[PATCH] md: fix sync_completed reporting for very large drives (>2TB)

=46rom: Rémi Rérolle

The values exported in the sync_completed file are unsigned long, which
overflows with very large drives, resulting in wrong values reported.

Since sync_completed uses sectors as unit, we'll start getting wrong
values with components larger than 2TB.

This patch simply replaces the use of unsigned long by unsigned long lo=
ng.

Signed-off-by: Rémi Rérolle
---
drivers/md/md.c | 4 ++--
1 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/md/md.c b/drivers/md/md.c
index 175c424..ea46b3b 100644
--- a/drivers/md/md.c
+++ b/drivers/md/md.c
@@ -3912,7 +3912,7 @@ static struct md_sysfs_entry md_sync_speed =3D __=
ATTR_RO(sync_speed);
static ssize_t
sync_completed_show(mddev_t *mddev, char *page)
{
- unsigned long max_sectors, resync;
+ unsigned long long max_sectors, resync;
=20
if (!test_bit(MD_RECOVERY_RUNNING, &mddev->recovery))
return sprintf(page, "none\n");
@@ -3923,7 +3923,7 @@ sync_completed_show(mddev_t *mddev, char *page)
max_sectors =3D mddev->dev_sectors;
=20
resync =3D mddev->curr_resync_completed;
- return sprintf(page, "%lu / %lu\n", resync, max_sectors);
+ return sprintf(page, "%llu / %llu\n", resync, max_sectors);
}
=20
static struct md_sysfs_entry md_sync_completed =3D __ATTR_RO(sync_comp=
leted);
--=20
1.7.1

--
To unsubscribe from this list: send the line "unsubscribe linux-raid" i=
n
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html

Re: [PATCH] md: fix sync_completed reporting for very large drives(>2TB)

Le 04/01/2011 16:44, Rémi Rérolle a écrit :
> From: Rémi Rérolle
>
> The values exported in the sync_completed file are unsigned long, whi=
ch
> overflows with very large drives, resulting in wrong values reported.
>
> Since sync_completed uses sectors as unit, we'll start getting wrong
> values with components larger than 2TB.
>
> This patch simply replaces the use of unsigned long by unsigned long =
long.
>

Hi Neil,

What is the status for this patch ?

Thanks in advance!

> Signed-off-by: Rémi Rérolle
> ---
> drivers/md/md.c | 4 ++--
> 1 files changed, 2 insertions(+), 2 deletions(-)
>
> diff --git a/drivers/md/md.c b/drivers/md/md.c
> index 175c424..ea46b3b 100644
> --- a/drivers/md/md.c
> +++ b/drivers/md/md.c
> @@ -3912,7 +3912,7 @@ static struct md_sysfs_entry md_sync_speed =3D =
__ATTR_RO(sync_speed);
> static ssize_t
> sync_completed_show(mddev_t *mddev, char *page)
> {
> - unsigned long max_sectors, resync;
> + unsigned long long max_sectors, resync;
>
> if (!test_bit(MD_RECOVERY_RUNNING,&mddev->recovery))
> return sprintf(page, "none\n");
> @@ -3923,7 +3923,7 @@ sync_completed_show(mddev_t *mddev, char *page)
> max_sectors =3D mddev->dev_sectors;
>
> resync =3D mddev->curr_resync_completed;
> - return sprintf(page, "%lu / %lu\n", resync, max_sectors);
> + return sprintf(page, "%llu / %llu\n", resync, max_sectors);
> }
>
> static struct md_sysfs_entry md_sync_completed =3D __ATTR_RO(sync_c=
ompleted);

--=20
Rémi
--
To unsubscribe from this list: send the line "unsubscribe linux-raid" i=
n
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html

Re: [PATCH] md: fix sync_completed reporting for very large drives(>2TB)

On Tue, 11 Jan 2011 11:37:54 +0100 R=E9mi R=E9rolle > wrote:

> Le 04/01/2011 16:44, R=E9mi R=E9rolle a =E9crit :
> > From: R=E9mi R=E9rolle
> >
> > The values exported in the sync_completed file are unsigned long, w=
hich
> > overflows with very large drives, resulting in wrong values reporte=
d.
> >
> > Since sync_completed uses sectors as unit, we'll start getting wron=
g
> > values with components larger than 2TB.
> >
> > This patch simply replaces the use of unsigned long by unsigned lon=
g long.
> >
>=20
> Hi Neil,
>=20
> What is the status for this patch ?
>=20
> Thanks in advance!

Perfect timing!!

The status is that I had completely forgotten about it, but I have just=
been
assembling some md patches for the current merge window, so I have just=
added
your patch - it should appear in -next in a day or so and go to Linus b=
y the
end of the week.

Thanks!

NeilBrown


>=20
> > Signed-off-by: R=E9mi R=E9rolle
> > ---
> > drivers/md/md.c | 4 ++--
> > 1 files changed, 2 insertions(+), 2 deletions(-)
> >
> > diff --git a/drivers/md/md.c b/drivers/md/md.c
> > index 175c424..ea46b3b 100644
> > --- a/drivers/md/md.c
> > +++ b/drivers/md/md.c
> > @@ -3912,7 +3912,7 @@ static struct md_sysfs_entry md_sync_speed =3D=
__ATTR_RO(sync_speed);
> > static ssize_t
> > sync_completed_show(mddev_t *mddev, char *page)
> > {
> > - unsigned long max_sectors, resync;
> > + unsigned long long max_sectors, resync;
> >
> > if (!test_bit(MD_RECOVERY_RUNNING,&mddev->recovery))
> > return sprintf(page, "none\n");
> > @@ -3923,7 +3923,7 @@ sync_completed_show(mddev_t *mddev, char *pag=
e)
> > max_sectors =3D mddev->dev_sectors;
> >
> > resync =3D mddev->curr_resync_completed;
> > - return sprintf(page, "%lu / %lu\n", resync, max_sectors);
> > + return sprintf(page, "%llu / %llu\n", resync, max_sectors);
> > }
> >
> > static struct md_sysfs_entry md_sync_completed =3D __ATTR_RO(sync=
_completed);
>=20

--
To unsubscribe from this list: send the line "unsubscribe linux-raid" i=
n
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html