aboutsummaryrefslogtreecommitdiff
diff options
context:
space:
mode:
authorTeresa Charlin <teresa.charlinreyes@arm.com>2020-08-28 15:06:57 +0100
committerTeresa Charlin <teresa.charlinreyes@arm.com>2020-08-28 15:06:57 +0100
commit5ccbdf1ac82013c8cb1c607252f72b6858e97a53 (patch)
tree9dcd2e99fb7fc9ae9a977e79179950fe884e5a13
parent686fd0e66384b887d84501df4fca1d5ab9170824 (diff)
downloadandroid-nn-driver-5ccbdf1ac82013c8cb1c607252f72b6858e97a53.tar.gz
Remove mention of lack of support for HAL1.3 in GpuAcc
Signed-off-by: Teresa Charlin <teresa.charlinreyes@arm.com> Change-Id: I2ac8b2dd1d012ddcd2f134f6d0d564f068caf601
-rw-r--r--NnapiSupport.txt4
1 files changed, 1 insertions, 3 deletions
diff --git a/NnapiSupport.txt b/NnapiSupport.txt
index 7e1999b4..989dc5f9 100644
--- a/NnapiSupport.txt
+++ b/NnapiSupport.txt
@@ -82,6 +82,4 @@ appropriately and the framework implements those operations using a CPU implemen
NOTE: By convention, only those tensor types have been listed above, which are fully supported across all
ArmNN backends.
- FLOAT16 input tensors are partially supported on most HAL 1.2 operators on the GpuAcc and
- CpuRef backends, however not on CpuAcc.
- - QUANT8_ASYMM_SIGNED has been added to the list in spite of not being supported in GpuAcc,
- as this data type was added as part of HAL 1.3, which is currently not supported by GpuAcc.
+ CpuRef backends, however not on CpuAcc. \ No newline at end of file