Alex Elder [Tue, 5 Mar 2013 00:29:06 +0000 (18:29 -0600)]
libceph: activate message data assignment checks
The mds client no longer tries to assign zero-length message data,
and the osd client no longer sets its data info more than once.
This allows us to activate assertions in the messenger to verify
these things never happen.
This resolves both of these:
http://tracker.ceph.com/issues/4263
http://tracker.ceph.com/issues/4284
Signed-off-by: Alex Elder <elder@inktank.com> Reviewed-by: Greg Farnum <greg@inktank.com>
Alex Elder [Tue, 5 Mar 2013 00:29:06 +0000 (18:29 -0600)]
libceph: set response data fields earlier
When an incoming message is destined for the osd client, the
messenger calls the osd client's alloc_msg method. That function
looks up which request has the tid matching the incoming message,
and returns the request message that was preallocated to receive the
response. The response message is therefore known before the
request is even started.
Between the start of the request and the receipt of the response,
the request and its data fields will not change, so there's no
reason we need to hold off setting them. In fact it's preferable
to set them just once because it's more obvious that they're
unchanging.
So set up the fields describing where incoming data is to land in a
response message at the beginning of ceph_osdc_start_request().
Define a helper function that sets these fields, and use it to
set the fields for both outgoing data in the request message and
incoming data in the response.
This resolves:
http://tracker.ceph.com/issues/4284
Signed-off-by: Alex Elder <elder@inktank.com> Reviewed-by: Josh Durgin <josh.durgin@inktank.com>
Alex Elder [Thu, 7 Mar 2013 21:38:26 +0000 (15:38 -0600)]
libceph: record message data byte length
Record the number of bytes of data in a page array rather than the
number of pages in the array. It can be assumed that the page array
is of sufficient size to hold the number of bytes indicated (and
offset by the indicated alignment).
Signed-off-by: Alex Elder <elder@inktank.com> Reviewed-by: Josh Durgin <josh.durgin@inktank.com>
Alex Elder [Thu, 14 Feb 2013 18:16:43 +0000 (12:16 -0600)]
libceph: isolate other message data fields
Define ceph_msg_data_set_pagelist(), ceph_msg_data_set_bio(), and
ceph_msg_data_set_trail() to clearly abstract the assignment of the
remaining data-related fields in a ceph message structure. Use the
new functions in the osd client and mds client.
This partially resolves:
http://tracker.ceph.com/issues/4263
Signed-off-by: Alex Elder <elder@inktank.com> Reviewed-by: Josh Durgin <josh.durgin@inktank.com>
Alex Elder [Thu, 14 Feb 2013 18:16:43 +0000 (12:16 -0600)]
libceph: isolate message page field manipulation
Define a function ceph_msg_data_set_pages(), which more clearly
abstracts the assignment page-related fields for data in a ceph
message structure. Use this new function in the osd client and mds
client.
Ideally, these fields would never be set more than once (with
BUG_ON() calls to guarantee that). At the moment though the osd
client sets these every time it receives a message, and in the event
of a communication problem this can happen more than once. (This
will be resolved shortly, but setting up these helpers first makes
it all a bit easier to work with.)
Rearrange the field order in a ceph_msg structure to group those
that are used to define the possible data payloads.
This partially resolves:
http://tracker.ceph.com/issues/4263
Signed-off-by: Alex Elder <elder@inktank.com> Reviewed-by: Josh Durgin <josh.durgin@inktank.com>
Alex Elder [Thu, 7 Mar 2013 21:38:25 +0000 (15:38 -0600)]
libceph: record byte count not page count
Record the byte count for an osd request rather than the page count.
The number of pages can always be derived from the byte count (and
alignment/offset) but the reverse is not true.
Signed-off-by: Alex Elder <elder@inktank.com> Reviewed-by: Josh Durgin <josh.durgin@inktank.com>
Alex Elder [Sat, 2 Mar 2013 00:00:16 +0000 (18:00 -0600)]
libceph: simplify new message initialization
Rather than explicitly initializing many fields to 0, NULL, or false
in a newly-allocated message, just use kzalloc() for allocating new
messages. This will become a much more convenient way of doing
things anyway for upcoming patches that abstract the data field.
Signed-off-by: Alex Elder <elder@inktank.com> Reviewed-by: Josh Durgin <josh.durgin@inktank.com>
Alex Elder [Thu, 7 Mar 2013 05:39:38 +0000 (23:39 -0600)]
libceph: advance pagelist with list_rotate_left()
While processing an outgoing pagelist (either the data pagelist or
trail) in a ceph message, the messenger cycles through each of the
pages on the list. This is accomplished in out_msg_pos_next(), if
the end of the first page on the list is reached, the first page is
moved to the end of the list.
There is a list operation, list_rotate_left(), which performs
exactly this operation, and by using it, what's really going on
becomes more obvious.
So replace these two list_move_tail() calls with list_rotate_left().
Signed-off-by: Alex Elder <elder@inktank.com> Reviewed-by: Josh Durgin <josh.durgin@inktank.com>
Alex Elder [Sat, 9 Mar 2013 00:51:04 +0000 (18:51 -0600)]
libceph: define and use in_msg_pos_next()
Define a new function in_msg_pos_next() to match out_msg_pos_next(),
and use it in place of code at the end of read_partial_message_pages()
and read_partial_message_bio().
Note that the page number is incremented and offset reset under
slightly different conditions from before. The result is
equivalent, however, as explained below.
Each time an incoming message is going to arrive, we find out how
much room is left--not surpassing the current page--and provide that
as the number of bytes to receive. So the amount we'll use is the
lesser of: all that's left of the entire request; and all that's
left in the current page.
If we received exactly how many were requested, we either reached
the end of the request or the end of the page. In the first case,
we're done, in the second, we move onto the next page in the array.
In all cases but (possibly) on the last page, after adding the
number of bytes received, page_pos == PAGE_SIZE. On the last page,
it doesn't really matter whether we increment the page number and
reset the page position, because we're done and we won't come back
here again. The code previously skipped over that last case,
basically. The new code handles that case the same as the others,
incrementing and resetting.
Signed-off-by: Alex Elder <elder@inktank.com> Reviewed-by: Josh Durgin <josh.durgin@inktank.com>
Alex Elder [Sat, 9 Mar 2013 00:51:03 +0000 (18:51 -0600)]
libceph: kill args in read_partial_message_bio()
There is only one caller for read_partial_message_bio(), and it
always passes &msg->bio_iter and &bio_seg as the second and third
arguments. Furthermore, the message in question is always the
connection's in_msg, and we can get that inside the called function.
So drop those two parameters and use their derived equivalents.
Signed-off-by: Alex Elder <elder@inktank.com> Reviewed-by: Josh Durgin <josh.durgin@inktank.com>
Sage Weil [Wed, 6 Mar 2013 22:57:03 +0000 (14:57 -0800)]
libceph: fix decoding of pgids
In 4f6a7e5ee1393ec4b243b39dac9f36992d161540 we effectively dropped support
for the legacy encoding for the OSDMap and incremental. However, we didn't
fix the decoding for the pgid.
Signed-off-by: Sage Weil <sage@inktank.com> Reviewed-by: Yehuda Sadeh <yehuda@inktank.com>
Alex Elder [Tue, 5 Mar 2013 15:25:10 +0000 (09:25 -0600)]
libceph: clean up skipped message logic
In ceph_con_in_msg_alloc() it is possible for a connection's
alloc_msg method to indicate an incoming message should be skipped.
By default, read_partial_message() initializes the skip variable
to 0 before it gets provided to ceph_con_in_msg_alloc().
The osd client, mon client, and mds client each supply an alloc_msg
method. The mds client always assigns skip to be 0.
The other two leave the skip value of as-is, or assigns it to zero,
except:
- if no (osd or mon) request having the given tid is found, in
which case skip is set to 1 and NULL is returned; or
- in the osd client, if the data of the reply message is not
adequate to hold the message to be read, it assigns skip
value 1 and returns NULL.
So the returned message pointer will always be NULL if skip is ever
non-zero.
Clean up the logic a bit in ceph_con_in_msg_alloc() to make this
state of affairs more obvious. Add a comment explaining how a null
message pointer can mean either a message that should be skipped or
a problem allocating a message.
This resolves:
http://tracker.ceph.com/issues/4324
Alex Elder [Thu, 14 Feb 2013 18:16:43 +0000 (12:16 -0600)]
libceph: separate read and write data
An osd request defines information about where data to be read
should be placed as well as where data to write comes from.
Currently these are represented by common fields.
Keep information about data for writing separate from data to be
read by splitting these into data_in and data_out fields.
This is the key patch in this whole series, in that it actually
identifies which osd requests generate outgoing data and which
generate incoming data. It's less obvious (currently) that an osd
CALL op generates both outgoing and incoming data; that's the focus
of some upcoming work.
This resolves:
http://tracker.ceph.com/issues/4127
Signed-off-by: Alex Elder <elder@inktank.com> Reviewed-by: Josh Durgin <josh.durgin@inktank.com>
Alex Elder [Thu, 14 Feb 2013 18:16:43 +0000 (12:16 -0600)]
libceph: distinguish page and bio requests
An osd request uses either pages or a bio list for its data. Use a
union to record information about the two, and add a data type
tag to select between them.
Signed-off-by: Alex Elder <elder@inktank.com> Reviewed-by: Josh Durgin <josh.durgin@inktank.com>
Alex Elder [Sat, 2 Mar 2013 00:00:15 +0000 (18:00 -0600)]
libceph: don't assign page info in ceph_osdc_new_request()
Currently ceph_osdc_new_request() assigns an osd request's
r_num_pages and r_alignment fields. The only thing it does
after that is call ceph_osdc_build_request(), and that doesn't
need those fields to be assigned.
Move the assignment of those fields out of ceph_osdc_new_request()
and into its caller. As a result, the page_align parameter is no
longer used, so get rid of it.
Note that in ceph_sync_write(), the value for req->r_num_pages had
already been calculated earlier (as num_pages, and fortunately
it was computed the same way). So don't bother recomputing it,
but because it's not needed earlier, move that calculation after the
call to ceph_osdc_new_request(). Hold off making the assignment to
r_alignment, doing it instead r_pages and r_num_pages are
getting set.
Similarly, in start_read(), nr_pages already holds the number of
pages in the array (and is calculated the same way), so there's no
need to recompute it. Move the assignment of the page alignment
down with the others there as well.
This and the next few patches are preparation work for:
http://tracker.ceph.com/issues/4127
Signed-off-by: Alex Elder <elder@inktank.com> Reviewed-by: Josh Durgin <josh.durgin@inktank.com>
(This is being reposted. The first one had a problem because it
erroneously added a similar change elsewhere; that change has been
dropped.)
The next patch in this series points out that the calculation for
the number of pages in an osd request is getting done twice. It
is not obvious, but the result of both calculations is identical.
This patch simplifies one of them--as a separate step--to make
it clear that the transformation in the next patch is valid.
In ceph_sync_write() there is some magic that computes page_align
for an osd request. But a little analysis shows it can be
simplified.
First, we have:
io_align = pos & ~PAGE_MASK;
which is used here:
page_align = (pos - io_align + buf_align) & ~PAGE_MASK;
Note (pos - io_align) simply rounds "pos" down to the nearest multiple
of the page size.
We also have:
buf_align = (unsigned long)data & ~PAGE_MASK;
Adding buf_align to that rounded-down "pos" value will stay within
the same page; the result will just be offset by the page offset for
the "data" pointer. The final mask therefore leaves just the value
of "buf_align".
One more simplification. Note that the result of calc_pages_for()
is invariant of which page the offset starts in--the only thing that
matters is the offset within the starting page. We will have
put the proper page offset to use into "page_align", so just use
that in calculating num_pages.
This resolves:
http://tracker.ceph.com/issues/4166
Signed-off-by: Alex Elder <elder@inktank.com> Reviewed-by: Josh Durgin <josh.durgin@inktank.com>
Alex Elder [Sat, 2 Mar 2013 00:00:15 +0000 (18:00 -0600)]
ceph: use calc_pages_for() in start_read()
There's a spot that computes the number of pages to allocate for a
page-aligned length by just shifting it. Use calc_pages_for()
instead, to be consistent with usage everywhere else. The result
is the same.
The reason for this is to make it clearer in an upcoming patch that
this calculation is duplicated.
Signed-off-by: Alex Elder <elder@inktank.com> Reviewed-by: Josh Durgin <josh.durgin@inktank.com>
Alex Elder [Sat, 2 Mar 2013 00:00:14 +0000 (18:00 -0600)]
libceph: define mds_alloc_msg() method
The only user of the ceph messenger that doesn't define an alloc_msg
method is the mds client. Define one, such that it works just like
it did before, and simplify ceph_con_in_msg_alloc() by assuming the
alloc_msg method is always present.
This and the next patch resolve:
http://tracker.ceph.com/issues/4322
Signed-off-by: Alex Elder <elder@inktank.com> Reviewed-by: Greg Farnum <greg@inktank.com>
Alex Elder [Sat, 2 Mar 2013 00:00:15 +0000 (18:00 -0600)]
libceph: rename ceph_calc_object_layout()
The purpose of ceph_calc_object_layout() is to fill in the pool
number and seed for a ceph_pg structure provided, based on a given
osd map and target object id.
Currently that function takes a file layout parameter, but the only
thing used out of that is its pool number.
Change the function so it takes a pool number rather than the full
file layout structure. Only update the ceph_pg if the pool is found
in the osd map. Get rid of few useless lines of code from the
function while there.
Since the function now very clearly just fills in the ceph_pg
structure it's provided, rename it ceph_calc_ceph_pg().
Signed-off-by: Alex Elder <elder@inktank.com> Reviewed-by: Josh Durgin <josh.durgin@inktank.com>
Alex Elder [Sat, 2 Mar 2013 00:00:15 +0000 (18:00 -0600)]
libceph: use (void *) for untyped data in osd ops
Two of the fields defining osd operations are defined using (char *)
while the data they represent are really untyped, not character
strings. Change them to have type (void *).
Signed-off-by: Alex Elder <elder@inktank.com> Reviewed-by: Josh Durgin <josh.durgin@inktank.com>
Alex Elder [Mon, 4 Mar 2013 17:08:29 +0000 (11:08 -0600)]
libceph: fix wrong opcode use in osd_req_encode_op()
The new cases added to osd_req_encode_op() caused a new sparse
error, which highlighted an existing problem that had been
overlooked since it was originally checked in. When an unsupported
opcode is found the destination rather than the source opcode was
being used in the error message. The two differ in their byte
order, and we want to be using the one in the source.
Alex Elder [Wed, 27 Feb 2013 16:26:25 +0000 (10:26 -0600)]
libceph: complete lingering requests only once
An osd request marked to linger will be re-submitted in the event
a connection to the target osd gets dropped. Currently, if there
is a callback function associated with a request it will be called
each time a request is submitted--which for lingering requests can
be more than once.
Change it so a request--including lingering ones--will get completed
(from the perspective of the user of the osd client) exactly once.
This resolves:
http://tracker.ceph.com/issues/3967
Signed-off-by: Alex Elder <elder@inktank.com> Reviewed-by: Josh Durgin <josh.durgin@inktank.com>
Yan, Zheng [Fri, 1 Mar 2013 02:55:39 +0000 (10:55 +0800)]
ceph: don't early drop Fw cap
ceph_aio_write() has an optimization that marks CEPH_CAP_FILE_WR
cap dirty before data is copied to page cache and inode size is
updated. The optimization avoids slow cap revocation caused by
balance_dirty_pages(), but introduces inode size update race. If
ceph_check_caps() flushes the dirty cap before the inode size is
updated, MDS can miss the new inode size. So just remove the
optimization.
Yan, Zheng [Mon, 18 Feb 2013 08:38:14 +0000 (16:38 +0800)]
ceph: use I_COMPLETE inode flag instead of D_COMPLETE flag
commit c6ffe10015 moved the flag that tracks if the dcache contents
for a directory are complete to dentry. The problem is there are
lots of places that use ceph_dir_{set,clear,test}_complete() while
holding i_ceph_lock. but ceph_dir_{set,clear,test}_complete() may
sleep because they call dput().
This patch basically reverts that commit. For ceph_d_prune(), it's
called with both the dentry to prune and the parent dentry are
locked. So it's safe to access the parent dentry's d_inode and
clear I_COMPLETE flag.
Yan, Zheng [Wed, 27 Feb 2013 01:26:09 +0000 (09:26 +0800)]
ceph: set mds_want according to cap import message
MDS ignores cap update message if migrate_seq mismatch, so when
receiving a cap import message with higher migrate_seq, set mds_want
according to the cap import message.
Alex Elder [Thu, 14 Feb 2013 18:16:43 +0000 (12:16 -0600)]
libceph: set page alignment in start_request()
The page alignment field for a request is currently set in
ceph_osdc_build_request(). It's not needed at that point
nor do either of its callers need that value assigned at
any point before they call ceph_osdc_start_request().
So move that assignment into ceph_osdc_start_request().
Signed-off-by: Alex Elder <elder@inktank.com> Reviewed-by: Josh Durgin <josh.durgin@inktank.com>
Alex Elder [Mon, 25 Feb 2013 23:35:46 +0000 (17:35 -0600)]
libceph: distinguish page array and pagelist count
Use distinct fields for tracking the number of pages in a message's
page array and in a message's page list. Currently only one or the
other is used at a time, but that will be changing soon.
Signed-off-by: Alex Elder <elder@inktank.com> Reviewed-by: Josh Durgin <josh.durgin@inktank.com>
Alex Elder [Sat, 16 Feb 2013 04:10:17 +0000 (22:10 -0600)]
libceph: don't pass request to calc_layout()
The only remaining reason to pass the osd request to calc_layout()
is to fill in its r_num_pages and r_page_alignment fields. Once it
fills those in, it doesn't do anything more with them.
We can therefore move those assignments into the caller, and get rid
of the "req" parameter entirely.
Note, however, that the only caller is ceph_osdc_new_request(),
and that immediately overwrites those fields with values based on
its passed-in page offset. So the assignment inside calc_layout()
was redundant anyway.
This resolves:
http://tracker.ceph.com/issues/4262
Signed-off-by: Alex Elder <elder@inktank.com> Reviewed-by: Josh Durgin <josh.durgin@inktank.com>
Alex Elder [Sat, 16 Feb 2013 04:10:17 +0000 (22:10 -0600)]
libceph: format target object name in caller
Move the formatting of the object name (oid) to use for an object
request into the caller of calc_layout(). This makes the "vino"
parameter no longer necessary, so get rid of it.
Signed-off-by: Alex Elder <elder@inktank.com> Reviewed-by: Josh Durgin <josh.durgin@inktank.com>
Alex Elder [Sat, 16 Feb 2013 04:10:17 +0000 (22:10 -0600)]
libceph: make ceph_msg->bio_seg be unsigned
The bio_seg field is used by the ceph messenger in iterating through
a bio. It should never have a negative value, so make it an
unsigned. (I contemplated making it unsigned short to match the
struct bio definition, but it offered no benefit.)
Change variables used to hold bio_seg values to all be unsigned as
well. Change two variable names in init_bio_iter() to match the
convention used everywhere else.
Signed-off-by: Alex Elder <elder@inktank.com> Reviewed-by: Josh Durgin <josh.durgin@inktank.com>
Alex Elder [Sat, 16 Feb 2013 04:10:17 +0000 (22:10 -0600)]
libceph: fix a osd request memory leak
If an invalid layout is provided to ceph_osdc_new_request(), its
call to calc_layout() might return an error. At that point in the
function we've already allocated an osd request structure, so we
need to free it (drop a reference) in the event such an error
occurs.
The only other value calc_layout() will return is 0, so make that
explicit in the successful case.
This resolves:
http://tracker.ceph.com/issues/4240
Signed-off-by: Alex Elder <elder@inktank.com> Reviewed-by: Josh Durgin <josh.durgin@inktank.com>
Sage Weil [Tue, 26 Feb 2013 18:39:09 +0000 (10:39 -0800)]
libceph: add support for HASHPSPOOL pool flag
The legacy behavior adds the pgid seed and pool together as the input for
CRUSH. That is problematic because each pool's PGs end up mapping to the
same OSDs: 1.5 == 2.4 == 3.3 == ...
Instead, if the HASHPSPOOL flag is set, we has the ps and pool together and
feed that into CRUSH. This ensures that two adjacent pools will map to
an independent pseudorandom set of OSDs.
Advertise our support for this via a protocol feature flag.
Signed-off-by: Sage Weil <sage@inktank.com> Reviewed-by: Alex Elder <elder@inktank.com>
Sage Weil [Tue, 26 Feb 2013 00:11:12 +0000 (16:11 -0800)]
libceph: update osd request/reply encoding
Use the new version of the encoding for osd requests and replies. In the
process, update the way we are tracking request ops and reply lengths and
results in the struct ceph_osd_request. Update the rbd and fs/ceph users
appropriately.
The main changes are:
- we keep pointers into the request memory for fields we need to update
each time the request is sent out over the wire
- we keep information about the result in an array in the request struct
where the users can easily get at it.
Signed-off-by: Sage Weil <sage@inktank.com> Reviewed-by: Alex Elder <elder@inktank.com>
Sage Weil [Tue, 26 Feb 2013 00:13:08 +0000 (16:13 -0800)]
libceph: calculate placement based on the internal data types
Instead of using the old ceph_object_layout struct, update our internal
ceph_calc_object_layout method to use the ceph_pg type. This allows us to
pass the full 32-bit precision of the pgid.seed to the callers. It also
allows some callers to avoid reaching into the request structures for the
struct ceph_object_layout fields.
Signed-off-by: Sage Weil <sage@inktank.com> Reviewed-by: Alex Elder <elder@inktank.com>
Sage Weil [Sat, 23 Feb 2013 18:41:09 +0000 (10:41 -0800)]
ceph: update support for PGID64, PGPOOL3, OSDENC protocol features
Support (and require) the PGID64, PGPOOL3, and OSDENC protocol features.
These have been present in ceph.git since v0.42, Feb 2012. Require these
features to simplify support; nobody is running older userspace.
Note that the new request and reply encoding is still not in place, so the new
code is not yet functional.
Signed-off-by: Sage Weil <sage@inktank.com> Reviewed-by: Alex Elder <elder@inktank.com>
Sage Weil [Sat, 23 Feb 2013 18:38:16 +0000 (10:38 -0800)]
libceph: decode into cpu-native ceph_pg type
Always decode data into our cpu-native ceph_pg type that has the correct
field widths. Limit any remaining uses of ceph_pg_v1 to dealing with the
legacy protocol.
Signed-off-by: Sage Weil <sage@inktank.com> Reviewed-by: Alex Elder <elder@inktank.com>
Alex Elder [Tue, 26 Feb 2013 20:23:07 +0000 (14:23 -0600)]
rbd: pass length, not op for osd completions
The only thing type-specific osd completion functions do with their
osd op parameter is (in some cases) extract the number of bytes
transferred from it. In the other cases, the xferred bytes field
is not used, and total message data transfer byte count (which may
well be zero) is used.
Just set the object request transfer count in the main osd request
callback function and provide that to the other routines. There is
then no longer any need to pass the op pointer to the type-specific
completion routines, so drop those parameters.
Stop doing anything with the total message data length.
Signed-off-by: Alex Elder <elder@inktank.com> Reviewed-by: Sage Weil <sage@inktank.com>
Alex Elder [Tue, 19 Feb 2013 18:25:57 +0000 (12:25 -0600)]
libceph: use a do..while loop in con_work()
This just converts a manually-implemented loop into a do..while loop
in con_work(). It also moves handling of EAGAIN inside the blocks
where it's already been determined an error code was returned.
Also update a few dout() calls near the affected code for
consistency.
Signed-off-by: Alex Elder <elder@inktank.com> Reviewed-by: Josh Durgin <josh.durgin@inktank.com>
Alex Elder [Tue, 19 Feb 2013 18:25:57 +0000 (12:25 -0600)]
libceph: use a flag to indicate a fault has occurred
This just rearranges the logic in con_work() a little bit so that a
flag is used to indicate a fault has occurred. This allows both the
fault and non-fault case to be handled the same way and avoids a
couple of nearly consecutive gotos.
Signed-off-by: Alex Elder <elder@inktank.com> Reviewed-by: Josh Durgin <josh.durgin@inktank.com>
Alex Elder [Tue, 19 Feb 2013 18:25:57 +0000 (12:25 -0600)]
libceph: separate non-locked fault handling
An error occurring on a ceph connection is treated as a fault,
causing the connection to be reset. The initial part of this fault
handling has to be done while holding the connection mutex, but
it must then be dropped for the last part.
Separate the part of this fault handling that executes without the
lock into its own function, con_fault_finish(). Move the call to
this new function, as well as call that drops the connection mutex,
into ceph_fault(). Rename that function con_fault() to reflect that
it's only handling the connection part of the fault handling.
The motivation for this was a warning from sparse about the locking
being done here. Rearranging things this way keeps all the mutex
manipulation within ceph_fault(), and this stops sparse from
complaining.
This partially resolves:
http://tracker.ceph.com/issues/4184
Alex Elder [Tue, 19 Feb 2013 18:25:57 +0000 (12:25 -0600)]
libceph: encapsulate connection backoff
Collect the code that tests for and implements a backoff delay for a
ceph connection into a new function, ceph_backoff().
Make the debug output messages in that part of the code report
things consistently by reporting a message in the socket closed
case, and by making the one for PREOPEN state report the connection
pointer like the rest.
Signed-off-by: Alex Elder <elder@inktank.com> Reviewed-by: Josh Durgin <josh.durgin@inktank.com>
Alex Elder [Tue, 19 Feb 2013 18:25:56 +0000 (12:25 -0600)]
libceph: eliminate sparse warnings
Eliminate most of the problems in the libceph code that cause sparse
to issue warnings.
- Convert functions that are never referenced externally to have
static scope.
- Pass NULL rather than 0 for a pointer argument in one spot in
ceph_monc_delete_snapid()
This partially resolves:
http://tracker.ceph.com/issues/4184
Alex Elder [Tue, 19 Feb 2013 18:25:56 +0000 (12:25 -0600)]
rbd: eliminate sparse warnings
Fengguang Wu reminded me that there were outstanding sparse reports
in the ceph and rbd code. This patch fixes these problems in rbd
that lead to those reports:
- Convert functions that are never referenced externally to have
static scope.
- Add a lockdep annotation to rbd_request_fn(), because it
releases a lock before acquiring it again.
This partially resolves:
http://tracker.ceph.com/issues/4184
Alex Elder [Wed, 20 Feb 2013 23:32:08 +0000 (17:32 -0600)]
rbd: normalize dout() calls
Add dout() calls to facilitate tracing of image and object requests.
Change a few existing calls so they use __func__ rather than the
hard-coded function name. Have calls always add ":" after the name
of the function, and prefix pointer values with a consistent tag
indicating what it represents. (Note that there remain some older
dout() calls that are left untouched by this patch.)
Issue a warning if rbd_osd_write_callback() ever gets a short write.
This resolves:
http://tracker.ceph.com/issues/4235
Signed-off-by: Alex Elder <elder@inktank.com> Reviewed-by: Josh Durgin <josh.durgin@inktank.com>
Alex Elder [Thu, 21 Feb 2013 16:10:06 +0000 (10:10 -0600)]
rbd: barriers are hard
Let's go shopping!
I'm afraid this may not have gotten it right: 07741308 rbd: add barriers near done flag operations
The smp_wmb() should have been done *before* setting the done flag,
to ensure all other data was valid before marking the object request
done.
Switch to use atomic_inc_return() here to set the done flag, which
allows us to verify we don't mark something done more than once.
Doing this also implies general barriers before and after the call.
And although a read memory barrier might have been sufficient before
reading the done flag, convert this to a full memory barrier just
to put this issue to bed.
This resolves:
http://tracker.ceph.com/issues/4238
Signed-off-by: Alex Elder <elder@inktank.com> Reviewed-by: Josh Durgin <josh.durgin@inktank.com>
Alex Elder [Thu, 21 Feb 2013 03:59:33 +0000 (21:59 -0600)]
rbd: ignore zero-length requests
The old request code simply ignored zero-length requests. We should
still operate that same way to avoid any changes in behavior. We
can implement handling for special zero-length requests separately
(see http://tracker.ceph.com/issues/4236).
Add some assertions based on this new constraint.
This resolves:
http://tracker.ceph.com/issues/4237
Signed-off-by: Alex Elder <elder@inktank.com> Reviewed-by: Josh Durgin <josh.durgin@inktank.com>
Sage Weil [Fri, 22 Feb 2013 23:31:00 +0000 (15:31 -0800)]
ceph: fix statvfs fr_size
Different versions of glibc are broken in different ways, but the short of
it is that for the time being, frsize should == bsize, and be used as the
multiple for the blocks, free, and available fields. This mirrors what is
done for NFS. The previous reporting of the page size for frsize meant
that newer glibc and df would report a very small value for the fs size.
Fixes http://tracker.ceph.com/issues/3793.
Signed-off-by: Sage Weil <sage@inktank.com> Reviewed-by: Greg Farnum <greg@inktank.com>
Alex Elder [Wed, 6 Feb 2013 19:11:38 +0000 (13:11 -0600)]
rbd: ignore result of ceph_copy_from_page_vector()
The result of ceph_copy_from_page_vector() is simply the length
argument it is provided.
This is called by rbd_obj_method_sync(), which returns the result if
it's non-negative. But we always either ignore or overwrite that
return value. So explicitly ignore what's returned by the copy
function, and have rbd_obj_method_sync() always return either a
negative errno or 0.
We also return the result of ceph_copy_from_page_vector() in
rbd_obj_read_sync(). There we still want to return the number of
bytes transferred, but we can use the value we already have in hand
rather than what ceph_copy_from_page_vector() provides.
Signed-off-by: Alex Elder <elder@inktank.com> Reviewed-by: Josh Durgin <josh.durgin@inktank.com>
Alex Elder [Wed, 6 Feb 2013 19:11:38 +0000 (13:11 -0600)]
rbd: prevent bytes transferred overflow
In rbd_obj_read_sync(), verify the number of bytes transferred won't
exceed what can be represented by a size_t before using it to
indicate the number of bytes to copy to the result buffer.
(The real motivation for this is to prepare for the next patch.)
Signed-off-by: Alex Elder <elder@inktank.com> Reviewed-by: Josh Durgin <josh.durgin@inktank.com>
Alex Elder [Wed, 6 Feb 2013 19:11:38 +0000 (13:11 -0600)]
libceph: use void pointers in page vector functions
The functions used for working with ceph page vectors are defined
with char pointers, but they're really intended to operate on
untyped data. Change the types of these function parameters
to (void *) to reflect this.
(Note that the functions now assume void pointer arithmetic works
like arithmetic on char pointers.)
Signed-off-by: Alex Elder <elder@inktank.com> Reviewed-by: Josh Durgin <josh.durgin@inktank.com>
Alex Elder [Wed, 6 Feb 2013 19:11:38 +0000 (13:11 -0600)]
ceph: remove a few bogus declarations
There are three ceph page vector functions declared in
"fs/ceph/super.h" that don't belong there. They're
probably left over from some long-ago code reorganization.
They're properly declared in "include/linux/ceph/libceph.h"
so just delete the ones in "super.h".
This and the next few commits resolve:
http://tracker.ceph.com/issues/4053
Signed-off-by: Alex Elder <elder@inktank.com> Reviewed-by: Josh Durgin <josh.durgin@inktank.com>
Alex Elder [Fri, 8 Feb 2013 15:55:48 +0000 (09:55 -0600)]
libceph: allow STAT osd operations
Add support for CEPH_OSD_OP_STAT operations in the osd client
and in rbd.
This operation sends no data to the osd; everything required is
encoded in identity of the target object.
The result will be ENOENT if the object doesn't exist. If it does
exist and no other error occurs the server returns the size and last
modification time of the target object as output data (in little
endian format). The size is a 64 bit unsigned and the time is
ceph_timespec structure (two unsigned 32-bit integers, representing
a seconds and nanoseconds value).
This resolves:
http://tracker.ceph.com/issues/4007
Signed-off-by: Alex Elder <elder@inktank.com> Reviewed-by: Josh Durgin <josh.durgin@inktank.com>
Alex Elder [Fri, 15 Feb 2013 17:42:30 +0000 (11:42 -0600)]
libceph: update ceph_mds_state_name() and ceph_mds_op_name()
Update ceph_mds_state_name() and ceph_mds_op_name() to include the
newly-added definitions in "ceph_fs.h", and to match its counterpart
in the user space code.
Signed-off-by: Alex Elder <elder@inktank.com> Reviewed-by: Josh Durgin <josh.durgin@inktank.com>
Alex Elder [Fri, 15 Feb 2013 17:42:30 +0000 (11:42 -0600)]
libceph: update ceph_fs.h
Update most of "include/linux/ceph/ceph_fs.h" to match its user
space counterpart in "src/include/ceph_fs.h" in the ceph tree.
Everything that has changed is either:
- added definitions (therefore no real effect on existing code)
- deleting unused symbols
- added or revised comments
There were some differences between the struct definitions for
ceph_mon_subscribe_item and the open field of ceph_mds_request_args;
those differences remain.
This and the next commit resolve:
http://tracker.ceph.com/issues/4165
Signed-off-by: Alex Elder <elder@inktank.com> Reviewed-by: Josh Durgin <josh.durgin@inktank.com>
Alex Elder [Fri, 15 Feb 2013 17:42:30 +0000 (11:42 -0600)]
libceph: remove dead code in osd_req_encode_op()
In osd_req_encode_op() there are a few cases that handle osd
opcodes that are never used in the kernel. The presence of
this code gives the impression it's correct (which really can't
be assumed), and may impose some unnecessary restrictions on
some upcoming refactoring of this code.
So delete this effectively dead code, and report uses of the
previously handled cases as unsupported.
Signed-off-by: Alex Elder <elder@inktank.com> Reviewed-by: Josh Durgin <josh.durgin@inktank.com>
Alex Elder [Fri, 15 Feb 2013 17:42:30 +0000 (11:42 -0600)]
libceph: update rados.h
Update most of "include/linux/ceph/rados.h" to match its user space
counterpart in "src/include/rados.h" in the ceph tree.
Almost everything that has changed is either:
- added or revised comments
- added definitions (therefore no real effect on existing code)
- defining the same value a different way (e.g., "1 << 0" vs "1")
The only exceptions are:
- The declaration of ceph_osd_state_name() was excluded; that
will be inserted in the next patch.
- ceph_osd_op_mode_read() and ceph_osd_op_mode_modify() are
defined differently, but they were never used in the kernel
- CEPH_OSD_FLAG_PEERSTAT is now CEPH_OSD_FLAG_PEERSTAT_OLD, but
that was never used in the kernel
Anything that was present in this file but not in its user space
counterpart was left intact here. I left the definitions of
EOLDSNAPC and EBLACKLISTED using numerical values here; I'm
not sure the right way to go with those.
This and the next two commits resolve:
http://tracker.ceph.com/issues/4164
Signed-off-by: Alex Elder <elder@inktank.com> Reviewed-by: Josh Durgin <josh.durgin@inktank.com>
Alex Elder [Fri, 15 Feb 2013 17:42:30 +0000 (11:42 -0600)]
libceph: kill ceph_osdc_wait_event()
There are no actual users of ceph_osdc_wait_event(). This would
have been one-shot events, but we no longer support those so just
get rid of this function.
Since this leaves nothing else that waits for the completion of an
event, we can get rid of the completion in a struct ceph_osd_event.
Signed-off-by: Alex Elder <elder@inktank.com> Reviewed-by: Josh Durgin <josh.durgin@inktank.com>
There is only one caller of ceph_osdc_create_event(), and it
provides 0 as its "one_shot" argument. Get rid of that argument and
just use 0 in its place.
Replace the code in handle_watch_notify() that executes if one_shot
is nonzero in the event with a BUG_ON() call.
While modifying "osd_client.c", give handle_watch_notify() static
scope.
Signed-off-by: Alex Elder <elder@inktank.com> Reviewed-by: Josh Durgin <josh.durgin@inktank.com>
Alex Elder [Fri, 15 Feb 2013 17:42:29 +0000 (11:42 -0600)]
libdeph: don't export ceph_osdc_init() or ceph_osdc_stop()
The only callers of ceph_osdc_init() and ceph_osdc_stop()
ceph_create_client() and ceph_destroy_client() (respectively)
and they are in the same kernel module as those two functions.
There's therefore no need to export those interfaces, so don't.
Signed-off-by: Alex Elder <elder@inktank.com> Reviewed-by: Josh Durgin <josh.durgin@inktank.com>
Alex Elder [Fri, 15 Feb 2013 17:42:29 +0000 (11:42 -0600)]
libceph: lock outside send_queued()
Two of the three callers of the osd client's send_queued() function
already hold the osd client mutex and drop it before the call.
Change send_queued() so it assumes the caller holds the mutex, and
update all callers accordingly. Rename it __send_queued() to match
the convention used elsewhere in the file with respect to the lock.
Signed-off-by: Alex Elder <elder@inktank.com> Reviewed-by: Josh Durgin <josh.durgin@inktank.com>
The "num_reply" parameter to ceph_osdc_new_request() is never
used inside that function, so get rid of it.
Note that ceph_sync_write() passes 2 for that argument, while all
other callers pass 1. It doesn't matter, but perhaps someone should
verify this doesn't indicate a problem.
Signed-off-by: Alex Elder <elder@inktank.com> Reviewed-by: Josh Durgin <josh.durgin@inktank.com>
There is only one caller of ceph_osdc_writepages(), and it always
passes 0 as its "flags" argument. Get rid of that argument and
replace its use in ceph_osdc_writepages() with 0.
Signed-off-by: Alex Elder <elder@inktank.com> Reviewed-by: Josh Durgin <josh.durgin@inktank.com>
There is only one caller of ceph_osdc_writepages(), and it always
passes 0 as its "dosync" argument. Get rid of that argument and
replace its use in ceph_osdc_writepages() with 0.
Signed-off-by: Alex Elder <elder@inktank.com> Reviewed-by: Josh Durgin <josh.durgin@inktank.com>
There is only one caller of ceph_osdc_writepages(), and it always
passes the value true as its "nofail" argument. Get rid of that
argument and replace its use in ceph_osdc_writepages() with the
constant value true.
This and a number of cleanup patches that follow resolve:
http://tracker.ceph.com/issues/4126
Signed-off-by: Alex Elder <elder@inktank.com> Reviewed-by: Josh Durgin <josh.durgin@inktank.com>
Linus Torvalds [Mon, 18 Feb 2013 17:58:02 +0000 (09:58 -0800)]
mm: fix pageblock bitmap allocation
Commit c060f943d092 ("mm: use aligned zone start for pfn_to_bitidx
calculation") fixed out calculation of the index into the pageblock
bitmap when a !SPARSEMEM zome was not aligned to pageblock_nr_pages.
However, the _allocation_ of that bitmap had never taken this alignment
requirement into accout, so depending on the exact size and alignment of
the zone, the use of that index could then access past the allocation,
resulting in some very subtle memory corruption.
This was reported (and bisected) by Ingo Molnar: one of his random
config builds would hang with certain very specific kernel command line
options.
In the meantime, commit c060f943d092 has been marked for stable, so this
fix needs to be back-ported to the stable kernels that backported the
commit to use the right alignment.
Linus Torvalds [Fri, 15 Feb 2013 20:12:55 +0000 (12:12 -0800)]
Merge tag 'stable/for-linus-3.8-rc7-tag-two' of git://git.kernel.org/pub/scm/linux/kernel/git/konrad/xen
Pull xen fixes from Konrad Rzeszutek Wilk:
"Two fixes:
- A simple bug-fix for redundant NULL check.
- CVE-2013-0228/XSA-42: x86/xen: don't assume %ds is usable in
xen_iret for 32-bit PVOPS
and two reverts:
- Revert the PVonHVM kexec. The patch introduces a regression with
older hypervisor stacks, such as Xen 4.1."
* tag 'stable/for-linus-3.8-rc7-tag-two' of git://git.kernel.org/pub/scm/linux/kernel/git/konrad/xen:
Revert "xen PVonHVM: use E820_Reserved area for shared_info"
Revert "xen/PVonHVM: fix compile warning in init_hvm_pv_info"
xen: remove redundant NULL check before unregister_and_remove_pcpu().
x86/xen: don't assume %ds is usable in xen_iret for 32-bit PVOPS.
Revert "[media] dvb_frontend: return -ENOTTY for unimplement IOCTL"
As reported by Klaus Schmidinger:
"In VDR I use an ioctl() call with FE_READ_UNCORRECTED_BLOCKS on a
device (using stb0899). After this call I check 'errno' for
EOPNOTSUPP to determine whether this device supports this call. This
used to work just fine, until a few months ago I noticed that my
devices using stb0899 didn't display their signal quality in VDR's OSD
any more. After further investigation I found that
ioctl(FE_READ_UNCORRECTED_BLOCKS) no longer returns EOPNOTSUPP, but
rather ENOTTY. And since I stop getting the signal quality in case
any unknown errno value appears, this broke my signal quality query
function."
Linus Torvalds [Fri, 15 Feb 2013 20:04:57 +0000 (12:04 -0800)]
Merge branch 'x86-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip
Pull one more x86 fix from Peter Anvin:
"Sigh. One more patch in the "please don't brick my Samsung" series"
* 'x86-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
efi: Clear EFI_RUNTIME_SERVICES rather than EFI_BOOT by "noefi" boot parameter
Linus Torvalds [Fri, 15 Feb 2013 20:04:08 +0000 (12:04 -0800)]
Merge tag '3.8-pci-fixes-3' of git://git.kernel.org/pub/scm/linux/kernel/git/helgaas/pci
Pull PCI fix from Bjorn Helgaas:
"This is another fix for v3.8. It fixes an oops that happens when a
Thunderbolt adapter is unplugged (remove device, poll for PME events
on no-longer-existing device, oops)."
* tag '3.8-pci-fixes-3' of git://git.kernel.org/pub/scm/linux/kernel/git/helgaas/pci:
PCI/PM: Clean up PME state when removing a device
projects / ceph-client.git / log