Message ID | 20231113170157.280181-1-zi.yan@sent.com (mailing list archive) |
---|---|
Headers | show |
Series | Enable >0 order folio memory compaction | expand |
On 13/11/2023 17:01, Zi Yan wrote: > From: Zi Yan <ziy@nvidia.com> > > Hi all, > > This patchset enables >0 order folio memory compaction, which is one of > the prerequisitions for large folio support[1]. It is on top of > mm-everything-2023-10-21-02-40 (also applies to mm-everything-2023-11-10-20-42 > without conflicts). > > Per patch vm-scalability results on a 8-CPU (Intel Xeon E5-2650 v4 @ 2.20GHz) > 16G VM are at the bottom and based on mm-everything-2023-10-21-02-40. > Each has 5 runs. Highlights: > > 1. mmap-pread-seq-mt: -18.3% if split folios during compaction and -7.7% > if migrating folios during compaction. > > 2. lru-file-readtwice: +41.5% if split folios during compaction and > +30.4% if migrating folios during compaction. > > 3. lru-file-mmap-read: +35.1% if split folios during compaction and > +32.3% if migrating folios during compaction. > > I will run experiments on bare metal and report results later. > > > Changelog > === > > From RFC: > 1. Enabled >0 order folio compaction in the first patch by splitting all > to-be-migrated folios. (per Huang, Ying) > > 2. Stopped isolating compound pages with order greater than cc->order > to avoid wasting effort, since cc->order gives a hint that no free pages > with order greater than it exist, thus migrating the compound pages will fail. > (per Baolin Wang) > > 3. Retained the folio check within lru lock. (per Baolin Wang) > > 4. Made isolate_freepages_block() generate order-sorted multi lists. > (per Johannes Weiner) > > > Overview > === > > To support >0 order folio compaction, the patchset changes how free pages used > for migration are kept during compaction. Free pages used to be split into > order-0 pages that are post allocation processed (i.e., PageBuddy flag cleared, > page order stored in page->private is zeroed, and page reference is set to 1). > Now all free pages are kept in a MAX_ORDER+1 array of page lists based > on their order without post allocation process. When migrate_pages() asks for > a new page, one of the free pages, based on the requested page order, is > then processed and given out. > > > Optimizations > === > > 1. Free page split is added to increase migration success rate in case > a source page does not have a matched free page in the free page lists. > Free page merge is possible but not implemented, since existing > PFN-based buddy page merge algorithm requires the identification of > buddy pages, but free pages kept for memory compaction cannot have > PageBuddy set to avoid confusing other PFN scanners. > > 2. Sort source pages in ascending order before migration is added to > reduce free page split. Otherwise, high order free pages might be > prematurely split, causing undesired high order folio migration failures. > > > TODOs > === > > 1. Refactor free page post allocation and free page preparation code so > that compaction_alloc() and compaction_free() can call functions instead > of hard coding. > > 2. Make isolate_migratepages_block() produce a sorted list for > compaction to avoid sort_folio_by_order() in Patch 4. > > > Feel free to give comments and ask questions. > > Thanks. > > vm-scalability results > === > > ========================================================================================= > compiler/kconfig/rootfs/runtime/tbox_group/test/testcase: > gcc-13/defconfig/debian/300s/qemu-vm/mmap-xread-seq-mt/vm-scalability > > commit: > 6.6.0-rc4-mm-everything-2023-10-21-02-40+ > 6.6.0-rc4-split-folio-in-compaction+ > 6.6.0-rc4-folio-migration-in-compaction+ > 6.6.0-rc4-folio-migration-free-page-split+ > 6.6.0-rc4-folio-migration-free-page-split-sort-src+ > > 6.6.0-rc4-mm-eve 6.6.0-rc4-split-folio-in-co 6.6.0-rc4-folio-migration-i 6.6.0-rc4-folio-migration-f 6.6.0-rc4-folio-migration-f > ---------------- --------------------------- --------------------------- --------------------------- --------------------------- > %stddev %change %stddev %change %stddev %change %stddev %change %stddev > \ | \ | \ | \ | \ > 12896955 +2.7% 13249322 -4.0% 12385175 ± 5% +1.1% 13033951 -0.4% 12845698 vm-scalability.throughput Hi Zi, Are you able to add any commentary to these results as I'm struggling to interpret them; Is a positive or negative change better (are they times or rates?). What are the stddev values? The title suggests percent but the values are huge - I'm trying to understand what the error bars look like - are the swings real or noise? Thanks, Ryan > > ========================================================================================= > compiler/kconfig/rootfs/runtime/tbox_group/test/testcase: > gcc-13/defconfig/debian/300s/qemu-vm/mmap-pread-seq/vm-scalability > > commit: > 6.6.0-rc4-mm-everything-2023-10-21-02-40+ > 6.6.0-rc4-split-folio-in-compaction+ > 6.6.0-rc4-folio-migration-in-compaction+ > 6.6.0-rc4-folio-migration-free-page-split+ > 6.6.0-rc4-folio-migration-free-page-split-sort-src+ > > 6.6.0-rc4-mm-eve 6.6.0-rc4-split-folio-in-co 6.6.0-rc4-folio-migration-i 6.6.0-rc4-folio-migration-f 6.6.0-rc4-folio-migration-f > ---------------- --------------------------- --------------------------- --------------------------- --------------------------- > %stddev %change %stddev %change %stddev %change %stddev %change %stddev > \ | \ | \ | \ | \ > 14626665 -5.5% 13825048 -1.7% 14384837 -1.9% 14345958 -2.0% 14338834 vm-scalability.throughput > > ========================================================================================= > compiler/kconfig/rootfs/runtime/tbox_group/test/testcase: > gcc-13/defconfig/debian/300s/qemu-vm/mmap-pread-seq-mt/vm-scalability > > commit: > 6.6.0-rc4-mm-everything-2023-10-21-02-40+ > 6.6.0-rc4-split-folio-in-compaction+ > 6.6.0-rc4-folio-migration-in-compaction+ > 6.6.0-rc4-folio-migration-free-page-split+ > 6.6.0-rc4-folio-migration-free-page-split-sort-src+ > > 6.6.0-rc4-mm-eve 6.6.0-rc4-split-folio-in-co 6.6.0-rc4-folio-migration-i 6.6.0-rc4-folio-migration-f 6.6.0-rc4-folio-migration-f > ---------------- --------------------------- --------------------------- --------------------------- --------------------------- > %stddev %change %stddev %change %stddev %change %stddev %change %stddev > \ | \ | \ | \ | \ > 14314960 -18.3% 11691969 ± 2% -7.7% 13217795 -8.8% 13057395 -10.8% 12764204 vm-scalability.throughput > > ========================================================================================= > compiler/kconfig/rootfs/runtime/tbox_group/test/testcase: > gcc-13/defconfig/debian/300s/qemu-vm/lru-file-readtwice/vm-scalability > > commit: > 6.6.0-rc4-mm-everything-2023-10-21-02-40+ > 6.6.0-rc4-split-folio-in-compaction+ > 6.6.0-rc4-folio-migration-in-compaction+ > 6.6.0-rc4-folio-migration-free-page-split+ > 6.6.0-rc4-folio-migration-free-page-split-sort-src+ > > 6.6.0-rc4-mm-eve 6.6.0-rc4-split-folio-in-co 6.6.0-rc4-folio-migration-i 6.6.0-rc4-folio-migration-f 6.6.0-rc4-folio-migration-f > ---------------- --------------------------- --------------------------- --------------------------- --------------------------- > %stddev %change %stddev %change %stddev %change %stddev %change %stddev > \ | \ | \ | \ | \ > 6498203 ± 6% +41.5% 9193356 +30.4% 8472024 +30.0% 8449054 ± 3% +32.3% 8596781 ± 2% vm-scalability.throughput > > ========================================================================================= > compiler/kconfig/rootfs/runtime/tbox_group/test/testcase: > gcc-13/defconfig/debian/300s/qemu-vm/lru-file-mmap-read/vm-scalability > > commit: > 6.6.0-rc4-mm-everything-2023-10-21-02-40+ > 6.6.0-rc4-split-folio-in-compaction+ > 6.6.0-rc4-folio-migration-in-compaction+ > 6.6.0-rc4-folio-migration-free-page-split+ > 6.6.0-rc4-folio-migration-free-page-split-sort-src+ > > 6.6.0-rc4-mm-eve 6.6.0-rc4-split-folio-in-co 6.6.0-rc4-folio-migration-i 6.6.0-rc4-folio-migration-f 6.6.0-rc4-folio-migration-f > ---------------- --------------------------- --------------------------- --------------------------- --------------------------- > %stddev %change %stddev %change %stddev %change %stddev %change %stddev > \ | \ | \ | \ | \ > 5440732 ± 10% +35.1% 7352631 ± 9% +32.3% 7200622 ± 5% +27.0% 6909673 ± 4% +25.9% 6849852 ± 5% vm-scalability.throughput > > ========================================================================================= > compiler/kconfig/rootfs/runtime/tbox_group/test/testcase: > gcc-13/defconfig/debian/300s/qemu-vm/lru-file-mmap-read-rand/vm-scalability > > commit: > 6.6.0-rc4-mm-everything-2023-10-21-02-40+ > 6.6.0-rc4-split-folio-in-compaction+ > 6.6.0-rc4-folio-migration-in-compaction+ > 6.6.0-rc4-folio-migration-free-page-split+ > 6.6.0-rc4-folio-migration-free-page-split-sort-src+ > > 6.6.0-rc4-mm-eve 6.6.0-rc4-split-folio-in-co 6.6.0-rc4-folio-migration-i 6.6.0-rc4-folio-migration-f 6.6.0-rc4-folio-migration-f > ---------------- --------------------------- --------------------------- --------------------------- --------------------------- > %stddev %change %stddev %change %stddev %change %stddev %change %stddev > \ | \ | \ | \ | \ > 1627989 +0.3% 1632141 +1.8% 1656877 +0.1% 1629816 -0.2% 1625213 vm-scalability.throughput > > ========================================================================================= > compiler/kconfig/rootfs/runtime/size/tbox_group/test/testcase: > gcc-13/defconfig/debian/300s/8T/qemu-vm/anon-wx-seq-mt/vm-scalability > > commit: > 6.6.0-rc4-mm-everything-2023-10-21-02-40+ > 6.6.0-rc4-split-folio-in-compaction+ > 6.6.0-rc4-folio-migration-in-compaction+ > 6.6.0-rc4-folio-migration-free-page-split+ > 6.6.0-rc4-folio-migration-free-page-split-sort-src+ > > 6.6.0-rc4-mm-eve 6.6.0-rc4-split-folio-in-co 6.6.0-rc4-folio-migration-i 6.6.0-rc4-folio-migration-f 6.6.0-rc4-folio-migration-f > ---------------- --------------------------- --------------------------- --------------------------- --------------------------- > %stddev %change %stddev %change %stddev %change %stddev %change %stddev > \ | \ | \ | \ | \ > 6738043 +1.4% 6835096 +1.4% 6834329 +1.0% 6805401 +1.2% 6820637 vm-scalability.throughput > > ========================================================================================= > compiler/kconfig/rootfs/runtime/size/tbox_group/test/testcase: > gcc-13/defconfig/debian/300s/8T/qemu-vm/anon-w-seq-mt/vm-scalability > > commit: > 6.6.0-rc4-mm-everything-2023-10-21-02-40+ > 6.6.0-rc4-split-folio-in-compaction+ > 6.6.0-rc4-folio-migration-in-compaction+ > 6.6.0-rc4-folio-migration-free-page-split+ > 6.6.0-rc4-folio-migration-free-page-split-sort-src+ > > 6.6.0-rc4-mm-eve 6.6.0-rc4-split-folio-in-co 6.6.0-rc4-folio-migration-i 6.6.0-rc4-folio-migration-f 6.6.0-rc4-folio-migration-f > ---------------- --------------------------- --------------------------- --------------------------- --------------------------- > %stddev %change %stddev %change %stddev %change %stddev %change %stddev > \ | \ | \ | \ | \ > 7644894 +0.2% 7661425 +0.4% 7676929 +1.0% 7719218 +0.3% 7664658 vm-scalability.throughput > > > ========================================================================================= > compiler/kconfig/rootfs/runtime/size/tbox_group/test/testcase: > gcc-13/defconfig/debian/300s/512G/qemu-vm/anon-wx-rand-mt/vm-scalability > > commit: > 6.6.0-rc4-mm-everything-2023-10-21-02-40+ > 6.6.0-rc4-split-folio-in-compaction+ > 6.6.0-rc4-folio-migration-in-compaction+ > 6.6.0-rc4-folio-migration-free-page-split+ > 6.6.0-rc4-folio-migration-free-page-split-sort-src+ > > 6.6.0-rc4-mm-eve 6.6.0-rc4-split-folio-in-co 6.6.0-rc4-folio-migration-i 6.6.0-rc4-folio-migration-f 6.6.0-rc4-folio-migration-f > ---------------- --------------------------- --------------------------- --------------------------- --------------------------- > %stddev %change %stddev %change %stddev %change %stddev %change %stddev > \ | \ | \ | \ | \ > 521941 +3.3% 539254 +3.3% 539213 +1.0% 527048 +3.4% 539588 vm-scalability.throughput > > ========================================================================================= > compiler/kconfig/rootfs/runtime/size/tbox_group/test/testcase: > gcc-13/defconfig/debian/300s/512G/qemu-vm/anon-w-rand/vm-scalability > > commit: > 6.6.0-rc4-mm-everything-2023-10-21-02-40+ > 6.6.0-rc4-split-folio-in-compaction+ > 6.6.0-rc4-folio-migration-in-compaction+ > 6.6.0-rc4-folio-migration-free-page-split+ > 6.6.0-rc4-folio-migration-free-page-split-sort-src+ > > 6.6.0-rc4-mm-eve 6.6.0-rc4-split-folio-in-co 6.6.0-rc4-folio-migration-i 6.6.0-rc4-folio-migration-f 6.6.0-rc4-folio-migration-f > ---------------- --------------------------- --------------------------- --------------------------- --------------------------- > %stddev %change %stddev %change %stddev %change %stddev %change %stddev > \ | \ | \ | \ | \ > 481028 +4.3% 501876 +2.4% 492713 +0.4% 482829 ± 2% +3.5% 497649 ± 2% vm-scalability.throughput > > ========================================================================================= > compiler/kconfig/rootfs/runtime/size/tbox_group/test/testcase: > gcc-13/defconfig/debian/300s/512G/qemu-vm/anon-w-rand-mt/vm-scalability > > commit: > 6.6.0-rc4-mm-everything-2023-10-21-02-40+ > 6.6.0-rc4-split-folio-in-compaction+ > 6.6.0-rc4-folio-migration-in-compaction+ > 6.6.0-rc4-folio-migration-free-page-split+ > 6.6.0-rc4-folio-migration-free-page-split-sort-src+ > > 6.6.0-rc4-mm-eve 6.6.0-rc4-split-folio-in-co 6.6.0-rc4-folio-migration-i 6.6.0-rc4-folio-migration-f 6.6.0-rc4-folio-migration-f > ---------------- --------------------------- --------------------------- --------------------------- --------------------------- > %stddev %change %stddev %change %stddev %change %stddev %change %stddev > \ | \ | \ | \ | \ > 477063 +4.0% 496373 +4.0% 496283 +1.4% 483966 +3.2% 492216 ± 2% vm-scalability.throughput > > ========================================================================================= > compiler/kconfig/rootfs/runtime/size/tbox_group/test/testcase: > gcc-13/defconfig/debian/300s/512G/qemu-vm/anon-cow-rand/vm-scalability > > commit: > 6.6.0-rc4-mm-everything-2023-10-21-02-40+ > 6.6.0-rc4-split-folio-in-compaction+ > 6.6.0-rc4-folio-migration-in-compaction+ > 6.6.0-rc4-folio-migration-free-page-split+ > 6.6.0-rc4-folio-migration-free-page-split-sort-src+ > > 6.6.0-rc4-mm-eve 6.6.0-rc4-split-folio-in-co 6.6.0-rc4-folio-migration-i 6.6.0-rc4-folio-migration-f 6.6.0-rc4-folio-migration-f > ---------------- --------------------------- --------------------------- --------------------------- --------------------------- > %stddev %change %stddev %change %stddev %change %stddev %change %stddev > \ | \ | \ | \ | \ > 458073 +3.3% 473056 +2.6% 469835 +0.3% 459266 +3.0% 471735 vm-scalability.throughput > > ========================================================================================= > compiler/kconfig/rootfs/runtime/size/tbox_group/test/testcase: > gcc-13/defconfig/debian/300s/512G/qemu-vm/anon-cow-rand-mt/vm-scalability > > commit: > 6.6.0-rc4-mm-everything-2023-10-21-02-40+ > 6.6.0-rc4-split-folio-in-compaction+ > 6.6.0-rc4-folio-migration-in-compaction+ > 6.6.0-rc4-folio-migration-free-page-split+ > 6.6.0-rc4-folio-migration-free-page-split-sort-src+ > > 6.6.0-rc4-mm-eve 6.6.0-rc4-split-folio-in-co 6.6.0-rc4-folio-migration-i 6.6.0-rc4-folio-migration-f 6.6.0-rc4-folio-migration-f > ---------------- --------------------------- --------------------------- --------------------------- --------------------------- > %stddev %change %stddev %change %stddev %change %stddev %change %stddev > \ | \ | \ | \ | \ > 468158 +0.8% 472115 +1.4% 474547 +1.1% 473307 +2.1% 478173 vm-scalability.throughput > > ========================================================================================= > compiler/kconfig/rootfs/runtime/size/tbox_group/test/testcase: > gcc-13/defconfig/debian/300s/2T/qemu-vm/shm-xread-seq/vm-scalability > > commit: > 6.6.0-rc4-mm-everything-2023-10-21-02-40+ > 6.6.0-rc4-split-folio-in-compaction+ > 6.6.0-rc4-folio-migration-in-compaction+ > 6.6.0-rc4-folio-migration-free-page-split+ > 6.6.0-rc4-folio-migration-free-page-split-sort-src+ > > 6.6.0-rc4-mm-eve 6.6.0-rc4-split-folio-in-co 6.6.0-rc4-folio-migration-i 6.6.0-rc4-folio-migration-f 6.6.0-rc4-folio-migration-f > ---------------- --------------------------- --------------------------- --------------------------- --------------------------- > %stddev %change %stddev %change %stddev %change %stddev %change %stddev > \ | \ | \ | \ | \ > 3489946 +1.4% 3540036 +0.4% 3505413 +1.7% 3549852 +1.4% 3539309 vm-scalability.throughput > > ========================================================================================= > compiler/kconfig/rootfs/runtime/size/tbox_group/test/testcase: > gcc-13/defconfig/debian/300s/2T/qemu-vm/shm-xread-seq-mt/vm-scalability > > commit: > 6.6.0-rc4-mm-everything-2023-10-21-02-40+ > 6.6.0-rc4-split-folio-in-compaction+ > 6.6.0-rc4-folio-migration-in-compaction+ > 6.6.0-rc4-folio-migration-free-page-split+ > 6.6.0-rc4-folio-migration-free-page-split-sort-src+ > > 6.6.0-rc4-mm-eve 6.6.0-rc4-split-folio-in-co 6.6.0-rc4-folio-migration-i 6.6.0-rc4-folio-migration-f 6.6.0-rc4-folio-migration-f > ---------------- --------------------------- --------------------------- --------------------------- --------------------------- > %stddev %change %stddev %change %stddev %change %stddev %change %stddev > \ | \ | \ | \ | \ > 2621783 +1.0% 2647480 +0.6% 2636399 +0.7% 2641150 +0.1% 2624301 vm-scalability.throughput > > ========================================================================================= > compiler/kconfig/rootfs/runtime/size/tbox_group/test/testcase: > gcc-13/defconfig/debian/300s/2T/qemu-vm/shm-pread-seq/vm-scalability > > commit: > 6.6.0-rc4-mm-everything-2023-10-21-02-40+ > 6.6.0-rc4-split-folio-in-compaction+ > 6.6.0-rc4-folio-migration-in-compaction+ > 6.6.0-rc4-folio-migration-free-page-split+ > 6.6.0-rc4-folio-migration-free-page-split-sort-src+ > > 6.6.0-rc4-mm-eve 6.6.0-rc4-split-folio-in-co 6.6.0-rc4-folio-migration-i 6.6.0-rc4-folio-migration-f 6.6.0-rc4-folio-migration-f > ---------------- --------------------------- --------------------------- --------------------------- --------------------------- > %stddev %change %stddev %change %stddev %change %stddev %change %stddev > \ | \ | \ | \ | \ > 3504293 +1.5% 3557664 -0.4% 3489639 +0.4% 3516842 +0.8% 3533944 vm-scalability.throughput > > ========================================================================================= > compiler/kconfig/rootfs/runtime/size/tbox_group/test/testcase: > gcc-13/defconfig/debian/300s/256G/qemu-vm/msync/vm-scalability > > commit: > 6.6.0-rc4-mm-everything-2023-10-21-02-40+ > 6.6.0-rc4-split-folio-in-compaction+ > 6.6.0-rc4-folio-migration-in-compaction+ > 6.6.0-rc4-folio-migration-free-page-split+ > 6.6.0-rc4-folio-migration-free-page-split-sort-src+ > > 6.6.0-rc4-mm-eve 6.6.0-rc4-split-folio-in-co 6.6.0-rc4-folio-migration-i 6.6.0-rc4-folio-migration-f 6.6.0-rc4-folio-migration-f > ---------------- --------------------------- --------------------------- --------------------------- --------------------------- > %stddev %change %stddev %change %stddev %change %stddev %change %stddev > \ | \ | \ | \ | \ > 919378 ± 6% +3.8% 954028 ± 10% +7.2% 985920 ± 6% +7.4% 987077 ± 5% +14.2% 1049473 ± 2% vm-scalability.throughput > > ========================================================================================= > compiler/kconfig/rootfs/runtime/size/tbox_group/test/testcase: > gcc-13/defconfig/debian/300s/256G/qemu-vm/msync-mt/vm-scalability > > commit: > 6.6.0-rc4-mm-everything-2023-10-21-02-40+ > 6.6.0-rc4-split-folio-in-compaction+ > 6.6.0-rc4-folio-migration-in-compaction+ > 6.6.0-rc4-folio-migration-free-page-split+ > 6.6.0-rc4-folio-migration-free-page-split-sort-src+ > > 6.6.0-rc4-mm-eve 6.6.0-rc4-split-folio-in-co 6.6.0-rc4-folio-migration-i 6.6.0-rc4-folio-migration-f 6.6.0-rc4-folio-migration-f > ---------------- --------------------------- --------------------------- --------------------------- --------------------------- > %stddev %change %stddev %change %stddev %change %stddev %change %stddev > \ | \ | \ | \ | \ > 2398063 +2.0% 2446844 -1.5% 2361096 ± 2% -3.1% 2324346 ± 4% -1.0% 2375055 ± 4% vm-scalability.throughput > > ========================================================================================= > compiler/kconfig/rootfs/runtime/size/tbox_group/test/testcase: > gcc-13/defconfig/debian/300s/128G/qemu-vm/truncate/vm-scalability > > commit: > 6.6.0-rc4-mm-everything-2023-10-21-02-40+ > 6.6.0-rc4-split-folio-in-compaction+ > 6.6.0-rc4-folio-migration-in-compaction+ > 6.6.0-rc4-folio-migration-free-page-split+ > 6.6.0-rc4-folio-migration-free-page-split-sort-src+ > > 6.6.0-rc4-mm-eve 6.6.0-rc4-split-folio-in-co 6.6.0-rc4-folio-migration-i 6.6.0-rc4-folio-migration-f 6.6.0-rc4-folio-migration-f > ---------------- --------------------------- --------------------------- --------------------------- --------------------------- > %stddev %change %stddev %change %stddev %change %stddev %change %stddev > \ | \ | \ | \ | \ > 2.084e+08 ± 2% +1.5% 2.114e+08 ± 3% +6.5% 2.22e+08 ± 3% -2.3% 2.035e+08 ± 2% -2.1% 2.04e+08 ± 5% vm-scalability.throughput > > > > [1] https://lore.kernel.org/linux-mm/f8d47176-03a8-99bf-a813-b5942830fd73@arm.com/ > > Zi Yan (4): > mm/compaction: enable compacting >0 order folios. > mm/compaction: add support for >0 order folio memory compaction. > mm/compaction: optimize >0 order folio compaction with free page > split. > mm/compaction: optimize >0 order folio compaction by sorting source > pages. > > mm/compaction.c | 280 ++++++++++++++++++++++++++++++++++++------------ > mm/internal.h | 7 +- > 2 files changed, 217 insertions(+), 70 deletions(-) >
On 21 Nov 2023, at 10:46, Ryan Roberts wrote: >> >> vm-scalability results >> === >> >> ========================================================================================= >> compiler/kconfig/rootfs/runtime/tbox_group/test/testcase: >> gcc-13/defconfig/debian/300s/qemu-vm/mmap-xread-seq-mt/vm-scalability >> >> commit: >> 6.6.0-rc4-mm-everything-2023-10-21-02-40+ >> 6.6.0-rc4-split-folio-in-compaction+ >> 6.6.0-rc4-folio-migration-in-compaction+ >> 6.6.0-rc4-folio-migration-free-page-split+ >> 6.6.0-rc4-folio-migration-free-page-split-sort-src+ >> >> 6.6.0-rc4-mm-eve 6.6.0-rc4-split-folio-in-co 6.6.0-rc4-folio-migration-i 6.6.0-rc4-folio-migration-f 6.6.0-rc4-folio-migration-f >> ---------------- --------------------------- --------------------------- --------------------------- --------------------------- >> %stddev %change %stddev %change %stddev %change %stddev %change %stddev >> \ | \ | \ | \ | \ >> 12896955 +2.7% 13249322 -4.0% 12385175 ± 5% +1.1% 13033951 -0.4% 12845698 vm-scalability.throughput > > Hi Zi, > > Are you able to add any commentary to these results as I'm struggling to > interpret them; Is a positive or negative change better (are they times or > rates?). What are the stddev values? The title suggests percent but the values > are huge - I'm trying to understand what the error bars look like - are the > swings real or noise? The metric is vm-scalability.throughput, so the larger the better. Some %stddev are not present since they are too small. For 6.6.0-rc4-folio-migration-in-compaction+, %stddev is greater than %change, so the change might be noise. Also, I talked to DavidH in last THP Cabal meeting about this. He suggested that there are a lot of noise in vm-scalability like what I have here and I should run more iterations and on bare metal. I am currently rerun them on a baremetal and more iterations on the existing VM and report the results later. Please note that the runs really take some time. In addition, I will find other fragmentation-related benchmarks, so we can see the impact on memory fragmentation. -- Best Regards, Yan, Zi
On 21/11/2023 16:45, Zi Yan wrote: > On 21 Nov 2023, at 10:46, Ryan Roberts wrote: > >>> >>> vm-scalability results >>> === >>> >>> ========================================================================================= >>> compiler/kconfig/rootfs/runtime/tbox_group/test/testcase: >>> gcc-13/defconfig/debian/300s/qemu-vm/mmap-xread-seq-mt/vm-scalability >>> >>> commit: >>> 6.6.0-rc4-mm-everything-2023-10-21-02-40+ >>> 6.6.0-rc4-split-folio-in-compaction+ >>> 6.6.0-rc4-folio-migration-in-compaction+ >>> 6.6.0-rc4-folio-migration-free-page-split+ >>> 6.6.0-rc4-folio-migration-free-page-split-sort-src+ >>> >>> 6.6.0-rc4-mm-eve 6.6.0-rc4-split-folio-in-co 6.6.0-rc4-folio-migration-i 6.6.0-rc4-folio-migration-f 6.6.0-rc4-folio-migration-f >>> ---------------- --------------------------- --------------------------- --------------------------- --------------------------- >>> %stddev %change %stddev %change %stddev %change %stddev %change %stddev >>> \ | \ | \ | \ | \ >>> 12896955 +2.7% 13249322 -4.0% 12385175 ± 5% +1.1% 13033951 -0.4% 12845698 vm-scalability.throughput >> >> Hi Zi, >> >> Are you able to add any commentary to these results as I'm struggling to >> interpret them; Is a positive or negative change better (are they times or >> rates?). What are the stddev values? The title suggests percent but the values >> are huge - I'm trying to understand what the error bars look like - are the >> swings real or noise? > > The metric is vm-scalability.throughput, so the larger the better. Some %stddev > are not present since they are too small. For 6.6.0-rc4-folio-migration-in-compaction+, > %stddev is greater than %change, so the change might be noise. Ahh got it - thanks! > > Also, I talked to DavidH in last THP Cabal meeting about this. He suggested that > there are a lot of noise in vm-scalability like what I have here and I should > run more iterations and on bare metal. I am currently rerun them on a baremetal > and more iterations on the existing VM and report the results later. Please > note that the runs really take some time. Ahh ok, I'll wait for the bare metal numbers and will disregard these for now. Thanks! > > In addition, I will find other fragmentation-related benchmarks, so we can see > the impact on memory fragmentation. > > -- > Best Regards, > Yan, Zi
On 13/11/2023 17:01, Zi Yan wrote: > From: Zi Yan <ziy@nvidia.com> > > Hi all, > > This patchset enables >0 order folio memory compaction, which is one of > the prerequisitions for large folio support[1]. It is on top of > mm-everything-2023-10-21-02-40 (also applies to mm-everything-2023-11-10-20-42 > without conflicts). > > Per patch vm-scalability results on a 8-CPU (Intel Xeon E5-2650 v4 @ 2.20GHz) > 16G VM are at the bottom and based on mm-everything-2023-10-21-02-40. > Each has 5 runs. Highlights: > > 1. mmap-pread-seq-mt: -18.3% if split folios during compaction and -7.7% > if migrating folios during compaction. > > 2. lru-file-readtwice: +41.5% if split folios during compaction and > +30.4% if migrating folios during compaction. > > 3. lru-file-mmap-read: +35.1% if split folios during compaction and > +32.3% if migrating folios during compaction. > > I will run experiments on bare metal and report results later. Hi Zi, I've taken this series into a backport on v6.1 that I maintain for my folios stuff - I had to do a fair bit of massaging to get it ported, so its been a good way to review the code. With the exception of the issue that Baolin has already pointed out, it looks good to me! So, assuming the forthcoming perf numbers don't show any problem: Reviewed-by: Ryan Roberts <ryan.roberts@arm.com> > > > Changelog > === > > From RFC: > 1. Enabled >0 order folio compaction in the first patch by splitting all > to-be-migrated folios. (per Huang, Ying) > > 2. Stopped isolating compound pages with order greater than cc->order > to avoid wasting effort, since cc->order gives a hint that no free pages > with order greater than it exist, thus migrating the compound pages will fail. > (per Baolin Wang) > > 3. Retained the folio check within lru lock. (per Baolin Wang) > > 4. Made isolate_freepages_block() generate order-sorted multi lists. > (per Johannes Weiner) > > > Overview > === > > To support >0 order folio compaction, the patchset changes how free pages used > for migration are kept during compaction. Free pages used to be split into > order-0 pages that are post allocation processed (i.e., PageBuddy flag cleared, > page order stored in page->private is zeroed, and page reference is set to 1). > Now all free pages are kept in a MAX_ORDER+1 array of page lists based > on their order without post allocation process. When migrate_pages() asks for > a new page, one of the free pages, based on the requested page order, is > then processed and given out. > > > Optimizations > === > > 1. Free page split is added to increase migration success rate in case > a source page does not have a matched free page in the free page lists. > Free page merge is possible but not implemented, since existing > PFN-based buddy page merge algorithm requires the identification of > buddy pages, but free pages kept for memory compaction cannot have > PageBuddy set to avoid confusing other PFN scanners. > > 2. Sort source pages in ascending order before migration is added to > reduce free page split. Otherwise, high order free pages might be > prematurely split, causing undesired high order folio migration failures. > > > TODOs > === > > 1. Refactor free page post allocation and free page preparation code so > that compaction_alloc() and compaction_free() can call functions instead > of hard coding. > > 2. Make isolate_migratepages_block() produce a sorted list for > compaction to avoid sort_folio_by_order() in Patch 4. > > > Feel free to give comments and ask questions. > > Thanks. > > vm-scalability results > === > > ========================================================================================= > compiler/kconfig/rootfs/runtime/tbox_group/test/testcase: > gcc-13/defconfig/debian/300s/qemu-vm/mmap-xread-seq-mt/vm-scalability > > commit: > 6.6.0-rc4-mm-everything-2023-10-21-02-40+ > 6.6.0-rc4-split-folio-in-compaction+ > 6.6.0-rc4-folio-migration-in-compaction+ > 6.6.0-rc4-folio-migration-free-page-split+ > 6.6.0-rc4-folio-migration-free-page-split-sort-src+ > > 6.6.0-rc4-mm-eve 6.6.0-rc4-split-folio-in-co 6.6.0-rc4-folio-migration-i 6.6.0-rc4-folio-migration-f 6.6.0-rc4-folio-migration-f > ---------------- --------------------------- --------------------------- --------------------------- --------------------------- > %stddev %change %stddev %change %stddev %change %stddev %change %stddev > \ | \ | \ | \ | \ > 12896955 +2.7% 13249322 -4.0% 12385175 ± 5% +1.1% 13033951 -0.4% 12845698 vm-scalability.throughput > > ========================================================================================= > compiler/kconfig/rootfs/runtime/tbox_group/test/testcase: > gcc-13/defconfig/debian/300s/qemu-vm/mmap-pread-seq/vm-scalability > > commit: > 6.6.0-rc4-mm-everything-2023-10-21-02-40+ > 6.6.0-rc4-split-folio-in-compaction+ > 6.6.0-rc4-folio-migration-in-compaction+ > 6.6.0-rc4-folio-migration-free-page-split+ > 6.6.0-rc4-folio-migration-free-page-split-sort-src+ > > 6.6.0-rc4-mm-eve 6.6.0-rc4-split-folio-in-co 6.6.0-rc4-folio-migration-i 6.6.0-rc4-folio-migration-f 6.6.0-rc4-folio-migration-f > ---------------- --------------------------- --------------------------- --------------------------- --------------------------- > %stddev %change %stddev %change %stddev %change %stddev %change %stddev > \ | \ | \ | \ | \ > 14626665 -5.5% 13825048 -1.7% 14384837 -1.9% 14345958 -2.0% 14338834 vm-scalability.throughput > > ========================================================================================= > compiler/kconfig/rootfs/runtime/tbox_group/test/testcase: > gcc-13/defconfig/debian/300s/qemu-vm/mmap-pread-seq-mt/vm-scalability > > commit: > 6.6.0-rc4-mm-everything-2023-10-21-02-40+ > 6.6.0-rc4-split-folio-in-compaction+ > 6.6.0-rc4-folio-migration-in-compaction+ > 6.6.0-rc4-folio-migration-free-page-split+ > 6.6.0-rc4-folio-migration-free-page-split-sort-src+ > > 6.6.0-rc4-mm-eve 6.6.0-rc4-split-folio-in-co 6.6.0-rc4-folio-migration-i 6.6.0-rc4-folio-migration-f 6.6.0-rc4-folio-migration-f > ---------------- --------------------------- --------------------------- --------------------------- --------------------------- > %stddev %change %stddev %change %stddev %change %stddev %change %stddev > \ | \ | \ | \ | \ > 14314960 -18.3% 11691969 ± 2% -7.7% 13217795 -8.8% 13057395 -10.8% 12764204 vm-scalability.throughput > > ========================================================================================= > compiler/kconfig/rootfs/runtime/tbox_group/test/testcase: > gcc-13/defconfig/debian/300s/qemu-vm/lru-file-readtwice/vm-scalability > > commit: > 6.6.0-rc4-mm-everything-2023-10-21-02-40+ > 6.6.0-rc4-split-folio-in-compaction+ > 6.6.0-rc4-folio-migration-in-compaction+ > 6.6.0-rc4-folio-migration-free-page-split+ > 6.6.0-rc4-folio-migration-free-page-split-sort-src+ > > 6.6.0-rc4-mm-eve 6.6.0-rc4-split-folio-in-co 6.6.0-rc4-folio-migration-i 6.6.0-rc4-folio-migration-f 6.6.0-rc4-folio-migration-f > ---------------- --------------------------- --------------------------- --------------------------- --------------------------- > %stddev %change %stddev %change %stddev %change %stddev %change %stddev > \ | \ | \ | \ | \ > 6498203 ± 6% +41.5% 9193356 +30.4% 8472024 +30.0% 8449054 ± 3% +32.3% 8596781 ± 2% vm-scalability.throughput > > ========================================================================================= > compiler/kconfig/rootfs/runtime/tbox_group/test/testcase: > gcc-13/defconfig/debian/300s/qemu-vm/lru-file-mmap-read/vm-scalability > > commit: > 6.6.0-rc4-mm-everything-2023-10-21-02-40+ > 6.6.0-rc4-split-folio-in-compaction+ > 6.6.0-rc4-folio-migration-in-compaction+ > 6.6.0-rc4-folio-migration-free-page-split+ > 6.6.0-rc4-folio-migration-free-page-split-sort-src+ > > 6.6.0-rc4-mm-eve 6.6.0-rc4-split-folio-in-co 6.6.0-rc4-folio-migration-i 6.6.0-rc4-folio-migration-f 6.6.0-rc4-folio-migration-f > ---------------- --------------------------- --------------------------- --------------------------- --------------------------- > %stddev %change %stddev %change %stddev %change %stddev %change %stddev > \ | \ | \ | \ | \ > 5440732 ± 10% +35.1% 7352631 ± 9% +32.3% 7200622 ± 5% +27.0% 6909673 ± 4% +25.9% 6849852 ± 5% vm-scalability.throughput > > ========================================================================================= > compiler/kconfig/rootfs/runtime/tbox_group/test/testcase: > gcc-13/defconfig/debian/300s/qemu-vm/lru-file-mmap-read-rand/vm-scalability > > commit: > 6.6.0-rc4-mm-everything-2023-10-21-02-40+ > 6.6.0-rc4-split-folio-in-compaction+ > 6.6.0-rc4-folio-migration-in-compaction+ > 6.6.0-rc4-folio-migration-free-page-split+ > 6.6.0-rc4-folio-migration-free-page-split-sort-src+ > > 6.6.0-rc4-mm-eve 6.6.0-rc4-split-folio-in-co 6.6.0-rc4-folio-migration-i 6.6.0-rc4-folio-migration-f 6.6.0-rc4-folio-migration-f > ---------------- --------------------------- --------------------------- --------------------------- --------------------------- > %stddev %change %stddev %change %stddev %change %stddev %change %stddev > \ | \ | \ | \ | \ > 1627989 +0.3% 1632141 +1.8% 1656877 +0.1% 1629816 -0.2% 1625213 vm-scalability.throughput > > ========================================================================================= > compiler/kconfig/rootfs/runtime/size/tbox_group/test/testcase: > gcc-13/defconfig/debian/300s/8T/qemu-vm/anon-wx-seq-mt/vm-scalability > > commit: > 6.6.0-rc4-mm-everything-2023-10-21-02-40+ > 6.6.0-rc4-split-folio-in-compaction+ > 6.6.0-rc4-folio-migration-in-compaction+ > 6.6.0-rc4-folio-migration-free-page-split+ > 6.6.0-rc4-folio-migration-free-page-split-sort-src+ > > 6.6.0-rc4-mm-eve 6.6.0-rc4-split-folio-in-co 6.6.0-rc4-folio-migration-i 6.6.0-rc4-folio-migration-f 6.6.0-rc4-folio-migration-f > ---------------- --------------------------- --------------------------- --------------------------- --------------------------- > %stddev %change %stddev %change %stddev %change %stddev %change %stddev > \ | \ | \ | \ | \ > 6738043 +1.4% 6835096 +1.4% 6834329 +1.0% 6805401 +1.2% 6820637 vm-scalability.throughput > > ========================================================================================= > compiler/kconfig/rootfs/runtime/size/tbox_group/test/testcase: > gcc-13/defconfig/debian/300s/8T/qemu-vm/anon-w-seq-mt/vm-scalability > > commit: > 6.6.0-rc4-mm-everything-2023-10-21-02-40+ > 6.6.0-rc4-split-folio-in-compaction+ > 6.6.0-rc4-folio-migration-in-compaction+ > 6.6.0-rc4-folio-migration-free-page-split+ > 6.6.0-rc4-folio-migration-free-page-split-sort-src+ > > 6.6.0-rc4-mm-eve 6.6.0-rc4-split-folio-in-co 6.6.0-rc4-folio-migration-i 6.6.0-rc4-folio-migration-f 6.6.0-rc4-folio-migration-f > ---------------- --------------------------- --------------------------- --------------------------- --------------------------- > %stddev %change %stddev %change %stddev %change %stddev %change %stddev > \ | \ | \ | \ | \ > 7644894 +0.2% 7661425 +0.4% 7676929 +1.0% 7719218 +0.3% 7664658 vm-scalability.throughput > > > ========================================================================================= > compiler/kconfig/rootfs/runtime/size/tbox_group/test/testcase: > gcc-13/defconfig/debian/300s/512G/qemu-vm/anon-wx-rand-mt/vm-scalability > > commit: > 6.6.0-rc4-mm-everything-2023-10-21-02-40+ > 6.6.0-rc4-split-folio-in-compaction+ > 6.6.0-rc4-folio-migration-in-compaction+ > 6.6.0-rc4-folio-migration-free-page-split+ > 6.6.0-rc4-folio-migration-free-page-split-sort-src+ > > 6.6.0-rc4-mm-eve 6.6.0-rc4-split-folio-in-co 6.6.0-rc4-folio-migration-i 6.6.0-rc4-folio-migration-f 6.6.0-rc4-folio-migration-f > ---------------- --------------------------- --------------------------- --------------------------- --------------------------- > %stddev %change %stddev %change %stddev %change %stddev %change %stddev > \ | \ | \ | \ | \ > 521941 +3.3% 539254 +3.3% 539213 +1.0% 527048 +3.4% 539588 vm-scalability.throughput > > ========================================================================================= > compiler/kconfig/rootfs/runtime/size/tbox_group/test/testcase: > gcc-13/defconfig/debian/300s/512G/qemu-vm/anon-w-rand/vm-scalability > > commit: > 6.6.0-rc4-mm-everything-2023-10-21-02-40+ > 6.6.0-rc4-split-folio-in-compaction+ > 6.6.0-rc4-folio-migration-in-compaction+ > 6.6.0-rc4-folio-migration-free-page-split+ > 6.6.0-rc4-folio-migration-free-page-split-sort-src+ > > 6.6.0-rc4-mm-eve 6.6.0-rc4-split-folio-in-co 6.6.0-rc4-folio-migration-i 6.6.0-rc4-folio-migration-f 6.6.0-rc4-folio-migration-f > ---------------- --------------------------- --------------------------- --------------------------- --------------------------- > %stddev %change %stddev %change %stddev %change %stddev %change %stddev > \ | \ | \ | \ | \ > 481028 +4.3% 501876 +2.4% 492713 +0.4% 482829 ± 2% +3.5% 497649 ± 2% vm-scalability.throughput > > ========================================================================================= > compiler/kconfig/rootfs/runtime/size/tbox_group/test/testcase: > gcc-13/defconfig/debian/300s/512G/qemu-vm/anon-w-rand-mt/vm-scalability > > commit: > 6.6.0-rc4-mm-everything-2023-10-21-02-40+ > 6.6.0-rc4-split-folio-in-compaction+ > 6.6.0-rc4-folio-migration-in-compaction+ > 6.6.0-rc4-folio-migration-free-page-split+ > 6.6.0-rc4-folio-migration-free-page-split-sort-src+ > > 6.6.0-rc4-mm-eve 6.6.0-rc4-split-folio-in-co 6.6.0-rc4-folio-migration-i 6.6.0-rc4-folio-migration-f 6.6.0-rc4-folio-migration-f > ---------------- --------------------------- --------------------------- --------------------------- --------------------------- > %stddev %change %stddev %change %stddev %change %stddev %change %stddev > \ | \ | \ | \ | \ > 477063 +4.0% 496373 +4.0% 496283 +1.4% 483966 +3.2% 492216 ± 2% vm-scalability.throughput > > ========================================================================================= > compiler/kconfig/rootfs/runtime/size/tbox_group/test/testcase: > gcc-13/defconfig/debian/300s/512G/qemu-vm/anon-cow-rand/vm-scalability > > commit: > 6.6.0-rc4-mm-everything-2023-10-21-02-40+ > 6.6.0-rc4-split-folio-in-compaction+ > 6.6.0-rc4-folio-migration-in-compaction+ > 6.6.0-rc4-folio-migration-free-page-split+ > 6.6.0-rc4-folio-migration-free-page-split-sort-src+ > > 6.6.0-rc4-mm-eve 6.6.0-rc4-split-folio-in-co 6.6.0-rc4-folio-migration-i 6.6.0-rc4-folio-migration-f 6.6.0-rc4-folio-migration-f > ---------------- --------------------------- --------------------------- --------------------------- --------------------------- > %stddev %change %stddev %change %stddev %change %stddev %change %stddev > \ | \ | \ | \ | \ > 458073 +3.3% 473056 +2.6% 469835 +0.3% 459266 +3.0% 471735 vm-scalability.throughput > > ========================================================================================= > compiler/kconfig/rootfs/runtime/size/tbox_group/test/testcase: > gcc-13/defconfig/debian/300s/512G/qemu-vm/anon-cow-rand-mt/vm-scalability > > commit: > 6.6.0-rc4-mm-everything-2023-10-21-02-40+ > 6.6.0-rc4-split-folio-in-compaction+ > 6.6.0-rc4-folio-migration-in-compaction+ > 6.6.0-rc4-folio-migration-free-page-split+ > 6.6.0-rc4-folio-migration-free-page-split-sort-src+ > > 6.6.0-rc4-mm-eve 6.6.0-rc4-split-folio-in-co 6.6.0-rc4-folio-migration-i 6.6.0-rc4-folio-migration-f 6.6.0-rc4-folio-migration-f > ---------------- --------------------------- --------------------------- --------------------------- --------------------------- > %stddev %change %stddev %change %stddev %change %stddev %change %stddev > \ | \ | \ | \ | \ > 468158 +0.8% 472115 +1.4% 474547 +1.1% 473307 +2.1% 478173 vm-scalability.throughput > > ========================================================================================= > compiler/kconfig/rootfs/runtime/size/tbox_group/test/testcase: > gcc-13/defconfig/debian/300s/2T/qemu-vm/shm-xread-seq/vm-scalability > > commit: > 6.6.0-rc4-mm-everything-2023-10-21-02-40+ > 6.6.0-rc4-split-folio-in-compaction+ > 6.6.0-rc4-folio-migration-in-compaction+ > 6.6.0-rc4-folio-migration-free-page-split+ > 6.6.0-rc4-folio-migration-free-page-split-sort-src+ > > 6.6.0-rc4-mm-eve 6.6.0-rc4-split-folio-in-co 6.6.0-rc4-folio-migration-i 6.6.0-rc4-folio-migration-f 6.6.0-rc4-folio-migration-f > ---------------- --------------------------- --------------------------- --------------------------- --------------------------- > %stddev %change %stddev %change %stddev %change %stddev %change %stddev > \ | \ | \ | \ | \ > 3489946 +1.4% 3540036 +0.4% 3505413 +1.7% 3549852 +1.4% 3539309 vm-scalability.throughput > > ========================================================================================= > compiler/kconfig/rootfs/runtime/size/tbox_group/test/testcase: > gcc-13/defconfig/debian/300s/2T/qemu-vm/shm-xread-seq-mt/vm-scalability > > commit: > 6.6.0-rc4-mm-everything-2023-10-21-02-40+ > 6.6.0-rc4-split-folio-in-compaction+ > 6.6.0-rc4-folio-migration-in-compaction+ > 6.6.0-rc4-folio-migration-free-page-split+ > 6.6.0-rc4-folio-migration-free-page-split-sort-src+ > > 6.6.0-rc4-mm-eve 6.6.0-rc4-split-folio-in-co 6.6.0-rc4-folio-migration-i 6.6.0-rc4-folio-migration-f 6.6.0-rc4-folio-migration-f > ---------------- --------------------------- --------------------------- --------------------------- --------------------------- > %stddev %change %stddev %change %stddev %change %stddev %change %stddev > \ | \ | \ | \ | \ > 2621783 +1.0% 2647480 +0.6% 2636399 +0.7% 2641150 +0.1% 2624301 vm-scalability.throughput > > ========================================================================================= > compiler/kconfig/rootfs/runtime/size/tbox_group/test/testcase: > gcc-13/defconfig/debian/300s/2T/qemu-vm/shm-pread-seq/vm-scalability > > commit: > 6.6.0-rc4-mm-everything-2023-10-21-02-40+ > 6.6.0-rc4-split-folio-in-compaction+ > 6.6.0-rc4-folio-migration-in-compaction+ > 6.6.0-rc4-folio-migration-free-page-split+ > 6.6.0-rc4-folio-migration-free-page-split-sort-src+ > > 6.6.0-rc4-mm-eve 6.6.0-rc4-split-folio-in-co 6.6.0-rc4-folio-migration-i 6.6.0-rc4-folio-migration-f 6.6.0-rc4-folio-migration-f > ---------------- --------------------------- --------------------------- --------------------------- --------------------------- > %stddev %change %stddev %change %stddev %change %stddev %change %stddev > \ | \ | \ | \ | \ > 3504293 +1.5% 3557664 -0.4% 3489639 +0.4% 3516842 +0.8% 3533944 vm-scalability.throughput > > ========================================================================================= > compiler/kconfig/rootfs/runtime/size/tbox_group/test/testcase: > gcc-13/defconfig/debian/300s/256G/qemu-vm/msync/vm-scalability > > commit: > 6.6.0-rc4-mm-everything-2023-10-21-02-40+ > 6.6.0-rc4-split-folio-in-compaction+ > 6.6.0-rc4-folio-migration-in-compaction+ > 6.6.0-rc4-folio-migration-free-page-split+ > 6.6.0-rc4-folio-migration-free-page-split-sort-src+ > > 6.6.0-rc4-mm-eve 6.6.0-rc4-split-folio-in-co 6.6.0-rc4-folio-migration-i 6.6.0-rc4-folio-migration-f 6.6.0-rc4-folio-migration-f > ---------------- --------------------------- --------------------------- --------------------------- --------------------------- > %stddev %change %stddev %change %stddev %change %stddev %change %stddev > \ | \ | \ | \ | \ > 919378 ± 6% +3.8% 954028 ± 10% +7.2% 985920 ± 6% +7.4% 987077 ± 5% +14.2% 1049473 ± 2% vm-scalability.throughput > > ========================================================================================= > compiler/kconfig/rootfs/runtime/size/tbox_group/test/testcase: > gcc-13/defconfig/debian/300s/256G/qemu-vm/msync-mt/vm-scalability > > commit: > 6.6.0-rc4-mm-everything-2023-10-21-02-40+ > 6.6.0-rc4-split-folio-in-compaction+ > 6.6.0-rc4-folio-migration-in-compaction+ > 6.6.0-rc4-folio-migration-free-page-split+ > 6.6.0-rc4-folio-migration-free-page-split-sort-src+ > > 6.6.0-rc4-mm-eve 6.6.0-rc4-split-folio-in-co 6.6.0-rc4-folio-migration-i 6.6.0-rc4-folio-migration-f 6.6.0-rc4-folio-migration-f > ---------------- --------------------------- --------------------------- --------------------------- --------------------------- > %stddev %change %stddev %change %stddev %change %stddev %change %stddev > \ | \ | \ | \ | \ > 2398063 +2.0% 2446844 -1.5% 2361096 ± 2% -3.1% 2324346 ± 4% -1.0% 2375055 ± 4% vm-scalability.throughput > > ========================================================================================= > compiler/kconfig/rootfs/runtime/size/tbox_group/test/testcase: > gcc-13/defconfig/debian/300s/128G/qemu-vm/truncate/vm-scalability > > commit: > 6.6.0-rc4-mm-everything-2023-10-21-02-40+ > 6.6.0-rc4-split-folio-in-compaction+ > 6.6.0-rc4-folio-migration-in-compaction+ > 6.6.0-rc4-folio-migration-free-page-split+ > 6.6.0-rc4-folio-migration-free-page-split-sort-src+ > > 6.6.0-rc4-mm-eve 6.6.0-rc4-split-folio-in-co 6.6.0-rc4-folio-migration-i 6.6.0-rc4-folio-migration-f 6.6.0-rc4-folio-migration-f > ---------------- --------------------------- --------------------------- --------------------------- --------------------------- > %stddev %change %stddev %change %stddev %change %stddev %change %stddev > \ | \ | \ | \ | \ > 2.084e+08 ± 2% +1.5% 2.114e+08 ± 3% +6.5% 2.22e+08 ± 3% -2.3% 2.035e+08 ± 2% -2.1% 2.04e+08 ± 5% vm-scalability.throughput > > > > [1] https://lore.kernel.org/linux-mm/f8d47176-03a8-99bf-a813-b5942830fd73@arm.com/ > > Zi Yan (4): > mm/compaction: enable compacting >0 order folios. > mm/compaction: add support for >0 order folio memory compaction. > mm/compaction: optimize >0 order folio compaction with free page > split. > mm/compaction: optimize >0 order folio compaction by sorting source > pages. > > mm/compaction.c | 280 ++++++++++++++++++++++++++++++++++++------------ > mm/internal.h | 7 +- > 2 files changed, 217 insertions(+), 70 deletions(-) >
On 21 Nov 2023, at 12:11, Ryan Roberts wrote: > On 21/11/2023 16:45, Zi Yan wrote: >> On 21 Nov 2023, at 10:46, Ryan Roberts wrote: >> >>>> >>>> vm-scalability results >>>> === >>>> >>>> ========================================================================================= >>>> compiler/kconfig/rootfs/runtime/tbox_group/test/testcase: >>>> gcc-13/defconfig/debian/300s/qemu-vm/mmap-xread-seq-mt/vm-scalability >>>> >>>> commit: >>>> 6.6.0-rc4-mm-everything-2023-10-21-02-40+ >>>> 6.6.0-rc4-split-folio-in-compaction+ >>>> 6.6.0-rc4-folio-migration-in-compaction+ >>>> 6.6.0-rc4-folio-migration-free-page-split+ >>>> 6.6.0-rc4-folio-migration-free-page-split-sort-src+ >>>> >>>> 6.6.0-rc4-mm-eve 6.6.0-rc4-split-folio-in-co 6.6.0-rc4-folio-migration-i 6.6.0-rc4-folio-migration-f 6.6.0-rc4-folio-migration-f >>>> ---------------- --------------------------- --------------------------- --------------------------- --------------------------- >>>> %stddev %change %stddev %change %stddev %change %stddev %change %stddev >>>> \ | \ | \ | \ | \ >>>> 12896955 +2.7% 13249322 -4.0% 12385175 ± 5% +1.1% 13033951 -0.4% 12845698 vm-scalability.throughput >>> >>> Hi Zi, >>> >>> Are you able to add any commentary to these results as I'm struggling to >>> interpret them; Is a positive or negative change better (are they times or >>> rates?). What are the stddev values? The title suggests percent but the values >>> are huge - I'm trying to understand what the error bars look like - are the >>> swings real or noise? >> >> The metric is vm-scalability.throughput, so the larger the better. Some %stddev >> are not present since they are too small. For 6.6.0-rc4-folio-migration-in-compaction+, >> %stddev is greater than %change, so the change might be noise. > > Ahh got it - thanks! > >> >> Also, I talked to DavidH in last THP Cabal meeting about this. He suggested that >> there are a lot of noise in vm-scalability like what I have here and I should >> run more iterations and on bare metal. I am currently rerun them on a baremetal >> and more iterations on the existing VM and report the results later. Please >> note that the runs really take some time. > > Ahh ok, I'll wait for the bare metal numbers and will disregard these for now. > Thanks! It seems that the unexpected big mmap-pread-seq-mt perf drop came from the mistake I made in patch 1. After fixing that, mmap-pread-seq-mt perf only drops 0.5%. The new results on top of 6.7.0-rc1-mm-everything-2023-11-15-00-17 are at the end of the email. I am preparing v2 and will send it out soon. ========================================================================================= compiler/kconfig/rootfs/runtime/tbox_group/test/testcase: gcc-13/defconfig/debian/300s/qemu-vm/mmap-xread-seq-mt/vm-scalability commit: 6.7.0-rc1-mm-everything-2023-11-15-00-17+ 6.7.0-rc1-split-folio-in-compaction+ 6.7.0-rc1-folio-migration-in-compaction+ 6.7.0-rc1-folio-migration-free-page-split+ 6.7.0-rc1-folio-migration-free-page-split-sort-src+ 6.7.0-rc1-mm-eve 6.7.0-rc1-split-folio-in-co 6.7.0-rc1-folio-migration-i 6.7.0-rc1-folio-migration-f 6.7.0-rc1-folio-migration-f ---------------- --------------------------- --------------------------- --------------------------- --------------------------- %stddev %change %stddev %change %stddev %change %stddev %change %stddev \ | \ | \ | \ | \ 13041962 +16.1% 15142976 +5.0% 13690666 ± 6% +6.7% 13920441 +5.5% 13762582 vm-scalability.throughput ========================================================================================= compiler/kconfig/rootfs/runtime/tbox_group/test/testcase: gcc-13/defconfig/debian/300s/qemu-vm/mmap-pread-seq/vm-scalability commit: 6.7.0-rc1-mm-everything-2023-11-15-00-17+ 6.7.0-rc1-split-folio-in-compaction+ 6.7.0-rc1-folio-migration-in-compaction+ 6.7.0-rc1-folio-migration-free-page-split+ 6.7.0-rc1-folio-migration-free-page-split-sort-src+ 6.7.0-rc1-mm-eve 6.7.0-rc1-split-folio-in-co 6.7.0-rc1-folio-migration-i 6.7.0-rc1-folio-migration-f 6.7.0-rc1-folio-migration-f ---------------- --------------------------- --------------------------- --------------------------- --------------------------- %stddev %change %stddev %change %stddev %change %stddev %change %stddev \ | \ | \ | \ | \ 14998168 -1.0% 14852803 -0.7% 14890569 -0.3% 14946766 -0.4% 14943302 vm-scalability.throughput ========================================================================================= compiler/kconfig/rootfs/runtime/tbox_group/test/testcase: gcc-13/defconfig/debian/300s/qemu-vm/lru-file-readtwice/vm-scalability commit: 6.7.0-rc1-mm-everything-2023-11-15-00-17+ 6.7.0-rc1-split-folio-in-compaction+ 6.7.0-rc1-folio-migration-in-compaction+ 6.7.0-rc1-folio-migration-free-page-split+ 6.7.0-rc1-folio-migration-free-page-split-sort-src+ 6.7.0-rc1-mm-eve 6.7.0-rc1-split-folio-in-co 6.7.0-rc1-folio-migration-i 6.7.0-rc1-folio-migration-f 6.7.0-rc1-folio-migration-f ---------------- --------------------------- --------------------------- --------------------------- --------------------------- %stddev %change %stddev %change %stddev %change %stddev %change %stddev \ | \ | \ | \ | \ 6750930 ± 7% +41.5% 9549570 ± 2% +31.6% 8883545 ± 2% +33.1% 8982606 ± 2% +30.7% 8821667 ± 3% vm-scalability.throughput ========================================================================================= compiler/kconfig/rootfs/runtime/tbox_group/test/testcase: gcc-13/defconfig/debian/300s/qemu-vm/lru-file-mmap-read/vm-scalability commit: 6.7.0-rc1-mm-everything-2023-11-15-00-17+ 6.7.0-rc1-split-folio-in-compaction+ 6.7.0-rc1-folio-migration-in-compaction+ 6.7.0-rc1-folio-migration-free-page-split+ 6.7.0-rc1-folio-migration-free-page-split-sort-src+ 6.7.0-rc1-mm-eve 6.7.0-rc1-split-folio-in-co 6.7.0-rc1-folio-migration-i 6.7.0-rc1-folio-migration-f 6.7.0-rc1-folio-migration-f ---------------- --------------------------- --------------------------- --------------------------- --------------------------- %stddev %change %stddev %change %stddev %change %stddev %change %stddev \ | \ | \ | \ | \ 5818610 ± 14% +43.1% 8329118 ± 2% +26.0% 7331427 ± 4% +23.2% 7170418 ± 4% +22.8% 7147458 ± 4% vm-scalability.throughput ========================================================================================= compiler/kconfig/rootfs/runtime/tbox_group/test/testcase: gcc-13/defconfig/debian/300s/qemu-vm/anon-r-rand/vm-scalability commit: 6.7.0-rc1-mm-everything-2023-11-15-00-17+ 6.7.0-rc1-split-folio-in-compaction+ 6.7.0-rc1-folio-migration-in-compaction+ 6.7.0-rc1-folio-migration-free-page-split+ 6.7.0-rc1-folio-migration-free-page-split-sort-src+ 6.7.0-rc1-mm-eve 6.7.0-rc1-split-folio-in-co 6.7.0-rc1-folio-migration-i 6.7.0-rc1-folio-migration-f 6.7.0-rc1-folio-migration-f ---------------- --------------------------- --------------------------- --------------------------- --------------------------- %stddev %change %stddev %change %stddev %change %stddev %change %stddev \ | \ | \ | \ | \ 1973808 -0.2% 1969617 -0.5% 1964376 +0.3% 1979245 +1.0% 1993677 vm-scalability.throughput ========================================================================================= compiler/kconfig/rootfs/runtime/size/tbox_group/test/testcase: gcc-13/defconfig/debian/300s/512G/qemu-vm/anon-w-rand/vm-scalability commit: 6.7.0-rc1-mm-everything-2023-11-15-00-17+ 6.7.0-rc1-split-folio-in-compaction+ 6.7.0-rc1-folio-migration-in-compaction+ 6.7.0-rc1-folio-migration-free-page-split+ 6.7.0-rc1-folio-migration-free-page-split-sort-src+ 6.7.0-rc1-mm-eve 6.7.0-rc1-split-folio-in-co 6.7.0-rc1-folio-migration-i 6.7.0-rc1-folio-migration-f 6.7.0-rc1-folio-migration-f ---------------- --------------------------- --------------------------- --------------------------- --------------------------- %stddev %change %stddev %change %stddev %change %stddev %change %stddev \ | \ | \ | \ | \ 479486 +0.3% 481036 +0.0% 479580 +2.3% 490310 +1.6% 487107 vm-scalability.throughput ========================================================================================= compiler/kconfig/rootfs/runtime/size/tbox_group/test/testcase: gcc-13/defconfig/debian/300s/512G/qemu-vm/anon-cow-rand/vm-scalability commit: 6.7.0-rc1-mm-everything-2023-11-15-00-17+ 6.7.0-rc1-split-folio-in-compaction+ 6.7.0-rc1-folio-migration-in-compaction+ 6.7.0-rc1-folio-migration-free-page-split+ 6.7.0-rc1-folio-migration-free-page-split-sort-src+ 6.7.0-rc1-mm-eve 6.7.0-rc1-split-folio-in-co 6.7.0-rc1-folio-migration-i 6.7.0-rc1-folio-migration-f 6.7.0-rc1-folio-migration-f ---------------- --------------------------- --------------------------- --------------------------- --------------------------- %stddev %change %stddev %change %stddev %change %stddev %change %stddev \ | \ | \ | \ | \ 453237 +1.2% 458580 +0.5% 455455 +2.6% 464830 +2.6% 465211 vm-scalability.throughput ========================================================================================= compiler/kconfig/rootfs/runtime/size/tbox_group/test/testcase: gcc-13/defconfig/debian/300s/256G/qemu-vm/msync/vm-scalability commit: 6.7.0-rc1-mm-everything-2023-11-15-00-17+ 6.7.0-rc1-split-folio-in-compaction+ 6.7.0-rc1-folio-migration-in-compaction+ 6.7.0-rc1-folio-migration-free-page-split+ 6.7.0-rc1-folio-migration-free-page-split-sort-src+ 6.7.0-rc1-mm-eve 6.7.0-rc1-split-folio-in-co 6.7.0-rc1-folio-migration-i 6.7.0-rc1-folio-migration-f 6.7.0-rc1-folio-migration-f ---------------- --------------------------- --------------------------- --------------------------- --------------------------- %stddev %change %stddev %change %stddev %change %stddev %change %stddev \ | \ | \ | \ | \ 1074404 ± 5% +78.7% 1920268 +78.7% 1920270 +77.2% 1903612 +78.0% 1912223 vm-scalability.throughput ========================================================================================= compiler/kconfig/rootfs/runtime/size/tbox_group/test/testcase: gcc-13/defconfig/debian/300s/256G/qemu-vm/msync-mt/vm-scalability commit: 6.7.0-rc1-mm-everything-2023-11-15-00-17+ 6.7.0-rc1-split-folio-in-compaction+ 6.7.0-rc1-folio-migration-in-compaction+ 6.7.0-rc1-folio-migration-free-page-split+ 6.7.0-rc1-folio-migration-free-page-split-sort-src+ 6.7.0-rc1-mm-eve 6.7.0-rc1-split-folio-in-co 6.7.0-rc1-folio-migration-i 6.7.0-rc1-folio-migration-f 6.7.0-rc1-folio-migration-f ---------------- --------------------------- --------------------------- --------------------------- --------------------------- %stddev %change %stddev %change %stddev %change %stddev %change %stddev \ | \ | \ | \ | \ 2568988 +5.3% 2706227 +5.3% 2706214 +6.3% 2729716 +5.3% 2704539 vm-scalability.throughput -- Best Regards, Yan, Zi
On 02/01/2024 20:50, Zi Yan wrote: > On 21 Nov 2023, at 12:11, Ryan Roberts wrote: > >> On 21/11/2023 16:45, Zi Yan wrote: >>> On 21 Nov 2023, at 10:46, Ryan Roberts wrote: >>> >>>>> >>>>> vm-scalability results >>>>> === >>>>> >>>>> ========================================================================================= >>>>> compiler/kconfig/rootfs/runtime/tbox_group/test/testcase: >>>>> gcc-13/defconfig/debian/300s/qemu-vm/mmap-xread-seq-mt/vm-scalability >>>>> >>>>> commit: >>>>> 6.6.0-rc4-mm-everything-2023-10-21-02-40+ >>>>> 6.6.0-rc4-split-folio-in-compaction+ >>>>> 6.6.0-rc4-folio-migration-in-compaction+ >>>>> 6.6.0-rc4-folio-migration-free-page-split+ >>>>> 6.6.0-rc4-folio-migration-free-page-split-sort-src+ >>>>> >>>>> 6.6.0-rc4-mm-eve 6.6.0-rc4-split-folio-in-co 6.6.0-rc4-folio-migration-i 6.6.0-rc4-folio-migration-f 6.6.0-rc4-folio-migration-f >>>>> ---------------- --------------------------- --------------------------- --------------------------- --------------------------- >>>>> %stddev %change %stddev %change %stddev %change %stddev %change %stddev >>>>> \ | \ | \ | \ | \ >>>>> 12896955 +2.7% 13249322 -4.0% 12385175 ± 5% +1.1% 13033951 -0.4% 12845698 vm-scalability.throughput >>>> >>>> Hi Zi, >>>> >>>> Are you able to add any commentary to these results as I'm struggling to >>>> interpret them; Is a positive or negative change better (are they times or >>>> rates?). What are the stddev values? The title suggests percent but the values >>>> are huge - I'm trying to understand what the error bars look like - are the >>>> swings real or noise? >>> >>> The metric is vm-scalability.throughput, so the larger the better. Some %stddev >>> are not present since they are too small. For 6.6.0-rc4-folio-migration-in-compaction+, >>> %stddev is greater than %change, so the change might be noise. >> >> Ahh got it - thanks! >> >>> >>> Also, I talked to DavidH in last THP Cabal meeting about this. He suggested that >>> there are a lot of noise in vm-scalability like what I have here and I should >>> run more iterations and on bare metal. I am currently rerun them on a baremetal >>> and more iterations on the existing VM and report the results later. Please >>> note that the runs really take some time. >> >> Ahh ok, I'll wait for the bare metal numbers and will disregard these for now. >> Thanks! > > It seems that the unexpected big mmap-pread-seq-mt perf drop came from the mistake I > made in patch 1. After fixing that, mmap-pread-seq-mt perf only drops 0.5%. The new > results on top of 6.7.0-rc1-mm-everything-2023-11-15-00-17 are at the end of the email. Good news! I don't see the results for mmap-pread-seq-mt below - perhaps you forgot to include it? > > I am preparing v2 and will send it out soon. > > ========================================================================================= > compiler/kconfig/rootfs/runtime/tbox_group/test/testcase: > gcc-13/defconfig/debian/300s/qemu-vm/mmap-xread-seq-mt/vm-scalability > > commit: > 6.7.0-rc1-mm-everything-2023-11-15-00-17+ > 6.7.0-rc1-split-folio-in-compaction+ > 6.7.0-rc1-folio-migration-in-compaction+ > 6.7.0-rc1-folio-migration-free-page-split+ > 6.7.0-rc1-folio-migration-free-page-split-sort-src+ > > 6.7.0-rc1-mm-eve 6.7.0-rc1-split-folio-in-co 6.7.0-rc1-folio-migration-i 6.7.0-rc1-folio-migration-f 6.7.0-rc1-folio-migration-f > ---------------- --------------------------- --------------------------- --------------------------- --------------------------- > %stddev %change %stddev %change %stddev %change %stddev %change %stddev > \ | \ | \ | \ | \ > 13041962 +16.1% 15142976 +5.0% 13690666 ± 6% +6.7% 13920441 +5.5% 13762582 vm-scalability.throughput I'm still not sure I'm interpretting this correctly; is %change always relative to 6.7.0-rc1-mm-everything-2023-11-15-00-17 or is it relative to the previous commit? If the former, then it looks like splitting the folios is actually faster than migrating them whole? > > ========================================================================================= > compiler/kconfig/rootfs/runtime/tbox_group/test/testcase: > gcc-13/defconfig/debian/300s/qemu-vm/mmap-pread-seq/vm-scalability > > commit: > 6.7.0-rc1-mm-everything-2023-11-15-00-17+ > 6.7.0-rc1-split-folio-in-compaction+ > 6.7.0-rc1-folio-migration-in-compaction+ > 6.7.0-rc1-folio-migration-free-page-split+ > 6.7.0-rc1-folio-migration-free-page-split-sort-src+ > > 6.7.0-rc1-mm-eve 6.7.0-rc1-split-folio-in-co 6.7.0-rc1-folio-migration-i 6.7.0-rc1-folio-migration-f 6.7.0-rc1-folio-migration-f > ---------------- --------------------------- --------------------------- --------------------------- --------------------------- > %stddev %change %stddev %change %stddev %change %stddev %change %stddev > \ | \ | \ | \ | \ > 14998168 -1.0% 14852803 -0.7% 14890569 -0.3% 14946766 -0.4% 14943302 vm-scalability.throughput > > ========================================================================================= > compiler/kconfig/rootfs/runtime/tbox_group/test/testcase: > gcc-13/defconfig/debian/300s/qemu-vm/lru-file-readtwice/vm-scalability > > commit: > 6.7.0-rc1-mm-everything-2023-11-15-00-17+ > 6.7.0-rc1-split-folio-in-compaction+ > 6.7.0-rc1-folio-migration-in-compaction+ > 6.7.0-rc1-folio-migration-free-page-split+ > 6.7.0-rc1-folio-migration-free-page-split-sort-src+ > > 6.7.0-rc1-mm-eve 6.7.0-rc1-split-folio-in-co 6.7.0-rc1-folio-migration-i 6.7.0-rc1-folio-migration-f 6.7.0-rc1-folio-migration-f > ---------------- --------------------------- --------------------------- --------------------------- --------------------------- > %stddev %change %stddev %change %stddev %change %stddev %change %stddev > \ | \ | \ | \ | \ > 6750930 ± 7% +41.5% 9549570 ± 2% +31.6% 8883545 ± 2% +33.1% 8982606 ± 2% +30.7% 8821667 ± 3% vm-scalability.throughput > > ========================================================================================= > compiler/kconfig/rootfs/runtime/tbox_group/test/testcase: > gcc-13/defconfig/debian/300s/qemu-vm/lru-file-mmap-read/vm-scalability > > commit: > 6.7.0-rc1-mm-everything-2023-11-15-00-17+ > 6.7.0-rc1-split-folio-in-compaction+ > 6.7.0-rc1-folio-migration-in-compaction+ > 6.7.0-rc1-folio-migration-free-page-split+ > 6.7.0-rc1-folio-migration-free-page-split-sort-src+ > > 6.7.0-rc1-mm-eve 6.7.0-rc1-split-folio-in-co 6.7.0-rc1-folio-migration-i 6.7.0-rc1-folio-migration-f 6.7.0-rc1-folio-migration-f > ---------------- --------------------------- --------------------------- --------------------------- --------------------------- > %stddev %change %stddev %change %stddev %change %stddev %change %stddev > \ | \ | \ | \ | \ > 5818610 ± 14% +43.1% 8329118 ± 2% +26.0% 7331427 ± 4% +23.2% 7170418 ± 4% +22.8% 7147458 ± 4% vm-scalability.throughput > > ========================================================================================= > compiler/kconfig/rootfs/runtime/tbox_group/test/testcase: > gcc-13/defconfig/debian/300s/qemu-vm/anon-r-rand/vm-scalability > > commit: > 6.7.0-rc1-mm-everything-2023-11-15-00-17+ > 6.7.0-rc1-split-folio-in-compaction+ > 6.7.0-rc1-folio-migration-in-compaction+ > 6.7.0-rc1-folio-migration-free-page-split+ > 6.7.0-rc1-folio-migration-free-page-split-sort-src+ > > 6.7.0-rc1-mm-eve 6.7.0-rc1-split-folio-in-co 6.7.0-rc1-folio-migration-i 6.7.0-rc1-folio-migration-f 6.7.0-rc1-folio-migration-f > ---------------- --------------------------- --------------------------- --------------------------- --------------------------- > %stddev %change %stddev %change %stddev %change %stddev %change %stddev > \ | \ | \ | \ | \ > 1973808 -0.2% 1969617 -0.5% 1964376 +0.3% 1979245 +1.0% 1993677 vm-scalability.throughput > > ========================================================================================= > compiler/kconfig/rootfs/runtime/size/tbox_group/test/testcase: > gcc-13/defconfig/debian/300s/512G/qemu-vm/anon-w-rand/vm-scalability > > commit: > 6.7.0-rc1-mm-everything-2023-11-15-00-17+ > 6.7.0-rc1-split-folio-in-compaction+ > 6.7.0-rc1-folio-migration-in-compaction+ > 6.7.0-rc1-folio-migration-free-page-split+ > 6.7.0-rc1-folio-migration-free-page-split-sort-src+ > > 6.7.0-rc1-mm-eve 6.7.0-rc1-split-folio-in-co 6.7.0-rc1-folio-migration-i 6.7.0-rc1-folio-migration-f 6.7.0-rc1-folio-migration-f > ---------------- --------------------------- --------------------------- --------------------------- --------------------------- > %stddev %change %stddev %change %stddev %change %stddev %change %stddev > \ | \ | \ | \ | \ > 479486 +0.3% 481036 +0.0% 479580 +2.3% 490310 +1.6% 487107 vm-scalability.throughput > > ========================================================================================= > compiler/kconfig/rootfs/runtime/size/tbox_group/test/testcase: > gcc-13/defconfig/debian/300s/512G/qemu-vm/anon-cow-rand/vm-scalability > > commit: > 6.7.0-rc1-mm-everything-2023-11-15-00-17+ > 6.7.0-rc1-split-folio-in-compaction+ > 6.7.0-rc1-folio-migration-in-compaction+ > 6.7.0-rc1-folio-migration-free-page-split+ > 6.7.0-rc1-folio-migration-free-page-split-sort-src+ > > 6.7.0-rc1-mm-eve 6.7.0-rc1-split-folio-in-co 6.7.0-rc1-folio-migration-i 6.7.0-rc1-folio-migration-f 6.7.0-rc1-folio-migration-f > ---------------- --------------------------- --------------------------- --------------------------- --------------------------- > %stddev %change %stddev %change %stddev %change %stddev %change %stddev > \ | \ | \ | \ | \ > 453237 +1.2% 458580 +0.5% 455455 +2.6% 464830 +2.6% 465211 vm-scalability.throughput > > ========================================================================================= > compiler/kconfig/rootfs/runtime/size/tbox_group/test/testcase: > gcc-13/defconfig/debian/300s/256G/qemu-vm/msync/vm-scalability > > commit: > 6.7.0-rc1-mm-everything-2023-11-15-00-17+ > 6.7.0-rc1-split-folio-in-compaction+ > 6.7.0-rc1-folio-migration-in-compaction+ > 6.7.0-rc1-folio-migration-free-page-split+ > 6.7.0-rc1-folio-migration-free-page-split-sort-src+ > > 6.7.0-rc1-mm-eve 6.7.0-rc1-split-folio-in-co 6.7.0-rc1-folio-migration-i 6.7.0-rc1-folio-migration-f 6.7.0-rc1-folio-migration-f > ---------------- --------------------------- --------------------------- --------------------------- --------------------------- > %stddev %change %stddev %change %stddev %change %stddev %change %stddev > \ | \ | \ | \ | \ > 1074404 ± 5% +78.7% 1920268 +78.7% 1920270 +77.2% 1903612 +78.0% 1912223 vm-scalability.throughput > > ========================================================================================= > compiler/kconfig/rootfs/runtime/size/tbox_group/test/testcase: > gcc-13/defconfig/debian/300s/256G/qemu-vm/msync-mt/vm-scalability > > commit: > 6.7.0-rc1-mm-everything-2023-11-15-00-17+ > 6.7.0-rc1-split-folio-in-compaction+ > 6.7.0-rc1-folio-migration-in-compaction+ > 6.7.0-rc1-folio-migration-free-page-split+ > 6.7.0-rc1-folio-migration-free-page-split-sort-src+ > > 6.7.0-rc1-mm-eve 6.7.0-rc1-split-folio-in-co 6.7.0-rc1-folio-migration-i 6.7.0-rc1-folio-migration-f 6.7.0-rc1-folio-migration-f > ---------------- --------------------------- --------------------------- --------------------------- --------------------------- > %stddev %change %stddev %change %stddev %change %stddev %change %stddev > \ | \ | \ | \ | \ > 2568988 +5.3% 2706227 +5.3% 2706214 +6.3% 2729716 +5.3% 2704539 vm-scalability.throughput > > > -- > Best Regards, > Yan, Zi
On 3 Jan 2024, at 4:12, Ryan Roberts wrote: > On 02/01/2024 20:50, Zi Yan wrote: >> On 21 Nov 2023, at 12:11, Ryan Roberts wrote: >> >>> On 21/11/2023 16:45, Zi Yan wrote: >>>> On 21 Nov 2023, at 10:46, Ryan Roberts wrote: >>>> >>>>>> >>>>>> vm-scalability results >>>>>> === >>>>>> >>>>>> ========================================================================================= >>>>>> compiler/kconfig/rootfs/runtime/tbox_group/test/testcase: >>>>>> gcc-13/defconfig/debian/300s/qemu-vm/mmap-xread-seq-mt/vm-scalability >>>>>> >>>>>> commit: >>>>>> 6.6.0-rc4-mm-everything-2023-10-21-02-40+ >>>>>> 6.6.0-rc4-split-folio-in-compaction+ >>>>>> 6.6.0-rc4-folio-migration-in-compaction+ >>>>>> 6.6.0-rc4-folio-migration-free-page-split+ >>>>>> 6.6.0-rc4-folio-migration-free-page-split-sort-src+ >>>>>> >>>>>> 6.6.0-rc4-mm-eve 6.6.0-rc4-split-folio-in-co 6.6.0-rc4-folio-migration-i 6.6.0-rc4-folio-migration-f 6.6.0-rc4-folio-migration-f >>>>>> ---------------- --------------------------- --------------------------- --------------------------- --------------------------- >>>>>> %stddev %change %stddev %change %stddev %change %stddev %change %stddev >>>>>> \ | \ | \ | \ | \ >>>>>> 12896955 +2.7% 13249322 -4.0% 12385175 ± 5% +1.1% 13033951 -0.4% 12845698 vm-scalability.throughput >>>>> >>>>> Hi Zi, >>>>> >>>>> Are you able to add any commentary to these results as I'm struggling to >>>>> interpret them; Is a positive or negative change better (are they times or >>>>> rates?). What are the stddev values? The title suggests percent but the values >>>>> are huge - I'm trying to understand what the error bars look like - are the >>>>> swings real or noise? >>>> >>>> The metric is vm-scalability.throughput, so the larger the better. Some %stddev >>>> are not present since they are too small. For 6.6.0-rc4-folio-migration-in-compaction+, >>>> %stddev is greater than %change, so the change might be noise. >>> >>> Ahh got it - thanks! >>> >>>> >>>> Also, I talked to DavidH in last THP Cabal meeting about this. He suggested that >>>> there are a lot of noise in vm-scalability like what I have here and I should >>>> run more iterations and on bare metal. I am currently rerun them on a baremetal >>>> and more iterations on the existing VM and report the results later. Please >>>> note that the runs really take some time. >>> >>> Ahh ok, I'll wait for the bare metal numbers and will disregard these for now. >>> Thanks! >> >> It seems that the unexpected big mmap-pread-seq-mt perf drop came from the mistake I >> made in patch 1. After fixing that, mmap-pread-seq-mt perf only drops 0.5%. The new >> results on top of 6.7.0-rc1-mm-everything-2023-11-15-00-17 are at the end of the email. > > Good news! I don't see the results for mmap-pread-seq-mt below - perhaps you > forgot to include it? The stats below only shows significant changes and mmap-pread-seq-mt delta is less than 5%, thus it is not shown. > >> >> I am preparing v2 and will send it out soon. >> >> ========================================================================================= >> compiler/kconfig/rootfs/runtime/tbox_group/test/testcase: >> gcc-13/defconfig/debian/300s/qemu-vm/mmap-xread-seq-mt/vm-scalability >> >> commit: >> 6.7.0-rc1-mm-everything-2023-11-15-00-17+ >> 6.7.0-rc1-split-folio-in-compaction+ >> 6.7.0-rc1-folio-migration-in-compaction+ >> 6.7.0-rc1-folio-migration-free-page-split+ >> 6.7.0-rc1-folio-migration-free-page-split-sort-src+ >> >> 6.7.0-rc1-mm-eve 6.7.0-rc1-split-folio-in-co 6.7.0-rc1-folio-migration-i 6.7.0-rc1-folio-migration-f 6.7.0-rc1-folio-migration-f >> ---------------- --------------------------- --------------------------- --------------------------- --------------------------- >> %stddev %change %stddev %change %stddev %change %stddev %change %stddev >> \ | \ | \ | \ | \ >> 13041962 +16.1% 15142976 +5.0% 13690666 ± 6% +6.7% 13920441 +5.5% 13762582 vm-scalability.throughput > > I'm still not sure I'm interpretting this correctly; is %change always relative > to 6.7.0-rc1-mm-everything-2023-11-15-00-17 or is it relative to the previous > commit? The former, always relative to 6.7.0-rc1-mm-everything-2023-11-15-00-17. > > If the former, then it looks like splitting the folios is actually faster than > migrating them whole? Yes, I will look into it when I am preparing the next version. > > >> >> ========================================================================================= >> compiler/kconfig/rootfs/runtime/tbox_group/test/testcase: >> gcc-13/defconfig/debian/300s/qemu-vm/mmap-pread-seq/vm-scalability >> >> commit: >> 6.7.0-rc1-mm-everything-2023-11-15-00-17+ >> 6.7.0-rc1-split-folio-in-compaction+ >> 6.7.0-rc1-folio-migration-in-compaction+ >> 6.7.0-rc1-folio-migration-free-page-split+ >> 6.7.0-rc1-folio-migration-free-page-split-sort-src+ >> >> 6.7.0-rc1-mm-eve 6.7.0-rc1-split-folio-in-co 6.7.0-rc1-folio-migration-i 6.7.0-rc1-folio-migration-f 6.7.0-rc1-folio-migration-f >> ---------------- --------------------------- --------------------------- --------------------------- --------------------------- >> %stddev %change %stddev %change %stddev %change %stddev %change %stddev >> \ | \ | \ | \ | \ >> 14998168 -1.0% 14852803 -0.7% 14890569 -0.3% 14946766 -0.4% 14943302 vm-scalability.throughput >> >> ========================================================================================= >> compiler/kconfig/rootfs/runtime/tbox_group/test/testcase: >> gcc-13/defconfig/debian/300s/qemu-vm/lru-file-readtwice/vm-scalability >> >> commit: >> 6.7.0-rc1-mm-everything-2023-11-15-00-17+ >> 6.7.0-rc1-split-folio-in-compaction+ >> 6.7.0-rc1-folio-migration-in-compaction+ >> 6.7.0-rc1-folio-migration-free-page-split+ >> 6.7.0-rc1-folio-migration-free-page-split-sort-src+ >> >> 6.7.0-rc1-mm-eve 6.7.0-rc1-split-folio-in-co 6.7.0-rc1-folio-migration-i 6.7.0-rc1-folio-migration-f 6.7.0-rc1-folio-migration-f >> ---------------- --------------------------- --------------------------- --------------------------- --------------------------- >> %stddev %change %stddev %change %stddev %change %stddev %change %stddev >> \ | \ | \ | \ | \ >> 6750930 ± 7% +41.5% 9549570 ± 2% +31.6% 8883545 ± 2% +33.1% 8982606 ± 2% +30.7% 8821667 ± 3% vm-scalability.throughput >> >> ========================================================================================= >> compiler/kconfig/rootfs/runtime/tbox_group/test/testcase: >> gcc-13/defconfig/debian/300s/qemu-vm/lru-file-mmap-read/vm-scalability >> >> commit: >> 6.7.0-rc1-mm-everything-2023-11-15-00-17+ >> 6.7.0-rc1-split-folio-in-compaction+ >> 6.7.0-rc1-folio-migration-in-compaction+ >> 6.7.0-rc1-folio-migration-free-page-split+ >> 6.7.0-rc1-folio-migration-free-page-split-sort-src+ >> >> 6.7.0-rc1-mm-eve 6.7.0-rc1-split-folio-in-co 6.7.0-rc1-folio-migration-i 6.7.0-rc1-folio-migration-f 6.7.0-rc1-folio-migration-f >> ---------------- --------------------------- --------------------------- --------------------------- --------------------------- >> %stddev %change %stddev %change %stddev %change %stddev %change %stddev >> \ | \ | \ | \ | \ >> 5818610 ± 14% +43.1% 8329118 ± 2% +26.0% 7331427 ± 4% +23.2% 7170418 ± 4% +22.8% 7147458 ± 4% vm-scalability.throughput >> >> ========================================================================================= >> compiler/kconfig/rootfs/runtime/tbox_group/test/testcase: >> gcc-13/defconfig/debian/300s/qemu-vm/anon-r-rand/vm-scalability >> >> commit: >> 6.7.0-rc1-mm-everything-2023-11-15-00-17+ >> 6.7.0-rc1-split-folio-in-compaction+ >> 6.7.0-rc1-folio-migration-in-compaction+ >> 6.7.0-rc1-folio-migration-free-page-split+ >> 6.7.0-rc1-folio-migration-free-page-split-sort-src+ >> >> 6.7.0-rc1-mm-eve 6.7.0-rc1-split-folio-in-co 6.7.0-rc1-folio-migration-i 6.7.0-rc1-folio-migration-f 6.7.0-rc1-folio-migration-f >> ---------------- --------------------------- --------------------------- --------------------------- --------------------------- >> %stddev %change %stddev %change %stddev %change %stddev %change %stddev >> \ | \ | \ | \ | \ >> 1973808 -0.2% 1969617 -0.5% 1964376 +0.3% 1979245 +1.0% 1993677 vm-scalability.throughput >> >> ========================================================================================= >> compiler/kconfig/rootfs/runtime/size/tbox_group/test/testcase: >> gcc-13/defconfig/debian/300s/512G/qemu-vm/anon-w-rand/vm-scalability >> >> commit: >> 6.7.0-rc1-mm-everything-2023-11-15-00-17+ >> 6.7.0-rc1-split-folio-in-compaction+ >> 6.7.0-rc1-folio-migration-in-compaction+ >> 6.7.0-rc1-folio-migration-free-page-split+ >> 6.7.0-rc1-folio-migration-free-page-split-sort-src+ >> >> 6.7.0-rc1-mm-eve 6.7.0-rc1-split-folio-in-co 6.7.0-rc1-folio-migration-i 6.7.0-rc1-folio-migration-f 6.7.0-rc1-folio-migration-f >> ---------------- --------------------------- --------------------------- --------------------------- --------------------------- >> %stddev %change %stddev %change %stddev %change %stddev %change %stddev >> \ | \ | \ | \ | \ >> 479486 +0.3% 481036 +0.0% 479580 +2.3% 490310 +1.6% 487107 vm-scalability.throughput >> >> ========================================================================================= >> compiler/kconfig/rootfs/runtime/size/tbox_group/test/testcase: >> gcc-13/defconfig/debian/300s/512G/qemu-vm/anon-cow-rand/vm-scalability >> >> commit: >> 6.7.0-rc1-mm-everything-2023-11-15-00-17+ >> 6.7.0-rc1-split-folio-in-compaction+ >> 6.7.0-rc1-folio-migration-in-compaction+ >> 6.7.0-rc1-folio-migration-free-page-split+ >> 6.7.0-rc1-folio-migration-free-page-split-sort-src+ >> >> 6.7.0-rc1-mm-eve 6.7.0-rc1-split-folio-in-co 6.7.0-rc1-folio-migration-i 6.7.0-rc1-folio-migration-f 6.7.0-rc1-folio-migration-f >> ---------------- --------------------------- --------------------------- --------------------------- --------------------------- >> %stddev %change %stddev %change %stddev %change %stddev %change %stddev >> \ | \ | \ | \ | \ >> 453237 +1.2% 458580 +0.5% 455455 +2.6% 464830 +2.6% 465211 vm-scalability.throughput >> >> ========================================================================================= >> compiler/kconfig/rootfs/runtime/size/tbox_group/test/testcase: >> gcc-13/defconfig/debian/300s/256G/qemu-vm/msync/vm-scalability >> >> commit: >> 6.7.0-rc1-mm-everything-2023-11-15-00-17+ >> 6.7.0-rc1-split-folio-in-compaction+ >> 6.7.0-rc1-folio-migration-in-compaction+ >> 6.7.0-rc1-folio-migration-free-page-split+ >> 6.7.0-rc1-folio-migration-free-page-split-sort-src+ >> >> 6.7.0-rc1-mm-eve 6.7.0-rc1-split-folio-in-co 6.7.0-rc1-folio-migration-i 6.7.0-rc1-folio-migration-f 6.7.0-rc1-folio-migration-f >> ---------------- --------------------------- --------------------------- --------------------------- --------------------------- >> %stddev %change %stddev %change %stddev %change %stddev %change %stddev >> \ | \ | \ | \ | \ >> 1074404 ± 5% +78.7% 1920268 +78.7% 1920270 +77.2% 1903612 +78.0% 1912223 vm-scalability.throughput >> >> ========================================================================================= >> compiler/kconfig/rootfs/runtime/size/tbox_group/test/testcase: >> gcc-13/defconfig/debian/300s/256G/qemu-vm/msync-mt/vm-scalability >> >> commit: >> 6.7.0-rc1-mm-everything-2023-11-15-00-17+ >> 6.7.0-rc1-split-folio-in-compaction+ >> 6.7.0-rc1-folio-migration-in-compaction+ >> 6.7.0-rc1-folio-migration-free-page-split+ >> 6.7.0-rc1-folio-migration-free-page-split-sort-src+ >> >> 6.7.0-rc1-mm-eve 6.7.0-rc1-split-folio-in-co 6.7.0-rc1-folio-migration-i 6.7.0-rc1-folio-migration-f 6.7.0-rc1-folio-migration-f >> ---------------- --------------------------- --------------------------- --------------------------- --------------------------- >> %stddev %change %stddev %change %stddev %change %stddev %change %stddev >> \ | \ | \ | \ | \ >> 2568988 +5.3% 2706227 +5.3% 2706214 +6.3% 2729716 +5.3% 2704539 vm-scalability.throughput >> >> >> -- >> Best Regards, >> Yan, Zi -- Best Regards, Yan, Zi
On 3 Jan 2024, at 10:51, Zi Yan wrote: > On 3 Jan 2024, at 4:12, Ryan Roberts wrote: > >> On 02/01/2024 20:50, Zi Yan wrote: >>> On 21 Nov 2023, at 12:11, Ryan Roberts wrote: >>> >>>> On 21/11/2023 16:45, Zi Yan wrote: >>>>> On 21 Nov 2023, at 10:46, Ryan Roberts wrote: >>>>> >>>>>>> >>>>>>> vm-scalability results >>>>>>> === >>>>>>> >>>>>>> ========================================================================================= >>>>>>> compiler/kconfig/rootfs/runtime/tbox_group/test/testcase: >>>>>>> gcc-13/defconfig/debian/300s/qemu-vm/mmap-xread-seq-mt/vm-scalability >>>>>>> >>>>>>> commit: >>>>>>> 6.6.0-rc4-mm-everything-2023-10-21-02-40+ >>>>>>> 6.6.0-rc4-split-folio-in-compaction+ >>>>>>> 6.6.0-rc4-folio-migration-in-compaction+ >>>>>>> 6.6.0-rc4-folio-migration-free-page-split+ >>>>>>> 6.6.0-rc4-folio-migration-free-page-split-sort-src+ >>>>>>> >>>>>>> 6.6.0-rc4-mm-eve 6.6.0-rc4-split-folio-in-co 6.6.0-rc4-folio-migration-i 6.6.0-rc4-folio-migration-f 6.6.0-rc4-folio-migration-f >>>>>>> ---------------- --------------------------- --------------------------- --------------------------- --------------------------- >>>>>>> %stddev %change %stddev %change %stddev %change %stddev %change %stddev >>>>>>> \ | \ | \ | \ | \ >>>>>>> 12896955 +2.7% 13249322 -4.0% 12385175 ± 5% +1.1% 13033951 -0.4% 12845698 vm-scalability.throughput >>>>>> >>>>>> Hi Zi, >>>>>> >>>>>> Are you able to add any commentary to these results as I'm struggling to >>>>>> interpret them; Is a positive or negative change better (are they times or >>>>>> rates?). What are the stddev values? The title suggests percent but the values >>>>>> are huge - I'm trying to understand what the error bars look like - are the >>>>>> swings real or noise? >>>>> >>>>> The metric is vm-scalability.throughput, so the larger the better. Some %stddev >>>>> are not present since they are too small. For 6.6.0-rc4-folio-migration-in-compaction+, >>>>> %stddev is greater than %change, so the change might be noise. >>>> >>>> Ahh got it - thanks! >>>> >>>>> >>>>> Also, I talked to DavidH in last THP Cabal meeting about this. He suggested that >>>>> there are a lot of noise in vm-scalability like what I have here and I should >>>>> run more iterations and on bare metal. I am currently rerun them on a baremetal >>>>> and more iterations on the existing VM and report the results later. Please >>>>> note that the runs really take some time. >>>> >>>> Ahh ok, I'll wait for the bare metal numbers and will disregard these for now. >>>> Thanks! >>> >>> It seems that the unexpected big mmap-pread-seq-mt perf drop came from the mistake I >>> made in patch 1. After fixing that, mmap-pread-seq-mt perf only drops 0.5%. The new >>> results on top of 6.7.0-rc1-mm-everything-2023-11-15-00-17 are at the end of the email. >> >> Good news! I don't see the results for mmap-pread-seq-mt below - perhaps you >> forgot to include it? > > The stats below only shows significant changes and mmap-pread-seq-mt delta is less > than 5%, thus it is not shown. > >> >>> >>> I am preparing v2 and will send it out soon. >>> >>> ========================================================================================= >>> compiler/kconfig/rootfs/runtime/tbox_group/test/testcase: >>> gcc-13/defconfig/debian/300s/qemu-vm/mmap-xread-seq-mt/vm-scalability >>> >>> commit: >>> 6.7.0-rc1-mm-everything-2023-11-15-00-17+ >>> 6.7.0-rc1-split-folio-in-compaction+ >>> 6.7.0-rc1-folio-migration-in-compaction+ >>> 6.7.0-rc1-folio-migration-free-page-split+ >>> 6.7.0-rc1-folio-migration-free-page-split-sort-src+ >>> >>> 6.7.0-rc1-mm-eve 6.7.0-rc1-split-folio-in-co 6.7.0-rc1-folio-migration-i 6.7.0-rc1-folio-migration-f 6.7.0-rc1-folio-migration-f >>> ---------------- --------------------------- --------------------------- --------------------------- --------------------------- >>> %stddev %change %stddev %change %stddev %change %stddev %change %stddev >>> \ | \ | \ | \ | \ >>> 13041962 +16.1% 15142976 +5.0% 13690666 ± 6% +6.7% 13920441 +5.5% 13762582 vm-scalability.throughput >> >> I'm still not sure I'm interpretting this correctly; is %change always relative >> to 6.7.0-rc1-mm-everything-2023-11-15-00-17 or is it relative to the previous >> commit? > > The former, always relative to 6.7.0-rc1-mm-everything-2023-11-15-00-17. > >> >> If the former, then it looks like splitting the folios is actually faster than >> migrating them whole? > > Yes, I will look into it when I am preparing the next version. > The reason seems to be that compaction ends early when migrating folios as a whole. It happens when a order-0 folio is migrated and there is no order-0 free page, then migrate_pages() returns -ENOMEM making compact_zone() stop compaction (for higher order folios, they would be split). This should be fixed by enabling free page split optimization, but the perf number does not say so. Let me dig more. -- Best Regards, Yan, Zi
From: Zi Yan <ziy@nvidia.com> Hi all, This patchset enables >0 order folio memory compaction, which is one of the prerequisitions for large folio support[1]. It is on top of mm-everything-2023-10-21-02-40 (also applies to mm-everything-2023-11-10-20-42 without conflicts). Per patch vm-scalability results on a 8-CPU (Intel Xeon E5-2650 v4 @ 2.20GHz) 16G VM are at the bottom and based on mm-everything-2023-10-21-02-40. Each has 5 runs. Highlights: 1. mmap-pread-seq-mt: -18.3% if split folios during compaction and -7.7% if migrating folios during compaction. 2. lru-file-readtwice: +41.5% if split folios during compaction and +30.4% if migrating folios during compaction. 3. lru-file-mmap-read: +35.1% if split folios during compaction and +32.3% if migrating folios during compaction. I will run experiments on bare metal and report results later. Changelog === From RFC: 1. Enabled >0 order folio compaction in the first patch by splitting all to-be-migrated folios. (per Huang, Ying) 2. Stopped isolating compound pages with order greater than cc->order to avoid wasting effort, since cc->order gives a hint that no free pages with order greater than it exist, thus migrating the compound pages will fail. (per Baolin Wang) 3. Retained the folio check within lru lock. (per Baolin Wang) 4. Made isolate_freepages_block() generate order-sorted multi lists. (per Johannes Weiner) Overview === To support >0 order folio compaction, the patchset changes how free pages used for migration are kept during compaction. Free pages used to be split into order-0 pages that are post allocation processed (i.e., PageBuddy flag cleared, page order stored in page->private is zeroed, and page reference is set to 1). Now all free pages are kept in a MAX_ORDER+1 array of page lists based on their order without post allocation process. When migrate_pages() asks for a new page, one of the free pages, based on the requested page order, is then processed and given out. Optimizations === 1. Free page split is added to increase migration success rate in case a source page does not have a matched free page in the free page lists. Free page merge is possible but not implemented, since existing PFN-based buddy page merge algorithm requires the identification of buddy pages, but free pages kept for memory compaction cannot have PageBuddy set to avoid confusing other PFN scanners. 2. Sort source pages in ascending order before migration is added to reduce free page split. Otherwise, high order free pages might be prematurely split, causing undesired high order folio migration failures. TODOs === 1. Refactor free page post allocation and free page preparation code so that compaction_alloc() and compaction_free() can call functions instead of hard coding. 2. Make isolate_migratepages_block() produce a sorted list for compaction to avoid sort_folio_by_order() in Patch 4. Feel free to give comments and ask questions. Thanks. vm-scalability results === ========================================================================================= compiler/kconfig/rootfs/runtime/tbox_group/test/testcase: gcc-13/defconfig/debian/300s/qemu-vm/mmap-xread-seq-mt/vm-scalability commit: 6.6.0-rc4-mm-everything-2023-10-21-02-40+ 6.6.0-rc4-split-folio-in-compaction+ 6.6.0-rc4-folio-migration-in-compaction+ 6.6.0-rc4-folio-migration-free-page-split+ 6.6.0-rc4-folio-migration-free-page-split-sort-src+ 6.6.0-rc4-mm-eve 6.6.0-rc4-split-folio-in-co 6.6.0-rc4-folio-migration-i 6.6.0-rc4-folio-migration-f 6.6.0-rc4-folio-migration-f ---------------- --------------------------- --------------------------- --------------------------- --------------------------- %stddev %change %stddev %change %stddev %change %stddev %change %stddev \ | \ | \ | \ | \ 12896955 +2.7% 13249322 -4.0% 12385175 ± 5% +1.1% 13033951 -0.4% 12845698 vm-scalability.throughput ========================================================================================= compiler/kconfig/rootfs/runtime/tbox_group/test/testcase: gcc-13/defconfig/debian/300s/qemu-vm/mmap-pread-seq/vm-scalability commit: 6.6.0-rc4-mm-everything-2023-10-21-02-40+ 6.6.0-rc4-split-folio-in-compaction+ 6.6.0-rc4-folio-migration-in-compaction+ 6.6.0-rc4-folio-migration-free-page-split+ 6.6.0-rc4-folio-migration-free-page-split-sort-src+ 6.6.0-rc4-mm-eve 6.6.0-rc4-split-folio-in-co 6.6.0-rc4-folio-migration-i 6.6.0-rc4-folio-migration-f 6.6.0-rc4-folio-migration-f ---------------- --------------------------- --------------------------- --------------------------- --------------------------- %stddev %change %stddev %change %stddev %change %stddev %change %stddev \ | \ | \ | \ | \ 14626665 -5.5% 13825048 -1.7% 14384837 -1.9% 14345958 -2.0% 14338834 vm-scalability.throughput ========================================================================================= compiler/kconfig/rootfs/runtime/tbox_group/test/testcase: gcc-13/defconfig/debian/300s/qemu-vm/mmap-pread-seq-mt/vm-scalability commit: 6.6.0-rc4-mm-everything-2023-10-21-02-40+ 6.6.0-rc4-split-folio-in-compaction+ 6.6.0-rc4-folio-migration-in-compaction+ 6.6.0-rc4-folio-migration-free-page-split+ 6.6.0-rc4-folio-migration-free-page-split-sort-src+ 6.6.0-rc4-mm-eve 6.6.0-rc4-split-folio-in-co 6.6.0-rc4-folio-migration-i 6.6.0-rc4-folio-migration-f 6.6.0-rc4-folio-migration-f ---------------- --------------------------- --------------------------- --------------------------- --------------------------- %stddev %change %stddev %change %stddev %change %stddev %change %stddev \ | \ | \ | \ | \ 14314960 -18.3% 11691969 ± 2% -7.7% 13217795 -8.8% 13057395 -10.8% 12764204 vm-scalability.throughput ========================================================================================= compiler/kconfig/rootfs/runtime/tbox_group/test/testcase: gcc-13/defconfig/debian/300s/qemu-vm/lru-file-readtwice/vm-scalability commit: 6.6.0-rc4-mm-everything-2023-10-21-02-40+ 6.6.0-rc4-split-folio-in-compaction+ 6.6.0-rc4-folio-migration-in-compaction+ 6.6.0-rc4-folio-migration-free-page-split+ 6.6.0-rc4-folio-migration-free-page-split-sort-src+ 6.6.0-rc4-mm-eve 6.6.0-rc4-split-folio-in-co 6.6.0-rc4-folio-migration-i 6.6.0-rc4-folio-migration-f 6.6.0-rc4-folio-migration-f ---------------- --------------------------- --------------------------- --------------------------- --------------------------- %stddev %change %stddev %change %stddev %change %stddev %change %stddev \ | \ | \ | \ | \ 6498203 ± 6% +41.5% 9193356 +30.4% 8472024 +30.0% 8449054 ± 3% +32.3% 8596781 ± 2% vm-scalability.throughput ========================================================================================= compiler/kconfig/rootfs/runtime/tbox_group/test/testcase: gcc-13/defconfig/debian/300s/qemu-vm/lru-file-mmap-read/vm-scalability commit: 6.6.0-rc4-mm-everything-2023-10-21-02-40+ 6.6.0-rc4-split-folio-in-compaction+ 6.6.0-rc4-folio-migration-in-compaction+ 6.6.0-rc4-folio-migration-free-page-split+ 6.6.0-rc4-folio-migration-free-page-split-sort-src+ 6.6.0-rc4-mm-eve 6.6.0-rc4-split-folio-in-co 6.6.0-rc4-folio-migration-i 6.6.0-rc4-folio-migration-f 6.6.0-rc4-folio-migration-f ---------------- --------------------------- --------------------------- --------------------------- --------------------------- %stddev %change %stddev %change %stddev %change %stddev %change %stddev \ | \ | \ | \ | \ 5440732 ± 10% +35.1% 7352631 ± 9% +32.3% 7200622 ± 5% +27.0% 6909673 ± 4% +25.9% 6849852 ± 5% vm-scalability.throughput ========================================================================================= compiler/kconfig/rootfs/runtime/tbox_group/test/testcase: gcc-13/defconfig/debian/300s/qemu-vm/lru-file-mmap-read-rand/vm-scalability commit: 6.6.0-rc4-mm-everything-2023-10-21-02-40+ 6.6.0-rc4-split-folio-in-compaction+ 6.6.0-rc4-folio-migration-in-compaction+ 6.6.0-rc4-folio-migration-free-page-split+ 6.6.0-rc4-folio-migration-free-page-split-sort-src+ 6.6.0-rc4-mm-eve 6.6.0-rc4-split-folio-in-co 6.6.0-rc4-folio-migration-i 6.6.0-rc4-folio-migration-f 6.6.0-rc4-folio-migration-f ---------------- --------------------------- --------------------------- --------------------------- --------------------------- %stddev %change %stddev %change %stddev %change %stddev %change %stddev \ | \ | \ | \ | \ 1627989 +0.3% 1632141 +1.8% 1656877 +0.1% 1629816 -0.2% 1625213 vm-scalability.throughput ========================================================================================= compiler/kconfig/rootfs/runtime/size/tbox_group/test/testcase: gcc-13/defconfig/debian/300s/8T/qemu-vm/anon-wx-seq-mt/vm-scalability commit: 6.6.0-rc4-mm-everything-2023-10-21-02-40+ 6.6.0-rc4-split-folio-in-compaction+ 6.6.0-rc4-folio-migration-in-compaction+ 6.6.0-rc4-folio-migration-free-page-split+ 6.6.0-rc4-folio-migration-free-page-split-sort-src+ 6.6.0-rc4-mm-eve 6.6.0-rc4-split-folio-in-co 6.6.0-rc4-folio-migration-i 6.6.0-rc4-folio-migration-f 6.6.0-rc4-folio-migration-f ---------------- --------------------------- --------------------------- --------------------------- --------------------------- %stddev %change %stddev %change %stddev %change %stddev %change %stddev \ | \ | \ | \ | \ 6738043 +1.4% 6835096 +1.4% 6834329 +1.0% 6805401 +1.2% 6820637 vm-scalability.throughput ========================================================================================= compiler/kconfig/rootfs/runtime/size/tbox_group/test/testcase: gcc-13/defconfig/debian/300s/8T/qemu-vm/anon-w-seq-mt/vm-scalability commit: 6.6.0-rc4-mm-everything-2023-10-21-02-40+ 6.6.0-rc4-split-folio-in-compaction+ 6.6.0-rc4-folio-migration-in-compaction+ 6.6.0-rc4-folio-migration-free-page-split+ 6.6.0-rc4-folio-migration-free-page-split-sort-src+ 6.6.0-rc4-mm-eve 6.6.0-rc4-split-folio-in-co 6.6.0-rc4-folio-migration-i 6.6.0-rc4-folio-migration-f 6.6.0-rc4-folio-migration-f ---------------- --------------------------- --------------------------- --------------------------- --------------------------- %stddev %change %stddev %change %stddev %change %stddev %change %stddev \ | \ | \ | \ | \ 7644894 +0.2% 7661425 +0.4% 7676929 +1.0% 7719218 +0.3% 7664658 vm-scalability.throughput ========================================================================================= compiler/kconfig/rootfs/runtime/size/tbox_group/test/testcase: gcc-13/defconfig/debian/300s/512G/qemu-vm/anon-wx-rand-mt/vm-scalability commit: 6.6.0-rc4-mm-everything-2023-10-21-02-40+ 6.6.0-rc4-split-folio-in-compaction+ 6.6.0-rc4-folio-migration-in-compaction+ 6.6.0-rc4-folio-migration-free-page-split+ 6.6.0-rc4-folio-migration-free-page-split-sort-src+ 6.6.0-rc4-mm-eve 6.6.0-rc4-split-folio-in-co 6.6.0-rc4-folio-migration-i 6.6.0-rc4-folio-migration-f 6.6.0-rc4-folio-migration-f ---------------- --------------------------- --------------------------- --------------------------- --------------------------- %stddev %change %stddev %change %stddev %change %stddev %change %stddev \ | \ | \ | \ | \ 521941 +3.3% 539254 +3.3% 539213 +1.0% 527048 +3.4% 539588 vm-scalability.throughput ========================================================================================= compiler/kconfig/rootfs/runtime/size/tbox_group/test/testcase: gcc-13/defconfig/debian/300s/512G/qemu-vm/anon-w-rand/vm-scalability commit: 6.6.0-rc4-mm-everything-2023-10-21-02-40+ 6.6.0-rc4-split-folio-in-compaction+ 6.6.0-rc4-folio-migration-in-compaction+ 6.6.0-rc4-folio-migration-free-page-split+ 6.6.0-rc4-folio-migration-free-page-split-sort-src+ 6.6.0-rc4-mm-eve 6.6.0-rc4-split-folio-in-co 6.6.0-rc4-folio-migration-i 6.6.0-rc4-folio-migration-f 6.6.0-rc4-folio-migration-f ---------------- --------------------------- --------------------------- --------------------------- --------------------------- %stddev %change %stddev %change %stddev %change %stddev %change %stddev \ | \ | \ | \ | \ 481028 +4.3% 501876 +2.4% 492713 +0.4% 482829 ± 2% +3.5% 497649 ± 2% vm-scalability.throughput ========================================================================================= compiler/kconfig/rootfs/runtime/size/tbox_group/test/testcase: gcc-13/defconfig/debian/300s/512G/qemu-vm/anon-w-rand-mt/vm-scalability commit: 6.6.0-rc4-mm-everything-2023-10-21-02-40+ 6.6.0-rc4-split-folio-in-compaction+ 6.6.0-rc4-folio-migration-in-compaction+ 6.6.0-rc4-folio-migration-free-page-split+ 6.6.0-rc4-folio-migration-free-page-split-sort-src+ 6.6.0-rc4-mm-eve 6.6.0-rc4-split-folio-in-co 6.6.0-rc4-folio-migration-i 6.6.0-rc4-folio-migration-f 6.6.0-rc4-folio-migration-f ---------------- --------------------------- --------------------------- --------------------------- --------------------------- %stddev %change %stddev %change %stddev %change %stddev %change %stddev \ | \ | \ | \ | \ 477063 +4.0% 496373 +4.0% 496283 +1.4% 483966 +3.2% 492216 ± 2% vm-scalability.throughput ========================================================================================= compiler/kconfig/rootfs/runtime/size/tbox_group/test/testcase: gcc-13/defconfig/debian/300s/512G/qemu-vm/anon-cow-rand/vm-scalability commit: 6.6.0-rc4-mm-everything-2023-10-21-02-40+ 6.6.0-rc4-split-folio-in-compaction+ 6.6.0-rc4-folio-migration-in-compaction+ 6.6.0-rc4-folio-migration-free-page-split+ 6.6.0-rc4-folio-migration-free-page-split-sort-src+ 6.6.0-rc4-mm-eve 6.6.0-rc4-split-folio-in-co 6.6.0-rc4-folio-migration-i 6.6.0-rc4-folio-migration-f 6.6.0-rc4-folio-migration-f ---------------- --------------------------- --------------------------- --------------------------- --------------------------- %stddev %change %stddev %change %stddev %change %stddev %change %stddev \ | \ | \ | \ | \ 458073 +3.3% 473056 +2.6% 469835 +0.3% 459266 +3.0% 471735 vm-scalability.throughput ========================================================================================= compiler/kconfig/rootfs/runtime/size/tbox_group/test/testcase: gcc-13/defconfig/debian/300s/512G/qemu-vm/anon-cow-rand-mt/vm-scalability commit: 6.6.0-rc4-mm-everything-2023-10-21-02-40+ 6.6.0-rc4-split-folio-in-compaction+ 6.6.0-rc4-folio-migration-in-compaction+ 6.6.0-rc4-folio-migration-free-page-split+ 6.6.0-rc4-folio-migration-free-page-split-sort-src+ 6.6.0-rc4-mm-eve 6.6.0-rc4-split-folio-in-co 6.6.0-rc4-folio-migration-i 6.6.0-rc4-folio-migration-f 6.6.0-rc4-folio-migration-f ---------------- --------------------------- --------------------------- --------------------------- --------------------------- %stddev %change %stddev %change %stddev %change %stddev %change %stddev \ | \ | \ | \ | \ 468158 +0.8% 472115 +1.4% 474547 +1.1% 473307 +2.1% 478173 vm-scalability.throughput ========================================================================================= compiler/kconfig/rootfs/runtime/size/tbox_group/test/testcase: gcc-13/defconfig/debian/300s/2T/qemu-vm/shm-xread-seq/vm-scalability commit: 6.6.0-rc4-mm-everything-2023-10-21-02-40+ 6.6.0-rc4-split-folio-in-compaction+ 6.6.0-rc4-folio-migration-in-compaction+ 6.6.0-rc4-folio-migration-free-page-split+ 6.6.0-rc4-folio-migration-free-page-split-sort-src+ 6.6.0-rc4-mm-eve 6.6.0-rc4-split-folio-in-co 6.6.0-rc4-folio-migration-i 6.6.0-rc4-folio-migration-f 6.6.0-rc4-folio-migration-f ---------------- --------------------------- --------------------------- --------------------------- --------------------------- %stddev %change %stddev %change %stddev %change %stddev %change %stddev \ | \ | \ | \ | \ 3489946 +1.4% 3540036 +0.4% 3505413 +1.7% 3549852 +1.4% 3539309 vm-scalability.throughput ========================================================================================= compiler/kconfig/rootfs/runtime/size/tbox_group/test/testcase: gcc-13/defconfig/debian/300s/2T/qemu-vm/shm-xread-seq-mt/vm-scalability commit: 6.6.0-rc4-mm-everything-2023-10-21-02-40+ 6.6.0-rc4-split-folio-in-compaction+ 6.6.0-rc4-folio-migration-in-compaction+ 6.6.0-rc4-folio-migration-free-page-split+ 6.6.0-rc4-folio-migration-free-page-split-sort-src+ 6.6.0-rc4-mm-eve 6.6.0-rc4-split-folio-in-co 6.6.0-rc4-folio-migration-i 6.6.0-rc4-folio-migration-f 6.6.0-rc4-folio-migration-f ---------------- --------------------------- --------------------------- --------------------------- --------------------------- %stddev %change %stddev %change %stddev %change %stddev %change %stddev \ | \ | \ | \ | \ 2621783 +1.0% 2647480 +0.6% 2636399 +0.7% 2641150 +0.1% 2624301 vm-scalability.throughput ========================================================================================= compiler/kconfig/rootfs/runtime/size/tbox_group/test/testcase: gcc-13/defconfig/debian/300s/2T/qemu-vm/shm-pread-seq/vm-scalability commit: 6.6.0-rc4-mm-everything-2023-10-21-02-40+ 6.6.0-rc4-split-folio-in-compaction+ 6.6.0-rc4-folio-migration-in-compaction+ 6.6.0-rc4-folio-migration-free-page-split+ 6.6.0-rc4-folio-migration-free-page-split-sort-src+ 6.6.0-rc4-mm-eve 6.6.0-rc4-split-folio-in-co 6.6.0-rc4-folio-migration-i 6.6.0-rc4-folio-migration-f 6.6.0-rc4-folio-migration-f ---------------- --------------------------- --------------------------- --------------------------- --------------------------- %stddev %change %stddev %change %stddev %change %stddev %change %stddev \ | \ | \ | \ | \ 3504293 +1.5% 3557664 -0.4% 3489639 +0.4% 3516842 +0.8% 3533944 vm-scalability.throughput ========================================================================================= compiler/kconfig/rootfs/runtime/size/tbox_group/test/testcase: gcc-13/defconfig/debian/300s/256G/qemu-vm/msync/vm-scalability commit: 6.6.0-rc4-mm-everything-2023-10-21-02-40+ 6.6.0-rc4-split-folio-in-compaction+ 6.6.0-rc4-folio-migration-in-compaction+ 6.6.0-rc4-folio-migration-free-page-split+ 6.6.0-rc4-folio-migration-free-page-split-sort-src+ 6.6.0-rc4-mm-eve 6.6.0-rc4-split-folio-in-co 6.6.0-rc4-folio-migration-i 6.6.0-rc4-folio-migration-f 6.6.0-rc4-folio-migration-f ---------------- --------------------------- --------------------------- --------------------------- --------------------------- %stddev %change %stddev %change %stddev %change %stddev %change %stddev \ | \ | \ | \ | \ 919378 ± 6% +3.8% 954028 ± 10% +7.2% 985920 ± 6% +7.4% 987077 ± 5% +14.2% 1049473 ± 2% vm-scalability.throughput ========================================================================================= compiler/kconfig/rootfs/runtime/size/tbox_group/test/testcase: gcc-13/defconfig/debian/300s/256G/qemu-vm/msync-mt/vm-scalability commit: 6.6.0-rc4-mm-everything-2023-10-21-02-40+ 6.6.0-rc4-split-folio-in-compaction+ 6.6.0-rc4-folio-migration-in-compaction+ 6.6.0-rc4-folio-migration-free-page-split+ 6.6.0-rc4-folio-migration-free-page-split-sort-src+ 6.6.0-rc4-mm-eve 6.6.0-rc4-split-folio-in-co 6.6.0-rc4-folio-migration-i 6.6.0-rc4-folio-migration-f 6.6.0-rc4-folio-migration-f ---------------- --------------------------- --------------------------- --------------------------- --------------------------- %stddev %change %stddev %change %stddev %change %stddev %change %stddev \ | \ | \ | \ | \ 2398063 +2.0% 2446844 -1.5% 2361096 ± 2% -3.1% 2324346 ± 4% -1.0% 2375055 ± 4% vm-scalability.throughput ========================================================================================= compiler/kconfig/rootfs/runtime/size/tbox_group/test/testcase: gcc-13/defconfig/debian/300s/128G/qemu-vm/truncate/vm-scalability commit: 6.6.0-rc4-mm-everything-2023-10-21-02-40+ 6.6.0-rc4-split-folio-in-compaction+ 6.6.0-rc4-folio-migration-in-compaction+ 6.6.0-rc4-folio-migration-free-page-split+ 6.6.0-rc4-folio-migration-free-page-split-sort-src+ 6.6.0-rc4-mm-eve 6.6.0-rc4-split-folio-in-co 6.6.0-rc4-folio-migration-i 6.6.0-rc4-folio-migration-f 6.6.0-rc4-folio-migration-f ---------------- --------------------------- --------------------------- --------------------------- --------------------------- %stddev %change %stddev %change %stddev %change %stddev %change %stddev \ | \ | \ | \ | \ 2.084e+08 ± 2% +1.5% 2.114e+08 ± 3% +6.5% 2.22e+08 ± 3% -2.3% 2.035e+08 ± 2% -2.1% 2.04e+08 ± 5% vm-scalability.throughput [1] https://lore.kernel.org/linux-mm/f8d47176-03a8-99bf-a813-b5942830fd73@arm.com/ Zi Yan (4): mm/compaction: enable compacting >0 order folios. mm/compaction: add support for >0 order folio memory compaction. mm/compaction: optimize >0 order folio compaction with free page split. mm/compaction: optimize >0 order folio compaction by sorting source pages. mm/compaction.c | 280 ++++++++++++++++++++++++++++++++++++------------ mm/internal.h | 7 +- 2 files changed, 217 insertions(+), 70 deletions(-)