From mboxrd@z Thu Jan  1 00:00:00 1970
Received: from eggs.gnu.org ([208.118.235.92]:45940)
	by lists.gnu.org with esmtp (Exim 4.71)
	(envelope-from <stefanha@gmail.com>) id 1S5x7E-0002z7-RG
	for qemu-devel@nongnu.org; Fri, 09 Mar 2012 05:32:10 -0500
Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71)
	(envelope-from <stefanha@gmail.com>) id 1S5x7A-0007kl-0p
	for qemu-devel@nongnu.org; Fri, 09 Mar 2012 05:32:04 -0500
Received: from mail-lpp01m010-f45.google.com ([209.85.215.45]:40703)
	by eggs.gnu.org with esmtp (Exim 4.71)
	(envelope-from <stefanha@gmail.com>) id 1S5x79-0007kH-Ms
	for qemu-devel@nongnu.org; Fri, 09 Mar 2012 05:31:59 -0500
Received: by lahe6 with SMTP id e6so1595292lah.4
	for <qemu-devel@nongnu.org>; Fri, 09 Mar 2012 02:31:57 -0800 (PST)
MIME-Version: 1.0
In-Reply-To: <1331226917-6658-15-git-send-email-pbonzini@redhat.com>
References: <1331226917-6658-1-git-send-email-pbonzini@redhat.com>
	<1331226917-6658-15-git-send-email-pbonzini@redhat.com>
Date: Fri, 9 Mar 2012 10:31:57 +0000
Message-ID: <CAJSP0QWmc-xQUYjaz16BACJPu2ZQL2VpTHWa+tSjiTHj0Gk3Ew@mail.gmail.com>
From: Stefan Hajnoczi <stefanha@gmail.com>
Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: quoted-printable
Subject: Re: [Qemu-devel] [RFC PATCH 14/17] block: support
 FALLOC_FL_PUNCH_HOLE trimming
List-Id: <qemu-devel.nongnu.org>
List-Unsubscribe: <https://lists.nongnu.org/mailman/options/qemu-devel>,
	<mailto:qemu-devel-request@nongnu.org?subject=unsubscribe>
List-Archive: <http://lists.nongnu.org/archive/html/qemu-devel>
List-Post: <mailto:qemu-devel@nongnu.org>
List-Help: <mailto:qemu-devel-request@nongnu.org?subject=help>
List-Subscribe: <https://lists.nongnu.org/mailman/listinfo/qemu-devel>,
	<mailto:qemu-devel-request@nongnu.org?subject=subscribe>
To: Paolo Bonzini <pbonzini@redhat.com>
Cc: qemu-devel@nongnu.org

On Thu, Mar 8, 2012 at 5:15 PM, Paolo Bonzini <pbonzini@redhat.com> wrote:
> This will enable discard on more filesystems, including ext4.
>
> Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
> ---
> =A0block/raw-posix.c | =A0 29 ++++++++++++++++++++++++-----
> =A01 files changed, 24 insertions(+), 5 deletions(-)
>
> diff --git a/block/raw-posix.c b/block/raw-posix.c
> index 2d1bc13..f23d92b 100644
> --- a/block/raw-posix.c
> +++ b/block/raw-posix.c
> @@ -52,6 +52,7 @@
> =A0#include <sys/param.h>
> =A0#include <linux/cdrom.h>
> =A0#include <linux/fd.h>
> +#include <linux/falloc.h>
> =A0#endif
> =A0#if defined (__FreeBSD__) || defined(__FreeBSD_kernel__)
> =A0#include <sys/disk.h>
> @@ -131,6 +132,7 @@ typedef struct BDRVRawState {
> =A0#endif
> =A0 =A0 uint8_t *aligned_buf;
> =A0 =A0 unsigned aligned_buf_size;
> + =A0 =A0bool has_discard : 1;
> =A0#ifdef CONFIG_XFS
> =A0 =A0 bool is_xfs : 1;
> =A0#endif
> @@ -257,11 +259,9 @@ static int raw_open_common(BlockDriverState *bs, con=
st char *filename,
> =A0 =A0 }
>
> =A0#ifdef CONFIG_XFS
> - =A0 =A0if (platform_test_xfs_fd(s->fd)) {
> - =A0 =A0 =A0 =A0s->is_xfs =3D 1;
> - =A0 =A0}
> + =A0 =A0s->is_xfs =3D platform_test_xfs_fd(s->fd);
> =A0#endif
> -
> + =A0 =A0s->has_discard =3D 1;
> =A0 =A0 return 0;
>
> =A0out_free_buf:
> @@ -605,15 +605,34 @@ static int xfs_discard(BDRVRawState *s, int64_t sec=
tor_num, int nb_sectors)
> =A0static coroutine_fn int raw_co_discard(BlockDriverState *bs,
> =A0 =A0 int64_t sector_num, int nb_sectors)
> =A0{
> -#ifdef CONFIG_XFS
> =A0 =A0 BDRVRawState *s =3D bs->opaque;
> + =A0 =A0int retval;
>
> + =A0 =A0if (s->has_discard =3D=3D 0) {
> + =A0 =A0 =A0 =A0return -ENOTSUP;
> + =A0 =A0}
> +#ifdef CONFIG_XFS
> =A0 =A0 if (s->is_xfs) {
> =A0 =A0 =A0 =A0 return xfs_discard(s, sector_num, nb_sectors);
> =A0 =A0 }
> =A0#endif
>
> +#ifdef FALLOC_FL_PUNCH_HOLE
> + =A0 =A0retval =3D fallocate(s->fd, FALLOC_FL_PUNCH_HOLE|FALLOC_FL_KEEP_=
SIZE,
> + =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 sector_num << 9, (int64_t)n=
b_sectors << 9);

I'm concerned about introducing blocking syscalls in coroutine code
paths.  This needs to be done asynchronously.

Stefan