Re: [PATCH] mtd: rawnand: qcom: update last code word register

[mdalam]

On 2021-02-24 22:06, Miquel Raynal wrote:
> Hello,
>
> mdalam@xxxxxxxxxxxxxx wrote on Wed, 24 Feb 2021 22:00:05 +0530:
>
> > On 2021-02-24 12:18, Miquel Raynal wrote:
> > > Hello,
> > >
> > > mdalam@xxxxxxxxxxxxxx wrote on Wed, 24 Feb 2021 10:09:48 +0530:
> > >
> > >> On 2021-02-24 01:13, mdalam@xxxxxxxxxxxxxx wrote:
> > >> > On 2021-02-23 22:04, Miquel Raynal wrote:
> > >> >> Hello,
> > >> >> >> Md Sadre Alam <mdalam@xxxxxxxxxxxxxx> wrote on Tue, 23 Feb 2021
> > >> >> 01:34:27 +0530:
> > >> >> >>> From QPIC version 2.0 onwards new register got added to read last
> > >> >> >>                                a new
> > >> >> >>> codeword. This change will add the READ_LOCATION_LAST_CW_n register.
> > >> >> >>             Add support for this READ_LOCATION_LAST_CW_n register.
> > >> >> >>> >>> For first three code word READ_LOCATION_n register will be
> > >> >>> use.For last code word READ_LOCATION_LAST_CW_n register will be
> > >> >>> use.
> > >> >> >> "
> > >> >> In the case of QPIC v2, codewords 0, 1 and 2 will be accessed through
> > >> >> READ_LOCATION_n, while codeword 3 will be accessed through
> > >> >> READ_LOCATION_LAST_CW_n.
> > >> >> "
> > >> >> >> When I read my own sentence, I feel that there is something wrong.
> > >> >> If there are only 4 codewords, I guess a QPIC v2 is able to use
> > >> >> READ_LOCATION_3 or READ_LOCATION_LAST_CW_0 interchangeably. Isn't it?
> > >> >> >> I guess the point of having these "last_cw_n" registers is to support
> > >> >> up to 8 codewords, am I wrong? If this the case, the current patch
> > >> >> completely fails doing that I don't get the point of such change.
> > >> >
> > >> > This register is only use to read last code word.
> > >> >
> > >> > I have address all the comments from all the previous sub sequent
> > >> > patches and pushed
> > >> > all patches in only one series.
> > >> >
> > >> > Please check.
> > >> >>   The registers READ_LOCATION & READ_LOCATION_LAST are not associated >> with number of code words.
> > >>   These two registers are used to access the location inside a code >> word.
> > >
> > > Ok. Can you please explain what is a location then? Or point me to a
> > > datasheet that explains it.
> >
> >    The location is the position inside a code word.
> >
> > >
> > > Bottom line question: why having READ_LOCATION_0, _1,... an
> > > READ_LOCATION_LAST_0, _1, etc?
> >
> >   READ_LOCATION_0, _1,... are used to extract multiple chunks from a 
> > code word.
> >
> >   e.g If we wanted to extract first 100 bytes from a code word then 
> > (0...99) READ_LOCATION_0 will be configured.
> >       if we wanted to extract next 100 bytes (100...199) then 
> > READ_LOCATION_1 will be configured.
> >
> >       same way for last code word READ_LOCATION_LAST_0, _1, will be 
> > used.
>
> Nice explanation, and thanks for the below figures. So I guess there
> is some kind of "small SRAM" that is
> directly addressable perhaps?
>
> I think I'm fine with your series now. Just a small nit: next time you
> send a series, please update the version number "[PATCH v6]"
> (automatically added with the -v6 parameter in git-format-patch). But
> no need to resend just for that.

  Thanks Miquel. So now no need to test these patches further. I have 
already tested these patches on IPQ5018 SoC with mtd_test module & 
nand-utils tool.

> Thanks,
> Miquèl