From c63a491d3737aec3c47c5e785d87021752ad9fa6 Mon Sep 17 00:00:00 2001 From: Artem Bityutskiy Date: Mon, 17 Dec 2007 13:21:07 +0200 Subject: UBI: add some more comments Signed-off-by: Artem Bityutskiy --- drivers/mtd/ubi/eba.c | 8 +++++++- 1 file changed, 7 insertions(+), 1 deletion(-) (limited to 'drivers/mtd/ubi') diff --git a/drivers/mtd/ubi/eba.c b/drivers/mtd/ubi/eba.c index b2b0f29bdc5..2ff34923e51 100644 --- a/drivers/mtd/ubi/eba.c +++ b/drivers/mtd/ubi/eba.c @@ -797,7 +797,7 @@ write_error: /* * ubi_eba_atomic_leb_change - change logical eraseblock atomically. * @ubi: UBI device description object - * @vol: volume escription object + * @vol: volume description object * @lnum: logical eraseblock number * @buf: data to write * @len: how many bytes to write @@ -955,6 +955,12 @@ int ubi_eba_copy_leb(struct ubi_device *ubi, int from, int to, /* * We may race with volume deletion/re-size, so we have to hold * @ubi->volumes_lock. + * + * Note, it is not a problem if we race with volume deletion or re-size + * here. If the volume is deleted or re-sized while we are moving an + * eraseblock which belongs to this volume, we'll end up with finding + * out that this LEB was unmapped at the end (see WL), and drop this + * PEB. */ spin_lock(&ubi->volumes_lock); vol = ubi->volumes[idx]; -- cgit v1.2.3-70-g09d2