Re: [PATCH v2 1/3] block: add QUEUE_FLAG_DAX for devices to advertise their DAX support

From: "Kani, Toshimitsu" <toshi.kani@hpe.com>
To: "yigal@plexistor.com" <yigal@plexistor.com>
Cc: "axboe@kernel.dk" <axboe@kernel.dk>,
	"linux-s390@vger.kernel.org" <linux-s390@vger.kernel.org>,
	"snitzer@redhat.com" <snitzer@redhat.com>,
	"linux-nvdimm@lists.01.org" <linux-nvdimm@lists.01.org>,
	"heiko.carstens@de.ibm.com" <heiko.carstens@de.ibm.com>,
	"linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>,
	"linux-raid@vger.kernel.org" <linux-raid@vger.kernel.org>,
	"dm-devel@redhat.com" <dm-devel@redhat.com>,
	"viro@zeniv.linux.org.uk" <viro@zeniv.linux.org.uk>,
	"schwidefsky@de.ibm.com" <schwidefsky@de.ibm.com>,
	"agk@redhat.com" <agk@redhat.com>
Subject: Re: [PATCH v2 1/3] block: add QUEUE_FLAG_DAX for devices to advertise their DAX support
Date: Thu, 23 Jun 2016 17:36:29 +0000	[thread overview]
Message-ID: <1466702773.3504.363.camel@hpe.com> (raw)
In-Reply-To: <CACTTzNYAxtY-d8+8Vo=Y2WpuXVaSkJk9PC7DcHXKJuU=gqvXFw@mail.gmail.com>

On Thu, 2016-06-23 at 19:31 +0300, Yigal Korman wrote:
> On Thu, Jun 23, 2016 at 2:54 AM, Toshi Kani <toshi.kani@hpe.com> wrote:
> > 
> > 
> > Currently, presence of direct_access() in block_device_operations
> > indicates support of DAX on its block device.  Because
> > block_device_operations is instantiated with 'const', this DAX
> > capablity may not be enabled conditinally.
> > 
> > In preparation for supporting DAX to device-mapper devices, add
> > QUEUE_FLAG_DAX to request_queue flags to advertise their DAX
> > support.  This will allow to set the DAX capability based on how
> > mapped device is composed.
> 
> Hi Toshi,
> This patch is very helpful!
> I think QUEUE_FLAG_DAX can also help with identifying dax devices in
> userspace.
> Perhaps you'd be willing to squash the patch below with this one or
> add it to your submission?

Hi Yigal,

Good idea.  Mike can probably take it into his tree, but I will include it
into the series if I needed to submit v3.

Acked-by: Toshi Kani <toshi.kani@hpe.com>

I have one minor comment below.

> [PATCH] block: expose QUEUE_FLAG_DAX in sysfs
> 
> There's currently no way to identify DAX enabled devices in userspace.
> 
> Signed-off-by: Yigal Korman <yigal@plexistor.com>
> ---
>  block/blk-sysfs.c | 15 +++++++++++++++
>  1 file changed, 15 insertions(+)
> 
> diff --git a/block/blk-sysfs.c b/block/blk-sysfs.c
> index 9920596..d55126d 100644
> --- a/block/blk-sysfs.c
> +++ b/block/blk-sysfs.c
> @@ -379,6 +379,15 @@ static ssize_t queue_wc_store(struct
> request_queue *q, const char *page,
>         return count;
>  }
> 
> +static ssize_t queue_dax_show(struct request_queue *q, char *page)
> +{
> +       int bit;
> +
> +       bit = test_bit(QUEUE_FLAG_DAX, &q->queue_flags);
> +
> +       return queue_var_show(bit, page);

This can be:
	return queue_var_show(blk_queue_dax(q), page);

Thanks,
-Toshi

> +}
> +
>  static struct queue_sysfs_entry queue_requests_entry = {
>         .attr = {.name = "nr_requests", .mode = S_IRUGO | S_IWUSR },
>         .show = queue_requests_show,
> @@ -516,6 +525,11 @@ static struct queue_sysfs_entry queue_wc_entry = {
>         .store = queue_wc_store,
>  };
> 
> +static struct queue_sysfs_entry queue_dax_entry = {
> +       .attr = {.name = "dax", .mode = S_IRUGO },
> +       .show = queue_dax_show,
> +};
> +
>  static struct attribute *default_attrs[] = {
>         &queue_requests_entry.attr,
>         &queue_ra_entry.attr,
> @@ -542,6 +556,7 @@ static struct attribute *default_attrs[] = {
>         &queue_random_entry.attr,
>         &queue_poll_entry.attr,
>         &queue_wc_entry.attr,
> +       &queue_dax_entry.attr,
>         NULL,
>  };
> 
_______________________________________________
Linux-nvdimm mailing list
Linux-nvdimm@lists.01.org
https://lists.01.org/mailman/listinfo/linux-nvdimm