Merge branch 'for-next/qspi'

Merge branch 'for-next/misc'

mtd: spi-nor: Add support for more Macronix devices ...

Add support for the Macronix mx66u51235f, mx66l1g45g and mx66l1g55g.

Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>

mtd: devices: m25p80: use the spi_mem_xx() API ...

This is the barebox adoption of the linux v4.16 patch

  4120f8d
  Author:     Boris Brezillon <bbrezillon@kernel.org>
  AuthorDate: Thu Apr 26 18:18:19 2018 +0200
  Commit:     Mark Brown <broonie@kernel.org>
  CommitDate: Fri May 11 11:33:51 2018 +0900

  mtd: spi-nor: Use the spi_mem_xx() API

  The spi_mem_xxx() API has been introduced to replace the
  spi_flash_read() one. Make use of it so we can get rid of
  spi_flash_read().

  Note that using spi_mem_xx() also simplifies the code because this API
  takes care of using the regular spi_sync() interface when the optimized
  ->mem_ops interface is not implemented by the controller.

  Signed-off-by: Boris Brezillon <boris.brezillon@bootlin.com>
  Reviewed-by: Frieder Schrempf <frieder.schrempf@exceet.de>
  Tested-by: Frieder Schrempf <frieder.schrempf@exceet.de>
  Signed-off-by: Mark Brown <broonie@kernel.org>

Signed-off-by: Steffen Trumtrar <s.trumtrar@pengutronix.de>
Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>

mtd: spi-nor: provide default erase_sector implementation ...

Based on the Linux v4.4 patch

commit c67cbb8
Author:     Brian Norris <computersforpeace@gmail.com>
AuthorDate: Tue Nov 10 12:15:27 2015 -0800
Commit:     Brian Norris <computersforpeace@gmail.com>
CommitDate: Thu Nov 19 13:34:44 2015 -0800

    mtd: spi-nor: provide default erase_sector implementation

    Some spi-nor drivers perform sector erase by duplicating their
    write_reg() command. Let's not require that the driver fill this out,
    and provide a default instead.

    Tested on m25p80.c and Medatek's MT8173 SPI NOR flash driver.

    Signed-off-by: Brian Norris <computersforpeace@gmail.com>

Signed-off-by: Steffen Trumtrar <s.trumtrar@pengutronix.de>
Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>

mtd: spi-nor: introduce SPI 1-2-2 and SPI 1-4-4 protocols ...

Sync the driver with Linux v4.12 and apply the patch

  commit cfc5604
  Author:     Cyrille Pitchen <cyrille.pitchen@atmel.com>
  AuthorDate: Tue Apr 25 22:08:46 2017 +0200
  Commit:     Cyrille Pitchen <cyrille.pitchen@wedev4u.fr>
  CommitDate: Mon May 15 21:56:17 2017 +0200

      mtd: spi-nor: introduce SPI 1-2-2 and SPI 1-4-4 protocols

      This patch changes the prototype of spi_nor_scan(): its 3rd parameter
      is replaced by a 'struct spi_nor_hwcaps' pointer, which tells the spi-nor
      framework about the actual hardware capabilities supported by the SPI
      controller and its driver.

      Besides, this patch also introduces a new 'struct spi_nor_flash_parameter'
      telling the spi-nor framework about the hardware capabilities supported by
      the SPI flash memory and the associated settings required to use those
      hardware caps.

      Then, to improve the readability of spi_nor_scan(), the discovery of the
      memory settings and the memory initialization are now split into two
      dedicated functions.

      1 - spi_nor_init_params()

      The spi_nor_init_params() function is responsible for initializing the
      'struct spi_nor_flash_parameter'. Currently this structure is filled with
      legacy values but further patches will allow to override some parameter
      values dynamically, for instance by reading the JESD216 Serial Flash
      Discoverable Parameter (SFDP) tables from the SPI memory.
      The spi_nor_init_params() function only deals with the hardware
      capabilities of the SPI flash memory: especially it doesn't care about
      the hardware capabilities supported by the SPI controller.

      2 - spi_nor_setup()

      The second function is called once the 'struct spi_nor_flash_parameter'
      has been initialized by spi_nor_init_params().
      With both 'struct spi_nor_flash_parameter' and 'struct spi_nor_hwcaps',
      the new argument of spi_nor_scan(), spi_nor_setup() computes the best
      match between hardware caps supported by both the (Q)SPI memory and
      controller hence selecting the relevant settings for (Fast) Read and Page
      Program operations.

      Signed-off-by: Cyrille Pitchen <cyrille.pitchen@atmel.com>
      Reviewed-by: Marek Vasut <marek.vasut@gmail.com>

Signed-off-by: Steffen Trumtrar <s.trumtrar@pengutronix.de>
Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>

mtd: spi-nor: remove unused write_enable from write_reg ...

Based on the Linux v4.2 commit:

  commit f9f3ce8
  Author:     Jagan Teki <jteki@openedev.com>
  AuthorDate: Wed Aug 19 15:26:44 2015 +0530
  Commit:     Brian Norris <computersforpeace@gmail.com>
  CommitDate: Fri Sep 11 16:04:55 2015 -0700

      mtd: spi-nor: Zap unneeded write_enable from write_reg

      The 'write_enable' argument is unused and unneeded, so remove it from
      the API.

      Signed-off-by: Jagan Teki <jteki@openedev.com>
      Cc: David Woodhouse <dwmw2@infradead.org>
      Cc: Han Xu <han.xu@freescale.com>
      [Brian: fixed for nxp-spifi.c]
      Signed-off-by: Brian Norris <computersforpeace@gmail.com>

Signed-off-by: Steffen Trumtrar <s.trumtrar@pengutronix.de>
Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>

spi: Extend the core to ease integration of SPI memory controllers ...

Sync with Linux v5.1-rc1.

This is the barebox adoption of the commit

  commit c36ff26
  Author: Boris Brezillon <boris.brezillon@bootlin.com>
  Date:   Thu Apr 26 18:18:14 2018 +0200

      spi: Extend the core to ease integration of SPI memory controllers

      Some controllers are exposing high-level interfaces to access various
      kind of SPI memories. Unfortunately they do not fit in the current
      spi_controller model and usually have drivers placed in
      drivers/mtd/spi-nor which are only supporting SPI NORs and not SPI
      memories in general.

      This is an attempt at defining a SPI memory interface which works for
      all kinds of SPI memories (NORs, NANDs, SRAMs).

      Signed-off-by: Boris Brezillon <boris.brezillon@bootlin.com>
      Reviewed-by: Frieder Schrempf <frieder.schrempf@exceet.de>
      Tested-by: Frieder Schrempf <frieder.schrempf@exceet.de>
      Signed-off-by: Mark Brown <broonie@kernel.org>

Signed-off-by: Steffen Trumtrar <s.trumtrar@pengutronix.de>
Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>

mtd: spi-nor: cadence: add cqspi_set_protocol ...

Sync up the driver with the original linux v4.7 driver version.

As only Quad-SPI mode is/was supported, the function was not
ported in the initial move to barebox.
To make future synchronizations with the kernel driver easier,
add this function and reorder the functions in the driver
accordingly.

Signed-off-by: Steffen Trumtrar <s.trumtrar@pengutronix.de>
Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>

UBI: remove Kconfig entry from include ...

CONFIG_MTD_UBI_WL_THRESHOLD can be set via Kconfig.
default there is also 4096. So remove the leftover here.

Signed-off-by: Jan Remmet <j.remmet@phytec.de>
Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>

mtd: spi-nor: add Winbond W25Q40BW ...

This 4MBit SPI chip can be found in Kindle Voyage devices

Signed-off-by: Alexander Kurz <akurz@blala.de>
Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>

Merge branch 'for-next/misc'

fs: devfs: Change .lseek callbacks to return 'int' ...

Returning requested offset from .lseek() callback doesn't really give
us any new information while bringing unnecessary
complications. Change all .lseek() types (both in struct struct
cdev_operations and in struct fs_driver_d) to return 'int' and adjust
the rest of the codebase accordingly.

Signed-off-by: Andrey Smirnov <andrew.smirnov@gmail.com>
Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>

devfs: Drop dev_lseek_default() ...

Only the following cdevs do not declare an .lseek() operation:

  - Console devices in common/console.c
  - Firmware framework in common/firmware.c
  - JTAG driver in drivers/misc/jtag.c
  - UBI in drivers/mtd/ubi/barebox.c

Of those four, first two are marked DEVFS_IS_CHARACTER_DEV and
implement only .write() operation and the last two don't implement
anything but .ioctl(). While there's probably no meaningful way to use
lseek() against any of those devices, there doesn't seem to be any
harm in allowing it either.

Change devfs_lseek() to ignore absense of .lseek() callback and drop
dev_lseek_default() and all references to it in the codebase.

Signed-off-by: Andrey Smirnov <andrew.smirnov@gmail.com>
Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>

mtd: peb: Add function to skip bad blocks ...

This adds a function that given a pointer to a PEB number increases the
number until the corresponding PEB is good. It also checks for the PEB
number being inside the mtd device.

Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>

ubi: prevent null pointer reference ...

After flashing a bogus UBI image, e.g. when the UBI is bigger than the
available flash size, it can happen that barebox hangs itself with a
stacktrace when the board code or the environment does a ubi_attach() on
boot:

    ERROR: ubi0 error: init_volumes: not enough PEBs, required 7478, available 4024
    unable to handle NULL pointer dereference at address 0x00000204
    pc : [<1fd1eaf2>]    lr : [<1fd15aad>]
    sp : 1feefd30  ip : 0000000a  fp : 1fd5d1d7
    r10: 00001d34  r9 : 00000000  r8 : 1fd5d540
    r7 : 14983148  r6 : 00000006  r5 : ffffffe4  r4 : 14983004
    r3 : 1fd82bfc  r2 : 021f0000  r1 : 0000000a  r0 : 00000000
    Flags: nzcv  IRQs off  FIQs off  Mode SVC_32
    [<1fd1eaf2>] (ubi_fastmap_destroy_checkmap+0x8/0xa) from [<1fd15aad>] (ubi_read_volume_table+0x4d9/0x71c)
    [<1fd15aad>] (ubi_read_volume_table+0x4d9/0x71c) from [<1fd1cbf7>] (ubi_attach+0x157/0x1f8)
    [<1fd1cbf7>] (ubi_attach+0x157/0x1f8) from [<1fd16ec3>] (ubi_attach_mtd_dev+0x4a3/0x954)
    [...]

After this change, barebox at least boots to a prompt where the faulty
flash can be repaired:

    ERROR: ubi0 error: init_volumes: not enough PEBs, required 7478, available 4024
    ERROR: ubi0 error: ubi_attach_mtd_dev: failed to attach mtd0, error -28
    [...]
    running /env/bin/init...
    barebox@boardname:/

Signed-off-by: Roland Hieber <rhi@pengutronix.de>
Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>

Revert "mtd: nand: Kill the chip->scan_bbt() hook" ...

This reverts commit 95ce697.

Merge branch 'for-next/mtd'

Merge branch 'for-next/imx'

mtd: core: Fix erase area alignment for non power of 2 erasesize ...

On Thu, Jan 10, 2019 at 09:32:07AM +0100, Sascha Hauer wrote:
> On Wed, Jan 09, 2019 at 12:28:14PM +0100, Ladislav Michl wrote:
> > Devices as AT45DB161 DataFlash uses non power of two page size (528)
> > while present alignment algorithm relies on erasesize being power
> > of 2.
> > Fix that by introducing helper functions rounding to any multiply.
> > Note that logic is sligthly changed to be consistent as ending
> > address is moved forward to include also last byte meant to be
> > erased while previous implementation moved it backward.
> >
> > Signed-off-by: Ladislav Michl <ladis@linux-mips.org>
> > ---
> >  drivers/mtd/core.c     | 28 ++++++++++++++++++++++------
> >  include/linux/kernel.h |  5 +++++
> >  2 files changed, 27 insertions(+), 6 deletions(-)
>
> Applied, thanks

Hi Sascha,

I was searching for paper bag, but was unable to find anything thick
enouh to cover behind :-/ This version is buggy and I noticed right
now when testing on another board. Corrected version follows, which
is also 192 bytes shorter for my ARM target. It seems next branch
at https://git.pengutronix.de/ is not yet updated, so perhaps my fault
won't affect anyone.

Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>

mtd: core: Fix erase area alignment for non power of 2 erasesize ...

Devices as AT45DB161 DataFlash uses non power of two page size (528)
while present alignment algorithm relies on erasesize being power
of 2.
Fix that by introducing helper functions rounding to any multiply.
Note that logic is sligthly changed to be consistent as ending
address is moved forward to include also last byte meant to be
erased while previous implementation moved it backward.

Signed-off-by: Ladislav Michl <ladis@linux-mips.org>
Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>

mtd: nand: omap: unbreak BCH8 support ...

We now tell the NAND layer that we have subpages with 512 bytes each.
This means the calculate_ecc / correct hooks are also called for each
subpage which means we cannot iterate over subpages in the calculate_ecc
callback.

Implement this correctly by removing the loop over subpages in the
calculate_ecc callback and by implementing a read_page callback.

With this the bch8_hw and bch8_hw_romcode were tested on AM335x, OMAP4
and AM3517. Number of bitflips is correctly reported as tested with the
nand_bitflip tool.

Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>

With fix from Teresa:

Using BCH rom code mode we start calculating with GPMC_BCH_RESULTx_1.
Fix offset for correct error correction.

Signed-off-by: Teresa Remmet <t.remmet@phytec.de>
Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>

mtd: nand: nand_base: Fix compile warning if CONFIG_MTD_WRITE is not enabled ...

drivers/mtd/nand/nand_base.c:337:12: warning: 'nand_default_block_markbad'
defined but not used [-Wunused-function]

Add __maybe_unused again to silence warning when CONFIG_MTD_WRITE is not enabled.

Signed-off-by: Teresa Remmet <t.remmet@phytec.de>
Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>

mtd: nand: detect OOB size for Toshiba 24nm raw SLC ...

Linux commit 60c673824561.

Signed-off-by: Ladislav Michl <ladis@linux-mips.org>
Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>

mtd: nand: Kill cellinfo ...

The only information used from cellinfo field is whenever flash is SLC
or MLC, therefore eliminate it completely. This patch is based on Linux
commit 7db906b79f69.

Signed-off-by: Ladislav Michl <ladis@linux-mips.org>
Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>

mtd: nand: Kill the chip->scan_bbt() hook ...

Linux commit e80eba758151 adapted for Barebox:

  None of the existing drivers are overloading the ->scan_bbt()
  method, let's get rid of it and replace calls to ->scan_bbt()
  by nand_create_bbt() ones.

Signed-off-by: Ladislav Michl <ladis@linux-mips.org>
Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>

mtd: nand_bbt: scan for next free bbt block if writing bbt fails ...

Linux commit 10ffd570f117 adapted for Barebox:

  If erasing or writing the BBT fails, we should mark the current BBT
  block as bad and use the BBT descriptor to scan for the next available
  unused block in the BBT. We should only return a failure if there isn't
  any space left.

Signed-off-by: Ladislav Michl <ladis@linux-mips.org>
Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>

mtd: nand_bbt: Move BBT block selection logic out of write_bbt() ...

Linux commit c3baf278d3bf adapted for Barebox:

  This clarifies the write_bbt() function by removing the write label
  and simplifying the error/exit path.

Signed-off-by: Ladislav Michl <ladis@linux-mips.org>
Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>

mtd: nand_bbt: unify/fix error handling in nand_scan_bbt() ...

Linux commit 83c59542d0af adapted for Barebox:

  Don't leak this->bbt, and return early if check_create() fails. It helps
  to have a single error path to avoid these problems.

Signed-off-by: Ladislav Michl <ladis@linux-mips.org>
Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>

mtd: nand_bbt: make nand_scan_bbt() static ...

Linux commit 17799359e7b adapted for Barebox:

  This implementation detail is no longer needed outside of nand_bbt.c.

Signed-off-by: Ladislav Michl <ladis@linux-mips.org>
Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>