Skip to contentSkip to conversation

Patch review

PATCH New Steve Sakoman 1 participant 31 messages Jan 5, 2021 Arm/Amlogic Meson SoC support

lore
Share & tools
lore
#0 lore raw

cover Patch review

Please review this next set of patches for dunfell and have comments back by end of day Thursday.

Passed a-full on autobuilder:

https://autobuilder.yoctoproject.org/typhoon/#/builders/83/builds/1722

The following changes since commit fcd335e2a721d8db6e2fdac483798ced0ff3e1ad:

openssl: Update to 1.1.1i (2021-01-01 11:55:26 +0000)

are available in the Git repository at:

git://git.openembedded.org/openembedded-core-contrib stable/dunfell-nut
http://cgit.openembedded.org/openembedded-core-contrib/log/?h=stable/dunfell-nut

Alexander Kanavin (1):
valgrind: exclude bar_bad/bar_bad_xml from ptests

Anuj Mittal (1):
mesa: add more details to elf-tls patch

Changqing Li (1):
libexif: fix CVE-2020-0198; CVE-2020-0452

Christopher Larson (2):
grub-efi-cfg: exclude OVERRIDES from build_efi_cfg vardeps
uboot-extlinux-config: exclude OVERRIDES from
do_create_extlinux_config vardeps

Diego Sueiro (1):
modutils-initscripts: Use depmod -a when modules.dep is empty

Dmitry Baryshkov (1):
perl: fix installation failure because of shell issue

Khem Raj (1):
initscripts: use quotes for shell variable comparision

Lee Chee Yang (2):
wic/direct/kparser: ensure fsuuid for vfat and msdos align with format
gdk-pixbuf: fix CVE-2020-29385

Maxime Roussin-Bélanger (1):
meta: add missing descriptions in some support recipes

Nathan Rossi (2):
ncurses: Prevent LDFLAGS being emitted in .pc files
coreutils: enable xattrs by default for nativesdk

Ovidiu Panait (1):
timezone: upgrade to 2020e

Peter Kjellerstedt (1):
apr-util: Only specify --with-dbm=gdbm if gdbm support is enabled

Richard Purdie (3):
oeqa/commands: Ensure sync can be found regardless of PATH
grub: Further reproducibility fix
man-db: Fix reproducibility issue

Robert Yang (5):
buildtools-tarball.bb: Fix PATH for environment setup script
ncurses: Make ncurses-tools depend on ncurses-terminfo-base
minicom: RDEPENDS on ncurses-terminfo-base archiver.bbclass: Fix --runall=deploy_archives for images
weston: Fix PACKAGECONFIG for remoting

Ross Burton (3):
ncurses: remove config.cache
wic-image-minimal: only depend on syslinux on x86 targets lib/oe/qa: handle the 'no specific instruction set' ELF e_machine
value

Scott Murray (2):
grub: fix "CVE:" line in one of the patches
patch: fix CVE-2019-20633

Tanu Kaskinen (1):
pulseaudio: Remove OE_LT_RPATH_ALLOW

Tomasz Dziendzielski (1):
populate_sdk_base: Fix condition syntax if SDK_RELOCATE_AFTER_INSTALL
is disabled

 .../recipes-test/images/wic-image-minimal.bb  |  5 +-
 meta/classes/archiver.bbclass                 |  1 +
 meta/classes/grub-efi-cfg.bbclass             |  1 +
 meta/classes/populate_sdk_base.bbclass        |  2 +-
 meta/classes/uboot-extlinux-config.bbclass    |  1 +
 meta/lib/oe/qa.py                             |  1 +
 meta/lib/oeqa/selftest/cases/oelib/elf.py     |  2 +-
 meta/lib/oeqa/utils/commands.py               |  3 +
 ...low-checking-primitives-where-we-do-.patch |  2 +-
 meta/recipes-bsp/grub/files/determinism.patch | 16 +++++
 meta/recipes-core/coreutils/coreutils_8.31.bb |  3 +
 .../initscripts/initscripts-1.0/checkroot.sh  |  2 +-
 meta/recipes-core/meta/buildtools-tarball.bb  |  2 +-
 ...-Do-not-include-LDFLAGS-in-generated.patch | 29 ++++++++
 meta/recipes-core/ncurses/files/config.cache  |  4 --
 meta/recipes-core/ncurses/ncurses.inc         |  5 +-
 meta/recipes-core/ncurses/ncurses_6.2.bb      |  4 +-
 .../patch/patch/CVE-2019-20633.patch          | 31 +++++++++
 meta/recipes-devtools/patch/patch_2.7.6.bb    |  1 +
 meta/recipes-devtools/perl/perl_5.30.1.bb     |  5 +-
 .../valgrind/valgrind/remove-for-aarch64      |  2 -
 .../valgrind/valgrind/remove-for-all          |  2 +
 .../valgrind/valgrind/run-ptest               | 10 +++
 .../valgrind/valgrind_3.15.0.bb               |  2 +
 meta/recipes-extended/man-db/man-db_2.9.0.bb  |  1 +
 .../recipes-extended/minicom/minicom_2.7.1.bb |  2 +
 meta/recipes-extended/timezone/timezone.inc   |  6 +-
 .../gdk-pixbuf/CVE-2020-29385.patch           | 55 ++++++++++++++++
 .../gdk-pixbuf/gdk-pixbuf_2.40.0.bb           |  1 +
 ...02-meson.build-make-TLS-ELF-optional.patch | 15 +++++
 meta/recipes-graphics/wayland/weston_8.0.0.bb |  2 +-
 .../modutils-initscripts/files/modutils.sh    |  2 +-
 .../pulseaudio/pulseaudio.inc                 |  5 --
 meta/recipes-support/apr/apr-util_1.6.1.bb    |  5 +-
 meta/recipes-support/apr/apr_1.7.0.bb         |  4 ++
 meta/recipes-support/aspell/aspell_0.60.8.bb  |  4 ++
 .../atk/at-spi2-core_2.34.0.bb                |  2 +
 meta/recipes-support/attr/acl_2.2.53.bb       |  2 +
 meta/recipes-support/attr/attr.inc            |  4 ++
 .../bash-completion/bash-completion_2.10.bb   |  2 +
 .../libexif/files/CVE-2020-0198.patch         | 66 +++++++++++++++++++
 .../libexif/files/CVE-2020-0452.patch         | 39 +++++++++++
 .../recipes-support/libexif/libexif_0.6.22.bb |  2 +
 meta/recipes-support/nettle/nettle_3.5.1.bb   |  3 +
 scripts/lib/wic/ksparser.py                   | 17 +++++
 scripts/lib/wic/plugins/imager/direct.py      |  7 ++
 46 files changed, 351 insertions(+), 31 deletions(-)
 create mode 100644 meta/recipes-core/ncurses/files/0003-gen-pkgconfig.in-Do-not-include-LDFLAGS-in-generated.patch
 delete mode 100644 meta/recipes-core/ncurses/files/config.cache
 create mode 100644 meta/recipes-devtools/patch/patch/CVE-2019-20633.patch
 create mode 100644 meta/recipes-devtools/valgrind/valgrind/remove-for-all
 create mode 100644 meta/recipes-gnome/gdk-pixbuf/gdk-pixbuf/CVE-2020-29385.patch
 create mode 100644 meta/recipes-support/libexif/files/CVE-2020-0198.patch
 create mode 100644 meta/recipes-support/libexif/files/CVE-2020-0452.patch
Signature · 1 line
2.17.1
#1 lore raw

PATCH 1/30 modutils-initscripts: Use depmod -a when modules.dep is empty

From: Diego Sueiro <[email protected]>

To correctly create the modules.dep when it is empty, we need to call depmod with '-a' instead of '-Ae' since the quick option uses the mtime to decide whether or not it needs to update. This way we can guarantee that it will be recreated completely and not only when adding new modules.

(cherry picked from commit 9292e9050fd95be4b556909e2775a45bce1ca79a)
---
 meta/recipes-kernel/modutils-initscripts/files/modutils.sh | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
meta/recipes-kernel/modutils-initscripts/files/modutils.sh +1 -1
index a78adf5729..df37bfe7a1 100755
@@ -18,7 +18,7 @@ LOAD_MODULE=modprobe
 
 if [ ! -f /lib/modules/`uname -r`/modules.dep ]; then
 	[ "$VERBOSE" != no ] && echo "Calculating module dependencies ..."
-	depmod -Ae
+	depmod -a
 fi
 
 loaded_modules=" "
Signature · 1 line
2.17.1
#2 lore raw

PATCH 2/30 buildtools-tarball.bb: Fix PATH for environment setup script

In reply to Steve Sakoman

From: Robert Yang <[email protected]>

It only added ${SDKPATHNATIVE}/usr/bin to PATH which didn't work when files were installed to other bin dirs such as /bin or /sbin, for example, nativesdk-pigz installs the files to /bin, now fix it to keep align with sdk's PATH.

(cherry picked from commit 67cac575a5696af5bad1aab888b65ea2686adff7)
---
 meta/recipes-core/meta/buildtools-tarball.bb | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
meta/recipes-core/meta/buildtools-tarball.bb +1 -1
index 434ffdc334..faf7108a86 100644
@@ -66,7 +66,7 @@ create_sdk_files_append () {
 	# Generate new (mini) sdk-environment-setup file
 	script=${1:-${SDK_OUTPUT}/${SDKPATH}/environment-setup-${SDK_SYS}}
 	touch $script
-	echo 'export PATH=${SDKPATHNATIVE}${bindir_nativesdk}:$PATH' >> $script
+	echo 'export PATH=${SDKPATHNATIVE}${bindir_nativesdk}:${SDKPATHNATIVE}${sbindir_nativesdk}:${SDKPATHNATIVE}${base_bindir_nativesdk}:${SDKPATHNATIVE}${base_sbindir_nativesdk}:$PATH' >> $script
 	echo 'export OECORE_NATIVE_SYSROOT="${SDKPATHNATIVE}"' >> $script
 	echo 'export GIT_SSL_CAINFO="${SDKPATHNATIVE}${sysconfdir}/ssl/certs/ca-certificates.crt"' >>$script
 	echo 'export SSL_CERT_FILE="${SDKPATHNATIVE}${sysconfdir}/ssl/certs/ca-certificates.crt"' >>$script
Signature · 1 line
2.17.1
#3 lore raw

PATCH 3/30 ncurses: Make ncurses-tools depend on ncurses-terminfo-base

In reply to Steve Sakoman

From: Robert Yang <[email protected]>

Fixed when ncurses-terminfo-base is not installed:

$ infocmp
infocmp: couldn't open terminfo file linux.

The required file is in ncurses-terminfo-base

(cherry picked from commit f3e0a6bf1b7d9009c253e3f97df8736ecf3aa79b)
---
 meta/recipes-core/ncurses/ncurses.inc | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
meta/recipes-core/ncurses/ncurses.inc +1 -1
index 4156bf4f7d..8c68f61d99 100644
@@ -306,7 +306,7 @@ FILES_${PN}-tools = "\
 "
 
 # 'reset' is a symlink to 'tset' which is in the 'ncurses' package
-RDEPENDS_${PN}-tools = "${PN}"
+RDEPENDS_${PN}-tools = "${PN} ${PN}-terminfo-base"
 
 FILES_${PN}-terminfo = "\
   ${datadir}/terminfo \
Signature · 1 line
2.17.1
#4 lore raw

PATCH 4/30 minicom: RDEPENDS on ncurses-terminfo-base

In reply to Steve Sakoman

From: Robert Yang <[email protected]>

Fixed when ncurses-terminfo-base is not installed:

$ minicom
No termcap database present!
(cherry picked from commit 15198ebe8a0fc58c2d1122b956fca092c66a0d41)
---
 meta/recipes-extended/minicom/minicom_2.7.1.bb | 2 ++
 1 file changed, 2 insertions(+)
meta/recipes-extended/minicom/minicom_2.7.1.bb +2 -0
index 6c539c553b..79d06d20f9 100644
@@ -26,3 +26,5 @@ do_install() {
 }
 
 RRECOMMENDS_${PN} += "lrzsz"
+
+RDEPENDS_${PN} += "ncurses-terminfo-base"
Signature · 1 line
2.17.1
#5 lore raw

PATCH 5/30 ncurses: remove config.cache

In reply to Steve Sakoman

From: Ross Burton <[email protected]>

Instead of having a config.cache that overrides the system site files, simply set the values in CACHED_CONFIGUREVARS. We can also drop the mkstemp check as the configure.ac assumes it works, leaving just nanosleep.

(cherry picked from commit c30c90e3adfa91407c37838c971e251f8482e2b8)
---
 meta/recipes-core/ncurses/files/config.cache | 4 ----
 meta/recipes-core/ncurses/ncurses.inc        | 3 ++-
 meta/recipes-core/ncurses/ncurses_6.2.bb     | 3 +--
 3 files changed, 3 insertions(+), 7 deletions(-)
 delete mode 100644 meta/recipes-core/ncurses/files/config.cache
meta/recipes-core/ncurses/files/config.cache deleted +0 -4
deleted file mode 100644
index 6a9217d5bb..0000000000
@@ -1,4 +0,0 @@
-#! /bin/sh
-
-cf_cv_func_nanosleep=yes
-cf_cv_func_mkstemp=yes
meta/recipes-core/ncurses/ncurses.inc +2 -1
index 8c68f61d99..630c5f26d6 100644
@@ -16,7 +16,8 @@ inherit autotools binconfig-disabled multilib_header pkgconfig
 SRC_URI = "git://salsa.debian.org/debian/ncurses.git;protocol=https"
 
 EXTRA_AUTORECONF = "-I m4"
-CONFIG_SITE =+ "${WORKDIR}/config.cache"
+
+CACHED_CONFIGUREVARS = "cf_cv_func_nanosleep=yes"
 
 EXTRASITECONFIG = "CFLAGS='${CFLAGS} -I${SYSROOT_DESTDIR}${includedir}'"
 
meta/recipes-core/ncurses/ncurses_6.2.bb +1 -2
index 723e685a9b..5c02db8541 100644
@@ -2,12 +2,11 @@ require ncurses.inc
 
 SRC_URI += "file://0001-tic-hang.patch \
            file://0002-configure-reproducible.patch \
-           file://config.cache \
            "
 # commit id corresponds to the revision in package version
 SRCREV = "a669013cd5e9d6434e5301348ea51baf306c93c4"
 S = "${WORKDIR}/git"
-EXTRA_OECONF += "--with-abi-version=5 --cache-file=${B}/config.cache"
+EXTRA_OECONF += "--with-abi-version=5"
 UPSTREAM_CHECK_GITTAGREGEX = "(?P<pver>\d+(\.\d+)+(\+\d+)*)"
 
 # This is needed when using patchlevel versions like 6.1+20181013
Signature · 1 line
2.17.1
#6 lore raw

PATCH 6/30 ncurses: Prevent LDFLAGS being emitted in .pc files

In reply to Steve Sakoman

From: Nathan Rossi <[email protected]>

By default ncurses includes the values of LDFLAGS in its output pkgconfig .pc files. This causes issues because OE includes options that are specific to either the build host, or build configuration. These options are not expected to be embedded in the pkgconfig output that is installed.

Specifically this change resolves issues with uninative, where uninative includes '-Wl,--dynamic-linker=' in LDFLAGS in order to force the building and execution of native binaries against the dynamic linker provided by uninative. This path is specific to TMPDIR at the time of build, such that the installed files (and the associated sstate) have this path. This prevents the sstate from being portable across build directories/hosts.

(cherry picked from commit ef960d14bd9cecb9a3b50994636fbd455f06104a)
---
 ...-Do-not-include-LDFLAGS-in-generated.patch | 29 +++++++++++++++++++
 meta/recipes-core/ncurses/ncurses_6.2.bb      |  1 +
 2 files changed, 30 insertions(+)
 create mode 100644 meta/recipes-core/ncurses/files/0003-gen-pkgconfig.in-Do-not-include-LDFLAGS-in-generated.patch
meta/recipes-core/ncurses/files/0003-gen-pkgconfig.in-Do-not-include-LDFLAGS-in-generated.patch new file +29 -0
new file mode 100644
index 0000000000..1eb17767a0
@@ -0,0 +1,29 @@
+From 3b3e87934bb6d8511261d7c3d6e39b4f71849272 Mon Sep 17 00:00:00 2001
+From: Nathan Rossi <[email protected]>
+Date: Mon, 14 Dec 2020 13:39:02 +1000
+Subject: [PATCH] gen-pkgconfig.in: Do not include LDFLAGS in generated pc
+ files
+
+Including the LDFLAGS in the pkgconfig output is problematic as OE
+includes build host specific paths and options (e.g. uninative and
+'-Wl,--dynamic-linker=').
+
+Upstream-Status: Inappropriate [OE Specific]
+Signed-off-by: Nathan Rossi <[email protected]>
+---
+ misc/gen-pkgconfig.in | 2 +-
+ 1 file changed, 1 insertion(+), 1 deletion(-)
+
+diff --git a/misc/gen-pkgconfig.in b/misc/gen-pkgconfig.in
+index 8f00b824b9..009d215663 100644
+--- a/misc/gen-pkgconfig.in
++++ b/misc/gen-pkgconfig.in
+@@ -80,7 +80,7 @@ if [ "$includedir" != "/usr/include" ]; then
+ fi
+ 
+ lib_flags=
+-for opt in -L$libdir @LDFLAGS@ @EXTRA_LDFLAGS@ @LIBS@
++for opt in -L$libdir @LIBS@
+ do
+ 	case $opt in
+ 	-l*) # LIBS is handled specially below
meta/recipes-core/ncurses/ncurses_6.2.bb +1 -0
index 5c02db8541..76f0cf97f4 100644
@@ -2,6 +2,7 @@ require ncurses.inc
 
 SRC_URI += "file://0001-tic-hang.patch \
            file://0002-configure-reproducible.patch \
+           file://0003-gen-pkgconfig.in-Do-not-include-LDFLAGS-in-generated.patch \
            "
 # commit id corresponds to the revision in package version
 SRCREV = "a669013cd5e9d6434e5301348ea51baf306c93c4"
Signature · 1 line
2.17.1
#7 lore raw

PATCH 7/30 archiver.bbclass: Fix --runall=deploy_archives for images

In reply to Steve Sakoman

From: Robert Yang <[email protected]>

Fixed:
INHERIT += "archiver"
COPYLEFT_LICENSE_INCLUDE = "*"

$ bitbake core-image-minimal --runall=deploy_archives
[snip]
KeyError: '/path/to/meta/recipes-core/images/core-image-minimal.bb:do_ar_original'
[snip]
(cherry picked from commit 59785a51110c450c7629218f6042f1d9d309618e)
---
 meta/classes/archiver.bbclass | 1 +
 1 file changed, 1 insertion(+)
meta/classes/archiver.bbclass +1 -0
index 1a3c190604..990f1d4674 100644
@@ -591,6 +591,7 @@ addtask do_dumpdata
 addtask do_ar_recipe
 addtask do_deploy_archives
 do_build[recrdeptask] += "do_deploy_archives"
+do_rootfs[recrdeptask] += "do_deploy_archives"
 do_populate_sdk[recrdeptask] += "do_deploy_archives"
 
 python () {
Signature · 1 line
2.17.1
#8 lore raw

PATCH 8/30 perl: fix installation failure because of shell issue

In reply to Steve Sakoman

From: Dmitry Baryshkov <[email protected]>

On one of my buildservers I noticed perl do_install failing with the following message:

| rm: cannot remove '<OEROOT>/tmp-rpb-glibc/work/armv8-2a-linaro-linux/perl/5.32.0-r0/image//usr/lib/perl5/5.32.0/*/CORE/libperl.so': No such file or directory

I tracked this down to shell being dash rather than bash not being able to expand this glob in the middle of the filename. So replace the glob expansion with the simpler one which works in all cases.

(cherry picked from commit d1ea1b5c12120abdd085dc4eb69120af9258a99b)
---
 meta/recipes-devtools/perl/perl_5.30.1.bb | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)
meta/recipes-devtools/perl/perl_5.30.1.bb +3 -2
index b53aff1216..ee6eb6ef0f 100644
@@ -146,8 +146,9 @@ do_install() {
     install lib/ExtUtils/typemap ${D}${libdir}/perl5/${PV}/ExtUtils/
 
     # Fix up shared library
-    rm ${D}/${libdir}/perl5/${PV}/*/CORE/libperl.so
-    ln -sf ../../../../libperl.so.${PERL_LIB_VER} $(echo ${D}/${libdir}/perl5/${PV}/*/CORE)/libperl.so
+    dir=$(echo ${D}/${libdir}/perl5/${PV}/*/CORE)
+    rm $dir/libperl.so
+    ln -sf ../../../../libperl.so.${PERL_LIB_VER} $dir/libperl.so
 }
 
 do_install_append_class-target() {
Signature · 1 line
2.17.1
#9 lore raw

PATCH 9/30 mesa: add more details to elf-tls patch

In reply to Steve Sakoman

From: Anuj Mittal <[email protected]>

Include link to upstream report and the last autobuilder failure [1] so it's easy to identify why this is needed in future.

[1] https://autobuilder.yoctoproject.org/typhoon/#/builders/64/builds/1160/steps/8/logs/step1c

(cherry picked from commit 28647f8e4fb5fe2e56f0609900187bc658ba9ccf)
---
 .../0002-meson.build-make-TLS-ELF-optional.patch  | 15 +++++++++++++++
 1 file changed, 15 insertions(+)
meta/recipes-graphics/mesa/files/0002-meson.build-make-TLS-ELF-optional.patch +15 -0
index cd35a1f850..a64f2faa85 100644
@@ -6,6 +6,21 @@ Subject: [PATCH] meson.build: make TLS ELF optional
 USE_ELF_TLS has replaced GLX_USE_TLS so this patch is the original "make
 TLS GLX optional again" patch updated to the latest mesa.
 
+For details, see:
+https://gitlab.freedesktop.org/mesa/mesa/-/issues/966
+
+This prevents runtime segfault on musl:
+
+Traceback (most recent call last):
+  File "/home/pokybuild/yocto-worker/musl-qemux86/build/meta/lib/oeqa/core/decorator/__init__.py", line 36, in wrapped_f
+    return func(*args, **kwargs)
+  File "/home/pokybuild/yocto-worker/musl-qemux86/build/meta/lib/oeqa/runtime/cases/parselogs.py", line 378, in test_parselogs
+    self.assertEqual(errcount, 0, msg=self.msg)
+AssertionError: 1 != 0 : Log: /home/pokybuild/yocto-worker/musl-qemux86/build/build/tmp/work/qemux86-poky-linux-musl/core-image-sato-sdk/1.0-r0/target_logs/Xorg.0.log
+-----------------------
+Central error: [    10.477] (EE) Failed to load /usr/lib/xorg/modules/extensions/libglx.so: Error relocating /usr/lib/libGL.so.1: alphasort: initial-exec TLS resolves to dynamic definition in /usr/lib/libGL.so.1
+***********************
+
 Upstream-Status: Inappropriate [configuration]
 Signed-off-by: Alistair Francis <[email protected]>
 
Signature · 1 line
2.17.1
#10 lore raw

PATCH 10/30 grub-efi-cfg: exclude OVERRIDES from build_efi_cfg vardeps

In reply to Steve Sakoman

From: Christopher Larson <[email protected]>

This function is primarily *appending* configuration entries to the overrides, it only gets it to ensure it's set at all, so doesn't belong in the vardeps. Having a dependency on OVERRIDES means you cannot use a bbclass like distrooverrides without changing the signatures of recipes using this class.

(cherry picked from commit 8fad5db8c3275a0dc9fdb37761f2e9381e1413da)
---
 meta/classes/grub-efi-cfg.bbclass | 1 +
 1 file changed, 1 insertion(+)
meta/classes/grub-efi-cfg.bbclass +1 -0
index 3a2cdd698b..ea21b3de3d 100644
@@ -120,3 +120,4 @@ python build_efi_cfg() {
 
     cfgfile.close()
 }
+build_efi_cfg[vardepsexclude] += "OVERRIDES"
Signature · 1 line
2.17.1
#11 lore raw

PATCH 11/30 uboot-extlinux-config: exclude OVERRIDES from do_create_extlinux_config vardeps

In reply to Steve Sakoman

From: Christopher Larson <[email protected]>

This function is primarily *appending* configuration entries to the overrides, it only gets it to ensure it's set at all, so doesn't belong in the vardeps. Having a dependency on OVERRIDES means you cannot use a bbclass like distrooverrides without changing the signatures of recipes using this class.

(cherry picked from commit fef74d3651d432977caef8fea54fc54bf2784a74)
---
 meta/classes/uboot-extlinux-config.bbclass | 1 +
 1 file changed, 1 insertion(+)
meta/classes/uboot-extlinux-config.bbclass +1 -0
index f4bf94be04..be285daa01 100644
@@ -153,5 +153,6 @@ python do_create_extlinux_config() {
 }
 UBOOT_EXTLINUX_VARS = "CONSOLE MENU_DESCRIPTION ROOT KERNEL_IMAGE FDTDIR FDT KERNEL_ARGS INITRD"
 do_create_extlinux_config[vardeps] += "${@' '.join(['UBOOT_EXTLINUX_%s_%s' % (v, l) for v in d.getVar('UBOOT_EXTLINUX_VARS').split() for l in d.getVar('UBOOT_EXTLINUX_LABELS').split()])}"
+do_create_extlinux_config[vardepsexclude] += "OVERRIDES"
 
 addtask create_extlinux_config before do_install do_deploy after do_compile
Signature · 1 line
2.17.1
#12 lore raw

PATCH 12/30 wic-image-minimal: only depend on syslinux on x86 targets

In reply to Steve Sakoman

From: Ross Burton <[email protected]>

Following other examples, only depend on syslinux when targetting x86.

(cherry picked from commit c58fcc1379ca5755a5b670f79b75e94370d4943c)
---
 meta-selftest/recipes-test/images/wic-image-minimal.bb | 5 ++++-
 1 file changed, 4 insertions(+), 1 deletion(-)
meta-selftest/recipes-test/images/wic-image-minimal.bb +4 -1
index e1da203b59..1cb019898d 100644
@@ -6,7 +6,10 @@ IMAGE_INSTALL = "packagegroup-core-boot"
 
 IMAGE_FSTYPES = "wic"
 
-WKS_FILE_DEPENDS = "syslinux syslinux-native dosfstools-native mtools-native gptfdisk-native"
+WKS_FILE_DEPENDS = "dosfstools-native mtools-native gptfdisk-native"
+WKS_FILE_DEPENDS_append_x86 = " syslinux-native syslinux"
+WKS_FILE_DEPENDS_append_x86-64 = " syslinux-native syslinux"
+WKS_FILE_DEPENDS_append_x86-x32 = " syslinux-native syslinux"
 
 LIC_FILES_CHKSUM = "file://${COREBASE}/meta/COPYING.MIT;md5=3da9cfbcb788c80a0384361b4de20420"
 
Signature · 1 line
2.17.1
#13 lore raw

PATCH 13/30 coreutils: enable xattrs by default for nativesdk

In reply to Steve Sakoman

From: Nathan Rossi <[email protected]>

When using coreutils nativesdk (e.g. with buildtools-tarball) for running oe-core builds, a number of recipes/classes/etc. expect xattr support. This requirement is also expressed by the existing PACKAGECONFIG_class-native default including xattrs.

(cherry picked from commit 80d7debffdeed165006b26dcb89cffafaaecca06)
---
 meta/recipes-core/coreutils/coreutils_8.31.bb | 3 +++
 1 file changed, 3 insertions(+)
meta/recipes-core/coreutils/coreutils_8.31.bb +3 -0
index bd7a8a2f35..7dd9e41def 100644
@@ -39,6 +39,9 @@ PACKAGECONFIG_class-target ??= "\
 # The lib/oe/path.py requires xattr
 PACKAGECONFIG_class-native ??= "xattr"
 
+# oe-core builds need xattr support
+PACKAGECONFIG_class-nativesdk ??= "xattr"
+
 # with, without, depends, rdepends
 #
 PACKAGECONFIG[acl] = "--enable-acl,--disable-acl,acl,"
Signature · 1 line
2.17.1
#14 lore raw

PATCH 14/30 valgrind: exclude bar_bad/bar_bad_xml from ptests

In reply to Steve Sakoman

From: Alexander Kanavin <[email protected]>

The tests' output is non-deterministic and sometimes
doesn't match the sample output. This has been reported at
https://bugs.kde.org/show_bug.cgi?id=430321
(see also an earlier related bug https://bugs.kde.org/show_bug.cgi?id=358213)

Until upstream figures out how to fix this, let's not run the tests.

(cherry picked from commit 7f7d2fa18267090891754d976cbc3e628324d3dd)
---
 .../valgrind/valgrind/remove-for-aarch64               |  2 --
 meta/recipes-devtools/valgrind/valgrind/remove-for-all |  2 ++
 meta/recipes-devtools/valgrind/valgrind/run-ptest      | 10 ++++++++++
 meta/recipes-devtools/valgrind/valgrind_3.15.0.bb      |  2 ++
 4 files changed, 14 insertions(+), 2 deletions(-)
 create mode 100644 meta/recipes-devtools/valgrind/valgrind/remove-for-all
meta/recipes-devtools/valgrind/valgrind/remove-for-aarch64 +0 -2
index a3a0c6e50f..93bfd45a4e 100644
@@ -31,8 +31,6 @@ drd/tests/annotate_static
 drd/tests/annotate_trace_memory
 drd/tests/annotate_trace_memory_xml
 drd/tests/atomic_var
-drd/tests/bar_bad
-drd/tests/bar_bad_xml
 drd/tests/bar_trivial
 drd/tests/bug-235681
 drd/tests/bug322621
meta/recipes-devtools/valgrind/valgrind/remove-for-all new file +2 -0
new file mode 100644
index 0000000000..d6a85c4735
@@ -0,0 +1,2 @@
+drd/tests/bar_bad
+drd/tests/bar_bad_xml
meta/recipes-devtools/valgrind/valgrind/run-ptest +10 -0
index 97b0a85dbf..7217dfca5d 100755
@@ -17,6 +17,12 @@ EXP_TOOLS="exp-bbv exp-dhat exp-sgcheck"
 GDB_BIN=@bindir@/gdb
 cd ${VALGRIND_LIB}/ptest && ./gdbserver_tests/make_local_links ${GDB_BIN}
 
+echo "Hide valgrind tests that are non-deterministic"
+echo "Reported at https://bugs.kde.org/show_bug.cgi?id=430321"
+for i in `cat remove-for-all`; do
+   mv $i.vgtest $i.IGNORE;
+done
+
 arch=`arch`
 if [ "$arch" = "aarch64" ]; then
    echo "Aarch64: Hide valgrind tests that result in defunct process and then out of memory"
@@ -44,6 +50,10 @@ if [ "$arch" = "aarch64" ]; then
    done
 fi
 
+echo "Restore valgrind tests that are non-deterministc"
+for i in `cat remove-for-all`; do
+   mv $i.IGNORE $i.vgtest;
+done
 
 passed=`grep PASS: ${LOG}|wc -l`
 failed=`grep FAIL: ${LOG}|wc -l`
meta/recipes-devtools/valgrind/valgrind_3.15.0.bb +2 -0
index a764d18177..e16a58fc62 100644
@@ -16,6 +16,7 @@ SRC_URI = "https://sourceware.org/pub/valgrind/valgrind-${PV}.tar.bz2 \
            file://Added-support-for-PPC-instructions-mfatbu-mfatbl.patch \
            file://run-ptest \
            file://remove-for-aarch64 \
+           file://remove-for-all \
            file://0004-Fix-out-of-tree-builds.patch \
            file://0005-Modify-vg_test-wrapper-to-support-PTEST-formats.patch \
            file://0001-Remove-tests-that-fail-to-build-on-some-PPC32-config.patch \
@@ -171,6 +172,7 @@ do_install_ptest() {
     # The scripts reference config.h so add it to the top ptest dir.
     cp ${B}/config.h ${D}${PTEST_PATH}
     install -D ${WORKDIR}/remove-for-aarch64 ${D}${PTEST_PATH}
+    install -D ${WORKDIR}/remove-for-all ${D}${PTEST_PATH}
 
     # Add an executable need by none/tests/bigcode
     mkdir ${D}${PTEST_PATH}/perf
Signature · 1 line
2.17.1
#15 lore raw

PATCH 15/30 meta: add missing descriptions in some support recipes

In reply to Steve Sakoman

From: Maxime Roussin-Bélanger <[email protected]>

(cherry picked from commit d2654384db4999c78bc3d98215a4eecdab63541b)
---
 meta/recipes-support/apr/apr_1.7.0.bb                        | 4 ++++
 meta/recipes-support/aspell/aspell_0.60.8.bb                 | 4 ++++
 meta/recipes-support/atk/at-spi2-core_2.34.0.bb              | 2 ++
 meta/recipes-support/attr/acl_2.2.53.bb                      | 2 ++
 meta/recipes-support/attr/attr.inc                           | 4 ++++
 meta/recipes-support/bash-completion/bash-completion_2.10.bb | 2 ++
 meta/recipes-support/nettle/nettle_3.5.1.bb                  | 3 +++
 7 files changed, 21 insertions(+)
meta/recipes-support/apr/apr_1.7.0.bb +4 -0
index 7073af8c98..c9b9bf0f50 100644
@@ -1,4 +1,8 @@
 SUMMARY = "Apache Portable Runtime (APR) library"
+DESCRIPTION = "The Apache Portable Runtime (APR) is a supporting library for the \
+Apache web server. It provides a set of APIs that map to the underlying \
+operating system (OS). Where the OS does not support a particular function, \
+APR will provide an emulation."
 HOMEPAGE = "http://apr.apache.org/"
 SECTION = "libs"
 DEPENDS = "util-linux"
meta/recipes-support/aspell/aspell_0.60.8.bb +4 -0
index 629987810a..f1d931b39c 100644
@@ -1,4 +1,8 @@
 SUMMARY = "GNU Aspell spell-checker"
+DESCRIPTION = "GNU Aspell is a spell-checker which can be used either as a \
+standalone application or embedded in other programs. Its main feature is that \
+it does a much better job of suggesting possible spellings than just about any \
+other spell-checker available for the English language"
 SECTION = "console/utils"
 
 LICENSE = "LGPLv2 | LGPLv2.1"
meta/recipes-support/atk/at-spi2-core_2.34.0.bb +2 -0
index 84e05e77fc..bcef8ef169 100644
@@ -1,4 +1,6 @@
 SUMMARY = "Assistive Technology Service Provider Interface (dbus core)"
+DESCRIPTION = "At-Spi2 is a protocol over DBus, toolkit widgets use it to \
+provide their content to screen readers such as Orca."
 HOMEPAGE = "https://wiki.linuxfoundation.org/accessibility/d-bus"
 LICENSE = "LGPL-2.1+"
 LIC_FILES_CHKSUM = "file://COPYING;md5=4fbd65380cdd255951079008b364516c"
meta/recipes-support/attr/acl_2.2.53.bb +2 -0
index 5bb50f77f7..b120c1f16f 100644
@@ -1,5 +1,7 @@
 SUMMARY = "Utilities for managing POSIX Access Control Lists"
 HOMEPAGE = "http://savannah.nongnu.org/projects/acl/"
+DESCRIPTION = "ACL allows you to provide different levels of access to files \
+and folders for different users."
 SECTION = "libs"
 
 LICENSE = "LGPLv2.1+ & GPLv2+"
meta/recipes-support/attr/attr.inc +4 -0
index f13a83a7b4..8515f96bf7 100644
@@ -1,4 +1,8 @@
 SUMMARY = "Utilities for manipulating filesystem extended attributes"
+DESCRIPTION = "A set of tools for manipulating extended attributes on filesystem \
+objects, in particular getfattr(1) and setfattr(1). An attr(1) command \
+is also provided which is largely compatible with the SGI IRIX tool of \
+the same name."
 HOMEPAGE = "http://savannah.nongnu.org/projects/attr/"
 SECTION = "libs"
 
meta/recipes-support/bash-completion/bash-completion_2.10.bb +2 -0
index 93e7d9dc3c..58e565dee5 100644
@@ -1,4 +1,6 @@
 SUMMARY = "Programmable Completion for Bash 4"
+DESCRIPTION = "bash completion extends bash's standard completion behavior to \
+achieve complex command lines with just a few keystrokes."
 HOMEPAGE = "https://github.com/scop/bash-completion"
 BUGTRACKER = "https://github.com/scop/bash-completion/issues"
 
meta/recipes-support/nettle/nettle_3.5.1.bb +3 -0
index d92db0ef95..a9550ed033 100644
@@ -1,5 +1,8 @@
 SUMMARY = "A low level cryptographic library"
 HOMEPAGE = "http://www.lysator.liu.se/~nisse/nettle/"
+DESCRIPTION = "It tries to solve a problem of providing a common set of \
+cryptographic algorithms for higher-level applications by implementing a \
+context-independent set of cryptographic algorithms"
 SECTION = "libs"
 LICENSE = "LGPLv3+ | GPLv2+"
 
Signature · 1 line
2.17.1
#16 lore raw

PATCH 16/30 pulseaudio: Remove OE_LT_RPATH_ALLOW

In reply to Steve Sakoman

From: Tanu Kaskinen <[email protected]>

I don't know what these variables were supposed to do, but there doesn't seem to be anything that would use these variables, so removing them should be safe. The PulseAudio recipe is the only place where these appear in the repository.

These lines were included already in the original PulseAudio recipe that was added in commit d9a4c588ef24e7e4019fbe5a2314addbcd3a6bb8 in 2007, and already at that time there didn't seem to be anything using these variables.

(cherry picked from commit 7a0b9209a366de7a2aed9b23a337b4ffdba61abe)
---
 meta/recipes-multimedia/pulseaudio/pulseaudio.inc | 5 -----
 1 file changed, 5 deletions(-)
meta/recipes-multimedia/pulseaudio/pulseaudio.inc +0 -5
index c7f3e67022..417eb1d9d3 100644
@@ -137,11 +137,6 @@ EXTRA_OECONF_append_armeb = "${@bb.utils.contains("TUNE_FEATURES", "neon", "", "
 
 export TARGET_PFPU = "${TARGET_FPU}"
 
-# TODO: Use more fine granular version
-#OE_LT_RPATH_ALLOW=":${libdir}/pulse-0.9:"
-OE_LT_RPATH_ALLOW = "any"
-OE_LT_RPATH_ALLOW[export]="1"
-
 set_cfg_value () {
 	sed -i -e "s/\(; *\)\?$2 =.*/$2 = $3/" "$1"
 	if ! grep -q "^$2 = $3\$" "$1"; then
Signature · 1 line
2.17.1
#17 lore raw

PATCH 17/30 lib/oe/qa: handle the 'no specific instruction set' ELF e_machine value

In reply to Steve Sakoman

From: Ross Burton <[email protected]>

[RP: Update OEQA selftest to match change]
(cherry picked from commit b7cfc0f51cc0b4866f913f6eae4fcc6f72d2578c)
---
 meta/lib/oe/qa.py                         | 1 +
 meta/lib/oeqa/selftest/cases/oelib/elf.py | 2 +-
 2 files changed, 2 insertions(+), 1 deletion(-)
meta/lib/oe/qa.py +1 -0
index ea831b930a..e8a854a302 100644
@@ -156,6 +156,7 @@ def elf_machine_to_string(machine):
     """
     try:
         return {
+            0x00: "Unset",
             0x02: "SPARC",
             0x03: "x86",
             0x08: "MIPS",
meta/lib/oeqa/selftest/cases/oelib/elf.py +1 -1
index d0a28090f2..5a5f9b4fdf 100644
@@ -21,6 +21,6 @@ class TestElf(TestCase):
         self.assertEqual(oe.qa.elf_machine_to_string(0xB7), "AArch64")
         self.assertEqual(oe.qa.elf_machine_to_string(0xF7), "BPF")
 
-        self.assertEqual(oe.qa.elf_machine_to_string(0x00), "Unknown (0)")
+        self.assertEqual(oe.qa.elf_machine_to_string(0x00), "Unset")
         self.assertEqual(oe.qa.elf_machine_to_string(0xDEADBEEF), "Unknown (3735928559)")
         self.assertEqual(oe.qa.elf_machine_to_string("foobar"), "Unknown ('foobar')")
Signature · 1 line
2.17.1
#18 lore raw

PATCH 18/30 initscripts: use quotes for shell variable comparision

In reply to Steve Sakoman

From: Khem Raj <[email protected]>

Helps to execute it with busybox shell

(cherry picked from commit 45ba0ca0352bca46f974d28781ac935d8e9ec3ea)
---
 meta/recipes-core/initscripts/initscripts-1.0/checkroot.sh | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
meta/recipes-core/initscripts/initscripts-1.0/checkroot.sh +1 -1
index 02f0351fcb..a63e71b780 100755
@@ -74,7 +74,7 @@ test "$VERBOSE" != no && echo "Activating swap"
 #
 # Check the root filesystem.
 #
-if test -f /fastboot || test $rootcheck = no
+if test -f /fastboot || test "$rootcheck" = "no"
 then
   test $rootcheck = yes && echo "Fast boot, no filesystem check"
 else
Signature · 1 line
2.17.1
#19 lore raw

PATCH 19/30 oeqa/commands: Ensure sync can be found regardless of PATH

In reply to Steve Sakoman

From: Richard Purdie <[email protected]>

Avoid command not found errors shown in selftest logs due to changes to PATH settings which also risks intermittent problems due to IO load.

(cherry picked from commit 40bcae01b0be2f293dea9ab42c6b7f8f47827cf5)
---
 meta/lib/oeqa/utils/commands.py | 3 +++
 1 file changed, 3 insertions(+)
meta/lib/oeqa/utils/commands.py +3 -0
index 8059cbce3e..8b3e12038d 100644
@@ -188,7 +188,10 @@ def runCmd(command, ignore_status=False, timeout=None, assert_error=True, sync=T
     # call sync around the tests to ensure the IO queue doesn't get too large, taking any IO
     # hit here rather than in bitbake shutdown.
     if sync:
+        p = os.environ['PATH']
+        os.environ['PATH'] = "/usr/bin:/bin:/usr/sbin:/sbin:" + p
         os.system("sync")
+        os.environ['PATH'] = p
 
     result.command = command
     result.status = cmd.status
Signature · 1 line
2.17.1
#20 lore raw

PATCH 20/30 weston: Fix PACKAGECONFIG for remoting

In reply to Steve Sakoman

From: Robert Yang <[email protected]>

Fix a typo: gstreamer-1.0 -> gstreamer1.0.

(cherry picked from commit d5b030dd9b9ada796f90919c97c887b00612cfc0)
---
 meta/recipes-graphics/wayland/weston_8.0.0.bb | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
meta/recipes-graphics/wayland/weston_8.0.0.bb +1 -1
index 8fef864827..0b383f25f3 100644
@@ -70,7 +70,7 @@ PACKAGECONFIG[colord] = "-Dcolor-management-colord=true,-Dcolor-management-color
 # Clients support
 PACKAGECONFIG[clients] = "-Dsimple-clients=all -Ddemo-clients=true,-Dsimple-clients= -Ddemo-clients=false"
 # Virtual remote output with GStreamer on DRM backend
-PACKAGECONFIG[remoting] = "-Dremoting=true,-Dremoting=false,gstreamer-1.0"
+PACKAGECONFIG[remoting] = "-Dremoting=true,-Dremoting=false,gstreamer1.0"
 # Weston with PAM support
 PACKAGECONFIG[pam] = "-Dpam=true,-Dpam=false,libpam"
 # Weston with screen-share support
Signature · 1 line
2.17.1
#21 lore raw

PATCH 21/30 apr-util: Only specify --with-dbm=gdbm if gdbm support is enabled

In reply to Steve Sakoman

From: Peter Kjellerstedt <[email protected]>

Support for gdbm was made optional in 3260ad9e, but it was still being used unconditionally.

(cherry picked from commit 09d303ca295dc27874c72b30c37a64d1fdf4c5c0)
---
 meta/recipes-support/apr/apr-util_1.6.1.bb | 5 ++---
 1 file changed, 2 insertions(+), 3 deletions(-)
meta/recipes-support/apr/apr-util_1.6.1.bb +2 -3
index 0dd8f025e8..f7d827a1d8 100644
@@ -19,10 +19,9 @@ SRC_URI = "${APACHE_MIRROR}/apr/${BPN}-${PV}.tar.gz \
 SRC_URI[md5sum] = "bd502b9a8670a8012c4d90c31a84955f"
 SRC_URI[sha256sum] = "b65e40713da57d004123b6319828be7f1273fbc6490e145874ee1177e112c459"
 
-EXTRA_OECONF = "--with-apr=${STAGING_BINDIR_CROSS}/apr-1-config \ 
+EXTRA_OECONF = "--with-apr=${STAGING_BINDIR_CROSS}/apr-1-config \
 		--without-odbc \
 		--without-pgsql \
-		--with-dbm=gdbm \
 		--without-sqlite2 \
 		--with-expat=${STAGING_DIR_HOST}${prefix}"
 
@@ -69,7 +68,7 @@ PACKAGECONFIG ??= "crypto gdbm"
 PACKAGECONFIG[ldap] = "--with-ldap,--without-ldap,openldap"
 PACKAGECONFIG[crypto] = "--with-openssl=${STAGING_DIR_HOST}${prefix} --with-crypto,--without-crypto,openssl"
 PACKAGECONFIG[sqlite3] = "--with-sqlite3=${STAGING_DIR_HOST}${prefix},--without-sqlite3,sqlite3"
-PACKAGECONFIG[gdbm] = "--with-gdbm=${STAGING_DIR_HOST}${prefix},--without-gdbm,gdbm"
+PACKAGECONFIG[gdbm] = "--with-dbm=gdbm --with-gdbm=${STAGING_DIR_HOST}${prefix},--without-gdbm,gdbm"
 
 #files ${libdir}/apr-util-1/*.so are not symlinks but loadable modules thus they are packaged in ${PN}
 FILES_${PN}     += "${libdir}/apr-util-1/apr*${SOLIBS} ${libdir}/apr-util-1/apr*${SOLIBSDEV}"
Signature · 1 line
2.17.1
#22 lore raw

PATCH 22/30 populate_sdk_base: Fix condition syntax if SDK_RELOCATE_AFTER_INSTALL is disabled

In reply to Steve Sakoman

From: Tomasz Dziendzielski <[email protected]>

If variable is set to empty string the comparison is "if [ -eq 1 ]" which fails with "[: -eq: unary operator expected".

(cherry picked from commit 36a2dc83fd0e1d6e2b8441e24a4cbc48a6c4fc19)
---
 meta/classes/populate_sdk_base.bbclass | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
meta/classes/populate_sdk_base.bbclass +1 -1
index 3e5b1359d6..39a7cadaf8 100644
@@ -256,7 +256,7 @@ fakeroot create_shar() {
 
 	rm -f ${T}/pre_install_command ${T}/post_install_command
 
-	if [ ${SDK_RELOCATE_AFTER_INSTALL} -eq 1 ] ; then
+	if [ "${SDK_RELOCATE_AFTER_INSTALL}" = "1" ] ; then
 		cp ${TOOLCHAIN_SHAR_REL_TMPL} ${T}/post_install_command
 	fi
 	cat << "EOF" >> ${T}/pre_install_command
Signature · 1 line
2.17.1
#23 lore raw

PATCH 23/30 grub: Further reproducibility fix

In reply to Steve Sakoman

From: Richard Purdie <[email protected]>

Fix the ordering in the generated unidata.c file to aid reproducibility.

[YOCTO #14167]

(cherry picked from commit 6d9c9f7604fd32ef926726a46ae053bbab6ccb4f)
---
 meta/recipes-bsp/grub/files/determinism.patch | 16 ++++++++++++++++
 1 file changed, 16 insertions(+)
meta/recipes-bsp/grub/files/determinism.patch +16 -0
index fac80e7b54..3c1f562c71 100644
@@ -8,6 +8,9 @@ Also, the output of the SOURCES lines in grub-core/Makefile.core.am, generated
 from grub-core/Makefile.core.def with gentpl.py is not deterministic due to
 missing sorting of the list used to generate it. Add such a sort.
 
+Also ensure the generated unidata.c file is deterministic by sorting the
+keys of the dict.
+
 Upstream-Status: Pending
 Richard Purdie <[email protected]>
 
@@ -38,3 +41,16 @@ Index: grub-2.04/gentpl.py
      return ''.join(r)
  
  def platform_conditional(platform, closure):
+Index: grub-2.04/util/import_unicode.py
+===================================================================
+--- grub-2.04.orig/util/import_unicode.py
++++ grub-2.04/util/import_unicode.py
+@@ -174,7 +174,7 @@ infile.close ()
+ 
+ outfile.write ("struct grub_unicode_arabic_shape grub_unicode_arabic_shapes[] = {\n ")
+ 
+-for x in arabicsubst:
++for x in sorted(arabicsubst):
+     try:
+         if arabicsubst[x]['join'] == "DUAL":
+             outfile.write ("{0x%x, 0x%x, 0x%x, 0x%x, 0x%x},\n " % (arabicsubst[x][0], arabicsubst[x][1], arabicsubst[x][2], arabicsubst[x][3], arabicsubst[x][4]))
Signature · 1 line
2.17.1
#24 lore raw

PATCH 24/30 timezone: upgrade to 2020e

In reply to Steve Sakoman

From: Ovidiu Panait <[email protected]>

Briefly:
Volgograd switches to Moscow time on 2020-12-27 at 02:00.

Changes to future timestamps

Volgograd changes time zone from +04 to +03 on 2020-12-27 at 02:00. (Thanks to Alexander Krivenyshev and Stepan Golosunov.)

Changes to past timestamps

Correct many pre-1986 transitions, fixing entries originally derived from Shanks. The fixes include:
- Australia: several 1917 through 1971 transitions
- Bahamas: several 1941 through 1945 transitions
- Bermuda: several 1917 through 1956 transitions
- Belize: several 1942 through 1968 transitions
- Ghana: several 1915 through 1956 transitions
- Israel and Palestine: several 1940 through 1985 transitions
- Kenya and adjacent: several 1908 through 1960 transitions
- Nigeria and adjacent: correcting LMT in Lagos, and several 1905 through 1919 transitions
- Seychelles: the introduction of standard time in 1907, not 1906
- Vanuatu: DST in 1973-1974, and a corrected 1984 transition
(Thanks to P Chan.)

Because of the Australia change, Australia/Currie (King Island) is no longer needed, as it is identical to Australia/Hobart for all timestamps since 1970 and was therefore created by mistake. Australia/Currie has been moved to the 'backward' file and its corrected data moved to the 'backzone' file.

Changes to past time zone abbreviations and DST flags

To better match legislation in Turks and Caicos, the 2015 shift to year-round observance of -04 is now modeled as AST throughout before returning to Eastern Time with US DST in 2018, rather than as maintaining EDT until 2015-11-01. (Thanks to P Chan.)

Changes to documentation

The zic man page now documents zic's coalescing of transitions when a zone falls back just before DST springs forward.

(cherry picked from commit 9d7643339b4db6609b2bad1f76599a54da9d17f2)
---
 meta/recipes-extended/timezone/timezone.inc | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)
meta/recipes-extended/timezone/timezone.inc +3 -3
index 5368464f30..d175f0d4a0 100644
@@ -6,7 +6,7 @@ SECTION = "base"
 LICENSE = "PD & BSD & BSD-3-Clause"
 LIC_FILES_CHKSUM = "file://LICENSE;md5=c679c9d6b02bc2757b3eaf8f53c43fba"
 
-PV = "2020d"
+PV = "2020e"
 
 SRC_URI =" http://www.iana.org/time-zones/repository/releases/tzcode${PV}.tar.gz;name=tzcode \
            http://www.iana.org/time-zones/repository/releases/tzdata${PV}.tar.gz;name=tzdata \
@@ -14,5 +14,5 @@ SRC_URI =" http://www.iana.org/time-zones/repository/releases/tzcode${PV}.tar.gz
 
 UPSTREAM_CHECK_URI = "http://www.iana.org/time-zones"
 
-SRC_URI[tzcode.sha256sum] = "6cf050ba28e8053029d3f32d71341d11a794c6b5dd51a77fc769d6dae364fad5"
-SRC_URI[tzdata.sha256sum] = "8d813957de363387696f05af8a8889afa282ab5016a764c701a20758d39cbaf3"
+SRC_URI[tzcode.sha256sum] = "3e10308976b09305d15cb4a32ff75483421f2063bfa24a9be366a027e7cd2902"
+SRC_URI[tzdata.sha256sum] = "0be1ba329eae29ae1b54057c3547b3e672f73b3ae7643aa87dac85122bec037e"
Signature · 1 line
2.17.1
#25 lore raw

PATCH 25/30 wic/direct/kparser: ensure fsuuid for vfat and msdos align with format

In reply to Steve Sakoman

From: Lee Chee Yang <[email protected]>

vfat/msdos filesystem should have fsuuid in format 0xYYYYYYYY where "0x" in front follow with 8 hexadecimal number in uppercase. In wic, when using custom fsuuid for vfat/msdos partition in wks, it is able to set the value in any length, with or without leading "0x". This can cause fsuuid missaligned when fstab updates, fstab expect exactly 10 character fsuuid for vfat/msdos partition and all in uppercase.

if custom fsuuid for vfat/msdos is set, check the length and format, error if it exceed the format size. Amend it so it is align with format 0xYYYYYYYY. This is done before image create and fstab update to ensure the fsuuid are same in all followup process. if custom fsuuid length less than expected, fill in "0".

[YOCTO #14161]

(cherry picked from commit d9686ae511ef10a504becfd81bfe296b788e1456)
---
 scripts/lib/wic/ksparser.py              | 17 +++++++++++++++++
 scripts/lib/wic/plugins/imager/direct.py |  7 +++++++
 2 files changed, 24 insertions(+)
scripts/lib/wic/ksparser.py +17 -0
index 76cc55b848..127ca79ade 100644
@@ -228,6 +228,23 @@ class KickStart():
                                 err = "%s:%d: SquashFS does not support LABEL" \
                                        % (confpath, lineno)
                                 raise KickStartError(err)
+                        if parsed.fstype == 'msdos' or parsed.fstype == 'vfat':
+                            if parsed.fsuuid:
+                                if parsed.fsuuid.upper().startswith('0X'):
+                                    if len(parsed.fsuuid) > 10:
+                                        err = "%s:%d: fsuuid %s given in wks kickstart file " \
+                                              "exceeds the length limit for %s filesystem. " \
+                                              "It should be in the form of a 32 bit hexadecimal" \
+                                              "number (for example, 0xABCD1234)." \
+                                              % (confpath, lineno, parsed.fsuuid, parsed.fstype)
+                                        raise KickStartError(err)
+                                elif len(parsed.fsuuid) > 8:
+                                    err = "%s:%d: fsuuid %s given in wks kickstart file " \
+                                          "exceeds the length limit for %s filesystem. " \
+                                          "It should be in the form of a 32 bit hexadecimal" \
+                                          "number (for example, 0xABCD1234)." \
+                                          % (confpath, lineno, parsed.fsuuid, parsed.fstype)
+                                    raise KickStartError(err)
                         if parsed.use_label and not parsed.label:
                             err = "%s:%d: Must set the label with --label" \
                                   % (confpath, lineno)
scripts/lib/wic/plugins/imager/direct.py +7 -0
index 55db826e93..0ca67b77d5 100644
@@ -343,6 +343,13 @@ class PartitionedImage():
                     part.fsuuid = '0x' + str(uuid.uuid4())[:8].upper()
                 else:
                     part.fsuuid = str(uuid.uuid4())
+            else:
+                #make sure the fsuuid for vfat/msdos align with format 0xYYYYYYYY
+                if part.fstype == 'vfat' or part.fstype == 'msdos':
+                    if part.fsuuid.upper().startswith("0X"):
+                        part.fsuuid = '0x' + part.fsuuid.upper()[2:].rjust(8,"0")
+                    else:
+                        part.fsuuid = '0x' + part.fsuuid.upper().rjust(8,"0")
 
     def prepare(self, imager):
         """Prepare an image. Call prepare method of all image partitions."""
Signature · 1 line
2.17.1
#26 lore raw

PATCH 26/30 man-db: Fix reproducibility issue

In reply to Steve Sakoman

From: Richard Purdie <[email protected]>

Executing a sequence like:

bitbake man-db
bitbake man-db -c clean
bitbake man-db -c package_write_deb -f

results in the ownership of files in the man-db package changing from "man/man" to "6/12". This is due to base-passwd not being installed in the recipe sysroot.

Add the missing dependency so even if a recipe "re-packages", the name of the user/group is found and the packages are reproducible.

[YOCTO #14172]

(cherry picked from commit 769f8be29a192fd4fa0c4752f851fa3728d71179)
---
 meta/recipes-extended/man-db/man-db_2.9.0.bb | 1 +
 1 file changed, 1 insertion(+)
meta/recipes-extended/man-db/man-db_2.9.0.bb +1 -0
index 493fe817e5..333fbfa76d 100644
@@ -12,6 +12,7 @@ SRC_URI[sha256sum] = "5d4aacd9e8876d6a3203a889860c3524c293c38f04111a3350deab8a6c
 
 DEPENDS = "libpipeline gdbm groff-native base-passwd"
 RDEPENDS_${PN} += "base-passwd"
+PACKAGE_WRITE_DEPS += "base-passwd"
 
 # | /usr/src/debug/man-db/2.8.0-r0/man-db-2.8.0/src/whatis.c:939: undefined reference to `_nl_msg_cat_cntr'
 USE_NLS_libc-musl = "no"
Signature · 1 line
2.17.1
#27 lore raw

PATCH 27/30 libexif: fix CVE-2020-0198; CVE-2020-0452

In reply to Steve Sakoman

From: Changqing Li <[email protected]>

(cherry picked from commit 46d837442ab216941df2d02f60c69155463e02d8)
---
 .../libexif/files/CVE-2020-0198.patch         | 66 +++++++++++++++++++
 .../libexif/files/CVE-2020-0452.patch         | 39 +++++++++++
 .../recipes-support/libexif/libexif_0.6.22.bb |  2 +
 3 files changed, 107 insertions(+)
 create mode 100644 meta/recipes-support/libexif/files/CVE-2020-0198.patch
 create mode 100644 meta/recipes-support/libexif/files/CVE-2020-0452.patch
meta/recipes-support/libexif/files/CVE-2020-0198.patch new file +66 -0
new file mode 100644
index 0000000000..2a48844cb2
@@ -0,0 +1,66 @@
+From ca71eda33fe8421f98fbe20eb4392473357c1c43 Mon Sep 17 00:00:00 2001
+From: Changqing Li <[email protected]>
+Date: Wed, 30 Dec 2020 10:22:47 +0800
+Subject: [PATCH] fixed another unsigned integer overflow
+
+first fixed by google in android fork,
+https://android.googlesource.com/platform/external/libexif/+/1e187b62682ffab5003c702657d6d725b4278f16%5E%21/#F0
+
+(use a more generic overflow check method, also check second overflow instance.)
+
+https://security-tracker.debian.org/tracker/CVE-2020-0198
+
+Upstream-Status: Backport[https://github.com/libexif/libexif/commit/ce03ad7ef4e8aeefce79192bf5b6f69fae396f0c]
+CVE: CVE-2020-0198
+
+Signed-off-by: Changqing Li <[email protected]>
+---
+ libexif/exif-data.c | 10 ++++++----
+ 1 file changed, 6 insertions(+), 4 deletions(-)
+
+diff --git a/libexif/exif-data.c b/libexif/exif-data.c
+index 8b280d3..34d58fc 100644
+--- a/libexif/exif-data.c
++++ b/libexif/exif-data.c
+@@ -47,6 +47,8 @@
+ #undef JPEG_MARKER_APP1
+ #define JPEG_MARKER_APP1 0xe1
+ 
++#define CHECKOVERFLOW(offset,datasize,structsize) (( offset >= datasize) || (structsize > datasize) || (offset > datasize - structsize ))
++
+ static const unsigned char ExifHeader[] = {0x45, 0x78, 0x69, 0x66, 0x00, 0x00};
+ 
+ struct _ExifDataPrivate
+@@ -327,7 +329,7 @@ exif_data_load_data_thumbnail (ExifData *data, const unsigned char *d,
+ 		exif_log (data->priv->log, EXIF_LOG_CODE_DEBUG, "ExifData", "Bogus thumbnail offset (%u).", o);
+ 		return;
+ 	}
+-	if (s > ds - o) {
++	if (CHECKOVERFLOW(o,ds,s)) {
+ 		exif_log (data->priv->log, EXIF_LOG_CODE_DEBUG, "ExifData", "Bogus thumbnail size (%u), max would be %u.", s, ds-o);
+ 		return;
+ 	}
+@@ -420,9 +422,9 @@ exif_data_load_data_content (ExifData *data, ExifIfd ifd,
+ 	}
+ 
+ 	/* Read the number of entries */
+-	if ((offset + 2 < offset) || (offset + 2 < 2) || (offset + 2 > ds)) {
++	if (CHECKOVERFLOW(offset, ds, 2)) {
+ 		exif_log (data->priv->log, EXIF_LOG_CODE_CORRUPT_DATA, "ExifData",
+-			  "Tag data past end of buffer (%u > %u)", offset+2, ds);
++			  "Tag data past end of buffer (%u+2 > %u)", offset, ds);
+ 		return;
+ 	}
+ 	n = exif_get_short (d + offset, data->priv->order);
+@@ -431,7 +433,7 @@ exif_data_load_data_content (ExifData *data, ExifIfd ifd,
+ 	offset += 2;
+ 
+ 	/* Check if we have enough data. */
+-	if (offset + 12 * n > ds) {
++	if (CHECKOVERFLOW(offset, ds, 12*n)) {
+ 		n = (ds - offset) / 12;
+ 		exif_log (data->priv->log, EXIF_LOG_CODE_DEBUG, "ExifData",
+ 				  "Short data; only loading %hu entries...", n);
+-- 
+2.17.1
+
meta/recipes-support/libexif/files/CVE-2020-0452.patch new file +39 -0
new file mode 100644
index 0000000000..a117b8b369
@@ -0,0 +1,39 @@
+From 302acd49eba0a125b0f20692df6abc6f7f7ca53e Mon Sep 17 00:00:00 2001
+From: Changqing Li <[email protected]>
+Date: Wed, 30 Dec 2020 10:18:51 +0800
+Subject: [PATCH] fixed a incorrect overflow check that could be optimized
+ away.
+
+inspired by:
+https://android.googlesource.com/platform/external/libexif/+/8e7345f3bc0bad06ac369d6cbc1124c8ceaf7d4b
+
+https://source.android.com/security/bulletin/2020-11-01
+
+CVE-2020-0452
+
+Upsteam-Status: Backport[https://github.com/libexif/libexif/commit/9266d14b5ca4e29b970fa03272318e5f99386e06]
+CVE: CVE-2020-0452
+
+Signed-off-by: Changqing Li <[email protected]>
+---
+ libexif/exif-entry.c | 4 ++--
+ 1 file changed, 2 insertions(+), 2 deletions(-)
+
+diff --git a/libexif/exif-entry.c b/libexif/exif-entry.c
+index 5de215f..3a6ce84 100644
+--- a/libexif/exif-entry.c
++++ b/libexif/exif-entry.c
+@@ -1371,8 +1371,8 @@ exif_entry_get_value (ExifEntry *e, char *val, unsigned int maxlen)
+ 	{
+ 		unsigned char *utf16;
+ 
+-		/* Sanity check the size to prevent overflow */
+-		if (e->size+sizeof(uint16_t)+1 < e->size) break;
++		/* Sanity check the size to prevent overflow. Note EXIF files are 64kb at most. */
++		if (e->size >= 65536 - sizeof(uint16_t)*2) break;
+ 
+ 		/* The tag may not be U+0000-terminated , so make a local
+ 		   U+0000-terminated copy before converting it */
+-- 
+2.17.1
+
meta/recipes-support/libexif/libexif_0.6.22.bb +2 -0
index a520d5c9f9..3b08dc52be 100644
@@ -8,6 +8,8 @@ def version_underscore(v):
     return "_".join(v.split("."))
 
 SRC_URI = "https://github.com/libexif/libexif/releases/download/libexif-${@version_underscore("${PV}")}-release/libexif-${PV}.tar.xz \
+           file://CVE-2020-0198.patch \
+           file://CVE-2020-0452.patch \
            "
 
 SRC_URI[sha256sum] = "5048f1c8fc509cc636c2f97f4b40c293338b6041a5652082d5ee2cf54b530c56"
Signature · 1 line
2.17.1
#28 lore raw

PATCH 28/30 gdk-pixbuf: fix CVE-2020-29385

In reply to Steve Sakoman

From: Lee Chee Yang <[email protected]>

Backport patch from
https://gitlab.gnome.org/GNOME/gdk-pixbuf/-/commit/bdd3acbd48a575d418ba6bf1b32d7bda2fae1c81

image file in upstream patch is for test purpose only, it cause error during do_patch so drop it:
File tests/test-images/fail/hang_114.gif: git binary diffs are not supported.

(cherry picked from commit 7cd401b7c60dba91f6fb10395ab4a10a267cf23d)
---
 .../gdk-pixbuf/CVE-2020-29385.patch           | 55 +++++++++++++++++++
 .../gdk-pixbuf/gdk-pixbuf_2.40.0.bb           |  1 +
 2 files changed, 56 insertions(+)
 create mode 100644 meta/recipes-gnome/gdk-pixbuf/gdk-pixbuf/CVE-2020-29385.patch
meta/recipes-gnome/gdk-pixbuf/gdk-pixbuf/CVE-2020-29385.patch new file +55 -0
new file mode 100644
index 0000000000..3fef2bc1eb
@@ -0,0 +1,55 @@
+From bdd3acbd48a575d418ba6bf1b32d7bda2fae1c81 Mon Sep 17 00:00:00 2001
+From: Robert Ancell <[email protected]>
+Date: Mon, 30 Nov 2020 12:26:12 +1300
+Subject: [PATCH 02/13] gif: Fix LZW decoder accepting invalid LZW code.
+
+The code value after a reset wasn't being validated, which means we would
+accept invalid codes. This could cause an infinite loop in the decoder.
+
+Fixes CVE-2020-29385
+
+Fixes https://gitlab.gnome.org/GNOME/gdk-pixbuf/-/issues/164
+
+Upstream-Status: Backport [https://gitlab.gnome.org/GNOME/gdk-pixbuf/-/commit/bdd3acbd48a575d418ba6bf1b32d7bda2fae1c81]
+CVE: CVE-2020-29385
+Signed-off-by: Chee Yang Lee <[email protected]>
+
+---
+ gdk-pixbuf/lzw.c                    |  13 +++++++------
+ 1 files changed, 7 insertions(+), 6 deletions(-)
+ create mode 100644 tests/test-images/fail/hang_114.gif
+
+diff --git a/gdk-pixbuf/lzw.c b/gdk-pixbuf/lzw.c
+index 9e052a6f7..105daf2b1 100644
+--- a/gdk-pixbuf/lzw.c
++++ b/gdk-pixbuf/lzw.c
+@@ -195,19 +195,20 @@ lzw_decoder_feed (LZWDecoder *self,
+                                 if (self->last_code != self->clear_code && self->code_table_size < MAX_CODES) {
+                                         if (self->code < self->code_table_size)
+                                                 add_code (self, self->code);
+-                                        else if (self->code == self->code_table_size)
++                                        else
+                                                 add_code (self, self->last_code);
+-                                        else {
+-                                                /* Invalid code received - just stop here */
+-                                                self->last_code = self->eoi_code;
+-                                                return output_length;
+-                                        }
+ 
+                                         /* When table is full increase code size */
+                                         if (self->code_table_size == (1 << self->code_size) && self->code_size < LZW_CODE_MAX)
+                                                 self->code_size++;
+                                 }
+ 
++                                /* Invalid code received - just stop here */
++                                if (self->code >= self->code_table_size) {
++                                        self->last_code = self->eoi_code;
++                                        return output_length;
++                                }
++
+                                 /* Convert codeword into indexes */
+                                 n_written += write_indexes (self, output + n_written, output_length - n_written);
+                         }
+-- 
+2.25.1
+
meta/recipes-gnome/gdk-pixbuf/gdk-pixbuf_2.40.0.bb +1 -0
index 0405fa78b5..54861e83c6 100644
@@ -24,6 +24,7 @@ SRC_URI = "${GNOME_MIRROR}/${BPN}/${MAJ_VER}/${BPN}-${PV}.tar.xz \
            file://0004-Do-not-run-tests-when-building.patch \
            file://0006-Build-thumbnailer-and-tests-also-in-cross-builds.patch \
            file://missing-test-data.patch \
+           file://CVE-2020-29385.patch \
            "
 
 SRC_URI_append_class-target = " \
Signature · 1 line
2.17.1
#29 lore raw

PATCH 29/30 grub: fix "CVE:" line in one of the patches

In reply to Steve Sakoman

From: Scott Murray <[email protected]>

The "CVE:" line in the patch for CVEs 2020-14309, CVE-2020-14310, and CVE-2020-14311 had commas between the CVE numbers, which resulted in CVE-2020-14310 not being picked up as patched by cve-check.bbclass's parsing. Remove the commas to match cve-check.bbclass's expectations.

(cherry picked from commit 396d5c0f9cffa4b54ae94738b1ef2b6fb545f082)
---
 ...1-malloc-Use-overflow-checking-primitives-where-we-do-.patch | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
meta/recipes-bsp/grub/files/CVE-2020-14309-CVE-2020-14310-CVE-2020-14311-malloc-Use-overflow-checking-primitives-where-we-do-.patch +1 -1
index 896a2145d4..7214ead9a7 100644
@@ -30,7 +30,7 @@ Signed-off-by: Peter Jones <[email protected]>
 Reviewed-by: Daniel Kiper <[email protected]>
 
 Upstream-Status: Backport
-CVE: CVE-2020-14309, CVE-2020-14310, CVE-2020-14311
+CVE: CVE-2020-14309 CVE-2020-14310 CVE-2020-14311
 
 Reference to upstream patch:
 https://git.savannah.gnu.org/cgit/grub.git/commit/?id=3f05d693d1274965ffbe4ba99080dc2c570944c6
Signature · 1 line
2.17.1
#30 lore raw

PATCH 30/30 patch: fix CVE-2019-20633

In reply to Steve Sakoman

From: Scott Murray <[email protected]>

* CVE detail: https://nvd.nist.gov/vuln/detail/CVE-2019-20633

* upstream tracking: https://savannah.gnu.org/bugs/index.php?56683

* Fixes potential for double free after incomplete fix for CVE-2018-6952
- src/pch.c (another_hunk): Avoid invalid memory access in context format diffs.

(cherry picked from commit be71dd2cc16a4c0d244a76a748f08ca0d9bfeba0)
---
 .../patch/patch/CVE-2019-20633.patch          | 31 +++++++++++++++++++
 meta/recipes-devtools/patch/patch_2.7.6.bb    |  1 +
 2 files changed, 32 insertions(+)
 create mode 100644 meta/recipes-devtools/patch/patch/CVE-2019-20633.patch
meta/recipes-devtools/patch/patch/CVE-2019-20633.patch new file +31 -0
new file mode 100644
index 0000000000..03988a179c
@@ -0,0 +1,31 @@
+From 15b158db3ae11cb835f2eb8d2eb48e09d1a4af48 Mon Sep 17 00:00:00 2001
+From: Andreas Gruenbacher <[email protected]>
+Date: Mon, 15 Jul 2019 19:10:02 +0200
+Subject: Avoid invalid memory access in context format diffs
+
+* src/pch.c (another_hunk): Avoid invalid memory access in context format
+diffs.
+
+CVE: CVE-2019-20633
+Upstream-Status: Backport[https://git.savannah.gnu.org/cgit/patch.git/patch/?id=15b158db3ae11cb835f2eb8d2eb48e09d1a4af48]
+Signed-off-by: Scott Murray <[email protected]>
+
+---
+ src/pch.c | 1 +
+ 1 file changed, 1 insertion(+)
+
+diff --git a/src/pch.c b/src/pch.c
+index a500ad9..cb54e03 100644
+--- a/src/pch.c
++++ b/src/pch.c
+@@ -1328,6 +1328,7 @@ another_hunk (enum diff difftype, bool rev)
+ 		  ptrn_prefix_context = context;
+ 		ptrn_suffix_context = context;
+ 		if (repl_beginning
++		    || p_end <= 0
+ 		    || (p_end
+ 			!= p_ptrn_lines + 1 + (p_Char[p_end - 1] == '\n')))
+ 		  {
+-- 
+cgit v1.2.1
+
meta/recipes-devtools/patch/patch_2.7.6.bb +1 -0
index b5897b357a..1997af0c25 100644
@@ -10,6 +10,7 @@ SRC_URI += "file://0001-Unset-need_charset_alias-when-building-for-musl.patch \
             file://0001-Invoke-ed-directly-instead-of-using-the-shell.patch \
             file://0001-Don-t-leak-temporary-file-on-failed-ed-style-patch.patch \
             file://0001-Don-t-leak-temporary-file-on-failed-multi-file-ed.patch \
+            file://CVE-2019-20633.patch \
 "
 
 SRC_URI[md5sum] = "4c68cee989d83c87b00a3860bcd05600"
Signature · 1 line
2.17.1