Re: [PATCH] mtd: spi-nor: add support for Microchip 25LC256

From: Marek Vasut <marek.vasut@gmail.com>
To: Radu Pirea <radu.pirea@microchip.com>,
	Boris Brezillon <boris.brezillon@bootlin.com>
Cc: linux-kernel@vger.kernel.org, linux-mtd@lists.infradead.org,
	Zhiqiang.Hou@nxp.com, pp@emlix.com, richard@nod.at,
	computersforpeace@gmail.com, dwmw2@infradead.org
Subject: Re: [PATCH] mtd: spi-nor: add support for Microchip 25LC256
Date: Fri, 18 May 2018 18:25:31 +0200	[thread overview]
Message-ID: <de2c119b-36f5-a219-7a9a-65668c61353c@gmail.com> (raw)
In-Reply-To: <ba6be822-b8ac-588f-ec11-d7cb5cddd1a3@microchip.com>

On 05/18/2018 03:00 PM, Radu Pirea wrote:
> 
> 
> On 05/18/2018 01:03 PM, Marek Vasut wrote:
>> On 05/18/2018 11:50 AM, Radu Pirea wrote:
>>>
>>>
>>> On 05/16/2018 04:47 PM, Marek Vasut wrote:
>>>> On 05/16/2018 12:05 PM, Radu Pirea wrote:
>>>>> On Wed, 2018-05-16 at 00:17 +0200, Marek Vasut wrote:
>>>>>> On 05/15/2018 06:22 PM, Radu Pirea wrote:
>>>>>>> On Fri, 2018-05-04 at 20:40 +0200, Boris Brezillon wrote:
>>>>>>>> On Fri, 4 May 2018 18:54:04 +0300
>>>>>>>> Radu Pirea <radu.pirea@microchip.com> wrote:
>>>>>>>>
>>>>>>>>> Added geometry description for Microchip 25LC256 memory.
>>>>>>>>
>>>>>>>> Same as for the dataflash stuff you posted a few weeks ago: I
>>>>>>>> don't
>>>>>>>> think this device belongs in the SPI NOR framework.
>>>>>>>
>>>>>>> Hi Boris,
>>>>>>>
>>>>>>> 25lc256 memory is similar with mr25h256, the only difference is the
>>>>>>> page size(64 vs 256). Because mr25h256 is already in SPI NOR
>>>>>>> framework
>>>>>>> I added here 25lc256.
>>>>>>
>>>>>> I think I must be reading the wrong datasheet, but can you show me
>>>>>> how
>>>>>> does it support things like READID opcode ?
>>>>>>
>>>>> Hi Marek,
>>>>>
>>>>> I read the datasheet for 25lc256 and for mr25h256 and none of them
>>>>> supports READID. Is this required for a chip to be included in spi-nor
>>>>> framework? I just followed the mr25h256 as an example.
>>>>
>>>> So I thought until you pointed out the MR25 devices.
>>>>
>>>> Does the 25LC device need erase or not ? I think the MR25s didn't,
>>>> but I
>>>> might be wrong.
>>>
>>> You are right. MR25s does not need erase and the same thing is true for
>>> 25LC.
>>
>> Oh. And the command set is (except for readid) comparable to SPI NOR ?
>>
>>>> Maybe the framework could support the 25LC afterall.
>>>>
>>> Yes, this was my impression too.
>>
>> I am still thinking that the AT25 driver might be a better fit for such
>> devices. Can you take a look ?
> 
> I tested the memory with at25 driver and it works with no line changed
> in driver. :)

So AT25 it is then ?

-- 
Best regards,
Marek Vasut