df6b3d0d3e681d9a1b3393417bd62b6275)
Signed-off-by: Anuj Mittal <anuj.mittal@intel.com>
4 years agomtd-utils: Remove duplicate assignments to alternative link names
Thomas Viehweger [Sat, 13 Feb 2021 22:04:03 +0000 (23:04 +0100)]
mtd-utils: Remove duplicate assignments to alternative link names

The two duplicate lines are not needed. The existence is confusing.

Signed-off-by: Thomas Viehweger <patchesThomas.Vie@web.de>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
(cherry picked from commit 36bdb4faa90dc18bc020481eba82ee570b968c39)
Signed-off-by: Anuj Mittal <anuj.mittal@intel.com>
4 years agonpm.bbclass: avoid building target nodejs for native npm recipes
Yoann Congal [Tue, 9 Feb 2021 15:53:46 +0000 (16:53 +0100)]
npm.bbclass: avoid building target nodejs for native npm recipes

The current recipe unconditionally RDEPENDS on nodejs (the target one).
When building on the "-native recipe" of "BBCLASSEXTEND native" recipe,
the target nodejs is unnecessarily built.

This patch fixes this by only RDEPENDS on nodejs when building for the target.

Signed-off-by: Yoann Congal <yoann.congal@smile.fr>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
(cherry picked from commit 92a9a86df9e3bcffb13d2f8b5dcbe7822170f734)
Signed-off-by: Anuj Mittal <anuj.mittal@intel.com>
4 years agoscreen: fix CVE-2021-26937
Scott Murray [Mon, 22 Feb 2021 02:52:06 +0000 (21:52 -0500)]
screen: fix CVE-2021-26937

Apply patch from Debian to fix CVE-2021-26937.

Signed-off-by: Scott Murray <scott.murray@konsulko.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
(cherry picked from commit 6d688cc8672b1830ba10966492ab3f8a12d32e17)
Signed-off-by: Anuj Mittal <anuj.mittal@intel.com>
4 years agogo: Update to 1.15.8
Khem Raj [Mon, 22 Feb 2021 09:14:21 +0000 (17:14 +0800)]
go: Update to 1.15.8

Signed-off-by: Khem Raj <raj.khem@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
(cherry picked from commit 2b724599e29808014f5d9c1fdc70d20abab9b3e5)
Signed-off-by: Lee Chee Yang <chee.yang.lee@intel.com>
Signed-off-by: Anuj Mittal <anuj.mittal@intel.com>
4 years agocve-check: add include/exclude layers
akuster [Mon, 8 Feb 2021 05:51:30 +0000 (05:51 +0000)]
cve-check: add include/exclude layers

There are times when exluding or including a layer
may be desired. This provide the framwork for that via
two variables. The default is all layers in bblayers.

CVE_CHECK_LAYER_INCLUDELIST
CVE_CHECK_LAYER_EXCLUDELIST

Signed-off-by: Armin Kuster <akuster808@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
(cherry picked from commit 5fdde65ef58b4c1048839e4f9462b34bab36fc22)
Signed-off-by: Anuj Mittal <anuj.mittal@intel.com>
4 years agocve-check.bbclass: add layer to cve log
akuster [Mon, 8 Feb 2021 05:51:29 +0000 (05:51 +0000)]
cve-check.bbclass: add layer to cve log

Lets include whcih layer a package belongs to and
add it to the cve logs

Signed-off-by: Armin Kuster <akuster808@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
(cherry picked from commit 00d965bb42dc427749a4c3985af56ceffff80457)
Signed-off-by: Anuj Mittal <anuj.mittal@intel.com>
4 years agodf.py: Add feature check for read-only-rootfs
Wes Lindauer [Thu, 18 Feb 2021 06:20:55 +0000 (01:20 -0500)]
df.py: Add feature check for read-only-rootfs

If the target is using a read-only rootfs, the available space on '/'
will be zero. This will cause the test to incorrectly fail and skipping
seems appropriate in this case.

Signed-off-by: Wes Lindauer <wesley.lindauer@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
(cherry picked from commit af1dbea3c9b9b42a3e6803b231e425423d70e210)
Signed-off-by: Anuj Mittal <anuj.mittal@intel.com>
4 years agogroff: Fix determinism issue
Richard Purdie [Thu, 18 Feb 2021 15:01:03 +0000 (15:01 +0000)]
groff: Fix determinism issue

Sometimes bison would regenerate source files and sometimes it would not
This is likely related to the patching of generated files by on of the
patches.

Drop those changes and force the files to regenerate in all cases since
we depend on bison-native anyway. This ensures the results are always
consistent.

Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
(cherry picked from commit 4449fa226e94f7124215c5ead43aadda7967f3af)
Signed-off-by: Anuj Mittal <anuj.mittal@intel.com>
4 years agolinux-yocto/5.4: update to v5.4.98
Bruce Ashfield [Thu, 18 Feb 2021 03:59:37 +0000 (22:59 -0500)]
linux-yocto/5.4: update to v5.4.98

Updating linux-yocto/5.4 to the latest korg -stable release that comprises
the following commits:

    5b9a4104c902 Linux 5.4.98
    3654a0ed0bdc squashfs: add more sanity checks in xattr id lookup
    d78a70667738 squashfs: add more sanity checks in inode lookup
    a814355e7057 squashfs: add more sanity checks in id lookup
    848bcb0a1d96 Fix unsynchronized access to sev members through svm_register_enc_region
    78e2f71b89b2 bpf: Fix 32 bit src register truncation on div/mod
    8589eda99cb1 regulator: Fix lockdep warning resolving supplies
    513fee2aee13 blk-cgroup: Use cond_resched() when destroy blkgs
    d1eb41833408 i2c: mediatek: Move suspend and resume handling to NOIRQ phase
    618b65dbde7a SUNRPC: Handle 0 length opaque XDR object data properly
    19b56e8433e7 SUNRPC: Move simple_get_bytes and simple_get_netobj into private header
    fa758032a546 iwlwifi: mvm: guard against device removal in reprobe
    2fa76f19dc15 iwlwifi: mvm: invalidate IDs of internal stations at mvm start
    c82793ef4f3b iwlwifi: pcie: fix context info memory leak
    b301eaf27f86 iwlwifi: pcie: add a NULL check in iwl_pcie_txq_unmap
    01742ade9286 iwlwifi: mvm: take mutex for calling iwl_mvm_get_sync_time()
    8f630ed7e98e iwlwifi: mvm: skip power command when unbinding vif during CSA
    589cf152fe47 ASoC: ak4458: correct reset polarity
    e96d10250227 pNFS/NFSv4: Try to return invalid layout in pnfs_layout_process()
    a5c70e57c4c1 chtls: Fix potential resource leak
    8b6d5013cd70 ASoC: Intel: Skylake: Zero snd_ctl_elem_value
    db272cd2bc9e mac80211: 160MHz with extended NSS BW in CSA
    26548561cb92 regulator: core: avoid regulator_resolve_supply() race condition
    03d76df5f164 af_key: relax availability checks for skb size calculation
    968b1b034136 tracing/kprobe: Fix to support kretprobe events on unloaded modules
    5e1942063dc3 Linux 5.4.97
    40af962eb1d4 usb: host: xhci: mvebu: make USB 3.0 PHY optional for Armada 3720
    76ab33055fbc net: sched: replaced invalid qdisc tree flush helper in qdisc_replace
    e65d331755de net: dsa: mv88e6xxx: override existent unicast portvec in port_fdb_add
    5d3007b6cc7b net: ip_tunnel: fix mtu calculation
    90d7459d24b8 neighbour: Prevent a dead entry from updating gc_list
    271ea7072901 igc: Report speed and duplex as unknown when device is runtime suspended
    7018edb19a92 md: Set prev_flush_start and flush_bio in an atomic way
    e857e21eb200 iommu/vt-d: Do not use flush-queue when caching-mode is on
    5fdf672759e9 Input: xpad - sync supported devices with fork on GitHub
    03d56dab56ae iwlwifi: mvm: don't send RFH_QUEUE_CONFIG_CMD with no queues
    2d5705150707 x86/apic: Add extra serialization for non-serializing MSRs
    bc1a3aeeff0f x86/build: Disable CET instrumentation in the kernel
    40d0fff29761 mm: thp: fix MADV_REMOVE deadlock on shmem THP
    56d61cd652dd mm, compaction: move high_pfn to the for loop scope
    90ef21e5806f mm: hugetlb: remove VM_BUG_ON_PAGE from page_huge_active
    af5508b1e862 mm: hugetlb: fix a race between isolating and freeing page
    3264a763174f mm: hugetlb: fix a race between freeing and dissolving the page
    108f56ed354f mm: hugetlbfs: fix cannot migrate the fallocated HugeTLB page
    75be4852490f ARM: footbridge: fix dc21285 PCI configuration accessors
    b2640b08c43c KVM: x86: Update emulator context mode if SYSENTER xfers to 64-bit mode
    6d3201c77be5 KVM: SVM: Treat SVM as unsupported when running as an SEV guest
    f9be9445e494 nvme-pci: avoid the deepest sleep state on Kingston A2000 SSDs
    f9034fcb27c0 drm/amd/display: Revert "Fix EDID parsing after resume from suspend"
    6844143e2198 mmc: core: Limit retries when analyse of SDIO tuples fails
    68c825bd2726 smb3: fix crediting for compounding when only one request in flight
    eaf2f835b52c smb3: Fix out-of-bounds bug in SMB2_negotiate()
    00f581964b66 cifs: report error instead of invalid when revalidating a dentry fails
    fd6dc98f66ef xhci: fix bounce buffer usage for non-sg list case
    f6a47f2ce090 genirq/msi: Activate Multi-MSI early when MSI_FLAG_ACTIVATE_EARLY is set
    c9654bbe52b5 libnvdimm/dimm: Avoid race between probe and available_slots_show()
    d0f36951ead4 kretprobe: Avoid re-registration of the same kretprobe earlier
    e80f9021d5be fgraph: Initialize tracing_graph_pause at task creation
    efa17285b338 mac80211: fix station rate table updates on assoc
    ecdd962c4b9b ovl: fix dentry leak in ovl_get_redirect
    6b9a2e5c0c42 usb: host: xhci-plat: add priv quirk for skip PHY initialization
    ddc682d33024 usb: xhci-mtk: break loop when find the endpoint to drop
    32410786279f usb: xhci-mtk: skip dropping bandwidth of unchecked endpoints
    f4e4f067f94c usb: xhci-mtk: fix unreleased bandwidth data
    1f9e9c1048b8 usb: dwc3: fix clock issue during resume in OTG mode
    9d058a06149b usb: dwc2: Fix endpoint direction check in ep_from_windex
    2a968ab0d2dd usb: renesas_usbhs: Clear pipe running flag in usbhs_pkt_pop()
    4d1d959348c1 USB: usblp: don't call usb_set_interface if there's a single alt
    522567fe540d USB: gadget: legacy: fix an error code in eth_bind()
    e57d70c59bb7 memblock: do not start bottom-up allocations with kernel_end
    7e6dcaeadc0e nvmet-tcp: fix out-of-bounds access when receiving multiple h2cdata PDUs
    f1c87b4b2c7b ARM: dts: sun7i: a20: bananapro: Fix ethernet phy-mode
    d97a821b2e9c r8169: fix WoL on shutdown if CONFIG_DEBUG_SHIRQ is set
    c545879e8080 net: mvpp2: TCAM entry enable should be written after SRAM data
    bf0507fb2073 net: lapb: Copy the skb before sending a packet
    1cef1d46add8 net/mlx5: Fix leak upon failure of rule creation
    67b7f73bbe3f i40e: Revert "i40e: don't report link up for a VF who hasn't enabled queues"
    6380ef64b9eb igc: check return value of ret_val in igc_config_fc_after_link_up
    ec68581f7479 igc: set the default return value to -IGC_ERR_NVM in igc_write_nvm_srwr
    5ce999efcaa7 arm64: dts: ls1046a: fix dcfg address range
    68e798fa3c0e rxrpc: Fix deadlock around release of dst cached on udp tunnel
    98650c3d0e33 um: virtio: free vu_dev only with the contained struct device
    02531b5549eb bpf, cgroup: Fix problematic bounds check
    9146fffc5d2a bpf, cgroup: Fix optlen WARN_ON_ONCE toctou
    4921f81ce65a arm64: dts: rockchip: fix vopl iommu irq on px30
    831132b13f0d arm64: dts: amlogic: meson-g12: Set FL-adj property value
    829bf438cb39 Input: i8042 - unbreak Pegatron C15B
    0d6e0a192e2e arm64: dts: qcom: c630: keep both touchpad devices enabled
    96dcfabef504 USB: serial: option: Adding support for Cinterion MV31
    5ad95c521fd5 USB: serial: cp210x: add new VID/PID for supporting Teraoka AD2000
    dfa820563c67 USB: serial: cp210x: add pid/vid for WSDA-200-USB

Signed-off-by: Bruce Ashfield <bruce.ashfield@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
(cherry picked from commit 98bb6a3cb528abe02b6edc291d05e240985ca80e)
Signed-off-by: Anuj Mittal <anuj.mittal@intel.com>
4 years agolinux-yocto/5.4: update to v5.4.96
Bruce Ashfield [Thu, 18 Feb 2021 03:59:35 +0000 (22:59 -0500)]
linux-yocto/5.4: update to v5.4.96

Updating linux-yocto/5.4 to the latest korg -stable release that comprises
the following commits:

    d4716ee8751b Linux 5.4.96
    b1a1c262e4b0 workqueue: Restrict affinity change to rescuer
    5b1e4fc2984e kthread: Extract KTHREAD_IS_PER_CPU
    2d7ca4a84b58 objtool: Don't fail on missing symbol table
    88240f7ac221 drm/amd/display: Change function decide_dp_link_settings to avoid infinite looping
    53c10bbf9186 drm/amd/display: Update dram_clock_change_latency for DCN2.1
    c6eb3dfdac44 selftests/powerpc: Only test lwm/stmw on big endian
    805e9cdb5793 nvme: check the PRINFO bit before deciding the host buffer length
    8e59209d53c9 udf: fix the problem that the disc content is not displayed
    2d1593543418 ALSA: hda: Add Cometlake-R PCI ID
    c03ecc192c8e scsi: ibmvfc: Set default timeout to avoid crash during migration
    02cc1ee3e8d1 mac80211: fix fast-rx encryption check
    efd061fc77f3 ASoC: SOF: Intel: hda: Resume codec to do jack detection
    e0f1ba38f788 scsi: fnic: Fix memleak in vnic_dev_init_devcmd2
    935fa0d5a5c5 scsi: libfc: Avoid invoking response handler twice if ep is already completed
    335bbffdd90c scsi: scsi_transport_srp: Don't block target in failfast state
    b1f680ffc25b x86: __always_inline __{rd,wr}msr()
    d1aed452c05f platform/x86: intel-vbtn: Support for tablet mode on Dell Inspiron 7352
    c99ac7213638 platform/x86: touchscreen_dmi: Add swap-x-y quirk for Goodix touchscreen on Estar Beauty HD tablet
    6f705e80fb87 phy: cpcap-usb: Fix warning for missing regulator_disable
    fd4c12f31209 net_sched: gen_estimator: support large ewma log
    27afc7128345 btrfs: backref, use correct count to resolve normal data refs
    66bcf5f6f989 btrfs: backref, only search backref entries from leaves of the same root
    c3089b06d6fe btrfs: backref, don't add refs from shared block when resolving normal backref
    21a0c97fb27c btrfs: backref, only collect file extent items matching backref offset
    1960c3d40b69 tcp: make TCP_USER_TIMEOUT accurate for zero window probes
    55cb8e232f9a arm64: Do not pass tagged addresses to __is_lm_address()
    b28387cf8f1c arm64: Fix kernel address detection of __is_lm_address()
    11084836e5fb ACPI: thermal: Do not call acpi_thermal_check() directly
    1410d2b68207 Revert "Revert "block: end bio with BLK_STS_AGAIN in case of non-mq devs and REQ_NOWAIT""
    5e8776df14fa ibmvnic: Ensure that CRQ entry read are correctly ordered
    bc4e7277cc93 net: switchdev: don't set port_obj_info->handled true when -EOPNOTSUPP
    9edebe46010c net: dsa: bcm_sf2: put device node before return
    e89428970c23 Linux 5.4.95
    e7aeca61cb9b tcp: fix TLP timer not set when CA_STATE changes from DISORDER to OPEN
    3e93b9efc3bb team: protect features update by RCU to avoid deadlock
    489e35c6829a ASoC: topology: Fix memory corruption in soc_tplg_denum_create_values()
    9a4d367b2783 NFC: fix possible resource leak
    c929c76e98b0 NFC: fix resource leak when target index is invalid
    b2f4a59a2216 rxrpc: Fix memory leak in rxrpc_lookup_local
    6d25d788efa4 iommu/vt-d: Don't dereference iommu_device if IOMMU_API is not built
    66f4f98ee363 iommu/vt-d: Gracefully handle DMAR units with no supported address widths
    0551a2fd456c selftests: forwarding: Specify interface when invoking mausezahn
    e4405451dd6e nvme-multipath: Early exit if no path is available
    64a4ec1850f7 can: dev: prevent potential information leak in can_fill_info()
    4dc2395d8f14 net/mlx5e: Reduce tc unsupported key print level
    a66705277baf net/mlx5e: E-switch, Fix rate calculation for overflow
    dbc13deeec6a net/mlx5: Fix memory leak on flow table creation error flow
    02ef126a002d igc: fix link speed advertising
    873d1a4740d6 i40e: acquire VSI pointer only after VF is initialized
    cf9276211563 mac80211: pause TX while changing interface type
    46c67a4c1a76 iwlwifi: pcie: reschedule in long-running memory reads
    563daf7c0f4a iwlwifi: pcie: use jiffies for memory read spin time limit
    f39fce916a7b pNFS/NFSv4: Fix a layout segment leak in pnfs_layout_process()
    a3c5fec1e09f ASoC: Intel: Skylake: skl-topology: Fix OOPs ib skl_tplg_complete
    748c2cd57615 RDMA/cxgb4: Fix the reported max_recv_sge value
    72797bfc5f77 firmware: imx: select SOC_BUS to fix firmware build
    347feca03881 ARM: dts: imx6qdl-kontron-samx6i: fix i2c_lcd/cam default status
    f7c6e6c9b62c arm64: dts: ls1028a: fix the offset of the reset register
    78fc9ef35d18 xfrm: Fix wraparound in xfrm_policy_addr_delta()
    f5e7db4fcd38 selftests: xfrm: fix test return value override issue in xfrm_policy.sh
    dac256de1fe2 xfrm: fix disable_xfrm sysctl when used on xfrm interfaces
    a7edea0fe85a xfrm: Fix oops in xfrm_replay_advance_bmp
    deb8d5dfeb63 netfilter: nft_dynset: add timeout extension to template
    347a1a20b195 ARM: imx: build suspend-imx6.S with arm instruction set
    61bdab3d770b xen-blkfront: allow discard-* nodes to be optional
    65543408f257 tee: optee: replace might_sleep with cond_resched
    ad3d896ef55c drm/i915: Check for all subplatform bits
    59546420c51b drm/nouveau/svm: fail NOUVEAU_SVM_INIT ioctl on unsupported devices
    37ef9b59f479 mt7601u: fix rx buffer refcounting
    77771158182f mt7601u: fix kernel crash unplugging the device
    2c4f52b9cddf arm64: dts: broadcom: Fix USB DMA address translation for Stingray
    6aceac245059 leds: trigger: fix potential deadlock with libata
    2c7b4b25293a xen: Fix XenStore initialisation for XS_LOCAL
    632a7728da9b KVM: Forbid the use of tagged userspace addresses for memslots
    ba668a507788 KVM: x86: get smi pending status correctly
    ff5f6de29faf KVM: nVMX: Sync unsync'd vmcs02 state to vmcs12 on migration
    2fc14cafefb3 KVM: x86/pmu: Fix UBSAN shift-out-of-bounds warning in intel_pmu_refresh()
    c547d39feb65 KVM: x86/pmu: Fix HW_REF_CPU_CYCLES event pseudo-encoding in intel_arch_events[]
    e1ae9aab8029 btrfs: fix possible free space tree corruption with online conversion
    d30cb3d348b8 drivers: soc: atmel: add null entry at the end of at91_soc_allowed_list[]
    082dc611fdc8 drivers: soc: atmel: Avoid calling at91_soc_init on non AT91 SoCs
    53fd4e4003a6 PM: hibernate: flush swap writer after marking
    7f9a267c67af s390/vfio-ap: No need to disable IRQ after queue reset
    9077bc37d2d1 net: usb: qmi_wwan: added support for Thales Cinterion PLSx3 modem family
    8aba60ebcfc3 wext: fix NULL-ptr-dereference with cfg80211's lack of commit()
    720032d3dc84 ARM: dts: imx6qdl-gw52xx: fix duplicate regulator naming
    b24dc0aa7e9b media: rc: ensure that uevent can be read directly after rc device register
    5d6fd0357057 ALSA: hda/via: Apply the workaround generically for Clevo machines
    f78803928481 ALSA: hda/realtek: Enable headset of ASUS B1400CEPE with ALC256
    8de2109f4670 kernel: kexec: remove the lock operation of system_transition_mutex
    93603a27fc31 ACPI: sysfs: Prefer "compatible" modalias
    587c6b75d7fd nbd: freeze the queue while we're adding connections
    b8fcb8f53995 IPv6: reply ICMP error if the first fragment don't include all headers
    1f58e378a17e ICMPv6: Add ICMPv6 Parameter Problem, code 3 definition

Signed-off-by: Bruce Ashfield <bruce.ashfield@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
(cherry picked from commit b39a6de2c1175e4554c688a44c905c24367fa57a)
Signed-off-by: Anuj Mittal <anuj.mittal@intel.com>
4 years agovalgrind: Disable ptest nlcontrolc for x86-64
Yi Fan Yu [Thu, 18 Feb 2021 16:29:13 +0000 (11:29 -0500)]
valgrind: Disable ptest nlcontrolc for x86-64

Test hangs after glibc 2.33 uprev.

Using gdb `p t[0]` to modify the timeout
argument no longer affects how long `select` wait.

https://bugs.kde.org/show_bug.cgi?id=432870
[YOCTO #14223]

Signed-off-by: Yi Fan Yu <yifan.yu@windriver.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
(cherry picked from commit bb991a24952ebf02cf0092c194d2b3a54a8b2e5b)
Signed-off-by: Anuj Mittal <anuj.mittal@intel.com>
4 years agogit: Fix determinism issue
Richard Purdie [Wed, 17 Feb 2021 17:57:13 +0000 (17:57 +0000)]
git: Fix determinism issue

Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
(cherry picked from commit 9ae740939f8315c64fe7571f912404127a29dc89)
Signed-off-by: Anuj Mittal <anuj.mittal@intel.com>
4 years agoxorg-minimal-fonts: Really fix determinism
Richard Purdie [Wed, 17 Feb 2021 10:56:19 +0000 (10:56 +0000)]
xorg-minimal-fonts: Really fix determinism

My previous fix wasn't correct as the file timestamps do vary by git checkout
or modification time and aren't correct here. Instead use a specific
date/time for the files to be deterministic.

Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
(cherry picked from commit 725a30a30052540a4b7fc2933396fe9eb946eeac)
Signed-off-by: Anuj Mittal <anuj.mittal@intel.com>
4 years agoxorg-fonts-minimal: Fix reproducibility
Richard Purdie [Mon, 15 Feb 2021 23:41:29 +0000 (23:41 +0000)]
xorg-fonts-minimal: Fix reproducibility

When installing the font files, preserve their timestamp rather than using
the current time which fixes reproducibility issues.

Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
(cherry picked from commit a0c4f4b0182a995f0eb2709cc9b3c852527ab936)
Signed-off-by: Anuj Mittal <anuj.mittal@intel.com>
4 years agorootfs_deb: handle aarch64 SDK_ARCH
Ross Burton [Wed, 17 Feb 2021 14:45:56 +0000 (14:45 +0000)]
rootfs_deb: handle aarch64 SDK_ARCH

This class maps from OE architectures to dpkg architectures, but wasn't
handling aarch64 correctly.

Signed-off-by: Ross Burton <ross.burton@arm.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
(cherry picked from commit 315fe8ab3b5bb8a709be98cd8998d4525fbe8725)
Signed-off-by: Anuj Mittal <anuj.mittal@intel.com>
4 years agolocal.conf.sample.extended: prefer INIT_MANAGER
Vivien Didelot [Thu, 18 Feb 2021 22:42:02 +0000 (17:42 -0500)]
local.conf.sample.extended: prefer INIT_MANAGER

Since the INIT_MANAGER variable has been introduced, there is no need
to append the distro features and set the init manager manually.

Replace the busybox/mdev and systemd examples with the 4 values
currently supported for the INIT_MANAGER variable.

Signed-off-by: Vivien Didelot <vdidelot@pbsc.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
(cherry picked from commit 514454651522f97590d1403c50effd9c79df827a)
Signed-off-by: Anuj Mittal <anuj.mittal@intel.com>
4 years agoRevert "sanity.bbclass: Add az fetcher to accepted URI protocols for mirrors sanity"
Richard Purdie [Tue, 9 Mar 2021 12:01:55 +0000 (12:01 +0000)]
Revert "sanity.bbclass: Add az fetcher to accepted URI protocols for mirrors sanity"

This reverts commit c0befa46a2494a1fee46bcee853db6430940ba8f.

Was added to the wrong branch accidentally.

4 years agosanity.bbclass: Add az fetcher to accepted URI protocols for mirrors sanity
Alejandro Hernandez Samaniego [Wed, 24 Feb 2021 17:28:15 +0000 (10:28 -0700)]
sanity.bbclass: Add az fetcher to accepted URI protocols for mirrors sanity

Signed-off-by: Alejandro Enedino Hernandez Samaniego <alhe@linux.microsoft.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
4 years agoacpica: Fix reproducibility issues
Joshua Watt [Tue, 16 Feb 2021 14:43:03 +0000 (08:43 -0600)]
acpica: Fix reproducibility issues

Fixes reproducibility when build acpica yacc sources by instructing
bison on how to do the file prefix mapping

Signed-off-by: Joshua Watt <JPEWhacker@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
(cherry picked from commit d13a1df7bd173fca3fc2cdf889d5eb487e35d6fd)
Signed-off-by: Anuj Mittal <anuj.mittal@intel.com>
4 years agobison: Fix up file name mapping
Joshua Watt [Tue, 16 Feb 2021 14:43:02 +0000 (08:43 -0600)]
bison: Fix up file name mapping

File name mapping was added to bison, but needs to be applied in more
cases. Add a patch to enable file name mapping for symbols

Signed-off-by: Joshua Watt <JPEWhacker@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
(cherry picked from commit 5a213506377b0db4e88ca094215eadf08c21a2e1)
Signed-off-by: Anuj Mittal <anuj.mittal@intel.com>
4 years agosystemd: Re-enable chvt as non-root user without polkit
Joshua Watt [Mon, 16 Nov 2020 14:38:26 +0000 (08:38 -0600)]
systemd: Re-enable chvt as non-root user without polkit

systemd 245 introduced a regression in behavior where they removed
support for non-root users to chvt from a service file. This prevents
running compositors (e.g. weston) as any user other than root. The
intention is for polkit to be used to allow this (and in fact the
default polkit rules that ship with systemd allow this). However, polkit
is a huge dependency to bring in for an embedded system, and isn't
support by OE-core.

The patch has been proposed upstream to restore the previous behavior of
allowing a non-root user to chvt to unbreak the regression without
requiring polkit.

Upstream-Status: Submitted [https://github.com/systemd/systemd/pull/17494]
Signed-off-by: Joshua Watt <JPEWhacker@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
(cherry picked from commit ef8cf581b680f8b8ee35d50c75891b4fd8b91f2f)
Signed-off-by: Anuj Mittal <anuj.mittal@intel.com>
4 years agolocal.conf.sample.extended: fix double 'of' typo
Vivien Didelot [Tue, 16 Feb 2021 18:47:00 +0000 (13:47 -0500)]
local.conf.sample.extended: fix double 'of' typo

Remove the redundant 'of' word in the INITRAMFS_IMAGE comment.

Signed-off-by: Vivien Didelot <vdidelot@pbsc.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
(cherry picked from commit de9544f4654510ac33821b7f170de3074205a221)
Signed-off-by: Anuj Mittal <anuj.mittal@intel.com>
4 years agoxmlto: Fix reproducibility
Richard Purdie [Mon, 15 Feb 2021 23:46:06 +0000 (23:46 +0000)]
xmlto: Fix reproducibility

Don't hardcode the host's grep path into xmlto.

Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
(cherry picked from commit a7d78971df193c321c309481749fc30cae77788c)
Signed-off-by: Anuj Mittal <anuj.mittal@intel.com>
4 years agowatchdog: Avoid reproducibility failures after fixing build
Richard Purdie [Tue, 16 Feb 2021 11:22:33 +0000 (11:22 +0000)]
watchdog: Avoid reproducibility failures after fixing build

Add some temporary bumps to versions to change the output hash so
the fix applies correctly. Can be dropped next time we update the
recipe and the output changes.

Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
(cherry picked from commit 69205edcdff865048e55a6b7feaf82064ebc10c5)
Signed-off-by: Anuj Mittal <anuj.mittal@intel.com>
4 years agowatchdog: Fix determinism issue from sendmail host path
Richard Purdie [Mon, 15 Feb 2021 21:34:25 +0000 (21:34 +0000)]
watchdog: Fix determinism issue from sendmail host path

The build was injecting the path of the host's sendmail binary. Set
this deterministically to match OE's path for it.

Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
(cherry picked from commit 476328b91ea4417160580d28df4fcc1147d85ae2)
Signed-off-by: Anuj Mittal <anuj.mittal@intel.com>
4 years agooeqa: reproducible: Add more logging
Joshua Watt [Sun, 14 Feb 2021 19:12:58 +0000 (13:12 -0600)]
oeqa: reproducible: Add more logging

The reproducible build tests can take a long time, so having more
logging messages at various points in the build can help debug where the
build is taking a long time.

Signed-off-by: Joshua Watt <JPEWhacker@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
(cherry picked from commit 6b792afe8759d62af8e713b86dad8f6721961a05)
Signed-off-by: Anuj Mittal <anuj.mittal@intel.com>
4 years agooeqa: reproducible: Fix SSTATE_MIRRORS variable
Joshua Watt [Sun, 14 Feb 2021 19:12:41 +0000 (13:12 -0600)]
oeqa: reproducible: Fix SSTATE_MIRRORS variable

The SSTATE_MIRRORS variable was misspelled, which allowed the "clean"
test build to pull from the mirror.

Signed-off-by: Joshua Watt <JPEWhacker@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
(cherry picked from commit e42497bd84d0bb370a9f7b0448bff29f01fd1b0c)
Signed-off-by: Anuj Mittal <anuj.mittal@intel.com>
4 years agobuildtools-extended-tarball: Add glibc-gconvs needed for build
Richard Purdie [Sun, 14 Feb 2021 14:41:13 +0000 (14:41 +0000)]
buildtools-extended-tarball: Add glibc-gconvs needed for build

When building vim it tries to rebuild files using iconv. If this fails
the build continues anyway but the output is not determnistic as builds
using a hosttools tarball are different from builds where there isn't a
hosttools tarball. Add the needed gconvs to the tarball when iconv is
present to become determistic and generate vim locales consistently.

Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
(cherry picked from commit b945652a088f430a2adec6b968cd00c5928d4272)
Signed-off-by: Anuj Mittal <anuj.mittal@intel.com>
4 years agoquilt: Be determnistic about column presence
Richard Purdie [Sun, 14 Feb 2021 14:40:04 +0000 (14:40 +0000)]
quilt: Be determnistic about column presence

I've seen local differences in the quilt output depending on whether the column
binary was available in the sysroot. Fix determinism issues by being specific
about configuration.

Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
(cherry picked from commit 0cd5fba8634bcc679518f98cc25be66a51081372)
Signed-off-by: Anuj Mittal <anuj.mittal@intel.com>
4 years agopackage_manager/deb: Fix image generation with package removal
Richard Purdie [Sun, 14 Feb 2021 14:38:21 +0000 (14:38 +0000)]
package_manager/deb: Fix image generation with package removal

When building SDKs with the deb backend you could see errors like:

Setting up nativesdk-python3-ndg-httpsclient (0.5.1-r0) ...
mkdir: cannot create directory ‘/usr/lib/opkg’: Permission denied
dpkg: error processing package nativesdk-python3-ndg-httpsclient (--configure):

which is due to environment misconfiguration when removing packages.

Fix this by setting the same environment variables as used for installation.

Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
(cherry picked from commit f23c7e319a192ada14bb9a82822ef2967309aaea)
Signed-off-by: Anuj Mittal <anuj.mittal@intel.com>
4 years agodeb: export INTERCEPT_DIR for remove actions
Richard Leitner [Thu, 29 Oct 2020 18:20:14 +0000 (19:20 +0100)]
deb: export INTERCEPT_DIR for remove actions

During the do_populate_sdk task apt-get purge is called by deb's remove
function. This fails with error messages similiar to the following one
if any of the included packages uses intercepts as the INTERCEPT_DIR
isn't exported:

.../*.postinst: line 4: /postinst_intercept: No such file or directory

Therefore fix it by exporting the INTERCEPT_DIR variable within the
remove function.

Signed-off-by: Richard Leitner <richard.leitner@skidata.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
(cherry picked from commit f18adf53dd4bf5dd3adef82b2dcc34a6cdfd0c89)
Signed-off-by: Anuj Mittal <anuj.mittal@intel.com>
4 years agovim: Fix a race over creation of the desktop files
Richard Purdie [Mon, 15 Feb 2021 23:32:04 +0000 (23:32 +0000)]
vim: Fix a race over creation of the desktop files

The LINGUAS file can be written by two different Makefile targets
and if they race, the desktop file contents isn't deterministic.

Fix the makfile to avoid this.

Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
(cherry picked from commit 416bc7b697764075fbf73683cd8bddf36d839244)
Signed-off-by: Anuj Mittal <anuj.mittal@intel.com>
4 years agovim: Improve determinism
Richard Purdie [Sun, 14 Feb 2021 13:29:11 +0000 (13:29 +0000)]
vim: Improve determinism

Add a couple of configure options to avoid determism issues in the vim build.
This can happen due to the addition of glib-2.0 to the native sysroot through
later task additions to the sysroot through indirect dependencies.

Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
(cherry picked from commit 914f86054f5ea0a115767c1b3d9cdb4c4ef9545b)
Signed-off-by: Anuj Mittal <anuj.mittal@intel.com>
4 years agoweston-init: Fix weston-keyboard path in weston.ini
Marek Vasut [Sun, 14 Feb 2021 14:40:45 +0000 (15:40 +0100)]
weston-init: Fix weston-keyboard path in weston.ini

The weston-keyboard executable is installed into /usr/libexec
instead of /usr/lib/weston , correct the path in weston.ini .

Signed-off-by: Marek Vasut <marex@denx.de>
Cc: Khem Raj <raj.khem@gmail.com>
Cc: Richard Purdie <richard.purdie@linuxfoundation.org>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
(cherry picked from commit 56dee47a5ddb1da66d30f894a282d0658dcc930c)
Signed-off-by: Anuj Mittal <anuj.mittal@intel.com>
4 years agocwautomacros: Ensure version is set deterministically
Richard Purdie [Sun, 14 Feb 2021 12:05:10 +0000 (12:05 +0000)]
cwautomacros: Ensure version is set deterministically

The makefile injects the current date into the version file. Do this
deterministically with SOURCE_DATE_EPOCH.

Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
(cherry picked from commit 31f2ad739ea776a1e11b5cef5434df188007c7bf)
Signed-off-by: Anuj Mittal <anuj.mittal@intel.com>
4 years agoopkg: Fix patch glitches
Richard Purdie [Tue, 9 Feb 2021 13:15:24 +0000 (13:15 +0000)]
opkg: Fix patch glitches

The original patch contained some text which shouldn't have been there
and used brackets in configure which isn't a great idea. Tweak the patch
to resolve this.

Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
(cherry picked from commit 63cbf187fe189c99645fe3afee8a6361a9a32cdc)
Signed-off-by: Anuj Mittal <anuj.mittal@intel.com>
4 years agoopkg: Fix build reproducibility issue
Richard Purdie [Tue, 2 Feb 2021 23:45:18 +0000 (23:45 +0000)]
opkg: Fix build reproducibility issue

A build date was leaking into the generated docs and makefile used for
ptests leading to reproducibility issues each time the month changed.

Add a patch to use SOURCE_DATE_EPOCH to derive it if available.

Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
(cherry picked from commit 6a9ca7aec4991eabd425e32fdf85f51bb1686b8b)
Signed-off-by: Anuj Mittal <anuj.mittal@intel.com>
4 years agoopenssh: fix CVE-2020-14145
Lee Chee Yang [Wed, 10 Feb 2021 12:50:46 +0000 (20:50 +0800)]
openssh: fix CVE-2020-14145

Signed-off-by: Lee Chee Yang <chee.yang.lee@intel.com>
Signed-off-by: Anuj Mittal <anuj.mittal@intel.com>
4 years agoqemu: fix CVE-2020-29443 CVE-2020-35517
Lee Chee Yang [Wed, 10 Feb 2021 12:50:47 +0000 (20:50 +0800)]
qemu: fix CVE-2020-29443 CVE-2020-35517

Signed-off-by: Lee Chee Yang <chee.yang.lee@intel.com>
Signed-off-by: Anuj Mittal <anuj.mittal@intel.com>
4 years agopseudo: Update to include fixes for glibc 2.33
Richard Purdie [Mon, 15 Feb 2021 20:52:57 +0000 (20:52 +0000)]
pseudo: Update to include fixes for glibc 2.33

Pull in the changes:
makewrappers: Fix glibc 2.33 fstatat usage issues
ports/linux: Add wrapper for fstatat/fstatat64 in glibc 2.33

Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
(cherry picked from commit dfcb1c5eb2690046f96c2bb6724e091028ddc3ec)
Signed-off-by: Anuj Mittal <anuj.mittal@intel.com>
4 years agoweston: remoting backend requires GStreamer base plugins
Joshua Watt [Fri, 12 Feb 2021 16:51:46 +0000 (10:51 -0600)]
weston: remoting backend requires GStreamer base plugins

The remoting backend of weston requires the GStreamer base plugins, so
add them to the PACAKGECONFIG depends list.

Signed-off-by: Joshua Watt <JPEWhacker@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
(cherry picked from commit 0b45994656fee7c4b7bdb5bd8571f0c61217a182)
Signed-off-by: Anuj Mittal <anuj.mittal@intel.com>
4 years agolibomxil: Fix up commercial license flag
Joshua Watt [Fri, 12 Feb 2021 16:51:57 +0000 (10:51 -0600)]
libomxil: Fix up commercial license flag

The commercial license flag on libomxil is set because it may include
the Adaptive Multi-Rate audio codec (AMR) using FFmepg, which is patent
encumbered.

It turns out this component is disabled by default in the recipe; add a
PACKAGECONFIG to enable it and trigger the "commercial" LICENSE_FLAGS on
it. This make the default build configuration clean unless a user
specifically asks for AMR support, and prevents them from marking the
recipe with the "commerical" flag unnecessarily which could hide
potential problems later on.

Signed-off-by: Joshua Watt <JPEWhacker@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
(cherry picked from commit 5f61e20002c2af93e2d6810574e23606925526ee)
Signed-off-by: Anuj Mittal <anuj.mittal@intel.com>
4 years agotcf-agent: Fix build on riscv32
Khem Raj [Tue, 9 Feb 2021 18:57:22 +0000 (10:57 -0800)]
tcf-agent: Fix build on riscv32

LCL_STOP_SERVICES needs tcf/cpudefs-mdep.h ported

Signed-off-by: Khem Raj <raj.khem@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
(cherry picked from commit ed5e0de938469a7fa4e6cd725d9e0c8325d890d3)
Signed-off-by: Anuj Mittal <anuj.mittal@intel.com>
4 years agoconnman: update to 1.39
akuster [Wed, 10 Feb 2021 04:32:07 +0000 (04:32 +0000)]
connman: update to 1.39

Bug fix only and includes two security fixes:

CVE-2021-26675
CVE-2021-26676

Changelog:
- Fix issue with scanning state synchronization and iwd.
- Fix issue with invalid key with 4-way handshake offloading.
- Fix issue with DNS proxy length checks to prevent buffer overflow.
- Fix issue with DHCP leaking stack data via uninitialized variable.

[Yocto #14231]

Signed-off-by: Armin Kuster <akuster808@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
(cherry picked from commit eb20fd47d738f469f7bbeb4b8d85040f9163722b)
Signed-off-by: Anuj Mittal <anuj.mittal@intel.com>
4 years agopseudo: Update for rename and faccessat fixes
Richard Purdie [Fri, 12 Feb 2021 15:32:38 +0000 (15:32 +0000)]
pseudo: Update for rename and faccessat fixes

Pull in:

  ports/rename/renameat: Avoid race when renaming files
  ports/unix: Add faccessat and faccessat2
  ports/access.c: Use EACCES, not EPERM

which includes a fix for rename race issues causing pseudo aborts.

Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
(cherry picked from commit 330c232e4f756296331f9026e91ac26fd45f0315)
Signed-off-by: Anuj Mittal <anuj.mittal@intel.com>
4 years agooe-pkgdata-util: Check if environment script is initialized
Dorinda [Mon, 1 Feb 2021 14:46:38 +0000 (09:46 -0500)]
oe-pkgdata-util: Check if environment script is initialized

Tinfoil doesn't behave well if environment is not initialized, this check ensures a proper error log if environment is not initialized.

[YOCTO #12096]

Signed-off-by: Dorinda Bassey <dorindabassey@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
(cherry picked from commit e88073e16f1b4cfd0f97c81a988640a84adad674)
Signed-off-by: Anuj Mittal <anuj.mittal@intel.com>
4 years agowic: debug mode to keep tmp directory
Lee Chee Yang [Mon, 8 Feb 2021 01:22:55 +0000 (09:22 +0800)]
wic: debug mode to keep tmp directory

files in wic tmp directory can be usefull for debugging, so do not remove
tmp directory when wic create run with debugging mode (-D or --debug).

also update wic.Wic.test_debug_short and wic.Wic.test_debug_long to
check for tmp directory.

[YOCTO#14216]

Signed-off-by: Lee Chee Yang <Chee.Yang.Lee@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
(cherry picked from commit a122e2418b67d38f691edcf8dd846c167d6b4fa9)
Signed-off-by: Anuj Mittal <anuj.mittal@intel.com>
4 years agoinitrdscripts: init-install-efi.sh install extra files for ESP
Chee Yang Lee [Tue, 2 Feb 2021 05:43:16 +0000 (13:43 +0800)]
initrdscripts: init-install-efi.sh install extra files for ESP

Currently the install script copy only few hard coded item while
setting up target ESP, kernel artifacts, all .efi in EFI/BOOT,
grub & boot cfg and loader.conf.
While ESP can be much complex, eg: contain multiple initrd.

Add a ESP folder to carry any other files to setup onto ESP.

Signed-off-by: Chee Yang Lee <chee.yang.lee@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
(cherry picked from commit 6eaca9cf20c42501fba27dea3a6446bad948e859)
Signed-off-by: Anuj Mittal <anuj.mittal@intel.com>
4 years agobuild-appliance-image: Update to gatesgarth head revision 2020-10.2 2020-10.2-gatesgarth
Richard Purdie [Tue, 16 Feb 2021 11:44:29 +0000 (11:44 +0000)]
build-appliance-image: Update to gatesgarth head revision

Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
4 years agopython3targetconfig.bbclass: Make py3 dep and tasks only for target recipes
Khem Raj [Tue, 17 Nov 2020 22:02:57 +0000 (14:02 -0800)]
python3targetconfig.bbclass: Make py3 dep and tasks only for target recipes

python3targetconfig append target python3 to dependencies
unconditionally, and here its inherited unconditionally too but
distutils3-base is inherited in BBCLASSEXTEND'ed recipes and other not-target
recipes as well. Hence the change added via 9c8f666097802cb594a759989edcf01603a22df3
is now bridging the native dependencies with target python3 and thats
resulting all sorts of rebuilds for multimachine builds e.g.

MACHINE=qemuarm bitbake python3-scons-native
MACHINE=qemumips bitbake python3-scons-native

results in rebuilds for python3-scons-native

bitbake-diffsigs shows

Hash for dependent task python/python3-scons-native_3.1.2.bb:do_populate_sysroot changed from 1cdb93193b416477df6faa137e83a967b433c7aa29033146b405153f73f36933 to 3cea1e7cbedd121ecb768fbc291cc4e4d7d3b5c0442897
0e3b97bd058d162065
    Hash for dependent task python/python3-scons-native_3.1.2.bb:do_install changed from 8d6018fd03ffc6060a04532dc39a5b7ccca1be026a69d069cb4fb11aef86dd89 to c5f1d173596a8e910f45a2b6e0b4dab96cd0102be4d62bd3156
229cb0f5ebb11
        Hash for dependent task python/python3-scons-native_3.1.2.bb:do_compile changed from e3ee4b52a15267e6ae7853ec19a666b2fb62608a597608793336382d1c45f8a0 to 1e582043dfe6b3e00aaa532f363ce6afb37652abe837dac
7cc9769194c43eae1
            Hash for dependent task python/python3-scons-native_3.1.2.bb:do_configure changed from 770a4d5a77a96ebd9e1e7368f710bca3f88e3b1266dffa3b2d0360b1e3a81e27 to a366982778b03eee5165c3117ee778f848acdfaa2
b346650fbdf114ac70ab57b
                Hash for dependent task python/python3-scons-native_3.1.2.bb:do_prepare_recipe_sysroot changed from 958910037856ff5d5eb2b5162b3cdd02a3a710fc543b933cfeba771ee095cb72 to 474333fb565f908992fd3716
4935aaecf31a79e867826fe634cde4f44171d8e7
                    Hash for dependent task python/python3_3.9.0.bb:do_populate_sysroot changed from 7ac1c4fcbb2eacf98d2c32d991751bd2f3c7d55e2e32f2c9e485e7f5975fecf8 to 25dcfe74a95af19cce8df7c29311cc5edbbf6ad
08777e46a6fa6e417c0445018

...

Therefore limit effects of this class only for target recipes.

Signed-off-by: Khem Raj <raj.khem@gmail.com>
Cc: Alexander Kanavin <alex.kanavin@gmail.com>
Cc: Martin Jansa <Martin.Jansa@gmail.com>
Cc: Jose Quaresma <quaresma.jose@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
(cherry picked from commit 59cc148de3fd19f5041727f072f087f741c506f6)
Signed-off-by: Anuj Mittal <anuj.mittal@intel.com>
4 years agogpgme: use python3targetconfig
Alexander Kanavin [Fri, 13 Nov 2020 19:48:54 +0000 (20:48 +0100)]
gpgme: use python3targetconfig

Signed-off-by: Alexander Kanavin <alex.kanavin@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
(cherry picked from commit 375d13fcb362b48e57ba8851b03f2b72dd44da11)
Signed-off-by: Anuj Mittal <anuj.mittal@intel.com>
4 years agometa: drop _PYTHON_SYSCONFIGDATA_NAME hacks
Alexander Kanavin [Fri, 13 Nov 2020 19:48:53 +0000 (20:48 +0100)]
meta: drop _PYTHON_SYSCONFIGDATA_NAME hacks

Signed-off-by: Alexander Kanavin <alex.kanavin@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
(cherry picked from commit 7901859e38de06c56b8535a8425e76cb114c57dc)
Signed-off-by: Anuj Mittal <anuj.mittal@intel.com>
4 years agodistutils3-base.bbclass: use python3targetconfig
Alexander Kanavin [Fri, 13 Nov 2020 19:48:52 +0000 (20:48 +0100)]
distutils3-base.bbclass: use python3targetconfig

Signed-off-by: Alexander Kanavin <alex.kanavin@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
(cherry picked from commit 10cdc26748e64394e829d919a15e899812bb2fe2)
Signed-off-by: Anuj Mittal <anuj.mittal@intel.com>
4 years agopython3-pycairo: use python3targetconfig
Alexander Kanavin [Fri, 13 Nov 2020 19:48:51 +0000 (20:48 +0100)]
python3-pycairo: use python3targetconfig

Signed-off-by: Alexander Kanavin <alex.kanavin@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
(cherry picked from commit 27d1dcf065ac2ccb57229eef54dd63b45d0fc5f9)
Signed-off-by: Anuj Mittal <anuj.mittal@intel.com>
4 years agopython3: split python target configuration into own class
Alexander Kanavin [Fri, 13 Nov 2020 19:48:50 +0000 (20:48 +0100)]
python3: split python target configuration into own class

Setting _PYTHON_SYSCONFIGDATA_NAME in python3native class globally was
problematic as it was leaking into host python environment, which
was causing tracebacks depending on host distro and action
(typically anything involving importing sysconfig module).

The new class sets the variable only in specific tasks where it is needed,
and should be inherited explicitly:
- use python3native to run scripts with native python
- use python3targetconfig to run scripts with native python
if those scripts need to access target config data (such
as correct installation directories). This also adds a dependency
on target python, so should be used carefully to avoid lengthening builds.

Signed-off-by: Alexander Kanavin <alex.kanavin@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
(cherry picked from commit 823cbf815d6984e813f0ae812f6a14469150eeff)
Signed-off-by: Anuj Mittal <anuj.mittal@intel.com>
4 years agouninative: Upgrade to 2.10
Michael Halstead [Sat, 6 Feb 2021 02:22:27 +0000 (18:22 -0800)]
uninative: Upgrade to 2.10

Final glibc 2.32 based uninative.

Signed-off-by: Michael Halstead <mhalstead@linuxfoundation.org>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
(cherry picked from commit 8b5d932a42ce9e3e801837bea9cf319c455d9ae5)
Signed-off-by: Anuj Mittal <anuj.mittal@intel.com>
4 years agopseudo: Update to work with glibc 2.33
Richard Purdie [Wed, 3 Feb 2021 23:41:16 +0000 (23:41 +0000)]
pseudo: Update to work with glibc 2.33

Update to a pseudo version which contains some heqader fixes for
glibc 2.33.

Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
(cherry picked from commit c897ac317926b132547578b1f6bd347fe5677dfc)
Signed-off-by: Anuj Mittal <anuj.mittal@intel.com>
4 years agoopenssh: Backport a fix to fix with glibc 2.33 on some platforms
Richard Purdie [Thu, 4 Feb 2021 21:31:00 +0000 (21:31 +0000)]
openssh: Backport a fix to fix with glibc 2.33 on some platforms

This fixes openssh failing to work on qemux86 with glibc 2.33 due to
seccomp and the fact new syscalls are used. Also likely fixes issues
on other platforms.

Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
(cherry picked from commit 22f8ce6e6d998c0539a40b2776b1a2abb4f44bb3)
Signed-off-by: Anuj Mittal <anuj.mittal@intel.com>
4 years agosystemd: change /bin/nologin to /sbin/nologin
Chen Qi [Fri, 5 Feb 2021 03:40:28 +0000 (11:40 +0800)]
systemd: change /bin/nologin to /sbin/nologin

Our nologin path is /sbin/nologin instead of /bin/nologin.

Signed-off-by: Chen Qi <Qi.Chen@windriver.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
(cherry picked from commit cd7f55e960e759d946d8b619b0a306e610f66356)
Signed-off-by: Anuj Mittal <anuj.mittal@intel.com>
4 years agolicense_image.bbclass: Don't attempt to symlink to the same file
Mike Looijmans [Fri, 5 Feb 2021 14:53:33 +0000 (15:53 +0100)]
license_image.bbclass: Don't attempt to symlink to the same file

Sometimes (that is, in all my builds) the lic_manifest_dir and
lic_manifest_symlink_dir end up pointing to the same file, resulting
in an error like this:
  Exception: FileExistsError: [Errno 17] File exists: '/.../tmp-glibc/deploy/licenses/my-image-tdkz15' -> '/.../tmp-glibc/deploy/licenses/my-image-tdkz15'

First check to see if this is the case before attempting to create
the link.

Signed-off-by: Mike Looijmans <mike.looijmans@topic.nl>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
(cherry picked from commit 50f83fb542065eaf7a20ac07b63ae06441ada180)
Signed-off-by: Anuj Mittal <anuj.mittal@intel.com>
4 years agoimage_types.bbclass: tar: use posix format instead of gnu
Martin Jansa [Fri, 5 Feb 2021 19:22:34 +0000 (20:22 +0100)]
image_types.bbclass: tar: use posix format instead of gnu

* gnu isn't compatible with  --xattrs used e.g. here:
https://github.com/advancedtelematic/meta-updater/blob/d3a832f66e8802cb45536ff278d5c77f946d341d/classes/image_types_ostree.bbclass#L16
causing do_image_tar failing with:

| tar: --xattrs can be used only on POSIX archives
| Try 'tar --help' or 'tar --usage' for more information.

* https://www.gnu.org/software/tar/manual/html_chapter/tar_8.html
  says about posix format:

  This is the most flexible and feature-rich format.
  It does not impose any restrictions on file sizes or file name lengths.
  This format is quite recent, so not all tar implementations are able to handle it properly.
  However, this format is designed in such a way that any tar implementation able to read `ustar'
  archives will be able to read most `posix' archives as well, with the only exception that any
  additional information (such as long file names etc.) will in such case be extracted as plain
  text files along with the files it refers to.

  This archive format will be the default format for future versions of GNU tar.

  and:

  The default format for GNU tar is defined at compilation time.
  You may check it by running tar --help, and examining the last lines of its output.
  Usually, GNU tar is configured to create archives in `gnu' format, however, future version will switch to `posix'.

* I've compared tar on centos7 and ubuntu-18.04:

bash-4.2$ cat /etc/centos-release
CentOS Linux release 7.9.2009 (Core)

bash-4.2$ tar --version
tar (GNU tar) 1.26
...

bash-4.2$ tar --help | tail -n 5
*This* tar defaults to:
--format=gnu -f- -b20 --quoting-style=escape --rmt-command=/etc/rmt
--rsh-command=/usr/bin/ssh
...

bitbake@e0ee76f81c2f:/$ grep VERSION /etc/os-release
VERSION="18.04.5 LTS (Bionic Beaver)"
VERSION_ID="18.04"
VERSION_CODENAME=bionic

bitbake@e0ee76f81c2f:/$ tar --version
tar (GNU tar) 1.29
...

bitbake@e0ee76f81c2f:/$ tar --help | tail -n 5
...
*This* tar defaults to:
--format=gnu -f- -b20 --quoting-style=escape --rmt-command=/usr/lib/tar/rmt
--rsh-command=/usr/bin/rsh

Both support posix format (as pax POSIX 1003.1-2001). But centos7 version is
already too old anyway, because it doesn't support --sort=name used since:
https://git.openembedded.org/openembedded-core/commit/?id=4fa68626bbcfd9795577e1426c27d00f4d9d1c17
and
https://git.openembedded.org/openembedded-core/commit/?id=f19e43dec63a86c200e04ba14393583588550380
says that 1.28 is the minium version now and
https://git.openembedded.org/openembedded-core/commit/?id=7a66434cf11b7f051699b774e4fccd6738351368
recommends to use install-buildtools for hosts with tar < 1.28

On the other side latest tumbleweed from:
https://hub.docker.com/r/opensuse/tumbleweed
with tar-1.33 alredy defaults to posix format:

b99dbb3d86dd:/ # head -n 3 /etc/os-release
NAME="openSUSE Tumbleweed"
ID="opensuse-tumbleweed"

b99dbb3d86dd:/ # tar --version
tar (GNU tar) 1.33
...

b99dbb3d86dd:/ # tar --help | tail -n 3
*This* tar defaults to:
--format=posix -f- -b20 --quoting-style=escape --rmt-command=/usr/bin/rmt
--rsh-command=/usr/bin/ssh

I've packaged some sample rootfs directory with both tars and the result is
identical (with --format=gnu as well as --format=posix).

with ubuntu:
tar --sort=name --format=gnu --numeric-owner -cf rootfs.ubuntu.gnu.tar -C rootfs .
tar --xattrs --xattrs-include=* --sort=name --format=posix --numeric-owner -cf rootfs.ubuntu.posix.tar -C rootfs .
tumbleweed:
tar --sort=name --format=gnu --numeric-owner -cf rootfs.tumbleweed.gnu.tar -C rootfs .
tar --xattrs --xattrs-include=* --sort=name --format=posix --numeric-owner -cf rootfs.tumbleweed.posix.tar -C rootfs .
centos7 (without --sort=name):
tar --format=gnu --numeric-owner -cf rootfs.centos7.gnu.tar -C rootfs .
tar --xattrs --xattrs-include=* --format=posix --numeric-owner -cf rootfs.centos7.posix.tar -C rootfs .

size is identical:
-rw-r--r-- 1 mjansa mjansa 2487480320 Feb  5 09:19 rootfs.ubuntu.gnu.tar
-rw-r--r-- 1 mjansa mjansa 2487480320 Feb  5 10:17 rootfs.centos7.gnu.tar
-rw-r--r-- 1 mjansa mjansa 2487480320 Feb  5 10:26 rootfs.tumbleweed.gnu.tar
-rw-r--r-- 1 mjansa mjansa 2579875840 Feb  5 10:15 rootfs.ubuntu.posix.tar
-rw-r--r-- 1 mjansa mjansa 2579875840 Feb  5 10:16 rootfs.centos7.posix.tar
-rw-r--r-- 1 mjansa mjansa 2579875840 Feb  5 10:26 rootfs.tumbleweed.posix.tar

but md5s aren't:
5e3880283379dd773ac054e20562fdea  rootfs.centos7.gnu.tar
abeaf992c780aa780a27be01365d26f5  rootfs.centos7.posix.tar
0c6ee59d87ab56583293262de110bca4  rootfs.tumbleweed.gnu.tar
1555bc7276eaba924bf82a13a010fd6d  rootfs.tumbleweed.posix.tar
553d802bba351e273191bd5b2a621b66  rootfs.ubuntu.gnu.tar
b6d7b43b30174686f6625ba3c7aefdc6  rootfs.ubuntu.posix.tar

diffoscope shows some differences when using gnu format:

$ diffoscope rootfs.tumbleweed.gnu.tar rootfs.ubuntu.gnu.tar
...
-00239890: 3030 3000 3030 3737 3637 0020 4b00 0000  000.007767. K...
+00239890: 3030 3000 3031 3135 3737 0020 4b00 0000  000.011577. K...
...
-00239900: 0075 7374 6172 2020 0000 0000 0000 0000  .ustar  ........
+00239900: 0075 7374 6172 2020 0072 6f6f 7400 0000  .ustar  .root...
...
-00239920: 0000 0000 0000 0000 0000 0000 0000 0000  ................
+00239920: 0000 0000 0000 0000 0072 6f6f 7400 0000  .........root...

with posix format there are also some differences shown by diffoscope:

$ diffoscope rootfs.tumbleweed.posix.tar rootfs.ubuntu.posix.tar
 016a4c00: 2e2f 7573 722f 6269 6e2f 5061 7848 6561  ./usr/bin/PaxHea
-016a4c10: 6465 7273 2f63 6861 7474 722e 6532 6673  ders/chattr.e2fs
-016a4c20: 7072 6f67 7300 0000 0000 0000 0000 0000  progs...........
+016a4c10: 6465 7273 2e32 322f 6368 6174 7472 2e65  ders.22/chattr.e
+016a4c20: 3266 7370 726f 6773 0000 0000 0000 0000  2fsprogs........
...
 03937000: 2e2f 7573 722f 6269 6e2f 5061 7848 6561  ./usr/bin/PaxHea
-03937010: 6465 7273 2f63 6f6e 7461 696e 6572 642d  ders/containerd-
-03937020: 6374 7200 0000 0000 0000 0000 0000 0000  ctr.............
+03937010: 6465 7273 2e32 322f 636f 6e74 6169 6e65  ders.22/containe
+03937020: 7264 2d63 7472 0000 0000 0000 0000 0000  rd-ctr..........

so cannot really say which format is better for reproducible tar
archives from different distros, but posix at least supports xattrs
and it's the format for future.

Signed-off-by: Martin Jansa <Martin.Jansa@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
(cherry picked from commit 3ecea58f2a3382d9f4b410d6ad7089111334cb6f)
Signed-off-by: Anuj Mittal <anuj.mittal@intel.com>
4 years agolibcroco: Added CVE
saloni [Fri, 5 Feb 2021 15:42:34 +0000 (21:12 +0530)]
libcroco: Added CVE

Added below CVE:
CVE-2020-12825
Link: CVE-2020-12825 [https://gitlab.gnome.org/Archive/libcroco/-/commit/6eb257e5c731c691eb137fca94e916ca73941a5a]
Link: https://gitlab.gnome.org/Archive/libcroco/-/issues/8
Signed-off-by: Saloni Jain <Saloni.Jain@kpit.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
(cherry picked from commit f8cee7386c556e1c5adb07a0aee385642b7a5568)
Signed-off-by: Anuj Mittal <anuj.mittal@intel.com>
4 years agolibgcrypt: Whitelisted CVEs
saloni [Fri, 5 Feb 2021 15:37:12 +0000 (21:07 +0530)]
libgcrypt: Whitelisted CVEs

Whitelisted below CVEs:

1. CVE-2018-12433
Link: https://security-tracker.debian.org/tracker/CVE-2018-12433
Link: https://nvd.nist.gov/vuln/detail/CVE-2018-12433
CVE-2018-12433 is marked disputed and ignored by NVD as it does
not impact crypt libraries for any distros and hence, can be safely
marked whitelisted.

2. CVE-2018-12438
Link: https://security-tracker.debian.org/tracker/CVE-2018-12438
Link: https://ubuntu.com/security/CVE-2018-12438
CVE-2018-12438 was reported for affecting openjdk crypt libraries
but there are no details available on which openjdk versions are
affected and does not directly affect libgcrypt or any specific
yocto distributions, hence, can be whitelisted.

Signed-off-by: Saloni Jain <Saloni.Jain@kpit.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
(cherry picked from commit 2943efe3f56d394308f9364b439c25f6a7613288)
Signed-off-by: Anuj Mittal <anuj.mittal@intel.com>
4 years agosudo: fix CVE-2021-3156
Anuj Mittal [Sat, 6 Feb 2021 07:57:55 +0000 (15:57 +0800)]
sudo: fix CVE-2021-3156

Signed-off-by: Anuj Mittal <anuj.mittal@intel.com>
4 years agosudo: fix CVE-2021-23240
Anuj Mittal [Sat, 6 Feb 2021 07:59:15 +0000 (15:59 +0800)]
sudo: fix CVE-2021-23240

Signed-off-by: Anuj Mittal <anuj.mittal@intel.com>
4 years agoqemu.inc: Should depend on qemu-system-native, not qemu-native
Richard Purdie [Sat, 30 Jan 2021 14:18:32 +0000 (14:18 +0000)]
qemu.inc: Should depend on qemu-system-native, not qemu-native

This looks like it was from before the recipe was split, we'd expect
the system qemu mode for running the images so the dependency should be
updated.

Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
(cherry picked from commit 3a4fed4ae0e8a0d1bd62ea5fa1ef12925e1f20f5)
Signed-off-by: Anuj Mittal <anuj.mittal@intel.com>
4 years agokernel.bbclass: fix deployment for initramfs images
Awais Belal [Fri, 29 Jan 2021 12:42:39 +0000 (17:42 +0500)]
kernel.bbclass: fix deployment for initramfs images

The do_bundle_initramfs() only processes kernel image
types that are found in KERNEL_IMAGETYPE_FOR_MAKE whereas
the build system can generate other types that are not
directly supported by the kernel build system. In which
case when we come to the deploy phase not all the images
mentioned in KERNEL_IMAGETYPES would have a respective
initramfs bundled image. An example is using vmlinux.gz
in KERNEL_IMAGETYPES and enabling initramfs and then we
see

install: cannot stat 'arch/arm64/boot/vmlinux.gz.initramfs': No such file or directory

So we align the deploy phase with bundle initramfs phase
and pick up relevant initramfs bundled images using
KERNEL_IMAGETYPE_FOR_MAKE instead of KERNEL_IMAGETYPES.

Signed-off-by: Awais Belal <awais_belal@mentor.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
(cherry picked from commit 526bdd88ccd758204452579333ba188e29270bde)
Signed-off-by: Anuj Mittal <anuj.mittal@intel.com>
4 years agopackage: Ensure do_packagedata is cleaned correctly
Richard Purdie [Wed, 3 Feb 2021 14:22:13 +0000 (14:22 +0000)]
package: Ensure do_packagedata is cleaned correctly

In an earlier commit, libprocps was split into a separate package leaving
no shlibs in the main package. A bug was seen where igt-gpu-tools wouldn't
build correctly in some cases as it thought the librbary was still in the
main package, throwing qa errors as a result.

The issue was due to an extra file being left in the sstate output of
the do_packagedata task in the shlibs2/ folder which contained the bad
shlibs information.

The reason for this was that the temporary directory used in this
task wasn't being cleaned so files which were deleted were not handled
correctly. Add a missing cleandirs entry to fix this.

Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
(cherry picked from commit 50f17d0a655a3a2556f9fcad67259101c2814a36)
Signed-off-by: Anuj Mittal <anuj.mittal@intel.com>
4 years agowic/selftest: test_permissions also test bitbake image
Lee Chee Yang [Thu, 4 Feb 2021 00:29:44 +0000 (08:29 +0800)]
wic/selftest: test_permissions also test bitbake image

existing test case test_permissions use Wic command as standalone
tools to create wic image and check that wic image for permissions.

add extra steps to the test case to also check against image build
using bitbake do_image_wic.

Signed-off-by: Lee Chee Yang <chee.yang.lee@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
(cherry picked from commit 551ce73a90757ba43501fe5cf9ac84a7b77de549)
Signed-off-by: Anuj Mittal <anuj.mittal@intel.com>
4 years agoopenssl: set CVE_VERSION_SUFFIX
Lee Chee Yang [Wed, 27 Jan 2021 09:03:54 +0000 (17:03 +0800)]
openssl: set CVE_VERSION_SUFFIX

Signed-off-by: Lee Chee Yang <chee.yang.lee@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
(cherry picked from commit 17df664a32a74f17baaef8c31ac23adec2d6255f)
Signed-off-by: Anuj Mittal <anuj.mittal@intel.com>
4 years agosstatesig: Add descriptive error message to getpwuid/getgrgid "uid/gid not found...
Tomasz Dziendzielski [Mon, 1 Feb 2021 00:32:56 +0000 (01:32 +0100)]
sstatesig: Add descriptive error message to getpwuid/getgrgid "uid/gid not found" KeyError

If path is not owned by any user installed on target it gives
insufficient error "getpwuid(): uid not found" which may be misleading.
This exception occurs if uid/gid of path was not found in PSEUDO_PASSWD
files, which simply means the path is owned by host user and there is
host user contamination.

Add more information to the exception message to make it easier for user
to debug.

[YOCTO #14031]

Signed-off-by: Tomasz Dziendzielski <tomasz.dziendzielski@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
(cherry picked from commit 38540b59ed4ec8632e30a5fd6364b010d9da8470)
Signed-off-by: Anuj Mittal <anuj.mittal@intel.com>
4 years agosanity.bbclass: Check if PSEUDO_IGNORE_PATHS and paths under pseudo control overlap
Dorinda [Mon, 1 Feb 2021 03:22:56 +0000 (22:22 -0500)]
sanity.bbclass: Check if PSEUDO_IGNORE_PATHS and paths under pseudo control overlap

Added a sanity check for when PSEUDO_IGNORE_PATHS and paths under pseudo control overlap to avoid random failures generated.

[YOCTO #14193]

Signed-off-by: Dorinda Bassey <dorindabassey@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
(cherry picked from commit 6e4bd8cabcdedf4b52345ef5eb421f71d0f19b1d)
Signed-off-by: Anuj Mittal <anuj.mittal@intel.com>
4 years agolinux-yocto/5.4: update to v5.4.94
Bruce Ashfield [Wed, 3 Feb 2021 03:24:05 +0000 (22:24 -0500)]
linux-yocto/5.4: update to v5.4.94

Updating linux-yocto/5.4 to the latest korg -stable release that comprises
the following commits:

    0fbca6ce4174 Linux 5.4.94
    315cd8fc2ad2 fs: fix lazytime expiration handling in __writeback_single_inode()
    5f8b8fccdfbc writeback: Drop I_DIRTY_TIME_EXPIRE
    2d8848edc96b dm integrity: conditionally disable "recalculate" feature
    43546b74ce6c tools: Factor HOSTCC, HOSTLD, HOSTAR definitions
    ab85b382dcf7 SMB3.1.1: do not log warning message if server doesn't populate salt
    0edc78af73d0 arm64: mm: use single quantity to represent the PA to VA translation
    b899d5b2a42a tracing: Fix race in trace_open and buffer resize call
    c4a23c852e80 io_uring: Fix current->fs handling in io_sq_wq_submit_work()
    336bb7dc5a1c HID: wacom: Correct NULL dereference on AES pen proximity
    ecd62d2e9ab4 futex: Handle faults correctly for PI futexes
    55ea172ce3eb futex: Simplify fixup_pi_state_owner()
    a3155c362ca0 futex: Use pi_state_update_owner() in put_pi_state()
    ceb83cf9ed67 rtmutex: Remove unused argument from rt_mutex_proxy_unlock()
    015b6a4c2564 futex: Provide and use pi_state_update_owner()
    65aad57cac8d futex: Replace pointless printk in fixup_owner()
    0dae88a92596 futex: Ensure the correct return value from futex_lock_pi()
    c27a2a1ecf69 Revert "mm/slub: fix a memory leak in sysfs_slab_add()"
    4afd772371d9 gpio: mvebu: fix pwm .get_state period calculation
    131f8d8a889a Linux 5.4.93
    f7020c437e13 tcp: fix TCP_USER_TIMEOUT with zero window
    945d182a046f tcp: do not mess with cloned skbs in tcp_add_backlog()
    ccc248b6444a net: dsa: b53: fix an off by one in checking "vlan->vid"
    ff64094dc718 net: Disable NETIF_F_HW_TLS_RX when RXCSUM is disabled
    3e5b335a55e9 net: mscc: ocelot: allow offloading of bridge on top of LAG
    b47a3c32c4c2 ipv6: set multicast flag on the multicast route
    b778940f2ab9 net_sched: reject silly cell_log in qdisc_get_rtab()
    4ed347901f08 net_sched: avoid shift-out-of-bounds in tcindex_set_parms()
    bc757ba6dc75 ipv6: create multicast route with RTPROT_KERNEL
    60fb547a3d5d udp: mask TOS bits in udp_v4_early_demux()
    da3711f42c68 kasan: fix incorrect arguments passing in kasan_add_zero_shadow
    0d190f53fa2f kasan: fix unaligned address is unhandled in kasan_remove_zero_shadow
    5a3890bad3a4 skbuff: back tiny skbs with kmalloc() in __netdev_alloc_skb() too
    49aaf012c478 lightnvm: fix memory leak when submit fails
    0ff55fc4d6a1 sh_eth: Fix power down vs. is_opened flag ordering
    fd2f5130ae98 net: dsa: mv88e6xxx: also read STU state in mv88e6250_g1_vtu_getnext
    4e1d17a1f73b sh: dma: fix kconfig dependency for G2_DMA
    8a0b8e26f79f netfilter: rpfilter: mask ecn bits before fib lookup
    99328b4b4408 x86/cpu/amd: Set __max_die_per_package on AMD
    6f8ba0ada139 pinctrl: ingenic: Fix JZ4760 support
    382ffe786647 driver core: Extend device_is_dependent()
    4e749a28c909 xhci: tegra: Delay for disabling LFPS detector
    a6a5d08170c2 xhci: make sure TRB is fully written before giving it to the controller
    7f3cfc7e378d usb: bdc: Make bdc pci driver depend on BROKEN
    f764f90b0c77 usb: udc: core: Use lock when write to soft_connect
    564f3c532642 usb: gadget: aspeed: fix stop dma register setting.
    f89a193fd9d3 USB: ehci: fix an interrupt calltrace error
    9a660760299b ehci: fix EHCI host controller initialization sequence
    5eda5db39e28 serial: mvebu-uart: fix tx lost characters at power off
    a8fade59466c stm class: Fix module init return on allocation failure
    5e4bacea58ca intel_th: pci: Add Alder Lake-P support
    c5885886c72c x86/mmx: Use KFPU_387 for MMX string operations
    d1a9cd1dc53c x86/topology: Make __max_die_per_package available unconditionally
    cdb4ce96fdd2 x86/fpu: Add kernel_fpu_begin_mask() to selectively initialize state
    cd1c4882ab43 irqchip/mips-cpu: Set IPI domain parent chip
    9a2f6007a228 cifs: do not fail __smb_send_rqst if non-fatal signals are pending
    745229c90301 iio: ad5504: Fix setting power-down state
    ddd1416f4413 can: peak_usb: fix use after free bugs
    a24476b37167 can: vxcan: vxcan_xmit: fix use after free bug
    ac48ef15826e can: dev: can_restart: fix use after free bug
    391187744436 selftests: net: fib_tests: remove duplicate log test
    237375005739 platform/x86: intel-vbtn: Drop HP Stream x360 Convertible PC 11 from allow-list
    57f0f0ddf9e4 i2c: octeon: check correct size of maximum RECV_LEN packet
    485e0255c19e powerpc: Fix alignment bug within the init sections
    cfea5cddeb71 scsi: megaraid_sas: Fix MEGASAS_IOC_FIRMWARE regression
    da3324ec5497 pinctrl: aspeed: g6: Fix PWMG0 pinctrl setting
    5625c3da7167 powerpc: Use the common INIT_DATA_SECTION macro in vmlinux.lds.S
    73a229119983 drm/nouveau/kms/nv50-: fix case where notifier buffer is at offset 0
    af91a2e7fb5e drm/nouveau/mmu: fix vram heap sizing
    ee2c9e58f430 drm/nouveau/i2c/gm200: increase width of aux semaphore owner fields
    38f35023fd30 drm/nouveau/privring: ack interrupts the same way as RM
    8c3d3b385ed8 drm/nouveau/bios: fix issue shadowing expansion ROMs
    f5dc9627ac04 drm/amd/display: Fix to be able to stop crc calculation
    9f6d85e20125 drm/amdgpu/psp: fix psp gfx ctrl cmds
    5b2266d62b54 riscv: defconfig: enable gpio support for HiFive Unleashed
    7eef73685871 dts: phy: fix missing mdio device and probe failure of vsc8541-01 device
    5fa6987258a7 x86/xen: Add xen_no_vector_callback option to test PCI INTX delivery
    a09d4e7acdbf xen: Fix event channel callback via INTX/GSI
    acc402fa5bf5 arm64: make atomic helpers __always_inline
    8ab3478335ad clk: tegra30: Add hda clock default rates to clock driver
    c074680653e2 HID: Ignore battery for Elan touchscreen on ASUS UX550
    9cec63a3aacb HID: logitech-dj: add the G602 receiver
    b1b943f5b65e riscv: Fix sifive serial driver
    cd0c46821aa5 riscv: Fix kernel time_init()
    5a1d7bb7d333 scsi: sd: Suppress spurious errors when WRITE SAME is being disabled
    68f99105752d scsi: qedi: Correct max length of CHAP secret
    97853a7eae80 scsi: ufs: Correct the LUN used in eh_device_reset_handler() callback
    b477f4371045 dm integrity: select CRYPTO_SKCIPHER
    8ebe26a1e236 HID: multitouch: Enable multi-input for Synaptics pointstick/touchpad device
    6af49167440a ASoC: Intel: haswell: Add missing pm_ops
    ad1df24b37d9 drm/i915/gt: Prevent use of engine->wa_ctx after error
    6b59bd9eea08 drm/syncobj: Fix use-after-free
    559c0ffedbe0 drm/atomic: put state on error path
    42d855f06d12 dm integrity: fix a crash if "recalculate" used without "internal_hash"
    a03ce9cc4bb8 dm: avoid filesystem lookup in dm_get_dev_t()
    cd3aa1495d8a mmc: sdhci-xenon: fix 1.8v regulator stabilization
    6acdefd0bd34 mmc: core: don't initialize block size from ext_csd if not present
    d8a487e673ab btrfs: send: fix invalid clone operations when cloning from the same file and root
    4d1cf8eeda5b btrfs: don't clear ret in btrfs_start_dirty_block_groups
    e1065331b730 btrfs: fix lockdep splat in btrfs_recover_relocation
    68718453159e btrfs: don't get an EINTR during drop_snapshot for reloc
    a826af1dea4a ACPI: scan: Make acpi_bus_get_device() clear return pointer on error
    08fa4ae93e95 ALSA: hda/via: Add minimum mute flag
    1607adf1ac41 ALSA: seq: oss: Fix missing error check in snd_seq_oss_synth_make_info()
    9c301133beda platform/x86: ideapad-laptop: Disable touchpad_switch for ELAN0634
    ea8d3c71313f platform/x86: i2c-multi-instantiate: Don't create platform device for INT3515 ACPI nodes
    60066d5181be i2c: bpmp-tegra: Ignore unknown I2C_M flags
    09f983f0c7fc Linux 5.4.92
    e2d69319b713 spi: cadence: cache reference clock rate during probe
    d04c7938d0f8 mac80211: check if atf has been disabled in __ieee80211_schedule_txq
    d46996cb4b16 mac80211: do not drop tx nulldata packets on encrypted links
    56e8947bcf81 tipc: fix NULL deref in tipc_link_xmit()
    55bac51762c3 net, sctp, filter: remap copy_from_user failure error
    52e0b20c8c57 rxrpc: Fix handling of an unsupported token type in rxrpc_read()
    5c466480d7d4 net: avoid 32 x truesize under-estimation for tiny skbs
    f6499a78e581 net: sit: unregister_netdevice on newlink's error path
    a3870cf8a7a2 net: stmmac: Fixed mtu channged by cache aligned
    c213d85cae39 rxrpc: Call state should be read with READ_ONCE() under some circumstances
    6d57b582fb35 net: dcb: Accept RTM_GETDCB messages carrying set-like DCB commands
    d52f5929d997 net: dcb: Validate netlink message in DCB handler
    814e04776211 esp: avoid unneeded kmap_atomic call
    0ff06dd1b949 rndis_host: set proper input size for OID_GEN_PHYSICAL_MEDIUM request
    c897c10e4334 net: mvpp2: Remove Pause and Asym_Pause support
    18c29e175e30 mlxsw: core: Increase critical threshold for ASIC thermal zone
    7680783452ce mlxsw: core: Add validation of transceiver temperature thresholds
    ff6d4e8da7c6 net: ipv6: Validate GSO SKB before finish IPv6 processing
    b41352a93c16 net: skbuff: disambiguate argument and member for skb_list_walk_safe helper
    aa350dbe3a1e net: introduce skb_list_walk_safe for skb segment walking
    760e9fd4f7ab netxen_nic: fix MSI/MSI-x interrupts
    982e763ea3c3 udp: Prevent reuseport_select_sock from reading uninitialized socks
    bd4793843c85 bpf: Fix helper bpf_map_peek_elem_proto pointing to wrong callback
    79ce12cfa56a bpf: Don't leak memory in bpf getsockopt when optlen == 0
    4aef760c28e8 nfsd4: readdirplus shouldn't return parent of export
    9b72d5ba50f1 spi: npcm-fiu: Disable clock in probe error path
    6ef67f59263e spi: npcm-fiu: simplify the return expression of npcm_fiu_probe()
    fa6de8d82d9c scsi: lpfc: Make lpfc_defer_acc_rsp static
    e82b58aa6471 scsi: lpfc: Make function lpfc_defer_pt2pt_acc static
    5e6b88828526 elfcore: fix building with clang
    ac29c052654f xen/privcmd: allow fetching resource sizes
    dd113b79ee7e compiler.h: Raise minimum version of GCC to 5.1 for arm64
    24cea7d70516 usb: ohci: Make distrust_firmware param default to false
    d26b3110041a Linux 5.4.91
    516bd00e5ac1 netfilter: nft_compat: remove flush counter optimization
    935114863364 netfilter: nf_nat: Fix memleak in nf_nat_init
    49fc6d92b484 netfilter: conntrack: fix reading nf_conntrack_buckets
    548e4168e68d ALSA: firewire-tascam: Fix integer overflow in midi_port_work()
    68e67535e26b ALSA: fireface: Fix integer overflow in transmit_midi_msg()
    2c3d03cdbd39 dm: eliminate potential source of excessive kernel log noise
    a34294774a32 net: sunrpc: interpret the return value of kstrtou32 correctly
    8b5107a74db3 iommu/vt-d: Fix unaligned addresses for intel_flush_svm_range_dev()
    c2226680343d mm, slub: consider rest of partial list if acquire_slab() fails
    cd9e901fe2fc drm/i915/dsi: Use unconditional msleep for the panel_on_delay when there is no reset-deassert MIPI-sequence
    9269296721b5 IB/mlx5: Fix error unwinding when set_has_smi_cap fails
    40a782293545 RDMA/mlx5: Fix wrong free of blue flame register on error
    e8c8d2319bd7 bnxt_en: Improve stats context resource accounting with RDMA driver loaded.
    3bcf35a7c05f RDMA/usnic: Fix memleak in find_free_vf_and_create_qp_grp
    da834a9bdc23 RDMA/restrack: Don't treat as an error allocation ID wrapping
    986fdc7685fa ext4: fix superblock checksum failure when setting password salt
    38992092b54e NFS: nfs_igrab_and_active must first reference the superblock
    6b3ae2030db9 NFS/pNFS: Fix a leak of the layout 'plh_outstanding' counter
    aa2399f55eff pNFS: Stricter ordering of layoutget and layoutreturn
    78c2ab7f5265 pNFS: Mark layout for return if return-on-close was not sent
    7d1241ae1dce pNFS: We want return-on-close to complete when evicting the inode
    69d121ca892c NFS4: Fix use-after-free in trace_event_raw_event_nfs4_set_lock
    c70f6e0ac9f9 nvme-tcp: fix possible data corruption with bio merges
    55a102004376 ASoC: Intel: fix error code cnl_set_dsp_D0()
    2392a54de8ba ASoC: meson: axg-tdmin: fix axg skew offset
    973900cd4614 ASoC: meson: axg-tdm-interface: fix loopback
    08eb8a735c11 dump_common_audit_data(): fix racy accesses to ->d_name
    d443cefd9f73 perf intel-pt: Fix 'CPU too large' error
    221dee1d0d4e ARM: picoxcell: fix missing interrupt-parent properties
    ba74e0f222c7 drm/msm: Call msm_init_vram before binding the gpu
    0251d3eb4480 ACPI: scan: add stub acpi_create_platform_device() for !CONFIG_ACPI
    bfdd0a3b86c3 usb: typec: Fix copy paste error for NVIDIA alt-mode description
    644baa95db2b drm/amdgpu: fix a GPU hang issue when remove device
    596b3423fddc nvmet-rdma: Fix list_del corruption on queue establishment failure
    4cb77b877fcc nvme-pci: mark Samsung PM1725a as IGNORE_DEV_SUBNQN
    242793c7ef2f selftests: fix the return value for UDP GRO test
    5fc06b706432 net: ethernet: fs_enet: Add missing MODULE_LICENSE
    15a8491cdcd4 misdn: dsp: select CONFIG_BITREVERSE
    635a658de303 arch/arc: add copy_user_page() to <asm/page.h> to fix build error on ARC
    bc68af1fdcac bfq: Fix computation of shallow depth
    2abc54579d1b lib/raid6: Let $(UNROLL) rules work with macOS userland
    1d05b91ab72e hwmon: (pwm-fan) Ensure that calculation doesn't discard big period values
    1229d433960c habanalabs: Fix memleak in hl_device_reset
    93aef8e6cc08 habanalabs: register to pci shutdown callback
    79df21218d63 ethernet: ucc_geth: fix definition and size of ucc_geth_tx_global_pram
    331a6438ebfd regulator: bd718x7: Add enable times
    d5f996bea464 btrfs: fix transaction leak and crash after RO remount caused by qgroup rescan
    c8dd8af4b35f netfilter: ipset: fixes possible oops in mtype_resize
    ca2fc0dc1cec ARC: build: move symlink creation to arch/arc/Makefile to avoid race
    6265a0f2410f ARC: build: add boot_targets to PHONY
    217d8ba22bce ARC: build: add uImage.lzma to the top-level target
    b9128252b9ee ARC: build: remove non-existing bootpImage from KBUILD_IMAGE
    5349b17c3df5 dm integrity: fix flush with external metadata device
    c553300f1453 cifs: fix interrupted close commands
    d17a9571e392 smb3: remove unused flag passed into close functions
    55a4dff288af ext4: don't leak old mountpoint samples
    2003c669df4c ext4: fix bug for rename with RENAME_WHITEOUT
    425faacff213 drm/i915/backlight: fix CPU mode backlight takeover on LPT
    72eb9fc82aea btrfs: tree-checker: check if chunk item end overflows
    82a948fc67ea r8152: Add Lenovo Powered USB-C Travel Hub
    ad5f19c7e9ce dm integrity: fix the maximum number of arguments
    5caac6317daf dm snapshot: flush merged data before committing metadata
    2017b99ec205 dm raid: fix discard limits for raid1
    4335af6c62fc mm/hugetlb: fix potential missing huge page size info
    c64366620d91 ACPI: scan: Harden acpi_device_add() against device ID overflows
    bc0b70f1d28c RDMA/ocrdma: Fix use after free in ocrdma_dealloc_ucontext_pd()
    f7a97dc302be MIPS: relocatable: fix possible boot hangup with KASLR enabled
    f5c2f7970683 MIPS: boot: Fix unaligned access with CONFIG_MIPS_RAW_APPENDED_DTB
    a650107de374 mips: lib: uncached: fix non-standard usage of variable 'sp'
    bda45bbc8e03 mips: fix Section mismatch in reference
    aeb64ef1f429 tracing/kprobes: Do the notrace functions check without kprobes on ftrace
    984f57e4258c x86/hyperv: check cpu mask after interrupt has been disabled
    1a202b9b9d23 ASoC: dapm: remove widget from dirty list on free
    82d1a5f6f2e5 btrfs: prevent NULL pointer dereference in extent_io_tree_panic
    bb562e6e0358 kbuild: enforce -Werror=return-type

Signed-off-by: Bruce Ashfield <bruce.ashfield@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
(cherry picked from commit 80d066f627225e9eefba84c799e9b27bc17526fc)
Signed-off-by: Anuj Mittal <anuj.mittal@intel.com>
4 years agolinux-yocto-rt/5.4: fix 5.4-stable caused build breakage
Bruce Ashfield [Wed, 3 Feb 2021 03:24:02 +0000 (22:24 -0500)]
linux-yocto-rt/5.4: fix 5.4-stable caused build breakage

5.4-stable included a backport of:

   Author: Eric W. Biederman <ebiederm@xmission.com>
   Date:   Thu Dec 3 14:11:13 2020 -0600

     rwsem: Implement down_read_interruptible

     [ Upstream commit 31784cff7ee073b34d6eddabb95e3be2880a425c ]

     In preparation for converting exec_update_mutex to a rwsem so that
     multiple readers can execute in parallel and not deadlock, add
     down_read_interruptible.  This is needed for perf_event_open to be
     converted (with no semantic changes) from working on a mutex to
     wroking on a rwsem.

Signed-off-by: Eric W. Biederman <ebiederm@xmission.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lkml.kernel.org/r/87k0tybqfy.fsf@x220.int.ebiederm.org
Signed-off-by: Sasha Levin <sashal@kernel.org>
We implement a -rt variant to fix the build issues.

Signed-off-by: Bruce Ashfield <bruce.ashfield@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
(cherry picked from commit e610fb7cc22447441f18a9b1bffe58aadb6aaab6)
Signed-off-by: Anuj Mittal <anuj.mittal@intel.com>
4 years agolinux-yocto/5.4: update to v5.4.90
Bruce Ashfield [Wed, 3 Feb 2021 03:24:00 +0000 (22:24 -0500)]
linux-yocto/5.4: update to v5.4.90

Updating linux-yocto/5.4 to the latest korg -stable release that comprises
the following commits:

    ceed81a883dc Linux 5.4.90
    6f484096196b regmap: debugfs: Fix a reversed if statement in regmap_debugfs_init()
    bbb2fee395e9 net: drop bogus skb with CHECKSUM_PARTIAL and offset beyond end of trimmed packet
    bd0051a5cb05 block: fix use-after-free in disk_part_iter_next
    c5fe50e18fcb KVM: arm64: Don't access PMCR_EL0 when no PMU is available
    f595e44b161a net: mvpp2: disable force link UP during port init procedure
    5b8d3c3a9fcb regulator: qcom-rpmh-regulator: correct hfsmps515 definition
    3582406b9c04 wan: ds26522: select CONFIG_BITREVERSE
    480c5e9c7e4c regmap: debugfs: Fix a memory leak when calling regmap_attach_dev
    c3c774886790 net/mlx5e: Fix two double free cases
    ce74b5a0689d net/mlx5e: Fix memleak in mlx5e_create_l2_table_groups
    a2b2ae3812e5 bpftool: Fix compilation failure for net.o with older glibc
    2992e3371a3a iommu/intel: Fix memleak in intel_irq_remapping_alloc
    006319327d21 lightnvm: select CONFIG_CRC32
    46c15eeb0a8a block: rsxx: select CONFIG_CRC32
    4834a984e456 wil6210: select CONFIG_CRC32
    b28378bc91d0 qed: select CONFIG_CRC32
    cc196d4604c9 dmaengine: xilinx_dma: fix mixed_enum_type coverity warning
    d0eaf8a8eff8 dmaengine: xilinx_dma: fix incompatible param warning in _child_probe()
    e6f247a5f927 dmaengine: xilinx_dma: check dma_async_device_register return value
    c15556cb344a dmaengine: mediatek: mtk-hsdma: Fix a resource leak in the error handling path of the probe function
    55503711adff i2c: i801: Fix the i2c-mux gpiod_lookup_table not being properly terminated
    12e8bcaef61a spi: stm32: FIFO threshold level - fix align packet size
    9ff4796e6fd9 cpufreq: powernow-k8: pass policy rather than use cpufreq_cpu_get()
    4dd15f9bc881 can: kvaser_pciefd: select CONFIG_CRC32
    82adac5ad13b can: m_can: m_can_class_unregister(): remove erroneous m_can_clk_stop()
    3b68980596fb can: tcan4x5x: fix bittiming const, use common bittiming from m_can driver
    b77e0283efdc dmaengine: dw-edma: Fix use after free in dw_edma_alloc_chunk()
    f6dd8c259ab8 i2c: sprd: use a specific timeout to avoid system hang up issue
    8d0cadc2ea64 ARM: OMAP2+: omap_device: fix idling of devices during probe
    003280bd8845 HID: wacom: Fix memory leakage caused by kfifo_alloc
    6f367fb1b7ee iio: imu: st_lsm6dsx: fix edge-trigger interrupts
    87ea51c90280 vmlinux.lds.h: Add PGO and AutoFDO input sections
    099340d3e758 exfat: Month timestamp metadata accidentally incremented
    bb039d45ebc5 x86/resctrl: Don't move a task to the same resource group
    628af07fc5cd x86/resctrl: Use an IPI instead of task_work_add() to update PQR_ASSOC MSR
    96fb3d28c885 chtls: Fix chtls resources release sequence
    fac9b53cfacb chtls: Added a check to avoid NULL pointer dereference
    38768ea1127d chtls: Replace skb_dequeue with skb_peek
    dcce456b2843 chtls: Fix panic when route to peer not configured
    44bed66b2be9 chtls: Remove invalid set_tcb call
    266ee00f402b chtls: Fix hardware tid leak
    ed62af62da41 net/mlx5e: ethtool, Fix restriction of autoneg with 56G
    cf59803ce4b3 net/mlx5: Use port_num 1 instead of 0 when delete a RoCE address
    3008c639c081 net: dsa: lantiq_gswip: Exclude RMII from modes that report 1 GbE
    fc1c907da5a1 s390/qeth: fix L2 header access in qeth_l3_osa_features_check()
    e6931e3eb084 nexthop: Unlink nexthop group entry in error path
    3cecab93f271 nexthop: Fix off-by-one error in error path
    f03b81e61ef5 octeontx2-af: fix memory leak of lmac and lmac->name
    12e10b12124c net: ip: always refragment ip defragmented packets
    41bfd4111257 net: fix pmtu check in nopmtudisc mode
    98fc9692ac3d tools: selftests: add test for changing routes with PTMU exceptions
    7694654168bb net: ipv6: fib: flush exceptions when purging route
    1cba7e270b16 net/sonic: Fix some resource leaks in error handling paths
    37e6368a8de6 net: vlan: avoid leaks on register_vlan_dev() failures
    4ff0737ebc76 net: stmmac: dwmac-sun8i: Balance internal PHY power
    5698f0921c9b net: stmmac: dwmac-sun8i: Balance internal PHY resource references
    fa020a28896c net: hns3: fix a phy loopback fail issue
    bddaf51d116c net: hns3: fix the number of queues actually used by ARQ
    d73f7e757526 net: cdc_ncm: correct overhead in delayed_ndp_size
    5597557244d4 vfio iommu: Add dma available capability
    335104082c21 x86/asm/32: Add ENDs to some functions and relabel with SYM_CODE_*
    a829146c3fdc Linux 5.4.89
    485e21729b1e scsi: target: Fix XCOPY NAA identifier lookup
    7795afa0d7a9 KVM: x86: fix shift out of bounds reported by UBSAN
    a9d49da7edf8 x86/mtrr: Correct the range check before performing MTRR type lookups
    a798b367a066 netfilter: nft_dynset: report EOPNOTSUPP on missing set feature
    5e401ea71676 netfilter: xt_RATEEST: reject non-null terminated string from userspace
    1dd6a790c220 netfilter: ipset: fix shift-out-of-bounds in htable_bits()
    e0281bb5a82d netfilter: x_tables: Update remaining dereference to RCU
    828f2a20f946 drm/i915: clear the gpu reloc batch
    ef8133b1b47e dmabuf: fix use-after-free of dmabuf's file->f_inode
    284be2b993ca Revert "device property: Keep secondary firmware node secondary by type"
    64d06c7f2fa2 btrfs: send: fix wrong file path when there is an inode with a pending rmdir
    0cb0b876f17f ALSA: hda/realtek: Add two "Intel Reference board" SSID in the ALC256.
    02e59692a6b1 ALSA: hda/realtek: Enable mute and micmute LED on HP EliteBook 850 G7
    d63a96f45c4f ALSA: hda/realtek - Fix speaker volume control on Lenovo C940
    30fd9778cf8f ALSA: hda/conexant: add a new hda codec CX11970
    121944484cc4 ALSA: hda/via: Fix runtime PM for Clevo W35xSS
    a5c7a456680f kvm: check tlbs_dirty directly
    10dcb79ec79e x86/mm: Fix leak of pmd ptlock
    d3e5db486fd8 USB: serial: keyspan_pda: remove unused variable
    bcffe2de9dde usb: gadget: configfs: Fix use-after-free issue with udc_name
    276828221852 usb: gadget: configfs: Preserve function ordering after bind failure
    b2bd36f54495 usb: gadget: Fix spinlock lockup on usb_function_deactivate
    ce507b55db29 USB: gadget: legacy: fix return error code in acm_ms_bind()
    7f875ea9883c usb: gadget: u_ether: Fix MTU size mismatch with RX packet size
    b89a5f39c2b5 usb: gadget: function: printer: Fix a memory leak for interface descriptor
    692ab0726460 usb: gadget: f_uac2: reset wMaxPacketSize
    7ac84fa85ba2 usb: gadget: select CONFIG_CRC32
    77a804dd6b46 ALSA: usb-audio: Fix UBSAN warnings for MIDI jacks
    5c263f16822f USB: usblp: fix DMA to stack
    41f15da2abd9 USB: yurex: fix control-URB timeout handling
    175f7a5fa7e6 USB: serial: option: add Quectel EM160R-GL
    1a59feb52dc4 USB: serial: option: add LongSung M5710 module support
    ac48b1dacb07 USB: serial: iuu_phoenix: fix DMA from stack
    8a051eaae708 usb: uas: Add PNY USB Portable SSD to unusual_uas
    a7b81d0d2e07 usb: usbip: vhci_hcd: protect shift size
    f7cc27eb358d USB: xhci: fix U1/U2 handling for hardware with XHCI_INTEL_HOST quirk set
    ea472d839133 usb: chipidea: ci_hdrc_imx: add missing put_device() call in usbmisc_get_init_data()
    a37a0667e1e0 usb: dwc3: ulpi: Use VStsDone to detect PHY regs access completion
    5b8e1be9e0c1 USB: cdc-wdm: Fix use after free in service_outstanding_interrupt().
    5445502a344b USB: cdc-acm: blacklist another IR Droid device
    eeae1d95ce4e usb: gadget: enable super speed plus
    70cf59b8ffb4 staging: mt7621-dma: Fix a resource leak in an error handling path
    c511f27e130e powerpc: Handle .text.{hot,unlikely}.* in linker script
    867c10a03f84 crypto: asym_tpm: correct zero out potential secrets
    ff7397add935 crypto: ecdh - avoid buffer overflow in ecdh_set_secret()
    9e60056b1f53 video: hyperv_fb: Fix the mmap() regression for v5.4.y and older
    84d488719b27 Bluetooth: revert: hci_h5: close serdev device and free hu in h5_close
    3417067b3111 kbuild: don't hardcode depmod path
    3f2a28930a7e net/sched: sch_taprio: ensure to reset/destroy all child qdiscs
    c41ea30c3839 ionic: account for vlan tag len in rx buffer len
    5c6eb887e192 vhost_net: fix ubuf refcount incorrectly when sendmsg fails
    8f64957fda12 net: usb: qmi_wwan: add Quectel EM160R-GL
    12ab7b627d43 CDC-NCM: remove "connected" log message
    171a2bce9d6c net: dsa: lantiq_gswip: Fix GSWIP_MII_CFG(p) register access
    c0883010d3b3 net: dsa: lantiq_gswip: Enable GSWIP_MII_CFG_EN also for internal PHYs
    07f26fc52b45 r8169: work around power-saving bug on some chip versions
    106ca9ca9acc net: hdlc_ppp: Fix issues when mod_timer is called while timer is running
    2b8aa896b151 erspan: fix version 1 check in gre_parse_header()
    606f5412ad86 net: hns: fix return value check in __lb_other_process()
    e40b5fc79110 net: sched: prevent invalid Scell_log shift count
    b16f883e71f3 ipv4: Ignore ECN bits for fib lookups in fib_compute_spec_dst()
    a018c071de14 net: mvpp2: fix pkt coalescing int-threshold configuration
    443a71031e49 tun: fix return value when the number of iovs exceeds MAX_SKB_FRAGS
    c076e1198554 net: ethernet: ti: cpts: fix ethtool output when no ptp_clock registered
    8602c20a9160 net-sysfs: take the rtnl lock when accessing xps_rxqs_map and num_tc
    1f6b04a2b282 net-sysfs: take the rtnl lock when storing xps_rxqs
    67ed54a63f43 net-sysfs: take the rtnl lock when accessing xps_cpus_map and num_tc
    fb14db9508c0 net-sysfs: take the rtnl lock when storing xps_cpus
    e43ec45d45af net: ethernet: Fix memleak in ethoc_probe
    56dc7908ed85 net/ncsi: Use real net-device for response handler
    dffef999e484 virtio_net: Fix recursive call to cpus_read_lock()
    5404192a8721 qede: fix offload for IPIP tunnel packets
    8009f6bb13a3 net: ethernet: mvneta: Fix error handling in mvneta_probe
    6d003fe7fe87 ibmvnic: continue fatal error reset after passive init
    3d16088a9668 net: mvpp2: Fix GoP port 3 Networking Complex Control configurations
    8548c9679939 atm: idt77252: call pci_disable_device() on error path
    2a006b4fa5cc ethernet: ucc_geth: set dev->max_mtu to 1518
    c2ca14cc6f55 ethernet: ucc_geth: fix use-after-free in ucc_geth_remove()
    af99cae96fdc net: systemport: set dev->max_mtu to UMAC_MAX_MTU_SIZE
    8dd98d5d2ba4 net: mvpp2: prs: fix PPPoE with ipv6 packet parse
    73445f29575a net: mvpp2: Add TCAM entry to drop flow control pause frames
    a5a6dc4dc293 iavf: fix double-release of rtnl_lock
    6aba31a7c72e i40e: Fix Error I40E_AQ_RC_EINVAL when removing VFs
    9ea03f6890ce proc: fix lookup in /proc/net subdirectories after setns(2)
    d2942e958f26 proc: change ->nlink under proc_subdir_lock
    59b10c8a59a1 depmod: handle the case of /sbin/depmod without /sbin in PATH
    663a0bcb3fa5 lib/genalloc: fix the overflow when size is too big
    19e0cf8fc481 scsi: scsi_transport_spi: Set RQF_PM for domain validation commands
    eb3e975ac2a3 scsi: ide: Do not set the RQF_PREEMPT flag for sense requests
    4ae3573c571e scsi: ufs-pci: Ensure UFS device is in PowerDown mode for suspend-to-disk ->poweroff()
    5f9c3d640505 scsi: ufs: Fix wrong print message in dev_err()
    515dc635eb76 workqueue: Kick a worker based on the actual activation of delayed works
    f3a4c8d50145 Linux 5.4.88
    0a49aaf4df29 mwifiex: Fix possible buffer overflows in mwifiex_cmd_802_11_ad_hoc_start
    117433236ae2 exec: Transform exec_update_mutex into a rw_semaphore
    d390fc97df62 rwsem: Implement down_read_interruptible
    1b75a263fbd9 rwsem: Implement down_read_killable_nested
    71b8355ba667 perf: Break deadlock involving exec_update_mutex
    732251cabeb3 fuse: fix bad inode
    06c672dd61b5 iio:imu:bmi160: Fix alignment and data leak issues
    7a736f41013e kdev_t: always inline major/minor helper functions
    61a0d8e437bb dmaengine: at_hdmac: add missing kfree() call in at_dma_xlate()
    20d5ee563bfd dmaengine: at_hdmac: add missing put_device() call in at_dma_xlate()
    f2a0b7677444 dmaengine: at_hdmac: Substitute kzalloc with kmalloc
    4d3ba541bede Revert "mtd: spinand: Fix OOB read"
    da5b4cf021b9 Revert "drm/amd/display: Fix memory leaks in S3 resume"

Signed-off-by: Bruce Ashfield <bruce.ashfield@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
(cherry picked from commit 5dcc8a5977725a9fe11ac13ebd16a7acc1eef37d)
Signed-off-by: Anuj Mittal <anuj.mittal@intel.com>
4 years agostaging: Clean up files installed into the sysroot
Richard Purdie [Wed, 27 Jan 2021 17:15:17 +0000 (17:15 +0000)]
staging: Clean up files installed into the sysroot

There are a variety of files being installed into $datadir which we
don't need. Pick the top "offenders" which amount of thousands of files
and simply don't install them. These include things like test data,
terminfo data, locale data for native tools and so on. This saves
copying these files into native and target sysroots and should improve
performance (smaller sstate, fewer files to copy around).

With this and the python recipe change, alsa-tools went from:

recipe-sysroot: 18357
recipe-sysroot-native: 14129

to

recipe-sysroot: 10809
recipe-sysroot-native: 8079

which is a decent improvement.

Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
(cherry picked from commit 366c72941fe1c24d0b1d96df46e13cb9eb4e79d6)
Signed-off-by: Anuj Mittal <anuj.mittal@intel.com>
4 years agopython3: Avoid installing test data into recipe-sysroot
Richard Purdie [Wed, 27 Jan 2021 17:12:46 +0000 (17:12 +0000)]
python3: Avoid installing test data into recipe-sysroot

There are several thousand files in the test directory which we don't need.
Adding these for the native and target sysroots is a crazy amount of files
to be throwing around needlessly. Delete the files from the sysroot side
of things to tidy up the sysroots and improve performance.

Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
(cherry picked from commit f6bced03011ad1663d68b0322a2f8aeb4d836646)
Signed-off-by: Anuj Mittal <anuj.mittal@intel.com>
4 years agoncurses: Don't put terminfo into the sysroot
Richard Purdie [Wed, 27 Jan 2021 12:13:28 +0000 (12:13 +0000)]
ncurses: Don't put terminfo into the sysroot

This recudes the file count from ~2850 to ~100 which is a huge win
for reducing build directory clutter, its unlikely anything uses the
terminfo data or man pages in the sysroot. This is especially helpful
as we usually end up with two copies of these sets of files.

Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
(cherry picked from commit 443633dfc20177ef88a388d96745675817510c99)
Signed-off-by: Anuj Mittal <anuj.mittal@intel.com>
4 years agoglibc: update to latest release/2.32/master branch
Steve Sakoman [Thu, 4 Feb 2021 20:34:51 +0000 (10:34 -1000)]
glibc: update to latest release/2.32/master branch

Remove patches for CVE-2019-25013 and CVE-2020-27618 since they are
present in the branch now. Add both CVEs to CVE_CHECK_WHITELIST.

760e1d28782 gconv: Fix assertion failure in ISO-2022-JP-3 module (bug 27256)
d3cb8f6222a aarch64: fix static PIE start code for BTI [BZ #27068]
082798622d8 __vfscanf_internal: fix aliasing violation (bug 26690)
33dc30bc838 aarch64: Use mmap to add PROT_BTI instead of mprotect [BZ #26831]
46e1e64fe3e elf: Pass the fd to note processing
b6eae83717d elf: Move note processing after l_phdr is updated
c6090dcebd1 aarch64: align address for BTI protection [BZ #26988]
610e2c51504 aarch64: Fix missing BTI protection from dependencies [BZ #26926]
4c619b3eed5 x86: Check IFUNC definition in unrelocated executable [BZ #20019]
87450ecf8a8 x86: Set header.feature_1 in TCB for always-on CET [BZ #27177]
2b4f67c2b33 Update for [BZ #27130] fix
1a24bbd43e4 x86-64: Avoid rep movsb with short distance [BZ #27130]
0d9793e82a1 Fix buffer overrun in EUC-KR conversion module (bz #24973)
1d49bede4d8 tests-mcheck: New variable to run tests with MALLOC_CHECK_=3
050022910be iconv: Accept redundant shift sequences in IBM1364 [BZ #26224]
ac0a6929c5d sh: Add sh4 fpu Implies folder
3ea24955bff struct _Unwind_Exception alignment should not depend on compiler flags
5c36293f067 resolv: Serialize processing in resolv/tst-resolv-txnid-collision
2dfa659a66f resolv: Handle transaction ID collisions in parallel queries (bug 26600)
05c025abca1 support: Provide a way to clear the RA bit in DNS server responses
f688bcd83de support: Provide a way to reorder responses within the DNS test server
eba0ce60588 Remove __warndecl
5337b2af4b8 Remove __warn_memset_zero_len [BZ #25399]
c6e794640c3 aarch64: Add unwind information to _start (bug 26853)
70ee5e8b573 aarch64: Fix DT_AARCH64_VARIANT_PCS handling [BZ #26798]
8813b2682e4 x86: Optimizing memcpy for AMD Zen architecture.
e61a8fd8fad Reversing calculation of __x86_shared_non_temporal_threshold
0b9460d22e2 sysvipc: Fix IPC_INFO and SHM_INFO handling [BZ #26636]
c4aeedea598 sysvipc: Fix IPC_INFO and MSG_INFO handling [BZ #26639]
9b139b6b81a sysvipc: Fix SEM_STAT_ANY kernel argument pass [BZ #26637]
81c5484d93a AArch64: Use __memcpy_simd on Neoverse N2/V1
0f8f0ed25c1 AArch64: Improve backwards memmove performance
23482f78866 Set version.h RELEASE to "stable" (Bug 26700)
69beb5cbf85 string: Fix strerrorname_np return value [BZ #26555]
fe62c4d173f intl: Handle translation output codesets with suffixes [BZ #26383]
386543bc449 NEWS: Update for [BZ #26534] fix
cebc01cbfd6 x86-64: Fix FMA4 detection in ifunc [BZ #26534]

Signed-off-by: Steve Sakoman <steve@sakoman.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
(cherry picked from commit 8d05c277c5350c4d968eb488788eac7978968ef7)
Signed-off-by: Anuj Mittal <anuj.mittal@intel.com>
4 years agonpm.bbclass: use python3 for npm config
Vyacheslav Yurkov [Thu, 28 Jan 2021 18:42:46 +0000 (19:42 +0100)]
npm.bbclass: use python3 for npm config

python2-native executable is not available in sysroot anymore, which
causes compilation of some nodejs modules to fail. Switch to python3 as a
default python version.

Signed-off-by: Vyacheslav Yurkov <uvv.mail@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
(cherry picked from commit d21f50ecf8e8683a92b7d234fa8225c2c1470595)
Signed-off-by: Anuj Mittal <anuj.mittal@intel.com>
4 years agorecipetool: create: only add npmsw url if required
Kamel Bouhara [Thu, 14 Jan 2021 07:12:35 +0000 (08:12 +0100)]
recipetool: create: only add npmsw url if required

Before adding a npmsw fetcher to a recipe we
should first check if the generated shrinkwrap file
contains dependencies.

Signed-off-by: Kamel Bouhara <kamel.bouhara@bootlin.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
(cherry picked from commit ef153ad36d0299e83a03af8f207686d0d8a238b3)
Signed-off-by: Anuj Mittal <anuj.mittal@intel.com>
4 years agonpm.bbclass: make shrinkwrap file optional
Kamel Bouhara [Thu, 14 Jan 2021 07:12:34 +0000 (08:12 +0100)]
npm.bbclass: make shrinkwrap file optional

Some packages don't have shrinkwrap file which
means no npmsw uri is provided in the recipe.

Signed-off-by: Kamel Bouhara <kamel.bouhara@bootlin.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
(cherry picked from commit 47760b0d7d66b2b68ee197d359f0b7b17374d742)
Signed-off-by: Anuj Mittal <anuj.mittal@intel.com>
4 years agoimage_types: Ensure tar archives are reproducible
Richard Purdie [Sat, 30 Jan 2021 17:57:13 +0000 (17:57 +0000)]
image_types: Ensure tar archives are reproducible

The tar output seems to vary depending on the version of tar used and distro
configuration. Be explict about the output format to avoid this and be
determinstic.

(From OE-Core rev: c56f3c9febc1732aa1302524c6c4da36f16bd1f7)

Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
(cherry picked from commit 9dbe0f69f874d3687ae1accc19116570bad86c04)
Signed-off-by: Anuj Mittal <anuj.mittal@intel.com>