Hello, Linus.
Late percpu pull request for v4.16-rc6.
* percpu allocator pool replenishing no longer triggers OOM or warning
messages. Also, the alloc interface now understands __GFP_NORETRY
and __GFP_NOWARN. This is to allow avoiding OOMs from userland
triggered actions like bpf map creation. Also added cond_resched()
in alloc loop.
* perpcu allocation now can be interrupted by kill sigs to avoid
deadlocking OOM killer.
* Added Dennis Zhou as a co-maintainer. He has rewritten the area map
allocator, understands most of the code base and has been responsive
for all bug reports.
Thanks.
The following changes since commit 7928b2cbe55b2a410a0f5c1f154610059c57b1b2:
Linux 4.16-rc1 (2018-02-11 15:04:29 -0800)
are available in the git repository at:
git://git.kernel.org/pub/scm/linux/kernel/git/tj/percpu.git for-4.16-fixes
for you to fetch changes up to b3a5d111994450909158929560906f2c1c6c1d85:
percpu_ref: Update doc to dissuade users from depending on internal RCU grace
periods (2018-03-19 10:09:44 -0700)
----------------------------------------------------------------
Dennis Zhou (3):
percpu: match chunk allocator declarations with definitions
percpu: add __GFP_NORETRY semantics to the percpu balancing path
percpu: allow select gfp to be passed to underlying allocators
Eric Dumazet (1):
percpu: add a schedule point in pcpu_balance_workfn()
Kirill Tkhai (1):
mm: Allow to kill tasks doing pcpu_alloc() and waiting for
pcpu_balance_workfn()
Tejun Heo (3):
percpu: add Dennis Zhou as a percpu co-maintainer
percpu: include linux/sched.h for cond_resched()
percpu_ref: Update doc to dissuade users from depending on internal RCU
grace periods
MAINTAINERS | 1 +
include/linux/percpu-refcount.h | 18 +++++++----
lib/percpu-refcount.c | 2 ++
mm/percpu-km.c | 8 ++---
mm/percpu-vm.c | 18 ++++++-----
mm/percpu.c | 67 +++++++++++++++++++++++++++--------------
6 files changed, 74 insertions(+), 40 deletions(-)
