Pierre Ossman [Sun, 13 Apr 2008 14:03:38 +0000 (16:03 +0200)]
sdhci: improve no card, no reset quirk
The quirk was meant to just inhibit some resets, but ended up blocking
all of them. Fortunately, this was just what was needed. Change the
comment to reflect reality.
Also, this issue has just been observed on Samsung laptops, so reduce
the number of chips the quirk affects.
Jarkko Lavinen [Wed, 26 Mar 2008 20:10:02 +0000 (16:10 -0400)]
MMC: OMAP: Do not busy wait for end of command for ever
The limit was a fixed 100k limit in the busy loop, which is not
accurate. It would better to have time limit for the worst case
which occurs when sending 80 cycles at 400 kHz and takes about
200 microseconds, so limit the max time spend in the busy loop
for some 250 microseconds.
Signed-off-by: Jarkko Lavinen <jarkko.lavinen@nokia.com> Signed-off-by: Pierre Ossman <drzeus@drzeus.cx>
Jarkko Lavinen [Wed, 26 Mar 2008 20:09:58 +0000 (16:09 -0400)]
MMC: OMAP: Start new commands from work queue instead of irq
Use work queues for starting new commands instead of starting them
directly from irq handler. The command scheduling needs to be delayed
a bit for some cards which should not be done from an interrupt.
Signed-off-by: Jarkko Lavinen <jarkko.lavinen@nokia.com> Signed-off-by: Pierre Ossman <drzeus@drzeus.cx>
Jarkko Lavinen [Wed, 26 Mar 2008 20:09:52 +0000 (16:09 -0400)]
MMC: OMAP: Lazy clock shutdown
MMCA spec says the mmc clock should be kept running for at least
8 cycles after the last RW request. Ensure this with lazy clock
disable after a request, or with an explicit delay before
switching a slot.
Signed-off-by: Jarkko Lavinen <jarkko.lavinen@nokia.com> Signed-off-by: Pierre Ossman <drzeus@drzeus.cx>
Jarkko Lavinen [Wed, 26 Mar 2008 20:09:42 +0000 (16:09 -0400)]
MMC: OMAP: Use tasklet instead of workqueue for cover switch notification
The cover waitqueue is occasionally scheduled twice from timer
and the interrupt and oops follows. It would have been possible
to fix this problem with spinlocks but using tasklet was a dropin
solution with no need for locking.
This path also adds some cleanups.
Signed-off-by: Jarkko Lavinen <jarkko.lavinen@nokia.com> Signed-off-by: Hiroshi DOYU <Hiroshi.DOYU@nokia.com> Signed-off-by: Pierre Ossman <drzeus@drzeus.cx>
Kyungmin Park [Wed, 26 Mar 2008 20:09:38 +0000 (16:09 -0400)]
MMC: OMAP: Check the get_cover_state function pointer if not set
If the get_cover_state is not set, it occurs the oops.
Signed-off-by: Kyungmin Park <kyungmin.park@samsung.com> Signed-off-by: Tony Lindgren <tony@atomide.com> Signed-off-by: Pierre Ossman <drzeus@drzeus.cx>
MMC: OMAP: Using setup_timer instead of init_timer
Using setup_timer() instead of init_timer() on omap.c file.
Signed-off-by: Carlos Eduardo Aguiar <carlos.aguiar@indt.org.br> Signed-off-by: Tony Lindgren <tony@atomide.com> Signed-off-by: Pierre Ossman <drzeus@drzeus.cx>
Jarkko Lavinen [Wed, 26 Mar 2008 20:09:29 +0000 (16:09 -0400)]
MMC: OMAP: Abort stuck commands
When a card is removed while it is being accessed, a command can get stuck so
that no timeout or end of command interrupt ever occurs. The command getting
stuck is almost always CDM12, but also the other commands can get stuck. Catch
a stuck command with a timer and try sending the initialization stream until
the controller starts running again and responds with the end of command
status.
Signed-off-by: Jarkko Lavinen <jarkko.lavinen@nokia.com> Signed-off-by: Carlos Eduardo Aguiar <carlos.aguiar@indt.org.br> Signed-off-by: Tony Lindgren <tony@atomide.com> Signed-off-by: Pierre Ossman <drzeus@drzeus.cx>
Juha Yrjola [Wed, 26 Mar 2008 20:09:26 +0000 (16:09 -0400)]
MMC: OMAP: General cleanup for MMC multislot support
General code cleanup, modifications at some dev_* functions and
other hacks at mmc_omap_irq() for MMC multislot support.
Signed-off-by: Juha Yrjola <juha.yrjola@solidboot.com> Signed-off-by: Carlos Eduardo Aguiar <carlos.aguiar@indt.org.br> Signed-off-by: Tony Lindgren <tony@atomide.com> Signed-off-by: Pierre Ossman <drzeus@drzeus.cx>
Juha Yrjola [Wed, 26 Mar 2008 20:09:22 +0000 (16:09 -0400)]
MMC: OMAP: Power functions modified to MMC multislot support
Modifications at power functions to MMC multislot support. This patch
also move board-specific code out of MMC OMAP driver.
Signed-off-by: Juha Yrjola <juha.yrjola@solidboot.com> Signed-off-by: Carlos Eduardo Aguiar <carlos.aguiar@indt.org.br> Signed-off-by: Tony Lindgren <tony@atomide.com> Signed-off-by: Pierre Ossman <drzeus@drzeus.cx>
Juha Yrjola [Wed, 26 Mar 2008 20:09:16 +0000 (16:09 -0400)]
MMC: OMAP: Fix timeout calculation for MMC multislot support
Fix the data timeout calculation for MMC multislot support.
Signed-off-by: Juha Yrjola <juha.yrjola@solidboot.com> Signed-off-by: Jarkko Lavinen <jarkko.lavinen@nokia.com> Signed-off-by: Carlos Eduardo Aguiar <carlos.aguiar@indt.org.br> Signed-off-by: Tony Lindgren <tony@atomide.com> Signed-off-by: Pierre Ossman <drzeus@drzeus.cx>
Juha Yrjola [Wed, 26 Mar 2008 20:09:12 +0000 (16:09 -0400)]
MMC: OMAP: New release dma and abort xfer functions
New functions to support MMC multislot:
mmc_omap_release_dma() and mmc_omap_abort_xfer().
Signed-off-by: Juha Yrjola <juha.yrjola@solidboot.com> Signed-off-by: Carlos Eduardo Aguiar <carlos.aguiar@indt.org.br> Signed-off-by: Tony Lindgren <tony@atomide.com> Signed-off-by: Pierre Ossman <drzeus@drzeus.cx>
Juha Yrjola [Wed, 26 Mar 2008 20:09:08 +0000 (16:09 -0400)]
MMC: OMAP: Add back cover switch support
This patch adds back MMC cover switch support in a way that
supports multiple slots.
Signed-off-by: Juha Yrjola <juha.yrjola@solidboot.com> Signed-off-by: Jarkko Lavinen <jarkko.lavinen@nokia.com> Signed-off-by: Carlos Eduardo Aguiar <carlos.aguiar@indt.org.br> Signed-off-by: Tony Lindgren <tony@atomide.com> Signed-off-by: Pierre Ossman <drzeus@drzeus.cx>
Juha Yrjola [Wed, 26 Mar 2008 20:08:57 +0000 (16:08 -0400)]
MMC: OMAP: Introduce new multislot structure and change driver to use it
Introduce new MMC multislot structure and change driver to use it.
Note that MMC clocking is now enabled in mmc_omap_select_slot()
and disabled in mmc_omap_release_slot().
Signed-off-by: Juha Yrjola <juha.yrjola@solidboot.com> Signed-off-by: Jarkko Lavinen <jarkko.lavinen@nokia.com> Signed-off-by: Carlos Eduardo Aguiar <carlos.aguiar@indt.org.br> Signed-off-by: Tony Lindgren <tony@atomide.com> Signed-off-by: Pierre Ossman <drzeus@drzeus.cx>
Tony Lindgren [Wed, 26 Mar 2008 20:08:53 +0000 (16:08 -0400)]
MMC: OMAP: Remove cover switch handling to allow adding multislot support
This patch removes the MMC cover switch handling temporarily
to make following multislot patches cleaner.
MMC cover switch handling will be added back in later patches
after adding basic multislot support.
Signed-off-by: Tony Lindgren <tony@atomide.com> Signed-off-by: Carlos Eduardo Aguiar <carlos.aguiar@indt.org.br> Signed-off-by: Pierre Ossman <drzeus@drzeus.cx>
According with commit 255d01af9a990fd5166f04ed0cc0b30b7b67e81e
from Linux-OMAP tree, the BYTEBLOCK capability was removed by Pierre Ossman.
MMC_CAP_BYTEBLOCK is not defined causing the compile error:
drivers/mmc/host/omap.c: In function `mmc_omap_probe':
drivers/mmc/host/omap.c:1077: error: `MMC_CAP_BYTEBLOCK' undeclared (first use in this function)
drivers/mmc/host/omap.c:1077: error: (Each undeclared identifier is reported only once
drivers/mmc/host/omap.c:1077: error: for each function it appears in.)
Signed-off-by: Francisco Alecrim <francisco.alecrim@indt.org.br> Signed-off-by: Carlos Eduardo Aguiar <carlos.aguiar@indt.org.br> Signed-off-by: Tony Lindgren <tony@atomide.com> Signed-off-by: Pierre Ossman <drzeus@drzeus.cx>
Tony Lindgren [Wed, 26 Mar 2008 20:08:45 +0000 (16:08 -0400)]
MMC: OMAP: Remove extra divisor increase
As noted by Kyungmin Park, the divisor calculation has
an unnecessary increase.
Signed-off-by: Tony Lindgren <tony@atomide.com> Signed-off-by: Carlos Eduardo Aguiar <carlos.aguiar@indt.org.br> Signed-off-by: Pierre Ossman <drzeus@drzeus.cx>
This patch removes some opcodes from host driver so there's no need on putting
'#include <linux/mmc/mmc.h>', that should not be needed in host drivers.
Signed-off-by: Carlos Eduardo Aguiar <carlos.aguiar@indt.org.br> Signed-off-by: Pierre Ossman <drzeus@drzeus.cx>
Robert P. J. Day [Fri, 28 Mar 2008 21:34:47 +0000 (14:34 -0700)]
mmc: use shorter, equivalent set_current_state()
Signed-off-by: Robert P. J. Day <rpjday@crashcourse.ca> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Pierre Ossman <drzeus@drzeus.cx>
Pierre Ossman [Sat, 8 Mar 2008 22:43:19 +0000 (23:43 +0100)]
mmc: set controller name early
Reorganise code so that mmc_hostname() works directly after allocation.
That way host drivers can use that name for resource allocations and
messages during probing.
Merge branch 'upstream-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/mfasheh/ocfs2
* 'upstream-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/mfasheh/ocfs2: (64 commits)
ocfs2/net: Add debug interface to o2net
ocfs2: Only build ocfs2/dlm with the o2cb stack module
ocfs2/cluster: Get rid of arguments to the timeout routines
ocfs2: Put tree in MAINTAINERS
ocfs2: Use BUG_ON
ocfs2: Convert ocfs2 over to unlocked_ioctl
ocfs2: Improve rename locking
fs/ocfs2/aops.c: test for IS_ERR rather than 0
ocfs2: Add inode stealing for ocfs2_reserve_new_inode
ocfs2: Add ac_alloc_slot in ocfs2_alloc_context
ocfs2: Add a new parameter for ocfs2_reserve_suballoc_bits
ocfs2: Enable cross extent block merge.
ocfs2: Add support for cross extent block
ocfs2: Move /sys/o2cb to /sys/fs/o2cb
sysfs: Allow removal of symlinks in the sysfs root
ocfs2: Reconnect after idle time out.
ocfs2/dlm: Cleanup lockres print
ocfs2/dlm: Fix lockname in lockres print function
ocfs2/dlm: Move dlm_print_one_mle() from dlmmaster.c to dlmdebug.c
ocfs2/dlm: Dumps the purgelist into a debugfs file
...
* git://git.kernel.org/pub/scm/linux/kernel/git/steve/gfs2-2.6-nmw: (49 commits)
[GFS2] fix assertion in log_refund()
[GFS2] fix GFP_KERNEL misuses
[GFS2] test for IS_ERR rather than 0
[GFS2] Invalidate cache at correct point
[GFS2] fs/gfs2/recovery.c: suppress warnings
[GFS2] Faster gfs2_bitfit algorithm
[GFS2] Streamline quota lock/check for no-quota case
[GFS2] Remove drop of module ref where not needed
[GFS2] gfs2_adjust_quota has broken unstuffing code
[GFS2] possible null pointer dereference fixup
[GFS2] Need to ensure that sector_t is 64bits for GFS2
[GFS2] re-support special inode
[GFS2] remove gfs2_dev_iops
[GFS2] fix file_system_type leak on gfs2meta mount
[GFS2] Allow bmap to allocate extents
[GFS2] Fix a page lock / glock deadlock
[GFS2] proper extern for gfs2/locking/dlm/mount.c:gdlm_ops
[GFS2] gfs2/ops_file.c should #include "ops_inode.h"
[GFS2] be*_add_cpu conversion
[GFS2] Fix bug where we called drop_bh incorrectly
...
This patch exposes o2net information via debugfs. The information includes
the list of sockets (sock_containers) as well as the list of outstanding
messages (send_tracking). Useful for o2dlm debugging.
(This patch is derived from an earlier one written by Zach Brown that
exposed the same information via /proc.)
[Mark: checkpatch fixes]
Signed-off-by: Sunil Mushran <sunil.mushran@oracle.com> Reviewed-by: Joel Becker <joel.becker@oracle.com> Signed-off-by: Mark Fasheh <mfasheh@suse.com>
Mark Fasheh [Fri, 4 Apr 2008 19:45:55 +0000 (12:45 -0700)]
ocfs2: Only build ocfs2/dlm with the o2cb stack module
fs/ocfs2/dlm/ocfs2_dlm.ko and fs/ocfs2/dlm/ocfs2_dlmfs.ko get built if
CONFIG_FS_OCFS2 is specified. This isn't quite how it should happen any more
- the "o2cb" dlm modules should only be built if CONFIG_FS_OCFS2_O2CB is
set, so update the dlm Makefile accordingly.
Signed-off-by: Mark Fasheh <mfasheh@suse.com> Acked-by: Randy Dunlap <randy.dunlap@oracle.com> Acked-by: Joel Becker <joel.becker@oracle.com>
Jeff Mahoney [Fri, 28 Mar 2008 23:44:13 +0000 (16:44 -0700)]
ocfs2/cluster: Get rid of arguments to the timeout routines
We keep seeing bug reports related to NULL pointer derefs in
o2net_set_nn_state(). When I originally wrote up the configurable timeout
patch, I had tried to plan for multiple clusters. This was silly.
The timeout routines all use o2nm_single_cluster so there's no point in
passing an argument at all. This patch removes the arguments and kills those
bugs dead.
Signed-off-by: Jeff Mahoney <jeffm@suse.com> Signed-off-by: Mark Fasheh <mfasheh@suse.com>
Julia Lawall [Fri, 28 Mar 2008 21:43:10 +0000 (14:43 -0700)]
fs/ocfs2/aops.c: test for IS_ERR rather than 0
The function ocfs2_start_trans always returns either a valid pointer or a
value made with ERR_PTR, so its result should be tested with IS_ERR, not
with a test for 0.
Signed-off-by: Julia Lawall <julia@diku.dk> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Mark Fasheh <mfasheh@suse.com>
Tao Ma [Wed, 5 Mar 2008 08:11:46 +0000 (16:11 +0800)]
ocfs2: Add inode stealing for ocfs2_reserve_new_inode
Inode allocation is modified to look in other nodes allocators during
extreme out of space situations. We retry our own slot when space is freed
back to the global bitmap, or whenever we've allocated more than 1024 inodes
from another slot.
Signed-off-by: Tao Ma <tao.ma@oracle.com> Signed-off-by: Mark Fasheh <mfasheh@suse.com>
Tao Ma [Mon, 3 Mar 2008 09:12:30 +0000 (17:12 +0800)]
ocfs2: Add ac_alloc_slot in ocfs2_alloc_context
In inode stealing, we no longer restrict the allocation to
happen in the local node. So it is neccessary for us to add
a new member in ocfs2_alloc_context to indicate which slot
we are using for allocation. We also modify the process of
local alloc so that this member can be used there also.
Signed-off-by: Tao Ma <tao.ma@oracle.com> Signed-off-by: Sunil Mushran <sunil.mushran@oracle.com> Signed-off-by: Mark Fasheh <mfasheh@suse.com>
Tao Ma [Mon, 3 Mar 2008 09:12:09 +0000 (17:12 +0800)]
ocfs2: Add a new parameter for ocfs2_reserve_suballoc_bits
In some cases(Inode stealing from other nodes), we may not want
ocfs2_reserve_suballoc_bits to allocate new groups from the
global_bitmap since it may already be full. So add a new parameter
for this.
Signed-off-by: Tao Ma <tao.ma@oracle.com> Signed-off-by: Sunil Mushran <sunil.mushran@oracle.com> Signed-off-by: Mark Fasheh <mfasheh@suse.com>
Tao Ma [Wed, 30 Jan 2008 06:21:32 +0000 (14:21 +0800)]
ocfs2: Enable cross extent block merge.
In ocfs2_figure_merge_contig_type, we judge whether there exists
a cross extent block merge and enable it by setting CONTIG_LEFT
and CONTIG_RIGHT accordingly.
Signed-off-by: Tao Ma <tao.ma@oracle.com> Signed-off-by: Mark Fasheh <mfasheh@suse.com>
Tao Ma [Wed, 30 Jan 2008 06:21:05 +0000 (14:21 +0800)]
ocfs2: Add support for cross extent block
In ocfs2_merge_rec_left, when we find the merge extent is "CONTIG_RIGHT"
with the first extent record of the next extent block, we will merge it to
the next extent block and change all the related extent blocks accordingly.
In ocfs2_merge_rec_right, when we find the merge extent is "CONTIG_LEFT"
with the last extent record of the previous extent block, we will merge
it to the prevoius extent block and change all the related extent blocks
accordingly.
As for CONTIG_LEFTRIGHT, we will handle CONTIG_RIGHT first so that when
the index is zero, the merge process will be more efficient and easier.
Signed-off-by: Tao Ma <tao.ma@oracle.com> Signed-off-by: Mark Fasheh <mfasheh@suse.com>
Mark Fasheh [Wed, 30 Jan 2008 01:08:26 +0000 (17:08 -0800)]
ocfs2: Move /sys/o2cb to /sys/fs/o2cb
/sys/fs is where we really want file system specific sysfs objects.
Ocfs2-tools has been updated to look in /sys/fs/o2cb. We can maintain
backwards compatibility with old ocfs2-tools by using a sysfs symlink. After
some time (2 years), the symlink can be safely removed. This patch also adds
documentation to make it easier for people to figure out what /sys/fs/o2cb
is used for.
Mark Fasheh [Tue, 29 Jan 2008 22:35:18 +0000 (14:35 -0800)]
sysfs: Allow removal of symlinks in the sysfs root
Allow callers of sysfs_remove_link() to pass a NULL kobj, in which case
sysfs_root will be used as the parent directory. This allows us to tear down
top level symlinks created via sysfs_create_link(), which already has
similar handling of a NULL parent object.
Signed-off-by: Mark Fasheh <mfasheh@suse.com> Acked-by: Greg Kroah-Hartman <gregkh@suse.de>
Tao Ma [Wed, 5 Mar 2008 07:50:12 +0000 (15:50 +0800)]
ocfs2: Reconnect after idle time out.
Currently, o2net connects to a node on hb_up and disconnects on
hb_down and net timeout.
It disconnects on net timeout is ok, but it should attempt to
reconnect back. This is because sometimes nodes get overloaded
enough that the network connection breaks but the disk hb does not.
And if we get into that situation, we either fence (unnecessarily)
or wait for its disk hb to die (and sometimes hang in the process).
So in this updated scheme, when the network disconnects, we keep
attempting to reconnect till we succeed or we get a disk hb down
event.
If the other node is really dead, then we will eventually get a
node down event. If not, we should be able to connect again and
continue.
Signed-off-by: Tao Ma <tao.ma@oracle.com> Signed-off-by: Mark Fasheh <mfasheh@suse.com>
Sunil Mushran [Fri, 14 Mar 2008 18:18:24 +0000 (11:18 -0700)]
ocfs2/dlm: Cleanup lockres print
A previous patch added KERN_NOTICE to printks printing the lockres that
cluttered the output. This patch removes the log level. For people concerned
with syslog clutter, please note we now use this facility to print lockres
only during an error.
Signed-off-by: Sunil Mushran <sunil.mushran@oracle.com> Signed-off-by: Mark Fasheh <mfasheh@suse.com>
Sunil Mushran [Mon, 10 Mar 2008 22:16:29 +0000 (15:16 -0700)]
ocfs2/dlm: Fix lockname in lockres print function
__dlm_print_one_lock_resource was printing lockname incorrectly.
Also, we now use printk directly instead of mlog as the latter prints
the line context which is not useful for this print.
Signed-off-by: Sunil Mushran <sunil.mushran@oracle.com> Signed-off-by: Joel Becker <joel.becker@oracle.com> Signed-off-by: Mark Fasheh <mfasheh@suse.com>
Sunil Mushran [Mon, 10 Mar 2008 22:16:25 +0000 (15:16 -0700)]
ocfs2/dlm: Move struct dlm_master_list_entry to dlmcommon.h
This patch moves some mle related definitions from dlmmaster.c
to dlmcommon.h. Future patches need these definitions to dump mle
debugging information.
Signed-off-by: Sunil Mushran <sunil.mushran@oracle.com> Signed-off-by: Joel Becker <joel.beckeroracle.com> Signed-off-by: Mark Fasheh <mfasheh@suse.com>
Sunil Mushran [Mon, 10 Mar 2008 22:16:21 +0000 (15:16 -0700)]
ocfs2/dlm: Link all lockres' to a tracking list
This patch links all the lockres' to a tracking list in dlm_ctxt.
We will use this in an upcoming patch that will walk the entire
list and to dump the lockres states to a debugfs file.
Signed-off-by: Sunil Mushran <sunil.mushran@oracle.com> Signed-off-by: Joel Becker <joel.becker@oracle.com> Signed-off-by: Mark Fasheh <mfasheh@suse.com>
Sunil Mushran [Mon, 10 Mar 2008 22:16:20 +0000 (15:16 -0700)]
ocfs2/dlm: Create slabcaches for lock and lockres
This patch makes the o2dlm allocate memory for lockres, lockname and lock
structures from slabcaches rather than kmalloc. This allows us to not only
make these allocs more efficient but also allows us to track the memory being
consumed by these structures.
Signed-off-by: Sunil Mushran <sunil.mushran@oracle.com> Signed-off-by: Joel Becker <joel.becker@oracle.com> Signed-off-by: Mark Fasheh <mfasheh@suse.com>
Joel Becker [Wed, 5 Mar 2008 01:58:56 +0000 (17:58 -0800)]
ocfs2: Allow selection of cluster plug-ins.
ocfs2 now supports plug-ins for the classic O2CB stack as well as
userspace cluster stacks in conjunction with fs/dlm. This allows zero,
one, or both of the plug-ins to be selected in Kconfig. For local mounts
(non-clustered), neither plug-in is needed. Both plugins can be loaded
at one time, the runtime will select the one needed for the cluster
systme in use.
Signed-off-by: Joel Becker <joel.becker@oracle.com> Signed-off-by: Mark Fasheh <mfasheh@suse.com>
Joel Becker [Wed, 5 Mar 2008 00:09:39 +0000 (16:09 -0800)]
ocfs2: Change mlog_bug_on to BUG_ON in ocfs2_lockid.h
The masklog code is in the o2cb stack, but ocfs2_lockid.h now needs to
be included by the user stack. The BUG() in ocfs2_lock_type_string()
does not need masklog support, so change it to a regular BUG_ON().
Signed-off-by: Joel Becker <joel.becker@oracle.com> Signed-off-by: Mark Fasheh <mfasheh@suse.com>
Joel Becker [Wed, 20 Feb 2008 23:39:44 +0000 (15:39 -0800)]
ocfs2: Add the 'set version' message to the ocfs2_control device.
The "SETV" message sets the filesystem locking protocol version as
negotiated by the client. The client negotiates based on the maximum
version advertised in /sys/fs/ocfs2/max_locking_protocol.
Signed-off-by: Joel Becker <joel.becker@oracle.com> Signed-off-by: Mark Fasheh <mfasheh@suse.com>
Joel Becker [Wed, 20 Feb 2008 22:44:34 +0000 (14:44 -0800)]
ocfs2: Add the local node id to the handshake.
This is the second part of the ocfs2_control handshake. After
negotiating the ocfs2_control protocol, the daemon tells the filesystem
what the local node id is via the SETN message.
Signed-off-by: Joel Becker <joel.becker@oracle.com> Signed-off-by: Mark Fasheh <mfasheh@suse.com>
Joel Becker [Tue, 19 Feb 2008 03:40:12 +0000 (19:40 -0800)]
ocfs2: Start the ocfs2_control handshake.
When a control daemon opens the ocfs2_control device, it must perform a
handshake to tell the filesystem it is something capable of monitoring
cluster status. Only after the handshake is complete will the filesystem
allow mounts.
This is the first part of the handshake. The daemon reads all supported
ocfs2_control protocols, then writes in the protocol it will use.
Signed-off-by: Joel Becker <joel.becker@oracle.com> Signed-off-by: Mark Fasheh <mfasheh@suse.com>
Joel Becker [Fri, 1 Feb 2008 23:17:30 +0000 (15:17 -0800)]
ocfs2: Add the 'cluster_stack' sysfs file.
Userspace can now query and specify the cluster stack in use via the
/sys/fs/ocfs2/cluster_stack file. By default, it is 'o2cb', which is
the classic stack. Thus, old tools that do not know how to modify this
file will work just fine. The stack cannot be modified if there is a
live filesystem.
ocfs2_cluster_connect() now takes the expected cluster stack as an
argument. This way, the filesystem and the stack glue ensure they are
speaking to the same backend.
If the stack is 'o2cb', the o2cb stack plugin is used. For any other
value, the fsdlm stack plugin is selected.
Signed-off-by: Joel Becker <joel.becker@oracle.com> Signed-off-by: Mark Fasheh <mfasheh@suse.com>
Joel Becker [Fri, 1 Feb 2008 23:08:23 +0000 (15:08 -0800)]
ocfs2: Add the USERSPACE_STACK incompat bit.
The filesystem gains the USERSPACE_STACK incomat bit and the
s_cluster_info field on the superblock. When a userspace stack is in
use, the name of the stack is stored on-disk for mount-time
verification.
The "cluster_stack" option is added to mount(2) processing. The mount
process needs to pass the matching stack name. If the passed name and
the on-disk name do not match, the mount is failed.
When using the classic o2cb stack, the incompat bit is *not* set and no
mount option is used other than the usual heartbeat=local. Thus, the
filesystem is compatible with older tools.
Signed-off-by: Joel Becker <joel.becker@oracle.com> Signed-off-by: Mark Fasheh <mfasheh@suse.com>
Joel Becker [Fri, 1 Feb 2008 07:56:17 +0000 (23:56 -0800)]
ocfs2: Create stack glue sysfs files.
Introduce a set of sysfs files that describe the current stack glue
state. The files live under /sys/fs/ocfs2. The locking_protocol file
displays the version of ocfs2's locking code. The
loaded_cluster_plugins file displays all of the currently loaded stack
plugins. When filesystems are mounted, the active_cluster_plugin file
will display the plugin in use.
Signed-off-by: Joel Becker <joel.becker@oracle.com> Signed-off-by: Mark Fasheh <mfasheh@suse.com>
Joel Becker [Fri, 1 Feb 2008 23:03:57 +0000 (15:03 -0800)]
ocfs2: Break out stackglue into modules.
We define the ocfs2_stack_plugin structure to represent a stack driver.
The o2cb stack code is split into stack_o2cb.c. This becomes the
ocfs2_stack_o2cb.ko module.
The stackglue generic functions are similarly split into the
ocfs2_stackglue.ko module. This module now provides an interface to
register drivers. The ocfs2_stack_o2cb driver registers itself. As
part of this interface, ocfs2_stackglue can load drivers on demand.
This is accomplished in ocfs2_cluster_connect().
ocfs2_cluster_disconnect() is now notified when a _hangup() is pending.
If a hangup is pending, it will not release the driver module and will
let _hangup() do that.
Signed-off-by: Joel Becker <joel.becker@oracle.com>
Joel Becker [Fri, 1 Feb 2008 23:02:36 +0000 (15:02 -0800)]
ocfs2: Create ocfs2_stack_operations and split out the o2cb stack.
Define the ocfs2_stack_operations structure. Build o2cb_stack_ops from
all of the o2cb-specific stack functions. Change the generic stack glue
functions to call the stack_ops instead of the o2cb functions directly.
The o2cb functions are moved to stack_o2cb.c. The headers are cleaned up
to where only needed headers are included.
In this code, stackglue.c and stack_o2cb.c refer to some shared
extern variables. When they become modules, that will change.
Signed-off-by: Joel Becker <joel.becker@oracle.com> Signed-off-by: Mark Fasheh <mfasheh@suse.com>
Joel Becker [Fri, 1 Feb 2008 22:51:03 +0000 (14:51 -0800)]
ocfs2: Split o2cb code from generic stack functions.
Split off the o2cb-specific funtionality from the generic stack glue
calls. This is a precurser to wrapping the o2cb functionality in an
operations vector.
Signed-off-by: Joel Becker <joel.becker@oracle.com> Signed-off-by: Mark Fasheh <mfasheh@suse.com>
Joel Becker [Wed, 30 Jan 2008 00:59:55 +0000 (16:59 -0800)]
ocfs2: Abstract out a debugging function for underlying dlms.
dlmglue.c was still referencing a raw o2dlm lksb in one instance. Let's
create a generic ocfs2_dlm_dump_lksb() function. This allows underlying
DLMs to print whatever they want about their lock.
We then move the o2dlm dump into stackglue.c where it belongs.
Signed-off-by: Joel Becker <joel.becker@oracle.com> Signed-off-by: Mark Fasheh <mfasheh@suse.com>
Joel Becker [Fri, 1 Feb 2008 22:45:08 +0000 (14:45 -0800)]
ocfs2: Remove CANCELGRANT from the view of dlmglue.
o2dlm has the non-standard behavior of providing a cancel callback
(unlock_ast) even when the cancel has failed (the locking operation
succeeded without canceling). This is called CANCELGRANT after the
status code sent to the callback. fs/dlm does not provide this
callback, so dlmglue must be changed to live without it.
o2dlm_unlock_ast_wrapper() in stackglue now ignores CANCELGRANT calls.
Because dlmglue no longer sees CANCELGRANT, ocfs2_unlock_ast() no longer
needs to check for it. ocfs2_locking_ast() must catch that a cancel was
tried and clear the cancel state.
Making these changes opens up a locking race. dlmglue uses the the
OCFS2_LOCK_BUSY flag to ensure only one thread is calling the dlm at any
one time. But dlmglue must unlock the lockres before calling into the
dlm. In the small window of time between unlocking the lockres and
calling the dlm, the downconvert thread can try to cancel the lock. The
downconvert thread is checking the OCFS2_LOCK_BUSY flag - it doesn't
know that ocfs2_dlm_lock() has not yet been called.
Because ocfs2_dlm_lock() has not yet been called, the cancel operation
will just be a no-op. There's nothing to cancel. With CANCELGRANT,
dlmglue uses the CANCELGRANT callback to clear up the cancel state.
When it comes around again, it will retry the cancel. Eventually, the
first thread will have called into ocfs2_dlm_lock(), and either the
lock or the cancel will succeed. The downconvert thread can then do its
downconvert.
Without CANCELGRANT, there is nothing to clean up the cancellation
state. The downconvert thread does not know to retry its operations.
More importantly, the original lock may be blocking on the other node
that is trying to cancel us. With neither able to make progress, the
ast is never called and the cancellation state is never cleaned up that
way. dlmglue is deadlocked.
The OCFS2_LOCK_PENDING flag is introduced to remedy this window. It is
set at the same time OCFS2_LOCK_BUSY is. Thus, the downconvert thread
can check whether the lock is cancelable. If not, it just loops around
to try again. Once ocfs2_dlm_lock() is called, the thread then clears
OCFS2_LOCK_PENDING and wakes the downconvert thread. Now, if the
downconvert thread finds the lock BUSY, it can safely try to cancel it.
Whether the cancel works or not, the state will be properly set and the
lock processing can continue.
Signed-off-by: Joel Becker <joel.becker@oracle.com> Signed-off-by: Mark Fasheh <mfasheh@suse.com>
Mark Fasheh [Wed, 30 Jan 2008 00:59:56 +0000 (16:59 -0800)]
ocfs2: Fill node number during cluster stack init
It doesn't make sense to query for a node number before connecting to the
cluster stack. This should be safe to do because node_num is only just
printed,
and we're actually only moving the setting of node num a small amount
further in the mount process.
[ Disconnect when node query fails -- Joel ]
Reviewed-by: Joel Becker <joel.becker@oracle.com> Signed-off-by: Mark Fasheh <mfasheh@suse.com>
Joel Becker [Wed, 30 Jan 2008 00:59:56 +0000 (16:59 -0800)]
ocfs2: Move o2hb functionality into the stack glue.
The last bit of classic stack used directly in ocfs2 code is o2hb.
Specifically, the check for heartbeat during mount and the call to
ocfs2_hb_ctl during unmount.
We create an extra API, ocfs2_cluster_hangup(), to encapsulate the call
to ocfs2_hb_ctl. Other stacks will just leave hangup() empty.
The check for heartbeat is moved into ocfs2_cluster_connect(). It will
be matched by a similar check for other stacks.
With this change, only stackglue.c includes cluster/ headers.
Signed-off-by: Joel Becker <joel.becker@oracle.com> Signed-off-by: Mark Fasheh <mfasheh@suse.com>
Joel Becker [Wed, 30 Jan 2008 23:38:24 +0000 (15:38 -0800)]
ocfs2: Abstract out node number queries.
ocfs2 asks the cluster stack for the local node's node number for two
reasons; to fill the slot map and to print it. While the slot map isn't
necessary for userspace cluster stacks, the printing is very nice for
debugging. Thus we add ocfs2_cluster_this_node() as a generic API to get
this value. It is anticipated that the slot map will not be used under a
userspace cluster stack, so validity checks of the node num only need to
exist in the slot map code. Otherwise, it just gets used and printed as an
opaque value.
[ Fixed up some "int" versus "unsigned int" issues and made osb->node_num
truly opaque. --Mark ]
Signed-off-by: Joel Becker <joel.becker@oracle.com> Signed-off-by: Mark Fasheh <mfasheh@suse.com>
Joel Becker [Fri, 1 Feb 2008 22:39:35 +0000 (14:39 -0800)]
ocfs2: Introduce the new ocfs2_cluster_connect/disconnect() API.
This step introduces a cluster stack agnostic API for initializing and
exiting. fs/ocfs2/dlmglue.c no longer uses o2cb/o2dlm knowledge to
connect to the stack. It is all handled in stackglue.c.
heartbeat.c no longer needs to know how it gets called.
ocfs2_do_node_down() is now a clean recovery trigger.
The big gotcha is the ordering of initializations and de-initializations done
underneath ocfs2_cluster_connect(). ocfs2_dlm_init() used to do all
o2dlm initialization in one block. Thus, the o2dlm functionality of
ocfs2_cluster_connect() is very straightforward. ocfs2_dlm_shutdown(),
however, did a few things between de-registration of the eviction
callback and actually shutting down the domain. Now de-registration and
shutdown of the domain are wrapped within the single
ocfs2_cluster_disconnect() call. I've checked the code paths to make
sure we can safely tear down things in ocfs2_dlm_shutdown() before
calling ocfs2_cluster_disconnect(). The filesystem has already set
itself to ignore the callback.
Signed-off-by: Joel Becker <joel.becker@oracle.com> Signed-off-by: Mark Fasheh <mfasheh@suse.com>
Joel Becker [Fri, 1 Feb 2008 20:15:37 +0000 (12:15 -0800)]
ocfs2: Use -errno instead of dlm_status for ocfs2_dlm_lock/unlock() API.
Change the ocfs2_dlm_lock/unlock() functions to return -errno values.
This is the first step towards elminiating dlm_status in
fs/ocfs2/dlmglue.c. The change also passes -errno values to
->unlock_ast().
[ Fix a return code in dlmglue.c and change the error translation table into
an array of ints. --Mark ]
Signed-off-by: Joel Becker <joel.becker@oracle.com> Signed-off-by: Mark Fasheh <mfasheh@suse.com>
Joel Becker [Wed, 30 Jan 2008 01:37:32 +0000 (17:37 -0800)]
ocfs2: Separate out dlm lock functions.
This is the first in a series of patches to isolate ocfs2 from the
underlying cluster stack. Here we wrap the dlm locking functions with
ocfs2-specific calls. Because ocfs2 always uses the same dlm lock status
callbacks, we can eliminate the callbacks from the filesystem visible
functions.
Signed-off-by: Joel Becker <joel.becker@oracle.com> Signed-off-by: Mark Fasheh <mfasheh@suse.com>
Joel Becker [Fri, 1 Feb 2008 20:06:54 +0000 (12:06 -0800)]
ocfs2: New slot map format
The old slot map had a few limitations:
- It was limited to one block, so the maximum slot count was 255.
- Each slot was signed 16bits, limiting node numbers to INT16_MAX.
- An empty slot was marked by the magic 0xFFFF (-1).
The new slot map format provides 32bit node numbers (UINT32_MAX), a
separate space to mark a slot in use, and extra room to grow. The slot
map is now bounded by i_size, not a block.
Signed-off-by: Joel Becker <joel.becker@oracle.com> Signed-off-by: Mark Fasheh <mfasheh@suse.com>
Joel Becker [Fri, 1 Feb 2008 20:04:48 +0000 (12:04 -0800)]
ocfs2: De-magic the in-memory slot map.
The in-memory slot map uses the same magic as the on-disk one. There is
a special value to mark a slot as invalid. It relies on the size of
certain types and so on.
Write a new in-memory map that keeps validity as a separate field. Outside
of the I/O functions, OCFS2_INVALID_SLOT now means what it is supposed to.
It also is no longer tied to the type size.
This also means that only the I/O functions refer to 16bit quantities.
Signed-off-by: Joel Becker <joel.becker@oracle.com> Signed-off-by: Mark Fasheh <mfasheh@suse.com>
Joel Becker [Fri, 1 Feb 2008 20:03:57 +0000 (12:03 -0800)]
ocfs2: Change the recovery map to an array of node numbers.
The old recovery map was a bitmap of node numbers. This was sufficient
for the maximum node number of 254. Going forward, we want node numbers
to be UINT32. Thus, we need a new recovery map.
Note that we can't keep track of slots here. We must write down the
node number to recovery *before* we get the locks needed to convert a
node number into a slot number.
The recovery map is now an array of unsigned ints, max_slots in size.
It moves to journal.c with the rest of recovery.
Because it needs to be initialized, we move all of recovery initialization
into a new function, ocfs2_recovery_init(). This actually cleans up
ocfs2_initialize_super() a little as well. Following on, recovery cleaup
becomes part of ocfs2_recovery_exit().
A number of node map functions are rendered obsolete and are removed.
Finally, waiting on recovery is wrapped in a function rather than naked
checks on the recovery_event. This is a cleanup from Mark.
Signed-off-by: Joel Becker <joel.becker@oracle.com> Signed-off-by: Mark Fasheh <mfasheh@suse.com>
Joel Becker [Fri, 1 Feb 2008 20:01:05 +0000 (12:01 -0800)]
ocfs2: Make ocfs2_slot_info private.
Just use osb_lock around the ocfs2_slot_info data. This allows us to
take the ocfs2_slot_info structure private in slot_info.c. All access
is now via accessors.
Signed-off-by: Joel Becker <joel.becker@oracle.com> Signed-off-by: Mark Fasheh <mfasheh@suse.com>
Mark Fasheh [Fri, 1 Feb 2008 19:59:09 +0000 (11:59 -0800)]
ocfs2: Move slot map access into slot_map.c
journal.c and dlmglue.c would refresh the slot map by hand. Instead, have
the update and clear functions do the work inside slot_map.c. The eventual
result is to make ocfs2_slot_info defined privately in slot_map.c
Signed-off-by: Joel Becker <joel.becker@oracle.com> Signed-off-by: Mark Fasheh <mfasheh@suse.com>
Merge branch 'for-linus' of git://oss.sgi.com:8090/xfs/xfs-2.6
* 'for-linus' of git://oss.sgi.com:8090/xfs/xfs-2.6: (87 commits)
[XFS] Fix merge failure
[XFS] The forward declarations for the xfs_ioctl() helpers and the
[XFS] Update XFS documentation for noikeep/ikeep.
[XFS] Update XFS Documentation for ikeep and ihashsize
[XFS] Remove unused HAVE_SPLICE macro.
[XFS] Remove CONFIG_XFS_SECURITY.
[XFS] xfs_bmap_compute_maxlevels should be based on di_forkoff
[XFS] Always use di_forkoff when checking for attr space.
[XFS] Ensure the inode is joined in xfs_itruncate_finish
[XFS] Remove periodic logging of in-core superblock counters.
[XFS] fix logic error in xfs_alloc_ag_vextent_near()
[XFS] Don't error out on good I/Os.
[XFS] Catch log unmount failures.
[XFS] Sanitise xfs_log_force error checking.
[XFS] Check for errors when changing buffer pointers.
[XFS] Don't allow silent errors in xfs_inactive().
[XFS] Catch errors from xfs_imap().
[XFS] xfs_bulkstat_one_dinode() never returns an error.
[XFS] xfs_iflush_fork() never returns an error.
[XFS] Catch unwritten extent conversion errors.
...
Merge branch 'next' of git://git.kernel.org/pub/scm/linux/kernel/git/djbw/async_tx
* 'next' of git://git.kernel.org/pub/scm/linux/kernel/git/djbw/async_tx:
dmaengine: ack to flags: make use of the unused bits in the 'ack' field
iop-adma: remove the workaround for missed interrupts on iop3xx
async_tx: kill ->device_dependency_added
async_tx: fix multiple dependency submission
fsldma: Split the MPC83xx event from MPC85xx and refine irq codes.
fsldma: Remove CONFIG_FSL_DMA_SELFTEST, keep fsl_dma_self_test() running always.
Merge branch 'upstream-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/jgarzik/libata-dev
* 'upstream-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/jgarzik/libata-dev: (79 commits)
ata-acpi: don't call _GTF for disabled drive
sata_mv add temporary 3 second init delay for SiliconImage PMs
sata_mv remove redundant edma init code
sata_mv add basic port multiplier support
sata_mv fix SOC flags, enable NCQ on SOC
sata_mv disable hotplug for now
sata_mv cosmetics
sata_mv hardreset rework
[libata] improve Kconfig help text for new PMP, SFF options
libata: make EH fail gracefully if no reset method is available
libata: Be a bit more slack about early devices
libata: cable logic
libata: move link onlineness check out of softreset methods
libata: kill dead code paths in reset path
pata_scc: fix build breakage
libata: make PMP support optional
libata: implement PMP helpers
libata: separate PMP support code from core code
libata: make SFF support optional
libata: don't use ap->ioaddr in non-SFF drivers
...
* git://git.kernel.org/pub/scm/linux/kernel/git/tglx/linux-2.6-hrt:
clocksource: make clocksource watchdog cycle through online CPUs
Documentation: move timer related documentation to a single place
clockevents: optimise tick_nohz_stop_sched_tick() a bit
locking: remove unused double_spin_lock()
hrtimers: simplify lockdep handling
timers: simplify lockdep handling
posix-timers: fix shadowed variables
timer_list: add annotations to workqueue.c
hrtimer: use nanosleep specific restart_block fields
hrtimer: add nanosleep specific restart_block member
projects / linux-2.6 / log