From: "Nicholas A. Bellinger" <nab@linux-iscsi.org>
To: target-devel <target-devel@vger.kernel.org>
Cc: linux-scsi <linux-scsi@vger.kernel.org>,
Christoph Hellwig <hch@lst.de>, Roland Dreier <roland@kernel.org>,
Kent Overstreet <koverstreet@google.com>,
Asias He <asias@redhat.com>,
"Michael S. Tsirkin" <mst@redhat.com>,
Or Gerlitz <ogerlitz@mellanox.com>,
Moussa Ba <moussaba@micron.com>,
Nicholas Bellinger <nab@linux-iscsi.org>
Subject: [PATCH 1/9] target: Add transport_cmd_check_stop write_pending bit
Date: Fri, 7 Jun 2013 21:34:16 +0000 [thread overview]
Message-ID: <1370640864-15568-2-git-send-email-nab@linux-iscsi.org> (raw)
In-Reply-To: <1370640864-15568-1-git-send-email-nab@linux-iscsi.org>
From: Nicholas Bellinger <nab@linux-iscsi.org>
This patch adds a new transport_cmd_check_stop() parameter for signaling
when TRANSPORT_WRITE_PENDING needs to be set.
This allows transport_generic_new_cmd() to avoid the extra lock acquire/release
of ->t_state_lock in the fast path for DMA_TO_DEVICE operations ahead of
transport_cmd_check_stop() + se_tfo->write_pending().
Cc: Christoph Hellwig <hch@lst.de>
Cc: Roland Dreier <roland@kernel.org>
Cc: Kent Overstreet <koverstreet@google.com>
Cc: Or Gerlitz <ogerlitz@mellanox.com>
Cc: Moussa Ba <moussaba@micron.com>
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>
---
drivers/target/target_core_transport.c | 19 +++++++++----------
1 files changed, 9 insertions(+), 10 deletions(-)
diff --git a/drivers/target/target_core_transport.c b/drivers/target/target_core_transport.c
index 4a79336..15afa83 100644
--- a/drivers/target/target_core_transport.c
+++ b/drivers/target/target_core_transport.c
@@ -445,11 +445,15 @@ static void target_remove_from_state_list(struct se_cmd *cmd)
spin_unlock_irqrestore(&dev->execute_task_lock, flags);
}
-static int transport_cmd_check_stop(struct se_cmd *cmd, bool remove_from_lists)
+static int transport_cmd_check_stop(struct se_cmd *cmd, bool remove_from_lists,
+ bool write_pending)
{
unsigned long flags;
spin_lock_irqsave(&cmd->t_state_lock, flags);
+ if (write_pending)
+ cmd->t_state = TRANSPORT_WRITE_PENDING;
+
/*
* Determine if IOCTL context caller in requesting the stopping of this
* command for LUN shutdown purposes.
@@ -514,7 +518,7 @@ static int transport_cmd_check_stop(struct se_cmd *cmd, bool remove_from_lists)
static int transport_cmd_check_stop_to_fabric(struct se_cmd *cmd)
{
- return transport_cmd_check_stop(cmd, true);
+ return transport_cmd_check_stop(cmd, true, false);
}
static void transport_lun_remove_cmd(struct se_cmd *cmd)
@@ -2116,12 +2120,7 @@ transport_generic_new_cmd(struct se_cmd *cmd)
target_execute_cmd(cmd);
return 0;
}
-
- spin_lock_irq(&cmd->t_state_lock);
- cmd->t_state = TRANSPORT_WRITE_PENDING;
- spin_unlock_irq(&cmd->t_state_lock);
-
- transport_cmd_check_stop(cmd, false);
+ transport_cmd_check_stop(cmd, false, true);
ret = cmd->se_tfo->write_pending(cmd);
if (ret == -EAGAIN || ret == -ENOMEM)
@@ -2325,7 +2324,7 @@ static int transport_lun_wait_for_tasks(struct se_cmd *cmd, struct se_lun *lun)
pr_debug("ConfigFS ITT[0x%08x] - CMD_T_STOP, skipping\n",
cmd->se_tfo->get_task_tag(cmd));
spin_unlock_irqrestore(&cmd->t_state_lock, flags);
- transport_cmd_check_stop(cmd, false);
+ transport_cmd_check_stop(cmd, false, false);
return -EPERM;
}
cmd->transport_state |= CMD_T_LUN_FE_STOP;
@@ -2433,7 +2432,7 @@ check_cond:
spin_unlock_irqrestore(&cmd->t_state_lock,
cmd_flags);
- transport_cmd_check_stop(cmd, false);
+ transport_cmd_check_stop(cmd, false, false);
complete(&cmd->transport_lun_fe_stop_comp);
spin_lock_irqsave(&lun->lun_cmd_lock, lun_flags);
continue;
--
1.7.2.5
next prev parent reply other threads:[~2013-06-07 21:47 UTC|newest]
Thread overview: 15+ messages / expand[flat|nested] mbox.gz Atom feed top
2013-06-07 21:34 [PATCH 0/9] target: Optimizations + cleanups for v3.11 Nicholas A. Bellinger
2013-06-07 21:34 ` Nicholas A. Bellinger [this message]
2013-06-08 4:00 ` [PATCH 1/9] target: Add transport_cmd_check_stop write_pending bit Jörn Engel
2013-06-08 6:29 ` Christoph Hellwig
2013-06-07 21:34 ` [PATCH 2/9] target: Drop unnecessary CMD_T_DEV_ACTIVE check from transport_lun_remove_cmd Nicholas A. Bellinger
2013-06-07 21:34 ` [PATCH 3/9] target: Remove legacy t_fe_count + avoid t_state_lock access in transport_put_cmd Nicholas A. Bellinger
2013-06-07 21:34 ` [PATCH 4/9] target: Avoid extra t_state_lock access in __target_execute_cmd Nicholas A. Bellinger
2013-06-08 6:32 ` Christoph Hellwig
2013-06-07 21:34 ` [PATCH 5/9] target: Drop unnecessary t_state_lock access for SCF_SUPPORTED_SAM_OPCODE assignment Nicholas A. Bellinger
2013-06-07 21:34 ` [PATCH 6/9] iscsi-target: Avoid unnecessary t_state_lock during unsolicited data-out check Nicholas A. Bellinger
2013-06-07 21:34 ` [PATCH 7/9] target: Drop legacy se_cmd->check_release bit Nicholas A. Bellinger
2013-06-07 21:34 ` [PATCH 8/9] vhost/scsi: Drop unnecessary wait_for_tasks=true usage with transport_generic_free_cmd Nicholas A. Bellinger
2013-06-27 1:59 ` Asias He
2013-06-07 21:34 ` [PATCH 9/9] vhost/scsi: Convert to se_cmd->cmd_kref TARGET_SCF_ACK_KREF usage Nicholas A. Bellinger
2013-06-27 1:59 ` Asias He
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=1370640864-15568-2-git-send-email-nab@linux-iscsi.org \
--to=nab@linux-iscsi.org \
--cc=asias@redhat.com \
--cc=hch@lst.de \
--cc=koverstreet@google.com \
--cc=linux-scsi@vger.kernel.org \
--cc=moussaba@micron.com \
--cc=mst@redhat.com \
--cc=ogerlitz@mellanox.com \
--cc=roland@kernel.org \
--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).