If a message queued for send gets revoked, zeroes are sent over the
wire instead of any unsent data.  This is done by constructing a
message and passing it to kernel_sendmsg() via ceph_tcp_sendmsg().

Since we are already working with a page in this case we can use
the sendpage interface instead.  Create a new ceph_tcp_sendpage()
helper that sets up flags to match the way ceph_tcp_sendmsg()
does now.
Signed-off-by: Alex Elder <elder@dreamhost.com>
Reviewed-by: Sage Weil <sage@newdream.net>

CRC's are computed for all messages between ceph entities.  The CRC
computation for the data portion of message can optionally be
disabled using the "nocrc" (common) ceph option.  The default is
for CRC computation for the data portion to be enabled.

Unfortunately, the code that implements this feature interprets the
feature flag wrong, meaning that by default the CRC's have *not*
been computed (or checked) for the data portion of messages unless
the "nocrc" option was supplied.

Fix this, in write_partial_msg_pages() and read_partial_message().
Also change the flag variable in write_partial_msg_pages() to be
"no_datacrc" to match the usage elsewhere in the file.

This fixes http://tracker.newdream.net/issues/2064Signed-off-by: Alex Elder <elder@dreamhost.com>
Reviewed-by: Sage Weil <sage@newdream.net>

Nothing too big here.
    - define the size of the buffer used for consuming ignored
      incoming data using a symbolic constant
    - simplify the condition determining whether to unmap the page
      in write_partial_msg_pages(): do it for crc but not if the
      page is the zero page
Signed-off-by: Alex Elder <elder@dreamhost.com>
Signed-off-by: Sage Weil <sage@newdream.net>

Make a small change in the code that counts down kvecs consumed by
a ceph_tcp_sendmsg() call.  Same functionality, just blocked out
a little differently.
Signed-off-by: Alex Elder <elder@dreamhost.com>
Signed-off-by: Sage Weil <sage@newdream.net>

Move blocks of code out of loops in read_partial_message_section()
and read_partial_message().  They were only was getting called at
the end of the last iteration of the loop anyway.
Signed-off-by: Alex Elder <elder@dreamhost.com>
Signed-off-by: Sage Weil <sage@newdream.net>

Calculate CRC in a separate step from rearranging the byte order
of the result, to improve clarity and readability.

Use offsetof() to determine the number of bytes to include in the
CRC calculation.

In read_partial_message(), switch which value gets byte-swapped,
since the just-computed CRC is already likely to be in a register.
Signed-off-by: Alex Elder <elder@dreamhost.com>
Signed-off-by: Sage Weil <sage@newdream.net>

Change the name (and type) of a few CRC-related Boolean local
variables so they contain the word "do", to distingish their purpose
from variables used for holding an actual CRC value.

Note that in the process of doing this I identified a fairly serious
logic error in write_partial_msg_pages():  the value of "do_crc"
assigned appears to be the opposite of what it should be.  No
attempt to fix this is made here; this change preserves the
erroneous behavior.  The problem I found is documented here:
    http://tracker.newdream.net/issues/2064Signed-off-by: Alex Elder <elder@dreamhost.com>
Signed-off-by: Sage Weil <sage@newdream.net>

Many ceph-related Boolean options offer the ability to both enable
and disable a feature.  For all those that don't offer this, add
a new option so that they do.

Note that ceph_show_options()--which reports mount options currently
in effect--only reports the option if it is different from the
default value.
Signed-off-by: Alex Elder <elder@dreamhost.com>
Signed-off-by: Sage Weil <sage@newdream.net>

This gathers a number of very minor changes:
    - use %hu when formatting the a socket address's address family
    - null out the ceph_msgr_wq pointer after the queue has been
      destroyed
    - drop a needless cast in ceph_write_space()
    - add a WARN() call in ceph_state_change() in the event an
      unrecognized socket state is encountered
    - rearrange the logic in ceph_con_get() and ceph_con_put() so
      that:
        - the reference counts are only atomically read once
	- the values displayed via dout() calls are known to
	  be meaningful at the time they are formatted
Signed-off-by: Alex Elder <elder@dreamhost.com>
Signed-off-by: Sage Weil <sage@newdream.net>

There is no real need for ceph_tcp_connect() to return the socket
pointer it creates, since it already assigns it to con->sock, which
is visible to the caller.  Instead, have it return an error code,
which tidies things up a bit.
Signed-off-by: Alex Elder <elder@dreamhost.com>
Signed-off-by: Sage Weil <sage@newdream.net>

Define a helper function to perform various cleanup operations.  Use
it both in the exit routine and in the init routine in the event of
an error.
Signed-off-by: Alex Elder <elder@dreamhost.com>
Signed-off-by: Sage Weil <sage@newdream.net>

The messenger workqueue has no need to be public.  So give it static
scope.
Signed-off-by: Alex Elder <elder@dreamhost.com>
Signed-off-by: Sage Weil <sage@newdream.net>

Encapsulate the operation of adding a new chunk of data to the next
open slot in a ceph_connection's out_kvec array.  Also add a "reset"
operation to make subsequent add operations start at the beginning
of the array again.

Use these routines throughout, avoiding duplicate code and ensuring
all calls are handled consistently.
Signed-off-by: Alex Elder <elder@dreamhost.com>
Signed-off-by: Sage Weil <sage@newdream.net>

One of the arguments to prepare_write_connect() indicates whether it
is being called immediately after a call to prepare_write_banner().
Move the prepare_write_banner() call inside prepare_write_connect(),
and reinterpret (and rename) the "after_banner" argument so it
indicates that prepare_write_connect() should *make* the call
rather than should know it has already been made.

This was split out from the next patch to highlight this change in
logic.
Signed-off-by: Alex Elder <elder@dreamhost.com>
Signed-off-by: Sage Weil <sage@newdream.net>

Currently an rbd device's id is released when it is removed, but it
is done before the code is run to clean up sysfs-related files (such
as /sys/bus/rbd/devices/1).

It's possible that an rbd is still in use after the rbd_remove()
call has been made.  It's essentially the same as an active inode
that stays around after it has been removed--until its final close
operation.  This means that the id shows up as free for reuse at a
time it should not be.

The effect of this was seen by Jens Rehpoehler, who:
    - had a filesystem mounted on an rbd device
    - unmapped that filesystem (without unmounting)
    - found that the mount still worked properly
    - but hit a panic when he attempted to re-map a new rbd device

This re-map attempt found the previously-unmapped id available.
The subsequent attempt to reuse it was met with a panic while
attempting to (re-)install the sysfs entry for the new mapped
device.

Fix this by holding off "putting" the rbd id, until the rbd_device
release function is called--when the last reference is finally
dropped.

Note: This fixes: http://tracker.newdream.net/issues/1907Signed-off-by: Alex Elder <elder@dreamhost.com>
Signed-off-by: Sage Weil <sage@newdream.net>

Here is another set of small code tidy-ups:
    - Define SECTOR_SHIFT and SECTOR_SIZE, and use these symbolic
      names throughout.  Tell the blk_queue system our physical
      block size, in the (unlikely) event we want to use something
      other than the default.
    - Delete the definition of struct rbd_info, which is never used.
    - Move the definition of dev_to_rbd() down in its source file,
      just above where it gets first used, and change its name to
      dev_to_rbd_dev().
    - Replace an open-coded operation in rbd_dev_release() to use
      dev_to_rbd_dev() instead.
    - Calculate the segment size for a given rbd_device just once in
      rbd_init_disk().
    - Use the '%zd' conversion specifier in rbd_snap_size_show(),
      since the value formatted is a size_t.
    - Switch to the '%llu' conversion specifier in rbd_snap_id_show().
      since the value formatted is unsigned.
Signed-off-by: Alex Elder <elder@dreamhost.com>

A few blocks of code are rearranged a bit here:
    - In rbd_header_from_disk():
	- Don't bother computing snap_count until we're sure the
	  on-disk header starts with a good signature.
	- Move a few independent lines of code so they are *after* a
	  check for a failed memory allocation.
	- Get rid of unnecessary local variable "ret".
    - Make a few other changes in rbd_read_header(), similar to the
      above--just moving things around a bit while preserving the
      functionality.
    - In rbd_rq_fn(), just assign rq in the while loop's controlling
      expression rather than duplicating it before and at the end of
      the loop body.  This allows the use of "continue" rather than
      "goto next" in a number of spots.
    - Rearrange the logic in snap_by_name().  End result is the same.
Signed-off-by: Alex Elder <elder@dreamhost.com>

Once rbd_bus_type is registered, it allows an "add" operation via
the /sys/bus/rbd/add bus attribute, and adding a new rbd device that
way establishes a connection between the device and rbd_root_dev.
But rbd_root_dev is not registered until after the rbd_bus_type
registration is complete.  This could (in principle anyway) result
in an invalid state.

Since rbd_root_dev has no tie to rbd_bus_type we can reorder these
two initializations and never be faced with this scenario.

In addition, unregister the device in the event the bus registration
fails at module init time.
Signed-off-by: Alex Elder <elder@dreamhost.com>
Signed-off-by: Sage Weil <sage@newdream.net>

The mon_addrs buffer in rbd_add is used to hold a copy of the
monitor IP addresses supplied via /sys/bus/rbd/add.  That is
passed to rbd_get_client(), which never modifies it (nor do
any of the functions it gets passed to thereafter)--the mon_addr
parameter to rbd_get_client() is a pointer to constant data, so it
can't be modifed.  Furthermore, rbd_get_client() has the length of
the mon_addrs buffer and that is used to ensure nothing goes beyond
its end.

Based on all this, there is no reason that a buffer needs to
be used to hold a copy of the mon_addrs provided via
/sys/bus/rbd/add.   Instead, the location within that passed-in
buffer can be provided, along with the length of the "token"
therein which represents the monitor IP's.

A small change to rbd_add_parse_args() allows the address within the
buffer to be passed back, and the length is already returned.  This
now means that, at least from the perspective of this interface,
there is no such thing as a list of monitor addresses that is too
long.
Signed-off-by: Alex Elder <elder@dreamhost.com>

The argument parsing routine already computes the size of the
mon_addrs buffer it extracts from the "command."  Pass it to the
caller so it can use it to provide the length to rbd_get_client().
Signed-off-by: Alex Elder <elder@dreamhost.com>

This is a bit gratuitous, but there are a few things that can be
verified at build time rather than run time, so do that.
Signed-off-by: Alex Elder <elder@dreamhost.com>

Make use of a few simple helper routines to parse the arguments
rather than sscanf().  This will treat both missing and too-long
arguments as invalid input (rather than silently truncating the
input in the too-long case).  In time this can also be used by
rbd_add() to use the passed-in buffer in place, rather than copying
its contents into new buffers.

It appears to me that the sscanf() previously used would not
correctly handle a supplied snapshot--the two final "%s" conversion
specifications were not separated by a space, and I'm not sure
how sscanf() handles that situation.  It may not be well-defined.
So that may be a bug this change fixes (but I didn't verify that).

The sizes of the mon_addrs and options buffers are now passed to
rbd_add_parse_args(), so they can be supplied to copy_token().
Signed-off-by: Alex Elder <elder@dreamhost.com>

Move the code that parses the arguments provided to rbd_add() (which
are supplied via /sys/bus/rbd/add) into a separate function.

Also rename the "mon_dev_name" variable in rbd_add() to be
"mon_addrs".   The variable represents a list of one or more
comma-separated monitor IP addresses, each with an optional port
number.  I think "mon_addrs" captures that notion a little better.
Signed-off-by: Alex Elder <elder@dreamhost.com>

If a couple pointers are initialized to NULL then a single
"out_nomem" label can be used for all of the memory allocation
failure cases in rbd_add().

Also, get rid of the "irc" local variable there.  There is no
real need for "rc" to be type ssize_t, and it can be used in
the spot "irc" was.
Signed-off-by: Alex Elder <elder@dreamhost.com>
Signed-off-by: Sage Weil <sage@newdream.net>

The length of the string containing the monitor address
specification(s) will never exceed the length of the string passed
in to rbd_add().  The same holds true for the ceph + rbd options
string.  So reduce the amount of memory allocated for these to
that length rather than the maximum (1024 bytes).
Signed-off-by: Alex Elder <elder@dreamhost.com>
Signed-off-by: Sage Weil <sage@newdream.net>

Since rbd_get_client() currently returns an error code.  It assigns
the rbd_client field of the rbd_device structure it is passed if
successful.  Instead, have it return the created rbd_client
structure and return a pointer-coded error if there is an error.
This makes the assignment of the client pointer more obvious at the
call site.
Signed-off-by: Alex Elder <elder@dreamhost.com>
Signed-off-by: Sage Weil <sage@newdream.net>

Here are a few very simple cleanups:
    - Add a "RBD_" prefix to the two driver name string definitions.
    - Move the definition of struct rbd_request below struct rbd_req_coll
      to avoid the need for an empty declaration of the latter.
    - Move and group the definitions of rbd_root_dev_release() and
      rbd_root_dev, as well as rbd_bus_type and rbd_bus_attrs[],
      close to the top of the file.  Arrange the latter so
      rbd_bus_type.bus_attrs can be initialized statically.
    - Get rid of an unnecessary local variable in rbd_open().
    - Rework some hokey logic in rbd_bus_add_dev(), so the value of
      "ret" at the end is either 0 or -ENOENT to avoid the need for
      the code duplication that was there.
    - Rename a goto target in rbd_add().
Signed-off-by: Alex Elder <elder@dreamhost.com>
Signed-off-by: Sage Weil <sage@newdream.net>

The spinlock used to protect rbd_client_list is named "node_lock".
Rename it to "rbd_client_list_lock" to make it more obvious what
it's for.
Signed-off-by: Alex Elder <elder@dreamhost.com>
Signed-off-by: Sage Weil <sage@newdream.net>

Since rbd_client_create() is only called in one place, move the
acquisition of the mutex around that call inside that function.
Signed-off-by: Alex Elder <elder@dreamhost.com>
Signed-off-by: Sage Weil <sage@newdream.net>

Since rbd_get_client() is only called in one place, move the
acquisition of the mutex around that call inside that function.

Furthermore, within rbd_get_client(), it appears the mutex only
needs to be held while calling rbd_client_create().  (Moving
the lock inside that function will wait for the next patch.)
Signed-off-by: Alex Elder <elder@dreamhost.com>
Signed-off-by: Sage Weil <sage@newdream.net>

In rbd_get_client(), if a client is reused, a number of things
get done while still holding the list lock unnecessarily.

This just moves a few things that need no lock protection outside
the lock.
Signed-off-by: Alex Elder <elder@dreamhost.com>
Signed-off-by: Sage Weil <sage@newdream.net>

It used to be that selecting a new unique identifier for an added
rbd device required searching all existing ones to find the highest
id is used.  A recent change made that unnecessary, but made it
so that id's used were monotonically non-decreasing.  It's a bit
more pleasant to have smaller rbd id's though, and this change
makes ids get allocated as they were before--each new id is one more
than the maximum currently in use.
Signed-off-by: Alex Elder <elder@dreamhost.com>
Signed-off-by: Sage Weil <sage@newdream.net>

The only time entries are added to or removed from the global
rbd_dev_list is exactly when a "put" or "get" operation is being
performed on a rbd_dev's id.  So just move the list management code
into get/put routines.
Signed-off-by: Alex Elder <elder@dreamhost.com>
Signed-off-by: Sage Weil <sage@newdream.net>

The rbd_dev_list is just a simple list of all the current
rbd_devices.  Using the ctl_mutex as a concurrency guard is
overkill.  Instead, use a spinlock for that specific purpose.

This also reduces the window that the ctl_mutex needs to be held in
rbd_add().
Signed-off-by: Alex Elder <elder@dreamhost.com>
Signed-off-by: Sage Weil <sage@newdream.net>

In order to select a new unique identifier for an added rbd device,
the list of all existing ones is searched and a value one greater
than the highest id is used.

The list search can be avoided by using an atomic variable that
keeps track of the current highest id.  Using a get/put model for
id's we can limit the boundless growth of id numbers a bit by
arranging to reuse the current highest id once it gets released.
Add these calls to "put" the id when an rbd is getting removed.

Note that this changes the pattern of device id's used--new values
will never be below the highest one seen so far (even if there
exists an unused lower one).  I assert this is OK because the key
property of an rbd id is its uniqueness, not its magnitude.

Regardless, a follow-on patch will restore the old way of doing
things, I just think this commit just makes the incremental change
to atomics a little easier to understand.
Signed-off-by: Alex Elder <elder@dreamhost.com>
Signed-off-by: Sage Weil <sage@newdream.net>

Move the loop that finds a new unique rbd id to use into
its own helper function.
Signed-off-by: Alex Elder <elder@dreamhost.com>
Signed-off-by: Sage Weil <sage@newdream.net>

There's already a constant for this anyway.

Since rbd_header_set_snap() is only used to set the rbd device
snap_name field, just do that within that function rather than
having it take the snap_name as an argument.
Signed-off-by: Alex Elder <elder@dreamhost.com>
Signed-off-by: Sage Weil <sage@newdream.net>

v2: Changed interface rbd_header_set_snap() so it explicitly updates
    the snap_name in the rbd_device.  Also added a BUILD_BUG_ON()
    to verify the size of the snap_name field is sufficient for
    SNAP_HEAD_NAME.

The rbd_device structure maintains a duplicate copy of the
ceph_client pointer maintained in its rbd_client structure.  There
appears to be no good reason for this, and its presence presents a
risk of them getting out of synch or otherwise misused.  So kill it
off, and use the rbd_client copy only.
Signed-off-by: Alex Elder <elder@dreamhost.com>
Signed-off-by: Sage Weil <sage@newdream.net>

ceph_parse_options() takes the address of a pointer as an argument
and uses it to return the address of an allocated structure if
successful.  With this interface is not evident at call sites that
the pointer is always initialized.  Change the interface to return
the address instead (or a pointer-coded error code) to make the
validity of the returned pointer obvious.
Signed-off-by: Alex Elder <elder@dreamhost.com>
Signed-off-by: Sage Weil <sage@newdream.net>

Some minor cleanups in "drivers/block/rbd.c:
    - Use the more meaningful "RBD_MAX_OBJ_NAME_LEN" in place if "96"
      in the definition of RBD_MAX_MD_NAME_LEN.
    - Use DEFINE_SPINLOCK() to define and initialize node_lock.
    - Drop a needless (char *) cast in parse_rbd_opts_token().
    - Make a few minor formatting changes.
Signed-off-by: Alex Elder <elder@dreamhost.com>
Signed-off-by: Sage Weil <sage@newdream.net>