[PATCH v4 2/3] mtd: mtk-nor: mtk serial flash controller driver

Brian Norris computersforpeace at gmail.com
Thu Oct 29 22:02:49 PDT 2015


Hi Bayi,

On Fri, Oct 30, 2015 at 10:12:39AM +0800, bayi.cheng wrote:
> Hi Brian,  The current station is as follows.
> 
> 1: put 0x9F to MTK_NOR_PRGDATA5_REG, and five 0x0 to
> MTK_NOR_PRGDATA4_REG ~ MTK_NOR_PRGDATA0_REG, then set (1 + 5) * 8 
> to MTK_NOR_CNT_REG, for this way, we can read five IDs.
> 
> 2: put 0x9F to MTK_NOR_PRGDATA5_REG, and five 0x0 to
> MTK_NOR_PRGDATA4_REG ~ MTK_NOR_PRGDATA0_REG, then set (1 + 5 + 1) * 8
> to MTK_NOR_CNT_REG, for this way, we can read six IDs.
> In this case, nor flash IDs can be read from MTK_NOR_SHREG5_REG to
> MTK_NOR_SHREG0_REG . Thanks!

Thanks for the update. Glad to hear you can read more bytes there; the
extra number of Shift Registers *looked* to me like you should be able
to read even more than 5...

So in my other email, I showed you how I generalized the TX/RX function,
so it can handle most of both the write_reg() and read_reg() functions.
I'm not 100% sure now that it has all of the RX path correct; it worked
for the cases I could test, but it's confusing when reading the manual
to figure out which SHREG register I should start from. But anyway, I
think it should only take some small adjustments to my patch to make it
handle things properly.

I'd really appreciate it if you could incorporate my feedback and
review/improve the ..._do_tx_rx() function I wrote, to make sure it
handles reading any arbitrary number of bytes (at least up to 6). So,
you might, for example, run some tests where you have spi-nor.c call

	nor->read_reg(nor, SPINOR_OP_RDID, id, idlen)

with varying values for 'idlen', and make sure they all work properly.

Let me know if you have any questions about comments either here, or on
v5.

Regards,
Brian



More information about the linux-arm-kernel mailing list