Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

[WIP] update tensorrt version to 8.5 #774

Open
wants to merge 3 commits into
base: main
Choose a base branch
from
Open

[WIP] update tensorrt version to 8.5 #774

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
7498f30
update tensorrt version to 8.5
zzpmiracle Nov 23, 2022
727a753
fix
zzpmiracle Nov 23, 2022
2c0623e
Update pytorch112_gpu.yml
zzpmiracle Nov 23, 2022
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
2 changes: 1 addition & 1 deletion .github/workflows/pytorch112_gpu.yml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -18,7 +18,7 @@ jobs:
develop_base_image: nvidia/cuda:11.3.0-cudnn8-devel-ubuntu20.04
runtime_base_image: nvidia/cuda:11.3.0-cudnn8-devel-ubuntu20.04
extra_build_args: --build-arg PYTHON_VERSION=PYTHON3.8 --build-arg ENABLE_FIND_FASTEST_APT_SOURCE=OFF
extra_envs: -e TORCH_BLADE_BUILD_TENSORRT_STATIC=ON
extra_envs: -e TORCH_BLADE_BUILD_TENSORRT_STATIC=OFF
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I'm curious why?

Copy link
Collaborator Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Just for product test. Our product better to provide static linked version, to reduce user's envoiroment request. But we met some compile error for TRT8.5 like (.text._ZN5torch5blade21GetTorchMlirWhiteListEv+0x17): relocation truncated to fit: R_X86_64_PC32 against .bss._ZGVZN5torch5blade21GetTorchMlirWhiteListEvE10white_list' in all static link build. This used to occur in torch1.12+cu113 only, so change this to fix this remaining problem. I'm testing some compiler flag locally such as -fPIC.

-e TORCH_BLADE_CI_BUILD_TORCH_VERSION=1.12.0+cu113
exec_command: bash ./scripts/ci/test_pytorch_blade.sh
deploy_command: TORCH_VERSION=1.12.0+cu113 bash ./scripts/ci/deploy_pytorch_blade.sh
Expand Down
2 changes: 1 addition & 1 deletion docker/scripts/install-tensorrt.sh
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -11,7 +11,7 @@
# limitations under the License.
set -ex

tensorrt_pkg=TensorRT-8.2.3.0.Linux.x86_64-gnu.cuda-11.4.cudnn8.2.tar.gz
tensorrt_pkg=TensorRT-8.5.1.7.Linux.x86_64-gnu.cuda-11.8.cudnn8.6.tar.gz
curl -sL https://pai-blade.oss-accelerate.aliyuncs.com/build_deps/tensorrt/${tensorrt_pkg} -o ${tensorrt_pkg}

tar xvfz $tensorrt_pkg -C /usr/local/ 1>/dev/null 2>&1
Expand Down
9 changes: 0 additions & 9 deletions third_party/bazel/tensorrt/trt.BUILD.tpl
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -191,14 +191,6 @@ cc_import(
visibility = ["//visibility:private"],
)

cc_import(
name = "proto_static_lib",
static_library = select({
":aarch64_linux": "lib/aarch64-linux-gnu/libprotobuf.a",
"//conditions:default": "lib/libprotobuf.a",
}),
visibility = ["//visibility:private"],
)

cc_library(
name = "nvonnxparser_headers",
Expand Down Expand Up @@ -245,7 +237,6 @@ cc_library(
"nvonnxparser_headers",
"nvonnxparser_static_lib",
"nvonnx_proto_static_lib",
"proto_static_lib",
],
)

Expand Down