Re: [Intel-gfx] [PATCH] softirq: Prevent looping on disabled tasklets
Hi Chris, [auto build test ERROR on linus/master] [also build test ERROR on v4.10-rc7 next-20170210] [if your patch is applied to the wrong git tree, please drop us a note to help improve the system] url: https://github.com/0day-ci/linux/commits/Chris-Wilson/softirq-Prevent-looping-on-disabled-tasklets/20170212-220435 config: x86_64-rhel (attached as .config) compiler: gcc-6 (Debian 6.2.0-3) 6.2.0 20160901 reproduce: # save the attached .config to linux build tree make ARCH=x86_64 All errors (new ones prefixed by >>): >> ERROR: "raise_softirq" [sound/pci/asihpi/snd-asihpi.ko] undefined! >> ERROR: "raise_softirq" [net/mac802154/mac802154.ko] undefined! ERROR: "raise_softirq" [net/mac80211/mac80211.ko] undefined! >> ERROR: "raise_softirq" [drivers/usb/atm/usbatm.ko] undefined! >> ERROR: "raise_softirq" [drivers/net/ethernet/jme.ko] undefined! >> ERROR: "raise_softirq" [drivers/media/pci/mantis/mantis_core.ko] undefined! >> ERROR: "raise_softirq" [drivers/hv/hv_vmbus.ko] undefined! >> ERROR: "raise_softirq" [drivers/firewire/firewire-ohci.ko] undefined! --- 0-DAY kernel test infrastructureOpen Source Technology Center https://lists.01.org/pipermail/kbuild-all Intel Corporation .config.gz Description: application/gzip
Re: [Intel-gfx] [PATCH] softirq: Prevent looping on disabled tasklets
Hi Chris, [auto build test ERROR on linus/master] [also build test ERROR on v4.10-rc7 next-20170210] [if your patch is applied to the wrong git tree, please drop us a note to help improve the system] url: https://github.com/0day-ci/linux/commits/Chris-Wilson/softirq-Prevent-looping-on-disabled-tasklets/20170212-220435 config: x86_64-rhel (attached as .config) compiler: gcc-6 (Debian 6.2.0-3) 6.2.0 20160901 reproduce: # save the attached .config to linux build tree make ARCH=x86_64 All errors (new ones prefixed by >>): >> ERROR: "raise_softirq" [sound/pci/asihpi/snd-asihpi.ko] undefined! >> ERROR: "raise_softirq" [net/mac802154/mac802154.ko] undefined! ERROR: "raise_softirq" [net/mac80211/mac80211.ko] undefined! >> ERROR: "raise_softirq" [drivers/usb/atm/usbatm.ko] undefined! >> ERROR: "raise_softirq" [drivers/net/ethernet/jme.ko] undefined! >> ERROR: "raise_softirq" [drivers/media/pci/mantis/mantis_core.ko] undefined! >> ERROR: "raise_softirq" [drivers/hv/hv_vmbus.ko] undefined! >> ERROR: "raise_softirq" [drivers/firewire/firewire-ohci.ko] undefined! --- 0-DAY kernel test infrastructureOpen Source Technology Center https://lists.01.org/pipermail/kbuild-all Intel Corporation .config.gz Description: application/gzip
Re: [Intel-gfx] [PATCH] softirq: Prevent looping on disabled tasklets
Hi Chris, [auto build test ERROR on linus/master] [also build test ERROR on v4.10-rc7 next-20170210] [if your patch is applied to the wrong git tree, please drop us a note to help improve the system] url: https://github.com/0day-ci/linux/commits/Chris-Wilson/softirq-Prevent-looping-on-disabled-tasklets/20170212-220435 config: x86_64-lkp (attached as .config) compiler: gcc-6 (Debian 6.2.0-3) 6.2.0 20160901 reproduce: # save the attached .config to linux build tree make ARCH=x86_64 All errors (new ones prefixed by >>): >> ERROR: "raise_softirq" [net/mac80211/mac80211.ko] undefined! --- 0-DAY kernel test infrastructureOpen Source Technology Center https://lists.01.org/pipermail/kbuild-all Intel Corporation .config.gz Description: application/gzip
Re: [Intel-gfx] [PATCH] softirq: Prevent looping on disabled tasklets
Hi Chris, [auto build test ERROR on linus/master] [also build test ERROR on v4.10-rc7 next-20170210] [if your patch is applied to the wrong git tree, please drop us a note to help improve the system] url: https://github.com/0day-ci/linux/commits/Chris-Wilson/softirq-Prevent-looping-on-disabled-tasklets/20170212-220435 config: x86_64-lkp (attached as .config) compiler: gcc-6 (Debian 6.2.0-3) 6.2.0 20160901 reproduce: # save the attached .config to linux build tree make ARCH=x86_64 All errors (new ones prefixed by >>): >> ERROR: "raise_softirq" [net/mac80211/mac80211.ko] undefined! --- 0-DAY kernel test infrastructureOpen Source Technology Center https://lists.01.org/pipermail/kbuild-all Intel Corporation .config.gz Description: application/gzip
Re: [PATCH] softirq: Prevent looping on disabled tasklets
On Sun, Feb 12, 2017 at 02:00:19PM +, Chris Wilson wrote:
> Disabling a tasklet causes it not to run during tasklet_action, but is
> put back onto the runnable tasklet list, and a new softirq raised. As
> the softirq is raised from within __do_softirq() this causing
> __do_softirq() to loop constantly until its timeslice expires and is
> transferred to the ksoftirq thread. ksoftirq then permanently spins,
> as on each action, the disabled tasklet keeps reraising the softirq.
>
> Break this vicious cycle by moving the softirq from the action to the
> final tasklet_enable().
>
> This behaviour appears to be historic (since the first git import).
> However, the looping until timeslice duration (to a max of 2ms) was
> first introduced in commit c10d73671ad3 ("softirq: reduce latencies"),
> with the restart limit restored in commit 34376a50fb1f ("Fix lockup
> related to stop_machine being stuck in __do_softirq.")
>
> Reported-by: Tvrtko Ursulin
> Signed-off-by: Chris Wilson
> Cc: Tvrtko Ursulin
> Cc: Thomas Gleixner
> Cc: Hannes Reinecke
> Cc: Jens Axboe
> Cc: Bjorn Helgaas
> Cc: Alexander Potapenko
> Cc: Chen Fan
> Cc: Ingo Molnar
> Cc: "Peter Zijlstra (Intel)"
> Cc: Sebastian Andrzej Siewior
> Cc: Johannes Thumshirn
> Cc: Emese Revfy
> Cc: Sagi Grimberg
> Cc: Eric Dumazet
> Cc: Tom Herbert
> Cc: Ben Hutchings
> ---
> include/linux/interrupt.h | 7 +--
> kernel/softirq.c | 2 --
> 2 files changed, 5 insertions(+), 4 deletions(-)
>
> diff --git a/include/linux/interrupt.h b/include/linux/interrupt.h
> index 53144e78a369..12750f00d00d 100644
> --- a/include/linux/interrupt.h
> +++ b/include/linux/interrupt.h
> @@ -613,8 +613,11 @@ static inline void tasklet_disable(struct tasklet_struct
> *t)
>
> static inline void tasklet_enable(struct tasklet_struct *t)
> {
> - smp_mb__before_atomic();
> - atomic_dec(>count);
> + if (!atomic_dec_and_test(>count))
> + return;
> +
> + if (test_bit(TASKLET_STATE_SCHED, >state))
> + raise_softirq(HI_SOFTIRQ | TASKLET_SOFTIRQ);
raise_softirq is not exported, so let's move tasklet_enable()
out-of-line and export that.
-Chris
--
Chris Wilson, Intel Open Source Technology Centre
Re: [PATCH] softirq: Prevent looping on disabled tasklets
On Sun, Feb 12, 2017 at 02:00:19PM +, Chris Wilson wrote:
> Disabling a tasklet causes it not to run during tasklet_action, but is
> put back onto the runnable tasklet list, and a new softirq raised. As
> the softirq is raised from within __do_softirq() this causing
> __do_softirq() to loop constantly until its timeslice expires and is
> transferred to the ksoftirq thread. ksoftirq then permanently spins,
> as on each action, the disabled tasklet keeps reraising the softirq.
>
> Break this vicious cycle by moving the softirq from the action to the
> final tasklet_enable().
>
> This behaviour appears to be historic (since the first git import).
> However, the looping until timeslice duration (to a max of 2ms) was
> first introduced in commit c10d73671ad3 ("softirq: reduce latencies"),
> with the restart limit restored in commit 34376a50fb1f ("Fix lockup
> related to stop_machine being stuck in __do_softirq.")
>
> Reported-by: Tvrtko Ursulin
> Signed-off-by: Chris Wilson
> Cc: Tvrtko Ursulin
> Cc: Thomas Gleixner
> Cc: Hannes Reinecke
> Cc: Jens Axboe
> Cc: Bjorn Helgaas
> Cc: Alexander Potapenko
> Cc: Chen Fan
> Cc: Ingo Molnar
> Cc: "Peter Zijlstra (Intel)"
> Cc: Sebastian Andrzej Siewior
> Cc: Johannes Thumshirn
> Cc: Emese Revfy
> Cc: Sagi Grimberg
> Cc: Eric Dumazet
> Cc: Tom Herbert
> Cc: Ben Hutchings
> ---
> include/linux/interrupt.h | 7 +--
> kernel/softirq.c | 2 --
> 2 files changed, 5 insertions(+), 4 deletions(-)
>
> diff --git a/include/linux/interrupt.h b/include/linux/interrupt.h
> index 53144e78a369..12750f00d00d 100644
> --- a/include/linux/interrupt.h
> +++ b/include/linux/interrupt.h
> @@ -613,8 +613,11 @@ static inline void tasklet_disable(struct tasklet_struct
> *t)
>
> static inline void tasklet_enable(struct tasklet_struct *t)
> {
> - smp_mb__before_atomic();
> - atomic_dec(>count);
> + if (!atomic_dec_and_test(>count))
> + return;
> +
> + if (test_bit(TASKLET_STATE_SCHED, >state))
> + raise_softirq(HI_SOFTIRQ | TASKLET_SOFTIRQ);
raise_softirq is not exported, so let's move tasklet_enable()
out-of-line and export that.
-Chris
--
Chris Wilson, Intel Open Source Technology Centre
[PATCH] softirq: Prevent looping on disabled tasklets
Disabling a tasklet causes it not to run during tasklet_action, but is
put back onto the runnable tasklet list, and a new softirq raised. As
the softirq is raised from within __do_softirq() this causing
__do_softirq() to loop constantly until its timeslice expires and is
transferred to the ksoftirq thread. ksoftirq then permanently spins,
as on each action, the disabled tasklet keeps reraising the softirq.
Break this vicious cycle by moving the softirq from the action to the
final tasklet_enable().
This behaviour appears to be historic (since the first git import).
However, the looping until timeslice duration (to a max of 2ms) was
first introduced in commit c10d73671ad3 ("softirq: reduce latencies"),
with the restart limit restored in commit 34376a50fb1f ("Fix lockup
related to stop_machine being stuck in __do_softirq.")
Reported-by: Tvrtko Ursulin
Signed-off-by: Chris Wilson
Cc: Tvrtko Ursulin
Cc: Thomas Gleixner
Cc: Hannes Reinecke
Cc: Jens Axboe
Cc: Bjorn Helgaas
Cc: Alexander Potapenko
Cc: Chen Fan
Cc: Ingo Molnar
Cc: "Peter Zijlstra (Intel)"
Cc: Sebastian Andrzej Siewior
Cc: Johannes Thumshirn
Cc: Emese Revfy
Cc: Sagi Grimberg
Cc: Eric Dumazet
Cc: Tom Herbert
Cc: Ben Hutchings
---
include/linux/interrupt.h | 7 +--
kernel/softirq.c | 2 --
2 files changed, 5 insertions(+), 4 deletions(-)
diff --git a/include/linux/interrupt.h b/include/linux/interrupt.h
index 53144e78a369..12750f00d00d 100644
--- a/include/linux/interrupt.h
+++ b/include/linux/interrupt.h
@@ -613,8 +613,11 @@ static inline void tasklet_disable(struct tasklet_struct
*t)
static inline void tasklet_enable(struct tasklet_struct *t)
{
- smp_mb__before_atomic();
- atomic_dec(>count);
+ if (!atomic_dec_and_test(>count))
+ return;
+
+ if (test_bit(TASKLET_STATE_SCHED, >state))
+ raise_softirq(HI_SOFTIRQ | TASKLET_SOFTIRQ);
}
extern void tasklet_kill(struct tasklet_struct *t);
diff --git a/kernel/softirq.c b/kernel/softirq.c
index 744fa611cae0..5a359eb1a541 100644
--- a/kernel/softirq.c
+++ b/kernel/softirq.c
@@ -527,7 +527,6 @@ static __latent_entropy void tasklet_action(struct
softirq_action *a)
t->next = NULL;
*__this_cpu_read(tasklet_vec.tail) = t;
__this_cpu_write(tasklet_vec.tail, &(t->next));
- __raise_softirq_irqoff(TASKLET_SOFTIRQ);
local_irq_enable();
}
}
@@ -563,7 +562,6 @@ static __latent_entropy void tasklet_hi_action(struct
softirq_action *a)
t->next = NULL;
*__this_cpu_read(tasklet_hi_vec.tail) = t;
__this_cpu_write(tasklet_hi_vec.tail, &(t->next));
- __raise_softirq_irqoff(HI_SOFTIRQ);
local_irq_enable();
}
}
--
2.11.0
[PATCH] softirq: Prevent looping on disabled tasklets
Disabling a tasklet causes it not to run during tasklet_action, but is
put back onto the runnable tasklet list, and a new softirq raised. As
the softirq is raised from within __do_softirq() this causing
__do_softirq() to loop constantly until its timeslice expires and is
transferred to the ksoftirq thread. ksoftirq then permanently spins,
as on each action, the disabled tasklet keeps reraising the softirq.
Break this vicious cycle by moving the softirq from the action to the
final tasklet_enable().
This behaviour appears to be historic (since the first git import).
However, the looping until timeslice duration (to a max of 2ms) was
first introduced in commit c10d73671ad3 ("softirq: reduce latencies"),
with the restart limit restored in commit 34376a50fb1f ("Fix lockup
related to stop_machine being stuck in __do_softirq.")
Reported-by: Tvrtko Ursulin
Signed-off-by: Chris Wilson
Cc: Tvrtko Ursulin
Cc: Thomas Gleixner
Cc: Hannes Reinecke
Cc: Jens Axboe
Cc: Bjorn Helgaas
Cc: Alexander Potapenko
Cc: Chen Fan
Cc: Ingo Molnar
Cc: "Peter Zijlstra (Intel)"
Cc: Sebastian Andrzej Siewior
Cc: Johannes Thumshirn
Cc: Emese Revfy
Cc: Sagi Grimberg
Cc: Eric Dumazet
Cc: Tom Herbert
Cc: Ben Hutchings
---
include/linux/interrupt.h | 7 +--
kernel/softirq.c | 2 --
2 files changed, 5 insertions(+), 4 deletions(-)
diff --git a/include/linux/interrupt.h b/include/linux/interrupt.h
index 53144e78a369..12750f00d00d 100644
--- a/include/linux/interrupt.h
+++ b/include/linux/interrupt.h
@@ -613,8 +613,11 @@ static inline void tasklet_disable(struct tasklet_struct
*t)
static inline void tasklet_enable(struct tasklet_struct *t)
{
- smp_mb__before_atomic();
- atomic_dec(>count);
+ if (!atomic_dec_and_test(>count))
+ return;
+
+ if (test_bit(TASKLET_STATE_SCHED, >state))
+ raise_softirq(HI_SOFTIRQ | TASKLET_SOFTIRQ);
}
extern void tasklet_kill(struct tasklet_struct *t);
diff --git a/kernel/softirq.c b/kernel/softirq.c
index 744fa611cae0..5a359eb1a541 100644
--- a/kernel/softirq.c
+++ b/kernel/softirq.c
@@ -527,7 +527,6 @@ static __latent_entropy void tasklet_action(struct
softirq_action *a)
t->next = NULL;
*__this_cpu_read(tasklet_vec.tail) = t;
__this_cpu_write(tasklet_vec.tail, &(t->next));
- __raise_softirq_irqoff(TASKLET_SOFTIRQ);
local_irq_enable();
}
}
@@ -563,7 +562,6 @@ static __latent_entropy void tasklet_hi_action(struct
softirq_action *a)
t->next = NULL;
*__this_cpu_read(tasklet_hi_vec.tail) = t;
__this_cpu_write(tasklet_hi_vec.tail, &(t->next));
- __raise_softirq_irqoff(HI_SOFTIRQ);
local_irq_enable();
}
}
--
2.11.0
