A large number of files include <flash.h> as it used to be how various
SPI flash related functions were found, or for other reasons entirely.
In order to migrate some further CONFIG symbols to Kconfig we need to
not include flash.h in cases where we don't have a NOR flash of some
sort enabled.  Furthermore, in cases where we are in common code and it
doesn't make sense to try and further refactor the code itself in to new
files we need to guard this inclusion.

Signed-off-by: Tom Rini <trini@konsulko.com>

Now, spi_flash_probe_bus_cs() relies on DT for spi speed and mode
and logically calls spi_get_bus_and_cs(). In case spi mode and speed are
not read from DT, make usage of spi_flash_probe() instead.

To sum-up:
 - Previous call tree was:
    spi_flash_probe() -> spi_flash_probe_bus_cs() -> spi_get_bus_and_cs()

 - Current call tree is:
    spi_flash_probe() -> _spi_get_bus_and_cs()
    spi_flash_probe_bus_cs() -> spi_get_bus_and_cs()

This patch impacts the following :
  - cmd/sf.c: if spi mode and/or speed is passed in argument of
    do_spi_flash_probe(), call spi_flash_probe() otherwise call
    spi_flash_probe_bus_cs().

  - drivers/net/fm/fm.c: as by default spi speed and mode was set to
    0 and a comment indicates that speed and mode are read from DT,
    use spi_flash_probe_bus_cs().

  - drivers/net/pfe_eth/pfe_firmware.c: spi speed and mode are not read
    from DT by all platforms using this driver, so keep legacy and replace
    spi_flash_probe_bus_cs() by spi_flash_probe();

  - drivers/net/sni_netsec.c : spi speed and mode are not read from DT,
    so replace spi_flash_probe_bus_cs() by spi_flash_probe().

  - drivers/usb/gadget/max3420_udc.c: Can't find any platform which make
    usage of this driver, nevertheless, keep legacy and replace
    spi_get_bus_and_cs() by _spi_get_bus_and_cs().

  - env/sf.c: a comment indicates that speed and mode are read
    from DT. So use spi_flash_probe_bus_cs().

Signed-off-by: Patrice Chotard <patrice.chotard@foss.st.com>

Cc: Marek Behun <marek.behun@nic.cz>
Cc: Jagan Teki <jagan@amarulasolutions.com>
Cc: Vignesh R <vigneshr@ti.com>
Cc: Joe Hershberger <joe.hershberger@ni.com>
Cc: Ramon Fried <rfried.dev@gmail.com>
Cc: Lukasz Majewski <lukma@denx.de>
Cc: Marek Vasut <marex@denx.de>
Cc: Wolfgang Denk <wd@denx.de>
Cc: Simon Glass <sjg@chromium.org>
Cc: Stefan Roese <sr@denx.de>
Cc: "Pali Rohár" <pali@kernel.org>
Cc: Konstantin Porotchkin <kostap@marvell.com>
Cc: Igal Liberman <igall@marvell.com>
Cc: Bin Meng <bmeng.cn@gmail.com>
Cc: Pratyush Yadav <p.yadav@ti.com>
Cc: Sean Anderson <seanga2@gmail.com>
Cc: Anji J <anji.jagarlmudi@nxp.com>
Cc: Biwen Li <biwen.li@nxp.com>
Cc: Priyanka Jain <priyanka.jain@nxp.com>
Cc: Chaitanya Sakinam <chaitanya.sakinam@nxp.com>

In some cases it makes sense to use env_sf_init_addr() also in SPL mode.
Allow it for boards by providing custom implementation of weak function
env_sf_get_env_addr(). When this function returns NULL it signals that
address is invalid, like config option CONFIG_ENV_ADDR.

There is no change in default behavior or in config options.

Signed-off-by: Pali Rohár <pali@kernel.org>

env_sf_init_addr() says the environment is valid even if it is assigning
default environment due to CRC failure. Change this to ENV_INVALID and
let the generic env_init() function, which calls this initializer,
assign the default environment.

Signed-off-by: Marek Behún <marek.behun@nic.cz>
Reviewed-by: Simon Glass <sjg@chromium.org>

The function env_sf_init_addr() assigns number literals (1) instead of
ENV_VALID to gd->env_valid. Fix this.

Signed-off-by: Marek Behún <marek.behun@nic.cz>
Reviewed-by: Simon Glass <sjg@chromium.org>

In the if clause we use tabs and in the else clause spaces. Let's use
spaces in the if clause too.

Signed-off-by: Marek Behún <marek.behun@nic.cz>
Reviewed-by: Simon Glass <sjg@chromium.org>

As the the SPI flash is probed and is released in each ENV sf function
the env_flash no more need to be static.

This patch move this device handle as local variable of each function and
simplify the associated code (env_flash is never == NULL when
setup_flash_device is called).

Signed-off-by: Patrick Delaunay <patrick.delaunay@foss.st.com>

Free the SPI resources by calling spi_flash_free() in each env sf
function to avoid issue for other SPI users.

Signed-off-by: Patrick Delaunay <patrick.delaunay@foss.st.com>

Add support of opts erase for env in SPI flash;
this opts is used by command 'env erase'.

This command only fills the env offset by 0x0 (bit flip to 0) and
the saved environment becomes invalid (with bad CRC).

It doesn't erase the sector here to avoid issue when the sector
is larger than the env (i.e. embedded when
CONFIG_ENV_SECT_SIZE > CONFIG_ENV_SIZE).

The needed sector erase will be managed in the next "env save" command,
using the opt ".save", before to update the environment in SPI flash.

Signed-off-by: Patrick Delaunay <patrick.delaunay@foss.st.com>

Remove CONFIG_IS_ENABLED(SAVEENV) as it is already tested in
the ENV_SAVE_PTR macro.

Signed-off-by: Patrick Delaunay <patrick.delaunay@foss.st.com>