Integrated: 8282086: Update jib profile to not set build to 0
On Fri, 18 Feb 2022 08:46:31 GMT, Magnus Ihse Bursie wrote: > The jib profile definition does not play well with > [JDK-8274980](https://bugs.openjdk.java.net/browse/JDK-8274980). When using 0 > as a marker for "no build number", we should not pass that on to configure, > since it will trigger a warning. This pull request has now been integrated. Changeset: cf6984dd Author:Magnus Ihse Bursie URL: https://git.openjdk.java.net/jdk/commit/cf6984ddaa5668e78d590c8ad1f2aec0632f0b28 Stats: 4 lines in 1 file changed: 3 ins; 0 del; 1 mod 8282086: Update jib profile to not set build to 0 Reviewed-by: erikj - PR: https://git.openjdk.java.net/jdk/pull/7527
Re: RFR: 8253757: Add LLVM-based backend for hsdis
On Fri, 18 Feb 2022 11:44:04 GMT, Magnus Ihse Bursie wrote: > Third time's a charm! After the two previous closed PRs for this issue, I > think this functionality is finally ready to enter mainline. :) > > This code is at it's core the same as the previous PR. The main C++ hsdis > implementation is still the one @luhenry wrote, with some changes. As in the > previous PR, I extracted the LLVM-specific part into a separate file. In > addition to the last PR, I have also fixed a warning, and added a patch > inspired by @nick-arm for getting past instructions unknown to LLVM. > > Thanks to the prototype written by @JornVernee (and his graciously providing > me with a working version of LLVM build for Windows), this PR now has full > support for LLVM on Windows (as well as Linux and macOS). > > Finally, I have cleaned up the integration in autoconf and Hsdis.gmk, and > written a thorough guide in the README on how to build witth LLVM, on Windows > and on saner platforms. :) > > I'm pretty sure this means that all comments and criticism in the previous PR > has been addressed. > > Huge thanks to everyone who has helped me with getting this PR in place. I > have a hard time remember a feature that has been so tricky to get in place, > for something to seemingly simple... Marked as reviewed by luhenry (Author). Thanks again for pushing that forward! It's always good to have an alternative especially for porting to new platforms and architectures. - PR: https://git.openjdk.java.net/jdk/pull/7531
Re: RFR: 8253757: Add LLVM-based backend for hsdis
On Fri, 18 Feb 2022 11:44:04 GMT, Magnus Ihse Bursie wrote: > Third time's a charm! After the two previous closed PRs for this issue, I > think this functionality is finally ready to enter mainline. :) > > This code is at it's core the same as the previous PR. The main C++ hsdis > implementation is still the one @luhenry wrote, with some changes. As in the > previous PR, I extracted the LLVM-specific part into a separate file. In > addition to the last PR, I have also fixed a warning, and added a patch > inspired by @nick-arm for getting past instructions unknown to LLVM. > > Thanks to the prototype written by @JornVernee (and his graciously providing > me with a working version of LLVM build for Windows), this PR now has full > support for LLVM on Windows (as well as Linux and macOS). > > Finally, I have cleaned up the integration in autoconf and Hsdis.gmk, and > written a thorough guide in the README on how to build witth LLVM, on Windows > and on saner platforms. :) > > I'm pretty sure this means that all comments and criticism in the previous PR > has been addressed. > > Huge thanks to everyone who has helped me with getting this PR in place. I > have a hard time remember a feature that has been so tricky to get in place, > for something to seemingly simple... Build changes look good. I can't comment on the hsdis implementation. - Marked as reviewed by erikj (Reviewer). PR: https://git.openjdk.java.net/jdk/pull/7531
Re: RFR: 8282086: Update jib profile to not set build to 0
On Fri, 18 Feb 2022 08:46:31 GMT, Magnus Ihse Bursie wrote: > The jib profile definition does not play well with > [JDK-8274980](https://bugs.openjdk.java.net/browse/JDK-8274980). When using 0 > as a marker for "no build number", we should not pass that on to configure, > since it will trigger a warning. Marked as reviewed by erikj (Reviewer). - PR: https://git.openjdk.java.net/jdk/pull/7527
RFR: 8253757: Add LLVM-based backend for hsdis
Third time's a charm! After the two previous closed PRs for this issue, I think this functionality is finally ready to enter mainline. :) This code is at it's core the same as the previous PR. The main C++ hsdis implementation is still the one @luhenry wrote, with some changes. As in the previous PR, I extracted the LLVM-specific part into a separate file. In addition to the last PR, I have also fixed a warning, and added a patch inspired by @nick-arm for getting past instructions unknown to LLVM. Thanks to the prototype written by @JornVernee (and his graciously providing me with a working version of LLVM build for Windows), this PR now has full support for LLVM on Windows (as well as Linux and macOS). Finally, I have cleaned up the integration in autoconf and Hsdis.gmk, and written a thorough guide in the README on how to build witth LLVM, on Windows and on saner platforms. :) I'm pretty sure this means that all comments and criticism in the previous PR has been addressed. Huge thanks to everyone who has helped me with getting this PR in place. I have a hard time remember a feature that has been so tricky to get in place, for something to seemingly simple... - Commit messages: - Update description on runtime requirements for LLVM - Fix warning on Windows (as opposed to hiding it) - Add LLVM backend to hsdis. Portions of this patch contributed by luhenry, jvernee and ngasson. Changes: https://git.openjdk.java.net/jdk/pull/7531/files Webrev: https://webrevs.openjdk.java.net/?repo=jdk=7531=00 Issue: https://bugs.openjdk.java.net/browse/JDK-8253757 Stats: 495 lines in 4 files changed: 490 ins; 0 del; 5 mod Patch: https://git.openjdk.java.net/jdk/pull/7531.diff Fetch: git fetch https://git.openjdk.java.net/jdk pull/7531/head:pull/7531 PR: https://git.openjdk.java.net/jdk/pull/7531
Re: macro expansion producing 'defined' has undefined behavior
Thanks a lot. It worked when I upgraded the openjdk version to 12. > 2022年2月18日 10:58，David Holmes 写道： > > Hi Simmias, > > On 14/02/2022 1:24 pm, 殷玉婷 wrote: >> Hi, >> When I ran >> make images in my local computer to compile the openjdk9u by guideline >> <>, I encountered the following problems: > > 9u is not expected to be built by recent Xcode, it was a much older release > (not sure if anyone is actually updating it at all): > > https://wiki.openjdk.java.net/display/Build/Supported+Build+Platforms > > You either need to downgrade your build tools, or else upgrade the OpenJDK > version you want to build. > > Cheers, > David > >> Compiling 61 files for BUILD_INTERIM_jdk.jdeps >> Compiling 457 files for BUILD_INTERIM_jdk.javadoc >> In file included from >> /Users/simmias/java/github/openjdk/jdk9u-master/hotspot/test/native/gc/g1/test_freeRegionList.cpp:25: >> In file included from >> /Users/simmias/java/github/openjdk/jdk9u-master/hotspot/src/share/vm/gc/g1/g1CollectedHeap.inline.hpp:28: >> In file included from >> /Users/simmias/java/github/openjdk/jdk9u-master/hotspot/src/share/vm/gc/g1/g1CollectedHeap.hpp:33: >> In file included from >> /Users/simmias/java/github/openjdk/jdk9u-master/hotspot/src/share/vm/gc/g1/g1ConcurrentMark.hpp:31: >> /Users/simmias/java/github/openjdk/jdk9u-master/hotspot/src/share/vm/gc/g1/heapRegionSet.hpp:126:5: >> error: macro expansion producing 'defined' has undefined behavior >> [-Werror,-Wexpansion-to-defined] >> #if HEAP_REGION_SET_FORCE_VERIFY >> ^ >> /Users/simmias/java/github/openjdk/jdk9u-master/hotspot/src/share/vm/gc/g1/heapRegionSet.hpp:53:38: >> note: expanded from macro 'HEAP_REGION_SET_FORCE_VERIFY' >> #define HEAP_REGION_SET_FORCE_VERIFY defined(ASSERT) >> ^ >> In file included from >> /Users/simmias/java/github/openjdk/jdk9u-master/hotspot/test/native/gc/g1/test_freeRegionList.cpp:25: >> In file included from >> /Users/simmias/java/github/openjdk/jdk9u-master/hotspot/src/share/vm/gc/g1/g1CollectedHeap.inline.hpp:28: >> In file included from >> /Users/simmias/java/github/openjdk/jdk9u-master/hotspot/src/share/vm/gc/g1/g1CollectedHeap.hpp:37: >> /Users/simmias/java/github/openjdk/jdk9u-master/hotspot/src/share/vm/gc/g1/g1HeapVerifier.hpp:67:5: >> error: macro expansion producing 'defined' has undefined behavior >> [-Werror,-Wexpansion-to-defined] >> #if HEAP_REGION_SET_FORCE_VERIFY >> ^ >> /Users/simmias/java/github/openjdk/jdk9u-master/hotspot/src/share/vm/gc/g1/heapRegionSet.hpp:53:38: >> note: expanded from macro 'HEAP_REGION_SET_FORCE_VERIFY' >> #define HEAP_REGION_SET_FORCE_VERIFY defined(ASSERT) >> ^ >> 2 errors generated. >> make[3]: *** >> [/Users/simmias/java/github/openjdk/jdk9u-master/build/macosx-x86_64-normal-server-release/hotspot/variant-server/libjvm/gtest/objs/test_freeRegionList.o] >> Error 1 >> make[3]: *** Waiting for unfinished jobs >> make[2]: *** [hotspot-server-libs] Error 2 >> make[2]: *** Waiting for unfinished jobs >> ?: API? >> ?: ??, ??? -Xlint:deprecation ? >> ERROR: Build failed for target 'images' in configuration >> 'macosx-x86_64-normal-server-release' (exit code 2) >> === Output from failing command(s) repeated here === >> * For target hotspot_variant-server_libjvm_gtest_objs_test_freeRegionList.o: >> In file included from >> /Users/simmias/java/github/openjdk/jdk9u-master/hotspot/test/native/gc/g1/test_freeRegionList.cpp:25: >> In file included from >> /Users/simmias/java/github/openjdk/jdk9u-master/hotspot/src/share/vm/gc/g1/g1CollectedHeap.inline.hpp:28: >> In file included from >> /Users/simmias/java/github/openjdk/jdk9u-master/hotspot/src/share/vm/gc/g1/g1CollectedHeap.hpp:33: >> In file included from >> /Users/simmias/java/github/openjdk/jdk9u-master/hotspot/src/share/vm/gc/g1/g1ConcurrentMark.hpp:31: >> /Users/simmias/java/github/openjdk/jdk9u-master/hotspot/src/share/vm/gc/g1/heapRegionSet.hpp:126:5: >> error: macro expansion producing 'defined' has undefined behavior >> [-Werror,-Wexpansion-to-defined] >> #if HEAP_REGION_SET_FORCE_VERIFY >> ^ >> /Users/simmias/java/github/openjdk/jdk9u-master/hotspot/src/share/vm/gc/g1/heapRegionSet.hpp:53:38: >> note: expanded from macro 'HEAP_REGION_SET_FORCE_VERIFY' >> #define HEAP_REGION_SET_FORCE_VERIFY defined(ASSERT) >> ^ >> In file included from >> /Users/simmias/java/github/openjdk/jdk9u-master/hotspot/test/native/gc/g1/test_freeRegionList.cpp:25: >> In file included from >> /Users/simmias/java/github/openjdk/jdk9u-master/hotspot/src/share/vm/gc/g1/g1CollectedHeap.inline.hpp:28: >>... (rest of output omitted) >> * All command lines available in >> /Users/simmias/java/github/openjdk/jdk9u-master/build/macosx-x86_64-normal-server-release/make-support/failure-logs. >> === End of repeated output === >> No indication of failed target found. >> Hint: Try searching the build log for ']
RFR: 8282086: Update jib profile to not set build to 0
The jib profile definition does not play well with [JDK-8274980](https://bugs.openjdk.java.net/browse/JDK-8274980). When using 0 as a marker for "no build number", we should not pass that on to configure, since it will trigger a warning. - Commit messages: - 8282086: Update jib profile to not set build to 0 Changes: https://git.openjdk.java.net/jdk/pull/7527/files Webrev: https://webrevs.openjdk.java.net/?repo=jdk=7527=00 Issue: https://bugs.openjdk.java.net/browse/JDK-8282086 Stats: 4 lines in 1 file changed: 3 ins; 0 del; 1 mod Patch: https://git.openjdk.java.net/jdk/pull/7527.diff Fetch: git fetch https://git.openjdk.java.net/jdk pull/7527/head:pull/7527 PR: https://git.openjdk.java.net/jdk/pull/7527
