cluster-devel.redhat.com archive mirror
 help / color / mirror / Atom feed
* [Cluster-devel] cluster/gfs-kernel/src/gfs super.c
@ 2006-08-02  1:28 wcheng
  0 siblings, 0 replies; 5+ messages in thread
From: wcheng @ 2006-08-02  1:28 UTC (permalink / raw)
  To: cluster-devel.redhat.com

CVSROOT:	/cvs/cluster
Module name:	cluster
Branch: 	RHEL4
Changes by:	wcheng at sourceware.org	2006-08-02 01:28:57

Modified files:
	gfs-kernel/src/gfs: super.c 

Log message:
	Bugzilla 199984: Increasing gt_statfs_slots tunable could significantly
	boost gfs "df" performance; e.g. set it to 128 from current default (64)
	could cut the "df" wall time in half with larger filesystem size. However,
	the kmalloc call within stat_gfs_async() has the possibility to fail due
	to increased gt_statfs_slots. There is really no need for this array to
	be on a piece of contiguous memory. Switch to vmalloc().

Patches:
http://sourceware.org/cgi-bin/cvsweb.cgi/cluster/gfs-kernel/src/gfs/super.c.diff?cvsroot=cluster&only_with_tag=RHEL4&r1=1.12.2.3&r2=1.12.2.4

--- cluster/gfs-kernel/src/gfs/super.c	2006/02/15 17:19:22	1.12.2.3
+++ cluster/gfs-kernel/src/gfs/super.c	2006/08/02 01:28:57	1.12.2.4
@@ -840,7 +840,7 @@
 
 	memset(sg, 0, sizeof(struct gfs_stat_gfs));
 
-	gha = kmalloc(slots * sizeof(struct gfs_holder), GFP_KERNEL);
+	gha = vmalloc(slots * sizeof(struct gfs_holder));
 	if (!gha)
 		return -ENOMEM;
 	memset(gha, 0, slots * sizeof(struct gfs_holder));
@@ -884,7 +884,7 @@
 		yield();
 	}
 
-	kfree(gha);
+	vfree(gha);
 
 	return error;
 }



^ permalink raw reply	[flat|nested] 5+ messages in thread

* [Cluster-devel] cluster/gfs-kernel/src/gfs super.c
@ 2006-08-02  1:37 wcheng
  0 siblings, 0 replies; 5+ messages in thread
From: wcheng @ 2006-08-02  1:37 UTC (permalink / raw)
  To: cluster-devel.redhat.com

CVSROOT:	/cvs/cluster
Module name:	cluster
Changes by:	wcheng at sourceware.org	2006-08-02 01:37:17

Modified files:
	gfs-kernel/src/gfs: super.c 

Log message:
	Bugzilla 199984: Increasing gt_statfs_slots tunable could significantly
	boost gfs "df" performance; e.g. set it to 128 from current default (64)
	could cut the "df" wall time in half with larger filesystem size. However,
	the kmalloc call within stat_gfs_async() has the possibility to fail due
	to increased gt_statfs_slots. There is really no need for this array to
	be on a piece of contiguous memory. Switch to vmalloc().

Patches:
http://sourceware.org/cgi-bin/cvsweb.cgi/cluster/gfs-kernel/src/gfs/super.c.diff?cvsroot=cluster&r1=1.19&r2=1.20

--- cluster/gfs-kernel/src/gfs/super.c	2006/07/11 17:29:24	1.19
+++ cluster/gfs-kernel/src/gfs/super.c	2006/08/02 01:37:16	1.20
@@ -840,7 +840,7 @@
 
 	memset(sg, 0, sizeof(struct gfs_stat_gfs));
 
-	gha = kmalloc(slots * sizeof(struct gfs_holder), GFP_KERNEL);
+	gha = vmalloc(slots * sizeof(struct gfs_holder));
 	if (!gha)
 		return -ENOMEM;
 	memset(gha, 0, slots * sizeof(struct gfs_holder));
@@ -884,7 +884,7 @@
 		yield();
 	}
 
-	kfree(gha);
+	vfree(gha);
 
 	return error;
 }



^ permalink raw reply	[flat|nested] 5+ messages in thread

* [Cluster-devel] cluster/gfs-kernel/src/gfs super.c
@ 2006-08-09 20:28 rpeterso
  0 siblings, 0 replies; 5+ messages in thread
From: rpeterso @ 2006-08-09 20:28 UTC (permalink / raw)
  To: cluster-devel.redhat.com

CVSROOT:	/cvs/cluster
Module name:	cluster
Changes by:	rpeterso at sourceware.org	2006-08-09 20:28:51

Modified files:
	gfs-kernel/src/gfs: super.c 

Log message:
	Fix compile error with vmalloc.

Patches:
http://sourceware.org/cgi-bin/cvsweb.cgi/cluster/gfs-kernel/src/gfs/super.c.diff?cvsroot=cluster&r1=1.20&r2=1.21

--- cluster/gfs-kernel/src/gfs/super.c	2006/08/02 01:37:16	1.20
+++ cluster/gfs-kernel/src/gfs/super.c	2006/08/09 20:28:51	1.21
@@ -18,6 +18,7 @@
 #include <asm/semaphore.h>
 #include <linux/completion.h>
 #include <linux/buffer_head.h>
+#include <linux/vmalloc.h>
 
 #include "gfs.h"
 #include "dio.h"



^ permalink raw reply	[flat|nested] 5+ messages in thread

* [Cluster-devel] cluster/gfs-kernel/src/gfs super.c
@ 2006-09-20  4:45 wcheng
  0 siblings, 0 replies; 5+ messages in thread
From: wcheng @ 2006-09-20  4:45 UTC (permalink / raw)
  To: cluster-devel.redhat.com

CVSROOT:	/cvs/cluster
Module name:	cluster
Branch: 	RHEL4
Changes by:	wcheng at sourceware.org	2006-09-20 04:45:48

Modified files:
	gfs-kernel/src/gfs: super.c 

Log message:
	Bugzilla 199984: This kmalloc to vmalloc change seems to cause panic in
	x86_64 machine. Cause unknown but will reset back until
	the problem is figured out.

Patches:
http://sourceware.org/cgi-bin/cvsweb.cgi/cluster/gfs-kernel/src/gfs/super.c.diff?cvsroot=cluster&only_with_tag=RHEL4&r1=1.12.2.4&r2=1.12.2.5

--- cluster/gfs-kernel/src/gfs/super.c	2006/08/02 01:28:57	1.12.2.4
+++ cluster/gfs-kernel/src/gfs/super.c	2006/09/20 04:45:48	1.12.2.5
@@ -840,7 +840,7 @@
 
 	memset(sg, 0, sizeof(struct gfs_stat_gfs));
 
-	gha = vmalloc(slots * sizeof(struct gfs_holder));
+	gha = kmalloc(slots * sizeof(struct gfs_holder), GFP_KERNEL);
 	if (!gha)
 		return -ENOMEM;
 	memset(gha, 0, slots * sizeof(struct gfs_holder));
@@ -884,7 +884,7 @@
 		yield();
 	}
 
-	vfree(gha);
+	kfree(gha);
 
 	return error;
 }



^ permalink raw reply	[flat|nested] 5+ messages in thread

* [Cluster-devel] cluster/gfs-kernel/src/gfs super.c
@ 2007-02-06 22:08 wcheng
  0 siblings, 0 replies; 5+ messages in thread
From: wcheng @ 2007-02-06 22:08 UTC (permalink / raw)
  To: cluster-devel.redhat.com

CVSROOT:	/cvs/cluster
Module name:	cluster
Branch: 	RHEL4
Changes by:	wcheng at sourceware.org	2007-02-06 22:08:37

Modified files:
	gfs-kernel/src/gfs: super.c 

Log message:
	Bugzilla 220622:
	
	Allow vmalloc in stat_gfs_async() (as in RHEL5) to alleviate statfs
	performance issue. This should allow statfs_slot tunable to take
	more asynchronous locks. Hopefully "df" and "ls" could run a little
	bit faster with this.

Patches:
http://sourceware.org/cgi-bin/cvsweb.cgi/cluster/gfs-kernel/src/gfs/super.c.diff?cvsroot=cluster&only_with_tag=RHEL4&r1=1.12.2.6&r2=1.12.2.7

--- cluster/gfs-kernel/src/gfs/super.c	2007/01/22 07:43:52	1.12.2.6
+++ cluster/gfs-kernel/src/gfs/super.c	2007/02/06 22:08:36	1.12.2.7
@@ -18,6 +18,7 @@
 #include <asm/semaphore.h>
 #include <linux/completion.h>
 #include <linux/buffer_head.h>
+#include <linux/vmalloc.h>
 
 #include "gfs.h"
 #include "dio.h"
@@ -841,7 +842,7 @@
 
 	memset(sg, 0, sizeof(struct gfs_stat_gfs));
 
-	gha = kmalloc(slots * sizeof(struct gfs_holder), GFP_KERNEL);
+	gha = vmalloc(slots * sizeof(struct gfs_holder));
 	if (!gha)
 		return -ENOMEM;
 	memset(gha, 0, slots * sizeof(struct gfs_holder));
@@ -885,7 +886,7 @@
 		yield();
 	}
 
-	kfree(gha);
+	vfree(gha);
 
 	return error;
 }



^ permalink raw reply	[flat|nested] 5+ messages in thread

end of thread, other threads:[~2007-02-06 22:08 UTC | newest]

Thread overview: 5+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2006-08-02  1:37 [Cluster-devel] cluster/gfs-kernel/src/gfs super.c wcheng
  -- strict thread matches above, loose matches on Subject: below --
2007-02-06 22:08 wcheng
2006-09-20  4:45 wcheng
2006-08-09 20:28 rpeterso
2006-08-02  1:28 wcheng

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).