From: Christoph Hellwig <hch@infradead.org>
To: Benny Halevy <bhalevy@panasas.com>
Cc: Christoph Hellwig <hch@infradead.org>,
Fred Isaman <iisaman@netapp.com>,
andros@netapp.com, trond.myklebust@netapp.com,
linux-nfs@vger.kernel.org
Subject: Re: [PATCH 08/16] pnfs: wave 3: lseg refcounting
Date: Tue, 15 Feb 2011 10:06:28 -0500 [thread overview]
Message-ID: <20110215150628.GA24358@infradead.org> (raw)
In-Reply-To: <4D5A94CD.4030508@panasas.com>
Btw, what's the point for deferring the free_lseg calls? It looks like it's
to avoid calling something that might block under i_lock, but looking around
the pnfs-submit branch it seems that root cause could be fixed trivially.
In common code *free_lseg* and *put_layout_hdr* do nothing but list
manipulations and kfrees. And in filelayout_free_lseg we have just kfrees
and a call to pnfs_put_deviceid which may sleep due to calling
synchronize_rcu. But synchronize_rcu is horribly inefficient to start with,
and you'd better be off using call_rcu to free the device id, which will
lead to much saner code and better performance.
next prev parent reply other threads:[~2011-02-15 15:06 UTC|newest]
Thread overview: 54+ messages / expand[flat|nested] mbox.gz Atom feed top
2011-02-14 19:18 [PATCH 0/16] pnfs wave 3 submission andros
2011-02-14 19:18 ` [PATCH 01/16] NFS remove unnecessary CONFIG_NFS_V4 from nfs_read_data andros
2011-02-15 9:16 ` Christoph Hellwig
2011-02-15 9:24 ` Taousif_Ansari-G5Y5guI6XLZWk0Htik3J/w
2011-02-15 14:51 ` Andy Adamson
2011-02-14 19:18 ` [PATCH 02/16] NFS put_layout_hdr can remove nfsi->layout andros
2011-02-14 19:18 ` [PATCH 03/16] NFS move nfs_client initialization into nfs_get_client andros
2011-02-16 2:58 ` Benny Halevy
2011-02-16 16:00 ` Andy Adamson
2011-02-14 19:18 ` [PATCH 04/16] pnfs: wave 3: send zero stateid seqid on v4.1 i/o andros
2011-02-14 19:18 ` [PATCH 05/16] pnfs: wave 3: new flag for state renewal check andros
2011-02-14 19:18 ` [PATCH 06/16] pnfs: wave 3: new flag for lease time check andros
2011-02-14 19:18 ` [PATCH 07/16] pnfs: wave 3: add MDS mount DS only check andros
2011-02-14 19:18 ` [PATCH 08/16] pnfs: wave 3: lseg refcounting andros
2011-02-15 9:25 ` Christoph Hellwig
2011-02-15 14:48 ` Fred Isaman
2011-02-15 14:58 ` Christoph Hellwig
2011-02-15 14:59 ` Benny Halevy
2011-02-15 15:06 ` Christoph Hellwig [this message]
2011-02-15 15:11 ` Fred Isaman
2011-02-15 16:02 ` Christoph Hellwig
2011-02-15 16:37 ` William A. (Andy) Adamson
2011-02-15 19:17 ` Andy Adamson
2011-02-15 19:29 ` Benny Halevy
2011-02-15 19:30 ` Andy Adamson
2011-02-15 15:07 ` Fred Isaman
2011-02-14 19:18 ` [PATCH 09/16] pnfs: wave 3: shift pnfs_update_layout locations andros
2011-02-14 23:14 ` Trond Myklebust
2011-02-15 14:41 ` Fred Isaman
2011-02-15 15:00 ` Trond Myklebust
2011-02-16 3:11 ` Benny Halevy
2011-02-14 19:18 ` [PATCH 10/16] pnfs: wave 3: coelesce across layout stripes andros
2011-02-14 23:42 ` Trond Myklebust
2011-02-15 14:43 ` William A. (Andy) Adamson
2011-02-15 15:03 ` Trond Myklebust
[not found] ` <1297782220.10103.13.camel-rJ7iovZKK19ZJLDQqaL3InhyD016LWXt@public.gmane.org>
2011-02-15 15:10 ` Andy Adamson
2011-02-14 19:18 ` [PATCH 11/16] pnfs: wave 3: generic read andros
2011-02-14 23:36 ` Trond Myklebust
2011-02-15 14:47 ` Andy Adamson
2011-02-16 3:16 ` Benny Halevy
2011-02-16 14:53 ` Andy Adamson
2011-02-16 15:09 ` Trond Myklebust
2011-02-16 15:52 ` Benny Halevy
2011-02-16 15:56 ` Andy Adamson
2011-02-16 15:57 ` Sager, Mike
2011-02-14 19:18 ` [PATCH 12/16] pnfs: wave 3: data server connection andros
2011-02-14 19:18 ` [PATCH 13/16] pnfs: wave 3: filelayout i/o helpers andros
2011-02-15 9:31 ` Christoph Hellwig
2011-02-15 15:12 ` Andy Adamson
2011-02-14 19:18 ` [PATCH 14/16] pnfs: wave 3: filelayout read andros
2011-02-14 19:18 ` [PATCH 15/16] pnfs: wave 3: filelayout async error handler andros
2011-02-14 19:18 ` [PATCH 16/16] pnfs: wave 3: turn off pNFS on ds connection failure andros
2011-02-14 22:39 ` [PATCH 0/16] pnfs wave 3 submission Trond Myklebust
2011-02-15 14:44 ` William A. (Andy) Adamson
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=20110215150628.GA24358@infradead.org \
--to=hch@infradead.org \
--cc=andros@netapp.com \
--cc=bhalevy@panasas.com \
--cc=iisaman@netapp.com \
--cc=linux-nfs@vger.kernel.org \
--cc=trond.myklebust@netapp.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).