[Cluster-devel] GFS2: Pull request (merge window)

Hi, Please consider pulling the following changes for the GFS2 file system. Bob Peterson The following changes since commit 9dffdb38d864ae89e16ff7b3a09451270736e35b: Merge tag 'staging-4.6-rc1' of

[Cluster-devel] [PATCH 4/9] GFS2: fs/gfs2/glock.c: Deinline do_error, save 1856 bytes

From: Denys Vlasenko This function compiles to 522 bytes of machine code. Error paths are not very time critical. Signed-off-by: Denys Vlasenko Signed-off-by: Bob Peterson --- fs/gfs2/glock.c | 2 +- 1 file changed, 1

[Cluster-devel] [PATCH 1/9] GFS2: ignore unlock failures after withdraw

From: Benjamin Marzinski After gfs2 has withdrawn the filesystem, it may still have many locks not in the unlocked state. If it is using lock_dlm, it will failed trying the unlocks since it has already unmounted the lock manager. Instead, it should set the

[Cluster-devel] [PATCH 3/9] gfs2: Use gfs2 wrapper to sync inode before calling generic_file_splice_read()

From: Abhi Das gfs2_file_splice_read() f_op grabs and releases the cluster-wide inode glock to sync the inode size to the latest. Without this, generic_file_splice_read() uses an older i_size value and can return EOF for valid offsets in the inode. Signed-off-by: Abhi Das

[Cluster-devel] [PATCH 8/9] GFS2: Remove allocation parms from gfs2_rbm_find

Struct gfs2_alloc_parms ap is never referenced in function gfs2_rbm_find, so this patch removes it. Signed-off-by: Bob Peterson --- fs/gfs2/rgrp.c | 16 ++-- 1 file changed, 6 insertions(+), 10 deletions(-) diff --git a/fs/gfs2/rgrp.c b/fs/gfs2/rgrp.c index

[Cluster-devel] [PATCH 2/9] GFS2: Get rid of dead code in inode_go_demote_ok

Function inode_go_demote_ok had some code that was only executed if gl_holders was not empty. However, if gl_holders was not empty, the only caller, demote_ok(), returns before inode_go_demote_ok would ever be called. Therefore, it's dead code, so I removed it. Signed-off-by: Bob Peterson

[Cluster-devel] [PATCH 9/9] GFS2: Refactor gfs2_remove_from_journal

This patch makes two simple changes to function gfs2_remove_from_journal. First, it removes the parameter that specifies the transaction. Since it's always passed in as current->journal_info, we might as well set that in the function rather than passing it in. Second, it changes the meta parameter

[Cluster-devel] [PATCH 0/9] GFS2: Pre-pull patch posting (merge window)

Hi, We've got nine patches this time: - Abhi Das has two patches that fix a GFS2 splice issue (and an adjustment). - Ben Marzinski has a patch which allows the proper unmount of a GFS2 file system after hitting a withdraw error. - I have a patch to fix a problem where GFS2 would dereference an

[Cluster-devel] [PATCH 5/9] GFS2: Don't dereference inode in gfs2_inode_lookup until it's valid

Function gfs2_inode_lookup was dereferencing the inode, and after, it checks for the value being NULL. We need to check that first. Signed-off-by: Bob Peterson --- fs/gfs2/inode.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/fs/gfs2/inode.c

[Cluster-devel] [PATCH 7/9] gfs2: use inode_lock/unlock instead of accessing i_mutex directly

From: Abhi Das i_mutex has been replaced by i_rwsem and directly accessing the non-existent i_mutex breaks the kernel build. Signed-off-by: Abhi Das Signed-off-by: Bob Peterson --- fs/gfs2/file.c | 6 +++--- 1 file changed, 3

[Cluster-devel] [PATCH 6/9] GFS2: Add calls to gfs2_holder_uninit in two error handlers

From: Daniel DeFreez This patch fixes two locations that do not call gfs2_holder_uninit if gfs2_glock_nq returns an error. Signed-off-by: Daniel DeFreez Signed-off-by: Bob Peterson --- fs/gfs2/aops.c | 3 ++- fs/gfs2/file.c |

Re: [Cluster-devel] [DLM PATCH] dlm_controld: outputs explicit info about stateful

Hi David, On 05/19/2016 02:50 AM, David Teigland wrote: On Wed, May 18, 2016 at 02:53:00PM +0800, Eric Ren wrote: Q1: what's stateful merged node? Q2: what if we add the stateful merged nodes to dlm_controld daemon cpg instead of fencing them? The details here are fundamental to the way

[Cluster-devel] [DLM PATCH] dlm_controld: outputs explicit info about stateful merging

When there are 3 or more partitions that merge, none may see enough clean nodes. Therefore, DLM would be stuck there forever unitl administrator manually reset/restart enough nodes to produce sufficient clean nodes. Therefore, output explicit information for higher code (e.g. pcmk) about the