linux-stable.git/drivers/mtd/ubi, branch linux-3.13.y

Merge tag 'for-linus-20131112' of git://git.infradead.org/linux-mtd

2013-11-14T03:31:43+00:00

Pull MTD changes from Brian Norris:
 - Unify some compile-time differences so that we have fewer uses of
   #ifdef CONFIG_OF in atmel_nand
 - Other general cleanups (removing unused functions, options,
   variables, fields; use correct interfaces)
 - Fix BUG() for new odd-sized NAND, which report non-power-of-2
   dimensions via ONFI
 - Miscellaneous driver fixes (SPI NOR flash; BCM47xx NAND flash; etc.)
 - Improve differentiation between SLC and MLC NAND -- this clarifies an
   ABI issue regarding the MTD "type" (in sysfs and in the MEMGETINFO
   ioctl), where the MTD_MLCNANDFLASH type was present but
   inconsistently used
 - Extend GPMI NAND to support multi-chip-select NAND for some platforms
 - Many improvements to the OMAP2/3 NAND driver, including an expanded
   DT binding to bring us closer to mainline support for some OMAP
   systems
 - Fix a deadlock in the error path of the Atmel NAND driver probe
 - Correct the error codes from MTD mmap() to conform to POSIX and the
   Linux Programmer's Manual.  This is an acknowledged change in the MTD
   ABI, but I can't imagine somebody relying on the non-standard -ENOSYS
   error code specifically.  Am I just being unimaginative? :)
 - Fix a few important GPMI NAND bugs (one regression from 3.12 and one
   long-standing race condition)
 - More? Read the log!

* tag 'for-linus-20131112' of git://git.infradead.org/linux-mtd: (98 commits)
  mtd: gpmi: fix the NULL pointer
  mtd: gpmi: fix kernel BUG due to racing DMA operations
  mtd: mtdchar: return expected errors on mmap() call
  mtd: gpmi: only scan two chips for imx6
  mtd: gpmi: Use devm_kzalloc()
  mtd: atmel_nand: fix bug driver will in a dead lock if no nand detected
  mtd: nand: use a local variable to simplify the nand_scan_tail
  mtd: nand: remove deprecated IRQF_DISABLED
  mtd: dataflash: Say if we find a device we don't support
  mtd: nand: omap: fix error return code in omap_nand_probe()
  mtd: nand_bbt: kill NAND_BBT_SCANALLPAGES
  mtd: m25p80: fixup device removal failure path
  mtd: mxc_nand: Include linux/of.h header
  mtd: remove duplicated include from mtdcore.c
  mtd: m25p80: add support for Macronix mx25l3255e
  mtd: nand: omap: remove selection of BCH ecc-scheme via KConfig
  mtd: nand: omap: updated devm_xx for all resource allocation and free calls
  mtd: nand: omap: use drivers/mtd/nand/nand_bch.c wrapper for BCH ECC instead of lib/bch.c
  mtd: nand: omap: clean-up ecc layout for BCH ecc schemes
  mtd: nand: omap2: clean-up BCHx_HW and BCHx_SW ECC configurations in device_probe
  ...

mtd: Move major number definitions to major.h

2013-11-07T07:32:59+00:00

This patch moves the char and block major number definitions
to major.h to be with the rest of the major numbers.
While doing this, include major.h in the files that need it.

Signed-off-by: Ezequiel Garcia 
Signed-off-by: Brian Norris

UBI: Add some asserts to ubi_attach_fastmap()

2013-10-03T16:42:51+00:00

Add more paranioa asserts to make it easier to detect
implementation errors.

Signed-off-by: Richard Weinberger 
Signed-off-by: Artem Bityutskiy

UBI: Fix memory leak in ubi_attach_fastmap() error path

2013-10-03T16:42:50+00:00

On error we have to free all three temporary lists.

Reported-by: Richard Genoud 
Signed-off-by: Richard Weinberger 
Signed-off-by: Artem Bityutskiy

UBI: simplify image sequence test

2013-10-03T16:42:50+00:00

The test:
if (!a && b)
  a = b;
can be symplified in:
if (!a)
  a = b;

And there's no need to test if ubi->image_seq is not null, because if it is,
it is set to image_seq.
So, we just test if image_seq is not null.

Signed-off-by: Richard Genoud 
Signed-off-by: Richard Weinberger 
Signed-off-by: Artem Bityutskiy

UBI: fastmap: fix backward compatibility with image_seq

2013-10-03T16:42:50+00:00

Some old UBI implementations (e.g. U-Boot) have not implemented the image
sequence feature.
So, when erase blocks are written, the image sequence in the ec header
is lost (set to zero).
UBI scan_all() takes this case into account (commits
32bc4820287a1a03982979515949e8ea56eac641 and
2eadaad67b2b6bd132eda105128d2d466298b8e3)

But fastmap scan functions (ubi_scan_fastmap() and scan_pool()) didn't.

This patch fixes the issue.

Signed-off-by: Richard Genoud 
Signed-off-by: Richard Weinberger 
Signed-off-by: Artem Bityutskiy

UBI: Call scan_all() with correct offset in error case

2013-10-03T16:42:49+00:00

If we find an invalid fastmap we have to scan from the very beginning.
Otherwise we leak the first 64 PEBs.

Reported-and-tested-by: Richard Genoud 
Signed-off-by: Richard Weinberger 
Signed-off-by: Artem Bityutskiy

UBI: Fix error path in scan_pool()

2013-10-03T16:42:49+00:00

We have to set "ret", not "err" in case of an error.

Reported-and-tested-by: Richard Genoud 
Signed-off-by: Richard Weinberger 
Signed-off-by: Artem Bityutskiy

UBI: fix refill_wl_user_pool()

2013-10-03T16:42:46+00:00

If no free PEBs are available refill_wl_user_pool() must not
return with -ENOSPC immediately.
It has to block till produce_free_peb() produced a free PEB.

Reported-and-Tested-by: Richard Genoud 
Signed-off-by: Richard Weinberger 
Signed-off-by: Artem Bityutskiy

UBI: Fix invalidate_fastmap()

2013-08-20T14:53:25+00:00

Onging tests uncovered that invalidate_fastmap() is broken.
It must not call ubi_wl_put_fm_peb() because all PEBs used
by the old fastmap have already been put back.

Signed-off-by: Richard Weinberger 
Signed-off-by: Artem Bityutskiy