Create a SPI NOR manufacturer driver for ESMT chips, and move the
ESMT definitions outside of core.c.
Signed-off-by: NBoris Brezillon <bbrezillon@kernel.org>
Signed-off-by: NTudor Ambarus <tudor.ambarus@microchip.com>

Create a SPI NOR manufacturer driver for Eon chips, and move the
Eon definitions outside of core.c.
Signed-off-by: NBoris Brezillon <bbrezillon@kernel.org>
Signed-off-by: NTudor Ambarus <tudor.ambarus@microchip.com>

Create a SPI NOR manufacturer driver for Atmel chips, and move the
Atmel definitions outside of core.c.
Signed-off-by: NBoris Brezillon <bbrezillon@kernel.org>
Signed-off-by: NTudor Ambarus <tudor.ambarus@microchip.com>

Declare a spi_nor_manufacturer struct and add basic building blocks to
move manufacturer specific code outside of the core.
Signed-off-by: NBoris Brezillon <bbrezillon@kernel.org>
Signed-off-by: NTudor Ambarus <tudor.ambarus@microchip.com>
Reviewed-by: NVignesh Raghavendra <vigneshr@ti.com>

Expose the flash_info struct and some function prototypes that
will be used by manufacturers.
Signed-off-by: NBoris Brezillon <bbrezillon@kernel.org>
Signed-off-by: NTudor Ambarus <tudor.ambarus@microchip.com>
Reviewed-by: NVignesh Raghavendra <vigneshr@ti.com>

It makes the core file a bit smaller and provides better separation
between the SFDP parsing and core logic.

Keep the core.h and sfdp.h definitions private in drivers/mtd/spi-nor/.
Both expose just the definitions that are required by the core and
manufacturer drivers. None of the SPI NOR controller drivers should
include them.
Signed-off-by: NTudor Ambarus <tudor.ambarus@microchip.com>
Reviewed-by: NBoris Brezillon <boris.brezillon@collabora.com>
Reviewed-by: NVignesh Raghavendra <vigneshr@ti.com>

Move all SPI NOR controller drivers to a controllers/ sub-directory
so that we only have SPI NOR related source files under
drivers/mtd/spi-nor/.

Rename spi-nor.c into core.c, we are about to split this file in multiple
source files (one per manufacturer, plus one for the SFDP parsing logic).
Signed-off-by: NBoris Brezillon <bbrezillon@kernel.org>
Signed-off-by: NTudor Ambarus <tudor.ambarus@microchip.com>
Reviewed-by: NVignesh Raghavendra <vigneshr@ti.com>

Rename (*set_4byte)() to (*set_4byte_addr_mode)() for a better
differentiation between the 4 byte address mode and opcodes.

Rename macronix_set_4byte() to spi_nor_set_4byte_addr_mode(), it will be
the only 4 byte address mode method exposed to the manufacturer drivers.

Here's how the manufacturers enter and exit the 4 byte address mode:
- eon, gidadevice, issi, macronix, xmc use EN4B/EX4B
- micron-st needs WEN. st_micron_set_4byte_addr_mode() will become
  a private method, as they are the only ones that need WEN before the
  EN4B/EX4B commands.
- newer spansion have a 4BAM opcode (this translates to a new, public
  command). Older spansion  flashes use the BRWR command (legacy in
  core.c -> spansion_set_4byte_addr_mode())
- winbond's method is hackish and may be reason for just a flash
  fixup hook -> private method
Signed-off-by: NTudor Ambarus <tudor.ambarus@microchip.com>
Reviewed-by: NVignesh Raghavendra <vigneshr@ti.com>

Replace the manufacturer prefix by something describing more precisely
what those functions do.
Signed-off-by: NBoris Brezillon <bbrezillon@kernel.org>
[tudor.ambarus@microchip.com: prepend spi_nor_ to all modified methods.]
Signed-off-by: NTudor Ambarus <tudor.ambarus@microchip.com>
Reviewed-by: NVignesh Raghavendra <vigneshr@ti.com>

- Don't use `tmp` for two purposes (return value, loop counter).
  Instead, use `i` for the loop counter, and `ret` for the return value.
- Don't use tabs between type and name in variable declarations,
  for consistency with other functions in spi-nor.c.
- Rewrite nested `if`s as `if (a && b)`.
- Remove `info` variable, and use spi_nor_ids[i] directly.
Signed-off-by: NJonathan Neuschäfer <j.neuschaefer@gmx.net>
[tudor.ambarus@microchip.com: change i's type from int to unsigned int,
reorder local variables]
Signed-off-by: NTudor Ambarus <tudor.ambarus@microchip.com>

Make use of the spi-mem direct mapping API to let advanced controllers
optimize read/write operations when they support direct mapping.

Based on the original patch by Boris Brezillon
<boris.brezillon@bootlin.com>.
Signed-off-by: NSergei Shtylyov <sergei.shtylyov@cogentembedded.com>
Reviewed-by: NBoris Brezillon <boris.brezillon@collabora.com>
Signed-off-by: NTudor Ambarus <tudor.ambarus@microchip.com>

spi_nor_spimem_xfer_data() being a helper function for the data reads/
writes contains 3 fragments that depend on the data direction; and I'm
going to add another one to call the SPI dirmap API...
I think this function should be split so that the common fragments are
put into 2 functions, spi_nor_spimem_bounce() and spi_nor_spimem_exec_op()
called from spi_nor_spimem_{read|write}_data(), and the data direction
dependent bits moved back into those read/write functions -- that way we
would be able to avoid *goto*s otherwise needed in the next patch adding
the SPI dirmap support...
Signed-off-by: NSergei Shtylyov <sergei.shtylyov@cogentembedded.com>
Signed-off-by: NTudor Ambarus <tudor.ambarus@microchip.com>

The shift variable of SR_BP is conclusive because the first bit of SR_BP
is fixed on all known flashes. Replace ffs operation with SR_BP_SHIFT.
Signed-off-by: NJungseung Lee <js07.lee@samsung.com>
Signed-off-by: NTudor Ambarus <tudor.ambarus@microchip.com>

The driver calls le32_to_cpu() to convert the little-endian tables
to  a CPU endianness, where le32_to_cpus() should have been called.
Was going to use that one... and then discovered a whole array
converter, le32_to_cpu_array()! :-)
Signed-off-by: NSergei Shtylyov <sergei.shtylyov@cogentembedded.com>
Signed-off-by: NTudor Ambarus <tudor.ambarus@microchip.com>

This was tested in single, dual and quad mode on a custom board with the
NXP FlexSPI controller.
Signed-off-by: NMichael Walle <michael@walle.cc>
Signed-off-by: NTudor Ambarus <tudor.ambarus@microchip.com>

Add support for the Winbond W25Q32JW-xM flashes. These have a
programmable QE bit. There is also the W25Q32JW-xQ variant which shares
the ID with the W25Q32DW and W25Q32FW parts. The W25Q32JW-xQ has the QE
bit hard strapped to 1, thus don't support the /HOLD and /WP pins.

This was tested in single, dual and quad mode on a custom board with the
NXP FlexSPI controller. Also the BP bits as well as the TB bit were
tested.
Signed-off-by: NMichael Walle <michael@walle.cc>
Signed-off-by: NTudor Ambarus <tudor.ambarus@microchip.com>

The commit 7b678c69 ("mtd: spi-nor: Merge spansion Quad Enable
methods") forgot to actually set the QE bit in some cases. Thus this
breaks quad mode accesses to flashes which support readback of the
status register-2. Fix it.

Fixes: 7b678c69 ("mtd: spi-nor: Merge spansion Quad Enable methods")
Signed-off-by: NMichael Walle <michael@walle.cc>
Reviewed-by: NTudor Ambarus <tudor.ambarus@microchip.com>
Signed-off-by: NMiquel Raynal <miquel.raynal@bootlin.com>

mtd->size is still unassigned when running spansion_post_sfdp_fixups()
hook, therefore use nor->params.size to determine the size of flash device.

This makes sure that 4-byte addressing opcodes are used on Spansion
flashes that are larger than 16MiB and don't have SFDP 4BAIT table
populated.

Fixes: 92094ebc ("mtd: spi-nor: Add spansion_post_sfdp_fixups()")
Signed-off-by: NVignesh Raghavendra <vigneshr@ti.com>
Reviewed-by: NTudor Ambarus <tudor.ambarus@microchip.com>
Signed-off-by: NMiquel Raynal <miquel.raynal@bootlin.com>

The ops aren't used in any SPI NOR controller. Therefore, remove them
altogether.
Signed-off-by: NMichael Walle <michael@walle.cc>
Signed-off-by: NTudor Ambarus <tudor.ambarus@microchip.com>

Adds support for sst26vf016b, a smaller variant of the sst26vf064b.
Tested on a sama5d3 SOC using the atmel,at91rm9200-spi controller.
The dual read could not be tested.

SPI_NOR_QUAD_READ is not added because it requires that the IOC bit
from the Configuration Register to be set to 1, which is not the case:
its default value at power-up is zero and we don't set it to one in
spi-nor either.
Signed-off-by: NJoseph Kust <joseph.kust@rockwellcollins.com>
Signed-off-by: NBrandon Maier <brandon.maier@rockwellcollins.com>
[tudor.ambarus@microchip.com: Drop quad read flag, amend commit description]
Signed-off-by: NTudor Ambarus <tudor.ambarus@microchip.com>

Micron flashes do not support 16 bit writes on the Status Register.
According to micron datasheets, when using the Write Status Register
(01h) command, the chip select should be driven LOW and held LOW until
the eighth bit of the last data byte has been latched in, after which
it must be driven HIGH. If CS is not driven HIGH, the command is not
executed, flag status register error bits are not set, and the write enable
latch remains set to 1. This fixes the lock operations on micron flashes.
Reported-by: NJohn Garry <john.garry@huawei.com>
Fixes: 39d1e334 ("mtd: spi-nor: Fix clearing of QE bit on lock()/unlock()")
Signed-off-by: NTudor Ambarus <tudor.ambarus@microchip.com>
Tested-by: NJohn Garry <john.garry@huawei.com>
Signed-off-by: NMiquel Raynal <miquel.raynal@bootlin.com>

Add MTD support for the Macronix MX25R3235F SPI NOR chip from Macronix.
The chip has 4MB of total capacity, divided into a total of 64 sectors,
each 64KB sized. The chip also supports 4KB large sectors.
Additionally, it supports dual and quad read modes.

Functionality was verified on an HPE/Aruba AP-303 board.
Signed-off-by: NDavid Bauer <mail@david-bauer.net>
Signed-off-by: NTudor Ambarus <tudor.ambarus@microchip.com>

Add USE_FSR flag to all variants of n25q entries that support Flag Status
Register.
Signed-off-by: NVignesh Raghavendra <vigneshr@ti.com>
Tested-by: John Garry <john.garry@huawei.com> #for n25q128a13
Signed-off-by: NTudor Ambarus <tudor.ambarus@microchip.com>

Add entries for mt25q*256a and mt25q*512a flashes. These are similar to
existing n25q variants but support stateless 4 byte addressing opcodes
Signed-off-by: NVignesh Raghavendra <vigneshr@ti.com>
Signed-off-by: NTudor Ambarus <tudor.ambarus@microchip.com>

mt25q family is different from n25q family of devices, even though manf
ID and device IDs are same. mt25q flash has bit 6 set in 5th byte of
READ ID response which can be used to distinguish it from n25q variant.
mt25q flashes support stateless 4 Byte addressing opcodes where as n25q
flashes don't. Therefore, have two separate entries for mt25qu512a and
n25q512a.

Fixes: 9607af6f ("mtd: spi-nor: Rename "n25q512a" to "mt25qu512a (n25q512a)"")
Signed-off-by: NVignesh Raghavendra <vigneshr@ti.com>
Signed-off-by: NTudor Ambarus <tudor.ambarus@microchip.com>

Tested on the Amlogic aml-libretech-pc platform which does not support
dual or quad modes
Signed-off-by: NJerome Brunet <jbrunet@baylibre.com>
[tudor.ambarus@microchip.com: order entry alphabetically]
Signed-off-by: NTudor Ambarus <tudor.ambarus@microchip.com>

For GD25Q256, wrong SR bit for top/bottom selection is being used.
Fix it to use appropriate bit.
Signed-off-by: NJungseung Lee <js07.lee@samsung.com>
Signed-off-by: NTudor Ambarus <tudor.ambarus@microchip.com>

There are some flashes to use bit 6 of status register for Top/Bottom (TB).
Use top/bottom bit variable instead of fixed value and support this case.

Set the Top/Bottom (TB) mask based on SPI_NOR_TB_SR_BIT6 flash_info flag.
We can't use a bigger granularity, for example to set TB_BIT6 per
manufacturer using a SNOR_F flag. The manufacturers don't have a common
rule in regards to the TB bit:

Winbond : Use the 6th bit from 32MB capacity
W25Q20EW, W25Q50BW, W25Q128V - TB(5)
W25Q256JV, W25M512JV - TB(6)

GigaDevice : Use the 6th bit from 32MB capacity
GD25Q16C, GD25Q32C, GD25LQ32D, GD25Q64C, GD25Q128 - TB(5)
GD25Q256 - TB(6)

Micron/STM : Keep to use 5th bit
M25PX64, N25Q128A, N25Q512A, MT25QL512ABB, MT25QL02GCBB - TB(5)

Spansion : Use the 6th bit from 16MB capacity
S25FL116K, S25FL132K, S25FL165K - TB(5)
S25FL128L, S25FL256L - TB(6)

We can't make a correlation between TB and BP3 either, i.e. assume that if
BP3 is defined then TB will be at BIT(6). Micron breaks this rule.
Signed-off-by: NJungseung Lee <js07.lee@samsung.com>
[tudor.ambarus@microchip.com: describe the reason for setting a
new flash_info flag.]
Signed-off-by: NTudor Ambarus <tudor.ambarus@microchip.com>

Each vendor uses different bits for SR_TB of flash.
To avoid ambiguity, rename SR_TB to indicate the bit used.
Signed-off-by: NJungseung Lee <js07.lee@samsung.com>
Signed-off-by: NTudor Ambarus <tudor.ambarus@microchip.com>

Fix typo s/JESB216/JESD216/
Signed-off-by: NJungseung Lee <js07.lee@samsung.com>
Signed-off-by: NTudor Ambarus <tudor.ambarus@microchip.com>

The w25q256 supports 4-byte opcodes so lets add the flag.
Tested on OpenWrt under 4.19.82 kernel on 8devices Habanero.
Signed-off-by: NRobert Marko <robimarko@gmail.com>
Signed-off-by: NTudor Ambarus <tudor.ambarus@microchip.com>

Set the default Quad Enable method for ISSI flashes. Used for
ISSI flashes (IS25WP256D-JMLE) that do not support SFDP tables
and can not determine the Quad Enable method by parsing BFPT.

Based on code originally written by Wesley Terpstra <wesley@sifive.com>
and/or Palmer Dabbelt <palmer@sifive.com>
https://github.com/riscv/riscv-linux/commit/c94e267766d62bc9a669611c3d0c8ed5ea26569bSigned-off-by: NSagar Shrikant Kadam <sagar.kadam@sifive.com>
[tudor.ambarus@microchip.com:
- rebase, split and adapt for latest spi-nor/next,
- use PMC CFI ID for ISSI. According to JEP106BA, "Programmable Micro Corp"
  changed its name to Integrated Silicon Solution (ISSI)]
Signed-off-by: NTudor Ambarus <tudor.ambarus@microchip.com>
Reviewed-by: NVignesh Raghavendra <vigneshr@ti.com>

Update the spi_nor_id table for is25wp256 (32MB) device from ISSI,
present on HiFive Unleashed dev board (Rev: A00).

Use the post bfpt fixup hook for the is25wp256 device, as done for
the is25lp256 device to overwrite the wrong address width advertised
by BFPT.
Signed-off-by: NSagar Shrikant Kadam <sagar.kadam@sifive.com>
[tudor.ambarus@microchip.com: rebase, split and adapt for latest spi-nor/next]
Signed-off-by: NTudor Ambarus <tudor.ambarus@microchip.com>
Reviewed-by: NVignesh Raghavendra <vigneshr@ti.com>

Add MTD support for w25q256jw SPI NOR chip from Winbond. This chip
supports dual/quad I/O mode with 512 blocks of memory organized in
64KB sectors. In addition to this, there is also small 4KB sectors
available for flexibility. The device has been validated using Thor96
board.

Cc: Marek Vasut <marek.vasut@gmail.com>
Cc: Tudor Ambarus <tudor.ambarus@microchip.com>
Cc: David Woodhouse <dwmw2@infradead.org>
Cc: Brian Norris <computersforpeace@gmail.com>
Cc: Miquel Raynal <miquel.raynal@bootlin.com>
Cc: Richard Weinberger <richard@nod.at>
Cc: Vignesh Raghavendra <vigneshr@ti.com>
Cc: linux-mtd@lists.infradead.org
Signed-off-by: NDarshak Patel <darshak.patel@einfochips.com>
[Mani: cleaned up for upstream]
Signed-off-by: NManivannan Sadhasivam <manivannan.sadhasivam@linaro.org>
Signed-off-by: NTudor Ambarus <tudor.ambarus@microchip.com>

When the controller is not under the SPI-MEM interface it may implement
the optional controller_ops->erase() method.

nor->spimem and nor->controller_ops are mutually exclusive. Move the
nor->controller_ops->erase != NULL check as an 'else if' case to
nor->spimem, in order to avoid the nor->controller_ops != NULL
check.
Reported-by: NDan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: NTudor Ambarus <tudor.ambarus@microchip.com>
Reviewed-by: NBoris Brezillon <boris.brezillon@collabora.com>

Expand the spi_nor_check() to make sure that nor->spimem and
nor->controller_ops are mutually exclusive.

Fixes: b35b9a10 ("mtd: spi-nor: Move m25p80 code in spi-nor.c")
Reported-by: NDan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: NTudor Ambarus <tudor.ambarus@microchip.com>
Reviewed-by: NBoris Brezillon <boris.brezillon@collabora.com>

Rename macronix_quad_enable() to a generic name:
spi_nor_sr1_bit6_quad_enable().

Prepend "spi_nor_" to "sr2_bit7_quad_enable". All SPI NOR generic
methods should be prepended by "spi_nor_".
Signed-off-by: NTudor Ambarus <tudor.ambarus@microchip.com>
Reviewed-by: NVignesh Raghavendra <vigneshr@ti.com>

Merge
    spansion_no_read_cr_quad_enable()
    spansion_read_cr_quad_enable()
into
    spi_nor_sr2_bit1_quad_enable().

Reduce code duplication by introducing spi_nor_write_16bit_cr_and_check().
The Configuration Register contains bits that can be updated in future:
FREEZE, CMP. Provide a generic method that allows updating all bits
of the Configuration Register.
Signed-off-by: NTudor Ambarus <tudor.ambarus@microchip.com>
Reviewed-by: NVignesh Raghavendra <vigneshr@ti.com>

JEDEC Basic Flash Parameter Table, 15th DWORD, bits 22:20,
refers to this bit as "bit 1 of the status register 2".
Rename the macro accordingly.
Signed-off-by: NTudor Ambarus <tudor.ambarus@microchip.com>
Reviewed-by: NVignesh Raghavendra <vigneshr@ti.com>

Test that all the bits from Status Register 1 and Status Register 2
were written correctly.
Signed-off-by: NTudor Ambarus <tudor.ambarus@microchip.com>
Reviewed-by: NVignesh Raghavendra <vigneshr@ti.com>