Re: [PATCH] spi: sun6i: Set SPI mode in prepare_message

[Mark Brown <broonie@kernel.org>]

[-- Attachment #1: Type: text/plain, Size: 1362 bytes --]

On Wed, Apr 22, 2026 at 01:01:11PM -0600, Kevin Mehall wrote:

> SS_LEVEL (aka SUN6I_TFR_CTL_CS_LEVEL) resets to 0x1 (CS high = inactive)
> and is also written in sun6i_spi_set_cs(), which is called in spi_setup() via
> spi_set_cs(). Thus it should be initialized before we get here the first time.
> Once SUN6I_TFR_CTL_CS_MANUAL is set, it is never cleared elsewhere in the
> driver, so in any case, this can only affect the first transfer.

I do see that the driver uses a reset controller over suspend, are you
sure that setup() will be called again on resume?

> However, what's not clear to me is why SUN6I_TFR_CTL_CS_MANUAL was in
> sun6i_spi_transfer_one() in the first place. sun6i_spi_set_cs() is writing to
> the same register, and it seems to me that setting that bit there would be a
> more logical place to do it, though I don't think there is any functional change
> vs what I have here. Let me know if you'd like me to move it to
> sun6i_spi_set_cs() instead and how that should be submitted (same patch?
> separate patch before the rest of this in a series? standalone patch to be
> applied after this one?).

Smaller patches are generally better so splitting is good.  It does seem
sensible to group the chip select operations together, so long as we
ensure that everything else is stable before the chip select gets
asserted we should be good.

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 488 bytes --]