Re: [PATCH v3 1/2] mtd: spi-nor: add support to non-uniform SFDP SPI NOR flash memories

[Boris Brezillon <boris.brezillon@bootlin.com>]

On Mon, 22 Oct 2018 12:46:27 +0200
Boris Brezillon <boris.brezillon@bootlin.com> wrote:

> On Mon, 22 Oct 2018 10:39:48 +0000
> Yogesh Narayan Gaur <yogeshnarayan.gaur@nxp.com> wrote:
> 
> > 
> > [    1.632190] Start [addr_width:00000000, read_dumy:08, read_opcode:00000000]                                                   
> > [    1.639148] spi_nor_get_map_in_use:2882 smpt[0]=08ff65fc                                                                      
> > [    1.644451] spi_nor_get_map_in_use:2882 smpt[1]=00000004                                                                      
> > [    1.649755] spi_nor_get_map_in_use:2882 smpt[2]=04ff65fc                                                                      
> > [    1.655057] spi_nor_get_map_in_use:2882 smpt[3]=00000002                                                                      
> > [    1.660360] spi_nor_get_map_in_use:2882 smpt[4]=02ff65fd                                                                      
> > [    1.665662] spi_nor_get_map_in_use:2882 smpt[5]=00000004                                                                      
> > [    1.670965] spi_nor_get_map_in_use:2882 smpt[6]=ff0201fe                                                                      
> > [    1.676267] spi_nor_get_map_in_use:2882 smpt[7]=00007ff1                                                                      
> > [    1.681571] spi_nor_get_map_in_use:2882 smpt[8]=00037ff4                                                                      
> > [    1.686874] spi_nor_get_map_in_use:2882 smpt[9]=03fbfff4                                                                      
> > [    1.692176] spi_nor_get_map_in_use:2882 smpt[10]=ff0203fe                                                                     
> > [    1.697566] spi_nor_get_map_in_use:2882 smpt[11]=03fbfff4                                                                     
> > [    1.702954] spi_nor_get_map_in_use:2882 smpt[12]=00037ff4                                                                     
> > [    1.708343] spi_nor_get_map_in_use:2882 smpt[13]=00007ff1                                                                     
> > [    1.713732] spi_nor_get_map_in_use:2882 smpt[14]=ff0005ff                                                                     
> > [    1.719120] spi_nor_get_map_in_use:2882 smpt[15]=03fffff4                                                                     
> > [    1.724509] smpt[0]=[addr_width:00000000, read_dumy:08, read_opcode:00000065]                                                 
> > [    1.731650] smpt[1]=[addr_width:00000000, read_dumy:08, read_opcode:00000000]                                                 
> > [    1.738791] smpt[2]=[addr_width:00000000, read_dumy:08, read_opcode:00000065]     
> 
> You still don't print read_dummy correctly: %0x8 -> %08x.
> 
> Can you add
> 
> 	if (!nor->addr_width)
> 		nor->addr_width = 3;
> 
> After the
> 
> 	nor->addr_width = spi_nor_smpt_addr_width(nor, smpt[i]);
> 
> line.

And you should also try to force ->read_dummy to 8, because according
to the spec, the default read_latency is 8 for this chip. With that in
place, you should get an map_id of 1, 3 or 5.

>                                               
> > [    1.745932] spi_nor_get_map_in_use:2911 map_id=0 smpt_len:16 i=:3                                                             
> > [    1.752018] End [addr_width:00000000, read_dumy:08, read_opcode:00000065]         
> > 
> > Also, one more thing when we are returning from the function, we are over-writing the values of addr_widthm read_dummy and read_opcode.
> > Is this correct?  
> 
> Yes, that's correct.
> 
> > 
> > out:
> >         nor->addr_width = addr_width;
> >         nor->read_dummy = read_dummy;
> >         nor->read_opcode = read_opcode;
> >         return ret;
> > }
> >   
> 
>