mirror of https://lore.kernel.org/lkml/
 help / color / mirror / Atom feed
* [PATCH][0/8] Arch agnostic completely out of line locks
@ 2004-09-03  0:02 Zwane Mwaikambo
  2004-09-03 14:48 ` Martin J. Bligh
  0 siblings, 1 reply; 3+ messages in thread
From: Zwane Mwaikambo @ 2004-09-03  0:02 UTC (permalink / raw)
  To: Linux Kernel
  Cc: Linus Torvalds, Andrew Morton, Matt Mackall,
	William Lee Irwin III, Keith Owens

This patch achieves out of line spinlocks by creating kernel/spinlock.c
and using the _raw_* inline locking functions. Now, as much as this is
supposed to be arch agnostic, there was still a fair amount of rummaging
about in archs, mostly for the cases where the arch already has out of
line locks and i wanted to avoid the extra call, saving that extra call
also makes lock profiling easier. PPC32/64 was an example of such an arch
and i have added the necessary profile_pc() function as an example.

Size differences are with CONFIG_PREEMPT enabled since we wanted to
determine how much could be saved by moving that lot out of line too.

ppc64 = 259897 bytes:
   text    data     bss     dec     hex filename
5489808 1962724  709064 8161596  7c893c vmlinux-after
5749577 1962852  709064 8421493  808075 vmlinux-before

sparc64 = 193368 bytes:
  text    data     bss     dec     hex filename
3472037  633712  308920 4414669  435ccd vmlinux-after
3665285  633832  308920 4608037  465025 vmlinux-before

i386 = 524115 bytes
   text    data     bss     dec     hex filename
5695619  870906  328112 6894637  69342d vmlinux-after
6221254  870634  326864 7418752  713380 vmlinux-before

x86-64 = 282446 bytes
   text    data     bss     dec     hex filename
4598025 1450644  523632 6572301  64490d vmlinux-after
4881679 1449436  523632 6854747  68985b vmlinux-before

It has been compile tested (UP, SMP, PREEMPT) on i386, x86-64, sparc,
sparc64, ppc64, ppc32 and runtime tested on i386 and x86-64. I still have
to get benchmarks done (most probably i386). The patch is currently
against 2.6.9-rc1-mm1 because we'd have to back out the i386/x86_64 out of
line lock patches since this does it in a different way, i can merge later
on.

Thanks,
	Zwane

^ permalink raw reply	[flat|nested] 3+ messages in thread

* Re: [PATCH][0/8] Arch agnostic completely out of line locks
  2004-09-03  0:02 [PATCH][0/8] Arch agnostic completely out of line locks Zwane Mwaikambo
@ 2004-09-03 14:48 ` Martin J. Bligh
  2004-09-03 15:48   ` Zwane Mwaikambo
  0 siblings, 1 reply; 3+ messages in thread
From: Martin J. Bligh @ 2004-09-03 14:48 UTC (permalink / raw)
  To: Zwane Mwaikambo, Linux Kernel
  Cc: Linus Torvalds, Andrew Morton, Matt Mackall,
	William Lee Irwin III, Keith Owens

> This patch achieves out of line spinlocks by creating kernel/spinlock.c
> and using the _raw_* inline locking functions.
... 
> Size differences are with CONFIG_PREEMPT enabled since we wanted to
> determine how much could be saved by moving that lot out of line too.

So does this have no performance impact at all? or has that not been measured?

M.


^ permalink raw reply	[flat|nested] 3+ messages in thread

* Re: [PATCH][0/8] Arch agnostic completely out of line locks
  2004-09-03 14:48 ` Martin J. Bligh
@ 2004-09-03 15:48   ` Zwane Mwaikambo
  0 siblings, 0 replies; 3+ messages in thread
From: Zwane Mwaikambo @ 2004-09-03 15:48 UTC (permalink / raw)
  To: Martin J. Bligh
  Cc: Linux Kernel, Linus Torvalds, Andrew Morton, Matt Mackall,
	William Lee Irwin III, Keith Owens

On Fri, 3 Sep 2004, Martin J. Bligh wrote:

> > This patch achieves out of line spinlocks by creating kernel/spinlock.c
> > and using the _raw_* inline locking functions.
> ...
> > Size differences are with CONFIG_PREEMPT enabled since we wanted to
> > determine how much could be saved by moving that lot out of line too.
>
> So does this have no performance impact at all? or has that not been measured?

No performance regression on i386 with a specific benchmark (bonnie++)

http://www.zwane.ca/results/cool-locks-stp

I'm going to be running benchmarks against this version of the patch, but
still collecting which to run.

Thanks,
	Zwane


^ permalink raw reply	[flat|nested] 3+ messages in thread

end of thread, other threads:[~2004-09-03 15:45 UTC | newest]

Thread overview: 3+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2004-09-03  0:02 [PATCH][0/8] Arch agnostic completely out of line locks Zwane Mwaikambo
2004-09-03 14:48 ` Martin J. Bligh
2004-09-03 15:48   ` Zwane Mwaikambo

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox

all inboxes | Powered by JetHome®