Re: [PATCH] kernel/timekeeping: move xtime_cache to be in the same cache line as the lock

[Richard Kennedy <richard@rsk.demon.co.uk>]

On Tue, 2010-01-26 at 15:28 -0800, Andrew Morton wrote:
> On Thu, 21 Jan 2010 15:39:21 +0000
> Richard Kennedy <richard@rsk.demon.co.uk> wrote:
> 
> > move xtime_cache to be in the same cache line as the lock
> >     
> > allowing current_kernel_time() to access only one cache line
> 
> Sentences start with capital letters, please.

Sorry about that, I will try harder in future ;)


> 
> I don't know how reliable this is.  I _think_ the compiler and linker
> are free to place variables of this nature in any old place.  Whether
> any of the current tools actually do that I don't know.  Note that one
> of these variables has file-static scope and the other does not, which
> perhaps increases the risk that the compiler or linker will go and
> fiddle with them.
> 
> To do this reliably one would need to put them in a struct:
> 
> time.h:
> 
> extern struct xtime_stuff {
> 	seqlock_t _xtime_lock,
> 	struct timespec _xtime_cache,
> } xtime_stuff;
> 
> #define xtime_lock xtime_stuff._xtime_lock
> 
> 
> timekeeping.c:
> 
> struct xtime_stuff {
> 	._xtime_lock = __SEQLOCK_UNLOCKED(xtime_stuff._xtime_lock),
> };
Thank you, yes that looks like a much better approach.
I can do this if it's needed, but John Stultz said he's going to kill
the xtime_cache anyway, so it may not be worth it?

However I do wonder if we should move all, or at least some, of the
variables protected by that xtime_lock into that structure? Then we can
manage their placement and they would be easier to find. After only a
brief look I see variables in ntp, tick &  timekeeping that seem to be
protected by that seqlock.

> > BTW on 64 bit timespec is a 16 byte structure so the aligned 16 doesn't
> > do much, and on 32bit timepec is 8bytes so this just seems to spread
> > these variables across more cache lines than necessary. Any ideas what
> > this is here for?
> 
> Dunno.  I had a bit of a peek in the git history but it got complicated
> and people rarely bother explaining things like this anyway :(
> 
regards
Richard