[SSI] openssi ChangeLog,1.264,1.265

Roger Tsang <[email protected]> Tue, 25 Jan 2011 06:16:08 +0000
Newsgroups gmane.linux.cluster.ssic.cvs
Message-ID <[email protected]>
Update of /cvsroot/ssic-linux/openssi
In directory sfp-cvsdas-3.v30.ch3.sourceforge.com:/tmp/cvs-serv26818

Modified Files:
      Tag: OPENSSI-FC
	ChangeLog 
Log Message:
update


Index: ChangeLog
===================================================================
RCS file: /cvsroot/ssic-linux/openssi/ChangeLog,v
retrieving revision 1.264
retrieving revision 1.265
diff -u -d -r1.264 -r1.265
--- ChangeLog	17 Jan 2011 07:23:28 -0000	1.264
+++ ChangeLog	25 Jan 2011 06:16:05 -0000	1.265
@@ -1,5 +1,79 @@
 2010/??/??	*** Release of OpenSSI 2.0.0rc1 ***
 
+2011/01/25	rogertsang <[email protected]>
+	* CVS Tags: OPENSSI-FC-1-9-6-PRE69
+
+	* Token Architecture:
+	cluster/ssi/token/tokseq.c
+	- insert_queue: always inline.
+	- tokseq_accept: indicate to compiler unlikely branch direction.
+	- tokseq_accept: access tqentry structure outside critical section.
+	  reduce contention over ts_lock in tokseq structure.
+	- tokseq_accept: reduce stack usage in local scope.
+	- tokseq_accept: immediately process items with seqnum 0 which could be
+	  items created while TSF_DOWN flag was set. we would see these items
+	  if we lost the race with tokseq_cleanup(). fix items stuck in queue.
+
+	* Remote File Block:
+	cluster/ssi/util/rmtfb.c
+	- __rmtfb_getcli_id: add comment before function.
+	- __rmtfb_getcli_id: use newly defined states for rfb_state field in
+	  rmtfb_cli structure.
+	- rmtfb_newcli: use newly defined states for rfb_state field in
+	  rmtfb_cli structure.
+
+	include/cluster/ssi/util/rmtfb.h
+	- define states RFB_READY and RFB_FAIL for rfb_state field in rmtfb_cli
+	  structure.
+
+	 cluster/ssi/token/tokseq.c       |   89 +++++++++++++++----------------
+	 cluster/ssi/util/rmtfb.c         |   25 ++++++--
+	 include/cluster/ssi/util/rmtfb.h |   10 ++-
+	 3 files changed, 70 insertions(+), 54 deletions(-)
+
+2011/01/25	rogertsang <[email protected]>
+	* CVS Tags: OPENSSI-FC-1-9-6-PRE68
+
+	* CFS:
+	cluster/ssi/cfs/cfs_chard.c
+	- cfs_sb_soft_down: purge caches before calling process_cnodes() to
+	  reduce amount of work.
+	- cfs_sb_start_down: remove unused inodes on hard mount to reduce
+	  amount of work during rebuild.
+
+	* Token Architecture:
+	cluster/ssi/token/tokseq.c
+	- tokseq_accept: fix memory leak if node goes down before we re-acquire
+	  ts_lock.
+
+	* Remote File Block:
+	cluster/ssi/util/rmtfb.c (#ifdef RMTFB_HASH_LOCKLESS)
+	- rmtfb_newcli: call rmtfb_remote_tty_ino() and update fields in file
+	  structure outside critical section. reduce rmtfb_newcli_lock
+	  contention.
+
+	* CFS supporting code:
+	fs/inode.c (#ifdef CONFIG_CFS)
+	- process_cnodes: traverse s_inodes list in super_block structure.
+	  avoid traversing other lists that contain inodes from other super
+	  blocks. reduce work.
+
+	fs/inode.c (#ifdef SHRINK_ICACHE_SB)
+	include/linux/dcache.h (#ifdef SHRINK_ICACHE_SB)
+	- shrink_icache_sb: remove function. not needed since during
+	  invalidate_inodes() if i_count is zero the inode cannot be LOCKed.
+
+	include/linux/config.h (#ifdef SHRINK_ICACHE_SB)
+	- undefine SHRINK_ICACHE_SB preprocessor macro. no longer used.
+
+	 cluster/ssi/cfs/cfs_chard.c |   29 +++++-------
+	 cluster/ssi/token/tokseq.c  |    8 +++
+	 cluster/ssi/util/rmtfb.c    |   28 ++++--------
+	 fs/inode.c                  |   83 ------------------------------------
+	 include/linux/config.h      |    1 
+	 include/linux/dcache.h      |    7 ---
+	 6 files changed, 31 insertions(+), 125 deletions(-)
+
 2011/01/17	rogertsang <[email protected]>
 	* CVS Tags: OPENSSI-FC-1-9-6-PRE67
 
@@ -18,10 +92,10 @@
 	  struct with same rfb_id. caller already checked. in one possible code
 	  path we were doing total of three checks.
 	- rmtfb_newcli: remove retry of __rmtfb_getcli_id(). SMP-safe.
-	- rmtfb_newcli: move rmtfb_setcli_re/open() till after we've added the
-	  rmtfb_cli structure to the hashed chained linked list to avoid racing
-	  with rmtfb_rclose(). fix -ERFB_TRYAGAIN error when rmtfb_rclose()
-	  lost the race.
+	- rmtfb_newcli: delay rmtfb_setcli_re/open() till after we've added the
+	  rmtfb_cli structure to the hash chained linked list to avoid racing
+	  with rmtfb_rclose(). fix -ERFB_TRYAGAIN error on next rmtfb_rclose()
+	  invocation when rmtfb_rclose() lost the race.
 	- rmtfb_newcli: add comment before function about circular reference
 	  between file struct and rmtfb_cli struct.
 	- __rmtfb_getcli_id: no need to return -ERFB_RELEASE error. just return
@@ -76,7 +150,7 @@
 	- proc_pid_make_inode: reduce redundant code.
 	- proc_pident_lookup, proc_lookupfd: No need to acquire VPROC MOVEMENT
 	  LOCK for revised proc_pid_make_inode().
-	- proc_task_readdir: no need for seperate code path for local listing.
+	- proc_task_readdir: no need for separate code path for local listing.
 	  remove redundant.
 	- vproc_migrate_local: no need to acquire VPROC MOVEMENT LOCK since
 	  setup_execnode_move() will check move header list for existing
@@ -91,7 +165,7 @@
 	  ensures PROC_I()->task is valid.
 
 	fs/proc/internal.h
-	- vproc_migrating: moved inline function from fs/proc/base.c
+	- vproc_migrating: moved inline function from fs/proc/base.c here.
 
 	fs/proc/internal.h (#ifdef VPROC_HOLD_ZERO_GET_TASK)
 	- proc_ssi_task: borrow pvp_pproc's task_struct reference while


------------------------------------------------------------------------------
Special Offer-- Download ArcSight Logger for FREE (a $49 USD value)!
Finally, a world-class log management solution at an even better price-free!
Download using promo code Free_Logger_4_Dev2Dev. Offer expires 
February 28th, so secure your free ArcSight Logger TODAY! 
http://p.sf.net/sfu/arcsight-sfd2d