Commit Graph

224 Commits

Author SHA1 Message Date
Brian Behlendorf 500e95c884 Revert "Disable vmalloc() direct reclaim"
This reverts commit 2092cf68d8.  The
use of the PF_MEMALLOC flag was always a hack to work around memory
reclaim deadlocks.  Those issues are believed to be resolved so this
workaround can be safely reverted.

Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2012-08-27 12:00:55 -07:00
Brian Behlendorf 617f79de6a Revert "Fix NULL deref in balance_pgdat()"
This reverts commit b8b6e4c453.  The
use of the PF_MEMALLOC flag was always a hack to work around memory
reclaim deadlocks.  Those issues are believed to be resolved so this
workaround can be safely reverted.

Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2012-08-27 12:00:55 -07:00
Brian Behlendorf bc03e07a7c Revert "Detect kernels that honor gfp flags passed to vmalloc()"
This reverts commit 36811b4430.
Which is no longer required because there is now SPL code in
place to safely handle the deadlocks the kernel patch was designed
to address.  Therefore we can unconditionally use vmalloc() and
drop all the PF_MEMALLOC code.

Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2012-08-27 12:00:55 -07:00
Brian Behlendorf d47e664ad4 Revert "Add TASKQ_NORECLAIM flag"
This reverts commit 372c257233.  The
use of the PF_MEMALLOC flag was always a hack to work around memory
reclaim deadlocks.  Those issues are believed to be resolved so this
workaround can be safely reverted.

Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2012-08-27 12:00:42 -07:00
Brian Behlendorf e2dcc6e2b8 Emergency slab objects
This patch is designed to resolve a deadlock which can occur with
__vmalloc() based slabs.  The issue is that the Linux kernel does
not honor the flags passed to __vmalloc().  This makes it unsafe
to use in a writeback context.  Unfortunately, this is a use case
ZFS depends on for correct operation.

Fixing this issue in the upstream kernel was pursued and patches
are available which resolve the issue.

  https://bugs.gentoo.org/show_bug.cgi?id=416685

However, these changes were rejected because upstream felt that
using __vmalloc() in the context of writeback should never be done.
Their solution was for us to rewrite parts of ZFS to accomidate
the Linux VM.

While that is probably the right long term solution, and it is
something we want to pursue, it is not a trivial task and will
likely destabilize the existing code.  This work has been planned
for the 0.7.0 release but in the meanwhile we want to improve the
SPL slab implementation to accomidate this expected ZFS usage.

This is accomplished by performing the __vmalloc() asynchronously
in the context of a work queue.  This doesn't prevent the posibility
of the worker thread from deadlocking.  However, the caller can now
safely block on a wait queue for the slab allocation to complete.

Normally this will occur in a reasonable amount of time and the
caller will be woken up when the new slab is available,.  The objects
will then get cached in the per-cpu magazines and everything will
proceed as usual.

However, if the __vmalloc() deadlocks for the reasons described
above, or is just very slow, then the callers on the wait queues
will timeout out.  When this rare situation occurs they will attempt
to kmalloc() a single minimally sized object using the GFP_NOIO flags.
This allocation will not deadlock because kmalloc() will honor the
passed flags and the caller will be able to make forward progress.

As long as forward progress can be maintained then even if the
worker thread is deadlocked the critical thread will make progress.
This will eventually allow the deadlocked worker thread to complete
and normal operation will resume.

These emergency allocations will likely be slow since they require
contiguous pages.  However, their use should be rare so the impact
is expected to be minimal.  If that turns out not to be the case in
practice further optimizations are possible.

One additional concern is if these emergency objects are long lived.
Right now they are simply tracked on a list which must be walked when
an object is freed.  Is they accumulate on a system and the list
grows freeing objects will become more expensive.  This could be
handled relatively easily by using a hash instead of a list, but that
optimization (if needed) is left for a follow up patch.

Additionally, these emeregency objects could be repacked in to existing
slabs as objects are freed if the kmem_cache_set_move() functionality
was implemented.  See issue https://github.com/zfsonlinux/spl/issues/26
for full details.  This work would also help reduce ZFS's memory
fragmentation problems.

The /proc/spl/kmem/slab file has had two new columns added at the
end.  The 'emerg' column reports the current number of these emergency
objects in use for the cache, and the following 'max' column shows
the historical worst case.  These value should give us a good idea
of how often these objects are needed.  Based on these values under
real use cases we can tune the default behavior.

Lastly, as a side benefit using a single work queue for the slab
allocations should reduce cpu contention on the global virtual address
space lock.   This should manifest itself as reduced cpu usage for
the system.

Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2012-08-27 12:00:42 -07:00
Prakash Surya 08850eddcb Avoid calling smp_processor_id in spl_magazine_age
The spl_magazine_age function had the implied assumption that it will
remain on its current cpu through its execution. In order to support
preempt enabled kernels, this assumption had to be removed.

The spl_kmem_magazine structure now holds the cpu id of the cpu it is
local to. This allows spl_magazine_age to use this field when scheduling
work to be done by the magazine's local cpu.

Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Issue #98
2012-08-24 09:43:22 -07:00
Richard Yao 15d0411297 Remove Makefile from non-toplevel .gitignore files
When building SPL support into the kernel, ./copy-builtin will copy
non-toplevel .gitignore files. These files list /Makefile, which causes
git-archive to omit ./module/{spl,splat}/Makefile. The absence of these
files result in build failures when SPL is selected. ZFS is unaffected
because it puts Makefile in the toplevel .gitignore, which is not
copied. We fix SPL by emulating that behavior.

Reported-by: Fabio Erculiani <lxnay@gentoo.org>
Signed-off-by: Richard Yao <ryao@cs.stonybrook.edu>
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Closes #152
2012-08-23 12:49:04 -07:00
Prakash Surya 9baf44bc17 Wrap trace_set_debug_header in trace_[get|put]_tcd
To properly support CONFIG_PREEMPT enabled kernels, we must refrain from
using a CPU index when preemption is enabled. As a result, this change
moves the trace_set_debug_header call (which calls smp_processor_id)
within trace_get_tcd and trace_put_tcd (which disable and enable
preemption respectively).

Signed-off-by: Prakash Surya <surya1@llnl.gov>
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Closes #160
2012-08-23 10:01:20 -07:00
Richard Yao 6576a1a70d Fix incorrect type in spl_kmem_cache_set_move() parameter
A preprocessor definition renders this harmless. However, it is a good
idea to change this to be consistent.

Signed-off-by: Richard Yao <ryao@cs.stonybrook.edu>
2012-08-01 16:35:18 -07:00
Etienne Dechamps a9f2397ee9 Determine the hostid on demand.
Currently, the SPL tries to determine the hostid at module load. The
hostid is usually determined by running the userland program "hostid"
during module initialization.

Unfortunately, when the module initializes, it may be way too soon to be
able to run any userland programs. This is especially true when the
module is compiled directly inside the kernel (built-in); in that case,
the SPL would try to run hostid when the kernel is still initializing,
which of course is doomed to fail.

This patch fixes the issue by deferring hostid generation until
something actually needs the hostid (that is, when zone_get_hostid() is
called), thus switching to a "on-initialization" model to a "on-demand"
(lazy loading) model. ZFS only needs the hostid when some pool
operations are requested, and this always happens way after the kernel
has finished initialization, thus solving the problem.

Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Issue zfsonlinux/zfs#851
2012-07-26 15:14:02 -07:00
Etienne Dechamps c167aadb27 Add script for builtin module building.
This commit introduces a "copy-builtin" script designed to prepare a
kernel source tree for building SPL as a builtin module. The script
makes a full copy of all needed files, thus making the kernel source
tree fully independent of the spl source package.

To achieve that, some compilation flags (-include, -I) have been moved
to module/Makefile. This Makefile is only used when compiling external
modules; when compiling builtin modules, a Kbuild file generated by the
configure-builtin script is used instead. This makes sure Makefiles
inside the kernel source tree does not contain references to the spl
source package.

Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Issue zfsonlinux/zfs#851
2012-07-26 15:13:09 -07:00
Etienne Dechamps 38b5ff4d07 Fix undefined reference on spl_mutex_spin_max().
Commit 3160d4f56b changed the set of
conditions under which spl_mutex_spin_max would be implemented as a
function by changing an #if in sys/mutex.h. The corresponding
implementation file spl-mutex.c, however, has not been updated to
reflect the change. This results in undefined reference errors on
spl_mutex_spin_max under the following condition:

((!CONFIG_SMP || CONFIG_DEBUG_MUTEXES) && HAVE_MUTEX_OWNER && HAVE_TASK_CURR)

This patch fixes the issue by using the same #if in sys/mutex.h and
spl-mutex.c.

Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Issue zfsonlinux/zfs#851
2012-07-26 14:54:53 -07:00
Etienne Dechamps 94aac9c9bc Use MODULE variable in module Makefile like zfs.
In zfs, each module Makefile contains a MODULE variable which contains
the name of the module, and the following declarations reference this
variable.

In spl, there is a MODULES variable which is never used. Rename it to
MODULE and use it like in zfs. This improves consistency between the two
build systems.

Signed-off-by: Prakash Surya <surya1@llnl.gov>
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Issue zfsonlinux/zfs#851
2012-07-26 14:53:48 -07:00
Brian Behlendorf e8267acd25 32-bit compat, hostid_read()
Explicitly cast the sizeof in hostid_read() to prevent the
following compiler warning on 32-bit systems.

  module/spl/spl-generic.c:490:10: error: format '%lu' expects
  argument of type 'long unsigned int', but argument 4 has type
  'unsigned int' [-Werror=format]

Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2012-07-20 11:14:04 -07:00
Richard Yao 36811b4430 Detect kernels that honor gfp flags passed to vmalloc()
zfsonlinux/spl@2092cf68d8 used
PF_MEMALLOC to workaround a bug in the Linux kernel where
allocations did not honor the gfp flags passed to vmalloc().
Unfortunately, PF_MEMALLOC has the side effect of permitting
allocations to allocate pages outside of ZONE_NORMAL. This
has been observed to result in the depletion of ZONE_DMA32.

A kernel patch is available in the Gentoo bug tracker for
this issue.

  https://bugs.gentoo.org/show_bug.cgi?id=416685

This negates any benefit PF_MEMALLOC provides, so we introduce
an autotools check to disable the use of PF_MEMALLOC on
systems with patched kernels.

Signed-off-by: Richard Yao <ryao@cs.stonybrook.edu>
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Closes #126
2012-07-11 11:44:27 -07:00
Richard Yao 973e8269bd Constify memory management functions
This prevents warnings in ZFS that were caused by changes necessary to
support PaX patched kernels. When debugging is enabled, these warnings
become build failures.

Signed-off-by: Richard Yao <ryao@cs.stonybrook.edu>
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Closes #131
2012-07-03 16:07:27 -07:00
Brian Behlendorf 44e406d712 PowerPC Compatibility
Usage of get_current() is not supported across all architectures.
The correct interface to use is the '#define current' which will
map to the appropriate function, usually current_thread_info().

Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Closes #119
2012-07-02 09:33:09 -07:00
Richard Yao e0093fea58 Linux 3.4 compat, __clear_close_on_exec replaces FD_CLR
torvalds/linux@1dce27c5aa introduced
__clear_close_on_exec() as a replacement for FD_CLR. Further commits
appear to have removed FD_CLR from the Linux source tree.  This
causes the following failure:

  error: implicit declaration of function '__FD_CLR'
  [-Werror=implicit-function-declaration]

To correct this we update the code to use the current
__clear_close_on_exec() interface for readability.  Then we introduce
an autotools check to determine if __clear_close_on_exec() is available.
If it isn't then we define some compatibility logic which used the older
FD_CLR() interface.

Signed-off-by: Richard Yao <ryao@gentoo.org>
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Closes #124
2012-06-13 16:18:51 -07:00
Brian Behlendorf eaac9ba510 Fix uninit variable in slab reclaim test
Gcc version 4.7.0 reports the delta.tv_sec in the slab reclaim test
as potentially unitialized.  In practice this will never occur but
to keep gcc happy we initialize the variable to zero.

Signed-off-by: Brian Behlendorf <behlendo@fedora-17-amd64.(none)>
2012-06-13 16:17:22 -07:00
Brian Behlendorf 2371321e8a Fix invalid context bug
In the module unload path the vm_file_cache was being destroyed
under a spin lock.  Because this operation might sleep it was
possible, although very very unlikely, that this could result
in a deadlock.

This issue was indentified by using a Linux debug kernel and
has been fixed by moving the kmem_cache_destroy() out from under
the spin lock.  There is no need to lock this operation here.

Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Closes zfsonlinux/zfs#771
2012-06-11 09:17:45 -07:00
Jorgen Lundman 93b0dc92ea Fix ARM 64-bit division
Correctly implementating 64-bit division for ARM requires more than
just providing the __aeabi_uldivmod() and __aeabi_ldivmod() symbols.
They are need to be implemented is such a way that the quotient and
remainder and left in specific registers after the division operation
completes.  This change updates the wrapper functions to accomplish
this according to the official ARM Run-time ABI.

Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Closes zfsonlinux/zfs#706
2012-05-22 09:27:11 -07:00
Brian Behlendorf 38d31a1e57 Remove Solaris module emulation
Originally I believed that these interfaces would be needed.
However, in practice it turned out that it was more straight
forward and maintainable to use the native Linux interfaces.
As such, this is all dead code and can be safely removed.

Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Closes #109
2012-05-18 13:57:44 -07:00
Prakash Surya a9a7a01cf5 Add SPLAT test to exercise slab direct reclaim
This test is designed to verify that direct reclaim is functioning as
expected.  We allocate a large number of objects thus creating a large
number of slabs.  We then apply memory pressure and expect that the
direct reclaim path can easily recover those slabs.  The registered
reclaim function will free the objects and the slab shrinker will call
it repeatedly until at least a single slab can be freed.

Note it may not be possible to reclaim every last slab via direct reclaim
without a failure because the shrinker_rwsem may be contended.  For this
reason, quickly reclaiming 3/4 of the slabs is considered a success.

This should all be possible within 10 seconds.  For reference, on a
system with 2G of memory this test takes roughly 0.2 seconds to run.
It may take longer on larger memory systems but should still easily
complete in the alloted 10 seconds.

Signed-off-by: Prakash Surya <surya1@llnl.gov>
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Closes #107
2012-05-07 11:55:59 -07:00
Brian Behlendorf b78d4b9d98 Ensure a minimum of one slab is reclaimed
To minimize the chance of triggering an OOM during direct reclaim.
The kmem caches have been improved to make a best effort to reclaim
at least one slab when a reclaim function is registered.  This helps
avoid the case where objects are released but they are spread over
multiple slabs so no memory gets reclaimed.

Care has been taken to avoid deadlocking if the reclaim function
is unable to make forward progress.  Additionally, the reclaim
function may be skipped entirely if there are already free slabs
which can be safely reaped.

Signed-off-by: Prakash Surya <surya1@llnl.gov>
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Closes #107
2012-05-07 11:54:28 -07:00
Brian Behlendorf 06089b9e19 Ensure direct reclaim forward progress
The Linux direct reclaim path uses this out of band value to
determine if forward progress is being made.  Normally this is
incremented by kmem_freepages() which is part of the various
Linux slab implementations.  However, since we are using none
of that infrastructure we're responsible for incrementing this
count.

If no forward progress is detected and a subsequent allocation
fails the OOM killer will be invoked.  If there was forward
progress additional reclaim will be attempted via the page
cache and registerd shrinker until the allocation succeeds.

Signed-off-by: Prakash Surya <surya1@llnl.gov>
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Closes #107
2012-05-07 11:54:19 -07:00
Prakash Surya c0e0fc14e3 Ignore slab cache age and delay in direct reclaim
When memory pressure triggers direct memory reclaim, a slabs age
and delay should not prevent it from being freed. This patch ensures
these values are ignored, allowing an empty slab to be freed in this
code path no matter the value of its age and delay.

This prevents needless scanning of the partial slabs and has been
observed to significantly reduce the total cpu usage.  In addition,
it should allow for snappier reclaim under memory pressure.

Signed-off-by: Prakash Surya <surya1@llnl.gov>
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Closes #102
2012-05-07 11:50:04 -07:00
Prakash Surya cef7605c34 Throttle number of freed slabs based on nr_to_scan
Previously, the SPL tried to maintain Solaris semantics by freeing
all available (empty) slabs from its slab caches when the shrinker
was called. This is not desirable when running on Linux. To make
the SPL shrinker more Linux friendly, the actual number of freed
slabs from each of the slab caches is now derived from nr_to_scan
and skc_slab_objs.

Additionally, an accounting bug was fixed in spl_slab_reclaim()
which could cause us to reclaim one more slab than requested.

Signed-off-by: Prakash Surya <surya1@llnl.gov>
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Closes #101
2012-05-07 11:46:15 -07:00
Jorgen Lundman ef6f91ce0c Add missing 64-bit divide for 32-bit ARM
Leverage the existing generic 64-bit division operations which
were originally implemented for x86 to support ARM.  All that is
required is to make the symbols available to the linker with the
expected names.

Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2012-05-03 10:07:54 -07:00
Prakash Surya 05b8f50c33 Update a comment to reflect new taskq internals
As of the removal of the taskq work list made in commit:

    commit 2c02b71b14
    Author: Prakash Surya <surya1@llnl.gov>
    Date:   Mon Dec 5 17:32:48 2011 -0800

        Replace tq_work_list and tq_threads in taskq_t

        To lay the ground work for introducing the taskq_dispatch_prealloc()
        interface, the tq_work_list and tq_threads fields had to be replaced
        with new alternatives in the taskq_t structure.

the comment above taskq_wait_check has been incorrect. This change is an
attempt at bringing that description more in line with the current
implementation. Essentially, references to the old task work list had to
be updated to reference the new taskq thread active list.

Signed-off-by: Prakash Surya <surya1@llnl.gov>
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Issue #65
2012-04-30 10:49:15 -07:00
Brian Behlendorf b29012b999 Remove condition variable names
Long ago I added support to the spl for condition variable names
because I thought they might be needed.  It turns out they aren't.
In fact the official Solaris cv_init(9F) man page discourages
their use in the kernel.

  cv_init(9F)
    Parameters
      name - Descriptive string. This is obsolete and should be
             NULL. (Non-NULL strings are legal, but they're a
             waste of kernel memory.)

Therefore, I'm removing them from the spl to reclaim this memory
and adding an ASSERT() to ensure no new consumers are added which
make use of the name.

Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2012-04-06 12:06:19 -07:00
Brian Behlendorf 0835057ee7 Add SPL_META_RELEASE to module load/unload messages
Include the ZFS_META_RELEASE in the module load/unload messages
to more clearly indicate exactly what version of the SPL has
been loaded.

Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2012-03-23 12:11:50 -07:00
Brian Behlendorf 9a8b7a7458 Add basic dynamic kstat support
Add the bare minimum functionality to support dynamic kstats.  A
complete kstat implementation should be done as part of issue #84.

Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Issue #84
2012-02-02 11:28:00 -08:00
Brian Behlendorf 4b2220f0b9 Add --enable-debug-log configure option
Until now the notion of an internal debug logging infrastructure
was conflated with enabling ASSERT()s.  This patch clarifies things
by cleanly breaking the two subsystem apart.  The result of this
is the following behavior.

--enable-debug      - Enable/disable code wrapped in ASSERT()s.
--disable-debug       ASSERT()s are used to check invariants and
                      are never required for correct operation.
                      They are disabled by default because they
                      may impact performance.

--enable-debug-log  - Enable/disable the debug log infrastructure.
--disable-debug-log   This infrastructure allows the spl code and
                      its consumer to log messages to an in-kernel
                      log.  The granularity of the logging can be
                      controlled by a debug mask.  By default the
                      mask disables most debug messages resulting
                      in a negligible performance impact.  Because
                      of this the debug log is enabled by default.

Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2012-02-02 11:27:54 -08:00
Ned Bass 3c6ed5410b Taskq locking optimizations
Testing has shown that tq->tq_lock can be highly contended when a
large number of small work items are dispatched.  The lock hold time
is reduced by the following changes:

1) Use exclusive threads in the work_waitq

When a single work item is dispatched we only need to wake a single
thread to service it.  The current implementation uses non-exclusive
threads so all threads are woken when the dispatcher calls wake_up().
If a large number of threads are in the queue this overhead can become
non-negligible.

2) Conditionally add/remove threads from work waitq

Taskq threads need only add themselves to the work wait queue if
there are no pending work items.

Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Issue #32
2012-01-19 14:42:49 -08:00
Ned Bass 0bb43ca282 Revert "Taskq locking optimizations"
This reverts commit ec2b41049f.

A race condition was introduced by which a wake_up() call can be lost
after the taskq thread determines there is no pending work items,
leading to deadlock:

1. taksq thread enables interrupts
2. dispatcher thread runs, queues work item, call wake_up()
3. taskq thread runs, adds self to waitq, sleeps

This could easily happen if an interrupt for an IO completion was
outstanding at the point where the taskq thread reenables interrupts,
just before the call to add_wait_queue_exclusive().  The handler would
run immediately within the race window.

Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Issue #32
2012-01-19 14:42:39 -08:00
Ned Bass ec2b41049f Taskq locking optimizations
Testing has shown that tq->tq_lock can be highly contended when a
large number of small work items are dispatched.  The lock hold time
is reduced by the following changes:

1) Use exclusive threads in the work_waitq

When a single work item is dispatched we only need to wake a single
thread to service it.  The current implementation uses non-exclusive
threads so all threads are woken when the dispatcher calls wake_up().
If a large number of threads are in the queue this overhead can become
non-negligible.

2) Conditionally add/remove threads from work waitq outside of tq_lock

Taskq threads need only add themselves to the work wait queue if there
are no pending work items.  Furthermore, the add and remove function
calls can be made outside of the taskq lock since the wait queues are
protected from concurrent access by their own spinlocks.

3) Call wake_up() outside of tq->tq_lock

Again, the wait queues are protected by their own spinlock, so the
dispatcher functions can drop tq->tq_lock before calling wake_up().

A new splat test taskq:contention was added in a prior commit to measure
the impact of these changes.  The following table summarizes the
results using data from the kernel lock profiler.

                        tq_lock time    %diff   Wall clock (s)  %diff
original:               39117614.10     0       41.72           0
exclusive threads:      31871483.61     18.5    34.2            18.0
unlocked add/rm waitq:  13794303.90     64.7    16.17           61.2
unlocked wake_up():     1589172.08      95.9    16.61           60.2

Each row reflects the average result over 5 test runs.
/proc/lock_stats was zeroed out before and collected after each run.
Column 1 is the cumulative hold time in microseconds for tq->tq_lock.
The tests are cumulative; each row reflects the code changes of the
previous rows.  %diff is calculated with respect to "original" as
100*(orig-new)/orig.

Although calling wake_up() outside of the taskq lock dramatically
reduced the taskq lock hold time, the test actually took slightly more
wall clock time.  This is because the point of contention shifts from
the taskq lock to the wait queue lock.  But the change still seems
worthwhile since it removes our taskq implementation as a bottleneck,
assuming the small increase in wall clock time to be statistical
noise.

Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Closes #32
2012-01-18 10:36:57 -08:00
Ned Bass cf5d23fa1e Add taskq contention splat test
Add a test designed to generate contention on the taskq spinlock by
using a large number of threads (100) to perform a large number (131072)
of trivial work items from a single queue.  This simulates conditions
that may occur with the zio free taskq when a 1TB file is removed from a
ZFS filesystem, for example.  This test should always pass.  Its purpose
is to provide a benchmark to easily measure the effectiveness of taskq
optimizations using statistics from the kernel lock profiler.

Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Issue #32
2012-01-18 10:36:51 -08:00
Darik Horn 966e5200d3 Fix `make distclean` for `--with-config=user`
Apply the same fix to SPL that was applied to ZFS earlier at:
zfsonlinux/zfs@d433c20651

Additionally quote @LINUX_SYMBOLS@ because it is a null substitution
in this configuration, which results in a `[ -f  ]` expression that
incorrectly evaluates to true.

  # ./configure --with-config=user
  # make distclean

  Making distclean in module
  make[1]: Entering directory `/spl/module'
  make -C  SUBDIRS=`pwd`  clean
  make: Entering an unknown directory
  make: *** SUBDIRS=/spl/module: No such file or directory.  Stop.

Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2012-01-17 10:06:00 -08:00
Brian Behlendorf 5f6c14b1ed Proxmox VE kernel compat, invalidate_inodes()
The Proxmox VE kernel contains a patch which renames the function
invalidate_inodes() to invalidate_inodes_check().  In the process
it adds a 'check' argument and a '#define invalidate_inodes(x)'
compatibility wrapper for legacy callers.  Therefore, if either
of these functions are exported invalidate_inodes() can be
safely used.

Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Closes #58
2011-12-21 14:29:45 -08:00
Prakash Surya 8f2503e0af Store copy of tqent_flags prior to servicing task
A preallocated taskq_ent_t's tqent_flags must be checked prior to
servicing the taskq_ent_t. Once a preallocated taskq entry is serviced,
the ownership of the entry is handed back to the caller of
taskq_dispatch, thus the entry's contents can potentially be mangled.

In particular, this is a problem in the case where a preallocated taskq
entry is serviced, and the caller clears it's tqent_flags field. Thus,
when the function returns and task_done is called, it looks as though
the entry is **not** a preallocated task (when in fact it **is** a
preallocated task).

In this situation, task_done will place the preallocated taskq_ent_t
structure onto the taskq_t's free list. This is a **huge** mistake. If
the taskq_ent_t is then freed by the caller of taskq_dispatch, the
taskq_t's free list will hold a pointer to garbage data. Even worse, if
nothing has over written the freed memory before the pointer is
dereferenced, it may still look as though it points to a valid list_head
belonging to a taskq_ent_t structure.

Thus, the task entry's flags are now copied prior to servicing the task.
This copy is then checked to see if it is a preallocated task, and
determine if the entry needs to be passed down to the task_done
function.

Signed-off-by: Prakash Surya <surya1@llnl.gov>
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Closes #71
2011-12-16 16:54:00 -08:00
Prakash Surya e7e5f78e7b Swap taskq_ent_t with taskqid_t in taskq_thread_t
The taskq_t's active thread list is sorted based on its
tqt_ent->tqent_id field. The list is kept sorted solely by inserting
new taskq_thread_t's in their correct sorted location; no other
means is used. This means that once inserted, if a taskq_thread_t's
tqt_ent->tqent_id field changes, the list runs the risk of no
longer being sorted.

Prior to the introduction of the taskq_dispatch_prealloc() interface,
this was not a problem as a taskq_ent_t actively being serviced under
the old interface should always have a static tqent_id field. Thus,
once the taskq_thread_t is added to the taskq_t's active thread list,
the taskq_thread_t's tqt_ent->tqent_id field would remain constant.

Now, this is no longer the case. Currently, if using the
taskq_dispatch_prealloc() interface, any given taskq_ent_t actively
being serviced _may_ have its tqent_id value incremented. This happens
when the preallocated taskq_ent_t structure is recursively dispatched.
Thus, a taskq_thread_t could potentially have its tqt_ent->tqent_id
field silently modified from under its feet. If this were to happen
to a taskq_thread_t on a taskq_t's active thread list, this would
compromise the integrity of the order of the list (as the list
_may_ no longer be sorted).

To get around this, the taskq_thread_t's taskq_ent_t pointer was
replaced with its own static copy of the tqent_id. So, as a taskq_ent_t
is pulled off of the taskq_t's pending list, a static copy of its
tqent_id is made and this copy is used to sort the active thread
list. Using a static copy is key in ensuring the integrity of the
order of the active thread list. Even if the underlying taskq_ent_t
is recursively dispatched (as has its tqent_id modified), this
static copy stored inside the taskq_thread_t will remain constant.

Signed-off-by: Prakash Surya <surya1@llnl.gov>
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Issue #71
2011-12-16 13:26:54 -08:00
Prakash Surya 699d5ee8a9 Exercise new taskq interface in splat-taskq tests
The splat-taskq test functions were slightly modified to exercise
the new taskq interface in addition to the old interface.  If the
old interface passes each of its tests, the new interface is
exercised.  Both sub tests (old interface and new interface) must
pass for each test as a whole to pass.

Signed-off-by: Prakash Surya <surya1@llnl.gov>
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Closes #65
2011-12-13 16:10:57 -08:00
Prakash Surya 44217f7aad Implement taskq_dispatch_prealloc() interface
This patch implements the taskq_dispatch_prealloc() interface which
was introduced by the following illumos-gate commit.  It allows for
a preallocated taskq_ent_t to be used when dispatching items to a
taskq.  This eliminates a memory allocation which helps minimize
lock contention in the taskq when dispatching functions.

    commit 5aeb94743e3be0c51e86f73096334611ae3a058e
    Author: Garrett D'Amore <garrett@nexenta.com>
    Date:   Wed Jul 27 07:13:44 2011 -0700

    734 taskq_dispatch_prealloc() desired
    943 zio_interrupt ends up calling taskq_dispatch with TQ_SLEEP

Signed-off-by: Prakash Surya <surya1@llnl.gov>
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Issue #65
2011-12-13 16:10:57 -08:00
Prakash Surya ac1e5b6033 Add Test: "Single task queue, recursive dispatch"
Added another splat taskq test to ensure tasks can be recursively
submitted to a single task queue without issue. When the
taskq_dispatch_prealloc() interface is introduced, this use case
can potentially cause a deadlock if a taskq_ent_t is dispatched
while its tqent_list field is not empty. This _should_ never be
a problem with the existing taskq_dispatch() interface.

Signed-off-by: Prakash Surya <surya1@llnl.gov>
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Issue #65
2011-12-13 16:10:57 -08:00
Prakash Surya 2c02b71b14 Replace tq_work_list and tq_threads in taskq_t
To lay the ground work for introducing the taskq_dispatch_prealloc()
interface, the tq_work_list and tq_threads fields had to be replaced
with new alternatives in the taskq_t structure.

The tq_threads field was replaced with tq_thread_list. Rather than
storing the pointers to the taskq's kernel threads in an array, they are
now stored as a list. In addition to laying the ground work for the
taskq_dispatch_prealloc() interface, this change could also enable taskq
threads to be dynamically created and destroyed as threads can now be
added and removed to this list relatively easily.

The tq_work_list field was replaced with tq_active_list. Instead of
keeping a list of taskq_ent_t's which are currently being serviced, a
list of taskq_threads currently servicing a taskq_ent_t is kept. This
frees up the taskq_ent_t's tqent_list field when it is being serviced
(i.e. now when a taskq_ent_t is being serviced, it's tqent_list field
will be empty).

Signed-off-by: Prakash Surya <surya1@llnl.gov>
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Issue #65
2011-12-13 16:10:50 -08:00
Prakash Surya 046a70c93b Replace struct spl_task with struct taskq_ent
The spl_task structure was renamed to taskq_ent, and all of
its fields were renamed to have a prefix of 'tqent' rather
than 't'. This was to align with the naming convention which
the ZFS code assumes.  Previously these fields were private
so the name never mattered.

Signed-off-by: Prakash Surya <surya1@llnl.gov>
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Issue #65
2011-12-13 12:28:09 -08:00
Prakash Surya ed948fa72b Add SPLAT_TEST_FINI call for SPLAT_TASKQ_TEST6_ID
This change adds the neglected SPLAT_TEST_FINI call for the
SPLAT_TASKQ_TEST6_ID, just as is done for the other 5 SPLAT_TASKQ_*
tests.

Signed-off-by: Prakash Surya <surya1@llnl.gov>
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Closes #64
2011-12-13 12:26:16 -08:00
Prakash Surya e05bec805b Fix a typo referencing an incorrect symbol
The splat_taskq_test4_common function was incorrectly referencing
the splat_taskq-test13_func symbol, when it meant to be using the
splat_taskq_test4_func symbol.

Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Closes #61
2011-11-21 16:52:36 -08:00
Brian Behlendorf 1114ae6ae7 Prepend spl_ to all init/fini functions
This is a bit of cleanup I'd been meaning to get to for a while
to reduce the chance of a type conflict.  Well that conflict
finally occurred with the kstat_init() function which conflicts
with a function in the 2.6.32-6-pve kernel.

Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Closes #56
2011-11-11 09:18:28 -08:00
Brian Behlendorf fe71c0e567 Linux 3.1 compat, shrink_*cache_memory
As of Linux 3.1 the shrink_dcache_memory and shrink_icache_memory
functions have been removed.  This same task is now accomplished
more cleanly with per super block shrinkers.  This unfortunately
leaves us no easy way to support the dnlc_reduce_cache() function.

This support has always been entirely optional.  So when no
reasonable interface is available allow the dnlc_reduce_cache()
function to effectively become a no-op.

The downside of this change is that it will prevent the zfs arc
meta data limts from being enforced.  However, the current zfs
implementation in this regard is already flawed and needs to
be reworked.  If the arc needs to enfore a meta data limit it
will need to be extended to coordinate directly with the zpl.
This will allow us to drop all this compatibility code and get
more fine grained control over the cache management.

Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Issue #52
2011-11-09 19:36:30 -08:00