[GIT PULL] RCU changes for v5.15

[GIT PULL] RCU changes for v5.15

[Paul E. McKenney]

Hello, Linus,

Please pull the latest RCU git tree from:

  git://git.kernel.org/pub/scm/linux/kernel/git/paulmck/linux-rcu.git core-rcu.2021.08.28a
  # HEAD: b770efc4608d24fb446b94e1087d9989425dd39b Merge branches 'doc.2021.07.20c', 'fixes.2021.08.06a', 'nocb.2021.07.20c', 'nolibc.2021.07.20c', 'tasks.2021.07.20c', 'torture.2021.07.27a' and 'torturescript.2021.07.27a' into HEAD

RCU changes for this cycle were:

o	Documentation updates.

o	Miscellaneous fixes.

o	Offloaded-callbacks updates

o	Updates to the nolibc library.

o	Tasks-RCU updates.

o	In-kernel torture-test updates.

o	Torture-test scripting, perhaps most notably the pinning of
	torture-test guest OSes so as to force differences in memory
	latency.  For example, in a two-socket system, a four-CPU guest
	OS will have one pair of its CPUs pinned to threads in a single
	core on one socket and the other pair pinned to threads in a
	single core on the other socket.  This approach proved able
	to force race conditions that earlier testing missed.  Some of
	these race conditions are still being tracked down.

							Thanx, Paul

----------------------------------------------------------------
Akira Yokosawa (2):
      Documentation/RCU: Fix emphasis markers
      Documentation/RCU: Fix nested inline markup

Bhaskar Chowdhury (1):
      tools: include: nolibc: Fix a typo occured to occurred in the file nolibc.h

Frederic Weisbecker (4):
      rcu/doc: Add a quick quiz to explain further why we need smp_mb__after_unlock_lock()
      rcu/nocb: Start moving nocb code to its own plugin file
      rcu/nocb: Remove NOCB deferred wakeup from rcutree_dead_cpu()
      rcu: Explain why rcu_all_qs() is a stub in preemptible TREE RCU

Haocheng Xie (1):
      docs: Fix a typo in Documentation/RCU/stallwarn.rst

Jiangong.Han (1):
      rcuscale: Console output claims too few grace periods

Joel Fernandes (Google) (1):
      rcu: Remove special bit at the bottom of the ->dynticks counter

Julian Wiedmann (1):
      rculist: Unify documentation about missing list_empty_rcu()

Liu Song (2):
      rcu: Remove useless "ret" update in rcu_gp_fqs_loop()
      rcu: Use per_cpu_ptr to get the pointer of per_cpu variable

Mark Brown (1):
      tools/nolibc: Implement msleep()

Paul E. McKenney (41):
      doc: Update stallwarn.rst with recent changes
      rcu-tasks: Add comments explaining task_struct strategy
      rcu-tasks: Mark ->trc_reader_nesting data races
      rcu-tasks: Mark ->trc_reader_special.b.need_qs data races
      rcu-tasks: Fix synchronize_rcu_rude() typo in comment
      torture: Add clocksource-watchdog testing to torture.sh
      torture: Make torture.sh accept --do-all and --donone
      torture: Enable KCSAN summaries over groups of torture-test runs
      torture: Create KCSAN summaries for torture.sh runs
      torture: Make kvm-recheck-scf.sh tolerate qemu-cmd comments
      torture: Make kvm-recheck-lock.sh tolerate qemu-cmd comments
      torture: Log more kvm-remote.sh information
      torture: Protect kvm-remote.sh directory trees from /tmp reaping
      torture: Make kvm-recheck.sh skip kcsan.sum for build-only runs
      torture: Move parse-console.sh call to PATH-aware scripts
      torture: Put kvm.sh batch-creation awk script into a temp file
      refscale: Add measurement of clock readout
      rcutorture: Preempt rather than block when testing task stalls
      locktorture: Mark statistics data races
      locktorture: Count lock readers
      scftorture: Add RPC-like IPI tests
      scftorture: Avoid NULL pointer exception on early exit
      torture: Make kvm.sh select per-scenario affinity masks
      torture: Don't redirect qemu-cmd comment lines
      torture: Make kvm-test-1-run-qemu.sh apply affinity
      rcutorture: Upgrade two-CPU scenarios to four CPUs
      torture: Use numeric taskset argument in jitter.sh
      torture: Consistently name "qemu*" test output files
      torture: Make kvm-test-1-run-batch.sh select per-scenario affinity masks
      torture: Don't use "test" command's "-a" argument
      torture: Add timestamps to kvm-test-1-run-qemu.sh output
      torture: Make kvm-test-1-run-qemu.sh check for reboot loops
      rcu: Weaken ->dynticks accesses and updates
      rcu: Mark accesses to ->rcu_read_lock_nesting
      rcu: Start timing stall repetitions after warning complete
      srcutiny: Mark read-side data races
      rcu: Mark lockless ->qsmask read in rcu_check_boost_fail()
      rcu: Make rcu_gp_init() and rcu_gp_fqs_loop noinline to conserve stack
      rcu: Mark accesses in tree_stall.h
      rcu: Print human-readable message for schedule() in RCU reader
      Merge branches 'doc.2021.07.20c', 'fixes.2021.08.06a', 'nocb.2021.07.20c', 'nolibc.2021.07.20c', 'tasks.2021.07.20c', 'torture.2021.07.27a' and 'torturescript.2021.07.27a' into HEAD

Sebastian Andrzej Siewior (2):
      rcu: Replace deprecated CPU-hotplug functions
      torture: Replace deprecated CPU-hotplug functions.

Sergey Senozhatsky (2):
      rcu/tree: Handle VM stoppage in stall detection
      rcu: Do not disable GP stall detection in rcu_cpu_stall_reset()

Yanfei Xu (2):
      rcu: Fix to include first blocked task in stall warning
      rcu: Fix stall-warning deadlock due to non-release of rcu_node ->lock

Zhouyi Zhou (1):
      rcu: Fix macro name CONFIG_TASKS_RCU_TRACE

 .../Memory-Ordering/Tree-RCU-Memory-Ordering.rst   |   29 +
 .../RCU/Design/Requirements/Requirements.rst       |    8 +-
 Documentation/RCU/checklist.rst                    |   24 +-
 Documentation/RCU/rcu_dereference.rst              |    6 +-
 Documentation/RCU/stallwarn.rst                    |   31 +-
 include/linux/rculist.h                            |   35 +-
 include/linux/rcupdate.h                           |    4 +-
 include/linux/rcutiny.h                            |    3 -
 include/linux/srcutiny.h                           |    8 +-
 kernel/locking/locktorture.c                       |   25 +-
 kernel/rcu/rcuscale.c                              |    4 +-
 kernel/rcu/rcutorture.c                            |    7 +-
 kernel/rcu/refscale.c                              |   36 +-
 kernel/rcu/srcutiny.c                              |    2 +-
 kernel/rcu/tasks.h                                 |   36 +-
 kernel/rcu/tree.c                                  |  107 +-
 kernel/rcu/tree_nocb.h                             | 1496 +++++++++++++++++++
 kernel/rcu/tree_plugin.h                           | 1506 +-------------------
 kernel/rcu/tree_stall.h                            |  111 +-
 kernel/scftorture.c                                |   78 +-
 kernel/sched/core.c                                |   11 +
 kernel/torture.c                                   |    6 +-
 tools/include/nolibc/nolibc.h                      |   15 +-
 tools/testing/selftests/rcutorture/bin/jitter.sh   |   10 +-
 .../selftests/rcutorture/bin/kcsan-collapse.sh     |    2 +-
 .../testing/selftests/rcutorture/bin/kvm-again.sh  |    4 +-
 .../selftests/rcutorture/bin/kvm-assign-cpus.sh    |  106 ++
 .../rcutorture/bin/kvm-get-cpus-script.sh          |   88 ++
 .../selftests/rcutorture/bin/kvm-recheck-lock.sh   |    2 +-
 .../selftests/rcutorture/bin/kvm-recheck-scf.sh    |    2 +-
 .../selftests/rcutorture/bin/kvm-recheck.sh        |    5 +-
 .../selftests/rcutorture/bin/kvm-remote-noreap.sh  |   30 +
 .../testing/selftests/rcutorture/bin/kvm-remote.sh |   20 +-
 .../rcutorture/bin/kvm-test-1-run-batch.sh         |   24 +
 .../rcutorture/bin/kvm-test-1-run-qemu.sh          |   49 +-
 .../selftests/rcutorture/bin/kvm-test-1-run.sh     |    2 +
 tools/testing/selftests/rcutorture/bin/kvm.sh      |   39 +-
 tools/testing/selftests/rcutorture/bin/torture.sh  |   37 +-
 .../selftests/rcutorture/configs/rcu/RUDE01        |    2 +-
 .../selftests/rcutorture/configs/rcu/TASKS01       |    2 +-
 .../selftests/rcutorture/configs/rcu/TASKS03       |    2 +-
 41 files changed, 2241 insertions(+), 1773 deletions(-)
 create mode 100644 kernel/rcu/tree_nocb.h
 create mode 100755 tools/testing/selftests/rcutorture/bin/kvm-assign-cpus.sh
 create mode 100755 tools/testing/selftests/rcutorture/bin/kvm-get-cpus-script.sh
 create mode 100755 tools/testing/selftests/rcutorture/bin/kvm-remote-noreap.sh

Re: [GIT PULL] RCU changes for v5.15

[Linus Torvalds]

On Sun, Aug 29, 2021 at 9:15 PM Paul E. McKenney <paulmck@kernel.org> wrote:
>
> Please pull the latest RCU git tree from:

Pulled.

I do note that you are one of the few people I pull from that haven't
converted to signed tags yet.

It's not technically a problem, since I trust the kernel.org
repositories, but when I notice I try to gently push people to use
signed tags even on kernel.org repositories. It's just a good idea.

So here's that gentle nudge.

I even have an old pgp key from you in my keyring, so I know you at
some point had one. And making git use them really is not a log more
than using "git tag -s" to create one, and pointing me at the tag
instead of the branch name.

         Linus

Re: [GIT PULL] RCU changes for v5.15

[pr-tracker-bot]

The pull request you sent on Sun, 29 Aug 2021 21:14:59 -0700:

> git://git.kernel.org/pub/scm/linux/kernel/git/paulmck/linux-rcu.git core-rcu.2021.08.28a

has been merged into torvalds/linux.git:
https://git.kernel.org/torvalds/c/4ca4256453effb885c1688633676682529593f82

Thank you!

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

Re: [GIT PULL] RCU changes for v5.15

[Paul E. McKenney]

On Mon, Aug 30, 2021 at 12:54:19PM -0700, Linus Torvalds wrote:
> On Sun, Aug 29, 2021 at 9:15 PM Paul E. McKenney <paulmck@kernel.org> wrote:
> >
> > Please pull the latest RCU git tree from:
> 
> Pulled.
> 
> I do note that you are one of the few people I pull from that haven't
> converted to signed tags yet.
> 
> It's not technically a problem, since I trust the kernel.org
> repositories, but when I notice I try to gently push people to use
> signed tags even on kernel.org repositories. It's just a good idea.
> 
> So here's that gentle nudge.
> 
> I even have an old pgp key from you in my keyring, so I know you at
> some point had one. And making git use them really is not a log more
> than using "git tag -s" to create one, and pointing me at the tag
> instead of the branch name.

Very well, I will use a signed tag for my next pull request to you.

If you would like to check ahead of time, I have created a
signed test tag named test.2021.08.30a on
git://git.kernel.org/pub/scm/linux/kernel/git/paulmck/linux-rcu.git

							Thanx, Paul