From gnu_andrew at member.fsf.org Wed Apr 1 04:20:34 2009 From: gnu_andrew at member.fsf.org (Andrew John Hughes) Date: Wed, 1 Apr 2009 11:20:34 +0000 Subject: HotSpot and OpenJDK6 Message-ID: <17c6771e0904010420s4edd7248raff8d66abdcd1dc0@mail.gmail.com> Hi all, Is there any news on a public stable tree for HotSpot b14? And when this will be used by OpenJDK6? Thanks, -- Andrew :-) Free Java Software Engineer Red Hat, Inc. (http://www.redhat.com) Support Free Java! Contribute to GNU Classpath and the OpenJDK http://www.gnu.org/software/classpath http://openjdk.java.net PGP Key: 94EFD9D8 (http://subkeys.pgp.net) Fingerprint: F8EF F1EA 401E 2E60 15FA 7927 142C 2591 94EF D9D8 From gnu_andrew at member.fsf.org Wed Apr 1 11:03:24 2009 From: gnu_andrew at member.fsf.org (Andrew John Hughes) Date: Wed, 1 Apr 2009 18:03:24 +0000 Subject: New project: getting rid of IcedTea local patches In-Reply-To: <17c6771e0904011055nfa6a0a5pec140dd568cc63e0@mail.gmail.com> References: <49D3917B.2010907@redhat.com> <49D3A8B1.4040201@sun.com> <17c6771e0904011055nfa6a0a5pec140dd568cc63e0@mail.gmail.com> Message-ID: <17c6771e0904011103j78adad05h59960f37ed2c32a2@mail.gmail.com> 2009/4/1 Andrew John Hughes : > 2009/4/1 Kelly O'Hair : >> Andrew Haley wrote: >>> >>> We at Red Hat have noticed that the list of IcedTea local patches has >>> been getting large, with some local patches that should have been >>> pushed upstream. ?Also, some IcedTea patches have been committed to >>> OpenJDK 7 but not OpenJDK 6. >>> >>> So, we're going to have a serious attempt to minimize the number of >>> IcedTea patches. ?We'll create a Wiki page at icedtea.classpath.org >>> that is a list of the patches, and for every one we're going to >>> discuss whether it should stay IcedTea local, be pushed upstream to >>> OpenJDK, or be deleted altogether. ?Then we will create OpenJDK >>> bugzilla entries for the patches that we think should be integrated >>> upstream and work with Sun engineers to get that done. >>> >>> I hope that it will be possible to reduce the number of patches to a >>> fraction of what we have at the moment, while improving the quality of >>> the upstream product. >>> >>> This is going to be quite a lot of work, so we'd appreciate help from >>> anyone. >>> >>> I hope this will meet with everyone's approval. >>> >>> Andrew. >> >> Sounds great. >> >> I can certainly help out with any build related (makefiles etc.) >> changes. >> >> -kto >> >> > > Good, because to my knowledge, those form the majority :) > However, my guess would also be that some require extensive reworking > to be generally acceptable (e.g. providing the option to build using > system libraries rather than forcing it). > -- > Andrew :-) > > Free Java Software Engineer > Red Hat, Inc. (http://www.redhat.com) > > Support Free Java! > Contribute to GNU Classpath and the OpenJDK > http://www.gnu.org/software/classpath > http://openjdk.java.net > > PGP Key: 94EFD9D8 (http://subkeys.pgp.net) > Fingerprint: F8EF F1EA 401E 2E60 15FA ?7927 142C 2591 94EF D9D8 > On the other hand, there is some low-hanging fruit. Andrew Haley reminded me of this patch earlier today: http://bugs.sun.com/bugdatabase/view_bug.do?bug_id=6781583 which went into OpenJDK7. The patch itself should be trivial to approve, given it has already been through the process once. The main issue is that it relies on OpenJDK6 being updated to hs14. Is there any news on this and of the public stable branch of this version? As to committing, do those on the OpenJDK commit list have access to all repositories (given patch approval obviously) or is it restricted in some way? As myself, Gary and Andrew Haley all are now on this list, it would be good to start the process of committing patches ourself once approved. Thanks, -- Andrew :-) Free Java Software Engineer Red Hat, Inc. (http://www.redhat.com) Support Free Java! Contribute to GNU Classpath and the OpenJDK http://www.gnu.org/software/classpath http://openjdk.java.net PGP Key: 94EFD9D8 (http://subkeys.pgp.net) Fingerprint: F8EF F1EA 401E 2E60 15FA 7927 142C 2591 94EF D9D8 From Kelly.Ohair at Sun.COM Wed Apr 1 11:36:06 2009 From: Kelly.Ohair at Sun.COM (Kelly O'Hair) Date: Wed, 01 Apr 2009 11:36:06 -0700 Subject: New project: getting rid of IcedTea local patches In-Reply-To: <17c6771e0904011103j78adad05h59960f37ed2c32a2@mail.gmail.com> References: <49D3917B.2010907@redhat.com> <49D3A8B1.4040201@sun.com> <17c6771e0904011055nfa6a0a5pec140dd568cc63e0@mail.gmail.com> <17c6771e0904011103j78adad05h59960f37ed2c32a2@mail.gmail.com> Message-ID: <49D3B416.4020907@sun.com> Andrew John Hughes wrote: > 2009/4/1 Andrew John Hughes : >> 2009/4/1 Kelly O'Hair : >>> Andrew Haley wrote: >>>> We at Red Hat have noticed that the list of IcedTea local patches has >>>> been getting large, with some local patches that should have been >>>> pushed upstream. Also, some IcedTea patches have been committed to >>>> OpenJDK 7 but not OpenJDK 6. >>>> >>>> So, we're going to have a serious attempt to minimize the number of >>>> IcedTea patches. We'll create a Wiki page at icedtea.classpath.org >>>> that is a list of the patches, and for every one we're going to >>>> discuss whether it should stay IcedTea local, be pushed upstream to >>>> OpenJDK, or be deleted altogether. Then we will create OpenJDK >>>> bugzilla entries for the patches that we think should be integrated >>>> upstream and work with Sun engineers to get that done. >>>> >>>> I hope that it will be possible to reduce the number of patches to a >>>> fraction of what we have at the moment, while improving the quality of >>>> the upstream product. >>>> >>>> This is going to be quite a lot of work, so we'd appreciate help from >>>> anyone. >>>> >>>> I hope this will meet with everyone's approval. >>>> >>>> Andrew. >>> Sounds great. >>> >>> I can certainly help out with any build related (makefiles etc.) >>> changes. >>> >>> -kto >>> >>> >> Good, because to my knowledge, those form the majority :) >> However, my guess would also be that some require extensive reworking >> to be generally acceptable (e.g. providing the option to build using >> system libraries rather than forcing it). >> -- >> Andrew :-) >> >> Free Java Software Engineer >> Red Hat, Inc. (http://www.redhat.com) >> >> Support Free Java! >> Contribute to GNU Classpath and the OpenJDK >> http://www.gnu.org/software/classpath >> http://openjdk.java.net >> >> PGP Key: 94EFD9D8 (http://subkeys.pgp.net) >> Fingerprint: F8EF F1EA 401E 2E60 15FA 7927 142C 2591 94EF D9D8 >> > > On the other hand, there is some low-hanging fruit. Andrew Haley > reminded me of this patch earlier today: > > http://bugs.sun.com/bugdatabase/view_bug.do?bug_id=6781583 > > which went into OpenJDK7. The patch itself should be trivial to > approve, given it has already been through the process once. The main > issue is that it relies on OpenJDK6 being updated to hs14. Is there > any news on this and of the public stable branch of this version? > I made sure there is a '6-open' version of this bug. Not sure what the state of hs14 in OpenJDK6 is, but if hs14 is 'done' we may need to apply this patch to OpenJDK6's hotspot, making it a hs14+ kind of. I'll pass this decision onto Joe. ;^) -kto > As to committing, do those on the OpenJDK commit list have access to > all repositories (given patch approval obviously) or is it restricted > in some way? As myself, Gary and Andrew Haley all are now on this > list, it would be good to start the process of committing patches > ourself once approved. > > Thanks, From gnu_andrew at member.fsf.org Wed Apr 1 12:27:18 2009 From: gnu_andrew at member.fsf.org (Andrew John Hughes) Date: Wed, 1 Apr 2009 19:27:18 +0000 Subject: New project: getting rid of IcedTea local patches In-Reply-To: <49D3B416.4020907@sun.com> References: <49D3917B.2010907@redhat.com> <49D3A8B1.4040201@sun.com> <17c6771e0904011055nfa6a0a5pec140dd568cc63e0@mail.gmail.com> <17c6771e0904011103j78adad05h59960f37ed2c32a2@mail.gmail.com> <49D3B416.4020907@sun.com> Message-ID: <17c6771e0904011227v201373cbw6bbc8e218fcb2d19@mail.gmail.com> 2009/4/1 Kelly O'Hair : > > Andrew John Hughes wrote: >> >> 2009/4/1 Andrew John Hughes : >>> >>> 2009/4/1 Kelly O'Hair : >>>> >>>> Andrew Haley wrote: >>>>> >>>>> We at Red Hat have noticed that the list of IcedTea local patches has >>>>> been getting large, with some local patches that should have been >>>>> pushed upstream. ?Also, some IcedTea patches have been committed to >>>>> OpenJDK 7 but not OpenJDK 6. >>>>> >>>>> So, we're going to have a serious attempt to minimize the number of >>>>> IcedTea patches. ?We'll create a Wiki page at icedtea.classpath.org >>>>> that is a list of the patches, and for every one we're going to >>>>> discuss whether it should stay IcedTea local, be pushed upstream to >>>>> OpenJDK, or be deleted altogether. ?Then we will create OpenJDK >>>>> bugzilla entries for the patches that we think should be integrated >>>>> upstream and work with Sun engineers to get that done. >>>>> >>>>> I hope that it will be possible to reduce the number of patches to a >>>>> fraction of what we have at the moment, while improving the quality of >>>>> the upstream product. >>>>> >>>>> This is going to be quite a lot of work, so we'd appreciate help from >>>>> anyone. >>>>> >>>>> I hope this will meet with everyone's approval. >>>>> >>>>> Andrew. >>>> >>>> Sounds great. >>>> >>>> I can certainly help out with any build related (makefiles etc.) >>>> changes. >>>> >>>> -kto >>>> >>>> >>> Good, because to my knowledge, those form the majority :) >>> However, my guess would also be that some require extensive reworking >>> to be generally acceptable (e.g. providing the option to build using >>> system libraries rather than forcing it). >>> -- >>> Andrew :-) >>> >>> Free Java Software Engineer >>> Red Hat, Inc. (http://www.redhat.com) >>> >>> Support Free Java! >>> Contribute to GNU Classpath and the OpenJDK >>> http://www.gnu.org/software/classpath >>> http://openjdk.java.net >>> >>> PGP Key: 94EFD9D8 (http://subkeys.pgp.net) >>> Fingerprint: F8EF F1EA 401E 2E60 15FA ?7927 142C 2591 94EF D9D8 >>> >> >> On the other hand, there is some low-hanging fruit. ?Andrew Haley >> reminded me of this patch earlier today: >> >> http://bugs.sun.com/bugdatabase/view_bug.do?bug_id=6781583 >> >> which went into OpenJDK7. ?The patch itself should be trivial to >> approve, given it has already been through the process once. ?The main >> issue is that it relies on OpenJDK6 being updated to hs14. ?Is there >> any news on this and of the public stable branch of this version? >> > > I made sure there is a '6-open' version of this bug. > Not sure what the state of hs14 in OpenJDK6 is, but if hs14 is > 'done' we may need to apply this patch to OpenJDK6's hotspot, > making it a hs14+ kind of. > I'll pass this decision onto Joe. ;^) > Thanks. To my knowledge, OpenJDK6 is still on hs11. With IcedTea6 (which builds against OpenJDK6 b14), we do an rm -rf hotspot and download the last hs14 changeset to use instead (fc6a5ae3fef5) from http://hg.openjdk.java.net/jdk7/hotspot/hotspot. We then apply three patches to the newly acquired b14 hotspot which achieve the same as the one attached to that bug report (these were developed independently, two of them sometime before that changeset). > -kto > >> As to committing, do those on the OpenJDK commit list have access to >> all repositories (given patch approval obviously) or is it restricted >> in some way? ?As myself, Gary and Andrew Haley all are now on this >> list, it would be good to start the process of committing patches >> ourself once approved. >> >> Thanks, > -- Andrew :-) Free Java Software Engineer Red Hat, Inc. (http://www.redhat.com) Support Free Java! Contribute to GNU Classpath and the OpenJDK http://www.gnu.org/software/classpath http://openjdk.java.net PGP Key: 94EFD9D8 (http://subkeys.pgp.net) Fingerprint: F8EF F1EA 401E 2E60 15FA 7927 142C 2591 94EF D9D8 From kelly.ohair at sun.com Wed Apr 1 17:18:40 2009 From: kelly.ohair at sun.com (kelly.ohair at sun.com) Date: Thu, 02 Apr 2009 00:18:40 +0000 Subject: hg: jdk6/jdk6/jdk: 6824976: Fix NAWK assignment in shell script jdk/make/java/java/genlocales.gmk Message-ID: <20090402001901.94F9DEEC9@hg.openjdk.java.net> Changeset: 4cfe3447dedc Author: ohair Date: 2009-04-01 10:41 -0700 URL: http://hg.openjdk.java.net/jdk6/jdk6/jdk/rev/4cfe3447dedc 6824976: Fix NAWK assignment in shell script jdk/make/java/java/genlocales.gmk Reviewed-by: darcy ! make/java/java/genlocales.gmk From martinrb at google.com Mon Apr 6 14:21:40 2009 From: martinrb at google.com (Martin Buchholz) Date: Mon, 6 Apr 2009 14:21:40 -0700 Subject: jdk-repo-only-build broken in openjdk6 Message-ID: <1ccfd1c10904061421p37271e8ej54414f4d2e9a131@mail.gmail.com> Hi build experts, I tried building just the openjdk6 jdk repo in (more or less) the usual way using 'cd jdk/make && make...' This works in openjdk7 but fails in openjdk6 (and has failed for quite a while) ------------------------------------------------------------ >>>Recursively making smartcardio all @ Sun Apr 5 16:04:57 PDT 2009 ... make[3]: Entering directory `/usr/local/google/home/martin/ws/openjdk6/jdk/make/sun/security/smartcardio' /bin/mkdir -p ../../../../build/linux-i586/tmp/sun/sun.security.smartcardio/j2pcsc/obj_g rm -f ../../../../build/linux-i586/tmp/sun/sun.security.smartcardio/j2pcsc/obj_g/.class.headers.i586 # Running javah: /usr/local/google/SLASH_JAVA/re/openjdk/6/promoted/latest/binaries/linux-i586/bin/javah -bootclasspath ../../../../build/linux-i586/classes -d ../../../../build/linux-i586/tmp/sun/sun.security.smartcardio/j2pcsc/CClassHeaders/ \ sun.security.smartcardio.PCSC sun.security.smartcardio.PlatformPCSC error: cannot access sun.security.smartcardio.PCSC class file for sun.security.smartcardio.PCSC not found javadoc: error - Class sun.security.smartcardio.PCSC not found. error: cannot access sun.security.smartcardio.PlatformPCSC class file for sun.security.smartcardio.PlatformPCSC not found javadoc: error - Class sun.security.smartcardio.PlatformPCSC not found. Error: No classes were specified on the command line. Try -help. make[3]: *** [../../../../build/linux-i586/tmp/sun/sun.security.smartcardio/j2pcsc/obj_g/.class.headers.i586] Error 15 make[3]: Leaving directory `/usr/local/google/home/martin/ws/openjdk6/jdk/make/sun/security/smartcardio' make[2]: *** [all] Error 1 make[2]: Leaving directory `/usr/local/google/home/martin/ws/openjdk6/jdk/make/sun/security' make[1]: *** [all] Error 1 make[1]: Leaving directory `/usr/local/google/home/martin/ws/openjdk6/jdk/make/sun' make: *** [all] Error 1 Command /usr/bin/nice /usr/bin/linux32 make SHELL=/bin/bash VARIANT=DBG sanity all docs images failed: rc=2 ------------------------------------------------------------ It's not obvious why a full control build doesn't encounter this issue. After much frustrating debugging, I found that it's due to the control build passing an absolute JDK_TOPDIR to the jdk repo sub-build. So I have a fine workaround. Define JDK_TOPDIR to the absolute path of the jdk directory. I still don't know where in the makefiles is a dependency on JDK_TOPDIR not being relative. The smartcardio Makefile appears to be blameless. This would be a P2 bug to fix, were it not for the fact that openjdk6 does not see a lot of development. Here's a snippet of perl I use in my build script: { # Work around bug in openjdk6, that requires an absolute JDK_TOPDIR use Cwd 'getcwd'; my $dir = getcwd(); while ($dir ne '') { sub isJdkDir($) { -r "$_[0]/make/common/Program.gmk" } last if isJdkDir "$dir/jdk"; if (isJdkDir $dir) { $ENV{JDK_TOPDIR} = $dir; print "JDK_TOPDIR=$dir\n" if $verbose; last; } $dir =~ s~/[^/]*$~~; } } Hope this helps. Martin From Tim.Bell at Sun.COM Mon Apr 6 14:33:02 2009 From: Tim.Bell at Sun.COM (Tim Bell) Date: Mon, 06 Apr 2009 14:33:02 -0700 Subject: jdk-repo-only-build broken in openjdk6 In-Reply-To: <1ccfd1c10904061421p37271e8ej54414f4d2e9a131@mail.gmail.com> References: <1ccfd1c10904061421p37271e8ej54414f4d2e9a131@mail.gmail.com> Message-ID: <49DA750E.4000204@sun.com> Hi Martin: > I tried building just the openjdk6 jdk repo in (more or less) the usual way > using 'cd jdk/make && make...' > > This works in openjdk7 but fails in openjdk6 (and has failed for quite a while) > > ------------------------------------------------------------ >>>> Recursively making smartcardio all @ Sun Apr 5 16:04:57 PDT 2009 ... > make[3]: Entering directory > `/usr/local/google/home/martin/ws/openjdk6/jdk/make/sun/security/smartcardio' > /bin/mkdir -p ../../../../build/linux-i586/tmp/sun/sun.security.smartcardio/j2pcsc/obj_g > rm -f ../../../../build/linux-i586/tmp/sun/sun.security.smartcardio/j2pcsc/obj_g/.class.headers.i586 > # Running javah: > /usr/local/google/SLASH_JAVA/re/openjdk/6/promoted/latest/binaries/linux-i586/bin/javah > -bootclasspath ../../../../build/linux-i586/classes -d > ../../../../build/linux-i586/tmp/sun/sun.security.smartcardio/j2pcsc/CClassHeaders/ > \ > sun.security.smartcardio.PCSC sun.security.smartcardio.PlatformPCSC > error: cannot access sun.security.smartcardio.PCSC > class file for sun.security.smartcardio.PCSC not found > javadoc: error - Class sun.security.smartcardio.PCSC not found. > error: cannot access sun.security.smartcardio.PlatformPCSC > class file for sun.security.smartcardio.PlatformPCSC not found > javadoc: error - Class sun.security.smartcardio.PlatformPCSC not found. > Error: No classes were specified on the command line. Try -help. > make[3]: *** [../../../../build/linux-i586/tmp/sun/sun.security.smartcardio/j2pcsc/obj_g/.class.headers.i586] > Error 15 > make[3]: Leaving directory > `/usr/local/google/home/martin/ws/openjdk6/jdk/make/sun/security/smartcardio' > make[2]: *** [all] Error 1 > make[2]: Leaving directory > `/usr/local/google/home/martin/ws/openjdk6/jdk/make/sun/security' > make[1]: *** [all] Error 1 > make[1]: Leaving directory > `/usr/local/google/home/martin/ws/openjdk6/jdk/make/sun' > make: *** [all] Error 1 > Command /usr/bin/nice /usr/bin/linux32 make SHELL=/bin/bash > VARIANT=DBG sanity all docs images failed: rc=2 > ------------------------------------------------------------ This fingerprint matches 6668260/6668018 Bug ID: 6668260 build in openjdk/jdk/make fails if openjdk/build exists Bug ID: 6668018 Incremental/partial OpenJDK builds from jdk/make subdirectories do not always work http://bugs.sun.com/bugdatabase/view_bug.do?bug_id=6668260 http://bugs.sun.com/bugdatabase/view_bug.do?bug_id=6668018 Did you previously do a control build at the top level? If so, another workaround is to rename that output directory. In your case, that would be: mv /usr/local/google/home/martin/ws/openjdk6/build \ /usr/local/google/home/martin/ws/openjdk6/build.ignore HTH - Tim > It's not obvious why a full control build doesn't encounter this issue. > After much frustrating debugging, I found that it's due to the control build > passing an absolute JDK_TOPDIR to the jdk repo sub-build. > > So I have a fine workaround. Define JDK_TOPDIR to the absolute path > of the jdk directory. I still don't know where in the makefiles is a > dependency on JDK_TOPDIR not being relative. The smartcardio Makefile > appears to be blameless. This would be a P2 bug to fix, > were it not for the fact that openjdk6 does not see a lot of development. > > Here's a snippet of perl I use in my build script: > > { > # Work around bug in openjdk6, that requires an absolute JDK_TOPDIR > use Cwd 'getcwd'; > my $dir = getcwd(); > while ($dir ne '') { > sub isJdkDir($) { -r "$_[0]/make/common/Program.gmk" } > last if isJdkDir "$dir/jdk"; > if (isJdkDir $dir) { > $ENV{JDK_TOPDIR} = $dir; > print "JDK_TOPDIR=$dir\n" if $verbose; > last; > } > $dir =~ s~/[^/]*$~~; > } > } > > Hope this helps. > > Martin From martinrb at google.com Mon Apr 6 14:55:03 2009 From: martinrb at google.com (Martin Buchholz) Date: Mon, 6 Apr 2009 14:55:03 -0700 Subject: jdk-repo-only-build broken in openjdk6 In-Reply-To: <49DA750E.4000204@sun.com> References: <1ccfd1c10904061421p37271e8ej54414f4d2e9a131@mail.gmail.com> <49DA750E.4000204@sun.com> Message-ID: <1ccfd1c10904061455o613d25c4p86e2faa6226928f2@mail.gmail.com> On Mon, Apr 6, 2009 at 14:33, Tim Bell wrote: > Hi Martin: > > Bug ID: 6668260 build in openjdk/jdk/make fails if openjdk/build exists > Bug ID: 6668018 Incremental/partial OpenJDK builds from jdk/make > subdirectories do not always work > > http://bugs.sun.com/bugdatabase/view_bug.do?bug_id=6668260 > http://bugs.sun.com/bugdatabase/view_bug.do?bug_id=6668018 > > Did you previously do a control build at the top level? Ah yes indeed - good clue! ?If so, another > workaround is to rename that output directory. ?In your case, that would be: > ?mv /usr/local/google/home/martin/ws/openjdk6/build \ > ? ? /usr/local/google/home/martin/ws/openjdk6/build.ignore I think my JDK_TOPDIR workaround is another clue, that should be added to the bug report. Since it appears that this bug is also in openjdk7, I would make it a P2, just because of the extreme frustration developers are likely to encounter. The Makefiles could simply make JDK_TOPDIR absolute, but that has the problem that FullPath is defined in a Makefile included using a path that uses JDK_TOPDIR. Martin > > HTH - Tim > > >> It's not obvious why a full control build doesn't encounter this issue. >> After much frustrating debugging, I found that it's due to the control >> build >> passing an absolute JDK_TOPDIR to the jdk repo sub-build. >> >> So I have a fine workaround. ?Define JDK_TOPDIR to the absolute path >> of the jdk directory. ?I still don't know where in the makefiles is a >> dependency on JDK_TOPDIR not being relative. ?The smartcardio Makefile >> appears to be blameless. ?This would be a P2 bug to fix, >> were it not for the fact that openjdk6 does not see a lot of development. >> >> Here's a snippet of perl I use in my build script: >> >> { >> ?# Work around bug in openjdk6, that requires an absolute JDK_TOPDIR >> ?use Cwd 'getcwd'; >> ?my $dir = getcwd(); >> ?while ($dir ne '') { >> ? ?sub isJdkDir($) { -r "$_[0]/make/common/Program.gmk" } >> ? ?last if isJdkDir "$dir/jdk"; >> ? ?if (isJdkDir $dir) { >> ? ? ?$ENV{JDK_TOPDIR} = $dir; >> ? ? ?print "JDK_TOPDIR=$dir\n" if $verbose; >> ? ? ?last; >> ? ?} >> ? ?$dir =~ s~/[^/]*$~~; >> ?} >> } >> >> Hope this helps. >> >> Martin > > From alex.menkov at sun.com Wed Apr 8 02:20:25 2009 From: alex.menkov at sun.com (alex.menkov at sun.com) Date: Wed, 08 Apr 2009 09:20:25 +0000 Subject: hg: jdk6/jdk6/jdk: 6806019: 38 JCK api/javax_sound/midi/ tests fails starting from jdk7 b46 Message-ID: <20090408092042.A1000E606@hg.openjdk.java.net> Changeset: 2b6b4a913149 Author: amenkov Date: 2009-04-08 13:10 +0400 URL: http://hg.openjdk.java.net/jdk6/jdk6/jdk/rev/2b6b4a913149 6806019: 38 JCK api/javax_sound/midi/ tests fails starting from jdk7 b46 Reviewed-by: kalli ! src/share/classes/com/sun/media/sound/SoftSynthesizer.java From gnu_andrew at member.fsf.org Wed Apr 8 12:58:05 2009 From: gnu_andrew at member.fsf.org (Andrew John Hughes) Date: Wed, 8 Apr 2009 20:58:05 +0100 Subject: Binary files in Nimbus Message-ID: <17c6771e0904081258x114ae22eyf03608e3f5e9c1a2@mail.gmail.com> >Peter, > >Earlier today you pushed Nimbus into the JDK 7 Swing forest [1]. I'm >happy to see Nimbus get in, but later on the #openjdk IRC channel Andrew >Hughes pointed out that the changes include seven binary jar files [2]. > >When we open-sourced JDK 7 we took great pains to eliminate all binary >files from the source tree, and as a matter of policy they are no longer >allowed. In retrospect we could've communicated this more often and more >widely, and we'll certainly do that going forward. I'm also going to >look into enhancing the jcheck extension to check for binary files. > >In the meantime, we need to get rid of these jar files. I suggest that >the best route forward is for me to do a rollback of the Nimbus changeset >on the server side and then for you to rework the code so that these jar >files are no longer required. > >I'd still like to see Nimbus in M3, and I'll support integration into a >later build as long as it's okay with SQE, but removing these jar files >is a higher priority. Aside from the fact that they're binary files, >some of them contain code that has not been approved by Sun Legal for >inclusion in OpenJDK. > >If anyone sees a problem with this plan, please let me know ASAP. > >- Mark > > >[1] http://hg.openjdk.java.net/jdk7/swing/jdk/rev/c5cd40f1f596 >[2] http://hg.openjdk.java.net/jdk7/swing/jdk/file/c5cd40f1f596/make/tools/swing-nimbus/libs/ My thanks to Mark for raising this. Blobs of unknown JAR code have been a problem in the past for including OpenJDK in the GNU/Linux distributions, and it would be a shame to take a backwards step on this. These ones do just appear to be http://jibx.sourceforge.net/, so providing some way to point to the JAR files instead should be sufficient. awt16.jar can be dumped, given OpenJDK (AFAIR) is supposed to be built with a 1.6 JDK. On the positive side, the patch applies to OpenJDK6 with a few simple modifications, and by copying across the JAR files manually to make/tools/swing-nimbus/libs, I was able to build with the following patch: https://bugs.openjdk.java.net/show_bug.cgi?id=100037 So anyone wanting to try Nimbus on OpenJDK6 today, feel free :) -- Andrew :-) Free Java Software Engineer Red Hat, Inc. (http://www.redhat.com) Support Free Java! Contribute to GNU Classpath and the OpenJDK http://www.gnu.org/software/classpath http://openjdk.java.net PGP Key: 94EFD9D8 (http://subkeys.pgp.net) Fingerprint: F8EF F1EA 401E 2E60 15FA 7927 142C 2591 94EF D9D8 From Joe.Darcy at Sun.COM Wed Apr 8 22:15:05 2009 From: Joe.Darcy at Sun.COM (Joseph D. Darcy) Date: Wed, 08 Apr 2009 22:15:05 -0700 Subject: Binary files in Nimbus In-Reply-To: <17c6771e0904081258x114ae22eyf03608e3f5e9c1a2@mail.gmail.com> References: <17c6771e0904081258x114ae22eyf03608e3f5e9c1a2@mail.gmail.com> Message-ID: <49DD8459.80705@sun.com> Andrew John Hughes wrote: >> Peter, >> >> Earlier today you pushed Nimbus into the JDK 7 Swing forest [1]. I'm >> happy to see Nimbus get in, but later on the #openjdk IRC channel Andrew >> Hughes pointed out that the changes include seven binary jar files [2]. >> >> When we open-sourced JDK 7 we took great pains to eliminate all binary >> files from the source tree, and as a matter of policy they are no longer >> allowed. In retrospect we could've communicated this more often and more >> widely, and we'll certainly do that going forward. I'm also going to >> look into enhancing the jcheck extension to check for binary files. >> >> In the meantime, we need to get rid of these jar files. I suggest that >> the best route forward is for me to do a rollback of the Nimbus changeset >> on the server side and then for you to rework the code so that these jar >> files are no longer required. >> >> I'd still like to see Nimbus in M3, and I'll support integration into a >> later build as long as it's okay with SQE, but removing these jar files >> is a higher priority. Aside from the fact that they're binary files, >> some of them contain code that has not been approved by Sun Legal for >> inclusion in OpenJDK. >> >> If anyone sees a problem with this plan, please let me know ASAP. >> >> - Mark >> >> >> [1] http://hg.openjdk.java.net/jdk7/swing/jdk/rev/c5cd40f1f596 >> [2] http://hg.openjdk.java.net/jdk7/swing/jdk/file/c5cd40f1f596/make/tools/swing-nimbus/libs/ >> > > My thanks to Mark for raising this. Blobs of unknown JAR code have > been a problem in the past for including OpenJDK in the GNU/Linux > distributions, and it would be a shame to take a backwards step on > this. > These ones do just appear to be http://jibx.sourceforge.net/, so > providing some way to point to the JAR files instead should be > sufficient. awt16.jar can be dumped, given OpenJDK (AFAIR) is > supposed to be built with a 1.6 JDK. > > On the positive side, the patch applies to OpenJDK6 with a few simple > modifications, and by copying across the JAR files manually to > make/tools/swing-nimbus/libs, I was able to build with the following > patch: > > https://bugs.openjdk.java.net/show_bug.cgi?id=100037 > > So anyone wanting to try Nimbus on OpenJDK6 today, feel free :) > Once the binary file problems are resolved, I'm open to having Nimbus ported to the OpenJDK 6 master repository. -Joe From gnu_andrew at member.fsf.org Thu Apr 9 00:50:34 2009 From: gnu_andrew at member.fsf.org (Andrew John Hughes) Date: Thu, 9 Apr 2009 08:50:34 +0100 Subject: Binary files in Nimbus In-Reply-To: <49DD8459.80705@sun.com> References: <17c6771e0904081258x114ae22eyf03608e3f5e9c1a2@mail.gmail.com> <49DD8459.80705@sun.com> Message-ID: <17c6771e0904090050x4d6f319fl3421daf5cec3cfb4@mail.gmail.com> 2009/4/9 Joseph D. Darcy : > Andrew John Hughes wrote: >>> >>> Peter, >>> >>> Earlier today you pushed Nimbus into the JDK 7 Swing forest [1]. ?I'm >>> happy to see Nimbus get in, but later on the #openjdk IRC channel Andrew >>> Hughes pointed out that the changes include seven binary jar files [2]. >>> >>> When we open-sourced JDK 7 we took great pains to eliminate all binary >>> files from the source tree, and as a matter of policy they are no longer >>> allowed. ?In retrospect we could've communicated this more often and more >>> widely, and we'll certainly do that going forward. ?I'm also going to >>> look into enhancing the jcheck extension to check for binary files. >>> >>> In the meantime, we need to get rid of these jar files. ?I suggest that >>> the best route forward is for me to do a rollback of the Nimbus changeset >>> on the server side and then for you to rework the code so that these jar >>> files are no longer required. >>> >>> I'd still like to see Nimbus in M3, and I'll support integration into a >>> later build as long as it's okay with SQE, but removing these jar files >>> is a higher priority. ?Aside from the fact that they're binary files, >>> some of them contain code that has not been approved by Sun Legal for >>> inclusion in OpenJDK. >>> >>> If anyone sees a problem with this plan, please let me know ASAP. >>> >>> - Mark >>> >>> >>> [1] http://hg.openjdk.java.net/jdk7/swing/jdk/rev/c5cd40f1f596 >>> [2] >>> http://hg.openjdk.java.net/jdk7/swing/jdk/file/c5cd40f1f596/make/tools/swing-nimbus/libs/ >>> >> >> My thanks to Mark for raising this. ?Blobs of unknown JAR code have >> been a problem in the past for including OpenJDK in the GNU/Linux >> distributions, and it would be a shame to take a backwards step on >> this. >> These ones do just appear to be http://jibx.sourceforge.net/, so >> providing some way to point to the JAR files instead should be >> sufficient. ?awt16.jar can be dumped, given OpenJDK (AFAIR) is >> supposed to be built with a 1.6 JDK. >> >> On the positive side, the patch applies to OpenJDK6 with a few simple >> modifications, and by copying across the JAR files manually to >> make/tools/swing-nimbus/libs, I was able to build with the following >> patch: >> >> https://bugs.openjdk.java.net/show_bug.cgi?id=100037 >> >> So anyone wanting to try Nimbus on OpenJDK6 today, feel free :) >> > > Once the binary file problems are resolved, I'm open to having Nimbus ported > to the OpenJDK 6 master repository. > > -Joe > Great! From a little testing yesterday, it already seems to work fine on current OpenJDK6, bar the binary issue. -- Andrew :-) Free Java Software Engineer Red Hat, Inc. (http://www.redhat.com) Support Free Java! Contribute to GNU Classpath and the OpenJDK http://www.gnu.org/software/classpath http://openjdk.java.net PGP Key: 94EFD9D8 (http://subkeys.pgp.net) Fingerprint: F8EF F1EA 401E 2E60 15FA 7927 142C 2591 94EF D9D8 From martinrb at google.com Thu Apr 9 10:23:24 2009 From: martinrb at google.com (martinrb at google.com) Date: Thu, 09 Apr 2009 17:23:24 +0000 Subject: hg: jdk6/jdk6/jdk: 6803402: Race condition in AbstractQueuedSynchronizer Message-ID: <20090409172338.B3382E712@hg.openjdk.java.net> Changeset: 8013150dfe8e Author: dl Date: 2009-04-09 10:17 -0700 URL: http://hg.openjdk.java.net/jdk6/jdk6/jdk/rev/8013150dfe8e 6803402: Race condition in AbstractQueuedSynchronizer Summary: Read fields in reverse initialization order Reviewed-by: martin ! src/share/classes/java/util/concurrent/locks/AbstractQueuedLongSynchronizer.java ! src/share/classes/java/util/concurrent/locks/AbstractQueuedSynchronizer.java From gnu_andrew at member.fsf.org Thu Apr 9 10:55:55 2009 From: gnu_andrew at member.fsf.org (Andrew John Hughes) Date: Thu, 9 Apr 2009 18:55:55 +0100 Subject: New project: getting rid of IcedTea local patches In-Reply-To: <49DE0442.6060802@sun.com> References: <49D3917B.2010907@redhat.com> <49DDF4C6.1070602@redhat.com> <49DE0442.6060802@sun.com> Message-ID: <17c6771e0904091055w7c14a9adg982e0ff37a55e0a4@mail.gmail.com> 2009/4/9 Alan Bateman : > Andrew Haley wrote: >> >> : >> The page is at http://icedtea.classpath.org/wiki/IcedTea_JDK6_Patches >> These are the patches in icedtea/patches and icedtea/patches/hotspot. >> >> >> * Every patch should be assigned to someone. >> >> * Every patch should be discussed. >> >> * If we can, we will push a patch upstream. ?If we can't the discussion >> ?should explain why we can't. >> >> Here we go... >> >> Andrew. >> > > I'm not very familiar with IcedTea but I cloned the repository to see what > these patches are about. Here are a few comments: > > icedtea-6728542-epoll.patch seems to be a change-set I pushed to jdk7 last > year. It seems very reasonable to push this into jdk6/jdk6/jdk. Go for it! > Yes, I've referenced the bug on the wiki now. Joe, can we commit this to 6? > icedtea-bytebuffer-compact.patch - this looks to be a partial fix to > 6593946. If so, 6593946 is already in jdk6/jdk6/jdk > (http://hg.openjdk.java.net/jdk6/jdk6/jdk/rev/3e7bfcdcf96f). > IcedTea6 is still based on the last build drop of jdk6, not the Mercurial repository. So this 6 week old patch won't be in the 25th of November b14 tarball. Interestingly, it also still exists in IcedTea7 as there were cases covered by the patch which didn't make it into the commit. I've attached that version. It was originally added by Keith Seitz while he was working on OpenJDK at Red Hat (according to the ChangeLog). > icedtea-io_util-overflow.patch - this looks like the initial fix for > 6788196. Martin has already pushed this to jdk6/jdk7/jdk > (http://hg.openjdk.java.net/jdk6/jdk6/jdk/rev/cc4d93e8ae7d). > Same case as the epoll patch, not in b14. > icedtea-jvmtiEnv.patch - I thought JvmtiEnv::GetLineNumberTable already > handled this case so I'm curious why this one is needed. > Added by Andrew Haley as part of work on Shark; perhaps he can explain. > icedtea-network-unreachable.patch - I'm pretty sure net-dev would be open to > this because it looks to be cases where the specific exceptions aren't > thrown in the timed read case. > >From our intern, Omair so this is covered by the SCA. Omair, maybe you could file a bug report for this (or link to one here if there is one)? > icedtea-sunsrc.patch - I'm not sure what this is about but it looks like it > removes the service configuration files. Does JDI still work? > Don't know about this one, all I can say is it was added by Joshua Sumali in late 2007. Perhaps someone at Red Hat then can shed some light on it? > icedtea-testenv.patch - the changes to > test/sun/net/InetAddress/nameservice/dns/cname.sh seems reasonable (pilot > error is why this one was missed when the other tests were updated). > Recent patch from Mark Wielaard. Mark? > icedtea-jtreg-hatrun.patch - I suspect this is 6545295. We should just fix > this test (serviceability-dev). > Again, this one is from Omair, so he should be able to provide more detail. > -Alan. > > PS: I'm not on distro-pkg-dev so this mail might need the moderator to > approve. > > Thanks again for the conf. call, very useful! Cheers, -- Andrew :-) Free Java Software Engineer Red Hat, Inc. (http://www.redhat.com) Support Free Java! Contribute to GNU Classpath and the OpenJDK http://www.gnu.org/software/classpath http://openjdk.java.net PGP Key: 94EFD9D8 (http://subkeys.pgp.net) Fingerprint: F8EF F1EA 401E 2E60 15FA 7927 142C 2591 94EF D9D8 From gnu_andrew at member.fsf.org Thu Apr 9 11:01:18 2009 From: gnu_andrew at member.fsf.org (Andrew John Hughes) Date: Thu, 9 Apr 2009 19:01:18 +0100 Subject: New project: getting rid of IcedTea local patches In-Reply-To: <17c6771e0904091055w7c14a9adg982e0ff37a55e0a4@mail.gmail.com> References: <49D3917B.2010907@redhat.com> <49DDF4C6.1070602@redhat.com> <49DE0442.6060802@sun.com> <17c6771e0904091055w7c14a9adg982e0ff37a55e0a4@mail.gmail.com> Message-ID: <17c6771e0904091101t634e419ctfaf5cc6c8b264a90@mail.gmail.com> 2009/4/9 Andrew John Hughes : > 2009/4/9 Alan Bateman : >> Andrew Haley wrote: >>> >>> : >>> The page is at http://icedtea.classpath.org/wiki/IcedTea_JDK6_Patches >>> These are the patches in icedtea/patches and icedtea/patches/hotspot. >>> >>> >>> * Every patch should be assigned to someone. >>> >>> * Every patch should be discussed. >>> >>> * If we can, we will push a patch upstream. ?If we can't the discussion >>> ?should explain why we can't. >>> >>> Here we go... >>> >>> Andrew. >>> >> >> I'm not very familiar with IcedTea but I cloned the repository to see what >> these patches are about. Here are a few comments: >> >> icedtea-6728542-epoll.patch seems to be a change-set I pushed to jdk7 last >> year. It seems very reasonable to push this into jdk6/jdk6/jdk. Go for it! >> > > Yes, I've referenced the bug on the wiki now. ?Joe, can we commit this to 6? > >> icedtea-bytebuffer-compact.patch - this looks to be a partial fix to >> 6593946. If so, 6593946 is already in jdk6/jdk6/jdk >> (http://hg.openjdk.java.net/jdk6/jdk6/jdk/rev/3e7bfcdcf96f). >> > > IcedTea6 is still based on the last build drop of jdk6, not the > Mercurial repository. ?So this 6 week old patch won't be in the 25th > of November b14 tarball. > Interestingly, it also still exists in IcedTea7 as there were cases > covered by the patch which didn't make it into the commit. > I've attached that version. ?It was originally added by Keith Seitz > while he was working on OpenJDK at Red Hat (according to the > ChangeLog). > >> icedtea-io_util-overflow.patch - this looks like the initial fix for >> 6788196. Martin has already pushed this to jdk6/jdk7/jdk >> (http://hg.openjdk.java.net/jdk6/jdk6/jdk/rev/cc4d93e8ae7d). >> > > Same case as the epoll patch, not in b14. > >> icedtea-jvmtiEnv.patch - I thought JvmtiEnv::GetLineNumberTable already >> handled this case so I'm curious why this one is needed. >> > > Added by Andrew Haley as part of work on Shark; perhaps he can explain. > >> icedtea-network-unreachable.patch - I'm pretty sure net-dev would be open to >> this because it looks to be cases where the specific exceptions aren't >> thrown in the timed read case. >> > > From our intern, Omair so this is covered by the SCA. ?Omair, maybe > you could file a bug report for this (or link to one here if there is > one)? > >> icedtea-sunsrc.patch - I'm not sure what this is about but it looks like it >> removes the service configuration files. Does JDI still work? >> > > Don't know about this one, all I can say is it was added by Joshua > Sumali in late 2007. ?Perhaps someone at Red Hat then can shed some > light on it? > >> icedtea-testenv.patch - the changes to >> test/sun/net/InetAddress/nameservice/dns/cname.sh seems reasonable (pilot >> error is why this one was missed when the other tests were updated). >> > > Recent patch from Mark Wielaard. ?Mark? > >> icedtea-jtreg-hatrun.patch - I suspect this is 6545295. We should just fix >> this test (serviceability-dev). >> > > Again, this one is from Omair, so he should be able to provide more detail. > >> -Alan. >> >> PS: I'm not on distro-pkg-dev so this mail might need the moderator to >> approve. >> >> > > Thanks again for the conf. call, very useful! > > Cheers, > -- > Andrew :-) > > Free Java Software Engineer > Red Hat, Inc. (http://www.redhat.com) > > Support Free Java! > Contribute to GNU Classpath and the OpenJDK > http://www.gnu.org/software/classpath > http://openjdk.java.net > > PGP Key: 94EFD9D8 (http://subkeys.pgp.net) > Fingerprint: F8EF F1EA 401E 2E60 15FA ?7927 142C 2591 94EF D9D8 > And then forgot the patch! -- Andrew :-) Free Java Software Engineer Red Hat, Inc. (http://www.redhat.com) Support Free Java! Contribute to GNU Classpath and the OpenJDK http://www.gnu.org/software/classpath http://openjdk.java.net PGP Key: 94EFD9D8 (http://subkeys.pgp.net) Fingerprint: F8EF F1EA 401E 2E60 15FA 7927 142C 2591 94EF D9D8 -------------- next part -------------- A non-text attachment was scrubbed... Name: icedtea-bytebuffer-compact.patch Type: application/octet-stream Size: 1333 bytes Desc: not available Url : http://mail.openjdk.java.net/pipermail/jdk6-dev/attachments/20090409/5bbf5b8b/attachment.obj From aph at redhat.com Thu Apr 9 11:04:07 2009 From: aph at redhat.com (Andrew Haley) Date: Thu, 09 Apr 2009 19:04:07 +0100 Subject: New project: getting rid of IcedTea local patches In-Reply-To: <17c6771e0904091055w7c14a9adg982e0ff37a55e0a4@mail.gmail.com> References: <49D3917B.2010907@redhat.com> <49DDF4C6.1070602@redhat.com> <49DE0442.6060802@sun.com> <17c6771e0904091055w7c14a9adg982e0ff37a55e0a4@mail.gmail.com> Message-ID: <49DE3897.5050902@redhat.com> Andrew John Hughes wrote: > 2009/4/9 Alan Bateman : >> icedtea-jvmtiEnv.patch - I thought JvmtiEnv::GetLineNumberTable already >> handled this case so I'm curious why this one is needed. > > Added by Andrew Haley as part of work on Shark; perhaps he can explain. GetLineNumberTable checks for a missing table but not for a zero-length one. I had a crash in Shark, but I can't remember exactly where. Andrew. From omajid at redhat.com Thu Apr 9 11:05:18 2009 From: omajid at redhat.com (Omair Majid) Date: Thu, 09 Apr 2009 14:05:18 -0400 Subject: New project: getting rid of IcedTea local patches In-Reply-To: <17c6771e0904091055w7c14a9adg982e0ff37a55e0a4@mail.gmail.com> References: <49D3917B.2010907@redhat.com> <49DDF4C6.1070602@redhat.com> <49DE0442.6060802@sun.com> <17c6771e0904091055w7c14a9adg982e0ff37a55e0a4@mail.gmail.com> Message-ID: <49DE38DE.1050505@redhat.com> Andrew John Hughes wrote: > 2009/4/9 Alan Bateman : >> Andrew Haley wrote: >>> : >>> The page is at http://icedtea.classpath.org/wiki/IcedTea_JDK6_Patches >>> These are the patches in icedtea/patches and icedtea/patches/hotspot. >>> >>> >>> * Every patch should be assigned to someone. >>> >>> * Every patch should be discussed. >>> >>> * If we can, we will push a patch upstream. If we can't the discussion >>> should explain why we can't. >>> >>> Here we go... >>> >>> Andrew. >>> >> I'm not very familiar with IcedTea but I cloned the repository to see what >> these patches are about. Here are a few comments: >> >> icedtea-6728542-epoll.patch seems to be a change-set I pushed to jdk7 last >> year. It seems very reasonable to push this into jdk6/jdk6/jdk. Go for it! >> > > Yes, I've referenced the bug on the wiki now. Joe, can we commit this to 6? > >> icedtea-bytebuffer-compact.patch - this looks to be a partial fix to >> 6593946. If so, 6593946 is already in jdk6/jdk6/jdk >> (http://hg.openjdk.java.net/jdk6/jdk6/jdk/rev/3e7bfcdcf96f). >> > > IcedTea6 is still based on the last build drop of jdk6, not the > Mercurial repository. So this 6 week old patch won't be in the 25th > of November b14 tarball. > Interestingly, it also still exists in IcedTea7 as there were cases > covered by the patch which didn't make it into the commit. > I've attached that version. It was originally added by Keith Seitz > while he was working on OpenJDK at Red Hat (according to the > ChangeLog). > >> icedtea-io_util-overflow.patch - this looks like the initial fix for >> 6788196. Martin has already pushed this to jdk6/jdk7/jdk >> (http://hg.openjdk.java.net/jdk6/jdk6/jdk/rev/cc4d93e8ae7d). >> > > Same case as the epoll patch, not in b14. > >> icedtea-jvmtiEnv.patch - I thought JvmtiEnv::GetLineNumberTable already >> handled this case so I'm curious why this one is needed. >> > > Added by Andrew Haley as part of work on Shark; perhaps he can explain. > >> icedtea-network-unreachable.patch - I'm pretty sure net-dev would be open to >> this because it looks to be cases where the specific exceptions aren't >> thrown in the timed read case. >> > > From our intern, Omair so this is covered by the SCA. Omair, maybe > you could file a bug report for this (or link to one here if there is > one)? > Just got around to filing this one today: http://bugs.openjdk.java.net/show_bug.cgi?id=100039 >> icedtea-sunsrc.patch - I'm not sure what this is about but it looks like it >> removes the service configuration files. Does JDI still work? >> > > Don't know about this one, all I can say is it was added by Joshua > Sumali in late 2007. Perhaps someone at Red Hat then can shed some > light on it? > >> icedtea-testenv.patch - the changes to >> test/sun/net/InetAddress/nameservice/dns/cname.sh seems reasonable (pilot >> error is why this one was missed when the other tests were updated). >> > > Recent patch from Mark Wielaard. Mark? > >> icedtea-jtreg-hatrun.patch - I suspect this is 6545295. We should just fix >> this test (serviceability-dev). >> > > Again, this one is from Omair, so he should be able to provide more detail. > Filed as http://bugs.openjdk.java.net/show_bug.cgi?id=100026 >> -Alan. >> >> PS: I'm not on distro-pkg-dev so this mail might need the moderator to >> approve. >> >> > > Thanks again for the conf. call, very useful! > > Cheers, Cheers, Omair From Alan.Bateman at Sun.COM Thu Apr 9 12:08:55 2009 From: Alan.Bateman at Sun.COM (Alan Bateman) Date: Thu, 09 Apr 2009 20:08:55 +0100 Subject: New project: getting rid of IcedTea local patches In-Reply-To: <17c6771e0904091055w7c14a9adg982e0ff37a55e0a4@mail.gmail.com> References: <49D3917B.2010907@redhat.com> <49DDF4C6.1070602@redhat.com> <49DE0442.6060802@sun.com> <17c6771e0904091055w7c14a9adg982e0ff37a55e0a4@mail.gmail.com> Message-ID: <49DE47C7.1040908@sun.com> Andrew John Hughes wrote: > : >> icedtea-bytebuffer-compact.patch - this looks to be a partial fix to >> 6593946. If so, 6593946 is already in jdk6/jdk6/jdk >> (http://hg.openjdk.java.net/jdk6/jdk6/jdk/rev/3e7bfcdcf96f). >> >> > > IcedTea6 is still based on the last build drop of jdk6, not the > Mercurial repository. So this 6 week old patch won't be in the 25th > of November b14 tarball. > Interestingly, it also still exists in IcedTea7 as there were cases > covered by the patch which didn't make it into the commit. > I've attached that version. It was originally added by Keith Seitz > while he was working on OpenJDK at Red Hat (according to the > ChangeLog). > Once the fix gets into a build (b16?) then I don't think you'll need the IcedTea6 patch. Also, it does seem that the IcedTea6 patch is only a partial fix in that I don't see the changes for the view buffers. I looked at the patch you attached in the last mail (is this the IcedTea patch to jdk7?). It seems to be a no-op, or at least just replaces setting of mark to -1 with calls to discardMark that does the same thing. Nothing wrong with it, just inconsistent with the existing code. -Alan. From gnu_andrew at member.fsf.org Thu Apr 9 12:27:03 2009 From: gnu_andrew at member.fsf.org (Andrew John Hughes) Date: Thu, 9 Apr 2009 20:27:03 +0100 Subject: New project: getting rid of IcedTea local patches In-Reply-To: <49DE47C7.1040908@sun.com> References: <49D3917B.2010907@redhat.com> <49DDF4C6.1070602@redhat.com> <49DE0442.6060802@sun.com> <17c6771e0904091055w7c14a9adg982e0ff37a55e0a4@mail.gmail.com> <49DE47C7.1040908@sun.com> Message-ID: <17c6771e0904091227rde7ae31g72602473b1bbf786@mail.gmail.com> 2009/4/9 Alan Bateman : > Andrew John Hughes wrote: >> >> : >>> >>> icedtea-bytebuffer-compact.patch - this looks to be a partial fix to >>> 6593946. If so, 6593946 is already in jdk6/jdk6/jdk >>> (http://hg.openjdk.java.net/jdk6/jdk6/jdk/rev/3e7bfcdcf96f). >>> >>> >> >> IcedTea6 is still based on the last build drop of jdk6, not the >> Mercurial repository. ?So this 6 week old patch won't be in the 25th >> of November b14 tarball. >> Interestingly, it also still exists in IcedTea7 as there were cases >> covered by the patch which didn't make it into the commit. >> I've attached that version. ?It was originally added by Keith Seitz >> while he was working on OpenJDK at Red Hat (according to the >> ChangeLog). >> > > Once the fix gets into a build (b16?) then I don't think you'll need the > IcedTea6 patch. Spot on, we need a new build drop (read tarball) to pick up the recent changes. There hasn't been one since Nov. 2008. Also, it does seem that the IcedTea6 patch is only a partial > fix in that I don't see the changes for the view buffers. Yeah, this is an old patch so I don't know all the origins, but it may have been pushed upstream and revised in the process, or even developed separately at Sun. One has clearMark(), one has discardMark IIRC. I looked at the > patch you attached in the last mail (is this the IcedTea patch to jdk7?). Yes, it's the leftovers after I took out the bits in upstream OpenJDK7 (last build against b50). > It > seems to be a no-op, or at least just replaces setting of mark to -1 with > calls to discardMark that does the same thing. Nothing wrong with it, just > inconsistent with the existing code. > Yeah I agree, it's not a must have. I meant to actually raise this at the time, but never got round to it obviously. Using discardMark is perhaps a little cleaner/maintainable, but has the downside of adding an extra method dispatch. > -Alan. > -- Andrew :-) Free Java Software Engineer Red Hat, Inc. (http://www.redhat.com) Support Free Java! Contribute to GNU Classpath and the OpenJDK http://www.gnu.org/software/classpath http://openjdk.java.net PGP Key: 94EFD9D8 (http://subkeys.pgp.net) Fingerprint: F8EF F1EA 401E 2E60 15FA 7927 142C 2591 94EF D9D8 From Alan.Bateman at Sun.COM Thu Apr 9 12:57:05 2009 From: Alan.Bateman at Sun.COM (Alan Bateman) Date: Thu, 09 Apr 2009 20:57:05 +0100 Subject: New project: getting rid of IcedTea local patches In-Reply-To: <49DE3897.5050902@redhat.com> References: <49D3917B.2010907@redhat.com> <49DDF4C6.1070602@redhat.com> <49DE0442.6060802@sun.com> <17c6771e0904091055w7c14a9adg982e0ff37a55e0a4@mail.gmail.com> <49DE3897.5050902@redhat.com> Message-ID: <49DE5311.5040108@sun.com> Andrew Haley wrote: > Andrew John Hughes wrote: > >> 2009/4/9 Alan Bateman : >> > > >>> icedtea-jvmtiEnv.patch - I thought JvmtiEnv::GetLineNumberTable already >>> handled this case so I'm curious why this one is needed. >>> >> Added by Andrew Haley as part of work on Shark; perhaps he can explain. >> > > GetLineNumberTable checks for a missing table but not for a zero-length > one. I had a crash in Shark, but I can't remember exactly where. > > Andrew. > I need to double-check the VM and JVM TI specs but off-hand, I believe a zero-length table is possible. I wonder if this was a case of the agent not handling the zero-length case. Anyway, if you do remember the details please send them on to serviceability-dev. -Alan. From abhijit.saha at sun.com Thu Apr 9 17:27:05 2009 From: abhijit.saha at sun.com (abhijit.saha at sun.com) Date: Fri, 10 Apr 2009 00:27:05 +0000 Subject: hg: jdk6/jdk6/jdk: 2 new changesets Message-ID: <20090410002737.C0F71E778@hg.openjdk.java.net> Changeset: d5332f2e03e6 Author: asaha Date: 2009-04-09 17:12 -0700 URL: http://hg.openjdk.java.net/jdk6/jdk6/jdk/rev/d5332f2e03e6 6828183: testcase from SSR09_01 into jdk6-open hangs Reviewed-by: prr ! test/java/awt/FontClass/CreateFont/BigFont.java ! test/java/awt/FontClass/CreateFont/bigfont.html ! test/java/awt/FontClass/CreateFont/fileaccess/FontFile.java + test/java/awt/FontClass/CreateFont/fileaccess/TestFontFile.sh Changeset: 4877cd701743 Author: asaha Date: 2009-04-09 17:22 -0700 URL: http://hg.openjdk.java.net/jdk6/jdk6/jdk/rev/4877cd701743 Merge From joe.darcy at sun.com Thu Apr 9 17:50:56 2009 From: joe.darcy at sun.com (joe.darcy at sun.com) Date: Fri, 10 Apr 2009 00:50:56 +0000 Subject: hg: jdk6/jdk6/jdk: 6819886: System.getProperty("os.name") reports Vista on Windows 7 Message-ID: <20090410005124.7031DE798@hg.openjdk.java.net> Changeset: ee3280d79578 Author: darcy Date: 2009-04-09 17:36 -0700 URL: http://hg.openjdk.java.net/jdk6/jdk6/jdk/rev/ee3280d79578 6819886: System.getProperty("os.name") reports Vista on Windows 7 Reviewed-by: jjg ! src/windows/native/java/lang/java_props_md.c From kelly.ohair at sun.com Wed Apr 15 18:41:56 2009 From: kelly.ohair at sun.com (kelly.ohair at sun.com) Date: Thu, 16 Apr 2009 01:41:56 +0000 Subject: hg: jdk6/jdk6/corba: 6830756: jprt configuration changes, remove jprt.config files, specify os revs wanted Message-ID: <20090416014157.9E72CEC58@hg.openjdk.java.net> Changeset: 04ef8357f30e Author: ohair Date: 2009-04-15 17:22 -0700 URL: http://hg.openjdk.java.net/jdk6/jdk6/corba/rev/04ef8357f30e 6830756: jprt configuration changes, remove jprt.config files, specify os revs wanted Reviewed-by: tbell - make/jprt.config ! make/jprt.properties From kelly.ohair at sun.com Wed Apr 15 18:50:35 2009 From: kelly.ohair at sun.com (kelly.ohair at sun.com) Date: Thu, 16 Apr 2009 01:50:35 +0000 Subject: hg: jdk6/jdk6/jaxp: 6830756: jprt configuration changes, remove jprt.config files, specify os revs wanted Message-ID: <20090416015038.035B3EC63@hg.openjdk.java.net> Changeset: b974a2a72ff1 Author: ohair Date: 2009-04-15 17:24 -0700 URL: http://hg.openjdk.java.net/jdk6/jdk6/jaxp/rev/b974a2a72ff1 6830756: jprt configuration changes, remove jprt.config files, specify os revs wanted Reviewed-by: tbell - make/jprt.config ! make/jprt.properties From kelly.ohair at sun.com Wed Apr 15 18:59:52 2009 From: kelly.ohair at sun.com (kelly.ohair at sun.com) Date: Thu, 16 Apr 2009 01:59:52 +0000 Subject: hg: jdk6/jdk6/langtools: 6830756: jprt configuration changes, remove jprt.config files, specify os revs wanted Message-ID: <20090416015955.003B0EC81@hg.openjdk.java.net> Changeset: 59ff92d0fdec Author: ohair Date: 2009-04-15 17:27 -0700 URL: http://hg.openjdk.java.net/jdk6/jdk6/langtools/rev/59ff92d0fdec 6830756: jprt configuration changes, remove jprt.config files, specify os revs wanted Reviewed-by: tbell - make/jprt.config ! make/jprt.properties - test/jprt.config From kelly.ohair at sun.com Wed Apr 15 19:08:43 2009 From: kelly.ohair at sun.com (kelly.ohair at sun.com) Date: Thu, 16 Apr 2009 02:08:43 +0000 Subject: hg: jdk6/jdk6/jaxws: 6830756: jprt configuration changes, remove jprt.config files, specify os revs wanted Message-ID: <20090416020845.591FEEC9D@hg.openjdk.java.net> Changeset: dcd5e14744cf Author: ohair Date: 2009-04-15 17:25 -0700 URL: http://hg.openjdk.java.net/jdk6/jdk6/jaxws/rev/dcd5e14744cf 6830756: jprt configuration changes, remove jprt.config files, specify os revs wanted Reviewed-by: tbell - make/jprt.config ! make/jprt.properties From Joe.Darcy at Sun.COM Wed Apr 15 19:17:36 2009 From: Joe.Darcy at Sun.COM (Joseph D. Darcy) Date: Wed, 15 Apr 2009 19:17:36 -0700 Subject: OpenJDK 6 build 16 source bundle posted Message-ID: <49E69540.80801@sun.com> Without further adieu, the source bundle for OpenJDK 6 b16 is available from http://download.java.net/openjdk/jdk6/ and the OpenJDK 6 Mercurial repositories (http://hg.openjdk.java.net/jdk6/jdk6) will be tagged with the build information accordingly. The previous source bundle was for b14. There was no distinct b15 per se; b15 was used to mark the transition from teamware to Mercurial. The b16 build is the first to be done purely in Mercurial. There were 45 bug fixes in this build ranging from security fixes to sound updates; OpenJDK 6 b16 has all relevant security fixes from the recent JDK synchronized security release. A few fixes of note, during a build importing binary plugs now defaults to false (6781572 "Default to IMPORT_BINARY_PLUGS=false for OpenJDK 6 builds") and several fixes were applied that should render a few IcedTea patches unnecessary: 6736248 EnumEditor bug. Class check incorrect 6733718 test /java/awt/FullScreen/UninitializedDisplayModeChangeTest/ fails 6593946 (bf) X-Buffer.compact() does not discard mark as specified 6778493 Fix (langtools) ant build to honor fcs MILESTONE setting The other non security bugs fixed in this build are: 6761791: Crash in the FontManager code due to use of JNIEnv saved by another thread 6512707: "incompatible types" after (unrelated) annotation processing 6632696: Writing to closed output files (writeBytes) leaks native memory (unix) 6788196: (porting) Bounds checks in io_util.c rely on undefined behaviour 6791458: FileInputStream/RandomAccessFile.read leaks memory if invoked on closed stream with len > 8k 6792066: src/share/native/java/io/io_util.c clean-ups 6819886: System.getProperty("os.name") reports Vista on Windows 7 6821031: Upgrade OpenJDK's LittleCMS version to 1.18 6800572: Removing elements from views of NavigableMap implementations does not always work correctly. 6801020: Concurrent Semaphore release may cause some require thread not signaled 6806019: 38 JCK api/javax_sound/midi/ tests fails starting from jdk7 b46 6803402: Race condition in AbstractQueuedSynchronizer 6793757: Fix formatting of copyright notices in Gervill 6794201: remove unused sources 6808724: UninitializedDisplayModeChangeTest/DisplayModeChanger.java has wrong legal notice 6821030: Merge OpenJDK Gervill with upstream sources, Q1CY2009 6823445: Gervill SoftChannel/ResetAllControllers jtreg test fails after portamento fix from last merge 6823446: Gervill SoftLowFrequencyOscillator fails when freq is set to 0 cent or 8.1758 Hz. 6824976: Fix NAWK assignment in shell script jdk/make/java/java/genlocales.gmk 6828183: testcase from SSR09_01 into jdk6-open hangs -Joe From kelly.ohair at sun.com Wed Apr 15 20:51:13 2009 From: kelly.ohair at sun.com (kelly.ohair at sun.com) Date: Thu, 16 Apr 2009 03:51:13 +0000 Subject: hg: jdk6/jdk6/jdk: 6830756: jprt configuration changes, remove jprt.config files, specify os revs wanted Message-ID: <20090416035133.CD91EECB4@hg.openjdk.java.net> Changeset: 536cbf2d9d0e Author: ohair Date: 2009-04-15 17:34 -0700 URL: http://hg.openjdk.java.net/jdk6/jdk6/jdk/rev/536cbf2d9d0e 6830756: jprt configuration changes, remove jprt.config files, specify os revs wanted Reviewed-by: tbell - make/jprt.config ! make/jprt.properties - test/jprt.config From kelly.ohair at sun.com Wed Apr 15 22:23:49 2009 From: kelly.ohair at sun.com (kelly.ohair at sun.com) Date: Thu, 16 Apr 2009 05:23:49 +0000 Subject: hg: jdk6/jdk6: 6830756: jprt configuration changes, remove jprt.config files, specify os revs wanted Message-ID: <20090416052349.5C518ECCC@hg.openjdk.java.net> Changeset: a78885a02070 Author: ohair Date: 2009-04-15 17:39 -0700 URL: http://hg.openjdk.java.net/jdk6/jdk6/rev/a78885a02070 6830756: jprt configuration changes, remove jprt.config files, specify os revs wanted Reviewed-by: tbell - make/jprt.config ! make/jprt.properties From gnu_andrew at member.fsf.org Thu Apr 16 02:00:59 2009 From: gnu_andrew at member.fsf.org (Andrew John Hughes) Date: Thu, 16 Apr 2009 10:00:59 +0100 Subject: OpenJDK 6 build 16 source bundle posted In-Reply-To: <49E69540.80801@sun.com> References: <49E69540.80801@sun.com> Message-ID: <17c6771e0904160200x76dc925fu353f18b93bd0f1a@mail.gmail.com> 2009/4/16 Joseph D. Darcy : > Without further adieu, the source bundle for OpenJDK 6 b16 is available from > > ? http://download.java.net/openjdk/jdk6/ > > and the OpenJDK 6 Mercurial repositories > (http://hg.openjdk.java.net/jdk6/jdk6) will be tagged with the build > information accordingly. > > The previous source bundle was for b14. ?There was no distinct b15 per se; > b15 was used to mark the transition from teamware to Mercurial. ?The b16 > build is the first to be done purely in Mercurial. > > There were 45 bug fixes in this build ranging from security fixes to sound > updates; OpenJDK 6 b16 has all relevant security fixes from the recent JDK > synchronized security release. ?A few fixes of note, during a build > importing binary plugs now defaults to false (6781572 "Default to > IMPORT_BINARY_PLUGS=false for OpenJDK 6 builds") and several fixes were > applied that should render a few IcedTea patches unnecessary: > > ?6736248 EnumEditor bug. Class check incorrect > ?6733718 test /java/awt/FullScreen/UninitializedDisplayModeChangeTest/ fails > ?6593946 (bf) X-Buffer.compact() does not discard mark as specified > ?6778493 Fix (langtools) ant build to honor fcs MILESTONE setting > > The other non security bugs fixed in this build are: > > 6761791: Crash in the FontManager code due to use of JNIEnv saved by another > thread > 6512707: "incompatible types" after (unrelated) annotation processing > 6632696: Writing to closed output files (writeBytes) leaks native memory > (unix) > 6788196: (porting) Bounds checks in io_util.c rely on undefined behaviour > 6791458: FileInputStream/RandomAccessFile.read leaks memory if invoked on > closed stream with len > 8k > 6792066: src/share/native/java/io/io_util.c clean-ups > 6819886: System.getProperty("os.name") reports Vista on Windows 7 > 6821031: Upgrade OpenJDK's LittleCMS version to 1.18 > 6800572: Removing elements from views of NavigableMap implementations does > not always work correctly. > 6801020: Concurrent Semaphore release may cause some require thread not > signaled > 6806019: 38 JCK api/javax_sound/midi/ tests fails starting from jdk7 b46 > 6803402: Race condition in AbstractQueuedSynchronizer > 6793757: Fix formatting of copyright notices in Gervill > 6794201: remove unused sources > 6808724: UninitializedDisplayModeChangeTest/DisplayModeChanger.java has > wrong legal notice > 6821030: Merge OpenJDK Gervill with upstream sources, Q1CY2009 > 6823445: Gervill SoftChannel/ResetAllControllers jtreg test fails after > portamento fix from last merge > 6823446: Gervill SoftLowFrequencyOscillator fails when freq is set to 0 cent > or 8.1758 Hz. > 6824976: Fix NAWK assignment in shell script > jdk/make/java/java/genlocales.gmk > 6828183: testcase from SSR09_01 into jdk6-open hangs > > -Joe > > Congratulations on the release. -- Andrew :-) Free Java Software Engineer Red Hat, Inc. (http://www.redhat.com) Support Free Java! Contribute to GNU Classpath and the OpenJDK http://www.gnu.org/software/classpath http://openjdk.java.net PGP Key: 94EFD9D8 (http://subkeys.pgp.net) Fingerprint: F8EF F1EA 401E 2E60 15FA 7927 142C 2591 94EF D9D8 From lcid-fire at gmx.net Thu Apr 16 14:01:12 2009 From: lcid-fire at gmx.net (LCID Fire) Date: Thu, 16 Apr 2009 23:01:12 +0200 Subject: Changing jdk order Message-ID: <49E79C98.1020308@gmx.net> On my Ubuntu system I have both the openjdk and sunjdk installed. Trying to generate jni header files in ant screws up totally when openjdk is installed (see http://bugs.launchpad.net/ubuntu/+source/openjdk-6/+bug/361578). Problem is to compile the java stuff I need some classes that come with openjdk. Currently it seems like the system uses openjdk before sunjdk - is there a way to switch/change the orders of the jdks? From doko at ubuntu.com Thu Apr 16 15:08:47 2009 From: doko at ubuntu.com (Matthias Klose) Date: Fri, 17 Apr 2009 00:08:47 +0200 Subject: Changing jdk order In-Reply-To: <49E79C98.1020308@gmx.net> References: <49E79C98.1020308@gmx.net> Message-ID: <49E7AC6F.1030207@ubuntu.com> please use an Ubuntu developers forum or the Ubuntu mailing lists for such questions. LCID Fire schrieb: > On my Ubuntu system I have both the openjdk and sunjdk installed. > Trying to generate jni header files in ant screws up totally when > openjdk is installed (see > http://bugs.launchpad.net/ubuntu/+source/openjdk-6/+bug/361578). Problem > is to compile the java stuff I need some classes that come with openjdk. > Currently it seems like the system uses openjdk before sunjdk - is there > a way to switch/change the orders of the jdks? see update-java-alternatives(8), or make sure to use the correct JAVA_HOME. Matthias From Joe.Darcy at Sun.COM Thu Apr 16 21:20:40 2009 From: Joe.Darcy at Sun.COM (Joseph D. Darcy) Date: Thu, 16 Apr 2009 21:20:40 -0700 Subject: New project: getting rid of IcedTea local patches In-Reply-To: <17c6771e0904091055w7c14a9adg982e0ff37a55e0a4@mail.gmail.com> References: <49D3917B.2010907@redhat.com> <49DDF4C6.1070602@redhat.com> <49DE0442.6060802@sun.com> <17c6771e0904091055w7c14a9adg982e0ff37a55e0a4@mail.gmail.com> Message-ID: <49E80398.6090601@sun.com> Catching up on email, Andrew John Hughes wrote: > 2009/4/9 Alan Bateman : > >> Andrew Haley wrote: >> [snip] >> I'm not very familiar with IcedTea but I cloned the repository to see what >> these patches are about. Here are a few comments: >> >> icedtea-6728542-epoll.patch seems to be a change-set I pushed to jdk7 last >> year. It seems very reasonable to push this into jdk6/jdk6/jdk. Go for it! >> >> > > Yes, I've referenced the bug on the wiki now. Joe, can we commit this to 6? > Yes! -Joe From jonathan.gibbons at sun.com Tue Apr 21 15:05:53 2009 From: jonathan.gibbons at sun.com (jonathan.gibbons at sun.com) Date: Tue, 21 Apr 2009 22:05:53 +0000 Subject: hg: jdk6/jdk6/langtools: 6759810: bad regression test causes source file to be deleted Message-ID: <20090421220556.0670FE45A@hg.openjdk.java.net> Changeset: 1fe711e6bce6 Author: jjg Date: 2009-04-21 11:53 -0700 URL: http://hg.openjdk.java.net/jdk6/jdk6/langtools/rev/1fe711e6bce6 6759810: bad regression test causes source file to be deleted Reviewed-by: darcy ! test/tools/javac/links/T.java ! test/tools/javac/links/links.sh From langel at redhat.com Wed Apr 22 08:42:59 2009 From: langel at redhat.com (Lillian Angel) Date: Wed, 22 Apr 2009 11:42:59 -0400 Subject: OpenJDK 6 build 16 source bundle posted In-Reply-To: <49E69540.80801@sun.com> References: <49E69540.80801@sun.com> Message-ID: <49EF3B03.1030705@redhat.com> Hi, Joseph D. Darcy wrote: > Without further adieu, the source bundle for OpenJDK 6 b16 is > available from > > http://download.java.net/openjdk/jdk6/ > > and the OpenJDK 6 Mercurial repositories > (http://hg.openjdk.java.net/jdk6/jdk6) will be tagged with the build > information accordingly. > > The previous source bundle was for b14. There was no distinct b15 per > se; b15 was used to mark the transition from teamware to Mercurial. > The b16 build is the first to be done purely in Mercurial. > > There were 45 bug fixes in this build ranging from security fixes to > sound updates; OpenJDK 6 b16 has all relevant security fixes from the > recent JDK synchronized security release. A few fixes of note, during > a build importing binary plugs now defaults to false (6781572 "Default > to IMPORT_BINARY_PLUGS=false for OpenJDK 6 builds") and several fixes > were applied that should render a few IcedTea patches unnecessary: > > 6736248 EnumEditor bug. Class check incorrect > 6733718 test > /java/awt/FullScreen/UninitializedDisplayModeChangeTest/ fails > 6593946 (bf) X-Buffer.compact() does not discard mark as specified > 6778493 Fix (langtools) ant build to honor fcs MILESTONE setting > > The other non security bugs fixed in this build are: > > 6761791: Crash in the FontManager code due to use of JNIEnv saved by > another thread > 6512707: "incompatible types" after (unrelated) annotation processing > 6632696: Writing to closed output files (writeBytes) leaks native > memory (unix) > 6788196: (porting) Bounds checks in io_util.c rely on undefined behaviour > 6791458: FileInputStream/RandomAccessFile.read leaks memory if invoked > on closed stream with len > 8k > 6792066: src/share/native/java/io/io_util.c clean-ups > 6819886: System.getProperty("os.name") reports Vista on Windows 7 > 6821031: Upgrade OpenJDK's LittleCMS version to 1.18 Actually, the 1.18 version was not included in this release. b16 still includes 1.16. Any thoughts? Cheers, Lillian > 6800572: Removing elements from views of NavigableMap implementations > does not always work correctly. > 6801020: Concurrent Semaphore release may cause some require thread > not signaled > 6806019: 38 JCK api/javax_sound/midi/ tests fails starting from jdk7 b46 > 6803402: Race condition in AbstractQueuedSynchronizer > 6793757: Fix formatting of copyright notices in Gervill > 6794201: remove unused sources > 6808724: UninitializedDisplayModeChangeTest/DisplayModeChanger.java > has wrong legal notice > 6821030: Merge OpenJDK Gervill with upstream sources, Q1CY2009 > 6823445: Gervill SoftChannel/ResetAllControllers jtreg test fails > after portamento fix from last merge > 6823446: Gervill SoftLowFrequencyOscillator fails when freq is set to > 0 cent or 8.1758 Hz. > 6824976: Fix NAWK assignment in shell script > jdk/make/java/java/genlocales.gmk > 6828183: testcase from SSR09_01 into jdk6-open hangs > > -Joe > From Joe.Darcy at Sun.COM Wed Apr 22 08:47:43 2009 From: Joe.Darcy at Sun.COM (Joseph D. Darcy) Date: Wed, 22 Apr 2009 08:47:43 -0700 Subject: OpenJDK 6 build 16 source bundle posted In-Reply-To: <49EF3B03.1030705@redhat.com> References: <49E69540.80801@sun.com> <49EF3B03.1030705@redhat.com> Message-ID: <49EF3C1F.4020002@sun.com> Lillian Angel wrote: > Hi, > > > Joseph D. Darcy wrote: >> Without further adieu, the source bundle for OpenJDK 6 b16 is >> available from >> >> http://download.java.net/openjdk/jdk6/ >> >> and the OpenJDK 6 Mercurial repositories >> (http://hg.openjdk.java.net/jdk6/jdk6) will be tagged with the build >> information accordingly. >> >> The previous source bundle was for b14. There was no distinct b15 >> per se; b15 was used to mark the transition from teamware to >> Mercurial. The b16 build is the first to be done purely in Mercurial. >> >> There were 45 bug fixes in this build ranging from security fixes to >> sound updates; OpenJDK 6 b16 has all relevant security fixes from the >> recent JDK synchronized security release. A few fixes of note, >> during a build importing binary plugs now defaults to false (6781572 >> "Default to IMPORT_BINARY_PLUGS=false for OpenJDK 6 builds") and >> several fixes were applied that should render a few IcedTea patches >> unnecessary: >> >> 6736248 EnumEditor bug. Class check incorrect >> 6733718 test >> /java/awt/FullScreen/UninitializedDisplayModeChangeTest/ fails >> 6593946 (bf) X-Buffer.compact() does not discard mark as specified >> 6778493 Fix (langtools) ant build to honor fcs MILESTONE setting >> >> The other non security bugs fixed in this build are: >> >> 6761791: Crash in the FontManager code due to use of JNIEnv saved by >> another thread >> 6512707: "incompatible types" after (unrelated) annotation processing >> 6632696: Writing to closed output files (writeBytes) leaks native >> memory (unix) >> 6788196: (porting) Bounds checks in io_util.c rely on undefined >> behaviour >> 6791458: FileInputStream/RandomAccessFile.read leaks memory if >> invoked on closed stream with len > 8k >> 6792066: src/share/native/java/io/io_util.c clean-ups >> 6819886: System.getProperty("os.name") reports Vista on Windows 7 >> 6821031: Upgrade OpenJDK's LittleCMS version to 1.18 > > Actually, the 1.18 version was not included in this release. b16 still > includes 1.16. Any thoughts? Phil? -Joe From Joe.Darcy at Sun.COM Wed Apr 22 08:54:10 2009 From: Joe.Darcy at Sun.COM (Joseph D. Darcy) Date: Wed, 22 Apr 2009 08:54:10 -0700 Subject: OpenJDK 6 build 16 source bundle posted In-Reply-To: <49EF3C1F.4020002@sun.com> References: <49E69540.80801@sun.com> <49EF3B03.1030705@redhat.com> <49EF3C1F.4020002@sun.com> Message-ID: <49EF3DA2.1080805@sun.com> Joseph D. Darcy wrote: > Lillian Angel wrote: >> Hi, >> >> >> Joseph D. Darcy wrote: >>> Without further adieu, the source bundle for OpenJDK 6 b16 is >>> available from >>> >>> http://download.java.net/openjdk/jdk6/ >>> >>> and the OpenJDK 6 Mercurial repositories >>> (http://hg.openjdk.java.net/jdk6/jdk6) will be tagged with the build >>> information accordingly. >>> >>> The previous source bundle was for b14. There was no distinct b15 >>> per se; b15 was used to mark the transition from teamware to >>> Mercurial. The b16 build is the first to be done purely in Mercurial. >>> >>> There were 45 bug fixes in this build ranging from security fixes to >>> sound updates; OpenJDK 6 b16 has all relevant security fixes from >>> the recent JDK synchronized security release. A few fixes of note, >>> during a build importing binary plugs now defaults to false (6781572 >>> "Default to IMPORT_BINARY_PLUGS=false for OpenJDK 6 builds") and >>> several fixes were applied that should render a few IcedTea patches >>> unnecessary: >>> >>> 6736248 EnumEditor bug. Class check incorrect >>> 6733718 test >>> /java/awt/FullScreen/UninitializedDisplayModeChangeTest/ fails >>> 6593946 (bf) X-Buffer.compact() does not discard mark as specified >>> 6778493 Fix (langtools) ant build to honor fcs MILESTONE setting >>> >>> The other non security bugs fixed in this build are: >>> >>> 6761791: Crash in the FontManager code due to use of JNIEnv saved by >>> another thread >>> 6512707: "incompatible types" after (unrelated) annotation processing >>> 6632696: Writing to closed output files (writeBytes) leaks native >>> memory (unix) >>> 6788196: (porting) Bounds checks in io_util.c rely on undefined >>> behaviour >>> 6791458: FileInputStream/RandomAccessFile.read leaks memory if >>> invoked on closed stream with len > 8k >>> 6792066: src/share/native/java/io/io_util.c clean-ups >>> 6819886: System.getProperty("os.name") reports Vista on Windows 7 >>> 6821031: Upgrade OpenJDK's LittleCMS version to 1.18 >> >> Actually, the 1.18 version was not included in this release. b16 >> still includes 1.16. Any thoughts? > > Phil? > > -Joe PS b16 does include this changeset, does it not? http://hg.openjdk.java.net/jdk6/jdk6/jdk/rev/2bdeec6199b7 -Joe From gnu_andrew at member.fsf.org Wed Apr 22 08:55:55 2009 From: gnu_andrew at member.fsf.org (Andrew John Hughes) Date: Wed, 22 Apr 2009 16:55:55 +0100 Subject: OpenJDK 6 build 16 source bundle posted In-Reply-To: <49EF3DA2.1080805@sun.com> References: <49E69540.80801@sun.com> <49EF3B03.1030705@redhat.com> <49EF3C1F.4020002@sun.com> <49EF3DA2.1080805@sun.com> Message-ID: <17c6771e0904220855o5890a5ek22abf568d8231838@mail.gmail.com> 2009/4/22 Joseph D. Darcy : > Joseph D. Darcy wrote: >> >> Lillian Angel wrote: >>> >>> Hi, >>> >>> >>> Joseph D. Darcy wrote: >>>> >>>> Without further adieu, the source bundle for OpenJDK 6 b16 is available >>>> from >>>> >>>> ? http://download.java.net/openjdk/jdk6/ >>>> >>>> and the OpenJDK 6 Mercurial repositories >>>> (http://hg.openjdk.java.net/jdk6/jdk6) will be tagged with the build >>>> information accordingly. >>>> >>>> The previous source bundle was for b14. ?There was no distinct b15 per >>>> se; b15 was used to mark the transition from teamware to Mercurial. ?The b16 >>>> build is the first to be done purely in Mercurial. >>>> >>>> There were 45 bug fixes in this build ranging from security fixes to >>>> sound updates; OpenJDK 6 b16 has all relevant security fixes from the recent >>>> JDK synchronized security release. ?A few fixes of note, during a build >>>> importing binary plugs now defaults to false (6781572 "Default to >>>> IMPORT_BINARY_PLUGS=false for OpenJDK 6 builds") and several fixes were >>>> applied that should render a few IcedTea patches unnecessary: >>>> >>>> ?6736248 EnumEditor bug. Class check incorrect >>>> ?6733718 test /java/awt/FullScreen/UninitializedDisplayModeChangeTest/ >>>> fails >>>> ?6593946 (bf) X-Buffer.compact() does not discard mark as specified >>>> ?6778493 Fix (langtools) ant build to honor fcs MILESTONE setting >>>> >>>> The other non security bugs fixed in this build are: >>>> >>>> 6761791: Crash in the FontManager code due to use of JNIEnv saved by >>>> another thread >>>> 6512707: "incompatible types" after (unrelated) annotation processing >>>> 6632696: Writing to closed output files (writeBytes) leaks native memory >>>> (unix) >>>> 6788196: (porting) Bounds checks in io_util.c rely on undefined >>>> behaviour >>>> 6791458: FileInputStream/RandomAccessFile.read leaks memory if invoked >>>> on closed stream with len > 8k >>>> 6792066: src/share/native/java/io/io_util.c clean-ups >>>> 6819886: System.getProperty("os.name") reports Vista on Windows 7 >>>> 6821031: Upgrade OpenJDK's LittleCMS version to 1.18 >>> >>> Actually, the 1.18 version was not included in this release. b16 still >>> includes 1.16. Any thoughts? >> >> Phil? >> >> -Joe > > PS b16 does include this changeset, does it not? > > http://hg.openjdk.java.net/jdk6/jdk6/jdk/rev/2bdeec6199b7 > > -Joe > There doesn't seem to be a b16 hg tag either, looking at that URL. -- Andrew :-) Free Java Software Engineer Red Hat, Inc. (http://www.redhat.com) Support Free Java! Contribute to GNU Classpath and the OpenJDK http://www.gnu.org/software/classpath http://openjdk.java.net PGP Key: 94EFD9D8 (http://subkeys.pgp.net) Fingerprint: F8EF F1EA 401E 2E60 15FA 7927 142C 2591 94EF D9D8 From langel at redhat.com Wed Apr 22 08:56:16 2009 From: langel at redhat.com (Lillian Angel) Date: Wed, 22 Apr 2009 11:56:16 -0400 Subject: OpenJDK 6 build 16 source bundle posted In-Reply-To: <49EF3DA2.1080805@sun.com> References: <49E69540.80801@sun.com> <49EF3B03.1030705@redhat.com> <49EF3C1F.4020002@sun.com> <49EF3DA2.1080805@sun.com> Message-ID: <49EF3E20.6040401@redhat.com> Joseph D. Darcy wrote: > Joseph D. Darcy wrote: >> Lillian Angel wrote: >>> Hi, >>> >>> >>> Joseph D. Darcy wrote: >>>> Without further adieu, the source bundle for OpenJDK 6 b16 is >>>> available from >>>> >>>> http://download.java.net/openjdk/jdk6/ >>>> >>>> and the OpenJDK 6 Mercurial repositories >>>> (http://hg.openjdk.java.net/jdk6/jdk6) will be tagged with the >>>> build information accordingly. >>>> >>>> The previous source bundle was for b14. There was no distinct b15 >>>> per se; b15 was used to mark the transition from teamware to >>>> Mercurial. The b16 build is the first to be done purely in Mercurial. >>>> >>>> There were 45 bug fixes in this build ranging from security fixes >>>> to sound updates; OpenJDK 6 b16 has all relevant security fixes >>>> from the recent JDK synchronized security release. A few fixes of >>>> note, during a build importing binary plugs now defaults to false >>>> (6781572 "Default to IMPORT_BINARY_PLUGS=false for OpenJDK 6 >>>> builds") and several fixes were applied that should render a few >>>> IcedTea patches unnecessary: >>>> >>>> 6736248 EnumEditor bug. Class check incorrect >>>> 6733718 test >>>> /java/awt/FullScreen/UninitializedDisplayModeChangeTest/ fails >>>> 6593946 (bf) X-Buffer.compact() does not discard mark as specified >>>> 6778493 Fix (langtools) ant build to honor fcs MILESTONE setting >>>> >>>> The other non security bugs fixed in this build are: >>>> >>>> 6761791: Crash in the FontManager code due to use of JNIEnv saved >>>> by another thread >>>> 6512707: "incompatible types" after (unrelated) annotation processing >>>> 6632696: Writing to closed output files (writeBytes) leaks native >>>> memory (unix) >>>> 6788196: (porting) Bounds checks in io_util.c rely on undefined >>>> behaviour >>>> 6791458: FileInputStream/RandomAccessFile.read leaks memory if >>>> invoked on closed stream with len > 8k >>>> 6792066: src/share/native/java/io/io_util.c clean-ups >>>> 6819886: System.getProperty("os.name") reports Vista on Windows 7 >>>> 6821031: Upgrade OpenJDK's LittleCMS version to 1.18 >>> >>> Actually, the 1.18 version was not included in this release. b16 >>> still includes 1.16. Any thoughts? >> >> Phil? >> >> -Joe > > PS b16 does include this changeset, does it not? > > http://hg.openjdk.java.net/jdk6/jdk6/jdk/rev/2bdeec6199b7 Apparently not. This change shows that has clearly been updated to 1.18 (see version in comment), but it does not appear in the b16 sources (still says 1.16): http://hg.openjdk.java.net/jdk6/jdk6/jdk/diff/2bdeec6199b7/src/share/native/sun/java2d/cmm/lcms/lcms.h Lillian From Phil.Race at Sun.COM Wed Apr 22 09:15:08 2009 From: Phil.Race at Sun.COM (Phil Race) Date: Wed, 22 Apr 2009 09:15:08 -0700 Subject: OpenJDK 6 build 16 source bundle posted In-Reply-To: <49EF3E20.6040401@redhat.com> References: <49E69540.80801@sun.com> <49EF3B03.1030705@redhat.com> <49EF3C1F.4020002@sun.com> <49EF3DA2.1080805@sun.com> <49EF3E20.6040401@redhat.com> Message-ID: <49EF428C.8040904@sun.com> All I know is that I pushed the change 4 whole weeks ago. How the b16 source bundle was created is not something into which I have any insight. -phil. Lillian Angel wrote: > Joseph D. Darcy wrote: >> Joseph D. Darcy wrote: >>> Lillian Angel wrote: >>>> Hi, >>>> >>>> >>>> Joseph D. Darcy wrote: >>>>> Without further adieu, the source bundle for OpenJDK 6 b16 is >>>>> available from >>>>> >>>>> http://download.java.net/openjdk/jdk6/ >>>>> >>>>> and the OpenJDK 6 Mercurial repositories >>>>> (http://hg.openjdk.java.net/jdk6/jdk6) will be tagged with the >>>>> build information accordingly. >>>>> >>>>> The previous source bundle was for b14. There was no distinct b15 >>>>> per se; b15 was used to mark the transition from teamware to >>>>> Mercurial. The b16 build is the first to be done purely in Mercurial. >>>>> >>>>> There were 45 bug fixes in this build ranging from security fixes >>>>> to sound updates; OpenJDK 6 b16 has all relevant security fixes >>>>> from the recent JDK synchronized security release. A few fixes of >>>>> note, during a build importing binary plugs now defaults to false >>>>> (6781572 "Default to IMPORT_BINARY_PLUGS=false for OpenJDK 6 >>>>> builds") and several fixes were applied that should render a few >>>>> IcedTea patches unnecessary: >>>>> >>>>> 6736248 EnumEditor bug. Class check incorrect >>>>> 6733718 test >>>>> /java/awt/FullScreen/UninitializedDisplayModeChangeTest/ fails >>>>> 6593946 (bf) X-Buffer.compact() does not discard mark as specified >>>>> 6778493 Fix (langtools) ant build to honor fcs MILESTONE setting >>>>> >>>>> The other non security bugs fixed in this build are: >>>>> >>>>> 6761791: Crash in the FontManager code due to use of JNIEnv saved >>>>> by another thread >>>>> 6512707: "incompatible types" after (unrelated) annotation processing >>>>> 6632696: Writing to closed output files (writeBytes) leaks native >>>>> memory (unix) >>>>> 6788196: (porting) Bounds checks in io_util.c rely on undefined >>>>> behaviour >>>>> 6791458: FileInputStream/RandomAccessFile.read leaks memory if >>>>> invoked on closed stream with len > 8k >>>>> 6792066: src/share/native/java/io/io_util.c clean-ups >>>>> 6819886: System.getProperty("os.name") reports Vista on Windows 7 >>>>> 6821031: Upgrade OpenJDK's LittleCMS version to 1.18 >>>> >>>> Actually, the 1.18 version was not included in this release. b16 >>>> still includes 1.16. Any thoughts? >>> >>> Phil? >>> >>> -Joe >> >> PS b16 does include this changeset, does it not? >> >> http://hg.openjdk.java.net/jdk6/jdk6/jdk/rev/2bdeec6199b7 > > Apparently not. This change shows that has clearly been updated to 1.18 > (see version in comment), but it does not appear in the b16 sources > (still says 1.16): > http://hg.openjdk.java.net/jdk6/jdk6/jdk/diff/2bdeec6199b7/src/share/native/sun/java2d/cmm/lcms/lcms.h > > > Lillian From Joe.Darcy at Sun.COM Wed Apr 22 09:37:51 2009 From: Joe.Darcy at Sun.COM (Joseph D. Darcy) Date: Wed, 22 Apr 2009 09:37:51 -0700 Subject: OpenJDK 6 build 16 source bundle posted In-Reply-To: <49EF428C.8040904@sun.com> References: <49E69540.80801@sun.com> <49EF3B03.1030705@redhat.com> <49EF3C1F.4020002@sun.com> <49EF3DA2.1080805@sun.com> <49EF3E20.6040401@redhat.com> <49EF428C.8040904@sun.com> Message-ID: <49EF47DF.5030408@sun.com> I'll check with our RE folks to see what is going on... -Joe Phil Race wrote: > All I know is that I pushed the change 4 whole weeks ago. > How the b16 source bundle was created is not something into which > I have any insight. > > -phil. > > Lillian Angel wrote: >> Joseph D. Darcy wrote: >>> Joseph D. Darcy wrote: >>>> Lillian Angel wrote: >>>>> Hi, >>>>> >>>>> >>>>> Joseph D. Darcy wrote: >>>>>> Without further adieu, the source bundle for OpenJDK 6 b16 is >>>>>> available from >>>>>> >>>>>> http://download.java.net/openjdk/jdk6/ >>>>>> >>>>>> and the OpenJDK 6 Mercurial repositories >>>>>> (http://hg.openjdk.java.net/jdk6/jdk6) will be tagged with the >>>>>> build information accordingly. >>>>>> >>>>>> The previous source bundle was for b14. There was no distinct >>>>>> b15 per se; b15 was used to mark the transition from teamware to >>>>>> Mercurial. The b16 build is the first to be done purely in >>>>>> Mercurial. >>>>>> >>>>>> There were 45 bug fixes in this build ranging from security fixes >>>>>> to sound updates; OpenJDK 6 b16 has all relevant security fixes >>>>>> from the recent JDK synchronized security release. A few fixes >>>>>> of note, during a build importing binary plugs now defaults to >>>>>> false (6781572 "Default to IMPORT_BINARY_PLUGS=false for OpenJDK >>>>>> 6 builds") and several fixes were applied that should render a >>>>>> few IcedTea patches unnecessary: >>>>>> >>>>>> 6736248 EnumEditor bug. Class check incorrect >>>>>> 6733718 test >>>>>> /java/awt/FullScreen/UninitializedDisplayModeChangeTest/ fails >>>>>> 6593946 (bf) X-Buffer.compact() does not discard mark as specified >>>>>> 6778493 Fix (langtools) ant build to honor fcs MILESTONE setting >>>>>> >>>>>> The other non security bugs fixed in this build are: >>>>>> >>>>>> 6761791: Crash in the FontManager code due to use of JNIEnv saved >>>>>> by another thread >>>>>> 6512707: "incompatible types" after (unrelated) annotation >>>>>> processing >>>>>> 6632696: Writing to closed output files (writeBytes) leaks native >>>>>> memory (unix) >>>>>> 6788196: (porting) Bounds checks in io_util.c rely on undefined >>>>>> behaviour >>>>>> 6791458: FileInputStream/RandomAccessFile.read leaks memory if >>>>>> invoked on closed stream with len > 8k >>>>>> 6792066: src/share/native/java/io/io_util.c clean-ups >>>>>> 6819886: System.getProperty("os.name") reports Vista on Windows 7 >>>>>> 6821031: Upgrade OpenJDK's LittleCMS version to 1.18 >>>>> >>>>> Actually, the 1.18 version was not included in this release. b16 >>>>> still includes 1.16. Any thoughts? >>>> >>>> Phil? >>>> >>>> -Joe >>> >>> PS b16 does include this changeset, does it not? >>> >>> http://hg.openjdk.java.net/jdk6/jdk6/jdk/rev/2bdeec6199b7 >> >> Apparently not. This change shows that has clearly been updated to >> 1.18 (see version in comment), but it does not appear in the b16 >> sources (still says 1.16): >> http://hg.openjdk.java.net/jdk6/jdk6/jdk/diff/2bdeec6199b7/src/share/native/sun/java2d/cmm/lcms/lcms.h >> >> >> Lillian From martinrb at google.com Wed Apr 22 12:00:33 2009 From: martinrb at google.com (Martin Buchholz) Date: Wed, 22 Apr 2009 12:00:33 -0700 Subject: OpenJDK 6 build 16 source bundle posted In-Reply-To: <49EF47DF.5030408@sun.com> References: <49E69540.80801@sun.com> <49EF3B03.1030705@redhat.com> <49EF3C1F.4020002@sun.com> <49EF3DA2.1080805@sun.com> <49EF3E20.6040401@redhat.com> <49EF428C.8040904@sun.com> <49EF47DF.5030408@sun.com> Message-ID: <1ccfd1c10904221200j47a927a3w65b3e821af082755@mail.gmail.com> I've been doing some more work on openjdk6 and there are some more inconsistencies. The top-level repo has a Makefile with history, but the old tarball openjdk-6-src-b11-10_jul_2008.tar.gz doesn't seem to contain this file. Which is authoritative, the old tarball or hg? An audit of jdk6 tarballs and the mercurial repo tags seems like a good idea. A command like diff -r --brief --exclude=.hg might be your friend. $ cd ~/ws/openjdk6; hg update -r jdk6-b11; hg log Makefile zsh: exit 1 0 files updated, 0 files merged, 0 files removed, 0 files unresolved changeset: 8:51c62af7b063 tag: jdk6-b04 user: ohair date: Fri Jan 30 16:24:38 2009 -0800 summary: 6755905: Changes for openjdk6 build 04 changeset: 6:f55431cdca29 tag: jdk6-b03 user: ohair date: Fri Jan 30 16:20:33 2009 -0800 summary: 6755902: Changes for openjdk6 build 03 changeset: 2:79e050893ef4 tag: jdk6-b00 user: ohair date: Fri Jan 30 16:03:02 2009 -0800 summary: 6755277: All initial changes to jdk7 to create openjdk 6 build 0 changeset: 0:d5ea358b429b user: duke date: Fri Jan 30 15:58:47 2009 -0800 summary: Initial load is jdk7-b23 On Wed, Apr 22, 2009 at 09:37, Joseph D. Darcy wrote: > I'll check with our RE folks to see what is going on... > > -Joe > > Phil Race wrote: >> >> All I know is that I pushed the change 4 whole weeks ago. >> How the b16 source bundle was created is not something into which >> I have any insight. >> >> -phil. >> >> Lillian Angel wrote: >>> >>> Joseph D. Darcy wrote: >>>> >>>> Joseph D. Darcy wrote: >>>>> >>>>> Lillian Angel wrote: >>>>>> >>>>>> Hi, >>>>>> >>>>>> >>>>>> Joseph D. Darcy wrote: >>>>>>> >>>>>>> Without further adieu, the source bundle for OpenJDK 6 b16 is >>>>>>> available from >>>>>>> >>>>>>> ? http://download.java.net/openjdk/jdk6/ >>>>>>> >>>>>>> and the OpenJDK 6 Mercurial repositories >>>>>>> (http://hg.openjdk.java.net/jdk6/jdk6) will be tagged with the build >>>>>>> information accordingly. >>>>>>> >>>>>>> The previous source bundle was for b14. ?There was no distinct b15 >>>>>>> per se; b15 was used to mark the transition from teamware to Mercurial. ?The >>>>>>> b16 build is the first to be done purely in Mercurial. >>>>>>> >>>>>>> There were 45 bug fixes in this build ranging from security fixes to >>>>>>> sound updates; OpenJDK 6 b16 has all relevant security fixes from the recent >>>>>>> JDK synchronized security release. ?A few fixes of note, during a build >>>>>>> importing binary plugs now defaults to false (6781572 "Default to >>>>>>> IMPORT_BINARY_PLUGS=false for OpenJDK 6 builds") and several fixes were >>>>>>> applied that should render a few IcedTea patches unnecessary: >>>>>>> >>>>>>> ?6736248 EnumEditor bug. Class check incorrect >>>>>>> ?6733718 test >>>>>>> /java/awt/FullScreen/UninitializedDisplayModeChangeTest/ fails >>>>>>> ?6593946 (bf) X-Buffer.compact() does not discard mark as specified >>>>>>> ?6778493 Fix (langtools) ant build to honor fcs MILESTONE setting >>>>>>> >>>>>>> The other non security bugs fixed in this build are: >>>>>>> >>>>>>> 6761791: Crash in the FontManager code due to use of JNIEnv saved by >>>>>>> another thread >>>>>>> 6512707: "incompatible types" after (unrelated) annotation processing >>>>>>> 6632696: Writing to closed output files (writeBytes) leaks native >>>>>>> memory (unix) >>>>>>> 6788196: (porting) Bounds checks in io_util.c rely on undefined >>>>>>> behaviour >>>>>>> 6791458: FileInputStream/RandomAccessFile.read leaks memory if >>>>>>> invoked on closed stream with len > 8k >>>>>>> 6792066: src/share/native/java/io/io_util.c clean-ups >>>>>>> 6819886: System.getProperty("os.name") reports Vista on Windows 7 >>>>>>> 6821031: Upgrade OpenJDK's LittleCMS version to 1.18 >>>>>> >>>>>> Actually, the 1.18 version was not included in this release. b16 still >>>>>> includes 1.16. Any thoughts? >>>>> >>>>> Phil? >>>>> >>>>> -Joe >>>> >>>> PS b16 does include this changeset, does it not? >>>> >>>> http://hg.openjdk.java.net/jdk6/jdk6/jdk/rev/2bdeec6199b7 >>> >>> Apparently not. This change shows that has clearly been updated to 1.18 >>> (see version in comment), but it does not appear in the b16 sources (still >>> says 1.16): >>> >>> http://hg.openjdk.java.net/jdk6/jdk6/jdk/diff/2bdeec6199b7/src/share/native/sun/java2d/cmm/lcms/lcms.h >>> >>> Lillian > > From Joe.Darcy at Sun.COM Wed Apr 22 14:11:13 2009 From: Joe.Darcy at Sun.COM (Joe Darcy) Date: Wed, 22 Apr 2009 14:11:13 -0700 Subject: OpenJDK 6 build 16 source bundle posted In-Reply-To: <49EF47DF.5030408@sun.com> References: <49E69540.80801@sun.com> <49EF3B03.1030705@redhat.com> <49EF3C1F.4020002@sun.com> <49EF3DA2.1080805@sun.com> <49EF3E20.6040401@redhat.com> <49EF428C.8040904@sun.com> <49EF47DF.5030408@sun.com> Message-ID: <49EF87F1.2010704@sun.com> Hello. There was a misunderstanding with RE; the posted b16 sources were based on an internal staging repository being used to hold the security fixes before they were made public. Fixes pushed to the public repo after that point were not included. I've asked RE to redo a version of b16 to reflect the current state of the OpenJDK 6 repositories, including the little cms changes. The new source bundled and corresponding tags in the repository should be available within a few days. Sorry for the confusion, -Joe On 04/22/09 09:37 AM, Joseph D. Darcy wrote: > I'll check with our RE folks to see what is going on... > > -Joe > > Phil Race wrote: >> All I know is that I pushed the change 4 whole weeks ago. >> How the b16 source bundle was created is not something into which >> I have any insight. >> >> -phil. >> >> Lillian Angel wrote: >>> Joseph D. Darcy wrote: >>>> Joseph D. Darcy wrote: >>>>> Lillian Angel wrote: >>>>>> Hi, >>>>>> >>>>>> >>>>>> Joseph D. Darcy wrote: >>>>>>> Without further adieu, the source bundle for OpenJDK 6 b16 is >>>>>>> available from >>>>>>> >>>>>>> http://download.java.net/openjdk/jdk6/ >>>>>>> >>>>>>> and the OpenJDK 6 Mercurial repositories >>>>>>> (http://hg.openjdk.java.net/jdk6/jdk6) will be tagged with the >>>>>>> build information accordingly. >>>>>>> >>>>>>> The previous source bundle was for b14. There was no distinct >>>>>>> b15 per se; b15 was used to mark the transition from teamware to >>>>>>> Mercurial. The b16 build is the first to be done purely in >>>>>>> Mercurial. >>>>>>> >>>>>>> There were 45 bug fixes in this build ranging from security >>>>>>> fixes to sound updates; OpenJDK 6 b16 has all relevant security >>>>>>> fixes from the recent JDK synchronized security release. A few >>>>>>> fixes of note, during a build importing binary plugs now >>>>>>> defaults to false (6781572 "Default to IMPORT_BINARY_PLUGS=false >>>>>>> for OpenJDK 6 builds") and several fixes were applied that >>>>>>> should render a few IcedTea patches unnecessary: >>>>>>> >>>>>>> 6736248 EnumEditor bug. Class check incorrect >>>>>>> 6733718 test >>>>>>> /java/awt/FullScreen/UninitializedDisplayModeChangeTest/ fails >>>>>>> 6593946 (bf) X-Buffer.compact() does not discard mark as >>>>>>> specified >>>>>>> 6778493 Fix (langtools) ant build to honor fcs MILESTONE setting >>>>>>> >>>>>>> The other non security bugs fixed in this build are: >>>>>>> >>>>>>> 6761791: Crash in the FontManager code due to use of JNIEnv >>>>>>> saved by another thread >>>>>>> 6512707: "incompatible types" after (unrelated) annotation >>>>>>> processing >>>>>>> 6632696: Writing to closed output files (writeBytes) leaks >>>>>>> native memory (unix) >>>>>>> 6788196: (porting) Bounds checks in io_util.c rely on undefined >>>>>>> behaviour >>>>>>> 6791458: FileInputStream/RandomAccessFile.read leaks memory if >>>>>>> invoked on closed stream with len > 8k >>>>>>> 6792066: src/share/native/java/io/io_util.c clean-ups >>>>>>> 6819886: System.getProperty("os.name") reports Vista on Windows 7 >>>>>>> 6821031: Upgrade OpenJDK's LittleCMS version to 1.18 >>>>>> >>>>>> Actually, the 1.18 version was not included in this release. b16 >>>>>> still includes 1.16. Any thoughts? >>>>> >>>>> Phil? >>>>> >>>>> -Joe >>>> >>>> PS b16 does include this changeset, does it not? >>>> >>>> http://hg.openjdk.java.net/jdk6/jdk6/jdk/rev/2bdeec6199b7 >>> >>> Apparently not. This change shows that has clearly been updated to >>> 1.18 (see version in comment), but it does not appear in the b16 >>> sources (still says 1.16): >>> http://hg.openjdk.java.net/jdk6/jdk6/jdk/diff/2bdeec6199b7/src/share/native/sun/java2d/cmm/lcms/lcms.h >>> >>> >>> Lillian > From Joe.Darcy at Sun.COM Wed Apr 22 14:47:02 2009 From: Joe.Darcy at Sun.COM (Joseph D. Darcy) Date: Wed, 22 Apr 2009 14:47:02 -0700 Subject: OpenJDK 6 build 16 source bundle posted In-Reply-To: <1ccfd1c10904221200j47a927a3w65b3e821af082755@mail.gmail.com> References: <49E69540.80801@sun.com> <49EF3B03.1030705@redhat.com> <49EF3C1F.4020002@sun.com> <49EF3DA2.1080805@sun.com> <49EF3E20.6040401@redhat.com> <49EF428C.8040904@sun.com> <49EF47DF.5030408@sun.com> <1ccfd1c10904221200j47a927a3w65b3e821af082755@mail.gmail.com> Message-ID: <49EF9056.9020705@sun.com> Martin, The Mercurial transition included the folding of the control workspace into the jdk; I don't know if the differences you are seeing are from that part of the switch which did not preserve file layout. Kelly's scripts to retcon the OpenJDK 6 repositories with the teamware workspaces did include the sort of file level checks you recommend, at least for files under /src. (A few binary files eventually removed from the jaxws workspace were not included at all in that Mercurial repository.) -Joe Martin Buchholz wrote: > I've been doing some more work on openjdk6 and there are some more > inconsistencies. > The top-level repo has a Makefile with history, > but the old tarball openjdk-6-src-b11-10_jul_2008.tar.gz > doesn't seem to contain this file. > Which is authoritative, the old tarball or hg? > > An audit of jdk6 tarballs and the mercurial repo tags seems like a good idea. > A command like > diff -r --brief --exclude=.hg > might be your friend. > > $ cd ~/ws/openjdk6; hg update -r jdk6-b11; hg log Makefile > zsh: exit 1 > 0 files updated, 0 files merged, 0 files removed, 0 files unresolved > changeset: 8:51c62af7b063 > tag: jdk6-b04 > user: ohair > date: Fri Jan 30 16:24:38 2009 -0800 > summary: 6755905: Changes for openjdk6 build 04 > > changeset: 6:f55431cdca29 > tag: jdk6-b03 > user: ohair > date: Fri Jan 30 16:20:33 2009 -0800 > summary: 6755902: Changes for openjdk6 build 03 > > changeset: 2:79e050893ef4 > tag: jdk6-b00 > user: ohair > date: Fri Jan 30 16:03:02 2009 -0800 > summary: 6755277: All initial changes to jdk7 to create openjdk 6 build 0 > > changeset: 0:d5ea358b429b > user: duke > date: Fri Jan 30 15:58:47 2009 -0800 > summary: Initial load is jdk7-b23 > > On Wed, Apr 22, 2009 at 09:37, Joseph D. Darcy wrote: > >> I'll check with our RE folks to see what is going on... >> >> -Joe >> >> Phil Race wrote: >> >>> All I know is that I pushed the change 4 whole weeks ago. >>> How the b16 source bundle was created is not something into which >>> I have any insight. >>> >>> -phil. >>> >>> Lillian Angel wrote: >>> >>>> Joseph D. Darcy wrote: >>>> >>>>> Joseph D. Darcy wrote: >>>>> >>>>>> Lillian Angel wrote: >>>>>> >>>>>>> Hi, >>>>>>> >>>>>>> >>>>>>> Joseph D. Darcy wrote: >>>>>>> >>>>>>>> Without further adieu, the source bundle for OpenJDK 6 b16 is >>>>>>>> available from >>>>>>>> >>>>>>>> http://download.java.net/openjdk/jdk6/ >>>>>>>> >>>>>>>> and the OpenJDK 6 Mercurial repositories >>>>>>>> (http://hg.openjdk.java.net/jdk6/jdk6) will be tagged with the build >>>>>>>> information accordingly. >>>>>>>> >>>>>>>> The previous source bundle was for b14. There was no distinct b15 >>>>>>>> per se; b15 was used to mark the transition from teamware to Mercurial. The >>>>>>>> b16 build is the first to be done purely in Mercurial. >>>>>>>> >>>>>>>> There were 45 bug fixes in this build ranging from security fixes to >>>>>>>> sound updates; OpenJDK 6 b16 has all relevant security fixes from the recent >>>>>>>> JDK synchronized security release. A few fixes of note, during a build >>>>>>>> importing binary plugs now defaults to false (6781572 "Default to >>>>>>>> IMPORT_BINARY_PLUGS=false for OpenJDK 6 builds") and several fixes were >>>>>>>> applied that should render a few IcedTea patches unnecessary: >>>>>>>> >>>>>>>> 6736248 EnumEditor bug. Class check incorrect >>>>>>>> 6733718 test >>>>>>>> /java/awt/FullScreen/UninitializedDisplayModeChangeTest/ fails >>>>>>>> 6593946 (bf) X-Buffer.compact() does not discard mark as specified >>>>>>>> 6778493 Fix (langtools) ant build to honor fcs MILESTONE setting >>>>>>>> >>>>>>>> The other non security bugs fixed in this build are: >>>>>>>> >>>>>>>> 6761791: Crash in the FontManager code due to use of JNIEnv saved by >>>>>>>> another thread >>>>>>>> 6512707: "incompatible types" after (unrelated) annotation processing >>>>>>>> 6632696: Writing to closed output files (writeBytes) leaks native >>>>>>>> memory (unix) >>>>>>>> 6788196: (porting) Bounds checks in io_util.c rely on undefined >>>>>>>> behaviour >>>>>>>> 6791458: FileInputStream/RandomAccessFile.read leaks memory if >>>>>>>> invoked on closed stream with len > 8k >>>>>>>> 6792066: src/share/native/java/io/io_util.c clean-ups >>>>>>>> 6819886: System.getProperty("os.name") reports Vista on Windows 7 >>>>>>>> 6821031: Upgrade OpenJDK's LittleCMS version to 1.18 >>>>>>>> >>>>>>> Actually, the 1.18 version was not included in this release. b16 still >>>>>>> includes 1.16. Any thoughts? >>>>>>> >>>>>> Phil? >>>>>> >>>>>> -Joe >>>>>> >>>>> PS b16 does include this changeset, does it not? >>>>> >>>>> http://hg.openjdk.java.net/jdk6/jdk6/jdk/rev/2bdeec6199b7 >>>>> >>>> Apparently not. This change shows that has clearly been updated to 1.18 >>>> (see version in comment), but it does not appear in the b16 sources (still >>>> says 1.16): >>>> >>>> http://hg.openjdk.java.net/jdk6/jdk6/jdk/diff/2bdeec6199b7/src/share/native/sun/java2d/cmm/lcms/lcms.h >>>> >>>> Lillian >>>> >> From martinrb at google.com Wed Apr 22 17:23:46 2009 From: martinrb at google.com (Martin Buchholz) Date: Wed, 22 Apr 2009 17:23:46 -0700 Subject: OpenJDK 6 build 16 source bundle posted In-Reply-To: <49EF9056.9020705@sun.com> References: <49E69540.80801@sun.com> <49EF3B03.1030705@redhat.com> <49EF3C1F.4020002@sun.com> <49EF3DA2.1080805@sun.com> <49EF3E20.6040401@redhat.com> <49EF428C.8040904@sun.com> <49EF47DF.5030408@sun.com> <1ccfd1c10904221200j47a927a3w65b3e821af082755@mail.gmail.com> <49EF9056.9020705@sun.com> Message-ID: <1ccfd1c10904221723q2bf34991l613a8fc97bee65d7@mail.gmail.com> Below is the output of a diff between mercurial fclone and the exploded old tarball They appear to differ in that: - top-level directory files are missing (no big deal unless maybe you're a lawyer) - it contains the control/make folding Actual file contents appear to be identical. Martin $ diff --brief -r . ~/ws/openjdk6/ Only in /home/martinrb/ws/openjdk6/: .hg Only in /home/martinrb/ws/openjdk6/: .hgignore Only in /home/martinrb/ws/openjdk6/: .hgtags Only in /home/martinrb/ws/openjdk6/: .jcheck Only in /home/martinrb/ws/openjdk6/: Makefile Only in /home/martinrb/ws/openjdk6/: README Only in .: README.html Only in .: control Only in /home/martinrb/ws/openjdk6/corba: .hg Only in /home/martinrb/ws/openjdk6/corba: .hgignore Only in /home/martinrb/ws/openjdk6/corba: .hgtags Only in /home/martinrb/ws/openjdk6/corba: .jcheck Only in /home/martinrb/ws/openjdk6/corba: ASSEMBLY_EXCEPTION Only in /home/martinrb/ws/openjdk6/corba: LICENSE Only in /home/martinrb/ws/openjdk6/corba: README Only in /home/martinrb/ws/openjdk6/corba: THIRD_PARTY_README Only in /home/martinrb/ws/openjdk6/corba: TRADEMARK Only in /home/martinrb/ws/openjdk6/hotspot: .hg Only in /home/martinrb/ws/openjdk6/hotspot: .hgignore Only in /home/martinrb/ws/openjdk6/hotspot: .hgtags Only in /home/martinrb/ws/openjdk6/hotspot: .jcheck Only in /home/martinrb/ws/openjdk6/hotspot: ASSEMBLY_EXCEPTION Only in /home/martinrb/ws/openjdk6/hotspot: LICENSE Only in /home/martinrb/ws/openjdk6/hotspot: README Only in /home/martinrb/ws/openjdk6/hotspot: THIRD_PARTY_README Only in /home/martinrb/ws/openjdk6/hotspot: TRADEMARK Only in /home/martinrb/ws/openjdk6/jaxp: .hg Only in /home/martinrb/ws/openjdk6/jaxp: .hgignore Only in /home/martinrb/ws/openjdk6/jaxp: .hgtags Only in /home/martinrb/ws/openjdk6/jaxp: .jcheck Only in /home/martinrb/ws/openjdk6/jaxp: ASSEMBLY_EXCEPTION Only in /home/martinrb/ws/openjdk6/jaxp: LICENSE Only in /home/martinrb/ws/openjdk6/jaxp: README Only in /home/martinrb/ws/openjdk6/jaxp: THIRD_PARTY_README Only in /home/martinrb/ws/openjdk6/jaxp: TRADEMARK Only in /home/martinrb/ws/openjdk6/jaxws: .hg Only in /home/martinrb/ws/openjdk6/jaxws: .hgignore Only in /home/martinrb/ws/openjdk6/jaxws: .hgtags Only in /home/martinrb/ws/openjdk6/jaxws: .jcheck Only in /home/martinrb/ws/openjdk6/jaxws: ASSEMBLY_EXCEPTION Only in /home/martinrb/ws/openjdk6/jaxws: LICENSE Only in /home/martinrb/ws/openjdk6/jaxws: README Only in /home/martinrb/ws/openjdk6/jaxws: THIRD_PARTY_README Only in /home/martinrb/ws/openjdk6/jaxws: TRADEMARK Only in /home/martinrb/ws/openjdk6/jdk: .hg Only in /home/martinrb/ws/openjdk6/jdk: .hgignore Only in /home/martinrb/ws/openjdk6/jdk: .hgtags Only in /home/martinrb/ws/openjdk6/jdk: .jcheck Only in /home/martinrb/ws/openjdk6/jdk: ASSEMBLY_EXCEPTION Only in /home/martinrb/ws/openjdk6/jdk: LICENSE Only in /home/martinrb/ws/openjdk6/jdk: README Only in /home/martinrb/ws/openjdk6/jdk: README-builds.html Only in /home/martinrb/ws/openjdk6/jdk: README.html Only in /home/martinrb/ws/openjdk6/jdk: THIRD_PARTY_README Only in /home/martinrb/ws/openjdk6/jdk: TRADEMARK Only in /home/martinrb/ws/openjdk6/langtools: .hg Only in /home/martinrb/ws/openjdk6/langtools: .hgignore Only in /home/martinrb/ws/openjdk6/langtools: .hgtags Only in /home/martinrb/ws/openjdk6/langtools: .jcheck Only in /home/martinrb/ws/openjdk6/langtools: ASSEMBLY_EXCEPTION Only in /home/martinrb/ws/openjdk6/langtools: LICENSE Only in /home/martinrb/ws/openjdk6/langtools: README Only in /home/martinrb/ws/openjdk6/langtools: THIRD_PARTY_README Only in /home/martinrb/ws/openjdk6/langtools: TRADEMARK Only in /home/martinrb/ws/openjdk6/: make On Wed, Apr 22, 2009 at 14:47, Joseph D. Darcy wrote: > Martin, > > The Mercurial transition included the folding of the control workspace into > the jdk; I don't know if the differences you are seeing are from that part > of the switch which did not preserve file layout. > > Kelly's scripts to retcon the OpenJDK 6 repositories with the teamware > workspaces did include the sort of file level checks you recommend, at least > for files under /src. ?(A few binary files eventually removed from the jaxws > workspace were not included at all in that Mercurial repository.) > > -Joe > > > Martin Buchholz wrote: >> >> I've been doing some more work on openjdk6 and there are some more >> inconsistencies. >> The top-level repo has a Makefile with history, >> but the old tarball openjdk-6-src-b11-10_jul_2008.tar.gz >> doesn't seem to contain this file. >> Which is authoritative, the old tarball or hg? >> >> An audit of jdk6 tarballs and the mercurial repo tags seems like a good >> idea. >> A command like >> diff -r --brief --exclude=.hg >> might be your friend. >> >> $ cd ~/ws/openjdk6; hg update -r jdk6-b11; hg log ?Makefile >> zsh: exit 1 >> 0 files updated, 0 files merged, 0 files removed, 0 files unresolved >> changeset: ? 8:51c62af7b063 >> tag: ? ? ? ? jdk6-b04 >> user: ? ? ? ?ohair >> date: ? ? ? ?Fri Jan 30 16:24:38 2009 -0800 >> summary: ? ? 6755905: Changes for openjdk6 build 04 >> >> changeset: ? 6:f55431cdca29 >> tag: ? ? ? ? jdk6-b03 >> user: ? ? ? ?ohair >> date: ? ? ? ?Fri Jan 30 16:20:33 2009 -0800 >> summary: ? ? 6755902: Changes for openjdk6 build 03 >> >> changeset: ? 2:79e050893ef4 >> tag: ? ? ? ? jdk6-b00 >> user: ? ? ? ?ohair >> date: ? ? ? ?Fri Jan 30 16:03:02 2009 -0800 >> summary: ? ? 6755277: All initial changes to jdk7 to create openjdk 6 >> build 0 >> >> changeset: ? 0:d5ea358b429b >> user: ? ? ? ?duke >> date: ? ? ? ?Fri Jan 30 15:58:47 2009 -0800 >> summary: ? ? Initial load is jdk7-b23 >> >> On Wed, Apr 22, 2009 at 09:37, Joseph D. Darcy wrote: >> >>> >>> I'll check with our RE folks to see what is going on... >>> >>> -Joe >>> >>> Phil Race wrote: >>> >>>> >>>> All I know is that I pushed the change 4 whole weeks ago. >>>> How the b16 source bundle was created is not something into which >>>> I have any insight. >>>> >>>> -phil. >>>> >>>> Lillian Angel wrote: >>>> >>>>> >>>>> Joseph D. Darcy wrote: >>>>> >>>>>> >>>>>> Joseph D. Darcy wrote: >>>>>> >>>>>>> >>>>>>> Lillian Angel wrote: >>>>>>> >>>>>>>> >>>>>>>> Hi, >>>>>>>> >>>>>>>> >>>>>>>> Joseph D. Darcy wrote: >>>>>>>> >>>>>>>>> >>>>>>>>> Without further adieu, the source bundle for OpenJDK 6 b16 is >>>>>>>>> available from >>>>>>>>> >>>>>>>>> ?http://download.java.net/openjdk/jdk6/ >>>>>>>>> >>>>>>>>> and the OpenJDK 6 Mercurial repositories >>>>>>>>> (http://hg.openjdk.java.net/jdk6/jdk6) will be tagged with the >>>>>>>>> build >>>>>>>>> information accordingly. >>>>>>>>> >>>>>>>>> The previous source bundle was for b14. ?There was no distinct b15 >>>>>>>>> per se; b15 was used to mark the transition from teamware to >>>>>>>>> Mercurial. ?The >>>>>>>>> b16 build is the first to be done purely in Mercurial. >>>>>>>>> >>>>>>>>> There were 45 bug fixes in this build ranging from security fixes >>>>>>>>> to >>>>>>>>> sound updates; OpenJDK 6 b16 has all relevant security fixes from >>>>>>>>> the recent >>>>>>>>> JDK synchronized security release. ?A few fixes of note, during a >>>>>>>>> build >>>>>>>>> importing binary plugs now defaults to false (6781572 "Default to >>>>>>>>> IMPORT_BINARY_PLUGS=false for OpenJDK 6 builds") and several fixes >>>>>>>>> were >>>>>>>>> applied that should render a few IcedTea patches unnecessary: >>>>>>>>> >>>>>>>>> ?6736248 EnumEditor bug. Class check incorrect >>>>>>>>> ?6733718 test >>>>>>>>> /java/awt/FullScreen/UninitializedDisplayModeChangeTest/ fails >>>>>>>>> ?6593946 (bf) X-Buffer.compact() does not discard mark as specified >>>>>>>>> ?6778493 Fix (langtools) ant build to honor fcs MILESTONE setting >>>>>>>>> >>>>>>>>> The other non security bugs fixed in this build are: >>>>>>>>> >>>>>>>>> 6761791: Crash in the FontManager code due to use of JNIEnv saved >>>>>>>>> by >>>>>>>>> another thread >>>>>>>>> 6512707: "incompatible types" after (unrelated) annotation >>>>>>>>> processing >>>>>>>>> 6632696: Writing to closed output files (writeBytes) leaks native >>>>>>>>> memory (unix) >>>>>>>>> 6788196: (porting) Bounds checks in io_util.c rely on undefined >>>>>>>>> behaviour >>>>>>>>> 6791458: FileInputStream/RandomAccessFile.read leaks memory if >>>>>>>>> invoked on closed stream with len > 8k >>>>>>>>> 6792066: src/share/native/java/io/io_util.c clean-ups >>>>>>>>> 6819886: System.getProperty("os.name") reports Vista on Windows 7 >>>>>>>>> 6821031: Upgrade OpenJDK's LittleCMS version to 1.18 >>>>>>>>> >>>>>>>> >>>>>>>> Actually, the 1.18 version was not included in this release. b16 >>>>>>>> still >>>>>>>> includes 1.16. Any thoughts? >>>>>>>> >>>>>>> >>>>>>> Phil? >>>>>>> >>>>>>> -Joe >>>>>>> >>>>>> >>>>>> PS b16 does include this changeset, does it not? >>>>>> >>>>>> http://hg.openjdk.java.net/jdk6/jdk6/jdk/rev/2bdeec6199b7 >>>>>> >>>>> >>>>> Apparently not. This change shows that has clearly been updated to 1.18 >>>>> (see version in comment), but it does not appear in the b16 sources >>>>> (still >>>>> says 1.16): >>>>> >>>>> >>>>> http://hg.openjdk.java.net/jdk6/jdk6/jdk/diff/2bdeec6199b7/src/share/native/sun/java2d/cmm/lcms/lcms.h >>>>> >>>>> Lillian >>>>> >>> >>> > > From kalli at midverk.is Wed Apr 22 17:37:03 2009 From: kalli at midverk.is (Karl Helgason) Date: Thu, 23 Apr 2009 00:37:03 +0000 Subject: [Request for review] 6823449: Gervill: ArrayIndexOutOfBoundsException thrown when trying to play too may voices at same time. Message-ID: <36EC82E93EB0AD40A4301DAD65432386F5A60E21E8@mail.midverk.is> Hi, I need code review for the fix: http://cr.openjdk.java.net/~kalli/6823449/webrev.01/ And I need somebody to change the bug description: http://bugs.sun.com/bugdatabase/view_bug.do?bug_id=6823449 Also I need new bug ids for these bugs/improvements: [New Bug]: Improve time-stamp support in Gervill to reduce jitter. [New Bug]: Add support for sending time-stamp information in RealTimeSequencer to reduce jitter in playback. Those two bugs reduces playback glitches when playing back midi files like described here: https://gervill.dev.java.net/servlets/ProjectForumMessageView?forumID=2923&messageID=26785 And also add new bug id for this bug: [New Bug]: Gervill: Add support to limit maximum CPU load by dynamically limit maximum voice polyphony. regards, Karl From Kelly.Ohair at Sun.COM Wed Apr 22 17:44:39 2009 From: Kelly.Ohair at Sun.COM (Kelly O'Hair) Date: Wed, 22 Apr 2009 17:44:39 -0700 Subject: OpenJDK 6 build 16 source bundle posted In-Reply-To: <1ccfd1c10904221723q2bf34991l613a8fc97bee65d7@mail.gmail.com> References: <49E69540.80801@sun.com> <49EF3B03.1030705@redhat.com> <49EF3C1F.4020002@sun.com> <49EF3DA2.1080805@sun.com> <49EF3E20.6040401@redhat.com> <49EF428C.8040904@sun.com> <49EF47DF.5030408@sun.com> <1ccfd1c10904221200j47a927a3w65b3e821af082755@mail.gmail.com> <49EF9056.9020705@sun.com> <1ccfd1c10904221723q2bf34991l613a8fc97bee65d7@mail.gmail.com> Message-ID: <49EFB9F7.7080908@sun.com> That looks exactly right for the old tarball created from the old TeamWare workspaces (which had a control workspace). The conversion to Mercurial folded the control workspace files into the top or enclosing Mercurial repository. The control/make/Makefile moved to ./Makefile and the control/make directory moved to ./make. Each repository gets it's own legal files because they can be individually downloaded or cloned. The repositories should be treated as the ultimate source. README-builds.html should be at the root... humm, it's in ./jdk/ that may get moved. -kto Martin Buchholz wrote: > Below is the output of a diff between mercurial fclone > and the exploded old tarball > > They appear to differ in that: > - top-level directory files are missing > (no big deal unless maybe you're a lawyer) > - it contains the control/make folding > > Actual file contents appear to be identical. > > Martin > > $ diff --brief -r . ~/ws/openjdk6/ > Only in /home/martinrb/ws/openjdk6/: .hg > Only in /home/martinrb/ws/openjdk6/: .hgignore > Only in /home/martinrb/ws/openjdk6/: .hgtags > Only in /home/martinrb/ws/openjdk6/: .jcheck > Only in /home/martinrb/ws/openjdk6/: Makefile > Only in /home/martinrb/ws/openjdk6/: README > Only in .: README.html > Only in .: control > Only in /home/martinrb/ws/openjdk6/corba: .hg > Only in /home/martinrb/ws/openjdk6/corba: .hgignore > Only in /home/martinrb/ws/openjdk6/corba: .hgtags > Only in /home/martinrb/ws/openjdk6/corba: .jcheck > Only in /home/martinrb/ws/openjdk6/corba: ASSEMBLY_EXCEPTION > Only in /home/martinrb/ws/openjdk6/corba: LICENSE > Only in /home/martinrb/ws/openjdk6/corba: README > Only in /home/martinrb/ws/openjdk6/corba: THIRD_PARTY_README > Only in /home/martinrb/ws/openjdk6/corba: TRADEMARK > Only in /home/martinrb/ws/openjdk6/hotspot: .hg > Only in /home/martinrb/ws/openjdk6/hotspot: .hgignore > Only in /home/martinrb/ws/openjdk6/hotspot: .hgtags > Only in /home/martinrb/ws/openjdk6/hotspot: .jcheck > Only in /home/martinrb/ws/openjdk6/hotspot: ASSEMBLY_EXCEPTION > Only in /home/martinrb/ws/openjdk6/hotspot: LICENSE > Only in /home/martinrb/ws/openjdk6/hotspot: README > Only in /home/martinrb/ws/openjdk6/hotspot: THIRD_PARTY_README > Only in /home/martinrb/ws/openjdk6/hotspot: TRADEMARK > Only in /home/martinrb/ws/openjdk6/jaxp: .hg > Only in /home/martinrb/ws/openjdk6/jaxp: .hgignore > Only in /home/martinrb/ws/openjdk6/jaxp: .hgtags > Only in /home/martinrb/ws/openjdk6/jaxp: .jcheck > Only in /home/martinrb/ws/openjdk6/jaxp: ASSEMBLY_EXCEPTION > Only in /home/martinrb/ws/openjdk6/jaxp: LICENSE > Only in /home/martinrb/ws/openjdk6/jaxp: README > Only in /home/martinrb/ws/openjdk6/jaxp: THIRD_PARTY_README > Only in /home/martinrb/ws/openjdk6/jaxp: TRADEMARK > Only in /home/martinrb/ws/openjdk6/jaxws: .hg > Only in /home/martinrb/ws/openjdk6/jaxws: .hgignore > Only in /home/martinrb/ws/openjdk6/jaxws: .hgtags > Only in /home/martinrb/ws/openjdk6/jaxws: .jcheck > Only in /home/martinrb/ws/openjdk6/jaxws: ASSEMBLY_EXCEPTION > Only in /home/martinrb/ws/openjdk6/jaxws: LICENSE > Only in /home/martinrb/ws/openjdk6/jaxws: README > Only in /home/martinrb/ws/openjdk6/jaxws: THIRD_PARTY_README > Only in /home/martinrb/ws/openjdk6/jaxws: TRADEMARK > Only in /home/martinrb/ws/openjdk6/jdk: .hg > Only in /home/martinrb/ws/openjdk6/jdk: .hgignore > Only in /home/martinrb/ws/openjdk6/jdk: .hgtags > Only in /home/martinrb/ws/openjdk6/jdk: .jcheck > Only in /home/martinrb/ws/openjdk6/jdk: ASSEMBLY_EXCEPTION > Only in /home/martinrb/ws/openjdk6/jdk: LICENSE > Only in /home/martinrb/ws/openjdk6/jdk: README > Only in /home/martinrb/ws/openjdk6/jdk: README-builds.html > Only in /home/martinrb/ws/openjdk6/jdk: README.html > Only in /home/martinrb/ws/openjdk6/jdk: THIRD_PARTY_README > Only in /home/martinrb/ws/openjdk6/jdk: TRADEMARK > Only in /home/martinrb/ws/openjdk6/langtools: .hg > Only in /home/martinrb/ws/openjdk6/langtools: .hgignore > Only in /home/martinrb/ws/openjdk6/langtools: .hgtags > Only in /home/martinrb/ws/openjdk6/langtools: .jcheck > Only in /home/martinrb/ws/openjdk6/langtools: ASSEMBLY_EXCEPTION > Only in /home/martinrb/ws/openjdk6/langtools: LICENSE > Only in /home/martinrb/ws/openjdk6/langtools: README > Only in /home/martinrb/ws/openjdk6/langtools: THIRD_PARTY_README > Only in /home/martinrb/ws/openjdk6/langtools: TRADEMARK > Only in /home/martinrb/ws/openjdk6/: make > > > On Wed, Apr 22, 2009 at 14:47, Joseph D. Darcy wrote: >> Martin, >> >> The Mercurial transition included the folding of the control workspace into >> the jdk; I don't know if the differences you are seeing are from that part >> of the switch which did not preserve file layout. >> >> Kelly's scripts to retcon the OpenJDK 6 repositories with the teamware >> workspaces did include the sort of file level checks you recommend, at least >> for files under /src. (A few binary files eventually removed from the jaxws >> workspace were not included at all in that Mercurial repository.) >> >> -Joe >> >> >> Martin Buchholz wrote: >>> I've been doing some more work on openjdk6 and there are some more >>> inconsistencies. >>> The top-level repo has a Makefile with history, >>> but the old tarball openjdk-6-src-b11-10_jul_2008.tar.gz >>> doesn't seem to contain this file. >>> Which is authoritative, the old tarball or hg? >>> >>> An audit of jdk6 tarballs and the mercurial repo tags seems like a good >>> idea. >>> A command like >>> diff -r --brief --exclude=.hg >>> might be your friend. >>> >>> $ cd ~/ws/openjdk6; hg update -r jdk6-b11; hg log Makefile >>> zsh: exit 1 >>> 0 files updated, 0 files merged, 0 files removed, 0 files unresolved >>> changeset: 8:51c62af7b063 >>> tag: jdk6-b04 >>> user: ohair >>> date: Fri Jan 30 16:24:38 2009 -0800 >>> summary: 6755905: Changes for openjdk6 build 04 >>> >>> changeset: 6:f55431cdca29 >>> tag: jdk6-b03 >>> user: ohair >>> date: Fri Jan 30 16:20:33 2009 -0800 >>> summary: 6755902: Changes for openjdk6 build 03 >>> >>> changeset: 2:79e050893ef4 >>> tag: jdk6-b00 >>> user: ohair >>> date: Fri Jan 30 16:03:02 2009 -0800 >>> summary: 6755277: All initial changes to jdk7 to create openjdk 6 >>> build 0 >>> >>> changeset: 0:d5ea358b429b >>> user: duke >>> date: Fri Jan 30 15:58:47 2009 -0800 >>> summary: Initial load is jdk7-b23 >>> >>> On Wed, Apr 22, 2009 at 09:37, Joseph D. Darcy wrote: >>> >>>> I'll check with our RE folks to see what is going on... >>>> >>>> -Joe >>>> >>>> Phil Race wrote: >>>> >>>>> All I know is that I pushed the change 4 whole weeks ago. >>>>> How the b16 source bundle was created is not something into which >>>>> I have any insight. >>>>> >>>>> -phil. >>>>> >>>>> Lillian Angel wrote: >>>>> >>>>>> Joseph D. Darcy wrote: >>>>>> >>>>>>> Joseph D. Darcy wrote: >>>>>>> >>>>>>>> Lillian Angel wrote: >>>>>>>> >>>>>>>>> Hi, >>>>>>>>> >>>>>>>>> >>>>>>>>> Joseph D. Darcy wrote: >>>>>>>>> >>>>>>>>>> Without further adieu, the source bundle for OpenJDK 6 b16 is >>>>>>>>>> available from >>>>>>>>>> >>>>>>>>>> http://download.java.net/openjdk/jdk6/ >>>>>>>>>> >>>>>>>>>> and the OpenJDK 6 Mercurial repositories >>>>>>>>>> (http://hg.openjdk.java.net/jdk6/jdk6) will be tagged with the >>>>>>>>>> build >>>>>>>>>> information accordingly. >>>>>>>>>> >>>>>>>>>> The previous source bundle was for b14. There was no distinct b15 >>>>>>>>>> per se; b15 was used to mark the transition from teamware to >>>>>>>>>> Mercurial. The >>>>>>>>>> b16 build is the first to be done purely in Mercurial. >>>>>>>>>> >>>>>>>>>> There were 45 bug fixes in this build ranging from security fixes >>>>>>>>>> to >>>>>>>>>> sound updates; OpenJDK 6 b16 has all relevant security fixes from >>>>>>>>>> the recent >>>>>>>>>> JDK synchronized security release. A few fixes of note, during a >>>>>>>>>> build >>>>>>>>>> importing binary plugs now defaults to false (6781572 "Default to >>>>>>>>>> IMPORT_BINARY_PLUGS=false for OpenJDK 6 builds") and several fixes >>>>>>>>>> were >>>>>>>>>> applied that should render a few IcedTea patches unnecessary: >>>>>>>>>> >>>>>>>>>> 6736248 EnumEditor bug. Class check incorrect >>>>>>>>>> 6733718 test >>>>>>>>>> /java/awt/FullScreen/UninitializedDisplayModeChangeTest/ fails >>>>>>>>>> 6593946 (bf) X-Buffer.compact() does not discard mark as specified >>>>>>>>>> 6778493 Fix (langtools) ant build to honor fcs MILESTONE setting >>>>>>>>>> >>>>>>>>>> The other non security bugs fixed in this build are: >>>>>>>>>> >>>>>>>>>> 6761791: Crash in the FontManager code due to use of JNIEnv saved >>>>>>>>>> by >>>>>>>>>> another thread >>>>>>>>>> 6512707: "incompatible types" after (unrelated) annotation >>>>>>>>>> processing >>>>>>>>>> 6632696: Writing to closed output files (writeBytes) leaks native >>>>>>>>>> memory (unix) >>>>>>>>>> 6788196: (porting) Bounds checks in io_util.c rely on undefined >>>>>>>>>> behaviour >>>>>>>>>> 6791458: FileInputStream/RandomAccessFile.read leaks memory if >>>>>>>>>> invoked on closed stream with len > 8k >>>>>>>>>> 6792066: src/share/native/java/io/io_util.c clean-ups >>>>>>>>>> 6819886: System.getProperty("os.name") reports Vista on Windows 7 >>>>>>>>>> 6821031: Upgrade OpenJDK's LittleCMS version to 1.18 >>>>>>>>>> >>>>>>>>> Actually, the 1.18 version was not included in this release. b16 >>>>>>>>> still >>>>>>>>> includes 1.16. Any thoughts? >>>>>>>>> >>>>>>>> Phil? >>>>>>>> >>>>>>>> -Joe >>>>>>>> >>>>>>> PS b16 does include this changeset, does it not? >>>>>>> >>>>>>> http://hg.openjdk.java.net/jdk6/jdk6/jdk/rev/2bdeec6199b7 >>>>>>> >>>>>> Apparently not. This change shows that has clearly been updated to 1.18 >>>>>> (see version in comment), but it does not appear in the b16 sources >>>>>> (still >>>>>> says 1.16): >>>>>> >>>>>> >>>>>> http://hg.openjdk.java.net/jdk6/jdk6/jdk/diff/2bdeec6199b7/src/share/native/sun/java2d/cmm/lcms/lcms.h >>>>>> >>>>>> Lillian >>>>>> >>>> >> From martinrb at google.com Wed Apr 22 18:21:33 2009 From: martinrb at google.com (Martin Buchholz) Date: Wed, 22 Apr 2009 18:21:33 -0700 Subject: OpenJDK 6 build 16 source bundle posted In-Reply-To: <49EFB9F7.7080908@sun.com> References: <49E69540.80801@sun.com> <49EF3C1F.4020002@sun.com> <49EF3DA2.1080805@sun.com> <49EF3E20.6040401@redhat.com> <49EF428C.8040904@sun.com> <49EF47DF.5030408@sun.com> <1ccfd1c10904221200j47a927a3w65b3e821af082755@mail.gmail.com> <49EF9056.9020705@sun.com> <1ccfd1c10904221723q2bf34991l613a8fc97bee65d7@mail.gmail.com> <49EFB9F7.7080908@sun.com> Message-ID: <1ccfd1c10904221821o24078189w40c6fa50a180f652@mail.gmail.com> On Wed, Apr 22, 2009 at 17:44, Kelly O'Hair wrote: > That looks exactly right for the old tarball created from the > old TeamWare workspaces (which had a control workspace). > The conversion to Mercurial folded the control workspace files into > the top or enclosing Mercurial repository. > The control/make/Makefile moved to ./Makefile and the > control/make directory moved to ./make. Thanks for the clarification, Kelly. I'm surprised by this reorganization of the old jdk6 builds, but I guess it doesn't matter too much going forward. Having the additional history in mercurial is very nice, but if I ever want to rebuild an old build, I'l use the tarballs. Martin From Joe.Darcy at Sun.COM Wed Apr 22 23:51:03 2009 From: Joe.Darcy at Sun.COM (Joseph D. Darcy) Date: Wed, 22 Apr 2009 23:51:03 -0700 Subject: [Request for review] 6823449: Gervill: ArrayIndexOutOfBoundsException thrown when trying to play too may voices at same time. In-Reply-To: <36EC82E93EB0AD40A4301DAD65432386F5A60E21E8@mail.midverk.is> References: <36EC82E93EB0AD40A4301DAD65432386F5A60E21E8@mail.midverk.is> Message-ID: <49F00FD7.5090407@sun.com> Karl Helgason wrote: > Hi, > > I need code review for the fix: > http://cr.openjdk.java.net/~kalli/6823449/webrev.01/ > > And I need somebody to change the bug description: > http://bugs.sun.com/bugdatabase/view_bug.do?bug_id=6823449 > Changed. > Also I need new bug ids for these bugs/improvements: > > [New Bug]: Improve time-stamp support in Gervill to reduce jitter. > 6833357 Improve time-stamp support in Gervill to reduce jitter. > [New Bug]: Add support for sending time-stamp information in RealTimeSequencer to reduce jitter in playback. > 6833358 Add support for sending time-stamp information in RealTimeSequencer to reduce jitter in playback. > Those two bugs reduces playback glitches when playing back midi files like described here: > https://gervill.dev.java.net/servlets/ProjectForumMessageView?forumID=2923&messageID=26785 > > And also add new bug id for this bug: > > [New Bug]: Gervill: Add support to limit maximum CPU load by dynamically limit maximum voice polyphony. > 6833360 Gervill: Add support to limit maximum CPU load by dynamically limit maximum voice polyphony. Cheers, -Joe From rogeriopimentel at hotmail.com Thu Apr 23 08:35:46 2009 From: rogeriopimentel at hotmail.com (Rogerio Pimentel) Date: Thu, 23 Apr 2009 12:35:46 -0300 Subject: JDK6 for embedded ARM Message-ID: Hi ! Has anybody compiled JDK6 for any embedded device? Is JDK6 a good choice to use in embedded systems? Regards, Rogerio _________________________________________________________________ Novo Windows Live: Messenger 2009 e muito mais. Descubra! http://www.windowslive.com.br -------------- next part -------------- An HTML attachment was scrubbed... URL: http://mail.openjdk.java.net/pipermail/jdk6-dev/attachments/20090423/e9c437ea/attachment.html From Kelly.Ohair at Sun.COM Thu Apr 23 08:57:45 2009 From: Kelly.Ohair at Sun.COM (Kelly O'Hair) Date: Thu, 23 Apr 2009 08:57:45 -0700 Subject: OpenJDK 6 build 16 source bundle posted In-Reply-To: <1ccfd1c10904221821o24078189w40c6fa50a180f652@mail.gmail.com> References: <49E69540.80801@sun.com> <49EF3C1F.4020002@sun.com> <49EF3DA2.1080805@sun.com> <49EF3E20.6040401@redhat.com> <49EF428C.8040904@sun.com> <49EF47DF.5030408@sun.com> <1ccfd1c10904221200j47a927a3w65b3e821af082755@mail.gmail.com> <49EF9056.9020705@sun.com> <1ccfd1c10904221723q2bf34991l613a8fc97bee65d7@mail.gmail.com> <49EFB9F7.7080908@sun.com> <1ccfd1c10904221821o24078189w40c6fa50a180f652@mail.gmail.com> Message-ID: <49F08FF9.6010109@sun.com> Martin Buchholz wrote: > On Wed, Apr 22, 2009 at 17:44, Kelly O'Hair wrote: >> That looks exactly right for the old tarball created from the >> old TeamWare workspaces (which had a control workspace). > > >> The conversion to Mercurial folded the control workspace files into >> the top or enclosing Mercurial repository. >> The control/make/Makefile moved to ./Makefile and the >> control/make directory moved to ./make. > > Thanks for the clarification, Kelly. > > I'm surprised by this reorganization of the old jdk6 builds, Remember, OpenJDK6 derived from OpenJDK7, which has this same layout. -kto > but I guess it doesn't matter too much going forward. > Having the additional history in mercurial is very nice, > but if I ever want to rebuild an old build, I'l use the tarballs. > > Martin From alex.menkov at sun.com Fri Apr 24 02:01:52 2009 From: alex.menkov at sun.com (Alex Menkov) Date: Fri, 24 Apr 2009 13:01:52 +0400 Subject: Approved: [Request for review] 6823449: Gervill: ArrayIndexOutOfBoundsException thrown when trying to play too may voices at same time. In-Reply-To: <36EC82E93EB0AD40A4301DAD65432386F5A60E21E8@mail.midverk.is> References: <36EC82E93EB0AD40A4301DAD65432386F5A60E21E8@mail.midverk.is> Message-ID: <49F18000.5030901@sun.com> approval for the fix for 6823449 Regards Alex Karl Helgason wrote: > Hi, > > I need code review for the fix: > http://cr.openjdk.java.net/~kalli/6823449/webrev.01/ > > And I need somebody to change the bug description: > http://bugs.sun.com/bugdatabase/view_bug.do?bug_id=6823449 > > Also I need new bug ids for these bugs/improvements: > > [New Bug]: Improve time-stamp support in Gervill to reduce jitter. > [New Bug]: Add support for sending time-stamp information in RealTimeSequencer to reduce jitter in playback. > > Those two bugs reduces playback glitches when playing back midi files like described here: > https://gervill.dev.java.net/servlets/ProjectForumMessageView?forumID=2923&messageID=26785 > > And also add new bug id for this bug: > > [New Bug]: Gervill: Add support to limit maximum CPU load by dynamically limit maximum voice polyphony. > > regards, > Karl From kalli at midverk.is Fri Apr 24 15:57:36 2009 From: kalli at midverk.is (kalli at midverk.is) Date: Fri, 24 Apr 2009 22:57:36 +0000 Subject: hg: jdk6/jdk6/jdk: 6823449: Gervill: ArrayIndexOutOfBoundsException thrown when trying to play too may voices at same time. Message-ID: <20090424225749.F06F8E7B7@hg.openjdk.java.net> Changeset: a42d6999734b Author: kalli Date: 2009-04-24 22:52 +0000 URL: http://hg.openjdk.java.net/jdk6/jdk6/jdk/rev/a42d6999734b 6823449: Gervill: ArrayIndexOutOfBoundsException thrown when trying to play too may voices at same time. Reviewed-by: amenkov ! src/share/classes/com/sun/media/sound/SoftChannel.java + test/javax/sound/midi/Gervill/SoftChannel/NoteOverFlowTest2.java From xerxes at zafena.se Mon Apr 27 06:51:09 2009 From: xerxes at zafena.se (=?ISO-8859-1?Q?Xerxes_R=E5nby?=) Date: Mon, 27 Apr 2009 15:51:09 +0200 Subject: JDK6 for embedded ARM Message-ID: <49F5B84D.6030908@zafena.se> Greetings Rogerio Rogerio Pimentel wrote:* * > Hi ! > > > > Has anybody compiled JDK6 for any embedded device? OpenJDK6 have been known to compile and run on embedded ARM systems since the summer 2008. Binarys are currently shipping with the Debian/Linux ARM "armel" distribution, the Ubuntu "armel" ARM distribution and various embedded Linux distributions based on the openembedded infrastructure. Most development are currently performed at the Icedtea repository, patches are maing their way into OpenJDK through the Icedtea project: http://icedtea.classpath.org/wiki/Main_Page http://openjdk.java.net/projects/icedtea/ OpenJDK6 for ARM can be built using the Icedtea build infrastructue in three different ways: 1. Hotspot+zero, (interpreter only) currently the most compatible way of running JDK6 on ARM, zero have quite quick upstart times yet dont have great top speed since it are only running in interpreted mode. 2. Hotspot+zero+shark (mixed mode JIT and interpreter) Shark are still under heavy developement and are not stable enough for real use , Shark on ARM can at the moment run "Hello World" and some other trivial java programs that finishes quickly. 3. Cacao (JIT only) instead of using hotspot it are possible to use the Cacao research vm, cacao are quite fast yet it lacks debugging support like JVMTI and have some upstart overhead since all methods must be jited before they are run. I have written a FAQ how to cross compile OpenJDK for many embedded ARM/Linux distributions using Icedtea, Jalimo, OpenEmbedded and BitBake. Binarys generated this way are known to work on Nokia internet tablets, Neo freerunner phones, The BUG and many embedded ARM developement boards. http://icedtea.classpath.org/wiki/CrossCompileFaq > Is JDK6 a good choice to use in embedded systems? > It all depends on your expectations. Yes! Embedded systems have recently got access to quite a lot of storage by using sd card memorys and processing power have increased a lot thus running a JavaSE like environment on embedded hardware are now more possible than ever. Being able to use software deployment strategies like java webstart on networked embedded hardware are a reality using JDK6. Being able to run the same applications as used on desktops on embedded hardware are possible! yay! No! Currently the main bottlenecks running JDK6 on embedded ARM hardware becomes obvious when running GUI intense swing applications. Unfortunally most embedded chip makers don't provide accelerated embedded graphics chips drivers as free software and thus most graphics functionality don't get the acceleration as they should thus keeps slowing down the cpu. Conclusion: Running embedded java servers on screenless devices like plugcomputers: go for it! Running tailored client java applications with slightly reduced swing/awt gui's are all ok as well. Running heavy swing applications currently don't work that well. > > > Regards, > > > > Rogerio > Cheers, and have a great day! Xerxes From gnu_andrew at member.fsf.org Wed Apr 29 07:07:41 2009 From: gnu_andrew at member.fsf.org (Andrew John Hughes) Date: Wed, 29 Apr 2009 15:07:41 +0100 Subject: [FYI] Backport fix for mitre join decorations from OpenJDK7 In-Reply-To: <49F85B90.4030002@redhat.com> References: <49F85B90.4030002@redhat.com> Message-ID: <17c6771e0904290707g5eed965frf4fd1790aa313bf3@mail.gmail.com> 2009/4/29 Omair Majid : > Hi, > > The attached patch fixes mitre join decorations (Sun bug 6812600[1]). It > backports the fix in OpenJDK7 [2] to IcedTea6. > > ChangeLog: > > 2009-04-29 ?Omair Majid ? > > ? ? ? * patches/icedtea-java2d-mitre-join.patch: New patch. Backport > ? ? ? fix for mitre join decorations from OpenJDK7 (S6812600) along > ? ? ? with its test. > ? ? ? * Makefile.am (ICEDTEA_PATCHES): Apply the above. > ? ? ? * HACKING: Document the above. > > Cheers, > Omair > > [1] http://bugs.sun.com/bugdatabase/view_bug.do?bug_id=6812600 > [2] http://hg.openjdk.java.net/jdk7/2d/jdk/rev/9318628e8eee > This should also be pushed upstream to OpenJDK6, if acceptable. CCing to jdk6 list. -- Andrew :-) Free Java Software Engineer Red Hat, Inc. (http://www.redhat.com) Support Free Java! Contribute to GNU Classpath and the OpenJDK http://www.gnu.org/software/classpath http://openjdk.java.net PGP Key: 94EFD9D8 (http://subkeys.pgp.net) Fingerprint: F8EF F1EA 401E 2E60 15FA 7927 142C 2591 94EF D9D8 From langel at redhat.com Wed Apr 29 08:30:23 2009 From: langel at redhat.com (Lillian Angel) Date: Wed, 29 Apr 2009 11:30:23 -0400 Subject: OpenJDK 6 build 16 source bundle posted In-Reply-To: <49EF87F1.2010704@sun.com> References: <49E69540.80801@sun.com> <49EF3B03.1030705@redhat.com> <49EF3C1F.4020002@sun.com> <49EF3DA2.1080805@sun.com> <49EF3E20.6040401@redhat.com> <49EF428C.8040904@sun.com> <49EF47DF.5030408@sun.com> <49EF87F1.2010704@sun.com> Message-ID: <49F8728F.7060403@redhat.com> Hi, Joe Darcy wrote: > Hello. > > There was a misunderstanding with RE; the posted b16 sources were > based on an internal staging repository being used to hold the > security fixes before they were made public. Fixes pushed to the > public repo after that point were not included. I've asked RE to redo > a version of b16 to reflect the current state of the OpenJDK 6 > repositories, including the little cms changes. The new source > bundled and corresponding tags in the repository should be available > within a few days. Any update on this? Is there a new source bundle? Thanks Lillian > > Sorry for the confusion, > > -Joe > > On 04/22/09 09:37 AM, Joseph D. Darcy wrote: >> I'll check with our RE folks to see what is going on... >> >> -Joe >> >> Phil Race wrote: >>> All I know is that I pushed the change 4 whole weeks ago. >>> How the b16 source bundle was created is not something into which >>> I have any insight. >>> >>> -phil. >>> >>> Lillian Angel wrote: >>>> Joseph D. Darcy wrote: >>>>> Joseph D. Darcy wrote: >>>>>> Lillian Angel wrote: >>>>>>> Hi, >>>>>>> >>>>>>> >>>>>>> Joseph D. Darcy wrote: >>>>>>>> Without further adieu, the source bundle for OpenJDK 6 b16 is >>>>>>>> available from >>>>>>>> >>>>>>>> http://download.java.net/openjdk/jdk6/ >>>>>>>> >>>>>>>> and the OpenJDK 6 Mercurial repositories >>>>>>>> (http://hg.openjdk.java.net/jdk6/jdk6) will be tagged with the >>>>>>>> build information accordingly. >>>>>>>> >>>>>>>> The previous source bundle was for b14. There was no distinct >>>>>>>> b15 per se; b15 was used to mark the transition from teamware >>>>>>>> to Mercurial. The b16 build is the first to be done purely in >>>>>>>> Mercurial. >>>>>>>> >>>>>>>> There were 45 bug fixes in this build ranging from security >>>>>>>> fixes to sound updates; OpenJDK 6 b16 has all relevant security >>>>>>>> fixes from the recent JDK synchronized security release. A few >>>>>>>> fixes of note, during a build importing binary plugs now >>>>>>>> defaults to false (6781572 "Default to >>>>>>>> IMPORT_BINARY_PLUGS=false for OpenJDK 6 builds") and several >>>>>>>> fixes were applied that should render a few IcedTea patches >>>>>>>> unnecessary: >>>>>>>> >>>>>>>> 6736248 EnumEditor bug. Class check incorrect >>>>>>>> 6733718 test >>>>>>>> /java/awt/FullScreen/UninitializedDisplayModeChangeTest/ fails >>>>>>>> 6593946 (bf) X-Buffer.compact() does not discard mark as >>>>>>>> specified >>>>>>>> 6778493 Fix (langtools) ant build to honor fcs MILESTONE setting >>>>>>>> >>>>>>>> The other non security bugs fixed in this build are: >>>>>>>> >>>>>>>> 6761791: Crash in the FontManager code due to use of JNIEnv >>>>>>>> saved by another thread >>>>>>>> 6512707: "incompatible types" after (unrelated) annotation >>>>>>>> processing >>>>>>>> 6632696: Writing to closed output files (writeBytes) leaks >>>>>>>> native memory (unix) >>>>>>>> 6788196: (porting) Bounds checks in io_util.c rely on undefined >>>>>>>> behaviour >>>>>>>> 6791458: FileInputStream/RandomAccessFile.read leaks memory if >>>>>>>> invoked on closed stream with len > 8k >>>>>>>> 6792066: src/share/native/java/io/io_util.c clean-ups >>>>>>>> 6819886: System.getProperty("os.name") reports Vista on Windows 7 >>>>>>>> 6821031: Upgrade OpenJDK's LittleCMS version to 1.18 >>>>>>> >>>>>>> Actually, the 1.18 version was not included in this release. b16 >>>>>>> still includes 1.16. Any thoughts? >>>>>> >>>>>> Phil? >>>>>> >>>>>> -Joe >>>>> >>>>> PS b16 does include this changeset, does it not? >>>>> >>>>> http://hg.openjdk.java.net/jdk6/jdk6/jdk/rev/2bdeec6199b7 >>>> >>>> Apparently not. This change shows that has clearly been updated to >>>> 1.18 (see version in comment), but it does not appear in the b16 >>>> sources (still says 1.16): >>>> http://hg.openjdk.java.net/jdk6/jdk6/jdk/diff/2bdeec6199b7/src/share/native/sun/java2d/cmm/lcms/lcms.h >>>> >>>> >>>> Lillian >> > From Joe.Darcy at Sun.COM Wed Apr 29 08:38:55 2009 From: Joe.Darcy at Sun.COM (Joseph D. Darcy) Date: Wed, 29 Apr 2009 08:38:55 -0700 Subject: [FYI] Backport fix for mitre join decorations from OpenJDK7 In-Reply-To: <17c6771e0904290707g5eed965frf4fd1790aa313bf3@mail.gmail.com> References: <49F85B90.4030002@redhat.com> <17c6771e0904290707g5eed965frf4fd1790aa313bf3@mail.gmail.com> Message-ID: <49F8748F.8060503@sun.com> Andrew John Hughes wrote: > 2009/4/29 Omair Majid : > >> Hi, >> >> The attached patch fixes mitre join decorations (Sun bug 6812600[1]). It >> backports the fix in OpenJDK7 [2] to IcedTea6. >> >> ChangeLog: >> >> 2009-04-29 Omair Majid >> >> * patches/icedtea-java2d-mitre-join.patch: New patch. Backport >> fix for mitre join decorations from OpenJDK7 (S6812600) along >> with its test. >> * Makefile.am (ICEDTEA_PATCHES): Apply the above. >> * HACKING: Document the above. >> >> Cheers, >> Omair >> >> [1] http://bugs.sun.com/bugdatabase/view_bug.do?bug_id=6812600 >> [2] http://hg.openjdk.java.net/jdk7/2d/jdk/rev/9318628e8eee >> >> > > This should also be pushed upstream to OpenJDK6, if acceptable. > CCing to jdk6 list. > Hello. That would be fine with me. -Joe From Joe.Darcy at Sun.COM Wed Apr 29 08:44:45 2009 From: Joe.Darcy at Sun.COM (Joseph D. Darcy) Date: Wed, 29 Apr 2009 08:44:45 -0700 Subject: OpenJDK 6 build 16 source bundle posted In-Reply-To: <49F8728F.7060403@redhat.com> References: <49E69540.80801@sun.com> <49EF3B03.1030705@redhat.com> <49EF3C1F.4020002@sun.com> <49EF3DA2.1080805@sun.com> <49EF3E20.6040401@redhat.com> <49EF428C.8040904@sun.com> <49EF47DF.5030408@sun.com> <49EF87F1.2010704@sun.com> <49F8728F.7060403@redhat.com> Message-ID: <49F875ED.1040207@sun.com> Lillian Angel wrote: > Hi, > > Joe Darcy wrote: >> Hello. >> >> There was a misunderstanding with RE; the posted b16 sources were >> based on an internal staging repository being used to hold the >> security fixes before they were made public. Fixes pushed to the >> public repo after that point were not included. I've asked RE to >> redo a version of b16 to reflect the current state of the OpenJDK 6 >> repositories, including the little cms changes. The new source >> bundled and corresponding tags in the repository should be available >> within a few days. > > Any update on this? Is there a new source bundle? > Hello. The new source bundle has been prepared and I've run regression tests on our internal builds and it looks fine. The bundle should be posted in the near future; I'll send email when it is. -Joe From langel at redhat.com Wed Apr 29 08:48:00 2009 From: langel at redhat.com (Lillian Angel) Date: Wed, 29 Apr 2009 11:48:00 -0400 Subject: OpenJDK 6 build 16 source bundle posted In-Reply-To: <49F875ED.1040207@sun.com> References: <49E69540.80801@sun.com> <49EF3B03.1030705@redhat.com> <49EF3C1F.4020002@sun.com> <49EF3DA2.1080805@sun.com> <49EF3E20.6040401@redhat.com> <49EF428C.8040904@sun.com> <49EF47DF.5030408@sun.com> <49EF87F1.2010704@sun.com> <49F8728F.7060403@redhat.com> <49F875ED.1040207@sun.com> Message-ID: <49F876B0.7040705@redhat.com> Joseph D. Darcy wrote: > Lillian Angel wrote: >> Hi, >> >> Joe Darcy wrote: >>> Hello. >>> >>> There was a misunderstanding with RE; the posted b16 sources were >>> based on an internal staging repository being used to hold the >>> security fixes before they were made public. Fixes pushed to the >>> public repo after that point were not included. I've asked RE to >>> redo a version of b16 to reflect the current state of the OpenJDK 6 >>> repositories, including the little cms changes. The new source >>> bundled and corresponding tags in the repository should be available >>> within a few days. >> >> Any update on this? Is there a new source bundle? >> > Hello. > > The new source bundle has been prepared and I've run regression tests > on our internal builds and it looks fine. The bundle should be posted > in the near future; I'll send email when it is. Great! Thanks for the quick reply. From Joe.Darcy at Sun.COM Thu Apr 30 20:05:14 2009 From: Joe.Darcy at Sun.COM (Joe Darcy) Date: Thu, 30 Apr 2009 20:05:14 -0700 Subject: OpenJDK 6 build 16 source bundle posted In-Reply-To: <49F875ED.1040207@sun.com> References: <49E69540.80801@sun.com> <49EF3B03.1030705@redhat.com> <49EF3C1F.4020002@sun.com> <49EF3DA2.1080805@sun.com> <49EF3E20.6040401@redhat.com> <49EF428C.8040904@sun.com> <49EF47DF.5030408@sun.com> <49EF87F1.2010704@sun.com> <49F8728F.7060403@redhat.com> <49F875ED.1040207@sun.com> Message-ID: <49FA66EA.3060901@sun.com> On 04/29/09 08:44 AM, Joseph D. Darcy wrote: > Lillian Angel wrote: >> Hi, >> >> Joe Darcy wrote: >>> Hello. >>> >>> There was a misunderstanding with RE; the posted b16 sources were >>> based on an internal staging repository being used to hold the >>> security fixes before they were made public. Fixes pushed to the >>> public repo after that point were not included. I've asked RE to >>> redo a version of b16 to reflect the current state of the OpenJDK 6 >>> repositories, including the little cms changes. The new source >>> bundled and corresponding tags in the repository should be available >>> within a few days. >> >> Any update on this? Is there a new source bundle? >> > Hello. > > The new source bundle has been prepared and I've run regression tests > on our internal builds and it looks fine. The bundle should be posted > in the near future; I'll send email when it is. > > -Joe Take two of the b16 source bundle is now at http://download.java.net/openjdk/jdk6/promoted/b16/openjdk-6-src-b16-24_apr_2009.tar.gz The repositories will be tagged accordingly soon. Sorry for the earlier mix-up, -Joe