linux/include
Feng Tang d2226ebd54 mm/hugetlb: add dedicated func to get 'allowed' nodemask for current process
Muchun Song found that after MPOL_PREFERRED_MANY policy was introduced in
commit b27abaccf8 ("mm/mempolicy: add MPOL_PREFERRED_MANY for multiple
preferred nodes"), the policy_nodemask_current()'s semantics for this new
policy has been changed, which returns 'preferred' nodes instead of
'allowed' nodes.

With the changed semantic of policy_nodemask_current, a task with
MPOL_PREFERRED_MANY policy could fail to get its reservation even though
it can fall back to other nodes (either defined by cpusets or all online
nodes) for that reservation failing mmap calles unnecessarily early.

The fix is to not consider MPOL_PREFERRED_MANY for reservations at all
because they, unlike MPOL_MBIND, do not pose any actual hard constrain.

Michal suggested the policy_nodemask_current() is only used by hugetlb,
and could be moved to hugetlb code with more explicit name to enforce the
'allowed' semantics for which only MPOL_BIND policy matters.

apply_policy_zone() is made extern to be called in hugetlb code and its
return value is changed to bool.

[1]. https://lore.kernel.org/lkml/20220801084207.39086-1-songmuchun@bytedance.com/t/

Link: https://lkml.kernel.org/r/20220805005903.95563-1-feng.tang@intel.com
Fixes: b27abaccf8 ("mm/mempolicy: add MPOL_PREFERRED_MANY for multiple preferred nodes")
Signed-off-by: Feng Tang <feng.tang@intel.com>
Reported-by: Muchun Song <songmuchun@bytedance.com>
Suggested-by: Michal Hocko <mhocko@suse.com>
Acked-by: Michal Hocko <mhocko@suse.com>
Reviewed-by: Muchun Song <songmuchun@bytedance.com>
Cc: Mike Kravetz <mike.kravetz@oracle.com>
Cc: Dave Hansen <dave.hansen@intel.com>
Cc: Ben Widawsky <bwidawsk@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
2022-09-11 20:25:50 -07:00
..
acpi Merge branch 'acpi-properties' 2022-08-11 19:21:03 +02:00
asm-generic Seventeen hotfixes. Mostly memory management things. Ten patches are 2022-08-28 14:49:59 -07:00
clocksource
crypto for-5.20/block-2022-08-04 2022-08-04 20:00:14 -07:00
drm
dt-bindings power supply and reset changes for the v6.0 series 2022-08-12 09:37:33 -07:00
keys
kunit
kvm
linux mm/hugetlb: add dedicated func to get 'allowed' nodemask for current process 2022-09-11 20:25:50 -07:00
math-emu
media SPDX changes for 6.0-rc1 2022-08-04 12:12:54 -07:00
memory
misc
net Merge git://git.kernel.org/pub/scm/linux/kernel/git/netfilter/nf 2022-08-24 19:18:10 -07:00
pcmcia
ras mm, hwpoison: enable memory error handling on 1GB hugepage 2022-08-08 18:06:44 -07:00
rdma dma-mapping updates 2022-08-06 10:56:45 -07:00
rv
scsi SCSI misc on 20220813 2022-08-13 13:41:48 -07:00
soc net: mscc: ocelot: keep ocelot_stat_layout by reg address, not offset 2022-08-17 21:58:32 -07:00
sound
target
trace mm/khugepaged: record SCAN_PMD_MAPPED when scan_pmd() finds hugepage 2022-09-11 20:25:46 -07:00
uapi userfaultfd: add /dev/userfaultfd for fine grained access control 2022-09-11 20:25:48 -07:00
ufs scsi: ufs: core: Enable link lost interrupt 2022-08-11 22:04:32 -04:00
vdso
video
xen x86/xen: Add support for HVMOP_set_evtchn_upcall_vector 2022-08-12 11:28:21 +02:00