Re: [PATCH 6/6] setup.c: use early_param() for early command line parsing

[Franck Bui-Huu <vagabon.xyz@gmail.com>]

Thiemo Seufer wrote:
> Franck Bui-Huu wrote:
>> Thiemo Seufer wrote:
>>> Franck Bui-Huu wrote:
>>>> There's no point to rewrite some logic to parse command line
>>>> to pass initrd parameters or to declare a user memory area.
>>>> We could use instead parse_early_param() that does the same
>>>> thing.
>>>>
>>>> NOTE ! This patch also changes the initrd semantic. Old code
>>>> was expecting "rd_start=xxx rd_size=xxx" which uses two
>>>> parameters. Now the code expects "initrd=xxx@yyy" which is
>>>> really simpler to parse and to use. No default config files
>>>> use these parameters anyways but not sure for bootloader's
>>>> users...
>>> This code is there precisely because most mips bootloaders use
>>> rd_start/rd_size.
>> OK, I guess we have to stick with this weird semantic...
>>
>>> It also is IMHO a bad idea to overload the
>>> semantics of initrd= with both file names and memory locations.
>> I wasn't aware of any file name usages. Can you give a pointer ?
> 
> Documentation/initrd.txt
> Documentation/filesystems/ramfs-rootfs-initramfs.txt
> 

I was asking for pointers on MIPS bootloaders which use
initrd=/path/to/initrd...

Anyways, you're talking about specific bootloader's parameters,
aren't you ? I don't know any MIPS bootloaders, but I wouldn't 
expect them to pass their own parameters to the kernel, that 
would be surprising...

What are you suggesting ? kernel_initrd ?

BTW, what do you think about rd_start/rd_size names ?

thanks

                Franck