Question about SPI-NOR micron-st.c driver and OSPI devices.

Question about SPI-NOR micron-st.c driver and OSPI devices.

[Mike Williamson]

Hello,

I am looking at the drivers/mtd/spi-not/micron-st.c file.

I was wondering why the mt35xu512aba_fixups, and in general the patch below,
were not also applied to the mt35xu02g device entry?  Best I can tell, the
datasheets for the two parts (same family, different densities) look nearly
identical and the

https://www.mail-archive.com/linux-kernel@vger.kernel.org/msg2333941.html

I ask because we are having some troubles with the mt35xu02g and I am
wondering if we need these fixups to run reliably in octal mode.

Thanks,
Mike Williamson

______________________________________________________
Linux MTD discussion mailing list
http://lists.infradead.org/mailman/listinfo/linux-mtd/

Re: Question about SPI-NOR micron-st.c driver and OSPI devices.

[Pratyush Yadav]

Hi Mike,

On Mon, Jan 16 2023, Mike Williamson wrote:

> Hello,
>
> I am looking at the drivers/mtd/spi-not/micron-st.c file.
>
> I was wondering why the mt35xu512aba_fixups, and in general the patch below,
> were not also applied to the mt35xu02g device entry?  Best I can tell, the
> datasheets for the two parts (same family, different densities) look nearly
> identical and the
>
> https://www.mail-archive.com/linux-kernel@vger.kernel.org/msg2333941.html
>
> I ask because we are having some troubles with the mt35xu02g and I am
> wondering if we need these fixups to run reliably in octal mode.

When I wrote the patches for Octal DTR support on MT35 flash family, I
only had the mt35xu512aba flash on me. And it was a new feature so I did
not want to add it to a flash I could not test, and risk breaking it. I
figured if someone really wanted Octal DTR support on the 02G part, they
can populate the fixups and test it themselves.

So this is what I suggest you do as well. Add the fixups, test it out,
and if it works, upstream the patch. While you are at it, perhaps also
convert it to use the PARSE_SFDP flag. See db391efe765c ("mtd: spi-nor:
spansion: Remove NO_SFDP_FLAGS from s28hs512t info") or a30f53d8bc0f
("mtd: spi-nor: issi: is25wp256: Init flash based on SFDP") for example.

-- 
Regards,
Pratyush Yadav



Amazon Development Center Germany GmbH
Krausenstr. 38
10117 Berlin
Geschaeftsfuehrung: Christian Schlaeger, Jonathan Weiss
Eingetragen am Amtsgericht Charlottenburg unter HRB 149173 B
Sitz: Berlin
Ust-ID: DE 289 237 879




______________________________________________________
Linux MTD discussion mailing list
http://lists.infradead.org/mailman/listinfo/linux-mtd/

Re: Question about SPI-NOR micron-st.c driver and OSPI devices.

[Mike Williamson]

Hi Pratyush,

On Tue, Jan 24, 2023 at 7:46 AM Pratyush Yadav <ptyadav@amazon.de> wrote:
>
> Hi Mike,
>
> On Mon, Jan 16 2023, Mike Williamson wrote:
>
> > Hello,
> >
> > I am looking at the drivers/mtd/spi-not/micron-st.c file.
> >
> > I was wondering why the mt35xu512aba_fixups, and in general the patch below,
> > were not also applied to the mt35xu02g device entry?  Best I can tell, the
> > datasheets for the two parts (same family, different densities) look nearly
> > identical and the
> >
> > https://www.mail-archive.com/linux-kernel@vger.kernel.org/msg2333941.html
> >
> > I ask because we are having some troubles with the mt35xu02g and I am
> > wondering if we need these fixups to run reliably in octal mode.
>
> When I wrote the patches for Octal DTR support on MT35 flash family, I
> only had the mt35xu512aba flash on me. And it was a new feature so I did
> not want to add it to a flash I could not test, and risk breaking it. I
> figured if someone really wanted Octal DTR support on the 02G part, they
> can populate the fixups and test it themselves.
>
> So this is what I suggest you do as well. Add the fixups, test it out,
> and if it works, upstream the patch.

I have tested using the fixups on the 02G part successfully, but we are
working out of a 5.10 branch TI is providing for a new SOC.  I'm not sure
when I will be able to test against mainline/next as I don't know the state
of the SOC drivers (for TI's AM62x).  I know they are working to get them
in, but I don't think they are there yet.

Knowing the motivation for leaving the fixups out is helpful. I have a bit of
confidence that the patch is likely the right fix for Octal DTR support.

> ... While you are at it, perhaps also
> convert it to use the PARSE_SFDP flag. See db391efe765c ("mtd: spi-nor:
> spansion: Remove NO_SFDP_FLAGS from s28hs512t info") or a30f53d8bc0f
> ("mtd: spi-nor: issi: is25wp256: Init flash based on SFDP") for example.
>

I will have a look.

Thanks,
Mike

______________________________________________________
Linux MTD discussion mailing list
http://lists.infradead.org/mailman/listinfo/linux-mtd/

Re: Question about SPI-NOR micron-st.c driver and OSPI devices.

[Pratyush Yadav]

Hi Mike,

On Tue, Jan 24 2023, Mike Williamson wrote:

> Hi Pratyush,
>
> On Tue, Jan 24, 2023 at 7:46 AM Pratyush Yadav <ptyadav@amazon.de> wrote:
>>
>> Hi Mike,
>>
>> On Mon, Jan 16 2023, Mike Williamson wrote:
>>
>> > Hello,
>> >
>> > I am looking at the drivers/mtd/spi-not/micron-st.c file.
>> >
>> > I was wondering why the mt35xu512aba_fixups, and in general the patch below,
>> > were not also applied to the mt35xu02g device entry?  Best I can tell, the
>> > datasheets for the two parts (same family, different densities) look nearly
>> > identical and the
>> >
>> > https://www.mail-archive.com/linux-kernel@vger.kernel.org/msg2333941.html
>> >
>> > I ask because we are having some troubles with the mt35xu02g and I am
>> > wondering if we need these fixups to run reliably in octal mode.
>>
>> When I wrote the patches for Octal DTR support on MT35 flash family, I
>> only had the mt35xu512aba flash on me. And it was a new feature so I did
>> not want to add it to a flash I could not test, and risk breaking it. I
>> figured if someone really wanted Octal DTR support on the 02G part, they
>> can populate the fixups and test it themselves.
>>
>> So this is what I suggest you do as well. Add the fixups, test it out,
>> and if it works, upstream the patch.
>
> I have tested using the fixups on the 02G part successfully, but we are
> working out of a 5.10 branch TI is providing for a new SOC.  I'm not sure
> when I will be able to test against mainline/next as I don't know the state
> of the SOC drivers (for TI's AM62x).  I know they are working to get them
> in, but I don't think they are there yet.

I think base support for AM62x has been in mainline since a033588ec66b
("arm64: dts: ti: Add support for AM62-SK"), which landed in v5.18. It
might not have all the devices/drivers supported yet but I think SPI NOR
should at least work. So you _should_ be able to boot mainline kernel on
AM62x SK, and you should be able to develop patches for mainline.

But if you only want support for this device on TI's tree then you need to
contact TI for that.

>
> Knowing the motivation for leaving the fixups out is helpful. I have a bit of
> confidence that the patch is likely the right fix for Octal DTR support.
>
>> ... While you are at it, perhaps also
>> convert it to use the PARSE_SFDP flag. See db391efe765c ("mtd: spi-nor:
>> spansion: Remove NO_SFDP_FLAGS from s28hs512t info") or a30f53d8bc0f
>> ("mtd: spi-nor: issi: is25wp256: Init flash based on SFDP") for example.
>>
>
> I will have a look.

-- 
Regards,
Pratyush Yadav



Amazon Development Center Germany GmbH
Krausenstr. 38
10117 Berlin
Geschaeftsfuehrung: Christian Schlaeger, Jonathan Weiss
Eingetragen am Amtsgericht Charlottenburg unter HRB 149173 B
Sitz: Berlin
Ust-ID: DE 289 237 879




______________________________________________________
Linux MTD discussion mailing list
http://lists.infradead.org/mailman/listinfo/linux-mtd/