Re: [RFC PATCH] mtd: spi-nor: write support for minor aligned partitions

[John Thomson]

On Mon, 4 Jan 2021, at 12:28, John Thomson wrote:
> --- a/drivers/mtd/mtdpart.c
> +++ b/drivers/mtd/mtdpart.c
> @@ -38,10 +38,11 @@ static struct mtd_info *allocate_partition(struct 
> mtd_info *parent,
>  	struct mtd_info *master = mtd_get_master(parent);
>  	int wr_alignment = (parent->flags & MTD_NO_ERASE) ?
>  			   master->writesize : master->erasesize;
> +	int wr_alignment_minor;

int wr_alignment_minor = 0;

> +	if (!(child->flags & MTD_NO_ERASE)) {
> 	wr_alignment = child->erasesize;
> +	if (child->erasesize_minor)

if (child->erasesize_minor && IS_ENABLED(CONFIG_MTD_SPI_NOR_USE_VARIABLE_ERASE)) {
Config test wrap wr_alignment_minor being set,
so that a partition on a minor boundary is only set writeable if the non-uniform erase path can be used.

> +		wr_alignment_minor = child->erasesize_minor;
> +	}

> +		if (wr_alignment_minor) {

smatch picked up a tested uninitialized symbol 'wr_alignment_minor' here,
initialise as 0.
Reported-by: kernel test robot <lkp@xxxxxxxxx>
Reported-by: Dan Carpenter <dan.carpenter@xxxxxxxxxx>

> +		if ((!wr_alignment_minor) || (wr_alignment_minor && remainder_minor != 0)) {

If it is safe to boolean test the ints and u32, I should use this consistently?
if ((!wr_alignment_minor) || (wr_alignment_minor && remainder_minor)) {
Or is it clearer to use the math tests?
if ((wr_alignment_minor == 0) || (wr_alignment_minor > 0 && remainder_minor > 0)) {

> --- a/drivers/mtd/spi-nor/core.c
> +++ b/drivers/mtd/spi-nor/core.c
> @@ -1225,7 +1225,11 @@ static u8 spi_nor_convert_3to4_erase(u8 opcode)
>  
>  static bool spi_nor_has_uniform_erase(const struct spi_nor *nor)
>  {
> +#ifdef CONFIG_MTD_SPI_NOR_USE_VARIABLE_ERASE

if (IS_ENABLED(CONFIG_MTD_SPI_NOR_USE_VARIABLE_ERASE)) {
and the closing brace better than the #ifdef here?

> +	return false;
> +#else
>  	return !!nor->params->erase_map.uniform_erase_type;
> +#endif
>  }
>  
>  static void spi_nor_set_4byte_opcodes(struct spi_nor *nor)


Otherwise, is this approach valid, or is there a better method I can use?

Cheers,
-- 
  John Thomson