lkml.org 
[lkml]   [2018]   [Jun]   [24]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
Date
From
SubjectRe: [PATCH 10/14] ubi: fastmap: Change a WARN_ON() to ubi_err()
On Wed, 13 Jun 2018 23:23:40 +0200
Richard Weinberger <richard@nod.at> wrote:

> This WARN_ON() was used while development of fastmap, now
> it can be a regular ubi_err().
>
> Signed-off-by: Richard Weinberger <richard@nod.at>
> ---
> drivers/mtd/ubi/fastmap.c | 14 ++++++++------
> 1 file changed, 8 insertions(+), 6 deletions(-)
>
> diff --git a/drivers/mtd/ubi/fastmap.c b/drivers/mtd/ubi/fastmap.c
> index dabeb01af24a..d5506152c9f7 100644
> --- a/drivers/mtd/ubi/fastmap.c
> +++ b/drivers/mtd/ubi/fastmap.c
> @@ -949,14 +949,16 @@ static int ubi_attach_fastmap(struct ubi_device *ubi,
> ubi_assert(list_empty(&free));
>
> /*
> - * If fastmap is leaking PEBs (must not happen), raise a
> - * fat warning and fall back to scanning mode.
> - * We do this here because in ubi_wl_init() it's too late
> - * and we cannot fall back to scanning.
> + * The orginal purpose of this check was detecting fastmap bugs were
> + * it missed blocks. Now it helps to detect mis-flashed UBIs.
> + * E.g. when a fastmap enabled UBI is copied to another target with
> + * different bad blocks.
> */
> - if (WARN_ON(count_fastmap_pebs(ai) != ubi->peb_count -
> - ai->bad_peb_count - fm->used_blocks))
> + if (count_fastmap_pebs(ai) != ubi->peb_count - ai->bad_peb_count -
> + fm->used_blocks) {

Nitpick, but can you align things like that:

if (count_fastmap_pebs(ai) != ubi->peb_count - ai->bad_peb_count -
fm->used_blocks) {

or

if (count_fastmap_pebs(ai) !=
ubi->peb_count - ai->bad_peb_count - fm->used_blocks) {

or even better, have an avail_peb_count variable:

avail_peb_count = ubi->peb_count - ai->bad_peb_count - fm->used_blocks;
if (count_fastmap_pebs(ai) != avail_peb_count)

> + ubi_err(ubi, "number of PEBs referenced by fastmap does not match MTD!");
> goto fail_bad;

Also, I'm not sure I get it. Why is that an error now that we have
preseeded fastmap? Isn't it expected to have a potential mismatch the
first time we boot, or is it already taken into account before we reach
this point?



> + }
>
> return 0;
>

\
 
 \ /
  Last update: 2018-06-24 15:05    [W:0.115 / U:38.200 seconds]
©2003-2018 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site