Dear RT Folks,
This is the RT stable review cycle of patch 4.14.292-rt138-rc1.
Please scream at me if I messed something up. Please test the patches
too.
The -rc release will be uploaded to kernel.org and will be deleted
when the final release is out. This is just a review release (or
release candidate).
The pre-releases will not be pushed to the git repository, only the
final release is.
If all goes well, this patch will be converted to the next main
release on 2022-10-06.
CONTEXT FOR THIS RC:
====================
==> v4.14.282-rt135
Last v4.14-rt in the repository
==> v4.14.282-rt136
Reverted the RT-specific changes to random() code:
6c44e13a906db Revert "drivers: random: Reduce preempt disabled region"
eb93f866c40dd Revert "random: Make it work on rt"
c88021dd7588c Revert "random: avoid preempt_disable()ed section"
c5bd012a413d7 Revert "char/random: don't print that the init is done"
==> v4,14,292-rt137
With the revert of the RT changes to random() I was able to merge up
to stable v4.14.292, where a new merge conflict appeared.
This kernel has the random() code broken on RT.
==> v4,14,292-rt138
Applied the two patches below and the kernel is up and running again.
random: Use local locks for crng context access
workqueue: update code to match v4.19.255-rt113
NOTE ABOUT v4,14,292-rt138:
===========================
This kernel works fine (tested with LTP and stress-ng, both with and
without lockdep enabled. But on poweroff or reboot there are tree
backtraces coming from the commit listed below, part of the all-in-one
"workqueue: update code to match v4.19.255-rt113" patch:
8c37f7c23c02f workqueue: Try to catch flush_work() without INIT_WORK().
This commits translates to the following statement in __flush_work(),
from kernel/workqueue.c:
if (WARN_ON(!work->func))
return false;
Should I remove this change? The fix requires a good amount of changes
and takes us farther away from stock v4.14 in a dangerous way.
And an even more important question, is this all-in-one patch
acceptable? Should I backport all the individual commits instead?
To build 4.14.292-rt138-rc1 directly, the following patches should be applied:
https://www.kernel.org/pub/linux/kernel/v4.x/linux-4.14.tar.xz
https://www.kernel.org/pub/linux/kernel/v4.x/patch-4.14.292.xz
https://www.kernel.org/pub/linux/kernel/projects/rt/4.14/older/patch-4.14.292-rt138-rc1.patch.xz
Signing key fingerprint:
4FDD B8C4 3C59 34BD 8BE9 2696 7203 D980 A448 C8F8
All keys used for the above files and repositories can be found on the
following git repository:
git://git.kernel.org/pub/scm/docs/kernel/pgpkeys.git
Enjoy!
Luis
Changes from v4.14.292-rt137:
Daniel Wagner (1):
random: Use local locks for crng context access
Luis Claudio R. Goncalves (2):
workqueue: update code to match v4.19.255-rt113
Linux 4.14.292-rt138-rc1
drivers/char/random.c | 11 +-
include/linux/swait.h | 14 +
include/linux/workqueue.h | 10 +-
kernel/workqueue.c | 623 ++++++++++++++++++++------------------
localversion-rt | 2 +-
5 files changed, 346 insertions(+), 314 deletions(-)
--
2.37.3
