- Broken with libc++ 6 and there's not enough time before the next bulk build.
- PORTEPOCH isn't necessary as if built it's good, and the update will return
rpcs3/util/atomic.cpp:55:51: error: no type named 'node_type' in 'std::__1::multimap<const void *, (anonymous namespace)::waiter, std::__1::less<const void *>, std::__1::allocator<std::__1::pair<const void *const, (anonymous namespace)::waiter> > >'; did you mean 'size_type'?
thread_local std::multimap<const void*, waiter>::node_type s_tls_waiter = []()
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~^~~~~~~~~
size_type
/usr/include/c++/v1/map:1507:60: note: 'size_type' declared here
typedef typename __alloc_traits::size_type size_type;
^
rpcs3/util/atomic.cpp:59:16: error: no member named 'extract' in 'std::__1::multimap<const void *, (anonymous namespace)::waiter, std::__1::less<const void *>, std::__1::allocator<std::__1::pair<const void *const, (anonymous namespace)::waiter> > >'
return dummy.extract(dummy.emplace(nullptr, &s_tls_waiter));
~~~~~ ^
rpcs3/util/atomic.cpp:55:61: error: cannot initialize a variable of type 'std::multimap<const void *, waiter>::size_type' (aka 'unsigned long') with an rvalue of type 'void'
thread_local std::multimap<const void*, waiter>::node_type s_tls_waiter = []()
^ ~~~~
rpcs3/util/atomic.cpp:92:69: error: no type named 'node_type' in 'std::__1::multimap<const void *, (anonymous namespace)::waiter, std::__1::less<const void *>, std::__1::allocator<std::__1::pair<const void *const, (anonymous namespace)::waiter> > >'; did you mean 'size_type'?
const auto ptls = static_cast<std::multimap<const void*, waiter>::node_type*>(found->second.tls_ptr);
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~^~~~~~~~~
size_type
/usr/include/c++/v1/map:1507:60: note: 'size_type' declared here
typedef typename __alloc_traits::size_type size_type;
^
rpcs3/util/atomic.cpp:93:21: error: no member named 'extract' in 'std::__1::multimap<const void *, (anonymous namespace)::waiter, std::__1::less<const void *>, std::__1::allocator<std::__1::pair<const void *const, (anonymous namespace)::waiter> > >'
*ptls = wmap.list.extract(found);
~~~~~~~~~ ^
rpcs3/util/atomic.cpp:94:7: error: member reference base type 'unsigned long' is not a structure or union
ptls->mapped().cond.notify_one();
~~~~^ ~~~~~~
6 errors generated.
powerpc64 requires the same Makefile fix as amd64.
Additionally, 684f7ec5d6 is necessary.
Approved by: linimon (mentor)
Differential Revision: https://reviews.freebsd.org/D21055
- fix qemu-user-static hangs with wc -l //
- add qemu-user-static redirection for riscv64, no support, yet.
Thanks to kevans, miki and others for fixing this nonsense.
Submitted by: kevans miki
as defined in Mk/bsd.default-versions.mk which has moved from GCC 8.3
to GCC 9.1 under most circumstances now after revision 507371.
This includes ports
- with USE_GCC=yes or USE_GCC=any,
- with USES=fortran,
- using Mk/bsd.octave.mk which in turn features USES=fortran, and
- with USES=compiler specifying openmp, nestedfct, c11, c++0x, c++11-lang,
c++11-lib, c++14-lang, c++17-lang, or gcc-c++11-lib
plus, everything INDEX-11 shows with a dependency on lang/gcc9 now.
PR: 238330
emulators/qemu-user-static
- Capture fix for keeping track of the thread counts being optimized.
9ac22b222b
Submitted by: Mark Millard <marklmi at yahoo.com>
Reviewed by: kevans
this dependency missed in terms of code, even if mentioned in the log,
in revision 504944 | gerald | 2019-06-23.
PR: 239170
Submitted by: chadf@triularity.org
regular 4.12 release that includes fixes for broken 64-bit prefix
initialization.
It also addresses a build failure for Wine Staging (the STAGING
option set in this port). [1]
PR: 239159 [1]
Reported by: John Hein <jcfyecrayz@liamekaens.com> [1]
infrastructure ports (linux_base-c6 and linux-c6*) install i386 CentOS 6
packages. Using c6_64 instead of c6 made them install mixed amd64/i386
CentOS 6 packages and allowed Linux application ports to install amd64
binaries when available. The i386-only configuration was only useful for
users of x11/nvidia-driver* which didn't provide Linux amd64 emulation
until recently.
With x11/nvidia-driver* now supporting Linux amd64 there's no longer a
need for i386-only Linux on amd64. Let DEFAULT_VERSIONS+=linux=c6 make
Linux infrastructure ports install mixed amd64/i386 CentOS 6 packages,
like c6_64 before, and remove support for c6_64 as a value.
Replace LINUX_ARCH with ARCH everywhere because they are always the same
now.
- Still more DLLs are built as PE files by default.
- Support for Plug & Play device drivers.
- Better support for the Visual Studio remote debugger.
- More support for enumerating display devices.
- Various bug fixes.
Cannonball is a program which allows you to play an enhanced version
of Yu Suzuki's seminal arcade racer, OutRun, on a variety of systems.
WWW: https://github.com/djyt/cannonball/wiki
PR: 238495
Submitted by: Santhosh Raju <santhosh.raju@gmail.com>
Not reproducible in a local 12.0 amd64 poudriere jail:
=======================<phase: stage >============================
===> Staging for yuzu-qt5-s20190702
[0/1] cd /wrkdirs/usr/ports/emulators/yuzu-qt5/work/.build && /usr/local/bin/ctest --force-new-ctest-process
Test project /wrkdirs/usr/ports/emulators/yuzu-qt5/work/.build
Start 1: tests
=>> Killing runaway build after 7200 seconds with no output
Reported by: pkg-fallout
Makefile
- pygtk2 is now gtk30 (upstream);
- python:2.7 became python:3.6+;
- option helpers changed to follow '5.13 Makefile Options'.
pkg-plist
- hatari_profile is installed by default (upstream);
- manpages and doc up to date, removed fr/ versions;
- all tools/ available, with respective docs.
* https://hatari.tuxfamily.org/doc/release-notes.txt
Reported by: swills
Differential Revision: https://reviews.freebsd.org/D19628
