Message ID | 1306940008-28454-1-git-send-email-bengardiner@nanometrics.ca |
---|---|
State | New, archived |
Headers | show |
On Wed, 2011-06-01 at 10:53 -0400, Ben Gardiner wrote: > When the patch 'UBIFS: intialize LPT earlier' was massaged before being applied > the context of the next patch 'UBIFS: fix-up free space earlier' was muddled > sufficiently for the order of allocate-lpt then free-space-fixup was reversed -- > I think. > > Put the free-space-fixup after allocate-lpt. > > Signed-off-by: Ben Gardiner <bengardiner@nanometrics.ca> I'll take a look at this and fold this change in a bit later, thanks.
On Wed, 2011-06-01 at 10:53 -0400, Ben Gardiner wrote: > When the patch 'UBIFS: intialize LPT earlier' was massaged before being applied > the context of the next patch 'UBIFS: fix-up free space earlier' was muddled > sufficiently for the order of allocate-lpt then free-space-fixup was reversed -- > I think. > > Put the free-space-fixup after allocate-lpt. > > Signed-off-by: Ben Gardiner <bengardiner@nanometrics.ca> > > --- > With this fixup applied on-top-of the ubifs-2.6 master I am able to mount a > ubinized image containing a UBIFS volume which has the free-space-fixup flag > set. No assertions are displayed. Thanks. Merged this patch to the 3rd one and pushed out.
Hi Artem, On Wed, Jun 1, 2011 at 11:14 AM, Artem Bityutskiy <dedekind1@gmail.com> wrote: > On Wed, 2011-06-01 at 10:53 -0400, Ben Gardiner wrote: >> When the patch 'UBIFS: intialize LPT earlier' was massaged before being applied >> the context of the next patch 'UBIFS: fix-up free space earlier' was muddled >> sufficiently for the order of allocate-lpt then free-space-fixup was reversed -- >> I think. >> >> Put the free-space-fixup after allocate-lpt. >> >> Signed-off-by: Ben Gardiner <bengardiner@nanometrics.ca> >> >> --- >> With this fixup applied on-top-of the ubifs-2.6 master I am able to mount a >> ubinized image containing a UBIFS volume which has the free-space-fixup flag >> set. No assertions are displayed. > > Thanks. Merged this patch to the 3rd one and pushed out. You're most welcome. Both branches master and linux-next are working as expected. Best Regards, Ben Gardiner --- Nanometrics Inc. http://www.nanometrics.ca
diff --git a/fs/ubifs/super.c b/fs/ubifs/super.c index 4722f31..70e8c92 100644 --- a/fs/ubifs/super.c +++ b/fs/ubifs/super.c @@ -1284,12 +1284,6 @@ static int mount_ubifs(struct ubifs_info *c) init_constants_master(c); - if (!c->ro_mount && c->space_fixup) { - err = ubifs_fixup_free_space(c); - if (err) - goto out_infos; - } - if ((c->mst_node->flags & cpu_to_le32(UBIFS_MST_DIRTY)) != 0) { ubifs_msg("recovery needed"); c->need_recovery = 1; @@ -1305,6 +1299,12 @@ static int mount_ubifs(struct ubifs_info *c) if (err) goto out_master; + if (!c->ro_mount && c->space_fixup) { + err = ubifs_fixup_free_space(c); + if (err) + goto out_master; + } + if (!c->ro_mount) { /* * Set the "dirty" flag so that if we reboot uncleanly we
When the patch 'UBIFS: intialize LPT earlier' was massaged before being applied the context of the next patch 'UBIFS: fix-up free space earlier' was muddled sufficiently for the order of allocate-lpt then free-space-fixup was reversed -- I think. Put the free-space-fixup after allocate-lpt. Signed-off-by: Ben Gardiner <bengardiner@nanometrics.ca> --- With this fixup applied on-top-of the ubifs-2.6 master I am able to mount a ubinized image containing a UBIFS volume which has the free-space-fixup flag set. No assertions are displayed. --- fs/ubifs/super.c | 12 ++++++------ 1 files changed, 6 insertions(+), 6 deletions(-)