Kernel/linux
1
0
Fork 0
mirror of https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git synced 2025-01-04 04:04:19 +00:00
Code Issues Actions Packages Projects Releases Wiki Activity
95599ef684
linux/include
History
Yu Zhao 95599ef684 mm/codetag: fix pgalloc_tag_split() 
The current assumption is that a large folio can only be split into
order-0 folios.  That is not the case for hugeTLB demotion, nor for THP
split: see commit c010d47f10 ("mm: thp: split huge page to any lower
order pages").

When a large folio is split into ones of a lower non-zero order, only the
new head pages should be tagged.  Tagging tail pages can cause imbalanced
"calls" counters, since only head pages are untagged by pgalloc_tag_sub()
and the "calls" counts on tail pages are leaked, e.g.,

  # echo 2048kB >/sys/kernel/mm/hugepages/hugepages-1048576kB/demote_size
  # echo 700 >/sys/kernel/mm/hugepages/hugepages-1048576kB/nr_hugepages
  # time echo 700 >/sys/kernel/mm/hugepages/hugepages-1048576kB/demote
  # echo 0 >/sys/kernel/mm/hugepages/hugepages-2048kB/nr_hugepages
  # grep alloc_gigantic_folio /proc/allocinfo

Before this patch:
  0  549427200  mm/hugetlb.c:1549 func:alloc_gigantic_folio

  real  0m2.057s
  user  0m0.000s
  sys   0m2.051s

After this patch:
  0          0  mm/hugetlb.c:1549 func:alloc_gigantic_folio

  real  0m1.711s
  user  0m0.000s
  sys   0m1.704s

Not tagging tail pages also improves the splitting time, e.g., by about
15% when demoting 1GB hugeTLB folios to 2MB ones, as shown above.

Link: https://lkml.kernel.org/r/20240906042108.1150526-2-yuzhao@google.com
Fixes: be25d1d4e8 ("mm: create new codetag references during page splitting")
Signed-off-by: Yu Zhao <yuzhao@google.com>
Acked-by: Suren Baghdasaryan <surenb@google.com>
Cc: Kent Overstreet <kent.overstreet@linux.dev>
Cc: Muchun Song <muchun.song@linux.dev>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
2024-09-09 16:39:18 -07:00
..
acpi ACPI: video: Add Dell UART backlight controller detection 2024-08-19 15:58:35 +02:00
asm-generic arch_numa: switch over to numa_memblks 2024-09-03 21:15:32 -07:00
clocksource
crypto
drm A revert for a previous TTM commit causing stuttering, 3 fixes for 2024-08-30 11:28:11 +10:00
dt-bindings I3C for 6.11 2024-07-27 10:53:06 -07:00
keys
kunit kunit: Introduce KUNIT_ASSERT_MEMEQ and KUNIT_ASSERT_MEMNEQ macros 2024-07-12 10:12:11 -06:00
kvm
linux mm/codetag: fix pgalloc_tag_split() 2024-09-09 16:39:18 -07:00
math-emu
media media: rc: add keymap for MyGica UTV3 remote 2024-06-29 12:20:05 +02:00
memory
misc
net netfilter pull request 24-08-28 2024-08-29 11:35:54 +02:00
pcmcia
ras mm/memory-failure: remove obsolete MF_MSG_DIFFERENT_COMPOUND 2024-07-12 15:52:22 -07:00
rdma RDMA/core: Introduce "name_assign_type" for an IB device 2024-07-04 07:59:53 +03:00
rv
scsi scsi: core: Fix the return value of scsi_logical_block_count() 2024-08-16 21:02:06 -04:00
soc net: mscc: ocelot: treat 802.1ad tagged traffic as 802.1Q-untagged 2024-08-16 09:59:32 +01:00
sound ASoC: Fixes for v6.11 2024-08-09 09:58:07 +02:00
target
trace x86: remove PG_uncached 2024-09-03 21:15:46 -07:00
uapi mm: remove PG_error 2024-09-01 20:26:05 -07:00
ufs scsi: ufs: core: Add a quirk for handling broken LSDBS field in controller capabilities register 2024-08-16 21:09:17 -04:00
vdso Random number generator updates for Linux 6.11-rc1. 2024-07-24 10:29:50 -07:00
video
xen Driver core changes for 6.11-rc1 2024-07-25 10:42:22 -07:00