Overview
Request 895276 revoked
- Fix libarmnnOnnxParser version
- Update to 21.05:
* Changelog: https://github.com/ARM-software/armnn/releases/tag/v21.05
* Main changes:
- new capabilities to allow users attain higher performance by:
* Making the Arm NN Core thread safe opening the possibility of
running multiple inferences on the same model in parallel
software threads.
* Allowing graphs on the GPU backend import their input and
output buffers either from correctly aligned main memory or
from kernel memory exposed as a dma_buf, thus reducing memory
usage and saving the time involved in copying data into and
out of the GPU memory space.
- In addition to this, support was added to allow the MobileBERT
network to be parsed and run.
- Finally three deprecated components: the Tensorflow Parser,
the Caffe Parser and the Arm NN Quantizer tool, were removed
- Add patch to fix include path:
* armnn-fix-include.patch
- Disable armnn-extratests as it fails to build with current version
- Update to 21.02:
* Changelog: https://github.com/ARM-software/armnn/releases/tag/v21.02
- Drop upstream patch:
* c5c40fe.diff
- Fix instalaltion of cmake files with an upstream patch:
* c5c40fe.diff
- Created by Guillaume_G
- In state revoked
- Supersedes 894805
- Open review for openSUSE:Backports:SLE-15-SP3:Staging:adi:1
- Open review for maxlin_factory
Please group with SR 894804 (ComputeLibrary)
@Guillaume_G Added tensorflow2 to the group so the unit tests don't fail but there are other issues.
From buildlog:
[...] libarmnnOnnxParser25.x86_64: E: shlib-policy-name-error (Badness: 10000) libarmnnOnnxParser24 Your package contains a single shared library but is not named after its SONAME. [...]
Request History
Guillaume_G created request
- Fix libarmnnOnnxParser version
- Update to 21.05:
* Changelog: https://github.com/ARM-software/armnn/releases/tag/v21.05
* Main changes:
- new capabilities to allow users attain higher performance by:
* Making the Arm NN Core thread safe opening the possibility of
running multiple inferences on the same model in parallel
software threads.
* Allowing graphs on the GPU backend import their input and
output buffers either from correctly aligned main memory or
from kernel memory exposed as a dma_buf, thus reducing memory
usage and saving the time involved in copying data into and
out of the GPU memory space.
- In addition to this, support was added to allow the MobileBERT
network to be parsed and run.
- Finally three deprecated components: the Tensorflow Parser,
the Caffe Parser and the Arm NN Quantizer tool, were removed
- Add patch to fix include path:
* armnn-fix-include.patch
- Disable armnn-extratests as it fails to build with current version
- Update to 21.02:
* Changelog: https://github.com/ARM-software/armnn/releases/tag/v21.02
- Drop upstream patch:
* c5c40fe.diff
- Fix instalaltion of cmake files with an upstream patch:
* c5c40fe.diff
licensedigger accepted review
ok
maxlin_factory set openSUSE:Backports:SLE-15-SP3:Staging:adi:1 as a staging project
Being evaluated by staging project "openSUSE:Backports:SLE-15-SP3:Staging:adi:1"
maxlin_factory staged request
Picked "openSUSE:Backports:SLE-15-SP3:Staging:adi:1"
maxlin_factory added maxlin_factory as a reviewer
maxlin_factory declined request
Leap 15.3 is gold, this request is too late for GM but can be a maintenance update, please resubmit to openSUSE:Leap:15.3. Thanks for understanding.
Guillaume_G revoked request