Re: [patch] uninline init_waitqueue_*() functions

[Ingo Molnar <mingo@elte.hu>]

* Andrew Morton <akpm@osdl.org> wrote:

> > > shrinks fs/select.o by eight bytes.  (More than I expected).  So 
> > > it does appear to be a space win, but a pretty slim one.
> > 
> > there are 855 calls to these functions in the allyesconfig vmlinux i 
> > did, and i measured a combined size reduction of 34791 bytes. That 
> > averages to a 40 bytes win per call site. (on i386.)
> > 
> 
> Yes, but that lumps all three together.  init_waitqueue_head() is 
> obviously the porky one.  And it's porkier with CONFIG_DEBUG_SPINLOCK 
> and CONFIG_LOCKDEP, which isn't the case to optimise for.

true. I redid my tests with both lockdep and debug-spinlocks turned off:

  text            data    bss     dec             filename
  21172153        6077270 3081864 30331287        vmlinux.x32.after
  21198222        6077106 3081864 30357192        vmlinux.x32.before

with 851 callsites that's a 30.6 bytes win per call site (total 26K) - 
still not bad at all.

it's a win even on 64-bit:

  text            data    bss     dec             filename
  21237025        6997266 3327600 31561891        vmlinux.x64.after
  21252773        6997090 3327600 31577463        vmlinux.x64.before

with 755 callsites that's still a 20.8 bytes win per call site (total 
15K).

> With the debug options turned off, even init_waitqueue_head() becomes 
> just three assignments, similar to init_waitqueue_entry() and 
> init_waitqueue_func_entry().  All pretty marginal.

but three assignments could mean 3 offsets embedded in the instructions. 
(for init_waitqueue_entry we also embedd the address of 
default_wake_function) Even 3 assignments can add up to a footprint that 
is far from marginal.

The rough rule of thumb for inlining is that anything that is larger 
than one C statement is probably too large for inlining. (but even 
1-line statements might be too fat at times)

	Ingo