From: Tejun Heo <tj@kernel.org>
To: Christoph Lameter <cl@linux.com>
Cc: Pekka Enberg <penberg@cs.helsinki.fi>,
David Rientjes <rientjes@google.com>,
Eric Dumazet <eric.dumazet@gmail.com>,
"H. Peter Anvin" <hpa@zytor.com>,
linux-kernel@vger.kernel.org,
Thomas Gleixner <tglx@linutronix.de>
Subject: Re: [slubllv7 04/17] x86: Add support for cmpxchg_double
Date: Wed, 15 Jun 2011 18:39:52 +0200 [thread overview]
Message-ID: <20110615163952.GE8141@htj.dyndns.org> (raw)
In-Reply-To: <alpine.DEB.2.00.1106150917520.768@router.home>
Hello, Christoph.
On Wed, Jun 15, 2011 at 09:26:15AM -0500, Christoph Lameter wrote:
> > Do we really need cmpxchg16b*() macros separately? Why not just
> > collapse them into cmpxchg_double*()? Also, it would be better if we
> > have the same level of VM_BUG_ON() checks as in percpu cmpxchg_double
> > ops. Maybe we should put them in a separate macro?
>
> The method here is to put all the high level checks in cmpxchg_double()
> and then do the low level asm stuff in cmpxchg16b macros. I think that is
> a good separation.
I don't know; then, I think the name better clearly indicate that
they're not used outside of implementation. I don't see merit in
keeping them separate.
> > > +#define system_has_cmpxchg_double() cpu_has_cx16
> >
> > Where's the fallback %false definition for the above feature macro for
> > archs which don't support cmpxchg_double? Also, is system_has_*()
> > conventional? Isn't arch_has_*() more conventional for this purpose?
>
> There is a convention for querying processor flags from core code?
At least generic ptrace code uses arch_has_block/single_step().
Probably better than introducing something completely new.
> The system_has_cmpxchg_double() is only used if the arch defines
> CONFIG_CMPXCHG_DOUBLE
Why? What's the benefit of that?
> This way it is done in the same way on 32 bit than on 64 bit. The use of
> cmpxchg64 also means that some of the parameters would have to be combined
> to form 64 bit ints from the 32 bit ones before __cmpxchg64 could be used.
>
> __cmpxchg64 has different parameter conventions.
But they all just deal with the starting addresses and the _local
version already has proper fallback implementation.
> > Another thing is that choosing different code path depending on
> > has_cmpxchg_double() would be quite messy and won't bode well with
> > many people. I agree that fallback implementation would be heavier
> > for SMP safe operations but some archs already do that for cmpxchg
> > (forgot which one). If we're gonna export this to generic code,
> > wouldn't it be better to implement proper generic fallbacks and
> > provide has_*() as hint?
>
> A generic fallback for cmpxchg_double would mean having to disable
> interrupts and then take a global spinlock. There are significant scaling
> problems with such an implementation.
>
> The fallback through the subsystem means that the subsystem can do locking
> that scales better. In the case of SLUB we fall back to a bit lock in the
> page struct which is a hot cache line in the hotpaths. This is the same
> approach as used before the lockless patches and we expect the performance
> on platforms not supporting cmpxchg_double to stay the same.
Yes, that's nice but you're introducing new operations and they should
meet the usual conventions and cmpxchg fallback on the arch which I
don't recall now already uses hashed lock so it's not like this is
completely new. As added, the interface basically requires extreme
ifdeffery which isn't good.
Thanks.
--
tejun
next prev parent reply other threads:[~2011-06-15 16:40 UTC|newest]
Thread overview: 41+ messages / expand[flat|nested] mbox.gz Atom feed top
2011-06-01 17:25 [slubllv7 00/17] SLUB: Lockless freelists for objects V7 Christoph Lameter
2011-06-01 17:25 ` [slubllv7 01/17] slub: Push irq disable into allocate_slab() Christoph Lameter
2011-06-01 17:25 ` [slubllv7 02/17] slub: Do not use frozen page flag but a bit in the page counters Christoph Lameter
2011-06-01 17:25 ` [slubllv7 03/17] slub: Move page->frozen handling near where the page->freelist handling occurs Christoph Lameter
2011-06-01 17:25 ` [slubllv7 04/17] x86: Add support for cmpxchg_double Christoph Lameter
2011-06-09 9:53 ` Pekka Enberg
2011-06-10 15:17 ` Christoph Lameter
2011-06-11 9:50 ` Pekka Enberg
2011-06-11 17:02 ` Christoph Lameter
2011-06-14 5:49 ` Pekka Enberg
2011-06-14 8:04 ` Ingo Molnar
2011-06-14 14:04 ` Christoph Lameter
2011-06-14 15:05 ` H. Peter Anvin
2011-06-15 8:55 ` Tejun Heo
2011-06-15 14:26 ` Christoph Lameter
2011-06-15 16:39 ` Tejun Heo [this message]
2011-06-15 17:19 ` Christoph Lameter
2011-06-25 23:49 ` [tip:x86/atomic] " tip-bot for Christoph Lameter
2011-06-01 17:25 ` [slubllv7 05/17] mm: Rearrange struct page Christoph Lameter
2011-06-09 9:57 ` Pekka Enberg
2011-06-09 16:45 ` Andrew Morton
2011-06-09 17:03 ` [PATCH] checkpatch: Add a "prefer __aligned" check Joe Perches
2011-06-01 17:25 ` [slubllv7 06/17] slub: Add cmpxchg_double_slab() Christoph Lameter
2011-07-11 19:55 ` Eric Dumazet
2011-07-12 15:59 ` Christoph Lameter
2011-07-12 16:06 ` Eric Dumazet
2011-07-12 16:47 ` Christoph Lameter
2011-07-12 18:40 ` H. Peter Anvin
2011-07-12 18:53 ` Christoph Lameter
2011-07-12 20:40 ` H. Peter Anvin
2011-06-01 17:25 ` [slubllv7 07/17] slub: explicit list_lock taking Christoph Lameter
2011-06-01 17:25 ` [slubllv7 08/17] slub: Pass kmem_cache struct to lock and freeze slab Christoph Lameter
2011-06-01 17:25 ` [slubllv7 09/17] slub: Rework allocator fastpaths Christoph Lameter
2011-06-01 17:25 ` [slubllv7 10/17] slub: Invert locking and avoid slab lock Christoph Lameter
2011-06-01 17:25 ` [slubllv7 11/17] slub: Disable interrupts in free_debug processing Christoph Lameter
2011-06-01 17:25 ` [slubllv7 12/17] slub: Avoid disabling interrupts in free slowpath Christoph Lameter
2011-06-01 17:25 ` [slubllv7 13/17] slub: Get rid of the another_slab label Christoph Lameter
2011-06-01 17:25 ` [slubllv7 14/17] slub: Add statistics for the case that the current slab does not match the node Christoph Lameter
2011-06-01 17:25 ` [slubllv7 15/17] slub: fast release on full slab Christoph Lameter
2011-06-01 17:25 ` [slubllv7 16/17] slub: Not necessary to check for empty slab on load_freelist Christoph Lameter
2011-06-01 17:26 ` [slubllv7 17/17] slub: slabinfo update for cmpxchg handling Christoph Lameter
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20110615163952.GE8141@htj.dyndns.org \
--to=tj@kernel.org \
--cc=cl@linux.com \
--cc=eric.dumazet@gmail.com \
--cc=hpa@zytor.com \
--cc=linux-kernel@vger.kernel.org \
--cc=penberg@cs.helsinki.fi \
--cc=rientjes@google.com \
--cc=tglx@linutronix.de \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).