Eviction changes allowing to evict (remap) cachelines while
holding hash bucket write lock instead of global metadata
write lock.
As eviction (replacement) is now tightly coupled with request,
each request uses eviction size equal to number of its
unmapped cachelines.
Evicting without global metadata write lock is possible
thanks to the fact that remaping is always performed
while exclusively holding cacheline (read or write) lock.
So for a cacheline on LRU list we acquire cacheline lock,
safely resolve hash and consequently write-lock hash bucket.
Since cacheline lock is acquired under hash bucket (everywhere
except for new eviction implementation), we are certain that
noone acquires cacheline lock behind our back. Concurrent
eviction threads are eliminated by holding eviction list
lock for the duration of critial locking operations.
Signed-off-by: Adam Rutkowski <adam.j.rutkowski@intel.com>
.. to make it clean that true means cleaner must lock
cachelines rather than the lock is already being held.
Signed-off-by: Adam Rutkowski <adam.j.rutkowski@intel.com>
Reformat function that calculates how long cache/core is dirty
Update `dirty_for` types in functional tests
Values stored in info structs fields (both in cache and core structs)
are unsigned 64-bits ints but `dirty_for`s were unsigned 32-bits ints.
Use existing function to transform returned value to seconds.
Replace seconds stored in metadata with seconds.
Replacement was done if old value of replaced field was equal to zero.
Acquiring monotonic high precission timestamp is potentially
slow and it makes sense to compare the field's value
to zero before calling atomic function.
Signed-off-by: Slawomir Jankowski <slawomir.jankowski@intel.com>
Cacheline concurrency functions have their interface changed
so that the cacheline concurrency private context is
explicitly on the parameter list, rather than being taken
from cache->device->concurrency.cache_line.
Cache pointer is no longer provided as a parameter to these
functions. Cacheline concurrency context now has a pointer
to cache structure (for logging purposes only).
The purpose of this change is to facilitate unit testing.
Signed-off-by: Adam Rutkowski <adam.j.rutkowski@intel.com>
Removing the logic for oportunistic partition overflow
reduction by evicting more cachelines than actually
required by the request being serviced.
Signed-off-by: Adam Rutkowski <adam.j.rutkowski@intel.com>
Min and max values, keept as an explicit number of cachelines, are tightly
coupled with particular cache. This might lead to errors and mismatches after
reattaching cache of different size.
To prevent those errors, min and max should be calculated dynamically.
Signed-off-by: Michal Mielewczyk <michal.mielewczyk@intel.com>
In case of cleaning metadata used to be flushed only when status of whole cache
line changed to clean.
This patch ensures that metadata flush is triggered after changin status of each
single sector is cache line.
Signed-off-by: Michal Mielewczyk <michal.mielewczyk@intel.com>
After second dirty write to cache line which was already dirty, metadata flush
was not triggered. In case of dirty shutdown, this led to data corruption.
Signed-off-by: Michal Mielewczyk <michal.mielewczyk@intel.com>
Cleaner doesn't set core object in req as it works in domain of cache
lines, which may belong to various cores. It this case should retrieve
core object not from the req, but from the map instead.
Signed-off-by: Robert Baldyga <robert.baldyga@intel.com>
When single request to cache was issued, stats updating function was called with
0 bytes as value to update. In case of many request issued to cache, stats were
updated only in case of error.
Signed-off-by: Michal Mielewczyk <michal.mielewczyk@intel.com>
Adding synchronization around metadata collision segment pages.
This part of metadata is modified when cacheline is mapped/unmapped
and when dirty status changes.
Synchronization on page level is required on top of cacheline
and hash bucket locks to assure metadata flush always reads
consistent state when copying entire collision table memory
page.
Signed-off-by: Adam Rutkowski <adam.j.rutkowski@intel.com>
The ocf_async_lock may be used in atomic context, thus we need
to replace synchronization primitives to non-sleeping variants.
Signed-off-by: Robert Baldyga <robert.baldyga@intel.com>
Core is not assigned to request in cleaner, so to increase it's stats it has to
be retrieved from mapping.
Signed-off-by: Michal Mielewczyk <michal.mielewczyk@intel.com>
Environment should provide calls for destroying primitives (i.e. env_mutex_destroy()) and OCF should call these functions in its cleanup paths.
Signed-off-by: Firas Medini <mdnfiras@yahoo.com>
Promotion policy is supposed to perform ALRU noise filtering by
eliminating one-hit wonders being added to cache and polluting it.
Signed-off-by: Jan Musial <jan.musial@intel.com>
Refactoring ocf_submit_cache_reqs to make it clear that
req->map is accessed at index derived from offset argument,
not necesarily starting at 0.
Signed-off-by: Adam Rutkowski <adam.j.rutkowski@intel.com>
Write-only cache mode is similar to writeback, however read
operations do not promote data to cache. Reads are mostly serviced
by the core device, only dirty sectors are fetched from the cache.
Signed-off-by: Adam Rutkowski <adam.j.rutkowski@intel.com>
ocf_request has always been first class citizen in OCF,
so lets place it along with another essential objects.
Signed-off-by: Robert Baldyga <robert.baldyga@intel.com>
This simplifies code by allowing to express programmer intent
explicitly and helps to avoid missing return statements (this patch
fixes at least one bug related to this).
Signed-off-by: Robert Baldyga <robert.baldyga@intel.com>
This simplifies cases when we want to call completion callback
and immediately return from void-returning function, by allowing
to explicitly express programmers intent. That way we can avoid
cases when return statement is missing by mistake (this patch
fixes at least one bug related to this).
Signed-off-by: Robert Baldyga <robert.baldyga@intel.com>
This is useful when reference counter is initialized in non-zeroed
memory (or assuming atomic variable is not properly initialized by
memseting to zero).
Signed-off-by: Adam Rutkowski <adam.j.rutkowski@intel.com>
In order to synchronize management operations with I/O OCF
maintains in-flight request counters. For example such ref
counters are used during ocf_mngt_detach to drain requests
accessing cache metadata (cache requests counter) and in
ocf_mngt_flush where we wait for outstanding requests sent
in write back mode (dirty requests counter).
Typically I/O threads increment cache/dirty counter when
creating request and decrement counter on request completion.
Management thread sets atomic variable to signal the start of
management operation. I/O threads react to this by changing
I/O requests mode so that the cache/dirty reference counter
is not incremented. As a result reference counter keeps getting
decremented. Management thread waits for the counter to drop to 0
and proceeds with management operation with assumption that no
cache/dirty requests are in progress.
This patch introduces a handy utility for requests reference
counting logic. ocf_refcnt_inc / dec are used to increment/
decrement counter. ocf_refcnt_freeze() makes subsequent
ocf_refcnt_inc() calls to return false, indicating that counter
cannot be incremented at this moment. ocf_refcnt_register_zero_cb
can be used to asynchronously wait for counter to drop to 0.
Signed-off-by: Adam Rutkowski <adam.j.rutkowski@intel.com>
For flush/purge entry points to be fully asynchronous we still
need to rework flush mutex and waiting for outstanding dirty
requests.
Signed-off-by: Adam Rutkowski <adam.j.rutkowski@intel.com>
Send cleaner IOs with appropriate queue set
This solves the issue of bottom adapter getting NULL in io->io_queue
Signed-off-by: Vitaliy Mysak <vitaliy.mysak@intel.com>
This allows to reuse same step functions giving them different parameters
on each step.
Additionally move pipeline to utils, to make it accessible to other
subsystems of OCF (e.g. metadata).
Signed-off-by: Robert Baldyga <robert.baldyga@intel.com>
NOTE: This is still not the real asynchronism. Metadata interfaces
are still not fully asynchronous.
Signed-off-by: Robert Baldyga <robert.baldyga@intel.com>
Signed-off-by: Michal Mielewczyk <michal.mielewczyk@intel.com>
NOTE: This patch only changes API that pretends to be asynchronous.
Most of management operations are still performed synchronously.
The real asynchronism will be introduced in the next patches.
Signed-off-by: Robert Baldyga <robert.baldyga@intel.com>
