From 0f5cb8cae15a76e8795e371f70a3bb87f7cc8bb5 Mon Sep 17 00:00:00 2001
From: Sai Enduri <saimanas.enduri@amd.com>
Date: Tue, 13 May 2025 01:49:33 -0700
Subject: [PATCH] Enable MI325X AMD CI. (#6259)

---
 .github/workflows/nightly-test-amd.yml |  5 ++++-
 .github/workflows/pr-test-amd.yml      | 30 ++++++++++++++++++++------
 2 files changed, 28 insertions(+), 7 deletions(-)

diff --git a/.github/workflows/nightly-test-amd.yml b/.github/workflows/nightly-test-amd.yml
index 5ea64f874..0a35d3aef 100644
--- a/.github/workflows/nightly-test-amd.yml
+++ b/.github/workflows/nightly-test-amd.yml
@@ -17,7 +17,10 @@ concurrency:
 jobs:
   nightly-test:
     if: github.repository == 'sgl-project/sglang' || github.event_name == 'pull_request'
-    runs-on: linux-mi300-gpu-2
+    strategy:
+      matrix:
+        runner: [linux-mi300-gpu-2, linux-mi325-gpu-2-nightly]
+    runs-on: ${{matrix.runner}}
     steps:
       - name: Checkout code
         uses: actions/checkout@v4
diff --git a/.github/workflows/pr-test-amd.yml b/.github/workflows/pr-test-amd.yml
index 24fb17961..dac584374 100644
--- a/.github/workflows/pr-test-amd.yml
+++ b/.github/workflows/pr-test-amd.yml
@@ -25,7 +25,10 @@ jobs:
   accuracy-test-1-gpu-amd:
     if: (github.repository == 'sgl-project/sglang' || github.event_name == 'pull_request') &&
       github.event.pull_request.draft == false
-    runs-on: linux-mi300-gpu-1
+    strategy:
+      matrix:
+        runner: [linux-mi300-gpu-1, linux-mi325-gpu-1]
+    runs-on: ${{matrix.runner}}
     steps:
       - name: Checkout code
         uses: actions/checkout@v4
@@ -65,7 +68,10 @@ jobs:
   accuracy-test-2-gpu-amd:
     if: (github.repository == 'sgl-project/sglang' || github.event_name == 'pull_request') &&
         github.event.pull_request.draft == false
-    runs-on: linux-mi300-gpu-2
+    strategy:
+      matrix:
+        runner: [linux-mi300-gpu-2, linux-mi325-gpu-2]
+    runs-on: ${{matrix.runner}}
     steps:
       - name: Checkout code
         uses: actions/checkout@v4
@@ -103,7 +109,10 @@ jobs:
   mla-test-1-gpu-amd:
     if: (github.repository == 'sgl-project/sglang' || github.event_name == 'pull_request') &&
       github.event.pull_request.draft == false
-    runs-on: linux-mi300-gpu-1
+    strategy:
+      matrix:
+        runner: [linux-mi300-gpu-1, linux-mi325-gpu-1]
+    runs-on: ${{matrix.runner}}
     steps:
       - name: Checkout code
         uses: actions/checkout@v4
@@ -141,7 +150,10 @@ jobs:
   performance-test-1-gpu-part-1-amd:
     if: (github.repository == 'sgl-project/sglang' || github.event_name == 'pull_request') &&
         github.event.pull_request.draft == false
-    runs-on: linux-mi300-gpu-1
+    strategy:
+      matrix:
+        runner: [linux-mi300-gpu-1, linux-mi325-gpu-1]
+    runs-on: ${{matrix.runner}}
     steps:
       - name: Checkout code
         uses: actions/checkout@v4
@@ -200,7 +212,10 @@ jobs:
   performance-test-1-gpu-part-2-amd:
     if: (github.repository == 'sgl-project/sglang' || github.event_name == 'pull_request') &&
         github.event.pull_request.draft == false
-    runs-on: linux-mi300-gpu-1
+    strategy:
+      matrix:
+        runner: [linux-mi300-gpu-1, linux-mi325-gpu-1]
+    runs-on: ${{matrix.runner}}
     steps:
       - name: Checkout code
         uses: actions/checkout@v4
@@ -248,7 +263,10 @@ jobs:
   bench-test-2-gpu-amd:
     if: (github.repository == 'sgl-project/sglang' || github.event_name == 'pull_request') &&
       github.event.pull_request.draft == false
-    runs-on: linux-mi300-gpu-2
+    strategy:
+      matrix:
+        runner: [linux-mi300-gpu-2, linux-mi325-gpu-2]
+    runs-on: ${{matrix.runner}}
     steps:
       - name: Checkout code
         uses: actions/checkout@v4