summaryrefslogtreecommitdiffstats
diff options
context:
space:
mode:
authorSascha Hauer <s.hauer@pengutronix.de>2017-03-24 08:56:48 +0100
committerSascha Hauer <s.hauer@pengutronix.de>2017-03-31 13:57:52 +0200
commit84895c1929e6ed38ac7b941f632892b42672d238 (patch)
tree30e4ea122d1d553af84661975524bd25dc385772
parent25f00d64ff067e6de53d40e0819c99bce3e50212 (diff)
downloaddt-utils-84895c1929e6ed38ac7b941f632892b42672d238.tar.gz
state: backend_circular: rewrite function doc
The commment talks about copies where buckets are meant and also claims we start at offset 0, which may not be true. Rewrite comment. Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>
-rw-r--r--src/barebox-state/backend_storage.c8
1 files changed, 3 insertions, 5 deletions
diff --git a/src/barebox-state/backend_storage.c b/src/barebox-state/backend_storage.c
index b7b6882..2b21481 100644
--- a/src/barebox-state/backend_storage.c
+++ b/src/barebox-state/backend_storage.c
@@ -227,11 +227,9 @@ const int desired_copies = 3;
* @param circular If false, use non-circular mode to write data that is compatible with the old on-flash format
* @return 0 on success, -errno otherwise
*
- * Starting from offset 0 this function tries to create circular buckets on
- * different offsets in the device. Different copies of the data are located in
- * different eraseblocks.
- * For MTD devices we use circular buckets to minimize the number of erases.
- * Circular buckets write new data always in the next free space.
+ * This function iterates over the eraseblocks and creates one bucket on
+ * each eraseblock until we have the number of desired buckets. Bad blocks
+ * will be skipped and the next block will be used.
*/
static int state_storage_mtd_buckets_init(struct state_backend_storage *storage,
struct mtd_info_user *meminfo, bool circular)