Re: [RFC PATCH 1/4] i2c: allow drivers to announce that they are IRQ safe

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



On Mon, Aug 20, 2018 at 07:54:30AM -0700, Tony Lindgren wrote:
> Hi,
> 
> * Stefan Lengfeld <contact@xxxxxxxxxxxxxxx> [180813 21:19]:
> > Making I2C transfers in atomic context is sometimes needed, e.g.  for
> > reboot handlers.
> 
> Just a generic comment related to where this might be needed.
> Let's try to avoid the slippery slope of making i2c work super
> early.. That will just make the race to the bottom initcall
> level and defferd probe situation worse :)
> 
> It's best to initialize everything late instead of trying to
> initialize something early except for what's needed for system
> timers and interrupts.
> 
> So from that point of view I don't have any issues limiting
> atomic i2c to shutdown time only.

I buy this.

There was only one occasion I recall where early, irqless I2C access was
desired and I can't even find the discussion now. The use case which
came back repeatedly is shutdown/reboot via PMIC.

Let's stick to that unless someone speaks up now having a good reason
for the early case.

Attachment: signature.asc
Description: PGP signature


[Index of Archives]     [Linux Arm (vger)]     [ARM Kernel]     [ARM MSM]     [Linux Tegra]     [Linux WPAN Networking]     [Linux Wireless Networking]     [Maemo Users]     [Linux USB Devel]     [Video for Linux]     [Linux Audio Users]     [Yosemite Trails]     [Linux Kernel]     [Linux SCSI]

  Powered by Linux