Re: [PATCH v2] spi: bcm63xx: fix premature CS deassertion on RX-only transactions

[Mark Brown]

On Tue, Nov 18, 2025 at 03:53:38AM +1100, Hang Zhou wrote:

> I initially picked 0xFF because in most case, the SPI bus is high when it's idle, but in practice 0x00 works just as well here.

> If you prefer, I'm happy to switch the dummy value to 0x00 for consistency and respin this as v3.

No, it's OK - if this hardware defaults to keeing the signal high then
0xff is a better choice.  Most hardware defaults to keeping the line
low.

Attachment: signature.asc
Description: PGP signature