linux.git/arch/powerpc/kernel/smp.c, branch v6.12

smp: Mark smp_prepare_boot_cpu() __init

2024-09-08T14:01:10+00:00

smp_prepare_boot_cpu() is only called during boot, hence mark it as
__init.

Signed-off-by: Bibo Mao 
Signed-off-by: Thomas Gleixner 
Reviewed-by: Huacai Chen 
Link: https://lore.kernel.org/all/20240907082720.452148-1-maobibo@loongson.cn

powerpc: Fix typos

2024-05-07T14:21:30+00:00

Fix typos, most reported by "codespell arch/powerpc".  Only touches
comments, no code changes.

Signed-off-by: Bjorn Helgaas 
Signed-off-by: Michael Ellerman 
Link: https://msgid.link/20240103231605.1801364-8-helgaas@kernel.org

powerpc/kdump: Split KEXEC_CORE and CRASH_DUMP dependency

2024-03-17T02:34:00+00:00

Remove CONFIG_CRASH_DUMP dependency on CONFIG_KEXEC. CONFIG_KEXEC_CORE
was used at places where CONFIG_CRASH_DUMP or CONFIG_CRASH_RESERVE was
appropriate. Replace with appropriate #ifdefs to support CONFIG_KEXEC
and !CONFIG_CRASH_DUMP configuration option. Also, make CONFIG_FA_DUMP
dependent on CONFIG_CRASH_DUMP to avoid unmet dependencies for FA_DUMP
with !CONFIG_KEXEC_CORE configuration option.

Signed-off-by: Hari Bathini 
Signed-off-by: Michael Ellerman 
Link: https://msgid.link/20240226103010.589537-4-hbathini@linux.ibm.com

sched/topology: Rename SD_SHARE_PKG_RESOURCES to SD_SHARE_LLC

2024-02-28T14:43:17+00:00

SD_SHARE_PKG_RESOURCES is a bit of a misnomer: its naming suggests that
it's sharing all 'package resources' - while in reality it's specifically
for sharing the LLC only.

Rename it to SD_SHARE_LLC to reduce confusion.

[ mingo: Rewrote the confusing changelog as well. ]

Suggested-by: Valentin Schneider 
Signed-off-by: Alex Shi 
Signed-off-by: Ingo Molnar 
Reviewed-by: Valentin Schneider 
Reviewed-by: Ricardo Neri 
Reviewed-by: Barry Song 
Link: https://lore.kernel.org/r/20240210113924.1130448-5-alexs@kernel.org

powerpc/smp: Dynamically build Powerpc topology

2023-12-15T02:51:34+00:00

Currently there are four Powerpc specific sched topologies.  These are
all statically defined.  However not all these topologies are used by
all Powerpc systems.

To avoid unnecessary degenerations by the scheduler, masks and flags
are compared. However if the sched topologies are build dynamically then
the code is simpler and there are greater chances of avoiding
degenerations.

Note:
Even X86 builds its sched topologies dynamically and proposed changes
are very similar to the way X86 is building its topologies.

Signed-off-by: Srikar Dronamraju 
Signed-off-by: Michael Ellerman 
Link: https://msgid.link/20231214180720.310852-6-srikar@linux.vnet.ibm.com

powerpc/smp: Avoid asym packing within thread_group of a core

2023-12-15T02:51:34+00:00

PowerVM Hypervisor will schedule at a core granularity. However each
core can have more than one thread_groups. For better utilization in
case of a shared processor, its preferable for the scheduler to pack to
the lowest core. However there is no benefit of moving a thread between
two thread groups of the same core.

Signed-off-by: Srikar Dronamraju 
Signed-off-by: Michael Ellerman 
Link: https://msgid.link/20231214180720.310852-5-srikar@linux.vnet.ibm.com

powerpc/smp: Add __ro_after_init attribute

2023-12-15T02:51:34+00:00

There are some variables that are only updated at boot time.
So add __ro_after_init attribute to such variables

Signed-off-by: Srikar Dronamraju 
Signed-off-by: Michael Ellerman 
Link: https://msgid.link/20231214180720.310852-4-srikar@linux.vnet.ibm.com

powerpc/smp: Disable MC domain for shared processor

2023-12-15T02:51:34+00:00

Like L2-cache info, coregroup information which is used to determine MC
sched domains is only present on dedicated LPARs. i.e PowerVM doesn't
export coregroup information for shared processor LPARs. Hence disable
creating MC domains on shared LPAR Systems.

Signed-off-by: Srikar Dronamraju 
Signed-off-by: Michael Ellerman 
Link: https://msgid.link/20231214180720.310852-3-srikar@linux.vnet.ibm.com

powerpc/smp: Enable Asym packing for cores on shared processor

2023-12-15T02:51:34+00:00

If there are shared processor LPARs, underlying Hypervisor can have more
virtual cores to handle than actual physical cores.

Starting with Power 9, a big core (aka SMT8 core) has 2 nearly
independent thread groups. On a shared processors LPARs, it helps to
pack threads to lesser number of cores so that the overall system
performance and utilization improves. PowerVM schedules at a big core
level. Hence packing to fewer cores helps.

Since each thread-group is independent, running threads on both the
thread-groups of a SMT8 core, should have a minimal adverse impact in
non over provisioned scenarios. These changes in this patchset will not
affect in the over provisioned scenario. If there are more threads than
SMT domains, then asym_packing will not kick-in

For example: Lets says there are two 8-core Shared LPARs that are
actually sharing a 8 Core shared physical pool, each running 8 threads
each. Then Consolidating 8 threads to 4 cores on each LPAR would help
them to perform better. This is because each of the LPAR will get
100% time to run applications and there will no switching required by
the Hypervisor.

To achieve this, enable SD_ASYM_PACKING flag at CACHE, MC and DIE level
when the system is running in shared processor mode and has big cores.

Signed-off-by: Srikar Dronamraju 
Signed-off-by: Michael Ellerman 
Link: https://msgid.link/20231214180720.310852-2-srikar@linux.vnet.ibm.com

Merge tag 'rcu-next-v6.7' of git://git.kernel.org/pub/scm/linux/kernel/git/frederic/linux-dynticks

2023-10-31T04:01:41+00:00

Pull RCU updates from Frederic Weisbecker:

 - RCU torture, locktorture and generic torture infrastructure updates
   that include various fixes, cleanups and consolidations.

   Among the user visible things, ftrace dumps can now be found into
   their own file, and module parameters get better documented and
   reported on dumps.

 - Generic and misc fixes all over the place. Some highlights:

     * Hotplug handling has seen some light cleanups and comments

     * An RCU barrier can now be triggered through sysfs to serialize
       memory stress testing and avoid OOM

     * Object information is now dumped in case of invalid callback
       invocation

     * Also various SRCU issues, too hard to trigger to deserve urgent
       pull requests, have been fixed

 - RCU documentation updates

 - RCU reference scalability test minor fixes and doc improvements.

 - RCU tasks minor fixes

 - Stall detection updates. Introduce RCU CPU Stall notifiers that
   allows a subsystem to provide informations to help debugging. Also
   cure some false positive stalls.

* tag 'rcu-next-v6.7' of git://git.kernel.org/pub/scm/linux/kernel/git/frederic/linux-dynticks: (56 commits)
  srcu: Only accelerate on enqueue time
  locktorture: Check the correct variable for allocation failure
  srcu: Fix callbacks acceleration mishandling
  rcu: Comment why callbacks migration can't wait for CPUHP_RCUTREE_PREP
  rcu: Standardize explicit CPU-hotplug calls
  rcu: Conditionally build CPU-hotplug teardown callbacks
  rcu: Remove references to rcu_migrate_callbacks() from diagrams
  rcu: Assume rcu_report_dead() is always called locally
  rcu: Assume IRQS disabled from rcu_report_dead()
  rcu: Use rcu_segcblist_segempty() instead of open coding it
  rcu: kmemleak: Ignore kmemleak false positives when RCU-freeing objects
  srcu: Fix srcu_struct node grpmask overflow on 64-bit systems
  torture: Convert parse-console.sh to mktemp
  rcutorture: Traverse possible cpu to set maxcpu in rcu_nocb_toggle()
  rcutorture: Replace schedule_timeout*() 1-jiffy waits with HZ/20
  torture: Add kvm.sh --debug-info argument
  locktorture: Rename readers_bind/writers_bind to bind_readers/bind_writers
  doc: Catch-up update for locktorture module parameters
  locktorture: Add call_rcu_chains module parameter
  locktorture: Add new module parameters to lock_torture_print_module_parms()
  ...