diff options
| author | Joseph Dobson <joseph.dobson@arm.com> | 2020-02-11 19:32:11 +0000 |
|---|---|---|
| committer | Gian Marco Iodice <gianmarco.iodice@arm.com> | 2020-05-10 13:34:11 +0000 |
| commit | 6f8b17dedb7b53b550e6210fd1c78c3a3e086271 (patch) | |
| tree | 6b040314802ea84f1ae84dda235c1af808863346 /arm_compute/runtime/IScheduler.h | |
| parent | 2886c757389c0ccca20a8689daf8180a730ecbc9 (diff) | |
| download | ComputeLibrary-6f8b17dedb7b53b550e6210fd1c78c3a3e086271.tar.gz | |
[ONCPUML-7] arm_compute support for ND parallelism
Currently 1D ranges of work are specified by the scheduler
via two integers, start and end. This limit opportunities
for advance parallelism and scheduling
This patch expands the interfaces to allow for ND parallism.
`GemmCommon::get_window_size` now returns an `NDRange` specifying the work
in N-dimensions rather than with the single integer it used prior (1D)
Execute now takes an `NDCoordinate` which specifies an `NDRange` with a start
position for that work along with an `NDCoordinate` to specify the thread location
In addition to expanding the interface to enable this functionality,
we have added the capability to SGEMM when the number of threads is high
this has the effective of allowing a much greater degree of parallelism
where te problem dimension would previously have limited the number of threads.
Change-Id: I3e1a8b7276216627bec4ff6f24ac2147552ea9fb
Signed-off-by: Joseph Dobson <joseph.dobson@arm.com>
Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/2962
Tested-by: Arm Jenkins <bsgcomp@arm.com>
Reviewed-by: Gian Marco Iodice <gianmarco.iodice@arm.com>
Comments-Addressed: Arm Jenkins <bsgcomp@arm.com>
Diffstat (limited to 'arm_compute/runtime/IScheduler.h')
| -rw-r--r-- | arm_compute/runtime/IScheduler.h | 10 |
1 files changed, 9 insertions, 1 deletions
diff --git a/arm_compute/runtime/IScheduler.h b/arm_compute/runtime/IScheduler.h index f68294016a..a5e20ee627 100644 --- a/arm_compute/runtime/IScheduler.h +++ b/arm_compute/runtime/IScheduler.h @@ -1,5 +1,5 @@ /* - * Copyright (c) 2017-2019 ARM Limited. + * Copyright (c) 2017-2020 ARM Limited. * * SPDX-License-Identifier: MIT * @@ -27,6 +27,7 @@ #include "arm_compute/core/CPP/CPPTypes.h" #include <functional> +#include <limits> namespace arm_compute { @@ -42,6 +43,13 @@ public: STATIC, /**< Split the workload evenly among the threads */ DYNAMIC, /**< Split the workload dynamically using a bucket system */ }; + + /** When arm_compute::ISchedular::Hints::_split_dimension is initialized with this value + * then the schedular is free to break down the problem space over as many dimensions + * as it wishes + */ + static constexpr unsigned int split_dimensions_all = std::numeric_limits<unsigned>::max(); + /** Scheduler hints * * Collection of preferences set by the function regarding how to split a given workload |