2013-10-11 07:24:05

by Shaohua Li

[permalink] [raw]
Subject: [patch 3/4] percpu_ida: add an API to return free tags

add an API to return free tags, blk-mq-tag will use it

Signed-off-by: Shaohua Li <[email protected]>
---
include/linux/percpu_ida.h | 1 +
lib/percpu_ida.c | 17 +++++++++++++++++
2 files changed, 18 insertions(+)

Index: master/include/linux/percpu_ida.h
===================================================================
--- master.orig/include/linux/percpu_ida.h 2013-10-11 12:15:06.996416710 +0800
+++ master/include/linux/percpu_ida.h 2013-10-11 12:15:06.992416749 +0800
@@ -76,4 +76,5 @@ static inline int percpu_ida_init(struct
int percpu_ida_for_each_free(struct percpu_ida *pool,
int (*fn)(int id, void *data), void *data);

+unsigned percpu_ida_free_tags(struct percpu_ida *pool, int cpu);
#endif /* __PERCPU_IDA_H__ */
Index: master/lib/percpu_ida.c
===================================================================
--- master.orig/lib/percpu_ida.c 2013-10-11 12:15:06.996416710 +0800
+++ master/lib/percpu_ida.c 2013-10-11 12:15:06.992416749 +0800
@@ -371,3 +371,20 @@ out:
return err;
}
EXPORT_SYMBOL_GPL(percpu_ida_for_each_free);
+
+/**
+ * percpu_ida_free_tags - return free tags number of a specific cpu or global pool
+ * @pool: pool related
+ * @cpu: specific cpu or global pool if @cpu == nr_cpu_ids
+ *
+ * Note: this just returns a snapshot of free tags number.
+ */
+unsigned percpu_ida_free_tags(struct percpu_ida *pool, int cpu)
+{
+ struct percpu_ida_cpu *remote;
+ if (cpu == nr_cpu_ids)
+ return pool->nr_free;
+ remote = per_cpu_ptr(pool->tag_cpu, cpu);
+ return remote->nr_free;
+}
+EXPORT_SYMBOL_GPL(percpu_ida_free_tags);


2013-10-11 20:35:49

by Kent Overstreet

[permalink] [raw]
Subject: Re: [patch 3/4] percpu_ida: add an API to return free tags

On Fri, Oct 11, 2013 at 03:18:05PM +0800, Shaohua Li wrote:
> add an API to return free tags, blk-mq-tag will use it

Can you explain how this is going to be used? Seems like something that
could be prone to misuse, try and convince me there isn't a better way
to do what it's for.

>
> Signed-off-by: Shaohua Li <[email protected]>
> ---
> include/linux/percpu_ida.h | 1 +
> lib/percpu_ida.c | 17 +++++++++++++++++
> 2 files changed, 18 insertions(+)
>
> Index: master/include/linux/percpu_ida.h
> ===================================================================
> --- master.orig/include/linux/percpu_ida.h 2013-10-11 12:15:06.996416710 +0800
> +++ master/include/linux/percpu_ida.h 2013-10-11 12:15:06.992416749 +0800
> @@ -76,4 +76,5 @@ static inline int percpu_ida_init(struct
> int percpu_ida_for_each_free(struct percpu_ida *pool,
> int (*fn)(int id, void *data), void *data);
>
> +unsigned percpu_ida_free_tags(struct percpu_ida *pool, int cpu);
> #endif /* __PERCPU_IDA_H__ */
> Index: master/lib/percpu_ida.c
> ===================================================================
> --- master.orig/lib/percpu_ida.c 2013-10-11 12:15:06.996416710 +0800
> +++ master/lib/percpu_ida.c 2013-10-11 12:15:06.992416749 +0800
> @@ -371,3 +371,20 @@ out:
> return err;
> }
> EXPORT_SYMBOL_GPL(percpu_ida_for_each_free);
> +
> +/**
> + * percpu_ida_free_tags - return free tags number of a specific cpu or global pool
> + * @pool: pool related
> + * @cpu: specific cpu or global pool if @cpu == nr_cpu_ids
> + *
> + * Note: this just returns a snapshot of free tags number.
> + */
> +unsigned percpu_ida_free_tags(struct percpu_ida *pool, int cpu)
> +{
> + struct percpu_ida_cpu *remote;
> + if (cpu == nr_cpu_ids)
> + return pool->nr_free;
> + remote = per_cpu_ptr(pool->tag_cpu, cpu);
> + return remote->nr_free;
> +}
> +EXPORT_SYMBOL_GPL(percpu_ida_free_tags);
>

2013-10-12 01:02:52

by Shaohua Li

[permalink] [raw]
Subject: Re: [patch 3/4] percpu_ida: add an API to return free tags

On Fri, Oct 11, 2013 at 01:35:36PM -0700, Kent Overstreet wrote:
> On Fri, Oct 11, 2013 at 03:18:05PM +0800, Shaohua Li wrote:
> > add an API to return free tags, blk-mq-tag will use it
>
> Can you explain how this is going to be used? Seems like something that
> could be prone to misuse, try and convince me there isn't a better way
> to do what it's for.

There are two usages. One is for sysfs info output, which can be used for
diagnosis. The other one is blk-mq wants to determine if it's possible a
request can be queued. Neither requires precise data. Yes, caller should be
aware returned data isn't very precise.

Thanks,
Shaohua