Since 2017 and maybe earlier people have experienced hangs in various
parts of Emacs. Two error cases which were easy to reproduce:
Emacs spinning when running org-capture in an X11 environment, and (more
recently reported) M-x list-packages failing to connect to online
mirrors.
Those hangs don't appear any more if we disable an old workaround for
our seemingly broken SIGIO implementation. Even if we still suffer from
problems in this area, right now the workaround seems to hurt us more
than help us.
Many thanks to mikeb@, Emmanuel Giraud, abieber@, Piotr Isajew, Gregoire
Jadi, krw@ and others for the reports and quick feedback. Special
thanks to fellow Gregoire Jadi for leading the latest bug hunt which
gave us this diff. <3
A recent commit brought base libexecinfo support for emacs, but it appears it
makes 'temacs' segfault during the build on powerpc. Disable the libexecinfo
support, only on this arch, to fix the build.
comment tweak and OK jca@ (maintainer)
From Brad
Since emacs is mostly the executable itself and the lisp code, don't
expect useful backtraces in case of problems. People should still
install the debug-emacs package and use egdb.
Fatal error 11: Segmentation fault
Backtrace:
0x13c54970bda <???> at emacs
0x13c5494f791 <???> at emacs
0x13c5497184e <???> at emacs
0x13c549718d2 <???> at emacs
0x13c549706e9 <???> at emacs
Segmentation fault (core dumped)
GNU poke is an interactive, extensible editor for binary data. Not limited
to editing basic entities such as bits and bytes, it provides a full-fledged
procedural, interactive programming language designed to describe data
structures and to operate on them.
ok bcallah@
other than the usual "python3/<blank>" python version selection and
remove setting MODPY_VERSION=${MODPY_DEFAULT_VERSION_3} again from the
affected ports.
if a port needs 2.x then set MODPY_VERSION=${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.
segfaults when trying to use some functionality. Required work to get it
working with -fno-common.
ok kn@, sthen@ also suggested fixing the -fno-common issue by tedu'ing it.
with two different versions of py-sip; the old one configured to only
produce the module for py-qt5 (in devel/py-sip) and the actual sip in
x11/py-sip-qt5, with broken update chain and which didn't entirely
make sense.
in particular: py-sip is now actually (a newer version of) sip again,
with py2 support dropped (current versions don't have it). other ports
using this are changed to the newer build system using "sip-build"
where possible.
for the future we will need to move to sip 6.x which drops the older
build system using the "sip5" or "sip" script; other ports (veusz, and
i think qgis) don't work with that yet.
- Remove uneeded patch
- Fix gobject-introspection overrides directory
- Fix for 64bit time_t
- Make AP_UnixPrefs::overlayEnvironmentPrefs() a noop; this function is totally
broken and leads to sigsegv.
This port hits segmentation faults right on startup (amd64, default/empty
`vm.malloc_conf'), both built with either `-fcommon' or fixes from FreeBSD.
No update since 2003.
Fails to build the now default "-fno-common".
OK gnezdo
Changelog: https://texmacs.org/tmweb/about/changes.en.html
- reverts back to DISTNAME and PKGNAME as in revision 1.15 now that
distfile extracts to -src again
- notes in README that octave plugin requires gnuplot
- compileall.py for a portcheck warning
- removes the patch for the TeXmacs interface to R
- updates README with R instructions
- removes REVISION
- removes two backported patches for qt 5.15.2
OK rsadowski@
