Re: [PATCH] mtd: spinand: Add support for GigaDevice GD5F1GQ4UFxxG

[Miquel Raynal <miquel.raynal@bootlin.com>]

Hi Schrempf,

Schrempf Frieder <frieder.schrempf@kontron.de> wrote on Tue, 14 May
2019 16:11:28 +0000:

> Hi Jeff,
> 
> On 14.05.19 17:42, Jeff Kletsky wrote:
> > On 5/13/19 6:56 AM, Schrempf Frieder wrote:
> >   
> >> Hi Jeff,
> >>
> >> I just noticed I hit the wrong button and my previous reply was only
> >> sent to the MTD list, so I'm resending with fixed recipients...
> >>
> >> On 10.05.19 14:17,lede@allycomm.com  wrote:  
> >>> From: Jeff Kletsky<git-commits@allycomm.com>
> >>>
> >>> The GigaDevice GD5F1GQ4UFxxG SPI NAND is in current production devices
> >>> and, while it has the same logical layout as the E-series devices,
> >>> it differs in the SPI interfacing in significant ways.
> >>>
> >>> To accommodate these changes, this patch also:
> >>>
> >>>     * Adds support for two-byte manufacturer IDs
> >>>     * Adds #define-s for three-byte addressing for read ops
> >>>
> >>> http://www.gigadevice.com/datasheet/gd5f1gq4xfxxg/
> >>>
> >>> Signed-off-by: Jeff Kletsky<git-commits@allycomm.com>  
> >> Maybe it would be better to split this patch into three parts:
> >> * Add support for two-byte device IDs
> >> * Add #define-s for three-byte addressing for read ops
> >> * Add support for GD5F1GQ4UFxxG
> >>
> >> Anyway the content looks good to me, so:
> >>
> >> Reviewed-by: Frieder Schrempf<frieder.schrempf@kontron.de>
> >>
> >> [...]  
> > 
> > Thanks for the time in review and good words!  
> 
> You're welcome!
> 
> > My apologies for an incomplete git-send-email config that left
> > me nameless in the headers.  
> 
> No problem, I guessed your name from the Signed-off-by tag ;)
> 
> > I wasn't sure if that was direction to submit as three patches
> > at this time, but would be happy to do so if the consensus is
> > that it the direction to follow.  
> 
> I think it's common to separate logical different changes. This makes it 
> easier to read.
> Also the preparation changes only touch the SPI NAND core. I guess 
> that's another reason why they should be separated from the 
> chip-specific changes.
> 
> > At least for me, I feel that the other two don't really stand
> > on their own without the context for their need.  
> 
> I don't think that's a problem. Just add a note to the commit message 
> that these core changes are needed to prepare for the GD5F1GQ4UFxxG support.
> 
> Thanks,
> Frieder

I agree with Frieder, if you don't mind, please split this commit in
three.

Thanks,
Miquèl