From: Volodymyr Babchuk <Volodymyr_Babchuk@epam.com>
To: "xen-devel@lists.xenproject.org" <xen-devel@lists.xenproject.org>
Cc: Volodymyr Babchuk <Volodymyr_Babchuk@epam.com>,
Stefano Stabellini <sstabellini@kernel.org>,
Julien Grall <julien@xen.org>,
Volodymyr Babchuk <Volodymyr_Babchuk@epam.com>,
George Dunlap <george.dunlap@citrix.com>,
Dario Faggioli <dfaggioli@suse.com>
Subject: [RFC PATCH 07/10] sched: core: remove ASSERT_NOT_IN_ATOMIC and disable preemption[!]
Date: Tue, 23 Feb 2021 02:34:57 +0000 [thread overview]
Message-ID: <20210223023428.757694-8-volodymyr_babchuk@epam.com> (raw)
In-Reply-To: <20210223023428.757694-1-volodymyr_babchuk@epam.com>
ASSERT_NOT_IN_ATOMIC() is very strict, because it checks that
local IRQs are disabled. But there is a case when this is okay:
when we finished handling IRQ in hypervisor mode we might want
to preempt current vCPU. In this case scheduler will be called
with local IRQs disabled.
On other hand, we want to ensure that scheduler code is not preempted
itself. So we need to disable preemption while doing scheduling.
WARNING! This patch works only for ARM code, because ARM code returns
after call to sched_context_switch() and it is able to enable
preemption back. In case of x86 further investigation required.
Signed-off-by: Volodymyr Babchuk <volodymyr_babchuk@epam.com>
---
xen/arch/arm/domain.c | 3 +++
xen/common/sched/core.c | 13 ++++++++++---
2 files changed, 13 insertions(+), 3 deletions(-)
diff --git a/xen/arch/arm/domain.c b/xen/arch/arm/domain.c
index bdd3d3e5b5..2ccf4449ea 100644
--- a/xen/arch/arm/domain.c
+++ b/xen/arch/arm/domain.c
@@ -335,6 +335,9 @@ static void continue_new_vcpu(struct vcpu *prev)
schedule_tail(prev);
+ /* This matches preempt_disable() in schedule() */
+ preempt_enable_no_sched();
+
if ( is_idle_vcpu(current) )
reset_stack_and_jump(idle_loop);
else if ( is_32bit_domain(current->domain) )
diff --git a/xen/common/sched/core.c b/xen/common/sched/core.c
index 7e075613d5..057b558367 100644
--- a/xen/common/sched/core.c
+++ b/xen/common/sched/core.c
@@ -2577,8 +2577,6 @@ static void sched_slave(void)
unsigned int cpu = smp_processor_id();
unsigned long flags;
- ASSERT_NOT_IN_ATOMIC();
-
rcu_read_lock(&sched_res_rculock);
lock = pcpu_schedule_lock_irqsave(cpu, &flags);
@@ -2643,7 +2641,7 @@ static void schedule(void)
int cpu = smp_processor_id();
unsigned int gran;
- ASSERT_NOT_IN_ATOMIC();
+ preempt_disable();
SCHED_STAT_CRANK(sched_run);
@@ -2665,6 +2663,9 @@ static void schedule(void)
rcu_read_unlock(&sched_res_rculock);
raise_softirq(SCHEDULE_SOFTIRQ);
+
+ preempt_enable_no_sched();
+
return sched_slave();
}
@@ -2681,7 +2682,10 @@ static void schedule(void)
cpumask_raise_softirq(mask, SCHED_SLAVE_SOFTIRQ);
next = sched_wait_rendezvous_in(prev, &lock, &flags, cpu, now);
if ( !next )
+ {
+ preempt_enable_no_sched();
return;
+ }
}
else
{
@@ -2695,6 +2699,9 @@ static void schedule(void)
vnext = sched_unit2vcpu_cpu(next, cpu);
sched_context_switch(vprev, vnext,
!is_idle_unit(prev) && is_idle_unit(next), now);
+
+ /* XXX: Move me */
+ preempt_enable_no_sched();
}
/* The scheduler timer: force a run through the scheduler */
--
2.29.2
next prev parent reply other threads:[~2021-02-23 2:35 UTC|newest]
Thread overview: 26+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-02-23 2:34 [RFC PATCH 00/10] Preemption in hypervisor (ARM only) Volodymyr Babchuk
2021-02-23 2:34 ` [RFC PATCH 01/10] sched: core: save IRQ state during locking Volodymyr Babchuk
2021-02-23 8:52 ` Jürgen Groß
2021-02-23 11:15 ` Volodymyr Babchuk
2021-02-24 18:29 ` Andrew Cooper
2021-02-23 2:34 ` [RFC PATCH 03/10] sched: credit2: " Volodymyr Babchuk
2021-02-23 2:34 ` [RFC PATCH 02/10] sched: rt: " Volodymyr Babchuk
2021-02-23 2:34 ` [RFC PATCH 04/10] preempt: use atomic_t to for preempt_count Volodymyr Babchuk
2021-02-23 2:34 ` [RFC PATCH 05/10] preempt: add try_preempt() function Volodymyr Babchuk
2021-02-23 2:34 ` Volodymyr Babchuk [this message]
2021-02-23 2:34 ` [RFC PATCH 06/10] arm: setup: disable preemption during startup Volodymyr Babchuk
2021-02-23 2:34 ` [RFC PATCH 08/10] arm: context_switch: allow to run with IRQs already disabled Volodymyr Babchuk
2021-02-23 2:34 ` [RFC PATCH 10/10] [HACK] alloc pages: enable preemption early Volodymyr Babchuk
2021-02-23 2:34 ` [RFC PATCH 09/10] arm: traps: try to preempt before leaving IRQ handler Volodymyr Babchuk
2021-02-23 9:02 ` [RFC PATCH 00/10] Preemption in hypervisor (ARM only) Julien Grall
2021-02-23 12:06 ` Volodymyr Babchuk
2021-02-24 10:08 ` Julien Grall
2021-02-24 20:57 ` Volodymyr Babchuk
2021-02-24 22:31 ` Julien Grall
2021-02-24 23:58 ` Volodymyr Babchuk
2021-02-25 0:39 ` Andrew Cooper
2021-02-25 12:51 ` Volodymyr Babchuk
2021-03-05 9:31 ` Volodymyr Babchuk
2021-02-24 18:07 ` Andrew Cooper
2021-02-24 23:37 ` Volodymyr Babchuk
2021-03-01 14:39 ` George Dunlap
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20210223023428.757694-8-volodymyr_babchuk@epam.com \
--to=volodymyr_babchuk@epam.com \
--cc=dfaggioli@suse.com \
--cc=george.dunlap@citrix.com \
--cc=julien@xen.org \
--cc=sstabellini@kernel.org \
--cc=xen-devel@lists.xenproject.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.