Re: [PATCH] custom PM support for 8250

[Vitaly Wool <vitalhome@rbcmail.ru>]

Yes, that's also something I was thinking of, but I wasn't sure enough 
in this radical change :)
Anyway, I do agree that this way looks better than the current one.
So, if you don't object against other changes (say, the suggested 
approach to set uart->pm), I can proceed with the changes you suggest 
and work it out to the new patch. ;)

Best regards,
   Vitaly

Russell King wrote:

>On Wed, Aug 31, 2005 at 03:10:09PM +0400, Vitaly Wool wrote:
>  
>
>>please find the patch that allows passing the pointer to custom power 
>>management routine (via platform_device) to 8250 serial driver.
>>Please note that the interface to the outer world (i. e. exported 
>>functions) remained the same.
>>    
>>
>
>I'd rather change the structure passed via the platform device to
>something like:
>
>struct platform_serial_data {
>	void	(*pm)(struct uart_port *port, unsigned int state, unsigned int old);
>	int	nr_ports;
>	struct plat_serial8250_port *ports;
>};
>
>which also eliminates the empty plat_serial8250_port terminator from
>all the serial8250 platform devices (which appears to have caused some
>folk problems.)
>
>It does mean that a set of 8250 ports (grouped by each platform device)
>have a common power management method - which seems a logical restriction.
>
>  
>