| Message ID | 1400249160-23186-1-git-send-email-anand.jain@oracle.com (mailing list archive) |
|---|---|
| State | New, archived |
| Headers | show |
2014-05-16 22:06 GMT+08:00 Anand Jain <anand.jain@oracle.com>: > BTRFS_IOC_FS_INFO return num_devices which does not include seed disks, > BTRFS_IOC_DEV_INFO fetches seed disk when probed. So in this case hits > the btrfs-progs bug: > get_fs_info() > :: > BUG_ON(ndevs >= fi_args->num_devices); > which is very easy to hit by using btrfs filesystem show. > > This patch will make BTRFS_IOC_DEV_INFO ioctl to provide disks only > of the FSID being probed (seed disks are under different FSID). > > which means when seed is still not deleted from the sprout the btrfs > filesystem show command will show disks them under their > respective FSIDs Just a quick comment: Newly created filesystem is based on existed seeding filesystem. Though they have different fsid, newly created filesystem devices should include seed devices, from seed device filesytem, they can not see newly created filesystem devices! No? Will the follow patch change the above behavior? Thanks, Wang > > Signed-off-by: Anand Jain <anand.jain@oracle.com> > --- > fs/btrfs/ioctl.c | 3 ++- > 1 files changed, 2 insertions(+), 1 deletions(-) > > diff --git a/fs/btrfs/ioctl.c b/fs/btrfs/ioctl.c > index ff27c08..902d279 100644 > --- a/fs/btrfs/ioctl.c > +++ b/fs/btrfs/ioctl.c > @@ -2584,7 +2584,8 @@ static long btrfs_ioctl_dev_info(struct btrfs_root *root, void __user *arg) > s_uuid = di_args->uuid; > > mutex_lock(&fs_devices->device_list_mutex); > - dev = btrfs_find_device(root->fs_info, di_args->devid, s_uuid, NULL); > + dev = btrfs_find_device(root->fs_info, di_args->devid, s_uuid, > + fs_devices->fsid); > > if (!dev) { > ret = -ENODEV; > -- > 1.7.1 > > -- > To unsubscribe from this list: send the line "unsubscribe linux-btrfs" in > the body of a message to majordomo@vger.kernel.org > More majordomo info at http://vger.kernel.org/majordomo-info.html -- To unsubscribe from this list: send the line "unsubscribe linux-btrfs" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
On 16/05/14 22:40, Shilong Wang wrote: > 2014-05-16 22:06 GMT+08:00 Anand Jain <anand.jain@oracle.com>: >> BTRFS_IOC_FS_INFO return num_devices which does not include seed disks, >> BTRFS_IOC_DEV_INFO fetches seed disk when probed. So in this case hits >> the btrfs-progs bug: >> get_fs_info() >> :: >> BUG_ON(ndevs >= fi_args->num_devices); >> which is very easy to hit by using btrfs filesystem show. >> >> This patch will make BTRFS_IOC_DEV_INFO ioctl to provide disks only >> of the FSID being probed (seed disks are under different FSID). >> >> which means when seed is still not deleted from the sprout the btrfs >> filesystem show command will show disks them under their >> respective FSIDs > > Just a quick comment: > > Newly created filesystem is based on existed seeding filesystem. > Though they have different fsid, newly created filesystem devices > should include seed devices, from seed device filesytem, they can not > see newly created filesystem devices! No? > > Will the follow patch change the above behavior? Yes, as the dev info list does not contain the seed disk so is the btrfs fi show output. Do you think its better to show seed and sprout "disks together" ? other way is indicate the seed fsid against the added disk until seed is deleted. Thanks. Anand > Thanks, > Wang >> >> Signed-off-by: Anand Jain <anand.jain@oracle.com> >> --- >> fs/btrfs/ioctl.c | 3 ++- >> 1 files changed, 2 insertions(+), 1 deletions(-) >> >> diff --git a/fs/btrfs/ioctl.c b/fs/btrfs/ioctl.c >> index ff27c08..902d279 100644 >> --- a/fs/btrfs/ioctl.c >> +++ b/fs/btrfs/ioctl.c >> @@ -2584,7 +2584,8 @@ static long btrfs_ioctl_dev_info(struct btrfs_root *root, void __user *arg) >> s_uuid = di_args->uuid; >> >> mutex_lock(&fs_devices->device_list_mutex); >> - dev = btrfs_find_device(root->fs_info, di_args->devid, s_uuid, NULL); >> + dev = btrfs_find_device(root->fs_info, di_args->devid, s_uuid, >> + fs_devices->fsid); >> >> if (!dev) { >> ret = -ENODEV; >> -- >> 1.7.1 >> >> -- >> To unsubscribe from this list: send the line "unsubscribe linux-btrfs" in >> the body of a message to majordomo@vger.kernel.org >> More majordomo info at http://vger.kernel.org/majordomo-info.html > -- > To unsubscribe from this list: send the line "unsubscribe linux-btrfs" in > the body of a message to majordomo@vger.kernel.org > More majordomo info at http://vger.kernel.org/majordomo-info.html > -- To unsubscribe from this list: send the line "unsubscribe linux-btrfs" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
2014-05-16 23:13 GMT+08:00 Anand Jain <Anand.Jain@oracle.com>: > > > On 16/05/14 22:40, Shilong Wang wrote: >> >> 2014-05-16 22:06 GMT+08:00 Anand Jain <anand.jain@oracle.com>: >>> >>> BTRFS_IOC_FS_INFO return num_devices which does not include seed disks, >>> BTRFS_IOC_DEV_INFO fetches seed disk when probed. So in this case hits >>> the btrfs-progs bug: >>> get_fs_info() >>> :: >>> BUG_ON(ndevs >= fi_args->num_devices); >>> which is very easy to hit by using btrfs filesystem show. >>> >>> This patch will make BTRFS_IOC_DEV_INFO ioctl to provide disks only >>> of the FSID being probed (seed disks are under different FSID). >>> >>> which means when seed is still not deleted from the sprout the btrfs >>> filesystem show command will show disks them under their >>> respective FSIDs >> >> >> Just a quick comment: >> >> Newly created filesystem is based on existed seeding filesystem. >> Though they have different fsid, newly created filesystem devices >> should include seed devices, from seed device filesytem, they can not >> see newly created filesystem devices! No? >> >> Will the follow patch change the above behavior? > > > Yes, as the dev info list does not contain the seed disk > so is the btrfs fi show output. Do you think its better > to show seed and sprout "disks together" ? other way is > indicate the seed fsid against the added disk until > seed is deleted. IMO, i think dev info list should contain seed disks, since new filesystem can access their data. Maybe a better way 'file show' will tell users which devices are seed devices.. > > Thanks. Anand > > >> Thanks, >> Wang >>> >>> >>> Signed-off-by: Anand Jain <anand.jain@oracle.com> >>> --- >>> fs/btrfs/ioctl.c | 3 ++- >>> 1 files changed, 2 insertions(+), 1 deletions(-) >>> >>> diff --git a/fs/btrfs/ioctl.c b/fs/btrfs/ioctl.c >>> index ff27c08..902d279 100644 >>> --- a/fs/btrfs/ioctl.c >>> +++ b/fs/btrfs/ioctl.c >>> @@ -2584,7 +2584,8 @@ static long btrfs_ioctl_dev_info(struct btrfs_root >>> *root, void __user *arg) >>> s_uuid = di_args->uuid; >>> >>> mutex_lock(&fs_devices->device_list_mutex); >>> - dev = btrfs_find_device(root->fs_info, di_args->devid, s_uuid, >>> NULL); >>> + dev = btrfs_find_device(root->fs_info, di_args->devid, s_uuid, >>> + fs_devices->fsid); >>> >>> if (!dev) { >>> ret = -ENODEV; >>> -- >>> 1.7.1 >>> >>> -- >>> To unsubscribe from this list: send the line "unsubscribe linux-btrfs" in >>> the body of a message to majordomo@vger.kernel.org >>> More majordomo info at http://vger.kernel.org/majordomo-info.html >> >> -- >> To unsubscribe from this list: send the line "unsubscribe linux-btrfs" in >> the body of a message to majordomo@vger.kernel.org >> More majordomo info at http://vger.kernel.org/majordomo-info.html >> > -- To unsubscribe from this list: send the line "unsubscribe linux-btrfs" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
diff --git a/fs/btrfs/ioctl.c b/fs/btrfs/ioctl.c index ff27c08..902d279 100644 --- a/fs/btrfs/ioctl.c +++ b/fs/btrfs/ioctl.c @@ -2584,7 +2584,8 @@ static long btrfs_ioctl_dev_info(struct btrfs_root *root, void __user *arg) s_uuid = di_args->uuid; mutex_lock(&fs_devices->device_list_mutex); - dev = btrfs_find_device(root->fs_info, di_args->devid, s_uuid, NULL); + dev = btrfs_find_device(root->fs_info, di_args->devid, s_uuid, + fs_devices->fsid); if (!dev) { ret = -ENODEV;
BTRFS_IOC_FS_INFO return num_devices which does not include seed disks, BTRFS_IOC_DEV_INFO fetches seed disk when probed. So in this case hits the btrfs-progs bug: get_fs_info() :: BUG_ON(ndevs >= fi_args->num_devices); which is very easy to hit by using btrfs filesystem show. This patch will make BTRFS_IOC_DEV_INFO ioctl to provide disks only of the FSID being probed (seed disks are under different FSID). which means when seed is still not deleted from the sprout the btrfs filesystem show command will show disks them under their respective FSIDs Signed-off-by: Anand Jain <anand.jain@oracle.com> --- fs/btrfs/ioctl.c | 3 ++- 1 files changed, 2 insertions(+), 1 deletions(-)