On Wed, 2012-07-18 at 22:55 +0300, Shmulik Ladkani wrote:
> However, if you DO NOT mark it bad, but instead go into RO mode, you
> should be able to later re-attach because the good_peb_count would fit
> (no shortage of PEBs).
Yeah, you are right, I'll return to the original patch.
--
Best
Hi Artem,
On Wed, 18 Jul 2012 13:28:37 +0300 Artem Bityutskiy wrote:
> The whole thing will become simpler if we first mark the PEB as bad
> unconditionally (because it _is_ bad), then grab the lock and do all the
> re-calculations.
On first glance that would sound the right thing to do.
On Wed, 2012-07-04 at 11:06 +0300, Shmulik Ladkani wrote:
> The existing mechanism of reserving PEBs for bad PEB handling has two
> flaws:
> - It is calculated as a percentage of good PEBs instead of total PEBs.
> - There's no limit on the amount of PEBs UBI reserves for future bad
> eraseblock
On Wed, 2012-07-18 at 13:28 +0300, Artem Bityutskiy wrote:
> > ubi->bad_peb_count += 1;
> > ubi->good_peb_count -= 1;
> > ubi_calculate_reserved(ubi);
> We do not need to call this function from here, right?
Err, sorry, it _is_ needed!
--
Best Regards,
Artem Bityutskiy
On Wed, 2012-07-04 at 11:06 +0300, Shmulik Ladkani wrote:
> @@ -1045,20 +1046,14 @@ static int erase_worker(struct ubi_device *ubi,
> struct ubi_work *wl_wrk,
> }
>
> spin_lock(>volumes_lock);
> - need = ubi->beb_rsvd_level - ubi->beb_rsvd_pebs + 1;
> - if (need > 0) {
> -
On Wed, 2012-07-04 at 11:06 +0300, Shmulik Ladkani wrote:
@@ -1045,20 +1046,14 @@ static int erase_worker(struct ubi_device *ubi,
struct ubi_work *wl_wrk,
}
spin_lock(ubi-volumes_lock);
- need = ubi-beb_rsvd_level - ubi-beb_rsvd_pebs + 1;
- if (need 0) {
-
On Wed, 2012-07-18 at 13:28 +0300, Artem Bityutskiy wrote:
ubi-bad_peb_count += 1;
ubi-good_peb_count -= 1;
ubi_calculate_reserved(ubi);
We do not need to call this function from here, right?
Err, sorry, it _is_ needed!
--
Best Regards,
Artem Bityutskiy
signature.asc
On Wed, 2012-07-04 at 11:06 +0300, Shmulik Ladkani wrote:
The existing mechanism of reserving PEBs for bad PEB handling has two
flaws:
- It is calculated as a percentage of good PEBs instead of total PEBs.
- There's no limit on the amount of PEBs UBI reserves for future bad
eraseblock
Hi Artem,
On Wed, 18 Jul 2012 13:28:37 +0300 Artem Bityutskiy dedeki...@gmail.com wrote:
The whole thing will become simpler if we first mark the PEB as bad
unconditionally (because it _is_ bad), then grab the lock and do all the
re-calculations.
On first glance that would sound the right
On Wed, 2012-07-18 at 22:55 +0300, Shmulik Ladkani wrote:
However, if you DO NOT mark it bad, but instead go into RO mode, you
should be able to later re-attach because the good_peb_count would fit
(no shortage of PEBs).
Yeah, you are right, I'll return to the original patch.
--
Best
On Mon, 9 Jul 2012 12:15:17 +0200 Richard Genoud
wrote:
> 2012/7/4 Shmulik Ladkani :
> > + /*
> > +* Calculate the actual number of PEBs currently needed to be
> > reserved
> > +* for future bad eraseblock handling.
> > +*/
> > + ubi->beb_rsvd_level =
2012/7/4 Shmulik Ladkani :
> diff --git a/drivers/mtd/ubi/misc.c b/drivers/mtd/ubi/misc.c
> index f6a7d7a..e9dcb83 100644
> --- a/drivers/mtd/ubi/misc.c
> +++ b/drivers/mtd/ubi/misc.c
> @@ -98,10 +98,18 @@ int ubi_check_volume(struct ubi_device *ubi, int vol_id)
> */
> void
2012/7/4 Shmulik Ladkani shmulik.ladk...@gmail.com:
diff --git a/drivers/mtd/ubi/misc.c b/drivers/mtd/ubi/misc.c
index f6a7d7a..e9dcb83 100644
--- a/drivers/mtd/ubi/misc.c
+++ b/drivers/mtd/ubi/misc.c
@@ -98,10 +98,18 @@ int ubi_check_volume(struct ubi_device *ubi, int vol_id)
*/
void
On Mon, 9 Jul 2012 12:15:17 +0200 Richard Genoud richard.gen...@gmail.com
wrote:
2012/7/4 Shmulik Ladkani shmulik.ladk...@gmail.com:
+ /*
+* Calculate the actual number of PEBs currently needed to be
reserved
+* for future bad eraseblock handling.
+*/
+
14 matches
Mail list logo
