Bug#852224: fixed in gpm 1.20.7-1
Hi,
On Sun, Aug 13, 2017 at 04:34:13AM +0000, Axel Beckert wrote:
> * Switch to source format "3.0 (quilt)" and rewrite debian/rules as dh
> v7 style minimal rules file with debhelper compatibility level 10.
> + Bump versioned debhelper build-dependency accordingly.
> + Drop build-dependencies on quilt, autoconf and autotools-dev.
> + Refresh 021_libgpm_dev_gpmctl_debug_msg.patch to remove fuzz.
> + Use dh-exec and executable debian/*.install files instead of
> $(INSTALL_PROGRAM) and $(INSTALL_DATA) in debian/rules.
> + Add patch to fix FTBFS with -Wformat-security.
> * Enable all hardening build flags.
> + Update 014_has_mouse_control.patch to properly pass compile flags.
[...]
> * Declare compliance with Debian Policy 4.0.1.
> + Use /run/ instead of /var/run/ in init script.
> * Import stable upstream release 1.20.7.
> + Refresh patches where necessary.
> + Manually extend 050_dont_link_libcurses to make package compile
> again independently of libncurses5-dev being installed or not.
> + Drop 020_daemon_quit_noverbose.patch,
> 021_libgpm_dev_gpmctl_debug_msg.patch,
> 022_libgpm_no_log_debug_msg.patch, 030_fd_set_negative_int.patch and
> 040_no_OPEN_MAX.patch, applied upstream.
> + Update debian/libgpm2.install wrt. to minor SONAME version.
> + No more symlink config.{sub,guess}, call ./autogen.sh instead.
[...]
> * Add an initial symbols file.
the upload had quite some changes: complete revamp of packaging, minor
SONAME bump, hardening build flags, a lots for fiddling getting the
exclusion of the ncurses dependency right again after importing the
new (years old) upstream release, initial symbols file, etc.
I've checked the new binaries against one package linked against
libgpm2, pdmenu, and it still worked fine.
But:
Since then I seem have some kind of shadow cursor on the console and
some strange warnings in the syslog:
Aug 13 05:05:53 c-cactus2 /usr/sbin/gpm[9430]: *** warning [daemon/old_main.c(254)]:
Aug 13 05:05:53 c-cactus2 /usr/sbin/gpm[9430]: Data on strange file descriptor 4
Aug 13 05:06:19 c-cactus2 /usr/sbin/gpm[9430]: *** info [daemon/processrequest.c(42)]:
Aug 13 05:06:19 c-cactus2 /usr/sbin/gpm[9430]: Request on 6 (console 1)
Aug 13 05:07:04 c-cactus2 /usr/sbin/gpm[9430]: *** warning [daemon/old_main.c(254)]:
Aug 13 05:07:04 c-cactus2 /usr/sbin/gpm[9430]: Data on strange file descriptor 4
But then again, I find them also from before I started working on the package:
Aug 02 18:19:43 c-cactus2 /usr/sbin/gpm[10034]: *** warning [daemon/old_main.c(254)]:
Aug 02 18:19:43 c-cactus2 /usr/sbin/gpm[10034]: Data on strange file descriptor 4
So I first uploaded it to experimental and will check it on more
devices, especially also different architectures to see if that's a
local or maybe even a non-reboot issue.
But I'd also appreciate some additional testers for that experimental
package.
Regards, Axel
--
,''`. | Axel Beckert <abe@debian.org>, http://people.debian.org/~abe/
: :' : | Debian Developer, ftp.ch.debian.org Admin
`. `' | 4096R: 2517 B724 C5F6 CA99 5329 6E61 2FF9 CD59 6126 16B5
`- | 1024D: F067 EA27 26B9 C3FC 1486 202E C09E 1D89 9593 0EDE
Reply to: