It calls icu-config to detect icu support in file makedefs of postfix.
The compile results are different depends on whether icu has been built
already.
postfix should provide Unicode support, so add dependency icu to solve
the problem.
Signed-off-by: Kai Kang <kai.kang@windriver.com>
Signed-off-by: Martin Jansa <Martin.Jansa@gmail.com>
Signed-off-by: Joe MacDonald <joe_macdonald@mentor.com>
Signed-off-by: Armin Kuster <akuster808@gmail.com>
Issue: LIN8-5177
When 'USE_OPENSSL=1' is configured for crda, key2pub.py doesn't
create "keys-ssl.c" with target's word size. Pass 'SITEINFO_BITS'
from building system to key2pub.py.
(LOCAL REV: NOT UPSTREAM) -- sent to oe-devel on 20161130
Signed-off-by: Haiqing Bai <Haiqing.Bai@windriver.com>
Signed-off-by: Martin Jansa <Martin.Jansa@gmail.com>
Signed-off-by: Joe MacDonald <joe_macdonald@mentor.com>
Signed-off-by: Armin Kuster <akuster808@gmail.com>
Fixed when len(TMPDIR) = 410:
aclocal: error: cannot open echo [snip]: Argument list too long
This is becuase it has a lot of m4 files, use relative path for them
can fix the problem.
Signed-off-by: Robert Yang <liezhi.yang@windriver.com>
Signed-off-by: Martin Jansa <Martin.Jansa@gmail.com>
Signed-off-by: Joe MacDonald <joe_macdonald@mentor.com>
Signed-off-by: Armin Kuster <akuster808@gmail.com>
Fixed when len(TMPDIR) = 410:
aclocal: error: cannot open echo [snip]: Argument list too long
This is becuase it has a lot of m4 files, use relative path for them
can fix the problem.
Signed-off-by: Robert Yang <liezhi.yang@windriver.com>
Signed-off-by: Martin Jansa <Martin.Jansa@gmail.com>
Signed-off-by: Joe MacDonald <joe_macdonald@mentor.com>
Signed-off-by: Armin Kuster <akuster808@gmail.com>
If no network connection appears within DefaultTimeoutStartSec
(~ 90s) after startup, nmbd fails to notify systemd and will
therefore get killed.
Signed-off-by: Andreas Oberritter <obi@opendreambox.org>
Signed-off-by: Martin Jansa <Martin.Jansa@gmail.com>
Signed-off-by: Joe MacDonald <joe_macdonald@mentor.com>
Signed-off-by: Armin Kuster <akuster808@gmail.com>
Replace the hardcoded path with base_bindir in init
script to fix the error:
/etc/init.d/samba.sh: line 15: /usr/bin/ps: No such file or directory
/etc/init.d/samba.sh: line 17: /usr/bin/sed: No such file or directory
/etc/init.d/samba.sh: line 16: /usr/bin/grep: No such file or directory
Signed-off-by: Jackie Huang <jackie.huang@windriver.com>
Signed-off-by: Martin Jansa <Martin.Jansa@gmail.com>
Signed-off-by: Joe MacDonald <joe_macdonald@mentor.com>
Signed-off-by: Armin Kuster <akuster808@gmail.com>
Backport a succeeding commit from net-snmp upstream to fix the issue
introduced by commit
<BUG#a2584: Fix snmptrap to use clientaddr from snmp.conf>.
The missing return will cause crash when binding to a non-exist IPv6
address.
Signed-off-by: Li Zhou <li.zhou@windriver.com>
Signed-off-by: Martin Jansa <Martin.Jansa@gmail.com>
Signed-off-by: Joe MacDonald <joe_macdonald@mentor.com>
Signed-off-by: Armin Kuster <akuster808@gmail.com>
Under IPv6 IP-multihomed environment, the socket does not bind to the
clientaddr indicated in snmp.conf when sending snmptrap and it might
choose a random one.
Backport the patch from net-snmp upstream to fix it.
Signed-off-by: Li Zhou <li.zhou@windriver.com>
Signed-off-by: Martin Jansa <Martin.Jansa@gmail.com>
Signed-off-by: Joe MacDonald <joe_macdonald@mentor.com>
Signed-off-by: Armin Kuster <akuster808@gmail.com>
By defining 'listen-address' in the global dnsmasq.conf file we
prevent other instances of dnsmasq from using the global config as we
will get an error that the address is already in use. Instead we pass
'local-service' when we start the main dnsmasq instance which will
listen on the loopback address as well as any other local
subnets. This will allow the main dnsmasq instance to play nice with
other instances we might start for network-manager, libvirt, lxc...
Signed-off-by: Mark Asselstine <mark.asselstine@windriver.com>
Signed-off-by: Martin Jansa <Martin.Jansa@gmail.com>
Signed-off-by: Joe MacDonald <joe_macdonald@mentor.com>
Signed-off-by: Armin Kuster <akuster808@gmail.com>
The dnsmasq package installs a ${sysconfigdir}/dnsmasq.d directory but
doesn't currently use any configuration files that might be placed
there. This goes against what the user would expect and doesn't match
what you see in other distros.
We could ensure this config directory is read when dnsmasq is started,
as we are doing here, or by including it in the main dnsmasq.conf
file. By doing it here we can have better reuseability of the main
dnsmasq.conf file by other instances of dnsmasq.
There are currently no files added to the dnsmasq.d directory for core
oe images, so this should not result in any runtime behavior changes.
Signed-off-by: Mark Asselstine <mark.asselstine@windriver.com>
Signed-off-by: Martin Jansa <Martin.Jansa@gmail.com>
Signed-off-by: Joe MacDonald <joe_macdonald@mentor.com>
Signed-off-by: Armin Kuster <akuster808@gmail.com>
curlpp is configured with boost by default. And it searches boost
directory on build machine. If boost is installed on build machine, then
causes qa issue:
| ERROR: curlpp-0.7.3-r0 do_configure: This autoconf log indicates
| errors, it looked at host include and/or library paths while
| determining system capabilities.
Pass the boost directory in ${STAGING_DIR_HOST} to configure script to
disable search on build machine to fix the issue.
Signed-off-by: Kai Kang <kai.kang@windriver.com>
Signed-off-by: Martin Jansa <Martin.Jansa@gmail.com>
Signed-off-by: Joe MacDonald <joe_macdonald@mentor.com>
Signed-off-by: Armin Kuster <akuster808@gmail.com>
Fix file-rdeps QA warning by replace shell interpreter of script
${bindir}/onnode with /bin/sh that it doesn't has any bashism.
File ${bindir}/ctdb_run_cluster_tests which is a link to
${bindir}/ctdb_run_tests is split into package samba. That causes wrong
runtime dependency that samba depends on samba-ctdb-tests. Split the
file to samba-ctdb-tests to fix the issue.
Script ctdb_run_tests calls command 'getopt', so add runtime dependency
util-linux-getopt. Install config file and test cases for
samba-ctdt-tests at same time.
Signed-off-by: Kai Kang <kai.kang@windriver.com>
Signed-off-by: Martin Jansa <Martin.Jansa@gmail.com>
Signed-off-by: Joe MacDonald <joe_macdonald@mentor.com>
Signed-off-by: Armin Kuster <akuster808@gmail.com>
Fix the QA warning:
WARNING: tsocks-1.8beta5-r0 do_package_qa: QA Issue: No GNU_HASH in the elf binary: '/opt/build-intel-x86/tmp/work/core2-64-linux/tsocks/1.8beta5-r0/packages-split/tsocks/usr/lib64/libtsocks.so.1.8' [ldflags]
Signed-off-by: Jason Wessel <jason.wessel@windriver.com>
Signed-off-by: Martin Jansa <Martin.Jansa@gmail.com>
Signed-off-by: Joe MacDonald <joe_macdonald@mentor.com>
Signed-off-by: Armin Kuster <akuster808@gmail.com>
The configure is not generated by autotools, and it's
designed to enable ipv6 by default and have option
--disable-ipv6 to disable it, but no --enable-ipv6,
so remove it.
Signed-off-by: Jackie Huang <jackie.huang@windriver.com>
Signed-off-by: Martin Jansa <Martin.Jansa@gmail.com>
Signed-off-by: Joe MacDonald <joe_macdonald@mentor.com>
Signed-off-by: Armin Kuster <akuster808@gmail.com>
The --with-readline means check readline automatically, and will cause build
failures when there is /usr/include/readline:
ERROR: This autoconf log indicates errors, it looked at host include and/or library paths while determining system capabilities.
Use --with-readline=${STAGING_INCDIR}/.. to fix the problem. And also add
--with-readline=no when PACKAGECONFIG is not set for readline.
Signed-off-by: Robert Yang <liezhi.yang@windriver.com>
Signed-off-by: Martin Jansa <Martin.Jansa@gmail.com>
Signed-off-by: Joe MacDonald <joe_macdonald@mentor.com>
(cherry picked from commit 06d15756bd)
Signed-off-by: Armin Kuster <akuster808@gmail.com>
1) Upgrade drbd-utils from 8.9.3 to 8.9.6.
2) Delete one patch, since the data has been changed.
0001-Makefile.in-don-t-compile-documentation.patch
3) Modify EXTRA_OECONF, since the documentation file is not suitable for cross-compile, so we do not compile it.
Signed-off-by: Wang Xin <wangxin2015.fnst@cn.fujitsu.com>
Signed-off-by: Martin Jansa <Martin.Jansa@gmail.com>
Highlight:
- fix CVE-2016-5180
Other changes, from https://c-ares.haxx.se/changelog.html
c-ares version 1.12.0 - Sep 29 2016
Changes:
api: add ARES_OPT_NOROTATE optmask value
Bug fixes:
CVE-2016-5180: ares_create_query single byte out of buffer write
configure: acknowledge --disable-tests [1]
man pages: fix typos detected by Lintian
test: add missing #includes for dns-proto.cc
test: avoid in6addr_* constants
test: Build with MinGW on AppVeyor
Makefile.m32: add support for extra flags
Makefile.m32: add support for CROSSPREFIX
configure: check if tests can get built before enabled
ares_library_cleanup: reset ares_realloc too
ahost.c: add cast to fix C++ compile
test: Only pass unused args to GoogleTest
build: commonize MSVC version detection
msvc_ver.inc: support Visual Studio 2015 Update 1, 2, 3
test: for AF_UNSPEC, return CNAME only for AAAA, but valid A record
ares_getnameinfo: explicitly clear struct servent before use
test: Update fuzzing function prototype
init: fix nsort initialization
test: add fuzzing check script to tests
web: http => https
read_tcp_data: remove superfluous NULL check
LICENSE.md: add a stand-alone license file
SECURITY.md: suggested "security process" for the project
ares_init_options: only propagate init failures from options [2]
headers: remove checks for and defines of variable sizes
test: fix gMock to work with gcc >= 6.x [3]
fix metadata:
We should set SUMMARY instead of DESCRIPTION.
Signed-off-by: André Draszik <git@andred.net>
Signed-off-by: Martin Jansa <Martin.Jansa@gmail.com>
Signed-off-by: Joe MacDonald <joe_macdonald@mentor.com>
With linux kernel v4.8, for file ./src/maint/zntune.c, glibc time.h is
included before linux time.h, when compile zntune.c, it break down by
the following errors:
redefinition of 'struct timespec' 'struct timeval'
'struct timezone' 'struct itimerval'
We should exclude the linux time.h by disable linux/atm_zatm.h and move
some useful definition in linux/atm_zatm.h to a new header file in linux-atm
source code to resolve it.
Signed-off-by: Dengke Du <dengke.du@windriver.com>
Signed-off-by: Martin Jansa <Martin.Jansa@gmail.com>
Signed-off-by: Joe MacDonald <joe_macdonald@mentor.com>
A nice tool to directly read, write and interpret ethernet PHY data.
Very useful when debugging PHY or MDIO problems, which ethtool does
not do.
Signed-off-by: Mike Looijmans <mike.looijmans@topic.nl>
Signed-off-by: Martin Jansa <Martin.Jansa@gmail.com>
Signed-off-by: Joe MacDonald <joe_macdonald@mentor.com>
* Upgrade geoip-perl from 1.45 to 1.50
* Update LIC_FILES_CHKSUM to check LICENSE file instead
of the previous README as no README file in 1.50
Signed-off-by: Mingli Yu <Mingli.Yu@windriver.com>
Signed-off-by: Martin Jansa <Martin.Jansa@gmail.com>
Signed-off-by: Joe MacDonald <joe_macdonald@mentor.com>
Solve the potential issue for previous commit
<arptables: add arptables systemd service file>: in a small build where
/etc/sysconfig directory doesn't exist, the action of adding default
configuration file (/etc/sysconfig/arptables) will fail.
Signed-off-by: Li Zhou <li.zhou@windriver.com>
Signed-off-by: Martin Jansa <Martin.Jansa@gmail.com>
Signed-off-by: Joe MacDonald <joe_macdonald@mentor.com>
Add arptables systemd service file.
If there isn't any configuration file (/etc/sysconfig/arptables),
add a default configuration file.
Signed-off-by: Li Zhou <li.zhou@windriver.com>
Signed-off-by: Martin Jansa <Martin.Jansa@gmail.com>
Signed-off-by: Joe MacDonald <joe_macdonald@mentor.com>
The below changes in kernel source, triggered iscsitarget
build fail with linux kernel v4.8.
1. An extra 'flags' argument has been passed to vfs_readv/vfs_writev
syscalls in v4.8. So, set this argument to "0" for now
(as there is no real need for that).
Ref: https://github.com/torvalds/linux/commit/793b80ef14af56d20c998265287648ad34239b6f
Solves:
-- snip --
TOPDIR/tmp-glibc/work/qemux86-oe-linux/iscsitarget/1.4.20.3+svn502-r0/iscsitarget-1.4.20.3+svn502/kernel/nthread.c: In function 'write_data':
TOPDIR/tmp-glibc/work/qemux86-oe-linux/iscsitarget/1.4.20.3+svn502-r0/iscsitarget-1.4.20.3+svn502/kernel/nthread.c:350:9: error: too few arguments to function 'vfs_writev'
res = vfs_writev(file, (struct iovec __user *) iop, count, &off);
^~~~~~~~~~
-- snip --
2. Redefine dropped PAGE_CACHE_* and page_cache_{get,release} definitions,
as they have been dropped with v4.8
Ref: https://github.com/torvalds/linux/commit/1fa64f198b9f8d6ec0f7aec7c18dc94684391140
Solves:
-- snip --
TOPDIR/tmp-glibc/work/qemux86-oe-linux/iscsitarget/1.4.20.3+svn502-r0/iscsitarget-1.4.20.3+svn502/kernel/param.c: In function 'sess_param_check':
TOPDIR/tmp-glibc/work/qemux86-oe-linux/iscsitarget/1.4.20.3+svn502-r0/iscsitarget-1.4.20.3+svn502/kernel/param.c:48:41: error: 'PAGE_CACHE_SIZE' undeclared (first use in this function)
(u32) ((ISCSI_CONN_IOV_MAX - 1) * PAGE_CACHE_SIZE));
^
-- snip --
3. Replace crypto_hash interfaces with crypto_ahash interfaces,
Ref: https://github.com/torvalds/linux/commit/896545098777564212b9e91af4c973f094649aa7
Ref: https://www.redhat.com/archives/dm-devel/2016-January/msg00244.html
Solves:
-- snip --
TOPDIR/tmp-glibc/work/qemux86-oe-linux/iscsitarget/1.4.20.3+svn502-r0/iscsitarget-1.4.20.3+svn502/kernel/digest.c: In function 'digest_init':
TOPDIR/tmp-glibc/work/qemux86-oe-linux/iscsitarget/1.4.20.3+svn502-r0/iscsitarget-1.4.20.3+svn502/kernel/digest.c:42:23: error: implicit declaration of function 'crypto_alloc_hash' [-Werror=implicit-function-declaration]
conn->rx_hash.tfm = crypto_alloc_hash("crc32c", 0,
^~~~~~~~~~~~~~~~~
TOPDIR/tmp-glibc/work/qemux86-oe-linux/iscsitarget/1.4.20.3+svn502-r0/iscsitarget-1.4.20.3+svn502/kernel/digest.c: In function 'digest_cleanup':
TOPDIR/tmp-glibc/work/qemux86-oe-linux/iscsitarget/1.4.20.3+svn502-r0/iscsitarget-1.4.20.3+svn502/kernel/digest.c:77:3: error: implicit declaration of function 'crypto_free_hash' [-Werror=implicit-function-declaration]
crypto_free_hash(conn->tx_hash.tfm);
^~~~~~~~~~~~~~~~
-- snip --
4. The earlier "rw" parameter has been set in "bi_rw" within
bio structure, hence remove "rw" argument.
Ref: https://github.com/torvalds/linux/commit/4e49ea4a3d276365bf7396c9b77b4d1d5923835a
Solves:
-- snip --
TOPDIR/tmp-glibc/work/qemux86-oe-linux/iscsitarget/1.4.20.3+svn502-r0/iscsitarget-1.4.20.3+svn502/kernel/block-io.c:140:14: warning: passing argument 1 of 'submit_bio' makes pointer from integer without a cast [-Wint-conversion]
submit_bio(rw, bio);
^~
-- snip --
5. The 'len' argument from sk_data_ready() callback has been removed in
linux kernel v4.3 and above.
Ref: https://github.com/torvalds/linux/commit/676d23690fb62b5d51ba5d659935e9f7d9da9f8e
Solves:
-- snip --
TOPDIR/tmp-glibc/work/qemux86-oe-linux/iscsitarget/1.4.20.3+svn502-r0/iscsitarget-1.4.20.3+svn502/kernel/conn.c: In function 'iet_socket_bind':
TOPDIR/tmp-glibc/work/qemux86-oe-linux/iscsitarget/1.4.20.3+svn502-r0/iscsitarget-1.4.20.3+svn502/kernel/conn.c:143:38: error: assignment from incompatible pointer type [-Werror=incompatible-pointer-types]
target->nthread_info.old_data_ready = conn->sock->sk->sk_data_ready;
^
TOPDIR/tmp-glibc/work/qemux86-oe-linux/iscsitarget/1.4.20.3+svn502-r0/iscsitarget-1.4.20.3+svn502/kernel/conn.c:144:32: error: assignment from incompatible pointer type [-Werror=incompatible-pointer-types]
conn->sock->sk->sk_data_ready = iet_data_ready;
^
-- snip --
6.
A. The "size" argument has been dropped from sock_recvmsg syscall in v4.8,
as all callers have it equal to msg_data_left(msg).
B. 'struct user_msghdr' is being used for userland-side msghdr instead
of 'struct msghdr', which is used for kernel-side msghdr in linux v3.19
and above, so typecase it while calling sock_recvmsg syscall.
Ref: https://github.com/torvalds/linux/commit/2da62906b1e298695e1bb725927041cd59942c98https://github.com/torvalds/linux/commit/666547ff591cebdedc4679bf6b1b3f3383a8dea3
Solves:
-- snip --
TOPDIR/tmp-glibc/work-shared/qemux86/kernel-source/include/linux/net.h:222:5: note: expected 'struct msghdr *' but argument is of type 'struct user_msghdr *'
int sock_recvmsg(struct socket *sock, struct msghdr *msg, int flags);
^~~~~~~~~~~~
TOPDIR/tmp-glibc/work/qemux86-oe-linux/iscsitarget/1.4.20.3+svn502-r0/iscsitarget-1.4.20.3+svn502/kernel/nthread.c:129:8: error: too many arguments to function 'sock_recvmsg'
res = sock_recvmsg(conn->sock, &msg, len, MSG_DONTWAIT | MSG_NOSIGNAL);
^~~~~~~~~~~~
-- snip --
Detailed error log is at: http://errors.yoctoproject.org/Errors/Details/83334/
Signed-off-by: Jagadeesh Krishnanjanappa <jkrishnanjanappa@mvista.com>
Signed-off-by: Martin Jansa <Martin.Jansa@gmail.com>
Signed-off-by: Joe MacDonald <joe_macdonald@mentor.com>
Kai Kang
Zheng Ruoqin
Chen Qi
Haiqing Bai
Robert Yang
Andreas Oberritter
Joe Slater
Jackie Huang
Khem Raj
Li Zhou
Mark Asselstine
Jason Wessel
Armin Kuster
Yi Zhao
Wang Xin
André Draszik
dengke.du@windriver.com
mike.looijmans@topic.nl
Oliver Graute
Ming Liu
Mingli Yu
Jagadeesh Krishnanjanappa