linux-toradex.git/fs/jffs2/wbuf.c, branch v2.6.32.26

JFFS2: add missing verify buffer allocation/deallocation

2009-09-03T14:01:34+00:00

The function jffs2_nor_wbuf_flash_setup() doesn't allocate the verify buffer
if CONFIG_JFFS2_FS_WBUF_VERIFY is defined, so causing a kernel panic when
that macro is enabled and the verify function is called. Similarly the
jffs2_nor_wbuf_flash_cleanup() must free the buffer if
CONFIG_JFFS2_FS_WBUF_VERIFY is enabled.
The following patch fixes the problem.
The following patch applies to 2.6.30 kernel.

Signed-off-by: Massimo Cirillo 
Signed-off-by: Artem Bityutskiy 
Signed-off-by: David Woodhouse 
Cc: stable@kernel.org

[JFFS2] Write buffer offset adjustment for NOR-ECC (Sibley) flash

2008-10-18T10:54:09+00:00

After choosing new c->nextblock, don't leave the wbuf offset field
occasionally pointing at the start of the next physical eraseblock.
This was causing a BUG() on NOR-ECC (Sibley) flash, where we start
writing after the cleanmarker.

Among other this fix should cover write buffer offset adjustment
after flushing the last page of an eraseblock.

Signed-off-by: Alexander Belyakov 
Signed-off-by: David Woodhouse

[JFFS2] Track parent inode for directories (for NFS export)

2008-05-01T17:47:17+00:00

To support NFS export, we need to know the parent inode of directories.
Rather than growing the jffs2_inode_cache structure, share space with
the nlink field -- which was always set to 1 for directories anyway.

Signed-off-by: David Woodhouse

[JFFS2] Invert last argument of jffs2_gc_fetch_inode(), make it boolean.

2008-05-01T16:24:28+00:00

We don't actually care about nlink; we only care whether the inode in
question is unlinked or not.

Signed-off-by: David Woodhouse

[JFFS2] semaphore->mutex conversion

2008-04-22T14:13:40+00:00

Signed-off-by: David Woodhouse

[JFFS2] add write verify on dataflash.

2008-04-22T11:35:50+00:00

Add the write verification buffer to the dataflash.  The mtd_dataflash has
the CONFIG_DATAFLASH_WRITE_VERIFY so is better a change to Kconfig.

Signed-off-by: Michael Trimarchi 
Signed-off-by: Andrew Morton 
Signed-off-by: David Woodhouse

[JFFS2] Correct cleanmarker checks -- we should use only 8 bytes

2007-08-09T09:28:20+00:00

Commit a7a6ace1406f95c3edb8365788f85984377f3832 revamped the OOB
handling but accidentally switched to 12-byte cleanmarkers, which is
incompatible with what 'flash_eraseall -j' will do. So using
flash_eraseall -j and then trying to mount the 'empty' flash will fail,
because the cleanmarkers aren't recognised.

Signed-off-by: David Woodhouse

[JFFS2] Add support for write-buffer verification.

2007-07-11T13:23:54+00:00

We've seen some evil corruption issues, where the corruption seems to be
introduced after the JFFS2 crc32 is calculated but before the NAND
controller calculates the ECC. So it's in RAM or in the PCI DMA
transfer; not on the flash. Attempt to catch it earlier by (optionally)
reading back from the flash immediately after writing it.

Signed-off-by: David Woodhouse

[JFFS2] Whitespace cleanups.

2007-07-10T09:01:22+00:00

Convert many spaces to tabs; one or two other minor cosmetic fixes.

Signed-off-by: David Woodhouse

[JFFS2] Don't advance c->wbuf_ofs to next eraseblock after wbuf flush

2007-05-05T08:52:49+00:00

After flushing the last page of an eraseblock, don't leave the
wbuf 'offset' field pointing at the start of the next physical
eraseblock. This was causing a BUG() on NOR-ECC (Sibley) flash, where
we start writing a little further in, after the cleanmarker.

Debugged by Alexander Belyakov 

Signed-off-by: David Woodhouse