While debugging timeouts happening in my application workload (ScyllaDB), I have
observed calls to open() taking a long time, ranging everywhere from 2 seconds -
the first ones that are enough to time out my application - to more than 30
seconds.
The problem seems to happen because XFS may block on pending metadata updates
under certain circumnstances, and that's confirmed with the following backtrace
taken by the offcputime tool (iovisor/bcc):
Inspecting my run with blktrace, I can see that the xfsaild kthread exhibit very
high "Dispatch wait" times, on the dozens of seconds range and consistent with
the open() times I have saw in that run.
Still from the blktrace output, we can after searching a bit, identify the
request that wasn't dispatched:
8,0 11 152 81.092472813 804 A WM 141698288 + 8 <- (8,1) 141696240
8,0 11 153 81.092472889 804 Q WM 141698288 + 8 [xfsaild/sda1]
8,0 11 154 81.092473207 804 G WM 141698288 + 8 [xfsaild/sda1]
8,0 11 206 81.092496118 804 I WM 141698288 + 8 ( 22911) [xfsaild/sda1]
<==== 'I' means Inserted (into the IO scheduler) ===================================>
8,0 0 289372 96.718761435 0 D WM 141698288 + 8 (15626265317) [swapper/0]
<==== Only 15s later the CFQ scheduler dispatches the request ======================>
As we can see above, in this particular example CFQ took 15 seconds to dispatch
this request. Going back to the full trace, we can see that the xfsaild queue
had plenty of opportunity to run, and it was selected as the active queue many
times. It would just always be preempted by something else (example):
8,0 1 0 81.117912979 0 m N cfq1618SN / insert_request
8,0 1 0 81.117913419 0 m N cfq1618SN / add_to_rr
8,0 1 0 81.117914044 0 m N cfq1618SN / preempt
8,0 1 0 81.117914398 0 m N cfq767A / slice expired t=1
8,0 1 0 81.117914755 0 m N cfq767A / resid=40
8,0 1 0 81.117915340 0 m N / served: vt=1948520448 min_vt=1948520448
8,0 1 0 81.117915858 0 m N cfq767A / sl_used=1 disp=0 charge=0 iops=1 sect=0
where cfq767 is the xfsaild queue and cfq1618 corresponds to one of the ScyllaDB
IO dispatchers.
The requests preempting the xfsaild queue are synchronous requests. That's a
characteristic of ScyllaDB workloads, as we only ever issue O_DIRECT requests.
While it can be argued that preempting ASYNC requests in favor of SYNC is part
of the CFQ logic, I don't believe that doing so for 15+ seconds is anyone's
goal.
Moreover, unless I am misunderstanding something, that breaks the expectation
set by the "fifo_expire_async" tunable, which in my system is set to the
default.
Looking at the code, it seems to me that the issue is that after we make
an async queue active, there is no guarantee that it will execute any request.
When the queue itself tests if it cfq_may_dispatch() it can bail if it sees SYNC
requests in flight. An incoming request from another queue can also preempt it
in such situation before we have the chance to execute anything (as seen in the
trace above).
This patch sets the must_dispatch flag if we notice that we have requests
that are already fifo_expired. This flag is always cleared after
cfq_dispatch_request() returns from cfq_dispatch_requests(), so it won't pin
the queue for subsequent requests (unless they are themselves expired)
Care is taken during preempt to still allow rt requests to preempt us
regardless.
Testing my workload with this patch applied produces much better results.
From the application side I see no timeouts, and the open() latency histogram
generated by systemtap looks much better, with the worst outlier at 131ms:
An NULL-pointer dereference happens in cachefiles_mark_object_inactive()
when it tries to read i_blocks so that it can tell the cachefilesd daemon
how much space it's making available.
The problem is that cachefiles_drop_object() calls
cachefiles_mark_object_inactive() after calling cachefiles_delete_object()
because the object being marked active staves off attempts to (re-)use the
file at that filename until after it has been deleted. This means that
d_inode is NULL by the time we come to try to access it.
To fix the problem, have the caller of cachefiles_mark_object_inactive()
supply the number of blocks freed up.
This fixes a bug where the permission was not properly checked in
overlayfs. The testcase is ltp/utimensat01.
It is also cleaner and safer to do the permission checking in the vfs
helper instead of the caller.
This patch introduces an additional ia_valid flag ATTR_TOUCH (since
touch(1) is the most obvious user of utimes(NULL)) that is passed into
notify_change whenever the conditions for this special permission checking
mode are met.
After backporting commit ee44b4bc054a ("dlm: use sctp 1-to-1 API")
series to a kernel with an older workqueue which didn't use RCU yet, it
was noticed that we are freeing the workqueues in dlm_lowcomms_stop()
too early as free_conn() will try to access that memory for canceling
the queued works if any.
This issue was introduced by commit 0d737a8cfd83 as before it such
attempt to cancel the queued works wasn't performed, so the issue was
not present.
This patch fixes it by simply inverting the free order.
This patch changes the p8_ghash driver to use ghash-generic as a fixed
fallback implementation. This allows the correct value of descsize to be
defined directly in its shash_alg structure and avoids problems with
incorrect buffer sizes when its state is exported or imported.
When zeroing blocks for DAX allocations, we also have to unmap aliases
in the block device mappings. Otherwise writeback can overwrite zeros
with stale data from block device page cache.
Signed-off-by: Jan Kara <email address hidden>
Signed-off-by: Theodore Ts'o <email address hidden>
Signed-off-by: Greg Kroah-Hartman <email address hidden>
Currently when doing a DAX hole punch with ext4 we fail to do a writeback.
This is because the logic around filemap_write_and_wait_range() in
ext4_punch_hole() only looks for dirty page cache pages in the radix tree,
not for dirty DAX exceptional entries.
Signed-off-by: Ross Zwisler <email address hidden>
Reviewed-by: Jan Kara <email address hidden>
Signed-off-by: Theodore Ts'o <email address hidden>
Signed-off-by: Greg Kroah-Hartman <email address hidden>