rpms/kernel/F-10 linux-2.6-kjournald-use-rt-io-priority.patch, 1.1, 1.2

[Chuck Ebbert]

Author: cebbert

Update of /cvs/pkgs/rpms/kernel/F-10
In directory cvs1.fedora.phx.redhat.com:/tmp/cvs-serv10594

Modified Files:
	linux-2.6-kjournald-use-rt-io-priority.patch 
Log Message:
dont use real-time priority for kjournald

linux-2.6-kjournald-use-rt-io-priority.patch:

Index: linux-2.6-kjournald-use-rt-io-priority.patch
===================================================================
RCS file: /cvs/pkgs/rpms/kernel/F-10/linux-2.6-kjournald-use-rt-io-priority.patch,v
retrieving revision 1.1
retrieving revision 1.2
diff -u -r1.1 -r1.2
--- linux-2.6-kjournald-use-rt-io-priority.patch	28 Mar 2009 01:38:24 -0000	1.1
+++ linux-2.6-kjournald-use-rt-io-priority.patch	24 Apr 2009 03:30:31 -0000	1.2
@@ -20,9 +20,10 @@
 this priority inversion behavior. With this patch, the latencies for atime
 updates (and similar operation) go down by a factor of 3x to 4x !
 
-[cebbert: changed to lowest real-time priority, fix up for 2.6.29]
+[cebbert: changed to highest normal priority, fix up for 2.6.29]
 
 Signed-off-by: Arjan van de Ven <arjan at linux.intel.com>
+Signed-off-by: Chuck Ebbert <cebbert at redhat.com>
 ---
  fs/jbd/journal.c       |   12 ++++++++++++
  1 files changed, 12 insertions(+), 0 deletions(-)
@@ -45,12 +46,12 @@
 +	 * kjournald is the process on which most other processes depend on
 +	 * for doing the filesystem portion of their IO. As such, there exists
 +	 * the equivalent of a priority inversion situation, where kjournald
-+	 * would get less priority as it should.
++	 * would get less priority than it should.
 +	 *
-+	 * For this reason we set to "lowest real time priority", which is higher
-+	 * than regular tasks, but not infinitely powerful.
++	 * For this reason we set to "highest normal priority", which is higher
++	 * than regular tasks, but not real-time.
 +	 */
-+	set_task_ioprio(current, IOPRIO_PRIO_VALUE(IOPRIO_CLASS_RT, 7));
++	set_task_ioprio(current, IOPRIO_PRIO_VALUE(IOPRIO_CLASS_BE, 0));
 +
 +	/*
  	 * And now, wait forever for commit wakeup events.