Skip to contentSkip to conversation

nss: Move to meta-oe

RFC PATCH In review Adrian Bunk 7 participants 26 messages Feb 23, 2020 – Mar 11, 2020

lore
Share & tools
lore
#0 lore raw

RFC PATCH 1/2 nss: Move to meta-oe

rpm was the last user in OE-core.

---
 meta/conf/distro/include/maintainers.inc      |   1 -
 ...figure-option-to-disable-ARM-HW-cryp.patch |  52 ----
 ...0001-nss-fix-support-cross-compiling.patch |  48 ---
 meta/recipes-support/nss/nss/blank-cert9.db   | Bin 28672 -> 0 bytes
 meta/recipes-support/nss/nss/blank-key4.db    | Bin 36864 -> 0 bytes
 .../nss/nss/disable-Wvarargs-with-clang.patch |  33 ---
 .../nss-fix-incorrect-shebang-of-perl.patch   | 110 -------
 .../nss/nss/nss-fix-nsinstall-build.patch     |  36 ---
 .../nss-no-rpath-for-cross-compiling.patch    |  26 --
 meta/recipes-support/nss/nss/nss.pc.in        |  11 -
 .../nss/nss/pqg.c-ULL_addend.patch            |  23 --
 meta/recipes-support/nss/nss/signlibs.sh      |  20 --
 .../recipes-support/nss/nss/system-pkcs11.txt |   5 -
 meta/recipes-support/nss/nss_3.50.bb          | 273 ------------------
 14 files changed, 638 deletions(-)
 delete mode 100644 meta/recipes-support/nss/nss/0001-freebl-add-a-configure-option-to-disable-ARM-HW-cryp.patch
 delete mode 100644 meta/recipes-support/nss/nss/0001-nss-fix-support-cross-compiling.patch
 delete mode 100644 meta/recipes-support/nss/nss/blank-cert9.db
 delete mode 100644 meta/recipes-support/nss/nss/blank-key4.db
 delete mode 100644 meta/recipes-support/nss/nss/disable-Wvarargs-with-clang.patch
 delete mode 100644 meta/recipes-support/nss/nss/nss-fix-incorrect-shebang-of-perl.patch
 delete mode 100644 meta/recipes-support/nss/nss/nss-fix-nsinstall-build.patch
 delete mode 100644 meta/recipes-support/nss/nss/nss-no-rpath-for-cross-compiling.patch
 delete mode 100644 meta/recipes-support/nss/nss/nss.pc.in
 delete mode 100644 meta/recipes-support/nss/nss/pqg.c-ULL_addend.patch
 delete mode 100644 meta/recipes-support/nss/nss/signlibs.sh
 delete mode 100644 meta/recipes-support/nss/nss/system-pkcs11.txt
 delete mode 100644 meta/recipes-support/nss/nss_3.50.bb
meta/conf/distro/include/maintainers.inc +0 -1
index 8f612ace39..e4b710d640 100644
@@ -526,7 +526,6 @@ RECIPE_MAINTAINER_pn-nfs-utils = "Robert Yang <[email protected]>"
 RECIPE_MAINTAINER_pn-ninja = "Khem Raj <[email protected]>"
 RECIPE_MAINTAINER_pn-npth = "Alexander Kanavin <[email protected]>"
 RECIPE_MAINTAINER_pn-nspr = "Armin Kuster <[email protected]>"
-RECIPE_MAINTAINER_pn-nss = "Armin Kuster <[email protected]>"
 RECIPE_MAINTAINER_pn-nss-myhostname = "Anuj Mittal <[email protected]>"
 RECIPE_MAINTAINER_pn-ofono = "Ross Burton <[email protected]>"
 RECIPE_MAINTAINER_pn-opensbi = "Alistair Francis <[email protected]>"
meta/recipes-support/nss/nss/0001-freebl-add-a-configure-option-to-disable-ARM-HW-cryp.patch deleted +0 -52

52 lines not on this page · raw message

meta/recipes-support/nss/nss/0001-nss-fix-support-cross-compiling.patch deleted +0 -48

48 lines not on this page · raw message

meta/recipes-support/nss/nss/blank-cert9.db deleted binary +0 -0
deleted file mode 100644
index 7d4bcf2582d510f7b51d4306706746178c41fbbc..0000000000000000000000000000000000000000
GIT binary patch
meta/recipes-support/nss/nss/blank-key4.db deleted binary +0 -0
deleted file mode 100644
index d47f08d04fe82197bc6a39ef9bf216b61c3dc77a..0000000000000000000000000000000000000000
GIT binary patch
meta/recipes-support/nss/nss/disable-Wvarargs-with-clang.patch deleted +0 -33
deleted file mode 100644
index de812d27ba..0000000000
@@ -1,33 +0,0 @@
-clang 3.9 add this warning to rightly flag undefined
-behavior, we relegate this to be just a warning instead
-of error and keep the behavior as it was. Right fix would
-be to not pass enum to the function with variadic arguments
-as last named argument
-
-Fixes errors like
-ocsp.c:2220:22: error: passing an object that undergoes default argument promotion to 'va_start' has undefined behavior [-Werror,-Wvarargs]
-        va_start(ap, responseType0);
-                     ^
-ocsp.c:2200:43: note: parameter of type 'SECOidTag' is declared here
-                                SECOidTag responseType0, ...)
-
-see
-https://www.securecoding.cert.org/confluence/display/cplusplus/EXP58-CPP.+Pass+an+object+of+the+correct+type+to+va_start
-for more details
-
-Signed-off-by: Khem Raj <[email protected]>
-Upstream-Status: Pending
-
-Index: nss-3.37.1/nss/coreconf/Werror.mk
-===================================================================
---- nss-3.37.1.orig/nss/coreconf/Werror.mk
-+++ nss-3.37.1/nss/coreconf/Werror.mk
-@@ -56,7 +56,7 @@ ifndef WARNING_CFLAGS
-     ifdef CC_IS_CLANG
-       # -Qunused-arguments : clang objects to arguments that it doesn't understand
-       #    and fixing this would require rearchitecture
--      WARNING_CFLAGS += -Qunused-arguments
-+      WARNING_CFLAGS += -Qunused-arguments -Wno-error=varargs
-       # -Wno-parentheses-equality : because clang warns about macro expansions
-       WARNING_CFLAGS += $(call disable_warning,parentheses-equality)
-       ifdef BUILD_OPT
meta/recipes-support/nss/nss/nss-fix-incorrect-shebang-of-perl.patch deleted +0 -110

110 lines not on this page · raw message

meta/recipes-support/nss/nss/nss-fix-nsinstall-build.patch deleted +0 -36
deleted file mode 100644
index 43c09d13ea..0000000000
@@ -1,36 +0,0 @@
-Fix nss multilib build on openSUSE 11.x 32bit
-
-While building lib64-nss on openSUSE 11.x 32bit, the nsinstall will
-fail with error:
-
-* nsinstall.c:1:0: sorry, unimplemented: 64-bit mode not compiled
-
-It caused by the '-m64' option which passed to host gcc.
-
-The nsinstall was built first while nss starting to build, it only runs
-on host to install built files, it doesn't need any cross-compling or
-multilib build options. Just clean the ARCHFLAG and LDFLAGS to fix this
-error.
-
-Upstream-Status: Pending
-
-Signed-off-by: Wenzong Fan <[email protected]>
-===================================================
-Index: nss-3.24/nss/coreconf/nsinstall/Makefile
-===================================================================
---- nss-3.24.orig/nss/coreconf/nsinstall/Makefile
-+++ nss-3.24/nss/coreconf/nsinstall/Makefile
-@@ -18,6 +18,13 @@ INTERNAL_TOOLS  = 1
- 
- include $(DEPTH)/coreconf/config.mk
- 
-+# nsinstall is unfit for cross-compiling/multilib-build since it was
-+# always run on local host to install built files. This change intends
-+# to clean the '-m64' from ARCHFLAG and LDFLAGS.
-+ARCHFLAG =
-+LDFLAGS =
-+# CFLAGS =
-+
- ifeq (,$(filter-out OS2 WIN%,$(OS_TARGET)))
- PROGRAM		=
- else
meta/recipes-support/nss/nss/nss-no-rpath-for-cross-compiling.patch deleted +0 -26
deleted file mode 100644
index 7661dc93a0..0000000000
@@ -1,26 +0,0 @@
-nss:no rpath for cross compiling
-
-Signed-off-by: Hongxu Jia <[email protected]>
-Upstream-Status: Inappropriate [configuration]
----
- nss/cmd/platlibs.mk | 4 ++--
- 1 file changed, 2 insertions(+), 2 deletions(-)
-
-diff --git a/nss/cmd/platlibs.mk b/nss/cmd/platlibs.mk
---- a/nss/cmd/platlibs.mk
-+++ b/nss/cmd/platlibs.mk
-@@ -18,9 +18,9 @@ endif
- 
- ifeq ($(OS_ARCH), Linux)
- ifeq ($(USE_64), 1)
--EXTRA_SHARED_LIBS += -Wl,-rpath,'$$ORIGIN/../lib64:/opt/sun/private/lib64:$$ORIGIN/../lib'
-+#EXTRA_SHARED_LIBS += -Wl,-rpath,'$$ORIGIN/../lib64:/opt/sun/private/lib64:$$ORIGIN/../lib'
- else
--EXTRA_SHARED_LIBS += -Wl,-rpath,'$$ORIGIN/../lib:/opt/sun/private/lib'
-+#EXTRA_SHARED_LIBS += -Wl,-rpath,'$$ORIGIN/../lib:/opt/sun/private/lib'
- endif
- endif
- 
--- 
-1.8.1.2
-
meta/recipes-support/nss/nss/nss.pc.in deleted +0 -11
deleted file mode 100644
index 402b4ecb33..0000000000
@@ -1,11 +0,0 @@
-prefix=OEPREFIX
-exec_prefix=OEEXECPREFIX
-libdir=OELIBDIR
-includedir=OEINCDIR
-
-Name: NSS
-Description: Network Security Services
-Version: %NSS_VERSION%
-Requires: nspr >= %NSPR_VERSION%
-Libs: -L${libdir} -lssl3 -lsmime3 -lnss3 -lsoftokn3 -lnssutil3
-Cflags: -IOEINCDIR
meta/recipes-support/nss/nss/pqg.c-ULL_addend.patch deleted +0 -23
deleted file mode 100644
index 3a817faaa6..0000000000
@@ -1,23 +0,0 @@
-nss does not build on mips with clang because wrong types are used?
-
-pqg.c:339:16: error: comparison of constant 18446744073709551615 with expression of type 'unsigned long' is always true [-Werror,-Wtautological-constant-out-of-range-compare]
-     if (addend < MP_DIGIT_MAX) {
-       ~~~~~~ ^ ~~~~~~~~~~~~
-
-Signed-off-by: Khem Raj <[email protected]>
-Upstream-Status: Pending
-Index: nss-3.37.1/nss/lib/freebl/pqg.c
-===================================================================
---- nss-3.37.1.orig/nss/lib/freebl/pqg.c
-+++ nss-3.37.1/nss/lib/freebl/pqg.c
-@@ -326,8 +326,8 @@ generate_h_candidate(SECItem *hit, mp_in
- 
- static SECStatus
- addToSeed(const SECItem *seed,
--          unsigned long addend,
--          int seedlen, /* g in 186-1 */
-+          unsigned long long  addend,
-+          int                 seedlen, /* g in 186-1 */
-           SECItem *seedout)
- {
-     mp_int s, sum, modulus, tmp;
meta/recipes-support/nss/nss/signlibs.sh deleted +0 -20
deleted file mode 100644
index a74e499f8c..0000000000
@@ -1,20 +0,0 @@
-#!/bin/sh
-
-# signlibs.sh
-#
-# (c)2010 Wind River Systems, Inc.
-#
-# regenerates the .chk files for the NSS libraries that require it
-# since the ones that are built have incorrect checksums that were
-# calculated on the host where they really need to be done on the
-# target
-
-CHK_FILES=`ls /lib*/*.chk /usr/lib*/*.chk 2>/dev/null`
-SIGN_BINARY=`which shlibsign`
-for I in $CHK_FILES
-do
-       DN=`dirname $I`
-       BN=`basename $I .chk`
-       FN=$DN/$BN.so
-       $SIGN_BINARY -i $FN
-done
meta/recipes-support/nss/nss/system-pkcs11.txt deleted +0 -5
deleted file mode 100644
index 1a264e9cc4..0000000000
@@ -1,5 +0,0 @@
-library=
-name=NSS Internal PKCS #11 Module
-parameters=configdir='sql:/etc/pki/nssdb' certPrefix='' keyPrefix='' secmod='secmod.db' flags= updatedir='' updateCertPrefix='' updateKeyPrefix='' updateid='' updateTokenDescription='' 
-NSS=Flags=internal,critical trustOrder=75 cipherOrder=100 slotParams=(1={slotFlags=[ECC,RSA,DSA,DH,RC2,RC4,DES,RANDOM,SHA1,MD5,MD2,SSL,TLS,AES,Camellia,SEED,SHA256,SHA512] askpw=any timeout=30})
-
meta/recipes-support/nss/nss_3.50.bb deleted +0 -273

273 lines not on this page · raw message

Signature · 1 line
2.17.1
#1 lore raw

RFC PATCH 2/2 nspr: Move to meta-oe

It was used only by nss.

---
 meta/conf/distro/include/maintainers.inc      |   1 -
 .../nspr/0001-md-Fix-build-with-musl.patch    |  31 ---
 .../nspr/nspr/0002-Add-nios2-support.patch    | 102 ---------
 ...remove-_BUILD_STRING-and-_BUILD_TIME.patch | 103 ---------
 .../nspr/nspr/fix-build-on-x86_64.patch       |  52 -----
 meta/recipes-support/nspr/nspr/nspr.pc.in     |  11 -
 .../nspr/nspr/remove-rpath-from-tests.patch   |  26 ---
 .../remove-srcdir-from-configure-in.patch     |  19 --
 meta/recipes-support/nspr/nspr_4.25.bb        | 197 ------------------
 9 files changed, 542 deletions(-)
 delete mode 100644 meta/recipes-support/nspr/nspr/0001-md-Fix-build-with-musl.patch
 delete mode 100644 meta/recipes-support/nspr/nspr/0002-Add-nios2-support.patch
 delete mode 100644 meta/recipes-support/nspr/nspr/Makefile.in-remove-_BUILD_STRING-and-_BUILD_TIME.patch
 delete mode 100644 meta/recipes-support/nspr/nspr/fix-build-on-x86_64.patch
 delete mode 100644 meta/recipes-support/nspr/nspr/nspr.pc.in
 delete mode 100644 meta/recipes-support/nspr/nspr/remove-rpath-from-tests.patch
 delete mode 100644 meta/recipes-support/nspr/nspr/remove-srcdir-from-configure-in.patch
 delete mode 100644 meta/recipes-support/nspr/nspr_4.25.bb
meta/conf/distro/include/maintainers.inc +0 -1
index e4b710d640..874c595f0d 100644
@@ -525,7 +525,6 @@ RECIPE_MAINTAINER_pn-nfs-export-root = "Robert Yang <[email protected]>"
 RECIPE_MAINTAINER_pn-nfs-utils = "Robert Yang <[email protected]>"
 RECIPE_MAINTAINER_pn-ninja = "Khem Raj <[email protected]>"
 RECIPE_MAINTAINER_pn-npth = "Alexander Kanavin <[email protected]>"
-RECIPE_MAINTAINER_pn-nspr = "Armin Kuster <[email protected]>"
 RECIPE_MAINTAINER_pn-nss-myhostname = "Anuj Mittal <[email protected]>"
 RECIPE_MAINTAINER_pn-ofono = "Ross Burton <[email protected]>"
 RECIPE_MAINTAINER_pn-opensbi = "Alistair Francis <[email protected]>"
meta/recipes-support/nspr/nspr/0001-md-Fix-build-with-musl.patch deleted +0 -31
deleted file mode 100644
index f3cd670026..0000000000
@@ -1,31 +0,0 @@
-From 147f3c2acbd96d44025cec11800ded0282327764 Mon Sep 17 00:00:00 2001
-From: Khem Raj <[email protected]>
-Date: Mon, 18 Sep 2017 17:22:43 -0700
-Subject: [PATCH] md: Fix build with musl
-
-The MIPS specific header <sgidefs.h> is not provided by musl
-linux kernel headers provide <asm/sgidefs.h> which has same definitions
-
-Signed-off-by: Khem Raj <[email protected]>
----
-Upstream-Status: Pending
-
- pr/include/md/_linux.cfg | 2 +-
- 1 file changed, 1 insertion(+), 1 deletion(-)
-
-diff --git a/pr/include/md/_linux.cfg b/pr/include/md/_linux.cfg
-index 640b19c..31296a8 100644
---- a/pr/include/md/_linux.cfg
-+++ b/pr/include/md/_linux.cfg
-@@ -499,7 +499,7 @@
- #elif defined(__mips__)
- 
- /* For _ABI64 */
--#include <sgidefs.h>
-+#include <asm/sgidefs.h>
- 
- #ifdef __MIPSEB__
- #define IS_BIG_ENDIAN 1
--- 
-2.14.1
-
meta/recipes-support/nspr/nspr/0002-Add-nios2-support.patch deleted +0 -102

102 lines not on this page · raw message

meta/recipes-support/nspr/nspr/Makefile.in-remove-_BUILD_STRING-and-_BUILD_TIME.patch deleted +0 -103

103 lines not on this page · raw message

meta/recipes-support/nspr/nspr/fix-build-on-x86_64.patch deleted +0 -52

52 lines not on this page · raw message

meta/recipes-support/nspr/nspr/nspr.pc.in deleted +0 -11
deleted file mode 100644
index 1f15d19cfa..0000000000
@@ -1,11 +0,0 @@
-os_libs=-lpthread -ldl
-prefix=OEPREFIX
-exec_prefix=OEEXECPREFIX
-libdir=OELIBDIR
-includedir=OEINCDIR
-
-Name: NSPR
-Description: The Netscape Portable Runtime
-Version: NSPRVERSION
-Libs: -L${libdir} -lplds4 -lplc4 -lnspr4 -lpthread -ldl
-Cflags: -I${includedir}/nspr
meta/recipes-support/nspr/nspr/remove-rpath-from-tests.patch deleted +0 -26
deleted file mode 100644
index 7ba59ed644..0000000000
@@ -1,26 +0,0 @@
-Author: Andrei Gherzan <[email protected]>
-Date:   Thu Feb 9 00:03:38 2012 +0200
-
-Avoid QA warnings by removing hardcoded rpath from binaries.
-
-[...]
-WARNING: QA Issue: package nspr contains bad RPATH {builddir}/tmp/work/armv5te-poky-linux-gnueabi/nspr-4.8.9-r1/nspr-4.8.9/mozilla/nsprpub/pr/tests/../../dist/lib
-in file {builddir}/tmp/work/armv5te-poky-linux-gnueabi/nspr-4.8.9-r1/packages-split/nspr/usr/lib/nspr/tests/multiwait
-[...]
-
-Signed-off-by: Andrei Gherzan <[email protected]>
-Upstream-Status: Pending
-
-Index: nspr/pr/tests/Makefile.in
-===================================================================
---- nspr.orig/pr/tests/Makefile.in
-+++ nspr/pr/tests/Makefile.in
-@@ -316,7 +316,7 @@ ifeq ($(OS_ARCH), SunOS)
- endif # SunOS
- 
- ifeq (,$(filter-out Linux GNU GNU_%,$(OS_ARCH)))
--    LDOPTS += -Xlinker -rpath $(ABSOLUTE_LIB_DIR)
-+    LDOPTS += -Xlinker
-     ifeq ($(USE_PTHREADS),1)
-         EXTRA_LIBS = -lpthread
-     endif
meta/recipes-support/nspr/nspr/remove-srcdir-from-configure-in.patch deleted +0 -19
deleted file mode 100644
index bde715c5dc..0000000000
@@ -1,19 +0,0 @@
-the $srcdir is not defined at the time of gnu-configurize.
-
-Upstream-Status: Inappropriate [OE-Core specific]
-
-Signed-off-by: Saul Wold <[email protected]>
-
-Index: nspr/configure.in
-===================================================================
---- nspr.orig/configure.in
-+++ nspr/configure.in
-@@ -8,7 +8,7 @@ AC_PREREQ(2.61)
- AC_INIT
- AC_CONFIG_SRCDIR([pr/include/nspr.h])
- 
--AC_CONFIG_AUX_DIR(${srcdir}/build/autoconf)
-+AC_CONFIG_AUX_DIR(build/autoconf)
- AC_CANONICAL_TARGET
- 
- dnl ========================================================
meta/recipes-support/nspr/nspr_4.25.bb deleted +0 -197

197 lines not on this page · raw message

Signature · 1 line
2.17.1
#2 lore raw
In reply to Adrian Bunk
Quoted 3 lines · Adrian Bunk

rpm was the last user in OE-core.

we should also assess external dependencies especially on libraries, there might be layers which do not depend on meta-oe but use nss or enable nss packageconfigs in core components like curl.

Quoted 811 lines
---
 meta/conf/distro/include/maintainers.inc      |   1 -
 ...figure-option-to-disable-ARM-HW-cryp.patch |  52 ----
 ...0001-nss-fix-support-cross-compiling.patch |  48 ---
 meta/recipes-support/nss/nss/blank-cert9.db   | Bin 28672 -> 0 bytes
 meta/recipes-support/nss/nss/blank-key4.db    | Bin 36864 -> 0 bytes
 .../nss/nss/disable-Wvarargs-with-clang.patch |  33 ---
 .../nss-fix-incorrect-shebang-of-perl.patch   | 110 -------
 .../nss/nss/nss-fix-nsinstall-build.patch     |  36 ---
 .../nss-no-rpath-for-cross-compiling.patch    |  26 --
 meta/recipes-support/nss/nss/nss.pc.in        |  11 -
 .../nss/nss/pqg.c-ULL_addend.patch            |  23 --
 meta/recipes-support/nss/nss/signlibs.sh      |  20 --
 .../recipes-support/nss/nss/system-pkcs11.txt |   5 -
 meta/recipes-support/nss/nss_3.50.bb          | 273 ------------------
 14 files changed, 638 deletions(-)
 delete mode 100644 meta/recipes-support/nss/nss/0001-freebl-add-a-configure-option-to-disable-ARM-HW-cryp.patch
 delete mode 100644 meta/recipes-support/nss/nss/0001-nss-fix-support-cross-compiling.patch
 delete mode 100644 meta/recipes-support/nss/nss/blank-cert9.db
 delete mode 100644 meta/recipes-support/nss/nss/blank-key4.db
 delete mode 100644 meta/recipes-support/nss/nss/disable-Wvarargs-with-clang.patch
 delete mode 100644 meta/recipes-support/nss/nss/nss-fix-incorrect-shebang-of-perl.patch
 delete mode 100644 meta/recipes-support/nss/nss/nss-fix-nsinstall-build.patch
 delete mode 100644 meta/recipes-support/nss/nss/nss-no-rpath-for-cross-compiling.patch
 delete mode 100644 meta/recipes-support/nss/nss/nss.pc.in
 delete mode 100644 meta/recipes-support/nss/nss/pqg.c-ULL_addend.patch
 delete mode 100644 meta/recipes-support/nss/nss/signlibs.sh
 delete mode 100644 meta/recipes-support/nss/nss/system-pkcs11.txt
 delete mode 100644 meta/recipes-support/nss/nss_3.50.bb
meta/conf/distro/include/maintainers.inc +0 -1
index 8f612ace39..e4b710d640 100644
@@ -526,7 +526,6 @@ RECIPE_MAINTAINER_pn-nfs-utils = "Robert Yang <[email protected]>"
 RECIPE_MAINTAINER_pn-ninja = "Khem Raj <[email protected]>"
 RECIPE_MAINTAINER_pn-npth = "Alexander Kanavin <[email protected]>"
 RECIPE_MAINTAINER_pn-nspr = "Armin Kuster <[email protected]>"
-RECIPE_MAINTAINER_pn-nss = "Armin Kuster <[email protected]>"
 RECIPE_MAINTAINER_pn-nss-myhostname = "Anuj Mittal <[email protected]>"
 RECIPE_MAINTAINER_pn-ofono = "Ross Burton <[email protected]>"
 RECIPE_MAINTAINER_pn-opensbi = "Alistair Francis <[email protected]>"
meta/recipes-support/nss/nss/0001-freebl-add-a-configure-option-to-disable-ARM-HW-cryp.patch deleted +0 -52
deleted file mode 100644
index c380c14491..0000000000
@@ -1,52 +0,0 @@
-From 5595e9651aca39af945931c73eb524a0f8bd130d Mon Sep 17 00:00:00 2001
-From: Alexander Kanavin <[email protected]>
-Date: Wed, 18 Dec 2019 12:29:50 +0100
-Subject: [PATCH] freebl: add a configure option to disable ARM HW crypto
-
-Not all current hardware supports it, particularly anything
-prior to armv8 does not.
-
-Upstream-Status: Pending
-Signed-off-by: Alexander Kanavin <[email protected]>
----
- nss/lib/freebl/Makefile | 3 +++
- 1 file changed, 3 insertions(+)
-
---- a/nss/lib/freebl/Makefile
-+++ b/nss/lib/freebl/Makefile
-@@ -125,6 +125,9 @@ else
-         DEFINES += -DNSS_X86
- endif
- endif
-+
-+ifdef NSS_USE_ARM_HW_CRYPTO
-+    DEFINES += -DNSS_USE_ARM_HW_CRYPTO
- ifeq ($(CPU_ARCH),aarch64)
-     DEFINES += -DUSE_HW_AES
-     EXTRA_SRCS += aes-armv8.c gcm-aarch64.c
-@@ -146,6 +149,7 @@ ifeq ($(CPU_ARCH),arm)
-         endif
-     endif
- endif
-+endif
-
- ifeq ($(OS_TARGET),OSF1)
-     DEFINES += -DMP_ASSEMBLY_MULTIPLY -DMP_NO_MP_WORD
---- a/nss/lib/freebl/gcm.c
-+++ b/nss/lib/freebl/gcm.c
-@@ -17,6 +17,7 @@
-
- #include <limits.h>
-
-+#ifdef NSS_USE_ARM_HW_CRYPTO
- /* old gcc doesn't support some poly64x2_t intrinsic */
- #if defined(__aarch64__) && defined(IS_LITTLE_ENDIAN) && \
-     (defined(__clang__) || defined(__GNUC__) && __GNUC__ > 6)
-@@ -25,6 +26,7 @@
- /* We don't test on big endian platform, so disable this on big endian. */
- #define USE_ARM_GCM
- #endif
-+#endif
-
- /* Forward declarations */
- SECStatus gcm_HashInit_hw(gcmHashContext *ghash);
meta/recipes-support/nss/nss/0001-nss-fix-support-cross-compiling.patch deleted +0 -48
deleted file mode 100644
index d5403397e7..0000000000
@@ -1,48 +0,0 @@
-From 0cf47ee432cc26a706864fcc09b2c3adc342a679 Mon Sep 17 00:00:00 2001
-From: Alexander Kanavin <[email protected]>
-Date: Wed, 22 Feb 2017 11:36:11 +0200
-Subject: [PATCH] nss: fix support cross compiling
-
-Let some make variables be assigned from outside makefile.
-
-Upstream-Status: Inappropriate [configuration]
-Signed-off-by: Hongxu Jia <[email protected]>
-Signed-off-by: Alexander Kanavin <[email protected]>
----
- nss/coreconf/arch.mk    | 2 +-
- nss/lib/freebl/Makefile | 6 ++++++
- 2 files changed, 7 insertions(+), 1 deletion(-)
-
-diff --git a/nss/coreconf/arch.mk b/nss/coreconf/arch.mk
-index 06c276f..9c1eb51 100644
---- a/nss/coreconf/arch.mk
-+++ b/nss/coreconf/arch.mk
-@@ -30,7 +30,7 @@ OS_TEST := $(shell uname -m)
- ifeq ($(OS_TEST),i86pc)
-     OS_RELEASE := $(shell uname -r)_$(OS_TEST)
- else
--    OS_RELEASE := $(shell uname -r)
-+    OS_RELEASE ?= $(shell uname -r)
- endif
-
- #
-diff --git a/nss/lib/freebl/Makefile b/nss/lib/freebl/Makefile
-index 0ce1425..ebeb411 100644
---- a/nss/lib/freebl/Makefile
-+++ b/nss/lib/freebl/Makefile
-@@ -36,6 +36,12 @@ ifdef USE_64
-       DEFINES += -DNSS_USE_64
- endif
-
-+ifeq ($(OS_TEST),mips)
-+ifndef USE_64
-+       DEFINES += -DNS_PTR_LE_32
-+endif
-+endif
-+
- ifdef USE_ABI32_FPU
-       DEFINES += -DNSS_USE_ABI32_FPU
- endif
---
-2.11.0
-
meta/recipes-support/nss/nss/blank-cert9.db deleted binary +0 -0
deleted file mode 100644
index 7d4bcf2582d510f7b51d4306706746178c41fbbc..0000000000000000000000000000000000000000
GIT binary patch
meta/recipes-support/nss/nss/blank-key4.db deleted binary +0 -0
deleted file mode 100644
index d47f08d04fe82197bc6a39ef9bf216b61c3dc77a..0000000000000000000000000000000000000000
GIT binary patch
meta/recipes-support/nss/nss/disable-Wvarargs-with-clang.patch deleted +0 -33
deleted file mode 100644
index de812d27ba..0000000000
@@ -1,33 +0,0 @@
-clang 3.9 add this warning to rightly flag undefined
-behavior, we relegate this to be just a warning instead
-of error and keep the behavior as it was. Right fix would
-be to not pass enum to the function with variadic arguments
-as last named argument
-
-Fixes errors like
-ocsp.c:2220:22: error: passing an object that undergoes default argument promotion to 'va_start' has undefined behavior [-Werror,-Wvarargs]
-        va_start(ap, responseType0);
-                     ^
-ocsp.c:2200:43: note: parameter of type 'SECOidTag' is declared here
-                                SECOidTag responseType0, ...)
-
-see
-https://www.securecoding.cert.org/confluence/display/cplusplus/EXP58-CPP.+Pass+an+object+of+the+correct+type+to+va_start
-for more details
-
-Signed-off-by: Khem Raj <[email protected]>
-Upstream-Status: Pending
-
-Index: nss-3.37.1/nss/coreconf/Werror.mk
-===================================================================
---- nss-3.37.1.orig/nss/coreconf/Werror.mk
-+++ nss-3.37.1/nss/coreconf/Werror.mk
-@@ -56,7 +56,7 @@ ifndef WARNING_CFLAGS
-     ifdef CC_IS_CLANG
-       # -Qunused-arguments : clang objects to arguments that it doesn't understand
-       #    and fixing this would require rearchitecture
--      WARNING_CFLAGS += -Qunused-arguments
-+      WARNING_CFLAGS += -Qunused-arguments -Wno-error=varargs
-       # -Wno-parentheses-equality : because clang warns about macro expansions
-       WARNING_CFLAGS += $(call disable_warning,parentheses-equality)
-       ifdef BUILD_OPT
meta/recipes-support/nss/nss/nss-fix-incorrect-shebang-of-perl.patch deleted +0 -110
deleted file mode 100644
index 547594d5b6..0000000000
@@ -1,110 +0,0 @@
-nss: fix incorrect shebang of perl
-
-Replace incorrect shebang of perl with `#!/usr/bin/env perl'.
-
-Signed-off-by: Hongxu Jia <[email protected]>
-Upstream-Status: Pending
----
- nss/cmd/smimetools/smime  | 2 +-
- nss/coreconf/cpdist.pl    | 2 +-
- nss/coreconf/import.pl    | 2 +-
- nss/coreconf/jniregen.pl  | 2 +-
- nss/coreconf/outofdate.pl | 2 +-
- nss/coreconf/release.pl   | 2 +-
- nss/coreconf/version.pl   | 2 +-
- nss/tests/clean_tbx       | 2 +-
- nss/tests/path_uniq       | 2 +-
- 9 files changed, 9 insertions(+), 9 deletions(-)
-
-diff --git a/nss/cmd/smimetools/smime b/nss/cmd/smimetools/smime
---- a/nss/cmd/smimetools/smime
-+++ b/nss/cmd/smimetools/smime
-@@ -1,4 +1,4 @@
--#!/usr/local/bin/perl
-+#!/usr/bin/env perl
-
- # This Source Code Form is subject to the terms of the Mozilla Public
- # License, v. 2.0. If a copy of the MPL was not distributed with this
-diff --git a/nss/coreconf/cpdist.pl b/nss/coreconf/cpdist.pl
-index 800edfb..652187f 100755
---- a/nss/coreconf/cpdist.pl
-+++ b/nss/coreconf/cpdist.pl
-@@ -1,4 +1,4 @@
--#! /usr/local/bin/perl
-+#!/usr/bin/env perl
- #
- # This Source Code Form is subject to the terms of the Mozilla Public
- # License, v. 2.0. If a copy of the MPL was not distributed with this
-diff --git a/nss/coreconf/import.pl b/nss/coreconf/import.pl
-index dd2d177..428eaa5 100755
---- a/nss/coreconf/import.pl
-+++ b/nss/coreconf/import.pl
-@@ -1,4 +1,4 @@
--#! /usr/local/bin/perl
-+#!/usr/bin/env perl
- #
- # This Source Code Form is subject to the terms of the Mozilla Public
- # License, v. 2.0. If a copy of the MPL was not distributed with this
-diff --git a/nss/coreconf/jniregen.pl b/nss/coreconf/jniregen.pl
-index 2039180..5f4f69c 100755
---- a/nss/coreconf/jniregen.pl
-+++ b/nss/coreconf/jniregen.pl
-@@ -1,4 +1,4 @@
--#!/usr/local/bin/perl
-+#!/usr/bin/env perl
- #
- # This Source Code Form is subject to the terms of the Mozilla Public
- # License, v. 2.0. If a copy of the MPL was not distributed with this
-diff --git a/nss/coreconf/outofdate.pl b/nss/coreconf/outofdate.pl
-index 33d80bb..01fc097 100755
---- a/nss/coreconf/outofdate.pl
-+++ b/nss/coreconf/outofdate.pl
-@@ -1,4 +1,4 @@
--#!/usr/local/bin/perl
-+#!/usr/bin/env perl
- #
- # This Source Code Form is subject to the terms of the Mozilla Public
- # License, v. 2.0. If a copy of the MPL was not distributed with this
-diff --git a/nss/coreconf/release.pl b/nss/coreconf/release.pl
-index 7cde19d..b5df2f6 100755
---- a/nss/coreconf/release.pl
-+++ b/nss/coreconf/release.pl
-@@ -1,4 +1,4 @@
--#! /usr/local/bin/perl
-+#!/usr/bin/env perl
- #
- # This Source Code Form is subject to the terms of the Mozilla Public
- # License, v. 2.0. If a copy of the MPL was not distributed with this
-diff --git a/nss/coreconf/version.pl b/nss/coreconf/version.pl
-index d2a4942..79359fe 100644
---- a/nss/coreconf/version.pl
-+++ b/nss/coreconf/version.pl
-@@ -1,4 +1,4 @@
--#!/usr/sbin/perl
-+#!/usr/bin/env perl
- #
- # This Source Code Form is subject to the terms of the Mozilla Public
- # License, v. 2.0. If a copy of the MPL was not distributed with this
-diff --git a/nss/tests/clean_tbx b/nss/tests/clean_tbx
-index 4de9555..a7def9f 100755
---- a/nss/tests/clean_tbx
-+++ b/nss/tests/clean_tbx
-@@ -1,4 +1,4 @@
--#! /bin/perl
-+#!/usr/bin/env perl
-
- #######################################################################
- #
-diff --git a/nss/tests/path_uniq b/nss/tests/path_uniq
-index f29f60a..08fbffa 100755
---- a/nss/tests/path_uniq
-+++ b/nss/tests/path_uniq
-@@ -1,4 +1,4 @@
--#! /bin/perl
-+#!/usr/bin/env perl
-
- ########################################################################
- #
---
-1.8.1.2
-
meta/recipes-support/nss/nss/nss-fix-nsinstall-build.patch deleted +0 -36
deleted file mode 100644
index 43c09d13ea..0000000000
@@ -1,36 +0,0 @@
-Fix nss multilib build on openSUSE 11.x 32bit
-
-While building lib64-nss on openSUSE 11.x 32bit, the nsinstall will
-fail with error:
-
-* nsinstall.c:1:0: sorry, unimplemented: 64-bit mode not compiled
-
-It caused by the '-m64' option which passed to host gcc.
-
-The nsinstall was built first while nss starting to build, it only runs
-on host to install built files, it doesn't need any cross-compling or
-multilib build options. Just clean the ARCHFLAG and LDFLAGS to fix this
-error.
-
-Upstream-Status: Pending
-
-Signed-off-by: Wenzong Fan <[email protected]>
-===================================================
-Index: nss-3.24/nss/coreconf/nsinstall/Makefile
-===================================================================
---- nss-3.24.orig/nss/coreconf/nsinstall/Makefile
-+++ nss-3.24/nss/coreconf/nsinstall/Makefile
-@@ -18,6 +18,13 @@ INTERNAL_TOOLS  = 1
-
- include $(DEPTH)/coreconf/config.mk
-
-+# nsinstall is unfit for cross-compiling/multilib-build since it was
-+# always run on local host to install built files. This change intends
-+# to clean the '-m64' from ARCHFLAG and LDFLAGS.
-+ARCHFLAG =
-+LDFLAGS =
-+# CFLAGS =
-+
- ifeq (,$(filter-out OS2 WIN%,$(OS_TARGET)))
- PROGRAM               =
- else
meta/recipes-support/nss/nss/nss-no-rpath-for-cross-compiling.patch deleted +0 -26
deleted file mode 100644
index 7661dc93a0..0000000000
@@ -1,26 +0,0 @@
-nss:no rpath for cross compiling
-
-Signed-off-by: Hongxu Jia <[email protected]>
-Upstream-Status: Inappropriate [configuration]
----
- nss/cmd/platlibs.mk | 4 ++--
- 1 file changed, 2 insertions(+), 2 deletions(-)
-
-diff --git a/nss/cmd/platlibs.mk b/nss/cmd/platlibs.mk
---- a/nss/cmd/platlibs.mk
-+++ b/nss/cmd/platlibs.mk
-@@ -18,9 +18,9 @@ endif
-
- ifeq ($(OS_ARCH), Linux)
- ifeq ($(USE_64), 1)
--EXTRA_SHARED_LIBS += -Wl,-rpath,'$$ORIGIN/../lib64:/opt/sun/private/lib64:$$ORIGIN/../lib'
-+#EXTRA_SHARED_LIBS += -Wl,-rpath,'$$ORIGIN/../lib64:/opt/sun/private/lib64:$$ORIGIN/../lib'
- else
--EXTRA_SHARED_LIBS += -Wl,-rpath,'$$ORIGIN/../lib:/opt/sun/private/lib'
-+#EXTRA_SHARED_LIBS += -Wl,-rpath,'$$ORIGIN/../lib:/opt/sun/private/lib'
- endif
- endif
-
---
-1.8.1.2
-
meta/recipes-support/nss/nss/nss.pc.in deleted +0 -11
deleted file mode 100644
index 402b4ecb33..0000000000
@@ -1,11 +0,0 @@
-prefix=OEPREFIX
-exec_prefix=OEEXECPREFIX
-libdir=OELIBDIR
-includedir=OEINCDIR
-
-Name: NSS
-Description: Network Security Services
-Version: %NSS_VERSION%
-Requires: nspr >= %NSPR_VERSION%
-Libs: -L${libdir} -lssl3 -lsmime3 -lnss3 -lsoftokn3 -lnssutil3
-Cflags: -IOEINCDIR
meta/recipes-support/nss/nss/pqg.c-ULL_addend.patch deleted +0 -23
deleted file mode 100644
index 3a817faaa6..0000000000
@@ -1,23 +0,0 @@
-nss does not build on mips with clang because wrong types are used?
-
-pqg.c:339:16: error: comparison of constant 18446744073709551615 with expression of type 'unsigned long' is always true [-Werror,-Wtautological-constant-out-of-range-compare]
-     if (addend < MP_DIGIT_MAX) {
-       ~~~~~~ ^ ~~~~~~~~~~~~
-
-Signed-off-by: Khem Raj <[email protected]>
-Upstream-Status: Pending
-Index: nss-3.37.1/nss/lib/freebl/pqg.c
-===================================================================
---- nss-3.37.1.orig/nss/lib/freebl/pqg.c
-+++ nss-3.37.1/nss/lib/freebl/pqg.c
-@@ -326,8 +326,8 @@ generate_h_candidate(SECItem *hit, mp_in
-
- static SECStatus
- addToSeed(const SECItem *seed,
--          unsigned long addend,
--          int seedlen, /* g in 186-1 */
-+          unsigned long long  addend,
-+          int                 seedlen, /* g in 186-1 */
-           SECItem *seedout)
- {
-     mp_int s, sum, modulus, tmp;
meta/recipes-support/nss/nss/signlibs.sh deleted +0 -20
deleted file mode 100644
index a74e499f8c..0000000000
@@ -1,20 +0,0 @@
-#!/bin/sh
-
-# signlibs.sh
-#
-# (c)2010 Wind River Systems, Inc.
-#
-# regenerates the .chk files for the NSS libraries that require it
-# since the ones that are built have incorrect checksums that were
-# calculated on the host where they really need to be done on the
-# target
-
-CHK_FILES=`ls /lib*/*.chk /usr/lib*/*.chk 2>/dev/null`
-SIGN_BINARY=`which shlibsign`
-for I in $CHK_FILES
-do
-       DN=`dirname $I`
-       BN=`basename $I .chk`
-       FN=$DN/$BN.so
-       $SIGN_BINARY -i $FN
-done
meta/recipes-support/nss/nss/system-pkcs11.txt deleted +0 -5
deleted file mode 100644
index 1a264e9cc4..0000000000
@@ -1,5 +0,0 @@
-library=
-name=NSS Internal PKCS #11 Module
-parameters=configdir='sql:/etc/pki/nssdb' certPrefix='' keyPrefix='' secmod='secmod.db' flags= updatedir='' updateCertPrefix='' updateKeyPrefix='' updateid='' updateTokenDescription=''
-NSS=Flags=internal,critical trustOrder=75 cipherOrder=100 slotParams=(1={slotFlags=[ECC,RSA,DSA,DH,RC2,RC4,DES,RANDOM,SHA1,MD5,MD2,SSL,TLS,AES,Camellia,SEED,SHA256,SHA512] askpw=any timeout=30})
-
meta/recipes-support/nss/nss_3.50.bb deleted +0 -274
deleted file mode 100644
index e9855d7a7e..0000000000
@@ -1,273 +0,0 @@
-SUMMARY = "Mozilla's SSL and TLS implementation"
-DESCRIPTION = "Network Security Services (NSS) is a set of libraries \
-designed to support cross-platform development of \
-security-enabled client and server applications. \
-Applications built with NSS can support SSL v2 and v3, \
-TLS, PKCS 5, PKCS 7, PKCS 11, PKCS 12, S/MIME, X.509 \
-v3 certificates, and other security standards."
-HOMEPAGE = "http://www.mozilla.org/projects/security/pki/nss/"
-SECTION = "libs"
-
-DEPENDS = "sqlite3 nspr zlib nss-native"
-DEPENDS_class-native = "sqlite3-native nspr-native zlib-native"
-
-LICENSE = "MPL-2.0 | (MPL-2.0 & GPL-2.0+) | (MPL-2.0 & LGPL-2.1+)"
-
-LIC_FILES_CHKSUM = "file://nss/COPYING;md5=3b1e88e1b9c0b5a4b2881d46cce06a18 \
-                    file://nss/lib/freebl/mpi/doc/LICENSE;md5=491f158d09d948466afce85d6f1fe18f \
-                    file://nss/lib/freebl/mpi/doc/LICENSE-MPL;md5=5d425c8f3157dbf212db2ec53d9e5132"
-
-VERSION_DIR = "${@d.getVar('BP').upper().replace('-', '_').replace('.', '_') + '_RTM'}"
-
-SRC_URI = "http://ftp.mozilla.org/pub/mozilla.org/security/nss/releases/${VERSION_DIR}/src/${BP}.tar.gz \
-           file://nss.pc.in \
-           file://signlibs.sh \
-           file://0001-nss-fix-support-cross-compiling.patch \
-           file://nss-no-rpath-for-cross-compiling.patch \
-           file://nss-fix-incorrect-shebang-of-perl.patch \
-           file://disable-Wvarargs-with-clang.patch \
-           file://pqg.c-ULL_addend.patch \
-           file://blank-cert9.db \
-           file://blank-key4.db \
-           file://system-pkcs11.txt \
-           file://nss-fix-nsinstall-build.patch \
-           file://0001-freebl-add-a-configure-option-to-disable-ARM-HW-cryp.patch \
-           "
-
-SRC_URI[md5sum] = "e0366615e12b147cebc136c915baea37"
-SRC_URI[sha256sum] = "185df319775243f5f5daa9d49b7f9cc5f2b389435be3247c3376579bee063ba7"
-
-UPSTREAM_CHECK_URI = "https://developer.mozilla.org/en-US/docs/Mozilla/Projects/NSS/NSS_Releases"
-UPSTREAM_CHECK_REGEX = "NSS_(?P<pver>.+)_release_notes"
-
-inherit siteinfo
-
-TD = "${S}/tentative-dist"
-TDS = "${S}/tentative-dist-staging"
-
-TARGET_CC_ARCH += "${LDFLAGS}"
-
-do_configure_prepend_libc-musl () {
-    sed -i -e '/-DHAVE_SYS_CDEFS_H/d' ${S}/nss/lib/dbm/config/config.mk
-}
-
-do_compile_prepend_class-native() {
-    export NSPR_INCLUDE_DIR=${STAGING_INCDIR_NATIVE}/nspr
-    export NSPR_LIB_DIR=${STAGING_LIBDIR_NATIVE}
-    export NSS_ENABLE_WERROR=0
-}
-
-do_compile_prepend_class-nativesdk() {
-    export LDFLAGS=""
-}
-
-do_compile_prepend_class-native() {
-    # Need to set RPATH so that chrpath will do its job correctly
-    RPATH="-Wl,-rpath-link,${STAGING_LIBDIR_NATIVE} -Wl,-rpath-link,${STAGING_BASE_LIBDIR_NATIVE} -Wl,-rpath,${STAGING_LIBDIR_NATIVE} -Wl,-rpath,${STAGING_BASE_LIBDIR_NATIVE}"
-}
-
-do_compile() {
-    export NSPR_INCLUDE_DIR=${STAGING_INCDIR}/nspr
-
-    export CROSS_COMPILE=1
-    export NATIVE_CC="${BUILD_CC}"
-    # Additional defines needed on Centos 7
-    export NATIVE_FLAGS="${BUILD_CFLAGS} -DLINUX -Dlinux"
-    export BUILD_OPT=1
-
-    export FREEBL_NO_DEPEND=1
-    export FREEBL_LOWHASH=1
-
-    export LIBDIR=${libdir}
-    export MOZILLA_CLIENT=1
-    export NS_USE_GCC=1
-    export NSS_USE_SYSTEM_SQLITE=1
-    export NSS_ENABLE_ECC=1
-
-    ${@bb.utils.contains("TUNE_FEATURES", "crypto", "export NSS_USE_ARM_HW_CRYPTO=1", "", d)}
-
-    export OS_RELEASE=3.4
-    export OS_TARGET=Linux
-    export OS_ARCH=Linux
-
-    if [ "${TARGET_ARCH}" = "powerpc" ]; then
-        OS_TEST=ppc
-    elif [ "${TARGET_ARCH}" = "powerpc64" ]; then
-        OS_TEST=ppc64
-    elif [ "${TARGET_ARCH}" = "mips" -o "${TARGET_ARCH}" = "mipsel" -o "${TARGET_ARCH}" = "mips64" -o "${TARGET_ARCH}" = "mips64el" ]; then
-        OS_TEST=mips
-    elif [ "${TARGET_ARCH}" = "aarch64_be" ]; then
-        OS_TEST="aarch64"
-    else
-        OS_TEST="${TARGET_ARCH}"
-    fi
-
-    if [ "${SITEINFO_BITS}" = "64" ]; then
-        export USE_64=1
-    elif [ "${TARGET_ARCH}" = "x86_64" -a "${SITEINFO_BITS}" = "32" ]; then
-        export USE_X32=1
-    fi
-
-    export NSS_DISABLE_GTESTS=1
-
-    # We can modify CC in the environment, but if we set it via an
-    # argument to make, nsinstall, a host program, will also build with it!
-    #
-    # nss pretty much does its own thing with CFLAGS, so we put them into CC.
-    # Optimization will get clobbered, but most of the stuff will survive.
-    # The motivation for this is to point to the correct place for debug
-    # source files and CFLAGS does that.  Nothing uses CCC.
-    #
-    export CC="${CC} ${CFLAGS}"
-    make -C ./nss CCC="${CXX} -g" \
-        OS_TEST=${OS_TEST} \
-        RPATH="${RPATH}"
-}
-
-do_compile[vardepsexclude] += "SITEINFO_BITS"
-
-do_install_prepend_class-nativesdk() {
-    export LDFLAGS=""
-}
-
-do_install() {
-    export CROSS_COMPILE=1
-    export NATIVE_CC="${BUILD_CC}"
-    export BUILD_OPT=1
-
-    export FREEBL_NO_DEPEND=1
-
-    export LIBDIR=${libdir}
-    export MOZILLA_CLIENT=1
-    export NS_USE_GCC=1
-    export NSS_USE_SYSTEM_SQLITE=1
-    export NSS_ENABLE_ECC=1
-
-    export OS_RELEASE=3.4
-    export OS_TARGET=Linux
-    export OS_ARCH=Linux
-
-    if [ "${TARGET_ARCH}" = "powerpc" ]; then
-        OS_TEST=ppc
-    elif [ "${TARGET_ARCH}" = "powerpc64" ]; then
-        OS_TEST=ppc64
-    elif [ "${TARGET_ARCH}" = "mips" -o "${TARGET_ARCH}" = "mipsel" -o "${TARGET_ARCH}" = "mips64" -o "${TARGET_ARCH}" = "mips64el" ]; then
-        OS_TEST=mips
-    elif [ "${TARGET_ARCH}" = "aarch64_be" ]; then
-        CPU_ARCH=aarch64
-        OS_TEST="aarch64"
-    else
-        OS_TEST="${TARGET_ARCH}"
-    fi
-    if [ "${SITEINFO_BITS}" = "64" ]; then
-        export USE_64=1
-    elif [ "${TARGET_ARCH}" = "x86_64" -a "${SITEINFO_BITS}" = "32" ]; then
-        export USE_X32=1
-    fi
-
-    export NSS_DISABLE_GTESTS=1
-
-    make -C ./nss \
-        CCC="${CXX}" \
-        OS_TEST=${OS_TEST} \
-        SOURCE_LIB_DIR="${TD}/${libdir}" \
-        SOURCE_BIN_DIR="${TD}/${bindir}" \
-        install
-
-    install -d ${D}/${libdir}/
-    for file in ${S}/dist/*.OBJ/lib/*.so; do
-        echo "Installing `basename $file`..."
-        cp $file  ${D}/${libdir}/
-    done
-
-    for shared_lib in ${TD}/${libdir}/*.so.*; do
-        if [ -f $shared_lib ]; then
-            cp $shared_lib ${D}/${libdir}
-            ln -sf $(basename $shared_lib) ${D}/${libdir}/$(basename $shared_lib .1oe)
-        fi
-    done
-    for shared_lib in ${TD}/${libdir}/*.so; do
-        if [ -f $shared_lib -a ! -e ${D}/${libdir}/$shared_lib ]; then
-            cp $shared_lib ${D}/${libdir}
-        fi
-    done
-
-    install -d ${D}/${includedir}/nss3
-    install -m 644 -t ${D}/${includedir}/nss3 dist/public/nss/*
-
-    install -d ${D}/${bindir}
-    for binary in ${TD}/${bindir}/*; do
-        install -m 755 -t ${D}/${bindir} $binary
-    done
-}
-
-do_install[vardepsexclude] += "SITEINFO_BITS"
-
-do_install_append() {
-    # Create empty .chk files for the NSS libraries at build time. They could
-    # be regenerated at target's boot time.
-    for file in libsoftokn3.chk libfreebl3.chk libnssdbm3.chk; do
-        touch ${D}/${libdir}/$file
-        chmod 755 ${D}/${libdir}/$file
-    done
-    install -D -m 755 ${WORKDIR}/signlibs.sh ${D}/${bindir}/signlibs.sh
-
-    install -d ${D}${libdir}/pkgconfig/
-    sed 's/%NSS_VERSION%/${PV}/' ${WORKDIR}/nss.pc.in | sed 's/%NSPR_VERSION%/4.9.2/' > ${D}${libdir}/pkgconfig/nss.pc
-    sed -i s:OEPREFIX:${prefix}:g ${D}${libdir}/pkgconfig/nss.pc
-    sed -i s:OEEXECPREFIX:${exec_prefix}:g ${D}${libdir}/pkgconfig/nss.pc
-    sed -i s:OELIBDIR:${libdir}:g ${D}${libdir}/pkgconfig/nss.pc
-    sed -i s:OEINCDIR:${includedir}/nss3:g ${D}${libdir}/pkgconfig/nss.pc
-}
-
-do_install_append_class-target() {
-    # It used to call certutil to create a blank certificate with empty password at
-    # build time, but the checksum of key4.db changes every time when certutil is called.
-    # It causes non-determinism issue, so provide databases with a blank certificate
-    # which are originally from output of nss in qemux86-64 build. You can get these
-    # databases by:
-    # certutil -N -d sql:/database/path/ --empty-password
-    install -d ${D}${sysconfdir}/pki/nssdb/
-    install -m 0644 ${WORKDIR}/blank-cert9.db ${D}${sysconfdir}/pki/nssdb/cert9.db
-    install -m 0644 ${WORKDIR}/blank-key4.db ${D}${sysconfdir}/pki/nssdb/key4.db
-    install -m 0644 ${WORKDIR}/system-pkcs11.txt ${D}${sysconfdir}/pki/nssdb/pkcs11.txt
-}
-
-PACKAGE_WRITE_DEPS += "nss-native"
-pkg_postinst_${PN} () {
-    if [ -n "$D" ]; then
-        for I in $D${libdir}/lib*.chk; do
-            DN=`dirname $I`
-            BN=`basename $I .chk`
-            FN=$DN/$BN.so
-            shlibsign -i $FN
-            if [ $? -ne 0 ]; then
-                exit 1
-            fi
-        done
-    else
-        signlibs.sh
-    fi
-}
-
-PACKAGES =+ "${PN}-smime"
-FILES_${PN}-smime = "\
-    ${bindir}/smime \
-"
-
-FILES_${PN} = "\
-    ${sysconfdir} \
-    ${bindir} \
-    ${libdir}/lib*.chk \
-    ${libdir}/lib*.so \
-    "
-
-FILES_${PN}-dev = "\
-    ${libdir}/nss \
-    ${libdir}/pkgconfig/* \
-    ${includedir}/* \
-    "
-
-RDEPENDS_${PN}-smime = "perl"
-
-BBCLASSEXTEND = "native nativesdk"
--

2.17.1

--
_______________________________________________
Openembedded-core mailing list
[email protected]
http://lists.openembedded.org/mailman/listinfo/openembedded-core

#3 lore raw
Quoted 8 lines · Khem Raj
Quoted 2 lines · Adrian Bunk

rpm was the last user in OE-core.

we should also assess external dependencies especially on libraries, there might be layers which do not depend on meta-oe but use nss or enable nss packageconfigs in core components like curl.
...

Is providing a crypto library in OE-core without providing security support better than not shipping it?

nss in warrior seems to lack fixes for at least 5 CVEs.

cu
Adrian

#4 lore raw

Adrian,

Quoted 11 lines · Adrian Bunk
Quoted 6 lines · Khem Raj
Quoted 1 line · Adrian Bunk

rpm was the last user in OE-core.

we should also assess external dependencies especially on libraries, there might be layers which do not depend on meta-oe but use nss or enable nss packageconfigs in core components like curl.
...

Is providing a crypto library in OE-core without providing security support better than not shipping it?

nss in warrior seems to lack fixes for at least 5 CVEs.

I don't see how that is relevant to the RFC?

nss package is not moving out for warrior nor zeus.

- armin

Quoted 3 lines

cu
Adrian

#5 lore raw
Quoted 9 lines · On Mon, Feb 24, 2020 at 08:32:24AM -0800, akuster808 wrote:

...

Quoted 7 lines · Adrian Bunk
Quoted 6 lines · Khem Raj
Quoted 1 line · Adrian Bunk

rpm was the last user in OE-core.

we should also assess external dependencies especially on libraries, there might be layers which do not depend on meta-oe but use nss or enable nss packageconfigs in core components like curl.
...

Quoted 7 lines
Quoted 4 lines

Is providing a crypto library in OE-core without providing security support better than not shipping it?

nss in warrior seems to lack fixes for at least 5 CVEs.

I don't see how that is relevant to the RFC?
...

It is a crypto library with a history of unfixed CVEs in supported stable Yocto releases.

Quoted 1 line

- armin

cu
Adrian

#6 lore raw
Quoted 5 lines · Adrian Bunk
Quoted 1 line

...

It is a crypto library with a history of unfixed CVEs in supported stable Yocto releases.

If the issue is unfixed CVEs, then I do not think it's particularly relevant which layer the recipe is in. Stable release maintainers are not expected to 'track and fix CVEs', that one is on users.

Alex

#7 lore raw
Quoted 11 lines · Alexander Kanavin
Quoted 5 lines · Adrian Bunk
Quoted 1 line

...

It is a crypto library with a history of unfixed CVEs in supported stable Yocto releases.

If the issue is unfixed CVEs, then I do not think it's particularly relevant which layer the recipe is in. Stable release maintainers are not expected to 'track and fix CVEs', that one is on users.

Yesterdays LTS announcement makes it clear that the Yocto project does provide regular security updates for supported stable branches:

<-- snip -->

Yocto Project releases are usually maintained for one year. Beyond this period, releases move to community support, which means they only receive occasional patches for critical defects and updates, and no regular defect fixes and security updates.

<-- snip -->

Quoted 1 line

Alex

cu
Adrian

#8 lore raw

You are misinterpreting the announcement. The security updates are provided by users as patches to the mailing list, maintainers merely collect and integrate them. There is no promise from the project to do anything else, and LTS doesn’t change that, it only extends the maintainer duty from one year to two. Moving a recipe in or out of core does not fundamentally change how much attention it gets w.r.t. security fixes.

Alex

Quoted 31 lines · Adrian Bunk
Quoted 11 lines · Alexander Kanavin
Quoted 5 lines · Adrian Bunk
Quoted 1 line

...

It is a crypto library with a history of unfixed CVEs in supported stable Yocto releases.

If the issue is unfixed CVEs, then I do not think it's particularly relevant which layer the recipe is in. Stable release maintainers are not expected to 'track and fix CVEs', that one is on users.

Yesterdays LTS announcement makes it clear that the Yocto project does provide regular security updates for supported stable branches:

<-- snip -->

Yocto Project releases are usually maintained for one year. Beyond this period, releases move to community support, which means they only receive occasional patches for critical defects and updates, and no regular defect fixes and security updates.

<-- snip -->

Quoted 1 line

Alex

cu
Adrian

#9 lore raw
Quoted 6 lines · Alexander Kanavin

You are misinterpreting the announcement. The security updates are provided by users as patches to the mailing list, maintainers merely collect and integrate them. There is no promise from the project to do anything else, and LTS doesn’t change that, it only extends the maintainer duty from one year to two. Moving a recipe in or out of core does not fundamentally change how much attention it gets w.r.t. security fixes.

The part in question does not even talk about LTS, it describes the status quo for the current stable releases with one year support.

Quoted 10 lines

...

Quoted 8 lines

<-- snip -->

Yocto Project releases are usually maintained for one year. Beyond this period, releases move to community support, which means they only receive occasional patches for critical defects and updates, and no regular defect fixes and security updates.

<-- snip -->

...

This announcement states pretty clearly that security updates are provided for stable branches, but not for community supported branches.

I am sure there will be an update to the announcement if this doesn't reflect current reality.

cu
Adrian

#10 lore raw
Quoted 3 lines · Adrian Bunk

I am sure there will be an update to the announcement if this doesn't reflect current reality.

Who is expected to do the actual work of tracking CVEs, making action points and performing the actions? The current reality is this: the security update work is done ad hoc by community, even for stable branches. There is no rigorous security process like in Debian, and no roles to follow in that process. This means that if no one bothers to make a patch, the security issue will remain unfixed, and this does happen often. If you are expecting anything else (e.g. that listed recipe maintainers should do something), you're setting yourself up to be disappointed.

Alex

#11 lore raw

Does YP provide security support for stable and LTS branches?

Quoted 5 lines · Alexander Kanavin
Quoted 2 lines · Adrian Bunk

I am sure there will be an update to the announcement if this doesn't reflect current reality.

Quoted 8 lines

Who is expected to do the actual work of tracking CVEs, making action points and performing the actions? The current reality is this: the security update work is done ad hoc by community, even for stable branches. There is no rigorous security process like in Debian, and no roles to follow in that process. This means that if no one bothers to make a patch, the security issue will remain unfixed, and this does happen often. If you are expecting anything else (e.g. that listed recipe maintainers should do something), you're setting yourself up to be disappointed.

All I am expecting is honesty.

If YP does not provide security support for supported stable branches, then public statements that community support would be worse than stable branches due to lack of security support are dishonest and offensive.

It also puts all users of Yocto stable and LTS releases and billions of devices at danger if the Yocto project announces security support but does not deliver.

The normal user expects that that the announced "usual defect fixes and updates for the extended period of two years" in LTS include the regular security updates that were claimed for stable branches earlier in the same announcement.

For cases where I am the user the only benefit of going through the pain of upgrading existing products from older releases to Yocto 3.1 would be 2 years of security support from upstream. Doing the upgrade and only discovering afterwards that it doesn't bring the benefit that was promised would make me <unprintable>.

Let me repeat that the only thing I am expecting is honesty, and all I am asking for is that if YP does not provide security support for stable and LTS branches this should be communicated clearly so that all users are aware.

Quoted 1 line

Alex

cu
Adrian

#12 lore raw

Re: Does YP provide security support for stable and LTS branches?

Taking offense or getting angry at the yocto project is entirely misdirected. The liability for insecure millions of devices does not lie with the yocto project, it lies with the OEMs. If the OEMs are unwilling to allocate manpower to work on security, there’s very little the yocto project can do. We are already struggling to find people to work on actual bugs, and there’s simply no manpower to do security properly. Still, stable releases *are* much better than community supported ones, they do get a lot of CVEs and bugs fixed, all I wanted to say is that the exhaustive process to fix everything that could be insecure is not there.

Alex

Quoted 50 lines · Adrian Bunk
Quoted 8 lines · Alexander Kanavin
Quoted 2 lines · Adrian Bunk

I am sure there will be an update to the announcement if this doesn't reflect current reality.

Who is expected to do the actual work of tracking CVEs, making action points and performing the actions? The current reality is this: the security update work is done ad hoc by community, even for stable

branches.

Quoted 2 lines

There is no rigorous security process like in Debian, and no roles to follow in that process. This means that if no one bothers to make a

patch,

Quoted 1 line

the security issue will remain unfixed, and this does happen often. If

you

Quoted 1 line

are expecting anything else (e.g. that listed recipe maintainers should

do

Quoted 1 line

something), you're setting yourself up to be disappointed.

All I am expecting is honesty.

If YP does not provide security support for supported stable branches, then public statements that community support would be worse than stable branches due to lack of security support are dishonest and offensive.

It also puts all users of Yocto stable and LTS releases and billions of devices at danger if the Yocto project announces security support but does not deliver.

The normal user expects that that the announced "usual defect fixes and updates for the extended period of two years" in LTS include the regular security updates that were claimed for stable branches earlier in the same announcement.

For cases where I am the user the only benefit of going through the pain of upgrading existing products from older releases to Yocto 3.1 would be 2 years of security support from upstream. Doing the upgrade and only discovering afterwards that it doesn't bring the benefit that was promised would make me <unprintable>.

Let me repeat that the only thing I am expecting is honesty, and all I am asking for is that if YP does not provide security support for stable and LTS branches this should be communicated clearly so that all users are aware.

Quoted 1 line

Alex

cu
Adrian

#13 lore raw

Re: Does YP provide security support for stable and LTS branches?

Quoted 2 lines · Alexander Kanavin

Taking offense or getting angry at the yocto project is entirely misdirected.

I am not angry if YP does not provide security support.

I am angry when YP is telling lies that it would provide security support, but does not actually provide it.

Quoted 3 lines

The liability for insecure millions of devices does not lie with the yocto project, it lies with the OEMs.
...

The liability for insecure millions of devices lies 100% with the Yocto Project if it claims to provide security support but does not actually provide it.

If a user has to decide today whether an upcoming product will run Ubuntu 20.04 LTS or Yocto 3.1 LTS, then it should be clear to the user whether or not choosing Yocto will provide upstream security support the same way as Ubuntu.

A user reading the YP LTS announcement expects security support similar to what Ubuntu is offering, and might only notice that this isn't true after a known vulnerability gets exploited on millions of devices.

If security support for YP stable and LTS releases is only on a community support basis and usually incomplete, then it is on YP to make that clear to all users instead of claiming the opposite - in other projects LTS does include security support, sometimes only security fixes are permitted.

This could be combined with a call for help for security support, an advantage of being honest would be that it becomes visible for users that there is a resource shortage.

Quoted 1 line

Alex

cu
Adrian

#14 lore raw

Re: [Openembedded-architecture] Does YP provide security support for stable and LTS branches?

Adrian,

Quoted 7 lines · Adrian Bunk
Quoted 2 lines · Alexander Kanavin

Taking offense or getting angry at the yocto project is entirely misdirected.

I am not angry if YP does not provide security support.

I am angry when YP is telling lies that it would provide security support, but does not actually provide it.

 I am sure its not the intent of the Yocto Project to miss lead folks.

Quoted 8 lines
Quoted 4 lines

The liability for insecure millions of devices does not lie with the yocto project, it lies with the OEMs.
...

The liability for insecure millions of devices lies 100% with the Yocto Project if it claims to provide security support but does not actually provide it.

Quoted 8 lines

If a user has to decide today whether an upcoming product will run Ubuntu 20.04 LTS or Yocto 3.1 LTS, then it should be clear to the user whether or not choosing Yocto will provide upstream security support the same way as Ubuntu.

A user reading the YP LTS announcement expects security support similar to what Ubuntu is offering, and might only notice that this isn't true after a known vulnerability gets exploited on millions of devices.

I didn't get that out of the announcement. It only reference in the LTS announcement  regarding Security was in context to Community support.

The new new LTS / Stable process wiki does not claim such claims either.

https://wiki.yoctoproject.org/wiki/Stable_Release_and_LTS

Quoted 6 lines

If security support for YP stable and LTS releases is only on a community support basis and usually incomplete, then it is on YP to make that clear to all users instead of claiming the opposite - in other projects LTS does include security support, sometimes only security fixes are permitted.

Can point out where those statements are made? Again, I am sure the Yocto Project would like to know.

Quoted 4 lines

This could be combined with a call for help for security support, an advantage of being honest would be that it becomes visible for users that there is a resource shortage.

There are weekly status reports  and minutes from various meeting that include the attendees list.  The attendees list has not grown. The build swat team was shut down as no one stepped up to help. That was an opportunity for the community to step up.

There is request for help regarding defects sent out every week and if it wasn't for WindRiver, Intel, Joshua Watt and Richard the open defects would go uncheck.

So the resource issue has been floating around on the various mailing lists for some time. What do you think needs to be done to make it even more visible?

regards,
armin

Quoted 8 lines
Quoted 2 lines

Alex

cu
Adrian
_______________________________________________
Openembedded-architecture mailing list
[email protected]
http://lists.openembedded.org/mailman/listinfo/openembedded-architecture

#15 lore raw

Re: [Openembedded-architecture] Does YP provide security support for stable and LTS branches?

On Wed, Mar 04, 2020 at 12:26:29PM -0800, akuster808 wrote:
...

Quoted 1 line · Adrian Bunk

...

Quoted 7 lines
Quoted 3 lines

This could be combined with a call for help for security support, an advantage of being honest would be that it becomes visible for users that there is a resource shortage.

There are weekly status reports  and minutes from various meeting that include the attendees list.  The attendees list has not grown. The build swat team was shut down as no one stepped up to help. That was an opportunity for the community to step up.

Yocto is setup as a B2B project providing a basis for embedded products, your "community" are mainly companies.

The "build everything yourself" niche for non-embedded is already occupied by Gentoo.

For most community companies there is no clear Return on Investment if they would use the opportunity to invest in upstream involvement.

Quoted 8 lines

There is request for help regarding defects sent out every week and if it wasn't for WindRiver, Intel, Joshua Watt and Richard the open defects would go uncheck.

So the resource issue has been floating around on the various mailing lists for some time. What do you think needs to be done to make it even more visible?
...

You are confusing developers with users.

Developers are following development discussions.

Users who are being paid for building a product based on Yocto read the user information at https://www.yoctoproject.org/
especially the great manuals written by Scott.

Development discussions are mostly irrelevant for users.

To make a non-Yocto example:
I am using Ubuntu LTS on my laptop.
The security support provided by Canonical is essential for me, and as part of the community I am using their bugtracker in rare cases where I have problems.
I have neither time nor interest in following upstream discussions regarding development of future Ubuntu releases.

Quoted 13 lines

...

Quoted 12 lines
Quoted 3 lines

The liability for insecure millions of devices does not lie with the yocto project, it lies with the OEMs.
...

The liability for insecure millions of devices lies 100% with the Yocto Project if it claims to provide security support but does not actually provide it.

If a user has to decide today whether an upcoming product will run Ubuntu 20.04 LTS or Yocto 3.1 LTS, then it should be clear to the user whether or not choosing Yocto will provide upstream security support the same way as Ubuntu.

Quoted 6 lines
Quoted 3 lines

A user reading the YP LTS announcement expects security support similar to what Ubuntu is offering, and might only notice that this isn't true after a known vulnerability gets exploited on millions of devices.

I didn't get that out of the announcement. It only reference in the LTS announcement  regarding Security was in context to Community support.
...

The wording is "releases move to community support, which means they only receive occasional patches for critical defects and updates, and no regular defect fixes and security updates".

When the move to community support means no regular security updates, this is a clear claim from YP that before the move there are regular security updates.

YP then claims for LTS that "These components will now receive the usual defect fixes and updates for the extended period of two years."

When YP states "A very important criterion for evaluating and adopting a software platform is support." the one kind of support that really matters for users of long term support releases is security support. This is the baseline users expect from anything called LTS, and the main reason why users want to use LTS releases.

If YP does not want to be responsible for insecure millions of devices, it is up to YP to not make incorrect claims and make it clear in announcements and user documentation if security support is not provided by YP.

This allows users to mitigate by allocating resources for security support of their products, instead of unknowingly shipping millions of insecure devices.

It would also allow YP to develop offers for community companies to pool smaller contributions together - 50 companies each paying 2k per year for pooled security support is cheaper for them than each of them locally providing the same security support.

cu
Adrian

#16 lore raw

Re: [Openembedded-architecture] Does YP provide security support for stable and LTS branches?

Quoted 2 lines · Adrian Bunk

For most community companies there is no clear Return on Investment if they would use the opportunity to invest in upstream involvement.

That isn't true. If you fix something yourself and hold the change you get to maintain it. If you work with upstream you can share the maintenance burden with the community going forward. That is a direct ROI, there are also more indirect benefits.

Quoted 28 lines
Quoted 22 lines
Quoted 16 lines
Quoted 3 lines

The liability for insecure millions of devices does not lie with the yocto project, it lies with the OEMs.
...

The liability for insecure millions of devices lies 100% with the Yocto
Project if it claims to provide security support but does not actually
provide it.

If a user has to decide today whether an upcoming product will run
Ubuntu 20.04 LTS or Yocto 3.1 LTS, then it should be clear to the user whether or not choosing Yocto will provide upstream security
support the same way as Ubuntu.

Quoted 6 lines

A user reading the YP LTS announcement expects security support similar
to what Ubuntu is offering, and might only notice that this isn't true
after a known vulnerability gets exploited on millions of devices.

Quoted 5 lines

I didn't get that out of the announcement. It only reference in the LTS
announcement regarding Security was in context to Community support.
...

Quoted 7 lines

The wording is "releases move to community support, which means they only receive occasional patches for critical defects and updates, and no regular defect fixes and security updates".

When the move to community support means no regular security updates, this is a clear claim from YP that before the move there are regular security updates.

I think you're being rather pedantic here and I'd suggest we go back to what the essence of this announcement is.

Today, our stable branch maintenance is done "ad-hoc" by volunteers. I/we can't ask them to do anything in any given time frame, its a best effort. I *hugely* appreciate what those people do but it has its limitations.

Even with the non-stable side of the project, the patch review/testing/merging and debugging which patch causes which regression basically falls onto one person, me. Yes, people help, its hugely appreciated when they do but I'm the only person paid to do it (amongst many other things) and I therefore know first hand how much work this is.

We want to extend the lifespan of some releases. We can't ask/force our volunteers to do that, its not reasonable. The project is therefore looking to try something different where we have someone with time dedicated to this.

There are certainly limitations to what will likely be one part time person can do, however we are hoping it improves on what we have today and may even set a mechanism by which the project can better operate for key functionality in future.

We're putting this into the context of the maintainer who coordinates things, not someone to write the individual security fixes as that is the only realistic way we can make this work. If people share their security patches they'll have a lot of work, if nobody shares such fixes, it will be minimal. Time will tell if people can/will share. I do think it makes sense to try.

Quoted 2 lines

YP then claims for LTS that "These components will now receive the usual defect fixes and updates for the extended period of two years."

So the current process is extended for a longer period. That seems quite clear.

Quoted 5 lines

When YP states "A very important criterion for evaluating and adopting a software platform is support." the one kind of support that really matters for users of long term support releases is security support. This is the baseline users expect from anything called LTS, and the main reason why users want to use LTS releases.

And we are trying to put something in place which better supports review and merging of patches into that LTS. That is a form of security support. We can't afford a team of 20 to work on it but it is an improvement from our current volunteer best efforts?

We're also not competing with member company offerings. If you want a full set of security guarantees, please use a company which offers that service, several project members will be happy to help (and charge accordingly).

Quoted 4 lines

If YP does not want to be responsible for insecure millions of devices, it is up to YP to not make incorrect claims and make it clear in announcements and user documentation if security support is not provided by YP.

I think the definition of "security support" is arguable to be different things but the intent of what we're trying to do here is clear. No, the person will not write the patches, the intent is to coordinate the maintenance of the branch. If there are huge security holes I would at least expect they can highlight the issues and then coordinate any help in fixing them. That in itself is a level of security support btw.

Quoted 8 lines

This allows users to mitigate by allocating resources for security support of their products, instead of unknowingly shipping millions of insecure devices.

It would also allow YP to develop offers for community companies to pool smaller contributions together - 50 companies each paying 2k per year for pooled security support is cheaper for them than each of them locally providing the same security support.

This is what we're effectively trying to do. The level of "security support" you can do with $100k/year is a maintainer/coordinator of the stable branch. Even writing a spec for that is hard, if you have one we'd love a copy btw.

Taking a step back, please realise that we are trying to improve what we can offer within the realms of what is practical. We can improve the maintenance function, we're not going to be able to write every security fix or provide guarantees.

Its incredibly hard to find funding to try and then organise what we're trying to do, it would be nice if you could try and help us support in doing so.

If we need to clarify how this is documented/presented, we should do so. Accusing people of misleading/lying/false information isn't helpful, or in fact accurate.

Cheers,

Richard

#17 lore raw

Re: [Openembedded-architecture] Does YP provide security support for stable and LTS branches?

Quoted 8 lines · Richard Purdie
Quoted 2 lines · Adrian Bunk

For most community companies there is no clear Return on Investment if they would use the opportunity to invest in upstream involvement.

That isn't true. If you fix something yourself and hold the change you get to maintain it. If you work with upstream you can share the maintenance burden with the community going forward. That is a direct ROI, there are also more indirect benefits.

I was responding to Armin talking about the build swat team. It is hard to see the ROI for a community company when participating in something like that.

Upstreaming of own local changes is a different story.

Quoted 5 lines

...
Today, our stable branch maintenance is done "ad-hoc" by volunteers. I/we can't ask them to do anything in any given time frame, its a best effort. I *hugely* appreciate what those people do but it has its limitations.

On developer lists the message is always "we are short on resources, please help". Which reaches the few people already involved in development.

The message to users is that everything is fine and what new things Yocto is additionally offering.

No regular user of Wikipedia would miss the regular fundraising where the project needs/wants money.

Quoted 11 lines

...

Quoted 7 lines

The wording is "releases move to community support, which means they only receive occasional patches for critical defects and updates, and no regular defect fixes and security updates".

When the move to community support means no regular security updates, this is a clear claim from YP that before the move there are regular security updates.

I think you're being rather pedantic here and I'd suggest we go back to what the essence of this announcement is.

The essence for me is that a few days after I am told on the developer list that 'track and fix CVEs' for stable releases is on users, YP makes an announcement that is worded to make users believe the opposite.

I am getting attacks like being accused of making "toxic accusations" for stating the fact that millions of devices being put at risk by YP misrepresenting the status of security support to users who are building products based on Yocto.

It is on YP to make it clear to users whether or not Yocto comes with the same set of security guarantees as distributions like Ubuntu or Debian. If it is the duty of every user of Yocto to track and fix CVEs, then this has to be stated clearly instead of implying the opposite. This gives users the opportunity to mitigate, instead of unknowingly shipping insecure products.

E.g. embedded products based on Ubuntu with its security support from Linux Foundation member Canonical are clearly better than embedded products based on any distribution without similar security support.

And while you are downloading an Ubuntu image from their website you are on a "Contribute with PayPal" page.

Quoted 6 lines

...

Quoted 4 lines

If YP does not want to be responsible for insecure millions of devices, it is up to YP to not make incorrect claims and make it clear in announcements and user documentation if security support is not provided by YP.

Quoted 7 lines

I think the definition of "security support" is arguable to be different things but the intent of what we're trying to do here is clear. No, the person will not write the patches, the intent is to coordinate the maintenance of the branch. If there are huge security holes I would at least expect they can highlight the issues and then coordinate any help in fixing them. That in itself is a level of security support btw.

This is a bit of a strawman, the problem is elsewhere.

Vulnerabilities are rarely OE/Yocto-specific, in practice security support means applying the same patches that other distributions are also applying.

When a security advisory will be published for Ubuntu 20.04 LTS there will be a patch available for usually approximately the same version of this software as is in Yocto 3.1 LTS.

Going back to where this discussion started:

If YP gives the impression that stable and LTS series are security supported by YP, then it is a problem when a crypto library like nss is not getting any security fixes in Yocto 2.7.

Debian stable ships the same upstream version and applied fixes for 5 CVEs last year. In practice security support means integrating these existing patches. And providing security support means that there is a person resposible for integrating such existing patches from elsewhere.

Quoted 5 lines

...
Its incredibly hard to find funding to try and then organise what we're trying to do, it would be nice if you could try and help us support in doing so.
...

Do you have constructive suggestions how people can help with funding who do not have access to money sources that could contribute 6 digit amounts to upstream Yocto?

Any support I could provide or might be able to organize would be 4 digits per year, and I am not aware of any existing way to pool such contributions easily for paying people for upstream YP work.

The best I could offer would be that I open a company that sends invoices for small contributions and pays people from that money. Which sounds like a clear non-starter for financing YP stable/LTS branch maintainance - this would likely have to be organized through the LF.

I might have use for continued warrior maintainance and might volunteer as community maintainer for that after the 12 months, but this doesn't generate funding for YP.

It is really hard to help you find funding when you are offering opportunities to fund YP only for big companies with huge budgets, but aren't offering any opportunity to give a 1k contribution to YP.

I fully understand that improving on that would be difficult, but this is a reason why most businesses are not able to help you even if they want to.

Quoted 3 lines

Cheers,

Richard

cu
Adrian

#18 lore raw

Re: [Openembedded-architecture] Does YP provide security support for stable and LTS branches?

Quoted 8 lines · Adrian Bunk

It is on YP to make it clear to users whether or not Yocto comes with the same set of security guarantees as distributions like Ubuntu or Debian.
If it is the duty of every user of Yocto to track and fix CVEs, then this has to be stated clearly instead of implying the opposite. This gives users the opportunity to mitigate, instead of unknowingly shipping insecure products.

Do you have any actual evidence for actual users shipping insecure products because they mistakenly believe Yocto takes care of security for them? This has been the situation from the start of the project, certainly this was the case 5 years ago when I joined it, and the only person ever to make an issue out of it is you. Everyone else seems to understand the deal they're getting by using Yocto without a commercial support contract.

Yes, there are millions of insecure yocto-based devices out there, but there reasons they are insecure have nothing to do with what you say.

Alex

#19 lore raw

Re: [Openembedded-architecture] Does YP provide security support for stable and LTS branches?

Quoted 11 lines · Alexander Kanavin
Quoted 8 lines · Adrian Bunk

It is on YP to make it clear to users whether or not Yocto comes with the same set of security guarantees as distributions like Ubuntu or Debian.
If it is the duty of every user of Yocto to track and fix CVEs, then this has to be stated clearly instead of implying the opposite. This gives users the opportunity to mitigate, instead of unknowingly shipping insecure products.

Quoted 2 lines

Do you have any actual evidence for actual users shipping insecure products because they mistakenly believe Yocto takes care of security for them?

Nothing to discuss in public.

Quoted 6 lines

This
has been the situation from the start of the project, certainly this was the case 5 years ago when I joined it, and the only person ever to make an issue out of it is you. Everyone else seems to understand the deal they're getting by using Yocto without a commercial support contract.
...

You are saying that 'track and fix CVEs' is on users. Let's check what YP is telling users.

Click on the "Is Yocto Project for you?" link on the YP frontpage:

https://www.yoctoproject.org/is-yocto-project-for-you/
13. Yocto Project follows a strict release schedule incorporating security patches in all supported releases. This predictability is crucial for projects that are based on Yocto Project and allows the development teams to plan their activities. Developers can choose which Yocto Project branch on which to base their activities as a function of their needs. The development branch will ensure access to the latest features while the stable branches will reduce the pace of changes. CVEs (common vulnerabilities and exposures) issues are supported for the latest 2 releases.

Quoted 1 line

Alex

cu
Adrian

#20 lore raw

Re: [Openembedded-architecture] Does YP provide security support for stable and LTS branches?

Hello,

Quoted 26 lines · Adrian Bunk
Quoted 10 lines · Alexander Kanavin
Quoted 8 lines · Adrian Bunk

It is on YP to make it clear to users whether or not Yocto comes with the same set of security guarantees as distributions like Ubuntu or Debian.
If it is the duty of every user of Yocto to track and fix CVEs, then this has to be stated clearly instead of implying the opposite. This gives users the opportunity to mitigate, instead of unknowingly shipping insecure products.

Quoted 2 lines

Do you have any actual evidence for actual users shipping insecure products because they mistakenly believe Yocto takes care of security for them?

Nothing to discuss in public.

Quoted 6 lines

This
has been the situation from the start of the project, certainly this was the case 5 years ago when I joined it, and the only person ever to make an issue out of it is you. Everyone else seems to understand the deal they're getting by using Yocto without a commercial support contract.
...

You are saying that 'track and fix CVEs' is on users. Let's check what YP is telling users.

Click on the "Is Yocto Project for you?" link on the YP frontpage:

Quoted 10 lines

https://www.yoctoproject.org/is-yocto-project-for-you/
13. Yocto Project follows a strict release schedule incorporating security patches in all supported releases. This predictability is crucial for projects that are based on Yocto Project and allows the development teams to plan their activities. Developers can choose which Yocto Project branch on which to base their activities as a function of their needs. The development branch will ensure access to the latest features while the stable branches will reduce the pace of changes. CVEs (common vulnerabilities and exposures) issues are supported for the latest 2 releases.

Adrian is making a point here, The Yocto Project by claiming that it supports security patches for Stable releases is misleading the Users!

I work with different customers and some of them think that by using and pulling the latest releases they will get the CVEs automatically fixed!

YP should state that CLEARLY! Of course it will impact the choice of going with Yocto or Not ( probably NOT in this case).

Quoted 5 lines
Quoted 3 lines

Alex

cu
Adrian

Mit freundlichen Grüßen / Kind regards

Signature · 11 lines
Ayoub Zaki
Embedded Systems Consultant

Vaihinger Straße 2/1
D-71634 Ludwigsburg


Mobile   : +4917662901545
Email    : [email protected]
Homepage : https://embexus.com
VAT No.  : DE313902634
#21 lore raw

Re: [Openembedded-architecture] Does YP provide security support for stable and LTS branches?

The intent is to say that security patches are eligible for inclusion into stable release updates (while e.g. version updates are not). If the wording is vague, it can be improved. I tend to agree that there has to be a more clear message that users must set up their own security process or pay someone to do it (especially considering that most products involve a lot more 3rd party layers and not just oe-core). Some of the fixing work can be reused from upstream but there is no promise that it covers everything.

Alex

Quoted 80 lines · Ayoub Zaki

Hello,

Quoted 11 lines · Adrian Bunk
Quoted 10 lines · Alexander Kanavin
Quoted 8 lines · Adrian Bunk

It is on YP to make it clear to users whether or not Yocto comes with the same set of security guarantees as distributions like Ubuntu or Debian.
If it is the duty of every user of Yocto to track and fix CVEs, then this has to be stated clearly instead of implying the opposite. This gives users the opportunity to mitigate, instead of unknowingly shipping insecure products.

Quoted 1 line
Quoted 1 line

Do you have any actual evidence for actual users shipping insecure

products

Quoted 6 lines
Quoted 1 line

because they mistakenly believe Yocto takes care of security for them?

Nothing to discuss in public.

Quoted 3 lines

This
has been the situation from the start of the project, certainly this was the case 5 years ago when I joined it, and the only person ever to make

an

Quoted 1 line
Quoted 1 line

issue out of it is you. Everyone else seems to understand the deal

they're

Quoted 7 lines
Quoted 2 lines

getting by using Yocto without a commercial support contract.
...

You are saying that 'track and fix CVEs' is on users. Let's check what YP is telling users.

Click on the "Is Yocto Project for you?" link on the YP frontpage:

Quoted 10 lines

https://www.yoctoproject.org/is-yocto-project-for-you/
13. Yocto Project follows a strict release schedule incorporating security patches in all supported releases. This predictability is crucial for projects that are based on Yocto Project and allows the development teams to plan their activities. Developers can choose which Yocto Project branch on which to base their activities as a function of their needs. The development branch will ensure access to the latest features while the stable branches will reduce the pace of changes. CVEs (common vulnerabilities and exposures) issues are supported for the latest 2 releases.

Adrian is making a point here, The Yocto Project by claiming that it supports security patches for Stable releases is misleading the Users!

I work with different customers and some of them think that by using and pulling the latest releases they will get the CVEs automatically fixed!

YP should state that CLEARLY! Of course it will impact the choice of going with Yocto or Not ( probably NOT in this case).

Quoted 5 lines
Quoted 3 lines

Alex

cu
Adrian

Mit freundlichen Grüßen / Kind regards

--
Ayoub Zaki
Embedded Systems Consultant

Vaihinger Straße 2/1
D-71634 Ludwigsburg

Mobile : +4917662901545
Email : [email protected]
Homepage : https://embexus.com
VAT No. : DE313902634

--
_______________________________________________
Openembedded-core mailing list
[email protected]
http://lists.openembedded.org/mailman/listinfo/openembedded-core

#22 lore raw

Re: [Openembedded-architecture] Does YP provide security support for stable and LTS branches?

In reply to Ayoub Zaki
Quoted 18 lines · Ayoub Zaki
Quoted 16 lines · Adrian Bunk
Quoted 1 line · Alexander Kanavin

On Sun, 8 Mar 2020 at 22:46, Adrian Bunk <[email protected]> wrote:

https://www.yoctoproject.org/is-yocto-project-for-you/
13. Yocto Project follows a strict release schedule incorporating security patches in all supported releases. This predictability is crucial for projects that are based on Yocto Project and allows the development teams to plan their activities. Developers can choose which
Yocto Project branch on which to base their activities as a function of
their needs. The development branch will ensure access to the latest
features while the stable branches will reduce the pace of changes. CVEs
(common vulnerabilities and exposures) issues are supported for the latest 2 releases.

Quoted 7 lines

Adrian is making a point here, The Yocto Project by claiming that it supports security patches for Stable releases is misleading the Users!

I work with different customers and some of them think that by using and pulling the latest releases they will get the CVEs automatically fixed!

If you use our latest codebase then you get the latest usptream versions and hence, yes, you get the fixes. It does depend what "latest" means in the contents of your comments above.

For the stable series we're putting in a commitment to review and include CVE fixes which are submitted to us. We have a pretty good track record of having them submitted and included.

That said, we cannot write a guarantee than all CVE fixes will be included, we're an open source project, not a company with an engineering team or a project with the scale of say Debian.

Quoted 2 lines

YP should state that CLEARLY! Of course it will impact the choice of going with Yocto or Not ( probably NOT in this case).

I agree we have to set realistic expectations, not entirely sure how to do that but hope the above clarifies whilst we try and update the docs/web pages and so on.

Cheers,

Richard

#23 lore raw

Re: [Openembedded-architecture] Does YP provide security support for stable and LTS branches?

In reply to Ayoub Zaki
Quoted 8 lines · Ayoub Zaki

Adrian is making a point here, The Yocto Project by claiming that it supports security patches for Stable releases is misleading the Users!

I work with different customers and some of them think that by using and pulling the latest releases they will get the CVEs automatically fixed!

YP should state that CLEARLY! Of course it will impact the choice of going with Yocto or Not ( probably NOT in this case).

What would the alternative to Yocto be, and what is their security policy? Does e.g. buildroot commit to fixing every known security issue (which is more than just known CVEs) in their releases?

Ross

#24 lore raw

Re: [Openembedded-architecture] Does YP provide security support for stable and LTS branches?

Hi,

Quoted 4 lines · Ross Burton
Quoted 3 lines · Ayoub Zaki

Adrian is making a point here, The Yocto Project by claiming that it supports security patches for Stable releases is misleading the Users!

Quoted 8 lines
Quoted 5 lines

I work with different customers and some of them think that by using and pulling the latest releases they will get the CVEs automatically fixed!

YP should state that CLEARLY! Of course it will impact the choice of going with Yocto or Not ( probably NOT in this case).

What would the alternative to Yocto be, and what is their security policy? Does e.g. buildroot commit to fixing every known security issue (which is more than just known CVEs) in their releases?

Security patches support is definitely for many companies a knock-out criterion. Probably in this case Debian or a commercial OSes like Qnx would be a choice for who can afford it.

Mit freundlichen Grüßen / Kind regards

Signature · 11 lines
Ayoub Zaki
Embedded Systems Consultant

Vaihinger Straße 2/1
D-71634 Ludwigsburg


Mobile   : +4917662901545
Email    : [email protected]
Homepage : https://embexus.com
VAT No.  : DE313902634
#25 lore raw

Re: [Openembedded-architecture] Does YP provide security support for stable and LTS branches?

Quoted 2 lines · Ayoub Zaki

Security patches support is definitely for many companies a knock-out criterion. Probably in this case Debian or a commercial OSes like Qnx would be a choice for who can afford it.

If someone is using Yocto commercially and willing to pay money, then they are entirely welcome to use a OSV such as Mentor/Wind River/Montavista/etc who *will* make firm commitments on security issues.

Ross