From: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
To: Jan Beulich <JBeulich@suse.com>
Cc: axboe@kernel.dk, xen-devel@lists.xensource.com,
Li Dongyang <lidongyang@suse.com>,
linux-kernel@vger.kernel.org
Subject: Re: [PATCH] xen/blkback: Squash the discard support for 'file' and 'phy' type.
Date: Wed, 14 Mar 2012 12:03:27 -0400 [thread overview]
Message-ID: <20120314160327.GC16960@phenom.dumpdata.com> (raw)
In-Reply-To: <4F60777502000078000783B6@nat28.tlf.novell.com>
On Wed, Mar 14, 2012 at 09:48:21AM +0000, Jan Beulich wrote:
> >>> On 13.03.12 at 23:52, Konrad Rzeszutek Wilk <konrad.wilk@oracle.com> wrote:
> > The only reason for the distinction was for the special case of
> > 'file' (which is assumed to be loopback device), was to reach inside
> > the loopback device, find the underlaying file, and call fallocate on it.
> > Fortunately "xen-blkback: convert hole punching to discard request on
> > loop devices" removes that use-case and we now based the discard
> > support based on blk_queue_discard(q) and extract all appropriate
> > parameters from the 'struct request_queue'.
> >
> > CC: Li Dongyang <lidongyang@novell.com>
> > CC: Jan Beulich <JBeulich@suse.com>
> > Signed-off-by: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
>
> Acked-by: Jan Beulich <jbeulich@suse.com>
>
> (with a few minor remarks below)
>
> > --- a/drivers/block/xen-blkback/blkback.c
> > +++ b/drivers/block/xen-blkback/blkback.c
> > @@ -419,21 +419,17 @@ static int dispatch_discard_io(struct xen_blkif *blkif,
> > int err = 0;
> > int status = BLKIF_RSP_OKAY;
> > struct block_device *bdev = blkif->vbd.bdev;
> > -
> > + unsigned long secure = 0;
>
> Mind keeping the blank line and dropping the pointless initializer (which
> future gcc is likely going to be warning about)?
<nods>
>
> > blkif->st_ds_req++;
> >
> > xen_blkif_get(blkif);
> > - if (blkif->blk_backend_type == BLKIF_BACKEND_PHY ||
> > - blkif->blk_backend_type == BLKIF_BACKEND_FILE) {
> > - unsigned long secure = (blkif->vbd.discard_secure &&
> > - (req->u.discard.flag & BLKIF_DISCARD_SECURE)) ?
> > - BLKDEV_DISCARD_SECURE : 0;
> > - err = blkdev_issue_discard(bdev,
> > - req->u.discard.sector_number,
> > - req->u.discard.nr_sectors,
> > - GFP_KERNEL, secure);
> > - } else
> > - err = -EOPNOTSUPP;
> > + secure = (blkif->vbd.discard_secure &&
> > + (req->u.discard.flag & BLKIF_DISCARD_SECURE)) ?
> > + BLKDEV_DISCARD_SECURE : 0;
> > +
> > + err = blkdev_issue_discard(bdev, req->u.discard.sector_number,
> > + req->u.discard.nr_sectors,
> > + GFP_KERNEL, secure);
> >
> > if (err == -EOPNOTSUPP) {
> > pr_debug(DRV_PFX "discard op failed, not supported\n");
> > --- a/drivers/block/xen-blkback/xenbus.c
> > +++ b/drivers/block/xen-blkback/xenbus.c
> > @@ -393,52 +393,37 @@ int xen_blkbk_discard(struct xenbus_transaction xbt,
> > struct backend_info *be)
> > char *type;
> > int err;
> > int state = 0;
> > + struct block_device *bdev = be->blkif->vbd.bdev;
> > + struct request_queue *q = bdev_get_queue(bdev);
> >
> > - type = xenbus_read(XBT_NIL, dev->nodename, "type", NULL);
> > - if (!IS_ERR(type)) {
> > - if (strncmp(type, "file", 4) == 0) {
> > - state = 1;
> > - blkif->blk_backend_type = BLKIF_BACKEND_FILE;
> > + if (blk_queue_discard(q)) {
> > + err = xenbus_printf(xbt, dev->nodename,
> > + "discard-granularity", "%u",
> > + q->limits.discard_granularity);
> > + if (err) {
> > + xenbus_dev_fatal(dev, err,
> > + "writing discard-granularity");
> > + goto kfree;
>
> Unrelated to the patch here, but failure to write any sort of extension
> data shouldn't be considered fatal - the backend can well work without
> these, and it should be left to the frontend to decide whether it wants
> to live without the unavailable extensions.
<nods>. Let me whip up another patch that removes these 'fatal' cases
and is based on this one.
>
> Jan
>
> > + }
> > + err = xenbus_printf(xbt, dev->nodename,
> > + "discard-alignment", "%u",
> > + q->limits.discard_alignment);
> > + if (err) {
> > + xenbus_dev_fatal(dev, err,
> > + "writing discard-alignment");
> > + goto kfree;
> > }
> > - if (strncmp(type, "phy", 3) == 0) {
> > - struct block_device *bdev = be->blkif->vbd.bdev;
> > - struct request_queue *q = bdev_get_queue(bdev);
> > - if (blk_queue_discard(q)) {
> > - err = xenbus_printf(xbt, dev->nodename,
> > - "discard-granularity", "%u",
> > - q->limits.discard_granularity);
> > - if (err) {
> > - xenbus_dev_fatal(dev, err,
> > - "writing discard-granularity");
> > - goto kfree;
> > - }
> > - err = xenbus_printf(xbt, dev->nodename,
> > - "discard-alignment", "%u",
> > - q->limits.discard_alignment);
> > - if (err) {
> > - xenbus_dev_fatal(dev, err,
> > - "writing discard-alignment");
> > - goto kfree;
> > - }
> > - state = 1;
> > - blkif->blk_backend_type = BLKIF_BACKEND_PHY;
> > - }
> > - /* Optional. */
> > - err = xenbus_printf(xbt, dev->nodename,
> > - "discard-secure", "%d",
> > - blkif->vbd.discard_secure);
> > - if (err) {
> > - xenbus_dev_fatal(dev, err,
> > + state = 1;
> > + /* Optional. */
> > + err = xenbus_printf(xbt, dev->nodename,
> > + "discard-secure", "%d",
> > + blkif->vbd.discard_secure);
> > + if (err) {
> > + xenbus_dev_fatal(dev, err,
> > "writting discard-secure");
> > - goto kfree;
> > - }
> > + goto kfree;
> > }
> > - } else {
> > - err = PTR_ERR(type);
> > - xenbus_dev_fatal(dev, err, "reading type");
> > - goto out;
> > }
> > -
> > err = xenbus_printf(xbt, dev->nodename, "feature-discard",
> > "%d", state);
> > if (err)
>
prev parent reply other threads:[~2012-03-14 16:03 UTC|newest]
Thread overview: 6+ messages / expand[flat|nested] mbox.gz Atom feed top
2012-03-13 22:52 [PATCH] xen/blkback: Squash the discard support for 'file' and 'phy' type Konrad Rzeszutek Wilk
2012-03-14 8:33 ` [Xen-devel] " Ian Campbell
2012-03-14 20:23 ` Konrad Rzeszutek Wilk
2012-03-15 7:43 ` Ian Campbell
2012-03-14 9:48 ` Jan Beulich
2012-03-14 16:03 ` Konrad Rzeszutek Wilk [this message]
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20120314160327.GC16960@phenom.dumpdata.com \
--to=konrad.wilk@oracle.com \
--cc=JBeulich@suse.com \
--cc=axboe@kernel.dk \
--cc=lidongyang@suse.com \
--cc=linux-kernel@vger.kernel.org \
--cc=xen-devel@lists.xensource.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).