Re: [RESEND RFC PATCH v3 20/20] mtd: spi-nor: Rework the disabling of block write protection

From: Boris Brezillon
Date: Mon Aug 26 2019 - 08:49:27 EST

On Mon, 26 Aug 2019 12:09:09 +0000
<Tudor.Ambarus@xxxxxxxxxxxxx> wrote:

> From: Tudor Ambarus <tudor.ambarus@xxxxxxxxxxxxx>
> spi_nor_unlock() unlocks blocks of memory or the entire flash memory
> array, if requested. clear_sr_bp() unlocks the entire flash memory
> array at boot time. This calls for some unification, clear_sr_bp() is
> just an optimization for the case when the unlock request covers the
> entire flash size.
> Merge the clear_sr_bp() and stm_lock/unlock logic and introduce
> spi_nor_unlock_all(), which makes an unlock request that covers the
> entire flash size.
> Get rid of the MFR handling and implement specific manufacturer
> default_init() fixup hooks.
> Move write_sr_cr() to avoid to add a forward declaration. Prefix
> new function with 'spi_nor_'.
> Note that this changes a bit the logic for the SNOR_MFR_ATMEL and
> SNOR_MFR_INTEL cases. Before this patch, the Atmel and Intel chips
> did not set the locking ops, but unlocked the entire flash at boot
> time, while now they are setting the locking ops to stm_locking_ops.
> This should work, since the the disable of the block protection at the
> boot time used the same Status Register bits to unlock the flash, as
> in the stm_locking_ops case.
> In future, we should probably add new hooks to
> 'struct spi_nor_flash_parameter' to describe how to interact with the
> Status and Configuration Registers in the form of:
> nor->params.ops->read_sr
> nor->params.ops->write_sr
> nor->params.ops->read_cr
> nor->params.ops->write_sr
> We can retrieve this info starting with JESD216 revB, by checking the
> 15th DWORD of Basic Flash Parameter Table, or with later revisions of
> the standard, by parsing the "Status, Control and Configuration Register
> Map for SPI Memory Devices".
> Suggested-by: Boris Brezillon <boris.brezillon@xxxxxxxxxxxxx>
> Signed-off-by: Tudor Ambarus <tudor.ambarus@xxxxxxxxxxxxx>

Reviewed-by: Boris Brezillon <boris.brezillon@xxxxxxxxxxxxx>

Though I'd recommend waiting a bit before applying that one. As
discussed privately, we might have problems when ->quad_enable is set
to spansion_read_cr_quad_enable or spansion_no_read_cr_quad_enable.