Re: [PATCH linux-next v6 4/8] mtd: spi-nor: use optimized commands for read/write/erase operations

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 




On Wed, Sep 9, 2015 at 3:24 PM, Cyrille Pitchen
<cyrille.pitchen@xxxxxxxxx> wrote:
> The op codes used by the spi-nor framework are now tuned depending on the
> memory manufacturer.
>
> Signed-off-by: Cyrille Pitchen <cyrille.pitchen@xxxxxxxxx>

To be honest, I'm not sure if it's worth it using the dual/quad I/O
commands on spansion, as they have different dummy cycle requirements
than the fast/dual/quad reads.

E.g. on S25FL128S, Fast/dual/quad is fine with 8 dummy cycles up to
104 MHz, but Quad I/O needs 1 cycle for <= 50 MHz, 4 for <= 90 MHz,
and 5 for <= 104 MHz. Which means we would need a lot more logic in
there or setting it from dts, and for existing users already using
dual/quad mode this would break it. And for m25p80 it would mean to
split the cmd + address into two transfers, all to save 20~30 clock
cycles, which will likely be dwarfed by the savings of the
dual/quadness of the data.

Apart from the fact that I would assume many spi-controllers will have
problems with dummy cycles that aren't a multiple of 8, and would
therefore require bitshifting the data when reading, which will
definitely eat up any time savings from the dual/quad written address.

And for the quad program command the datasheet says it only has any
performance improvement if running at less than 12 MHz, so probably
not much to gain for most users.


Jonas
--
To unsubscribe from this list: send the line "unsubscribe devicetree" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html



[Index of Archives]     [Device Tree Compilter]     [Device Tree Spec]     [Linux Driver Backports]     [Video for Linux]     [Linux USB Devel]     [Linux PCI Devel]     [Linux Audio Users]     [Linux Kernel]     [Linux SCSI]     [XFree86]     [Yosemite Backpacking]
  Powered by Linux