for arm64, it just needs a few defines so that Mono recognizes the platform as
a valid one. The garbage collector patch is copied from amd64, the ucontext
defines simply have to use the correct sigcontext members, and for BoringSSL
we need to provide the setup function that tells BoringSSL which crypto accel
is supported by the CPU. For now this only enables NEON, which we have on all
OpenBSD/arm64 machines.
"makes sense, need to be sure it doesn't break existing working archs" sthen@
Tested by myself on arm64 by playing games/openra through iked(8) IPsec
Tested by myself on amd64 and i386
ok robert@ naddy@
if a port needs 2.x then set MODPY_VERSION=${MODPY_DEFAULT_VERSION_2}.
This commit doesn't change any versions currently used; it may be that
some ports have MODPY_DEFAULT_VERSION_2 but don't require it, those
should be cleaned up in the course of updating ports where possible.
Python module ports providing py3-* packages should still use
FLAVOR=python3 so that we don't have a mixture of dependencies some
using ${MODPY_FLAVOR} and others not.
is the value used by MacOS and not OpenBSD, however some ports are distributed
in a binary form and they freak out if there is a mismatch between the mono
runtime that they were compiled with and the one that they are running with.
the long-term solution is to get msbuild ported and compile depotdownloader
with our own mono runtime, until then, break IPv6 , nobody is using it anyway ;)
Note that this should only be used in well-defined cases that otherwise throw
array/list exceptions since it opens possibilities for undefined behavior.
ok robert@ (maintainer)
there may be some missing as my unpacked ports source is a little out of date
but this should catch the main things people might run into
the struct was reordered a second time in sysctl.h r1.192 to improve
compatibility but amd64 snapshot packages made it out before that happened
so the bumps are still needed
Follow the upstream recommendations for packagers and switch to
multi-packages:
devel/gettext -> devel/gettext,-runtime
devel/gettext-tools -> devel/gettext,-tools
(new) devel/gettext,-textstyle
mono already tries to detect whether mincore(2) is available, but one
place was missing a check. Upstream should probably add #ifdef checks
for all the sys/mman.h functions wrapped in this file, to replace the
existing "#ifndef HAIKU" checks.
ok sthen@
the default case gets fixed. this is not yet the default upstream
but ther plan is to do this soon
this update includes a hack that is being investigated upstream as well
in https://github.com/mono/mono/issues/6359
- apply a patch from Thomas Frohwein:
Use SIGUSR1 as the abort signal in mono instead of SIGTTIN,
because SIGTTIN gets delivered to all processes/threads and
that is not what we want here.
The port is still broken though, so leaving it as BROKEN.
fixed:
starting from 5.x, mono uses the csc (rosyln) compiler by default,
instead of mcs and csc is completely broken on OpenBSD for now, however
mcs can be used to build mono itself and other code as well.
there is an ongoing effort to fix this so i am commiting this so
that it is easier to work on it
