[Cluster-devel] [GFS2 PATCH] GFS2: Re-add a call to log_flush_wait when flushing the journal

2014-03-12 Thread Bob Peterson
to infinite loops in journal replay, both in the kernel code and fsck.gfs2 code. This patch re-adds that call. Regards, Bob Peterson Red Hat File Systems Signed-off-by: Bob Peterson rpete...@redhat.com --- diff --git a/fs/gfs2/log.c b/fs/gfs2/log.c index edbd461..4a14d50 100644 --- a/fs/gfs2

[Cluster-devel] [GFS2 PATCH] [TRY #2] GFS2: inline function gfs2_set_mode

2014-03-19 Thread Bob Peterson
Hi, Here is a revised patch based on Steve's feedback: This patch eliminates function gfs2_set_mode which was only called in one place, and always returned 0. Regards, Bob Peterson Red Hat File Systems Signed-off-by: Bob Peterson rpete...@redhat.com --- diff --git a/fs/gfs2/acl.c b/fs/gfs2

[Cluster-devel] [GFS2 PATCH] [TRY #2] GFS2: Prevent recovery before the local journal is set

2014-05-02 Thread Bob Peterson
Hi, This patch uses a completion to prevent dlm's recovery process from referencing and trying to recover a journal before a journal has been opened. Regards, Bob Peterson Red Hat File Systems Signed-off-by: Bob Peterson rpete...@redhat.com --- diff --git a/fs/gfs2/incore.h b/fs/gfs2/incore.h

Re: [Cluster-devel] [PATCH 2/5] gfs2-utils: Expressly expunge 'expert mode'

2014-05-07 Thread Bob Peterson
test conditions, like very small resource groups. I don't think Nate uses this in any of his tests, but it might be worth asking. I've used it to try to recreate certain failing scenarios, but I haven't needed it in years. Are we sure we want to get rid of it? Regards, Bob Peterson Red Hat File

[Cluster-devel] [gfs2-utils patch] gfs2_edit: Print block types with log descriptors

2014-06-11 Thread Bob Peterson
in 0x70dc1a7 in 0x70eaf7a in 0x70dbfa9 in 0x70ead7c in In this case, in indicates indirect blocks, and rb indicates a rgrp bitmap. This is useful for getting a bigger picture of what that journal descriptor contains and represents. Regards, Bob Peterson Red Hat File Systems Signed

[Cluster-devel] [gfs2-utils patch] gfs2_edit: print LB (log descriptor continuation blocks) for GFS2

2014-06-11 Thread Bob Peterson
Hi, This patch allows gfs2_edit to print metadata blocks of type LB (log descriptor continuation blocks) for GFS2. Prior to this, only GFS1's continuation blocks would print. Regards, Bob Peterson Red Hat File Systems Signed-off-by: Bob Peterson rpete...@redhat.com --- diff --git a/gfs2/edit

[Cluster-devel] [GFS2 PATCH] GFS2: Only wait for demote when last holder is dequeued

2014-06-20 Thread Bob Peterson
is especially apparent when dealing with nested flocks. Regards, Bob Peterson Red Hat File Systems Signed-off-by: Bob Peterson rpete...@redhat.com --- diff --git a/fs/gfs2/glock.c b/fs/gfs2/glock.c index aec7f73..22ce2f0 100644 --- a/fs/gfs2/glock.c +++ b/fs/gfs2/glock.c @@ -1128,7 +1128,9 @@ void

[Cluster-devel] [GFS2 PATCH] GFS2: Allow caching of glocks for flock

2014-06-26 Thread Bob Peterson
where the flocks are held in Shared (SH) mode. Regards, Bob Peterson Red Hat File Systems Signed-off-by: Bob Peterson rpete...@redhat.com --- diff --git a/fs/gfs2/file.c b/fs/gfs2/file.c index 4fc3a30..0740a57 100644 --- a/fs/gfs2/file.c +++ b/fs/gfs2/file.c @@ -981,7 +981,7 @@ static int

[Cluster-devel] [GFS2 PATCH] GFS2: Allow caching of glocks for flock

2014-06-26 Thread Bob Peterson
where the flocks are held in Shared (SH) mode. Regards, Bob Peterson Red Hat File Systems Signed-off-by: Bob Peterson rpete...@redhat.com --- diff --git a/fs/gfs2/file.c b/fs/gfs2/file.c index 4fc3a30..0740a57 100644 --- a/fs/gfs2/file.c +++ b/fs/gfs2/file.c @@ -981,7 +981,7 @@ static int

Re: [Cluster-devel] [RFC PATCH] dlm: Remove unused conf from lm_grant

2014-07-01 Thread Bob Peterson
/ holiday this week, but he will be back on Thursday and Friday (which is a holiday). I'm all for getting rid of useless parameters, and I've done so on several occasions in GFS2. Regards, Bob Peterson Red Hat File Systems

[Cluster-devel] [gfs2-utils PATCH] fsck.gfs2: File read-ahead

2014-07-10 Thread Bob Peterson
Hi, This patch introduces file read-ahead to pass1. Regards, Bob Peterson Red Hat File Systems Signed-off-by: Bob Peterson rpete...@redhat.com --- gfs2/fsck/metawalk.c | 65 ++-- gfs2/fsck/metawalk.h | 1 + gfs2/fsck/pass1.c| 1 + 3 files

Re: [Cluster-devel] [gfs2-utils PATCH] fsck.gfs2: time each of the passes

2014-07-11 Thread Bob Peterson
- Original Message - Hi, On 11/07/14 18:41, Bob Peterson wrote: - Original Message - (snip) Use timersub() here perhaps? Otherwise looks good, Steve. Hi Steve, Thanks for the suggestion. How about this version? Yes, that looks better. There is probably

[Cluster-devel] [GFS2 PATCH] GFS2: Change maxlen variables to size_t

2014-08-06 Thread Bob Peterson
Hi, This patch changes some variables (especially maxlen in function gfs2_block_map) from unsigned int to size_t. We need 64-bit arithmetic for very large files (e.g. 1PB) where the variables otherwise get shifted to all 0's. Regards, Bob Peterson Red Hat File Systems Signed-off-by: Bob

[Cluster-devel] [GFS2 PATCH] GFS2: Request demote when a try flock fails

2014-08-20 Thread Bob Peterson
. Then it tries again, for an increasing amount of time. Subsequent attempts to gain the try lock don't use _1CB so that only one callback is issued. Regards, Bob Peterson Red Hat File Systems Signed-off-by: Bob Peterson rpete...@redhat.com --- diff --git a/fs/gfs2/file.c b/fs/gfs2/file.c index

Re: [Cluster-devel] [PATCH 00/19] gfs2-utils: Introduce extent allocation and speed up journal creation

2014-09-02 Thread Bob Peterson
, running fsck.gfs2, and seeing if it can properly repair it. Another concern is gfs2_convert. I don't know if it makes any assumptions about the master directory, but it's much less likely. I think it just assumes the file system is healthy. But fsck.gfs2 is a concern. Regards, Bob Peterson Red

[Cluster-devel] [GFS2 PATCH] GFS2: call d_add for a NULL entry when an inode lookup fails

2014-09-09 Thread Bob Peterson
. Regards, Bob Peterson Red Hat File Systems Signed-off-by: Bob Peterson rpete...@redhat.com --- diff --git a/fs/gfs2/inode.c b/fs/gfs2/inode.c index e62e594..9317ddc 100644 --- a/fs/gfs2/inode.c +++ b/fs/gfs2/inode.c @@ -840,8 +840,10 @@ static struct dentry *__gfs2_lookup(struct inode *dir, struct

[Cluster-devel] [GFS2 PATCH] GFS2: Make rename not save dirent location

2014-09-29 Thread Bob Peterson
is released as per previous behavior. Regards, Bob Peterson Red Hat File Systems Signed-off-by: Bob Peterson rpete...@redhat.com --- diff --git a/fs/gfs2/dir.c b/fs/gfs2/dir.c index 1a349f9..5d4261f 100644 --- a/fs/gfs2/dir.c +++ b/fs/gfs2/dir.c @@ -2100,8 +2100,13 @@ int gfs2_diradd_alloc_required

[Cluster-devel] [DLM PATCH] DLM: Don't wait for resource library lookups if NOLOOKUP is specified

2014-10-01 Thread Bob Peterson
of local locks distributes the locks between the nodes (at least until nodes enter or leave the cluster), which tends to make each node keep to itself when doing allocations. Thus, dlm communications are kept to a minimum, which results in significantly faster block allocations. Regards, Bob Peterson

Re: [Cluster-devel] [DLM PATCH] DLM: Don't wait for resource library lookups if NOLOOKUP is specified

2014-10-03 Thread Bob Peterson
patches depending on those results. Regards, Bob Peterson Red Hat File Systems

Re: [Cluster-devel] [DLM PATCH] DLM: Don't wait for resource library lookups if NOLOOKUP is specified

2014-10-03 Thread Bob Peterson
/rgrp.c?h=RHEL4 Regards, Bob Peterson Red Hat File Systems

[Cluster-devel] [GFS2 PATCH] GFS2: Change minimum reservation size to 64 blocks

2014-10-04 Thread Bob Peterson
FILES = 735760 Patch description: The minimum block reservation size was 32 blocks. This patch doubles it to 64 blocks. Regards, Bob Peterson Red Hat File Systems Signed-off-by: Bob Peterson rpete...@redhat.com --- diff --git a/fs/gfs2/rgrp.h b/fs/gfs2/rgrp.h index 5d8f085..e2058a7 100644

[Cluster-devel] [GFS2 PATCH] GFS2: Block reservation doubling scheme

2014-10-08 Thread Bob Peterson
it's now adaptive, but before I did that, I'd have to run some performance tests. What do you think? Regards, Bob Peterson Red Hat File Systems Patch text: This patch introduces a new block reservation doubling scheme. If we get

[Cluster-devel] [GFS2 PATCH] GFS2: Set of distributed preferences for rgrps

2014-10-08 Thread Bob Peterson
35sec Run 5 time: 2hr 5min 55sec Regards, Bob Peterson Red Hat File Systems Signed-off-by: Bob Peterson rpete...@redhat.com --- diff --git a/fs/gfs2/incore.h b/fs/gfs2/incore.h index f98fa37..c9f4f4c 100644 --- a/fs/gfs2/incore.h +++ b/fs/gfs2/incore.h @@ -97,6 +97,7 @@ struct gfs2_rgrpd { #define

Re: [Cluster-devel] [GFS2 PATCH] GFS2: Block reservation doubling scheme

2014-10-09 Thread Bob Peterson
COUNT FOR OUTPUT FILES = 341769 EXTENT COUNT FOR OUTPUT FILES = 348676 EXTENT COUNT FOR OUTPUT FILES = 348079 So I'm looking for opinions: (a) Stick with the original reservation doubling patch, or (b) Go with the additive version. (c) Any other ideas? Regards, Bob Peterson Red Hat File Systems

Re: [Cluster-devel] [GFS2 PATCH] GFS2: Block reservation doubling scheme

2014-10-14 Thread Bob Peterson
for my tests. Regards, Bob Peterson Red Hat File Systems

[Cluster-devel] [GFS2 PATCH] GFS2: Speed up fiemap function by skipping holes

2014-10-14 Thread Bob Peterson
on the current metapath buffers. Therefore, fiemap may call blockmap several times to get the hole size. The hole size is determined by a new function. Regards, Bob Peterson Red Hat File Systems Signed-off-by: Bob Peterson rpete...@redhat.com --- diff --git a/fs/gfs2/bmap.c b/fs/gfs2/bmap.c index f0b945a

[Cluster-devel] [GFS2 PATCH 3/4] GFS2: Only increase rs_sizehint

2014-10-20 Thread Bob Peterson
If an application does a sequence of (1) big write, (2) little write we don't necessarily want to reset the size hint based on the smaller size. The fact that they did any big writes implies they may do more, and therefore we should try to allocate bigger block reservations, even if the last few

[Cluster-devel] [GFS2 PATCH 2/4] GFS2: Make block reservations more persistent

2014-10-20 Thread Bob Peterson
Before this patch, whenever a struct file (opened to allow writes) was closed, the multi-block reservation structure associated with the inode was deleted. That's a problem, especially when there are multiple writers. Applications that do open-write-close will suffer from greater levels of

[Cluster-devel] [GFS2 PATCH 0/4] Patches to reduce GFS2 fragmentation

2014-10-20 Thread Bob Peterson
with the first patch alone. This patch set includes all four patches. Bob Peterson (4): GFS2: Set of distributed preferences for rgrps GFS2: Make block reservations more persistent GFS2: Only increase rs_sizehint GFS2: If we use up our block reservation, request more next time fs/gfs2/file.c

[Cluster-devel] [GFS2 PATCH 1/4] GFS2: Set of distributed preferences for rgrps

2014-10-20 Thread Bob Peterson
This patch tries to use the journal numbers to evenly distribute which node prefers which resource group for block allocations. This is to help performance. --- fs/gfs2/incore.h | 2 ++ fs/gfs2/lock_dlm.c | 2 ++ fs/gfs2/ops_fstype.c | 1 + fs/gfs2/rgrp.c | 66

[Cluster-devel] [PATCH][try6] VFS: new want_holesize and got_holesize buffer_head flags for fiemap

2014-10-21 Thread Bob Peterson
a block-by-block search for the next valid block. Regards, Bob Peterson Red Hat File Systems Signed-off-by: Bob Peterson rpete...@redhat.com --- fs/ioctl.c | 7 ++- include/linux/buffer_head.h | 4 2 files changed, 10 insertions(+), 1 deletion(-) diff --git a/fs/ioctl.c b

[Cluster-devel] [PATCH] GFS2: Speed up fiemap function by skipping holes

2014-10-21 Thread Bob Peterson
is determined by a new function. Regards, Bob Peterson Red Hat File Systems Signed-off-by: Bob Peterson rpete...@redhat.com --- fs/gfs2/bmap.c | 70 -- 1 file changed, 68 insertions(+), 2 deletions(-) diff --git a/fs/gfs2/bmap.c b/fs/gfs2/bmap.c

Re: [Cluster-devel] [PATCH][try6] VFS: new want_holesize and got_holesize buffer_head flags for fiemap

2014-10-22 Thread Bob Peterson
of GFS2). (Which in the case of a 1PB sparse file is non-trivial, since it involves several levels of metadata indirection). Combining it with get_block made it much more efficient. Making a separate get_block_map_fiemap() function just seems like an exercise in redundancy. Regards, Bob Peterson Red

Re: [Cluster-devel] [PATCH][try6] VFS: new want_holesize and got_holesize buffer_head flags for fiemap

2014-10-23 Thread Bob Peterson
- Original Message - On Wed, Oct 22, 2014 at 08:28:53AM -0400, Bob Peterson wrote: Yes, I thought about that. One of my early prototypes had a separate function used by fiemap. Function __generic_block_fiemap would call get_block() which returned an indication of a hole as it does

[Cluster-devel] [GFS2 PATCH 1/3] GFS2: Set of distributed preferences for rgrps

2014-10-24 Thread Bob Peterson
This patch tries to use the journal numbers to evenly distribute which node prefers which resource group for block allocations. This is to help performance. --- fs/gfs2/incore.h | 1 + fs/gfs2/rgrp.c | 66 +++- 2 files changed, 62

[Cluster-devel] [GFS2 PATCH 3/3] GFS2: If we use up our block reservation, request more next time

2014-10-24 Thread Bob Peterson
If we run out of blocks for a given multi-block allocation, we obviously did not reserve enough. We should reserve more blocks for the next reservation to reduce fragmentation. This patch increases the size hint for reservations when they run out. --- fs/gfs2/rgrp.c | 3 +++ fs/gfs2/rgrp.h | 1 +

[Cluster-devel] [GFS2 PATCH 0/3] Patches to reduce GFS2 fragmentation

2014-10-24 Thread Bob Peterson
COUNT FOR OUTPUT FILES = 419736 EXTENT COUNT FOR OUTPUT FILES = 419040 EXTENT COUNT FOR OUTPUT FILES = 422774 EXTENT COUNT FOR OUTPUT FILES = 409281 (Average 419844) Regards, Bob Peterson Red Hat File Systems Signed-off-by: Bob Peterson rpete...@redhat.com --- Bob Peterson (3): GFS2: Set

[Cluster-devel] [GFS2 PATCH 2/3] GFS2: Only increase rs_sizehint

2014-10-24 Thread Bob Peterson
If an application does a sequence of (1) big write, (2) little write we don't necessarily want to reset the size hint based on the smaller size. The fact that they did any big writes implies they may do more, and therefore we should try to allocate bigger block reservations, even if the last few

Re: [Cluster-devel] [GFS2 PATCH 1/3] GFS2: Set of distributed preferences for rgrps

2014-10-27 Thread Bob Peterson
a few times for various tests. Regards, Bob Peterson Red Hat File Systems

[Cluster-devel] [GFS2 PATCH 0/3] Patches to reduce GFS2 fragmentation

2014-10-29 Thread Bob Peterson
6 time: 2hr 10min 58sec EXTENT COUNT FOR OUTPUT FILES = 424329 EXTENT COUNT FOR OUTPUT FILES = 430663 EXTENT COUNT FOR OUTPUT FILES = 430186 EXTENT COUNT FOR OUTPUT FILES = 421635 EXTENT COUNT FOR OUTPUT FILES = 416706 EXTENT COUNT FOR OUTPUT FILES = 418887 Regards, Bob Peterson Red Hat

[Cluster-devel] [GFS2 PATCH 3/3] GFS2: If we use up our block reservation, request more next time

2014-10-29 Thread Bob Peterson
If we run out of blocks for a given multi-block allocation, we obviously did not reserve enough. We should reserve more blocks for the next reservation to reduce fragmentation. This patch increases the size hint for reservations when they run out. --- fs/gfs2/rgrp.c | 3 +++ fs/gfs2/rgrp.h | 1 +

[Cluster-devel] [GFS2 PATCH 1/3] GFS2: Set of distributed preferences for rgrps

2014-10-29 Thread Bob Peterson
This patch tries to use the journal numbers to evenly distribute which node prefers which resource group for block allocations. This is to help performance. --- fs/gfs2/incore.h | 1 + fs/gfs2/rgrp.c | 66 +++- 2 files changed, 62

[Cluster-devel] [GFS2 PATCH 2/3] GFS2: Only increase rs_sizehint

2014-10-29 Thread Bob Peterson
If an application does a sequence of (1) big write, (2) little write we don't necessarily want to reset the size hint based on the smaller size. The fact that they did any big writes implies they may do more, and therefore we should try to allocate bigger block reservations, even if the last few

[Cluster-devel] [gfs2-utils PATCH] fsck.gfs2: Detect and correct corrupt journals

2014-11-13 Thread Bob Peterson
be tolerated, the problem is detected and the journal may be cleared. Regards, Bob Peterson Red Hat File Systems Signed-off-by: Bob Peterson rpete...@redhat.com --- gfs2/fsck/fs_recovery.c | 54 +++-- 1 file changed, 43 insertions(+), 11 deletions

Re: [Cluster-devel] [PATCH 1/2] fsck.gfs2: Improve reporting of pass timings

2014-11-14 Thread Bob Peterson
matches the existing output format. Signed-off-by: Andrew Price anpr...@redhat.com --- ACK Bob Peterson Red Hat File Systems

Re: [Cluster-devel] [PATCH 2/2] mkfs.gfs2: Revert default resource group size

2014-11-14 Thread Bob Peterson
; opts-jsize = GFS2_DEFAULT_JSIZE; opts-qcsize = GFS2_DEFAULT_QCSIZE; - opts-rgsize = GFS2_MAX_RGSIZE; + opts-rgsize = GFS2_DEFAULT_RGSIZE; opts-lockproto = lock_dlm; opts-locktable = ; opts-confirm = 1; -- 1.9.3 ACK Bob Peterson Red Hat File Systems

[Cluster-devel] [GFS2 PATCH] GFS2: Simplify directory filling, hashing and unhashing functions

2014-11-19 Thread Bob Peterson
Hi, This patch implements a new cookie encoding scheme for encoding, decoding and filling directory entries. This new scheme prevents duplicate cookies, so it eliminates the need to sort the entries to prevent duplicates. Thus, it simplies the code greatly. Regards, Bob Peterson Red Hat File

Re: [Cluster-devel] [PATCH 1/3] gfs2: bugger off early if O_CREAT open finds a directory

2014-11-19 Thread Bob Peterson
of a directory that already exists) but it looks like it already behaves that way. So I guess so. It may warrant further investigation. Bob Peterson Red Hat File Systems

Re: [Cluster-devel] [PATCH 1/3] gfs2: bugger off early if O_CREAT open finds a directory

2014-11-19 Thread Bob Peterson
- Original Message - On Wed, Nov 19, 2014 at 03:33:29PM -0500, Bob Peterson wrote: - Original Message - Signed-off-by: Al Viro v...@zeniv.linux.org.uk --- fs/gfs2/inode.c |5 + 1 file changed, 5 insertions(+) diff --git a/fs/gfs2/inode.c b/fs/gfs2

Re: [Cluster-devel] gfs2: use kvfree() instead of open-coding it

2014-11-20 Thread Bob Peterson
, const char *msg, int error) -- 1.7.10.4 ACK Bob Peterson Red Hat File Systems

Re: [Cluster-devel] gfs2: gfs2_dir_get_hash_table(): avoiding deferred vfree() is easy here...

2014-11-20 Thread Bob Peterson
) - kvfree(hc); - else + if (likely(!ip-i_hash_cache)) { ip-i_hash_cache = hc; + hc = NULL; + } spin_unlock(inode-i_lock); + kvfree(hc); return ip-i_hash_cache; } -- 1.7.10.4 ACK Bob Peterson Red Hat File Systems

[Cluster-devel] [GFS2 PATCH] GFS2: Write the log descriptor entries in the correct order

2014-11-26 Thread Bob Peterson
: This patch changes the order in which the blocks are processed by function gfs2_before_commit. Before, it was processing them in the wrong order such that the blocks would be layed out and written in reverse order. This changes the order to the correct order. Regards, Bob Peterson Red Hat File

Re: [Cluster-devel] [GFS2 PATCH] GFS2: Write the log descriptor entries in the correct order

2014-11-26 Thread Bob Peterson
- Original Message - As you can see, the blocks are reverse-sorted. This patch changes the order so that they're written in sorted order. Log headers written with the patch look more like this: Does that mean that this patch got the ordering the wrong way around then?

Re: [Cluster-devel] [fsck.gfs2 PATCH] fsck.gfs2: Change block_map to match bitmap

2015-01-26 Thread Bob Peterson
- Original Message - Hi Bob, On 22/01/15 20:41, Bob Peterson wrote: Hi, This patch changes the old block_map structure for fsck.gfs2 to the simpler bitmap structure so that we have a 1:1 correspondence. This was done to reduce memory requirements of fsck.gfs2. I'm curious

[Cluster-devel] [fsck.gfs2 PATCH] fsck.gfs2: Fix journal sequence number reporting problem

2015-01-23 Thread Bob Peterson
Hi, This bug was spotted by coverity. The fsck.gfs2 should include a line to report the number of journal sequence errors fixed. It was coded improperly so that it was never displayed. This patch fixes the code so that the message will be printed properly. Regards, Bob Peterson Red Hat File

[Cluster-devel] [fsck.gfs2 PATCH] fsck.gfs2: Fix a use-after-free in pass2

2015-01-15 Thread Bob Peterson
Hi, This patch fixes a rare code path that's doing use-after-free I spotted in pass2. Regards, Bob Peterson Red Hat File Systems Signed-off-by: Bob Peterson rpete...@redhat.com --- diff --git a/gfs2/fsck/pass2.c b/gfs2/fsck/pass2.c index 27b7336..4ea322a 100644 --- a/gfs2/fsck/pass2.c +++ b

[Cluster-devel] [fsck.gfs2 PATCH] fsck.gfs2: Reprocess nodes if anything changed

2015-01-15 Thread Bob Peterson
, but if the block is reused, it may be repurposed as a leaf block. That may confuse subsequent processing. Another reason to call reprocess_inode is when the goal block changes to a value outside the resource group, due to block allocations for repairs. Regards, Bob Peterson Red Hat File Systems Signed

[Cluster-devel] [fsck.gfs2 PATCH] fsck.gfs2: fix double-free bug

2015-01-15 Thread Bob Peterson
value back to NULL to ensure it's not reused. This is just a precaution I spotted while debugging, and it's worth doing. Regards, Bob Peterson Red Hat File Systems Signed-off-by: Bob Peterson rpete...@redhat.com --- diff --git a/gfs2/fsck/fs_recovery.c b/gfs2/fsck/fs_recovery.c index 3d8fe98

[Cluster-devel] [fsck.gfs2 patch] fsck.gfs2: Rebuild system files if they don't have the SYS bit set

2015-01-19 Thread Bob Peterson
Hi, This patch checks for the GFS2_DIF_SYSTEM bit on system dinodes. If the bit is not set, the dinode is assumed to be corrupt and rebuilt. If the jindex is rebuilt, it needs to be re-read. Regards, Bob Peterson Red Hat File Systems Signed-off-by: Bob Peterson rpete...@redhat.com --- diff

[Cluster-devel] [fsck.gfs2 PATCH] fsck.gfs2: Check the integrity of the journal index

2015-01-19 Thread Bob Peterson
. This allows us to remove a dummied-up rgrp kludge in the code. However, if the replayed journal block is part of an rgrp, we need to refresh the rgrp based on the values rewritten from the journal. Regards, Bob Peterson Red Hat File Systems Signed-off-by: Bob Peterson rpete...@redhat.com --- diff

[Cluster-devel] [fsck.gfs2 PATCH] fsck.gfs2: rgrp block count reform

2015-01-19 Thread Bob Peterson
today's blockmap that's 4:1. Regards, Bob Peterson Red Hat File Systems Signed-off-by: Bob Peterson rpete...@redhat.com --- diff --git a/gfs2/fsck/pass5.c b/gfs2/fsck/pass5.c index b2e8adf..2b8536d 100644 --- a/gfs2/fsck/pass5.c +++ b/gfs2/fsck/pass5.c @@ -12,94 +12,6 @@ #include fsck.h #include

[Cluster-devel] [GFS2 PATCH] GFS2: Eliminate a nonsense goto

2015-01-22 Thread Bob Peterson
Hi, This patch just removes a goto that did nothing. Regards, Bob Peterson Red Hat File Systems Signed-off-by: Bob Peterson rpete...@redhat.com --- diff --git a/fs/gfs2/inode.c b/fs/gfs2/inode.c index 9054002..73c72253 100644 --- a/fs/gfs2/inode.c +++ b/fs/gfs2/inode.c @@ -543,10 +543,7

Re: [Cluster-devel] [fsck.gfs2 PATCH] fsck.gfs2: Change block_map to match bitmap

2015-02-12 Thread Bob Peterson
- Original Message - - Original Message - Hi Bob, On 22/01/15 20:41, Bob Peterson wrote: Hi, This patch changes the old block_map structure for fsck.gfs2 to the simpler bitmap structure so that we have a 1:1 correspondence. This was done to reduce memory

[Cluster-devel] [GFS2 PATCH] GFS2: Allocate reservation during write_begin if needed

2015-02-17 Thread Bob Peterson
Hi, This patch adds a call to function gfs2_rs_alloc to make sure a reservation structure has been allocated before attempting to reserve blocks. Regards, Bob Peterson Red Hat File Systems Signed-off-by: Bob Peterson rpete...@redhat.com --- fs/gfs2/aops.c | 3 +++ 1 file changed, 3

Re: [Cluster-devel] [GFS2 PATCH] GFS2: Allocate reservation during write_begin if needed

2015-02-17 Thread Bob Peterson
or gfs2_file_aio_write. In the failing scenario, it's going through sendfile. I suppose I could patch sendfile as an alternative, but the advantage here is that this patch will do it only if a block allocation is needed. Regards, Bob Peterson Red Hat File Systems

Re: [Cluster-devel] [PATCH] Add myself (Bob Peterson) as a maintainer of GFS2

2015-02-17 Thread Bob Peterson
- Original Message - Hi Bob, On Mon, 16 Feb 2015 09:20:56 -0500 (EST) Bob Peterson rpete...@redhat.com wrote: This patch adds Bob Peterson as a maintainer of the GFS2 file system. It also changes the development repository to a shared location rather than Steve Whitehouse's

[Cluster-devel] [GFS2 PATCH] GFS2: Allocate reservation during splice_write

2015-02-19 Thread Bob Peterson
Hi, This patch adds a GFS2-specific function for splice_write which first calls function gfs2_rs_alloc to make sure a reservation structure has been allocated before attempting to reserve blocks. Regards, Bob Peterson Red Hat File Systems Signed-off-by: Bob Peterson rpete...@redhat.com

[Cluster-devel] [PATCH] Add myself (Bob Peterson) as a maintainer of GFS2

2015-02-16 Thread Bob Peterson
Hi, This patch adds Bob Peterson as a maintainer of the GFS2 file system. It also changes the development repository to a shared location rather than Steve Whitehouse's private location. Regards, Bob Peterson Red Hat File Systems Signed-off-by: Bob Peterson rpete...@redhat.com

Re: [Cluster-devel] [PATCH 4/8] gfs2: Convert to using -get_state callback

2015-02-18 Thread Bob Peterson
ACK Looks right. Bob Peterson Red Hat File Systems

Re: [Cluster-devel] [PATCH 2/3] fsck.gfs2: Simplify bad_journalname

2015-01-28 Thread Bob Peterson
), + len, filename); } /** -- 1.9.3 Nice trick (strangely, I've never used that construct), but we could just as well get rid of the whole function altogether. Bob Peterson Red Hat File Systems

Re: [Cluster-devel] [PATCH 3/3] gfs2-utils build: Add a configure script summary

2015-01-28 Thread Bob Peterson
- Original Message - Print a nicely formatted summary of some of the more interesting configure options. Required some tweaking of earlier configure stages for accuracy. Signed-off-by: Andrew Price anpr...@redhat.com --- Hi, ACK Bob Peterson Red Hat File Systems

Re: [Cluster-devel] [PATCH 1/3] fsck.gfs2: Fix 'initializer element is not constant' build error

2015-01-28 Thread Bob Peterson
anyway, the initializers can be removed to achieve the intended result. Signed-off-by: Andrew Price anpr...@redhat.com --- ACK Bob Peterson Red Hat File Systems

[Cluster-devel] [fsck.gfs2 patch] fsck.gfs2: Revise undo processing

2015-01-09 Thread Bob Peterson
of that indirect block. This patch allows it to finish out the rest of the undo work properly. Regards, Bob Peterson Red Hat File Systems Signed-off-by: Bob Peterson rpete...@redhat.com --- diff --git a/gfs2/fsck/metawalk.c b/gfs2/fsck/metawalk.c index cd224fe..1875b24 100644 --- a/gfs2/fsck

[Cluster-devel] [fsck.gfs2 patch] fsck.gfs2: Change basic dentry checks for too long of file names

2015-01-09 Thread Bob Peterson
Peterson Red Hat File Systems Signed-off-by: Bob Peterson rpete...@redhat.com --- diff --git a/gfs2/fsck/pass2.c b/gfs2/fsck/pass2.c index 1559d8e..bdcf77c 100644 --- a/gfs2/fsck/pass2.c +++ b/gfs2/fsck/pass2.c @@ -468,7 +468,8 @@ static int basic_dentry_checks(struct gfs2_inode *ip, struct

[Cluster-devel] [fsck.gfs2 patch] fsck.gfs2: Print out block number when pass3 finds a bad directory

2015-01-09 Thread Bob Peterson
Hi, This patch changes pass3 so that it prints out the directory inode number when it finds a directory containing a bad block. Regards, Bob Peterson Red Hat File Systems Signed-off-by: Bob Peterson rpete...@redhat.com --- diff --git a/gfs2/fsck/pass3.c b/gfs2/fsck/pass3.c index 9582b5b

[Cluster-devel] [fsck.gfs2 patch] fsck.gfs2: Adjust when hash table is doubled

2015-01-09 Thread Bob Peterson
into account the fact that the hash table could just unexpectly double, throwing off all the pointers. This patch attempts to correct that by fixing up the pointers as necessary. Regards, Bob Peterson Red Hat File Systems Signed-off-by: Bob Peterson rpete...@redhat.com --- diff --git a/gfs2/fsck

[Cluster-devel] [fsck.gfs2 patch] fsck.gfs2: remove duplicate designation during undo

2015-01-09 Thread Bob Peterson
D2. Note, though, that if the corruption is found later in the cycle (after undo is possible) the duplicate reference MUST remain and be resolved by pass1b. Regards, Bob Peterson Red Hat File Systems Signed-off-by: Bob Peterson rpete...@redhat.com --- diff --git a/gfs2/fsck/pass1.c b/gfs2/fsck

Re: [Cluster-devel] 3.18.5 kernel panic: fs/gfs2/acl.c:76

2015-02-09 Thread Bob Peterson
tree? Then you can get the credit. Regards, Bob Peterson Red Hat File Systems

Re: [Cluster-devel] [GFS2 PATCH 0/4] fallocate quota fixes

2015-02-12 Thread Bob Peterson
/gfs2/rgrp.h | 10 +- fs/gfs2/xattr.c | 2 +- 9 files changed, 89 insertions(+), 34 deletions(-) -- 1.8.1.4 Hi, ACK. Looks okay to me. Regards, Bob Peterson Red Hat File Systems

[Cluster-devel] [GFS2 PATCH] GFS2: Eliminate __gfs2_glock_remove_from_lru

2015-01-05 Thread Bob Peterson
Hi, Since the only caller of function __gfs2_glock_remove_from_lru locks the same spin_lock as gfs2_glock_remove_from_lru, the functions can be combined. Regards, Bob Peterson Red Hat File Systems Signed-off-by: Bob Peterson rpete...@redhat.com --- diff --git a/fs/gfs2/glock.c b/fs/gfs2

Re: [Cluster-devel] [GFS2 0/3] fallocate and quota fixes

2015-03-18 Thread Bob Peterson
+++--- fs/gfs2/rgrp.c | 20 ++ fs/gfs2/rgrp.h | 3 ++- fs/gfs2/xattr.c | 2 +- 10 files changed, 133 insertions(+), 65 deletions(-) -- 1.8.1.4 Hi, ACK to all three. Now pushed to the for-next branch of the linux-gfs2.git git tree. Regards, Bob Peterson

Re: [Cluster-devel] [PATCH 1/1] gfs2: incorrect check for debugfs returns

2015-03-30 Thread Bob Peterson
Song cson...@gatech.edu --- fs/gfs2/glock.c | 47 --- 1 file changed, 28 insertions(+), 19 deletions(-) Hi, I've now added this patch to the for-next branch of the linux-gfs2.git tree. Thanks. Regards, Bob Peterson Red Hat File Systems

Re: [Cluster-devel] [PATCH] GFS2: Improve readability of gfs2_alloc_inode

2015-03-03 Thread Bob Peterson
; - } + if (!ip) + return NULL; + + ip-i_flags = 0; + ip-i_gl = NULL; + ip-i_rgd = NULL; + ip-i_res = NULL; return ip-i_inode; } -- 1.9.3 ACK, Bob Peterson Red Hat File Systems

Re: [Cluster-devel] [PATCH RHEL6] libgfs2: Use a matching context mount option in mount_gfs2_meta

2015-02-27 Thread Bob Peterson
; -- 1.9.3 Hi, Is this a memory leak (albeit a small one) or did I miss something? I don't see where the memory allocate by strndup is ever freed. Regards, Bob Peterson Red Hat File Systems

[Cluster-devel] Announcing a new GFS2 development git repository and branch

2015-02-23 Thread Bob Peterson
new patches. To clone the new tree, do something like this: git clone git://git.kernel.org/pub/scm/linux/kernel/git/gfs2/linux-gfs2.git There are currently three patches pushed to the new for-next branch: 0108044 Add myself (Bob Peterson) as a maintainer of GFS2 (Bob Peterson) 9717c00 GFS2

Re: [Cluster-devel] [PATCH] GFS2: gfs2_set_acl(): Cache no acl as well

2015-02-23 Thread Bob Peterson
Hi, Now in the for-next tree of the linux-gfs2.git repo. Regards, Bob Peterson - Original Message - When removing a default acl or setting an access acl that is entirely represented in the file mode, we end up with acl == NULL in gfs2_set_acl(). In that case, bring gfs2 in line

Re: [Cluster-devel] [GFS2 0/3] fallocate and quota fixes

2015-02-25 Thread Bob Peterson
to me. ACK to the series. Regards, Bob Peterson Red Hat File Systems

Re: [Cluster-devel] linux-next: Tree for Apr 20 (gfs2)

2015-04-24 Thread Bob Peterson
! -- ~Randy Hi, My apologies. I've pushed an addendum patch that fixes the problem to the for-next branch of the linux-gfs2.git git tree. Regards, Bob Peterson

Re: [Cluster-devel] [PATCH] mkfs.gfs2: Allow longer cluster names

2015-04-22 Thread Bob Peterson
changed, 13 insertions(+), 5 deletions(-) Hi, ACK Bob Peterson Red Hat File Systems

Re: [Cluster-devel] [PATCH] GFS2: mark the journal idle to fix ro mounts

2015-04-24 Thread Bob Peterson
sense to me. Regards, Bob Peterson Red Hat File Systems

Re: [Cluster-devel] [PATCH] GFS2: mark the journal idle to fix ro mounts

2015-05-01 Thread Bob Peterson
=for-nextid=086cc672e1cb600b9c17688a4aa44560db858c03 Regards, Bob Peterson Red Hat File Systems

Re: [Cluster-devel] [PATCH 02/12] trivial: GFS2: inode.c: indent with TABs, not spaces

2015-05-01 Thread Bob Peterson
- Original Message - Follow the same style used for the other functions in the same file. Signed-off-by: Antonio Ospite a...@ao2.it Cc: Steven Whitehouse swhit...@redhat.com Cc: Bob Peterson rpete...@redhat.com Cc: cluster-devel@redhat.com --- fs/gfs2/inode.c | 4 ++-- 1 file

Re: [Cluster-devel] [PATCH v2] GFS2: add support for rename2 and RENAME_EXCHANGE

2015-05-05 Thread Bob Peterson
://git.kernel.org/cgit/linux/kernel/git/gfs2/linux-gfs2.git/commit/fs/gfs2?h=for-nextid=a63b7bbc2175901d79fa36ba734499655c077f0d Regards, Bob Peterson Red Hat File Systems

Re: [Cluster-devel] [GFS2] gfs2: handle NULL rgd in set_rgrp_preferences

2015-05-05 Thread Bob Peterson
of the linux-gfs2 git tree. FYI: I removed Resolves: for upstream. Regards, Bob Peterson Red Hat File Systems

Re: [Cluster-devel] [PATCH v2] GFS2: make sure S_NOSEC flag isn't overwritten

2015-05-05 Thread Bob Peterson
in the for-next branch of the linux-gfs2 tree: https://git.kernel.org/cgit/linux/kernel/git/gfs2/linux-gfs2.git/commit/fs/gfs2?h=for-nextid=01e64ee40ad741037352d1d6202eaa432f833eb4 Regards, Bob Peterson Red Hat File Systems

Re: [Cluster-devel] [PATCH V3 linux-next] gfs2: convert simple_str to kstr

2015-05-05 Thread Bob Peterson
Frederick f...@skynet.be Hi, Thanks. The v3 version of the patch is now in the for-next branch of the linux-gfs2 tree: https://git.kernel.org/cgit/linux/kernel/git/gfs2/linux-gfs2.git/commit/fs/gfs2?h=for-nextid=e50ead480fac63ede9e0b656cd29c1820f7af9de Regards, Bob Peterson Red Hat File Systems

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

2015-04-14 Thread Bob Peterson
Hi, Please consider pulling the following changes, Bob Peterson GFS2: merge window Here is a list of patches we've accumulated for GFS2 for the current upstream merge window. Most of the patches fix GFS2 quotas, which were

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

2015-04-15 Thread Bob Peterson
- Original Message - On Tue, Apr 14, 2015 at 10:47 AM, Bob Peterson rpete...@redhat.com wrote: 12 files changed, 184 insertions(+), 95 deletions(-) Oh, and this was incorrect. You had apparently limited the statistics to the fs/gfs2 directory, and thus missed the changes

[Cluster-devel] GFS2: Patches pulled, linux-gfs2.git rebased

2015-04-15 Thread Bob Peterson
Hi, Linus has pulled the latest set of patches from the for-next branch of the linux-gfs2.git tree, so I rebased the tree based on Linus's master. So the current tree is back to having no unmerged GFS2 patches. Regards, Bob Peterson Red Hat File Systems

[Cluster-devel] [GFS2 PATCH] GFS2: Use average srttb value in congestion calculations

2015-04-15 Thread Bob Peterson
processing occurring on a different CPU. Signed-off-by: Bob Peterson rpete...@redhat.com --- diff --git a/fs/gfs2/rgrp.c b/fs/gfs2/rgrp.c index 6af2396..f39eedc 100644 --- a/fs/gfs2/rgrp.c +++ b/fs/gfs2/rgrp.c @@ -1850,14 +1850,19 @@ static bool gfs2_rgrp_congested(const struct gfs2_rgrpd *rgd, int

<    1   2   3   4   5   6   7   8   9   10   >