leandrof/linux
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

UBI: cleanup comments about corrupted PEBs

Browse Source 
Just make them a bit more readable and explanatory.

Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
This commit is contained in:
Artem Bityutskiy 2010-10-29 08:34:50 +03:00
parent 6c1e875ca6
commit fef2deb31f
1 changed files with 34 additions and 20 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

54
drivers/mtd/ubi/scan.c
View File

@ -39,32 +39,46 @@
* eraseblocks are put to the @free list and the physical eraseblock to be * eraseblocks are put to the @free list and the physical eraseblock to be
* erased are put to the @erase list. * erased are put to the @erase list.
* *
* About corruptions
* ~~~~~~~~~~~~~~~~~
*
* UBI protects EC and VID headers with CRC-32 checksums, so it can detect
* whether the headers are corrupted or not. Sometimes UBI also protects the
* data with CRC-32, e.g., when it executes the atomic LEB change operation, or
* when it moves the contents of a PEB for wear-leveling purposes.
*
* UBI tries to distinguish between 2 types of corruptions. * UBI tries to distinguish between 2 types of corruptions.
* 1. Corruptions caused by power cuts. These are harmless and expected *
* corruptions and UBI tries to handle them gracefully, without printing too * 1. Corruptions caused by power cuts. These are expected corruptions and UBI
* many warnings and error messages. The idea is that we do not lose * tries to handle them gracefully, without printing too many warnings and
* important data in these case - we may lose only the data which was being * error messages. The idea is that we do not lose important data in these case
* written to the media just before the power cut happened, and the upper * - we may lose only the data which was being written to the media just before
* layers (e.g., UBIFS) are supposed to handle these situations. UBI puts * the power cut happened, and the upper layers (e.g., UBIFS) are supposed to
* these PEBs to the head of the @erase list and they are scheduled for * handle such data losses (e.g., by using the FS journal).
* erasure. *
* When UBI detects a corruption (CRC-32 mismatch) in a PEB, and it looks like
* the reason is a power cut, UBI puts this PEB to the @erase list, and all
* PEBs in the @erase list are scheduled for erasure later.
* *
* 2. Unexpected corruptions which are not caused by power cuts. During * 2. Unexpected corruptions which are not caused by power cuts. During
* scanning, such PEBs are put to the @corr list and UBI preserves them. * scanning, such PEBs are put to the @corr list and UBI preserves them.
* Obviously, this lessens the amount of available PEBs, and if at some * Obviously, this lessens the amount of available PEBs, and if at some point
* point UBI runs out of free PEBs, it switches to R/O mode. UBI also loudly * UBI runs out of free PEBs, it switches to R/O mode. UBI also loudly informs
* informs about such PEBs every time the MTD device is attached. * about such PEBs every time the MTD device is attached.
* *
* However, it is difficult to reliably distinguish between these types of * However, it is difficult to reliably distinguish between these types of
* corruptions and UBI's strategy is as follows. UBI assumes (2.) if the VID * corruptions and UBI's strategy is as follows. UBI assumes corruption type 2
* header is corrupted and the data area does not contain all 0xFFs, and there * if the VID header is corrupted and the data area does not contain all 0xFFs,
* were not bit-flips or integrity errors while reading the data area. Otherwise * and there were no bit-flips or integrity errors while reading the data area.
* UBI assumes (1.). The assumptions are: * Otherwise UBI assumes corruption type 1. So the decision criteria are as
* o if the data area contains only 0xFFs, there is no data, and it is safe * follows.
* to just erase this PEB. * o If the data area contains only 0xFFs, there is no data, and it is safe
* o if the data area has bit-flips and data integrity errors (ECC errors on * to just erase this PEB - this is corruption type 1.
* o If the data area has bit-flips or data integrity errors (ECC errors on
* NAND), it is probably a PEB which was being erased when power cut * NAND), it is probably a PEB which was being erased when power cut
* happened. * happened, so this is corruption type 1. However, this is just a guess,
* which might be wrong.
* o Otherwise this it corruption type 2.
*/ */
#include <linux/err.h> #include <linux/err.h>