The code to flush buffers in the umount code is a bit iffy: we first
flush all delwri buffers out, but then might be able to queue up a
new one when logging the sb counts. On a normal shutdown that one
would get flushed out when doing the synchronous superblock write in
xfs_unmountfs_writesb, but we skip that one if the filesystem has
been shut down.
Fix this by moving the delwri list flushing until just before unmounting
the log, and while we're at it also remove the superflous delwri list
and buffer lru flusing for the rt and log device that can never have
cached or delwri buffers.
Signed-off-by: Christoph Hellwig <hch@lst.de> Reported-by: Amit Sahrawat <amit.sahrawat83@gmail.com> Tested-by: Amit Sahrawat <amit.sahrawat83@gmail.com> Signed-off-by: Alex Elder <aelder@sgi.com>
Directories are only updated transactionally, which means fsync only
needs to flush the log the inode is currently dirty, but not bother
with checking for dirty data, non-transactional updates, and most
importanly doesn't have to flush disk caches except as part of a
transaction commit.
While the first two optimizations can't easily be measured, the
latter actually makes a difference when doing lots of fsync that do
not actually have to commit the inode, e.g. because an earlier fsync
already pushed the log far enough.
The new xfs_dir_fsync is identical to xfs_nfs_commit_metadata except
for the prototype, but I'm not sure creating a common helper for the
two is worth it given how simple the functions are.
Signed-off-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Alex Elder <aelder@sgi.com>
Dave Chinner [Fri, 30 Sep 2011 04:45:03 +0000 (04:45 +0000)]
xfs: reduce the number of log forces from tail pushing
The AIL push code will issue a log force on ever single push loop
that it exits and has encountered pinned items. It doesn't rescan
these pinned items until it revisits the AIL from the start. Hence
we only need to force the log once per walk from the start of the
AIL to the target LSN.
Dave Chinner [Fri, 30 Sep 2011 04:45:02 +0000 (04:45 +0000)]
xfs: Don't allocate new buffers on every call to _xfs_buf_find
Stats show that for an 8-way unlink @ ~80,000 unlinks/s we are doing
~1 million cache hit lookups to ~3000 buffer creates. That's almost
3 orders of magnitude more cahce hits than misses, so optimising for
cache hits is quite important. In the cache hit case, we do not need
to allocate a new buffer in case of a cache miss, so we are
effectively hitting the allocator for no good reason for vast the
majority of calls to _xfs_buf_find. 8-way create workloads are
showing similar cache hit/miss ratios.
The result is profiles that look like this:
samples pcnt function DSO
_______ _____ _______________________________ _________________
Further, there is a fair bit of work involved in initialising a new
buffer once a cache miss has occurred and we currently do that under
the rbtree spinlock. That increases spinlock hold time on what are
heavily used trees.
To fix this, remove the initialisation of the buffer from
_xfs_buf_find() and only allocate the new buffer once we've had a
cache miss. Initialise the buffer immediately after allocating it in
xfs_buf_get, too, so that is it ready for insert if we get another
cache miss after allocation. This minimises lock hold time and
avoids unnecessary allocator churn. The resulting profiles look
like:
samples pcnt function DSO
_______ _____ ___________________________ _________________
There is no reason to keep a reference to the inode even if we unlock
it during transaction commit because we never drop a reference between
the ijoin and commit. Also use this fact to merge xfs_trans_ijoin_ref
back into xfs_trans_ijoin - the third argument decides if an unlock
is needed now.
I'm actually starting to wonder if allowing inodes to be unlocked
at transaction commit really is worth the effort. The only real
benefit is that they can be unlocked earlier when commiting a
synchronous transactions, but that could be solved by doing the
log force manually after the unlock, too.
Signed-off-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Alex Elder <aelder@sgi.com>
In xfs_ioc_trim it is possible that computing the last allocation group
to discard might overflow for big start & len values, because the result
might be bigger then xfs_agnumber_t which is 32 bit long. Fix this by not
allowing the start and end block of the range to be beyond the end of the
file system.
Note that if the start is beyond the end of the file system we have to
return -EINVAL, but in the "end" case we have to truncate it to the fs
size.
Also introduce "end" variable, rather than using start+len which which
might be more confusing to get right as this bug shows.
Signed-off-by: Lukas Czerner <lczerner@redhat.com> Reviewed-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Alex Elder <aelder@sgi.com>
xfs: pass bmalloca to xfs_bmap_add_extent_hole_real
All the parameters passed to xfs_bmap_add_extent_hole_real() are in
the xfs_bmalloca structure now. Just pass the bmalloca parameter to
the function instead of 8 separate parameters.
Signed-off-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Alex Elder <aelder@sgi.com>
xfs: pass bmalloca to xfs_bmap_add_extent_delay_real
All the parameters passed to xfs_bmap_add_extent_delay_real() are in
the xfs_bmalloca structure now. Just pass the bmalloca parameter to
the function instead of 8 separate parameters.
Signed-off-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Alex Elder <aelder@sgi.com>
Dave Chinner [Sun, 18 Sep 2011 20:40:57 +0000 (20:40 +0000)]
xfs: move firstblock and bmap freelist cursor into bmalloca structure
Rather than passing the firstblock and freelist structure around,
embed it into the bmalloca structure and remove it from the function
parameters.
This also enables the minleft parameter to be set only once in
xfs_bmapi_write(), and the freelist cursor directly queried in
xfs_bmapi_allocate to clear it when the lowspace algorithm is
activated.
Signed-off-by: Dave Chinner <dchinner@redhat.com> Signed-off-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Alex Elder <aelder@sgi.com>
Dave Chinner [Sun, 18 Sep 2011 20:40:56 +0000 (20:40 +0000)]
xfs: move extent records into bmalloca structure
Rather that putting extent records on the stack and then pointing to
them in the bmalloca structure which is in the same stack frame, put
the extent records directly in the bmalloca structure. This reduces
the number of args that need to be passed around.
Signed-off-by: Dave Chinner <dchinner@redhat.com> Signed-off-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Alex Elder <aelder@sgi.com>
There is no real need to the xfs_bmap_add_extent, as the callers
know what kind of extents they need to it. Removing it means
duplicating the extents to btree conversion logic in three places,
but overall it's still much simpler code and quite a bit less code.
Signed-off-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Alex Elder <aelder@sgi.com>
Dave Chinner [Sun, 18 Sep 2011 20:40:52 +0000 (20:40 +0000)]
xfs: rename xfs_bmapi to xfs_bmapi_write
Now that all the read-only users of xfs_bmapi have been converted to
use xfs_bmapi_read(), we can remove all the read-only handling cases
from xfs_bmapi().
Once this is done, rename xfs_bmapi to xfs_bmapi_write to reflect
the fact it is for allocation only. This enables us to kill the
XFS_BMAPI_WRITE flag as well.
Also clean up xfs_bmapi_write to the style used in the newly added
xfs_bmapi_read/delay functions.
Signed-off-by: Dave Chinner <dchinner@redhat.com> Signed-off-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Alex Elder <aelder@sgi.com>
Dave Chinner [Sun, 18 Sep 2011 20:40:51 +0000 (20:40 +0000)]
xfs: factor unwritten extent map manipulations out of xfs_bmapi
To further improve the readability of xfs_bmapi(), factor the
unwritten extent conversion out into a separate function. This
removes large block of logic from the xfs_bmapi() code loop and
makes it easier to see the operational logic flow for xfs_bmapi().
Signed-off-by: Dave Chinner <dchinner@redhat.com> Signed-off-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Alex Elder <aelder@sgi.com>
Dave Chinner [Sun, 18 Sep 2011 20:40:50 +0000 (20:40 +0000)]
xfs: factor extent allocation out of xfs_bmapi
To further improve the readability of xfs_bmapi(), factor the extent
allocation out into a separate function. This removes a large block
of logic from the xfs_bmapi() code loop and makes it easier to see
the operational logic flow for xfs_bmapi().
Signed-off-by: Dave Chinner <dchinner@redhat.com> Signed-off-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Alex Elder <aelder@sgi.com>
xfs: do not use xfs_bmap_add_extent for adding delalloc extents
We can just call xfs_bmap_add_extent_hole_delay directly to add a
delayed allocated regions to the extent tree, instead of going
through all the complexities of xfs_bmap_add_extent that aren't
needed for this simple case.
Signed-off-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Alex Elder <aelder@sgi.com>
Delalloc reservations are much simpler than allocations, so give
them a separate bmapi-level interface. Using the previously added
xfs_bmapi_reserve_delalloc we get a function that is only minimally
more complicated than xfs_bmapi_read, which is far from the complexity
in xfs_bmapi. Also remove the XFS_BMAPI_DELAY code after switching
over the only user to xfs_bmapi_delay.
Signed-off-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Alex Elder <aelder@sgi.com>
xfs: factor delalloc reservations out of xfs_bmapi
Move the reservation of delayed allocations, and addition of delalloc
regions to the extent trees into a new helper function. For now
this adds some twisted goto logic to xfs_bmapi, but that will be
cleaned up in the following patches.
Signed-off-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Alex Elder <aelder@sgi.com>
Dave Chinner [Sun, 18 Sep 2011 20:40:45 +0000 (20:40 +0000)]
xfs: introduce xfs_bmapi_read()
xfs_bmapi() currently handles both extent map reading and
allocation. As a result, the code is littered with "if (wr)"
branches to conditionally do allocation operations if required.
This makes the code much harder to follow and causes significant
indent issues with the code.
Given that read mapping is much simpler than allocation, we can
split out read mapping from xfs_bmapi() and reuse the logic that
we have already factored out do do all the hard work of handling the
extent map manipulations. The results in a much simpler function for
the common extent read operations, and will allow the allocation
code to be simplified in another commit.
Once xfs_bmapi_read() is implemented, convert all the callers of
xfs_bmapi() that are only reading extents to use the new function.
Signed-off-by: Dave Chinner <dchinner@redhat.com> Signed-off-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Alex Elder <aelder@sgi.com>
Dave Chinner [Sun, 18 Sep 2011 20:40:44 +0000 (20:40 +0000)]
xfs: factor extent map manipulations out of xfs_bmapi
To further improve the readability of xfs_bmapi(), factor the pure
extent map manipulations out into separate functions. This removes
large blocks of logic from the xfs_bmapi() code loop and makes it
easier to see the operational logic flow for xfs_bmapi().
Signed-off-by: Dave Chinner <dchinner@redhat.com> Signed-off-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Alex Elder <aelder@sgi.com>
xfs: remove impossible to read code in xfs_bmap_add_extent_delay_real
We already have the worst case blocks reserved, so xfs_icsb_modify_counters
won't fail in xfs_bmap_add_extent_delay_real. In fact we've had an assert
to catch this case since day and it never triggered. So remove the code
to try smaller reservations, and just return the error for that case in
addition to keeping the assert.
Signed-off-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Alex Elder <aelder@sgi.com>
xfs: remove the first extent special case in xfs_bmap_add_extent
Both xfs_bmap_add_extent_hole_delay and xfs_bmap_add_extent_hole_real
already contain code to handle the case where there is no extent to
merge with, which is effectively the same as the code duplicated here.
Signed-off-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Alex Elder <aelder@sgi.com>
An attribute of inode can be fetched via xfs_vn_getattr() in XFS.
Currently it returns EIO, not negative value, when it failed. As a
result, the system call returns not negative value even though an
error occured. The stat(2), ls and mv commands cannot handle this
error and do not work correctly.
This patch fixes this bug, and returns -EIO, not EIO when an error
is detected in xfs_vn_getattr().
Signed-off-by: Mitsuo Hayasaka <mitsuo.hayasaka.hu@hitachi.com> Reviewed-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Alex Elder <aelder@sgi.com>
Currently a buffered reader or writer can add pages to the pagecache
while we are waiting for the iolock in xfs_file_dio_aio_write. Prevent
this by re-checking mapping->nrpages after we got the iolock, and if
nessecary upgrade the lock to exclusive mode. To simplify this a bit
only take the ilock inside of xfs_file_aio_write_checks.
Signed-off-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Dave Chinner <dchinner@redhat.com> Signed-off-by: Alex Elder <aelder@sgi.com>
xfs: avoid synchronous transactions when deleting attr blocks
Currently xfs_attr_inactive causes a synchronous transactions if we are
removing a file that has any extents allocated to the attribute fork, and
thus makes XFS extremely slow at removing files with out of line extended
attributes. The code looks a like a relict from the days before the busy
extent list, but with the busy extent list we avoid reusing data and attr
extents that have been freed but not commited yet, so this code is just
as superflous as the synchronous transactions for data blocks.
Signed-off-by: Christoph Hellwig <hch@lst.de> Reported-by: Bernd Schubert <bernd.schubert@itwm.fraunhofer.de> Reviewed-by: Dave Chinner <dchinner@redhat.com> Signed-off-by: Alex Elder <aelder@sgi.com>
We now have an i_dio_count filed and surrounding infrastructure to wait
for direct I/O completion instead of i_icount, and we have never needed
to iocount waits for buffered I/O given that we only set the page uptodate
after finishing all required work. Thus remove i_iocount, and replace
the actually needed waits with calls to inode_dio_wait.
Signed-off-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Dave Chinner <dchinner@redhat.com> Signed-off-by: Alex Elder <aelder@sgi.com>
xfs: wait for I/O completion when writing out pages in xfs_setattr_size
The current code relies on the xfs_ioend_wait call later on to make sure
all I/O actually has completed. The xfs_ioend_wait call will go away soon,
so prepare for that by using the waiting filemap function.
Signed-off-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Dave Chinner <dchinner@redhat.com> Signed-off-by: Alex Elder <aelder@sgi.com>
There is no reason to queue up ioends for processing in user context
unless we actually need it. Just complete ioends that do not convert
unwritten extents or need a size update from the end_io context.
Signed-off-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Dave Chinner <dchinner@redhat.com> Signed-off-by: Alex Elder <aelder@sgi.com>
We really shouldn't complete AIO or DIO requests until we have finished
the unwritten extent conversion and size update. This means fsync never
has to pick up any ioends as all work has been completed when signalling
I/O completion.
Signed-off-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Dave Chinner <dchinner@redhat.com> Signed-off-by: Alex Elder <aelder@sgi.com>
Unify the ways we add buffers to the delwri queue by always calling
xfs_buf_delwri_queue directly. The xfs_bdwrite functions is removed and
opencoded in its callers, and the two places setting XBF_DELWRI while a
buffer is locked and expecting xfs_buf_unlock to pick it up are converted
to call xfs_buf_delwri_queue directly, too. Also replace the
XFS_BUF_UNDELAYWRITE macro with direct calls to xfs_buf_delwri_dequeue
to make the explicit queuing/dequeuing more obvious.
Signed-off-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Dave Chinner <dchinner@redhat.com> Signed-off-by: Alex Elder <aelder@sgi.com>
xfs: move more delwri setup into xfs_buf_delwri_queue
Do not transfer a reference held by the caller to the buffer on the list,
or decrement it in xfs_buf_delwri_queue, but instead grab a new reference
if needed, and let the caller drop its own reference. Also move setting
of the XBF_DELWRI and XBF_ASYNC flags into xfs_buf_delwri_queue, and
only do it if needed. Note that for now xfs_buf_unlock already has
XBF_DELWRI, but that will change in the following patches.
Signed-off-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Dave Chinner <dchinner@redhat.com> Signed-off-by: Alex Elder <aelder@sgi.com>
xfs: remove the unlock argument to xfs_buf_delwri_queue
We can just unlock the buffer in the caller, and the decrement of b_hold
would also be needed in the !unlock, we just never hit that case currently
given that the caller handles that case.
Signed-off-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Dave Chinner <dchinner@redhat.com> Signed-off-by: Alex Elder <aelder@sgi.com>
xfs: remove delwri buffer handling from xfs_buf_iorequest
We cannot ever reach xfs_buf_iorequest for a buffer with XBF_DELWRI set,
given that all write handlers make sure that the buffer is remove from
the delwri queue before, and we never do reads with the XBF_DELWRI flag
set (which the code would not handle correctly anyway).
Signed-off-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Dave Chinner <dchinner@redhat.com> Signed-off-by: Alex Elder <aelder@sgi.com>
Currently we always redirty an inode that was attempted to be written out
synchronously but has been cleaned by an AIL pushed internall, which is
rather bogus. Fix that by doing the i_update_core check early on and
return 0 for it. Also include async calls for it, as doing any work for
those is just as pointless. While we're at it also fix the sign for the
EIO return in case of a filesystem shutdown, and fix the completely
non-sensical locking around xfs_log_inode.
Signed-off-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Dave Chinner <dchinner@redhat.com> Signed-off-by: Alex Elder <aelder@sgi.com>
During umount we do not add a dirty inode to the lru and wait for it to
become clean first, but force writeback of data and metadata with
I_WILL_FREE set. Currently there is no way for XFS to detect that the
inode has been redirtied for metadata operations, as we skip the
mark_inode_dirty call during teardown. Fix this by setting i_update_core
nanually in that case, so that the inode gets flushed during inode reclaim.
Alternatively we could enable calling mark_inode_dirty for inodes in
I_WILL_FREE state, and let the VFS dirty tracking handle this. I decided
against this as we will get better I/O patterns from reclaim compared to
the synchronous writeout in write_inode_now, and always marking the inode
dirty in some way from xfs_mark_inode_dirty is a better safetly net in
either case.
Signed-off-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Dave Chinner <dchinner@redhat.com> Signed-off-by: Alex Elder <aelder@sgi.com>
Dave Chinner [Thu, 25 Aug 2011 07:17:02 +0000 (07:17 +0000)]
xfs: don't serialise adjacent concurrent direct IO appending writes
For append write workloads, extending the file requires a certain
amount of exclusive locking to be done up front to ensure sanity in
things like ensuring that we've zeroed any allocated regions
between the old EOF and the start of the new IO.
For single threads, this typically isn't a problem, and for large
IOs we don't serialise enough for it to be a problem for two
threads on really fast block devices. However for smaller IO and
larger thread counts we have a problem.
Take 4 concurrent sequential, single block sized and aligned IOs.
After the first IO is submitted but before it completes, we end up
with this state:
And the IO is done without exclusive locking because offset <=
ip->i_size. When we submit IO 2, we see offset > ip->i_size, and
grab the IO lock exclusive, because there is a chance we need to do
EOF zeroing. However, there is already an IO in progress that avoids
the need for IO zeroing because offset <= ip->i_new_size. hence we
could avoid holding the IO lock exlcusive for this. Hence after
submission of the second IO, we'd end up this state:
There is no need to grab the i_mutex of the IO lock in exclusive
mode if we don't need to invalidate the page cache. Taking these
locks on every direct IO effective serialises them as taking the IO
lock in exclusive mode has to wait for all shared holders to drop
the lock. That only happens when IO is complete, so effective it
prevents dispatch of concurrent direct IO writes to the same inode.
And so you can see that for the third concurrent IO, we'd avoid
exclusive locking for the same reason we avoided the exclusive lock
for the second IO.
Fixing this is a bit more complex than that, because we need to hold
a write-submission local value of ip->i_new_size to that clearing
the value is only done if no other thread has updated it before our
IO completes.....
Signed-off-by: Dave Chinner <dchinner@redhat.com> Signed-off-by: Alex Elder <aelder@sgi.com>
Dave Chinner [Thu, 25 Aug 2011 07:17:01 +0000 (07:17 +0000)]
xfs: don't serialise direct IO reads on page cache checks
There is no need to grab the i_mutex of the IO lock in exclusive
mode if we don't need to invalidate the page cache. Taking these
locks on every direct IO effective serialises them as taking the IO
lock in exclusive mode has to wait for all shared holders to drop
the lock. That only happens when IO is complete, so effective it
prevents dispatch of concurrent direct IO reads to the same inode.
Fix this by taking the IO lock shared to check the page cache state,
and only then drop it and take the IO lock exclusively if there is
work to be done. Hence for the normal direct IO case, no exclusive
locking will occur.
Signed-off-by: Dave Chinner <dchinner@redhat.com> Tested-by: Joern Engel <joern@logfs.org> Reviewed-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Alex Elder <aelder@sgi.com>
The code really requires the current source directory to be in the
header search path. We already do this if building with an object
tree separate from the source, but it needs to be added manually
if building inside the source. The cflags addition for it accidentally
got removed when collapsing the xfs directory structure.
Signed-off-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Dave Chinner <david@fromorbit.com> Signed-off-by: Alex Elder <aelder@sgi.com>
Use the move from Linux 2.6 to Linux 3.x as an excuse to kill the
annoying subdirectories in the XFS source code. Besides the large
amount of file rename the only changes are to the Makefile, a few
files including headers with the subdirectory prefix, and the binary
sysctl compat code that includes a header under fs/xfs/ from
kernel/.
Signed-off-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Alex Elder <aelder@sgi.com>
xfs: Check the return value of xfs_buf_read() for NULL
Check the return value of xfs_buf_read() for NULL and return ENOMEM
if it is NULL. This is necessary in a few spots to avoid subsequent
code blindly dereferencing the null buffer pointer.
Signed-off-by: Chandra Seetharaman <sekharan@us.ibm.com> Signed-off-by: Alex Elder <aelder@sgi.com>
Given that xfs_bwrite actually does the shutdown already after
waiting for the b_iodone completion and given that we actually
found that calling xfs_force_shutdown from inside
xfs_buf_iodone_callbacks was a major contributor the problem
it better to drop this call.
Signed-off-by: Ajeet Yadav <ajeet.yadav.77@gmail.com> Reviewed-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Alex Elder <aelder@sgi.com>
Alex Elder [Fri, 22 Jul 2011 16:04:41 +0000 (16:04 +0000)]
xfs: set cursor in xfs_ail_splice() even when AIL was empty
In xfs_ail_splice(), if a cursor is provided it is updated to
point to the last item on the list being spliced into the AIL.
But if the AIL was found to be empty, the cursor (if provided)
is just initialized instead.
There is no reason the empty AIL case needs to be treated any
differently. And treating it the same way allows this code
to be rearranged a bit, with a somewhat tidier result.
Signed-off-by: Alex Elder <aelder@sgi.com> Reviewed-by: Dave Chinner <dchinner@redhat.com>
Commit d006199e72a9 ("serial: sh-sci: Regtype probing doesn't need to be
fatal.") made sci_init_single() return when sci_probe_regmap() succeeds,
although it should return when sci_probe_regmap() fails. This causes
systems using the serial sh-sci driver to crash during boot.
Fix the problem by using the right return condition.
Signed-off-by: Rafael J. Wysocki <rjw@sisk.pl> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Linus Torvalds [Sun, 7 Aug 2011 21:07:03 +0000 (14:07 -0700)]
arm: remove "optimized" SHA1 routines
Since commit 1eb19a12bd22 ("lib/sha1: use the git implementation of
SHA-1"), the ARM SHA1 routines no longer work. The reason? They
depended on the larger 320-byte workspace, and now the sha1 workspace is
just 16 words (64 bytes). So the assembly version would overwrite the
stack randomly.
The optimized asm version is also probably slower than the new improved
C version, so there's no reason to keep it around. At least that was
the case in git, where what appears to be the same assembly language
version was removed two years ago because the optimized C BLK_SHA1 code
was faster.
Reported-and-tested-by: Joachim Eastwood <manabian@gmail.com> Cc: Andreas Schwab <schwab@linux-m68k.org> Cc: Nicolas Pitre <nico@fluxnic.net> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Al Viro [Sun, 7 Aug 2011 17:55:11 +0000 (18:55 +0100)]
fix rcu annotations noise in cred.h
task->cred is declared as __rcu, and access to other tasks' ->cred is,
indeed, protected. Access to current->cred does not need rcu_dereference()
at all, since only the task itself can change its ->cred. sparse, of
course, has no way of knowing that...
Add force-cast in current_cred(), make current_fsuid() et.al. use it.
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Linus Torvalds [Sun, 7 Aug 2011 16:53:20 +0000 (09:53 -0700)]
vfs: rename 'do_follow_link' to 'should_follow_link'
Al points out that the do_follow_link() helper function really is
misnamed - it's about whether we should try to follow a symlink or not,
not about actually doing the following.
Ari Savolainen [Sat, 6 Aug 2011 16:43:07 +0000 (19:43 +0300)]
Fix POSIX ACL permission check
After commit 3567866bf261: "RCUify freeing acls, let check_acl() go ahead in
RCU mode if acl is cached" posix_acl_permission is being called with an
unsupported flag and the permission check fails. This patch fixes the issue.
Signed-off-by: Ari Savolainen <ari.m.savolainen@gmail.com> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Linus Torvalds [Sun, 7 Aug 2011 05:56:03 +0000 (22:56 -0700)]
Merge branch 'for-linus' of git://git.open-osd.org/linux-open-osd
* 'for-linus' of git://git.open-osd.org/linux-open-osd:
ore: Make ore its own module
exofs: Rename raid engine from exofs/ios.c => ore
exofs: ios: Move to a per inode components & device-table
exofs: Move exofs specific osd operations out of ios.c
exofs: Add offset/length to exofs_get_io_state
exofs: Fix truncate for the raid-groups case
exofs: Small cleanup of exofs_fill_super
exofs: BUG: Avoid sbi realloc
exofs: Remove pnfs-osd private definitions
nfs_xdr: Move nfs4_string definition out of #ifdef CONFIG_NFS_V4
Linus Torvalds [Sun, 7 Aug 2011 05:45:50 +0000 (22:45 -0700)]
vfs: optimize inode cache access patterns
The inode structure layout is largely random, and some of the vfs paths
really do care. The path lookup in particular is already quite D$
intensive, and profiles show that accessing the 'inode->i_op->xyz'
fields is quite costly.
We already optimized the dcache to not unnecessarily load the d_op
structure for members that are often NULL using the DCACHE_OP_xyz bits
in dentry->d_flags, and this does something very similar for the inode
ops that are used during pathname lookup.
It also re-orders the fields so that the fields accessed by 'stat' are
together at the beginning of the inode structure, and roughly in the
order accessed.
The effect of this seems to be in the 1-2% range for an empty kernel
"make -j" run (which is fairly kernel-intensive, mostly in filename
lookup), so it's visible. The numbers are fairly noisy, though, and
likely depend a lot on exact microarchitecture. So there's more tuning
to be done.
Linus Torvalds [Sun, 7 Aug 2011 05:41:50 +0000 (22:41 -0700)]
vfs: renumber DCACHE_xyz flags, remove some stale ones
Gcc tends to generate better code with small integers, including the
DCACHE_xyz flag tests - so move the common ones to be first in the list.
Also just remove the unused DCACHE_INOTIFY_PARENT_WATCHED and
DCACHE_AUTOFS_PENDING values, their users no longer exists in the source
tree.
And add a "unlikely()" to the DCACHE_OP_COMPARE test, since we want the
common case to be a nice straight-line fall-through.
* git://git.kernel.org/pub/scm/linux/kernel/git/davem/net:
net: Compute protocol sequence numbers and fragment IDs using MD5.
crypto: Move md5_transform to lib/md5.c
Boaz Harrosh [Sun, 7 Aug 2011 02:26:31 +0000 (19:26 -0700)]
exofs: Rename raid engine from exofs/ios.c => ore
ORE stands for "Objects Raid Engine"
This patch is a mechanical rename of everything that was in ios.c
and its API declaration to an ore.c and an osd_ore.h header. The ore
engine will later be used by the pnfs objects layout driver.
* File ios.c => ore.c
* Declaration of types and API are moved from exofs.h to a new
osd_ore.h
* All used types are prefixed by ore_ from their exofs_ name.
* Shift includes from exofs.h to osd_ore.h so osd_ore.h is
independent, include it from exofs.h.
Other than a pure rename there are no other changes. Next patch
will move the ore into it's own module and will export the API
to be used by exofs and later the layout driver
Boaz Harrosh [Fri, 5 Aug 2011 22:06:04 +0000 (15:06 -0700)]
exofs: ios: Move to a per inode components & device-table
Exofs raid engine was saving on memory space by having a single layout-info,
single pid, and a single device-table, global to the filesystem. Then passing
a credential and object_id info at the io_state level, private for each
inode. It would also devise this contraption of rotating the device table
view for each inode->ino to spread out the device usage.
This is not compatible with the pnfs-objects standard, demanding that
each inode can have it's own layout-info, device-table, and each object
component it's own pid, oid and creds.
So: Bring exofs raid engine to be usable for generic pnfs-objects use by:
* Define an exofs_comp structure that holds obj_id and credential info.
* Break up exofs_layout struct to an exofs_components structure that holds a
possible array of exofs_comp and the array of devices + the size of the
arrays.
* Add a "comps" parameter to get_io_state() that specifies the ids creds
and device array to use for each IO.
This enables to keep the layout global, but the device-table view, creds
and IDs at the inode level. It only adds two 64bit to each inode, since
some of these members already existed in another form.
* ios raid engine now access layout-info and comps-info through the passed
pointers. Everything is pre-prepared by caller for generic access of
these structures and arrays.
At the exofs Level:
* Super block holds an exofs_components struct that holds the device
array, previously in layout. The devices there are in device-table
order. The device-array is twice bigger and repeats the device-table
twice so now each inode's device array can point to a random device
and have a round-robin view of the table, making it compatible to
previous exofs versions.
* Each inode has an exofs_components struct that is initialized at
load time, with it's own view of the device table IDs and creds.
When doing IO this gets passed to the io_state together with the
layout.
While preforming this change. Bugs where found where credentials with the
wrong IDs where used to access the different SB objects (super.c). As well
as some dead code. It was never noticed because the target we use does not
check the credentials.
Boaz Harrosh [Tue, 16 Nov 2010 18:09:58 +0000 (20:09 +0200)]
exofs: Add offset/length to exofs_get_io_state
In future raid code we will need to know the IO offset/length
and if it's a read or write to determine some of the array
sizes we'll need.
So add a new exofs_get_rw_state() API for use when
writeing/reading. All other simple cases are left using the
old way.
The major change to this is that now we need to call
exofs_get_io_state later at inode.c::read_exec and
inode.c::write_exec when we actually know these things. So this
patch is kept separate so I can test things apart from other
changes.
David S. Miller [Thu, 4 Aug 2011 03:50:44 +0000 (20:50 -0700)]
net: Compute protocol sequence numbers and fragment IDs using MD5.
Computers have become a lot faster since we compromised on the
partial MD4 hash which we use currently for performance reasons.
MD5 is a much safer choice, and is inline with both RFC1948 and
other ISS generators (OpenBSD, Solaris, etc.)
Furthermore, only having 24-bits of the sequence number be truly
unpredictable is a very serious limitation. So the periodic
regeneration and 8-bit counter have been removed. We compute and
use a full 32-bit sequence number.
For ipv6, DCCP was found to use a 32-bit truncated initial sequence
number (it needs 43-bits) and that is fixed here as well.
Reported-by: Dan Kaminsky <dan@doxpara.com> Tested-by: Willy Tarreau <w@1wt.eu> Signed-off-by: David S. Miller <davem@davemloft.net>
* git://git.kernel.org/pub/scm/linux/kernel/git/sfrench/cifs-2.6:
cifs: cope with negative dentries in cifs_get_root
cifs: convert prefixpath delimiters in cifs_build_path_to_root
CIFS: Fix missing a decrement of inFlight value
cifs: demote DFS referral lookup errors to cFYI
Revert "cifs: advertise the right receive buffer size to the server"
Linus Torvalds [Sat, 6 Aug 2011 19:22:30 +0000 (12:22 -0700)]
Merge branch 'stable/bug.fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/konrad/xen
* 'stable/bug.fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/konrad/xen:
xen/trace: Fix compile error when CONFIG_XEN_PRIVILEGED_GUEST is not set
xen: Fix misleading WARN message at xen_release_chunk
xen: Fix printk() format in xen/setup.c
xen/tracing: it looks like we wanted CONFIG_FTRACE
xen/self-balloon: Add dependency on tmem.
xen/balloon: Fix compile errors - missing header files.
xen/grant: Fix compile warning.
xen/pciback: remove duplicated #include
John Stanley [Thu, 4 Aug 2011 00:41:00 +0000 (20:41 -0400)]
savagedb: Fix typo causing regression in savage4 series video chip detection
Two additional savage4 variants were added, but the S3_SAVAGE4_SERIES
macro was incompletely modified, resulting in a false positive detection
of a savage4 card regardless of which savage card is actually present.
For non-savage4 series cards, such as a Savage/IX-MV card, this results
in garbled video and/or a hard-hang at boot time. Fix this by changing
an '||' to an '&&' in the S3_SAVAGE4_SERIES macro.
Signed-off-by: John P. Stanley <jpsinthemix@verizon.net> Reviewed-by: Tormod Volden <debian.tormod@gmail.com>
[ The macros have incomplete parenthesis too, but whatever .. -Linus ] Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Josh Triplett [Wed, 3 Aug 2011 19:19:07 +0000 (12:19 -0700)]
CodingStyle: Document the exception of not splitting user-visible strings, for grepping
Patch reviewers now recommend not splitting long user-visible strings,
such as printk messages, even if they exceed 80 columns. This avoids
breaking grep. However, that recommendation did not actually appear
anywhere in Documentation/CodingStyle.
See, for example, the thread at
http://news.gmane.org/find-root.php?message_id=%3c1312215262.11635.15.camel%40Joe%2dLaptop%3e
Linus Torvalds [Sat, 6 Aug 2011 18:51:33 +0000 (11:51 -0700)]
vfs: show O_CLOEXE bit properly in /proc/<pid>/fdinfo/<fd> files
The CLOEXE bit is magical, and for performance (and semantic) reasons we
don't actually maintain it in the file descriptor itself, but in a
separate bit array. Which means that when we show f_flags, the CLOEXE
status is shown incorrectly: we show the status not as it is now, but as
it was when the file was opened.
Fix that by looking up the bit properly in the 'fdt->close_on_exec' bit
array.
Uli needs this in order to re-implement the pfiles program:
"For normal file descriptors (not sockets) this was the last piece of
information which wasn't available. This is all part of my 'give
Solaris users no reason to not switch' effort. I intend to offer the
code to the util-linux-ng maintainers."
Linus Torvalds [Sat, 6 Aug 2011 18:43:08 +0000 (11:43 -0700)]
oom_ajd: don't use WARN_ONCE, just use printk_once
WARN_ONCE() is very annoying, in that it shows the stack trace that we
don't care about at all, and also triggers various user-level "kernel
oopsed" logic that we really don't care about. And it's not like the
user can do anything about the applications (sshd) in question, it's a
distro issue.
For ChromiumOS, we use SHA-1 to verify the integrity of the root
filesystem. The speed of the kernel sha-1 implementation has a major
impact on our boot performance.
To improve boot performance, we investigated using the heavily optimized
sha-1 implementation used in git. With the git sha-1 implementation, we
see a 11.7% improvement in boot time.
10 reboots, remove slowest/fastest.
Before:
Mean: 6.58 seconds Stdev: 0.14
After (with git sha-1, this patch):
Mean: 5.89 seconds Stdev: 0.07
The other cool thing about the git SHA-1 implementation is that it only
needs 64 bytes of stack for the workspace while the original kernel
implementation needed 320 bytes.
Signed-off-by: Mandeep Singh Baines <msb@chromium.org> Cc: Ramsay Jones <ramsay@ramsay1.demon.co.uk> Cc: Nicolas Pitre <nico@cam.org> Cc: Herbert Xu <herbert@gondor.apana.org.au> Cc: David S. Miller <davem@davemloft.net> Cc: linux-crypto@vger.kernel.org Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
projects / karo-tx-linux.git / log