Bump to the v9 bugfix release that explicitly fails whenever mount
points are found in /lib* or /usr/lib*. While this is far from perfect,
it prevents unexpected failures later on.
Bug: https://bugs.gentoo.org/643194
Includes 105 patches taken from WX_3_0_BRANCH, plus one from master only that we
need to not litter DT_NEEDED with wxWebView library without --as-needed (it was
part of squashed wxGTK-3.0.2.0-webview-fixes.patch before). Also includes my patch
to install wx/evtloopsrc.h in core-only builds.
Most notably the patchset adds support for webkit2gtk API (as provided by security
safe net-libs/webkit-gtk:4), and lots of gtk3 port fixes, hopefully making 3.0-gtk3
SLOT good enough for mass transition without problematic runtime regressions.
The patchset touches lines right above the first Makefile.in changes in the
collision patch, so that patch needed to be adjusted to not fail on top of the
patchset. Ideally the collision patch would be revised and moved to the same
approach as 3.0-gtk3 has with seds - or vice-versa, but at least same approach..
Compiling amule[-X] and veracrypt[-X] against wxGTK[-X] is untested; please test
and close the relevant bugs (617440 and 605018) if this is fixed now.
Bug: https://bugs.gentoo.org/629122
Bug: https://bugs.gentoo.org/617440
Bug: https://bugs.gentoo.org/605018
Package-Manager: Portage-2.3.19, Repoman-2.3.6
gtk2 supporting webkit-gtk versions have not received security fixes for years, so
don't allow subjecting oneself to such an unsafe possibility. The only in-tree
wxGTK:3.0[webkit] consumer was ported to wxGTK:3.0-gtk3[webkit], which will have
a security safe version in a subsequent commit.
Bug: https://bugs.gentoo.org/629122
Package-Manager: Portage-2.3.19, Repoman-2.3.6
Properly sed the .mo versioning and patch wxTranslations::AddStdCatalog
to use the slotted translation domain.
Closes: https://bugs.gentoo.org/618624
Package-Manager: Portage-2.3.19, Repoman-2.3.6
wxWidgets doesn't use full autotools (at least not automake), and it maintains
an autoconf'ed configure script inside its code repository, which gets (usually)
updated after each configure.in update.
Because we don't eautoreconf to see a slightly dumb warning about configure.in
vs configure.ac, remove the renaming, as the code is buggy and breaks on some
non-default PORTAGE_TMPDIR values. I don't intend to ever add back an eautoreconf
if at all possible, but if it is, make sure such renaming isn't buggy like this.
Also note that other Makefile.in's seem to refer to the preset configure.in's by
name (with .in ending), so renaming them is probably wrong as well via such a
global "find" approach, at least without patching the references too.
The eautoreconf and buggy renaming was already removed on bump for 3.0 SLOT.
Closes: https://bugs.gentoo.org/614982
Package-Manager: Portage-2.3.19, Repoman-2.3.6
This is a bug of my own doing, not from kuzetsa, as I edited the commit myself
to bump to 3.0.3 instead of 3.0.3.1, latter of which has a subrelease number
and all was well then.
Package-Manager: Portage-2.3.19, Repoman-2.3.6
SLOT=3.0 only, 3.0-gtk3 already was, so this is done based on that to also reduce delta.
Also removes some extra whitespaces that don't seem to give any indent value to me
(though some are kept, that do line things up a bit better).
Package-Manager: Portage-2.3.19, Repoman-2.3.6
We don't use wxPython tarballs anymore, as wxPython is focusing on their
Phoenix project (wxPython4) and not releasing these neat tarballs anymore.
Package-Manager: Portage-2.3.19, Repoman-2.3.6
The CPU microcode for Intel Haswell-X, Skylake-X and Broadwell-X
chipsets was updated to report both branch prediction control via CPUID
flag and ability to control branch prediction via an MSR register.
Required for kernel mitigation against CVE-2017-5715.
Bug: https://bugs.gentoo.org/643430
Package-Manager: Portage-2.3.19, Repoman-2.3.6
TL;DR: if you run 3ware cards on a 3.8 or newer kernel, you need to
upgrade the utility to correctly write card variables, as the old
version will sometimes segfault.
Tested on hardware models: 9750-4i, 9650SE-8LPML, 9690SA.
Per email:
> I've noticed recently that sys-block/tw_cli will segfault on modern
> systems when setting variables on a card (at least on a 9750-4i).
> ...
> when you go to run a battery capacity test, it'll complete the capacity
> test command, but segfault before it can update the timestamp on the
> card to record the last capacity test date. That causes the date to
> become "01-Jan-1970", and then dmesg will get flooded with "Battery
> capacity test overdue" messages. That's what prompted me to open a case
> with Broadcom, and then try to find some kind of workaround.
Reported-by: Joshua Kinard <kumba@gentoo.org>
Signed-off-by: Robin H. Johnson <robbat2@gentoo.org>
Package-Manager: Portage-2.3.16, Repoman-2.3.6
RepoMan-Options: --force
This bump includes the fix for the "mailsploit" vulnerability, as well as
others that may or may not affect linux systems. More info can be found
at https://www.mozilla.org/en-US/thunderbird/52.5.2/releasenotes/
Package-Manager: Portage-2.3.13, Repoman-2.3.3
Bernard Cafarelli
Chris Reffett
Ian Whyman
reagentoo
Michał Górny
Alon Bar-Lev
Alexey Sokolov
Mikle Kolyada
Andreas K. Hüttel
Brian Evans
Mart Raudsepp
kuzetsa
Fabian Groffen
Patrice Clement
Thomas Deutschmann
Robin H. Johnson
Alice Ferrazzi
Ian Stakenvicius
Andrey Grozin
Lars Wendler
Jeroen Roovers