On Fri, Oct 20, 2017 at 10:59:45PM +0900, Masahiro Yamada wrote: > 2017-10-20 22:54 GMT+09:00 Greg KH <[email protected]>: > > On Fri, Oct 20, 2017 at 10:47:28PM +0900, Masahiro Yamada wrote: > >> 2017-10-20 22:34 GMT+09:00 Greg KH <[email protected]>: > >> > On Mon, Oct 09, 2017 at 03:26:31PM +0200, [email protected] > >> > wrote: > >> >> From: Masahiro Yamada <[email protected]> > >> >> > >> >> nvmem_register() copies all the members of nvmem_config to > >> >> nvmem_device. So, nvmem_config is one-time use data during > >> >> probing. There is no point to keep it until the driver detach. > >> >> Using stack should be no problem because nvmem_config is pretty > >> >> small. > >> > > >> > Same objection as previous patch, what is wrong with it as-is? > >> > > >> > >> It is wasting memory. > > > > The memory is freed again, after the call, right? > > > > I assume your "as-is" meant devm_kzalloc(). > > The memory is freed when the driver is detached. > > It is generally quite long time > that this struct occupies the memory for no good reason.
Ah, ok, a simple kmalloc() might be better, but ok, if you are _sure_ it's ok taking that much stack space up... :) thanks, greg k-h
