Age | Commit message (Collapse) | Author |
|
Added volatile to the for loop counter in direct_convolution1x1.cl, direct_convolution3x3.cl and direct_convolution5x5.cl
This fix seems to solve the problem on Samsung S8 about the mismatches and clEnqueueMapBuffer.
Change-Id: I51687ec94bb897af2698ceab1133c988821e4c4e
Reviewed-on: http://mpd-gerrit.cambridge.arm.com/90601
Reviewed-by: Anthony Barbier <anthony.barbier@arm.com>
Tested-by: Kaizen <jeremy.johnson+kaizengerrit@arm.com>
|
|
GraphContext hold all the information about the hints that need to be
passed in the nodes. As these might expand, it serves as a centralized
class for such information.
Change-Id: I0b5527630fb97cc5fa500db0bac8307ff2ea36e6
Reviewed-on: http://mpd-gerrit.cambridge.arm.com/90300
Tested-by: Kaizen <jeremy.johnson+kaizengerrit@arm.com>
Reviewed-by: Anthony Barbier <anthony.barbier@arm.com>
|
|
Change-Id: Ia99d26768d780d6d0fee9f5e2ed88f3276ecf3d0
Reviewed-on: http://mpd-gerrit.cambridge.arm.com/89908
Reviewed-by: Anthony Barbier <anthony.barbier@arm.com>
Tested-by: Kaizen <jeremy.johnson+kaizengerrit@arm.com>
Reviewed-by: Georgios Pinitas <georgios.pinitas@arm.com>
Reviewed-by: Pablo Tello <pablo.tello@arm.com>
|
|
non-internal builds
Change-Id: Id4d03880ddec9c29d515724134c654221804b513
Reviewed-on: http://mpd-gerrit.cambridge.arm.com/90392
Reviewed-by: Pablo Tello <pablo.tello@arm.com>
Reviewed-by: Gian Marco Iodice <gianmarco.iodice@arm.com>
Tested-by: Kaizen <jeremy.johnson+kaizengerrit@arm.com>
|
|
Change toolchain for arm64-v8a and armv7a and disable
examples when building for bare_metal.
Change-Id: I09c22cc9aad4d32a8e521c36fec98acad2b4d855
Reviewed-on: http://mpd-gerrit.cambridge.arm.com/90207
Tested-by: Kaizen <jeremy.johnson+kaizengerrit@arm.com>
Reviewed-by: Anthony Barbier <anthony.barbier@arm.com>
|
|
issue (see COMPMID-602)
Change-Id: I8e30f13105afa155df4e7b4e764f5da218d486de
Reviewed-on: http://mpd-gerrit.cambridge.arm.com/90053
Tested-by: Kaizen <jeremy.johnson+kaizengerrit@arm.com>
Reviewed-by: Steven Niu <steven.niu@arm.com>
|
|
Adds ThreadInfo in the OMPScheduler.
Change-Id: I760ff9ea8af0431e024b4a4b1fc77ff88980f51a
Reviewed-on: http://mpd-gerrit.cambridge.arm.com/90158
Reviewed-by: Michalis Spyrou <michalis.spyrou@arm.com>
Reviewed-by: Anthony Barbier <anthony.barbier@arm.com>
Tested-by: Anthony Barbier <anthony.barbier@arm.com>
|
|
Change-Id: I8035cfd2b7fd62bd9f79f2a7d4b59f3e2bbe9163
Reviewed-on: http://mpd-gerrit.cambridge.arm.com/89988
Tested-by: Kaizen <jeremy.johnson+kaizengerrit@arm.com>
Reviewed-by: Georgios Pinitas <georgios.pinitas@arm.com>
|
|
-Adds grouping support in convolution layer
-Adds Normalization layer node in graph
-Adds alexnet example
-Fixes FullyConnectedLayer output autoconfigure (works only for 1d batch
space)
Change-Id: I5bd75f9a8b08cfd68f7c34745150266c2bc4221f
Reviewed-on: http://mpd-gerrit.cambridge.arm.com/89518
Tested-by: Kaizen <jeremy.johnson+kaizengerrit@arm.com>
Reviewed-by: Anthony Barbier <anthony.barbier@arm.com>
|
|
Change-Id: I8ae44187624deeab3d40d878e7b34ff651f1dad0
Reviewed-on: http://mpd-gerrit.cambridge.arm.com/89834
Reviewed-by: Georgios Pinitas <georgios.pinitas@arm.com>
Tested-by: Kaizen <jeremy.johnson+kaizengerrit@arm.com>
|
|
Change-Id: Ieff79255b013cfcc1bae871276ab94722ab0b7c6
Reviewed-on: http://mpd-gerrit.cambridge.arm.com/90002
Reviewed-by: Pablo Tello <pablo.tello@arm.com>
Tested-by: Kaizen <jeremy.johnson+kaizengerrit@arm.com>
|
|
Change-Id: I7798a56677d541338a73e3888ed0a2cfe0375794
Reviewed-on: http://mpd-gerrit.cambridge.arm.com/89726
Tested-by: Kaizen <jeremy.johnson+kaizengerrit@arm.com>
Reviewed-by: Georgios Pinitas <georgios.pinitas@arm.com>
|
|
Change-Id: Iefa628de780b76977578a250c17f27962664eae0
Reviewed-on: http://mpd-gerrit.cambridge.arm.com/88094
Tested-by: Kaizen <jeremy.johnson+kaizengerrit@arm.com>
Reviewed-by: Pablo Tello <pablo.tello@arm.com>
|
|
The maximum workgroup size depends on the kernel and the device, rather
than being a property of the device. The present patch fixes the case
when a kernel is queued with no workgroup size and the default workgroup
size is used instead.
A previous patch introduced a maximum workgroup size that depended on
the device but ignored the kernel. In OpenCL the maximum workgroup size
we query from the device is an upper bound of the actual maximum that
we can query for a given kernel running on the same device. For some
kernels the values will match, but for others we will get a lower value
when querying for an specific kernel (i.e. if the kernel uses a high
number of registers).
Change-Id: I3bed6bde80ddc4f0ddb8f82c80903774aa1999b6
Reviewed-on: http://mpd-gerrit.cambridge.arm.com/89471
Tested-by: Kaizen <jeremy.johnson+kaizengerrit@arm.com>
Reviewed-by: Anthony Barbier <anthony.barbier@arm.com>
|
|
Change-Id: I54dd435258a2d0ff486ded64b23654bab6b80f3f
Reviewed-on: http://mpd-gerrit.cambridge.arm.com/89373
Tested-by: Kaizen <jeremy.johnson+kaizengerrit@arm.com>
Reviewed-by: Georgios Pinitas <georgios.pinitas@arm.com>
|
|
levels of library
Until now we had:
core = core_obj
arm_compute = core_obj + arm_compute_obj
graph = core_obj + arm_compute_obj + graph_obj
But if an application link against more than one of these libraries then bad things happen.
Added version strings in the runtime library too (As it used to only be in Core objects)
Updated doxygen for how to compile the examples
Change-Id: I7aad6ecf75cfa8dca59f2ea093e13fb0314a3eb4
Reviewed-on: http://mpd-gerrit.cambridge.arm.com/89743
Reviewed-by: Gian Marco Iodice <gianmarco.iodice@arm.com>
Tested-by: Kaizen <jeremy.johnson+kaizengerrit@arm.com>
|
|
In case the library has been built just for NEON, we should not run the example
Change-Id: I99f8d3efd20276dfeb73ca8362872111be1b1fd7
Reviewed-on: http://mpd-gerrit.cambridge.arm.com/89682
Reviewed-by: Anthony Barbier <anthony.barbier@arm.com>
Tested-by: Kaizen <jeremy.johnson+kaizengerrit@arm.com>
|
|
Change-Id: I3bd91a9f234b8b42a96454c0d2e45f6098c543a5
Reviewed-on: http://mpd-gerrit.cambridge.arm.com/89481
Reviewed-by: Georgios Pinitas <georgios.pinitas@arm.com>
Tested-by: Kaizen <jeremy.johnson+kaizengerrit@arm.com>
|
|
Change-Id: I7bbe0f19a7255a37a6b0ce8c4d0bb57cf21c4bfb
Reviewed-on: http://mpd-gerrit.cambridge.arm.com/89677
Reviewed-by: Anthony Barbier <anthony.barbier@arm.com>
Tested-by: Kaizen <jeremy.johnson+kaizengerrit@arm.com>
|
|
Change-Id: I8e3fece5b786b5529e8f1320fb5794e43756ec4a
Reviewed-on: http://mpd-gerrit.cambridge.arm.com/89274
Tested-by: Kaizen <jeremy.johnson+kaizengerrit@arm.com>
Reviewed-by: Georgios Pinitas <georgios.pinitas@arm.com>
Reviewed-by: Anthony Barbier <anthony.barbier@arm.com>
|
|
Added explanation about how to build the example graph_lenet.cpp
Change-Id: I1da0683f6284a77e0f3e11250a96bc9f0d7f0b97
Reviewed-on: http://mpd-gerrit.cambridge.arm.com/89690
Tested-by: Kaizen <jeremy.johnson+kaizengerrit@arm.com>
Reviewed-by: Anthony Barbier <anthony.barbier@arm.com>
|
|
Missing include path from some build commands
Change-Id: I28e87938f90d0507d8ce0814968c252955b49023
Reviewed-on: http://mpd-gerrit.cambridge.arm.com/89669
Reviewed-by: Anthony Barbier <anthony.barbier@arm.com>
Reviewed-by: Georgios Pinitas <georgios.pinitas@arm.com>
Tested-by: Kaizen <jeremy.johnson+kaizengerrit@arm.com>
|
|
Change-Id: Iba380f0edd5a7d759a5a52545263a25c966626c7
Reviewed-on: http://mpd-gerrit.cambridge.arm.com/89564
Reviewed-by: Anthony Barbier <anthony.barbier@arm.com>
Tested-by: Kaizen <jeremy.johnson+kaizengerrit@arm.com>
Reviewed-by: Georgios Pinitas <georgios.pinitas@arm.com>
|
|
Change-Id: I566a41061b75c3a1dad5374fcdc84372e6cfbe89
Reviewed-on: http://mpd-gerrit.cambridge.arm.com/89670
Tested-by: Kaizen <jeremy.johnson+kaizengerrit@arm.com>
Reviewed-by: Anthony Barbier <anthony.barbier@arm.com>
|
|
Change-Id: If70bba00bb4451251cbc21058dfecff2c0d106de
Reviewed-on: http://mpd-gerrit.cambridge.arm.com/89641
Tested-by: Kaizen <jeremy.johnson+kaizengerrit@arm.com>
Reviewed-by: Georgios Pinitas <georgios.pinitas@arm.com>
Reviewed-by: Anthony Barbier <anthony.barbier@arm.com>
|
|
Fixed some Doxygen issues in the introduction and tests sections
Switched MaliCounter to use ARM_COMPUTE_ERROR in order to get more context in case of failure
Switched to use ARM_COMPUTE_ERROR instead of ERROR_ON in PMU and MaliCounter so that checks work in release mode too. (As these checks are to detect platform issues rather than programming errors)
Change-Id: Ibbefa5a87d93c5c3aa5e190680052e4e99987dfe
Reviewed-on: http://mpd-gerrit.cambridge.arm.com/89662
Reviewed-by: Gian Marco Iodice <gianmarco.iodice@arm.com>
Reviewed-by: Georgios Pinitas <georgios.pinitas@arm.com>
Tested-by: Kaizen <jeremy.johnson+kaizengerrit@arm.com>
|
|
Change-Id: I1f495aa9506fc1d86926f20f01b3d06f4c4dd578
Reviewed-on: http://mpd-gerrit.cambridge.arm.com/89580
Tested-by: Kaizen <jeremy.johnson+kaizengerrit@arm.com>
Reviewed-by: Anthony Barbier <anthony.barbier@arm.com>
|
|
map() and unmap() are not inline functions and can cause issues at compile time:
i.e.
multiple definition of `arm_compute::utils::unmap(arm_compute::CLTensor&)'
Change-Id: I2da3e70030d896a06feb995fa6ebf7a718fe0af6
Reviewed-on: http://mpd-gerrit.cambridge.arm.com/89579
Reviewed-by: Georgios Pinitas <georgios.pinitas@arm.com>
Tested-by: Kaizen <jeremy.johnson+kaizengerrit@arm.com>
Reviewed-by: Anthony Barbier <anthony.barbier@arm.com>
|
|
Change-Id: I919c3352bf22bb41df2928be291be754c0518d3f
Reviewed-on: http://mpd-gerrit.cambridge.arm.com/89536
Reviewed-by: Pablo Tello <pablo.tello@arm.com>
Reviewed-by: Georgios Pinitas <georgios.pinitas@arm.com>
Tested-by: Kaizen <jeremy.johnson+kaizengerrit@arm.com>
|
|
Change-Id: I685a68e7bc8d2cdff19851d839f244206b3d5790
Reviewed-on: http://mpd-gerrit.cambridge.arm.com/89391
Tested-by: Kaizen <jeremy.johnson+kaizengerrit@arm.com>
Reviewed-by: Georgios Pinitas <georgios.pinitas@arm.com>
Reviewed-by: Anthony Barbier <anthony.barbier@arm.com>
|
|
Change-Id: Iae85c195a4ecdaf6df1d0055f3ad09582d128dc1
Reviewed-on: http://mpd-gerrit.cambridge.arm.com/88669
Reviewed-by: Pablo Tello <pablo.tello@arm.com>
Tested-by: Kaizen <jeremy.johnson+kaizengerrit@arm.com>
|
|
Change-Id: Ief165b1d583a70cbe35aae93f05ddfe962196323
Reviewed-on: http://mpd-gerrit.cambridge.arm.com/89503
Reviewed-by: Georgios Pinitas <georgios.pinitas@arm.com>
Tested-by: Kaizen <jeremy.johnson+kaizengerrit@arm.com>
|
|
Change-Id: Icc6194a311af0e96978e6be2cc4c5da9d7fb0bcc
Reviewed-on: http://mpd-gerrit.cambridge.arm.com/89493
Reviewed-by: Georgios Pinitas <georgios.pinitas@arm.com>
Tested-by: Kaizen <jeremy.johnson+kaizengerrit@arm.com>
Reviewed-by: Michalis Spyrou <michalis.spyrou@arm.com>
Reviewed-by: Steven Niu <steven.niu@arm.com>
|
|
Incorrect logic was leading to an object being registered for managing
but was never needed.
Change-Id: I1682f7840f5d47ed856d60d3c0ee458a097e6b98
Reviewed-on: http://mpd-gerrit.cambridge.arm.com/89496
Reviewed-by: Gian Marco Iodice <gianmarco.iodice@arm.com>
Tested-by: Kaizen <jeremy.johnson+kaizengerrit@arm.com>
|
|
Change-Id: I2eacba2c87bce84b7f6b69a734ff775473f990bc
Reviewed-on: http://mpd-gerrit.cambridge.arm.com/89401
Reviewed-by: Steven Niu <steven.niu@arm.com>
Tested-by: Kaizen <jeremy.johnson+kaizengerrit@arm.com>
|
|
Change-Id: I52c95de004d4508f74f7d984695c0abd24c72525
Reviewed-on: http://mpd-gerrit.cambridge.arm.com/89385
Reviewed-by: Gian Marco Iodice <gianmarco.iodice@arm.com>
Tested-by: Kaizen <jeremy.johnson+kaizengerrit@arm.com>
|
|
Change-Id: I2fbb6dda1c281627a4d64dce3b4c4d2ebaa8d022
Reviewed-on: http://mpd-gerrit.cambridge.arm.com/89289
Reviewed-by: Pablo Tello <pablo.tello@arm.com>
Tested-by: Kaizen <jeremy.johnson+kaizengerrit@arm.com>
Reviewed-by: Anthony Barbier <anthony.barbier@arm.com>
|
|
Change-Id: I42d1f2fa2eb6a241df22f716bbca7fda71ba3eff
Reviewed-on: http://mpd-gerrit.cambridge.arm.com/89217
Tested-by: Kaizen <jeremy.johnson+kaizengerrit@arm.com>
Reviewed-by: Anthony Barbier <anthony.barbier@arm.com>
|
|
The empty string used by default for config_id caused a memory corruption on Android.
Change-Id: Ib9c764c8a8f7009936071f8c8fec1653bb000f90
Reviewed-on: http://mpd-gerrit.cambridge.arm.com/89262
Tested-by: Kaizen <jeremy.johnson+kaizengerrit@arm.com>
Reviewed-by: Georgios Pinitas <georgios.pinitas@arm.com>
Reviewed-by: Anthony Barbier <anthony.barbier@arm.com>
|
|
Change-Id: I0d96ceb9d9d1d077bec09330cda4fbe6d81ce641
Reviewed-on: http://mpd-gerrit.cambridge.arm.com/88476
Tested-by: Kaizen <jeremy.johnson+kaizengerrit@arm.com>
Reviewed-by: Pablo Tello <pablo.tello@arm.com>
|
|
Change-Id: I4924ab1de17adc3b880a5cc22f2497abbc8e221b
Reviewed-on: http://mpd-gerrit.cambridge.arm.com/85820
Tested-by: Kaizen <jeremy.johnson+kaizengerrit@arm.com>
Reviewed-by: Steven Niu <steven.niu@arm.com>
|
|
example
Change-Id: I31839993bd75789566045c0db53dbf6dc3ae14e9
Reviewed-on: http://mpd-gerrit.cambridge.arm.com/89202
Reviewed-by: Gian Marco Iodice <gianmarco.iodice@arm.com>
Reviewed-by: Georgios Pinitas <georgios.pinitas@arm.com>
Tested-by: Anthony Barbier <anthony.barbier@arm.com>
|
|
Change-Id: Ia37515fb8238a03699d75751b877d5aaff5ba1a0
Reviewed-on: http://mpd-gerrit.cambridge.arm.com/89174
Reviewed-by: Moritz Pflanzer <moritz.pflanzer@arm.com>
Tested-by: Anthony Barbier <anthony.barbier@arm.com>
|
|
Change-Id: Idba0b30bfb27866a46a22388014ab81432ea28dc
Reviewed-on: http://mpd-gerrit.cambridge.arm.com/86196
Reviewed-by: Anthony Barbier <anthony.barbier@arm.com>
Tested-by: Kaizen <jeremy.johnson+kaizengerrit@arm.com>
|
|
The optimization works on tensors with width <= 8 and height <= 8.
The new code is 0.5 faster than the old one as it uses fewer instrunctions to compute the same result.
Change-Id: I408d6c73ebd3d266bdaaf92fcb6bcdd58f239977
Reviewed-on: http://mpd-gerrit.cambridge.arm.com/88642
Tested-by: Kaizen <jeremy.johnson+kaizengerrit@arm.com>
Reviewed-by: Anthony Barbier <anthony.barbier@arm.com>
|
|
Change-Id: Idd6f67f6f7dbe1110e012b5fa012d112125a42bc
Reviewed-on: http://mpd-gerrit.cambridge.arm.com/89101
Tested-by: Kaizen <jeremy.johnson+kaizengerrit@arm.com>
Reviewed-by: Georgios Pinitas <georgios.pinitas@arm.com>
|
|
Change-Id: I4e21ad98d029e360010c5927f04b716527700a00
Reviewed-on: http://mpd-gerrit.cambridge.arm.com/88888
Reviewed-by: Anthony Barbier <anthony.barbier@arm.com>
Tested-by: Kaizen <jeremy.johnson+kaizengerrit@arm.com>
|
|
The specified output size of the failing test case was invalid.
Additionally the kernel has been cleaned up and asserts have been added
in case of invalid configurations.
Change-Id: I198f3574f003b71968e4081a54cf102d748af5c1
Reviewed-on: http://mpd-gerrit.cambridge.arm.com/88821
Tested-by: Kaizen <jeremy.johnson+kaizengerrit@arm.com>
Reviewed-by: Steven Niu <steven.niu@arm.com>
Reviewed-by: Anthony Barbier <anthony.barbier@arm.com>
|
|
Change-Id: I34f94f99cb05f0eabafee13c5e623ee779b72360
Reviewed-on: http://mpd-gerrit.cambridge.arm.com/83741
Tested-by: Kaizen <jeremy.johnson+kaizengerrit@arm.com>
Reviewed-by: Anthony Barbier <anthony.barbier@arm.com>
Reviewed-by: Pablo Tello <pablo.tello@arm.com>
|
|
Change-Id: Ic0a9dbd8e646abbf8d9ea52e497a5fe60e499cc7
Reviewed-on: http://mpd-gerrit.cambridge.arm.com/88883
Tested-by: Kaizen <jeremy.johnson+kaizengerrit@arm.com>
Reviewed-by: Anthony Barbier <anthony.barbier@arm.com>
|