[PATCH v5 8/8] mtd: spi-nor: macronix: Add support for mx25uw51245g with RWW

[Tudor Ambarus]

On 3/31/23 20:37, Miquel Raynal wrote:
> Hi Tudor,
> 

Hi!

> tudor.ambarus at linaro.org wrote on Wed, 29 Mar 2023 12:12:29 +0100:
> 
>> On 3/29/23 06:48, liao jaime wrote:
>>> Hi Miquel
>>>   
>>>>
>>>> Hi Tudor,
>>>>
>>>> tudor.ambarus at linaro.org wrote on Tue, 28 Mar 2023 17:31:46 +0100:
>>>>  
>>>>> On 3/28/23 16:41, Miquel Raynal wrote:  
>>>>>> Describe this new part and provide the RWW flag for it.
>>>>>>
>>>>>> There is no public datasheet, but here are the sfdp tables plus base
>>>>>> testing to show it works.
>>>>>>
>>>>>> mx25uw51245g
>>>>>> c2813a
>>>>>> macronix
>>>>>> 53464450080104fd00070114400000ff8701011c900000ff0a0001080001
>>>>>> 00ff05000105200100ff84000102340100ff0000000000000000ffffffff
>>>>>> ffffffffe5208affffffff1f00ff00ff00ff00ffeeffffffffff00ffffff
>>>>>> 00ff0c2010d800ff00ff87790100821200e27704674630b030b0f4bdd55c
>>>>>> 000000ff101000200000000000007ca14800000000008888000000000000
>>>>>> 00400fd1fff30fd1fff300050090000500b1002b0095002b0096727103b8
>>>>>> 727103b80000000090a3188200c069960000000000000000727100987271
>>>>>> 00b8727100990000000072710098727100f872710099727100f900000000
>>>>>> 00000000011501d0727106d8000086500000060100000000020001030002
>>>>>> 00000000060100000000000072060002000000eec0697272717100d8f7f6
>>>>>> 000a00001445988043060f0021dcffff
>>>>>> 047a884cf44d9ffc2a94d3ab37b48c63  /sys/bus/spi/devices/spi0.0/spi-nor/sfdp  
>>>>>
>>>>> which hash alg was used, was it sha1? No need to resubmnit just for this.  
>>>>
>>>> Actually git just ignored all the lines starting with '#'
>>>> like if they were comments and I did not notice...
>>>>
>>>> I will amend my commit logs and resend.
>>>>  
>>>>>>
>>>>>> 6+0 records in
>>>>>> 6+0 records out
>>>>>> Copied 6291456 bytes from qspi_test to address 0x00000000 in flash
>>>>>> Erased 6291456 bytes from address 0x00000000 in flash
>>>>>> Copied 6291456 bytes from address 0x00000000 in flash to qspi_read
>>>>>> 0000000 ffff ffff ffff ffff ffff ffff ffff ffff
>>>>>> *
>>>>>> 0600000
>>>>>> Copied 6291456 bytes from qspi_test to address 0x00000000 in flash
>>>>>> Copied 6291456 bytes from address 0x00000000 in flash to qspi_read
>>>>>> d24a9523db829a0df688f34b8dc76a1383b74024  qspi_test
>>>>>> d24a9523db829a0df688f34b8dc76a1383b74024  qspi_read
>>>>>>
>>>>>> Signed-off-by: Miquel Raynal <miquel.raynal at bootlin.com>
>>>>>> ---
>>>>>>  drivers/mtd/spi-nor/macronix.c | 4 ++++
>>>>>>  1 file changed, 4 insertions(+)
>>>>>>
>>>>>> diff --git a/drivers/mtd/spi-nor/macronix.c b/drivers/mtd/spi-nor/macronix.c
>>>>>> index 6853ec9ae65d..b65d41e5f749 100644
>>>>>> --- a/drivers/mtd/spi-nor/macronix.c
>>>>>> +++ b/drivers/mtd/spi-nor/macronix.c
>>>>>> @@ -82,6 +82,10 @@ static const struct flash_info macronix_nor_parts[] = {
>>>>>>     { "mx25u51245g", INFO(0xc2253a, 0, 64 * 1024, 1024)
>>>>>>             NO_SFDP_FLAGS(SECT_4K | SPI_NOR_DUAL_READ | SPI_NOR_QUAD_READ)
>>>>>>             FIXUP_FLAGS(SPI_NOR_4B_OPCODES) },
>>>>>> +   { "mx25uw51245g", INFOB(0xc2813a, 0, 16 * 1024, 1024, 4)  
>>>>>
>>>>> Sector size and nsectors shall be discovered at parse SFDP time.
>>>>> Does INFOB(0xc2813a, 0, 0, 0, 4) work for you?  
>>>>
>>>> I'll check.
>>>>  
>>>>>  
>>>>>> +           PARSE_SFDP  
>>>>>
>>>>> Yaay
>>>>>  
>>>>>> +           FLAGS(SPI_NOR_RWW)
>>>>>> +           FIXUP_FLAGS(SPI_NOR_4B_OPCODES) },  
>>>>>
>>>>> Is SPI_NOR_4B_OPCODES flag really needed?  
>>>>
>>>> I have no idea, I took that from older code.
>>>>
>>>> Jaime, what do you think?  
>>> I think Tudor hopes that SFDP table could determine how many address bytes.  
>>
>> No, no.
>>
>> :) and hope and programming don't get along.
>>
>> Does this flash define the 4bait table? Can you add a print in
>> parse_4bait table and check if SNOR_F_4B_OPCODES is set? If yes, the
>> fixup flag is not needed.
>>
>> SNOR_F_4B_OPCODES is an indication that the flash supports dedicated
>> opcodes for 4byte address, without the need to change the address mode
>> to 4 byte. When 4B_OPCODES are supported, we prefer not changing the
>> address mode, because when changing to 4 byte address mode, the state is
>> kept at reset (or power loss) and flashes might not be able to boot at
>> lower layers (u-boot for example).
> 
> The flag is set, no need for hardcoding it.
> 
> Excellent, I'll prepare the patch.
> 
> While I was testing I figured out: the bank size calculation happens
> too early, so if we discover the size with the SFDP table, that won't
> work. I'll fix it, feel free to squash the patch if you prefer.
> 
Indeed. Michael or me are going to fix the n_sectors and sector_size
handling soon. Separate patch is fine, I don't change git history unless
really needed.