Age | Commit message (Collapse) | Author |
|
- const fix in the CLKernels part 2
Change-Id: Ia12845e291b4137cbaf76eb8438e381c4fd0368a
Signed-off-by: Manuel Bottini <manuel.bottini@arm.com>
Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/3071
Reviewed-by: Michele Di Giorgio <michele.digiorgio@arm.com>
Comments-Addressed: Arm Jenkins <bsgcomp@arm.com>
Tested-by: Arm Jenkins <bsgcomp@arm.com>
|
|
- Only CLKernels have been updated
Change-Id: Ife55b847c2e39e712a186eb6ca452503d5b66937
Signed-off-by: Manuel Bottini <manuel.bottini@arm.com>
Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/3001
Reviewed-by: Michele Di Giorgio <michele.digiorgio@arm.com>
Reviewed-by: Michalis Spyrou <michalis.spyrou@arm.com>
Comments-Addressed: Arm Jenkins <bsgcomp@arm.com>
Tested-by: Arm Jenkins <bsgcomp@arm.com>
|
|
Split out the parts of ToolchainSupport coming from <memory> and
the parts coming from <string> into their own new header files.
This accounts for 99% of uses of ToolchainSupport, which means that
expensive header files such as arm_neon.h don't need to be included
everywhere.
Knocks about 10% of compilation time off kernel files.
Signed-off-by: Matthew Bentham <matthew.bentham@arm.com>
Change-Id: I2ae718fe766b5ff28608812b0f686f30eeac1b21
Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/2852
Tested-by: Arm Jenkins <bsgcomp@arm.com>
Reviewed-by: Michele Di Giorgio <michele.digiorgio@arm.com>
|
|
Change-Id: I447030e69b9e565f2f81529a41af8c5e7ece7ecf
Signed-off-by: Giorgio Arena <giorgio.arena@arm.com>
Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/2702
Comments-Addressed: Arm Jenkins <bsgcomp@arm.com>
Reviewed-by: Georgios Pinitas <georgios.pinitas@arm.com>
Tested-by: Arm Jenkins <bsgcomp@arm.com>
|
|
When assert is set to 0 the output was not initialized
Change-Id: Icf894d021f2b8932dabc55ae31c1195623be9e33
Signed-off-by: Manuel Bottini <manuel.bottini@arm.com>
Reviewed-on: https://review.mlplatform.org/c/2574
Tested-by: Arm Jenkins <bsgcomp@arm.com>
Reviewed-by: Michele Di Giorgio <michele.digiorgio@arm.com>
|
|
Change-Id: I93ad3e5b9531ce1699214ff6e657a76ffdaacedd
Signed-off-by: Manuel Bottini <manuel.bottini@arm.com>
Reviewed-on: https://review.mlplatform.org/c/2396
Comments-Addressed: Arm Jenkins <bsgcomp@arm.com>
Reviewed-by: Michele Di Giorgio <michele.digiorgio@arm.com>
Tested-by: Arm Jenkins <bsgcomp@arm.com>
|
|
Change-Id: I050f1f84e214e61f7cbb0197a672b68a4940edae
Signed-off-by: Michele Di Giorgio <michele.digiorgio@arm.com>
Reviewed-on: https://review.mlplatform.org/c/2158
Comments-Addressed: Arm Jenkins <bsgcomp@arm.com>
Reviewed-by: Manuel Bottini <manuel.bottini@arm.com>
Tested-by: Arm Jenkins <bsgcomp@arm.com>
Reviewed-by: Giorgio Arena <giorgio.arena@arm.com>
|
|
Avoid querying device's default lws on kernel enqueue as this is already
cached in the kernel during configuration.
Change-Id: Ia26ecb712caeb8f042356815e0cfd23522764d27
Signed-off-by: Georgios Pinitas <georgios.pinitas@arm.com>
Reviewed-on: https://review.mlplatform.org/c/1803
Reviewed-by: Michele Di Giorgio <michele.digiorgio@arm.com>
Comments-Addressed: Arm Jenkins <bsgcomp@arm.com>
Tested-by: Arm Jenkins <bsgcomp@arm.com>
|
|
CLGEMMLowpQuantizeDownInt32ToUint8ScaleByFloat
Since we perform an element-wise operation, it is not necessary to pass the output_depth3d.
Change-Id: Ibfa07a0706e902acf59b444aa61e18a348162ea9
|
|
OpenCL
COMPMID-1424 - Add dot product support for CLDepthwise QASYMM8 3x3 NHWC non-unit stride
With this patch we are able to improve the performance of MobileNet v1-qasymm8 by 37 %
Tried to use the dot product instruction in CLDepthwise QASYMM8 3x3 NHWC non-unit stride
but I have not seen any benefit (maybe because we have few arithemtic operation and we
do not have more load instructions). However Depthwise convolution has been improved by
30%
Change-Id: Id768a99c2e53a04276707e427af5d0ec93419ada
Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/155082
Tested-by: bsgcomp <bsgcomp@arm.com>
Reviewed-by: Georgios Pinitas <georgios.pinitas@arm.com>
|
|
Change-Id: I637add70310d2da4d82b236a6352af9d33be17a1
Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/149706
Reviewed-by: Isabella Gottardi <isabella.gottardi@arm.com>
Reviewed-by: Michele DiGiorgio <michele.digiorgio@arm.com>
Tested-by: bsgcomp <bsgcomp@arm.com>
|
|
This causes problems when ACL is used as a shared library on Android.
Fixes some problems related to creation / destruction order between the Graph's CL backend and core / runtime
Change-Id: I716d63fd42f4586df1ffbb6fa97e4db06d3a781b
Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/143228
Tested-by: Jenkins <bsgcomp@arm.com>
Reviewed-by: Michele DiGiorgio <michele.digiorgio@arm.com>
Reviewed-by: Gian Marco Iodice <gianmarco.iodice@arm.com>
|
|
Change-Id: Ib57d4f7177cc6179302bda7ad870acb8bd3825f5
Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/112115
Tested-by: BSG Visual Compute Jenkins server to access repositories on http://mpd-gerrit.cambridge.arm.com <bsgcomp@arm.com>
Reviewed-by: Anthony Barbier <anthony.barbier@arm.com>
|
|
Change-Id: Id69df4ce98d1d89bdf9c9aa5c4d909659909b30f
Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/110456
Tested-by: BSG Visual Compute Jenkins server to access repositories on http://mpd-gerrit.cambridge.arm.com <bsgcomp@arm.com>
Reviewed-by: Georgios Pinitas <georgios.pinitas@arm.com>
Reviewed-by: Anthony Barbier <anthony.barbier@arm.com>
|
|
DoD:
- Implement NEON kernel for quantizing down the gemmlowp result. The
result should be scaled by a fixedpoint number
- Implement OpenCL kernel for quantizing down the gemmlowp result. The
result should be scaled by a fixedpoint number
- Add test for validating the result
Required for:
- Integration of GEMMLowp in Android NN
- Convolution quantized
- Fully connected quantized
Change-Id: Ia963d25d695471e963961fb49a5600e78374ac4f
Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/110981
Tested-by: BSG Visual Compute Jenkins server to access repositories on http://mpd-gerrit.cambridge.arm.com <bsgcomp@arm.com>
Reviewed-by: Georgios Pinitas <georgios.pinitas@arm.com>
Reviewed-by: Anthony Barbier <anthony.barbier@arm.com>
|