Kent writes:

Hey Jens - I've been busy torture testing and chasing bugs, here's the
fruits of my labors. These are all fairly small fixes, some of them
quite important.

The alloc kthread should've been using try_to_freeze() - and also there
was the potential for the alloc kthread to get woken up after it had
shut down, which would have been bad.
Signed-off-by: Kent Overstreet <kmo@daterainc.com>

Part of the job of garbage collection is to add up however many sectors
of live data it finds in each bucket, but that doesn't work very well if
it doesn't reset GC_SECTORS_USED() when it starts. Whoops.

This wouldn't have broken anything horribly, but allocation tries to
preferentially reclaim buckets that are mostly empty and that's not
gonna work with an incorrect GC_SECTORS_USED() value.
Signed-off-by: Kent Overstreet <kmo@daterainc.com>
Cc: linux-stable <stable@vger.kernel.org> # >= v3.10

The journal replay code starts by finding something that looks like a
valid journal entry, then it does a binary search over the unchecked
region of the journal for the journal entries with the highest sequence
numbers.

Trouble is, the logic was wrong - journal_read_bucket() returns true if
it found journal entries we need, but if the range of journal entries
we're looking for loops around the end of the journal - in that case
journal_read_bucket() could return true when it hadn't found the highest
sequence number we'd seen yet, and in that case the binary search did
the wrong thing. Whoops.
Signed-off-by: Kent Overstreet <kmo@daterainc.com>
Cc: linux-stable <stable@vger.kernel.org> # >= v3.10

Stopping a cache set is supposed to make it stop attached backing
devices, but somewhere along the way that code got lost. Fixing this
mainly has the effect of fixing our reboot notifier.
Signed-off-by: Kent Overstreet <kmo@daterainc.com>
Cc: linux-stable <stable@vger.kernel.org> # >= v3.10

If we stopped a bcache device when we were already detaching (or
something like that), bcache_device_unlink() would try to remove a
symlink from sysfs that was already gone because the bcache dev kobject
had already been removed from sysfs.

So keep track of whether we've removed stuff from sysfs.
Signed-off-by: Kent Overstreet <kmo@daterainc.com>
Cc: linux-stable <stable@vger.kernel.org> # >= v3.10

Whoops - bcache's flush/FUA was mostly correct, but flushes get filtered
out unless we say we support them...
Signed-off-by: Kent Overstreet <kmo@daterainc.com>
Cc: linux-stable <stable@vger.kernel.org> # >= v3.10

There is a missing NULL check after the kzalloc().
Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>

In the far-too-complicated closure code - closures can have destructors,
for probably dubious reasons; they get run after the closure is no
longer waiting on anything but before dropping the parent ref, intended
just for freeing whatever memory the closure is embedded in.

Trouble is, when remaining goes to 0 and we've got nothing more to run -
we also have to unlock the closure, setting remaining to -1. If there's
a destructor, that unlock isn't doing anything - nobody could be trying
to lock it if we're about to free it - but if the unlock _is needed...
that check for a destructor was racy. Argh.
Signed-off-by: Kent Overstreet <kmo@daterainc.com>
Cc: linux-stable <stable@vger.kernel.org> # >= v3.10

Linux 3.10-rc7

Pull this in early to avoid doing it with the bcache merge,
since there are a number of changes to bcache between my old
base (3.10-rc1) and the new pull request.

Some of bcache's utility code has made it into the rest of the kernel,
so drop the bcache versions.

Bcache used to have a workaround for allocating from a bio set under
generic_make_request() (if you allocated more than once, the bios you
already allocated would get stuck on current->bio_list when you
submitted, and you'd risk deadlock) - bcache would mask out __GFP_WAIT
when allocating bios under generic_make_request() so that allocation
could fail and it could retry from workqueue. But bio_alloc_bioset() has
a workaround now, so we can drop this hack and the associated error
handling.
Signed-off-by: Kent Overstreet <koverstreet@google.com>

Signed-off-by: Kent Overstreet <kmo@daterainc.com>

This code has rotted and it hasn't been used in ages anyways.
Signed-off-by: Kent Overstreet <kmo@daterainc.com>

Signed-off-by: Kent Overstreet <kmo@daterainc.com>

Journal writes need to be marked FUA, not just REQ_FLUSH. And btree node
writes have... weird ordering requirements.
Signed-off-by: Kent Overstreet <koverstreet@google.com>

Allow to change the AL layout with an resize operation. For that
the reisze command gets two new fields: al_stripes and al_stripe_size.

In order to make the operation crash save:
1) Lock out all IO and MD-IO
2) Write the super block with MDF_PRIMARY_IND clear
3) write the bitmap to the new location (all zeros, since
   we allow only while connected)
4) Initialize the new AL-area
5) Write the super block with the restored MDF_PRIMARY_IND.
6) Unfreeze all IO

Since the AL-layout has no influence on the protocol, this operation
needs to be beforemed on both sides of a resource (if intended).
Signed-off-by: Andreas Gruenbacher <agruen@linbit.com>
Signed-off-by: Philipp Reisner <philipp.reisner@linbit.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>

Signed-off-by: Andreas Gruenbacher <agruen@linbit.com>
Signed-off-by: Philipp Reisner <philipp.reisner@linbit.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>

In case the connection was established and lost again before
the a fence-peer handler returns, ignore the exit code of this
instance. (And use the exit code of the later started instance)
Signed-off-by: Andreas Gruenbacher <agruen@linbit.com>
Signed-off-by: Philipp Reisner <philipp.reisner@linbit.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>

Signed-off-by: Andreas Gruenbacher <agruen@linbit.com>
Signed-off-by: Philipp Reisner <philipp.reisner@linbit.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>

Fix to return a negative error code from the error handling
case instead of 0, as returned elsewhere in this function.
Signed-off-by: Wei Yongjun <yongjun_wei@trendmicro.com.cn>
Signed-off-by: Lars Ellenberg <lars.ellenberg@linbit.com>
Signed-off-by: Andreas Gruenbacher <agruen@linbit.com>
Signed-off-by: Philipp Reisner <philipp.reisner@linbit.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>

Signed-off-by: Andreas Gruenbacher <agruen@linbit.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>

Merge branch 'stable/for-jens-3.10' of git://git.kernel.org/pub/scm/linux/kernel/git/konrad/xen into for-3.11/drivers

Konrad writes:

It has the 'feature-max-indirect-segments' implemented in both backend
and frontend. The current problem with the backend and frontend is that the
segment size is limited to 11 pages. It means we can at most squeeze in 44kB per
request. The ring can hold 32 (next power of two below 36) requests, meaning we
can do 1.4M of outstanding requests. Nowadays that is not enough.

The problem in the past was addressed in two ways - but neither one went upstream.
The first solution to this proposed by Justin from Spectralogic was to negotiate
the segment size.  This means that the ‘struct blkif_sring_entry’ is now a variable size.
It can expand from 112 bytes (cover 11 pages of data - 44kB) to 1580 bytes
(256 pages of data - so 1MB). It is a simple extension by just making the array in the
request expand from 11 to a variable size negotiated. But it had limits: this extension
still limits the number of segments per request to 255 (as the total number must be
specified in the request, which only has an 8-bit field for that purpose).

The other solution (from Intel - Ronghui) was to create one extra ring that only has the
‘struct blkif_request_segment’ in them. The ‘struct blkif_request’ would be changed to have
an index in said ‘segment ring’. There is only one segment ring. This means that the size of
the initial ring is still the same. The requests would point to the segment and enumerate out
how many of the indexes it wants to use. The limit is of course the size of the segment.
If one assumes a one-page segment this means we can in one request cover ~4MB.

Those patches were posted as RFC and the author never followed up on the ideas on changing
it to be a bit more flexible.

There is yet another mechanism that could be employed  (which these patches implement) - and it
borrows from VirtIO protocol. And that is the ‘indirect descriptors’. This very similar to
what Intel suggests, but with a twist. The twist is to negotiate how many of these
'segment' pages (aka indirect descriptor pages) we want to support (in reality we negotiate
how many entries in the segment we want to cover, and we module the number if it is
bigger than the segment size).

This means that with the existing 36 slots in the ring (single page) we can cover:
32 slots * each blkif_request_indirect covers: 512 * 4096 ~= 64M. Since we ample space
in the blkif_request_indirect to span more than one indirect page, that number (64M)
can be also multiplied by eight = 512MB.

Roger Pau Monne took the idea and implemented them in these patches. They work
great and the corner cases (migration between backends with and without this extension)
work nicely. The backend has a limit right now off how many indirect entries
it can handle: one indirect page, and at maximum 256 entries (out of 512 - so  50% of the page
is used). That comes out to 32 slots * 256 entries in a indirect page * 1 indirect page
per request * 4096 = 32MB.

This is a conservative number that can change in the future. Right now it strikes
a good balance between giving excellent performance, memory usage in the backend, and
balancing the needs of many guests.

In the patchset there is also the split of the blkback structure to be per-VBD.
This means that the spinlock contention we had with many guests trying to do I/O and
all the blkback threads hitting the same lock has been eliminated.

Also there are bug-fixes to deal with oddly sized sectors, insane amounts on
th ring, and also a security fix (posted earlier).

Mention udev autoregistration, symlinks.  Write down some sysfs paths.
Signed-off-by: Gabriel de Perthuis <g2p.code@gmail.com>
Signed-off-by: Kent Overstreet <koverstreet@google.com>

Signed-off-by: Gabriel de Perthuis <g2p.code@gmail.com>
Signed-off-by: Kent Overstreet <koverstreet@google.com>

Signed-off-by: Gabriel de Perthuis <g2p.code@gmail.com>

Correct spelling typo in documentation/bcache.txt
Signed-off-by: Masanari Iida <standby24x7@gmail.com>
Signed-off-by: Kent Overstreet <koverstreet@google.com>

Now that we're tracking dirty data per stripe, we can add two
optimizations for raid5/6:

 * If a stripe is already dirty, force writes to that stripe to
   writeback mode - to help build up full stripes of dirty data

 * When flushing dirty data, preferentially write out full stripes first
   if there are any.
Signed-off-by: Kent Overstreet <koverstreet@google.com>

To make background writeback aware of raid5/6 stripes, we first need to
track the amount of dirty data within each stripe - we do this by
breaking up the existing sectors_dirty into per stripe atomic_ts
Signed-off-by: Kent Overstreet <koverstreet@google.com>

Previously, dirty_data wouldn't get initialized until the first garbage
collection... which was a bit of a problem for background writeback (as
the PD controller keys off of it) and also confusing for users.

This is also prep work for making background writeback aware of raid5/6
stripes.
Signed-off-by: Kent Overstreet <koverstreet@google.com>

The old lazy sorting code was kind of hacky - rewrite in a way that
mathematically makes more sense; the idea is that the size of the sets
of keys in a btree node should increase by a more or less fixed ratio
from smallest to biggest.
Signed-off-by: Kent Overstreet <koverstreet@google.com>

Old gcc doesnt like the struct hack, and it is kind of ugly. So finish
off the work to convert pr_debug() statements to tracepoints, and delete
pkey()/pbtree().
Signed-off-by: Kent Overstreet <koverstreet@google.com>

The tracepoints were reworked to be more sensible, and fixed a null
pointer deref in one of the tracepoints.

Converted some of the pr_debug()s to tracepoints - this is partly a
performance optimization; it used to be that with DEBUG or
CONFIG_DYNAMIC_DEBUG pr_debug() was an empty macro; but at some point it
was changed to an empty inline function.

Some of the pr_debug() statements had rather expensive function calls as
part of the arguments, so this code was getting run unnecessarily even
on non debug kernels - in some fast paths, too.
Signed-off-by: Kent Overstreet <koverstreet@google.com>

The most significant change is that btree reads are now done
synchronously, instead of asynchronously and doing the post read stuff
from a workqueue.

This was originally done because we can't block on IO under
generic_make_request(). But - we already have a mechanism to punt cache
lookups to workqueue if needed, so if we just use that we don't have to
deal with the complexity of doing things asynchronously.

The main benefit is this makes the locking situation saner; we can hold
our write lock on the btree node until we're finished reading it, and we
don't need that btree_node_read_done() flag anymore.

Also, for writes, btree_write() was broken out into btree_node_write()
and btree_leaf_dirty() - the old code with the boolean argument was dumb
and confusing.

The prio_blocked mechanism was improved a bit too, now the only counter
is in struct btree_write, we don't mess with transfering a count from
struct btree anymore.

This required changing garbage collection to block prios at the start
and unblock when it finishes, which is cleaner than what it was doing
anyways (the old code had mostly the same effect, but was doing it in a
convoluted way)

And the btree iter btree_node_read_done() uses was converted to a real
mempool.
Signed-off-by: Kent Overstreet <koverstreet@google.com>

Using a workqueue when we just want a single thread is a bit silly.
Signed-off-by: Kent Overstreet <koverstreet@google.com>

Signed-off-by: Gabriel de Perthuis <g2p.code+bcache@gmail.com>
Signed-off-by: Kent Overstreet <koverstreet@google.com>

An old version of gcc was complaining about using a const int as the
size of a stack allocated array. Which should be fine - but using
ARRAY_SIZE() is better, anyways.

Also, refactor the code to use scnprintf().
Signed-off-by: Kent Overstreet <koverstreet@google.com>

bio_alloc_bioset returns NULL on failure. This fix adds a missing check
for potential NULL pointer dereferencing.
Signed-off-by: Kumar Amit Mehta <gmate.amit@gmail.com>
Signed-off-by: Kent Overstreet <koverstreet@google.com>

With the introduction of indirect segments we can receive requests
with a number of segments bigger than the maximum number of allowed
iovecs in a bios, so make sure that blkback doesn't try to allocate a
bios with more iovecs than BIO_MAX_PAGES
Signed-off-by: Roger Pau Monné <roger.pau@citrix.com>
Cc: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
Signed-off-by: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>