Repository navigation
docs: state the real glibc floor of the s390x jar; follow-ups in TODO - #568
Merged
Merged
Conversation
README said only "A JDK 8+ JVM" for cpu-linux-s390x, and CHANGELOG item 5 promised glibc 2.28 for all of Linux. Both were wrong for s390x. The jar references GLIBC_2.38 and GLIBCXX_3.4.32, because it is cross-compiled on ubuntu-latest (GCC 13.3, glibc 2.39 sysroot) instead of in a manylinux image. The x86-64 and aarch64 jars stay at 2.27 / 3.4.25. Measured 2026-10-25 on the CI artifacts of Publish runs 40217818548 and 40227501358, loading as a consumer does (llama + cpu-linux-s390x on the classpath, LlamaLoader.initialize(), s390x emulated by qemu in Docker): - fails on RHEL 8.10 (GLIBC_2.38 not found, libstdc++ too), RHEL 9 and Ubuntu 22.04; - loads on Ubuntu 24.04; - ldd is clean on SLES 15 SP6, Debian 13 and Ubuntu 25.04; - the x86-64 and aarch64 jars load on RHEL 8. No build change before the release. TODO.md records the follow-ups: - pin the runner to ubuntu-24.04; - add a GLIBC_CEILING entry at (2, 38); - reach the 2.28 floor through manylinux_2_28_s390x or a glibc 2.28 sysroot. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01AJGh28Gu8pLWYbTCqG1LZf
bernardladenthin
had a problem deploying
to
maven-central
October 25, 2026 20:48 — with
GitHub Actions
Failure
bernardladenthin
had a problem deploying
to
startgate
October 25, 2026 20:48 — with
GitHub Actions
Error
bernardladenthin
had a problem deploying
to
maven-central
October 25, 2026 20:48 — with
GitHub Actions
Failure
|
Caution Review failedThe pull request is closed. Walkthrough
Merge Risk:
|
|
This branch had an error being deployed
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.



Documentation only; the build and the shipped artifacts do not change.
What was wrong
cpu-linux-s390xsaid only "A JDK 8+ JVM".What the CI artifacts actually require
Measured on the artifacts of Publish runs 40217818548 and 40227501358:
cpu-linux-x86-64cpu-linux-aarch64cpu-linux-s390xLoad test
The test loads the way a consumer does:
llamaandcpu-linux-s390xfromllama-jarson the classpath, thenLlamaLoader.initialize(). It ran in Docker, with s390x emulated by qemu.UnsatisfiedLinkError … GLIBC_2.38 not found. libstdc++ is missingGLIBCXX_3.4.26–3.4.32there as well.lddfinds no missing versions on SLES 15 SP6, Debian 13 and Ubuntu 25.04.Why it is a side effect, not a decision
GLIBC_CEILINGlists only the manylinux directories. Every other Linux build "inherits the floor of the ubuntu runner", and CLAUDE.md justifies that for GPU modules only.ctestruns under qemu against the same glibc 2.39 sysroot the build used.Changes
ubuntu-24.04;"Linux/s390x/cpu": (2, 38)toGLIBC_CEILING;manylinux_2_28_s390xor a glibc 2.28 sysroot, with the gcc-toolset treatment for libstdc++ and acceptance on cfarm191/192.check-natives.pyreports 0 disagreements.🤖 Generated with Claude Code
https://claude.ai/code/session_01AJGh28Gu8pLWYbTCqG1LZf
Summary by CodeRabbit