Hi Bryant,
On Thu, 2017-02-23 at 11:49 -0600, Bryant G. Ly wrote:
> > On Wed, 2017-02-08 at 14:25 -0800, Bart Van Assche wrote:
> >> Signed-off-by: Bart Van Assche <bart.vanassche@xxxxxxxxxxx>
> >> Cc: Christoph Hellwig <hch@xxxxxx>
> >> ---
> >> fs/configfs/item.c | 6 ++++++
> >> include/linux/configfs.h | 1 +
> >> 2 files changed, 7 insertions(+)
> >>
> >> diff --git a/fs/configfs/item.c b/fs/configfs/item.c
> >> index 8b2a994042dd..e3501b9bbb60 100644
> >> --- a/fs/configfs/item.c
> >> +++ b/fs/configfs/item.c
> >> @@ -138,6 +138,12 @@ struct config_item *config_item_get(struct config_item *item)
> >> }
> >> EXPORT_SYMBOL(config_item_get);
> >>
> >> +struct config_item *config_item_get_unless_zero(struct config_item *item)
> >> +{
> >> + return item && kref_get_unless_zero(&item->ci_kref) ? item : NULL;
> >> +}
> >> +EXPORT_SYMBOL(config_item_get_unless_zero);
> >> +
> >> static void config_item_cleanup(struct config_item *item)
> >> {
> >> struct config_item_type *t = item->ci_type;
> >> diff --git a/include/linux/configfs.h b/include/linux/configfs.h
> >> index 2319b8c108e8..406e16dabc28 100644
> >> --- a/include/linux/configfs.h
> >> +++ b/include/linux/configfs.h
> >> @@ -75,6 +75,7 @@ extern void config_item_init_type_name(struct config_item *item,
> >> struct config_item_type *type);
> >>
> >> extern struct config_item * config_item_get(struct config_item *);
> >> +extern struct config_item * config_item_get_unless_zero(struct config_item *);
> >> extern void config_item_put(struct config_item *);
> >>
> >> struct config_item_type {
> > A reasonable idea, but this needs review from other configfs folks.
> >
> > Dropping for now.
> >
> > --
> > To unsubscribe from this list: send the line "unsubscribe target-devel" in
> > the body of a message to majordomo@xxxxxxxxxxxxxxx
> > More majordomo info at http://vger.kernel.org/majordomo-info.html
> >
> I actually have been testing this patch since we see some hanging in
> regards to configfs, and so far haven't hit any problems.
>
> We've been running this since this was posted and so far it looks
> good. So if configfs folks are okay with it, I'd like this patch and
> target: Introduce target_get_device() and target_put_device()
> accepted.
>
Btw, this particular patch is a prerequisite for avoiding calling
target_depend_item() -> configfs_depend_item() while holding
g_device_mutex in target_core_xcopy.c:target_xcopy_locate_se_dev_e4().
Namely, it only effects EXTENDED_COPY operation when looking up a remote
se_device by EVPD 0x83 device identifiers, and is not related to other
configfs stuff.
If you've got a particular configfs issue, please feel free to post it
to the list and I'll comment in a separate thread.
--
To unsubscribe from this list: send the line "unsubscribe target-devel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
