Hello everyone, Before the Kernel Summit, I think it's good idea to post a new AutoNUMA24 and to go through a new review cycle. The last review cycle has been fundamental in improving the patchset. Thanks! The objective of AutoNUMA is to be able to perform as close as possible to (and sometime faster than) the NUMA hard CPU/memory bindings setups, without requiring the administrator to manually setup any NUMA hard bind. I hope everyone sees this is an hard problem, and what one thinks will work great in theory, when tested in practice, it may not run so great. But I'd like to remind that all research is good and valuable. All approaches to solve the problem are worthwhile, regardless if they work better/worse. sched-numa rewrite is also a very interesting approach and I hope everyone agrees that it's wonderful that both ways to solve the problem are being researched. Whatever will be merged upstream in the end won't change the fact that all work done to try to solve this hard problem is very valuable and worthwhile. git clone --reference linux -b autonuma24 git://git.kernel.org/pub/scm/linux/kernel/git/andrea/aa.git Development autonuma branch: git clone --reference linux -b autonuma git://git.kernel.org/pub/scm/linux/kernel/git/andrea/aa.git To update: git fetch git checkout -f origin/autonuma PDF with some benchmark results: http://www.kernel.org/pub/linux/kernel/people/andrea/autonuma/autonuma-vs-sched-numa-rewrite-20120817.pdf http://www.kernel.org/pub/linux/kernel/people/andrea/autonuma/autonuma_bench-20120530.pdf Changelog from AutoNUMA19 to AutoNUMA24: o Improved lots of comments and header commit messages. o Rewritten from scratch the comment at the top of kernel/sched/numa.c as the old comment wasn't well received in upstream reviews. Tried to describe the algorithm from a global view now. o Added ppc64 support. o Improved patch splitup. o Lots of code cleanups and variable renames to make the code more readable. o Try to take advantage of task_autonuma_nid before the knuma_scand is complete. o Moved some performance tuning sysfs tweaks under DEBUG_VM so they won't be visible on production kernels. o Enabled by default the working set mode for the mm_autonuma data collection. o Halved the size of the mm_autonuma structure. o scan_sleep_pass_millisecs now is more intuitive (you can can set it to 10000 to mean one pass every 10 sec, in the previous release it had to be set to 5000 to one pass every 10 sec). o Removed PF_THREAD_BOUND to allow CPU isolation. Turned the VM_BUG_ON verifying the hard binding into a WARN_ON_ONCE so the knuma_migrated can be moved by root anywhere safely. o Optimized autonuma_possible() to avoid checking num_possible_nodes() every time. o Added the math on the last_nid statistical effects from sched-numa rewrite which also introduced the last_nid logic of AutoNUMA. o Now handle systems with holes in the NUMA nodemask. Lots of num_possible_nodes() replaced with nr_node_ids (nr_node_ids not so nice name for such information). o Fixed a bug affecting KSM. KSM failed to merge pages mapped with a pte_numa pte, now it passes LTP fine. LTP found it. o Fixed repeated CPU scheduler migrate in sched_autonuma_balance() (the idle load balancing sometime was faster and it put the task back to its previous CPU before it had a chance to be scheduled on the destination CPU). o More... Changelog from AutoNUMA-alpha14 to AutoNUMA19: o sched_autonuma_balance callout location removed from schedule() now it runs in the softirq along with CFS load_balancing o lots of documentation about the math in the sched_autonuma_balance algorithm o fixed a bug in the fast path detection in sched_autonuma_balance that could decrease performance with many nodes o reduced the page_autonuma memory overhead to from 32 to 12 bytes per page o fixed a crash in __pmd_numa_fixup o knuma_numad won't scan VM_MIXEDMAP|PFNMAP (it never touched those ptes anyway) o fixed a crash in autonuma_exit o fixed a crash when split_huge_page returns 0 in knuma_migratedN as the page has been freed already o assorted cleanups and probably more Changelog from alpha13 to alpha14: o page_autonuma introduction, no memory wasted if the kernel is booted on not-NUMA hardware. Tested with flatmem/sparsemem on x86 autonuma=y/n and sparsemem/vsparsemem on x86_64 with autonuma=y/n. "noautonuma" kernel param disables autonuma permanently also when booted on NUMA hardware (no /sys/kernel/mm/autonuma, and no page_autonuma allocations, like cgroup_disable=memory) o autonuma_balance only runs along with run_rebalance_domains, to avoid altering the usual scheduler runtime. autonuma_balance gives a "kick" to the scheduler after a rebalance (it overrides the load balance activity if needed). It's not yet tested on specjbb or more schedule intensive benchmark, hopefully there's no NUMA regression. For intensive compute loads not involving a flood of scheduling activity this doesn't show any performance regression, and it avoids altering the strict schedule performance. It goes in the direction of being less intrusive with the stock scheduler runtime. Note: autonuma_balance still runs from normal context (not softirq context like run_rebalance_domains) to be able to wait on process migration (avoid _nowait), but most of the time it does nothing at all. Changelog from alpha11 to alpha13: o autonuma_balance optimization (take the fast path when process is in the preferred NUMA node) TODO: o THP native migration (orthogonal and also needed for cpuset/migrate_pages(2)/numa/sched). Andrea Arcangeli (35): autonuma: make set_pmd_at always available autonuma: export is_vma_temporary_stack() even if CONFIG_TRANSPARENT_HUGEPAGE=n autonuma: define _PAGE_NUMA_PTE and _PAGE_NUMA_PMD autonuma: pte_numa() and pmd_numa() autonuma: teach gup_fast about pmd_numa autonuma: introduce kthread_bind_node() autonuma: mm_autonuma and task_autonuma data structures autonuma: define the autonuma flags autonuma: core autonuma.h header autonuma: CPU follows memory algorithm autonuma: add page structure fields autonuma: knuma_migrated per NUMA node queues autonuma: autonuma_enter/exit autonuma: call autonuma_setup_new_exec() autonuma: alloc/free/init task_autonuma autonuma: alloc/free/init mm_autonuma autonuma: prevent select_task_rq_fair to return -1 autonuma: teach CFS about autonuma affinity autonuma: memory follows CPU algorithm and task/mm_autonuma stats collection autonuma: default mempolicy follow AutoNUMA autonuma: call autonuma_split_huge_page() autonuma: make khugepaged pte_numa aware autonuma: retain page last_nid information in khugepaged autonuma: numa hinting page faults entry points autonuma: reset autonuma page data when pages are freed autonuma: link mm/autonuma.o and kernel/sched/numa.o autonuma: add CONFIG_AUTONUMA and CONFIG_AUTONUMA_DEFAULT_ENABLED autonuma: page_autonuma autonuma: autonuma_migrate_head[0] dynamic size autonuma: bugcheck page_autonuma fields on newly allocated pages autonuma: shrink the per-page page_autonuma struct size autonuma: boost khugepaged scanning rate autonuma: make the AUTONUMA_SCAN_PMD_FLAG conditional to CONFIG_HAVE_ARCH_AUTONUMA_SCAN_PMD autonuma: add knuma_migrated/allow_first_fault in sysfs autonuma: add mm_autonuma working set estimation Vaidyanathan Srinivasan (1): autonuma: powerpc port arch/Kconfig | 6 + arch/powerpc/Kconfig | 6 + arch/powerpc/include/asm/pgtable.h | 48 +- arch/powerpc/include/asm/pte-hash64-64k.h | 4 +- arch/powerpc/mm/numa.c | 3 +- arch/x86/Kconfig | 2 + arch/x86/include/asm/paravirt.h | 2 - arch/x86/include/asm/pgtable.h | 65 ++- arch/x86/include/asm/pgtable_types.h | 28 + arch/x86/mm/gup.c | 13 +- arch/x86/mm/numa.c | 6 +- arch/x86/mm/numa_32.c | 3 +- fs/exec.c | 7 + include/asm-generic/pgtable.h | 12 + include/linux/autonuma.h | 72 ++ include/linux/autonuma_flags.h | 168 +++ include/linux/autonuma_list.h | 100 ++ include/linux/autonuma_sched.h | 50 + include/linux/autonuma_types.h | 169 +++ include/linux/huge_mm.h | 6 +- include/linux/kthread.h | 1 + include/linux/memory_hotplug.h | 3 +- include/linux/mm_types.h | 5 + include/linux/mmzone.h | 38 + include/linux/page_autonuma.h | 59 + include/linux/sched.h | 3 + init/main.c | 2 + kernel/fork.c | 18 + kernel/kthread.c | 21 + kernel/sched/Makefile | 1 + kernel/sched/core.c | 1 + kernel/sched/fair.c | 86 ++- kernel/sched/numa.c | 604 ++++++++++ kernel/sched/sched.h | 19 + mm/Kconfig | 17 + mm/Makefile | 1 + mm/autonuma.c | 1727 +++++++++++++++++++++++++++++ mm/autonuma_list.c | 169 +++ mm/huge_memory.c | 78 ++- mm/memory.c | 31 + mm/memory_hotplug.c | 2 +- mm/mempolicy.c | 12 +- mm/mmu_context.c | 3 + mm/page_alloc.c | 7 +- mm/page_autonuma.c | 248 +++++ mm/sparse.c | 126 ++- 46 files changed, 4014 insertions(+), 38 deletions(-) create mode 100644 include/linux/autonuma.h create mode 100644 include/linux/autonuma_flags.h create mode 100644 include/linux/autonuma_list.h create mode 100644 include/linux/autonuma_sched.h create mode 100644 include/linux/autonuma_types.h create mode 100644 include/linux/page_autonuma.h create mode 100644 kernel/sched/numa.c create mode 100644 mm/autonuma.c create mode 100644 mm/autonuma_list.c create mode 100644 mm/page_autonuma.c -- To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to majordomo@xxxxxxxxx. For more info on Linux MM, see: http://www.linux-mm.org/ . Don't email: <a href=mailto:"dont@xxxxxxxxx";> email@xxxxxxxxx </a>