Impact: refactor

Make a distinct frontend xenbus, in preparation for adding a backend xenbus.
Signed-off-by: NIan Campbell <ian.campbell@citrix.com>
Signed-off-by: NJeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
[corresponds to 2fd433a4188f in git://git.kernel.org/pub/scm/linux/kernel/git/jeremy/xen.git
 with adjustments to reflect changes in the code which is moved]
Signed-off-by: NKonrad Rzeszutek Wilk <konrad.wilk@oracle.com>

The new interface creates directories per mapped image
and under each it creates a subdir per available snapshot.
This allows keeping a cleaner interface within the sysfs
guidelines. The ABI documentation was updated too.
Acked-by: NGreg Kroah-Hartman <gregkh@suse.de>
Signed-off-by: NYehuda Sadeh <yehuda@hq.newdream.net>
Signed-off-by: NSage Weil <sage@newdream.net>

The big kernel lock has been removed from all these files at some point,
leaving only the #include.

Remove this too as a cleanup.
Signed-off-by: NArnd Bergmann <arnd@arndb.de>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

The recent patch to fix the removal of a non-existing proc
directory introduced this build problem for !CONFIG_PROC_FS:

drivers/block/cciss.c:4929: error: 'proc_cciss' undeclared (first use in this function)

Fix it by moving proc_cciss outside of the CONFIG_PROC_FS scope.
Reported-by: NRandy Dunlap <randy.dunlap@oracle.com>
Signed-off-by: NJens Axboe <jaxboe@fusionio.com>

Move the mid-layer's ->queuecommand() invocation from being locked
with the host lock to being unlocked to facilitate speeding up the
critical path for drivers who don't need this lock taken anyway.

The patch below presents a simple SCSI host lock push-down as an
equivalent transformation.  No locking or other behavior should change
with this patch.  All existing bugs and locking orders are preserved.

Additionally, add one parameter to queuecommand,
	struct Scsi_Host *
and remove one parameter from queuecommand,
	void (*done)(struct scsi_cmnd *)

Scsi_Host* is a convenient pointer that most host drivers need anyway,
and 'done' is redundant to struct scsi_cmnd->scsi_done.

Minimal code disturbance was attempted with this change.  Most drivers
needed only two one-line modifications for their host lock push-down.
Signed-off-by: NJeff Garzik <jgarzik@redhat.com>
Acked-by: NJames Bottomley <James.Bottomley@suse.de>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Geert, my crosstool don't produce warning below. I guess this has to do
something with compiler version.

- Geert noticed following warning during compilation.

  drivers/block/amiflop.c:1344: warning: ‘rq’ may be used uninitialized in
  this function
  drivers/block/ataflop.c:1402: warning: ‘rq’ may be used uninitialized in
  this function

- Initialize rq to NULL to fix the warning. If we can't find a suitable request
  to dispatch, this function should return NULL instead of a possibly garbage
  pointer.

- Cross compile tested only. Don't have hardware to test it.
Reported-by: NGeert Uytterhoeven <geert@linux-m68k.org>
Signed-off-by: NVivek Goyal <vgoyal@redhat.com>
Signed-off-by: NJens Axboe <jaxboe@fusionio.com>

REQ_HARDBARRIER is dead now, so remove the leftovers.  What's left
at this point is:

 - various checks inside the block layer.
 - sanity checks in bio based drivers.
 - now unused bio_empty_barrier helper.
 - Xen blockfront use of BLKIF_OP_WRITE_BARRIER - it's dead for a while,
   but Xen really needs to sort out it's barrier situaton.
 - setting of ordered tags in uas - dead code copied from old scsi
   drivers.
 - scsi different retry for barriers - it's dead and should have been
   removed when flushes were converted to FS requests.
 - blktrace handling of barriers - removed.  Someone who knows blktrace
   better should add support for REQ_FLUSH and REQ_FUA, though.
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Signed-off-by: NJens Axboe <jaxboe@fusionio.com>

Convert direct reads of an inode's i_size to using i_size_read().

i_size_{read,write} use a seqcount to protect reads from accessing
incomple writes.  Concurrent i_size_write()s require mutual exclussion
to protect the seqcount that is used by i_size_{read,write}.  But
i_size_read() callers do not need to use additional locking.
Signed-off-by: NMike Snitzer <snitzer@redhat.com>
Acked-by: NNeilBrown <neilb@suse.de>
Acked-by: NLars Ellenberg <lars.ellenberg@linbit.com>
Signed-off-by: NJens Axboe <jaxboe@fusionio.com>

Randy reports that he gets the following stack trace when
removing the cciss module:

[  109.164277] Pid: 3463, comm: rmmod Not tainted 2.6.37-rc1 #7
[  109.164280] Call Trace:
[  109.164292]  [<ffffffff8107eb8d>] warn_slowpath_common+0xc6/0xf3
[  109.164299]  [<ffffffff8107ecaa>] warn_slowpath_fmt+0x5b/0x6b
[  109.164307]  [<ffffffff8155175b>] ? _raw_spin_unlock+0x40/0x4b
[  109.164313]  [<ffffffff8123dd1e>] remove_proc_entry+0x156/0x35e
[  109.164320]  [<ffffffff812cd91b>] ? do_raw_spin_unlock+0xff/0x10f
[  109.164327]  [<ffffffff8113823d>] ? trace_hardirqs_on+0x10/0x4a
[  109.164333]  [<ffffffff8155162d>] ? _raw_spin_unlock_irq+0x4c/0x7b
[  109.164339]  [<ffffffff8154d4d1>] ? wait_for_common+0x145/0x15e
[  109.164345]  [<ffffffff81075337>] ? default_wake_function+0x0/0x22
[  109.164357]  [<ffffffffa0615a8f>] cciss_cleanup+0xa9/0xc7 [cciss]
[  109.164365]  [<ffffffff810d3cb0>] sys_delete_module+0x2d6/0x368
[  109.164371]  [<ffffffff8155036b>] ? lockdep_sys_exit_thunk+0x35/0x67
[  109.164377]  [<ffffffff810fdfaf>] ? audit_syscall_entry+0x172/0x1a5
[  109.164383]  [<ffffffff815502f5>] ? trace_hardirqs_on_thunk+0x3a/0x3f
[  109.164389]  [<ffffffff8100ea72>] system_call_fastpath+0x16/0x1b
[  109.164394] ---[ end trace 88e8568246ed0b1d ]---

which will happen if you don't actually have an HP CISS adapter,
since it'll do an uncondional removal of a proc directory it
never attempted to create in that case.
Reported-by: NRandy Dunlap <randy.dunlap@oracle.com>
Tested-by: NRandy Dunlap <randy.dunlap@oracle.com>
Signed-off-by: NJens Axboe <jaxboe@fusionio.com>

While scanning the floopy code due to c093ee4f ("floppy: fix
use-after-free in module load failure path"), I found one more instance
of trying to access disk->queue pointer after doing put_disk() on
gendisk.  For some reason , floppy moule still loads/unloads fine.  The
object is probably still around with right pointer values.

 o There seems to be one more instance of trying to cleanup the request
   queue after we have called put_disk() on associated gendisk.

 o This fix is more out of code inspection.  Even without this fix for
   some reason I am able to load/unload floppy module without any
   issues.

 o Floppy module loads/unloads fine after the fix.
Signed-off-by: NVivek Goyal <vgoyal@redhat.com>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Commit 48821184 ("floppy: switch to one queue per drive instead of
sharing a queue") introduced a use-after-free.  We do "put_disk()" on
the disk device _before_ we then clean up the queue associated with that
disk.

Move the put_disk() down to avoid dereferencing a free'd data structure.

Cc: Jens Axboe <jaxboe@fusionio.com>
Cc: Vivek Goyal <vgoyal@redhat.com>
Reported-and-tested-by: NRandy Dunlap <randy.dunlap@oracle.com>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Some(?) Xen block backends fail BLKIF_OP_WRITE_BARRIER requests, which
Linux uses as a cache flush operation.  In that case, disable use
of FLUSH.
Signed-off-by: NJeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
Cc: Daniel Stodden <daniel.stodden@citrix.com>

The BLKIF_OP_WRITE_BARRIER is a full ordered barrier, so we can use it
to implement FUA as well as a plain FLUSH.
Signed-off-by: NJeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
Acked-by: NChristoph Hellwig <hch@lst.de>

Signed-off-by: NJeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>

Implement a flush as a full barrier, since we have nothing weaker.
Signed-off-by: NJeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
Acked-by: NChristoph Hellwig <hch@lst.de>

This patch removes the controller overlap between cciss and hpsa. It was
decided that no overlap should exist. All new controllers will use the hpsa
SCSI based driver.
Signed-off-by: NJens Axboe <jaxboe@fusionio.com>

Structure IOCTL_Command_struct is copied to userland with
some padding fields at the end of the struct unitialized.
It leads to leaking of contents of kernel stack memory.
Signed-off-by: NVasiliy Kulikov <segooon@gmail.com>
Signed-off-by: NJens Axboe <jaxboe@fusionio.com>

As described in https://bugzilla.kernel.org/show_bug.cgi?id=19922

: I had an AoE device go down overnight, and while a server was trying to
: write to it, it was also writing this message to its logs:
:
: 209                 printk(KERN_INFO "aoe: device %ld.%d is not up\n",
: 210                         d->aoemajor, d->aoeminor);
:
: The message appeared many times per second, and over several hours
: produced about 7.5 gigabytes of log files, filling up all free space on
: the root filesystem.

Cc: "Ed L. Cashin" <ecashin@coraid.com>
Suggested-by: NRoman Mamedov <roman@rm.pp.ru>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NJens Axboe <jaxboe@fusionio.com>

If CONFIG_LBDAF=y, `sector_t' becomes `u64' instead of `unsigned long':

drivers/block/z2ram.c: In function ¡do_z2_request¢:
drivers/block/z2ram.c:83: warning: format %lu expects type `long unsigned int', but argument 2 has type `sector_t'

Hence always cast it to `unsigned long long' for printing.  Also do the
pr_err() dance, while we're at it.
Signed-off-by: NGeert Uytterhoeven <geert@linux-m68k.org>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NJens Axboe <jaxboe@fusionio.com>

flush_scheduled_work() is deprecated and scheduled to be removed.
Directly cancel aoedev->work on free instead of depending on
flush_scheduled_works().
Signed-off-by: NTejun Heo <tj@kernel.org>
Cc: "Ed L. Cashin" <ecashin@coraid.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NJens Axboe <jaxboe@fusionio.com>

Failure to create drbd_ee_mempool appears not to get checked.  Looks like
a copy-and-paste problem to me.
Signed-off-by: NNicolas Kaiser <nikai@nikai.net>
Cc: Lars Ellenberg <drbd-dev@lists.linbit.com>
Cc: Philipp Reisner <philipp.reisner@linbit.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NJens Axboe <jaxboe@fusionio.com>

In autoclear mode bdev is NULL but the sysfs
entry should be destroyed otherwise this warning appears:

WARNING: at fs/sysfs/dir.c:451 sysfs_add_one+0x82/0x95()
sysfs: cannot create duplicate filename '/devices/virtual/block/loop0/loop'

Fixes commit ee862730Signed-off-by: NMilan Broz <mbroz@redhat.com>
Signed-off-by: NJens Axboe <jaxboe@fusionio.com>

Ensure kmap_atomic() usage is strictly nested
Signed-off-by: NPeter Zijlstra <a.p.zijlstra@chello.nl>
Reviewed-by: NRik van Riel <riel@redhat.com>
Acked-by: NChris Metcalf <cmetcalf@tilera.com>
Cc: David Howells <dhowells@redhat.com>
Cc: Hugh Dickins <hughd@google.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Russell King <rmk@arm.linux.org.uk>
Cc: Ralf Baechle <ralf@linux-mips.org>
Cc: David Miller <davem@davemloft.net>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

We would prefer not to have any overlap between the two drivers.
Remove the cciss_allow_hpsa option, as it it is no longer needed.
Signed-off-by: NStephen M. Cameron <scameron@beardog.cce.hp.com>
Signed-off-by: NJens Axboe <jaxboe@fusionio.com>

Signed-off-by: NStephen M. Cameron <scameron@beardog.cce.hp.com>
Signed-off-by: NJens Axboe <jaxboe@fusionio.com>

This is to conserve memory in a memory-limited kdump scenario
Signed-off-by: NStephen M. Cameron <scameron@beardog.cce.hp.com>
Signed-off-by: NJens Axboe <jaxboe@fusionio.com>

and use the doorbell reset method if available (which doesn't
lock up the controller if you properly save and restore all
the PCI registers that you're supposed to.)
Signed-off-by: NStephen M. Cameron <scameron@beardog.cce.hp.com>
Signed-off-by: NJens Axboe <jaxboe@fusionio.com>

After a reset, we should first wait for the board to become "not ready",
and then wait for it to become "ready", instead of immediately
waiting for it to become "ready", and do this waiting *after*
restoring PCI config space registers.
Signed-off-by: NStephen M. Cameron <scameron@beardog.cce.hp.com>
Signed-off-by: NJens Axboe <jaxboe@fusionio.com>

Signed-off-by: NPhilipp Reisner <philipp.reisner@linbit.com>
Signed-off-by: NLars Ellenberg <lars.ellenberg@linbit.com>

Signed-off-by: NPhilipp Reisner <philipp.reisner@linbit.com>
Signed-off-by: NLars Ellenberg <lars.ellenberg@linbit.com>

Signed-off-by: NPhilipp Reisner <philipp.reisner@linbit.com>
Signed-off-by: NLars Ellenberg <lars.ellenberg@linbit.com>

This commit changes prefix for some of the USB mass storage
class related macros (ie. USB_SC_ for subclass and USB_PR_
for class).
Signed-off-by: NMichal Nazarewicz <mina86@mina86.com>
Cc: Alan Stern <stern@rowland.harvard.edu>
Cc: Matthew Wilcox <willy@linux.intel.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@suse.de>

That assertion's condition needed adjustment for today's semantics
Signed-off-by: NPhilipp Reisner <philipp.reisner@linbit.com>
Signed-off-by: NLars Ellenberg <lars.ellenberg@linbit.com>

If we don't rate limit it, and you happen to log err level messages via
serial console, an IO error on a disconnected Primary may cause serious
unresponsiveness.
Signed-off-by: NPhilipp Reisner <philipp.reisner@linbit.com>
Signed-off-by: NLars Ellenberg <lars.ellenberg@linbit.com>

This codepath used to be called only for failed kmalloc GFP_ATOMIC,
but is now also triggered by other things.
Signed-off-by: NPhilipp Reisner <philipp.reisner@linbit.com>
Signed-off-by: NLars Ellenberg <lars.ellenberg@linbit.com>

If we get an IO-error during an activity log transaction,
if we failed to write the bitmap of the evicted extent,
we must not write the transaction itself.
If we failed to write the transaction,
we must not even submit the corresponding bio,
as its extent is not yet marked in the activity log.

Otherwise, if this was a disconneted Primary (degraded cluster), which
now lost its disk as well, and we later re-attach the same backend
storage, we possibly "forget" to resync some parts of the disk that
potentially have been changed.

On the receiving side, when receiving from a peer with unhealthy disk,
checking for pdsk == D_DISKLESS is not enough, we need to set out of
sync and do AL transactions for everything pdsk < D_INCONSISTENT on the
receiving side.
Signed-off-by: NPhilipp Reisner <philipp.reisner@linbit.com>
Signed-off-by: NLars Ellenberg <lars.ellenberg@linbit.com>

If we have contention in drbd_al_begin_iod (heavy randon IO),
an administrative request to detach the disk may deadlock
for similar reasons as the recently fixed deadlock if detaching
because of IO-error.

The approach taken here is to either go through the intermediate
cleanup state D_FAILED, or first lock out application io,
don't just go directly to D_DISKLESS.

We need an additional state bit (WAS_IO_ERROR) to distinguish
the -> D_FAILED because of IO-error from other failures.

Sanitize D_ATTACHING -> D_FAILED to D_ATTACHING -> D_DISKLESS.
If only attaching, ldev may be missing still, but would be referenced
from within the after_state_ch for -> D_FAILED, potentially
dereferencing a NULL pointer.
Signed-off-by: NPhilipp Reisner <philipp.reisner@linbit.com>
Signed-off-by: NLars Ellenberg <lars.ellenberg@linbit.com>

If those messages ever get logged, clearly state that they are
actually failed ASSERTS, so our regression tests can pick them up
from the logs more easily.
Signed-off-by: NPhilipp Reisner <philipp.reisner@linbit.com>
Signed-off-by: NLars Ellenberg <lars.ellenberg@linbit.com>

Every code path changing the current UUID needs to get it on stable
storage anyways. Flush it to disk right there, remove the now obsolte
explicit drbd_md_sync statements in the other code paths.
Signed-off-by: NPhilipp Reisner <philipp.reisner@linbit.com>
Signed-off-by: NLars Ellenberg <lars.ellenberg@linbit.com>

The driver doesn't handle empty flushes. Disable barrier/flush
write support until this is fixed up.
Signed-off-by: NJens Axboe <jaxboe@fusionio.com>