Message ID | 1326214623-15938-1-git-send-email-richard@nod.at |
---|---|
State | Accepted |
Commit | 1f4f43475f1295dc41c18684e59deea9b6d46045 |
Headers | show |
On Tue, 2012-01-10 at 17:57 +0100, Richard Weinberger wrote: > UBI_LAYOUT_VOLUME_TYPE and UBI_LAYOUT_VOLUME_ALIGN are currently > defined but not used. > Use them at least once. 8-) > > Signed-off-by: Richard Weinberger <richard@nod.at> Hi Richard, This will break backward compatibility - older and current kernels which do not have this patch will error out when the find a volume without a name. The right way to do this would be to change the UBI format version, but you would need a really strong reason for this.
Am 14.01.2012 16:11, schrieb Artem Bityutskiy: > On Tue, 2012-01-10 at 17:57 +0100, Richard Weinberger wrote: >> UBI_LAYOUT_VOLUME_TYPE and UBI_LAYOUT_VOLUME_ALIGN are currently >> defined but not used. >> Use them at least once. 8-) >> >> Signed-off-by: Richard Weinberger <richard@nod.at> > > Hi Richard, > > This will break backward compatibility - older and current kernels which > do not have this patch will error out when the find a volume without a > name. The right way to do this would be to change the UBI format > version, but you would need a really strong reason for this. > Sorry, this comment makes absolutely no sense. Did you mean my other patch? But still, the other patch also does *not* allow volumes without names. Confused, //richard
On Sat, 2012-01-14 at 16:34 +0100, Richard Weinberger wrote: > Am 14.01.2012 16:11, schrieb Artem Bityutskiy: > > On Tue, 2012-01-10 at 17:57 +0100, Richard Weinberger wrote: > >> UBI_LAYOUT_VOLUME_TYPE and UBI_LAYOUT_VOLUME_ALIGN are currently > >> defined but not used. > >> Use them at least once. 8-) > >> > >> Signed-off-by: Richard Weinberger <richard@nod.at> > > > > Hi Richard, > > > > This will break backward compatibility - older and current kernels which > > do not have this patch will error out when the find a volume without a > > name. The right way to do this would be to change the UBI format > > version, but you would need a really strong reason for this. > > > > Sorry, this comment makes absolutely no sense. > > Did you mean my other patch? > But still, the other patch also does *not* allow volumes without names. Yes, sorry, ignore this one - I've pushed your patch. (Embarrassed)
diff --git a/drivers/mtd/ubi/vtbl.c b/drivers/mtd/ubi/vtbl.c index 9ad18da..93cb2f1 100644 --- a/drivers/mtd/ubi/vtbl.c +++ b/drivers/mtd/ubi/vtbl.c @@ -322,7 +322,7 @@ retry: goto out_free; } - vid_hdr->vol_type = UBI_VID_DYNAMIC; + vid_hdr->vol_type = UBI_LAYOUT_VOLUME_TYPE; vid_hdr->vol_id = cpu_to_be32(UBI_LAYOUT_VOLUME_ID); vid_hdr->compat = UBI_LAYOUT_VOLUME_COMPAT; vid_hdr->data_size = vid_hdr->used_ebs = @@ -632,7 +632,7 @@ static int init_volumes(struct ubi_device *ubi, const struct ubi_scan_info *si, return -ENOMEM; vol->reserved_pebs = UBI_LAYOUT_VOLUME_EBS; - vol->alignment = 1; + vol->alignment = UBI_LAYOUT_VOLUME_ALIGN; vol->vol_type = UBI_DYNAMIC_VOLUME; vol->name_len = sizeof(UBI_LAYOUT_VOLUME_NAME) - 1; memcpy(vol->name, UBI_LAYOUT_VOLUME_NAME, vol->name_len + 1);
UBI_LAYOUT_VOLUME_TYPE and UBI_LAYOUT_VOLUME_ALIGN are currently defined but not used. Use them at least once. 8-) Signed-off-by: Richard Weinberger <richard@nod.at> --- drivers/mtd/ubi/vtbl.c | 4 ++-- 1 files changed, 2 insertions(+), 2 deletions(-)