review request
cover review request
Please review these changes for hardknott. Builds cleanly on autobuilder except for one intermittent and unrelated change.
Thanks,
Anuj
The following changes since commit 3357bbf0dad31306d5e16ad306d3e931042eec61:
go: Use dl.google.com for SRC_URI (2021-04-26 08:44:53 +0800)
are available in the Git repository at:
git://push.openembedded.org/openembedded-core-contrib anujm/hardknott
Alexander Kanavin (2):
oeqa: tear down oeqa decorators if one of them raises an exception in
setup
meta/lib/oeqa/core/tests/cases/timeout.py: add a testcase for the
previous fix
Chen Qi (3):
glib-2.0: fix CVE-2021-28153
weston: fix build failure due to race condition
rsync: fix CVE-2020-14387
Kai Kang (2):
kernel-yocto.bbclass: chdir to ${WORKDIR} for do_kernel_checkout cmake.bbclass: remove ${B} before cmake_do_configure
Richard Purdie (2):
yocto-check-layer: Avoid bug when iterating and autoadding
dependencies
patchelf: Backport fix from upstream for note section overlap error
Stefan Ghinea (1):
libssh2: fix build failure with option no-ecdsa
meta/classes/cmake.bbclass | 8 +- meta/classes/kernel-yocto.bbclass | 2 +- meta/lib/oeqa/core/case.py | 9 +- meta/lib/oeqa/core/decorator/oetimeout.py | 5 +- meta/lib/oeqa/core/tests/cases/timeout.py | 13 + meta/lib/oeqa/core/tests/test_decorators.py | 6 + ...outputstream-Fix-a-typo-in-a-comment.patch | 32 ++ ...-using-g_test_bug_base-in-file-tests.patch | 47 +++ ...outputstream-Factor-out-a-flag-check.patch | 60 ++++ ...stream-Fix-CREATE_REPLACE_DESTINATIO.patch | 294 ++++++++++++++++++ ...stream-Add-a-missing-O_CLOEXEC-flag-.patch | 60 ++++ meta/recipes-core/glib-2.0/glib-2.0_2.66.7.bb | 10 + ...c83653ce1b5fc201ff6db93b966394766814.patch | 44 +++ .../patchelf/patchelf_0.12.bb | 1 + ...-the-hostname-in-the-certificate-whe.patch | 31 ++ meta/recipes-devtools/rsync/rsync_3.2.3.bb | 1 + ...001-meson.build-fix-incorrect-header.patch | 32 ++ meta/recipes-graphics/wayland/weston_9.0.0.bb | 1 + ...EC-macro-outside-of-if-check-549-550.patch | 112 +++++++ meta/recipes-support/libssh2/libssh2_1.9.0.bb | 1 + scripts/yocto-check-layer | 3 + 21 files changed, 762 insertions(+), 10 deletions(-) create mode 100644 meta/recipes-core/glib-2.0/glib-2.0/0001-glocalfileoutputstream-Fix-a-typo-in-a-comment.patch create mode 100644 meta/recipes-core/glib-2.0/glib-2.0/0002-tests-Stop-using-g_test_bug_base-in-file-tests.patch create mode 100644 meta/recipes-core/glib-2.0/glib-2.0/0003-glocalfileoutputstream-Factor-out-a-flag-check.patch create mode 100644 meta/recipes-core/glib-2.0/glib-2.0/0004-glocalfileoutputstream-Fix-CREATE_REPLACE_DESTINATIO.patch create mode 100644 meta/recipes-core/glib-2.0/glib-2.0/0005-glocalfileoutputstream-Add-a-missing-O_CLOEXEC-flag-.patch create mode 100644 meta/recipes-devtools/patchelf/patchelf/6edec83653ce1b5fc201ff6db93b966394766814.patch create mode 100644 meta/recipes-devtools/rsync/files/0001-rsync-ssl-Verify-the-hostname-in-the-certificate-whe.patch create mode 100644 meta/recipes-graphics/wayland/weston/0001-meson.build-fix-incorrect-header.patch create mode 100644 meta/recipes-support/libssh2/files/0001-kex.c-move-EC-macro-outside-of-if-check-549-550.patch
Signature · 1 line
2.30.2
PATCH 1/10 glib-2.0: fix CVE-2021-28153
From: Chen Qi <[email protected]>
Backport patches to fix CVE-2021-28153.
--- ...outputstream-Fix-a-typo-in-a-comment.patch | 32 ++ ...-using-g_test_bug_base-in-file-tests.patch | 47 +++ ...outputstream-Factor-out-a-flag-check.patch | 60 ++++ ...stream-Fix-CREATE_REPLACE_DESTINATIO.patch | 294 ++++++++++++++++++ ...stream-Add-a-missing-O_CLOEXEC-flag-.patch | 60 ++++ meta/recipes-core/glib-2.0/glib-2.0_2.66.7.bb | 10 + 6 files changed, 503 insertions(+) create mode 100644 meta/recipes-core/glib-2.0/glib-2.0/0001-glocalfileoutputstream-Fix-a-typo-in-a-comment.patch create mode 100644 meta/recipes-core/glib-2.0/glib-2.0/0002-tests-Stop-using-g_test_bug_base-in-file-tests.patch create mode 100644 meta/recipes-core/glib-2.0/glib-2.0/0003-glocalfileoutputstream-Factor-out-a-flag-check.patch create mode 100644 meta/recipes-core/glib-2.0/glib-2.0/0004-glocalfileoutputstream-Fix-CREATE_REPLACE_DESTINATIO.patch create mode 100644 meta/recipes-core/glib-2.0/glib-2.0/0005-glocalfileoutputstream-Add-a-missing-O_CLOEXEC-flag-.patch
meta/recipes-core/glib-2.0/glib-2.0/0001-glocalfileoutputstream-Fix-a-typo-in-a-comment.patch new file +32 -0
@@ -0,0 +1,32 @@ +From 48dd0d030a2b5240457472d40d8691b80bf5fa78 Mon Sep 17 00:00:00 2001 +From: Philip Withnall <[email protected]> +Date: Wed, 24 Feb 2021 17:33:38 +0000 +Subject: [PATCH 1/5] glocalfileoutputstream: Fix a typo in a comment + +Signed-off-by: Philip Withnall <[email protected]> + +CVE: CVE-2021-28153 + +Upstream-Status: Backport [https://gitlab.gnome.org/GNOME/glib/-/issues/2325] + +Signed-off-by: Chen Qi <[email protected]> +--- + gio/glocalfileoutputstream.c | 2 +- + 1 file changed, 1 insertion(+), 1 deletion(-) + +diff --git a/gio/glocalfileoutputstream.c b/gio/glocalfileoutputstream.c +index f34c3e4..e3d31d6 100644 +--- a/gio/glocalfileoutputstream.c ++++ b/gio/glocalfileoutputstream.c +@@ -854,7 +854,7 @@ handle_overwrite_open (const char *filename, + mode = mode_from_flags_or_info (flags, reference_info); + + /* We only need read access to the original file if we are creating a backup. +- * We also add O_CREATE to avoid a race if the file was just removed */ ++ * We also add O_CREAT to avoid a race if the file was just removed */ + if (create_backup || readable) + open_flags = O_RDWR | O_CREAT | O_BINARY; + else +-- +2.17.1 +
meta/recipes-core/glib-2.0/glib-2.0/0002-tests-Stop-using-g_test_bug_base-in-file-tests.patch new file +47 -0
47 lines not on this page · raw message
meta/recipes-core/glib-2.0/glib-2.0/0003-glocalfileoutputstream-Factor-out-a-flag-check.patch new file +60 -0
60 lines not on this page · raw message
meta/recipes-core/glib-2.0/glib-2.0/0004-glocalfileoutputstream-Fix-CREATE_REPLACE_DESTINATIO.patch new file +294 -0
294 lines not on this page · raw message
meta/recipes-core/glib-2.0/glib-2.0/0005-glocalfileoutputstream-Add-a-missing-O_CLOEXEC-flag-.patch new file +60 -0
60 lines not on this page · raw message
meta/recipes-core/glib-2.0/glib-2.0_2.66.7.bb +10 -0
@@ -50,6 +50,16 @@ SRC_URI += "\ file://0028-gresource-Fix-a-pointer-mismatch-with-an-atomic-load.patch \ file://0029-docs-Document-not-to-use-volatile-qualifiers.patch \ " + +# Fix CVE-2021-28153 +SRC_URI += "\ + file://0001-glocalfileoutputstream-Fix-a-typo-in-a-comment.patch \ + file://0002-tests-Stop-using-g_test_bug_base-in-file-tests.patch \ + file://0003-glocalfileoutputstream-Factor-out-a-flag-check.patch \ + file://0004-glocalfileoutputstream-Fix-CREATE_REPLACE_DESTINATIO.patch \ + file://0005-glocalfileoutputstream-Add-a-missing-O_CLOEXEC-flag-.patch \ +" + SRC_URI_append_class-native = " file://relocate-modules.patch" SRC_URI[sha256sum] = "09f158769f6f26b31074e15b1ac80ec39b13b53102dfae66cfe826fb2cc65502"
Signature · 1 line
2.30.2
PATCH 2/10 libssh2: fix build failure with option no-ecdsa
In reply to Anuj MittalFrom: Stefan Ghinea <[email protected]>
libssh2 fails at do_compile if DEPRECATED_CRYPTO_FLAGS = "no-ecdsa" is set in recipe:
../src/.libs/libssh2.so: undefined reference to
`LIBSSH2_KEX_METHOD_EC_SHA_HASH_CREATE_VERIFY'
References:
https://github.com/libssh2/libssh2/issues/549
Upstream patches:
https://github.com/libssh2/libssh2/commit/1f76151c92e1b52e9c24ebf06adc77fbd6c062bc
--- ...EC-macro-outside-of-if-check-549-550.patch | 112 ++++++++++++++++++ meta/recipes-support/libssh2/libssh2_1.9.0.bb | 1 + 2 files changed, 113 insertions(+) create mode 100644 meta/recipes-support/libssh2/files/0001-kex.c-move-EC-macro-outside-of-if-check-549-550.patch
meta/recipes-support/libssh2/files/0001-kex.c-move-EC-macro-outside-of-if-check-549-550.patch new file +112 -0
@@ -0,0 +1,112 @@ +From 1f76151c92e1b52e9c24ebf06adc77fbd6c062bc Mon Sep 17 00:00:00 2001 +From: Will Cosgrove <[email protected]> +Date: Tue, 26 Jan 2021 11:41:21 -0800 +Subject: [PATCH] kex.c: move EC macro outside of if check #549 (#550) + +File: kex.c + +Notes: +Moved the macro LIBSSH2_KEX_METHOD_EC_SHA_HASH_CREATE_VERIFY outside of the LIBSSH2_ECDSA since it's also now used by the ED25519 code. + +Sha 256, 384 and 512 need to be defined for all backends now even if they aren't used directly. I believe this is already the case, but just a heads up. + +Credit: +Stefan-Ghinea + +Upstream-Status: Backport + +Reference to upstream patch: +https://github.com/libssh2/libssh2/commit/1f76151c92e1b52e9c24ebf06adc77fbd6c062bc + +Signed-off-by: Stefan Ghinea <[email protected]> +--- + src/kex.c | 66 +++++++++++++++++++++++++++---------------------------- + 1 file changed, 33 insertions(+), 33 deletions(-) + +diff --git a/src/kex.c b/src/kex.c +index cb16639..19ab6ec 100644 +--- a/src/kex.c ++++ b/src/kex.c +@@ -1885,39 +1885,6 @@ kex_method_diffie_hellman_group_exchange_sha256_key_exchange + } + + +-#if LIBSSH2_ECDSA +- +-/* kex_session_ecdh_curve_type +- * returns the EC curve type by name used in key exchange +- */ +- +-static int +-kex_session_ecdh_curve_type(const char *name, libssh2_curve_type *out_type) +-{ +- int ret = 0; +- libssh2_curve_type type; +- +- if(name == NULL) +- return -1; +- +- if(strcmp(name, "ecdh-sha2-nistp256") == 0) +- type = LIBSSH2_EC_CURVE_NISTP256; +- else if(strcmp(name, "ecdh-sha2-nistp384") == 0) +- type = LIBSSH2_EC_CURVE_NISTP384; +- else if(strcmp(name, "ecdh-sha2-nistp521") == 0) +- type = LIBSSH2_EC_CURVE_NISTP521; +- else { +- ret = -1; +- } +- +- if(ret == 0 && out_type) { +- *out_type = type; +- } +- +- return ret; +-} +- +- + /* LIBSSH2_KEX_METHOD_EC_SHA_HASH_CREATE_VERIFY + * + * Macro that create and verifies EC SHA hash with a given digest bytes +@@ -2027,6 +1994,39 @@ kex_session_ecdh_curve_type(const char *name, libssh2_curve_type *out_type) + } \ + + ++#if LIBSSH2_ECDSA ++ ++/* kex_session_ecdh_curve_type ++ * returns the EC curve type by name used in key exchange ++ */ ++ ++static int ++kex_session_ecdh_curve_type(const char *name, libssh2_curve_type *out_type) ++{ ++ int ret = 0; ++ libssh2_curve_type type; ++ ++ if(name == NULL) ++ return -1; ++ ++ if(strcmp(name, "ecdh-sha2-nistp256") == 0) ++ type = LIBSSH2_EC_CURVE_NISTP256; ++ else if(strcmp(name, "ecdh-sha2-nistp384") == 0) ++ type = LIBSSH2_EC_CURVE_NISTP384; ++ else if(strcmp(name, "ecdh-sha2-nistp521") == 0) ++ type = LIBSSH2_EC_CURVE_NISTP521; ++ else { ++ ret = -1; ++ } ++ ++ if(ret == 0 && out_type) { ++ *out_type = type; ++ } ++ ++ return ret; ++} ++ ++ + /* ecdh_sha2_nistp + * Elliptic Curve Diffie Hellman Key Exchange + */ +-- +2.17.1 +
meta/recipes-support/libssh2/libssh2_1.9.0.bb +1 -0
@@ -11,6 +11,7 @@ SRC_URI = "http://www.libssh2.org/download/${BP}.tar.gz \ file://CVE-2019-17498.patch \ file://0001-configure-Conditionally-undefine-backend-m4-macro.patch \ file://run-ptest \ + file://0001-kex.c-move-EC-macro-outside-of-if-check-549-550.patch \ " SRC_URI_append_ptest = " file://0001-Don-t-let-host-enviroment-to-decide-if-a-test-is-bui.patch"
Signature · 1 line
2.30.2
PATCH 3/10 yocto-check-layer: Avoid bug when iterating and autoadding dependencies
In reply to Anuj MittalFrom: Richard Purdie <[email protected]>
If iterating a layer with multiple components and auto-adding dependencies the tests can break since layers are never removed and order isn't guaranteed to account for that.
Fix this by resetting the layer list back to the original list each time before auto-adding the dependencies in each case.
This fixes scanning of meta-openembedded in particular where the sublayers may not be added in order of minimal dependency.
--- scripts/yocto-check-layer | 3 +++ 1 file changed, 3 insertions(+)
scripts/yocto-check-layer +3 -0
@@ -138,6 +138,9 @@ def main(): layer['type'] == LayerType.ERROR_BSP_DISTRO: continue + # Reset to a clean backup copy for each run + shutil.copyfile(bblayersconf + '.backup', bblayersconf) + if check_bblayers(bblayersconf, layer['path'], logger): logger.info("%s already in %s. To capture initial signatures, layer under test should not present " "in BBLAYERS. Please remove %s from BBLAYERS." % (layer['name'], bblayersconf, layer['name']))
Signature · 1 line
2.30.2
PATCH 4/10 kernel-yocto.bbclass: chdir to ${WORKDIR} for do_kernel_checkout
In reply to Anuj MittalFrom: Kai Kang <[email protected]>
It chdirs to ${S} at the beginning of task do_kernel_checkout. Then it removes ${S} when it still resides in ${S}. It may fail to run the task do_kernel_checkout when bitbake is called by third-part wrapper script. So chdir to ${WORKDIR} by default for do_kernel_checkout. And it will chdir to ${S} afterwards in task do_kernel_checkout.
--- meta/classes/kernel-yocto.bbclass | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
meta/classes/kernel-yocto.bbclass +1 -1
@@ -378,7 +378,7 @@ do_kernel_checkout() { # checkout and clobber any unimportant files git checkout -f ${machine_branch} } -do_kernel_checkout[dirs] = "${S}" +do_kernel_checkout[dirs] = "${S} ${WORKDIR}" addtask kernel_checkout before do_kernel_metadata after do_symlink_kernsrc addtask kernel_metadata after do_validate_branches do_unpack before do_patch
Signature · 1 line
2.30.2
PATCH 5/10 cmake.bbclass: remove ${B} before cmake_do_configure
In reply to Anuj MittalFrom: Kai Kang <[email protected]>
It is fallible to remove ${B} in directory ${B} itself. And it does fail when call bitbake by third-party wrapper script.
Use flag 'cleandirs' to remove ${B} first if build out of source tree.
--- meta/classes/cmake.bbclass | 8 +++----- 1 file changed, 3 insertions(+), 5 deletions(-)
meta/classes/cmake.bbclass +3 -5
@@ -149,16 +149,14 @@ addtask generate_toolchain_file after do_patch before do_configure CONFIGURE_FILES = "CMakeLists.txt" +do_configure[cleandirs] = "${@d.getVar('B') if d.getVar('S') != d.getVar('B') else ''}" + cmake_do_configure() { if [ "${OECMAKE_BUILDPATH}" ]; then bbnote "cmake.bbclass no longer uses OECMAKE_BUILDPATH. The default behaviour is now out-of-tree builds with B=WORKDIR/build." fi - if [ "${S}" != "${B}" ]; then - rm -rf ${B} - mkdir -p ${B} - cd ${B} - else + if [ "${S}" = "${B}" ]; then find ${B} -name CMakeFiles -or -name Makefile -or -name cmake_install.cmake -or -name CMakeCache.txt -delete fi
Signature · 1 line
2.30.2
PATCH 6/10 oeqa: tear down oeqa decorators if one of them raises an exception in setup
In reply to Anuj MittalFrom: Alexander Kanavin <[email protected]>
Some of the decorators need proper cleanup, such as OETimeout
which sets a signal handler that needs to be cleared via teardown. If this is not done then the signal gets called later with unpredictable effects.
This can be seen if there's a test that is skipped via a decorator and sets a timeout at the same time: the timeout isn't cleared, and is invoked later in a completely unrelated context. The test case for this is added in the next commit.
--- meta/lib/oeqa/core/case.py | 9 +++++++-- meta/lib/oeqa/core/decorator/oetimeout.py | 5 +++-- 2 files changed, 10 insertions(+), 4 deletions(-)
meta/lib/oeqa/core/case.py +7 -2
@@ -43,8 +43,13 @@ class OETestCase(unittest.TestCase): clss.tearDownClassMethod() def _oeSetUp(self): - for d in self.decorators: - d.setUpDecorator() + try: + for d in self.decorators: + d.setUpDecorator() + except: + for d in self.decorators: + d.tearDownDecorator() + raise self.setUpMethod() def _oeTearDown(self):
meta/lib/oeqa/core/decorator/oetimeout.py +3 -2
@@ -24,5 +24,6 @@ class OETimeout(OETestDecorator): def tearDownDecorator(self): signal.alarm(0) - signal.signal(signal.SIGALRM, self.alarmSignal) - self.logger.debug("Removed SIGALRM handler") + if hasattr(self, 'alarmSignal'): + signal.signal(signal.SIGALRM, self.alarmSignal) + self.logger.debug("Removed SIGALRM handler")
Signature · 1 line
2.30.2
PATCH 7/10 meta/lib/oeqa/core/tests/cases/timeout.py: add a testcase for the previous fix
In reply to Anuj MittalFrom: Alexander Kanavin <[email protected]>
This is the sequence that didn't properly operate:
- a test case that skips and isn't executed
- a second test case that is skipped via a dependency decorator, and sets a timeout
- a third test case that takes longer than the timeout from the second test case
Without the fix, the timeout is not cleared, and the third test case is erroneously aborted. With the fix, the timeout is cleared and the third test case is able to complete.
--- meta/lib/oeqa/core/tests/cases/timeout.py | 13 +++++++++++++ meta/lib/oeqa/core/tests/test_decorators.py | 6 ++++++ 2 files changed, 19 insertions(+)
meta/lib/oeqa/core/tests/cases/timeout.py +13 -0
@@ -8,6 +8,7 @@ from time import sleep from oeqa.core.case import OETestCase from oeqa.core.decorator.oetimeout import OETimeout +from oeqa.core.decorator.depends import OETestDepends class TimeoutTest(OETestCase):
@@ -19,3 +20,15 @@ class TimeoutTest(OETestCase): def testTimeoutFail(self): sleep(2) self.assertTrue(True, msg='How is this possible?') + + + def testTimeoutSkip(self): + self.skipTest("This test needs to be skipped, so that testTimeoutDepends()'s OETestDepends kicks in") + + @OETestDepends(["timeout.TimeoutTest.testTimeoutSkip"]) + @OETimeout(3) + def testTimeoutDepends(self): + self.assertTrue(False, msg='How is this possible?') + + def testTimeoutUnrelated(self): + sleep(6)
meta/lib/oeqa/core/tests/test_decorators.py +6 -0
@@ -133,5 +133,11 @@ class TestTimeoutDecorator(TestBase): msg = "OETestTimeout didn't restore SIGALRM" self.assertIs(alarm_signal, signal.getsignal(signal.SIGALRM), msg=msg) + def test_timeout_cancel(self): + tests = ['timeout.TimeoutTest.testTimeoutSkip', 'timeout.TimeoutTest.testTimeoutDepends', 'timeout.TimeoutTest.testTimeoutUnrelated'] + msg = 'Unrelated test failed to complete' + tc = self._testLoader(modules=self.modules, tests=tests) + self.assertTrue(tc.runTests().wasSuccessful(), msg=msg) + if __name__ == '__main__': unittest.main()
Signature · 1 line
2.30.2
PATCH 8/10 weston: fix build failure due to race condition
In reply to Anuj MittalFrom: Chen Qi <[email protected]>
The wayland.c actually include 'xdg-shell-client-protocol.h' instead of the server one, so fix it. Otherwise, it's possible to get build failure due to race condition.
--- ...001-meson.build-fix-incorrect-header.patch | 32 +++++++++++++++++++ meta/recipes-graphics/wayland/weston_9.0.0.bb | 1 + 2 files changed, 33 insertions(+) create mode 100644 meta/recipes-graphics/wayland/weston/0001-meson.build-fix-incorrect-header.patch
meta/recipes-graphics/wayland/weston/0001-meson.build-fix-incorrect-header.patch new file +32 -0
@@ -0,0 +1,32 @@ +From a2ba4714a6872e547621d29d9ddcb0f374b88cf6 Mon Sep 17 00:00:00 2001 +From: Chen Qi <[email protected]> +Date: Tue, 20 Apr 2021 20:42:18 -0700 +Subject: [PATCH] meson.build: fix incorrect header + +The wayland.c actually include 'xdg-shell-client-protocol.h' instead of +the server one, so fix it. Otherwise, it's possible to get build failure +due to race condition. + +Upstream-Status: Pending + +Signed-off-by: Chen Qi <[email protected]> +--- + libweston/backend-wayland/meson.build | 2 +- + 1 file changed, 1 insertion(+), 1 deletion(-) + +diff --git a/libweston/backend-wayland/meson.build b/libweston/backend-wayland/meson.build +index 7e82513..29270b5 100644 +--- a/libweston/backend-wayland/meson.build ++++ b/libweston/backend-wayland/meson.build +@@ -10,7 +10,7 @@ srcs_wlwl = [ + fullscreen_shell_unstable_v1_protocol_c, + presentation_time_protocol_c, + presentation_time_server_protocol_h, +- xdg_shell_server_protocol_h, ++ xdg_shell_client_protocol_h, + xdg_shell_protocol_c, + ] + +-- +2.30.2 +
meta/recipes-graphics/wayland/weston_9.0.0.bb +1 -0
@@ -11,6 +11,7 @@ SRC_URI = "https://wayland.freedesktop.org/releases/${BPN}-${PV}.tar.xz \ file://xwayland.weston-start \ file://0001-weston-launch-Provide-a-default-version-that-doesn-t.patch \ file://0001-tests-include-fcntl.h-for-open-O_RDWR-O_CLOEXEC-and-.patch \ + file://0001-meson.build-fix-incorrect-header.patch \ " SRC_URI_append_libc-musl = " file://dont-use-plane-add-prop.patch "
Signature · 1 line
2.30.2
PATCH 9/10 patchelf: Backport fix from upstream for note section overlap error
In reply to Anuj MittalFrom: Richard Purdie <[email protected]>
Backport a patch from upstream to fix an error:
patchelf: cannot normalize PT_NOTE segment: non-contiguous SHT_NOTE sections
seen on our ubuntu1604 autobuilder worker.
--- ...c83653ce1b5fc201ff6db93b966394766814.patch | 44 +++++++++++++++++++ .../patchelf/patchelf_0.12.bb | 1 + 2 files changed, 45 insertions(+) create mode 100644 meta/recipes-devtools/patchelf/patchelf/6edec83653ce1b5fc201ff6db93b966394766814.patch
meta/recipes-devtools/patchelf/patchelf/6edec83653ce1b5fc201ff6db93b966394766814.patch new file +44 -0
@@ -0,0 +1,44 @@ +From 6edec83653ce1b5fc201ff6db93b966394766814 Mon Sep 17 00:00:00 2001 +From: rmnull <[email protected]> +Date: Tue, 18 Aug 2020 20:22:52 +0530 +Subject: [PATCH] mark phdrs synced with sections, avoid rechecking it when + syncing note sections to segments. + +This also serves as a bug fix when a previously synced note segment +overlaps with another section and creates a false alarm. + +Upstream-Status: Backport +--- + src/patchelf.cc | 5 ++++- + 1 file changed, 4 insertions(+), 1 deletion(-) + +diff --git a/src/patchelf.cc b/src/patchelf.cc +index 05ec793..622f0b6 100644 +--- a/src/patchelf.cc ++++ b/src/patchelf.cc +@@ -669,6 +669,7 @@ void ElfFile<ElfFileParamNames>::writeReplacedSections(Elf_Off & curOff, + memset(contents + rdi(shdr.sh_offset), 'X', rdi(shdr.sh_size)); + } + ++ std::set<unsigned int> noted_phdrs = {}; + for (auto & i : replacedSections) { + std::string sectionName = i.first; + auto & shdr = findSection(sectionName); +@@ -721,7 +722,7 @@ void ElfFile<ElfFileParamNames>::writeReplacedSections(Elf_Off & curOff, + shdr.sh_addralign = orig_shdr.sh_addralign; + + for (unsigned int j = 0; j < phdrs.size(); ++j) +- if (rdi(phdrs[j].p_type) == PT_NOTE) { ++ if (rdi(phdrs[j].p_type) == PT_NOTE && noted_phdrs.find(j) == noted_phdrs.end()) { + Elf_Off p_start = rdi(phdrs[j].p_offset); + Elf_Off p_end = p_start + rdi(phdrs[j].p_filesz); + Elf_Off s_start = rdi(orig_shdr.sh_offset); +@@ -739,6 +740,8 @@ void ElfFile<ElfFileParamNames>::writeReplacedSections(Elf_Off & curOff, + phdrs[j].p_offset = shdr.sh_offset; + phdrs[j].p_vaddr = phdrs[j].p_paddr = shdr.sh_addr; + phdrs[j].p_filesz = phdrs[j].p_memsz = shdr.sh_size; ++ ++ noted_phdrs.insert(j); + } + } +
meta/recipes-devtools/patchelf/patchelf_0.12.bb +1 -0
@@ -6,6 +6,7 @@ LICENSE = "GPLv3" SRC_URI = "git://github.com/NixOS/patchelf;protocol=https \ file://handle-read-only-files.patch \ + file://6edec83653ce1b5fc201ff6db93b966394766814.patch \ " SRCREV = "8d3a16e97294e3c5521c61b4c8835499c9918264"
Signature · 1 line
2.30.2
PATCH 10/10 rsync: fix CVE-2020-14387
In reply to Anuj MittalFrom: Chen Qi <[email protected]>
Backport patch to fix CVE-2020-14387.
--- ...-the-hostname-in-the-certificate-whe.patch | 31 +++++++++++++++++++ meta/recipes-devtools/rsync/rsync_3.2.3.bb | 1 + 2 files changed, 32 insertions(+) create mode 100644 meta/recipes-devtools/rsync/files/0001-rsync-ssl-Verify-the-hostname-in-the-certificate-whe.patch
meta/recipes-devtools/rsync/files/0001-rsync-ssl-Verify-the-hostname-in-the-certificate-whe.patch new file +31 -0
@@ -0,0 +1,31 @@ +From fbe85634d88e82fbb439ae2a5d1aca8b8c309bea Mon Sep 17 00:00:00 2001 +From: Matt McCutchen <[email protected]> +Date: Wed, 26 Aug 2020 12:16:08 -0400 +Subject: [PATCH] rsync-ssl: Verify the hostname in the certificate when using + openssl. + +CVE: CVE-2020-14387 + +Upstream-Status: Backport [https://git.samba.org/?p=rsync.git;a=commit;h=c3f7414] + +Signed-off-by: Chen Qi <[email protected]> +--- + rsync-ssl | 2 +- + 1 file changed, 1 insertion(+), 1 deletion(-) + +diff --git a/rsync-ssl b/rsync-ssl +index 8101975..46701af 100755 +--- a/rsync-ssl ++++ b/rsync-ssl +@@ -129,7 +129,7 @@ function rsync_ssl_helper { + fi + + if [[ $RSYNC_SSL_TYPE == openssl ]]; then +- exec $RSYNC_SSL_OPENSSL s_client $caopt $certopt -quiet -verify_quiet -servername $hostname -connect $hostname:$port ++ exec $RSYNC_SSL_OPENSSL s_client $caopt $certopt -quiet -verify_quiet -servername $hostname -verify_hostname $hostname -connect $hostname:$port + elif [[ $RSYNC_SSL_TYPE == gnutls ]]; then + exec $RSYNC_SSL_GNUTLS --logfile=/dev/null $gnutls_cert_opt $gnutls_opts $hostname:$port + else +-- +2.17.1 +
meta/recipes-devtools/rsync/rsync_3.2.3.bb +1 -0
@@ -14,6 +14,7 @@ SRC_URI = "https://download.samba.org/pub/${BPN}/src/${BP}.tar.gz \ file://rsyncd.conf \ file://makefile-no-rebuild.patch \ file://determism.patch \ + file://0001-rsync-ssl-Verify-the-hostname-in-the-certificate-whe.patch \ " SRC_URI[sha256sum] = "becc3c504ceea499f4167a260040ccf4d9f2ef9499ad5683c179a697146ce50e"
Signature · 1 line
2.30.2