[PATCH v2 2/2] mtd: spi-nor: Add Renesas AT25QL128A serial nor flash

[Biju Das]

Hi Tudor Ambarus,

I have done quick test with these generic patches. Please find my observation
below.

> Subject: Re: [PATCH v2 2/2] mtd: spi-nor: Add Renesas AT25QL128A serial nor
> flash
> 
> On 7/15/22 13:57, Biju Das wrote:
> 
> Hi, Biju!
> 
> > EXTERNAL EMAIL: Do not click links or open attachments unless you know
> > the content is safe
> >
> > Add support for Renesas AT25QL128A serial nor flash.
> > Details of flash chip can be found here [1]
> >
> > [1]
> >
> > Signed-off-by: Biju Das <biju.das.jz at bp.renesas.com>
> > ---
> > $ xxd -p sfdp
> > 53464450060101ff00060110300000ff1f00010280000001ffffffffffff
> > ffffffffffffffffffffffffffffffffffffe520f1ffffffff0744eb086b
> > 083b80bbfeffffffffff00ffffff42eb0c200f5210d800ff3362d5008429
> > 01ceeca1073d7a757a75f7a2d55c19f61cffe810c080ffffffffffffffff
> > ffffffffffffffff501650190000ffff
> >
> > $ md5sum
> > /sys/devices/platform/soc/10060000.spi/rpc-if-spi/spi_master/spi1/spi1
> > .0/spi-nor/sfdp 23e3ec56b5b8f986d0488ba4727239dd
> > /sys/devices/platform/soc/10060000.spi/rpc-if-spi/spi_master/spi1/spi1
> > .0/spi-nor/sfdp $ cat
> > /sys/devices/platform/soc/10060000.spi/rpc-if-spi/spi_master/spi1/spi1
> > .0/spi-nor/jedec_id
> > 1f4218
> > $ cat
> > /sys/devices/platform/soc/10060000.spi/rpc-if-spi/spi_master/spi1/spi1
> > .0/spi-nor/partname
> > at25ql128a
> > $ cat
> > /sys/devices/platform/soc/10060000.spi/rpc-if-spi/spi_master/spi1/spi1
> > .0/spi-nor/manufacturer
> > atmel

With generic flash driver patches I get below results
----------------------------------------------------

xxd -p sfdp
53464450060101ff00060110300000ff1f00010280000001ffffffffffff
ffffffffffffffffffffffffffffffffffffe520f1ffffffff0744eb086b
083b80bbfeffffffffff00ffffff42eb0c200f5210d800ff3362d5008429
01ceeca1073d7a757a75f7a2d55c19f61cffe810c080ffffffffffffffff
ffffffffffffffff501650190000ffff

This result is same as above.

cat /sys/devices/platform/soc/10060000.spi/rpc-if-spi/spi_master/spi1/spi1.0/spi-nor/jedec_id
1f42181f4218

This result is not matching 1f4218->1f42181f4218

root at smarc-rzg2ul:~# cat /sys/devices/platform/soc/10060000.spi/rpc-if-spi/spi_master/spi1/spi1.0/spi-nor/partname
spi-nor-generic

This result is not matching at25ql128a-> spi-nor-generic

root at smarc-rzg2ul:~# cat /sys/devices/platform/soc/10060000.spi/rpc-if-spi/spi_master/spi1/spi1.0/spi-nor/manufacturer
cat: /sys/devices/platform/soc/10060000.spi/rpc-if-spi/spi_master/spi1/spi1.0/spi-nor/manufacturer: No such file or directory

This result is not matching atmel-> cat: /sys/devices/platform/soc/10060000.spi/rpc-if-spi/spi_master/spi1/spi1.0/spi-nor/manufacturer: No such file or directory

How do we get proper partname/manufacturer details with generic flash driver?

Cheers,
Biju


> > v2->v1:
> >  * Added flash definitions into Atmel table.
> > RFC->v1:
> >  * Moved the flash definitions to atmel.c
> >  * Remove macro related to locking as it is untested.
> >  * Replaced INFO macro with SNOR_ID3
> >
> > RFC:
> >  *
> > ---
> >  drivers/mtd/spi-nor/atmel.c | 1 +
> >  1 file changed, 1 insertion(+)
> >
> > diff --git a/drivers/mtd/spi-nor/atmel.c b/drivers/mtd/spi-nor/atmel.c
> > index 656dd80a0be7..80c24d138b27 100644
> > --- a/drivers/mtd/spi-nor/atmel.c
> > +++ b/drivers/mtd/spi-nor/atmel.c
> > @@ -186,6 +186,7 @@ static const struct flash_info atmel_nor_parts[] = {
> >                 .fixups = &atmel_nor_global_protection_fixups },
> >         { "at25sl321",  INFO(0x1f4216, 0, 64 * 1024, 64)
> >                 NO_SFDP_FLAGS(SECT_4K | SPI_NOR_DUAL_READ |
> > SPI_NOR_QUAD_READ) },
> > +       { "at25ql128a", SNOR_ID3(0x1f4218) },
> 
> As the flash is defined now, all its characteristics can be discovered by
> parsing SFDP. So instead of adding a new flash_info entry in the already big
> array that we have, I would like us to focus more on a generic flash driver
> which acts as a backup when the flash is not found in the flash_info array.
> We can discover the flash particularities by parsing SFDP and we can probe
> the flash based on that. Michael has already suggested this, would you
> please help review it?
> 
> 
> >         { "at26f004",   INFO(0x1f0400, 0, 64 * 1024,  8)
> >                 NO_SFDP_FLAGS(SECT_4K) },
> >         { "at26df081a", INFO(0x1f4501, 0, 64 * 1024, 16)
> > --
> > 2.25.1
> >