On Wed, Aug 4, 2021 at 4:28 PM Richard Cochran [off-list ref] wrote:
quoted
@@ -87,8 +87,8 @@ config E1000E_HWTS
config IGB
tristate "Intel(R) 82575/82576 PCI-Express Gigabit Ethernet support"
depends on PCI
- imply PTP_1588_CLOCK
- select I2C
+ depends on PTP_1588_CLOCK_OPTIONAL
+ depends on I2C
This little i2c bit sneaks in, but I guess you considered any possible
trouble with it?
Good catch!
I did need this with v2, as it was causing a circular dependency against
(IIRC) CONFIG_MLXSW_I2C, but I'm fairly sure it's not needed any
more after everything else uses 'depends on' now.
I'm happy to resend a v4 without that change, as it doesn't belong in here,
or we just leave it because it is correct after all, depending on what the Intel
ethernet people prefer.
Acked-by: Richard Cochran <richardcochran@gmail.com>
Thanks,
Arnd