Hi!
> We are still discussing some of the details of the message formats, and
> what we are coming up with definitely does NOT look like hwat you are
> proposing here.
Really? We want to stick structure there. But that can not be done
with one step... patch would be too big and too disruptive.
I introduced typedef there. Once typedef is introduced, switching to
structure *can* be done with one patch.
> More specifically, this list:
>
> +#define PMSG_APM_SUSPEND ((__force pm_message_t) 3)
> +#define PMSG_APM_STANDBY ((__force pm_message_t) 3)
> +#define PMSG_SYSTEM_HALT ((__force pm_message_t) 3)
> +#define PMSG_SYSTEM_REBOOT ((__force pm_message_t) 3)
> +#define PMSG_SYSTEM_SHUTDOWN ((__force pm_message_t) 3)
> +#define PMSG_SUSPEND_RAM ((__force pm_message_t) 3)
> +#define PMSG_SUSPEND_DISK ((__force pm_message_t) 3)
> +#define PMSG_DEV_DETACH ((__force pm_message_t) 3)
> +#define PMSG_ON ((__force pm_message_t) 0)
>
> Is just not useful at this point. We decided we would _NOT_ pass the
> system state down to devices (STD, STR, ...) but an abstract
> FREEZE/IDLE/SUSPEND message, where is that reflected in the above ?
I assumed that I'd go to
#define PMSG_APM_SUSPEND { PM_SUSPEND, FL_APM | FL_SUSPEND_TO_RAM }
in future.
You keep saying that patches are easy; I do not agree at this
point. It will need to be series of patches that do not break the
tree...
Pavel
--
People were complaining that M$ turns users into beta-testers...
...jr ghea gurz vagb qrirybcref, naq gurl frrz gb yvxr vg gung jnl!
