| Age | Commit message (Collapse) | Author |
|---|
|
Use a latch value to detect read/write tearing of the individual reads.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
|
|
Oops.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
|
|
Use the seqno as a seqlock so that we hopefully only read a coherent set
of results for a single update.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
|
|
Since we are tracking the request depth outside of the balancer, we can
apply the qd thottle unconditionally.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
|
|
Slightly more fine-grained than a single on/off switch.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
|
|
An improved version of the rt balancer that tracks the average latency
on each engine, rather than the last instantaneous execution latency.
This makes it much less sensitive to rapid changes, which is both a
positive and a negative.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
|
|
Include the next batch in our estimated run time (that is the time to
execute the remaining queued requests plus ours). This is a better
estimate of queue length and also provides a bit of history to guide
judgement after a sync point.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
|
|
Discard the low cycles when comparing queue length between engines as we
expect a small amount of instabilty. The threshold is chosen to be
around 50us.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
|
|
In order to measure external load (from other clients), we need a
periodic marker along each queue.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
|
|
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
|
|
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
|
|
Stronger checks for some of the fields like dependencies out
of range and invalid values.
Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
|
|
Instead of numbers passed to -b also allow strings.
Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
|
|
It makes more sense for the sequence of random batch durations
to be different between different clients.
Also add a new command line option (-S) to enable forcing the
same sequence if so can be useful for testing.
Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Suggested-by: Chris Wilson <chris@chris-wilson.co.uk>
|
|
New command option -R enables the initial VCS engine assignment
to be round-robinned across all clients.
Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
|
|
This one flips a coin when it fails to decide where to schedule.
It works well with the hd12 workload, either for one or multiple
clients, on which other balancers fail at the moment.
Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
|
|
Last step has to be a batch for the implicit sync to make sense.
Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
|
|
This enables media balancing workloads to be ran on GT2 SKUs.
Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
|
|
Multiple dependencies separated by forward slashes are now supported.
Some media workloads also updated to use this for better efficiency.
Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
|
|
Put the batch in exec object array last instead of moving it
after checking have there been any dependencies requested.
Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
|
|
One copy and paste error sneaked in the no reloc changes.
Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Reported-by: Chris Wilson <chris@chris-wilson.co.uk>
|
|
Tool which emits batch buffers to engines with configurable
sequences, durations, contexts, dependencies and userspace waits.
Unfinished but shows promise so sending out for early feedback.
v2:
* Load workload descriptors from files. (also -w)
* Help text.
* Calibration control if needed. (-t)
* NORELOC | LUT to eb flags.
* Added sample workload to wsim/workload1.
v3:
* Multiple parallel different workloads (-w -w ...).
* Multi-context workloads.
* Variable (random) batch length.
* Load balancing (round robin and queue depth estimation).
* Workloads delays and explicit sync steps.
* Workload frequency (period) control.
v4:
* Fixed queue-depth estimation by creating separate batches
per engine when qd load balancing is on.
* Dropped separate -s cmd line option. It can turn itself on
automatically when needed.
* Keep a single status page and lie about the write hazard
as suggested by Chris.
* Use batch_start_offset for controlling the batch duration.
(Chris)
* Set status page object cache level. (Chris)
* Moved workload description to a README.
* Tidied example workloads.
* Some other cleanups and refactorings.
v5:
* Master and background workloads (-W / -w).
* Single batch per step is enough even when balancing. (Chris)
* Use hars_petruska_f54_1_random IGT functions and see to zero
at start. (Chris)
* Use WC cache domain when WC mapping. (Chris)
* Keep seqnos 64-bytes apart in the status page. (Chris)
* Add workload throttling and queue-depth throttling commands.
(Chris)
v6:
* Added two more workloads.
* Merged RT balancer from Chris.
v7:
* Merged NO_RELOC patch from Chris.
* Added missing RT balancer to help text.
TODO list:
* Fence support.
* Batch buffer caching (re-use pool).
* Better error handling.
* Less 1980's workload parsing.
* More workloads.
* Threads?
* ... ?
Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Cc: Chris Wilson <chris@chris-wilson.co.uk>
Cc: "Rogozhkin, Dmitry V" <dmitry.v.rogozhkin@intel.com>
|
