copy. This should have the following effects:
. Fix problems experienced by programmes that dynamically create their
own copy of the JVM and are linked against the system's zlib (e.g.,
eclipse).
. Reduce the potential for zlib based security problems affecting the
JDK.
Submitted by: mi@
in the systems libz.so. This conflict broke applications such as
Eclipse which is linked with libz.so (via gtk+ I believe).
This is a slightly modified version of the submitter's patch.
A better solution may be to link with the system's libz.so and remove
the JDK's internal zlib code altogether, but I'd like to test that a
little more first. Until then this solves the problem.
. Bump PORTREVISION since Eclipse seems to be quite widely used.
Submitted by: "Arno J. Klaassen" <arno@heho.snv.jussieu.fr>
valid one doesn't currently exist.
. Add a pkg-deinstall which removes the symbolic link if this port owns it.
. Produce pkg-install and pkg-deinstall with SUB_FILES and SUB_LIST rather
than manually using ${SED} ourselves.
Approved by: maintainer timeout
http://java.sun.com/j2se/1.5.0/docs/guide/awt/1.5/xawt.html
it has some advantages over XToolkit. Its also the default on Linux
and Solaris will be switching to it. Some people have reported that
it fixes a crash in the browser plugin for them.
Please let me know straight away if this causes problems, particularly
with Swing, as it hasn't been extensively tested. The web page
mentioned above explains how to switch the toolkits dynamically so you
can compare them.
Submitted by: Huang wen hui <hwh@gddsn.org.cn>
(the awt_LoadLibrary.c patch)
Approved by: phantom (maintainer)
. Many patches are now unnecessary as they are included in the new
patchset.
. The browser plugin and Java Web Start is enabled on i386 (there are
64 bit issues with both the plugin and Mozilla/Firefox which prevent
enabling it on amd64).
. Update the amount of disk space needed.
. Update the status of the port.
. Disable building the shared class data archive. This broke the build
on amd64 and appears to also be problematic on some i386 versions
(4.11 is broken at least). It will reappear in future, probably
initially on a limited set of FreeBSD versions and architectures
(6.0/i386 is reported to work).
Reviewed by: freebsd-java@
Approved by: maintainer timeout (1 week)
encoding. It is even easier to do the same thing to jdk14 and jdk13,
where only one charset-interface exists (jdk15 has two with the old one
considered obsolete).
Approved by: Alexey Zelkin (maintainer)
in or below the current working directory. Fixes a security problem with
jar(1).
This fix may change to be compatible with whatever fix Sun applies when
they release the next version of 1.5.
. Bump PORTREVISION for this fix.
Security: http://vuxml.FreeBSD.org/18e5428f-ae7c-11d9-837d-000e0c2e438a.html
Reviewed by: maintainer timeout
. /etc/localtime is a symlink.
. /etc/localtime contains a time zone not recognised by the JDK.
Submitted by: Kurt Miller <truk@optonline.net>
Reviewed by: maintainer timeout
hence the path for the shared libraries doesn't always work on FreeBSD.
It definitely fails on FreeBSD 4.11 and FreeBSD 6-CURRENT under the
tested environments. In fact, the dladdr(3) man page even warns of
these problems. While there is work under way to fix this, it isn't
available yet.
Given that situation, switch to trying /proc/curproc/file, which is
similar to what Linux does, and if that fails, drop back to checking
argv[0] and iterating through $PATH as in jdk 1.4. Both these methods
work correctly in testing.
Reported by: das
Reviewed by: maintainer timeout
unfortunately sending it to stdout. When using such a JDK to bootstrap
this line ends up at the head of generated classes, leaving them
uncompilable. Add a filter to the class generation to strip out such
lines with egrep.
A similar patch is present in the jdk14 port and prevents a semi-common
class of error reports.
Approved by: phantom (maintainer)
files that are generated by the post-install script (which runs after
the dynamic packing list has been generated).
Approved by: portmgr (krion), phantom (maintainer)
This ensures that this command is run before the files in the package
are deleted (which is necessary for it to correctly delete the symbolic
links created by registervm).
Approved by: phantom (maintainer)
family -- first public patchset of native Sun JDK 1.5.0 port.
Most valuable addition of this patchset is native amd64 support.
And special thanks goes to Daniel Seuffert <ds@freeBSD.org> for
making it possible by providing amd64 hardware.
This patchset was tested on following configurations: i386/4.10,
i386/5.3, amd64/5.3. 5.3-RELEASE support is quite strong and
shown no huge visible problems over last week.
But even mentioning above note, keep in mind -- THIS IS ALPHA
PATCHSET and suitable for testers/developers ONLY!
Known issues are including (but for sure not limited to):
. Browser plugin support is missing
. JVMTI, JDWP and JMX are not tested yet
. FreeBSD i386/4.10 support is suffering from hidden memory
allocation failres (ideas and patches are welcome)
NOTE ABOUT BOOTSTRAPING: It's possible to bootstrap jdk 1.5.0 using
jdk 1.4.2 (either native or linux one). There's no need to have
java/linux_jdk15 installed and working.
Supported by: FreeBSD Foundation
a generated file will be overwritten with a warning, causing the
build to fail. There is a check for linprocfs in pre-build, but it
seems as though this problem can somehow trigger anyway, based on
semi-regular reports to the mailing lists.
PR: 74999
Approved by: phantom
existing the Solaris base, and similarly to what happened with NSPR, made
a bad assumption on undefined behavior. This broke locking in various
places in Java, for example, causing the the debugging support to be
totally broken. It is worth someone who knows the Java codebase taking
a look to see what other things could have been broken by this on
FreeBSD 5.x+.
The assumption is that pthread_mutex_trylock(3) on a default-type
mutex will fail with EBUSY. This assumption is wrong for our
libpthread, which returns EDEADLK if the owner thread is trying to
acquire the mutex again with trylock. The behavior of performing a
locking operation on a self-locked default-type mutex is explicitly
undefined for pthread_mutex_lock(3).
The POSIX specification is still not very clear. It defines
pthread_mutex_trylock(3) in terms of pthread_mutex_lock(3) yet
does not say what the defined behavior should be for a self-locked
pthread_mutex_trylock(3) for any of the various mutex types, so it is
ambiguous whether the result is clearly undefined or clearly to return
EBUSY.
It is a one line change whether or not to make libpthread return
EDEADLK in this case, where it seems that most implementations do not.
Reference: http://www.opengroup.org/onlinepubs/009695399/functions/pthread_mutex_lock.html
The HotSpot code (ab)uses named enums as ints in a number of places.
The problem with this is that according the the C++ spec, the compiler
(essentially) only needs to use an integral type wide enough to hold
the values defined in the enum. Earlier versions of gcc appear to have
just used an int whether they could have got away with a narrower type
or not, hence the code worked as expected. gcc 3.4 now appears to
implement this part of the spec, so using an enum blindly as an int
causes various problems due to overflow.
In this case the enum, Bytecodes::Code, appears to be a genuine enum,
its just assumed to be wide enough to hold an arbitrary int in various
places in the code. The correct fix would be to track down all those
places in the code and fix them. Since there are quite a lot of these
places and 5.3 is close to release for now we just add a value to the
enum set to INT_MAX, forcing the compiler to use at least an int for the
type.
Sleuth work, discussion and code suggestions: peadar
The HotSpot code (ab)uses named enums as ints in a number of places.
The problem with this is that according the the C++ spec, the compiler
(essentially) only needs to use an integral type wide enough to hold
the values defined in the enum. Earlier versions of gcc appear to have
just used an int whether they could have got away with a narrower type
or not, hence the code worked as expected. gcc 3.4 now appears to
implement this part of the spec, so using an enum blindly as an int
causes various problems due to overflow.
This case is particularly bogus since the enums are merely to define
a named integral type within a class (VMReg::Name doesn't even have
any values enumerated in the declaration). So, convert these two
enums to simply be typedef'ed ints.
Sleuth work, discussion and code suggestions: peadar
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.
. Fix some warnings regarding formats in debugging printf's.
(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 if they
aren't defined.
override the MAKEFLAGS ARCH value in the main HotSpot Makefile. Fix
this by passing in a blank MAKEFLAGS up front so there is nothing to
(try to) override.
Submitted by: truckman
Requested by: kris
