public inbox for linux-kernel@vger.kernel.org
 help / color / mirror / Atom feed
From: green@linuxhacker.ru
To: Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
	devel@driverdev.osuosl.org,
	Andreas Dilger <andreas.dilger@intel.com>
Cc: Linux Kernel Mailing List <linux-kernel@vger.kernel.org>,
	Oleg Drokin <green@linuxhacker.ru>
Subject: [PATCH 11/20] staging/lustre/obdclass: Move AT controls from sysctl to sysfs
Date: Mon,  6 Jul 2015 12:48:49 -0400	[thread overview]
Message-ID: <1436201338-14263-12-git-send-email-green@linuxhacker.ru> (raw)
In-Reply-To: <1436201338-14263-1-git-send-email-green@linuxhacker.ru>

From: Oleg Drokin <green@linuxhacker.ru>

Adaptive Timeouts controls are being moved from /proc/sys/lustre
to /sys/fs/lustre

Signed-off-by: Oleg Drokin <green@linuxhacker.ru>
---
 .../lustre/lustre/obdclass/linux/linux-sysctl.c    | 45 ++++--------------
 drivers/staging/lustre/sysfs-fs-lustre             | 53 ++++++++++++++++++++++
 2 files changed, 63 insertions(+), 35 deletions(-)

diff --git a/drivers/staging/lustre/lustre/obdclass/linux/linux-sysctl.c b/drivers/staging/lustre/lustre/obdclass/linux/linux-sysctl.c
index 6af1cb2..82b3c39 100644
--- a/drivers/staging/lustre/lustre/obdclass/linux/linux-sysctl.c
+++ b/drivers/staging/lustre/lustre/obdclass/linux/linux-sysctl.c
@@ -138,44 +138,14 @@ LUSTRE_RW_ATTR(max_dirty_mb);
 LUSTRE_STATIC_UINT_ATTR(debug_peer_on_timeout, &obd_debug_peer_on_timeout);
 LUSTRE_STATIC_UINT_ATTR(dump_on_timeout, &obd_dump_on_timeout);
 LUSTRE_STATIC_UINT_ATTR(dump_on_eviction, &obd_dump_on_eviction);
+LUSTRE_STATIC_UINT_ATTR(at_min, &at_min);
+LUSTRE_STATIC_UINT_ATTR(at_max, &at_max);
+LUSTRE_STATIC_UINT_ATTR(at_extra, &at_extra);
+LUSTRE_STATIC_UINT_ATTR(at_early_margin, &at_early_margin);
+LUSTRE_STATIC_UINT_ATTR(at_history, &at_history);
 
 #ifdef CONFIG_SYSCTL
 static struct ctl_table obd_table[] = {
-	{
-		.procname = "at_min",
-		.data     = &at_min,
-		.maxlen   = sizeof(int),
-		.mode     = 0644,
-		.proc_handler = &proc_dointvec,
-	},
-	{
-		.procname = "at_max",
-		.data     = &at_max,
-		.maxlen   = sizeof(int),
-		.mode     = 0644,
-		.proc_handler = &proc_dointvec,
-	},
-	{
-		.procname = "at_extra",
-		.data     = &at_extra,
-		.maxlen   = sizeof(int),
-		.mode     = 0644,
-		.proc_handler = &proc_dointvec,
-	},
-	{
-		.procname = "at_early_margin",
-		.data     = &at_early_margin,
-		.maxlen   = sizeof(int),
-		.mode     = 0644,
-		.proc_handler = &proc_dointvec,
-	},
-	{
-		.procname = "at_history",
-		.data     = &at_history,
-		.maxlen   = sizeof(int),
-		.mode     = 0644,
-		.proc_handler = &proc_dointvec,
-	},
 	{}
 };
 
@@ -197,6 +167,11 @@ static struct attribute *lustre_attrs[] = {
 	&lustre_sattr_debug_peer_on_timeout.u.attr,
 	&lustre_sattr_dump_on_timeout.u.attr,
 	&lustre_sattr_dump_on_eviction.u.attr,
+	&lustre_sattr_at_min.u.attr,
+	&lustre_sattr_at_max.u.attr,
+	&lustre_sattr_at_extra.u.attr,
+	&lustre_sattr_at_early_margin.u.attr,
+	&lustre_sattr_at_history.u.attr,
 	NULL,
 };
 
diff --git a/drivers/staging/lustre/sysfs-fs-lustre b/drivers/staging/lustre/sysfs-fs-lustre
index ea9982b..873e2cf 100644
--- a/drivers/staging/lustre/sysfs-fs-lustre
+++ b/drivers/staging/lustre/sysfs-fs-lustre
@@ -90,6 +90,59 @@ Description:
 		This is useful if yout debug buffer typically rolls over
 		 by the time you notice the eviction event.
 
+What:		/sys/fs/lustre/at_min
+Date:		July 2015
+Contact:	"Oleg Drokin" <oleg.drokin@intel.com>
+Description:
+		Controls minimum adaptive timeout in seconds. If you encounter
+		a case where clients timeout due to server-reported processing
+		time being too short, you might consider increasing this value.
+		One common case of this if the underlying network has
+		unpredictable long delays.
+		Default: 0
+
+What:		/sys/fs/lustre/at_max
+Date:		July 2015
+Contact:	"Oleg Drokin" <oleg.drokin@intel.com>
+Description:
+		Controls maximum adaptive timeout in seconds. If at_max timeout
+		is reached for an RPC, the RPC will time out.
+		Some genuinuely slow network hardware might warrant increasing
+		this value.
+		Setting this value to 0 disables Adaptive Timeouts
+		functionality and old-style obd_timeout value is then used.
+		Default: 600
+
+What:		/sys/fs/lustre/at_extra
+Date:		July 2015
+Contact:	"Oleg Drokin" <oleg.drokin@intel.com>
+Description:
+		Controls how much extra time to request for unfinished requests
+		in processing in seconds. Normally a server-side parameter, it
+		is also used on the client for responses to various LDLM ASTs
+		that are handled with a special server thread on the client.
+		This is a way for the servers to ask the clients not to time
+		out the request that reached current servicing time estimate
+		yet and give it some more time.
+		Default: 30
+
+What:		/sys/fs/lustre/at_early_margin
+Date:		July 2015
+Contact:	"Oleg Drokin" <oleg.drokin@intel.com>
+Description:
+		Controls when to send the early reply for requests that are
+		about to timeout as an offset to the estimated service time in
+		seconds..
+		Default: 5
+
+What:		/sys/fs/lustre/at_history
+Date:		July 2015
+Contact:	"Oleg Drokin" <oleg.drokin@intel.com>
+Description:
+		Controls for how many seconds to remember slowest events
+		encountered by adaptive timeouts code.
+		Default: 600
+
 What:		/sys/fs/lustre/llite/<fsname>-<uuid>/blocksize
 Date:		May 2015
 Contact:	"Oleg Drokin" <oleg.drokin@intel.com>
-- 
2.1.0


  parent reply	other threads:[~2015-07-06 16:49 UTC|newest]

Thread overview: 33+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2015-07-06 16:48 [PATCH 00/20] Lustre: final procfs bits removal green
2015-07-06 16:48 ` [PATCH 01/20] staging/lustre/lov: Move target sysfs symlink removal to object freeing green
2015-07-06 16:48 ` [PATCH 02/20] staging/lustre: make ldebugfs_remove recursive green
2015-07-06 16:48 ` [PATCH 03/20] staging/lustre/ldlm: In ldlm_pools_fini make sure there was init first green
2015-07-06 16:48 ` [PATCH 04/20] staging/lustre/obdclass: fix class_procfs_init error return value green
2015-07-06 16:48 ` [PATCH 05/20] staging/lustre: remove alloc_fail_rate sysctl green
2015-07-06 16:48 ` [PATCH 06/20] staging/lustre: Remove now obsolete memory tracking sysctls green
2015-07-06 16:48 ` [PATCH 07/20] staging/lustre: Remove unneeded ldlm_timeout control green
2015-07-06 16:48 ` [PATCH 08/20] staging/lustre/obdclass: move sysctl timeout to sysfs green
2015-07-06 16:48 ` [PATCH 09/20] staging/lustre/obdclass: move max_dirty_mb from sysctl " green
2015-07-06 16:48 ` [PATCH 10/20] staging/lustre/obdclass: move debug controls " green
2015-07-06 16:48 ` green [this message]
2015-07-06 16:48 ` [PATCH 12/20] staging/lustre: Get rid of remaining /proc/sys/lustre plumbing green
2015-07-06 16:48 ` [PATCH 13/20] staging/lustre/libcfs: move /proc/sys/lnet to debugfs green
2015-07-06 16:48 ` [PATCH 14/20] staging/lustre/libcfs: Remove redundant lnet debugfs variables green
2015-07-06 16:48 ` [PATCH 15/20] staging/lustre/libcfs: get rid of debugfs/lnet/console_backoff green
2015-07-08  8:28   ` Dan Carpenter
2015-07-08  8:30     ` Dan Carpenter
2015-07-08 13:43     ` Oleg Drokin
2015-07-06 16:48 ` [PATCH 16/20] staging/lustre/libcfs: Remove redundant enums and sysctl moduleparams green
2015-07-06 16:48 ` [PATCH 17/20] staging/lustre/libcfs: Remove unneeded lnet watchdog_ratelimit sysctl green
2015-07-06 16:48 ` [PATCH 18/20] staging/lustre/libcfs: get rid of debugfs/lnet/debug_mb green
2015-07-08  8:45   ` Dan Carpenter
2015-07-14  2:43     ` Greg Kroah-Hartman
2015-07-14  2:45       ` Oleg Drokin
2015-07-14  3:04         ` Greg Kroah-Hartman
2015-07-14  2:46   ` Greg Kroah-Hartman
2015-07-14  2:51     ` Oleg Drokin
2015-07-14  3:52       ` Greg Kroah-Hartman
2015-07-14  2:49   ` Greg Kroah-Hartman
2015-07-14  2:52     ` Oleg Drokin
2015-07-06 16:48 ` [PATCH 19/20] staging/lustre/libcfs: get rid of debugfs/lnet/console_{min,max}_delay_centisecs green
2015-07-06 16:48 ` [PATCH 20/20] staging/lustre/libcfs: remove unused portal_enter_debugger variable green

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=1436201338-14263-12-git-send-email-green@linuxhacker.ru \
    --to=green@linuxhacker.ru \
    --cc=andreas.dilger@intel.com \
    --cc=devel@driverdev.osuosl.org \
    --cc=gregkh@linuxfoundation.org \
    --cc=linux-kernel@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