HotSpot 16
Andrew John Hughes
gnu_andrew at member.fsf.org
Fri Jan 15 11:11:39 PST 2010
2010/1/14 Andrew John Hughes <gnu_andrew at member.fsf.org>:
> 2010/1/14 Andrew John Hughes <gnu_andrew at member.fsf.org>:
>> 2010/1/14 Matthias Klose <doko at ubuntu.com>:
>>> On 13.01.2010 22:37, Andrew John Hughes wrote:
>>>>
>>>> I've now bumped IcedTea6 HEAD and 1.7 to build the latest hs16 (b13)
>>>> as I believe is used in the proprietary JDK 6 update 18.
>>>>
>>>> I've confirmed that both normal and Zero builds complete. Shark also
>>>> builds but can't bootstrap itself; see
>>>> http://icedtea.classpath.org/bugzilla/show_bug.cgi?id=430 As both
>>>> Shark and hs16 are experimental, I don't think that's a release
>>>> stopper.
>>>
>>> so update hotspot two days before a planned release and then declare the
>>> regressions as non-stoppers?
>>>
>>
>> It's a minor update to a version of HotSpot that is _not the default_,
>> and which fully bootstraps with both normal and Zero builds. I don't
>> recall Shark working with hs16 before so how is this a regression?
>>
>> FWIW, Shark also can't bootstrap itself on 7 either
>> (http://icedtea.classpath.org/bugzilla/show_bug.cgi?id=348) and that's
>> been the case for some time. If I thought there was going to be a fix
>> for this in the next week or so, I would see the point in delaying the
>> release. Are you offering to fix this?
>>
>> As is, I don't see how there is anything to gain from waiting
>> indefinitely for a fix to a pretty minor build option (hs16+shark).
>> Shark is still a work-in-progress and anyone wanting to ship
>> Shark+hs16 binaries would be quite mad.
>>
>> As I said before, I've updated hs16 because if we're going to include
>> the option to build with it at all, we should use the same version Sun
>> are shipping and that will end up in b18.
>>
>
> On IRC, we found that this is a regression from when we first added
> hs16 (though not 1.6 obviously).
>
> These are the actual changes between the two hs16 revisions:
>
> changeset: 941:6bdfda9a7120
> user: trims
> date: Fri Oct 02 14:16:39 2009 -0700
> summary: 6887914: Bump the HS16 build number to 10
>
> changeset: 942:6de2c9c36168
> user: tonyp
> date: Tue Oct 20 19:55:13 2009 -0400
> summary: 6893095: G1: bulk G1 backports to hs16
>
> changeset: 943:47ca7aa20d97
> user: asaha
> date: Wed Oct 21 15:43:35 2009 -0700
> summary: 6858208: jvm crash when specifying TypeProfileWidth=0 on jdk 6.0
>
> changeset: 944:aa001a20bd61
> parent: 941:6bdfda9a7120
> user: ysr
> date: Fri Oct 16 02:05:46 2009 -0700
> summary: 6888898: CMS: ReduceInitialCardMarks unsafe in the
> presence of cms precleaning
>
> changeset: 945:164505f3849e
> user: ysr
> date: Tue Oct 20 00:00:23 2009 -0700
> summary: 6892749: assert(UseParNewGC || UseSerialGC ||
> UseConcMarkSweepGC, "...") fails
>
> changeset: 946:d43a7bf86266
> parent: 945:164505f3849e
> parent: 942:6de2c9c36168
> user: ysr
> date: Wed Oct 21 12:35:50 2009 -0700
> summary: Merge
>
> changeset: 947:aa83c08db78c
> parent: 946:d43a7bf86266
> parent: 943:47ca7aa20d97
> user: ysr
> date: Wed Oct 21 19:13:54 2009 -0700
> summary: Merge
>
> changeset: 948:b7e7fc226009
> parent: 946:d43a7bf86266
> parent: 943:47ca7aa20d97
> user: ysr
> date: Wed Oct 21 18:57:28 2009 -0700
> summary: Merge
>
> changeset: 949:0537635c7c16
> parent: 948:b7e7fc226009
> parent: 947:aa83c08db78c
> user: ysr
> date: Wed Oct 21 21:06:27 2009 -0700
> summary: Merge
>
> changeset: 950:06264910ac5f
> user: trims
> date: Wed Oct 21 21:31:53 2009 -0700
> summary: 6893995: Bump the HS16 build number to 11
>
> changeset: 951:38b152ff0cac
> user: trims
> date: Wed Oct 21 21:34:37 2009 -0700
> summary: 6893996: change default JPRT release target to be 6u18
>
> changeset: 952:0799687b7385
> user: phh
> date: Thu Oct 29 11:45:09 2009 -0400
> summary: 6887571: Increase default heap config sizes
>
> changeset: 953:b0651468aff7
> user: ysr
> date: Wed Oct 28 11:16:42 2009 -0700
> summary: 6818264: Heap dumper unexpectedly adds .hprof suffix
>
> changeset: 954:511f4c69c1be
> user: johnc
> date: Thu Oct 29 09:42:26 2009 -0700
> summary: 6889740: G1: OpenDS fails with "unhandled exception in
> compiled code"
>
> changeset: 955:684c526959a9
> parent: 953:b0651468aff7
> user: phh
> date: Tue Nov 03 15:17:42 2009 -0500
> summary: 6852078: HSX 14/16 in jdk 5.0: api/javax_management
> api/org_omg jck tests crashes or make tnameserv crash
>
> changeset: 956:c5388a988d5a
> parent: 954:511f4c69c1be
> parent: 955:684c526959a9
> user: phh
> date: Wed Nov 04 07:07:14 2009 -0800
> summary: Merge
>
> changeset: 957:4385d4e7497b
> user: ysr
> date: Tue Nov 03 16:43:16 2009 -0800
> summary: 6801625: CDS: HeapDump tests crash with internal error in
> compactingPermGenGen.cpp
>
> changeset: 958:248f0ad25d21
> user: kvn
> date: Fri Nov 06 10:07:50 2009 -0800
> summary: 6896084: VM does not reserve protected page below heap
> for compressed oops implicit null checks
>
> changeset: 959:3a2dae936e36
> user: kvn
> date: Fri Nov 06 10:23:54 2009 -0800
> summary: 6893554: SPECjvm2008 mpegaudio fails with SecurityException
>
> changeset: 960:395db6268de2
> user: cfang
> date: Fri Nov 06 14:38:53 2009 -0800
> summary: 6876276: assert(!is_visited,"visit only once")
>
> changeset: 961:26d0c7276c9e
> parent: 954:511f4c69c1be
> user: phh
> date: Fri Nov 06 19:57:55 2009 -0500
> summary: 6898160: Need serviceability support for new vm argument
> type 'uint64_t'
>
> changeset: 962:a76f17a8fd77
> parent: 960:395db6268de2
> parent: 961:26d0c7276c9e
> user: phh
> date: Fri Nov 06 17:44:04 2009 -0800
> summary: Merge
>
> changeset: 963:788af96b55e3
> user: apetrusenko
> date: Thu Oct 22 07:43:49 2009 -0700
> summary: 6858886: G1: guarantee(_next_marked_bytes <=
> used(),"invariant") at heapRegion.hpp:359
>
> changeset: 964:e3c995ac8078
> user: tonyp
> date: Fri Oct 23 14:34:27 2009 -0400
> summary: 6886024: G1: assert(recent_avg_pause_time_ratio() < 1.00,"All GC?")
>
> changeset: 965:2572dd3d6d7c
> user: tonyp
> date: Wed Sep 30 14:50:51 2009 -0400
> summary: 6890137: G1: revamp reachable object dump
>
> changeset: 966:850627f5fd81
> user: apetrusenko
> date: Tue Oct 27 02:42:24 2009 -0700
> summary: 6870843: G1: G1 GC memory leak
>
> changeset: 967:9f7d1f6201ab
> user: johnc
> date: Fri Nov 06 11:10:05 2009 -0800
> summary: 6895788: G1: SATB and update buffer allocation code
> allocates too much space
>
> changeset: 968:1a1b644fe617
> parent: 962:a76f17a8fd77
> user: never
> date: Thu Nov 12 09:24:21 2009 -0800
> summary: 6892658: C2 should optimize some stringbuilder patterns
>
> changeset: 969:6571190a0304
> parent: 968:1a1b644fe617
> parent: 967:9f7d1f6201ab
> user: never
> date: Thu Nov 12 14:27:25 2009 -0800
> summary: Merge
>
> changeset: 970:0ad684e27697
> parent: 962:a76f17a8fd77
> user: phh
> date: Thu Nov 12 18:07:55 2009 -0500
> summary: 6900899: vm fails to start when -Xmx value is less than
> OldSize + NewSize
>
> changeset: 971:da3df85c0bcd
> parent: 969:6571190a0304
> parent: 970:0ad684e27697
> user: phh
> date: Thu Nov 12 15:25:37 2009 -0800
> summary: Merge
>
> changeset: 972:dba6aafa42ff
> parent: 969:6571190a0304
> user: jmasa
> date: Thu Nov 12 15:01:43 2009 -0800
> summary: 6898857: [Regression] -XX:NewRatio with
> -XX:+UseConcMarkSweepGC causes fatal error
>
> changeset: 973:54d471e8741d
> parent: 972:dba6aafa42ff
> parent: 971:da3df85c0bcd
> user: jmasa
> date: Thu Nov 12 17:36:13 2009 -0800
> summary: Merge
>
> changeset: 974:834503c17fc1
> user: trims
> date: Thu Nov 12 19:44:32 2009 -0800
> summary: 6900997: Bump the HS16 build number to 12
>
> changeset: 975:938b86f7a7fc
> user: cfang
> date: Tue Nov 24 15:52:43 2009 -0800
> summary: 6892079: live value must not be garbage failure after fix
> for 6854812
>
> changeset: 976:c5cb255ec7a3
> user: trims
> date: Wed Nov 25 12:43:26 2009 -0800
> summary: 6821003: Update hotspot windows os_win32 for windows 7
>
> changeset: 977:5a6085c47686
> user: trims
> date: Wed Apr 08 15:16:14 2009 -0700
> summary: 6828069: Change JDK_MINOR_VER to 6 for 6Update HS versions
>
> changeset: 978:5f14026df877
> user: asaha
> date: Fri Jul 10 14:00:02 2009 -0700
> summary: 6842999: Update hotspot windows os_win32 for windows 2008 R2
>
> changeset: 979:43e3f86b638d
> parent: 975:938b86f7a7fc
> user: cfang
> date: Wed Nov 25 12:09:02 2009 -0800
> summary: 6904191: OptimizeStringConcat should be product instead
> of experimental
>
> changeset: 980:c44e78b9aa0a
> parent: 978:5f14026df877
> parent: 979:43e3f86b638d
> user: trims
> date: Wed Nov 25 17:01:17 2009 -0800
> summary: Merge
>
> changeset: 981:b6e6e189e1db
> user: trims
> date: Wed Nov 25 19:15:17 2009 -0800
> summary: 6904996: Bump the HS16 build number to 13
>
> changeset: 982:49bd3de668dc
> user: dholmes
> date: Tue Dec 01 19:10:24 2009 -0500
> summary: 6822370: ReentrantReadWriteLock: threads hung when there
> are no threads holding onto the lock (Netra x4450)
>
> changeset: 983:9adb2f184e47
> user: ysr
> date: Thu Dec 03 15:01:57 2009 -0800
> summary: 6906727: UseCompressedOops: some card-marking fixes
> related to object arrays
>
> changeset: 984:62926c7f67a3
> user: ysr
> date: Tue Dec 08 15:12:17 2009 -0800
> summary: 6908208: UseCompressedOops: array_size() returns
> incorrect size for MAX_INT object array following 6906727
>
> I don't see anything obvious though. The only thing I can think of is
> some option was turned on that Shark doesn't support. As 7 fails too,
> I'm not optimistic of a quick fix, so we may just have to go for an
> appropriate warning that hs16+Shark are not supported for this
> release.
>
>>> the shark build fails for me as well when using the new hotspot as bootstrap
>>> compiler (on ix86 configuring --with-additional-vms=shark), so this is also
>>> seen with other bootstrap compilers.
>>
>> What other bootstrap compilers? Other than the arch., that sounds
>> pretty much like what I did.
>>
>
> From IRC; sounds like Matthias is hitting the same or a similar
> problem, but the --with-additional-vms hits it at a different point
> because Shark is invoked later.
>
>>>
>>
>>
>>
>> --
>> 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
>>
>
>
>
> --
> 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
>
Ok, I'm going to summarise things in this email because I'm getting
confused now as to what works and what doesn't:
The only one missing below is the ARM port as these builds are all on x86_64.
Normal builds
==========
1. --enable-vms=cacao,shark
$ /home/andrew/build/addvm-icedtea6/bin/java -version
java version "1.6.0_17"
OpenJDK Runtime Environment (IcedTea6 1.8pre-r1a118a8c23ff) (build 1.6.0_17-b17)
OpenJDK 64-Bit Server VM (build 14.0-b16, mixed mode)
$ /home/andrew/build/addvm-icedtea6/bin/java -version -zero
java version "1.6.0_17"
OpenJDK Runtime Environment (IcedTea6 1.8pre-r1a118a8c23ff) (build 1.6.0_17-b17)
OpenJDK 64-Bit Shark VM (build 14.0-b16, mixed mode)
$ /home/andrew/build/addvm-icedtea6/bin/java -version -cacao
java version "1.6.0_17"
IcedTea Runtime Environment (IcedTea6 1.8pre-r1a118a8c23ff) (build 1.6.0_17-b17)
CACAO (build 0.99.4, compiled mode)
2. --enable-shark
Crashes on stage 2:dbh
[javac] # Internal Error (os_linux_zero.cpp:236), pid=22828,
tid=47264282425104
[javac] # Error: caught unhandled signal 11
[javac] #
[javac] # JRE version: 6.0_17-b17
[javac] # Java VM: OpenJDK 64-Bit Shark VM (14.0-b16 mixed mode
linux-amd64 )
[javac] # An error report file with more information is saved as:
[javac] # /mnt/builder/shark6/openjdk/langtools/hs_err_pid22828.log
$ /mnt/builder/shark6/bootstrap/jdk1.6.0/bin/java -version
java version "1.6.0_17"
OpenJDK Runtime Environment (IcedTea6 1.8pre-rd9e57470d35d) (build 1.6.0_17-b17)
OpenJDK 64-Bit Shark VM (build 14.0-b16, mixed mode)
3. --enable-zero
$ /mnt/builder/zero6/openjdk/build/linux-amd64/j2sdk-image/bin/java -version
java version "1.6.0_17"
OpenJDK Runtime Environment (IcedTea6 1.8pre-rd9e57470d35d) (build 1.6.0_17-b17)
OpenJDK 64-Bit Zero VM (build 14.0-b16, interpreted mode)
Builds using --with-hotspot-build=hs16:
===========================
1. --enable-shark:
Crashes on stage 2:
[javac] Compiling 244 source files to
/mnt/builder/shark6/openjdk/build/linux-amd64/langtools/build/classes
[javac] #
[javac] # A fatal error has been detected by the Java Runtime Environment:
[javac] #
[javac] # Internal Error (os_linux_zero.cpp:236), pid=22007,
tid=47864850761488
[javac] # Error: caught unhandled signal 11
[javac] #
[javac] # JRE version: 6.0_17-b17
[javac] # Java VM: OpenJDK 64-Bit Shark VM (16.0-b13 mixed mode
linux-amd64 )
2. --enable-vms=cacao,zero
Succeeds
$ /home/andrew/build/addvm-icedtea6/bin/java -version
java version "1.6.0_17"
OpenJDK Runtime Environment (IcedTea6 1.8pre-r1a118a8c23ff) (build 1.6.0_17-b17)
OpenJDK 64-Bit Server VM (build 16.0-b13, mixed mode)
$ /home/andrew/build/addvm-icedtea6/bin/java -zero -version
java version "1.6.0_17"
OpenJDK Runtime Environment (IcedTea6 1.8pre-r1a118a8c23ff) (build 1.6.0_17-b17)
OpenJDK 64-Bit Zero VM (build 16.0-b13, interpreted mode)
$ /home/andrew/build/addvm-icedtea6/bin/java -cacao -version
java version "1.6.0_17"
IcedTea Runtime Environment (IcedTea6 1.8pre-r1a118a8c23ff) (build 1.6.0_17-b17)
CACAO (build 0.99.4, compiled mode)
3. --enable-zero
/mnt/builder/zero6/openjdk/build/linux-amd64/bin/java -version
java version "1.6.0_17"
OpenJDK Runtime Environment (IcedTea6 1.8pre-r1a118a8c23ff) (build 1.6.0_17-b17)
OpenJDK 64-Bit Zero VM (build 16.0-b13, interpreted mode)
1.6.2+hg
=======
1. --enable-shark
java version "1.6.0_0"
OpenJDK Runtime Environment (IcedTea6 1.6.2-r66c327225114) (build 1.6.0_0-b16)
OpenJDK 64-Bit Shark VM (build 14.0-b16, mixed mode)
HotSpot jtreg:
Passed: compiler/6646019/Test.java
FAILED: compiler/6646020/Tester.java
Passed: compiler/6659207/Test.java
Passed: compiler/6661247/Test.java
Passed: compiler/6663621/IVTest.java
Passed: compiler/6663848/Tester.java
Passed: compiler/6689060/Test.java
Passed: compiler/6695810/Test.java
Passed: compiler/6700047/Test6700047.java
Passed: compiler/6711100/Test.java
Passed: compiler/6712835/Test6712835.java
Passed: compiler/6714694/Tester.java
FAILED: compiler/6716441/Tester.java
Passed: compiler/6724218/Test.java
Passed: compiler/6726999/Test.java
FAILED: compiler/6741738/Tester.java
Passed: compiler/6756768/Test6756768_2.java
Passed: compiler/6756768/Test6756768.java
Passed: compiler/6757316/Test6757316.java
Passed: compiler/6758234/Test6758234.java
Passed: compiler/6775880/Test.java
Passed: compiler/6795362/Test6795362.java
Passed: compiler/6805724/Test6805724.java
Passed: runtime/6819213/TestBootNativeLibraryPath.java
Test results: passed: 21; failed: 3
So Shark has regressed to not bootstrapping with either HotSpot in 1.7.
--
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
More information about the distro-pkg-dev
mailing list