Changes in smp.c

6142be7e - powerpc: Split systemcfg struct definitions out from vdso

Thu, 10 Oct 2024 07:01:29 +0000

powerpc: Split systemcfg struct definitions out from vdsoThe systemcfg data has nothing to do anymore with the vdso.Split it into a dedicated header file.Signed-off-by: Thomas Weißschuh Signed-off-by: Thomas Gleixner Link: https://lore.kernel.org/all/20241010-vdso-generic-base-v1-27-b64f0842d512@linutronix.de List of files: /linux-6.15/arch/powerpc/kernel/smp.c

1184674d - powerpc: Split systemcfg data out of vdso data page

Thu, 10 Oct 2024 07:01:28 +0000

powerpc: Split systemcfg data out of vdso data pageThe systemcfg data only has minimal overlap with the vdso data.Splitting the two avoids mapping the implementation-defined vdso datainto /proc/ppc64/systemcfg.It is also a preparation for the standardization of vdso data storage.The only field actually used by both systemcfg and vdso istb_ticks_per_sec and it is only changed once during time_init().Initialize it in both structures there.Signed-off-by: Thomas Weißschuh Signed-off-by: Thomas Gleixner Link: https://lore.kernel.org/all/20241010-vdso-generic-base-v1-26-b64f0842d512@linutronix.de List of files: /linux-6.15/arch/powerpc/kernel/smp.c

1d070854 - smp: Mark smp_prepare_boot_cpu() __init

Sat, 07 Sep 2024 08:27:20 +0000

smp: Mark smp_prepare_boot_cpu() __initsmp_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 List of files: /linux-6.15/arch/powerpc/kernel/smp.c

0ddbbb89 - powerpc: Fix typos

Wed, 03 Jan 2024 23:16:04 +0000

powerpc: Fix typosFix typos, most reported by "codespell arch/powerpc". Only touchescomments, no code changes.Signed-off-by: Bjorn Helgaas Signed-off-by: Michael Ellerman Link: https://msgid.link/20240103231605.1801364-8-helgaas@kernel.org List of files: /linux-6.15/arch/powerpc/kernel/smp.c

5c4233cc - powerpc/kdump: Split KEXEC_CORE and CRASH_DUMP dependency

Mon, 26 Feb 2024 10:30:10 +0000

powerpc/kdump: Split KEXEC_CORE and CRASH_DUMP dependencyRemove CONFIG_CRASH_DUMP dependency on CONFIG_KEXEC. CONFIG_KEXEC_COREwas used at places where CONFIG_CRASH_DUMP or CONFIG_CRASH_RESERVE wasappropriate. Replace with appropriate #ifdefs to support CONFIG_KEXECand !CONFIG_CRASH_DUMP configuration option. Also, make CONFIG_FA_DUMPdependent on CONFIG_CRASH_DUMP to avoid unmet dependencies for FA_DUMPwith !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 List of files: /linux-6.15/arch/powerpc/kernel/smp.c

54de4427 - sched/topology: Rename SD_SHARE_PKG_RESOURCES to SD_SHARE_LLC

Sat, 10 Feb 2024 11:39:23 +0000

sched/topology: Rename SD_SHARE_PKG_RESOURCES to SD_SHARE_LLCSD_SHARE_PKG_RESOURCES is a bit of a misnomer: its naming suggests thatit's sharing all 'package resources' - while in reality it's specificallyfor 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 List of files: /linux-6.15/arch/powerpc/kernel/smp.c

c4697571 - powerpc/smp: Dynamically build Powerpc topology

Thu, 14 Dec 2023 18:07:15 +0000

powerpc/smp: Dynamically build Powerpc topologyCurrently there are four Powerpc specific sched topologies. These areall statically defined. However not all these topologies are used byall Powerpc systems.To avoid unnecessary degenerations by the scheduler, masks and flagsare compared. However if the sched topologies are build dynamically thenthe code is simpler and there are greater chances of avoidingdegenerations.Note:Even X86 builds its sched topologies dynamically and proposed changesare 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 List of files: /linux-6.15/arch/powerpc/kernel/smp.c

0e93f1c7 - powerpc/smp: Avoid asym packing within thread_group of a core

Thu, 14 Dec 2023 18:07:14 +0000

powerpc/smp: Avoid asym packing within thread_group of a corePowerVM Hypervisor will schedule at a core granularity. However eachcore can have more than one thread_groups. For better utilization incase of a shared processor, its preferable for the scheduler to pack tothe lowest core. However there is no benefit of moving a thread betweentwo 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 List of files: /linux-6.15/arch/powerpc/kernel/smp.c

fd535a85 - powerpc/smp: Add __ro_after_init attribute

Thu, 14 Dec 2023 18:07:13 +0000

powerpc/smp: Add __ro_after_init attributeThere are some variables that are only updated at boot time.So add __ro_after_init attribute to such variablesSigned-off-by: Srikar Dronamraju Signed-off-by: Michael Ellerman Link: https://msgid.link/20231214180720.310852-4-srikar@linux.vnet.ibm.com List of files: /linux-6.15/arch/powerpc/kernel/smp.c

0e1c1986 - powerpc/smp: Disable MC domain for shared processor

Thu, 14 Dec 2023 18:07:12 +0000

powerpc/smp: Disable MC domain for shared processorLike L2-cache info, coregroup information which is used to determine MCsched domains is only present on dedicated LPARs. i.e PowerVM doesn'texport coregroup information for shared processor LPARs. Hence disablecreating 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 List of files: /linux-6.15/arch/powerpc/kernel/smp.c

aa80c634 - powerpc/smp: Enable Asym packing for cores on shared processor

Thu, 14 Dec 2023 18:07:11 +0000

powerpc/smp: Enable Asym packing for cores on shared processorIf there are shared processor LPARs, underlying Hypervisor can have morevirtual cores to handle than actual physical cores.Starting with Power 9, a big core (aka SMT8 core) has 2 nearlyindependent thread groups. On a shared processors LPARs, it helps topack threads to lesser number of cores so that the overall systemperformance and utilization improves. PowerVM schedules at a big corelevel. Hence packing to fewer cores helps.Since each thread-group is independent, running threads on both thethread-groups of a SMT8 core, should have a minimal adverse impact innon over provisioned scenarios. These changes in this patchset will notaffect in the over provisioned scenario. If there are more threads thanSMT domains, then asym_packing will not kick-inFor example: Lets says there are two 8-core Shared LPARs that areactually sharing a 8 Core shared physical pool, each running 8 threadseach. Then Consolidating 8 threads to 4 cores on each LPAR would helpthem to perform better. This is because each of the LPAR will get100% time to run applications and there will no switching required bythe Hypervisor.To achieve this, enable SD_ASYM_PACKING flag at CACHE, MC and DIE levelwhen 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 List of files: /linux-6.15/arch/powerpc/kernel/smp.c

f577cd57 - sched/topology: Rename 'DIE' domain to 'PKG'

Wed, 12 Jul 2023 14:10:56 +0000

sched/topology: Rename 'DIE' domain to 'PKG'While reworking the x86 topology code Thomas tripped over creating a 'DIE' domainfor the package mask. :-)Since these names are CONFIG_SCHED_DEBUG=y only, rename them to make thename less ambiguous.[ Shrikanth Hegde: rename on s390 as well. ][ Valentin Schneider: also rename it in the comments. ][ mingo: port to recent kernels & find all remaining occurances. ]Reported-by: Thomas Gleixner Signed-off-by: Peter Zijlstra (Intel) Signed-off-by: Ingo Molnar Acked-by: Valentin Schneider Acked-by: Mel Gorman Acked-by: Heiko Carstens Acked-by: Gautham R. Shenoy Acked-by: Vincent Guittot Link: https://lore.kernel.org/r/20230712141056.GI3100107@hirez.programming.kicks-ass.net List of files: /linux-6.15/arch/powerpc/kernel/smp.c

448e9f34 - rcu: Standardize explicit CPU-hotplug calls

Fri, 08 Sep 2023 20:36:01 +0000

rcu: Standardize explicit CPU-hotplug callsrcu_report_dead() and rcutree_migrate_callbacks() have their headers inrcupdate.h while those are pure rcutree calls, like the other CPU-hotplugfunctions.Also rcu_cpu_starting() and rcu_report_dead() have different namingconventions while they mirror each other's effects.Fix the headers and propose a naming that relates both functions andaligns with the prefix of other rcutree CPU-hotplug functions.Reviewed-by: Paul E. McKenney Signed-off-by: Frederic Weisbecker List of files: /linux-6.15/arch/powerpc/kernel/smp.c

73c58e7e - powerpc: Add HOTPLUG_SMT support

Wed, 05 Jul 2023 14:51:42 +0000

powerpc: Add HOTPLUG_SMT supportAdd support for HOTPLUG_SMT, which enables the generic sysfs SMT supportfiles in /sys/devices/system/cpu/smt, as well as the "nosmt" bootparameter.Implement the recently added hooks to allow partial SMT states, allowany number of threads per core.Tie the config symbol to HOTPLUG_CPU, which enables it on the majorplatforms that support SMT. If there are other platforms that want theSMT support that can be tweaked in future.Signed-off-by: Michael Ellerman [ldufour: remove topology_smt_supported][ldufour: remove topology_smt_threads_supported][ldufour: select CONFIG_SMT_NUM_THREADS_DYNAMIC][ldufour: update kernel-parameters.txt]Signed-off-by: Laurent Dufour Link: https://msgid.link/20230705145143.40545-10-ldufour@linux.ibm.com List of files: /linux-6.15/arch/powerpc/kernel/smp.c

c3c2e937 - powerpc: Account mm_cpumask and active_cpus in init_mm

Wed, 24 May 2023 06:08:18 +0000

powerpc: Account mm_cpumask and active_cpus in init_mminit_mm mm_cpumask and context.active_cpus is not maintained at bootand hotplug. This seems to be harmless because init_mm does not have auserspace and so never gets user TLBs flushed, but it looks odd and itprevents some sanity checks being added.Signed-off-by: Nicholas Piggin Signed-off-by: Michael Ellerman Link: https://msgid.link/20230524060821.148015-2-npiggin@gmail.com List of files: /linux-6.15/arch/powerpc/kernel/smp.c

8608f14b - powerpc: Annotate accesses to ipi message flags

Wed, 10 May 2023 03:31:13 +0000

powerpc: Annotate accesses to ipi message flagsIPI message flags are observed and consequently consumed in thesmp_ipi_demux_relaxed function, which handles these message sourcesuntil it observes none more arriving. Mark the checked loop guard withREAD_ONCE, to signal to KCSAN that the read is known to be volatile, andthat non-determinism is expected. Mark write for message source insmp_muxed_ipi_set_message().Signed-off-by: Rohan McLure Reviewed-by: Nicholas Piggin Signed-off-by: Michael Ellerman Link: https://msgid.link/20230510033117.1395895-8-rmclure@linux.ibm.com List of files: /linux-6.15/arch/powerpc/kernel/smp.c

0f613bfa - locking/atomic: treewide: use raw_atomic*_()

Mon, 05 Jun 2023 07:01:15 +0000

locking/atomic: treewide: use raw_atomic*_()Now that we have raw_atomic*_() definitions, there's no need to usearch_atomic*_() definitions outside of the low-level atomicdefinitions.Move treewide users of arch_atomic*_() over to the equivalentraw_atomic*_().There should be no functional change as a result of this patch.Signed-off-by: Mark Rutland Signed-off-by: Peter Zijlstra (Intel) Reviewed-by: Kees Cook Link: https://lore.kernel.org/r/20230605070124.3741859-19-mark.rutland@arm.com List of files: /linux-6.15/arch/powerpc/kernel/smp.c

514ca14e - start_kernel: Add __no_stack_protector function attribute

Mon, 17 Apr 2023 22:00:05 +0000

start_kernel: Add __no_stack_protector function attributeBack during the discussion ofcommit a9a3ed1eff36 ("x86: Fix early boot crash on gcc-10, third try")we discussed the need for a function attribute to control the omissionof stack protectors on a per-function basis; at the time Clang hadsupport for no_stack_protector but GCC did not. This was fixed ingcc-11. Now that the function attribute is available, let's start usingit.Callers of boot_init_stack_canary need to use this function attributeunless they're compiled with -fno-stack-protector, otherwise the canarystored in the stack slot of the caller will differ upon the call toboot_init_stack_canary. This will lead to a call to __stack_chk_fail()then panic.Link: https://gcc.gnu.org/bugzilla/show_bug.cgi?id=94722Link: https://lore.kernel.org/all/20200316130414.GC12561@hirez.programming.kicks-ass.net/Tested-by: Nathan Chancellor Acked-by: Michael Ellerman (powerpc)Acked-by: Miguel Ojeda Acked-by: Peter Zijlstra (Intel) Signed-off-by: Nick Desaulniers Link: https://lore.kernel.org/r/20230412-no_stackp-v2-1-116f9fe4bbe7@google.comSigned-off-by: Josh Poimboeuf Signed-off-by: ndesaulniers@google.com List of files: /linux-6.15/arch/powerpc/kernel/smp.c

aa464ba9 - lazy tlb: introduce lazy tlb mm refcount helper functions

Fri, 03 Feb 2023 07:18:34 +0000

lazy tlb: introduce lazy tlb mm refcount helper functionsAdd explicit _lazy_tlb annotated functions for lazy tlb mm refcounting. This makes the lazy tlb mm references more obvious, and allows therefcounting scheme to be modified in later changes. There is nofunctional change with this patch.Link: https://lkml.kernel.org/r/20230203071837.1136453-3-npiggin@gmail.comSigned-off-by: Nicholas Piggin Acked-by: Linus Torvalds Cc: Andy Lutomirski Cc: Catalin Marinas Cc: Christophe Leroy Cc: Dave Hansen Cc: Michael Ellerman Cc: Nadav Amit Cc: Peter Zijlstra Cc: Rik van Riel Cc: Will Deacon Signed-off-by: Andrew Morton List of files: /linux-6.15/arch/powerpc/kernel/smp.c

4c8c3c7f - treewide: Trace IPIs sent via smp_send_reschedule()

Tue, 07 Mar 2023 14:35:56 +0000

treewide: Trace IPIs sent via smp_send_reschedule()To be able to trace invocations of smp_send_reschedule(), rename thearch-specific definitions of it to arch_smp_send_reschedule() and wrap itinto an smp_send_reschedule() that contains a tracepoint.Changes to include the declaration of the tracepoint were driven by thefollowing coccinelle script: @func_use@ @@ smp_send_reschedule(...); @include@ @@ #include @no_include depends on func_use && !include@ @@ #include <...> + + #include [csky bits][riscv bits]Signed-off-by: Valentin Schneider Signed-off-by: Peter Zijlstra (Intel) Acked-by: Guo Ren Acked-by: Palmer Dabbelt Link: https://lore.kernel.org/r/20230307143558.294354-6-vschneid@redhat.com List of files: /linux-6.15/arch/powerpc/kernel/smp.c