On an idle system with large amount of cpus it might happen that
klp_update_patch_state() is not reached in do_idle() for a long periods
of time. With debug messages enabled log is filled with:
[ 499.442643] livepatch: klp_try_switch_task: swapper/63:0 is running
without any signs of progress. Ending up with "failed to complete
transition".
On s390 LPAR with 128 cpus not a single transition is able to complete
and livepatch kselftests fail. Tests on idling x86 kvm instance with 128
cpus demonstrate similar symptoms with and without CONFIG_NO_HZ.
To deal with that, since runqueue is already locked in
klp_try_switch_task() identify idling cpus and trigger rescheduling
potentially waking them up and making sure idle tasks break out of
do_idle() inner loop and reach klp_update_patch_state(). This helps to
speed up transition time while avoiding unnecessary extra system load.
Signed-off-by: Vasily Gorbik <[email protected]>
---
Previous discussion and RFC PATCH:
lkml.kernel.org/r/patch.git-b76842ceb035.your-ad-here.call-01625661932-ext-1304@work.hours
kernel/livepatch/transition.c | 2 ++
1 file changed, 2 insertions(+)
diff --git a/kernel/livepatch/transition.c b/kernel/livepatch/transition.c
index 3a4beb9395c4..c5832b2dd081 100644
--- a/kernel/livepatch/transition.c
+++ b/kernel/livepatch/transition.c
@@ -308,6 +308,8 @@ static bool klp_try_switch_task(struct task_struct *task)
rq = task_rq_lock(task, &flags);
if (task_running(rq, task) && task != current) {
+ if (is_idle_task(task))
+ resched_curr(rq);
snprintf(err_buf, STACK_ERR_BUF_SIZE,
"%s: %s:%d is running\n", __func__, task->comm,
task->pid);
--
2.25.4
On Thu 2021-09-09 11:16:01, Vasily Gorbik wrote:
> On an idle system with large amount of cpus it might happen that
> klp_update_patch_state() is not reached in do_idle() for a long periods
> of time. With debug messages enabled log is filled with:
> [ 499.442643] livepatch: klp_try_switch_task: swapper/63:0 is running
>
> without any signs of progress. Ending up with "failed to complete
> transition".
>
> On s390 LPAR with 128 cpus not a single transition is able to complete
> and livepatch kselftests fail. Tests on idling x86 kvm instance with 128
> cpus demonstrate similar symptoms with and without CONFIG_NO_HZ.
>
> To deal with that, since runqueue is already locked in
> klp_try_switch_task() identify idling cpus and trigger rescheduling
> potentially waking them up and making sure idle tasks break out of
> do_idle() inner loop and reach klp_update_patch_state(). This helps to
> speed up transition time while avoiding unnecessary extra system load.
>
> Signed-off-by: Vasily Gorbik <[email protected]>
> ---
> Previous discussion and RFC PATCH:
> lkml.kernel.org/r/patch.git-b76842ceb035.your-ad-here.call-01625661932-ext-1304@work.hours
>
> kernel/livepatch/transition.c | 2 ++
> 1 file changed, 2 insertions(+)
>
> diff --git a/kernel/livepatch/transition.c b/kernel/livepatch/transition.c
> index 3a4beb9395c4..c5832b2dd081 100644
> --- a/kernel/livepatch/transition.c
> +++ b/kernel/livepatch/transition.c
> @@ -308,6 +308,8 @@ static bool klp_try_switch_task(struct task_struct *task)
Please, update also the comment above klp_try_switch_task(). I would
write something like:
/*
* Try to safely switch a task to the target patch state. If it's currently
* running, or it's sleeping on a to-be-patched or to-be-unpatched function, or
* if the stack is unreliable, return false.
*
* Idle tasks are switched in the main loop when running.
*/
> rq = task_rq_lock(task, &flags);
>
> if (task_running(rq, task) && task != current) {
This would deserve a comment, for example:
/*
* Idle task might stay running for a long time. Switch them
* in the main loop.
*/
> + if (is_idle_task(task))
> + resched_curr(rq);
> snprintf(err_buf, STACK_ERR_BUF_SIZE,
> "%s: %s:%d is running\n", __func__, task->comm,
> task->pid);
Otherwise, it looks good to me. With the two comments:
Reviewed-by: Petr Mladek <[email protected]>
Best Regards,
Petr
On Thu, 9 Sep 2021, Vasily Gorbik wrote:
> On an idle system with large amount of cpus it might happen that
> klp_update_patch_state() is not reached in do_idle() for a long periods
> of time. With debug messages enabled log is filled with:
> [ 499.442643] livepatch: klp_try_switch_task: swapper/63:0 is running
>
> without any signs of progress. Ending up with "failed to complete
> transition".
>
> On s390 LPAR with 128 cpus not a single transition is able to complete
> and livepatch kselftests fail. Tests on idling x86 kvm instance with 128
> cpus demonstrate similar symptoms with and without CONFIG_NO_HZ.
>
> To deal with that, since runqueue is already locked in
> klp_try_switch_task() identify idling cpus and trigger rescheduling
> potentially waking them up and making sure idle tasks break out of
> do_idle() inner loop and reach klp_update_patch_state(). This helps to
> speed up transition time while avoiding unnecessary extra system load.
>
> Signed-off-by: Vasily Gorbik <[email protected]>
Seems reasonable to me.
Acked-by: Miroslav Benes <[email protected]>
M
On Thu, Sep 09, 2021 at 11:16:01AM +0200, Vasily Gorbik wrote:
> On an idle system with large amount of cpus it might happen that
> klp_update_patch_state() is not reached in do_idle() for a long periods
> of time. With debug messages enabled log is filled with:
> [ 499.442643] livepatch: klp_try_switch_task: swapper/63:0 is running
>
> without any signs of progress. Ending up with "failed to complete
> transition".
>
> On s390 LPAR with 128 cpus not a single transition is able to complete
> and livepatch kselftests fail. Tests on idling x86 kvm instance with 128
> cpus demonstrate similar symptoms with and without CONFIG_NO_HZ.
>
> To deal with that, since runqueue is already locked in
> klp_try_switch_task() identify idling cpus and trigger rescheduling
> potentially waking them up and making sure idle tasks break out of
> do_idle() inner loop and reach klp_update_patch_state(). This helps to
> speed up transition time while avoiding unnecessary extra system load.
>
> Signed-off-by: Vasily Gorbik <[email protected]>
Looks ok to me, but we should get an ack from Ingo or Peter since
livepatch would be calling another private scheduler interface.
> ---
> Previous discussion and RFC PATCH:
> lkml.kernel.org/r/patch.git-b76842ceb035.your-ad-here.call-01625661932-ext-1304@work.hours
>
> kernel/livepatch/transition.c | 2 ++
> 1 file changed, 2 insertions(+)
>
> diff --git a/kernel/livepatch/transition.c b/kernel/livepatch/transition.c
> index 3a4beb9395c4..c5832b2dd081 100644
> --- a/kernel/livepatch/transition.c
> +++ b/kernel/livepatch/transition.c
> @@ -308,6 +308,8 @@ static bool klp_try_switch_task(struct task_struct *task)
> rq = task_rq_lock(task, &flags);
>
> if (task_running(rq, task) && task != current) {
> + if (is_idle_task(task))
> + resched_curr(rq);
> snprintf(err_buf, STACK_ERR_BUF_SIZE,
> "%s: %s:%d is running\n", __func__, task->comm,
> task->pid);
> --
> 2.25.4
>
--
Josh