Add alma10 sysroot #92

h-vetinari · 2025-09-17T16:05:33Z

https://repo.almalinux.org/vault/10.0/BaseOS/ exists but is still empty. It should be filled once alma 10.1 is out, which might happen in November. In the meantime, use packages from https://repo.almalinux.org/almalinux. Later, we should simply be able to revert 0de3517.

Fixes #91

…5.09.17.06.41.22 Other tools: - conda-build 25.7.0 - rattler-build 0.47.0 - rattler-build-conda-compat 1.4.6

conda-forge-admin · 2025-09-17T16:07:18Z

Hi! This is the friendly automated conda-forge-linting service.

I just wanted to let you know that I linted all conda-recipes in your PR (recipe/meta.yaml) and found it was in an excellent condition.

I do have some suggestions for making it better though...

For recipe/meta.yaml:

ℹ️ The recipe is not parsable by parser conda-souschef (grayskull). This parser is not currently used by conda-forge, but may be in the future. We are collecting information to see which recipes are compatible with grayskull.
ℹ️ The recipe is not parsable by parser conda-recipe-manager. The recipe can only be automatically migrated to the new v1 format if it is parseable by conda-recipe-manager.
ℹ️ License exception is not an SPDX exception.

Documentation on acceptable licenses can be found here.

_{This message was generated by GitHub Actions workflow run https://github.com/conda-forge/conda-forge-webservices/actions/runs/19406394618. Examine the logs at this URL for more detail.}

…5.11.09.23.10.12 Other tools: - conda-build 25.9.0 - rattler-build 0.49.0 - rattler-build-conda-compat 1.4.9

h-vetinari · 2025-11-15T23:20:39Z

Ah right

Could not solve for environment specs
The following package could not be installed
└─ sysroot_linux-ppc64le =2.39 hf8f1b53_2 is not installable because it requires
   └─ __glibc >=2.39 *, which is missing on the system.

we need the images first

h-vetinari · 2025-11-16T11:03:22Z

OK, the reason for the failures here is that the packaging in alma10 changed; the long story short (AFAICT) is that RHEL 10 seems to have cleaned up some leftovers of the /usr-merge, also on the packaging side.

In alma9, the glibc bits were still packaged in /lib64 (even though /lib64 is already symlinked to /usr/lib64 in the actual image)

$ rpm -qlp ./glibc-2.34-125.el9_5.8.alma.1.x86_64.rpm
[...]
/lib64/ld-linux-x86-64.so.2
/lib64/libBrokenLocale.so.1
/lib64/libSegFault.so
/lib64/libanl.so.1
/lib64/libc.so.6
/lib64/libc_malloc_debug.so.0
/lib64/libdl.so.2
/lib64/libm.so.6
/lib64/libmvec.so.1
/lib64/libnss_compat.so.2
/lib64/libnss_dns.so.2
/lib64/libnss_files.so.2
/lib64/libpthread.so.0
/lib64/libresolv.so.2
/lib64/librt.so.1
/lib64/libthread_db.so.1
/lib64/libutil.so.1
/sbin/ldconfig
/usr/lib64/audit
/usr/lib64/audit/sotruss-lib.so
/usr/lib64/gconv
/usr/lib64/gconv/ANSI_X3.110.so
/usr/lib64/gconv/CP1252.so
/usr/lib64/gconv/ISO8859-1.so
/usr/lib64/gconv/ISO8859-15.so
/usr/lib64/gconv/UNICODE.so
/usr/lib64/gconv/UTF-16.so
/usr/lib64/gconv/UTF-32.so
/usr/lib64/gconv/UTF-7.so
/usr/lib64/gconv/gconv-modules
/usr/lib64/gconv/gconv-modules.cache
/usr/lib64/gconv/gconv-modules.d
/usr/lib64/libmemusage.so
/usr/lib64/libpcprofile.so
/usr/libexec/getconf
/usr/libexec/getconf/POSIX_V6_LP64_OFF64
/usr/libexec/getconf/POSIX_V7_LP64_OFF64
/usr/libexec/getconf/XBS5_LP64_OFF64
/usr/sbin/iconvconfig
[...]

whereas in RHEL 10 everything gets delivered in /usr/lib64:

$ rpm -qlp ./glibc-2.39-46.el10_0.alma.1.x86_64.rpm
[...]
/usr/lib64/audit
/usr/lib64/audit/sotruss-lib.so
/usr/lib64/gconv
/usr/lib64/gconv/CP1252.so
/usr/lib64/gconv/ISO8859-1.so
/usr/lib64/gconv/ISO8859-15.so
/usr/lib64/gconv/UNICODE.so
/usr/lib64/gconv/UTF-16.so
/usr/lib64/gconv/UTF-32.so
/usr/lib64/gconv/UTF-7.so
/usr/lib64/gconv/gconv-modules
/usr/lib64/gconv/gconv-modules.cache
/usr/lib64/gconv/gconv-modules.d
/usr/lib64/ld-linux-x86-64.so.2
/usr/lib64/libBrokenLocale.so.1
/usr/lib64/libanl.so.1
/usr/lib64/libc.so.6
/usr/lib64/libdl.so.2
/usr/lib64/libm.so.6
/usr/lib64/libmvec.so.1
/usr/lib64/libnss_compat.so.2
/usr/lib64/libnss_dns.so.2
/usr/lib64/libnss_files.so.2
/usr/lib64/libpcprofile.so
/usr/lib64/libpthread.so.0
/usr/lib64/libresolv.so.2
/usr/lib64/librt.so.1
/usr/lib64/libthread_db.so.1
/usr/lib64/libutil.so.1
/usr/libexec/getconf
/usr/libexec/getconf/POSIX_V6_LP64_OFF64
/usr/libexec/getconf/POSIX_V7_LP64_OFF64
/usr/libexec/getconf/XBS5_LP64_OFF64
/usr/sbin/iconvconfig
/usr/sbin/ldconfig
[...]

h-vetinari · 2025-11-16T11:19:54Z

Good that I added some extra tests; one of them caught

$SP_DIR/conda_build/build.py:1537: UserWarning: file $PREFIX/x86_64-conda-linux-gnu/sysroot/lib is a symlink with no target
$SP_DIR/conda_build/build.py:1537: UserWarning: file $PREFIX/x86_64-conda-linux-gnu/sysroot/usr/bin/ld.so is a symlink with no target
$SP_DIR/conda_build/build.py:1537: UserWarning: file $PREFIX/x86_64-conda-linux-gnu/sysroot/usr/lib64/libBrokenLocale.so is a symlink with no target
$SP_DIR/conda_build/build.py:1537: UserWarning: file $PREFIX/x86_64-conda-linux-gnu/sysroot/usr/lib64/libanl.so is a symlink with no target
$SP_DIR/conda_build/build.py:1537: UserWarning: file $PREFIX/x86_64-conda-linux-gnu/sysroot/usr/lib64/libmvec.so is a symlink with no target
$SP_DIR/conda_build/build.py:1537: UserWarning: file $PREFIX/x86_64-conda-linux-gnu/sysroot/usr/lib64/libresolv.so is a symlink with no target
$SP_DIR/conda_build/build.py:1537: UserWarning: file $PREFIX/x86_64-conda-linux-gnu/sysroot/usr/lib64/libthread_db.so is a symlink with no target

(or would have caught it, had the test not failed otherwise 😅)

beckermr

We have at least historrically kept branches on the upstream feedstock for each sysroot version. We should do that here and so the PR needs to be redirect to a branch. Feel free to make one.

h-vetinari · 2025-11-16T12:39:38Z

We have at least historically kept branches on the upstream feedstock for each sysroot version.

Look at the branches here

These correspond to

branch	Distro	glibc version
`v2.12`	CentOS 6	2.12
`v2.17`	CentOS 7	2.17
`v2.28`	Alma 8	2.28
`main`	Alma 9	2.34

In other words, already now the newest version doesn't have a special branch. This is regular feedstock behaviour if there are several maintenance branches.

We should do that here and so the PR needs to be redirect to a branch. Feel free to make one.

What we should do here is to merge this PR to main as usual, and once that happens, create a v2.34 branch that matches the state before this PR (i.e. the main branch as of right now). I'm happy to create that branch once this is in.

…5.11.15.23.15.53 Other tools: - conda-build 25.9.0 - rattler-build 0.49.0 - rattler-build-conda-compat 1.4.9

beckermr · 2025-11-16T12:59:27Z

Ack yes you are right @h-vetinari. We should make a branch for alma9 from the current main! Thank you!

incorrect!

h-vetinari · 2025-11-16T13:44:37Z

OK, this should be ready now! PTAL :)

h-vetinari · 2025-11-16T13:47:41Z

Branch v2.34 is there too now: https://github.com/conda-forge/linux-sysroot-feedstock/tree/v2.34

beckermr

Thank you!

beckermr · 2025-11-16T21:15:43Z

recipe/meta.yaml

-    sha256: 50808a6a71fd9a3d5b3db2604f825af8aa6cf0020dd8f1d0edca1b0bc77729f9  # [cross_target_platform == "linux-ppc64le"]
-    sha256: e7a9df0d9eb747b1dacf13636a4defb42123a0aa3d166adff9970736d1be073c  # [cross_target_platform == "linux-s390x"]
-
-  - folder: binary-glibc-headers                                              # [cross_target_platform in ("linux-64", "linux-s390x")]


I'd be nice if they stopped moving stuff around.

beckermr · 2025-11-16T21:18:12Z

recipe/meta.yaml

-        - test -f $PREFIX/{{ target_machine }}-conda-linux-gnu/sysroot/lib/libc.so.6
-        - test -f $PREFIX/{{ target_machine }}-conda-linux-gnu/sysroot/sbin/ldconfig
-        - test -f $PREFIX/{{ target_machine }}-conda-linux-gnu/sysroot/usr/lib/crt1.o
+        # the files below are just a sample of the full content


Thanks for expanding the tests!

h-vetinari · 2025-11-16T21:45:46Z

Thanks for the review!

h-vetinari added 5 commits September 17, 2025 18:00

adapt update.py for alma10

55b6d9d

vault is not populated yet

0de3517

no more glibc-headers in alma10

2700ee4

regenerate meta.yaml

8a4716d

MNT: Re-rendered with conda-smithy 3.52.2 and conda-forge-pinning 202…

7e8c90b

…5.09.17.06.41.22 Other tools: - conda-build 25.7.0 - rattler-build 0.47.0 - rattler-build-conda-compat 1.4.6

h-vetinari requested review from beckermr, isuruf and scopatz as code owners September 17, 2025 16:05

This was referenced Oct 17, 2025

Status Report (2025-10-17) Quansight-Labs/conda-ecosystem-sta-mgmt#45

Closed

Status Report (2025-10-24) Quansight-Labs/conda-ecosystem-sta-mgmt#47

Closed

h-vetinari mentioned this pull request Nov 2, 2025

Status Report (2025-10-31) Quansight-Labs/conda-ecosystem-sta-mgmt#91

Closed

h-vetinari mentioned this pull request Nov 9, 2025

Status Report (2025-11-07) Quansight-Labs/conda-ecosystem-sta-mgmt#94

Closed

MNT: Re-rendered with conda-smithy 3.53.0 and conda-forge-pinning 202…

c5d086e

…5.11.09.23.10.12 Other tools: - conda-build 25.9.0 - rattler-build 0.49.0 - rattler-build-conda-compat 1.4.9

h-vetinari mentioned this pull request Nov 14, 2025

Status Report (2025-11-14) Quansight-Labs/conda-ecosystem-sta-mgmt#97

Closed

This was referenced Nov 15, 2025

add alma10 images conda-forge/docker-images#319

Merged

RISC-V Support? conda-forge/conda-forge.github.io#1744

Open

move missing_dso_whitelist to respective output

c8782e1

h-vetinari force-pushed the alma10 branch 2 times, most recently from 9ebca9b to 75f30eb Compare November 16, 2025 11:08

h-vetinari force-pushed the alma10 branch from fcd7de5 to 2d629d8 Compare November 16, 2025 12:05

beckermr previously requested changes Nov 16, 2025

View reviewed changes

h-vetinari added 4 commits November 16, 2025 23:47

packaging now reflects /usr-merge

58b0035

use newest image

9ca45e2

MNT: Re-rendered with conda-smithy 3.53.1 and conda-forge-pinning 202…

5945b0d

…5.11.15.23.15.53 Other tools: - conda-build 25.9.0 - rattler-build 0.49.0 - rattler-build-conda-compat 1.4.9

fail on error

8a86142

Ensure paths without /usr keeps working; expand tests

868f64d

h-vetinari force-pushed the alma10 branch from b001a3e to 868f64d Compare November 16, 2025 12:53

h-vetinari added 2 commits November 17, 2025 00:04

also keep /bin/* & /sbin/* working

914f404

remove s390x workaround

a6b010d

beckermr approved these changes Nov 16, 2025

View reviewed changes

h-vetinari merged commit ce487d2 into conda-forge:main Nov 16, 2025
7 checks passed

h-vetinari deleted the alma10 branch November 16, 2025 21:45

h-vetinari mentioned this pull request Nov 17, 2025

roll-out of glibc 2.39 / RHEL10-derived infrastructure conda-forge/conda-forge.github.io#2657

Open

4 tasks

h-vetinari mentioned this pull request Nov 24, 2025

Add riscv sysroot; switch to rockylinux #93

Merged

h-vetinari mentioned this pull request Dec 26, 2025

allow DEFAULT_LINUX_VERSION=alma10 conda-forge/conda-forge-pinning-feedstock#8095

Open

Uh oh!

Add alma10 sysroot #92

Add alma10 sysroot #92

Uh oh!

Conversation

h-vetinari commented Sep 17, 2025

Uh oh!

conda-forge-admin commented Sep 17, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

h-vetinari commented Nov 15, 2025

Uh oh!

h-vetinari commented Nov 16, 2025

Uh oh!

h-vetinari commented Nov 16, 2025

Uh oh!

beckermr left a comment

Choose a reason for hiding this comment

Uh oh!

h-vetinari commented Nov 16, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

beckermr commented Nov 16, 2025

Uh oh!

h-vetinari commented Nov 16, 2025

Uh oh!

h-vetinari commented Nov 16, 2025

Uh oh!

beckermr left a comment

Choose a reason for hiding this comment

Uh oh!

beckermr Nov 16, 2025

Choose a reason for hiding this comment

Uh oh!

beckermr Nov 16, 2025

Choose a reason for hiding this comment

Uh oh!

Uh oh!

h-vetinari commented Nov 16, 2025

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

3 participants

conda-forge-admin commented Sep 17, 2025 •

edited

Loading

h-vetinari commented Nov 16, 2025 •

edited

Loading