2016-12-10 18:04:11

by Joel Fernandes

[permalink] [raw]
Subject: [PATCH v2] llist: Clarify comments about when locking is needed

llist.h comments are a bit confusing about when locking is needed versus when
it isn't. Clarify these comments a bit more by being a bit more descriptive
about why locking is needed for llist_del_first.

Cc: Huang Ying <[email protected]>
Cc: Ingo Molnar <[email protected]>
Cc: Will Deacon <[email protected]>
Cc: Paul McKenney <[email protected]>
Acked-by: Mathieu Desnoyers <[email protected]>
Signed-off-by: Joel Fernandes <[email protected]>
---
v2 changes:
Minor changes to comment and commit message based on Mathieu's suggestions
(https://lkml.org/lkml/2016/12/10/39)

include/linux/llist.h | 37 +++++++++++++++++++++----------------
1 file changed, 21 insertions(+), 16 deletions(-)

diff --git a/include/linux/llist.h b/include/linux/llist.h
index fd4ca0b..31822bb 100644
--- a/include/linux/llist.h
+++ b/include/linux/llist.h
@@ -3,28 +3,33 @@
/*
* Lock-less NULL terminated single linked list
*
- * If there are multiple producers and multiple consumers, llist_add
- * can be used in producers and llist_del_all can be used in
- * consumers. They can work simultaneously without lock. But
- * llist_del_first can not be used here. Because llist_del_first
- * depends on list->first->next does not changed if list->first is not
- * changed during its operation, but llist_del_first, llist_add,
- * llist_add (or llist_del_all, llist_add, llist_add) sequence in
- * another consumer may violate that.
- *
- * If there are multiple producers and one consumer, llist_add can be
- * used in producers and llist_del_all or llist_del_first can be used
- * in the consumer.
- *
- * This can be summarized as follow:
+ * Cases where locking is not needed:
+ * If there are multiple producers and multiple consumers, llist_add can be
+ * used in producers and llist_del_all can be used in consumers simultaneously
+ * without locking. Also a single consumer can use llist_del_first while
+ * multiple producers simultaneously use llist_add, without any locking.
+ *
+ * Cases where locking is needed:
+ * If we have multiple consumers with llist_del_first used in one consumer, and
+ * llist_del_first or llist_del_all used in other consumers, then a lock is
+ * needed. This is because llist_del_first depends on list->first->next not
+ * changing, but without lock protection, there's no way to be sure about that
+ * if a preemption happens in the middle of the delete operation and on being
+ * preempted back, the list->first is the same as before causing the cmpxchg in
+ * llist_del_first to succeed. For example, while a llist_del_first operation
+ * is in progress in one consumer, then a llist_del_first, llist_add,
+ * llist_add (or llist_del_all, llist_add, llist_add) sequence in another
+ * consumer may cause violations.
+ *
+ * This can be summarized as follows:
*
* | add | del_first | del_all
* add | - | - | -
* del_first | | L | L
* del_all | | | -
*
- * Where "-" stands for no lock is needed, while "L" stands for lock
- * is needed.
+ * Where, a particular row's operation can happen concurrently with a column's
+ * operation, with "-" being no lock needed, while "L" being lock is needed.
*
* The list entries deleted via llist_del_all can be traversed with
* traversing function such as llist_for_each etc. But the list
--
2.8.0.rc3.226.g39d4020


2016-12-10 18:14:14

by Mathieu Desnoyers

[permalink] [raw]
Subject: Re: [PATCH v2] llist: Clarify comments about when locking is needed

----- On Dec 10, 2016, at 7:03 PM, Joel Fernandes [email protected] wrote:

> llist.h comments are a bit confusing about when locking is needed versus when
> it isn't. Clarify these comments a bit more by being a bit more descriptive
> about why locking is needed for llist_del_first.

As I stated in my earlier review, please remove a couple of "a bit"
from the changelog.

Thanks,

Mathieu

>
> Cc: Huang Ying <[email protected]>
> Cc: Ingo Molnar <[email protected]>
> Cc: Will Deacon <[email protected]>
> Cc: Paul McKenney <[email protected]>
> Acked-by: Mathieu Desnoyers <[email protected]>
> Signed-off-by: Joel Fernandes <[email protected]>
> ---
> v2 changes:
> Minor changes to comment and commit message based on Mathieu's suggestions
> (https://lkml.org/lkml/2016/12/10/39)
>
> include/linux/llist.h | 37 +++++++++++++++++++++----------------
> 1 file changed, 21 insertions(+), 16 deletions(-)
>
> diff --git a/include/linux/llist.h b/include/linux/llist.h
> index fd4ca0b..31822bb 100644
> --- a/include/linux/llist.h
> +++ b/include/linux/llist.h
> @@ -3,28 +3,33 @@
> /*
> * Lock-less NULL terminated single linked list
> *
> - * If there are multiple producers and multiple consumers, llist_add
> - * can be used in producers and llist_del_all can be used in
> - * consumers. They can work simultaneously without lock. But
> - * llist_del_first can not be used here. Because llist_del_first
> - * depends on list->first->next does not changed if list->first is not
> - * changed during its operation, but llist_del_first, llist_add,
> - * llist_add (or llist_del_all, llist_add, llist_add) sequence in
> - * another consumer may violate that.
> - *
> - * If there are multiple producers and one consumer, llist_add can be
> - * used in producers and llist_del_all or llist_del_first can be used
> - * in the consumer.
> - *
> - * This can be summarized as follow:
> + * Cases where locking is not needed:
> + * If there are multiple producers and multiple consumers, llist_add can be
> + * used in producers and llist_del_all can be used in consumers simultaneously
> + * without locking. Also a single consumer can use llist_del_first while
> + * multiple producers simultaneously use llist_add, without any locking.
> + *
> + * Cases where locking is needed:
> + * If we have multiple consumers with llist_del_first used in one consumer, and
> + * llist_del_first or llist_del_all used in other consumers, then a lock is
> + * needed. This is because llist_del_first depends on list->first->next not
> + * changing, but without lock protection, there's no way to be sure about that
> + * if a preemption happens in the middle of the delete operation and on being
> + * preempted back, the list->first is the same as before causing the cmpxchg in
> + * llist_del_first to succeed. For example, while a llist_del_first operation
> + * is in progress in one consumer, then a llist_del_first, llist_add,
> + * llist_add (or llist_del_all, llist_add, llist_add) sequence in another
> + * consumer may cause violations.
> + *
> + * This can be summarized as follows:
> *
> * | add | del_first | del_all
> * add | - | - | -
> * del_first | | L | L
> * del_all | | | -
> *
> - * Where "-" stands for no lock is needed, while "L" stands for lock
> - * is needed.
> + * Where, a particular row's operation can happen concurrently with a column's
> + * operation, with "-" being no lock needed, while "L" being lock is needed.
> *
> * The list entries deleted via llist_del_all can be traversed with
> * traversing function such as llist_for_each etc. But the list
> --
> 2.8.0.rc3.226.g39d4020

--
Mathieu Desnoyers
EfficiOS Inc.
http://www.efficios.com

2016-12-10 18:20:45

by Joel Fernandes

[permalink] [raw]
Subject: Re: [PATCH v2] llist: Clarify comments about when locking is needed

On Sat, Dec 10, 2016 at 10:15 AM, Mathieu Desnoyers
<[email protected]> wrote:
> ----- On Dec 10, 2016, at 7:03 PM, Joel Fernandes [email protected] wrote:
>
>> llist.h comments are a bit confusing about when locking is needed versus when
>> it isn't. Clarify these comments a bit more by being a bit more descriptive
>> about why locking is needed for llist_del_first.
>
> As I stated in my earlier review, please remove a couple of "a bit"
> from the changelog.
>

I'm sorry I missed that. I will update it correctly in the next rev.

Regards,
Joel



> Thanks,
>
> Mathieu
>
>>
>> Cc: Huang Ying <[email protected]>
>> Cc: Ingo Molnar <[email protected]>
>> Cc: Will Deacon <[email protected]>
>> Cc: Paul McKenney <[email protected]>
>> Acked-by: Mathieu Desnoyers <[email protected]>
>> Signed-off-by: Joel Fernandes <[email protected]>
>> ---
>> v2 changes:
>> Minor changes to comment and commit message based on Mathieu's suggestions
>> (https://lkml.org/lkml/2016/12/10/39)
>>
>> include/linux/llist.h | 37 +++++++++++++++++++++----------------
>> 1 file changed, 21 insertions(+), 16 deletions(-)
>>
>> diff --git a/include/linux/llist.h b/include/linux/llist.h
>> index fd4ca0b..31822bb 100644
>> --- a/include/linux/llist.h
>> +++ b/include/linux/llist.h
>> @@ -3,28 +3,33 @@
>> /*
>> * Lock-less NULL terminated single linked list
>> *
>> - * If there are multiple producers and multiple consumers, llist_add
>> - * can be used in producers and llist_del_all can be used in
>> - * consumers. They can work simultaneously without lock. But
>> - * llist_del_first can not be used here. Because llist_del_first
>> - * depends on list->first->next does not changed if list->first is not
>> - * changed during its operation, but llist_del_first, llist_add,
>> - * llist_add (or llist_del_all, llist_add, llist_add) sequence in
>> - * another consumer may violate that.
>> - *
>> - * If there are multiple producers and one consumer, llist_add can be
>> - * used in producers and llist_del_all or llist_del_first can be used
>> - * in the consumer.
>> - *
>> - * This can be summarized as follow:
>> + * Cases where locking is not needed:
>> + * If there are multiple producers and multiple consumers, llist_add can be
>> + * used in producers and llist_del_all can be used in consumers simultaneously
>> + * without locking. Also a single consumer can use llist_del_first while
>> + * multiple producers simultaneously use llist_add, without any locking.
>> + *
>> + * Cases where locking is needed:
>> + * If we have multiple consumers with llist_del_first used in one consumer, and
>> + * llist_del_first or llist_del_all used in other consumers, then a lock is
>> + * needed. This is because llist_del_first depends on list->first->next not
>> + * changing, but without lock protection, there's no way to be sure about that
>> + * if a preemption happens in the middle of the delete operation and on being
>> + * preempted back, the list->first is the same as before causing the cmpxchg in
>> + * llist_del_first to succeed. For example, while a llist_del_first operation
>> + * is in progress in one consumer, then a llist_del_first, llist_add,
>> + * llist_add (or llist_del_all, llist_add, llist_add) sequence in another
>> + * consumer may cause violations.
>> + *
>> + * This can be summarized as follows:
>> *
>> * | add | del_first | del_all
>> * add | - | - | -
>> * del_first | | L | L
>> * del_all | | | -
>> *
>> - * Where "-" stands for no lock is needed, while "L" stands for lock
>> - * is needed.
>> + * Where, a particular row's operation can happen concurrently with a column's
>> + * operation, with "-" being no lock needed, while "L" being lock is needed.
>> *
>> * The list entries deleted via llist_del_all can be traversed with
>> * traversing function such as llist_for_each etc. But the list
>> --
>> 2.8.0.rc3.226.g39d4020
>
> --
> Mathieu Desnoyers
> EfficiOS Inc.
> http://www.efficios.com

2016-12-12 01:15:36

by Huang, Ying

[permalink] [raw]
Subject: Re: [PATCH v2] llist: Clarify comments about when locking is needed

Joel Fernandes <[email protected]> writes:

> llist.h comments are a bit confusing about when locking is needed versus when
> it isn't. Clarify these comments a bit more by being a bit more descriptive
> about why locking is needed for llist_del_first.
>
> Cc: Huang Ying <[email protected]>
> Cc: Ingo Molnar <[email protected]>
> Cc: Will Deacon <[email protected]>
> Cc: Paul McKenney <[email protected]>
> Acked-by: Mathieu Desnoyers <[email protected]>
> Signed-off-by: Joel Fernandes <[email protected]>

Acked-by: "Huang, Ying" <[email protected]>

Best Regards,
Huang, Ying

> ---
> v2 changes:
> Minor changes to comment and commit message based on Mathieu's suggestions
> (https://lkml.org/lkml/2016/12/10/39)
>
> include/linux/llist.h | 37 +++++++++++++++++++++----------------
> 1 file changed, 21 insertions(+), 16 deletions(-)
>
> diff --git a/include/linux/llist.h b/include/linux/llist.h
> index fd4ca0b..31822bb 100644
> --- a/include/linux/llist.h
> +++ b/include/linux/llist.h
> @@ -3,28 +3,33 @@
> /*
> * Lock-less NULL terminated single linked list
> *
> - * If there are multiple producers and multiple consumers, llist_add
> - * can be used in producers and llist_del_all can be used in
> - * consumers. They can work simultaneously without lock. But
> - * llist_del_first can not be used here. Because llist_del_first
> - * depends on list->first->next does not changed if list->first is not
> - * changed during its operation, but llist_del_first, llist_add,
> - * llist_add (or llist_del_all, llist_add, llist_add) sequence in
> - * another consumer may violate that.
> - *
> - * If there are multiple producers and one consumer, llist_add can be
> - * used in producers and llist_del_all or llist_del_first can be used
> - * in the consumer.
> - *
> - * This can be summarized as follow:
> + * Cases where locking is not needed:
> + * If there are multiple producers and multiple consumers, llist_add can be
> + * used in producers and llist_del_all can be used in consumers simultaneously
> + * without locking. Also a single consumer can use llist_del_first while
> + * multiple producers simultaneously use llist_add, without any locking.
> + *
> + * Cases where locking is needed:
> + * If we have multiple consumers with llist_del_first used in one consumer, and
> + * llist_del_first or llist_del_all used in other consumers, then a lock is
> + * needed. This is because llist_del_first depends on list->first->next not
> + * changing, but without lock protection, there's no way to be sure about that
> + * if a preemption happens in the middle of the delete operation and on being
> + * preempted back, the list->first is the same as before causing the cmpxchg in
> + * llist_del_first to succeed. For example, while a llist_del_first operation
> + * is in progress in one consumer, then a llist_del_first, llist_add,
> + * llist_add (or llist_del_all, llist_add, llist_add) sequence in another
> + * consumer may cause violations.
> + *
> + * This can be summarized as follows:
> *
> * | add | del_first | del_all
> * add | - | - | -
> * del_first | | L | L
> * del_all | | | -
> *
> - * Where "-" stands for no lock is needed, while "L" stands for lock
> - * is needed.
> + * Where, a particular row's operation can happen concurrently with a column's
> + * operation, with "-" being no lock needed, while "L" being lock is needed.
> *
> * The list entries deleted via llist_del_all can be traversed with
> * traversing function such as llist_for_each etc. But the list