[UBUNTU 22.04] GLIBC: Adding new s390 platform IBM z16

[bugproxy]

SRU Justification:
==================

[Impact]

 * This is a hardware enablement SRU to improve IBM z16 support.

 * A fatal glibc error is dumped if glibc was build with an
   architecture level set for z16, but run on an older machine.
   (See dl-hwcap-check.h.)

 * Hence z16 is added as new s390 platform,

 * the new IBM z16 is added to platform string array,

 * the macro _DL_PLATFORMS_COUNT is incremented.

 * '_dl_hwcaps_subdir' is extended by "z16" if HWCAP_S390_VXRS_PDE2
   is set,

 * and 'tst-glibc-hwcaps.c' is extended in order to test z16 via
   new marker5.

[Test Plan]

 * An IBM zSystems (or LinuxONE) system,
   with Ubuntu 22.04 LTS installed.

 * Install the updated libc6 package on different zSystems
   hardware generations for regression testing, since
   the patched libc6 should not cause any changed behavior
   or impact on existing IBM zSystems hardware.

 * Test of the glibc-hwcaps feature is covered by tst-glibc-hwcaps.c test
   from glibc-testsuite.

 * In short: Different shared libraries are placed in the z16/z15/...
   subdirectories.
   A runtime-check tests whether the correct library was loaded on
   the particular machine.

 * IBM is going to test this feature on a z16 as soon as a patched
   libc6 package is available.

[Where problems could occur]

 * There could be problems with the newly introduced marker5
   (libmarkermod5-1.so) in sysdeps/s390/s390-64/Makefile.
   These should largely occur already at build time,
   or in 'tst-glibc-hwcaps.c'.

 * The increased 'DL_PLATFORMS_COUNT' shouldn't harm.

 * Problems could occur is the hardware capabilities (hwcap) are not properly
   identified at 'dl-hwcap-check.h' and 'dl-hwcaps-subdirs.c'.

 * Wrong (not existing) capabilities might be assumed that could
   eventually lead to crashes in attempts to use them.

 * The majority of the changes are under sysdeps/s390 and with that
   hardware specific, other are in script and Makefile.

[Other Info]

 * IBM z16 exploitation is introduced with 22.04 LTS, hence
   this patch is not needed for older Ubuntu releases.
__________

After IBM z16 has been announced, we need to add the new s390 platform in glibc.

S390: Add new s390 platform z16.
https://sourceware.org/git/?p=glibc.git;a=commit;h=2376944b9e5c0364b9fb473e4d8dabca31b57167

Backport to glibc 2.35 release branch:
https://sourceware.org/git/?p=glibc.git;a=commit;h=6583d534458e7f8beec8d0766b6db0bdaae6749c

[Frank Heimes (fheimes)]

@IBM am I right assuming that this only affects 22.04's glibc version 2.35 (since older glibc versions (for example of 20.10: glibc 2.34 and 20.04: glibc 2.31) don't contain the z16 related code that is affected?

[bugproxy (bugproxy)]

------- Comment From <email address hidden> 2022-05-10 05:33 EDT-------
Hi fheimes,

the first part of z16 support went into glibc-2.34:
glibc commit "S390: Add new hwcap values."
https://sourceware.org/git/?p=glibc.git;a=commit;h=25251c0707fe34f30a27381a5fabc35435a96621

Thus 20.10: glibc 2.34 is also affected.
Backport to glibc 2.34 release branch:
S390: Add new s390 platform z16.
https://sourceware.org/git/?p=glibc.git;a=commit;h=0c03cb54c808173d8e7ba96f6152dfcf627ac496

For 20.04: glibc 2.31, there is nothing to do.

[Frank Heimes (fheimes)]

Ok, thx - I'll then mark impish/20.10, jammy/22.04 and kinetic/22.10 as affected releases ...

[Michael Hudson-Doyle (mwhudson)]

impish will be EOL in a few weeks and I do not expect to upload a glibc before then.

[Michael Hudson-Doyle (mwhudson)]

Could someone update this bug to follow the SRU template? https://wiki.ubuntu.com/StableReleaseUpdates#SRU_Bug_Template

I'm presuming someone from IBM who has access to z16 will have to test this for us.

[bugproxy (bugproxy)]

------- Comment From <email address hidden> 2022-07-05 04:17 EDT-------
[Impact]

The first part of z16 support went into glibc-2.34:
- glibc commit "S390: Add new hwcap values."
https://sourceware.org/git/?p=glibc.git;a=commit;h=25251c0707fe34f30a27381a5fabc35435a96621

After IBM z16 has been announced, we need to add the new s390 platform in glibc:
- glibc upstream commit "S390: Add new s390 platform z16."
https://sourceware.org/git/?p=glibc.git;a=commit;h=2376944b9e5c0364b9fb473e4d8dabca31b57167
- Backport to glibc 2.35 release branch:
https://sourceware.org/git/?p=glibc.git;a=commit;h=6583d534458e7f8beec8d0766b6db0bdaae6749c

Beneath adding the platform name, the glibc-hwcaps subdirectories feature is extended. The z16 subfolder is now also searched for "z16" specific libraries if support is indicated by the HWCAPs.

[Test Plan]

Testing of the glibc-hwcaps feature is done by tst-glibc-hwcaps.c test from glibc-testsuite. In short: Different shared libraries are placed in the z16/z15/... subdirectories. A runtime-check tests whether the correct library was loaded on the particular machine.
Yes, IBM will test this feature on a z16 as soon as the libc packages are updated.

[Where problems could occur]

The mentioned patch does only change s390. Other architectures are not involved.

[Brian Murray (brian-murray)]

Hello bugproxy, or anyone else affected,

Accepted glibc into jammy-proposed. The package will build now and be available at https://launchpad.net/ubuntu/+source/glibc/2.35-0ubuntu3.1 in a few hours, and then in the -proposed repository.

Please help us by testing this new package. See https://wiki.ubuntu.com/Testing/EnableProposed for documentation on how to enable and use -proposed. Your feedback will aid us getting this update out to other Ubuntu users.

If this package fixes the bug for you, please add a comment to this bug, mentioning the version of the package you tested, what testing has been performed on the package and change the tag from verification-needed-jammy to verification-done-jammy. If it does not fix the bug for you, please add a comment stating that, and change the tag to verification-failed-jammy. In either case, without details of your testing we will not be able to proceed.

Further information regarding the verification process can be found at https://wiki.ubuntu.com/QATeam/PerformingSRUVerification . Thank you in advance for helping!

N.B. The updated package will be released to -updates after the bug(s) fixed by this package have been verified and the package has been in -proposed for a minimum of 7 days.

[bugproxy (bugproxy)]

------- Comment From <email address hidden> 2022-07-12 05:14 EDT-------
I've just installed the package on z15/z16:
$ apt list --installed libc6
libc6/jammy-proposed,now 2.35-0ubuntu3.1 s390x [installed]

I've placed some test-libraries into these paths:
$ find /usr/lib/s390x-linux-gnu/glibc-hwcaps -type f
/usr/lib/s390x-linux-gnu/glibc-hwcaps/z13/libtest-glibc-hwcaps-lib.so
/usr/lib/s390x-linux-gnu/glibc-hwcaps/z14/libtest-glibc-hwcaps-lib.so
/usr/lib/s390x-linux-gnu/glibc-hwcaps/z15/libtest-glibc-hwcaps-lib.so
/usr/lib/s390x-linux-gnu/glibc-hwcaps/z16/libtest-glibc-hwcaps-lib.so

Depending on the machine-level, the correct library is used, e.g.:
- on z16, /usr/lib/s390x-linux-gnu/glibc-hwcaps/z16/libtest-glibc-hwcaps-lib.so
- on z16, if .../glibc-hwcaps/z16/... is not present: /usr/lib/s390x-linux-gnu/glibc-hwcaps/z15/libtest-glibc-hwcaps-lib.so

- on z15, /usr/lib/s390x-linux-gnu/glibc-hwcaps/z15/libtest-glibc-hwcaps-lib.so

Thanks for integrating the patches. This bug is successfully verified for jammy.

[Frank Heimes (fheimes)]

Thx for the verification!

[Ubuntu SRU Bot (ubuntu-sru-bot)]

All autopkgtests for the newly accepted glibc (2.35-0ubuntu3.1) for jammy have finished running.
The following regressions have been reported in tests triggered by the package:

macaulay2/1.19.1+ds-6 (armhf)
oss4/4.2-build2010-5ubuntu9 (amd64)
packer/1.6.6+ds1-4 (s390x)
network-manager/1.36.6-0ubuntu2 (arm64)
hilive/2.0a-3build3 (arm64)
openjdk-lts/11.0.15+10-0ubuntu0.22.04.1 (armhf)
ruby-mysql2/0.5.3-3ubuntu4 (s390x, arm64, ppc64el, armhf)
prometheus/2.31.2+ds1-1ubuntu1 (armhf)
umockdev/0.17.7-1 (s390x)
mbedtls/2.28.0-1build1 (s390x)
golang-v2ray-core/4.34.0-5 (armhf)
slixmpp/1.7.1-1build1 (s390x)
nwchem/7.0.2-3 (arm64)
pappl/1.0.3-2 (s390x)
golang-github-bmatsuo-lmdb-go/1.8.0+git20170215.a14b5a3-2 (amd64)
ruby-standalone/3.0~1 (s390x)
seqkit/2.1.0+ds-1 (s390x)
pandas/1.3.5+dfsg-3 (s390x)
netplan.io/0.104-0ubuntu2 (arm64)
pyfai/0.21.1+dfsg1-1build1 (ppc64el)
iptables/1.8.7-1ubuntu5 (i386, s390x)
golang-github-influxdata-tail/1.0.0+git20180327.c434825-4 (s390x, ppc64el)
opensaml/3.2.1-1 (arm64)
node-iconv/3.0.1+~3.0.0-1 (armhf)
ubuntu-image/2.2+22.04ubuntu3 (s390x)
fenix/0.92a.dfsg1-12.1 (i386)
notary/0.7.0+ds1-1 (arm64)
rustc/1.59.0+dfsg1-1~ubuntu2~22.04.1 (arm64)
tmux/3.2a-4build1 (s390x)
genshi/0.7.6-1build1 (s390x)

Please visit the excuses page listed below and investigate the failures, proceeding afterwards as per the StableReleaseUpdates policy regarding autopkgtest regressions [1].

https://people.canonical.com/~ubuntu-archive/proposed-migration/jammy/update_excuses.html#glibc

[1] https://wiki.ubuntu.com/StableReleaseUpdates#Autopkgtest_Regressions

Thank you!

[Launchpad Janitor (janitor)]

This bug was fixed in the package glibc - 2.35-0ubuntu3.1

---------------
glibc (2.35-0ubuntu3.1) jammy; urgency=medium

  * debian/maint: add a script to manage backports of patches from upstream
    maintenance branch.
  * Cherry-pick patches from upstream maintenance branch:
    - 0001-S390-Add-new-s390-platform-z16.patch (LP: #1971612)
    - 0002-powerpc-Fix-VSX-register-number-on-__strncpy_power9-.patch (LP: #1978130)

 -- Michael Hudson-Doyle <email address hidden> Thu, 07 Jul 2022 11:23:23 +1200

[Łukasz Zemczak (sil2100)]

The verification of the Stable Release Update for glibc has completed successfully and the package is now being released to -updates. Subsequently, the Ubuntu Stable Release Updates Team is being unsubscribed and will not receive messages about this bug report. In the event that you encounter a regression using the package from -updates please report a new bug using ubuntu-bug and tag the bug report regression-update so we can easily find any regressions.

[Frank Heimes (fheimes)]

Since libc6 2.36-0ubuntu1 (that incl. the patches mentioned here) is in kinetic-proposed,
I'm updating the status for kinetic to Fix Committed.

[Frank Heimes (fheimes)]

ignoring impish, since it's eol

2.36-0ubuntu2 is still in kinetic-proposed, hence status Fix Committed

[Frank Heimes (fheimes)]

Meanwhile glibc 2.36-0ubuntu2 migrated from kinetic-proposed to release.
Hence updating (and with that closing) this ticket to Fix Released.