On Thu, Sep 11, 2014 at 3:45 AM, Boaz Harrosh <boaz@xxxxxxxxxxxxx> wrote: > On 09/11/2014 02:01 AM, Dan Williams wrote: > <> >> >> Imagine you want to deploy a policy like "use half of the memory >> provided by the dimm in slot3, i.e. the only one with a battery". >> That sort of thing gets unwieldy in a command line string compared to >> a description table format that we can update at will. >> > > Actually it is easy to do, why? I do this here in the lab all the time. > with a "command line" with this code you see here. > > [DDR3 NvDIMM which means I need memmap=16G\$32G on Kernel command line. > Then: modprobe pmem map=8G@32G,4G@44G,... > and so on Just as a simple example where 2/4-3/4 addresses are not used. > You can have holes in the middle or what ever you want. This here is just > a table in comma-separated format. If we need like flags in future we can > extend the format to nn@ss:flags, but I do no have any 3rd column yet] The point I am getting at is not requiring a priori knowledge of the physical memory map of a system. Rather, place holder variables to enable simple dynamic discovery. > And again I have in the pipe a dynamic interface added on top of > the module param one. So it will all be there soon. Without reverting > the old one. Why start on step 2 when we haven't got agreement on step 1? -- To unsubscribe from this list: send the line "unsubscribe linux-fsdevel" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
