Eric Chanudet 188f87f264 mm/mm_init: use node's number of cpus in deferred_page_init_max_threads ... x86_64 is already using the node's cpu as maximum threads. Make that the default for all archs setting DEFERRED_STRUCT_PAGE_INIT. This returns to the behavior prior making the function arch-specific with commit ecd0965069 ("mm: make deferred init's max threads arch-specific"). Setting DEFERRED_STRUCT_PAGE_INIT and testing on a few arm64 platforms shows faster deferred_init_memmap completions: | | x13s | SA8775p-ride | Ampere R137-P31 | Ampere HR330 | | | Metal, 32GB | VM, 36GB | VM, 58GB | Metal, 128GB | | | 8cpus | 8cpus | 8cpus | 32cpus | |---------|-------------|--------------|-----------------|--------------| | threads | ms (%) | ms (%) | ms (%) | ms (%) | |---------|-------------|--------------|-----------------|--------------| | 1 | 108 (0%) | 72 (0%) | 224 (0%) | 324 (0%) | | cpus | 24 (-77%) | 36 (-50%) | 40 (-82%) | 56 (-82%) | Michael Ellerman reported: : On a machine here (1TB, 40 cores, 4KB pages) the existing code gives: : : [ 0.500124] node 2 deferred pages initialised in 210ms : [ 0.515790] node 3 deferred pages initialised in 230ms : [ 0.516061] node 0 deferred pages initialised in 230ms : [ 0.516522] node 7 deferred pages initialised in 230ms : [ 0.516672] node 4 deferred pages initialised in 230ms : [ 0.516798] node 6 deferred pages initialised in 230ms : [ 0.517051] node 5 deferred pages initialised in 230ms : [ 0.523887] node 1 deferred pages initialised in 240ms : : vs with the patch: : : [ 0.379613] node 0 deferred pages initialised in 90ms : [ 0.380388] node 1 deferred pages initialised in 90ms : [ 0.380540] node 4 deferred pages initialised in 100ms : [ 0.390239] node 6 deferred pages initialised in 100ms : [ 0.390249] node 2 deferred pages initialised in 100ms : [ 0.390786] node 3 deferred pages initialised in 110ms : [ 0.396721] node 5 deferred pages initialised in 110ms : [ 0.397095] node 7 deferred pages initialised in 110ms : : Which is a nice speedup. [echanude@redhat.com: v3] Link: https://lkml.kernel.org/r/20240528185455.643227-4-echanude@redhat.com Link: https://lkml.kernel.org/r/20240522203758.626932-4-echanude@redhat.com Signed-off-by: Eric Chanudet <echanude@redhat.com> Tested-by: Michael Ellerman <mpe@ellerman.id.au> (powerpc) Reviewed-by: Baoquan He <bhe@redhat.com> Acked-by: Alexander Gordeev <agordeev@linux.ibm.com> Acked-by: Mike Rapoport (IBM) <rppt@kernel.org> Cc: Andy Lutomirski <luto@kernel.org> Cc: Borislav Petkov (AMD) <bp@alien8.de> Cc: Dave Hansen <dave.hansen@linux.intel.com> Cc: "H. Peter Anvin" <hpa@zytor.com> Cc: Ingo Molnar <mingo@redhat.com> Cc: Nicholas Piggin <npiggin@gmail.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Thomas Gleixner <tglx@linutronix.de> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>		2024-07-03 19:29:58 -07:00
..
alpha	mseal: wire up mseal syscall	2024-05-23 19:40:26 -07:00
arc	bpf-for-netdev	2024-05-27 16:26:30 -07:00
arm	SoC fixes for 6.10	2024-06-22 07:58:21 -07:00
arm64	hardening fixes for v6.10-rc6	2024-06-28 16:11:02 -07:00
csky	syscalls: mmap(): use unsigned offset type consistently	2024-06-25 15:57:38 +02:00
hexagon	hexagon: fix fadvise64_64 calling conventions	2024-06-25 15:57:37 +02:00
loongarch	mm: implement update_mmu_tlb() using update_mmu_tlb_range()	2024-07-03 19:29:57 -07:00
m68k	mseal: wire up mseal syscall	2024-05-23 19:40:26 -07:00
microblaze	syscalls: mmap(): use unsigned offset type consistently	2024-06-25 15:57:38 +02:00
mips	mm: implement update_mmu_tlb() using update_mmu_tlb_range()	2024-07-03 19:29:57 -07:00
nios2	Kbuild updates for v6.10	2024-05-18 12:39:20 -07:00
openrisc	openrisc: Move FPU state out of pt_regs	2024-04-15 15:20:39 +01:00
parisc	parisc: use generic sys_fanotify_mark implementation	2024-06-25 15:57:25 +02:00
powerpc	powerpc: restore some missing spu syscalls	2024-06-25 15:57:26 +02:00
riscv	mm: implement update_mmu_tlb() using update_mmu_tlb_range()	2024-07-03 19:29:57 -07:00
s390	hardening fixes for v6.10-rc6	2024-06-28 16:11:02 -07:00
sh	sh: rework sync_file_range ABI	2024-06-25 15:57:26 +02:00
sparc	sparc: fix compat recv/recvfrom syscalls	2024-06-25 15:57:25 +02:00
um	This pull request contains the following changes for UML:	2024-05-25 13:17:48 -07:00
x86	mm/mm_init: use node's number of cpus in deferred_page_init_max_threads	2024-07-03 19:29:58 -07:00
xtensa	mm: implement update_mmu_tlb() using update_mmu_tlb_range()	2024-07-03 19:29:57 -07:00
.gitignore
Kconfig	Revert "mm: mmap: allow for the maximum number of bits for randomizing mmap_base by default"	2024-06-17 12:57:03 -07:00