qemu now requires gcc >= 7.5, which was the last release in the 7.x
series. We do not have symbols for a gcc dot-version dependency, nor do
we want to add one.
So, add a dependency on gcc >= 8, for both the host and target variants.
In addition:
- Add a missing comment in package/qemu/Config.in.host explaining the
host gcc requirements if host gcc < 8.
Fixes:
http://autobuild.buildroot.org/results/32b7fee1f8cda2290fd4bd8ac9fe78bacb25b652
Signed-off-by: Adam Duskett <aduskett@gmail.com>
[yann.morin.1998@free.fr:
- target qemu also impacted
- propagate to libvirt, gst1-python
- reword commit log that 7.5 is the last 7.x release
]
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
Build of eap-sim-file or eap-simaka-reauth fails without eap-sim since
commit 89f20c1c3f:
make[6]: *** No rule to make target '../../../../src/libsimaka/libsimaka.la', needed by 'libstrongswan-eap-sim-file.la'. Stop.
make[6]: *** No rule to make target '../../../../src/libsimaka/libsimaka.la', needed by 'libstrongswan-eap-simaka-reauth.la'. Stop.
So select eap-sim or eap-aka if needed to enable build of simaka library
Fixes:
- http://autobuild.buildroot.org/results/0c550721c20946e5b9d07951a9698b88e9c53e7f
- http://autobuild.buildroot.org/results/354a022f789067edfe7ebd464991d1695a697a7a
Signed-off-by: Fabrice Fontaine <fontaine.fabrice@gmail.com>
Signed-off-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>
python2 support has been dropped since version 1.10.0 and
69fd5a3ed2
resulting in the following build failure since bump to version 2.0.1 in
commit 6703b94f24:
Traceback (most recent call last):
File "setup.py", line 22, in <module>
from html.parser import HTMLParser
ImportError: No module named html.parser
Fixes:
- http://autobuild.buildroot.org/results/fe011ccb7eac1be9762919f9feda35114e752d8a
Signed-off-by: Fabrice Fontaine <fontaine.fabrice@gmail.com>
Signed-off-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>
Drop second patch (already in version)
Signed-off-by: Fabrice Fontaine <fontaine.fabrice@gmail.com>
Signed-off-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>
python 2 support has been dropped since version 21.0 resulting in the
following build failure since bump to version 21.2.4 in commit
3491f34d9e:
File "setup.py", line 7
def read(rel_path: str) -> str:
^
Fixes:
- http://autobuild.buildroot.org/results/ed74f4ca96332454af329fac23505000c379d4a3
Signed-off-by: Fabrice Fontaine <fontaine.fabrice@gmail.com>
Signed-off-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>
Drop python3-pip now that python3-cryptography has been removed
Signed-off-by: Fabrice Fontaine <fontaine.fabrice@gmail.com>
Signed-off-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>
python3-cryptography is not needed anymore now that python-cryptography
has a host variant
Signed-off-by: Fabrice Fontaine <fontaine.fabrice@gmail.com>
Signed-off-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>
Since commit 0c7e30b43a,
python-cryptography depends on python3. So python3-cryptography became
redundant at that point. As python3-cryptography is a host-only package,
while python-cryptography is a python-only package, add a host variant
to python-cryptography to be able to drop python3-cryptography.
Update ca-certificates to use host-python-cryptography instead of
host-python3-cryptography.
While at it, drop host-python3-pip and host-python3-six as these
dependencies are not needed. Indeed, pip was never a dependency of
cryptography and six is not a dependency with python3 and will be
dropped in version 3.4 and
e66db8079d
Signed-off-by: Fabrice Fontaine <fontaine.fabrice@gmail.com>
Signed-off-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>
squash! package/python-cryptography: add host variant
Commit 33c69c02fb forgot to propagate
host gcc >= 7 dependency to reverse dependencies of host-qemu resulting
in the following build failure:
ERROR: You need at least GCC v7.5 or Clang v6.0 (or XCode Clang v10.0)
Fixes:
- http://autobuild.buildroot.org/results/14cd0250d55c5ca5dc1d297aee1b927fda889531
Signed-off-by: Fabrice Fontaine <fontaine.fabrice@gmail.com>
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
- Remove upstreamed patches
- Add BR2_PACKAGE_PYTHON_CERTIFI and BR2_PACKAGE_QHULL as dependencies
- make DEPENDENCIES one per line and sort alphabetically
- Add a new file: setup.cfg. This file is needed to force matplotlib to use
the system-provided freetype and qhull, and to disable lto.
The setup.cfg file is copied to the source directory before configuring.
LTO must be disabled or else compile errors such as:
"Relocation R_AARCH64_ADR_PREL_PG_HI21 against symbol `_ZSt3hexRSt8ios_base'
which may bind externally can not be used when making a shared object;
recompile with -fPIC"
Signed-off-by: Adam Duskett <aduskett@gmail.com>
Signed-off-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>
LICENSE copyright year was updated.
Signed-off-by: Adam Duskett <aduskett@gmail.com>
Signed-off-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>
Replace BR2_PYTHON3_HOST_DEPENDENCY by host-python3 to avoid the
following build failure raised since bump to version 1.17 in commit
32b2a03745:
Traceback (most recent call last):
File "../py/makeversionhdr.py", line 117, in <module>
make_version_header(sys.argv[1])
File "../py/makeversionhdr.py", line 79, in make_version_header
info = get_version_info_from_docs_conf()
File "../py/makeversionhdr.py", line 67, in get_version_info_from_docs_conf
for line in f:
File "/usr/lib/python3.5/encodings/ascii.py", line 26, in decode
return codecs.ascii_decode(input, self.errors)[0]
UnicodeDecodeError: 'ascii' codec can't decode byte 0xc2 in position 2154: ordinal not in range(128)
Indeed, this build failure is not reproduced by host-python3 (in version
3.9.7) so it is probably an issue with python 3.5 (which is 6-years old).
Fixes:
- http://autobuild.buildroot.org/results/96f7b81f5d14715b0b6673a8016a31e9e1552d0b
Signed-off-by: Fabrice Fontaine <fontaine.fabrice@gmail.com>
Reviewed-by: Chris Packham <judge.packham@gmail.com>
Signed-off-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>
- disable doc generation (via sphinx) for host build
Reduces host-qemu build time from (on a system with sphinx installed):
real 2m5,522s
user 9m41,292s
sys 1m9,732s
to:
real 1m9,183s
user 8m40,131s
sys 1m9,533s
Signed-off-by: Peter Seiderer <ps.report@gmx.net>
Signed-off-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>
Seen with the IconvGNU transcoder when parsing "<aaa.xsdopengis.net/gml\x96".
The reason is that XMLString::transcode(repText2, manager) throws a TranscodingException
which causes the tmp1 string to leak.
Upstream: 1bdf6d8ba8
Signed-off-by: Matthew Weber <matthew.weber@collins.com>
Signed-off-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>
- Drop second patch (already in version)
- Update hash of README (changes not related to license)
https://brltty.app/doc/ChangeLog.txt
Signed-off-by: Fabrice Fontaine <fontaine.fabrice@gmail.com>
Signed-off-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>
gnuradio suffers from gcc bug 43744 but gqrx that depends on it doesn't
take into account the gcc bug. So let's add it as:
'depends on !BR2_TOOLCHAIN_HAS_GCC_BUG_43744 # gnuradio'
Signed-off-by: Giulio Benetti <giulio.benetti@benettiengineering.com>
Signed-off-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>
There seems to still be a similar parallel build issue, but with this
patch applied, it occurs much less frequently: from a 1/3rf failure
rate, I am now experiencing failures under the 1/10th mark.
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
Cc: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
Cc: Kory Maincent <kory.maincent@bootlin.com>
Reviewed-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
Fixes the following security issues:
16.15.0:
- ASTERISK-29057: pjsip: Crash on call rejection during high load
16.15.1:
- AST-2020-003: Remote crash in res_pjsip_diversion
A crash can occur in Asterisk when a SIP message is received that has a
History-Info header, which contains a tel-uri.
https://downloads.asterisk.org/pub/security/AST-2020-003.pdf
- AST-2020-004: Remote crash in res_pjsip_diversion
A crash can occur in Asterisk when a SIP 181 response is received that has
a Diversion header, which contains a tel-uri.
https://downloads.asterisk.org/pub/security/AST-2020-004.pdf
16.16.0:
- ASTERISK-29219: res_pjsip_diversion: Crash if Tel URI contains History-Info
16.16.1:
- AST-2021-001: Remote crash in res_pjsip_diversion
If a registered user is tricked into dialing a malicious number that sends
lots of 181 responses to Asterisk, each one will cause a 181 to be sent
back to the original caller with an increasing number of entries in the
“Supported” header. Eventually the number of entries in the header
exceeds the size of the entry array and causes a crash.
https://downloads.asterisk.org/pub/security/AST-2021-001.pdf
- AST-2021-002: Remote crash possible when negotiating T.38
When re-negotiating for T.38 if the initial remote response was delayed
just enough Asterisk would send both audio and T.38 in the SDP. If this
happened, and the remote responded with a declined T.38 stream then
Asterisk would crash.
https://downloads.asterisk.org/pub/security/AST-2021-002.pdf
- AST-2021-003: Remote attacker could prematurely tear down SRTP calls
An unauthenticated remote attacker could replay SRTP packets which could
cause an Asterisk instance configured without strict RTP validation to
tear down calls prematurely.
https://downloads.asterisk.org/pub/security/AST-2021-003.pdf
- AST-2021-004: An unsuspecting user could crash Asterisk with multiple
hold/unhold requests
Due to a signedness comparison mismatch, an authenticated WebRTC client
could cause a stack overflow and Asterisk crash by sending multiple
hold/unhold requests in quick succession.
https://downloads.asterisk.org/pub/security/AST-2021-004.pdf
- AST-2021-005: Remote Crash Vulnerability in PJSIP channel driver
Given a scenario where an outgoing call is placed from Asterisk to a
remote SIP server it is possible for a crash to occur.
https://downloads.asterisk.org/pub/security/AST-2021-005.pdf
16.16.2:
- AST-2021-006: Crash when negotiating T.38 with a zero port
When Asterisk sends a re-invite initiating T.38 faxing and the endpoint
responds with a m=image line and zero port, a crash will occur in
Asterisk.
This is a reoccurrence of AST-2019-004.
https://downloads.asterisk.org/pub/security/AST-2021-006.pdf
16.17.0:
- ASTERISK-29203 / AST-2021-002 — Another scenario is causing a crash
- ASTERISK-29260: sRTP Replay Protection ignored; even tears down long calls
- ASTERISK-29227: res_pjsip_diversion: sending multiple 181 responses causes
memory corruption and crash
16.19.1:
- AST-2021-007: Remote Crash Vulnerability in PJSIP channel driver
When Asterisk receives a re-INVITE without SDP after having sent a BYE
request a crash will occur. This occurs due to the Asterisk channel no
longer being present while code assumes it is.
https://downloads.asterisk.org/pub/security/AST-2021-007.pdf
- AST-2021-008: Remote crash when using IAX2 channel driver
If the IAX2 channel driver receives a packet that contains an unsupported
media format it can cause a crash to occur in Asterisk.
https://downloads.asterisk.org/pub/security/AST-2021-008.pdf
- AST-2021-009: pjproject/pjsip: crash when SSL socket destroyed during
handshake
Depending on the timing, it’s possible for Asterisk to crash when using a
TLS connection if the underlying socket parent/listener gets destroyed
during the handshake.
https://downloads.asterisk.org/pub/security/AST-2021-009.pdf
16.20.0:
- ASTERISK-29415: Crash in PJSIP TLS transport
- ASTERISK-29381: chan_pjsip: Remote denial of service by an authenticated
user
In addition, a large number of bugfixes.
Drop now upstreamed
0006-AC_HEADER_STDC-causes-a-compile-failure-with-autoconf-2-70.patch.
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
Bump to Linux kernel 5.14.13 and u-boot 2021.10.
Signed-off-by: Giulio Benetti <giulio.benetti@benettiengineering.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
Bump to Linux kernel 5.14.13 and u-boot 2021.10.
Signed-off-by: Giulio Benetti <giulio.benetti@benettiengineering.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
Bump to Linux kernel 5.14.13 and u-boot 2021.10.
Signed-off-by: Giulio Benetti <giulio.benetti@benettiengineering.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
Bump to Linux kernel 5.14.13 and u-boot 2021.10.
Signed-off-by: Giulio Benetti <giulio.benetti@benettiengineering.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
Bump to Linux kernel 5.14.13 and u-boot 2021.10
Signed-off-by: Giulio Benetti <giulio.benetti@benettiengineering.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
Bump to Linux kernel 5.14.13 and u-boot 2021.10.
Signed-off-by: Giulio Benetti <giulio.benetti@benettiengineering.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
This bumps to linux 5.14.13, linux-headers 5.14 and u-boot version
2021.10 and drop local uboot patch where CONFIG_SYS_BOOTM_LEN was set to
16MB because on u-boot 2021.10 it's now set to 64MB as default to make
room for Linux images.
Signed-off-by: Giulio Benetti <giulio.benetti@benettiengineering.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
This bumps to linux 5.14.13, linux-headers 5.14 and u-boot version
2021.10 and drop local uboot patch where CONFIG_SYS_BOOTM_LEN was set to
16MB because on u-boot 2021.10 it's now set to 64MB as default to make
room for Linux images.
Signed-off-by: Giulio Benetti <giulio.benetti@benettiengineering.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
- Fix CVE-2021-41159: Improper client input validation for gateway
connections allows to overwrite memory
- Fix CVE-2021-41160: Improper region checks in all clients allow out of
bound write to memory
https://github.com/FreeRDP/FreeRDP/releases/tag/2.4.1
Signed-off-by: Fabrice Fontaine <fontaine.fabrice@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
- bump version to 20211016
- add host-python3-cryptography dependency (needed since upstream
commit [1])
[1] 8033d52259
Signed-off-by: Peter Seiderer <ps.report@gmx.net>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
- the last version bump of package/python-cffi (790c10d) ignored
the comment 'Please keep in sync with package/python3-cffi/python3-cffi.mk',
so catch up now
Fixes:
ERROR: No hash found for cffi-1.14.2.tar.gz
Signed-off-by: Peter Seiderer <ps.report@gmx.net>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
Switch site to github to get latest release
https://github.com/festvox/flite/releases/tag/v2.2
Signed-off-by: Fabrice Fontaine <fontaine.fabrice@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
