Patch review
cover Patch review
Please review this set of changes for kirkstone and have comments back by end of day Thursday, December 14
Passed a-full on autobuilder:
https://autobuilder.yoctoproject.org/typhoon/#/builders/83/builds/6324
The following changes since commit 09ecafaf0e128c4dea062d359de37cbef461aed2:
native: Clear TUNE_FEATURES/ABIEXTENSION (2023-12-07 08:09:37 -1000)
are available in the Git repository at:
https://git.openembedded.org/openembedded-core-contrib stable/kirkstone-nut
https://git.openembedded.org/openembedded-core-contrib/log/?h=stable/kirkstone-nut
Alexander Kanavin (1):
gstreamer1.0-plugins-base: enable glx/opengl support
Archana Polampalli (1):
bluez5: fix CVE-2023-45866
Mikko Rapeli (1):
openssh: drop sudo from ptest dependencies
Vijay Anusuri (2):
avahi: backport CVE-2023-1981 & CVE's follow-up patches
gnutls: Backport fix for CVE-2023-5981
meta/recipes-connectivity/avahi/avahi_0.8.bb | 10 +-
.../avahi/files/CVE-2023-1981.patch | 58 +++++
...023-38469.patch => CVE-2023-38469-1.patch} | 0
.../avahi/files/CVE-2023-38469-2.patch | 65 ++++++
...023-38470.patch => CVE-2023-38470-1.patch} | 0
.../avahi/files/CVE-2023-38470-2.patch | 52 +++++
...023-38471.patch => CVE-2023-38471-1.patch} | 0
.../avahi/files/CVE-2023-38471-2.patch | 52 +++++
.../avahi/files/CVE-2023-38472.patch | 44 ++--
meta/recipes-connectivity/bluez5/bluez5.inc | 1 +
.../bluez5/bluez5/CVE-2023-45866.patch | 56 +++++
.../openssh/openssh/run-ptest | 2 +-
.../openssh/openssh_8.9p1.bb | 2 +-
.../gstreamer1.0-plugins-base_1.20.7.bb | 6 +-
.../gnutls/gnutls/CVE-2023-5981.patch | 206 ++++++++++++++++++
meta/recipes-support/gnutls/gnutls_3.7.4.bb | 1 +
16 files changed, 526 insertions(+), 29 deletions(-)
create mode 100644 meta/recipes-connectivity/avahi/files/CVE-2023-1981.patch
rename meta/recipes-connectivity/avahi/files/{CVE-2023-38469.patch => CVE-2023-38469-1.patch} (100%)
create mode 100644 meta/recipes-connectivity/avahi/files/CVE-2023-38469-2.patch
rename meta/recipes-connectivity/avahi/files/{CVE-2023-38470.patch => CVE-2023-38470-1.patch} (100%)
create mode 100644 meta/recipes-connectivity/avahi/files/CVE-2023-38470-2.patch
rename meta/recipes-connectivity/avahi/files/{CVE-2023-38471.patch => CVE-2023-38471-1.patch} (100%)
create mode 100644 meta/recipes-connectivity/avahi/files/CVE-2023-38471-2.patch
create mode 100644 meta/recipes-connectivity/bluez5/bluez5/CVE-2023-45866.patch
create mode 100644 meta/recipes-support/gnutls/gnutls/CVE-2023-5981.patchSignature · 1 line
2.34.1
PATCH 1/5 avahi: backport CVE-2023-1981 & CVE's follow-up patches
From: Vijay Anusuri <[email protected]>
import patches from ubuntu to fix
CVE-2023-1981
CVE-2023-38469-2
CVE-2023-38470-2
CVE-2023-38471-2
Upstream-Status: Backport [import from ubuntu https://git.launchpad.net/ubuntu/+source/avahi/tree/debian/patches?h=ubuntu/jammy-security
Upstream commit
https://github.com/lathiat/avahi/commit/a2696da2f2c50ac43b6c4903f72290d5c3fa9f6f
&
https://github.com/lathiat/avahi/commit/c6cab87df290448a63323c8ca759baa516166237
&
https://github.com/lathiat/avahi/commit/94cb6489114636940ac683515417990b55b5d66c
&
https://github.com/lathiat/avahi/commit/b675f70739f404342f7f78635d6e2dcd85a13460]
---
meta/recipes-connectivity/avahi/avahi_0.8.bb | 10 ++-
.../avahi/files/CVE-2023-1981.patch | 58 +++++++++++++++++
...023-38469.patch => CVE-2023-38469-1.patch} | 0
.../avahi/files/CVE-2023-38469-2.patch | 65 +++++++++++++++++++
...023-38470.patch => CVE-2023-38470-1.patch} | 0
.../avahi/files/CVE-2023-38470-2.patch | 52 +++++++++++++++
...023-38471.patch => CVE-2023-38471-1.patch} | 0
.../avahi/files/CVE-2023-38471-2.patch | 52 +++++++++++++++
.../avahi/files/CVE-2023-38472.patch | 44 ++++++-------
9 files changed, 256 insertions(+), 25 deletions(-)
create mode 100644 meta/recipes-connectivity/avahi/files/CVE-2023-1981.patch
rename meta/recipes-connectivity/avahi/files/{CVE-2023-38469.patch => CVE-2023-38469-1.patch} (100%)
create mode 100644 meta/recipes-connectivity/avahi/files/CVE-2023-38469-2.patch
rename meta/recipes-connectivity/avahi/files/{CVE-2023-38470.patch => CVE-2023-38470-1.patch} (100%)
create mode 100644 meta/recipes-connectivity/avahi/files/CVE-2023-38470-2.patch
rename meta/recipes-connectivity/avahi/files/{CVE-2023-38471.patch => CVE-2023-38471-1.patch} (100%)
create mode 100644 meta/recipes-connectivity/avahi/files/CVE-2023-38471-2.patchmeta/recipes-connectivity/avahi/avahi_0.8.bb +7 -3
@@ -26,9 +26,13 @@ SRC_URI = "https://github.com/lathiat/avahi/releases/download/v${PV}/avahi-${PV} file://0001-Fix-opening-etc-resolv.conf-error.patch \ file://handle-hup.patch \ file://local-ping.patch \ - file://CVE-2023-38471.patch \ - file://CVE-2023-38470.patch \ - file://CVE-2023-38469.patch \ + file://CVE-2023-1981.patch \ + file://CVE-2023-38469-1.patch \ + file://CVE-2023-38469-2.patch \ + file://CVE-2023-38470-1.patch \ + file://CVE-2023-38470-2.patch \ + file://CVE-2023-38471-1.patch \ + file://CVE-2023-38471-2.patch \ file://CVE-2023-38472.patch \ file://CVE-2023-38473.patch \ "
meta/recipes-connectivity/avahi/files/CVE-2023-1981.patch new file +58 -0
58 lines not on this page · raw message
meta/recipes-connectivity/avahi/files/CVE-2023-38469-1.patch renamed from meta/recipes-connectivity/avahi/files/CVE-2023-38469.patch +0 -0
meta/recipes-connectivity/avahi/files/CVE-2023-38469-2.patch new file +65 -0
65 lines not on this page · raw message
meta/recipes-connectivity/avahi/files/CVE-2023-38470-1.patch renamed from meta/recipes-connectivity/avahi/files/CVE-2023-38470.patch +0 -0
meta/recipes-connectivity/avahi/files/CVE-2023-38470-2.patch new file +52 -0
52 lines not on this page · raw message
meta/recipes-connectivity/avahi/files/CVE-2023-38471-1.patch renamed from meta/recipes-connectivity/avahi/files/CVE-2023-38471.patch +0 -0
meta/recipes-connectivity/avahi/files/CVE-2023-38471-2.patch new file +52 -0
52 lines not on this page · raw message
meta/recipes-connectivity/avahi/files/CVE-2023-38472.patch +22 -22
68 lines not on this page · raw message
Signature · 1 line
2.34.1
PATCH 2/5 bluez5: fix CVE-2023-45866
In reply to Steve SakomanFrom: Archana Polampalli <[email protected]>
Bluetooth HID Hosts in BlueZ may permit an unauthenticated Peripheral role HID Device to initiate and establish an encrypted connection, and accept HID keyboard reports,potentially permitting injection of HID messages when no user interaction has occurred in the Central role to authorize such access. An example affected package is bluez 5.64-0ubuntu1 in Ubuntu 22.04LTS. NOTE: in some cases, a CVE-2020-0556 mitigation would have already addressed this Bluetooth HID Hosts issue.
References:
https://nvd.nist.gov/vuln/detail/CVE-2023-45866
Upstream patches:
https://git.kernel.org/pub/scm/bluetooth/bluez.git/commit/profiles/input?id=25a471a83e02e1effb15d5a488b3f0085eaeb675
--- meta/recipes-connectivity/bluez5/bluez5.inc | 1 + .../bluez5/bluez5/CVE-2023-45866.patch | 56 +++++++++++++++++++ 2 files changed, 57 insertions(+) create mode 100644 meta/recipes-connectivity/bluez5/bluez5/CVE-2023-45866.patch
meta/recipes-connectivity/bluez5/bluez5.inc +1 -0
@@ -54,6 +54,7 @@ SRC_URI = "${KERNELORG_MIRROR}/linux/bluetooth/bluez-${PV}.tar.xz \ ${@bb.utils.contains('DISTRO_FEATURES', 'systemd', '', 'file://0001-Allow-using-obexd-without-systemd-in-the-user-sessio.patch', d)} \ file://0001-tests-add-a-target-for-building-tests-without-runnin.patch \ file://0001-test-gatt-Fix-hung-issue.patch \ + file://CVE-2023-45866.patch \ " S = "${WORKDIR}/bluez-${PV}"
meta/recipes-connectivity/bluez5/bluez5/CVE-2023-45866.patch new file +56 -0
56 lines not on this page · raw message
Signature · 1 line
2.34.1
PATCH 3/5 gnutls: Backport fix for CVE-2023-5981
In reply to Steve SakomanFrom: Vijay Anusuri <[email protected]>
Upstream-Status: Backport [import from debian https://launchpad.net/ubuntu/+archive/primary/+sourcefiles/gnutls28/3.7.3-4ubuntu1.3/gnutls28_3.7.3-4ubuntu1.3.debian.tar.xz
Upstream-Commit: https://gitlab.com/gnutls/gnutls/-/commit/29d6298d0b04cfff970b993915db71ba3f580b6d]
References:
https://ubuntu.com/security/CVE-2023-5981
--- .../gnutls/gnutls/CVE-2023-5981.patch | 206 ++++++++++++++++++ meta/recipes-support/gnutls/gnutls_3.7.4.bb | 1 + 2 files changed, 207 insertions(+) create mode 100644 meta/recipes-support/gnutls/gnutls/CVE-2023-5981.patch
meta/recipes-support/gnutls/gnutls/CVE-2023-5981.patch new file +206 -0
@@ -0,0 +1,206 @@ +Backport of: + +From 29d6298d0b04cfff970b993915db71ba3f580b6d Mon Sep 17 00:00:00 2001 +From: Daiki Ueno <[email protected]> +Date: Mon, 23 Oct 2023 09:26:57 +0900 +Subject: [PATCH] auth/rsa_psk: side-step potential side-channel + +This removes branching that depends on secret data, porting changes +for regular RSA key exchange from +4804febddc2ed958e5ae774de2a8f85edeeff538 and +80a6ce8ddb02477cd724cd5b2944791aaddb702a. This also removes the +allow_wrong_pms as it was used sorely to control debug output +depending on the branching. + +Signed-off-by: Daiki Ueno <[email protected]> + +Upstream-Status: Backport [import from debian https://launchpad.net/ubuntu/+archive/primary/+sourcefiles/gnutls28/3.7.3-4ubuntu1.3/gnutls28_3.7.3-4ubuntu1.3.debian.tar.xz +Upstream-Commit: https://gitlab.com/gnutls/gnutls/-/commit/29d6298d0b04cfff970b993915db71ba3f580b6d] +CVE: CVE-2023-5981 +Signed-off-by: Vijay Anusuri <[email protected]> +--- + lib/auth/rsa.c | 2 +- + lib/auth/rsa_psk.c | 90 ++++++++++++++++++---------------------------- + lib/gnutls_int.h | 4 --- + lib/priority.c | 1 - + 4 files changed, 35 insertions(+), 62 deletions(-) + +--- a/lib/auth/rsa.c ++++ b/lib/auth/rsa.c +@@ -207,7 +207,7 @@ proc_rsa_client_kx(gnutls_session_t sess + session->key.key.size); + /* After this point, any conditional on failure that cause differences + * in execution may create a timing or cache access pattern side +- * channel that can be used as an oracle, so treat very carefully */ ++ * channel that can be used as an oracle, so tread carefully */ + + /* Error handling logic: + * In case decryption fails then don't inform the peer. Just use the +--- a/lib/auth/rsa_psk.c ++++ b/lib/auth/rsa_psk.c +@@ -264,14 +264,13 @@ _gnutls_proc_rsa_psk_client_kx(gnutls_se + { + gnutls_datum_t username; + psk_auth_info_t info; +- gnutls_datum_t plaintext; + gnutls_datum_t ciphertext; + gnutls_datum_t pwd_psk = { NULL, 0 }; + int ret, dsize; +- int randomize_key = 0; + ssize_t data_size = _data_size; + gnutls_psk_server_credentials_t cred; + gnutls_datum_t premaster_secret = { NULL, 0 }; ++ volatile uint8_t ver_maj, ver_min; + + cred = (gnutls_psk_server_credentials_t) + _gnutls_get_cred(session, GNUTLS_CRD_PSK); +@@ -327,71 +326,47 @@ _gnutls_proc_rsa_psk_client_kx(gnutls_se + } + ciphertext.size = dsize; + +- ret = +- gnutls_privkey_decrypt_data(session->internals.selected_key, 0, +- &ciphertext, &plaintext); +- if (ret < 0 || plaintext.size != GNUTLS_MASTER_SIZE) { +- /* In case decryption fails then don't inform +- * the peer. Just use a random key. (in order to avoid +- * attack against pkcs-1 formatting). +- */ +- gnutls_assert(); +- _gnutls_debug_log +- ("auth_rsa_psk: Possible PKCS #1 format attack\n"); +- if (ret >= 0) { +- gnutls_free(plaintext.data); +- } +- randomize_key = 1; +- } else { +- /* If the secret was properly formatted, then +- * check the version number. +- */ +- if (_gnutls_get_adv_version_major(session) != +- plaintext.data[0] +- || (session->internals.allow_wrong_pms == 0 +- && _gnutls_get_adv_version_minor(session) != +- plaintext.data[1])) { +- /* No error is returned here, if the version number check +- * fails. We proceed normally. +- * That is to defend against the attack described in the paper +- * "Attacking RSA-based sessions in SSL/TLS" by Vlastimil Klima, +- * Ondej Pokorny and Tomas Rosa. +- */ +- gnutls_assert(); +- _gnutls_debug_log +- ("auth_rsa: Possible PKCS #1 version check format attack\n"); +- } +- } ++ ver_maj = _gnutls_get_adv_version_major(session); ++ ver_min = _gnutls_get_adv_version_minor(session); + ++ premaster_secret.data = gnutls_malloc(GNUTLS_MASTER_SIZE); ++ if (premaster_secret.data == NULL) { ++ gnutls_assert(); ++ return GNUTLS_E_MEMORY_ERROR; ++ } ++ premaster_secret.size = GNUTLS_MASTER_SIZE; + +- if (randomize_key != 0) { +- premaster_secret.size = GNUTLS_MASTER_SIZE; +- premaster_secret.data = +- gnutls_malloc(premaster_secret.size); +- if (premaster_secret.data == NULL) { +- gnutls_assert(); +- return GNUTLS_E_MEMORY_ERROR; +- } +- +- /* we do not need strong random numbers here. +- */ +- ret = gnutls_rnd(GNUTLS_RND_NONCE, premaster_secret.data, +- premaster_secret.size); +- if (ret < 0) { +- gnutls_assert(); +- goto cleanup; +- } +- } else { +- premaster_secret.data = plaintext.data; +- premaster_secret.size = plaintext.size; ++ /* Fallback value when decryption fails. Needs to be unpredictable. */ ++ ret = gnutls_rnd(GNUTLS_RND_NONCE, premaster_secret.data, ++ premaster_secret.size); ++ if (ret < 0) { ++ gnutls_assert(); ++ goto cleanup; + } + ++ gnutls_privkey_decrypt_data2(session->internals.selected_key, 0, ++ &ciphertext, premaster_secret.data, ++ premaster_secret.size); ++ /* After this point, any conditional on failure that cause differences ++ * in execution may create a timing or cache access pattern side ++ * channel that can be used as an oracle, so tread carefully */ ++ ++ /* Error handling logic: ++ * In case decryption fails then don't inform the peer. Just use the ++ * random key previously generated. (in order to avoid attack against ++ * pkcs-1 formatting). ++ * ++ * If we get version mismatches no error is returned either. We ++ * proceed normally. This is to defend against the attack described ++ * in the paper "Attacking RSA-based sessions in SSL/TLS" by ++ * Vlastimil Klima, Ondej Pokorny and Tomas Rosa. ++ */ ++ + /* This is here to avoid the version check attack + * discussed above. + */ +- +- premaster_secret.data[0] = _gnutls_get_adv_version_major(session); +- premaster_secret.data[1] = _gnutls_get_adv_version_minor(session); ++ premaster_secret.data[0] = ver_maj; ++ premaster_secret.data[1] = ver_min; + + /* find the key of this username + */ +--- a/lib/gnutls_int.h ++++ b/lib/gnutls_int.h +@@ -974,7 +974,6 @@ struct gnutls_priority_st { + bool _no_etm; + bool _no_ext_master_secret; + bool _allow_key_usage_violation; +- bool _allow_wrong_pms; + bool _dumbfw; + unsigned int _dh_prime_bits; /* old (deprecated) variable */ + +@@ -992,7 +991,6 @@ struct gnutls_priority_st { + (x)->no_etm = 1; \ + (x)->no_ext_master_secret = 1; \ + (x)->allow_key_usage_violation = 1; \ +- (x)->allow_wrong_pms = 1; \ + (x)->dumbfw = 1 + + #define ENABLE_PRIO_COMPAT(x) \ +@@ -1001,7 +999,6 @@ struct gnutls_priority_st { + (x)->_no_etm = 1; \ + (x)->_no_ext_master_secret = 1; \ + (x)->_allow_key_usage_violation = 1; \ +- (x)->_allow_wrong_pms = 1; \ + (x)->_dumbfw = 1 + + /* DH and RSA parameters types. +@@ -1126,7 +1123,6 @@ typedef struct { + bool no_etm; + bool no_ext_master_secret; + bool allow_key_usage_violation; +- bool allow_wrong_pms; + bool dumbfw; + + /* old (deprecated) variable. This is used for both srp_prime_bits +--- a/lib/priority.c ++++ b/lib/priority.c +@@ -690,7 +690,6 @@ gnutls_priority_set(gnutls_session_t ses + COPY_TO_INTERNALS(no_etm); + COPY_TO_INTERNALS(no_ext_master_secret); + COPY_TO_INTERNALS(allow_key_usage_violation); +- COPY_TO_INTERNALS(allow_wrong_pms); + COPY_TO_INTERNALS(dumbfw); + COPY_TO_INTERNALS(dh_prime_bits); +
meta/recipes-support/gnutls/gnutls_3.7.4.bb +1 -0
@@ -23,6 +23,7 @@ SRC_URI = "https://www.gnupg.org/ftp/gcrypt/gnutls/v${SHRT_VER}/gnutls-${PV}.tar file://arm_eabi.patch \ file://CVE-2022-2509.patch \ file://CVE-2023-0361.patch \ + file://CVE-2023-5981.patch \ " SRC_URI[sha256sum] = "e6adbebcfbc95867de01060d93c789938cf89cc1d1f6ef9ef661890f6217451f"
Signature · 1 line
2.34.1
PATCH 4/5 gstreamer1.0-plugins-base: enable glx/opengl support
In reply to Steve SakomanFrom: Alexander Kanavin <[email protected]>
This is required by latest webkit when built with x11 support.
(From OE-Core rev: 024edebf6f722ae4d05411be348730d9eeb3bd7c)
--- .../gstreamer/gstreamer1.0-plugins-base_1.20.7.bb | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-)
meta/recipes-multimedia/gstreamer/gstreamer1.0-plugins-base_1.20.7.bb +4 -2
@@ -21,7 +21,8 @@ inherit gobject-introspection # opengl packageconfig factored out to make it easy for distros # and BSP layers to choose OpenGL APIs/platforms/window systems -PACKAGECONFIG_GL ?= "${@bb.utils.contains('DISTRO_FEATURES', 'opengl', 'gles2 egl', '', d)}" +PACKAGECONFIG_X11 = "${@bb.utils.contains('DISTRO_FEATURES', 'x11', 'opengl glx', '', d)}" +PACKAGECONFIG_GL ?= "${@bb.utils.contains('DISTRO_FEATURES', 'opengl', 'gles2 egl ${PACKAGECONFIG_X11}', '', d)}" PACKAGECONFIG ??= " \ ${GSTREAMER_ORC} \
@@ -32,7 +33,7 @@ PACKAGECONFIG ??= " \ " OPENGL_APIS = 'opengl gles2' -OPENGL_PLATFORMS = 'egl' +OPENGL_PLATFORMS = 'egl glx' X11DEPENDS = "virtual/libx11 libsm libxrender libxv" X11ENABLEOPTS = "-Dx11=enabled -Dxvideo=enabled -Dxshm=enabled"
@@ -61,6 +62,7 @@ PACKAGECONFIG[gles2] = ",,virtual/libgles2" # OpenGL platform packageconfigs PACKAGECONFIG[egl] = ",,virtual/egl" +PACKAGECONFIG[glx] = ",,virtual/libgl" # OpenGL window systems (except for X11) PACKAGECONFIG[gbm] = ",,virtual/libgbm libgudev libdrm"
Signature · 1 line
2.34.1
PATCH 5/5 openssh: drop sudo from ptest dependencies
In reply to Steve SakomanFrom: Mikko Rapeli <[email protected]>
The tests don't actually need sudo on core-image-ptest-openssh. Based on logs seen in
https://bugzilla.yoctoproject.org/show_bug.cgi?id=15178 it seems that socket errors from sudo are creeping into stderr which are failing the banner ptest from openssh. Removing sudo should help removing the stderr messages and possibly cure the banner test failures.
(From OE-Core rev: 47e754f483b674b207bfddcc8d4c5d9a3008e102)
--- meta/recipes-connectivity/openssh/openssh/run-ptest | 2 +- meta/recipes-connectivity/openssh/openssh_8.9p1.bb | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-)
meta/recipes-connectivity/openssh/openssh/run-ptest +1 -1
@@ -5,7 +5,7 @@ export SKIP_UNIT=1 cd regress sed -i "/\t\tagent-ptrace /d" Makefile -make -k BUILDDIR=`pwd`/.. .OBJDIR=`pwd` .CURDIR=`pwd` SUDO="sudo" tests \ +make -k BUILDDIR=`pwd`/.. .OBJDIR=`pwd` .CURDIR=`pwd` SUDO="" tests \ | sed -u -e 's/^skipped/SKIP: /g' -e 's/^ok /PASS: /g' -e 's/^failed/FAIL: /g' SSHAGENT=`which ssh-agent`
meta/recipes-connectivity/openssh/openssh_8.9p1.bb +1 -1
@@ -170,7 +170,7 @@ RDEPENDS:${PN}-sshd += "${PN}-keygen ${@bb.utils.contains('DISTRO_FEATURES', 'pa # conflict with each other RDEPENDS:${PN}-dev = "" # gdb would make attach-ptrace test pass rather than skip but not worth the build dependencies -RDEPENDS:${PN}-ptest += "${PN}-sftp ${PN}-misc ${PN}-sftp-server make sed sudo coreutils" +RDEPENDS:${PN}-ptest += "${PN}-sftp ${PN}-misc ${PN}-sftp-server make sed coreutils" RPROVIDES:${PN}-ssh = "ssh" RPROVIDES:${PN}-sshd = "sshd"
Signature · 1 line
2.34.1