From: Sergey Senozhatsky <sergey.senozhatsky.work@gmail.com>
To: Joonsoo Kim <js1304@gmail.com>
Cc: Sergey Senozhatsky <sergey.senozhatsky.work@gmail.com>,
Andrew Morton <akpm@linux-foundation.org>,
Minchan Kim <minchan@kernel.org>,
Sergey Senozhatsky <sergey.senozhatsky@gmail.com>,
linux-kernel@vger.kernel.org, kernel-team@lge.com
Subject: Re: [PATCH v4 2/4] zram: implement deduplication in zram
Date: Wed, 26 Apr 2017 15:14:57 +0900 [thread overview]
Message-ID: <20170426061457.GF673@jagdpanzerIV.localdomain> (raw)
In-Reply-To: <20170426060816.GD29773@js1304-desktop>
On (04/26/17 15:08), Joonsoo Kim wrote:
> > > +struct zram_hash {
> > > + spinlock_t lock;
> > > + struct rb_root rb_root;
> > > };
> >
> > just a note.
> >
> > we can easily have N CPUs spinning on ->lock for __zram_dedup_get() lookup,
> > which can invole a potentially slow zcomp_decompress() [zlib, for example,
> > with 64k pages] and memcmp(). the larger PAGE_SHIFT is, the more serialized
> > IOs become. in theory, at least.
> >
> > CPU0 CPU1 ... CPUN
> >
> > __zram_bvec_write() __zram_bvec_write() __zram_bvec_write()
> > zram_dedup_find() zram_dedup_find() zram_dedup_find()
> > spin_lock(&hash->lock);
> > spin_lock(&hash->lock); spin_lock(&hash->lock);
> > __zram_dedup_get()
> > zcomp_decompress()
> > ...
> >
> >
> > so may be there is a way to use read-write lock instead on spinlock for hash
> > and reduce write/read IO serialization.
>
> In fact, dedup release hash->lock before doing zcomp_decompress(). So,
> above contention cannot happen.
oh, my bad. you are right. somehow I didn't spot the unlock.
-ss
next prev parent reply other threads:[~2017-04-26 6:15 UTC|newest]
Thread overview: 25+ messages / expand[flat|nested] mbox.gz Atom feed top
2017-04-26 0:52 [PATCH v4 0/4] zram: implement deduplication in zram js1304
2017-04-26 0:52 ` [PATCH v4 1/4] zram: introduce zram_entry to prepare dedup functionality js1304
2017-04-26 0:52 ` [PATCH v4 2/4] zram: implement deduplication in zram js1304
2017-04-26 2:14 ` Sergey Senozhatsky
2017-04-26 5:57 ` Joonsoo Kim
2017-04-26 6:29 ` Minchan Kim
2017-04-26 6:59 ` Joonsoo Kim
2017-04-26 7:21 ` Sergey Senozhatsky
2017-04-26 7:39 ` Minchan Kim
2017-04-26 7:12 ` Sergey Senozhatsky
2017-04-26 2:37 ` Sergey Senozhatsky
2017-04-26 5:59 ` Joonsoo Kim
2017-04-26 6:04 ` Sergey Senozhatsky
2017-04-26 4:02 ` Sergey Senozhatsky
2017-04-26 6:04 ` Joonsoo Kim
2017-04-26 6:21 ` Sergey Senozhatsky
2017-04-27 6:57 ` Joonsoo Kim
2017-04-27 7:46 ` Sergey Senozhatsky
2017-05-02 5:31 ` Joonsoo Kim
2017-04-26 4:28 ` Sergey Senozhatsky
2017-04-26 6:08 ` Joonsoo Kim
2017-04-26 6:14 ` Sergey Senozhatsky [this message]
2017-04-26 0:52 ` [PATCH v4 3/4] zram: make deduplication feature optional js1304
2017-04-26 0:52 ` [PATCH v4 4/4] zram: compare all the entries with same checksum for deduplication js1304
2017-04-27 7:49 ` [PATCH v4 0/4] zram: implement deduplication in zram Sergey Senozhatsky
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=20170426061457.GF673@jagdpanzerIV.localdomain \
--to=sergey.senozhatsky.work@gmail.com \
--cc=akpm@linux-foundation.org \
--cc=js1304@gmail.com \
--cc=kernel-team@lge.com \
--cc=linux-kernel@vger.kernel.org \
--cc=minchan@kernel.org \
--cc=sergey.senozhatsky@gmail.com \
/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).