[GIT PULL] RCU changes for v6.14

[GIT PULL] RCU changes for v6.14

[Uladzislau Rezki (Sony)]

Hi Linus,

Please pull the latest RCU git tree from:

git://git.kernel.org/pub/scm/linux/kernel/git/rcu/linux.git tags/rcu.release.v6.14

# HEAD: 4b5c2205526cc1579b840893b98eb7545220f7cf:

Merge branches 'fixes.2024.12.14a', 'rcutorture.2024.12.14a', 'srcu.2024.12.14a' and 'torture-test.2024.12.14a' into rcu-merge.2024.12.14a

fixes.2024.12.14a: RCU fixes
rcutorture.2024.12.14a: Torture-test updates
srcu.2024.12.14a: SRCU updates
torture-test.2024.12.14a: Adding an extra test, fixes

One merge conflicts was detected by linux-next:

* https://lore.kernel.org/lkml/20241220133738.1beae531@canb.auug.org.au/T/

The conflict resolution from linux-next look good to me, plus I make
my own resolutions at branch merge/rcu.2024.12.21a for your reference.

----------------------------------------------------------------

RCU pull request for v6.14

This pull request contains the following branches:

fixes.2024.12.14a: Misc fixes, check if IRQs are disabled in rcu_exp_need_qs(),
    instrument KCSAN exclusive-writer assertions, add extra WARN_ON_ONCE() check,
    set the cpu_no_qs.b.exp under lock, warn if callback enqueued on offline CPU.

rcutorture.2024.12.14a: Torture-test updates, add rcutorture.preempt_duration kernel
    module parameter, make the TREE03 scenario do preemption, improve pooling timeouts
    for rcu_torture_writer(), improve output of "Failure/close-call rcutorture reader
    segments", add some reader-state debugging checks, update doc of polled APIs, add
    extra diagnostics for per-reader-segment preemption.

srcu.2024.12.14a: SRCU updates, improve doc for srcu_read_lock() in terms of return
    value, fix typo in comments, remove redundant GP sequence checks in the
    srcu_funnel_gp_start.

torture-test.2024.12.14a: Add an extra test for sched_clock(), improve testing
    on unresponsive systems.

----------------------------------------------------------------

Feng Lee (1):
  srcu: Remove redundant GP sequence checks in srcu_funnel_gp_start

Frederic Weisbecker (1):
  rcu: Report callbacks enqueued on offline CPU blind spot

Paul E. McKenney (29):
  refscale: Add test for sched_clock()
  torture: Make kvm-remote.sh give up on unresponsive system
  torture: Add dowarn argument to torture_sched_setaffinity()
  rcutorture: Add random real-time preemption
  rcutorture: Make the TREE03 scenario do preemption
  rcutorture: Decorate failing reader segments with CPU ID
  rcutorture: Use finer-grained timeouts for rcu_torture_writer() polling
  rcutorture: Add ->cond_sync_exp_full function to rcu_ops structure
  rcutorture: Check preemption for failing reader
  rcutorture: Decorate failing reader segments with last CPU ID
  rcutorture: Add full read-side contexts to "busted" torture type
  rcutorture: Pretty-print rcutorture reader segments
  rcutorture: Make rcutorture_one_extend() check reader state
  rcutorture: Ignore attempts to test preemption and forward progress
  rcutorture: Add documentation for recent conditional and polled APIs
  rcutorture: Add parameters to control polled/conditional wait interval
  rcutorture: Add preempt_count() to rcutorture_one_extend_check() diagnostics
  rcutorture: Read CPU ID for decoration protected by both reader types
  rcutorture: Add per-reader-segment preemption diagnostics
  rcutorture: Use symbols for SRCU reader flavors
  rcu: Make rcu_report_exp_cpu_mult() caller acquire lock
  rcu: Move rcu_report_exp_rdp() setting of ->cpu_no_qs.b.exp under lock
  rcu: Replace open-coded rcu_exp_need_qs() from rcu_exp_handler() with call
  rcu: Make preemptible rcu_exp_handler() check idempotency
  rcu: Add KCSAN exclusive-writer assertions for rdp->cpu_no_qs.b.exp
  rcu: Add lockdep_assert_irqs_disabled() to rcu_exp_need_qs()
  MAINTAINERS: Update RCU git tree
  srcu: Guarantee non-negative return value from srcu_read_lock()
  srcu: Fix typo s/srcu_check_read_flavor()/__srcu_check_read_flavor()/

 Documentation/admin-guide/kernel-parameters.txt                |   74 ++++++++++++++++++++
 MAINTAINERS                                                    |   10 +-
 Makefile                                                       |    2
 arch/x86/kernel/cpu/amd.c                                      |    2
 include/linux/rcupdate_wait.h                                  |   11 +++
 include/linux/srcu.h                                           |   15 ++--
 include/linux/srcutree.h                                       |    2
 include/linux/torture.h                                        |    2
 kernel/bpf/lpm_trie.c                                          |   44 +++++-------
 kernel/locking/locktorture.c                                   |    6 -
 kernel/rcu/Kconfig.debug                                       |   15 ++++
 kernel/rcu/rcutorture.c                                        |    2
 kernel/rcu/refscale.c                                          |   40 ++++++++++-
 kernel/rcu/srcutree.c                                          |    3
 kernel/rcu/tree.c                                              |    3
 kernel/rcu/tree_exp.h                                          |   20 +++--
 kernel/rcu/tree_plugin.h                                       |    1
 kernel/rcu/update.c                                            |    4 -
 scripts/mod/modpost.c                                          |    2
 scripts/package/install-extmod-build                           |    4 -
 tools/testing/selftests/bpf/.gitignore                         |    1
 tools/testing/selftests/bpf/Makefile                           |    2
 tools/testing/selftests/bpf/map_tests/lpm_trie_map_basic_ops.c |   10 --
 tools/testing/selftests/rcutorture/bin/kvm-remote.sh           |   25 +++++-
 tools/testing/selftests/rcutorture/configs/rcu/TREE03.boot     |    1
 include/linux/srcu.h                                           |   21 +++--
 include/linux/srcutree.h                                       |    8 --
 kernel/bpf/lpm_trie.c                                          |   87 +++++++++++++++---------
 kernel/rcu/Kconfig.debug                                       |   16 ++++
 kernel/rcu/rcutorture.c                                        |  279 +++++++++++++++++++++++++++++++++++++++++++++++++++++++++++-----------------
 kernel/rcu/srcutree.c                                          |    5 -
 kernel/rcu/tree_exp.h                                          |  128 ++++++++++++++++++++++-------------
 kernel/rcu/tree_plugin.h                                       |    1
 tools/testing/selftests/bpf/map_tests/lpm_trie_map_basic_ops.c |  395 +++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
 37 files changed, 1033 insertions(+), 235 deletions(-)

Re: [GIT PULL] RCU changes for v6.14

[Linus Torvalds]

On Fri, 17 Jan 2025 at 09:24, Uladzislau Rezki (Sony) <urezki@gmail.com> wrote:
>
> Please pull the latest RCU git tree from:
>
> git://git.kernel.org/pub/scm/linux/kernel/git/rcu/linux.git tags/rcu.release.v6.14

Hmm. I have pulled this, and the resulting shortlog matches what you
say I should have gotten.

But your diffstat is odd and looks very wrong.

Things like this:

>  tools/testing/selftests/bpf/map_tests/lpm_trie_map_basic_ops.c |  395 +++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

were already in my tree from before, and aren't new (and your shortlog
doesn't mention them).

Also, look at that line. It's long. It's not supposed to be that long.
You've presumably generated this (incorrect) diffstat in a very wide
terminal, and then cut-and-pasted that very wide output into the
email.

Please don't do that. Yes, git will default to wide output if the
output is a wide terminal, but if you either pipe it into something
else (may I suggest "xsel", so that it goes directly into your X
clipboard).

Or use "--stat=80" to limit the width manually to 80 columns.

But the strange formatting is the less worrisome issue. The diffstat
just being plain wrong is the more alarming thing.

             Linus

Re: [GIT PULL] RCU changes for v6.14

[pr-tracker-bot]

The pull request you sent on Fri, 17 Jan 2025 18:24:40 +0100:

> git://git.kernel.org/pub/scm/linux/kernel/git/rcu/linux.git tags/rcu.release.v6.14

has been merged into torvalds/linux.git:
https://git.kernel.org/torvalds/c/9f3ee94e705a5b2fe352befb37e499163f98b9b6

Thank you!

-- 
Deet-doot-dot, I am a bot.
https://korg.docs.kernel.org/prtracker.html

Re: [GIT PULL] RCU changes for v6.14

[Uladzislau Rezki]

On Tue, Jan 21, 2025 at 02:51:15PM -0800, Linus Torvalds wrote:
> On Fri, 17 Jan 2025 at 09:24, Uladzislau Rezki (Sony) <urezki@gmail.com> wrote:
> >
> > Please pull the latest RCU git tree from:
> >
> > git://git.kernel.org/pub/scm/linux/kernel/git/rcu/linux.git tags/rcu.release.v6.14
> 
> Hmm. I have pulled this, and the resulting shortlog matches what you
> say I should have gotten.
> 
> But your diffstat is odd and looks very wrong.
> 
> Things like this:
> 
> >  tools/testing/selftests/bpf/map_tests/lpm_trie_map_basic_ops.c |  395 +++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
> 
> were already in my tree from before, and aren't new (and your shortlog
> doesn't mention them).
> 
> Also, look at that line. It's long. It's not supposed to be that long.
> You've presumably generated this (incorrect) diffstat in a very wide
> terminal, and then cut-and-pasted that very wide output into the
> email.
> 
> Please don't do that. Yes, git will default to wide output if the
> output is a wide terminal, but if you either pipe it into something
> else (may I suggest "xsel", so that it goes directly into your X
> clipboard).
> 
> Or use "--stat=80" to limit the width manually to 80 columns.
> 
> But the strange formatting is the less worrisome issue. The diffstat
> just being plain wrong is the more alarming thing.
>
Thank you for taking this in!

My script that generates a diff-stat is not perfect, i will fix it.
Sorry this is my bad.

--
Uladzislau Rezki