Also allow /proc itself (rather than /emul/linux/proc) to be mounted
"-o linux" -- this is a common configuration that allows mounting of only
one procfs usable by both NetBSD and Linux binaries.
package recently broke because of this (it still worked in pkgsrc-2005Q4,
so something must have changed in pkgsrc but I have no idea what exactly).
A workaround to fix this: just sed(1) the ^M's away.
Ok with veego@ (who brought this up).
Notable changes:
- DragonFly support (thanks to asmodai and joerg)
- Add DONT_ENABLE_IPV6 knob (turned into pkgsrc "inet6" option)
- Many small improvements and fixes
> there were some problems during deinstallation so I had to split PLIST
> into two files, one which contains the files and a second one which
> contains the @dirrm Just rename PLIST to PLIST.common and move the @dirrm
> lines into the new file PLIST.common_end. This is needed because the
> @dirrm would be executed while there are still some files left from the
> PLIST.jce or PLIST.plugin filelist. I also removed the '@exec ${MKDIR}
> %D/java/jdk-1.4.2/jre/lib/applet' line, while I did that.
>
> typo in options.mk: SCR instead of SRC and so PLIST.plugin wouldn't be
> included into the generated .PLIST
All files/directories are uninstalled correctly now.
> removed the pre-build target and added a BUILDLINK_PASSTHRU_RPATHDIRS
> line, jlam told me about that trick.
The jdk14 package can be build as non-root user now.
Set PKGVERSION to 1.4.2.6.1 to distinguish it from the p6 patch.
Also, add the updated cacert file and some ns plugin related patches
(currently unused) from the FreeBSD port.
changes:
2004-07-21 Wednesday 17:58 glewis
* j2se/src/share/native/java/io/io_util.h:
. At least on FreeBSD, O_SYNC and O_DSYNC aren't both defined
(for FreeBSD 4.x neither are defined and for FreeBSD 5.x
O_DSYNC isn't defined). This caused them to be defined to
some bogus values. In particular, O_SYNC would be defined
as 0x800, which is O_EXCL (at least on FreeBSD 4.x). The
result being that the RandomAccessFile class would fail to
open an existing file if you specified "s" as part of the mode.
Fix this by defining O_SYNC and O_DSYNC to O_FSYNC on BSD if
they aren't defined. This shouldn't hurt Net or Open.
2004-07-08 Thursday 20:21 glewis
*
j2se/src/share/native/sun/awt/font/fontmanager/fontobjects/fontObject.cpp:
. In the ReadChunk() function, change an assert() to be a "test for a
condition and return NULL". Take account of the NULL in the
appropriate place (which is somewhat worrisome in itself since
ReadChunk() has always had the possibility of returning NULL).
This makes loading a font file a little more resilient to specially
crafted font data which can be used, for example, by an applet to
crash the browser plugin by triggering the assert(). Such an applet
was mentioned on Bugtraq:
http://www.securityfocus.com/archive/1/367331/2004-06-26/2004-07-02/0
and can be found at
http://www.illegalaccess.org/cms/?q=node/view/9
This change stops the browser plugin from crashing. However, I'm
somewhat unfamiliar with this part of the code though, so any review
would be appreciated.
. While I'm here, fix some warnings regarding types in debugging printf's.
2004-07-08 Thursday 20:10 glewis
* j2se/src/share/native/java/lang/fdlibm/src/k_tan.c:
. Merge in a fix recently added to FreeBSD's math library (taken from
fdlibm-5.3):
"Merge a bugfix from FDLIBM 5.3 to ensure that the error in tan()
is always less than 1 ulp."
I've not yet run this through the compatibility test suite, so this
change may be backed out if the suite relies on the broken behaviour.
2004-07-08 Thursday 20:09 glewis
* j2se/src/share/native/java/lang/fdlibm/src/e_pow.c:
. Merge in a fix recently added to FreeBSD's math library (taken from
fdlibm-5.3):
"Fixed the sign of the result in some overflow and underflow cases (ones
where the exponent is an odd integer and the base is negative)."
I've not yet run this through the compatibility test suite, so this
change may be backed out if the suite relies on the broken behaviour.
2004-07-08 Thursday 20:04 glewis
* deploy/src/javaws/src/solaris/native/system_md.c:
. The implementation of sysTempnam() formerly just called tempnam(), which,
on FreeBSD at least, would provoke warnings about security. So, on *BSD,
implement sysTempnam() using mkstemp(3).
2004-04-03 Saturday 05:54 glewis
* j2se/src/solaris/native/java/net/NetworkInterface.c:
. Don't leak a file descriptor in the IPv6 case.
Submitted by: Pavlin Radoslavov <pavlin@icir.org>
This is Sun's Java[tm] Development Kit, version 1.4.2, made buildable
and usable natively on NetBSD/i386 by Greg Lewis and a host of others.
===== From the patchset README:
This is a set of source code patches for the JDK 1.4.2 SCSL source.
The resulting JDK should be suitable for most uses, although using it in
a production environment is at your own risk. Please note that this
patchset only provides support for recent versions of FreeBSD and NetBSD.
The patchset is available at
http://www.eyesbeyond.com/freebsddom/java/jdk14.html
Please read the file THANKS for attributions.
Instructions to build the JDK are contained in the file BUILD.
You should also examine the ChangeLog for the changes that have been
made so far, and notes on them.
