| From: |
| Peter Zijlstra <peterz-AT-infradead.org> |
| To: |
| tglx-AT-linutronix.de, mingo-AT-kernel.org |
| Subject: |
| [PATCH 0/9] sched: Migrate disable support |
| Date: |
| Mon, 21 Sep 2020 18:35:57 +0200 |
| Message-ID: |
| <20200921163557.234036895@infradead.org> |
| Cc: |
| linux-kernel-AT-vger.kernel.org, bigeasy-AT-linutronix.de, qais.yousef-AT-arm.com, swood-AT-redhat.com, peterz-AT-infradead.org, valentin.schneider-AT-arm.com, juri.lelli-AT-redhat.com, vincent.guittot-AT-linaro.org, dietmar.eggemann-AT-arm.com, rostedt-AT-goodmis.org, bsegall-AT-google.com, mgorman-AT-suse.de, bristot-AT-redhat.com, vincent.donnefort-AT-arm.com |
| Archive-link: |
| Article |
Hi,
Here's my take on migrate_disable(). It avoids growing a second means of
changing the affinity, documents how the things violates locking rules but
still mostly works.
It also avoids blocking completely, so no more futex band-aids required.
Also, no more atomics/locks on the fast path.
I also put in a comment that explains how the whole concept is fundamentally
flawed but a necessary evil to get PREEMPT_RT going -- for now.
Somewhat tested with PREEMPT_RT.
---
include/linux/cpuhotplug.h | 1
include/linux/preempt.h | 60 +++
include/linux/sched.h | 4
include/linux/sched/hotplug.h | 2
include/linux/stop_machine.h | 5
kernel/cpu.c | 9
kernel/sched/core.c | 673 +++++++++++++++++++++++++++++++-----------
kernel/sched/deadline.c | 5
kernel/sched/sched.h | 24 +
kernel/stop_machine.c | 23 +
lib/dump_stack.c | 2
lib/smp_processor_id.c | 5
12 files changed, 635 insertions(+), 178 deletions(-)
