[Bug 3443] New: [IcedTea7] OpenJDK on aarch64 has fatal error "Field too big for insn"

classic Classic list List threaded Threaded
9 messages Options
Reply | Threaded
Open this post in threaded view
|

[Bug 3443] New: [IcedTea7] OpenJDK on aarch64 has fatal error "Field too big for insn"

bugzilla-daemon
Bug ID 3443
Summary [IcedTea7] OpenJDK on aarch64 has fatal error "Field too big for insn"
Product IcedTea
Version 7-hg
Hardware aarch64
OS All
Status NEW
Severity major
Priority P5
Component IcedTea
Assignee gnu.andrew@redhat.com
Reporter gnu.andrew@redhat.com
CC unassigned@icedtea.classpath.org

https://bugzilla.redhat.com/show_bug.cgi?id=1482244

# A fatal error has been detected by the Java Runtime Environment:
#
#  Internal Error (assembler_aarch64.hpp:226), pid=14956, tid=281472602731008
#  guarantee(chk == -1 || chk == 0) failed: Field too big for insn
#
# JRE version: OpenJDK Runtime Environment (7.0_151) (build
1.7.0_151-mockbuild_2017_08_02_08_31-b00)
# Java VM: OpenJDK 64-Bit Server VM (24.151-b00 mixed mode linux-aarch64
compressed oops)
# Derivative: IcedTea 2.6.10
# Distribution: Red Hat Enterprise Linux Server release 7.4 (Maipo), package
rhel-2.6.11.1.el7_4-aarch64 u151-b00
# Failed to write core dump. Core dumps have been disabled. To enable core
dumping, try "ulimit -c unlimited" before starting Java again

"8144028: Use AArch64 bit-test instructions in C2" had to be backported for one
of the security fixes. It looks like we also need:

8145438: Guarantee failures since 8144028: Use AArch64 bit-test instructions in
C2
Summary: Implement short and long versions of bit test instructions.
Reviewed-by: kvn


You are receiving this mail because:
  • You are on the CC list for the bug.
Reply | Threaded
Open this post in threaded view
|

[Bug 3443] [IcedTea7] OpenJDK on aarch64 has fatal error "Field too big for insn"

bugzilla-daemon
[hidden email] changed bug 3443
What Removed Added
Status NEW ASSIGNED
Target Milestone --- 2.6.12


You are receiving this mail because:
  • You are on the CC list for the bug.
Reply | Threaded
Open this post in threaded view
|

[Bug 3443] [IcedTea7] OpenJDK on aarch64 has fatal error "Field too big for insn"

bugzilla-daemon
In reply to this post by bugzilla-daemon
[hidden email] changed bug 3443
What Removed Added
URL   https://bugs.openjdk.java.net/browse/JDK-8145438


You are receiving this mail because:
  • You are on the CC list for the bug.
Reply | Threaded
Open this post in threaded view
|

Re: [Bug 3443] New: [IcedTea7] OpenJDK on aarch64 has fatal error "Field too big for insn"

Andrew Haley
In reply to this post by bugzilla-daemon
On 21/08/17 19:03, [hidden email] wrote:
> "8144028: Use AArch64 bit-test instructions in C2" had to be backported for one
> of the security fixes.

8144028 is just a performance fix.  How is it needed by one of the security
fixes?  Which one?

--
Andrew Haley
Java Platform Lead Engineer
Red Hat UK Ltd. <https://www.redhat.com>
EAC8 43EB D3EF DB98 CC77 2FAD A5CD 6035 332F A671
Reply | Threaded
Open this post in threaded view
|

[Bug 3443] [IcedTea7] OpenJDK on aarch64 has fatal error "Field too big for insn"

bugzilla-daemon
In reply to this post by bugzilla-daemon

Comment # 1 on bug 3443 from [hidden email]
details:
http://icedtea.classpath.org//hg/icedtea7-forest/hotspot?cmd=changeset;node=46d12689c870
author: aph
date: Tue Aug 22 18:17:20 2017 +0100

    8145438, PR3443, RH1482244: Guarantee failures since 8144028: Use AArch64
bit-test instructions in C2
    Summary: Implement short and long versions of bit test instructions.
    Reviewed-by: kvn


You are receiving this mail because:
  • You are on the CC list for the bug.
Reply | Threaded
Open this post in threaded view
|

[Bug 3443] [IcedTea7] OpenJDK on aarch64 has fatal error "Field too big for insn"

bugzilla-daemon
In reply to this post by bugzilla-daemon

Comment # 2 on bug 3443 from [hidden email]
http://icedtea.classpath.org/hg/icedtea7/rev/e9659e584b7c

Bump to icedtea-2.7.0pre14.

Upstream changes:
- Bump to icedtea-2.7.0pre14
- S7117357: Warnings in sun.instrument, tools and other sun.* classes
- S7117570: Warnings in sun.mangement.* and its subpackages
- S7143230: fix warnings in java.util.jar, sun.tools.jar, zipfs demo, etc.
- S8022440: suppress deprecation warnings in sun.rmi
- S8024069: replace_in_map() should operate on parent maps
- S8026796: Make replace_in_map() on parent maps generic
- S8030787: [Parfait] JNI-related warnings from b119 for
jdk/src/share/native/sun/awt/image
- S8030875: Macros for checking and returning on exceptions
- S8031737: CHECK_NULL and CHECK_EXCEPTION macros cleanup
- S8034912: backport of 8031737 to jdk8u breaks linux buld.
- S8035629: [parfait] JNI exc pending in
jdk/src/windows/native/sun/windows/ShellFolder2.cpp
- S8037287: Windows build failed after JDK-8030787
- S8048703: ReplacedNodes dumps it's content to tty
- S8080492: [Parfait] Uninitialised variable in
jdk/src/java/desktop/windows/native/libawt/
- S8139870: sun.management.LazyCompositeData.isTypeMatched() fails for
composite types with items of ArrayType
- S8143377: Test PKCS8Test.java fails
- S8144028, PR3431: Use AArch64 bit-test instructions in C2
- S8145438, PR3443, RH1482244: Guarantee failures since 8144028: Use AArch64
bit-test instructions in C2
- S8149450: LdapCtx.processReturnCode() throwing Null Pointer Exception
- S8152537, PR3431: aarch64: Make use of CBZ and CBNZ when comparing unsigned
values with zero.
- S8155690: Update libPNG library to the latest up-to-date
- S8156804: Better constraint checking
- S8162461: Hang due to JNI up-call made whilst holding JNI critical lock
- S8163958: Improved garbage collection
- S8165231: java.nio.Bits.unaligned() doesn't return true on ppc
- S8165367: Additional tests for JEP 288: Disable SHA-1 Certificates
- S8167228: Update to libpng 1.6.28
- S8169209: Improved image post-processing steps
- S8169392: Additional jar validation steps
- S8170966: Right parenthesis issue
- S8172204: Better Thread Pool execution
- S8172461: Service Registration Lifecycle
- S8172465: Better handling of channel groups
- S8172469: Transform Transformer Exceptions
- S8173145: Menu is activated after using mnemonic Alt/Key combination
- S8173286: Better reading of text catalogs
- S8173697: Less Active Activations
- S8173770: Image conversion improvements
- S8174098: Better image fetching
- S8174105: Better naming attribution
- S8174113: Better sourcing of code
- S8174164: SafePointNode::_replaced_nodes breaks with irreducible loops
- S8174729: Race Condition in java.lang.reflect.WeakCache
- S8174770: Check registry registration location
- S8174873: Improved certificate procesing
- S8175097: [TESTBUG] 8174164 fix missed the test
- S8175106: Higher quality DSA operations
- S8175110: Higher quality ECDSA operations
- S8175251: Failed to load RSA private key from pkcs12
- S8176055: JMX diagnostic improvements
- S8176067: Proper directory lookup processing
- S8176731: JCK tests in api/javax_xml/transform/ spec conformance started
failing after 8172469
- S8176760: Better handling of PKCS8 material
- S8176769: Remove accidental spec change in jdk8u
- S8177449: (tz) Support tzdata2017b
- S8178135: Additional elliptic curve support
- S8178996: [macos] JComboBox doesn't display popup in mixed JavaFX Swing
Application on 8u131 and Mac OS 10.12
- S8179014: JFileChooser with Windows look and feel crashes on win 10
- S8179887: Build failure with glibc >= 2.24: error: 'int readdir_r(DIR*,
dirent*, dirent**)' is deprecated
- S8180582: The bind to rmiregistry is rejected by registryFilter even though
registryFilter is set
- S8181420: PPC: Image conversion improvements
- S8181591: 8u141 L10n resource file update
- S8182054: Improve wsdl support
- S8183551, PR3431: AArch64: Image conversion improvements
- S8184119: Incorrect return processing for the LF editor of
MethodHandles.permuteArguments
- S8184993: Jar file verification failing with SecurityException: digest
missing xxx
- S8185501: Missing import in JAXP code
- S8185502: No overflow operator on OpenJDK 7
- S8185716: OpenJDK 7 PPC64 port uses a different ins_encode format in ppc.ad

ChangeLog:

2017-09-08 Andrew John Hughes <[hidden email]>

Bump to icedtea-2.7.0pre14.
* Makefile.am:
(JDK_UPDATE_VERSION): Bump to 151.
(BUILD_VERSION): Reset to b01.
(CORBA_CHANGESET): Update to icedtea-2.7.0pre14.
(JAXP_CHANGESET): Likewise.
(JAXWS_CHANGESET): Likewise.
(JDK_CHANGESET): Likewise.
(LANGTOOLS_CHANGESET): Likewise.
(OPENJDK_CHANGESET): Likewise.
(CORBA_SHA256SUM): Likewise.
(JAXP_SHA256SUM): Likewise.
(JAXWS_SHA256SUM): Likewise.
(JDK_SHA256SUM): Likewise.
(LANGTOOLS_SHA256SUM): Likewise.
(OPENJDK_SHA256SUM): Likewise.
* NEWS: Updated.
* configure.ac: Bump to 2.7.0pre14.
* hotspot.map.in: Update to icedtea-2.7.0pre14.
* patches/boot/ecj-diamond.patch:
Regenerated. Add cases in
java.util.jar.Attributes, java.util.jar.JarVerifier,
javax.imageio.spi.ServiceRegistry,
sun.management.ConnectorAddressLink,
sun.management.counter.perf.PerfInstrumentation,
sun.management.Flag,
sun.management.GarbageCollectionNotifInfoCompositeData,
sun.management.HotspotCompilation,
sun.management.HotspotDiagnostic,
sun.management.HotspotThread,
sun.management.jmxremote.ConnectorBootstrap,
sun.management.ManagementFactoryHelper,
sun.management.MappedMXBeanType,
sun.management.NotificationEmitterSupport,
sun.management.RuntimeImpl,
sun.net.RegisterDomain,
sun.tools.jar.CommandLine,
sun.tools.jar.Manifest and sun.tools.jar.SignatureFile
* patches/boot/ecj-multicatch.patch:
Add cases in
com.sun.imageio.plugins.jpeg.JPEGImageReader,
sun.management.Agent,
sun.management.GarbageCollectionNotifInfoCompositeData,
sun.management.GcInfoCompositeData,
sun.rmi.server.UnicastRef, sun.rmi.server.UnicastServerRef,
javax.xml.transform.TransformerException,
com.sun.xml.internal.ws.util.xml.XmlUtil,
sun.rmi.registry.RegistryImpl_Skel and
sun.rmi.transport.DGCImpl_Stub
* patches/boot/ecj-stringswitch.patch:
Regenerated.
* patches/boot/ecj-trywithresources.patch:
Add cases in
sun.management.jmxremote.ConnectorBootstrap.


You are receiving this mail because:
  • You are on the CC list for the bug.
Reply | Threaded
Open this post in threaded view
|

[Bug 3443] [IcedTea7] OpenJDK on aarch64 has fatal error "Field too big for insn"

bugzilla-daemon
In reply to this post by bugzilla-daemon

Comment # 3 on bug 3443 from [hidden email]
details:
http://icedtea.classpath.org//hg/release/icedtea7-2.6?cmd=changeset;node=83d470e904b6
author: Andrew John Hughes <[hidden email]>
date: Sun Nov 19 16:45:14 2017 +0000

    Bump to 2.6.12pre01.

    Upstream changes:
      - Bump to icedtea-2.6.12pre01
      - PR3480, RH1486025: ECC and NSS JVM crash
      - S8075484, PR3474, RH1490713: SocketInputStream.socketRead0 can hang
even with soTimeout set
      - S8138745, PR3465, RH1484399: Implement ExitOnOutOfMemory and
CrashOnOutOfMemory in HotSpot
      - S8145438, PR3443, RH1482244: Guarantee failures since 8144028: Use
AArch64 bit-test instructions in C2
      - S8185164, PR3433: GetOwnedMonitorInfo() returns incorrect owned monitor
      - S8188030, PR3460, RH1484079: AWT java apps fail to start when some
minimal fonts are present

    ChangeLog:

    2017-11-18  Andrew John Hughes  <[hidden email]>

        Bump to 2.6.12pre01.
        * Makefile.am:
        (CORBA_CHANGESET): Update to icedtea-2.6.12pre01.
        (JAXP_CHANGESET): Likewise.
        (JAXWS_CHANGESET): Likewise.
        (JDK_CHANGESET): Likewise.
        (LANGTOOLS_CHANGESET): Likewise.
        (OPENJDK_CHANGESET): Likewise.
        (CORBA_SHA256SUM): Likewise.
        (JAXP_SHA256SUM): Likewise.
        (JAXWS_SHA256SUM): Likewise.
        (JDK_SHA256SUM): Likewise.
        (LANGTOOLS_SHA256SUM): Likewise.
        (OPENJDK_SHA256SUM): Likewise.
        * NEWS: Updated.
        * configure.ac: Bump to 2.6.12pre01.
        * hotspot.map.in: Update to icedtea-2.6.12pre01.


You are receiving this mail because:
  • You are on the CC list for the bug.
Reply | Threaded
Open this post in threaded view
|

[Bug 3443] [IcedTea7] OpenJDK on aarch64 has fatal error "Field too big for insn"

bugzilla-daemon
In reply to this post by bugzilla-daemon

Comment # 4 on bug 3443 from [hidden email]
details:
http://icedtea.classpath.org//hg/release/icedtea7-forest-2.6/hotspot?cmd=changeset;node=d48ba758f16d
author: aph
date: Tue Aug 22 18:17:20 2017 +0100

    8145438, PR3443, RH1482244: Guarantee failures since 8144028: Use AArch64
bit-test instructions in C2
    Summary: Implement short and long versions of bit test instructions.
    Reviewed-by: kvn


You are receiving this mail because:
  • You are on the CC list for the bug.
Reply | Threaded
Open this post in threaded view
|

[Bug 3443] [IcedTea7] OpenJDK on aarch64 has fatal error "Field too big for insn"

bugzilla-daemon
In reply to this post by bugzilla-daemon
[hidden email] changed bug 3443
What Removed Added
Status ASSIGNED RESOLVED
Resolution --- FIXED

Comment # 5 on bug 3443 from [hidden email]
Fixed in icedtea-2.7.0pre14 and icedtea-2.6.12pre01


You are receiving this mail because:
  • You are on the CC list for the bug.