Freeze the triton version in vllm-gaudi image to 3.1.0 (opea-project#1463)

The new triton version 3.2.0 can't work with vllm-gaudi. Freeze the triton version in vllm-gaudi image to 3.1.0.

Issue create for vllm-fork: HabanaAI/vllm-fork#732
Signed-off-by: chensuyue <suyue.chen@intel.com>

Author: chensuyue

.github/workflows/_example-workflow.yml

@@ -78,7 +78,8 @@ jobs:
               cd vllm && git rev-parse HEAD && cd ../
           fi
           if [[ $(grep -c "vllm-gaudi:" ${docker_compose_path}) != 0 ]]; then
-               git clone --depth 1 --branch v0.6.4.post2+Gaudi-1.19.0 https://github.com/HabanaAI/vllm-fork.git
+              git clone --depth 1 --branch v0.6.4.post2+Gaudi-1.19.0 https://github.com/HabanaAI/vllm-fork.git
+              sed -i 's/triton/triton==3.1.0/g' vllm-fork/requirements-hpu.txt
           fi
           git clone --depth 1 --branch ${{ inputs.opea_branch }} https://github.com/opea-project/GenAIComps.git
           cd GenAIComps && git rev-parse HEAD && cd ../

AgentQnA/tests/step1_build_images.sh

@@ -43,6 +43,7 @@ function build_vllm_docker_image() {
     fi
     cd ./vllm-fork
     git checkout v0.6.4.post2+Gaudi-1.19.0
+    sed -i 's/triton/triton==3.1.0/g' requirements-hpu.txt
     docker build --no-cache -f Dockerfile.hpu -t opea/vllm-gaudi:ci --shm-size=128g . --build-arg https_proxy=$https_proxy --build-arg http_proxy=$http_proxy
     if [ $? -ne 0 ]; then
         echo "opea/vllm-gaudi:ci failed"

AgentQnA/tests/test_compose_on_gaudi.sh

@@ -1,7 +1,6 @@
 #!/bin/bash
 # Copyright (C) 2024 Intel Corporation
 # SPDX-License-Identifier: Apache-2.0
-
 set -xe
 
 WORKPATH=$(dirname "$PWD")

ChatQnA/tests/test_compose_guardrails_on_gaudi.sh

@@ -29,6 +29,7 @@ function build_docker_images() {
     cd $WORKPATH/docker_image_build
     git clone --depth 1 --branch ${opea_branch} https://github.com/opea-project/GenAIComps.git
     git clone --depth 1 --branch v0.6.4.post2+Gaudi-1.19.0 https://github.com/HabanaAI/vllm-fork.git
+    sed -i 's/triton/triton==3.1.0/g' vllm-fork/requirements-hpu.txt
 
     echo "Build all the images with --no-cache, check docker_image_build.log for details..."
     service_list="chatqna-guardrails chatqna-ui dataprep retriever vllm-gaudi guardrails nginx"

ChatQnA/tests/test_compose_on_gaudi.sh

@@ -29,6 +29,7 @@ function build_docker_images() {
     cd $WORKPATH/docker_image_build
     git clone --depth 1 --branch ${opea_branch} https://github.com/opea-project/GenAIComps.git
     git clone --depth 1 --branch v0.6.4.post2+Gaudi-1.19.0 https://github.com/HabanaAI/vllm-fork.git
+    sed -i 's/triton/triton==3.1.0/g' vllm-fork/requirements-hpu.txt
 
     echo "Build all the images with --no-cache, check docker_image_build.log for details..."
     service_list="chatqna chatqna-ui dataprep retriever vllm-gaudi nginx"

ChatQnA/tests/test_compose_without_rerank_on_gaudi.sh

@@ -29,6 +29,7 @@ function build_docker_images() {
     cd $WORKPATH/docker_image_build
     git clone --depth 1 --branch ${opea_branch} https://github.com/opea-project/GenAIComps.git
     git clone --depth 1 --branch v0.6.4.post2+Gaudi-1.19.0 https://github.com/HabanaAI/vllm-fork.git
+    sed -i 's/triton/triton==3.1.0/g' vllm-fork/requirements-hpu.txt
 
     echo "Build all the images with --no-cache, check docker_image_build.log for details..."
     service_list="chatqna-without-rerank chatqna-ui dataprep retriever vllm-gaudi nginx"