I'm pleased to announce emc 2.4.2, a new release of emc2 with a
substantial number of bugs fixed.
The new packages are available for Ubuntu 8.04 "Hardy Heron" in the
package repositories. If you installed 2.4.0, this will be presented as
an upgrade. If you have 2.3.5 or earlier, it will not automatically be
installed as an upgrade. For more information and installation/upgrade
instructions, please see
http://wiki.linuxcnc.org/cgi-bin/emcinfo.pl?UpdatingTo2.4
Git users can get this release by checking out the v2.4.2 tag, or follow
bugfixes on the v2.4_branch. Source tarballs are available from the
package repository, http://linuxcnc.org/hardy/dists/hardy/emc2.4/source
Many thanks to everyone who helped us with reports, information, or code
related to the problems fixed in this release.
Changes since 2.4.1:
* axis: fix crash when environment variable AXIS_NO_HAL is set
* classicladder: modbus reliability and performance improvements
* components: fix tan(), atan() implementations. Before, using these
functions would make all math in realtime components behave unpredictably.
* comp: fix repeated MODULE_LICENSE declarations
* documentation: many improvements
* emccalib: fix crash 'unmatched open quote in list' for lines like
loadrt hal_parport cfg="0x378 out "
* emcrsh: fix crash when invoked with a bad commandline
* emcsh: report effect tool offsets on all axes
* emcsh: enable emc_tool_offset to access all tool offsets
* emcsh: fix wrong conversion of ABC axis values as though they were
linear axes
* gcode: fix a problem with inverse time mode and m66 leading to very slow
moves
* gcode: save state of g92 across runs. In the first run after upgrading
to 2.4.2, g92 will be disabled. In subsequent runs, the state of g92
at startup will be the the same one stored in the var file at shutdown.
* gcode: fix G1 followed by G5, G5.x
* gcode: improve accuracy of NURBS paths. An important calculation of the
direction of a NURBS path at a given point has been improved. This makes
the resulting path more closely follow the mathematically precise NURBS
curve.
* gcode: improve accuracy of G5, G5.1 spline paths by converting them to
NURBS paths.
* gcode: detect missing P, Q words with G5
* gcode: avoid division by zero in NURBS
* hal_input: fix keys and buttons with unknown names
* image-to-gcode: fix crash when using lace bounding
* mesa: fix failure to load firmwares on 7i43
* mesa: fix failure to load firmwares on 5i23, 3x20, and other PLX9054-based
cards
* mesa: fix stepgen slowly moving in negative direction, losing position
* mesa: fix stepgen mode setting problem
* motion: fix threading and tapping with spindle turning backward (M4)
* pncconf: mesa 3x20 support
* pncconf: write configuration files without "base thread". This may improve
behavior on step&direction systems configured with pncconf.
* pncconf: many additional bugfixes
* pickconfig: fix icon for desktop shortcuts
* pickconfig: fix permissions on desktop shortcuts
* pickconfig: don't allow user to run sample configuration without copying.
This has not worked right for a long time, so there's no value in offering
the option.
* ppmc: improve error message when no board is present
* ppmc: fix board detection when cooperating with linux parport_pc driver
* sample configurations:
* ppmc: sample configs now use axis user interface
* ppmc: hal samples for spindle speed display, pendant
* smithy: 1240 and 622leadshine sample configuration improvements
* sample gcode:
* lathe_pawn: set radius mode in startup codes
* stepgen: fix ctrl_type for kernels 2.6.30 and newer
Jeff
------------------------------------------------------------------------------
This SF.net email is sponsored by Sprint
What will you do first with EVO, the first 4G phone?
Visit sprint.com/first -- http://p.sf.net/sfu/sprint-com-first
_______________________________________________
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users
