From: Christoph Hellwig <hch@infradead.org>
To: "Nicholas A. Bellinger" <nab@linux-iscsi.org>
Cc: "Nicholas A. Bellinger" <nab@daterainc.com>,
target-devel <target-devel@vger.kernel.org>,
linux-scsi <linux-scsi@vger.kernel.org>,
Hannes Reinecke <hare@suse.de>,
Sagi Grimberg <sagig@mellanox.com>,
Andy Grover <agrover@redhat.com>
Subject: Re: [RFC 00/22] target: se_node_acl LUN list RCU conversion
Date: Thu, 2 Apr 2015 01:56:07 -0700 [thread overview]
Message-ID: <20150402085607.GA31017@infradead.org> (raw)
In-Reply-To: <1427953047.9176.292.camel@haakon3.risingtidesystems.com>
On Wed, Apr 01, 2015 at 10:37:27PM -0700, Nicholas A. Bellinger wrote:
> On Wed, 2015-04-01 at 00:04 -0700, Christoph Hellwig wrote:
> > On Tue, Mar 31, 2015 at 11:51:56PM -0700, Nicholas A. Bellinger wrote:
> > > Since last week, enable/disable device_list code has been converted to
> > > use mempool + call_rcu() and performs the RCU pointer swap in
> > > se_node_acl->lun_entry_hlist[] under se_node_acl->lun_entry_mutex.
> >
> > Why use a mempool there?
>
> Would be nice to guarantee a certain number of new se_dev_entry
> allocations always succeed as existing code expects.
mempools are for I/O path that make guaranteed progress. While the
callers of core_enable_device_list_for_node are in the control path,
and not in a very deep callstack, fairly shallow below the configfs
interface. This is not a use case for mempools, as there is no need
to guarantee progress in deep I/O callstacks.
> The use of t10_pr_registration->pr_reg_deve is still wrong though, as
> it's held for the lifetime of the pr_reg, and not just while
> pr_ref_count is non zero for REGISTER w/ I_PORT + REGISTER_AND_MOVE
> special cases.
Seems like we should just take the reference properly.
> I've already fixed this for v2 by using se_dev_entry->pr_reg for
> signaling when a PR registration is active.
At least in the version you posted ->pr_reg is just used as a boolean
flag, in which case we might want to keep the real def_pr_registered
flag.
> > The refcount is still there in the good old atomic_t-based version,
> > which is perfectl fine for the PRIN and PROUT subcommands which aren't
> > the I/O fast path.
>
> Agreed the percpu_ref is overkill and percpu_ref_init() is problematic
> because it can still -ENOMEM during se_dev_entry creation..
>
> Switching to a normal kref with a blocking completion probably makes the
> most sense, in order to avoid potentially spinning on atomic_read() if
> APTPL metadata write-out ends up blocking during the two special PR
> cases.
The nice part about using the dynamically allocated dev entry is that
the shutdown path doesn't need to be sync. You can just use a normal
kref and don't care when it actually gets freed. One thing I didn't
do in my branch but which would e useful is to rename -> pr_ref
to just ->ref as it's a generic refcount - for now it's only used by
the PR code, but there is nothing specific to it in there.
next prev parent reply other threads:[~2015-04-02 8:56 UTC|newest]
Thread overview: 37+ messages / expand[flat|nested] mbox.gz Atom feed top
2015-03-27 8:04 [RFC 00/22] target: se_node_acl LUN list RCU conversion Nicholas A. Bellinger
2015-03-27 8:04 ` [RFC 01/22] target: Convert transport_lookup_*_lun to RCU reader Nicholas A. Bellinger
2015-03-29 6:38 ` Sagi Grimberg
2015-03-27 8:04 ` [RFC 02/22] target: Convert enable/disable ->device_list to RCU updater Nicholas A. Bellinger
2015-03-29 6:51 ` Sagi Grimberg
2015-03-27 8:04 ` [RFC 03/22] target/device: Convert se_node_acl->device_list access to RCU reader Nicholas A. Bellinger
2015-03-27 8:04 ` [RFC 04/22] target/configfs: Convert mappedlun link/unlink " Nicholas A. Bellinger
2015-03-27 8:04 ` [RFC 05/22] target/configfs: Convert SCSI MIB attrs " Nicholas A. Bellinger
2015-03-27 8:04 ` [RFC 06/22] target/spc: Convert REPORT_LUN + MODE_SENSE " Nicholas A. Bellinger
2015-03-27 8:04 ` [RFC 07/22] target/pscsi: Convert MODE_SENSE special case " Nicholas A. Bellinger
2015-03-27 8:04 ` [RFC 08/22] target/pr: Convert se_dev_entry to percpu-refcount for RCU Nicholas A. Bellinger
2015-03-27 8:04 ` [RFC 09/22] target/pr: Convert registration check to RCU pointer Nicholas A. Bellinger
2015-03-27 8:05 ` [RFC 10/22] target/pr: Change alloc_registration to avoid pr_reg_tg_pt_lun Nicholas A. Bellinger
2015-03-27 8:05 ` [RFC 11/22] target: Convert UNIT_ATTENTION logic to RCU reader Nicholas A. Bellinger
2015-03-27 8:05 ` [RFC 12/22] target: Convert se_tpg->tpg_lun_lock to ->tpg_lun_mutex Nicholas A. Bellinger
2015-03-27 8:05 ` [RFC 13/22] target: Convert se_tpg->acl_node_lock to ->acl_node_mutex Nicholas A. Bellinger
2015-03-27 8:05 ` [RFC 14/22] target: Convert se_node_acl->lun_entry_lock to ->lun_entry_mutex Nicholas A. Bellinger
2015-03-27 8:05 ` [RFC 15/22] target: Convert core_tpg_deregister to use list splice Nicholas A. Bellinger
2015-03-27 8:05 ` [RFC 16/22] target: Drop se_lun->lun_acl_list Nicholas A. Bellinger
2015-03-27 8:05 ` [RFC 17/22] target: Drop core_tpg_clear_object_luns Nicholas A. Bellinger
2015-03-27 8:05 ` [RFC 18/22] target: Rename TPG initiator_node_acl to target_* prefix Nicholas A. Bellinger
2015-03-27 8:05 ` [RFC 19/22] target: Rename TPG register/deregister " Nicholas A. Bellinger
2015-03-27 8:05 ` [RFC 20/22] target: Rename LUN lookup/add/remove " Nicholas A. Bellinger
2015-03-27 8:05 ` [RFC 21/22] target: Rename se_node_acl->lun_entry_hlist " Nicholas A. Bellinger
2015-03-27 8:05 ` [RFC 22/22] target: Rename se_port/se_device export " Nicholas A. Bellinger
2015-03-30 12:08 ` [RFC 00/22] target: se_node_acl LUN list RCU conversion Christoph Hellwig
2015-03-30 19:16 ` Andy Grover
2015-03-30 19:21 ` Christoph Hellwig
2015-03-30 19:35 ` Andy Grover
2015-03-31 9:23 ` Christoph Hellwig
2015-04-01 6:51 ` Nicholas A. Bellinger
2015-04-01 7:04 ` Christoph Hellwig
2015-04-02 5:37 ` Nicholas A. Bellinger
2015-04-02 8:56 ` Christoph Hellwig [this message]
2015-04-02 20:57 ` Nicholas A. Bellinger
2015-04-07 6:17 ` Christoph Hellwig
2015-04-08 6:31 ` Nicholas A. Bellinger
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=20150402085607.GA31017@infradead.org \
--to=hch@infradead.org \
--cc=agrover@redhat.com \
--cc=hare@suse.de \
--cc=linux-scsi@vger.kernel.org \
--cc=nab@daterainc.com \
--cc=nab@linux-iscsi.org \
--cc=sagig@mellanox.com \
--cc=target-devel@vger.kernel.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).