[PATCH] RFC: mtd/spi-nor: add checking for the spi_nor_read

[Hou Zhiqiang]

> -----Original Message-----
> From: Bean Huo 霍斌斌 (beanhuo) [mailto:beanhuo at micron.com]
> Sent: 2015年8月13日 23:39
> To: Hou Zhiqiang-B48286; linux-mtd at lists.infradead.org;
> computersforpeace at gmail.com; dwmw2 at infradead.org;
> jonatas.rech at datacom.ind.br; Jane.Wan at gainspeed.com;
> shijie.huang at intel.com; valentin.longchamp at keymile.com;
> hkallweit1 at gmail.com; hramrach at gmail.com
> Cc: Hu Mingkai-B21284
> Subject: RE: [PATCH] RFC: mtd/spi-nor: add checking for the spi_nor_read
> 
> 
> Hi,
> Is it convenient to tell me that which vendor's spi nor?
> For current spi nor structure, write operation has been already split
> transaction length By page size(256 Bytes). But seems "retlen" not
> reasonable.
> 
> Spi nor read operation indeed does not take spi controller's transaction
> ability into Account.but for spi nor, one-time read length is non-
> restrictive.
> 

I'm sorry for my unclear explanation.
Yes, for SPI NOR Flash, the one-time read length is non-restrictive, but for the SPI
Controller there are some limits, e.g. 64KiB one-time transfer length limit for 
Freescale eSPI controller, and as you known the spi nor read operation indeed does
not take spi controller's transaction ability into account.

> --Beanhuo

B.R
Zhiqiang