linux-fsdevel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Peter Zijlstra <peterz@infradead.org>
To: Kent Overstreet <kent.overstreet@gmail.com>
Cc: linux-kernel@vger.kernel.org, linux-fsdevel@vger.kernel.org,
	Andrew Morton <akpm@linux-foundation.org>,
	Dave Chinner <dchinner@redhat.com>,
	darrick.wong@oracle.com, tytso@mit.edu,
	linux-btrfs@vger.kernel.org, clm@fb.com, jbacik@fb.com,
	viro@zeniv.linux.org.uk, willy@infradead.org
Subject: Re: [PATCH 03/10] locking: bring back lglocks
Date: Fri, 18 May 2018 13:03:39 +0200	[thread overview]
Message-ID: <20180518110339.GG12217@hirez.programming.kicks-ass.net> (raw)
In-Reply-To: <20180518101353.GA15403@kmo-pixel>

On Fri, May 18, 2018 at 06:13:53AM -0400, Kent Overstreet wrote:
> On Fri, May 18, 2018 at 11:51:02AM +0200, Peter Zijlstra wrote:
> > On Fri, May 18, 2018 at 03:49:04AM -0400, Kent Overstreet wrote:
> > > bcachefs makes use of them - also, add a proper lg_lock_init()
> > 
> > Why?! lglocks are horrid things, we got rid of them for a reason. They
> > have terrifying worst case preemption off latencies.
> 
> Ah. That was missing from your commit message.

Yeah, sorry, sometimes it's hard to state what is obvious to oneself :/

> > Why can't you use something like per-cpu rwsems?
> 
> Well,
> 
>  a) in my use case, lg_global_lock() pretty much isn't used in normal operation,
>     it's only called when starting mark and sweep gc (which is not needed
>     anymore and disabled by default, it'll eventually get rolled into online
>     fsck) and for device resize
> 
>  b) I'm using it in conjection with percpu counters, and technically yes I
>     certainly _could_ use per-cpu sleepable locks (mutexes would make more sense
>     for me than rwsems), there's a bit of a clash there and it's going to be a
>     bit ugly and messy and it's more work for me. (this_cpu_ptr() no longer
>     makes any sense in that case, so it'd mean auditing and converting all the
>     code that touches the relevant data structures).

Well, lg is a reader-writer style lock per definition, as you want
concurrency on the local and full exclusion against the global, so I'm
not sure how mutexes fit into this.

In any case, have a look at percpu_down_read_preempt_disable() and
percpu_up_read_preempt_enable(); they're a bit of a hack but they should
work for you I think.

They will sleep at down_read, but the entire actual critical section
will be with preemption disabled -- therefore it had better be short and
bounded, and the RT guys will thank you for not using spinlock_t under
it (use raw_spinlock_t if you have to).

The (global) writer side will block and be preemptible like normal.

> If you're really that dead set against lglocks I might just come up with a new
> lock with similar semantics, that doesn't break this_cpu_ptr() but sleeps if the
> global lock is held...

See above, we already have this ;-)

  reply	other threads:[~2018-05-18 11:03 UTC|newest]

Thread overview: 43+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2018-05-18  7:48 [PATCH 00/10] RFC: assorted bcachefs patches Kent Overstreet
2018-05-18  7:49 ` [PATCH 01/10] mm: pagecache add lock Kent Overstreet
2018-05-18 13:13   ` Matthew Wilcox
2018-05-18 15:53     ` Christoph Hellwig
2018-05-18 17:45       ` Kent Overstreet
2018-05-23 23:55         ` Notes on locking for pagacache consistency (was: [PATCH 01/10] mm: pagecache add lock) Kent Overstreet
2018-05-20 22:45       ` [PATCH 01/10] mm: pagecache add lock Kent Overstreet
2018-05-23 15:22         ` Christoph Hellwig
2018-05-23 17:12           ` Kent Overstreet
2018-05-18  7:49 ` [PATCH 02/10] mm: export find_get_pages() Kent Overstreet
2018-05-18 16:00   ` Christoph Hellwig
2018-05-18  7:49 ` [PATCH 03/10] locking: bring back lglocks Kent Overstreet
2018-05-18  9:51   ` Peter Zijlstra
2018-05-18 10:13     ` Kent Overstreet
2018-05-18 11:03       ` Peter Zijlstra [this message]
2018-05-18 11:39         ` Kent Overstreet
2018-05-18  7:49 ` [PATCH 04/10] locking: export osq_lock()/osq_unlock() Kent Overstreet
2018-05-18  9:52   ` Peter Zijlstra
2018-05-18 10:18     ` Kent Overstreet
2018-05-18 11:08       ` Peter Zijlstra
2018-05-18 11:32         ` Kent Overstreet
2018-05-18 11:40           ` Peter Zijlstra
2018-05-18 12:40             ` Kent Overstreet
2018-05-18  7:49 ` [PATCH 05/10] don't use spin_lock_irqsave() unnecessarily Kent Overstreet
2018-05-18 16:01   ` Christoph Hellwig
2018-05-18  7:49 ` [PATCH 06/10] Generic radix trees Kent Overstreet
2018-05-18 16:02   ` Christoph Hellwig
2018-05-18 17:38     ` Kent Overstreet
2018-05-18  7:49 ` [PATCH 07/10] bcache: optimize continue_at_nobarrier() Kent Overstreet
2018-05-18  7:49 ` [PATCH 08/10] bcache: move closures to lib/ Kent Overstreet
2018-05-18 16:02   ` Christoph Hellwig
2018-05-18  7:49 ` [PATCH 09/10] closures: closure_wait_event() Kent Overstreet
2018-05-18  7:49 ` [PATCH 10/10] Dynamic fault injection Kent Overstreet
2018-05-18 16:02   ` Christoph Hellwig
2018-05-18 17:37     ` Kent Overstreet
2018-05-18 19:05   ` Andreas Dilger
2018-05-18 19:10     ` Kent Overstreet
2018-05-18 20:54       ` Andreas Dilger
2018-05-18  7:55 ` [PATCH 00/10] RFC: assorted bcachefs patches Kent Overstreet
2018-05-18 17:45 ` Josef Bacik
2018-05-18 17:49   ` Kent Overstreet
2018-05-18 18:03     ` Josef Bacik
2018-05-18 18:28       ` Kent Overstreet

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=20180518110339.GG12217@hirez.programming.kicks-ass.net \
    --to=peterz@infradead.org \
    --cc=akpm@linux-foundation.org \
    --cc=clm@fb.com \
    --cc=darrick.wong@oracle.com \
    --cc=dchinner@redhat.com \
    --cc=jbacik@fb.com \
    --cc=kent.overstreet@gmail.com \
    --cc=linux-btrfs@vger.kernel.org \
    --cc=linux-fsdevel@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=tytso@mit.edu \
    --cc=viro@zeniv.linux.org.uk \
    --cc=willy@infradead.org \
    /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).