| Patch | Description | Author | Forwarded | Bugs | Origin | Last update |
|---|---|---|---|---|---|---|
| jdk-8307977-proposed.patch | [PATCH] 8226919: Fix dynamic attach in Linux for non-container environments | =?UTF-8?q?Sebastian=20L=C3=B6vdahl?= <sebastian.lovdahl@hibox.tv> | yes | debian upstream | 2023-04-18 | |
| use_clock64_fns_armhf.patch | Dynamically link with *64 versions of clock_* functions On 32-bit archs that implement a 64-bit time_t, clock_* functions linked via dlsym() should be the *64 versions of them. For example, on such archs, we should try to link with __clock_gettime64 first and fallback to linking with clock_gettime if the *64 version is not available. |
Pushkar Kulkarni <pushkar.kulkarni@canonical.com> | no | |||
| jdk-8334502.patch | use int instead of time_t to print timezone info printf() call uses %02d format specifier that requires an int parameter. diff --git a/src/hotspot/share/runtime/os.cpp b/src/hotspot/share/runtime/os.cpp index bdf93e1d3b4..d928dc55dce 100644 |
Vladimir Petko <vladimir.petko@canonical.com> | yes | upstream | 2024-06-19 | |
| system-pcsclite.diff | no | |||||
| hotspot-mips-align.diff | no | |||||
| icc_loading_with_symlink.diff | no | |||||
| icedtea-override-redirect-compiz.diff | no | |||||
| libpcsclite-dlopen.diff | no | |||||
| jexec.diff | no | |||||
| default-jvm-cfg.diff | no | |||||
| adlc-parser.diff | # DP: fixes an uninitialized memory issue in adlc | no | ||||
| multiple-pkcs11-library-init.diff | # HG changeset patch # User andrew # Date 1352129932 0 # Node ID e9c857dcb964dbfa5eef3a3590244cb4d999cf7a # Parent 1406789608b76d0906881979335d685855f44190 Allow multiple PKCS11 library initialisation to be a non-critical error. |
no | ||||
| s390x-thread-stack-size.diff | no | |||||
| s390x-opt.diff | no | |||||
| jdk-getAccessibleValue.diff | no | |||||
| jdk-i18n-pt_BR.diff | no | |||||
| disable-doclint-by-default.diff | no | |||||
| 8199220.diff | # HG changeset patch # User enevill # Date 1520877434 0 # Mon Mar 12 17:57:14 2018 +0000 # Node ID 81265b23c488c7fd03ebaee04d1802fc0ada737e # Parent ecd91135d645bed6620abbd76ce30a906b6d7d44 8199220: Zero build broken after 8195103, 8191102 and 8189871 |
no | ||||
| machine-flag.diff | no | |||||
| zero-x32.diff | no | |||||
| hotspot-disable-exec-shield-workaround.diff | no | |||||
| atk-wrapper-security.diff | no | |||||
| dnd-files.diff | no | |||||
| hotspot-libpath.diff | no | |||||
| reproducible-properties-timestamp.diff | Makes the timestamp in the properties files header reproducible when SOURCE_DATE_EPOCH is specified | Emmanuel Bourg <ebourg@apache.org> | no | |||
| reproducible-javadoc-timestamp.diff | Makes the timestamp in the javadoc files reproducible when SOURCE_DATE_EPOCH is specified | Emmanuel Bourg <ebourg@apache.org> | no | |||
| Don-t-optimize-fdlibm-fork-for-Zero-on-linux-sparc-Z.patch | [PATCH] Don't optimize fdlibm-fork for Zero on linux-sparc (Zero) | John Paul Adrian Glaubitz <glaubitz@physik.fu-berlin.de> | no | 2018-12-21 | ||
| keep-gtk2-as-default.patch | Revert upstream default to GTK2 OpenJDK 11 uses GTK3 by default for the GTK Look and Feel but it still has too many pending issues and artifacts and is not on par with GTK2 support. This patch is the reverse of the upstream commit which ensures that GTK2 is tried before GTK3. When GTK3 becomes better supported we should drop this and update the dlopen_jre_depends variable in debian/rules accordingly. https://bugs.java.com/bugdatabase/view_bug.do?bug_id=JDK-8198654 |
not-needed | upstream | upstream, http://hg.openjdk.java.net/jdk-updates/jdk11u/rev/3e3696a308e1 | 2019-03-27 | |
| riscv64.diff | # HG changeset patch # User enevill # Date 1521985117 -3600 # Sun Mar 25 14:38:37 2018 +0100 # Node ID 7d1ae835ef5194e7ae5b56b51f944029fe602d11 # Parent 1b1de4b263c81853719f6bb0385fe23bc4e35f6c 8199138: Add RISC-V support to Zero |
no | ||||
| reproducible-character-data.diff | Makes the generated character data classes reproducible (removes a timestamp and trims a build path captured in the comments) | Emmanuel Bourg <ebourg@apache.org> | no | |||
| reproducible-module-info.diff | Makes the generated module-info.java files reproducible (removes a captured build path) | Emmanuel Bourg <ebourg@apache.org> | no | |||
| reproducible-copyright-headers.diff | Makes the generated copyright headers reproducible | Emmanuel Bourg <ebourg@apache.org> | no | |||
| reproducible-build-user.diff | Makes the build user invariant to improve the reproducibility (it appears in the interval VM version in libjvm.so) | Emmanuel Bourg <ebourg@apache.org> | no | |||
| reproducible-build-jmod.diff | jlink: Hash of module differs to expected hash recorded in java.base The cause is the use of dh_strip_nondeterminism late in the build process. This reorganises the jmod files, which in turn changes their SHA256 checksums. This would not be a problem, except that the checksums are saved in java.base.jmod *before* the use of dh_strip_nondeterminism. Performing this stripping immediately after each jmod file is created results in the checksums being consistent throughout. |
Julian Gilbey <jdg@debian.org> | not-needed | debian | ||
| hotspot-test-fix-npe-on-missing-file.patch | Test fails with NPE on dangling link We link libatk on OpenJDK's lib/ but it is a dangling link unless libatk is installed. Since we don't build depend on it the ModeJDKTest needs to be fixed to recurse only when the source exists. |
Tiago Stürmer Daitx <tiago.daitx@ubuntu.com> | no | Ubuntu | 2020-11-09 | |
| disable-thumb-assertion.patch | Unable to obtain frame pointer on Thumb architecture ARM Thumb has os::current_frame() disabled. Disable associated test. | Vladimir Petko <vladimir.petko@canonical.com> | yes | upstream | 2023-03-30 | |
| update-assertion-for-armhf.patch | armhf does not support frame unwinding, simplify assertion os::current_frame() is stubbed out on armhf and this test fails. Simplify assertion as there is no "from xxxx" output on armhf. |
Vladimir Petko <vladimir.petko@canonical.com> | yes | upstream | ||
| log-generated-classes-test.patch | Skip test when getFileStore() fails Due to https://bugs.openjdk.org/browse/JDK-8166162 getFileStore() fails with exception failing the test suite. Skip test when getFileStore() throws. |
Vladimir Petko <vladimir.petko@canonical.com> | not-needed | upstream | 2023-03-27 | |
| update-permission-test.patch | Add missing permissions for TestNG 7.7 TestNG has added more configuration properties, allow them in the minimal policy | Vladimir Petko <vladimir.petko@canonical.com> | no | 2023-03-24 | ||
| ldap-timeout-test-use-ip.patch | Replace hostname with a non-existant ip address In Ubuntu build environment we do not have DNS resolver, causing an unexpected test failure. Use a non-existant IP address instead of hostname. |
Vladimir Petko <vladimir.petko@canonical.com> | no | 2023-03-27 | ||
| test-use-ip-address.patch | Use IP address in tests Ubuntu build machines are not set up for dns. Use IP addresses to get correct test outcomes. |
Vladimir Petko <vladimir.petko@canonical.com> | not-needed | 2023-03-27 | ||
| nss-limited-ecc-tests.patch | no | |||||
| loong64-autoconf-config.diff | no | |||||
| jdk-8336529-proposed.patch | (fs) UnixFileAttributeViews setTimes() failing on armhf, Ubuntu noble time_t transition in Debian/Ubuntu left 32 bit time_t symbols in glibc. Looking up 'futimens' via dlsym returns 32 bit version of the function. This is causing failure to set last modified time (e.g. instead of year 2017 we get 1976 in the test). Using the function directly correctly calls 64 bit versions. When we lookup functions with time_t arguments through dlsym() calls we should use 64 bit versions. |
Vladimir Petko <vladimir.petko@canonical.com> | yes | upstream | upstream, https://github.com/openjdk/jdk/pull/20208 | 2024-07-17 |