| Message ID | 20250214175709.76029-1-ryncsn@gmail.com (mailing list archive) |
|---|---|
| Headers | show |
| Series | mm, swap: remove swap slot cache | expand |
Hi Kairui, On 02/15/25 at 01:57am, Kairui Song wrote: > From: Kairui Song <kasong@tencent.com> > > Slot cache was initially introduced by commit 67afa38e012e ("mm/swap: > add cache for swap slots allocation") to reduce the lock contention > of si->lock. Thanks for adding me in CC. While I got failure to apply this series to the latest mainline kernel, could you tell what is the base commit of this pathcset? Thanks Baoquan
On Sat, Feb 15, 2025 at 6:28 PM Baoquan He <bhe@redhat.com> wrote: > > Hi Kairui, > > On 02/15/25 at 01:57am, Kairui Song wrote: > > From: Kairui Song <kasong@tencent.com> > > > > Slot cache was initially introduced by commit 67afa38e012e ("mm/swap: > > add cache for swap slots allocation") to reduce the lock contention > > of si->lock. > > Thanks for adding me in CC. While I got failure to apply this series > to the latest mainline kernel, could you tell what is the base commit > of this pathcset? > > Thanks > Baoquan > Hi Baoquan, It's based on Andrews's mm-unstable here: git://git.kernel.org/pub/scm/linux/kernel/git/akpm/mm.git I've just re-checked, there should be no conflict. Sorry I didn't include this info in the cover letter, mm development is rapid so usually I send patch based on mm-unstable.
On 02/15/25 at 09:34pm, Kairui Song wrote: > On Sat, Feb 15, 2025 at 6:28 PM Baoquan He <bhe@redhat.com> wrote: > > > > Hi Kairui, > > > > On 02/15/25 at 01:57am, Kairui Song wrote: > > > From: Kairui Song <kasong@tencent.com> > > > > > > Slot cache was initially introduced by commit 67afa38e012e ("mm/swap: > > > add cache for swap slots allocation") to reduce the lock contention > > > of si->lock. > > > > Thanks for adding me in CC. While I got failure to apply this series > > to the latest mainline kernel, could you tell what is the base commit > > of this pathcset? > > > > Thanks > > Baoquan > > > > Hi Baoquan, > > It's based on Andrews's mm-unstable here: > git://git.kernel.org/pub/scm/linux/kernel/git/akpm/mm.git > > I've just re-checked, there should be no conflict. Sorry I didn't > include this info in the cover letter, mm development is rapid so > usually I send patch based on mm-unstable. Thanks, applied to akpm-mm/mm-unstable branch cleanly.
From: Kairui Song <kasong@tencent.com> Slot cache was initially introduced by commit 67afa38e012e ("mm/swap: add cache for swap slots allocation") to reduce the lock contention of si->lock. Previous series "mm, swap: rework of swap allocator locks" [1] removed swap slot cache for freeing path as freeing path no longer touches si->lock in most cased. Allocation path also have slight to none contention on si->lock since that series, but slot cache still helps to reduce other overheads, like counters and the plist. This series removes the slot cache from allocation path too, by using the cluster as allocation fast path and also reduce other overheads. Now slot cache is completely gone, the code is much simplified without obvious feature or performance change, also clean up related workaround. Also this should avoid other potential issues, e.g. the long pinning of swap slots: swap slot cache pins swap slots with HAS_CACHE, causing reclaim or allocation fail to use these slots on scanning. The only behavior change is the swap device allocation rotation mechanism, as explained in the patch "mm, swap: use percpu cluster as allocation fast path". Test results are looking good after deleting the swap slot cache: - vm-scalability with: `usemem --init-time -O -y -x -R -31 1G`, 12G memory cgroup using simulated pmem as SWAP (32G pmem, 32 CPUs), 16 test runs for each case, measuring the total throughput: Before (KB/s) (stdev) After (KB/s) (stdev) Random (4K): 424907.60 (24410.78) 414745.92 (34554.78) Random (64K): 163308.82 (11635.72) 167314.50 (18434.99) Sequential (4K, !-R): 6150056.79 (103205.90) 6321469.06 (115878.16) - Build linux kernel with make -j96, using 4K folio with 1.5G memory cgroup limit and 64K folio with 2G memory cgroup limit, on top of tmpfs, 12 test runs, measuring the system time: Before (s) (stdev) After (s) (stdev) make -j96 (4K): 6445.69 (61.95) 6408.80 (69.46) make -j96 (64K): 6841.71 (409.04) 6437.99 (435.55) The performance is unchanged, slightly better in some cases. [1] https://lore.kernel.org/linux-mm/20250113175732.48099-1-ryncsn@gmail.com/ Kairui Song (7): mm, swap: avoid reclaiming irrelevant swap cache mm, swap: drop the flag TTRS_DIRECT mm, swap: avoid redundant swap device pinning mm, swap: don't update the counter up-front mm, swap: use percpu cluster as allocation fast path mm, swap: remove swap slot cache mm, swap: simplify folio swap allocation include/linux/swap.h | 21 +-- include/linux/swap_slots.h | 28 ---- mm/Makefile | 2 +- mm/shmem.c | 21 +-- mm/swap.h | 6 - mm/swap_slots.c | 295 ---------------------------------- mm/swap_state.c | 79 ++-------- mm/swapfile.c | 315 ++++++++++++++++++------------------- mm/vmscan.c | 16 +- mm/zswap.c | 6 + 10 files changed, 196 insertions(+), 593 deletions(-) delete mode 100644 include/linux/swap_slots.h delete mode 100644 mm/swap_slots.c