By using the generic list_lru code, we can now separate the
per sb quota list locking from the lru locking. The lru
lock is made into the inner-most lock.
As a result of this new lock order, we may occasionally see
items on the per-sb quota list which are "dead" so that the
two places where we traverse that list are updated to take
account of that.
As a result of this patch, the gfs2 quota shrinker is now
NUMA zone aware, and we are also laying the foundations for
further improvments in due course.
Signed-off-by: Steven Whitehouse <swhiteho@redhat.com>
Signed-off-by: Abhijith Das <adas@redhat.com>
Tested-by: Abhijith Das <adas@redhat.com>
Cc: Dave Chinner <dchinner@redhat.com>
This patch adds reflink support to the quota data cache. It
looks a bit strange because we still don't have a sensible
split in the lookup by id and the lru list. That is coming in
later patches though.
The intent here is just to swap the current ref count for
reflinks in all cases with as little as possible other change.
Signed-off-by: Steven Whitehouse <swhiteho@redhat.com>
Signed-off-by: Abhijith Das <adas@redhat.com>
Tested-by: Abhijith Das <adas@redhat.com>
Currently glocks have an atomic reference count and also a spinlock
which covers various internal fields, such as the state. This intent of
this patch is to replace the spinlock and the atomic reference count
with a lockref structure. This contains a spinlock which we can continue
to use as before, and a reference counter which is used in conjuction
with the spinlock to replace the previous atomic counter.
As a result of this there are some new rules for reference counting on
glocks. We need to distinguish between reference count changes under
gl_spin (which are now just increment or decrement of the new counter,
provided the count cannot hit zero) and those which are outside of
gl_spin, but which now take gl_spin internally.
The conversion is relatively straight forward. There is probably some
further clean up which can be done, but the priority at this stage is to
make the change in as simple a manner as possible.
A consequence of this change is that the reference count is being
decoupled from the lru list processing. This should allow future
adoption of the lru_list code with glocks in due course.
The reason for using the "dead" state and not just relying on 0 being
the "invalid state" is so that in due course 0 ref counts can be
allowable. The intent is to eventually be able to remove the ref count
changes which are currently hidden away in state_change().
Signed-off-by: Steven Whitehouse <swhiteho@redhat.com>
Now that gfs2_quota_sync can be potentially called from multiple
threads, we should protect this bit of code, and the sync generation
number in particular in order to ensure that there are no races
when syncing quotas.
Signed-off-by: Steven Whitehouse <swhiteho@redhat.com>
Cc: Abhijith Das <adas@redhat.com>
There is no need for a paramater which relates to the internals
of quota to be exposed to users. The only possible use would be
to turn it up so large that the memory allocation fails. So lets
remove it and set it to a sensible value which ensures that we
don't ask for multipage allocations.
Currently the size of struct gfs2_holder means that the caluclated
value is identical to the previous default value, so there should
be no functional change.
Signed-off-by: Steven Whitehouse <swhiteho@redhat.com>
Cc: Abhijith Das <adas@redhat.com>
This patch adds a structure to contain allocation parameters with
the intention of future expansion of this structure. The idea is
that we should be able to add more information about the allocation
in the future in order to allow the allocator to make a better job
of placing the requests on-disk.
There is no functional difference from applying this patch.
Signed-off-by: Steven Whitehouse <swhiteho@redhat.com>
This is a respin of the original patch. As Steve pointed out, the
introduction of field bii makes it easy to eliminate bi itself.
This revised patch does just that, replacing bi with bii.
This patch adds a new field to the rbm structure, called bii,
which is an index into the array of bitmaps for an rgrp.
This replaces *bi which was a pointer to the bitmap.
This is being done for further optimizations.
Signed-off-by: Bob Peterson <rpeterso@redhat.com>
Signed-off-by: Steven Whitehouse <swhiteho@redhat.com>
This patch introduces a new field in the bitmap structure called
bi_blocks. Its purpose is to save us from constantly multiplying
bi_len by the constant GFS2_NBBY. It also paves the way for more
optimization in a future patch.
Signed-off-by: Bob Peterson <rpeterso@redhat.com>
Signed-off-by: Steven Whitehouse <swhiteho@redhat.com>
This patch adds a bool indicating whether the demote
request was originated locally or remotely. This is then
used by the iopen ->go_callback() to make 100% sure that
it will only respond to remote callbacks.
Since ->evict_inode() uses GL_NOCACHE when it attempts to
get an exclusive lock on the iopen lock, this may result
in extra scheduling of the workqueue in case that the
exclusive promotion request failed. This patch prevents
that from happening.
Signed-off-by: Steven Whitehouse <swhiteho@redhat.com>
In order to allow transactions and log flushes to happen at the same
time, gfs2 needs to move the transaction accounting and active items
list code into the gfs2_trans structure. As a first step toward this,
this patch removes the gfs2_ail structure, and handles the active items
list in the gfs_trans structure. This keeps gfs2 from allocating an ail
structure on log flushes, and gives us a struture that can later be used
to store the transaction accounting outside of the gfs2 superblock
structure.
With this patch, at the end of a transaction, gfs2 will add the
gfs2_trans structure to the superblock if there is not one already.
This structure now has the active items fields that were previously in
gfs2_ail. This is not necessary in the case where the transaction was
simply used to add revokes, since these are never written outside of the
journal, and thus, don't need an active items list.
Also, in order to make sure that the transaction structure is not
removed while it's still in use by gfs2_trans_end, unlocking the
sd_log_flush_lock has to happen slightly later in ending the
transaction.
Signed-off-by: Benjamin Marzinski <bmarzins@redhat.com>
Signed-off-by: Steven Whitehouse <swhiteho@redhat.com>
The temp lvb bitmap was on the stack, which could
be an alignment problem for __set_bit_le. Use
kmalloc for it instead.
Signed-off-by: David Teigland <teigland@redhat.com>
Signed-off-by: Steven Whitehouse <swhiteho@redhat.com>
Pull user namespace and namespace infrastructure changes from Eric W Biederman:
"This set of changes starts with a few small enhnacements to the user
namespace. reboot support, allowing more arbitrary mappings, and
support for mounting devpts, ramfs, tmpfs, and mqueuefs as just the
user namespace root.
I do my best to document that if you care about limiting your
unprivileged users that when you have the user namespace support
enabled you will need to enable memory control groups.
There is a minor bug fix to prevent overflowing the stack if someone
creates way too many user namespaces.
The bulk of the changes are a continuation of the kuid/kgid push down
work through the filesystems. These changes make using uids and gids
typesafe which ensures that these filesystems are safe to use when
multiple user namespaces are in use. The filesystems converted for
3.9 are ceph, 9p, afs, ocfs2, gfs2, ncpfs, nfs, nfsd, and cifs. The
changes for these filesystems were a little more involved so I split
the changes into smaller hopefully obviously correct changes.
XFS is the only filesystem that remains. I was hoping I could get
that in this release so that user namespace support would be enabled
with an allyesconfig or an allmodconfig but it looks like the xfs
changes need another couple of days before it they are ready."
* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/ebiederm/user-namespace: (93 commits)
cifs: Enable building with user namespaces enabled.
cifs: Convert struct cifs_ses to use a kuid_t and a kgid_t
cifs: Convert struct cifs_sb_info to use kuids and kgids
cifs: Modify struct smb_vol to use kuids and kgids
cifs: Convert struct cifsFileInfo to use a kuid
cifs: Convert struct cifs_fattr to use kuid and kgids
cifs: Convert struct tcon_link to use a kuid.
cifs: Modify struct cifs_unix_set_info_args to hold a kuid_t and a kgid_t
cifs: Convert from a kuid before printing current_fsuid
cifs: Use kuids and kgids SID to uid/gid mapping
cifs: Pass GLOBAL_ROOT_UID and GLOBAL_ROOT_GID to keyring_alloc
cifs: Use BUILD_BUG_ON to validate uids and gids are the same size
cifs: Override unmappable incoming uids and gids
nfsd: Enable building with user namespaces enabled.
nfsd: Properly compare and initialize kuids and kgids
nfsd: Store ex_anon_uid and ex_anon_gid as kuids and kgids
nfsd: Modify nfsd4_cb_sec to use kuids and kgids
nfsd: Handle kuids and kgids in the nfs4acl to posix_acl conversion
nfsd: Convert nfsxdr to use kuids and kgids
nfsd: Convert nfs3xdr to use kuids and kgids
...
- Change qd_id in struct gfs2_qutoa_data to struct kqid.
- Remove the now unnecessary QDF_USER bit field in qd_flags.
- Propopoage this change through the code generally making
things simpler along the way.
Cc: Steven Whitehouse <swhiteho@redhat.com>
Signed-off-by: "Eric W. Biederman" <ebiederm@xmission.com>
This patch reinstates the ack system which withdraw should be using. It
appears to have been accidentally forgotten when the lock module was
merged into GFS2, due to two different sysfs files having the same name.
Reported-by: David Teigland <teigland@redhat.com>
Signed-off-by: Steven Whitehouse <swhiteho@redhat.com>
Instead of using a list of buffers to write ahead of the journal
flush, this now uses a list of inodes and calls ->writepages
via filemap_fdatawrite() in order to achieve the same thing. For
most use cases this results in a shorter ordered write list,
as well as much larger i/os being issued.
The ordered write list is sorted by inode number before writing
in order to retain the disk block ordering between inodes as
per the previous code.
The previous ordered write code used to conflict in its assumptions
about how to write out the disk blocks with mpage_writepages()
so that with this updated version we can also use mpage_writepages()
for GFS2's ordered write, writepages implementation. So we will
also send larger i/os from writeback too.
Signed-off-by: Steven Whitehouse <swhiteho@redhat.com>
The freeze code has not been looked at a lot recently. Upstream has
moved on, and this is an attempt to catch us back up again. There
is a vfs level interface for the freeze code which can be called
from our (obsolete, but kept for backward compatibility purposes)
sysfs freeze interface. This means freezing this way vs. doing it
from the ioctl should now work in identical fashion.
As a result of this, the freeze function is only called once
and we can drop our own special purpose code for counting the
number of freezes.
Signed-off-by: Steven Whitehouse <swhiteho@redhat.com>
This patch copies the body of gfs2_trans_add_bh into the two newly
added gfs2_trans_add_data and gfs2_trans_add_meta functions. We can
then move the .lo_add functions from lops.c into trans.c and call
them directly.
As a result of this, we no longer need to use the .lo_add functions
at all, so that is removed from the log operations structure.
Signed-off-by: Steven Whitehouse <swhiteho@redhat.com>
The lksb struct already contains a pointer to the lvb,
so another directly from the glock struct is not needed.
Signed-off-by: David Teigland <teigland@redhat.com>
Signed-off-by: Steven Whitehouse <swhiteho@redhat.com>
Save the effort of allocating, reading and writing
the lvb for most glocks that do not use it.
Signed-off-by: David Teigland <teigland@redhat.com>
Signed-off-by: Steven Whitehouse <swhiteho@redhat.com>
When unmounting, gfs2 does a full dlm_unlock operation on every
cached lock. This can create a very large amount of work and can
take a long time to complete. However, the vast majority of these
dlm unlock operations are unnecessary because after all the unlocks
are done, gfs2 leaves the dlm lockspace, which automatically clears
the locks of the leaving node, without unlocking each one individually.
So, gfs2 can skip explicit dlm unlocks, and use dlm_release_lockspace to
remove the locks implicitly. The one exception is when the lock's lvb is
being used. In this case, dlm_unlock is called because it may update the
lvb of the resource.
Signed-off-by: David Teigland <teigland@redhat.com>
Signed-off-by: Steven Whitehouse <swhiteho@redhat.com>
[Editorial: This is a nit, but has been a minor irritation for a long time:]
This patch renames glops structure item for go_xmote_th to go_sync.
The functionality is unchanged; it's just for readability.
Signed-off-by: Bob Peterson <rpeterso@redhat.com>
Signed-off-by: Steven Whitehouse <swhiteho@redhat.com>
This patch is a rewrite of function gfs2_rbm_from_block. Rather than
looping to find the right bitmap, the code now does a few simple
math calculations.
I compared the performance of both algorithms side by side and the new
algorithm is noticeably faster. Sample instrumentation output from a
"fast" machine:
5 million calls: millisec spent: Orig: 166 New: 113
5 million calls: millisec spent: Orig: 189 New: 114
In addition, I ran postmark (on a somewhat slowr CPU) before the after
the new algorithm was put in place and postmark showed a decent
improvement:
Before the new algorithm:
-------------------------
Time:
645 seconds total
584 seconds of transactions (171 per second)
Files:
150087 created (232 per second)
Creation alone: 100000 files (2083 per second)
Mixed with transactions: 50087 files (85 per second)
49995 read (85 per second)
49991 appended (85 per second)
150087 deleted (232 per second)
Deletion alone: 100174 files (7705 per second)
Mixed with transactions: 49913 files (85 per second)
Data:
273.42 megabytes read (434.08 kilobytes per second)
852.13 megabytes written (1.32 megabytes per second)
With the new algorithm:
-----------------------
Time:
599 seconds total
530 seconds of transactions (188 per second)
Files:
150087 created (250 per second)
Creation alone: 100000 files (1886 per second)
Mixed with transactions: 50087 files (94 per second)
49995 read (94 per second)
49991 appended (94 per second)
150087 deleted (250 per second)
Deletion alone: 100174 files (6260 per second)
Mixed with transactions: 49913 files (94 per second)
Data:
273.42 megabytes read (467.42 kilobytes per second)
852.13 megabytes written (1.42 megabytes per second)
Signed-off-by: Bob Peterson <rpeterso@redhat.com>
Signed-off-by: Steven Whitehouse <swhiteho@redhat.com>
With the recently added block reservation code, an additional function
was added to search for free blocks. This had a restriction of only being
able to search for aligned extents of free blocks. As a result the
allocation patterns when reserving blocks were suboptimal when the
existing allocation of blocks for an inode was not aligned to the same
boundary.
This patch resolves that problem by adding the ability for gfs2_rbm_find
to search for extents of a particular minimum size. We can then use
gfs2_rbm_find for both looking for reservations, and also looking for
free blocks on an individual basis when we actually come to do the
allocation later on. As a result we only need a single set of code
to deal with both situations.
The function gfs2_rbm_from_block() is moved up rgrp.c so that it
occurs before all of its callers.
Many thanks are due to Bob for helping track down the final issue in
this patch. That fix to the rb_tree traversal and to not share
block reservations from a dirctory to its children is included here.
Signed-off-by: Steven Whitehouse <swhiteho@redhat.com>
Signed-off-by: Bob Peterson <rpeterso@redhat.com>
This patch improves the tracing of block reservations by
removing some corner cases and also providing more useful
detail in the traces.
A new field is added to the reservation structure to contain
the inode number. This is used since in certain contexts it is
not possible to access the inode itself to obtain this information.
As a result we can then display the inode number for all tracepoints
and also in case we dump the resource group.
The "del" tracepoint operation has been removed. This could be called
with the reservation rgrp set to NULL. That resulted in not printing
the device number, and thus making the information largely useless
anyway. Also, the conditional on the rgrp being NULL can then be
removed from the tracepoint. After this change, all the block
reservation tracepoint calls will be called with the rgrp information.
The existing ins,clm and tdel calls to the block reservation tracepoint
are sufficient to track the entire life of the block reservation.
In gfs2_block_alloc() the error detection is updated to print out
the inode number of the problematic inode. This can then be compared
against the information in the glock dump,tracepoints, etc.
Signed-off-by: Steven Whitehouse <swhiteho@redhat.com>
This is part of a series of patches which are introducing the
gfs2_rbm structure throughout the block allocation code. The
main aim of this part is to create a search function which can
deal directly with struct gfs2_rbm. In this case it specifies
the initial position at which to start the search and also the
point at which the search terminates.
The net result of this is to clean up the search code and make
it rather more readable, and the various possible exceptions which
may occur during the search are partitioned into their own functions.
There are some bug fixes too. We should not be checking the reservations
while allocating extents - the time for that is when we are searching
for where to put the extent, not when we've already made that decision.
Also, rgblk_search had two uses, and in only one of those cases did
it make sense to check for reservations. This is fixed in the new
gfs2_rbm_find function, which has a cleaner interface.
The reservation checking has been improved by always checking for
contiguous reservations, and returning the first free block after
all contiguous reservations. This is done under the spin lock to
ensure consistancy of the tree.
The allocation of extents is now in all cases done by the existing
allocation code, and if there is an active reservation, that is updated
after the fact. Again this is done under the spin lock, since it entails
changing the lookup key for the reservation in question.
Signed-off-by: Steven Whitehouse <swhiteho@redhat.com>
This patch introduces a new structure, gfs2_rbm, which is a
tuple of a resource group, a bitmap within the resource group
and an offset within that bitmap. This is designed to make
manipulating these sets of variables easier. There is also a
new helper function which converts this representation back
to a disk block address.
In addition, the rbtree nodes which are used for the reservations
were not being correctly initialised, which is now fixed. Also,
the tracing was not passing through the inode where it should
have been. That is mostly fixed aside from one corner case. This
needs to be revisited since there can also be a NULL rgrp in
some cases which results in the device being incorrect in the
trace.
This is intended to be the first step towards cleaning up some
of the allocation code, and some further bug fixes.
Signed-off-by: Steven Whitehouse <swhiteho@redhat.com>
The rs_requested field is left over from the original allocation
code, however this should have been a parameter passed to the
various functions from gfs2_inplace_reserve() and not a member of the
reservation structure as the value is not required after the
initial allocation.
This also helps simplify the code since we no longer need to set
the rs_requested to zero. Also the gfs2_inplace_release()
function can also be simplified since the reservation structure
will always be defined when it is called, and the only remaining
task is to unlock the rgrp if required. It can also now be
called unconditionally too, resulting in a further simplification.
Signed-off-by: Steven Whitehouse <swhiteho@redhat.com>
This patch reduces GFS2 file fragmentation by pre-reserving blocks. The
resulting improved on disk layout greatly speeds up operations in cases
which would have resulted in interlaced allocation of blocks previously.
A typical example of this is 10 parallel dd processes, each writing to a
file in a common dirctory.
The implementation uses an rbtree of reservations attached to each
resource group (and each inode).
Signed-off-by: Bob Peterson <rpeterso@redhat.com>
Signed-off-by: Steven Whitehouse <swhiteho@redhat.com>
Instead of reading in the resource groups when gfs2 is checking
for free space to allocate from, gfs2 can store the necessary infromation
in the resource group's lvb. Also, instead of searching for unlinked
inodes in every resource group that's checked for free space, gfs2 can
store the number of unlinked but inodes in the lvb, and only check for
unlinked inodes if it will find some.
The first time a resource group is locked, the lvb must initialized.
Since this involves counting the unlinked inodes in the resource group,
this takes a little extra time. But after that, if the resource group
is locked with GL_SKIP, the buffer head won't be read in unless it's
actually needed.
Enabling the resource groups lvbs is done via the rgrplvb mount option. If
this option isn't set, the lvbs will still be set and updated, but they won't
be verfied or used by the filesystem. To safely turn on this option, all of
the nodes mounting the filesystem must be running code with this patch, and
the filesystem must have been completely unmounted since they were updated.
Signed-off-by: Benjamin Marzinski <bmarzins@redhat.com>
Signed-off-by: Steven Whitehouse <swhiteho@redhat.com>
This patch moves the ancillary quota data structures into the
block reservations structure. This saves GFS2 some time and
effort in allocating and deallocating the qadata structure.
Signed-off-by: Bob Peterson <rpeterso@redhat.com>
Signed-off-by: Steven Whitehouse <swhiteho@redhat.com>
This set includes some minor fixes and improvements.
The one large patch addresses the special "nodir" mode,
which has been a long neglected proof of concept, but
with these fixes seems to be quite usable. It allows
the resource master to be assigned statically instead of
dynamically, which can improve performance if there is
little locality and most resources are shared.
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.12 (GNU/Linux)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=mHOh
-----END PGP SIGNATURE-----
Merge tag 'dlm-3.5' of git://git.kernel.org/pub/scm/linux/kernel/git/teigland/linux-dlm
Pull dlm updates from David Teigland:
"This set includes some minor fixes and improvements. The one large
patch addresses the special "nodir" mode, which has been a long
neglected proof of concept, but with these fixes seems to be quite
usable. It allows the resource master to be assigned statically
instead of dynamically, which can improve performance if there is
little locality and most resources are shared."
* tag 'dlm-3.5' of git://git.kernel.org/pub/scm/linux/kernel/git/teigland/linux-dlm:
dlm: NULL dereference on failure in kmem_cache_create()
gfs2: fix recovery during unmount
dlm: fixes for nodir mode
dlm: improve error and debug messages
dlm: avoid unnecessary search in search_rsb
dlm: limit rcom debug messages
dlm: fix waiter recovery
dlm: prevent connections during shutdown
The "nodir" mode (statically assign master nodes instead
of using the resource directory) has always been highly
experimental, and never seriously used. This commit
fixes a number of problems, making nodir much more usable.
- Major change to recovery: recover all locks and restart
all in-progress operations after recovery. In some
cases it's not possible to know which in-progess locks
to recover, so recover all. (Most require recovery
in nodir mode anyway since rehashing changes most
master nodes.)
- Change the way nodir mode is enabled, from a command
line mount arg passed through gfs2, into a sysfs
file managed by dlm_controld, consistent with the
other config settings.
- Allow recovering MSTCPY locks on an rsb that has not
yet been turned into a master copy.
- Ignore RCOM_LOCK and RCOM_LOCK_REPLY recovery messages
from a previous, aborted recovery cycle. Base this
on the local recovery status not being in the state
where any nodes should be sending LOCK messages for the
current recovery cycle.
- Hold rsb lock around dlm_purge_mstcpy_locks() because it
may run concurrently with dlm_recover_master_copy().
- Maintain highbast on process-copy lkb's (in addition to
the master as is usual), because the lkb can switch
back and forth between being a master and being a
process copy as the master node changes in recovery.
- When recovering MSTCPY locks, flag rsb's that have
non-empty convert or waiting queues for granting
at the end of recovery. (Rename flag from LOCKS_PURGED
to RECOVER_GRANT and similar for the recovery function,
because it's not only resources with purged locks
that need grant a grant attempt.)
- Replace a couple of unnecessary assertion panics with
error messages.
Signed-off-by: David Teigland <teigland@redhat.com>
This patch eliminates the gfs2_log_element data structure and
rolls its two components into the gfs2_bufdata. This makes the code
easier to understand and makes it easier to migrate to a rbtree
to keep the list sorted.
Signed-off-by: Bob Peterson <rpeterso@redhat.com>
Signed-off-by: Steven Whitehouse <swhiteho@redhat.com>
This patch eliminates gfs2 superblock variable sd_log_le_rg which
is no longer used.
Signed-off-by: Bob Peterson <rpeterso@redhat.com>
Signed-off-by: Steven Whitehouse <swhiteho@redhat.com>
This is another clean up in the logging code. This per-transaction
list was largely unused. Its main function was to ensure that the
number of buffers in a transaction was correct, however that counter
was only used to check the number of buffers in the bd_list_tr, plus
an assert at the end of each transaction. With the assert now changed
to use the calculated buffer counts, we can remove both bd_list_tr and
its associated counter.
This should make the code easier to understand as well as shrinking
a couple of structures.
Signed-off-by: Steven Whitehouse <swhiteho@redhat.com>
Prior to this patch, we have two ways of sending i/o to the log.
One of those is used when we need to allocate both the data
to be written itself and also a buffer head to submit it. This
is done via sb_getblk and friends. This is used mostly for writing
log headers.
The other method is used when writing blocks which have some
in-place counterpart. This is the case for all the metadata
blocks which are journalled, and when journaled data is in use,
for unescaped journalled data blocks.
This patch replaces both of those two methods, and about half
a dozen separate i/o submission points with a single i/o
submission function. We also go direct to bio rather than
using buffer heads, since this allows us to build i/o
requests of the maximum size for the block device in
question. It also reduces the memory required for flushing
the log, which can be very useful in low memory situations.
Signed-off-by: Steven Whitehouse <swhiteho@redhat.com>
Over time, we've slowly eliminated the use of sd_rindex_mutex.
Up to this point, it was only used in two places: function
gfs2_ri_total (which totals the file system size by reading
and parsing the rindex file) and function gfs2_rindex_update
which updates the rgrps in memory. Both of these functions have
the rindex glock to protect them, so the rindex is unnecessary.
Since gfs2_grow writes to the rindex via the meta_fs, the mutex
is in the wrong order according to the normal rules. This patch
eliminates the mutex entirely to avoid the problem.
Signed-off-by: Bob Peterson <rpeterso@redhat.com>
Signed-off-by: Steven Whitehouse <swhiteho@redhat.com>
The stats are divided into two sets: those relating to the
super block and those relating to an individual glock. The
super block stats are done on a per cpu basis in order to
try and reduce the overhead of gathering them. They are also
further divided by glock type.
In the case of both the super block and glock statistics,
the same information is gathered in each case. The super
block statistics are used to provide default values for
most of the glock statistics, so that newly created glocks
should have, as far as possible, a sensible starting point.
The statistics are divided into three pairs of mean and
variance, plus two counters. The mean/variance pairs are
smoothed exponential estimates and the algorithm used is
one which will be very familiar to those used to calculation
of round trip times in network code.
The three pairs of mean/variance measure the following
things:
1. DLM lock time (non-blocking requests)
2. DLM lock time (blocking requests)
3. Inter-request time (again to the DLM)
A non-blocking request is one which will complete right
away, whatever the state of the DLM lock in question. That
currently means any requests when (a) the current state of
the lock is exclusive (b) the requested state is either null
or unlocked or (c) the "try lock" flag is set. A blocking
request covers all the other lock requests.
There are two counters. The first is there primarily to show
how many lock requests have been made, and thus how much data
has gone into the mean/variance calculations. The other counter
is counting queueing of holders at the top layer of the glock
code. Hopefully that number will be a lot larger than the number
of dlm lock requests issued.
So why gather these statistics? There are several reasons
we'd like to get a better idea of these timings:
1. To be able to better set the glock "min hold time"
2. To spot performance issues more easily
3. To improve the algorithm for selecting resource groups for
allocation (to base it on lock wait time, rather than blindly
using a "try lock")
Due to the smoothing action of the updates, a step change in
some input quantity being sampled will only fully be taken
into account after 8 samples (or 4 for the variance) and this
needs to be carefully considered when interpreting the
results.
Knowing both the time it takes a lock request to complete and
the average time between lock requests for a glock means we
can compute the total percentage of the time for which the
node is able to use a glock vs. time that the rest of the
cluster has its share. That will be very useful when setting
the lock min hold time.
The other point to remember is that all times are in
nanoseconds. Great care has been taken to ensure that we
measure exactly the quantities that we want, as accurately
as possible. There are always inaccuracies in any
measuring system, but I hope this is as accurate as we
can reasonably make it.
Signed-off-by: Steven Whitehouse <swhiteho@redhat.com>
If the first mounter fails to recover one of the journals
during mount, the mount should fail.
Signed-off-by: David Teigland <teigland@redhat.com>
Signed-off-by: Steven Whitehouse <swhiteho@redhat.com>
Previously, a spectator mount would not even attempt to do
journal recovery for a failed node. This meant that if all
mounted nodes were spectators, everyone would be stuck after
a node failed, all waiting for recovery to be performed.
This is unnecessary since the failed node had a clean journal.
Instead, allow a spectator mount to do a partial "read only"
recovery, which means it will check if the failed journal is
clean, and if so, report a successful recovery. If the failed
journal is not clean, it reports that journal recovery failed.
This makes it work the same as a read only mount on a read only
block device.
Signed-off-by: David Teigland <teigland@redhat.com>
Signed-off-by: Steven Whitehouse <swhiteho@redhat.com>
This new method of managing recovery is an alternative to
the previous approach of using the userland gfs_controld.
- use dlm slot numbers to assign journal id's
- use dlm recovery callbacks to initiate journal recovery
- use a dlm lock to determine the first node to mount fs
- use a dlm lock to track journals that need recovery
Signed-off-by: David Teigland <teigland@redhat.com>
Signed-off-by: Steven Whitehouse <swhiteho@redhat.com>
This patch separates the code pertaining to allocations into two
parts: quota-related information and block reservations.
This patch also moves all the block reservation structure allocations to
function gfs2_inplace_reserve to simplify the code, and moves
the frees to function gfs2_inplace_release.
Signed-off-by: Bob Peterson <rpeterso@redhat.com>
Signed-off-by: Steven Whitehouse <swhiteho@redhat.com>
This patch removes the vestigial variable al_alloced from
the gfs2_alloc structure. This is another baby step toward
multi-block reservations.
My next planned step is to decouple the quota variables
from the gfs2_alloc structure so we can use a different
method for allocations.
Signed-off-by: Bob Peterson <rpeterso@redhat.com>
Signed-off-by: Steven Whitehouse <swhiteho@redhat.com>
The two variables being initialised in gfs2_inplace_reserve
to track the file & line number of the caller are never
used, so we might as well remove them.
If something does go wrong, then a stack trace is probably
more useful anyway.
Signed-off-by: Steven Whitehouse <swhiteho@redhat.com>
GFS2's fallocate code currently goes through the page cache. Since it's only
writing to the end of the file or to holes in it, it doesn't need to, and it
was causing issues on low memory environments. This patch pulls in some of
Steve's block allocation work, and uses it to simply allocate the blocks for
the file, and zero them out at allocation time. It provides a slight
performance increase, and it dramatically simplifies the code.
Signed-off-by: Benjamin Marzinski <bmarzins@redhat.com>
Signed-off-by: Steven Whitehouse <swhiteho@redhat.com>
This means that after the initial allocation for any inode, the
last used resource group is cached in the inode for future use.
This drastically reduces the number of lookups of resource
groups in the common case, and this the contention on that
data structure.
The allocation algorithm is the same as previously, except that we
always check to see if the goal block is within the cached rgrp
first before going to the rbtree to look one up.
Signed-off-by: Steven Whitehouse <swhiteho@redhat.com>
Since we have ruled out supporting online filesystem shrink,
it is possible to make the resource group list append only
during the life of a super block. This gives several benefits:
Firstly, we only need to read new rindex elements as they are added
rather than needing to reread the whole rindex file each time one
element is added.
Secondly, the rindex glock can be held for much shorter periods of
time, and is completely removed from the fast path for allocations.
The lock is taken in shared mode only when updating the resource
groups when the first allocation occurs, and after a grow has
taken place.
Thirdly, this results in a reduction in code size, and everything
gets a lot simpler to understand in this area.
Signed-off-by: Steven Whitehouse <swhiteho@redhat.com>
Here is an update of Bob's original rbtree patch which, in addition, also
resolves the rather strange ref counting that was being done relating to
the bitmap blocks.
Originally we had a dual system for journaling resource groups. The metadata
blocks were journaled and also the rgrp itself was added to a list. The reason
for adding the rgrp to the list in the journal was so that the "repolish
clones" code could be run to update the free space, and potentially send any
discard requests when the log was flushed. This was done by comparing the
"cloned" bitmap with what had been written back on disk during the transaction
commit.
Due to this, there was a requirement to hang on to the rgrps' bitmap buffers
until the journal had been flushed. For that reason, there was a rather
complicated set up in the ->go_lock ->go_unlock functions for rgrps involving
both a mutex and a spinlock (the ->sd_rindex_spin) to maintain a reference
count on the buffers.
However, the journal maintains a reference count on the buffers anyway, since
they are being journaled as metadata buffers. So by moving the code which deals
with the post-journal accounting for bitmap blocks to the metadata journaling
code, we can entirely dispense with the rather strange buffer ref counting
scheme and also the requirement to journal the rgrps.
The net result of all this is that the ->sd_rindex_spin is left to do exactly
one job, and that is to look after the rbtree or rgrps.
This patch is designed to be a stepping stone towards using RCU for the rbtree
of resource groups, however the reduction in the number of uses of the
->sd_rindex_spin is likely to have benefits for multi-threaded workloads,
anyway.
The patch retains ->go_lock and ->go_unlock for rgrps, however these maybe also
be removed in future in favour of calling the functions directly where required
in the code. That will allow locking of resource groups without needing to
actually read them in - something that could be useful in speeding up statfs.
In the mean time though it is valid to dereference ->bi_bh only when the rgrp
is locked. This is basically the same rule as before, modulo the references not
being valid until the following journal flush.
Signed-off-by: Steven Whitehouse <swhiteho@redhat.com>
Signed-off-by: Bob Peterson <rpeterso@redhat.com>
Cc: Benjamin Marzinski <bmarzins@redhat.com>
If we have got far enough through the inode allocation code
path that an inode has already been allocated, then we must
call iput to dispose of it, if an error occurs during a
later part of the process. This will always be the final iput
since there will be no other references to the inode.
Unlike when the inode has been unlinked, its block state will
be GFS2_BLKST_INODE rather than GFS2_BLKST_UNLINKED so we need
to skip the test in ->evict_inode() for this one case in order
to ensure that it will be deallocated correctly. This patch adds
a new flag in order to ensure that this will happen correctly.
Signed-off-by: Steven Whitehouse <swhiteho@redhat.com>
This patch is a performance improvement for GFS2 in a clustered
environment. It makes the glock hold time self-adjusting.
Signed-off-by: Bob Peterson <rpeterso@redhat.com>
Signed-off-by: Steven Whitehouse <swhiteho@redhat.com>
