distfile has been updated at 2004/April/10th
and according to http://crd.lbl.gov/~xiaoye/SuperLU/
o Include "symmetric mode"
b) change portversion
Submitted by: kris via bento
Note: two of the sites listed in MASTER_SITE_PORTS_JP do not seem to be
keeping current on these ports. These are (together with the date that
they last kept up):
ftp://ports.jp.FreeBSD.org/pub/FreeBSD-jp/ports-jp/LOCAL_PORTS (Feb 2003)
ftp://ftp.ics.es.osaka-u.ac.jp/pub/mirrors/FreeBSD-jp/ports-jp/LOCAL_PORTS (Oct 2001)
as to why :-) ), the distfile survey reports that Nathaniel's original
site for this software is long gone. A Google search shows the mirror
that I added to be the best option, and that the most recent distfile is
from 1994. The only other links refer to security problems; although
all the known ones have been fixed, Debian claims that:
We have been devoting some effort to trying to avoid shipping metamail
in the future. It became unmaintainable and these are probably not the
last of the vulnerabilities.
Accordingly, mark it as 'deprecated'. If anyone is using this code,
they should prepare to adopt the port and address any security concerns
going forward.
The approach is slightly different from that taken by nethack34 and
prevents the manual page conflicts produced by that approach.
. While here, clean up the Makefile a little w.r.t. order and clean up the
packing list to not produce warnings when the log or record are removed.
With this change, you can have all nethack versions from ports installed
at once. However, this isn't as pleasant as it could be, and I'd like to
address this in the future by merging some changes from here into
nethack33 (in particular) and nethack34.
- Workaround internal compiller error on -STABLE [1]
- More build patches incl. fix for build with antiqe STL on -STABLE
Reported by: Lee Harr <missive@hotmail.com> [1]
Notes: - Since 2.01a26 cdrecord has a new option "-tao" (Track At Once)
for what was the default write mode in previous versions.
Currently, cdrecord still falls back to writing in TAO mode if
no writing mode has been specified but this might change in a
future version (infact, the release notes for 2.01a26 already
say that cdrecord won't burn a CD if no write mode has been
specified).
- Cdrecord 2.01a26 was broken in that it only wrote in TAO mode,
which was fixed in 2.01a27.
For changes since 2.01a25 see:
ftp://ftp.berlios.de/pub/cdrecord/alpha/AN-2.01a26
and
ftp://ftp.berlios.de/pub/cdrecord/alpha/AN-2.01a27
- Add a workaround for what appears to be a problem with gmake and certain
locales on -current. [1]
- Correct a typo in pkg-descr.mkisofs inherited from the cdrtools ABOUT
file.
- Remove the workaround for the broken mlockall(2) on EOL'ed FreeBSD 5.1.
- No longer compile with -DSOURCE_MODIFIED, the cdrecord sources are no
longer altered by this port except for changing the location of the
configuration file, which is done in compliance with the terms in
cdrecord/defaults.c.
- Make the direct ATAPI transport (cooked_ioctl interface) of cdda2wav
(e.g. when used via `cdda2wav -D /dev/acd0`) work on FreeBSD after
ata(4) was GEOM'ifed and the CDIOCREADAUDIO ioctl removed.
Note: For qualitiy reasons (e.g. to be able to fix/work around broken
TOCs etc.) cdda2wav should be used with its generic_scsi interface
(e.g. via `cdda2wav -D 2,0,0` and using atapicam(4) if you have an
ATAPI device). However, there are some ATAPI devices around where
the latter simply doesn't work (look at e.g. PR 62507 or the audit-
trail of PR 60233 for examples).
- Properly initialise the verbosity level of cdda2wav when the cooked_ioctl
interface is used so it's disabled by default (the outcome of this bug is
actually compiler dependend, with GCC 2.95.4 verbosity is off by default,
with GCC 3.3.3 it's on).
Note: The patches for the last two items where sent to Heiko Eissfeldt, the
author of cdda2wav, and from his replies I was of the opinion that he
"accepted" them and they would be incorporated in cdrtools 2.01a26.
However, they still didn't show up in the cdrtools sources and I so
far didn't get any reply to my email asking if there would be anything
wrong with them.
Requested by: pav [1]
Submitted by: maintainer
anti-virus engine (6.23.0.3 -> 6.25.0.2), current virus definition file
(6.23.0.49 -> 6.25.0.1) and an updated antivir.conf.
If you are using the auto-update facility you should already have the
updated AV engine binary and VDF.
- Look for the libmilter of the Sendmail port and the GnuPG binary in
LOCALBASE instead of PREFIX.
Submitted by: Marius Strobl <marius@alchemy.franken.de>
